BULETINUL INSTITUTULUI POLITEHNIC DIN IAŞI - Universitatea ...
BULETINUL INSTITUTULUI POLITEHNIC DIN IAŞI - Universitatea ...
BULETINUL INSTITUTULUI POLITEHNIC DIN IAŞI - Universitatea ...
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
<strong>BULETINUL</strong><br />
<strong>INSTITUTULUI</strong><br />
<strong>POLITEHNIC</strong><br />
<strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
UNIVERSITATEA TEHNICĂ „GHEORGHE ASACHI”, <strong>IAŞI</strong><br />
Tomul LVI (LX)<br />
Fasc. 2<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
2010
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
PUBLISHED BY<br />
”GHEORGHE ASACHI” TECHNICAL UNIVERSITY OF <strong>IAŞI</strong><br />
Editorial Office: Bd. D. Mangeron 63, 700050, Iaşi, ROMANIA<br />
Tel. 40-232-278683; Fax: 40-232-237666; e-mail: polytech@mail.tuiasi.ro<br />
Editorial Board<br />
President: Prof. dr. eng. Ion Giurmă, Member of the Academy of Agricultural<br />
Sciences and Forest, Rector of the“Gheorghe Asachi” Technical University of Iaşi<br />
Editor-in-Chief: Prof. dr. eng. Carmen Teodosiu, Vice-Rector of the<br />
“Gheorghe Asachi” Technical University of Iaşi<br />
Honorary Editors of the Bulletin: Prof. dr. eng. Alfred Braier,<br />
Prof. dr. eng. Hugo Rosman,<br />
Prof. dr. eng. Mihail Voicu, Corresponding Member of the Romanian Academy,<br />
President of the “Gheorghe Asachi” Technical University of Iaşi<br />
Editors in Chief of the MACHINE CONSTRUCTION Section<br />
Prof. dr. eng. Radu Ibănescu, Assoc. Prof. dr. eng. Aristotel Popescu<br />
Honorary Editors: Prof. dr. eng. Gheorghe Nagîţ, Prof. dr. eng. Cezar Oprişan<br />
Associated Editor: Prof. dr. eng. Eugen Axinte<br />
Editorial Advisory Board<br />
Prof. dr. eng. Nicuşor Amariei, „Gheorghe Asachi” Prof. dr. eng. Nouraş-Barbu Lupulescu, University<br />
Technical University of Iaşi, Romania<br />
Transilvania of Braşov, Romania<br />
Assoc.Prof.dr.eng. Aristomenis Antoniadis, Technical Prof. dr. eng. Francisco Javier Santos Martin,<br />
University of Crete, Greece<br />
University of Valladolid, Spain<br />
Prof. dr. eng. Virgil Atanasiu, „Gheorghe Asachi”<br />
Technical University of Iaşi, Romania<br />
Prof. dr. eng.Fabio Miani, University of Udine, Italy<br />
Prof. dr. eng. Manuel San Juan Blanco, University of Prof. dr. eng.Mircea Mihailide, „Gheorghe Asachi”<br />
Valladolid, Spain<br />
Technical University of Iaşi, Romania<br />
Prof. dr. eng. Petru Berce, Technical University of Cluj Prof. dr. eng. Sevasti Mitsi, Aristotle University of<br />
Napoca, Romania<br />
Thessaloniki Salonic, Greece<br />
Prof. dr. eng. Ion Bostan, Technical University of Prof. dr. eng. Gheorghe Nagîţ, „Gheorghe Asachi”<br />
Chişinău, Repablic of Moldova<br />
Technical University of Iaşi, Romania<br />
Prof. dr. eng. Walter Calles, Hochschule für Technik und Prof. dr. eng.Vasile Neculăiasa, „Gheorghe Asachi”<br />
Wirtschaft des Saarlandes, Saarbrücken, Germany<br />
Technical University of Iaşi, Romania<br />
Prof. dr. eng. Doru Călăraşu, „Gheorghe Asachi” Prof. dr. eng. Dumitru Olaru, „Gheorghe Asachi”<br />
Technical University of Iaşi, Romania<br />
Technical University of Iaşi, Romania<br />
Prof. dr. eng. Francisco Chinesta, Ecole Centrale de Prof. dr. eng. Cezar Oprişan, „Gheorghe Asachi”<br />
Nantes, France<br />
Technical University of Iaşi, Romania<br />
Assoc.Prof.dr.eng. Conçalves Coelho, University Nova of Prof. dr. eng. Juan Pablo Contreras Samper,<br />
Lisbon, Portugal<br />
University of Cadiz, Spain<br />
Assoc.Prof.dr.eng. Mircea Cozmîncă, „Gheorghe Asachi” Prof. dr. eng.Loredana Santo, University „Tor Vergata”,<br />
Technical University of Iaşi, Romania<br />
Rome, Italy<br />
Prof. dr. eng. Spiridon Creţu, „Gheorghe Asachi” Prof. dr. eng.Cristina Siligardi, University of Modena,<br />
Technical University of Iaşi, Romania<br />
Italy<br />
Prof. dr. eng. Gheorghe Dumitraşcu, „Gheorghe Asachi” Prof. dr. eng.Fernando José Neto da Silva, University<br />
Technical University of Iaşi, Romania<br />
of Aveiro, Portugal<br />
Prof. dr. eng. Cătălin Fetecău, University „Dunărea de<br />
Jos” of Galaţi, Romania<br />
Prof. dr. eng. Filipe Silva, University of Minho, Portugal<br />
Prof. dr. eng. Mihai Gafitanu, „Gheorghe Asachi” Prof. dr. eng. Laurenţiu Slătineanu, Technical<br />
Technical University of Iaşi, Romania<br />
University of Iaşi, Romania<br />
Prof. dr. eng. Radu Gaiginschi, „Gheorghe Asachi” Lecturer dr.eng. Birgit Kjærside Storm, Aalborg<br />
Technical University of Iaşi, Romania<br />
Universitet Esbjerg, Denmark<br />
Prof.dr.ir.Dirk Lefeber, Vrije Universiteit Brussels, Belgium Prof. dr. eng. Ezio Spessa, Politecnico di Torino, Italy<br />
Prof. dr. eng.Dorel Leon, „Gheorghe Asachi” Technical Prof. dr. eng. Alexei Toca, Technical University of<br />
University of Iaşi, Romania<br />
Chişinău, Repablic of Moldova<br />
Prof. dr. eng.James A. Liburdy, Oregon State University, Prof. dr. eng.Roberto Teti, University „Federico II”,<br />
Corvallis, Oregon, SUA<br />
Naples, Italy<br />
Prof. dr. eng. dr. H.C. Peter Lorenz, Hochschule für Prof. dr. eng.Hans-Bernhard Woyand, Bergische<br />
Technik und Wirtschaft, Saarbrücken, Germany<br />
University Wuppertal, Germany
Papers presented at<br />
THE INTERNATIONAL CONFERENCE on<br />
DESIGN, TECNOLOGIES & MANAGEMENT<br />
IN MANUFACTURING<br />
Iaşi, May 14 th – 16 th , 2010<br />
organized by the<br />
FACULTY OF MACHINE MANUFACTURING &<br />
INDUSTRIAL MANAGEMENT<br />
Papers published with the support of<br />
NATIONAL AUTHORITY for SCIENTIFIC RESEARCHERS<br />
EDITORIAL BOARD<br />
MACHINE CONSTRUCTION<br />
Fascicle 2<br />
Conf.univ.dr.ing. Irina Cozmîncă<br />
Prof.univ.dr.ing. Radu Ibănescu<br />
Conf.univ.dr.ing. Vasile V. Merticaru
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
BULLETIN OF THE POLYTECHNIC INSTITUTE OF <strong>IAŞI</strong><br />
Publicat de<br />
UNIVERSITATEA TEHNICĂ „GHEORGHE ASACHI” <strong>DIN</strong> <strong>IAŞI</strong><br />
Tomul LVI(LX), Fasc. 2 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
S U M A R<br />
ANTÓNIO M. GONÇALVES-COELHO, GABRIELA NEŞTIAN și<br />
ANTÓNIO MOURÃO, Modelul matricei de proiectare în cazul<br />
soluțiilor redundante de proiectare (engl., rez. rom.)……........................<br />
TAXIARCHIS BELIS și ARISTOMENIS ANTONIADIS, Model pentru<br />
evaluarea uzurii dinților frezelor de danturat pe baza determinării<br />
așchiilor 3D (engl., rez. rom.). ..................................................................<br />
NIKOLAOS TAPOGLOU și ARISTOMENIS ANTONIADIS, Determinarea<br />
prin simulare CAD a componentelor forței de așchiere la frezarea<br />
danturilor (engl., rez. rom.)........................................................................<br />
EUGEN STRĂJESCU, CONSTANTIN DOGARIU, OLIMPIA PAVLOV și<br />
DUMITRU DUMITRU, Contribuţii privind controlul informatizat al<br />
cuţitelor roată (engl., rez. rom.).................................................................<br />
SILVIU BERBINSCHI, VIRGIL TEODOR, NICOLAE DUMITRAŞCU și<br />
NICOLAE OANCEA, Contribuţii la elaborarea unei metode grafice<br />
pentru profilarea sculelor care generează prin înfăşurare. I. Algoritm.<br />
(engl., rez. rom.) .......................................................................................<br />
Pag.<br />
1<br />
9<br />
21<br />
31<br />
41
SILVIU BERBINSCHI, VIRGIL TEODOR, NICOLAE DUMITRAŞCU și<br />
NICOLAE OANCEA, Contribuţii la elaborarea unei metode grafice<br />
pentru profilarea sculelor care generează prin înfăşurare. II. Aplicație<br />
pentru profilarea sculei-cremalieră (engl., rez. rom.)................................<br />
CĂTĂLIN FETECĂU, DANIEL-VIOREL VLAD și COSTEL MOCANU,<br />
Simularea procesului de strunjire folosind analiza cu element finit<br />
(engl., rez. rom.)........................................................................................<br />
MIRCEA COZMÎNCĂ, CRISTIAN CROITORU și CĂTĂLIN<br />
UNGUREANU, Cercetări experimentale pentru validarea unei noi<br />
metode de evaluare a forțelor de așchiere (engl., rez. rom.) .....................<br />
ANA-MARIA MATEI și MARIUS NICOLAE MILEA, Forțele de așchiere<br />
la frezarea frontală în funcție de forțele dezvoltate la nivelul unui dinte<br />
(engl., rez. rom.)........................................................................................<br />
MARIUS-IONUŢ RÎPANU, GHEORGHE NAGÎŢ, IOLANDA-ELENA<br />
MANOLE și ANDREI WEINGOLD, Aspecte comparative privind<br />
croirea la operaţia de decupare-perforare pe prese clasice și centre de<br />
presare prevăzute cu comandă numerică (engl., rez. rom.).......................<br />
IUSTINA ELENA ROTMAN, PETRU DUȘA și RADU ADRIAN BACIU<br />
LUPAȘCU, Considerații teoretice și experimentale privind<br />
determinarea efectului de divergență (engl., rez. rom.) ...........................<br />
CĂTĂLIN UNGUREANU, RADU IBĂNESCU și IRINA COZMÎNCĂ,<br />
Sistem de măsurare computerizat (engl., rez. rom.).................................<br />
BIRGIT KJÆRSIDE STORM, Lipirea cu adezivi a aluminiului tratat<br />
superficial (engl., rez. rom.)......................................................................<br />
IOANA PETRE, DAN PETRE, CRISTINA FILIP și LAVINIA NEAGOE,<br />
Aplicaţii industriale ale muşchilor pneumatici (engl., rez. rom.) .............<br />
MIHĂIȚĂ HORO<strong>DIN</strong>CĂ, Noi resurse ale cercetării experimentale asistate<br />
de calculator a puterii electrice absorbite în sistemele de fabricație<br />
(engl., rez. rom.)........................................................................................<br />
ION BOSTAN, VALERIU DULGHERU și ANATOL SOCHIREANU,<br />
Dezvoltarea integrată CAE a transmisiilor precesionale utilizând<br />
platforma Autodesk Inventor (engl., rez. rom.) .......................................<br />
ILEANA FULGA și EUGEN STRĂJESCU, Propuneri de optimizare a<br />
funcționării morilor fluidice cu jeturi în spirală (engl., rez. rom.)............<br />
DORU CĂLĂRAŞU, IRINA TIŢA, DAN SCURTU și BOGDAN<br />
CIOBANU, Analiza dinamică a mecanismului mecano-hidraulic de<br />
protecţie a turbinelor eoliene cu ax orizontal de mică putere prin<br />
basculare în plan vertical (engl., rez. rom.)...............................................<br />
CONSTANTIN CHIRIȚĂ, ADRIAN HANGANU și DANIEL CALFA,<br />
Cercetări privind sistemele hidraulice pentru deplasarea maselor mari<br />
pe distante mici, cu frecvenţă redusă (engl., rez. rom.)............................<br />
49<br />
57<br />
65<br />
75<br />
83<br />
91<br />
97<br />
105<br />
117<br />
125<br />
135<br />
143<br />
153<br />
161
EROL MURAD, CĂTALIN DUMITRESCU, GEORGETA HARAGA și<br />
LILIANA DUMITRESCU, Sistem de măsurare pneumatică a masei de<br />
apă extrasă în procesele de uscare convectivă (engl., rez. rom.)...............<br />
MIHAI FLORIN MĂNESCU și VALERIU PANAITESCU, Tehnologii<br />
folosite pentru monitorizarea defecţiunilor structurale apărute în<br />
funcţionarea grupurilor eoliene (engl., rez. rom.)......................................<br />
AURORA ALEXANDRESCU, A<strong>DIN</strong>A SIMONA ALEXANDRESCU și<br />
ADRIAN CONSTANTIN ALEXANDRESCU, Reabilitarea stației de<br />
pompare pentru alimentare cu apă (engl., rez. rom.).................................<br />
ILARE BORDEAȘU, MIRCEA OCTAVIAN POPOVICIU, DRAGOȘ<br />
NOVAC, LIVIU MARSAVINA, RADU NEGRU, MIRCEA VODĂ,<br />
VICTOR BĂLĂȘOIU și MARIAN BĂRAN, Contribuţii în evaluarea<br />
durabilităţii arborilor hidroagregatelor axiale orizontale (engl., rez.<br />
rom.) .........................................................................................................<br />
TEODOR MILOŞ, MIRCEA BĂRGLĂZAN, EUGEN DOBÂNDĂ,<br />
ADRIANA MANEA, RODICA BĂDĂRĂU și DANIEL STROIŢĂ,<br />
Traseul optim al unei conducte de aducţiune utilizând algoritmul<br />
Bellman-Kalaba (engl., rez. rom.) ............................................................<br />
DĂNUŢ ZAHARIEA și MIHAELA TUDORACHE, Analiza structurală a<br />
cuplajelor fixe de tip manşon cu ştifturi cilindrice (engl., rez. rom.) .......<br />
DĂNUŢ ZAHARIEA și MARIUS STACHIE, Analiza structurală a arcurilor<br />
bimetalice lamelare (engl., rez. rom.) .......................................................<br />
IRÈNE ALEXANDRESCU, HANS-JOACHIM FRANKE și THOMAS<br />
VIETOR, Managementul cunoașterii în configurarea produselor<br />
complexe personalizate în fazele incipiente ale dezvoltării acestora<br />
(engl., rez. rom.)........................................................................................<br />
PETRU DUŞA și IULIANA LAURA TARANOVSCHI, Cercetări cu privire<br />
la activitatea de inovare din mediul tehnic (engl., rez.<br />
rom.)..........................................................................................................<br />
ANDREI MIHALACHE, GHEORGHE NAGÎŢ și MARIUS-IONUŢ<br />
RÎPANU, Avantaje şi puncte slabe ale diferitelor tehnici de inginerie<br />
inversă (engl., rez. rom.) ...........................................................................<br />
ROBERTO LOPEZ, MANUEL SAN JUAN, FRANCISCO SANTOS,<br />
OSCAR MARTÍN și FLORIN NEGOESCU, Termografie aplicată în<br />
frezarea osoasă (engl., rez. rom.)...............................................................<br />
OLGA MARINA MONTES și VASILE V. MERTICARU, Studiu asupra<br />
oportunităţii unei noi fabrici de reciclare a sticlei, pentru o dezvoltare<br />
regională durabilă (engl., rez. rom.)..........................................................<br />
CĂTĂLIN DUMITRAȘ, CARMEN LOGHIN, SULEYMAN YALDIZ,<br />
MEHMET SAHIN și LUMINIȚA CIOBANU, Modelarea 3D a<br />
suprafețelor textile cu destinația optimizării curgerii fluidelor (engl.,<br />
rez. rom.)....................................................................................................<br />
167<br />
177<br />
189<br />
197<br />
205<br />
215<br />
221<br />
227<br />
237<br />
245<br />
251<br />
259<br />
267
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
BULLETIN OF THE POLYTECHNIC INSTITUTE OF <strong>IAŞI</strong><br />
Published by the<br />
„GHEORGHE ASACHI” TECHNICAL UNIVERSITY OF <strong>IAŞI</strong><br />
Tomul LVI(LX), Fasc. 2 2010<br />
Section<br />
MACHINE CONSTRUCTION<br />
C O N T E N T S<br />
ANTÓNIO M. GONÇALVES-COELHO, GABRIELA NEŞTIAN and<br />
ANTÓNIO MOURÃO, On the Pattern of the Design Matrix in<br />
Redundant Design Solutions (English, Romanian summary). ….............<br />
TAXIARCHIS BELIS and ARISTOMENIS ANTONIADIS, Hobbing Wear<br />
Prediction Model Based on 3D Chips Determination (English,<br />
Romanian summary).................................................................................<br />
NIKOLAOS TAPOGLOU and ARISTOMENIS ANTONIADIS, CAD-<br />
Based Calculation of Cutting Force Components in Gear Hobbing<br />
(English, Romanian summary)..................................................................<br />
EUGEN STRĂJESCU, CONSTANTIN DOGARIU, OLIMPIA PAVLOV<br />
and DUMITRU DUMITRU, Contributions Concerning The Computer<br />
Aided Control of the Fellows' Cutter (English, Romanian<br />
summary)...................................................................................................<br />
SILVIU BERBINSCHI, VIRGIL TEODOR, NICOLAE DUMITRAŞCU<br />
and NICOLAE OANCEA, Contributions to the Elaborations of a<br />
Graphical Method for Profiling of Tools which Generate by<br />
Enveloping. I. Algorithms (English, Romanian summary).......................<br />
Pag.<br />
1<br />
9<br />
21<br />
31<br />
41
SILVIU BERBINSCHI, VIRGIL TEODOR, NICOLAE DUMITRAŞCU<br />
and NICOLAE OANCEA, Contributions to the Elaborations of a<br />
Graphical Method for Profiling of Tools which Generate by<br />
Enveloping. II. Application for Rack-Gear Tool’s Profiling (English,<br />
Romanian summary).................................................................................<br />
CĂTĂLIN FETECĂU, DANIEL-VIOREL VLAD and COSTEL<br />
MOCANU, The Numerical Simulation of Turning Process using Finite<br />
Element Modeling (English, Romanian summary)...................................<br />
MIRCEA COZMÎNCĂ, CRISTIAN CROITORU and CĂTĂLIN<br />
UNGUREANU, Experimental Researches Regarding a New Method<br />
for Cutting Forces Evaluation (English, Romanian summary)................<br />
ANA-MARIA MATEI and MARIUS NICOLAE MILEA, Face Milling<br />
Forces Depending on the Forces Developed on a Single-Tooth (English,<br />
Romanian summary).................................................................................<br />
MARIUS-IONUŢ RÎPANU, GHEORGHE NAGÎŢ, IOLANDA-ELENA<br />
MANOLE and ANDREI WEINGOLD, Comparative Aspects<br />
Regarding the Nesting for Blanking-Punching Operation on Classical<br />
Presses and Numerical Commanded Pressing Centers (English,<br />
Romanian summary).................................................................................<br />
IUSTINA ELENA ROTMAN, PETRU DUȘA and RADU ADRIAN BACIU<br />
LUPAȘCU, Theoretical and Experimental Considerations on<br />
Determining the Effect of Divergence (English, Romanian<br />
summary)...................................................................................................<br />
CĂTĂLIN UNGUREANU, RADU IBĂNESCU and IRINA COZMÎNCĂ,<br />
Computerized Measurement System (English, Romanian summary).....<br />
BIRGIT KJÆRSIDE STORM, Adhesive Bonding of Surface Treated<br />
Aluminium (English, Romanian summary)..............................................<br />
IOANA PETRE, DAN PETRE, CRISTINA FILIP, and LAVINIA<br />
NEAGOE, Industrial Applications of the Pneumatic Muscles (English,<br />
Romanian summary).................................................................................<br />
MIHĂIȚĂ HORO<strong>DIN</strong>CĂ, Some New Resources on Computer Assisted<br />
Experimental Research of the Absorbed Electric Power in<br />
Manufacturing Systems (English, Romanian summary)...........................<br />
ION BOSTAN, VALERIU DULGHERU and ANATOL SOCHIREANU,<br />
Integrated CAE Development of Precessional Drives using Autodesk<br />
Inventor Platform (English, Romanian summary)....................................<br />
ILEANA FULGA and EUGEN STRĂJESCU, Proposals for the<br />
Improvement of the Fluidic Spiral Jetmills' Activity (English,<br />
Romanian summary).................................................................................<br />
DORU CĂLĂRAŞU, IRINA TIŢA, DAN SCURTU and BOGDAN<br />
CIOBANU, Dynamic Analysis of Mechanical-Hydraulic Protection<br />
Mechanisms of Low Power Horizontal Axis Wind Turbines through<br />
Vertical Tilting (English, Romanian summary)........................................<br />
49<br />
57<br />
65<br />
75<br />
83<br />
91<br />
97<br />
105<br />
117<br />
125<br />
135<br />
143<br />
153
CONSTANTIN CHIRIȚĂ, ADRIAN HANGANU and DANIEL CALFA,<br />
Research on Hydraulically Systems which Move Heavy Masses on<br />
Small Distances with Lower Frequencies (English, Romanian<br />
summary)...................................................................................................<br />
EROL MURAD, CĂTALIN DUMITRESCU, GEORGETA HARAGA and<br />
LILIANA DUMITRESCU, Pneumatic Metering System for Amount of<br />
Water Extracted in Convective Drying Processes (English, Romanian<br />
summary)...................................................................................................<br />
MIHAI FLORIN MĂNESCU and VALERIU PANAITESCU, Technologies<br />
for Monitoring Structural Damages Arising in the Functioning of Wind<br />
Turbines (English, Romanian summary)...................................................<br />
AURORA ALEXANDRESCU, A<strong>DIN</strong>A SIMONA ALEXANDRESCU and<br />
ADRIAN CONSTANTIN ALEXANDRESCU, Pumping Station<br />
Exoneration for Water Supply (English, Romanian summary).................<br />
ILARE BORDEASU, MIRCEA OCTAVIAN POPOVICIU, DRAGOS<br />
NOVAC, LIVIU MARSAVINA, RADU NEGRU, MIRCEA VODA,<br />
VICTOR BALASOIU and MARIAN BĂRAN, Contributions regarding<br />
Durability Evaluation of Horizontal Axial Hydraulic Turbines Shafts<br />
(English, Romanian summary)..................................................................<br />
TEODOR MILOŞ, MIRCEA BĂRGLĂZAN, EUGEN DOBÂNDĂ,<br />
ADRIANA MANEA, RODICA BĂDĂRĂU and DANIEL STROIŢĂ,<br />
Optimal Routes of Pipeline Supply using the Bellman-Kalaba<br />
Algorithm (English, Romanian summary)................................................<br />
DĂNUŢ ZAHARIEA and MIHAELA TUDORACHE, Structural Analysis<br />
of a Sleeve Rigid Coupling with Cylindrical Pins (English, Romanian<br />
summary)...................................................................................................<br />
DĂNUŢ ZAHARIEA and MARIUS STACHIE, Structural Analysis of a<br />
Bimetallic Strip Thermostat (English, Romanian summary)....................<br />
IRÈNE ALEXANDRESCU, HANS-JOACHIM FRANKE and THOMAS<br />
VIETOR, Knowledge Management for the Configuration in Early<br />
Phases of Complex Custom Products (English, Romanian summary)...<br />
PETRU DUŞA and IULIANA LAURA TARANOVSCHI, Researches<br />
Regarding to Innovative Activity in Technical Environment (English,<br />
Romanian summary).................................................................................<br />
ANDREI MIHALACHE, GHEORGHE NAGÎŢ and MARIUS-IONUŢ<br />
RÎPANU, Advantages and Weak Points of Different Reverse<br />
Engineering (RE) Techniques (English, Romanian summary).................<br />
ROBERTO LOPEZ, MANUEL SAN JUAN, FRANCISCO SANTOS,<br />
OSCAR MARTÍN and FLORIN NEGOESCU, Thermography Applied<br />
to Bone Drilling (English, Romanian summary).......................................<br />
OLGA MARINA MONTES and VASILE V. MERTICARU, Study on the<br />
Opportunity of a New Glass Recycling Factory for Regional<br />
Sustainable Development (English, Romanian summary)........................<br />
161<br />
167<br />
177<br />
189<br />
197<br />
205<br />
215<br />
221<br />
227<br />
237<br />
245<br />
251<br />
259
CĂTĂLIN DUMITRAȘ, CARMEN LOGHIN, SULEYMAN YALDIZ,<br />
MEHMET SAHIN și LUMINIȚA CIOBANU, Modeling 3D Surface of<br />
Textile Structures for Fluid Flow Improvement (English, Romanian<br />
summary)...................................................................................................<br />
267
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
ON THE PATTERN OF THE DESIGN MATRIX IN<br />
REDUNDANT DESIGN SOLUTIONS<br />
BY<br />
ANTÓNIO M. GONÇALVES-COELHO 1 , GABRIELA NEŞTIAN 2 ,<br />
and ANTÓNIO MOURÃO 1<br />
Abstract. Axiomatic Design was created with the aim of building a<br />
systematic model for engineering education and practice, taking into<br />
account the initial hypothesis that there are fundamental principles that<br />
govern good design practice. According to this design theory, the design<br />
solutions can be classified as uncoupled, decoupled or coupled, depending<br />
on the way their design matrices are populated. Uncoupled solutions are the<br />
best, decoupled solutions are acceptable, and coupled solutions are poor<br />
design and should be avoided. Redundant designs make up a specific class<br />
of design solutions, in which the number of functional requirements is<br />
lesser than the number of design parameters. This paper discusses how the<br />
design matrix could be populated, so that a redundant design could be either<br />
uncoupled or decoupled.<br />
Keywords: Axiomatic Design, Design Matrix, Redundant Design, Ideal<br />
Design.<br />
1. Introduction<br />
Axiomatic Design (AD) was created in the late 1970’s by Nam P. Suh with<br />
the aim of building a systematic model for engineering education and practice<br />
under the initial hypothesis that there are fundamental principles that govern<br />
good design practice [1].<br />
According to AD, any “design object” — being it a product, a process or any<br />
other technical system — can be described by a vector in each one of four<br />
design domains (see Fig. 1). The design process starts at the customer domain<br />
with the definition of the customer needs (CNs). Mapping between the customer
2 Antonio M. Gonçalves-Coelho et al.<br />
and the functional domains allows finding the functional requirements (FRs).<br />
Once this is done, another mapping translates the FRs into design parameters<br />
(DPs), i.e. the set of properties that physically describe the design object.<br />
Finally, mapping from the physical to the process domain leads to the process<br />
variables (PVs), which outline how to make the design object [1].<br />
Fig. 1 – The Design Domains [1].<br />
The left-to-right mapping between any two contiguous domains can be<br />
represented by a “design equation” of the form:<br />
(1)<br />
{} Y = [] A {} X ; Aij =<br />
∂Yi ∂X j<br />
; i = 1, ..., m; j = 1,...,n ,<br />
where {Y} is a vector that represents the set of m requirements that should be<br />
accomplished, {X} is a vector representing the set of n parameters of the design<br />
object that is expected to fulfil the requirements, and [A] is the design matrix.<br />
Usually, any prospective design equation is bounded by constraints [1].<br />
Eq. (1) is not unique, and different {X} vectors would represent different<br />
design solutions that are characterized by distinct design matrices, which<br />
patterns would make the difference between “good” and “poor” design. The<br />
good or the poor quality of any design solution is ruled by the Independence<br />
Axiom, which states that, in good design, the selected parameters {X} should be<br />
such that the requirements {Y} are fulfilled independently. As a result, the ideal<br />
design solution should have the same number of requirements and parameters<br />
(m = n) and the design matrix should be diagonal, case of which the design<br />
solution is called “uncoupled” [1]. A triangular design matrix is acceptable as<br />
well and corresponds to a “decoupled” design [1]. Any other pattern of square<br />
design matrix corresponds to a “coupled” design, which should be recognized as<br />
poor and as such should be avoided [1].<br />
For any design solution where m > n, AD’s Theorem 1 states that either the<br />
design is coupled or some of its FRs can never be fulfilled [1]. An example of<br />
such a design can be found elsewhere [2]. In the case of m < n, AD’s Theorem 3<br />
states that the design is either redundant or coupled [1]. At last, the specific case<br />
of a design with a single requirement (m = 1) is worth to mention. In this case,<br />
the design would be either uncoupled (if n = 1), or redundant (if n > 1). In fact,
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 3<br />
the decoupled and the coupled conditions are impossible to attain in any singlerequirement<br />
design because there is no different FRs to couple.<br />
The present work contains an analysis of the pattern of the design matrix of<br />
redundant designs with more than one requirement (m > 1), a matter to which<br />
the researchers have not paid sufficient attention so far.<br />
2. The Key Characteristics of Redundant Designs<br />
Fig. 2 depicts the design of a simple clamping device with one only<br />
customer need: the clamping action, which can be attained by adjusting one<br />
only FR — the distance d.<br />
Fig. 2 – Example of a simple redundant design.<br />
The adjustment of d can be achieved by suitably setting the values of two<br />
design parameters: the position of the cam in the end of the hand lever (see Fig.<br />
1), which is denoted by angle α, and the angular position of the threaded rod,<br />
which is represented by angle β. Thus, the design equation for this redundant<br />
design with one requirement and two parameters is:<br />
(2) d<br />
⎡<br />
⎣<br />
{}= A 11 A 12<br />
⎧<br />
⎤<br />
⎪ α ⎫⎪<br />
⎦ ⎨ ⎬ .<br />
β<br />
⎩⎪ ⎭⎪<br />
A more general case is the Eq. (3) that represents a design with arbitrary<br />
numbers of requirements m and parameters n, with m < n. The equation relates<br />
to a redundant design, and its quality depends on the pattern of the design<br />
matrix, as per Theorem 3.<br />
To better understand how the pattern of the design matrix of Eq. (3) could<br />
characterize a good or a poor design, let us consider the three coexisting designs
4 Antonio M. Gonçalves-Coelho et al.<br />
represented by Eq. (4), Eq. (5) and Eq. (6), where Akj denotes the possible nonzero<br />
elements of the related design matrices.<br />
(3)<br />
(4)<br />
(5)<br />
⎧<br />
⎪<br />
⎨<br />
⎪<br />
⎩⎪<br />
⎧<br />
⎪<br />
⎨<br />
⎪<br />
⎩⎪<br />
FR 1<br />
FR 2<br />
FR 3<br />
FR 1<br />
FR 2<br />
FR 3<br />
⎧<br />
⎪<br />
⎪<br />
⎪<br />
⎫ ⎡<br />
⎪ ⎢<br />
⎬ = ⎢<br />
⎪ ⎢<br />
⎭⎪<br />
⎣⎢<br />
A11 A21 A31 A12 A22 A32 A13 A23 A33 0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
⎤ ⎪<br />
⎥ ⎪<br />
⎥ ⎨<br />
⎥ ⎪<br />
⎦⎥<br />
⎪<br />
⎪<br />
⎪<br />
⎪<br />
⎩<br />
⎧<br />
⎪<br />
⎪<br />
⎪<br />
⎫ ⎡<br />
⎪ ⎢<br />
⎬ = ⎢<br />
⎪ ⎢<br />
⎭⎪<br />
⎣⎢<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
A14 A24 A34 A15 A25 A35 A16 A26 A36 0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
⎤ ⎪<br />
⎥ ⎪<br />
⎥ ⎨<br />
⎥ ⎪<br />
⎦⎥<br />
⎪<br />
⎪<br />
⎪<br />
⎪<br />
⎩<br />
DP 1<br />
DP 2<br />
DP 3<br />
DP 4<br />
DP 5<br />
DP 6<br />
DP 7<br />
DP 8<br />
DP 1<br />
DP 2<br />
DP 3<br />
DP 4<br />
DP 5<br />
DP 6<br />
DP 7<br />
DP 8<br />
⎫<br />
⎪<br />
⎪<br />
⎪<br />
⎪<br />
⎪<br />
⎬ ,<br />
⎪<br />
⎪<br />
⎪<br />
⎪<br />
⎪<br />
⎭<br />
⎫<br />
⎪<br />
⎪<br />
⎪<br />
⎪<br />
⎪<br />
⎬ ,<br />
⎪<br />
⎪<br />
⎪<br />
⎪<br />
⎪<br />
⎭
(6)<br />
⎧<br />
⎪<br />
⎨<br />
⎪<br />
⎩⎪<br />
FR 1<br />
FR 2<br />
FR 3<br />
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 5<br />
⎧<br />
⎪<br />
⎪<br />
⎪<br />
⎫ ⎡<br />
⎪ ⎢<br />
⎬ = ⎢<br />
⎪ ⎢<br />
⎭⎪<br />
⎣⎢<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
0<br />
A17 A27 A37 A18 A28 A38 ⎤ ⎪<br />
⎥ ⎪<br />
⎥ ⎨<br />
⎥ ⎪<br />
⎦⎥<br />
⎪<br />
⎪<br />
⎪<br />
⎪<br />
⎩<br />
One can see that DP1, DP2 and DP3 are the only possible contributing<br />
parameters in Eq. (4). As for the design of Eq. (5), just DP4, DP5 and DP6 are<br />
significant. At last, in what concerns to Eq. (6), only DP7 and DP8 contribute. In<br />
other words, each one of the above-defined coexisting designs can fulfil all the<br />
FRs of Eq. (3) using entirely different subsets of the DPs included in the latter<br />
equation, in such a manner that all the DPs are taken into account.<br />
Now, one can figure out that the design of Eq. (3) could be achieved by<br />
merging the designs of Eq. (4), Eq. (5) and Eq. (6).<br />
The design matrices of Eq. (4) and Eq. (5) could be reduced to block<br />
matrices of size m x m by eliminating the zero elements of their design matrices.<br />
Therefore, the surrogate “condensed design equations” could be obtained<br />
from those equations by eliminating the non-relevant DPs, and by using the all<br />
the existing FRs and the non-zero block matrices. For example, Eq. (7) is the<br />
condensed equation that is obtained from Eq. (5).<br />
The same treatment could be done to Eq. (6), but in this case we would<br />
obtain a non-square block matrix of size (n mod m) x m.<br />
⎧ FR ⎫ ⎡<br />
1 A14 A15 A ⎤ ⎧<br />
16 DP ⎫<br />
4<br />
⎪ ⎪ ⎢<br />
⎥ ⎪ ⎪<br />
(7)<br />
⎨ FR2 ⎬ = ⎢ A24 A25 A26 ⎥ ⎨ DP5 ⎬ .<br />
⎪ FR ⎪ ⎢<br />
⎩⎪<br />
3 ⎭⎪<br />
A34 A35 A<br />
⎥ ⎪<br />
⎣⎢<br />
36 ⎦⎥<br />
DP ⎪<br />
⎩⎪<br />
6 ⎭⎪<br />
As a result, the redundant design of Eq. (3) could be considered “suitable<br />
design” if the condensed equations obtained from Eq. (4), Eq. (5) and Eq. (6)<br />
are not coupled. The coupled condition is excluded in Eq. (4) and Eq. (5) if the<br />
relevant DPs are chosen so that their non-zero block matrices are either<br />
triangular or diagonal.<br />
As for Eq. (6), the coupled condition is excluded if its non-zero block<br />
matrix is populated in such a manner that the condensed equations correspond<br />
to uncoupled or decoupled designs. If this is not the case, one can selectively<br />
“freeze” as many DPs as required, so that the condensed design become<br />
uncoupled or decoupled, as exemplified elsewhere [3].<br />
DP 1<br />
DP 2<br />
DP 3<br />
DP 4<br />
DP 5<br />
DP 6<br />
DP 7<br />
DP 8<br />
⎫<br />
⎪<br />
⎪<br />
⎪<br />
⎪<br />
⎪<br />
.<br />
⎬<br />
⎪<br />
⎪<br />
⎪<br />
⎪<br />
⎪<br />
⎭
6 Antonio M. Gonçalves-Coelho et al.<br />
4. Conclusion<br />
The key conclusion of this paper can be summarized as a new theorem: Let<br />
us suppose a design with m requirements and n parameters, with m < n. Its<br />
design matrix can be partitioned in (n div m) square block matrices of size<br />
m x m, and an extra non-square block matrix of size (n mod m) x m, in such a<br />
manner that the block matrices have not common elements. Such a design is<br />
acceptable if each block matrix describes either an uncoupled or a decoupled<br />
design. Otherwise, the design is of the coupled type.<br />
Received: January, 30, 2010 1 UNIDEMI, Universidade Nova de Lisboa,<br />
Faculdade de Ciências e Tecnologia<br />
Departamento de Engenharia Mecânica e Industrial<br />
Monte de Caparica, Portugal<br />
e-mail: goncalves.coelho@fct.unl.pt<br />
ajfm@fct.unl.pt<br />
2 Ministerul Comunicaţiilor şi Societăţii Informaţionale,<br />
Bucureşti, România<br />
e-mail: gabriela.nestian@mcsi.ro<br />
R E F E R E N C E S<br />
1. S u h N. P., The Principles of Design, Oxford Univ. Press, N. Y., 1990.<br />
2. G o n ç a l v e s – C o e l h o A. M., M o u r ã o A. J. F., Axiomatic Design: The<br />
Meaning of the First Axiom. In: A. Toca, O. Pruteanu (Eds.), Tehnologii<br />
Moderne, Calitate, Restructurare: Culegere de Lucrări Ştiinţifice, 3, pp. 341-344,<br />
<strong>Universitatea</strong> Tehnică a Moldovei, Chişinău, 2003.<br />
3. F r a d i n h o J., G o n ç a l v e s – C o e l h o A. M., M o u r ã o A. J. F., An<br />
axiomatic design approach for the cost optimisation of industrial coupled<br />
designs: A case study. In: Gonçalves-Coelho A.M. (Ed.), Proc. 5 th International<br />
Conference on Axiomatic Design - ICAD 2009, pp. 201-207, Campus de<br />
Caparica, 2009.<br />
MODELUL MATRICEI DE PROIECTARE IN CAZUL SOLUTIILOR<br />
REDUNDANTE DE PROIECTARE<br />
(Rezumat)<br />
Proiectarea axiomatică a apărut din necesitatea creării unui model sistematic de<br />
educare şi practică inginerească, pornind de la principiile fundamentale de bune practici<br />
ale proiectării. Conform acestei teorii, în funcţie de modul de populare al matricelor de<br />
proiectare, soluţiile de proiectare pot fi clasificate în soluţii necuplate, decuplate sau<br />
cuplate. Soluţiile de tip necuplat sunt cele mai bune, soluţiile de tip decuplat sunt
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 7<br />
acceptabile iar soluţiile cuplate nu sunt satisfăcătoare din punct de vedere tehnic şi ar<br />
trebui evitate. Matricele diagonale corespund soluţiilor necuplate, matricele<br />
triunghiulare corespund soluţiilor decuplate iar toate celelalte tipuri de matrice pătratice<br />
corespund soluţiilor cuplate. În cazul în care numărul cerinţelor funcţionale este mai<br />
mare decât numărul parametrilor de proiectare, unele cerinţe nu pot fi îndeplinite<br />
niciodată şi soluţiile sunt de tip cuplat. Proiectarea redundantă creează o clasă specifică<br />
de soluţii de proiectare în care numărul cerinţelor funcţionale este mai mic decât<br />
numărul de parametri de proiectare.<br />
Această lucrare prezintă modalităţi de populare ale matricei de proiectare, astfel<br />
încât un proiect de tip redundant să poată deveni de tip necuplat sau decuplat.<br />
In concluzie putem formula o nouă teoremă: Presupunând că avem un produs sau un<br />
proces a cărui matrice de proiectare are m cerinţe şi n parametri, unde m < n, atunci<br />
matricea de proiectare poate fi descompusă în (n div m) submatrici pătratice de ordinul<br />
m x m şi o submatrice nepătratică de ordinul (n mod m) x m, astfel încât submatricile să<br />
nu aibă elemente comune. O astfel de soluţie este acceptabilă dacă fiecare bloc descrie o<br />
soluţie necuplată sau decuplată, dacă nu, atunci soluţia este de tip cuplat.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
HOBBING WEAR PREDICTION MODEL<br />
BASED ON 3D CHIPS DETERMINATION<br />
BY<br />
TAXIARCHIS BELIS and ARISTOMENIS ANTONIADIS<br />
Abstract. Gear hobbing is a machining process widely used in the industry<br />
for massive production of external gears. In this process the variant chip formation<br />
on each generating position causes uneven wear distribution on the hob teeth. This<br />
paper presents a new method for the determination of wear parameters based on<br />
3D chips. The 3D chip characteristics are fed to an existing wear model in order to<br />
calculate the hob wear distribution more precisely.<br />
Key words: gear hobbing, tool wear, simulation<br />
1. Introduction<br />
Gear hobbing is one of the most efficient generating processes for cutting<br />
external cylindrical gears. Although, hobbing cutters are still quite expensive<br />
due to their complex geometry. Thus their extensive exploitation is very<br />
important for industry. In gear hobbing the chip formation varies for each<br />
cutting tooth due to the fact that every tooth always cuts the same generating<br />
position. As a result, different wear laws are developed leading to uneven wear<br />
distribution on the hob teeth. Thus, the need to adopt an effective wear<br />
prediction model in gear hobbing arises.<br />
In this paper 3D chip determination is studied in order to feed more accurate<br />
data to an existing wear prediction model. Moreover total wear distribution on<br />
the hob teeth can be calculated. This information can be used to optimize<br />
tangential shifting conditions in order to maximize tool life and prolong the<br />
time interval before hob cutter resharpening.<br />
Figure 1 presents the basic nomenclature of the hob cutter that has been<br />
used in the wear simulation program developed in the present work. As it can be<br />
seen in the upper part of the figure, three distinct motions are required by this<br />
cutting method those being the workpiece revolution, the tool rotation, and the
10 Taxiarchis Belis and Aristomenis Antodiadis<br />
tool tangential displacement. Considering the above kinematic, two different<br />
hobbing types may be applied according to the direction of the axial feed, the<br />
climb and the up-cut hobbing respectively. In the present paper the tooth’s<br />
profile generation complies with <strong>DIN</strong> 3972 [1].<br />
Fig. 1 – Basic kinematics and essential parameters of gear hobbing.<br />
2. State of the Art<br />
As mentioned before, wear prediction in gear hobbing is a challenging task<br />
because of the large amount of wear influencing data and the complex<br />
kinematics of the process. Many studies have been published, introducing<br />
different methods for wear determination. These studies can fall under two<br />
categories. The first includes methods based on wear calculation in individual<br />
generating positions [2], [3].These methods are based on the simulated chip<br />
geometry and cutting conditions. Lately, FEM-based methods for wear<br />
determination have been introduced [4]. These methods take into account<br />
occurring stresses in the tool-chip contact areas. Nevertheless a FEM-based<br />
simulation needs a considerable amount of time before it can be realised.<br />
The wear model presented in [2] is used in the present paper,. The<br />
contribution of the proposed method is the precise 3D chips determination
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 11<br />
because the chip characteristics can be calculated with the accuracy provided by<br />
a CAD system.<br />
3. Hob Wear Simulation Process<br />
In order to achieve uniform wear distribution in the majority of the cutting<br />
teeth, a hob tangential shifting is required. In the left part of Figure 2, simulated<br />
chips and the corresponding gaps formed at the indicated generating positions,<br />
are illustrated. These chips are categorized into groups according to their shape<br />
and the cutting direction of the process. In the right part of Figure 2 the wear<br />
laws of several teeth after tangential shifts can be observed.<br />
Fig. 2 –Determination of the flank wear at individual hob teeth<br />
considering the shift data.
12 Taxiarchis Belis and Aristomenis Antodiadis<br />
In the first column the wear laws of teeth number 1, 2, 3 and 4 and the total<br />
wear distribution on the hob after a certain number of cuts are presented. The<br />
second column shows the wear laws after the first tangential displacement equal<br />
to the tool axial pitch ε. After this sifting the cutting tooth i is placed at the<br />
generating position of the tooth i+1. Thus the tooth number 1 that was cutting in<br />
the generating position number 1 quits, whereas the tooth number 5 cuts for the<br />
first time in the generating position 4.<br />
For example, tooth number 4 was cutting initially in the generating position<br />
4. After AS* number of cuts, tooth 4 reaches a certain flank wear depth. In the<br />
first tangential shift, tooth 4 cuts in the generating position 3, obeying the wear<br />
law that governs that generating position. In this position, flank wear increases<br />
with a different rate for another AS* number of cuts. Finally, in the last<br />
tangential shift, tooth 4 cuts in generating position number 2, obeying a<br />
different wear law.<br />
As a result, in the bottom side of figure 2, the normalized flank wear<br />
distribution on the hob after two tangential displacements is presented.<br />
Consequently the wear distribution becomes uniform and the tool exploitation is<br />
enhanced.<br />
Fig. 3 – The effect of chip geometry and shape on the tool wear development.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 13<br />
3.1. HOBWEAR Formulation<br />
Figure 3 presents wear laws for different chip groups. There are five chip<br />
groups concerning the chip flow obstruction intensity, introduced by [3]. The<br />
determination of the chip group involves prior knowledge of the cutting<br />
direction and the examined gear flank. As it can be easily noticed, the wear rate<br />
in group I is more intense. On the other hand, cutting with chip group 0 causes<br />
less flank tooth wear and the achieved number of cuts increases.<br />
This can be explained by the fact that in group I there is intense chip flow<br />
obstruction. In the region of the tooth head chip flow obstruction phenomena<br />
are the most intense due to collision of chip distinct sections. In the bottom side<br />
of the figure the included equivalent chip thickness and length are further<br />
explained.<br />
Figure 4 presents the general structure of the program developed. The left<br />
part of the figure shows the categorized data input. Data such as chip group,<br />
equivalent thickness and length are loaded into the program from an external<br />
source. The undeformed chip geometries are calculated in various generating<br />
and revolving positions with the aid of HOB3D [2], [3].<br />
Fig. 4 – HOBWEAR simulation algorithm.<br />
The wear behaviour on the hob teeth is primarily influenced by two sets of<br />
parameters. The first set refers to machining data and the geometry of hob and
14 Taxiarchis Belis and Aristomenis Antodiadis<br />
gear. The second set refers to the machine tool, the cutting and working material<br />
combination and the used cooling lubricant. Every possible combination of the<br />
above parameters results in different coefficients in the model. Wear<br />
coefficients are experimentally defined for a variety of materials and geometries<br />
widely used in gear industry.<br />
The middle column of the figure shows the flow chart for wear calculation<br />
and optimal tangential displacement determination. The program has the ability<br />
to calculate wear progress at all generating positions including transient areas.<br />
In order to shift the hob cutter, all the group gears have to be processed. At the<br />
bottom of the figure the outputs of the program are presented. The output<br />
includes wear progress graphs for all generating positions, and total wear<br />
distribution graphs for all tangential displacements. The developed program also<br />
calculates the maximum flank wear and the corresponding number of cuts.<br />
3.2. 3D Chips Determination<br />
As illustrated on figure 5, this paper introduces a novel way of calculating<br />
equivalent chip dimensions. The equivalent chip thickness calculation is more<br />
accurate due to the fact that is measured on the 3D chip’s cross-sections. Detail<br />
A in the upper right part of the figure, shows the chip cross-section in plane 1.<br />
The calculation of thickness in tooth’s head and in arc distance a f<br />
) is hereby<br />
determined. The equation predicting the wear progress in the individual<br />
generating positions and the equations of equivalent chip dimensions introduced<br />
in [2] are presented in the bottom side of the figure.<br />
Fig. 5 – Determination of equivalent chip dimensions.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 15<br />
The wear determination model includes two stages. The first stage involves<br />
the employment of hobbing data in order to calculate the chip thickness, length<br />
and group for every cutting position of all generating positions. Categorization<br />
of chips in groups is determined by the program FRSWEAR [3].<br />
Fig. 6 – Calculations of the equivalent chip thickness and length based on 3D chips.<br />
The measurement of the total length of the chip and the length compressed<br />
on the head’s corner are quite precise. As shown in the left bottom side of figure<br />
6, the first step for measuring the chip’s length compressed on the tooth’s<br />
head’s corner, is to cut the section of the chip from the left flank to the tool head<br />
center H, as the hob’s profile is revolving. In this cross-section we can easily
16 Taxiarchis Belis and Aristomenis Antodiadis<br />
and precisely measure the chip’s thickness and length in head hH and lH,TF in<br />
CAD environment. This data is fed to the equation illustrated in figure 5 in<br />
order to calculate the arc length a f<br />
) . Determination of a f<br />
) is very important due<br />
to the fact that maximum wear depth has been observed by researchers in this<br />
region [2], [3], [7]-[9]. The calculation method of chip thickness hf is similar to<br />
hH. After hs and lH are determined for every generating position, data can be fed<br />
to the wear prediction model. All the calculations above can be simplified by<br />
the use of parametric design in CAD environment. In the first step the arc a f<br />
) is<br />
set as parameter. In the next step the section of the chip from point K to F as the<br />
hob profile revolves is cut. In the chip produced hH and lH,TF can be easily<br />
measure and the arc a f<br />
) calculated precisely. The parameter is set at the<br />
calculated value and the chips regenerated automatically. Thus, chip thickness<br />
hf can be easily measured. The final step includes calculation of the equivalent<br />
thickness hs. The upper part of figure 6 presents the variation of the chip<br />
thickness for all the revolving planes in several generating positions. These<br />
diagrams have been generated from the sequence of the chip thickness results<br />
hH as mentioned above.<br />
3.3. Development of the Wear Prediction Program<br />
The gear flank wear determination program is implemented in MATLAB<br />
high-level matrix array language.<br />
Fig. 7 – The Graphical User Interface of the wear simulation program HOBWEAR.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 17<br />
The user has the ability to select gear and hob material from a variety of the<br />
most popular ones in industry. This selection changes the wear coefficients used<br />
to determine the flank wear. An automatic report generator has been<br />
incorporated for better organization of the program results.<br />
4. Simulation Results<br />
Wear graphs for each individual generating position is the first output of the<br />
program developed, as presented in figure 8. As it can be seen, teeth number 10,<br />
11 and 12 have almost reached the wear limit of 0,6 mm. For wear width below<br />
0,2 mm the relation between number of cuts and flank wear is linear.<br />
The normalization progress of total wear distribution on the hob cutter from<br />
tangential shifting number 5 to 8 is illustrated in Figure 9. Optimization of<br />
tangential shifting in gear hobbing has been studied by [10]. The credibility of<br />
the above wear prediction model has been verified with the aid of a wide variety<br />
of cutting experiments [2], [3].<br />
Figure 10 presents the wear distribution on the cutting teeth of a hob cutter<br />
for two cases. In both cases, 18 gears of the same width have been cut. In the<br />
first case, 3 tangential shifts took place with 1 ε displacement per shift. In the<br />
second case, 2 tangential shifts took place with 2 ε displacement per shift. It is<br />
evident that in the second case the maximum wear depth on the hob cutter is<br />
less than in the first one.<br />
Fig. 8 – Wear progress in individual generating positions.
18 Taxiarchis Belis and Aristomenis Antodiadis<br />
Fig. 9 – Total wear distribution on the hob cutter after shifting.<br />
As it can be seen in the second case, the wear distribution is more<br />
uniform due to the fact that more teeth are involved in the cut. Authors in [10]<br />
have studied the optimal selection of the shift displacement and number of<br />
shifts and introduced a nomogram for various shifting conditions.<br />
Fig. 10 – Cutting of 18 gears with different tangential displacement parameters.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 19<br />
5. Conclusion<br />
This paper suggests a model which simulates the wear progress on the<br />
hobbing cutter. The end-user has the ability to optimize tangential shifting<br />
conditions, owing to this model, in order to minimize the total gear<br />
manufacturing cost.<br />
Acknowledgements. The authors wish to thank the Research Committee of the<br />
Technical University of Crete for their financial support (via basic research project<br />
2009).<br />
Received:March 10, 2010 Technical University of Crete,<br />
Department of Production Engineering & Management<br />
Chania, Crete, Greece<br />
e-mail: tbelis@isc.tuc.gr<br />
antoniadis@dpem.tuc.gr<br />
R E F E R E N C E S<br />
1. D I N 3972, Bezugsprofile von Verzahnwerkzeugen fuer Evolventenverzahnungen<br />
Nach <strong>DIN</strong> 867“, 1981.<br />
2. B o u z a k i s K., K o m p o g i a n n i s S., A n t o n i a d i s A., V i d a k i s N., Gear<br />
Hobbing Cutting Process Simulation and Tool Wear Prediction Models. ASME<br />
Journal of Manufacturing Science and Engineering, 124, 1, 42-51 (2002).<br />
3. B o u z a k i s K. D., Konzept und technologishe Grundlagen zur automatisierten<br />
Erstellung optimaler Bearbeitungsdaten beim Waelzfraesen, Habilitation, TH<br />
Aachen, 1980.<br />
4. F r i d e r i k o s O., Simulation of Chip Formation and Flow in Gear Hobbing Using<br />
the Finite Element Method, Ph.D. Thesis, Aristoteles University of Thessaloniki,<br />
Greece, 2008.<br />
5. D i m i t r i o u V., V i d a k i s, N., A n t o n i a d i s A., Advanced Computer Aided<br />
Design Simulation of Gear Hobbing by Means of 3-Dimensional Kinematics<br />
Modeling, ASME Journal of Manufacturing Science and Engineering, 129, 911-<br />
918., (2007).<br />
6. D i m i t r i o u V., A n t o n i a d i s A., CAD-based Simulation of the Hobbing<br />
Process for the Manufacturing of Spur and Helical Gears, International Journal<br />
of Advanced Manufacturing Technology, 41, 3-4, 347-357 (2008).<br />
7. A n t o n i a d i s A., Determination of the Impact Tool Stresses During Gear<br />
Hobbing and Determination of Cutting Forces During Hobbing of Hardened<br />
Gears, Ph.D. Thesis, Aristoteles University of Thessaloniki, 1989.<br />
8. A n t o n i a d i s A., V i d a k i s N., B i l a l i s, N., Fatique Fracture Investigation of<br />
Cemented Carbide Tools in Gear Hobbing. Part 1: FEM Modeling of Fly
20 Taxiarchis Belis and Aristomenis Antodiadis<br />
Hobbing and Computational Interpretation of Experimental Results. ASME<br />
Journal of Manufacturing Science and Engineering, 124, 4, 784-791, (2002).<br />
9. A n t o n i a d i s A., V i d a k i s N., B i l a l i s N., Fatique Fracture Investigation<br />
of Cemented Carbide Tools in Gear Hobbing. Part 2: The Effect of Cutting<br />
Parameters on the Level of Tool Stresses – A Quantitative Parametric Analysis.<br />
ASME Journal of Manufacturing Science and Engineering, 124, 4, 792-798,<br />
(2002).<br />
10. B o u z a k i s K., A n t o n i a d i s A., Optimizing of Tangential Tool Shift in Gear<br />
Hobbing, Annals of the CIRP, 44, 1 (1995).<br />
MODEL PENTRU EVALUAREA UZURII <strong>DIN</strong>ȚILOR FREZELOR DE<br />
DANTURAT PE BAZA DETERMINĂRII AȘCHIILOR 3D<br />
(Rezumat)<br />
Frezarea danturii este un procedeu de prelucrare larg răspândit în industrie, în<br />
special pentru producția de serie a danturilor exterioare. În acest proces, modul diferit<br />
de formare a așchiei în fiecare poziție de generare a profilului duce la uzura neuniformă<br />
a dinților frezei. Lucrarea prezintă o nouă metodă de determinare a parametrilor uzurii,<br />
frezelor pe baza așchiilor 3D. Caracteristicile așchiilor 3D au fost stabilite pornind de la<br />
un model existent de uzură, în vederea determinării cu mai multă acuratețe a distribuției<br />
uzurii la nivelul dinților frezei.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
CAD-BASED CALCULATION OF CUTTING FORCE<br />
COMPONENTS IN GEAR HOBBING<br />
BY<br />
NIKOLAOS TAPOGLOU and ARISTOMENIS ANTONIADIS<br />
Abstract. One of the most commonly used gear manufacturing process<br />
especially for external is gear hobbing. The optimisation of the process of gear<br />
hobbing is of great importance in the modern industry. This paper presents a novel<br />
simulation program called HOB3D that can simulate the cutting process in a<br />
commercial CAD environment, thus producing results with the optimal precision.<br />
The outputs of the program include the 3D chip and gap geometry as well as the<br />
developing cutting forces which were validated using experiments.<br />
Key words: gear hobbing, cutting forces, simulation.<br />
1. Introduction<br />
One of the key components of any torque transmission system is high<br />
precision involute gears. Gears can be constructed with a wide variety of<br />
methods; Gear hobbing is the one mainly used in the modern industry. Gear<br />
hobbing kinematics consists of three relative motions between the cutting tool<br />
and the workpiece. This makes it difficult to simulate with analytical models.<br />
HOB3D is a novel simulation program based on a commercial CAD<br />
environment which can simulate the cutting process and provide results<br />
including 3D solid chips and gaps as well as predicting the developing cutting<br />
forces.<br />
2. State of the Art<br />
The research conducted in the area of gear hobbing can be divided into two<br />
categories: gear hobbing process simulation and wear prediction. In the first<br />
field a series of simulation models have been developed using CAD [1],[2],<br />
FEA [3] or analytical [4]-[7] based models. Experiment based models [8], [9]
22 Nikolaos Tapoglou and Aristomenis Antoniadis<br />
are used in the field of wear prediction order to calculate the wear of the cutting<br />
tool and optimize the cutting so as to obtain uniform wear along the cutting tool.<br />
Cutting forces prediction is an area of great interest also. Research conducted in<br />
this area is based on Kienzle-Victor’s equations and depend on geometry of<br />
chips.<br />
3. Gear Hobbing and HOB3D Simulation Process<br />
Gear hobbing kinematics is based on three relative motions between the<br />
cutting tool and the workgear. These motions must be synchronised in order to<br />
produce high quality helical and spur gears. As presented in Fig. 1 the work<br />
gear rotates round its axis while at the same time the hob rotates round its own<br />
axis and moves parallel to the gear axis. The hob is positioned in an angle<br />
relative to the gear. The magnitude of this angle is relative to the hob helix<br />
angle and the helix angle to the gear produced correspondingly.<br />
Fig. 1 – Gear hobbing.<br />
Gear hobbing process is affected by a series of parameters which be divided<br />
in three categories: hob, gear and process parameters. The first include module<br />
(m), external diameter (dh), number of origins (z1) and number of columns (ni)<br />
of the hob. Gear parameters are number of teeth (z2), helix angle (ha) and gear<br />
width. Finally, process parameters include axial feed (fa) and cutting speed (v).<br />
A series of parameters can be calculated from those above mentioned, those<br />
being distance e, the helix angle of the hob (γ), gear diameter (dg) and depth of<br />
cut (t).<br />
In order to simulate the process of gear hobbing, new software has been<br />
developed. The proposed simulation model has been embedded in a commercial<br />
CAD program thus taking advantage of its accuracy resulting in more detailed<br />
calculations.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 23<br />
The new simulation code called HOB3D uses three coordinate systems in<br />
order to calculate the results required. The first (1) is positioned on the<br />
examined cutting edge and has the x axis parallel to the hob’s axis, the y axis<br />
perpendicular to x axis and finally z axis perpendicular to the prior two.<br />
Coordinate systems (2) and (3) have axis z running through the workgear’s axis.<br />
In coordinate system (2), x axis is always rotating in order to point to the gap,<br />
while the axes of coordinate system (3) are fixed.<br />
Fig. 2 – HOB3D flowchart.
24 Nikolaos Tapoglou and Aristomenis Antoniadis<br />
Fig. 2 presents the flowchart of this simulation model. As it is illustrated,<br />
after all the input data, shown on the left side of the figure, has been defined, the<br />
initial workpiece is developed in the CAD environment and the number of the<br />
effective teeth is determined accordingly. The gear hobbing simulation process<br />
is executed for all these teeth. The first step in the simulation process is the<br />
positioning of the cutting edge on the 3D space.<br />
This positioning is repeated along the path that is covered by the cutting<br />
edge. The profiles that arise are combined in order to form a 3D surface. The<br />
next step is the creation of an assembly that includes the workgear and the 3D<br />
surface. Afterwards, Boolean operations are used in order to generate the 3D<br />
chip and the 3D gap. The above described process is repeated for all active teeth<br />
in every rotation of the cutting tool. Finally, after the end of the simulation<br />
process the cutting forces on every one of the active teeth of the hob are<br />
calculated. These forces are added up and the total cutting forces are calculated.<br />
4. HOB3D Formulation<br />
4.1. Simulation Formulation<br />
In HOB3D all the movements involved in the process are transferred to the<br />
hob cutting tooth motion. Furthermore, the simulation process is carried out on<br />
one gap of the gear thus reducing the simulation time. In order to identify the<br />
cutting teeth, those are numbered. The tooth which has the local axis Y1 parallel<br />
to local axis X2 when it passes through the center of the gap is named Tooth 0,<br />
the tooth that passes after that is named Tooth 1 and the tooth previous to that is<br />
named Tooth-1 and so on.<br />
The simulation process is illustrated on Fig. 3. As it is presented, the hob<br />
profile according to <strong>DIN</strong>3972 [10] must be designed first. This profile is<br />
presented on the top left frame of Fig. 3 and positioned on the 3D space, as<br />
illustrated on the next frame of fig. 3. The positioning is repeated for a series of<br />
times until the tooth is out of the cut.<br />
After this step, there is a series of profiles correctly positioned on the 3D<br />
space. Every profile represents a revolving position of the hob and its<br />
positioning includes hob rotation, workpiece rotation and feed rate. All these<br />
profiles are combined in order to form a 3D surface like the one seen in the last<br />
frame of the first line of fig. 3. Then, the 3D surface is assembled with the 3D<br />
gap produced from the previous tooth. The final stage of the simulation process<br />
is the extraction of the final gap geometry and the non-deformed chip geometry.<br />
This is achieved with the aid of Boolean operations, as supported by the used<br />
CAD environment.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 25<br />
Fig. 3 – HOB3D simulation process.<br />
4.2. Force Component Calculation Formulation<br />
After the simulation is completed, the cutting forces can be calculated. The<br />
force calculation code is performed in accordance with Kienzle-Victor’s<br />
equations. The calculation process is based on sections made on every one of<br />
the solid chips.<br />
A series of sections are made for every chip. Each section is made in the<br />
plane of the hob’s cutting edge. The planes are visible on the top left frame of<br />
Fig. 4. After the creation of the section on the 3D chip, the crossection is<br />
discritised along the cutting edge. This discritisation is presented on the top<br />
right frame of fig. 4.<br />
The three force components are calculated according to the elementary chip<br />
width and thickness for every elementary chip section. The three force<br />
components are rotated in order to match the local coordinate system (1) and<br />
then added up in order to produce the total cutting forces on every crossection.<br />
After all the sections are calculated the total forces on the tooth are obtained, as<br />
presented on the bottom of Fig. 4.
26 Nikolaos Tapoglou and Aristomenis Antoniadis<br />
Fig. 4 – Force calculation algorithm.<br />
4.3. Simulation Results<br />
HOB3D has been used in order to produce the 3D solid chips in the full cut<br />
phase of gear hobbing. These chips were analysed so that the cutting forces are<br />
calculated. The crossections made on the chip are presented on the right side of<br />
Fig. 5 whereas the chip thickness on six of the crossections is illustrated on the<br />
left side of the same figure.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 27<br />
Fig. 5 – Chip thickness of the solid chip.<br />
The cutting forces components for the above chip are presented in the next<br />
Fig. 6 which are calculated in accordance with the system 1.<br />
Fig. 6 – Cutting forces on one generating position.
28 Nikolaos Tapoglou and Aristomenis Antoniadis<br />
5. Verification<br />
The verification of the force calculation module was conducted in two<br />
phases. First, the cutting forces simulated on specific teeth were compared to<br />
the ones measured by B o u z a k i s [4] while other cutting forces calculated by<br />
HOB3D were compared to the ones measured by G u t m a n n [5]. Fig. 7<br />
illustrates the results of the first phase of the verification. As it can be seen, each<br />
column of the figure illustrates the measured and calculated cutting forces on<br />
one generating position measured on the coordinate system of the 3D gap (2). In<br />
most of the cases the simulation code predicts not only the form but also the<br />
magnitude of the cutting forces.<br />
Fig. 7 – Comparison between calculated and measured cutting forces.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 29<br />
The second step of the verification is the comparison between measured and<br />
calculated forces for all the cutting teeth simultaneously. Fig. 8 presents the<br />
comparison between the measured and the calculated cutting forces in all three<br />
directions of the coordinate system, as illustrated on the top left side of the<br />
figure.<br />
Fig. 8 – Comparison between calculated and measured cutting forces.<br />
4. Conclusions<br />
1. A novel simulation model for gear hobbing, developed in a commercial<br />
CAD environment is presented in this paper. The model can simulate the<br />
manufacturing of helical as well as spur gears<br />
2. The developed model can predict the cutting forces components with<br />
great accuracy. The calculated forces have been verified with the aid of cutting<br />
experiments.<br />
Received: March 10, 2010 Technical University of Crete,<br />
Department of Production Engineering&Management,<br />
Chania, Crete, Greece,<br />
e-mail: ntapoglou@isc.tuc.gr
30 Nikolaos Tapoglou and Aristomenis Antoniadis<br />
R E F E R E N C E S<br />
1. D i m i t r i o u V., V i d a k i s N., A n t o n i a d i s A., Advanced Computer Aided<br />
Design Simulation of Gear Hobbing by Means of 3-Dimensional Kinematics<br />
Modeling, ASME J. of Manuf. Science and Eng., 129, 911-918., (2007).<br />
2. D i m i t r i o u V., A n t o n i a d i s A., CAD-based Simulation of the Hobbing<br />
Process for the Manufacturing of Spur and Helical Gears, Int. J. of Advanced<br />
Manuf. Technology, 41, 3-4, 347-357 (2008).<br />
3. F r i d e r i k o s O., Simulation of Chip Formation and Flow in Gear Hobbing Using<br />
the Finite Element Method, Ph.D. Thesis, Aristoteles University of Thessaloniki,<br />
Greece 2008.<br />
4. B o u z a k i s K. D., Konzept und technologishe Grundlagen zur automatisierten<br />
Erstellung optimaler Bearbeitungsdaten beim Waelzfraese, Habilitation, TH<br />
Aachen 1980.<br />
5. G u t m a n n P., Zerspankraftberechnung beim Waelzfraesen, Ph.d. thesis, TH<br />
Aachen, 1988.<br />
6. A n t o n i a d i s A., Determination of the Impact Tool Stresses During Gear<br />
Hobbing and Determination of Cutting Forces During Hobbing of Hardened<br />
Gears, Ph.d. thesis, Aristoteles University of Thessaloniki, 1989.<br />
7. A n t o n i a d i s A., V i d a k i s N., B i l a l i s N., Fatique Fracture Investigation of<br />
Cemented Carbide Tools in Gear Hobbing. Part 1: FEM Modeling of Fly<br />
Hobbing and Computational Interpretation of Experimental Results. ASME J. of<br />
Manuf. Science and Engineering, 124, 4, 784-791, 2002.<br />
8. A n t o n i a d i s A., V i d a k i s N., B i l a l i s N., Fatique Fracture Investigation of<br />
Cemented Carbide Tools in Gear Hobbing. Part 2: The Effect of Cutting<br />
Parameters on the Level of Tool Stresses – A Quantitative Parametric Analysis,<br />
ASME J. of Manufacturing Science and Engineering, 124, 4, 792-798, 2002.<br />
9. B o u z a k i s K., K o m p o g i a n n i s S., A n t o n i a d i s A., V i d a k i s N., Gear<br />
Hobbing Cutting Process Simulation and Tool Wear Prediction Models. ASME<br />
J.of Manuf. Science and Engineering, 124, 1, 42-51 2002.<br />
10. D I N 3972, Bezugsprofile von Verzahnwerkzeugen fuer Evolventenverzahnungen<br />
Nach <strong>DIN</strong> 867“, 1981.<br />
DETERMINAREA PRIN SIMULARE CAD A COMPONENTELOR<br />
FORȚEI DE AȘCHIERE LA FREZAREA DANTURILOR<br />
(Rezumat)<br />
Unul dintre cele mai uzuale procedee de prelucrare a danturilor, în special a celor exterioare,<br />
este frezarea. Optimizarea acestui tip de proces de danturare prezintă la ora actuală o importanță<br />
deosebită în industrie. Lucrarea prezintă un nou program de simulare, numit HOB3D, care<br />
permite simularea procesului de așchiere într-un mediu CAD comercial, cu rezultate optime în<br />
ceea ce privește precizia de prelucrare. Elementele rezultante ale programului de simulare se<br />
referă atât la așchia 3D și la geometria golului dintre dinții frezei, cât și la nivelul forțelor de<br />
așchiere dezvoltate, elemente care au fost validate experimental.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCȚII DE MAȘINI<br />
CONTRIBUTIONS CONCERNING THE COMPUTER<br />
AIDED CONTROL OF THE FELLOWS' CUTTER<br />
BY<br />
EUGEN STRĂJESCU 1 , CONSTANTIN DOGARIU 1 ,<br />
OLIMPIA PAVLOV 2 and DUMITRU DUMITRU 3<br />
Abstract. In the paper are presented the bases of a methodology for the<br />
determination of all the geometric and constructive elements of the cutting tools<br />
for the fellows cutter, tools having curve edges, starting from the getting of a solid<br />
model 3D of the measured tool. There are shown the possibilities to obtain a 3D<br />
model and the mathematical model of the geometric parameters control.<br />
Key words: fellows' cutter, control, cutting tools, 3D model, geometry,<br />
curved edges.<br />
1. Introduction<br />
Preoccupations concerning the increase of the standard, standard type and<br />
special cutting tools’ production and preoccupations concerning new tools with<br />
superior constructive functional performances represent a major tendency in the<br />
manufacturing of machines. The software resources implicates the existence in<br />
the system of a collection of organized information in order to assure not only<br />
the complete design of the cutting tool but also the analysis, the syntheses,<br />
comparison, modeling, simulation optimization, visualization, the optional<br />
presentation of the partial results etc. So the modern design is made in order to<br />
obtain a solid 3D model that is later automatically detailed in sections and views<br />
2D. The existence of the solid model signify the spectacular increase of the<br />
information's’ number permitting practically the total knowledge of the<br />
positions of any point or surfaces from the element’s construction and of the<br />
angles between directions or planes. This demarche ameliorates the tackling of<br />
the cutting tools' design by means of the concurrent engineering.
32 Eugen Străjescu et al.<br />
The modern means for the assisted design (Catia, Solid Works, Ideas, Solid<br />
Edge) are capable to make automatically sections in any point and after any<br />
direction, pointing automatically distances between points or angles’ values.<br />
In this way we can avoid the situations in which we are choosing for the<br />
cutting tool a value for an angle at a peak, but in other sections or points the real<br />
values are outside the domain of availability. Starting from these observations,<br />
we want to develop a new method for the control of the manufactured cutting<br />
tool, starting from the idea of the obtention of a solid model.<br />
This demarche is extremely utile in the admission of the concurrent<br />
engineering.<br />
2. Using the Solid Model for the Cutting Tools' Control<br />
2.1. Possibilities to obtain the 3D Models of the Real Pieces<br />
2.1.1. The getting of the 3D model of the cutting tool by photography. The<br />
software D Sculptor presents a relative new in the large landscape of the CAD<br />
software and the software for treatment of the images and corps 2D and 3D. It<br />
permits the creation on the computer photorealistic models for a large game of<br />
objects using habitual photos, easy and fast. We do not need special hardware<br />
elements, but only a normal<br />
computer and a photo camera.<br />
A digital camera is required,<br />
but it is possible to use<br />
scanned photos. The main<br />
screen of the program is<br />
presented in the fig. 1.<br />
The basic processes for<br />
the model’s construction are<br />
presented down:<br />
• it is positioned the object to<br />
acquire the model on the<br />
calibration plane;<br />
• it is photographed the object<br />
from many angles;<br />
Fig. 1 – The main screen of D Sculptor software.<br />
• the photos are imported in<br />
the software D Sculptor;<br />
• D Sculptor detects automa-ticaly the model;<br />
• D Sculptor is calculating the three-dimensional model.<br />
• The obtained three-dimensional model (3D) (fig. 2) can be used with software. In<br />
time D Sculptor 2.0 was improved at a technical level and from the point of<br />
view of the interface.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 33<br />
It is possible to create models faster than before, and the last version of D<br />
Sculptor 2.0 Professional has a superior accuracy.<br />
2.1.2. Generation of the 3D model by scanning. A 3D scanner is a modern<br />
device, intense developed during the last years, device that analyses a real<br />
object or an average in order to collect data concerning the form, the texture and<br />
the color of the object.<br />
The acquired data can be used to digitally construct 3D models, utile for a<br />
large game of applications. In the most situations, a single scanning do not<br />
produces a complete model of the subject. Multiple scanning, some time<br />
hundreds, from many directions are necessary usually to obtain information<br />
about all the sides of the object.<br />
These scans must be introduced in a common reference system, processes<br />
named usually alignment or recording. The scans will fusion in order to create a<br />
complete model.<br />
Now the scanners are of two types:<br />
scanners 3D with contact and without<br />
contact. The scanners non contact can be<br />
active or passive. It exists a large gamut<br />
of technologies for the two categories. In<br />
techniques are imposed the non contact<br />
scanners, with laser.<br />
The essential problem of the getting<br />
of a 3D model by scanning is represented<br />
by the great number of necessaries scans<br />
and by the possible necessity to complete<br />
Fig. 2. – The solid model resulted<br />
after the images’ treatment.<br />
form of the scanned tool is more complex.<br />
the acquired model using programs for<br />
assisted design (Catia, SolidWorks etc.).<br />
This last activity is more necessary if the<br />
For these reasons we can affirm that for the simple tools the proposed<br />
method is not efficient, because the completion of the model need too much<br />
time for routine measurements, but becomes very efficient for the complex<br />
cutting tools, measurable with difficulties, at which we obtain the geometric<br />
parameters baffling to measure (and in the main cases with a low accuracy).<br />
3. The Control of the Geometry Based on the 3D Model<br />
After the getting of the 3D model using the software D Sculptor 2.0 or by<br />
scanning, the model is exported in the Solid Works software, resulting the up<br />
presented (fig. 2).
34 Eugen Străjescu et al.<br />
On the obtained model are choose points on the main edge and on the<br />
secondary one and there are given command for the construction of the planes<br />
in which are measured the requested geometric elements.<br />
3.1. The Determination of the Fellows' Cutters' Geometry.<br />
The fellows' cutters are tools destined to the manufacturing by mortising the<br />
internal and external gear of the spur gears with right, inclined of "in V" teeth,<br />
representing from that reason a high degree of universality, as consequence of<br />
the cutting edges' access in areas inaccessible for other kind of tools (for<br />
example, side mills, hobbing cutters, etc.). Also, the fellows' cutters are the<br />
single cutting tools that can machine cylindrical gear wheels with interior<br />
denture, by rolling.<br />
3.1.1. Geometric parameters. The clearance angle at the tooth' peak presents<br />
a big importance, because from that value depends the bigness of the profile<br />
correction in time after successive sharpening.<br />
Also, by the value of this angle depends the size of the lateral clearance<br />
angle on the two flanks of the tooth that result much more little. So, it is<br />
considered the lateral clearance angle at the level of the divided circ,<br />
respectively if the tooth is cut after the division cylinder, the intersection curve<br />
between this cylinder and the involute clearance angle of the tooth will be a<br />
helix. Because the disposition on the involute flanks of the points in which the<br />
values of the angles α, γ, κ and λ are determined, it results that in any point we<br />
obtain different values that must very between convenient limits.<br />
The actual methods do not permit the determination of these angles, but<br />
3 0 31'<br />
Fig. 3 – The determination of the angles αv and γv at the peak.<br />
2 0
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 35<br />
only of the values from the peak of the tooth (Fig. 3). The minimal values (that<br />
must have a specified value) are determined on the base of mathematical<br />
relations,<br />
but the real angles' variations not controlled.<br />
If the intersection of the tooth with planes parallel with a medial plane of<br />
the tooth is realized (posterior planes, Fig. 4), we obtain different values for the<br />
angles αp and γp (the plane Pp<br />
- Pp) in the points of<br />
intersection between these<br />
planes and the cutting edge<br />
from the involute flank. The<br />
points in which the<br />
determination is made can be<br />
however dense (from example<br />
0.01 mm), but such frequency<br />
is not necessary, because the<br />
variation limits and the<br />
variation mode can be<br />
observed in only some points,<br />
and that fact simplify<br />
determination operations.<br />
the<br />
Fig. 4 – The intersection of the<br />
tooth<br />
with paralel planes.<br />
The Solid Works computer<br />
program, by the facilities<br />
that are offered, give directly<br />
under the form of a quote the<br />
values obtained for the angles<br />
determined in these planes.<br />
Using a program that permits<br />
the drawing of the mentioned<br />
planes at determined<br />
distances, with a pre<br />
established step, becomes<br />
Fig. 5 – Intersectio ns with orthogonal planes.<br />
possible to obtain the desired<br />
values in every point on the<br />
edge, and with these values it<br />
is possible<br />
to draw a chart<br />
(Fig. 6).<br />
Similarly, it is possible to<br />
construct normal planes at the<br />
cutting edge's tangent in the<br />
anterior determined points on<br />
that edge, or orthogonal planes (Fig. 5). Obviously, in these planes are<br />
determined the angles αn and γn, or the angles αO and γO.<br />
The determination<br />
conditions<br />
are similar with the before precise conditions.
36 Eugen Străjescu et al.<br />
3.5<br />
3<br />
2.5<br />
2<br />
1.5<br />
1<br />
0.5<br />
0<br />
4.5<br />
4<br />
3.5<br />
3<br />
2.5<br />
2<br />
1.5<br />
1<br />
0.5<br />
0<br />
3.31 3.31 3.31 3.31<br />
2<br />
2.11<br />
M1 M2 M3 M4<br />
α p 2 2.11 3.31 4.27<br />
γ p 3.31 3.31 3.31 3.31<br />
Fig. 6 – Variation chart of the angles αp and γp along the cutting edge<br />
on the involute flank of the fellows' cutter.<br />
3.31<br />
2<br />
2.91<br />
0.94<br />
Excepted the presented determinations, that are essential for a good service<br />
of the fellows' cutter, it is possible to determine another angles, e,g. in sections<br />
with quidam planes, as in Fig. 8, or with front planes (Pf - Pf) as in Fig. 10.<br />
In this case too, the angles are directly posted by the Solid Works computer<br />
program. The values for the angles for the determinations from<br />
the Fig. 8 and<br />
Fig. 10 are presented in the charts from the Fig. 9 and Fig. 11.<br />
2.52<br />
0.95<br />
4.27<br />
M1 M2 M3 M4<br />
α n 2 0.94 2.52 2.24<br />
γ n 3.31 2.91 0.95 0.97<br />
Fig. 7 – Variation chart of the angles αO, γO (noted in the figure with αn, γn).<br />
2.24<br />
0.97<br />
α p<br />
γ p<br />
α n<br />
γ n
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 37<br />
Fig. 8 – Sections with certain parallel planes with the intersections' presentation<br />
7<br />
6<br />
5<br />
4<br />
3<br />
2<br />
1<br />
0<br />
3.36 3.38 3.31<br />
2.91<br />
2.02<br />
4.44<br />
M1 M2 M3 M4<br />
α 0 2.02 2.91 4.44 6.44<br />
γ 0 3.36 3.38 3.31 3.52<br />
We can observe the anomaly of the variation of the normal back rake angle<br />
γO. A lot of verification proves that the model is correct, so it is to research the<br />
real situation.<br />
Fig. 9 – The charts of the angles αO and γO<br />
A proposal for a complete control of cutting tools made on the basis on the<br />
getting of a 3D solid model is new and brings the first advantage that it is<br />
possible to obtain all the angles searched at any point of the cutting edges or<br />
active planes.<br />
6.44<br />
3.52<br />
α 0<br />
γ 0
38 Eugen Străjescu et al.<br />
2.5<br />
2<br />
1.5<br />
1<br />
0.5<br />
0<br />
Fig. 10 – Intersections with frontal planes in different points.<br />
2.32<br />
0.48<br />
2.05 2.01<br />
0.56<br />
0.62<br />
M1 M2 M3 M4<br />
αf 2.32 2.05 2.01 1.93<br />
γ f 0.48 0.56 0.62 0.74<br />
Fig. 11 – Values for the angles αf şi γf in frontal plans Pf - Pf.<br />
1.93<br />
0.74<br />
αf<br />
γ f
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 39<br />
4. Conclusion<br />
The method is especially useful for complex tools, small, with the active<br />
surfaces and curved cutting edges, to which access control with current<br />
instruments is very slowness or impossible and where the definition of<br />
theoretical planes control angles is difficult to apply. The other method of<br />
obtaining the 3D model through photography has its limits. Further research<br />
will develop methodologies for control of complex tools for very large or very<br />
small, with curved surfaces, methodologies able to change the angles of the long<br />
edges or surfaces with steps as small.<br />
Received: March 25, 2010 1 "<strong>POLITEHNIC</strong>A" University from Bucharest ,<br />
Department of Machines and Production Systems<br />
e-mail: eugen_strajescu@yahoo.com<br />
2 S.C. MUNPLAST S.A.<br />
e-mail: ostefu@yahoo.com<br />
3 "VALAHIA" University from Targoviste<br />
e-mail: ddumitru@yahoo.com<br />
R E F E R E N C E S<br />
.<br />
1. Minciu C., Stră jescu E., ş.a., Scule aşchietoare, Îndrumar de proiectare.<br />
Editura Tehnică, Bucureşti, 1995.<br />
2. S t r ă jescu, E., Proiectarea sculelor aschietoare, Litografia I.P.Bucureşti, l984.<br />
3. E n a c h e , Ş t., Stră jescu, E., Minciu, C., Metode şi programe pentru<br />
proiectarea asistată a sculelor aşchietoare. Litografia I.P.Bucureşti, 1988.<br />
4. Strajescu E., Pavlov O., Metodologie de control asistat de calculator al<br />
sculelor aşchietoare pe baza unor metode neconvenionale de obţinere a<br />
modelelor 3D. Conferinţa ICEEMS, Braşov, 2005.<br />
5. S t r ă jescu E., Pavlov O., Dumitru, D., C ontributions Concerning the<br />
Informatic Control of the Cutting Tools, International Conference on<br />
Manufacturing Science and Education - MSE Sibiu, 2009.<br />
CONTRIBUŢII PRIVIND CONTROLUL INFORMATIZAT<br />
AL CUŢITELOR ROATĂ<br />
(Rezumat)<br />
În lucrare se prezintă bazele unei metodologii pentru controlul informatizat al tuturor<br />
elementelor geometrice şi constructive ale sculelor aşchietoare în general, şi al cuţitelor roată<br />
pentru mortezat roţi dinţate cilindricve în special. Metodologia are la bază obţinerea unui model<br />
solid, iar acest lucru se poate face fie prin fotografiere, folosindu-se programe potrivite, fie prin<br />
scanare 3D, fie în faza de proiectare. Metoda propusă permite controlul absolut al tuturor<br />
parametrilor geometrici şi constructivi, în orice punct şi în orice plan, precum şi trasarea<br />
graficelor de variaţie. Se poate imagina o metodă prin care un program de proiectare asistată de<br />
tipul Solid Concept, parametrizat, să modifice unghiurile sculei pornind de la valoarea admisibilă<br />
a unui anume unghi într-un anume plan.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
CONTRIBUTIONS TO THE ELABORATIONS OF A<br />
GRAPHICAL METHOD FOR PROFILING OF TOOLS<br />
WHICH GENERATE BY ENVELOPING<br />
I. ALGORITHMS<br />
SILVIU BERBINSCHI, VIRGIL TEODOR,<br />
NICOLAE DUMITRAŞCU and NICOLAE OANCEA<br />
Abstract. The generation of the ordered curls profile (surfaces) associated<br />
with the centrodes of the rack-gear tool represent an fundamental problem for the<br />
profiles reciprocally enveloping associated with a couple of rolling centrodes. The<br />
knowing, in principle, of the generating rack-gear allow the construction of the<br />
tool and, also, the determination of the axial profile of the worm mill for the<br />
ordered curl profile’s generation. They are known and used more solutions for the<br />
solving of this problem: analytical solutions, based on the fundamentals theorems<br />
of gearing; complementary theorems, obtained from the fundamentals theorem;<br />
graphical methods based on graphical environments capabilities. In this paper, is<br />
proposed an algorithm for the approach of issue of ordered curl profiles based on<br />
the CATIA design environment capabilities. It is proposed a general algorithm<br />
which allows making applications, ended with the numerical determination of the<br />
rack-gear reciprocally enveloping with ordered curl profiles. They are solved<br />
profiles known in discrete form, substituted by spline approximation. The method<br />
allows drawing the gearing lines as so as the interference trajectories. The<br />
obtained results are comparing with results obtained by analytical methods.<br />
Key words: enveloping surfaces, rack-gear tool, graphical design method.<br />
BY<br />
1. Introduction<br />
The profiling of tools which generated by enveloping by the rolling method<br />
— rack gear tool and gear shaped tool — may be make by some methods:
42 Silviu Berbinschi et al<br />
- analytical methods, based on fundamental methods of surfaces enwrapping<br />
– first Olivier theorem, Gohman theorem, normals method, W i l l i s [1], [2];<br />
- complementary analytical methods – “minimum distance” method, the<br />
“substitutive circles family” method, the “in-plane generating trajectories”<br />
method [3]-[5];<br />
- graphical-analytical methods [6];<br />
- graphical methods, using the capabilities of CAD software [7].<br />
We mention that the methods proposed and used for the study of<br />
reciprocally enveloping surfaces respect the enveloping fundamental theorem.<br />
The proposed solutions leads to comparable results, in most cases<br />
identically, for the crossing profile tool’s shape, which generated by rolling<br />
ordered curls profiles associated with a couple of rolling centrodes.<br />
Literature include multiples applications for various domains as methods of<br />
tool’s profiling, revolution surfaces, for generating helical surfaces [8]; the<br />
modelling of the cylindrical surfaces with disk tools [9]; generation of the<br />
helical compressors rotors [10].<br />
In this paper, is proposed a method for profiling of the rack-gear tools<br />
reciprocally enveloping with ordered curl profiles, based on the capabilities of<br />
the CATIA design environment.<br />
2. Generation Kinematics<br />
In principle, the tool’s profiling determination problem for rack-gear tool<br />
reciprocally enveloping with an ordered curls profile, associated with a circular<br />
centrode, assume to respect the kinematics of the generating process, see figure<br />
1.<br />
Fig. 1 – Couple of rolling centrodes.<br />
The two rolling centrodes, C1 —circle, associated with the profiles ordered<br />
curls and C2 —straight line, associated with the rack-gear tool, are in rolling<br />
movement, so, is respected the condition:
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 43<br />
λ = R ⋅ ϕ ,<br />
(1) rp 1<br />
where λ is the linear velocity in the translation of C2 centrode;<br />
R ⋅ ϕ — the value of velocity in point O1—the gearing pole, from C1<br />
rp<br />
1<br />
centrode, in the rotation movement around z axis;<br />
ϕ1 — variable angular parameter.<br />
In the rotation of C1 centrode the translation movements along the C2<br />
centrode and the rotation around Z axis are evenness.<br />
They are defined the reference systems:<br />
xyz is the global reference system, with z axis overlapped to the rotation axis<br />
of the C1 centrode;<br />
XYZ — mobile reference system, initially overlapped to the global reference<br />
system, joined with the ordered curls profile Σ;<br />
ξηζ — mobile reference system, joined with C2 centrode of rack-gear tool,<br />
with axis parallel and of the same sense with the global reference system xyz.<br />
The kinematics of the rolling process of the two centrodes, C1 and C2,<br />
tangents in point O1 — gearing pole— presume that the velocities of points<br />
belongs to the two centrodes, temporally situated in point O1, to be equals.<br />
In this way, the global motion of the ξηζ reference system, joined with<br />
centrode C2, is described by the transformation,<br />
(2) x = ξ + a ,<br />
where:<br />
(3)<br />
⎛ξ⎞ ⎛x⎞ ⎜ ⎟ ⎜ ⎟<br />
ξ =<br />
⎜<br />
η<br />
⎟<br />
; x =<br />
⎜<br />
y<br />
⎟<br />
⎜ζ⎟ ⎜z⎟ ⎝ ⎠ ⎝ ⎠<br />
represent the matrix of the current points in the space ξηζ, respectively xyz;<br />
(4)<br />
⎛ 0 ⎞<br />
⎜ ⎟<br />
a = ⎜ −λ⎟<br />
⎜ R ⎟<br />
⎝−rp ⎠<br />
is the matrix formed with the coordinates of point O1, in the global reference<br />
system, with λ instantaneous velocity in the translation movement of C1<br />
centrode and Rrp the value of circular centrode C1 (rolling radius).<br />
Also, the revolution movement of C1 centrode is described by transformation
44 Silviu Berbinschi et al<br />
⎛x⎞ ⎛X⎞ ⎜ ⎟ ⎜ ⎟<br />
⎜<br />
y<br />
⎟<br />
= 1 1 ⋅<br />
⎜<br />
Y<br />
⎟<br />
⎜z⎟ ⎜Z ⎟<br />
⎝ ⎠ ⎝ ⎠<br />
T<br />
(5) ω ( ϕ )<br />
⎛X⎞ ⎜ ⎟<br />
where<br />
⎜<br />
Y<br />
⎟<br />
is the matrix of the current point in space XYZ, and<br />
⎜Z⎟ ⎝ ⎠<br />
⎛1 0 0 ⎞<br />
⎜ ⎟<br />
= ⎜ ⎟<br />
⎜ ⎟<br />
(6) ω1( ϕ1) 0 cos( ϕ1) sin(<br />
ϕ1)<br />
0 −sin(<br />
ϕ ) cos(<br />
ϕ )<br />
⎝ 1 1 ⎠<br />
is the rotation transformation matrix, around X axis, with angle ϕ1<br />
(counterclockwise rotation).<br />
The assembly of equations (3) and (6), with the respect of rolling condition<br />
(1), determine the relative motion,<br />
⎛ξ⎞ ⎛1 0 0 ⎞ ⎛X⎞ ⎛ 0 ⎞<br />
⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟<br />
⎜ ⎟<br />
= ⎜ ⎟⋅<br />
⎜<br />
Y<br />
⎟<br />
−⎜ −λ<br />
⎟<br />
⎜ ⎟ ⎜0 −sin 1 cos ⎟ ⎜<br />
1 Z ⎟ ⎜−R⎟ ⎝ ⎠ ⎝ ⎠ ⎝ ⎠ ⎝ rp ⎠<br />
(7) η 0 cos( ϕ1) sin(<br />
ϕ1)<br />
ζ ( ϕ ) ( ϕ )<br />
while the Σ profile, belongs to the profile’s ordered curl, associated with the C1<br />
centrode, in form,<br />
⎛ 0 ⎞<br />
⎜ ⎟<br />
Σ= ⎜ ⎟<br />
⎜ ⎟<br />
⎝ ⎠<br />
(8) Y( u)<br />
Z ( u)<br />
with u variable parameter, describe a profile’s family in the rack-gear space:<br />
(9) ( )<br />
( u,<br />
1 ) = 0;<br />
( , 1) ( ) cos( 1) ( ) sin(<br />
1)<br />
( , ) ( ) sin( ) ( ) cos(<br />
)<br />
ξ ϕ<br />
u ϕ1<br />
Y u Z u Rrp<br />
1<br />
u Y u Z u R<br />
;<br />
Σ η ϕ = ϕ − ϕ + ⋅ ϕ<br />
ζ ϕ = ϕ + ϕ − .<br />
1 1 1<br />
The enveloping of the profile family (9) represents the rack-gear tool’s<br />
profile.<br />
rp
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 45<br />
Often, the profile (8) may be replaced by the equations of one surface,<br />
cylindrical or cylindrical helical, of which belongs the profile (8), as crossing<br />
in-plane profile section of Σ surface (the section plane is a perpendicularly plane<br />
on X axis), surface which is described by equations on form:<br />
(10)<br />
( )<br />
( , ) ;<br />
( , ) ,<br />
X = X u, t ;<br />
Σ Y = Y u t<br />
Z = Z u t<br />
with u and t independent variable parameters.<br />
To the profile’s family (9) was associated the enveloping condition<br />
specifically for one of the enwrapping fundamental theorems (first Olivier<br />
theorem, Gohman theorem, “normals method” [1], [2]) or one of the<br />
complementary theorem (“minimum distance”, “substitutive circles family”,<br />
“in-plane generating trajectories”) in order to determine the envelope of profile<br />
family as generating rack-gear profile [3], [4].<br />
Also, are known graphical solutions for profiling tools which generate by<br />
enwrapping method, ordered curls profile associated with a rolling centrodes<br />
couple (the “regions” method [6], [7]).<br />
3. Kinematics Method in the CATIA Design Environment<br />
Is proposed a new solution for the profiling of rack-gear tool reciprocally<br />
enveloping with an ordered curl profiles associated with a couple of rolling<br />
centrodes, using the capabilities of CATIA software, by making an kinematics<br />
entity which will simulate the rolling movement of centrodes: C1 — rolling<br />
circle with radius Rrp, associated with the profiles curl; C2 — straight line,<br />
associated with the rack-gear.<br />
The proposed solution has the advantages that use the capabilities of a very<br />
versatile product, which may offer rigorous numerical results.<br />
Also, being a graphical method, the rough errors, due firstly to the passing<br />
curves, which may be erroneously considered as profile’s zone, are easy to<br />
identify and deleted from analysis.<br />
The proposed solve is based on the capabilities of Part environment where<br />
are synthesized the elements of a mechanism able to simulate the enveloping<br />
condition, in this case, the normal condition. These elements, created in the Part<br />
environment, are inserted in a file of Assembly environment, assuring the<br />
position of the mechanism elements in the start position, in following in the<br />
DMU Kinematics environment are defined the kinematics couples.<br />
The mechanism motion is made by the command Simulation, establishing<br />
the intermediate position number (Shots) and creating with the command Replay
46 Silviu Berbinschi et al<br />
a movie of the mechanism successive positions.<br />
With command Trace is draw the trajectory of any point which belongs to<br />
one element of the mechanism regarding any another element, including the<br />
global reference system, determining the gearing line between the profile to be<br />
generate and the profile of the rack-gear tool.<br />
These trajectories represent spline curves, constructed by successive points<br />
obtained by the mechanism rolling. The coordinates of these points may be<br />
exported as text file or any spreadsheet, see figure 2.<br />
Fig. 2 – Generating algorithms in CATIA environment.<br />
4. Conclusions<br />
1. The rolling process kinematics for two centrodes becomes possible to be<br />
described in the graphical design environment as CATIA.<br />
2. The precision for these profile description, in this graphical environment,<br />
the geometrical forms and the kinematics of these allow a rigorous analysis of<br />
the reciprocally enveloping surfaces.<br />
3. A software suit specifically for the graphical design environment was<br />
synthesised in the end of paper, for the tool’s profiling reciprocally of an<br />
ordered curl of surfaces (profiles). Specifically applications will allow<br />
establishing the proposed method quality.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 47<br />
Acknowledgements. The authors gratefully acknowledge the financial support of the<br />
Romanian Ministry of Education, Research and Innovation through grant<br />
PN_II_ID_791/2008.<br />
Received:January, 20, 2010 Dunărea de Jos University of Galaţi,<br />
Manufacturing Science and Engineering Department<br />
Galaţi, Romania,<br />
e-mail: nicolae.oancea@ugal.ro<br />
R E F E R E N C E S<br />
1. L i t v i n F.L., Theory of Gearing Reference Publication 1212, NASA. Scientific and<br />
Technical Information Division, Washington D.C., 1984.<br />
2. O a n c e a N., Generarea suprafeţelor prin înfăşurare, Vol. I, Teoreme<br />
fundamentale, Editura Fundaţiei Universitare “Dunărea de Jos”, Galaţi, 2004,<br />
ISBN 973-627-106-4.<br />
3. O a n c e a N., Generarea suprafeţelor prin înfăşurare, Vol. II, Teoreme<br />
complementare, Editura Fundaţiei Universitare “Dunărea de Jos”, Galaţi, 2004,<br />
ISBN 973-627-106-4;<br />
4. T e o d o r V., O a n c e a N., D i m a M., Profilarea sculelor prin metode analitice,<br />
Editura Fundaţiei Universitare “Dunărea de Jos”, Galaţi, 2004, ISBN (10)<br />
973-627-333-4;<br />
5. M i n c i u C., C r o i t o r u S., I l i e S., Aspects regarding generation of non-involute<br />
gear profiles, Proceedings of the International Conference on Manufacturing<br />
Systems, ICMaS 2006, Ed. Academiei Române, Bucuresti, ISSN 1842-3183, pp.<br />
311-314;<br />
6. B a i c u I., O a n c e a N., Profilarea sculelor prin modelare solidă, Ed.<br />
Tehnica-Info, Chişinău, 2002, ISBN 9975-63-172-X;<br />
7. D i m a M., O a n c e a N., T e o d o r V., Modelarea schemelor de aşchiere la<br />
danturare, Ed. Cermi, Iaşi, 2007, ISBN 978-973-667-270-5;<br />
8. M o h a n, L. V., S h u n m u g a m, M. S., An orthogonal array based optimization<br />
algorithm for computer-aided measurement of worm surface, Int. J. Adv. Manuf.<br />
Technol. (2006) 30: 434–443;<br />
9. P o t t m a n n H., W i e n, T. R a n d r u p, O d e n s e, Rotational and Helical<br />
Surface Approximation for Reverse Engineering, Computing, 60, 307-322<br />
(1998);<br />
10. I v a n o v V., N a n k o v G., K i r o v V., CAD orientated mathematical model for<br />
determination of profile helical surfaces, International Journal of Machine Tools<br />
& Manufacture”, Elsevier Science, Pergamon, 38, 8, pp.1001-1015, UK, (1998).
48 Silviu Berbinschi et al<br />
CONTRIBUŢII LA ELABORAREA UNEI METODE GRAFICE PENTRU<br />
PROFILAREA SCULELOR CARE GENEREAZĂ PRIN ÎNFĂŞURARE<br />
I. ALGORITM<br />
(Rezumat)<br />
Generarea vârtejurilor ordonate de profiluri (suprafeţe), asociate unor centroide, cu<br />
scule de tip cremalieră, reprezintă o problemă fundamentală pentru profiluri reciproc<br />
înfăşurătoare asociate unui cuplu de centroide în rulare. Cunoaşterea principială a<br />
formei cremalierei generatoare permite construcţia sculei-pieptene şi, de asemenea,<br />
determinarea profilului sculei melc pentru generarea vârtejurilor ordonate de profiluri.<br />
Sunt cunoscute şi utilizate multiple soluţii pentru abordarea unei astfel de probleme:<br />
soluţiile analitice, bazate pe teoremele fundamentale ale angrenării; teoremele<br />
complementare, derivate din metodele fundamentale, metode grafice, bazate pe<br />
posibilităţile mediilor de proiectare grafică — CAD.<br />
În lucrare se propune un algoritm general pentru abordarea problematicii<br />
profilurilor (suprafeţe) bazat pe facilităţile mediului de proiectare CATIA. Se propune<br />
un algoritm general, în baza căruia se realizează aplicaţia, finalizată cu determinarea<br />
numerică (grafică) a formei cremalierei reciproc înfăşurătoare vârtejului ordonat de<br />
profiluri.<br />
Sunt tratate profiluri cunoscute analitic sau sub formă de coordonate discrete,<br />
substituite prin forme de aproximare spline.<br />
Metodica permite tratarea liniilor de angrenare precum şi traiectoriile de<br />
interferenţă a profilurilor. Se compară rezultatele obţinute prin această metodă cu<br />
rezultatele obţinute prin metode analitice.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
CONTRIBUTIONS TO THE ELABORATIONS OF A<br />
GRAPHICAL METHOD FOR PROFILING OF TOOLS WHICH<br />
GENERATE BY ENVELOPING<br />
II. APLICATION FOR RACK-GEAR TOOL’S PROFILING<br />
SILVIU BERBINSCHI, VIRGIL TEODOR,<br />
NICOLAE DUMITRAŞCU and NICOLAE OANCEA<br />
Abstract. In this paper is presented an application of the general algorithm<br />
regarding the tool’s profiling which generate by enveloping, by rolling method,<br />
respectively the rack-gear tool generating an ordered curl of profiles. They are<br />
presented the successive steps of this process and, also, numerical examples<br />
regarding the crossing profile of the rack-gear tool. The method quality is proofed<br />
by comparing the obtained results with results obtained for the same problem,<br />
using one of the analytical methods, the Willis method, and software developed in<br />
Java.<br />
Key words: enveloping surfaces, rack-gear tool, graphical design method.<br />
BY<br />
1. Introduction<br />
It was present the general algorithm for profiling the tools generating by<br />
enveloping, by the rolling method, in the CATIA design environment.<br />
In following are presented an application of the proposed algorithm for<br />
rack-gear tool’s profiling generating ordered curl profiles, for a hexagonal shaft,<br />
as so as, the comparison for the profile determination of the rack-gear tool with<br />
results obtained by analytical methods, based on specialised software, realised<br />
in Java [6], [7].<br />
In the paper entitled “Contributions to the Elaborations of a Graphical Method for<br />
Profiling of Tools Which Generate by Enveloping. I. Algorithm” is proposed a new
50 Silviu Berbinschi et al<br />
solution for the profiling of rack-gear tool reciprocally enveloping with an<br />
ordered curl profiles associated with a couple of rolling centrodes, using the<br />
capabilities of CATIA software, by making an kinematics entity which will<br />
simulate the rolling movement of the two centrodes [8].<br />
2. Application for Rack-Gear Profiling<br />
In order to obtain the tool’s profile for various polygonal shafts, for various<br />
side lengths, is need to follow two stages. The first stage is to choose the side<br />
length and the diameter of the rolling circle as input data. This thing may be<br />
done introducing in an text or Excel file the value, in following, the CATIA<br />
software will automate modify the whole mechanism, with the new values. In<br />
the same stage is created the mechanism, the rolling of this and the tool’s profile<br />
determination. The coordinates of points belongs to profile will be exported in a<br />
text or Excel file.<br />
The second stage consist in the partial remake of the mechanism couples,<br />
the rolling of this for the new tool’s profile, for the previous modified values<br />
and the export of points in a new file.<br />
2.1. First stage<br />
2.1.1. Data input and mechanism elements construction. In this step are<br />
created a folder with all files needed for the mechanism.<br />
First of all for the input data is created a text of excel file with the names<br />
and the values of the input parameters, namely the piece shaft radius Ra, the<br />
polygonal side length L and the rolling circle radius of blank Rrp. These are the<br />
parameters which may be modified by user. This file will be linked with the<br />
parameters from the CATIA file.<br />
In the following example is considered a hexagonal shaft with L=50 mm<br />
and the including radius Ra=50 mm. In order to simplify the representation we<br />
consider the rolling radius Rrp=50 mm. The shaft sketch and the associated<br />
centrode are represented in figure 1. The mechanism elements are successive<br />
created, in the Part environment, are suggestive named and inserted in an<br />
assembly file, created in the Assembly environment, named GenCremaliera.<br />
The Part type files are: baza; piesa; cremaliera and tachet.<br />
The Baza file contains the standing part of the mechanism and represent<br />
more lines needed to mount others elements, see figure 2.<br />
From a point named OriginePiesa is construct a line named AxaArbore,<br />
representing the piece’s axis and next another line which will be as guide for the<br />
rolling line named GhidajDreaptaRulare. The segment having the rolling radius<br />
length and which contain the gearing pole is created in order to ease the
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 51<br />
mechanism assembling. The point PolulAngrenarii is situated on the rolling<br />
circle with radius Rrp.<br />
Fig. 1 – The shaft sketch.<br />
This file contain a parameters table (Design Table) named DateIntrare<br />
which link the line and point’s parameters from the Part file (from CATIA)<br />
with the input data from the file DateIntrare.txt.<br />
The file Piesa contain, firstly, the geometrical elements of the profile for<br />
which is determined the tool’s profile, see figure 2. Another sketch from this<br />
file is CercRulare which contain a circle with radius Rrp. The two values Ra and<br />
Rrp are linked with the text file DateIntrare.txt. The file Cremaliera contain the<br />
rolling straight line (DreaptaRulare) representing a circular pitch of the curl to<br />
be generated and having the length as function of the rolling radius Rrp,<br />
controlled by file DateIntrare.txt. The file Tachet have the straight lines which<br />
are tangent and respectively normal at the piece’s profile in the contact points.<br />
Fig. 2 – Assembly file with kinematics couples and tool’s profile.
52 Silviu Berbinschi et al<br />
2.1.2. Mechanism assembling and the creation of kinematics couples. In the<br />
assembly<br />
file GenCremaliera are inserted all the above described elements,<br />
with the command Insert Existing Component.<br />
These have geometrical and dimensional constraints in order to be<br />
positioned in accordance with the mechanism functionality. After inserting the<br />
constraints, the GenCremaliera contain the whole mechanism in the initial<br />
position. In the Assembly environment the mechanism elements were positioned<br />
but doesn’t exist yet any couple between them.<br />
The kinematics couples are created in the DMU Kinematics environment. It<br />
is created the mechanism named Generare.Scula.Cremaliera and is established<br />
the standing element, Baza.<br />
After the kinematics couples creation the mechanism function may be<br />
simulated and look as in figure 2.<br />
2.1.3. Mechanism rolling and determination of tool’s profile. In order to<br />
rolling<br />
this mechanism is created a simulation with command Simulation and<br />
are established the intermediate position number, Shot, number which will have<br />
influence to the tool’s profile precision. For this example we considered 500<br />
steps, so it will result 500 points. With this simulation is determined the tool’s<br />
profile, see Figure 3.<br />
0<br />
1<br />
2<br />
3<br />
4<br />
5<br />
6<br />
7<br />
ζ<br />
η<br />
Fig. 3 – Tool’s profile in the rack-gear reference system.<br />
2.1.4. Points coordinates extraction. After the determination of the tool’s<br />
profile, these may be exported in a text file using the command Design Table,<br />
from the Part environment.<br />
This command will generate a text file PuncteProfil.txt, see Table 1.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 53<br />
Table 1<br />
Points coordinates on tool’s profile in CATIA program<br />
Nr. Crt. η [mm] ζ [mm] Nr. Crt. η [mm] ζ [mm] Nr. Crt. η [mm] ζ [mm]<br />
1 0 0 170 17.07 -5.96 340 36.18 -5.81<br />
17 1 .41 -0 .78 187 18.96 -6.24 357 38.04 -5.44<br />
34 2.98 -1.57 204 20.86 -6.45 374 39.88 -5<br />
51 4.59 -2.31 221 22.78 -6.6 391 41.69 - 4.51<br />
68 6.26 -3.01 238 24.71 -6.68 408 43.46 -3.96<br />
85 7.98 -3.64 255 26.63 -6.7 425 45.2 -3.35<br />
102 9.73 -4.23 272 28.56 -6.65 442 46.89 -2.69<br />
119 11.52 -4.75 289 30.48 -6.54 459 48.53 -1.97<br />
136 13.34 -5.21 306 32.39 -6.36 476 50.13 -1.21<br />
153 15.19 -5.62 323 34.29 -6.11 501 52.36 0<br />
2.2. Second stage<br />
In the second stage if the user modify the DateIntrare.txt<br />
file saving the<br />
modification of the input parameters, the CATIA software detect this and<br />
warning<br />
by an Update message.<br />
After this message the user have to update the assembly file and the<br />
kinematics file in order to remake<br />
the kinematics couples which may be<br />
modified.<br />
3. Method Quality Verification<br />
It is proposed in following the verification of this method<br />
by comparing the<br />
numerical results obtained with graphical method<br />
regarding the results obtained<br />
by one of the analytical methods: Gohman method; “normals” method;<br />
“minimum distance” method etc [3].<br />
For the polygonal shaft is defined the analytical form of the profile to be<br />
generated, see figure 4:<br />
(1)<br />
X = 0;<br />
Σ Y = u;<br />
Z =<br />
a,
54 Silviu Berbinschi et al<br />
with u variable parameter, measured along the profile and a constant value<br />
depending the profile form.<br />
The Σ profile family is on form:<br />
(2) ( )<br />
Fig. 4 – Straight lined profile.<br />
ξ = 0;<br />
Σ η = ucosϕ − asinϕ + R ϕ;<br />
ϕ<br />
ζ = usinϕ + acos ϕ −R<br />
,<br />
with ϕ variable parameter.<br />
The<br />
specifically enveloping condition, in the “in-plane generation trajectory”<br />
methods is:<br />
(3)<br />
or<br />
(4)<br />
cosϕ<br />
−usinϕ<br />
− acosϕ<br />
+ Rrp<br />
= ,<br />
sinϕ ucosϕ − asinϕ<br />
u = R sinϕ<br />
.<br />
The equations<br />
assembly (2) and (4), for u variable between limits<br />
umin<br />
=−0.5⋅ Rrp ; umax = 0.5⋅Rrp , for a hexagonal shaft, represent the rack-gear<br />
tool’s profile reciprocally enwrapping with the shaft’s profile.<br />
In Table 2 are presented the tool’s profile coordinates determined by the<br />
analytical<br />
method [6], [7].<br />
In Figure 5 are presented the profiles form and the errors obtained at<br />
profiling, regarding the coordinates<br />
transformation:<br />
rp<br />
rp<br />
rp
(5)<br />
Bul. Inst. P olit. Iaşi, t. LVI (LX), f. 2, 2010<br />
55<br />
ζ = −ξ<br />
CATIA JAVA<br />
η = η + 26.1799.<br />
CATIA JAVA<br />
Table 2<br />
Points coordinates on tool’s profile—normal method, in Java program<br />
Crt.<br />
no. ξ [mm] η [mm] Crt. no. ξ [mm] η [mm] Crt. no. ξ [mm] η [mm]<br />
1 1.40E-06 -26.1799 248 6.698025 -0.28286 496 0.199296 25.83104<br />
2 0.05004 -26.093 249 6.698477 -0.1695 497 0.149679 25.9186<br />
3 0.099941 -26.0059 250 6.698703 -0.05562 498 0.099923 26.00594<br />
4 0.149712 -25.9185 251 6.698701 0.057729 499 0.050028<br />
26.09306<br />
5 0.19934 -25.831 252 6.698471 0.171609 500 8.88E-07 26.17994<br />
M M M M M M M<br />
Fig. 5 – Errors obtained.<br />
4. Conclusion<br />
1. The proposed algorith m use the CATIA design environment capabilities.<br />
2. The Tachet mechanism used in the problem solving is universal,<br />
excepting the profiles with singular points.<br />
3. The profiling precision of rack-gear tool is comparable with results<br />
obt ained by analytical methods, based on which was created dedicated software,<br />
in present<br />
paper, in Java language.<br />
Acknowledgements. The authors gratefully acknowledge the financial support of the<br />
Romanian Ministry of Education, Research and Innovation through grant<br />
PN_II_ID_656/2007.<br />
Received:January,<br />
20, 2010 Dunărea de Jos University of Galaţi,<br />
Manufacturing Science and Engineering Department<br />
Galaţi, Romania,<br />
e-mail: nicolae.oancea@ugal.ro<br />
;
56 Silviu Berbinschi et al<br />
R E F E R E N C E S<br />
1. L i t v i n F. L., Theory of Gearing Reference Publication 1212, NASA. Scientific<br />
and Technical Information Division, Washington D.C., 1984;<br />
2. O a n c e a N., Generarea suprafeţelor prin înfăşurare, Vol. I, Teoreme<br />
fundamentale, Editura Fundaţiei Universitare “Dunărea<br />
de Jos”, Galaţi, 2004,<br />
ISBN 973-627-106-4;<br />
3. O a n c e a N., Generarea suprafeţelor prin înfăşurare, Vol. II, Teoreme<br />
complementare, Editura Fundaţiei Universitare “Dunărea de Jos”, Galaţi, 2004,<br />
ISBN 973-627-106-4;<br />
4. T e o d o r V., O a n c e a N., D i m a M., Profilarea sculelor prin metode analitice,<br />
Editura Fundaţiei Universitare “Dunărea de Jos”, Galaţi, 2004, ISBN (10)<br />
973-627-333-4;<br />
5. D i m a M., O a n c e a N., T e o d o r V., Modelarea schemelor de aşchiere la<br />
danturare, Editura Cermi, Iaşi, 2007, ISBN 978-973-667-270-5;<br />
6. C u c u M., O a n c e a N., T e o d o r V., Discretly Known Reciprocally Enwrapping<br />
Surfaces Representation Model - Software’s Descriptions, Buletinul Institutului<br />
Politehnic din Iaşi, Publicat de <strong>Universitatea</strong> Tehnica”Gh.Asachi” Iaşi, LII<br />
(LVI), Fasc.5A, Secţia Construcţii de Maşini (2006), pp. 229-232;<br />
7. C u c u M., O a n c e a N.,T e o d o r V., Metoda tangentelor – profilarea<br />
sculei-cremalieră pentru profiluri circulare, Tehnologii Moderne<br />
Calitate<br />
Restructurare (2007), <strong>Universitatea</strong> Tehnică a Moldovei, 2, Chişinău, pp. 86-89,<br />
ISBN 978-975-45-034-8, ISBN 978-9975-45-035-2.<br />
8. * * * CATIA Version 5.19. DS.- Documentation<br />
CONTRIBUŢII LA ELABORAREA UNEI METODE GRAFICE PENTRU<br />
PROFILAREA SCULELOR CARE GENEREAZĂ PRIN ÎNFĂŞURARE<br />
I. APLICATIE PENTRU PROFILAREA SCULEI-CREMALIERA<br />
(Rezumat)<br />
În lucrare este prezentată o aplicaţie a algoritmului general prezentat, privind<br />
profilarea sculelor care generează prin înfăşurare, prin metoda rulării, în speţă<br />
scula-cremalieră generatoare a unui vârtej ordonat de profiluri. Sunt prezentaţi<br />
paşii<br />
succesivi ai procesului de profilare a sculei cremalieră, de asemenea, un exemplu<br />
numeric privind forma profilului transversal al sculei<br />
cremalieră.<br />
Calitatea metodei propusă este dovedi tă prin comparaţia rezultatelor obţinute, în<br />
cadrul<br />
aceleiaşi probleme concrete de profilare, utilizând însă una dintre metodele<br />
analitice — metoda Willis, şi un produs soft dezvoltat în Java.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCȚII DE MAȘINI<br />
THE NUMERICAL SIMULATION OF TURNING PROCESS<br />
USING FINITE ELEMENT MODELING<br />
BY<br />
CATALIN FETECAU, DANIEL-VIOREL VLAD<br />
and COSTEL MOCANU<br />
Abstract. In this paper we present a study on numerical simulation of turning<br />
process using finite element analysis program AdvantEdge TM for AISI 1060<br />
carbon steel material. The main objective is to determine the influence of different<br />
cutting parameters (depth of cut, feed rate and cutting speed) on cutting force, von<br />
Mises stress and temperature in the longitudinal turning process. The predict of<br />
cutting forces by numerical simulation is important for the optimization of the<br />
process. The simulations were conducted without coolant.<br />
Key words: turning, numerical simulation, cutting forces, temperature.<br />
1. Introduction<br />
The turning process is probably the most important process of all cutting<br />
processes. It is therefore very important to analyze the process for to<br />
optimization it.<br />
Cutting force has a value and direction of action that depend on the quality<br />
of processed material, on the cutting regime elements, geometric parameters of<br />
cutting tools, cooling-lubricating liquids [4].<br />
The resultant cutting force is defined by the relation:<br />
(1) R = ( F + F + F ) ≈(1.1…1.18)·Fx,<br />
where:<br />
Fz=(0.2…0.3)·Fx,<br />
Fy=(0.25…0.4)·Fx.<br />
2<br />
x<br />
2<br />
y<br />
2<br />
z
58 Cătălin Fetecău et al.<br />
n<br />
piece<br />
Fz<br />
tool<br />
Fx<br />
Fy<br />
chip<br />
feed rate<br />
Fig. 1 – 3D cutting force components according AdvantEdge TM .<br />
2. Numerical Simulation of Cutting Process<br />
2.1. The Mathematical Model<br />
Computer simulation of cutting process was achieved with AdvantEdge TM<br />
program. Elasto-plastic behavior of material during the cutting process is<br />
characterized by a nonlinear relationship between stresses and strains. In the<br />
mathematical modeling of surface flow movement, in software, are<br />
implemented a series of mathematical models, where the most important being<br />
model Power Law and model Drucker Prager [5].<br />
Model Power Law defined by the relation<br />
(2) p<br />
p<br />
σ( ε ε&,<br />
T ) = g(<br />
ε ) ⋅ Γ(<br />
ε&<br />
) ⋅ Θ(<br />
T )<br />
p<br />
where: ( )<br />
, ,<br />
g ε is strain hardening, Γ ( ε& ) - strain rate sensitivity, ( T )<br />
softening.<br />
Model Drucker-Prager defined by the relation<br />
(3) p<br />
σ ( ε<br />
p<br />
J1<br />
, & ε,<br />
T ) = G(<br />
ε , J1)<br />
⋅Γ(<br />
& ε ) ⋅Θ(<br />
T )<br />
where: G( , J ) p<br />
sensitivity, ( T )<br />
, ,<br />
t<br />
Θ - thermal<br />
ε 1 is strain hardening plus hydrostatic pressure, Γ (ε& ) - strain rate<br />
p<br />
Θ - thermal softening, ε - the plastic strain.<br />
p<br />
The function g( ε ) for the Drucker-Prager model is defined as
p<br />
(4) g(<br />
)<br />
(5) p<br />
g(<br />
)<br />
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 59<br />
1 n<br />
⎛ p ⎞<br />
⎜ ε<br />
p p<br />
ε = σ 0 1 + ⎟ , if ε < ε cut ;<br />
⎜ p ⎟<br />
⎝ ε 0 ⎠<br />
1 n<br />
⎛ p ⎞<br />
⎜ εcut<br />
0 1 ⎟<br />
p p<br />
ε = σ + , if ε ≥ ε ,<br />
⎜ p<br />
cut<br />
⎟<br />
⎝ ε0<br />
⎠<br />
p<br />
p<br />
where: σ 0 is the initial yield stress, ε - the plastic strain, ε 0 - the reference<br />
p<br />
plastic strain, ε cut - the cut-off strain, n - the strain hardening exponent.<br />
The rate sensitivity function Γ ( ε& ) for the Drucker-Prager model is defined<br />
as<br />
⎛ & ε ⎞<br />
⎜<br />
+<br />
⎟<br />
⎝ & ε 0 ⎠<br />
1<br />
(6) Γ(<br />
& ε ) = ⎜1<br />
⎟ , if & ε ≤ & εcut<br />
;<br />
(7)<br />
1<br />
1<br />
m<br />
⎛ 1 1 ⎞<br />
⎜ −<br />
⎟<br />
⎝ m1<br />
m2<br />
⎠<br />
⎛ & ε ⎞ m2<br />
⎛ & ε ⎞<br />
Γ(<br />
& ) =<br />
⎜ +<br />
⎟<br />
⎜ cut<br />
ε 1 1+<br />
⎟ , if & ε > & εcut<br />
.<br />
⎝ & ε0<br />
⎠ ⎝ & ε0<br />
⎠<br />
( )<br />
The thermal function Θ T for the Drucker-Prager model is defined as<br />
2 3 4 5<br />
(8) ( )<br />
Θ T = c + c T + c T + c T + c T + c T . if T < T ,<br />
(9) ( T ) = Θ(<br />
T )<br />
0<br />
1<br />
2<br />
3<br />
T −Tcut<br />
Θ cut −<br />
if T ≥ Tcut<br />
,<br />
T −T<br />
where: c0 through c5<br />
are coefficients for the polynomial fit, T - the<br />
temperature,<br />
temperature.<br />
Tcut - the linear cut-off temperature, T melt - the melting<br />
melt<br />
4<br />
cut<br />
2.2. Finite Element Machining Model<br />
The turning simulations were carried out on cylindrical extruded bar with 50<br />
mm diameter and 80mm length for AISI 1060. Length of cut L=20mm. The<br />
cutting tests were conducted without coolant.<br />
Order to study the simulation results, we set a plan of experiments as Table<br />
1. We set the tool’s parameters (tool’s angle and material), process parameters<br />
(feed rate, f, cutting speed, v and depth of cut, t). The tool’s material is carbide<br />
plates which had the parameters according to the Table 1.<br />
5<br />
cut
60 Cătălin Fetecău et al.<br />
Table 1<br />
The plan of the experiments<br />
PROCESS PARAMETERS<br />
TOOL<br />
feed, f, speed, v, depth, t,<br />
PARAMETERS<br />
mm/rev rev/min mm<br />
Clearance angle, α, º 7 0.083<br />
Side rake angle, γ, º -5 0.208<br />
Lead angle, χ, º 45 0.416<br />
Nose radius, r, mm 0.4 0.5<br />
78.54<br />
157.08<br />
314.16<br />
The mechanical properties and chemical composition of work-piece<br />
material, AISI 1060, are show in Table 2.<br />
Table 2<br />
The chemical composition and mechanical properties of AISI 1060<br />
Chemical composition, % Yield Ultimate Tensile Hardness<br />
Material<br />
Cu Mn P S<br />
strenght,<br />
MPa<br />
Strenght<br />
MPa Bhn<br />
AISI<br />
1060<br />
0.605 0.75 0.04 0.05 370 625 179<br />
For meshing were used triangular elements with 3 nodes, both for piece and<br />
for the tool. AdvantEdge TM program automatically generates the finite element<br />
network after that specified a maximum length of finite element, mean that the<br />
smallest and most finite element of edge length [5].<br />
Fig. 4 – The finite element network for the whole piece-tool-chip.<br />
The Druker-Prager model was used to simulate the process of chip breaking.<br />
1
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 61<br />
Fig. 5 – The finite element network for the whole piece-tool-chip with continuous chips<br />
of plastic strain for AISI 1060, f=0.208 mm/rev; v=157.08 m/min.<br />
Fig. 6 – Von Mises Stress f=0.208 mm/rev; v=157.08 m/min.<br />
3. The results of numerical simulations<br />
In according with the plan of experiments, it was made numerical<br />
simulation for each cutting regime. For different values of the feed rate and the<br />
cutting speed the average values of tangential force Fx in according with<br />
AdvantEdge TM are presented in Table 3.<br />
Table 3<br />
The average values of tangential force, Fx, N<br />
Feed, f, Cutting speed, v, m/min<br />
mm/rev 314,16 157,08 78,54<br />
0.083 306.177 308.298 315.186<br />
0.208 544.703 557.041 544.703<br />
0.416 880.12 950.191 1073.24<br />
0.5 1015.5 1135.32 1205.5
62 Cătălin Fetecău et al.<br />
The value of tangential force increases when feed rate increase as shows in Fig.<br />
no. 7.<br />
F [N]<br />
1400<br />
1200<br />
1000<br />
800<br />
600<br />
400<br />
200<br />
0<br />
0 0.1 0.2 0.3 0.4 0.5 0.6<br />
v1<br />
v2<br />
v3<br />
f [mm/rev]<br />
Fig. 7 – The effect of feed rate on cutting force.<br />
For the simulation made, the average values of temperature of cutting<br />
tool are presented in Table 4.<br />
Table 4<br />
The values of the temperature T, ºC<br />
Feed, f,<br />
Cutting speed, v, m/min<br />
mm/rev 314,16 157,08 78,54<br />
0.083 571.408 480.065 427.476<br />
0.208 699.709 589.149 496.942<br />
0.416 714.676 611.191 699.709<br />
0.5 729.953 622.045 729.953<br />
Fig. 8 – The temperature distribution in the cutting zone.<br />
In the FEM software AdvantEdge TM , the Johnson-Cook constitutive model<br />
were used to predict the thermal behaviour in longitudinal turning process [1].<br />
This software allows the temperature distribution, cutting forces and von<br />
Misses stress to be predicted to appropriate measurements [2].
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 63<br />
With AdvantEdge TM is possible to determine optimum machining<br />
parameters with lower costs and in short time without experimental processes<br />
[3].<br />
4. Conclusion<br />
1. This paper presents a study on numerical simulation of turning process<br />
using finite element analysis for AISI 1060 material. Simulation was done using<br />
AdvantEdge TM program. Based on numerical simulations we determine the<br />
influence of cutting parameters (depth of cut, feed rate and cutting speed ) on<br />
cutting force, von Mises stress and temperature from cutting zone.<br />
2. Based on numerical simulations to determine the size variation of cutting<br />
force and temperature parameters based on cutting regime.<br />
3. It was also determined variation von Mises stress across-chip track-and to<br />
investigate the formation of the chips.<br />
4. Future work will be as based on results of the numerical simulation will<br />
to make the experiments and will compare the results with those of simulation.<br />
Acknowledgements. The work of Drd. Daniel-Viorel Vlad was supported by Project<br />
SOP HRD – EFICIENT 61445".<br />
Received: March 20, 2010 ”Dunarea de Jos” University of Galati,<br />
Department of Machine Manyfacturing<br />
Galati, Romania,<br />
e-mail: catalin.fetecau@ugal.ro<br />
R E F E R E N C E S<br />
1. G r z e s i k W., N i e s l o n y P., FEM–based thermal modelling of the cutting<br />
process using power law temperature dependent concept. Archives of Materials<br />
Science and Engineering, 29, 2 (2008), pp. 105-108.<br />
2. Ő z e l T., Modeling of hard part machining: effect of insert edge preparation in<br />
CBN cutting tool. Journal of Materials Processing Technology, 141, (2003) pp.<br />
284-293.<br />
3. D a v i m J. P., F a r i a P., M a r a n h ã o C., C a r l o s C. A., Finite element<br />
simulation of precision machining on AISI 1045, 2008, 5º Congresso Luso-<br />
Moçambicano de Engenharia, Maputo, Moçambique, pp.1-7(ref:25A003).<br />
4. D a v i m J. P., M a r a n h ã o C. A study of plastic strain and plastic strain rate in<br />
machining of steel AISI 1045 using FEM analysis, Material and Design., 30,<br />
2009, pp. 160-165.
64 Cătălin Fetecău et al.<br />
5. *** Third Wave Systems AdvantEdge TM FEM. User’s manual, Minneapolis, 2008.<br />
6. G r z e s i k W., B a r t o s z u k M., N i e s ł o n y P., Finite element modelling of<br />
temperature distribution in the cutting zone in turning processes with differently<br />
coated tools, 13 th International Scientific Conference on Achievements in<br />
Mechanical and Materials Engineering, 2005.<br />
7. H a g l u n d A. J., K i s h a w y H. A., R o g e r s R. J., , An exploration of friction<br />
models for the chip–tool interface using an Arbitrary Lagrangian–Eulerian finite<br />
element model, Wear. 2008, 265, pp. 542-460.<br />
SIMULAREA PROCESULUI DE STRUNJIRE FOLOSIND<br />
ANALIZA CU ELEMENT FINIT<br />
(Rezumat)<br />
In această lucrare se prezintă un studiu privind simularea numerică a procesului de<br />
strunjire folosind programul de analiză cu elemente finite AdvantEdge TM în cazul,<br />
oţelului OL60 (AISI 1060). Obiectivul principal este de a determina înfluenţa<br />
parametrilor regimului de aşchiere (adâncime de aşchiere, avans şi vizeza de aşchiere)<br />
asupra componentei tangenţiale a forţei de aşchiere, Fx, a tensiunii von Mises şi a<br />
temperaturii din zona de aşchiere.<br />
Comportarea elasto-plastică a materialului în timpul procesului de aşchiere este<br />
caracterizată de o relaţie neliniară dintre tensiuni şi deformaţii.<br />
În vederea modelării matematice a deplasării suprafeţei de curgere, în programul<br />
AdvantEdge TM sunt implementate o serie de modele matematice, cele mai importante<br />
fiind: Model Power Low şi Druker Prager Model. Pentru simulările efectuate , am<br />
folosit modelul Druker Prager.<br />
Forma aşchiei reprezintă un indicator al condiţiilor de aşchiere, arătând gradul de<br />
deformare plastică suferit de stratul de material detaşat. Forma aşchiei depinde de natura<br />
materialului prelucrat, geometria sculei, regimul de aşchiere etc. Se deosebesc aşchii de<br />
rupere şi aşchii de deforrnare plastică.<br />
Rezultatele obţinute vor sta la baza realizării experimentelor fizice, astfel încât să<br />
putem confrunta rezultatele obţinute din simulare cu cele din experiment.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCȚII DE MAȘINI<br />
EXPERIMENTAL RESEARCHES REGAR<strong>DIN</strong>G A NEW<br />
METHOD FOR CUTTING FORCES EVALUATION<br />
BY<br />
MIRCEA COZMÎNCĂ, CRISTIAN CROITORU<br />
and CĂTĂLIN UNGUREANU<br />
Abstract. The paper presents some preliminary experimental researches<br />
regarding the cutting forces level when cutting four types of steel. The aim of<br />
these tests is to finalize the calculus models for cutting force components Fz, Fx, Fy on a cutting tooth. The new models are considering the area of the chip’s<br />
theoretical cross section (txs), the compression yield point of the cut material and<br />
the chips contraction coefficient, as presented in our previous papers. The chips<br />
contraction coefficient includes all the working parameters’ influences and also<br />
the interdependencies between them. Beside those elements, the calculus models<br />
contain a theoretical component, depending on the constructive angles of the tool,<br />
γ, K and λ and the friction coefficient between chips and the cutting tool’s rake<br />
face.<br />
Key words: model, chips contraction, cutting force, experimental.<br />
1. Introduction<br />
The previous papers [1]-[4] were presented the new calculus models for<br />
cutting force evaluation (Eqs. 1), that could replace the traditional equations for<br />
Fz, Fx, Fy components. The theoretical quantities Cz, Cx, Cy in Eqs. (1) result<br />
using the Eqs.(2) and (3), if the chips deviation angle η=0 [5], and with Eqs.(4)<br />
if η≠0.<br />
(1)<br />
F<br />
F<br />
F<br />
z<br />
x<br />
y<br />
= C<br />
= C<br />
z<br />
= C<br />
x<br />
y<br />
σ<br />
σ<br />
σ<br />
o<br />
o<br />
o<br />
t s<br />
t s<br />
t s<br />
C<br />
C<br />
C<br />
1+<br />
m<br />
d<br />
z<br />
1+<br />
m<br />
d<br />
x<br />
1+<br />
m<br />
d<br />
y<br />
,<br />
,<br />
.
66 Mircea Cozmîncă et al.<br />
(2)<br />
(3)<br />
(4)<br />
C<br />
C<br />
C<br />
z<br />
x<br />
y<br />
= cosγ<br />
cos<br />
k<br />
k<br />
k<br />
k<br />
C<br />
C<br />
C<br />
1<br />
2<br />
3<br />
4<br />
z<br />
x<br />
y<br />
= k cosλ,<br />
1<br />
= k<br />
2<br />
= k cos K − k k .<br />
3<br />
= cosγ<br />
= cosγ<br />
sin K + k k<br />
N0<br />
N0<br />
= sin λ cos K,<br />
= sin λ sin K.<br />
1<br />
1<br />
3<br />
4<br />
( 1+<br />
μ tgγ<br />
N ) ,<br />
0<br />
( μ − tgγ<br />
) ,<br />
N0<br />
( λ + η)[<br />
1+<br />
μtgγ<br />
] ,<br />
( λ + η)<br />
cos K[<br />
1+<br />
tgγ<br />
] + cosγ<br />
sin(<br />
K −η<br />
)[ μ − tgγ<br />
] ,<br />
= cosγ<br />
sin<br />
= cosγ<br />
cos(<br />
K −η<br />
)[ μ − tgγ<br />
] − cos K sin(<br />
λ + η)<br />
sin K[<br />
1+<br />
μtgγ<br />
].<br />
In Eqs.(3), the rake angle γ is measured in the plane normal to the active<br />
cutting edge and in Eqs.(4) this angle is measured in the plane including the<br />
chips’ real flow direction (η≠0). In the same equations, μ represents the friction<br />
coefficient between the chips and the tooth’s rake face, considering μ=0.5...0.6<br />
in designing activities [5].<br />
Using the models in Eqs.(1) ask for the quantities mx, mz, my, that are<br />
amplifying the cutting forces level through the chips contraction coefficient.<br />
Therefore, experimental tests are necessary for the forces Fz, Fx, Fy and also for<br />
the chips contraction coefficient Cd, when cutting with a single tooth.<br />
2. Experimental Tests to Validate the Models for Fz, Fx, Fy<br />
In order to validate the models in Eqs.(1) for the force components Fz, Fx, Fy<br />
on the cutting tooth, a Kistler dynamometer is used to measure the main force<br />
component Fz when cutting four different types of steel. In the same time, the Cd<br />
values are determining by measurement of the chips lengths. These chips are<br />
resulted from certain cutting lengths limited by four longitudinal and equidistant<br />
channels executed on the workpiece and refilled with brass wedges, in order to<br />
ensure the continuity of the cutting process [5].<br />
The experimental values for Cd result using the model in Eq.(5), where the<br />
cutting lenth L = 0.25πd–c and the chip’s length, La, result by measuring a<br />
number of n chips. The quantity c represents the channel width, measured on<br />
the workpiece circumference of diameter D.<br />
,
(5)<br />
Bul. Inst . Polit. Iaşi, t. LVI (LX), f. 2, 2010<br />
67<br />
C<br />
d<br />
L<br />
=<br />
L<br />
a<br />
,<br />
The experimental tests were conducted using five lathe tools (Co...C4)<br />
having different values of the main constructive angles, as presented in Table 1.<br />
This table includes also the calculated values for k1... k4 and Cz, Cx, Cy, based on<br />
Eqs.(2) and (3).<br />
Table 1<br />
Lathe tools’ constructive geometry<br />
Lathe<br />
tool<br />
Main constructive geometry<br />
γ λ K α r<br />
[°] [°] [°] [°] [mm]<br />
k1 Values calculated with Eqs. (2) and (3)<br />
k2 k3 k4 Cz Cx Cy Co 0 0 70 10 0 1 0.5 0 0 1 0.47 0.17<br />
C1 -10 0 70 10 0 0.898 0.67 0 0 0.898 0.63 0.23<br />
C2 0 -10 70 10 0 1 0.5 -0.06 -0.16 0.985 0.41 0.33<br />
C3 0 10 70 10 0 1 0.5 0.06 0.16 0.985 0.53 0.01<br />
C4 10 0 70 10 0 1.07 0.32 0 0 1.07 0.3 0.01<br />
The specific working conditions and the experimental results (Fz, Cd, mz)<br />
are presented in Tables 2 – 6. The values of mz were obtained using Eq.(6),<br />
where Cz is according to Table 1 and σo is a steel type characteristic.<br />
(6)<br />
F<br />
zexp<br />
z<br />
L<br />
a<br />
o<br />
=<br />
n<br />
∑<br />
i=<br />
1<br />
L<br />
= C σ t s C<br />
ai<br />
1+<br />
mz<br />
d<br />
Table 2<br />
OL44 – The influences of tool´s geometry and cutting feed<br />
Cut<br />
material/<br />
Tool´s type<br />
σo [daN/<br />
mm 2 ]<br />
t<br />
[mm]<br />
s<br />
[mm/<br />
rev]<br />
n<br />
[rev/<br />
min]<br />
L<br />
[mm]<br />
La<br />
[mm]<br />
D<br />
[mm]<br />
Cd<br />
Fz<br />
[daN]<br />
mz<br />
OL44/ Co 29 2<br />
0.1<br />
0.25<br />
630<br />
630<br />
42.5<br />
42.5<br />
16.52<br />
17.38<br />
58.5<br />
58.5<br />
2.57<br />
2.44<br />
54.9<br />
113.5<br />
1.38<br />
1.30<br />
OL44/ C1 29 2<br />
0.1<br />
0.25<br />
630<br />
630<br />
38.5<br />
38.5<br />
13.03<br />
14.18<br />
53.5<br />
53.5<br />
2.91<br />
2.72<br />
63.4<br />
125.8<br />
1.34<br />
1.27<br />
OL44/ C2 29 2<br />
0.1<br />
0.25<br />
630<br />
630<br />
41.5<br />
41.5<br />
14.25<br />
15.11<br />
57.3<br />
57.3<br />
2.91<br />
2.75<br />
58.16<br />
123.6<br />
1.17<br />
1.14<br />
OL44/ C3 29 2<br />
0.1<br />
0.25<br />
630<br />
630<br />
41.5<br />
41.5<br />
15.22<br />
16.18<br />
57.3<br />
57.3<br />
2.73<br />
2.57<br />
47.9<br />
118.0<br />
1.12<br />
1.24<br />
OL44/ C4 29 2<br />
0.1<br />
0.25<br />
630<br />
630<br />
41.5<br />
41.5<br />
16.47<br />
17.33<br />
57.3<br />
57.3<br />
2.52<br />
2.39<br />
47.65<br />
118.0<br />
1.21<br />
1.32<br />
n.<br />
.
68 Mircea Cozmîncă et al.<br />
Table 3<br />
40C10 – The influences of tool´s geometry and cutting feed<br />
Cut<br />
material/<br />
Tool´s type<br />
σo [daN/<br />
mm 2 ]<br />
t<br />
[mm]<br />
s<br />
[mm/<br />
rev]<br />
n<br />
[rev/<br />
min]<br />
L<br />
[mm]<br />
La [mm]<br />
D<br />
[mm]<br />
Cd Fz [daN]<br />
mz<br />
40C10/ Co 78 2<br />
0.1<br />
0.25<br />
630<br />
630<br />
36.56<br />
36.56<br />
19.86<br />
20.74<br />
51.0<br />
51.0<br />
1.84<br />
1.76<br />
52.5<br />
110.0<br />
0.98<br />
0.82<br />
40C10/ C1 78 2<br />
0.1<br />
0.25<br />
630<br />
630<br />
37.03<br />
37.03<br />
17.49<br />
19.38<br />
51.6<br />
51.6<br />
2.12<br />
1.91<br />
59.0<br />
117.3<br />
0.91<br />
0.86<br />
40C10/ C2 78 2<br />
0.1<br />
0.25<br />
630<br />
630<br />
36.56<br />
36.56<br />
20.32<br />
22.84<br />
51.0<br />
51.0<br />
1.80<br />
1.60<br />
55.0<br />
113.0<br />
1.16<br />
1.29<br />
40C10/ C3 78 2<br />
0.1<br />
0.25<br />
630<br />
630<br />
36.56<br />
36.56<br />
23.8<br />
26.36<br />
51.0<br />
51.0<br />
1.54<br />
1.39<br />
54.7<br />
104.7<br />
1.95<br />
2.05<br />
40C10/ C4 78 2<br />
0.1<br />
0.25<br />
630<br />
630<br />
36.56<br />
36.56<br />
21.14<br />
22.62<br />
51.0<br />
51.0<br />
1.73<br />
1.62<br />
40.4<br />
103.4<br />
0.60<br />
0.88<br />
Cut<br />
material/<br />
Tool´s type<br />
MoCM14/<br />
Co<br />
MoCM14/<br />
C1<br />
MoCM14/<br />
C 2<br />
MoCM14/<br />
C3<br />
MoCM14/<br />
C 4<br />
Table 4<br />
MoCM14 – The influences of tool´s geometry and cutting feed<br />
σo [daN/<br />
mm 2 ]<br />
t<br />
[mm]<br />
s<br />
[mm/<br />
rev]<br />
n<br />
[rev/<br />
min]<br />
L<br />
[mm]<br />
La [mm]<br />
D<br />
[mm]<br />
Cd<br />
Fz<br />
[daN]<br />
mz<br />
74 1<br />
0.1<br />
0.25<br />
400<br />
400<br />
51.09<br />
51.09<br />
30.53<br />
32.16<br />
69.5<br />
69.5<br />
1.67<br />
1.59<br />
57.9<br />
116.4<br />
2.99<br />
2.96<br />
74 1<br />
0.1<br />
0.25<br />
400<br />
400<br />
50.54<br />
50.54<br />
23.47<br />
25.12<br />
68.8<br />
68.8<br />
2.15<br />
2.01<br />
60.9<br />
134.3<br />
1.88<br />
1.98<br />
74 1<br />
0.1<br />
0.25<br />
400<br />
400<br />
50.14<br />
50.14<br />
28.11<br />
29.96<br />
68.3<br />
68.3<br />
1.78<br />
1.67<br />
58.9<br />
119.5<br />
2.60<br />
2.64<br />
74 1<br />
0.1<br />
0.25<br />
400<br />
400<br />
50.14<br />
50.14<br />
18.08<br />
20.14<br />
68.3<br />
68.3<br />
2.77<br />
2.49<br />
57.1<br />
118.0<br />
1.01<br />
1.04<br />
74 1<br />
0.1<br />
0.25<br />
400<br />
400<br />
50.14<br />
50.14<br />
21.84<br />
24.13<br />
68.3<br />
68.3<br />
2.30<br />
2.10<br />
50.3<br />
102.6<br />
1.22<br />
1.24<br />
Table 5<br />
OSC10 – The influences of tool´s geometry and cutting feed<br />
Cut material/<br />
Tool´s type<br />
σo [daN/<br />
mm 2 ]<br />
t<br />
[mm]<br />
s<br />
[mm/<br />
rev]<br />
n<br />
[rev/<br />
min]<br />
L<br />
[mm]<br />
La [mm]<br />
D<br />
[mm]<br />
Cd<br />
Fz [daN]<br />
mz<br />
OSC10/ Co 52 2<br />
0.1<br />
0.25<br />
500<br />
500<br />
57.9<br />
57.9<br />
25.6<br />
28.3<br />
78.2<br />
78.2<br />
2.26<br />
2.05<br />
59.4<br />
119.2<br />
1.13<br />
1.12<br />
OSC10/ C1 52 2<br />
0.1<br />
0.25<br />
500<br />
500<br />
57.0<br />
57.0<br />
22.5<br />
24.3<br />
77.0<br />
77.0<br />
2.53<br />
2.35<br />
65.3<br />
137.2<br />
1.09<br />
1.07<br />
OSC10/ C2 52 2<br />
0.1<br />
0.25<br />
500<br />
500<br />
57.0<br />
57.0<br />
23.9<br />
25.9<br />
77.0<br />
77.0<br />
2.38<br />
2.20<br />
61.2<br />
128.6<br />
1.05<br />
1.04<br />
OSC10/ C3 52 2<br />
0.1<br />
0.25<br />
500<br />
500<br />
57.0<br />
57.0<br />
27.6<br />
29.9<br />
77.0<br />
77.0<br />
2.06<br />
1.90<br />
57.8<br />
113.9<br />
1.39<br />
1.32<br />
OSC10/ C4 52 2<br />
0.1<br />
0.25<br />
500<br />
500<br />
57.9<br />
57.9<br />
30.0<br />
32.8<br />
78.2<br />
78.2<br />
1.93<br />
1.76<br />
54.4<br />
106.4<br />
1.41<br />
1.36
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 69<br />
Table 6<br />
The influence of main cutting speed<br />
Cut material/<br />
Tool´s type<br />
σo [daN/<br />
mm 2 ]<br />
t<br />
[mm]<br />
s<br />
[mm/<br />
rev]<br />
v<br />
[m/<br />
min]<br />
Cd Fz [daN]<br />
mz<br />
85 2.62 60 1.43<br />
OL44/ Co 29 2 0.1 115 2.57 55 1.38<br />
165 2.50 52 1.37<br />
80 2-35 68 1.19<br />
OSC10/ Co 52 2 0.1 122.5 2.26 63 1.21<br />
164 2.20 61 1.24<br />
87 1.67 30 1.72<br />
MoCM14/ Co 74 1 0.1 120 1.52 27 2.08<br />
160 1.48 25 2.09<br />
80 1.90 65 1.21<br />
40C10/ C3 78 2 0.1 102 1.86 63 1.24<br />
Constant parameters:<br />
150 1.80 58 1.22<br />
K=70°; α=10°; γ=0°; λ=0°; wear hα=0; cutting in air<br />
For the most ductile steel type (OL44) and the most resistant one (40C10),<br />
the experimental results regarding the influences of the constructive angles,<br />
cutting feed and main cutting speed are illustrated in the diagrams in Figs. 1-4.<br />
OL44<br />
λ=0°<br />
s=0.1 mm/rev<br />
Fig.1a – The influence of angle γ for OL44 steel.
70 Mircea Cozmîncă et al.<br />
OL44<br />
γ=0°<br />
s=0.1 mm/rev<br />
Fig.1b – The influence of angle λ for OL44 steel.<br />
40C10<br />
λ=0°<br />
s=0.1 mm/rev<br />
Fig.2a – The influence of angle γ for 40C10 steel.<br />
40C10<br />
γ=0°<br />
s=0.1 mm/rev<br />
Fig.2b – The influence of angle λ for 40C10 steel.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 71<br />
Fig.3a – The influence of feed values for OL44 steel.<br />
OL44<br />
γ=0°<br />
λ=0°<br />
OL44<br />
γ=0°; λ=0°<br />
s=0.1 mm/rev<br />
Fig.3b – The influence of speed values for OL44 steel.<br />
40C10<br />
γ=0°<br />
λ=0°<br />
Fig.4a – The influence of feed values for 40C10 steel.
72 Mircea Cozmîncă et al.<br />
40C10<br />
γ=0°; λ=0°<br />
s=0.1 mm/rev<br />
Fig.4b – The influence of speed values for 40C10 steel<br />
From the diagrams in Figs.1 – 4 result that Fz and Cd are decreasing when<br />
the angles γ, λ and cutting speed values are growing. When the cutting feed is<br />
increasing, the force Fz grows and the coefficient Cd decrease. For OL44 steel,<br />
the Fz values are smaller and the Cd are larger than those resulted for 40C10<br />
steel.<br />
For OL44 steel, the mz exponent is relatively constant in the 1.2 – 1.4 range.<br />
For 40C10 steel, the mz values are variable in a larger domain, 0.8 – 1.22. For<br />
λ=0° and γ=0° the mz exponent may achieve even the value of 1.95 (Fig.2b).<br />
This is the reason why mz it is necessary to be reached for every material type.<br />
The chemical composition, the nature and the structural micro constituents´<br />
distribution of the cut material determine the plastic deformation capability.<br />
Consequently, the cutting forces level, the values of Cd and the mz values<br />
through which Cd amplify the cutting forces values are indirectly influenced.<br />
4. Conclusions<br />
1. The experimental researches regarding the values of cutting force Fz and<br />
coefficient Cd for the four steel types were stressing the known dependencies<br />
from angles γ and λ, cutting feed and main cutting speed.<br />
2. The level of influence of the constructive angles γ and λ, of cutting feed<br />
and speed takes different values depending on the material being cut.<br />
3. Experimental tests are necessary for each material, to determine the<br />
values for mz, mx, my exponents in order to use the models in Eqs. (1)-(4) for<br />
the force components Fz, Fx, Fy on a cutting tooth.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 73<br />
Received: ”Gheorghe Asachi” Technical University,<br />
Department of Machine Tools<br />
Iasi, Romania<br />
e-mail: cozminca@tcm.tuiasi.ro<br />
R E F E R E N C E S<br />
1. C o z m î n c ă M., C o z m î n c ă I., P o e n a r u S., A new Model for Estimating<br />
the Force Components Fz, Fx, Fy when Cutting Metals with Single Tooth Tools,<br />
Buletinul Inst. Polit. Iași, editat de Univ. Tehnică “Gheorghe Asachi”, LV(LIX),<br />
Construcții de mașini, 1, 2009, pp.1-7.<br />
2. C o z m î n c ă M., C o z m î n c ă I., P o e n a r u S., Variation of the Ratios<br />
between Forces Fz, Fx, Fy and the Plastic Strain Force in Single Tooth Cutting.<br />
Buletinul Inst. Polit. Iași, editat de Univ. Tehnică “Gheorghe Asachi”, LV(LIX),<br />
Construcții de mașini, 1, 2009, pp.8-14.<br />
3. P o e n a r u S., C o z m î n c ă M., Model for the Deformation Force at Metals<br />
Cutting with Single Tooth Tools. Buletinul Inst. Polit. Iași, editat de Univ.<br />
Tehnică “Gheorghe Asachi”, LV(LIX), Construcții de mașini, 2, 2009, pp.1-7.<br />
4. C o z m î n c ă M., C o n s t a n t i n e s c u C., Bazele așchierii. Ed “Gh. Asachi”,<br />
Iaşi, 1995.<br />
CERCETĂRI EXPERIMENTALE PENTRU VALIDAREA UNEI<br />
NOI METODE DE EVALUARE A FORȚELOR DE AȘCHIERE<br />
(Rezumat)<br />
În lucrare se prezintă modul de realizare a unor cercetări experimentale preliminare<br />
în vederea stabilirii mărimii forțelor de așchiere, la strunjirea a patru mărci de oțel, în<br />
scopul definitivării modelelelor de calcul pentru componentele Fz, Fx, Fy ale forței de<br />
așchiere la nivelul uni dinte așchietor. Noile modele de calcul au în vedere aria secțiunii<br />
transversale a așchiei nedeformate (txs), limita de curgere a materialului așchiat și<br />
coeficientul de deformare plastică a așchiilor. Coeficientul de deformare plastică a<br />
așchiilor include toate influențele parametrilor de lucru, precum și interdependențele<br />
dintre aceste influențe. Pe lîngă aceste elemente, modelele de calcul propuse cuprind și<br />
o componentă teoretică, în funcție de unghiurile constructive ale dintelui, γ, K, λ și<br />
coeficientul de frecare dintre așchii și suprafața de degajare a dintelui așchietor.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
FACE MILLING FORCES DEPEN<strong>DIN</strong>G ON THE<br />
FORCES DEVELOPED ON A SINGLE-TOOTH<br />
BY<br />
ANA-MARIA MATEI and MARIUS NICOLAE MILEA<br />
Abstract. This paper presents some theoretical models for the evaluation of<br />
cutting forces components in face milling, developed on a single−tooth level. The<br />
cutting forces at face milling depend on the every variant of face milling process<br />
(teeth number which simultaneously cut) and the relative position of cutting teeth<br />
and the material being cut (cut - down milling and cut - up milling). Regarding to<br />
this, the determination of cutting force components at face milling FZ, FX, FY is<br />
based on the forces for a single tooth, the cutting tooth position beside the XYZ<br />
coordinates system of the tool and the number of teeth that simultaneously cut.<br />
Key words: cutting, face milling, forces.<br />
1. Introduction<br />
The cutting forces developed during the milling process are affected by a<br />
series of parameters, including cutting feed, the milling depth and width, tooth<br />
number, tooth geometry, the relative position of cutting teeth and the material<br />
being cut, and the hardness of the material being cut. Therefore, the most<br />
calculus models of face milling forces are including these parameters, and also<br />
some specific elements of the tools’ circular motion [1].<br />
This paper presents new theoretical models for the evaluation of cutting<br />
force components at face milling, considering besides the parameters described<br />
above, also the specific elements for each variant of face milling process:<br />
complete (full), symmetrical incomplete or unsymmetrical incomplete, and the<br />
relative position of cutting teeth and the material being cut: cut-down milling<br />
and cut-up milling.<br />
The face milling forces are also depending on the number of teeth that<br />
simultaneously cut (odd or even number of teeth) [2].
76 Ana-Maria Matei and Marius-Nicolae Milea<br />
2. Theoretical Models of Cutting Force Components in Face Milling<br />
depending on the Cutting Forces developed on a Single-Tooth<br />
The face milling forces components acting on the tool FZ , FX si FY are<br />
depending on the cutting forces components developed on a single tooth level<br />
Fz , Fx si Fy , which are influenced by the cutting tooth geometry (γ, k, λ), the<br />
standard compression yield point of the material being cut, the cross - sectional<br />
area of the chip, the chips’ friction coefficient μ, and the chips´ contraction<br />
coefficient Cd [3,4].<br />
Considering the specific elements of the symmetrical or unsymmetrical<br />
milling, there are several possibilities of modelling the cutting force<br />
components in face milling.<br />
2.1. Theoretical Models of Cutting Force Components<br />
in Symmetrical Face Milling<br />
Fig. 1 shows the geometrical models of complete and incomplete face<br />
milling. The values of specific elements for each milling variant result as<br />
follows: the contact angle Ψ = 180°, t = D, the number of teeth that<br />
simultaneously cut zs = z/2 at complete face milling and Ψ < 180°, t < D,<br />
z t<br />
= arcsin at incomplete face milling [2].<br />
z s<br />
π<br />
D<br />
The theoretical models for the evaluation of cutting force components in<br />
face milling are developed in the light of two simplifying assumptions, as<br />
follows:<br />
a) The cutting force components on a tooth, respectively Fz – the main cutting<br />
force (the tangential force), Fx – the radial force and Fy – the axial force, take<br />
values depending on the cross-sectional area of undeformed chip;<br />
b) The tooth z1 is the one to which the tooth coordinates system xyz<br />
corresponds to the mill coordinates system (FZ = Fz, FX = Fx, FY = Fy), and the<br />
other teeth that simultaneously cut are placed equidistant (φ = 2π/z).<br />
The values of cutting force components in symmetrical face milling depend<br />
on zs and Fz, Fx , Fy , as follows:<br />
a) For symmetrical milling with odd zs the Eq. (1) results.<br />
(1)<br />
F<br />
F<br />
Z<br />
X<br />
( zs<br />
/ 2)<br />
−1<br />
( zs<br />
/ 2)<br />
−1<br />
2π<br />
⎡ 2π<br />
⎤<br />
= Fz<br />
+ 2 ∑ Fz<br />
cos( zi<br />
) = Fz<br />
⎢1<br />
+ 2 ∑ Fz<br />
cos( zi<br />
) ⎥ ,<br />
1 z ⎣ 1 z ⎦<br />
( zs<br />
/ 2)<br />
−1<br />
( zs<br />
/ 2)<br />
−1<br />
2π<br />
⎡ 2π<br />
⎤<br />
= Fx<br />
+ 2 ∑ Fx<br />
cos( zi<br />
) = Fx<br />
⎢1<br />
+ 2 ∑ Fx<br />
cos( zi<br />
) ⎥ ,<br />
1 z ⎣ 1 z ⎦<br />
F = F ⋅ z .<br />
Y<br />
y<br />
s
(2)<br />
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 77<br />
Fig.1 – Forces model in symmetrical face milling.<br />
b) For symmetrical milling with even zs the Eq. (2) results.<br />
F<br />
F<br />
Z<br />
X<br />
= 2<br />
= 2<br />
z / 2<br />
s<br />
∑<br />
1<br />
z / 2<br />
s<br />
∑<br />
1<br />
2π<br />
Fz<br />
cos( zi<br />
) = 2Fz<br />
z<br />
2π<br />
Fx<br />
cos( zi<br />
) = 2Fx<br />
z<br />
F = F ⋅ z .<br />
Y<br />
y<br />
s<br />
z / 2<br />
s<br />
∑<br />
1<br />
z / 2<br />
s<br />
∑<br />
1<br />
2π<br />
cos( zi<br />
) ,<br />
z<br />
2π<br />
cos( zi<br />
) ,<br />
z<br />
From Eqs. (1) and (2) result that in symmetrical face milling the forces FZ,<br />
FX and FY depend on the components Fz, Fx and Fy developed on a single-tooth<br />
level, the number of teeth that simultaneously cut (zs) and the cutting tooth<br />
position beside the XYZ coordinates system of the tool. The axial component FY<br />
doesn’t depend on the relative position of the cutting tooth.<br />
2.2. Theoretical Models of Cutting Force Components in<br />
Unsymmetrical Face Milling<br />
In this case, the values of cutting force components FZ, FX, FY are influenced<br />
by the relative position of the tool and the material being cut, resulting different<br />
equations for cut-up (conventional) milling and cut-down (climb) milling.<br />
Fig. 2 shows the geometrical models of unsymmetrical cut-up milling and its<br />
three possible variants:<br />
a) Unsymmetrical face milling with t' = D/2, Ψ = 90°, and z = z / 4 ;<br />
s<br />
1
78 Ana-Maria Matei and Marius-Nicolae Milea<br />
b) Unsymmetrical face milling with t" < D/2, Ψ < 90° and<br />
z 2t<br />
zs = arccos( 1−<br />
) ;<br />
2 2π<br />
D<br />
c) Unsymmetrical face milling with t > D/2 , where t = t' + t"' and t"' = t -D/2,<br />
z 2t<br />
Ψ > 90° and z s = z , where<br />
3 s + z 1 s z = arcsin( −1)<br />
4<br />
s<br />
.<br />
4 2π<br />
D<br />
Fig. 2 – Forces model in unsymmetrical cut-up face milling.<br />
According to the geometrical model of face milling process shown in Fig. 2,<br />
the cutting force components FZ, FX and FY result using the Eqs.(3).<br />
a) zs = z / 4<br />
(3)<br />
1<br />
F<br />
F<br />
Z<br />
X<br />
=<br />
=<br />
z / 4<br />
∑<br />
0<br />
z / 4<br />
∑<br />
z 2t<br />
b) zs = arccos( 1−<br />
)<br />
2 2π<br />
D<br />
(4)<br />
F<br />
F<br />
Z<br />
X<br />
=<br />
=<br />
z<br />
0<br />
s2<br />
∑<br />
z<br />
0<br />
s2<br />
∑<br />
0<br />
2π<br />
Fz<br />
sin( zi)<br />
−<br />
z<br />
2π<br />
Fx<br />
sin( zi)<br />
+<br />
z<br />
Y<br />
y<br />
z / 4<br />
∑<br />
0<br />
z / 4<br />
∑<br />
0<br />
F = F ⋅ z .<br />
2π<br />
Fz<br />
sin( zi)<br />
−<br />
z<br />
2π<br />
Fx<br />
sin( zi)<br />
+<br />
z<br />
z<br />
s<br />
1<br />
s2<br />
∑<br />
z<br />
0<br />
s2<br />
∑<br />
0<br />
F = F ⋅ z .<br />
Y<br />
y<br />
s<br />
2<br />
2π<br />
Fx<br />
cos( zi)<br />
,<br />
z<br />
2π<br />
Fz<br />
cos( zi)<br />
,<br />
z<br />
2π<br />
Fx<br />
cos( zi)<br />
,<br />
z<br />
2π<br />
Fz<br />
cos( zi)<br />
,<br />
z
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 79<br />
c) For the unsymmetrical milling with t > D/2, we suppose that for a<br />
number of teeth that simultaneously cut z = z + z , the values of cutting<br />
force components will be obtained as the sum of the values of cutting force<br />
components corresponding to a tooth z , Eqs. (3), and a tooth z , Eqs. (5).<br />
(5)<br />
F<br />
F<br />
Z<br />
X<br />
=<br />
=<br />
z<br />
−1<br />
s4<br />
∑<br />
z<br />
1<br />
s4<br />
∑<br />
s1<br />
2π<br />
Fz<br />
cos( zi)<br />
+<br />
z<br />
−1<br />
1<br />
2π<br />
Fx<br />
cos( zi)<br />
−<br />
z<br />
Y<br />
y<br />
s<br />
3<br />
z<br />
s<br />
s4<br />
∑<br />
z<br />
1<br />
−1<br />
1<br />
−1<br />
s4<br />
∑<br />
s<br />
1<br />
F = F ⋅ z .<br />
4<br />
s<br />
4<br />
s4<br />
2π<br />
Fx<br />
sin( zi<br />
) ,<br />
z<br />
2π<br />
Fz<br />
sin( zi<br />
) ,<br />
z<br />
The values of cutting force components corresponding to the total number<br />
of teeth that simultaneously cut ( z ) result using the Eqs. (6).<br />
(6)<br />
F<br />
F<br />
Z<br />
X<br />
=<br />
=<br />
z / 4<br />
∑<br />
0<br />
z / 4<br />
∑<br />
0<br />
2π<br />
Fz<br />
sin( zi<br />
) +<br />
z<br />
2π<br />
Fx<br />
sin( zi<br />
) +<br />
z<br />
s<br />
3<br />
z<br />
−1<br />
s4<br />
∑<br />
z<br />
1<br />
+<br />
−1<br />
s4<br />
∑<br />
2π<br />
Fz<br />
cos( zi<br />
) −<br />
z<br />
z<br />
−1<br />
s4<br />
∑<br />
1<br />
x<br />
1<br />
z −1<br />
z / 4<br />
∑<br />
2π<br />
Fx<br />
sin( zi<br />
)<br />
z<br />
2π<br />
F cos( zi<br />
) +<br />
z<br />
s4<br />
2π<br />
− ∑ Fz<br />
sin( zi<br />
)<br />
1 z<br />
F = F ⋅ z .<br />
Y<br />
y<br />
s<br />
3<br />
0<br />
z / 4<br />
∑<br />
0<br />
2π<br />
Fx<br />
cos( zi<br />
)<br />
z<br />
,<br />
2π<br />
Fz<br />
cos( zi<br />
)<br />
z<br />
,<br />
It can be observed that in case of cut-up face milling with cu t' = D/2, and t''<br />
< D/2, the same relations for the cutting force components FZ, FX, FY are<br />
obtained, distinguishing only by the number of teeth that simultaneously cut.<br />
Fig. 3 is illustrating the geometrical models of unsymmetrical cut-down<br />
milling and the same three possible variants:
80 Ana-Maria Matei and Marius-Nicolae Milea<br />
Fig. 3 – Forces model in unsymmetrical cut-down face milling.<br />
For all variants of unsymmetrical cut-down face milling, the cutting force<br />
components FZ, FX, FY are given by the Eqs.(7) – (8).<br />
a) zs = z / 4<br />
1<br />
z / 4<br />
z / 4 2π<br />
2π<br />
FZ<br />
= ∑ Fz<br />
cos( zi<br />
) + ∑ Fx<br />
sin( zi)<br />
,<br />
z<br />
z<br />
(7)<br />
F<br />
X<br />
=<br />
1<br />
z / 4<br />
∑<br />
z 2t<br />
b) zs = arccos( 1−<br />
)<br />
2 2π<br />
D<br />
F<br />
Z<br />
=<br />
z<br />
1<br />
s2<br />
∑<br />
0<br />
2π<br />
Fx<br />
cos( zi)<br />
−<br />
z<br />
Y<br />
y<br />
1<br />
z / 4<br />
∑<br />
s<br />
1<br />
F = F ⋅ z .<br />
2π<br />
Fz<br />
cos( zi<br />
) +<br />
z<br />
z<br />
1<br />
s2<br />
∑<br />
0<br />
2π<br />
Fz<br />
sin( zi)<br />
,<br />
z<br />
2π<br />
Fx<br />
sin( zi<br />
) ,<br />
z<br />
z<br />
(8)<br />
s<br />
z<br />
2<br />
s2<br />
2π<br />
2π<br />
FX<br />
= ∑ Fx<br />
cos( zi)<br />
− ∑ Fz<br />
sin( zi)<br />
,<br />
0 z<br />
0 z<br />
FY = Fy<br />
⋅ zs<br />
.<br />
2<br />
c) Similarly to cut-up milling, for the unsymmetrical cut-down milling<br />
with t > D/2, we use the same method for determining the values of components<br />
FZ , FX , FY . First, will be determined the values of cutting force components<br />
corresponding to a number of teeth that simultaneously cut z , Eqs. (9).<br />
s4
(9)<br />
F<br />
F<br />
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 81<br />
Z<br />
X<br />
=<br />
=<br />
z<br />
−1<br />
s4<br />
∑<br />
z<br />
0<br />
−1<br />
s4<br />
∑<br />
0<br />
2π<br />
Fz<br />
sin( zi)<br />
−<br />
z<br />
2π<br />
Fx<br />
sin( zi<br />
) +<br />
z<br />
Y<br />
y<br />
z<br />
−1<br />
s4<br />
∑<br />
z<br />
F = F ⋅ z<br />
0<br />
−1<br />
s4<br />
∑<br />
s<br />
0<br />
4<br />
2π<br />
Fx<br />
cos( zi)<br />
z<br />
2π<br />
Fz<br />
cos( zi)<br />
z<br />
The values of cutting force components corresponding to a total number of<br />
teeth that simultaneously cut ( z ) result using the Eqs.(10).<br />
(10)<br />
F<br />
F<br />
Z<br />
X<br />
=<br />
=<br />
z / 4<br />
∑<br />
1<br />
z / 4<br />
∑<br />
1<br />
s3<br />
2π<br />
Fz<br />
cos( zi<br />
) +<br />
z<br />
−<br />
2π<br />
Fx<br />
cos( zi<br />
) +<br />
z<br />
z<br />
−1<br />
s4<br />
∑<br />
z<br />
0<br />
z −1<br />
s4<br />
∑<br />
z<br />
0<br />
−1<br />
s4<br />
∑<br />
0<br />
z −1<br />
2π<br />
F sin( zi<br />
) +<br />
z<br />
2π<br />
Fx<br />
cos( zi<br />
)<br />
z<br />
2π<br />
Fx<br />
sin( zi<br />
) −<br />
z<br />
s4<br />
2π<br />
+ ∑ Fz<br />
cos( zi<br />
)<br />
0 z<br />
F = F ⋅ z .<br />
Y<br />
y<br />
s3<br />
z / 4<br />
∑<br />
1<br />
z / 4<br />
∑<br />
1<br />
2π<br />
Fx<br />
sin( zi<br />
)<br />
z<br />
,<br />
2π<br />
Fz<br />
sin( zi<br />
)<br />
z<br />
,<br />
Similarly to cut-up milling, it can be observed that in case of cut-down face<br />
milling with cu t' = D/2, and t'' < D/2, the same relations for the cutting force<br />
components FZ, FX , FY are obtained, distinguishing only by the number of teeth<br />
that simultaneously cut.<br />
3. Conclusions<br />
1. The theoretical models presented in the paper are concentrating the main<br />
elements that influence the level of cutting force at face milling. These specific<br />
elements refer to the tooth geometry (γ, K, λ), standard compression yield point<br />
of the material being cut, the cross-sectional area of the theoretical chip, the<br />
milling width, the contraction chips coefficient, the constructive geometry of the<br />
tooth (γ, K, λ), the number of teeth that simultaneously cut, the cutter diameter<br />
and teeth number.
82 Ana-Maria Matei and Marius-Nicolae Milea<br />
2. Using those models to evaluate the cutting force provides extra accuracy<br />
because we are considering the forces on a tooth, the number of teeth that<br />
simultaneously cut, and also the relative position of the tool and material being<br />
cut for all possible variants of the face milling process.<br />
Received: January 29, 2010 “Gheorghe Asachi” Technical University of Iaşi<br />
Department of Machine Tools<br />
e-mail: anca_reea@yahoo.com<br />
R E F E R E N C E S<br />
1. C o z m î n c ă M., C o n s t a n t i n e s c u C, Bazele aşchierii, Ed. Gh. Asachi,<br />
Iaşi, 1995.<br />
2. C o z m î n c ă M. et al., About the cutting forces at face milling, Buletinul<br />
Institutului Politehnic din Iaşi, 2, Construcţii de Maşini, (2009).<br />
3. C o z m î n c ă M. et al., A new model for estimating the force components Fz, Fx and<br />
Fy when cutting metals with single tooth tools, Buletinul Institutului Politehnic<br />
din Iaşi, 1, Construcţii de Maşini, (2009).<br />
4. C o z m î n c ă M. et al., Variation of the ratios between forces Fz, Fx and Fy and the<br />
plastic strain force in single tooth cutting , Buletinul Institutului Politehnic din<br />
Iaşi, 1, Construcţii de Maşini, (2009).<br />
FORȚELE DE AȘCHIERE LA FREZAREA FRONTALĂ ÎN FUNCȚIE DE<br />
FORȚELE DEZVOLTATE LA NIVELUL UNUI <strong>DIN</strong>TE<br />
(Rezumat)<br />
În lucrare sunt prezentate modele teoretice de evaluare a componentelor forţei<br />
de aşchiere la frezarea frontală în funcţie de forţele dezvoltate la nivelul unui dinte.<br />
Forţele la frezarea frontală depind atât de varianta de frezare, poziţia relativă a dinţilor<br />
aşchietori faţă de semifabricat. poziţia dintelui aşchietor în raport cu sistemul de<br />
coordonate XYZ al sculei cît şi numărul de dinţi care aşchiază simultan.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
COMPARATIVE ASPECTS REGAR<strong>DIN</strong>G THE NESTING<br />
FOR BLANKING-PUNCHING OPERATION ON<br />
CLASSICAL PRESSES AND NUMERICAL<br />
COMANDED PRESSING CENTERS<br />
BY<br />
MARIUS - IONUŢ RÎPANU 1 GHEORGHE NAGÎŢ 1 ,<br />
IOLANDA - ELENA MANOLE 1 and ANDREI WEINGOLD 2<br />
Abstract. The nesting operation represents an important chapter in the design<br />
of stamping technology of parts or blanks in the area of cold plastic forming<br />
processes, with a large applicability in industry. The type of nesting has a very<br />
important role in determining the material consumption, the number of blanks cut<br />
concomitant, in determining the type and the construction of the stamping device<br />
or of the punch. The goal of this paper is to elaborate a comparative study<br />
between nesting for the classical presses stamping and for the CNC pressing<br />
centres. In order to obtain this goal, in this paper a case study, along with the<br />
results obtained and the associated conclusions are presented.<br />
Key words: punching, blanking, stamping press, CNC, nesting.<br />
1. Introduction<br />
The nesting of the blanks is a problem of special importance in the area of<br />
cold plastic forming processes. By nesting the blanks we understand the optimal<br />
cutting of metal sheets into individual blanks or strips. We understand as well<br />
the judiciously positioning or placing on metal sheets or strips the plan<br />
components as structured technological forms used to detach the pieces in order<br />
to get a minimum amount of waste scrap and an optimum coefficient of material<br />
utilization [1].
84 Marius Ionuţ Rîpanu et al.<br />
The objective of this paper is to elaborate a comparative study upon the<br />
nesting for the classical presses and nesting for the numerical commanded<br />
centres in order to emphasize the advantages and the disadvantages of both<br />
operations and, not the last, based on the effectuated researches, to obtain high<br />
quality and high precision parts at the smallest cost possible.<br />
2. Particularities of the Nesting Operation<br />
The cold plastic forming researches and especially the nesting for numerical<br />
controlled pressing centres researches made by J a c k s o n and M i t t a l [2] are<br />
based on blanking-punching using a very well planned algorithm, method that<br />
leads to the automatical generation of the CNC program. R a g g e n b a s s and R<br />
e i s s n e r [2] have researched the connection between pressing and laser, with<br />
applicability on the processing centres with CNC. In Twente University was<br />
studied and developed the processing of the metal sheets domain through<br />
planning and managing the factors that influence the machinability of metal<br />
sheets using NC [2], [3]. Some of the most important CAD/CAM programs that<br />
are used to perform this technological process are: RADAN, WICAM, Nesting<br />
Software, EditCNC, CNCezPRO TM , and some of the pressing centre with NC<br />
are TRUMPF, AMADA ARIES, Finn-Power, Mazak, Bystronic, etc.<br />
The nesting of the blanks for classical presses takes into account certain<br />
coefficients that depend on the nature of the material, the placing of the part<br />
and the nesting style of it (with waste, without waste, with little waste).<br />
During the nesting of the parts for the NC pressing centres it must be taken<br />
into consideration the type of the material of the metal sheet. The software used<br />
for modelling, simulating and generation of the numerical control scheme will<br />
consider all the factors and parameters that may influence the process<br />
productivity.<br />
3. Comparative Study<br />
For this study was chosen a pinchbeck part made of CuZn15, SR EN<br />
1653:2003, having the configuration from Fig. 1 and the thickness of 1 mm.<br />
The processing of this part was followed both for the classical presses and the<br />
CNC pressing centres. For the classical nesting, three variants of placing were<br />
considered, as in Figure 2. For those variants were calculated the coefficients<br />
with values of 53.77%, 54.16 % and 30.75%.<br />
In case of using the NC cutting centres, CAD/CAM software were used in<br />
order to realise and integrate the designing, modelling, simulating, planning and<br />
development functions of the CNC program for processing the metal sheet,<br />
choosing the optimum cutting variant with minimum waste.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 85<br />
Fig. 1 – The processed part.<br />
a) b)<br />
c)<br />
Fig. 2 – Nesting variants.<br />
In the present case, for cutting the 6.6 m 2 of the metal sheet, it was chosen<br />
the RADAN software of designing, modeling and simulating with the AMADA<br />
ARIES 245 pressing center. Following there are presented some sequences of<br />
the program that calculated and cut the part from Fig. 1.
86 Marius Ionuţ Rîpanu et al.<br />
Fig. 3 – The choice of the pressing centre for the processing.<br />
.<br />
Fig.4 – Standard sizes of metal sheets with their coefficients<br />
of utilization and the number of cut parts per sheet.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 87<br />
Fig.5 – Position of the parts on 850 x 300 mm sheet metal.<br />
Fig.6 – The cutting path of the parts.
88 Marius Ionuţ Rîpanu et al.<br />
There are presented a part of the steps of the processing program. The<br />
generated program processes a number of 98 pieces from a 3000 x 1500 mm<br />
sheet metal. At the end, as it is noticeable, the program shows exactly the<br />
number of program blocks, the number of tool changes, the programmed stops<br />
and the run time.<br />
Line43: B2<br />
Line44: N43X – 2.5Y119.5T12<br />
Line45: N44Y40.5<br />
Line46: N45X202.5<br />
Line47: N46Y119.5<br />
Line48: N47X407.5<br />
Line49: N48Y40.5<br />
Line50: N49X612.5<br />
Line51: N50Y119.5<br />
Line52: N51X817.5<br />
Line53 N52Y40.5<br />
Line54: N53G98X17.5Y59.98I205.J0.P3K0<br />
Line56: N55G98X201.5Y162.52T1 called from Line 61<br />
Line57: N56G72X – 2.Y40.02 called from Line 61<br />
Line58: N57G66I204.J0.P – 85.Q – 5. called from Line 61<br />
Line59: N58M00 called from Line 61<br />
Line64: N63G98X17.5Y100.I0.J0.P0.K0<br />
Line65: N64G50<br />
Amada Aries 245(RA1)<br />
Verification Analysis<br />
Total number of program blocks 65<br />
Total number of program characters 25 544<br />
Number of sub programs 0<br />
Number of tool changes 3<br />
Number of programmed stops 4<br />
Number of repositions 0<br />
Run time 1813 secs = 30.21 min.<br />
In the figures below are presented the steps that must be covered in order to<br />
cut the part. Therefore, in Fig. 3, there are presented the processing centres that<br />
use the RADAN software. From these ones, it was chosen for the part to be cut<br />
using the AMADA ARIES 245, using as well the specified material with 1 mm<br />
thickness.<br />
In figure 4, based on the part geometry and dimensions, are presented four<br />
types of standard metal sheets that may be processed by the machine. There are<br />
also presented the usage coefficients of the material and the exact number of
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 89<br />
parts that may result from the process and the number of metal sheets that are<br />
needed to cover the approximately 6.6 m 2 .<br />
It is noticeable that the coefficient of utilization of the material varies based<br />
on the dimensions of the sheet and the number of parts increases proportionally<br />
with the sheet dimensions. In Fig. 5 is presented the placing of the part on the<br />
metal sheet, the anchorage of the metal sheet on the pressing centre and the<br />
distances between the pieces that are about to be cut. In the last figure, Fig. 6 is<br />
presented the technological flow of cutting and the modality in which the tool<br />
approaches and clips step by step the part.<br />
4. Conclusions<br />
1. The CNC pressing centres are much more efficient than the classical<br />
pressing methods, the evidence of this fact being the calculation made for the<br />
two types of processing of blanks and the coefficient of utilization of the<br />
material.<br />
2. For the CNC pressing centres the coefficient of utilization varies<br />
depending on the dimensions of the metal sheets while, for the classical<br />
pressing, if it would have been used a longer metal strip the usage coefficient<br />
would have been the same. I this case it is more appropriate to use a CNC<br />
pressing centre.<br />
3. The number of parts obtained from CNC pressing centres is much<br />
higher than the number of parts obtained with the classical presses.<br />
4. The CNC pressing centres are characterised by the easiness of<br />
calculation at changing the setting of a parameter that has direct influence over<br />
the processing of the blanks, due to the presence of the CAD/CAM system<br />
presented.<br />
5. An important advantage of the CNC pressing centres, unlike the<br />
classical presses, at the processing of simple configuration parts (like the one<br />
presented above) is the geometry and the fabrication costs of the tools used in<br />
production.<br />
Received: March, 22, 2010 1 “Gheorghe Asachi” Technical University,<br />
Department of Machine Manufacturing Technology,<br />
e-mail: ripanumariusionut@yahoo.com<br />
2 S.C. BMTech S.R.L of Vladeni,<br />
Iasi – Romania,<br />
e-mail: andrei.weingold@bmtech.ro
90 Marius Ionuţ Rîpanu et al.<br />
R E F E R E N C E S<br />
1. B r a h a V., N a g i t G h., N e g o e s c u F., Tehnologia Presarii la Rece, Editura<br />
Tehnica, Stiintifica si Didactica CERMI, Iasi, 2003, pp 138-163.<br />
2. P a n M., R a o Y., An integrated knowledge based system for sheet metal cutting–<br />
punching combination processing, Knowledge-Based Systems 22, 2009, 368–<br />
375.<br />
3. R a o Y., H u a n g G., P e i g e n L., S h a o X., D a o y u a n Y., An integrated<br />
manufacturing information system for mass sheet metal cutting, International<br />
Journal Advanced Manufacturing Technology 33, 2007, 436 – 448.<br />
4. E n d o J., O h b a S.,A n z a i T., Virtual manufacturing for sheet metal processing,<br />
Journal of Materials Processing Technology, 60 (1996, 191–196.<br />
5. C h a t u r v e d i S., A l l a d a V., Integrated Manufacturing System for Precision<br />
Press Tooling, International Journal Advanced Manufacturing Technology, 15,<br />
(1999) 356–365.<br />
ASPECTE COMPARATIVE PRIVIND CROIREA<br />
LA OPERAŢIA DE DECUPARE-PERFORARE PE PRESE CLASICE<br />
ŞI CENTRE DE PRESARE PREVĂZUTE CU COMANDĂ NUMERICĂ<br />
(Rezumat)<br />
Operaţia de croire reprezintă un capitol important în proiectarea tehnologiei de<br />
ştantare a pieselor sau semifabricatelor din domeniul deformării plastice la rece având o<br />
aplicabilitate foarte mare în industrie. Croirea joacă un rol important în determinarea<br />
consumului de material, numărului de semifabricate ştantate concomitent, în<br />
determinarea tipului şi construcţiei ştanţei sau matriţei. Scopul acestei lucrări este de a<br />
elabora un studiu comparativ între croirea pieselor pe presele clasice şi croirea acestora<br />
pe centrele de comandă prevăzute cu comandă numerică, de a scoate în evidenţa<br />
avantajele şi dezavantajele acestei operaţii şi nu în ultimul rând, pe baza cercetărilor<br />
făcute, să putem obţine piese de o calitate superiaoră şi precizie ridicată la un cost cat<br />
mai redus. In acest sens, in lucrare este prezentat un studiu de caz, impreuna cu<br />
rezultatele obtinute si concluziile aferente.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
THEORETICAL AND EXPERIMENTAL CONSIDERATIONS<br />
ON DETERMINING THE EFFECT OF DIVERGENCE<br />
BY<br />
IUSTINA ELENA ROTMAN, PETRU DUSA<br />
and RADU ADRIAN BACIU LUPASCU<br />
Abstract. This paper presents the experimental studies conducted on the basis<br />
of ultrasound examination of the parts with different geometries and structures.<br />
The main objective of this work is to eliminate disturbances in order to perform an<br />
examination and a correct interpretation in case of the A-scan representation. The<br />
experimental results and conclusions drawn follow the analysis of the factors<br />
involved in the process, which negatively influences the interpretation of signals<br />
from the investigation display device.<br />
Key words: transducer, device, back wall, divergence, discontinuity.<br />
1. Introduction<br />
The Ultrasonic Testing Method of materials is practiced by more than 50<br />
years [1], [3].Since the first examination in detecting defects using ultrasonic<br />
oscillations of different materials, it became a classic test method based on<br />
measurements. That method takes in consideration all those factors that have a<br />
certain impact [4]. It is expected that in our days the ultrasonic testing to give<br />
results with minimum tolerance, if that method is sustained by a large variety of<br />
instruments and by one advance technical execution [4].<br />
The structure transformations of the material underlayers by manufacturing<br />
process lead to changes for the workpiece general properties [2]. After the type<br />
of defects are detecting, that will be compared with contractual standard. The<br />
contractual standard would decide if the detected defect could be accepted, not<br />
accepted, or remediable [2]. In many cases some of defects could be accepted if<br />
their values respect standard limits. Once detected the defects types, they must
92 Iustina Elena Rotman et al<br />
be confronted to the standards or to the contractual technical documentations to<br />
establish if they are admissible, inadmissible or remediable [2]. In many cases<br />
some influences can be neglected without to exceeding the measure limits<br />
prescribed in the technical documentation [2], [4]. This presume the exactly<br />
knowledge of the influence factors and of the results interpretation mode.<br />
2. Method of Investigation<br />
2.1. The Ultrasonic Testing Method<br />
Performing the non-destructive control with ultrasound effectuation consist<br />
principally in transmitting ultrasound signals (waves), produced by a waves<br />
generator, through the examined pieces [2], [3]. The ultrasounds signals are<br />
reflected by any surface and by any defect from the interior of the piece [1]-[5]<br />
that was exposed to testing, see fig.1.<br />
Fig. 1 – The ultrasonic testing method [5] Fig. 2 – The geometry examination [2]<br />
The ultrasonic equipment used in this research is presented in the table<br />
below:<br />
USM<br />
35X<br />
Table 1<br />
The necessary instruments of control US<br />
Device Transducer Piece Coupled<br />
USLT<br />
2000<br />
Reference<br />
model: B2S<br />
Aluminum bare<br />
Ø 90 x 318 mm<br />
Axle OL47<br />
Ø 900 x 1500mm<br />
2.1.1. The analysis of the investigation. Often in industrial practice there<br />
are problems of interpreting the image obtained on the screen investigation<br />
device. This paper presents an experimental study over the determination of the<br />
divergence marginal effect and the elimination of disturbances in order to obtain<br />
a correct examination and interpretation. Its objective is to highlight the journey<br />
Oil
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 93<br />
of ultrasonic waves from transducer through the examined piece, the reflection<br />
modalities of the waves and the way displaying on the screen of the device.<br />
When the diameter of transducer is comparable to the width of the piece, due to<br />
the divergence and partial reflection of the beam from the lateral surfaces of the<br />
piece, along the back wall echo and additional signal is formed, more or less<br />
distinct, named divergence marginal echo [2]. The necessary condition for<br />
eliminated this effect is:<br />
(1)<br />
B−Dp ⎡ ⎛ λ ⎞⎤<br />
≥ tg ⎢arcsin ⎜1, 22 ⎟⎥,<br />
2l<br />
⎣ ⎝ Dp⎠⎦<br />
where: B – the diameter of the piece; Dp<br />
– the diameter of the transducer; l –<br />
the length of the piece; λ – the divergence beam of fascicle. Also the back wall<br />
echo can be confused with a fault echo. These confusions occur due to the piece<br />
geometry, to the transducer diameter which is comparable to the diameter of the<br />
piece and to the differences in length. To eliminate these interpretation errors of<br />
and disturbances we will present several different cases.<br />
3. Experimental Results<br />
The experimental studies were performed with the devices type<br />
Krautkramer (USM 35 X and USLT 2000) in order to check possible errors in<br />
measurement. The pieces subjected to ultrasonic testing are: an aluminum bar<br />
with dimensions Ø 90 x 318 mm and one axle in OL47, forged and heat treated.<br />
By conducting the experiments we achieved the following results.<br />
3.1. First case. In this case the back wall echo can be confused with an<br />
echo of a defect (See fig. 3). These echoes can be called divergent effects.<br />
Fig. 3 – Ultrasonic testing display. Fig. 4 – The divergent effect.
94 Iustina Elena Rotman et al<br />
This measurement error is due to amplification (61 dB) that distorts the back<br />
wall echo base and to the transducer which is positioned near the edge of the<br />
sample.<br />
3.2. Second case. It may be noted that in this case the effect of divergence<br />
and misinterpretation can be eliminated by positioning the transducer on the<br />
axial direction of the sample. In this manner, the repetition rate of the flaw<br />
(PRF) does not influence the accuracy of the signal and the interpretation is<br />
objective. Bottom echo is very clear. (See fig. 5).<br />
Fig. 5 – Ultrasonic testing display. Fig. 6 – The ultrasonic beam shape .<br />
3.3. Third case. Analyzing the measurements made on both samples we can<br />
conclude that the divergence effect can be eliminated by reducing the repetition<br />
rate of the flaw.<br />
a) b)<br />
Fig. 7 –Ultrasonic testing display:<br />
a) – The parasite echo; b) – The additional echo disappears<br />
As we can see in Fig. 7a), parasites echoes are present and at a PRF9 (See<br />
Fig.7b)) these additional effects disappear. In this kind of situations it is<br />
recommended to get more measurements at a minimum level of PRF.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 95<br />
These cases study presented show how to determine the marginal effect of<br />
divergence and to eliminate stray echoes in order to obtain effective results.<br />
4. Conclusions<br />
Behind the effectuated researches one can evidence the following<br />
conclusions:<br />
1. Considering the applicability of this method and its expansion<br />
possibilities, it is necessary to analyze the factors that influence ultrasonic<br />
nondestructive testing.<br />
2. The position of transducer directly influences the results of testing. It is<br />
preferable that measurements are effectuated in different points of the<br />
investigated surface.<br />
3. When the transducer is near the edge of the piece, the back wall echo<br />
becomes smaller and it can be interpreted as a defect. This leads to<br />
misinterpretation.<br />
4. To eliminate stray echoes that appear on the display device is<br />
necessary for the repetition rate of the flaw (PRF) is minimized<br />
Acknowledgements. This paper was realised with the support of BRAIN “Doctoral<br />
scholarships as an investment in intelligence” project, financed by the European Social<br />
Found and Romanian Government.<br />
Received:March 3, 2010 “Gheorghe Asachi”Technical University,<br />
Department of Machine Manufacturing Technology<br />
Iasi, Romania<br />
e-mail: yustinikrotman@yahoo.com<br />
R E F E R E N C E S<br />
1. B e r k e M., Nondestructive Material Testing with Ultrasonics, Introduction to the<br />
Basic Principles, The e-Journal of Nondestructive Testing & Ultrasonics, 5,<br />
2000.<br />
2. V o i c u I. S a f t a, Defectoscopie nedistructiva industriala, Ed. Sudura, Timisoara,<br />
2001.<br />
3. K r a u t k r a m e r J. & H., Ultrasonic testing of materials, Springer Verlag Berlin<br />
Heidelberg New York, 1983.<br />
4. P e t c u l e s c u P., Ultrasounds Fundamental. Aplications, Ed. Univ. Ovidius,<br />
Constanta, 2002.
96 Iustina Elena Rotman et al<br />
5. * * * Ultrasound and Ultrasonic Testing. NDT Resource Center. Available from:<br />
http://www.ndt-ed.org/EducationResources/HighSchool/Sound/ultrasound.htm,<br />
Accessed: 07/02/2010.<br />
CONSIDERATII TEORETICE SI EXPERIMENTALE PRIVIND<br />
DETERMINAREA EFECTULUI DE DIVERGENTA<br />
(Rezumat)<br />
Aceasta lucrare prezinta un studiu experimental ce urmareste evidentierea<br />
parcursului ultrasunetelor de la traductor catre piesa de examinat, modul de reflexie al<br />
acestora si de afisare pe ecranul instrumentului de investigat cu scopul de a obtine<br />
rezultate eficiente. Obiectivul principal al acestei lucrari este determinarea efectului<br />
marginal de divergenta si eliminarea factorilor perturbatori ce pot influenta negativ<br />
rezultatele testarii cu ultrasunete. Studiile de caz prezentate si concluziile finale<br />
indeplinesc obiectivele propuse si pot contribui la dezvoltarea bazei de cunostinte.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
COMPUTERIZED MEASUREMENT SYSTEM<br />
BY<br />
CĂTĂLIN UNGUREANU 1 , RADU IBĂNESCU 2<br />
and IRINA COZMÎNCĂ 1<br />
Abstract. All technological machining systems parts made real surfaces<br />
certain deviations from the theoretical and nominal areas. The role of quality<br />
control is to decide whether or not these deviations fall within the allowable<br />
accuracy classes that imposed parts. The paper presents a computerized system for<br />
determining and recording deviations in shape of the real parts<br />
Key words: measurements, inductive transducers, data acquisition.<br />
1. Introduction<br />
Accuracy is the degree of form areas of agreement between actual shape of<br />
the areas resulting from the processing and documentation of performance<br />
prescribed form. The causes of form deviations are multiple of which can be<br />
mentioned:<br />
• elastic deformation of pieces during processing;<br />
• tighten the fixing the wrong parts in devices;<br />
• failure of machine tools guidance;<br />
• wear or deformation of system machine-tool - cutting tools elements.<br />
The effects of shape irregularities are of the least desired, with negative<br />
implications for economic and technical performance of pieces processed:<br />
• edit the sliding friction between the parts of joints,<br />
• reduce static and dynamic tightness of parts and joints,<br />
• edit character fits,<br />
• they increase wear and may lead to locking parts in contact<br />
• decrease the durability of parts in contact.
98 Cătălin Ungureanu et al.<br />
Deviations in shape are defined as deviations of actual surface shape of the<br />
surface adjacent to, or deviation from the actual profile shape as the adjacent<br />
profile. Profile is adjacent to the same profile as the profile geometry, the outer<br />
tangential to the actual profile, and placed so that the distance between it and<br />
the actual profile to be minimum. Maximum permissible deviation form as is<br />
tolerance.<br />
Traditional means of measuring the deviations in shape are instrumentality<br />
mechanical comparators, scale value of 0.01 mm or 0.005 mm. They have the<br />
advantages of simplicity and low cost, but recording and graphic representation<br />
of measurement data for a production series is difficult. These disadvantages are<br />
removed if for control is used a measuring system equipped with a displacement<br />
transducer, data acquisition card and computer [1], [2].<br />
2. Experimentals<br />
Experimental facility is shown in Figure 1.<br />
Fig.1 – Experimental facility.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 99<br />
Transducer was chosen and the whole system was built base on<br />
recommendations from the literature [3] – [5].<br />
Part of the measure is placed on two supports adjustable so that it can be<br />
defined by two points equal height verification. This determines a straight line,<br />
to which will determine the actual right - edge play. For measurement: using<br />
an inductive transducer type Hottinger Baldwin Messtechnik WA T-10mm,<br />
supplied with 20 V DC voltage from a source Hameg HM 8040-2. Output is<br />
taken from a data acquisition card National Instruments USB 6009.<br />
Measurement chain is completed with a PC. No other mode of signal<br />
conditioning is required. Virtual instrument was developed programming<br />
environment LabVIEW 8.2 Student Edition.<br />
The block diagram of the application is presented in Fig. 2 [6] and involves<br />
the DAQ Assistant Express VI, the Waveform Chart for displaying the<br />
acquired signal and also the Build Table and the Write to Measurement File<br />
blocks for writing the acquired signal values in a table and in a file for further<br />
analysis.<br />
Fig. 2 – The block diagram of the application [6].<br />
The DAQ Assistant may be founded in the Express Functions palette of<br />
the block diagram and it is needed to create and configure a task that reads a<br />
voltage level in a specified range from a DAQ device. After the DAQ Assistant<br />
is placed on the block diagram, a dialog box window is launched in order to<br />
configure the data acquisition system. The dialog box displays a list of channels<br />
of the installed DAQ device and from this list the physical channel to which the<br />
instrument connects the signal must be selected. The others configurations are<br />
related to terminal configuration (differential or single ended), acquisition mode<br />
(continuous or not), signal input range and rate of acquisition. The Write to<br />
Measurement File Express VI is used for storing the information about the data
100 Cătălin Ungureanu et al.<br />
VI generates in a specific file (*.lvm) in the default LabVIEW Data folder<br />
installed by LabVIEW in the default file folder of the operating system.<br />
Experimental data were represented graphic with a Matlab application.<br />
Calibration curve was presented in [6]. Diagram thus obtained is shown in<br />
Fig.3.<br />
d [mm]<br />
0.02<br />
0.01<br />
0<br />
-0.01<br />
-0.02<br />
0 50 100 150<br />
l [mm]<br />
Fig. 3 – Experimental diagram.<br />
In these diagrams “l” represents the length of the measured piece in<br />
mm and “d” the form deviation also in mm.<br />
Trough numerous measurement data recorded it is possible that the<br />
chart should be influenced by measurement noise. To remove these<br />
noises, was made a filtration of data with a Matlab application using<br />
smooth function [7]. The default syntax of smooth function is<br />
yf = smooth (y,span),<br />
where y is a column vector containing the acquired data and yf is the<br />
result column vector after a moving average filter is applied. The smooth<br />
function implements a lowpass filter with filter coefficients equal to the<br />
reciprocal of the span. The span must be odd but the default span for the
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 101<br />
moving average is 5 and in this case the first few elements of yf are<br />
calculated as follows:<br />
yf(1) = y(1)<br />
yf(2) = (y(1) + y(2) + y(3))/3<br />
yf(3) = (y(1) + y(2) + y(3) + y(4) + y(5))/5<br />
yf(4) = (y(2) + y(3) + y(4) + y(5) + y(6))/5 and so on.<br />
The diagram of the whole acquired data filtered for a span equal with<br />
3 is shown in Fig. 4.<br />
d [mm]<br />
0.02<br />
0.01<br />
0<br />
-0.01<br />
-0.02<br />
0 50 100 150<br />
l [mm]<br />
Fig. 4 – The filtered values for a span equal with 3.<br />
It is noted that the chart is much more precise and accurate<br />
assessment can be made on the actual shape of the part to be measured.<br />
The filtration effect is observed more clearly if the graph is made for<br />
filtered and the unfiltered data only for first portion of the controlled<br />
piece. In Fig. 5 is represented by the dotted line unfiltered data and<br />
filtered data with continuous line for a span equal with 3 and in Fig.6 for<br />
a span equal with 5.
102 Cătălin Ungureanu et al.<br />
d [mm]<br />
d [mm]<br />
0.02<br />
0.01<br />
0<br />
-0.01<br />
-0.02<br />
0 10 20 30 40 50<br />
l [mm]<br />
0.02<br />
0.01<br />
-0.01<br />
Fig. 5 – The filtered and unfiltered values for a span equal with 3.<br />
0<br />
-0.02<br />
0 10 20 30 40 50<br />
l [mm]<br />
Fig. 6 – The filtered and unfiltered values for a span equal with 5.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 103<br />
4. Conclusions<br />
1. The computerized system for measuring deviations of form<br />
enables the rapid and effective control of the geometric quality of parts in<br />
machine building.<br />
2. It can measure the absolute value of deviation and may draw a<br />
diagram of the actual shape of the part to control.<br />
3. However, because of noise measurement is required filtering<br />
acquired data. You can use low-pass filter easily facilitated by MatLab<br />
software package. It remains to be studied in future what is the best level<br />
of filtering.<br />
Received: February 20, 2010<br />
R E F E R E N C E S<br />
1 Technical University “Gh. Asachi,<br />
Department of Machine Tools<br />
Iaşi, Romania,<br />
e-mail: cungurea@yahoo.com<br />
2 Department of Theoretical Mechanics<br />
e-mail: ribanesc@yahoo.com<br />
1. K e j í k P., K l u s e r C., B i s c h o f b e r g e r R., P o p o v i c R. S. A low-cost<br />
inductive proximity sensor for industrial applications, Sensors and Actuators A:<br />
Physical, 110, 1-3 (2004) pp. 93-97.<br />
2. M a r č i č M., A new inductive displacement transducer, Sensors and Actuators A:<br />
Physical, 70, 3, 30 (1998), pp. 223-237.<br />
3. P a r k J., M c k a y S., Practical data acquisition for instrumentation and control<br />
systems, Newnes, Elsevier 2003.<br />
4. R i p k a P., T i p e k A., Modern Sensors Handbook, ISTE Ltd, London, 2007<br />
5. S i n c l a i r I.R., Sensors and transducers, Newnes, Butterwork Heinemann,<br />
Oxford, 2001.<br />
6. U n g u r e a n u C., I b a n e s c u R., C o z m i n c a I., Virtual Instrument for<br />
Linear Position Measurement, Buletinul Institutului Politehnic din Iasi, publicat<br />
de <strong>Universitatea</strong> Tehnica “Gh. Asachi” Iasi, LV(LIX), 4, 2009, pag.54-62.<br />
7. www.mathworks.com<br />
,
104 Cătălin Ungureanu et al.<br />
SISTEM DE MĂSURARE COMPUTERIZAT<br />
(Rezumat)<br />
În lucrare este prezentat un sistem de măsurare a abaterilor de formă geometrică.<br />
Piesa de măsurat este amplasată pe două reazeme reglabile, astfel încât dreapta<br />
adiacentă să fie paralelă cu placa de control pe care se efecuează măsurarea. Pentru<br />
măsurare se utilizează un traductor inductiv, alimentat de la o sursă de curent continuu.<br />
Se utilizează o cartelă de achiziţie de date, un PC şi mediul de programare LabVIEW.<br />
Cu datele astfel obţinute se pot trasa diagrame ale formei reale a piesei. Datorită<br />
zgomotelor de măsurare, forma reală a piesei este mai greu de observat. Pentru a<br />
înlătura influenţa zgomotelor de măsurare se recomandă o filtrare de tip ”trece jos” a<br />
profilului obţinut. Aceasta se poate face cu ajutorul funcţiei smooth a programului<br />
Matlab, cu nivele diferite de filtrare.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
ADHESIVE BON<strong>DIN</strong>G OF SURFACE TREATED ALUMINIUM<br />
BY<br />
BIRGIT KJÆRSIDE STORM<br />
Abstract. Adhesive Bonding of aluminium and aluminium alloys can be carried<br />
out so a strong bonding is the result. The perfect bonding between two aluminium<br />
parts will be a bonding, where the material will break in the aluminium part, because<br />
in that case the bonding is stronger than the aluminium. For making a strong bonding<br />
a structural adhesive shall be used. A structural adhesive can be made of epoxy,<br />
polyurethane, acrylate, cyanoacrylate and a few other types. In some cases an<br />
anaerobe adhesive can cure up with aluminium. There is interaction between the<br />
adhesive, the aluminium alloy and the surface treatment. Aluminium creates in<br />
atmospheric air immediately an oxide layer on the surface. Anodized aluminium<br />
reacts with contaminations and CO2, which gives a surface with a low surface tension,<br />
though aluminium will have a high surface tension. The low surface tension of the<br />
reacted surface cause a bad wetting of the surface with the adhesive, and an<br />
insufficient wetting causes bad adhesion between the aluminium and the adhesive. For<br />
making a good bonding a good wetting is necessary. The sufficient wetting can be<br />
made on the aluminium by making a surface treatment. The surface treatment can be<br />
an anodizing. For obtaining the best adhesion on anodized surface the adhesive shall<br />
be carried out before the sealing immediately after the anodizing. The fresh anodized<br />
aluminium has a bigger surface and open pores, which cause a mechanical bonding in<br />
addition to the adhesion. The non sealed aluminium has reactive groups, which in<br />
some cases can react with reactive groups in the adhesive and can cause a chemical<br />
bonding as well. The non-sealed anodized aluminium can be treated with a primer,<br />
which can react with the active groups and can run into the pores and fill out the<br />
porousity. Before the primer has cured up totally the adhesive shall be added to the<br />
system for making a chemical reaction between the primer and the adhesive. Using<br />
such a system can give a higher strength in the bonding.<br />
Key words: aluminium, anodizing, adhesive, adhesive bonding, surface<br />
treatment, wetting<br />
1. Introduction<br />
Adhesive Bonding of aluminium and aluminium alloys can be carried out so<br />
a strong bonding is the result. The perfect bonding between two aluminium
106 Birgit Kjærside Storm<br />
parts will be a bonding, where the material will break in the aluminium part,<br />
because in that case the bonding is stronger than the aluminium.<br />
Anodized aluminium can - as aluminium with other surface treatments - be<br />
bonded by an adhesive bonding. For obtaining a strong bonding, which means a<br />
bonding which has a least the same strength as the aluminium alloy, there need<br />
to be a good adhesion between the anodized aluminium surface and the<br />
adhesive. For adhesive bonding there are various theories. No single theory<br />
explains adhesion in general and in reality it is probably a combination between<br />
the different theoretical explanations.<br />
There are six theories of adhesion. They are physical adsorption theory,<br />
chemical bonding theory, diffusion theory, electrostatic theory, mechanical<br />
interlocking theory and weak boundary layer theory.<br />
For anodized aluminium the physical adsorption theory and mechanical<br />
interlocking are the main methods for making the bonding, but the chemical<br />
bonding theory can occur if the surface is treated in a special way.<br />
The physical adsorption need to be present, if a bonding shall be strong<br />
enough. For creating the physical adsorption it is necessary to have a sufficient<br />
wetting of the surface.<br />
The physical adsorption gives the adsorption forces in the interphase<br />
between the metal or the anodized layer and the adhesive. The larger interphase<br />
between the two interfaces the higher strength will be obtained in the adhesive<br />
bonding.<br />
The mechanical interlocking will always be possible and in many cases, it is<br />
the only connection between the metal/anodized layer and the adhesive. This is<br />
the case, if the adhesive is not able to wet the surface.<br />
The mechanical interlocking is good in many ways, because if the surface is<br />
larger the area over which the physical adsorption can occur is larger. If the<br />
surface is rough the forces created in the interphase will give forces in different<br />
directions, which also will give a better physical adhesion.<br />
The chemical bonding in the interphase is the absolute most attractive force<br />
to obtain, if it can be possible. In some cases it is possible to choose an<br />
adhesive, which can react chemical with the surface and especially this can be<br />
possible in a fresh made anodized aluminium surface, where reactive groups<br />
still are present.<br />
2. Wetting and HSP<br />
The surface shall be able to be wetted by the adhesive, if a strong bonding<br />
shall be created. If it is possible to obtain a good wetting, the possibility to<br />
increase the physical adsorption in the interphase is present. For obtaining a<br />
good wetting the metal/anodized metal surface must have a surface tension<br />
which is higher than the surface tension of the adhesive. In figure 1 the wetting
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 107<br />
of a surface and the determination of the surface tension by the droplet method<br />
is sketched. The surface tension of as well the metal/anodized metal and the<br />
adhesives depends on the secondary bonding forces in the material. For<br />
aluminium and anodized aluminium the surface will react with CO2. This<br />
reaction causes a low surface tension of the surface, which gives a bad wetting<br />
with the adhesive.<br />
The surface tension can be measured or it can be calculated from knowledge<br />
about the solubility parameters. The solubility parameters can be measured<br />
indirectly or they can be calculated.<br />
Fig. 1 – Surface tension and determination of surface tension [5].<br />
The solubility parameters can be a one-dimensional parameter or a three<br />
dimensional parameter. The solubility parameter is defined as<br />
(1) δtot 2 = ∆Evap/Vl,<br />
δtot is the total is the total solubility, ∆Evap the heat evaporation and Vl is the<br />
molar volume.<br />
The one dimensional solubility parameter gives not the total answer for the<br />
secondary bonding forces. Therefore Charles Hansen defined the Hansen<br />
Solubility Parameters, HSP, which are three dimensional solubility parameters.<br />
(2) δtot 2 = δd 2 + δp 2 + δh 2 ,<br />
δd is the contribution from the dispersion forces, δp is the contribution from the<br />
polar forces and δh is the contribution from the hydrogen forces.<br />
The three contributions will be put into a three dimensional coordinate<br />
system, where the dispersion forces will be printed with the double value of the<br />
two others. Each material, chemical or solvent will besides a three dimensional<br />
solubility parameter, which will be the centre in a sphere, also have an action<br />
radius, R0. The action radius is the area around the centre where the<br />
material/chemical is able to be dissolve or in other way to interact with other<br />
materials. The HSP can be set up in a three dimensional diagram and the<br />
wetting possibility can in that way directly be observed by studying if there are<br />
overlap by the spheres. In Fig. 2, two spheres are set up besides each other.
108 Birgit Kjærside Storm<br />
They are not able to cover each other and the materials behind them are not able<br />
to wet each other.<br />
Fig. 2 – Two spheres in a HSP diagram.<br />
The two spheres are not touching each other. The two materials behind are not compatible.<br />
If wetting of a material with another material shall be possible, the solubility<br />
spheres for the two materials shall at least touch each other. To calculate if<br />
wetting of a material with another material for example an adhesive is possible<br />
the RED – relative energy density – value can be calculated. The RED value is<br />
defined as:<br />
(3) RED = Ra/R0,<br />
Ra is the distance between the centre of the two spheres and R0 is the centre of<br />
the material, which shall be compared with the other sphere. If the RED number<br />
is 1, there is no compatibility/solubility between the two<br />
compared materials.<br />
As mentioned above the surface tension, γ, can be measured or be calculated<br />
from the HSP. Different researchers have found the correlation between the<br />
surface tension and the HSP. The most used is<br />
(4) γ = 0,0688V 1/3 [δD 2 + k (δP 2 + δH 2 )]<br />
γ is the surface tension and k is constant depending on the liquids involved.<br />
Other similar correlations are used.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 109<br />
3. The surface of Anodized Aluminium<br />
Aluminium alloys and anodized aluminium will theoretically have a high<br />
surface tension, so wetting of the surface with an adhesive would not be a<br />
problem, because the adhesives will have a surface tension much lower.<br />
Aluminium and anodized aluminium will react with CO2 and possible<br />
contaminations in the air and will there fore in practical have a surface tension,<br />
which is so low that wetting with the adhesives is not possible, because the<br />
actually surface tension is lower for aluminium surface than for the adhesive.<br />
If the surface is new created the surface tension will still be high and wetting<br />
will not be a problem. For solving the problem with not anodized aluminium the<br />
aluminium surface is often chromatized or phosphatized to obtain a good wetting.<br />
The surface of anodized aluminium itself is possible to wet with an adhesive<br />
especially if it is new treated and that will say that as soon after the anodizing the<br />
wetting will take place the better adhesion will be obtained. The bigger surface<br />
the better wetting is possible and that means also that if the anodized surface is<br />
used for bonding before sealing. The fresh anodized aluminium surface is the best<br />
surface for adhesive bonding, because the surface there has reactive groups,<br />
which will give possibility for chemical bonding in the interphase. The reactive<br />
groups give a bigger contribution to the surface tension because of more hydrogen<br />
bonds and the physical adsorption in the interphase will increase. The mechanical<br />
interlocking will also increase because of the porousity in the anodized surface.<br />
After sealing the surface is like un-anodized aluminium and the advantages<br />
obtained by the anodizing will disappear for the bonding.<br />
4. Surface Treatments<br />
Before bonding the surface need to be clean. The surface shall as mentioned<br />
above be with so many reactive groups as possible and the surface tension shall<br />
be as big as possible. The possibility for mechanical interlocking shall be<br />
present if possible.<br />
If the surface is contaminated with oil or grease from production or cutting<br />
process the surface tension will be the same as the surface tension of oil and<br />
grease, which will say very low, because those materials will almost have a<br />
contribution form the dispersion forces, very few if any from the polar forces<br />
and no from the hydrogen forces. Such a surface is impossible to bond on. The<br />
grease and oil shall be removed with an organic solvent as f.ex. isopropyl<br />
alcohol or CO2. The surface can if it is non anodized aluminium or is a sealed<br />
anodized aluminium be treated with a Scotch Brite for obtaining a bigger<br />
surface for mechanical interlocking. The treatment with Scotch Brite shall take<br />
place before the cleaning for removing dust from the surface.
110 Birgit Kjærside Storm<br />
A fresh anodized surface does not need any treatment if it has not been<br />
touch before bonding.<br />
5. Adhesive Bonding<br />
An adhesive bonding shall be constructed so the forces on the bonding is the<br />
best possible for the bonding. An adhesive bonding like a shear stress and<br />
dislike a direct tension. In Fig. 3 the two types of forces are sketched.<br />
Fig. 3 – The figure shows to the left, the shear force and to the right, the tension [3].<br />
The adhesive bonding dislikes a peeling, but in practice it is difficult to<br />
create a bonding, where peeling is not possible. The peeling is the weakest point<br />
for an adhesive bonding and is often the limitation for making the bonding. If it<br />
is possible the bonding shall be constructed so the forces are shear forces and<br />
peeling cannot occur.<br />
6. Structural Adhesives<br />
Adhesives used for aluminium will be a structural adhesive. With a<br />
structural adhesive means an adhesive with a high strength. The perfect bonding<br />
will be a bonding where the bonding will have a strength which is the same as<br />
the aluminium. The bonding shall not break in interphase between the<br />
aluminium and the adhesive but in the aluminium or in the adhesive layer. If the<br />
break will be in the interphase, the wetting has normally not been good enough,<br />
or the cleaning of the surface has not been sufficient, or the forces in the surface<br />
of either aluminium or the adhesive have not been sufficient.<br />
Structural adhesives can be many different polymers. All of them are<br />
polymers which will cure up under cross linking. All of them will have an<br />
opening time. The opening time – the time from the adhesive has been added to<br />
the product until it will start the cross linking - depends on the curing system<br />
and can be regulated to be short or long depending on the application of the<br />
product. There are of course limitations for how short and how long the opening<br />
time for the adhesive can be. The structural adhesives have no green strength,<br />
which means that for several applications it is necessary to use a tool to fix the<br />
products until the adhesive has obtained the strength. The structural adhesives<br />
are epoxy adhesives, polyurethane adhesives, acrylic adhesives or other<br />
adhesives which can cross link. In Fig. 4 the curing reaction of an epoxy
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 111<br />
adhesive is shown. There are many different types of curing reactions for epoxy<br />
and the shown is only an example. In Fig. 5 the reaction of a polyurethane<br />
adhesive is shown.<br />
Fig. 4 – Curing of epoxy with a primary amine [4].<br />
Fig. 5 – Formation of biuret from diisocyanat and urea [2].<br />
A structural adhesive can also be a cyanoacrylate, but for working with a<br />
quick curing adhesive can give problems in a production. On the market there<br />
are many different adhesives of all the mentioned types. They are different in<br />
opening time and in strength and flexibility. They can have different colours as<br />
well, and often the colours are used as an indicator for seeing if the adhesive is<br />
placed all over where it shall be.<br />
The properties of the adhesive depends on the polymer and the cross linking<br />
system in the adhesive. Aromatic groups in the adhesive give higher strength<br />
and aliphatic groups give higher flexibility. The more aromatic groups and the<br />
higher density of the aromatic groups the higher strength will be obtained. The<br />
more aliphatic groups there are in the polymer, the better flexibility there will be<br />
possible to obtain. The density of the reactive groups in the adhesive will also<br />
give possibility for strength and flexibility. The more reactive groups for cross<br />
linking there are, the higher rate of cross linking can be obtained and the higher
112 Birgit Kjærside Storm<br />
rate of crosslinking the higher strength will be obtained. Opposite will a low<br />
density of reactive groups for crosslinking gives a lower strength and a higher<br />
flexibility of the adhesive. Bonding with a high strength and stiffness is<br />
necessary under a shear stress and especially if a peeling is a risk. If the load<br />
will be a dynamical load the bonding need to have some flexibility as well as<br />
strength and the adhesive need certain flexibility.<br />
The epoxy adhesives can general obtain the highest strength, because I most<br />
epoxy adhesives the density of aromatic groups is high. Most of the epoxy<br />
resins are made on basis of bis-phenol A, which contains two aromatic groups.<br />
In Fig. 6 the formation of the epoxy resion bisphenol-A-diglycidylether<br />
(DGEBA) is shown. Aliphatic groups can be added and they will change the<br />
strength and the flexibility of the cured adhesive. Adding aliphatic groups can<br />
also be used for changing the opening time of the adhesive, because a longer<br />
aliphatic chain will decrease the density of reactive epoxide groups and there<br />
fore change the reaction time. Epoxy adhesives are normal two component<br />
systems. With mixing the two reactants the curing process will start. Increasing<br />
temperature will decrease the reaction time. With changing in reaction<br />
temperature the curing process – the cross linking process – can change and the<br />
properties of the adhesive can change.<br />
Fig. 6 – Formation of DGEBA from bisphenol A and epichlorhydrin [4].<br />
The polyurethane adhesives can as well obtain a good strength. Most of the<br />
polyurethane adhesives can obtain a good flexibility, especially if they are made<br />
on basis of an aliphatic isocyanate. The polyurethane adhesives can be one- or<br />
two component systems. Two component systems will start the reaction, when<br />
the isocyanate and the polyol – the two active components – are mixed with<br />
each other. Again increased temperature will decrease the reaction time and<br />
increased temperature can change the properties of the adhesives. One<br />
component polyurethane adhesives will start the reaction under influence of<br />
humidity in the atmosphere. The humidity will react with one of the<br />
components in the adhesive and make it active for reacting with the other<br />
component. If the humidity is low the reaction will not take place as expected.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 113<br />
The acrylic adhesives cure up with a radical reaction over a double bond. For<br />
initiating the reaction an initiator system shall be reacted. This can happen under<br />
influence of adding the initiator or under adding energy f.ex. with UV irradiation<br />
for starting the reaction. The acrylics can be one or two component adhesives.<br />
The rate of cross linking depends on the curing system. The rate of the cross<br />
linking depends also on the temperature and on the time in which the system will<br />
react.<br />
Fig. 7 – The HSP plot of DP460 from 3M. DP460 is an epoxy based adhesive.<br />
Fig. 8 – The HSP plot of Sikaflex-360HC. Sikaflex-360HC is a polyurethane based adhesive.
114 Birgit Kjærside Storm<br />
The HSP can be measured for the different adhesives. In Fig. 7 the HSP plot<br />
of the epoxy based adhesive DP 460 from 3M is shown and in Fig. 8 the HSP plot<br />
of the polyurethane adhesive Sikaflex from Sika Industries is shown.<br />
From comparing the HSP for an adhesive with the HSP for a surface –<br />
aluminium or anodized aluminium – it can be foreseen if it can be possible to<br />
wet the surface with the adhesive and if it can be possible to make a reaction<br />
between the adhesive and the surface. Fig. 9 shows the HSP plot for an unsealed<br />
anodized aluminium surface.<br />
Fig. 9 - The HSP plot of an unsealed anodized aluminium.<br />
The anodizing has been made in the laboratory<br />
For an unsealed anodized surface it can with some polyurethane adhesives<br />
and some epoxy adhesives be possible to make a chemical bonding in the<br />
interphase. Such an adhesive can be used as a primer if it can be possible to find<br />
another adhesive, which can give the strength of the bonding and can wet the<br />
primer.<br />
7. Results<br />
For several systems combined with anodized aluminium and a structural<br />
adhesive the strength and the flexibility has been measured. The adhesives used<br />
in the experiments have been commercial adhesives and the anodizing process<br />
has been carried out either in a laboratory or in a commercial system. For the<br />
laboratory made anodizing the process has been made with parameters and<br />
chemicals as near the commercial systems as possible.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 115<br />
The strength of the bonding depends of the adhesive and of the surfaces. In<br />
a group of experiments where the aluminium alloy 6063 with a tensile strength<br />
of 100 MPa was used the strength of the bonding on plate of 20 mm and the<br />
bonding area of 20 x 20 mm 2 have for anodized unsealed aluminium been<br />
measured to app. 7400 N with DP460 and to 3700 N for the flexible Sikaflex -<br />
360HC. The strength of bonding with DP460 (epoxy resin) can be up to 10000<br />
N on the area 20x20 mm 2 .<br />
8. Conclusions<br />
It is possible to bond aluminium with an adhesive and obtain a good<br />
strength and flexibility. The adhesive bonding is best for a shear stress and<br />
sensitive for especially a peeling. The adhesive used will always be a structural<br />
adhesive. Epoxy adhesives, polyurethane adhesives and acrylics adhesives are<br />
the most common used adhesives for aluminium.<br />
For obtaining a good bonding the surface of the aluminium needs to be<br />
wetted by the adhesive. If a surface shall be wetted by an adhesive the surface<br />
tension of the surface need to be higher than the surface tension of the adhesive.<br />
For estimating the possibility of wetting of the surface with an adhesive the<br />
HSP can be used as a tool.<br />
Adhesive bonding give a better strength for anodized aluminium if the<br />
bonding is made on unsealed anodized aluminium. The unsealed anodized<br />
aluminium is porous and will give a better mechanical interlocking together<br />
with a physical adsorption in the interphase between the anodized layer and the<br />
adhesive. Use of an adhesive as a primer can increase the strength in the<br />
interphase. If it is possible a primer or an adhesive, which can give a chemical<br />
reaction between the unsealed anodized aluminium, shall be used for obtaining<br />
the highest strength of the bonding.<br />
Received:January, 30,l 2010 Aalborg University,<br />
Esbjerg Institute of Technology<br />
Esbjerg, Denmark<br />
e-mail: bks@bio.aau.dk<br />
R E F E R E N C E S<br />
1. H a n s e n Ch. M., Handbook of Solubility Parameters – A User´s Handbook, CRC<br />
Press, 1 st edition, 2000.<br />
2. O e r t e l G., Polyurethane Handbook, Hanser Publishers, 2 nd edition, 1994.<br />
3. P e t r i e E. D., Handbook of adhesives and sealants, McGraw-Hill, 2000.
116 Birgit Kjærside Storm<br />
4. P i z z i A. et al., Handbook of Adhesive technology, Marcel Dekker, Inc, 2 nd edition,<br />
2003.<br />
5. www.specialchem.adhesives.com<br />
LIPIREA CU ADEZIVI A ALUMINIULUI TRATAT SUPERFICIAL<br />
(Rezumat)<br />
Lipirea cu adezivi a aluminiului şi aliajelor de aluminium poate fi realizată astfel încât<br />
să se obţină îmbinări rezistente. O lipire perfectă a două piese din aluminiu va fi acea<br />
lipire la care ruperea îmbinării are loc în piesa de aluminiu, lipitura fiind mai rezistentă ca<br />
aluminiul. Pentru a obţine o lipitură de mare rezistenţă, se vor folosi adezivi structurali,<br />
care pot fi de tip epoxi, poliuretan, acrilat, cianoacrilat etc. În unele cazuri, un adeziv<br />
anaerob poate interacţiona cu aliajul de aluminiu şi cu tratamentul de suprafaţă.<br />
Aluminiul, în aer atmosferic, creează imediat un strat de oxid la suprafaţă. Aluminiul<br />
anodizat reacţionează cu contaminatorii şi cu CO2 rezultând o suprafaţă cu tensiuni<br />
superficiale reduse, deşi aluminiul are tensiuni superficiale ridicate. Asemenea tensiuni<br />
superficiale reduse vor determina o umectare slabă a suprafeţei cu adeziv, o umectare<br />
insuficientă determinând o adeziune necorespunzătoare între aluminiu şi adeziv. Pentru o<br />
lipire de calitate este necesară o bună umectare. O umectare suficientă poate fi obţinută<br />
prin aplicare de tratamente superficiale. Un asemenea tratament poate fi unul de<br />
anodizare. Pentru a obţine adeziune optimă pe suprafaţa anodizată, adezivul trebuie aplicat<br />
imediat după anodizare. Aluminiul proaspăt anodizat are suprafaţă de contact mai mare şi<br />
pori deschişi, rezultând o lipire mecanică suplimentară adeziunii. Aluminiul poate avea<br />
grupuri reactive care uneori pot reacţiona cu grupuri reactive din adeziv, rezultând şi o<br />
lipire chimică. Aluminiul anodizat poate fi tratat cu o amorsă, care poate reacţiona cu<br />
grupurile active şi poate intra în pori şi umple porozităţile. În asemenea cazuri, adezivul se<br />
aplică înainte de epuizarea amorsei, pentru a avea loc reacţie chimică între amorsă şi<br />
adeziv. Asemenea soluţii pot asigura rezistenţă sporită îmbinărilor cu adezivi.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
INDUSTRIAL APPLICATIONS OF<br />
THE PNEUMATIC MUSCLES<br />
BY<br />
IOANA PETRE 1 , DAN PETRE 2 , CRISTINA FILIP 1 ,<br />
and LAVINIA NEAGOE 1<br />
Abstract. Pneumatic muscles were, lately, used in many applications<br />
especially in the field of industrial robots, because of the evolution of the<br />
technology. The characteristics of the pneumatic muscles make them being easy<br />
to use and with great performances. Pneumatic muscles have applications in<br />
robotics, biorobotics, biomechanics, artificial limb replacement and industry. In<br />
this paper I will present some industrial applications of the pneumatic muscles.<br />
The utilization of the pneumatic muscle is soon to become a good alternative for<br />
the present day electric or mechanically drives because of their characteristics and<br />
because of the evolution of the industrial development.<br />
Key words: pneumatic muscle, actuator, industrial application.<br />
1. Introduction<br />
Although the pneumatic muscles have been conceived since 1930 by S.<br />
Garasiev, a Russian inventor, only for a few years, they have been used in<br />
different applications. Lately, pneumatic muscles became a better choice than<br />
present day electric or other drives. They are usually used in factory floor<br />
automation and nowadays, in industrial robotics as a main motion power source.<br />
The current paper will focus on the pneumatic muscle with it’s applicability<br />
in the industry. These actuators are usually cylindrical in shape and they are<br />
composed by an interior inflatable tube typically made by neoprene rubber<br />
wrapped in a multilayer tissue typically made by nylon. The tube, under the<br />
action of compressed air, increases its diameter and decreases its length; the<br />
stroke resulted is in direct relation with the pressure of the compressed air<br />
passed into the muscle.
118 Ioana Petre et al.<br />
The utilization of the pneumatics offers many advantages, the most important<br />
being the low weight and the inherent compliant behavior of its actuators.<br />
Compliance ensures a soft touch and safe interaction. In contrast with<br />
pneumatic muscle actuator, hydraulic and electric drives have a very rigid<br />
behavior and can only be made to act in a compliant manner through the use of<br />
relatively complex feedback control strategies. [1]<br />
2. Pneumatic Muscle<br />
2.1. History<br />
Pneumatic muscle is an actuator system based on an inflatable and flexible<br />
membrane operated by pressurized air (Fig. 1).<br />
Fig. 1 – Pneumatic muscles and their function principle<br />
(source http://www.festo.com/hm2001/eng/1001.htm).<br />
Pneumatic muscles were first conceived in 1930 by G a r a s i e v, a Russian<br />
inventor [2]. According to B a l w in [3], J. L. McKibben introduced it to<br />
motorize pneumatic arm orthotics for helping control handicapped hands: due to<br />
the similarity in length-load curves between this artificial muscle and skeletal<br />
muscle, it seemed an ideal choice for this purpose [4], [5].<br />
The artificial muscle, which construction is simple, was made of a rubber<br />
inner tube covered with a shell braided according to helical weaving. The<br />
muscle was closed by two ends, one being the air input and the other the force<br />
attachment point. When the inner tube was pressurized, the muscle inflated and<br />
contracted [6].<br />
The Bridgestone rubber company (Japan) proposed a redesigned and more<br />
powerful version of the pneumatic muscle in the 1980s under the name of<br />
Rubbertuators and used them to power an industrial use robot arm, Soft Arm.<br />
At the present, McKibben-like muscles are being brought to the market by<br />
Festo Ag. & Co.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 119<br />
2.2. Generalities<br />
The pneumatic muscles have many characteristics that make them being<br />
easy to use and with great performances. Some of these characteristics are:<br />
shock - absorbing, adjustability, simulating capability, storage capable,<br />
safeness, lightweight, natural compliance and shock resistance.<br />
Favourable response to commands, known as compliance is in direct<br />
relation to air compressibility, and hence the pneumatic muscle can be<br />
influenced by controlling/adjusting the command pressure.<br />
The function principle consists in the fact that, under the action of<br />
compressed air, the pneumatic muscle, which is blocked at one end, shortens its<br />
lengths and expands its diameter. As the volume of the internal tube increases<br />
due to the increase in pressure, the actuator shortens with a certain stroke.<br />
Pneumatic muscle construction is based on an interior tube, made from<br />
neoprene rubber wrapped in braided sleeves made of nylon with strengthening<br />
and protecting role. The braided sleeves act to constrain the expansion for<br />
maintaining the cylindrical shape. As we can see in the figure 1, the angle of the<br />
enveloping tissue, denoted by α, is one in relaxed state and differs in contracted<br />
state. It has the value of 25.4º in the relaxed state of the muscle and of 54.7º at<br />
maximum contraction. [4]<br />
In Fig. 2 is presented the working principle of a pneumatic muscle.<br />
Fig. 2 – Working principle of a Fig. 3 – Dependence of the force on the<br />
pneumatic muscle envelope angle and on the working pressure [7]<br />
(1)<br />
The force F developed by pneumatic muscle is given by Eq. (1) [4].<br />
π<br />
= ⋅ ⋅<br />
4 d p F<br />
2<br />
⎡ 2<br />
3⋅<br />
cos α − 1⎤<br />
⋅ ⎢ 2 ⎥<br />
⎢⎣<br />
1 − cos α ⎥⎦<br />
,<br />
where p is the working pressure and d the interior diameter of the pneumatic<br />
muscle. Upon completion of the maximum stroke the developed force is equal<br />
to zero. Equation (1) allows plotting of the graph featuring the force developed
120 Ioana Petre et al.<br />
by a pneumatic muscle versus the enveloping angle and feed pressure (Fig. 3)<br />
[5].<br />
The advantages of the pneumatic muscles utilization are: power to weight<br />
ratios in excess of 1 kW/kg; a varying force-displacement relation at constant<br />
gas pressure, an adjustable compliance; the absence of friction and hysteresis;<br />
the ability to operate at a wide range of gas pressures, and thus to develop both<br />
very low and very high pulling forces; the possibility of direct connection to a<br />
robotic joint; cheaper to buy and install than other actuators and pneumatic<br />
cylinders; smooth and natural movement; fast -full contraction.<br />
Disadvantages are: the force which can be applied is only tensile in nature;<br />
its total displacement is only about 20% to 30% of its initial length; friction<br />
between the netting and the tube leads to a substantial hysteresis in the forcelength<br />
characteristics; rubber is often needed to avoid the tube from bursting;<br />
rubber deformation will lower the force output of this type of muscle up to 60%<br />
[8].<br />
3. Industrial Applications of Pneumatic Muscles<br />
Pneumatic muscles were, lately, used in many applications especially in the<br />
field of industrial robots, because of the evolution of the technology.<br />
The major application of pneumatic muscles is in the field of industrial<br />
robots, usually in those that mimic human actions. Because of their analogy<br />
with human skeletal muscles, they are able to perform many functions specific<br />
to human hand.<br />
As example in this area we can specify the stepping robot WAP-1 is the first<br />
biped robot designed in 1969 by Ichiro Kato from Waseda University of Tokyo.<br />
(Fig. 4a)<br />
a b<br />
Fig. 4 a – WAP-1(source www.androidworld.com) ; b - Humanoid robot manufactured<br />
by Shadow robotic company (source www.shadowrobot.com)
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 121<br />
The humanoid bipedal walking machine developed by Shadow Robot<br />
Company is a robot that can function in a human environment to do various<br />
repetitive things. (Fig. 4b)<br />
The humanoid muscle-robot developed by Festo AG & Co. in collaboration<br />
with the department of bionics and evolutionary technology of the Technical<br />
University of Berlin (Fig. 5). This robot is a feasible extension and place holder<br />
of men capable of operating in places either too dangerous or inaccessible for<br />
human beings. It can operate from terrestrial and deep sea operations to tasks<br />
carried out in outer space.<br />
Fig. 5 – Humanoid muscle-robot moving (source www.festo.com).<br />
Pneumatic muscles have been used in construction of artificial limbs.<br />
At the bio robotic lab of University of Washington the limb as shown figure<br />
6a was developed. The major requirements of their research team were:<br />
continuous operation, low weight, quieter operation, user satisfaction, no<br />
maintenance.<br />
a b<br />
Fig. 6a – Artificial limb developed at the bio robotics Lab, University of Washington.<br />
b – Shadow Leg (source www.shadowrobot.com)
122 Ioana Petre et al.<br />
Another application for the pneumatic muscle is the leg made for Shadow<br />
Robot Company by David Buckley from North Carolina A & T University.<br />
(Fig. 6b)<br />
The dexterous hand was developed by the Shadow robotic company. The<br />
hands operate just like human hands with five fingers. It contains an integrated<br />
bank of 40 Air Muscles which make it move. The figure 7 illustrates this fact.<br />
Fig. 7 – The Shadow Hand (source www.shadowrobot.com).<br />
Caldwell used 18 small McKibben Muscles to power a dexterous fourfingered<br />
manipulator. Hannaford built an anthropomorphic arm, having fifteen<br />
McKibben Muscles. The Soft Arm, developed by Bridgestone Co. has a<br />
shoulder, an upper arm, a lower arm and wrist, and a useful payload of<br />
maximum 3 kg. Yoshinada used hydraulically actuated McKibben Muscles to<br />
power an underwater manipulator [1].<br />
New Application Areas are: Simulator Technology, High Speed cutting<br />
processes, Aeronautical Technology, Wood-working, Metal-working, Medical/<br />
Biomedical, Mobile Applications, Building / Construction, Mining, Process/<br />
Water, Amusement/ Recreation, Medical / Surgical / Hospital<br />
4. Conclusions<br />
1. Even though pneumatic muscles are not capable of offering an<br />
extremely wide range of operations, in the case of humanoid robots they offer a<br />
wide range of possibilities. Their low assembly weight and high power-toweight<br />
ratio make the pneumatic muscles to be considered for use in mobile<br />
robotics [1].
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 123<br />
2. The utilization of the pneumatic muscle is soon to become a good<br />
alternative for the present day electric or mechanically drives because of their<br />
characteristics and because of the evolution of the industrial development.<br />
Received: March 15, 2010 1 Transilvania University of Braşov,<br />
Technological Engineering Faculty,<br />
Economic Engineering and Production Systems Department,<br />
Braşov, România<br />
e-mail: ioana.petre@unitbv.ro<br />
2 Material Science and Engineering Faculty,<br />
Materials Engineering and Welding Department,<br />
e-mail: petredanbv@yahoo.com<br />
R E F E R E N C E S<br />
1. D a e r d e n F., L e f e b e r D., The concept and design of pleated pneumatic<br />
artificial muscles, International Journal of Fluid Power, 2, 3, pp. 41–50, 2001.<br />
2. M a r c i n č i n J., P a l k o A., Negative pressure artificial muscle—An<br />
unconventional drive of robotic and handling systems, Transactions of the<br />
University of Košice, pp. 350– 354, Riecansky Science Publishing Co, Slovak<br />
Republic, 1993.<br />
3. B a l d w i n H. A., Realizable models of muscle function, Proceedings of the First<br />
Rock Biomechanics Symposium, pp. 139–148, New York, 1969.<br />
4. S c h u l t e H. F., The characteristics of the McKibben Artificial Muscle, The<br />
Application of External Power in Prosthetics and Orthotics, pp. 94–115,<br />
National Academy of Sciences–National Research Council, Publication 874,<br />
Lake Arrowhead, 1961.<br />
5. D a v i s S.T., C a l d w e l l D.G., The biomimetic design of a robot primate using<br />
pneumatic muscle actuators, 4th International Conference on Climbing and<br />
Walking Robots CLAWAR, 2001.<br />
6. K e n n e t h, K u K.K., B r a d b e e r R., Static Model of the Shadow Muscle under<br />
Pneumatic Testing, 2006<br />
http://www.ee.cityu.edu.hk/~rtbrad/muscles%20riupeeec%202006.pdf<br />
7. D e a c o n e s c u A. , D e a c o n e s c u T., Contribution to the Behavioural Study<br />
of Pneumatically Actuated Artificial Muscles. 6 th International Conference of<br />
DAAAM Baltic Industrial Engineering, Tallinn, Estonia, Vol. 1, pag. 215-220,<br />
2008<br />
8. * * * Air Muscles 2008 [Online] Available at:<br />
http:// www.techalone.com 23/01/2010] – [Accessed 27.12.2009]<br />
http:// www.festo.com – [Accessed 27.12.2009]<br />
http://www.shadowrobot.com – [Accessed 27.12.2009]
124 Ioana Petre et al.<br />
APLICAŢII INDUSTRIALE ALE MUŞCHILOR PNEUMATICI<br />
(Rezumat)<br />
Muşchiul pneumatic este un sistem bazat pe o membrană care se contractă, şi care,<br />
sub acţiunea aerului comprimat îşi măreşte diametrul şi îşi micşorează lungimea.<br />
Lungimea cursei depinde in mod direct de nivelul presiunii alimentate. Un muşchi<br />
pneumatic include un tub interior care este realizat dintr-un material elastic, de obicei<br />
neopren. Acest tub este acoperit de un ţesut cu mai multe straturi realizat din nylon,<br />
pentru a-i da rezistenţă şi pentru a-l proteja de influenţele din mediul de lucru.<br />
Caracteristici precum capacitatea de a absorbi şocurile, greutatea redusă, rezistenţa<br />
la şocuri, gabarit si masa redusa pe unitatea de putere (1KW/kg), elasticitate<br />
(comportare ca de arc) datorata pe de-o parte compresibilităţii aerului si pe de alta<br />
variaţiei forţei cu deplasarea, amortizarea şocurilor datorate impactului, posibilităţi de<br />
conectare uşoara, siguranţa (fără pericol de electrocutare sau incendiu) fac din muşchii<br />
pneumatici instrumente fezabile pentru utilizarea uşoară in domenii diverse şi cu<br />
performanţe deosebite. Domeniile de aplicare ale muşchilor pneumatici se referă la<br />
robotică, biorobotică, biomecanică, dispozitive de protezare si sustinere a scheletului<br />
osos şi industrie.<br />
În această lucrare se prezintă câteva aplicaţii industriale ale muşchilor pneumatici. Deşi<br />
prima apariţie a muşchilor pneumatici a fost în 1930 fiind inventaţi S. Garasiev, iar apoi<br />
în 1950 muşchii cu membrană împletită introduşi de J. L. McKibben aceştia au cunoscut<br />
o utilizare mai intensa abia in ultima perioadă.<br />
Câteva exemple in domeniul roboţilor industriali sunt: braţul acţionat pneumatic<br />
realizat de J.L.Mc Kibben, robotul WAP 1 realizat la <strong>Universitatea</strong> Waseda din Tokyo,<br />
robotul humanoid conceput de Festo in colaborare cu <strong>Universitatea</strong> Tehnică din Berlin,<br />
robotul humanoid realizat de Compania Shadow, proteza realizată de <strong>Universitatea</strong> din<br />
Washington, piciorul şi mâna realizate pentru Compania Shadow, care efectuează<br />
mişcări similare cu cele efectuate de corpul uman. Desigur că aplicaţii sunt multe si nu<br />
au fost toate menţionate in acest articol, însa trebuie reţinut că muşchii pneumatici,<br />
datorită caracteristicilor lor, sunt utilizaţi în special în domeniul roboţilor mobili.<br />
Aceştia înlocuiesc cu succes motoarele electrice sau hidraulice fiind mai uşor de folosit<br />
şi îndeplinind aceleaşi funcţii.<br />
Dezvoltarea continuă a tehnologiei a dus la apariţia unor noi arii de aplicaţie pentru<br />
muşchii pneumatici, cum ar fi: tehnologia simulării, procese de tăiere cu viteză mare,<br />
tehnologia aeronautică, lucrul cu lemnul, lucrul cu metalul, industria medicală,<br />
construcţii, industria minieră.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
SOME NEW RESOURCES ON COMPUTER ASSISTED<br />
EXPERIMENTAL RESEARCH OF THE ABSORBED ELECTRIC<br />
POWER IN MANUFACTURING SYSTEMS<br />
BY<br />
MIHĂIȚĂ HORO<strong>DIN</strong>CĂ<br />
Abstract. This paper presents some theoretical considerations and experimental<br />
results on computer aided research of manufacturing systems (e.g. machine-tools)<br />
using the evolution of the absorbed electric power (EP) of the driving motors. Based<br />
on an original computer aided data acquisition and processing method, the electric<br />
power evolution monitoring (and its components, especially the EP real part) should<br />
be now a very useful tool in experimental research (kinematic chains condition,<br />
working processes loading, and so on). In a new approach the numerical describing<br />
with high resolution of real EP evolution can be used also to evaluate the<br />
performances in dynamic conditions, especially using the evolution in frequency<br />
domain (by power spectral density analysis). Some phenomena in kinematic chains<br />
was observed with a certain scientific priority, using the real EP monitoring<br />
capabilities, such as the behaviour of driving belts and the behaviour of the elastic<br />
system of the rotor on the electric driving motor. The real EP monitoring can be<br />
used to supervise any other electric actuated equipment.<br />
Key words: manufacturing systems, experimental research, real electric power,<br />
computer aided monitoring, data processing.<br />
1. Introduction<br />
The working processes (WP) on manufacturing systems (MS) or any other<br />
working equipment needs mechanical energy, usually delivered by an electric<br />
driving motor (EDM) and transported by a kinematic chain. By mechanical<br />
loading point of view the behaviour of MS is mirrored in the real part of electric<br />
power (EP) evolution (active electric power, AEP). The EDM seems to be a<br />
very appropriate sensor of mechanical loading, useful in MS and WP<br />
monitoring and diagnosis or even in WP active control. The state of the art [2],<br />
[3] indicates that generally for monitoring it is used just the electric current
126 Mihăiță Horodincă<br />
absorbed by the EDM, but for MS driven with alternating voltage EDM, the<br />
best characterization of loading is done using AEP evolution. In some papers<br />
[4], [5] the AEP it is taken into account, but the research resources are not<br />
completely exploited. This paper presents some experimental research results<br />
based on a new approach on computer aided AEP monitoring for MS powered<br />
by EDM.<br />
2. EP Monitoring Principle on EDM and Experimental Conditions<br />
Fig. 1 describes a computer assisted monitoring bench test developed for an<br />
asynchronous EDM supplied on a three phase 50 Hz frequency sinusoidal<br />
voltage, symmetrical network (3×380V). A single phase is used for computer<br />
data acquisition (instantaneous current i(t) IC and voltage u(t) IV in time<br />
Fig. 1 – Experimental features of the EP bench test.<br />
evolutions [6] provided by two transformers CT and VT) via a numerical<br />
oscilloscope (ADC 212/50, Pico Technology Limited, UK). The acquired data<br />
are processed in a personal computer. The instantaneous electric power p(t)<br />
evolution (IEP) is given by: p(t)=3·u(t)·i(t). The AEP (one value on each period<br />
T) is given by:<br />
(1)<br />
T<br />
1<br />
3<br />
P = ∫ p(<br />
t)<br />
⋅ dt = ⋅ ∫u(<br />
t)<br />
⋅i(<br />
t)<br />
⋅ dt = 3⋅U<br />
⋅ I ⋅cosϕ<br />
,<br />
T T<br />
0<br />
T<br />
0<br />
where: T=20 ms is the period (reciprocal of the frequency f), U - the effective<br />
(root mean square, RMS value) voltage, I - the effective current, φ - the angle of
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 127<br />
phase (AP) between current ant voltage. Using the numerical values u[tk] and<br />
i[tk] of the voltage and the current, the numerical evolution of the AEP can be<br />
described by calculus (based on Eq. (1), as average value on the period T) with:<br />
(2)<br />
n⋅(<br />
l+<br />
1)<br />
n⋅(<br />
l+<br />
1)<br />
3<br />
3<br />
P[ tl<br />
] ≈ ∑(<br />
u[<br />
tk<br />
] ⋅i[<br />
t k]<br />
⋅ Δ t)<br />
= ⋅ ∑(<br />
u[<br />
tk<br />
] ⋅i[<br />
t k])<br />
,<br />
T<br />
n<br />
k=<br />
n⋅l+<br />
1<br />
k=<br />
n⋅l+<br />
1<br />
where: P[tl] is the numerical value of the AEP (tl=l·T, l=1,2….m, m values for a<br />
registration of m·T total duration), n=T/Δt, Δt is the reciprocal of the sampling<br />
rate for u[tk] and i[tk], tk=k·Δt. If Δt→0 then in Eq (2) the symbol ≈ becomes =.<br />
The sampling rate of AEP is equal with frequency f value (50 s -1 ). Using the<br />
effective values U[tl] and I[tl] (the amplitudes of the voltage and current divided<br />
by square root of 2), the AP and REP (reactive electric power Q) are given by:<br />
(3)<br />
P[<br />
tl<br />
]<br />
ϕ [ tl<br />
] ≈ arccos<br />
and Q[ tl<br />
] ≈ 3⋅U<br />
[ tl<br />
] ⋅ I[<br />
tl<br />
] ⋅sin(<br />
ϕ[<br />
tl<br />
]) .<br />
3⋅U<br />
[ t ] ⋅ I[<br />
t ]<br />
l<br />
l<br />
The experimental research was done on a Romanian lathe SNA 360 (used as<br />
MS, see Fig. 1) with the kinematics of the main shaft gearbox described in Fig.<br />
Fig. 2 – Overview on the kinematics of the SNA 360 lathe main shaft<br />
gearbox used as MS in experimental research.<br />
2. An asynchronous EDM with 5.5 KW and 1440 rpm (speed in rotations per<br />
minute) is used to drive it. Two different configurations of the gearbox were<br />
used (see Kd1 and Kd2), each one is able to actuate the main spindle (shaft) Ms<br />
on 1000 rpm or 1600 rpm.
128 Mihăiță Horodincă<br />
3. Some Experimental Research Results on Monitoring<br />
Let be first a simple experiment. The belt drive transmission BT1 is<br />
switched-off (the belt was removed), no loading on EDM’s rotor anymore.<br />
Fig. 3 – The AEP evolution during EDM starting procedure<br />
(no loading on the EDM’s rotor).<br />
The evolution of AEP during the EDM starting procedure (star-triangle<br />
Fig. 4 – Gearbox start-stop clutching dynamics (on Kd1<br />
configuration) mirrored in AEP evolution.<br />
connection) is given in Fig. 3. A strong transitory regime (high power, short
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 129<br />
time, sees Bs area) is observed, because of the dynamics of accelerated rotor<br />
motion, it needs a big amount of AEP stored as kinetic energy. In D the star<br />
connection is changed automatically in triangle connection, also a transitory<br />
regime (marked as Bt) occurs.<br />
Fig. 5 – Numerical filtered (with p=10) evolutions for AEP<br />
(given in Fig. 4) and REP.<br />
There is an interesting behavior of the EDM mirrored in AEP evolution, see the<br />
regions I and II. There are two free responses on 5.2 and 10.8 Hz frequency, due<br />
to the elastic system associated to the rotor (mechanical inertia, magnetic field<br />
stiffness and air friction damping).<br />
A second experiment describes (Fig. 4) the mirroring of the clutching<br />
dynamics of the gearbox (on Kd1). In A the MC3 clutch is engaged, the gearbox<br />
start the motion and the AEP consumption increases during a transitory regime<br />
B (with a 4.16 W·h average electrical energy stored as kinetic energy). The<br />
peak B1 indicates that the BT1’s driving belt start the sliding (the dynamic<br />
torque is bigger than the friction torque between the belt and the pulleys). In B2<br />
the sliding disappears and immediately after (in C) starts the steady-state regime<br />
on 1000 rpm. Here the absorbed AEP is used to cover the waste of mechanical<br />
energy (dry and viscous friction).<br />
The AEP evolution seems to be noisily, but it will be proved later (Fig.7)<br />
that it is very useful for gearbox diagnosis. In D the clutch is disengaged (by<br />
switching-off the electrical supply) so the AEP suddenly decreases. The level of<br />
power in E is for a short time lower than in A, the difference (183 W) describes<br />
the viscous friction in MC3 clutch.
130 Mihăiță Horodincă<br />
For certain experimental purposes the AEP evolution can be numerical<br />
filtered using a sliding average low pass filter described with the equation:<br />
p xin[<br />
i]<br />
xout[<br />
j]<br />
= ∑ where: p is the filter parameter, xin[i] - the input signal, xout[j]<br />
i=<br />
1 p<br />
- the output signal, both in numerical format, j=i+p. Figure 5 shows the result<br />
of filtering for AEP evolution given in Fig.4 (with p=10).<br />
Also is presented the filtered evolution of REP in the same experimental<br />
conditions. As it is well known from the electrical network theory, the variation<br />
of REP (1,036 VAR peak to peak) is smaller than the AEP variation (4,805 W).<br />
The evolution of AEP during a third experiment is described in Fig. 6.<br />
There is a dynamic behaviour with negative absorbed AEP. Before A, the<br />
gearbox moves in steady-state idle running regime on 1600 rpm (using Kd2<br />
kinematic chain with MC2 clutch engaged). In A the clutch MC3 is engaged<br />
(and MC2 is automatically switched-off), so Kd2 becomes active and Kd1<br />
inactive. Here occurs a transitory regime B, with negative AEP (the angle φ ><br />
π/2, see Eq. (1)), for a short time the EDM works as a brake, it convert the<br />
Fig. 6 – Clutching dynamics with negative AEP.<br />
available kinetic energy from gearbox in electrical energy (with a total amount<br />
of 0.464 W·h) delivered on the electrical network. After that the AEP increases<br />
with an overshoot in C, the generator becomes again electric motor. In D the<br />
steady-state on 1000 rpm is installed. In E the MC3 clutch is switched-off, so<br />
just the shaft I is driven. The AEP consumption in steady-state idle regime is<br />
smaller than in Fig.4 because of the gearbox heating (the viscosity of the<br />
lubricant so the viscous friction decreases).
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 131<br />
4. Gearbox Diagnosis using the AEP Processing in Frequency Domain<br />
There is also a very important experimental research resource useful in<br />
gearbox diagnosis. The AEP or IEP evolution in steady-state it is a mixture of<br />
many variable components. Each variable signal’s component describes the<br />
behavior of a gearbox’s part (belts, spindles, gears, et cetera). The identification<br />
of these components can be done [1] by computer assisted AEP or IEP power<br />
spectral density (PSD) analysis using Fast Fourier Transform (FFT). The<br />
correlation between signal’s components and gearbox’s parts can be done using<br />
the frequency and the speed of rotation. Figure 7 presents a zoom on the PSD of<br />
the AEP evolution in steady state on 1000 rpm (already described before; see on<br />
Fig. 4 the encircled area).<br />
On Fig. 7 there are a lot of different peaks, each one described by the frequency<br />
and the amplitude (the higher peak’s amplitude are not clearly indicated because<br />
Fig. 7 – The evolution of power spectral density of the<br />
AEP signal absorbed on 1000 rpm (steady-state).<br />
of zoom). The PSD amplitude is proportional with the amplitude of signal<br />
component. Each peak has a number, for each one there is a short description<br />
about the frequency and the correspondence with the gearbox parts (see Fig. 2).<br />
The peak 5 is associated with the spindle II behaviour, the peak 7 with the<br />
spindles III and V, the peak 9 with the spindle I and the peak 10 with the<br />
EDM’s rotor (24.84 Hz frequency, for 1490.4 rpm the real speed of rotation).<br />
The peak 9 is generated by the driving belt pulley run-out error motion (65 μm).<br />
There are also some harmonics of these peaks. There is a very interesting<br />
research result, with certain scientific priority, here is very well mirrored the
132 Mihăiță Horodincă<br />
behaviour of the driving belt from BT1. The belt generates a very strong<br />
variable component of AEP, with the fundamental frequency on 5.4 Hz (first<br />
peak on Fig. 7) and 6 harmonics (see the peaks 4, 6, 9, 11, 14 and 18). It is<br />
generated by the belt’s stiffness variation (the belt is a little bit damaged, it has a<br />
small tear). The fundamental’s frequency fFB is given by:<br />
π ⋅ D1<br />
π ⋅125mm<br />
(4)<br />
fFB<br />
= f EDM ⋅ = 24.<br />
84Hz<br />
⋅ = 5.<br />
38Hz<br />
.<br />
L<br />
1812mm<br />
c<br />
Fig. 8 – The BT1’s belt behaviour mirroring in PSD<br />
analysis of the IEP signal evolution.<br />
where: fEDM is the rotation frequency of the EDM’s rotor, D1- the diameter of<br />
the belt pulley of the EDM’s rotor, Lc- the belts’ length. A new experiment was<br />
done, with PSD analysis on IEP evolution during the steady-state regime, with<br />
all the clutches disengaged, see Fig. 8. The BT1’s belt behaviour is better<br />
indicated with the harmonic F and the harmonics H1,H2…H6. Because the<br />
mechanical loading of the EDM is smaller, the speed of rotation is bigger so the<br />
frequency in Fig. 8 increases. The first harmonic is bigger than the fundamental<br />
because of the EDM’s rotor elastic system resonance.<br />
5. Conclusions<br />
1. The paper proves that the computer-assisted monitoring of EP parameters<br />
on manufacturing systems (or any other electric actuated equipment) can be an<br />
available experimental research procedure.<br />
2. The computer assisted EP monitoring uses very simple procedures of data<br />
acquisition and processing. Using these research features, it is possible to detect<br />
and study a lot of new static and dynamic phenomena from kinematic chain.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 133<br />
Acknowledgements. The author would like to thank Prof. Costache D r u t u. For, he<br />
is the one who, long time ago, gave me the basic idea of some of my researches.<br />
Received: “Gheorghe Asachi” Technical University<br />
Department of Machine-Tools<br />
Iasi, Romania,<br />
e-mail: horodinca@tuiasi.ro<br />
R E F E R E N C E S<br />
1. B i r a n A., B r e i n e r M., Matlab for Engineers, Addison-Wesley Pub. (SD), May<br />
1995, ISBN-13: 978-0201565249.<br />
2. D i m l a E., D i m l a., Sensor signals for tool-wear monitoring in metal cutting<br />
operations – a review of methods, Int. J.of Machine Tools & Manufacture, 40, 8,<br />
(2000), pp. 1073-1098, ISSN:0890-6955.<br />
3. D o n g-W o o, C., S a n g J. L., C h o n g N a m C., The state of machining process<br />
monitoring research in Korea, Int. J. of Machine Tools & Manufacture, 39, 11,<br />
November, 1999, pp. 1697-1715.<br />
4. H w a-Y o u n g K., J u n g-H w a n A., Chip disposal state monitoring in drilling<br />
using neural network based spindle motor power sensing, Int. J. of Machine<br />
Tools & Manufacture, 42, 10 (2002), pp 1113-1119.<br />
5. M i n g L., T e t Y., S a e e d R., Z h i x i n H., Fuzzy control of spindle power in<br />
end milling process, International Journal of Machine Tools & Manufacture, 42,<br />
14 (2002), pp. 1487-1496, ISSN:0890-6955.<br />
6. H o r o d i n c ă M., Utilizarea parametrilor energetici în monitorizarea, diagnoza şi<br />
conducerea sistemelor de prelucrare, Ed. Performantica, Iaşi, 2004.<br />
NOI RESURSE ALE CERCETĂRII EXPERIMENTALE<br />
ASISTATE DE CALCULATOR A PUTERII ELECTRICE<br />
ABSORBITE ÎN SISTEMELE DE FABRICAȚIE<br />
(Rezumat)<br />
Lucrarea prezintă o serie de noi oportunităţi ale cercetării experimentale a<br />
sistemelor de fabricaţie (monitorizare şi diagnoză) pe baza analizei evoluţiei temporale<br />
şi în domeniul frecvenţă a parametrilor încărcării energetice (putere activă PA şi<br />
instantanee PI) ai electromotoarelor de acţionare. Se prezintă exemplificări legate de<br />
monitorizarea proceselor tranzitorii şi de utilizarea analizei spectrale a semnalelor PA şi<br />
PI, pentru cercetarea comportării lanţurilor cinematice ale mişcării principale.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
INTEGRATED CAE DEVELOPMENT OF PRECESSIONAL<br />
DRIVES USING AUTODESK INVENTOR PLATFORM<br />
BY<br />
ION BOSTAN, VALERIU DULGHERU<br />
and ANATOL SOCHIREANU<br />
Abstract: The paper presents the modelling and simulation of processional<br />
drives designed in two variants capable of high transmission ratio and torque for one<br />
stage compact construction. The constructions were designed in Inventor and also as<br />
multi body systems in MotionInventor. The simulations of the drives provide<br />
information concerning positions, velocities, accelerations, point trajectories, forces<br />
and moments, energies, as well as contact forces at the contact between gear teeth<br />
and satellite teeth and other data concerning the system. The bearings of the two<br />
drives were modelled as multi body systems in SolidDynamics for defining an<br />
approach for studying the noise emission during running through simulation with<br />
possibility of extension to the contact between teeth.<br />
Key words: precessional transmissions, Inventor, 3D multi body model<br />
SolidDynamics<br />
1. Introduction<br />
The engineering methods based on computer permitted to develop a new<br />
type of processional transmissions with multi-couple meshed teeth, which, from<br />
the technological point of view, can be manufactured by means of a new<br />
method of processing conical teeth with convex-concave profile.<br />
It appeared the necessity of elaboration of new profiles adequate to the<br />
spheroid-spatial motion of the gears, which would ensure high performances to<br />
the processional transmission.<br />
Considering the necessity of achieving the transfer function continuity and<br />
gear multiplicity some objectives were taken into account. One of them is the<br />
integrated method of design, modelling and simulation using powerful means of
136 Ion Bostan et al<br />
creation and management of parametrical models of the mechanical assemblies<br />
on the basis of CAD-CAE.<br />
2. General Information<br />
The ever-growing requirements, especially, concerning the bearing capacity,<br />
the kinematical accuracy and the kinematical possibilities, impose the necessity to<br />
develop a new type of planetary gearing with distinct performances. The gearing<br />
improving is one solution of the problem. Novicov - Wildhaber, Symarc and other<br />
gearings have increased considerably the bearing capacity of gear. Another<br />
direction of developing gears is the design of new types of mechanical gears.<br />
The creative search of designers has crowned with the elaboration of a new<br />
type of gearing – harmonic gear. W. Musser, an American engineer, patented the<br />
action principle of the harmonic gear in 1959. Starting that year W. Musser<br />
patented a big number of diverse constructive diagrams for harmonic gears (teeth,<br />
friction and tapped gears) and couplings, and demonstrated the possibilities of the<br />
new construction principle of mechanical gears. Thus, in 1961, the harmonic<br />
drive was produced at one of the American companies, for the first time at<br />
industrial scale. Harmonic drive are compact and possess increased bearing<br />
capacity; they provide high kinematical accuracy and possibility to transmit<br />
motion in sealed mediums, which is one of the basic advantages of harmonic<br />
drive. As their disadvantages we can mention reduced reliability of the flexible<br />
element (and, thus, of the gear, on the whole), reduced working order at high<br />
speeds, and also some technological difficulties.<br />
By the end of the 70s Prof. Ion Bostan designed a new type of gears, new in<br />
principle, - processional planetary gears with multiple gearing (B o s t a n, 1991).<br />
Over 20 years, research was carried out comprising the total range of problems<br />
from the idea to the implementation: fundamental theory of the precessional gearengineering<br />
calculation procedures-Know-How manufacturing technologiesapplications.<br />
The research results have been published in over 450 scientific<br />
papers, in about 150 patents, in 3 monographs and in one design guidebook.<br />
The absolute multiplicity of the processional gear (up to 100% of teeth pairs,<br />
geared simultaneously, compared to 5-7% in classical gears) provides increased<br />
bearing capacity and kinematical accuracy, small dimensions and mass. In<br />
addition to the above said, extended kinematical possibilities (±8÷3600 compared<br />
to 79÷300 in sinusoidal gears), reduced acoustic emission and solution of all<br />
technological problems, as advantages, open new perspectives for precessional<br />
planetary gears utilisation in various fields of mechanical engineering (B o s t a n,<br />
1991). On the whole, processional planetary gears can be divided in two basic<br />
groups:<br />
- power processional planetary gears;<br />
- kinematical processional planetary gears.
As the planetary processional<br />
transmission is the new transmission, in<br />
the beginning of effectuation the<br />
simulation and calculation with<br />
programs CAE it is necessary to<br />
calculate theoretically base parameters.<br />
On fig. 1 is shown kinematical scheme<br />
of planetary processional transmission,<br />
but on fig. 2 is shown the design of<br />
planetary processional transmission. The<br />
main elements of transmission make:<br />
crankshaft 1, block satellite 2, fixed<br />
wheel gear 3 and mobile wheel gear 4.<br />
The important point this that, having<br />
such design (single-stage) is possible to<br />
receive transfer rate up to 3600, using<br />
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 137<br />
designs of double-stage it is possible to receive transfer rate up to 14 million.<br />
Obviously that, having such big transfer rate, there are big loadings on a tooth.<br />
This problem is solved by that, all teeth it is participate in the gearing, and<br />
loading is transferred by half from them. Meaning that, all teeth are in gearing at<br />
the big loadings, the kinematic error is small.<br />
Fig. 1 – Kinematical scheme of<br />
planetary processional transmission.<br />
Principle of functioning of the planetary processional transmission the<br />
following: crankshaft 1 with inclined section to specify a block satellite 2<br />
spatially spherical movement, block satellite by means of roller crown<br />
interaction with fixed wheel gear 3 and mobile wheel gear 4 (having a special<br />
structure generated by means of the equations), in turn a mobile wheel gear 4 it<br />
is rigidly connected with output shaft of a reducer, transfers the moment and<br />
speed of rotation. The direction of rotation of an input shaft and output shaft can<br />
be in one or in different directions. It is visible and in calculation of transfer rate<br />
if it has positive number we have an identical rotation. The transfer rate of a<br />
planetary processional transmission is defined by the relation (1).<br />
1 3 6 2 7 4 5<br />
Fig. 2 – Design of planetary processional transmission.<br />
(1)<br />
Zg Z<br />
1 a<br />
i =−<br />
,<br />
ZbZg − Zg Za<br />
2 1<br />
where: 1 , 2 are<br />
Z g Z g<br />
number of a roller the<br />
crown satellite g1<br />
and g2<br />
;<br />
Z a and Zb<br />
is represented<br />
number of a teethes the<br />
cog-wheels a and b .
138 Ion Bostan et al<br />
3. Calculation by CAE Simulation of Kinetostatics<br />
Parameters of Processional Transmission<br />
3.1. 3D Model elaboration of Planetary Processional Transmission<br />
Calculation of planetary processional transmission by a simulation is carried<br />
out using the simplified 3D model created in program Motion Inventor 2004+<br />
but (4) from which it is possible to determine and check up dynamic loadings in<br />
bearings which have been designed earlier.<br />
Dynamic processes in planetary processional transmission derive, to a great<br />
extent, from the interaction of conical rollers of the satellite crowns with<br />
generating surfaces of central wheel teeth. The bearing capacity defined by gear<br />
forces (static and dynamic), the noise emission and the transmission<br />
vibroactivity, depend on the gear dynamic processes on the whole. With<br />
account of these important factors in the elaboration of 3D model of the initial<br />
processional gearing, the linear contour of central wheel teeth profile (fig. 3, a)<br />
were designed applying parametric equations:<br />
(2)<br />
Where:<br />
X = k Z + d ;<br />
Y k Z d ;<br />
(k d<br />
=<br />
k d )<br />
−<br />
m m m m<br />
1E 2 1E 2<br />
m<br />
1E =<br />
m<br />
1<br />
m<br />
1E −<br />
m<br />
1<br />
m<br />
Z1E<br />
m m m m<br />
1 1 − 2 2<br />
m2 m2<br />
k1 + k2 + 1<br />
(k d − k d ) + (k + k + 1) ⋅(R −d−d )<br />
Z ,<br />
m m m m 2 m2 m2 2 m2 m2<br />
m<br />
1E =<br />
1 1 2 2 1 2<br />
m2 m2<br />
k1 + k2 + 1<br />
D 1 2<br />
⎛ ⎞<br />
• • •<br />
m m m m m m 2 m<br />
X1D⎜X1D X1D+ Y1DY1D⎟ + Z1D X1D<br />
+<br />
m m<br />
1 =<br />
⎝ ⎠<br />
2 = −<br />
• •<br />
m<br />
m⎛ m m m m ⎞<br />
X1D<br />
Z1DX1DY1D−Y1D X1D<br />
k ; k<br />
⎜ ⎟<br />
⎝ ⎠<br />
2 m m<br />
( Rcos D β + dY 1 1D)<br />
•<br />
2 m<br />
m Rcos D β X1D<br />
1 = • •<br />
⎛ m m m m⎞ X1DY1D− X1DY1D m<br />
2 =<br />
m<br />
X1D<br />
d ; d .<br />
⎜ ⎟<br />
⎝ ⎠<br />
m m m ( kY 1 1D Z1D)<br />
The 3D model of the central wheel (fig. 3, b) was designed by using CAD<br />
Autodesk Inventor (5) (B o s t a n et al, 2007).<br />
;
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 139<br />
a. b.<br />
Fig. 3 – 3D linear contour of the central wheel teeth (a),<br />
and the model 3D of the central wheel (b).<br />
The 3D model for calculation is shown on fig. 4, it includes a crankshaft 1,<br />
a fixed wheel gear 2, the block satellite 3, a mobile wheel gear 4 rigidly<br />
connected with output shaft 5.<br />
Fig. 4 – 3D Model of planetary processional transmission.<br />
The dynamic model has been created on the basis of rigid model. As the<br />
initial data has been specified speed on input shaft [deg/sec] and the moment of<br />
torsion on the output shaft [Nm].<br />
Kinematic joints fig. 5, have been enclosed according to movements in<br />
gearing. The crankshaft and the case of a reducer by means of cylindrical roller<br />
bearings are connected by a cylindrical joint to an opportunity of a selfcentering.<br />
A crankshaft and the block satellite by means of two tapered bearings<br />
with pair rotation in points of the appendix of loading on bearings (point-line),<br />
pair rotation (revolution) rollers on an axis of the block satellite and 3D contact<br />
of rollers to a mobile and fixed wheel gear.
140 Ion Bostan et al<br />
Fig. 5 – Kinematics joints in the planetary precessional transmission.<br />
3.2. Precessional Transmission Kinetostatic Simulation<br />
The simulation of model processional gearings has been created in some<br />
stage. At the first stage has been executed the kinematic analysis, from<br />
definition of the following parameters: transfer ratio, absolute angular speed of<br />
the block satellite, relative angular speed of the block satellite, angular speed on<br />
the output shaft. The received results are shown on Fig. 6.<br />
Fig. 6 – Kinematic calculation of planetary processional transmission.<br />
At the following stage has been executed the kinetostatics analysis with<br />
calculation and simulation of total loadings in gearings.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 141<br />
An important advantage of the processional planetary transmission consists<br />
in the insurance of absolute multiplicity (ε=100%) of the central wheel teeth<br />
and pinion simultaneous gearing. This advantage provides small dimensions<br />
and mass (reduced material consumption) and small cost of the final product.<br />
a. b.<br />
Fig. 7 – Normal forces among the teeth and<br />
their distribution among simultaneously geared teeth.<br />
The vertical line on the diagram (Fig. 7) shows the precession phase<br />
reported symbolically to the gearing time of the pinion tooth with the central<br />
wheel tooth (as reference the time was taken, because as CAE input parameters<br />
the angular ω velocity was introduced). On the diagrams one can see the<br />
distribution of load among the teeth which corresponds to the respective<br />
precession phase (blue colour – distribution of load on the active profile of<br />
central wheel tooth, red colour – per one precession cycle). It is possible to state<br />
from the diagrams that despite the precession phase, the load among teeth is<br />
uniform.<br />
Due to load transmission from the input shaft to the output shaft by a big<br />
number of teeth couples geared simultaneously ⎛ (Z −<br />
⎜ = 4 1)<br />
ε<br />
⎞<br />
⎝ 2 ⎟ the normal<br />
⎠<br />
forces at teeth contact are much smaller as in the classical transmissions with in<br />
volute transmission.<br />
4. Conclusions<br />
1. Taking into account the fact that in processional planetary gearings the<br />
(Z4 − 1)<br />
teeth couples transmit the load simultaneously we can conclude that<br />
2<br />
the bearing capacity of the processional gear is much bigger than that of the<br />
classical in volute gear (in which only 5...7 % of wheel teeth gear<br />
simultaneously).
142 Ion Bostan et al<br />
2. The elaboration of the planetary processional transmission dynamic<br />
sample (CAE) based on developed 3D model allows verification of kineticstatic<br />
parameters previously defined.<br />
Received: January, 15, 2010 Technical University of Moldova<br />
Department of Theory of Mechanisms and Machine Parts<br />
e-mail: dulgheru@mail.utm.md<br />
R E F E R E N C E S<br />
1. B o s t a n I. Precessionnye eredaci s Mnogoparnym Zacepleniem.(Ed.) Ştiinţa, 356p.<br />
1991. ISBN 5-376-01005-8, Chişinău.<br />
2. B o s t a n I., D u l g h e r u V., G r i g o r a ş Ş., Planetary, precessional and<br />
harmonic transmissions, Bucureşti - Chişinău, 1997, 198 p.<br />
3. B o s t a n I., I o n e s c u F l., D u l g h e r u V., A. S o c h i r e a n u., Kinetostatic<br />
Analysis of the Sphere - Spatial Mechanisms by using 3D-Models and -<br />
Simulations. Meridian Ingineresc, Revue of the Technical University of Moldova<br />
and Moldavian Society of Engineers, 2, ISSN 1683-853-X, (2004). pp. 59-61.<br />
4. http://www.solid-dynamics.com (Motion Inventor)<br />
5. www.autodesk.com (Autodesk Inventor).<br />
DEZVOLTAREA INTEGRATĂ CAE A TRANSMISIILOR PRECESIONALE<br />
UTILIZÂND PLATFORMA AUTODESK INVENTOR<br />
(Rezumat)<br />
În lucrare se prezintă modelarea şi simularea transmisiilor planetare precesionale<br />
proiectate în două variante cu raport de transmitere mare şi capacitate portantă ridicată<br />
într-o singură treaptă. Construcţiile transmisiilor precesionale au fost proiectate în softul<br />
Inventor şi simulate ca sisteme multicorp în MotionInventor. Simulările transmisiilor au<br />
permis obţinerea informaţiilor privind poziţia, vitezele, acceleraţiile, traiectoriile<br />
punctelor, forţe şi momente, energii, de asemenea, forţele de contact la contactul dintre<br />
dinţii roţilor dinţate şi ai satelitului şi alte date privind sistemul. Angrenajele a două<br />
transmisii au fost modelate ca sisteme multicorp în SolidDynamics pentru studiul<br />
emisiei de zgomot în timpul lucrului prin simulare, cu posibilitatea extinderii la<br />
contactul dintre dinţi.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCȚII DE MAȘINI<br />
PROPOSALS FOR THE IMPROVEMENT OF<br />
THE FLUIDIC SPIRAL JETMILLS' ACTIVITY<br />
BY<br />
ILEANA FULGA 1 and EUGEN STRAJESCU 2<br />
Abstract. In this paper is presented a documentary study about the particle<br />
size reduction. There are presented the authors' realizations concerning the<br />
optimization of the spiral jetmills produced in the APTM Enterprises, based on a<br />
new mathematical and experimental model of the flows, developed by the authors.<br />
There are presented conclusions about the research directions to be attacked in the<br />
future.<br />
Key words: micronization, spiral jetmills, fluidic mills, optimization.<br />
1. Introduction<br />
The fluidic mills are the single equipments that permit the dry grinding and<br />
the get of an ultra fine granulometry (1-25 μm), indifferent by the product<br />
resistance and by the hardness of the product's crystals.<br />
The grinding is produced by the collision between the particles that are<br />
accelerated at very high speeds in flows and gas jets, or by their impact with<br />
solid surfaces (fixed or mobile) from the interior of the milling room.<br />
We consider important to signalize that the real or as a rough guide speeds of<br />
the entrainment and grinding particles' jets are considered secret and can not be<br />
found in the literature. At this moment exist in the world 7 fabricants that<br />
produce fluidic mills. In fact, there are many societies that commercialize<br />
fluidic mills, but they are procuring the equipments offered from one of the<br />
mentioned fabricants.<br />
All the fluidic mills fabricants consider strictly confidential any data<br />
concerning the sizing and the calculus of these equipments from the point of<br />
view of the fluidic mechanics, refusing to give any theoretical or experimental<br />
value concerning the geometry of the fluidic nozzle the jets' speeds and<br />
particles' speeds or values bound of the fluidic grinding process.
144 Ileana Fulga and Eugen Strajescu<br />
2. Clarifications about the Possibilities to improve the Spiral Jetmills'<br />
Activity, resulted from the Mathematical Model<br />
2.1. The Supersonic Jet's Protection<br />
The construction of the Venturi tubes by where the material is introduced<br />
and of the nozzles from where the compressed air is blowing off represents an<br />
important factor for a good activity of the fluidic spiral jetmill. The super sonic<br />
nozzle (the one by where is insufflated fluid under pressure, realizing important<br />
debits) must have a determined length. The prolongation of the active area of<br />
the nozzle behind the peripheral wall, or nozzles with inclined termination, and<br />
not perpendicular on the axis's direction of the nozzle could realize a<br />
compressibility effect manifested in an opposite sense, because the boost<br />
pressure decrease, the circulation is reduced and the jet is displaced in a<br />
direction more nigh to the tangent.<br />
Of course this kind of position must be considered as "off-set" beside the<br />
desired equilibrium position, considered as "ideal". It could be imposed in that<br />
case the "protection" of the protuberant peak of the nozzle by an aerodynamic<br />
modeling of the peripheral wall of the milling room, or by the practice of some<br />
clefts on the same wall.<br />
2.2. The Geometry of the Supersonic Jet<br />
It is proved that the employment of the convergent - divergent nozzles, at<br />
necessary and effective pressures do not present great advantages in the great<br />
majority of the practical applications.<br />
It is recommended for the reduction of the losing, the construction of jets as<br />
short as possible, with an alignment near the perfection, with the same radial<br />
angle. Because deviations from the alignment between different jets although<br />
minima, create important tourbillions, unequal, in the transversal flux,<br />
discontinuities in the peripheral circulation of the particles and tourbillions in<br />
the central classification vortex.<br />
3. Practical Applications<br />
3.1. The Optimization of the StarJet and CosmoJet Mills<br />
The APTM Enterprise realized a great number of researches in order to be<br />
able to optimize the activity of the produced spiral jetmills, presented in a great<br />
number of scientific papers [1] - [10] and in a doctoral dissertation.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 145<br />
Fig. 1 – The analyzis of the evolution of the APTM spiral jetmills,<br />
considering the application of the process' optimization.<br />
The results presented in this doctoral dissertation were transposed in
146 Ileana Fulga and Eugen Strajescu<br />
Fig. 2 – Example of the optimized technological flux for the fabrication of the<br />
cosmetic micronized powders (make-up, cheek powder).<br />
practice in the enterprise APTM, for a new jetmill family, named "StarJet-<br />
Millennium" (Fig. 1). An optimized technological process is shown in Fig. 3)
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 147<br />
The first three typo dimension of this family (4" - 8" and 12") applied an<br />
important number of the optimization algorithms of the milling process that<br />
form the research object of this research. It results implicitly the fact that these<br />
algorithms (fig. 2) where already verified in practice for moreover micronized<br />
powders categories (that are partly confidential).<br />
Fig. 3 – The example of the spiral jetmills' optimization algorithm applied for<br />
the definition of the parameters recomended in industry and for the qualitative<br />
differentiation of the cosmethic powders.
148 Ileana Fulga and Eugen Strajescu<br />
The effect of the micronization about the dispersed pigments can be<br />
observed the besting the case of the sample of Prussian Blue (fig. 4): being in a<br />
non micronized stadium this pigment presents agglomerates at 100-150 μm,<br />
whiles after the micronization it is to observe a granulometry contained<br />
between 20-25 μm.<br />
Fig. 4 – Pigment (Prusia blue) nonmicronizes = 150 μm (left) and<br />
micronized = 20 μm (right).<br />
Fig. 5 – Example of cosmethic powders micronized with optimization<br />
algorithms; visualization on microscope with phase contrast.<br />
Even without the microscope or another magnifier, from the photos'<br />
observation it is to note the intensification of the chromatic coverage of the<br />
pigment (fig. 2), because after the micronization, the specific surface of the<br />
product is majorized for 2000 - 3000 times, and particles are very uniform from<br />
the point of view of the granulometry.<br />
Another object of the optimization is the getting of a micronization with a<br />
plain classification room, in order to not charge to frequent the mill (fig. 6).
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 149<br />
Fig. 6 – Example for an optimized production for<br />
the filling up of the classificatory.<br />
We can observe the fact that after the micronization, all the materials<br />
are grinded at granulometry about 15-30 μm, with a remarkable<br />
uniformity. We make the assignation that in this paper where presented<br />
only the non confidential elements that show the optimization.<br />
We present (fig. 5) an example of a technological process for the getting of<br />
cosmetic powders micronized with optimization algorithms. In fig. 6 and 7 we<br />
present micronized powders realized with optimization algorithms<br />
(visualization on microscope with phase contrast). We can affirm that the
150 Ileana Fulga and Eugen Strajescu<br />
micronized powders are an especial good quality, representing a revolution in<br />
the cosmetic and pharmaceutical industries.<br />
Fig. 6 – Non micronized cosmethic pigments,<br />
with a granulometry of 120 - 350 μm.<br />
Fig. 7 – Micronized cosmethic pigments, with a granulometry of 15 - 30 μm.<br />
Mills from the family StarJet optimized with algorithms similar with the<br />
presented algorithms were delivered and perfectly work from the year 2002, in<br />
many enterprises for decorative cosmetic products of the societies Avon (USA),<br />
Natura (Brazilia), Belcorp (Columbia), EverBilena (Taiwan), Swan Cosmetics<br />
(Mexic), Cosmetica (Canada), GammaCroma (Italia), Hellenica (Grecia),<br />
GiPiccos (Italia), etc.<br />
The up presented algorithms, partially used in production, ask for verifications<br />
and ulterior confirmations, forming the object of some future studies.<br />
4. Conclusions<br />
1. The research from what doctoral dissertation introduced an original threedimensional<br />
model of the fluidic running in the spiral jetmills that describes<br />
very veridical the complex jet in this kind of mills.<br />
2. The results are obtained as suit of a very large gamut of theoretical and<br />
experimental researches for the determination of the jets' speed.<br />
3. The optimization algorithms were verified on models and in realty on 37<br />
mills having different typo dimensions, in many fields (cosmetic,<br />
pharmaceutics, and metallurgies) in above 20 countries.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 151<br />
4. From the experimental researches it was obtained important conclusions<br />
that permit to optimize the spiral jetmills in at least two directions.<br />
5. The first one is the micronization operation conditions and the best<br />
geometry of the mill.<br />
6. A second one is the optimization of the operation conditions in order to<br />
realize the filling of the classification part of the mill. In this case, the<br />
productivity of the mill increases with important values.<br />
7. The researches for the optimization in the case of the spiral jet mills must<br />
go on,, because we appreciate that it exists a lot of unexplored ways and<br />
possibilities.<br />
Acknowledgements. The authors would like to thank APTM Enterprises for the<br />
support and constructive comments and also to the staff of the Laboratory of APTM.<br />
Received: March 20, 2010 1 APTM Enterprises,<br />
Lugano, Swiss<br />
e-mail: i@fulga.com<br />
2 <strong>POLITEHNIC</strong>A University of Bucharest,<br />
Department of Machines and Production Systems<br />
Bucharest, Romania<br />
e-mail: eugen_strajescu@yahoo.com<br />
R E F E R E N C E S<br />
1. Fulga I., Huddleston K., Cosmetics cross an even finer line, Powder<br />
Reporter, vol. 1, UK, 1998.<br />
2. Fulga I., Stră jescu E., Sandu I. Gh., Experimental Model of the fluidic<br />
Mills with Spiral Jets, Academic Journal of Manufacturing Engineering, Editura<br />
Politehnica, Timişoara, România, 53 - 57, ISSN 1583 – 7904 (2006).<br />
3. Fulga I., Stră jescu E., Theoretical and Experimental Contributions<br />
Concerning the Logistics of the Research Systems of the Fluidic Mills with Spiral<br />
Jets, Buletinul Institutului Politehnic, Iaşi, România, Construcţii de Maşini, LII<br />
(LVII) , 5, 313 - 320, ISSN 1011 - 2855 (2007).<br />
4. S t r ă jescu E., Fulga I., The Visualization of the Fluidic Fluxes in the Mills<br />
with Jets, Industrial Engineering Journal RECENT, 8, 3b (21b), 361 - 366, ISSN<br />
1582 - 0246. Braşov, România (2007).
152 Ileana Fulga and Eugen Strajescu<br />
5. F u l g a I . , S t r ă jescu E., Contributions to the Fluid Dynamics in Jetmills by a<br />
New Aerodinamic Model of Supersonic Flow, Buletinul Ştiinţific U.P.B., Serie<br />
D, 74, 1, ISSN 1454 - 2358, România (2008).<br />
6. Fulga I., Gadonna J. P., Remolue M., Froidevaux L., Ajana<br />
A., Chayvialle L., Improvements in MDI technology with total process<br />
containment, European Aerosol Conference, EAC, Belgium, 2005.<br />
7. F u l g a I . , S t r ă jescu E., Probleme actuale şi de perspectivå în construcţia<br />
utilajelor pentru măcinare şi în tehno-logia măcinării, A XI-a Conferinţă<br />
ştiinţifică cu participare internaţională TEHNOMUS XIII, <strong>Universitatea</strong> "Ştefan<br />
cel Mare" Suceava, România, Facultatea de Inginerie Mecanică, pg. 94-103,<br />
ISBN 973-666-154-7, 2005.<br />
8. F u l g a I . , S t r ă jescu E. - The Present Stage and the Perspectives of Researches<br />
in the Area of Fluid Grinding Operation Equipment with Spiral Jets,<br />
International Conference on Manufacturing Systems, Buletinul Institutului<br />
Politehnic Iaşi, România, secţia Construcţii de maşini, LI (LV), 5, ISSN 1582-<br />
6392 (2005).<br />
9. F u l g a I . , S t r ă jescu E. - Contributions Concerning the Perfecting of the<br />
Crushing Technologies of the Cosmetic Particles and of the Metallic Carbides in<br />
the Wheels with Fluid Jets, International Conference “Technologies and Quality<br />
for Sustained Developement”, pg. 357 - 360, ISBN 973-720-035-7, 2006.<br />
10. F u l g a I . , S t r ă jescu E. - Experimental Model of the fluidic Mills with Spiral<br />
Jets. (Long Abstract), The 3rd International Conference on Manufacturing<br />
Science and Education, MSE, Sibiu, România, pg. 79 - 80, ISSN 1843 - 2522,<br />
2007.<br />
PROPUNERI DE OPTIMIZARE A FUNCȚIONĂRII<br />
MORILOR FLUIDICE CU JETURI ÎN SPIRALĂ<br />
(Rezumat)<br />
În lucrare se prezintă succint concluziile unui studiu anterior asupra realizării unui<br />
model matematic privind comportarea jeturilor supersonice ale morilor fluidice cu jeturi<br />
în spirală. Pe baza lor ,se enunţă unele posibilităţi de optimizare a funcţionării acestor<br />
tipuri de mori, precum şi rezultate comparative ale măcinării unor pulberi cosmetice.<br />
Rezultă din lucrare şi faptul că micronizarea produselor simplifică substanţial<br />
tehnologiile de măcinare, ceea ce micşorează costurile producerii pulberilor de orice fel.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
DYNAMIC ANALYSIS OF MECHANICAL-HYDRAULIC<br />
PROTECTION MECHANISMS OF LOW POWER HORIZONTAL<br />
AXIS WIND TURBINES THROUGH VERTICAL TILTING<br />
BY<br />
DORU CĂLĂRAŞU, IRINA TIŢA, DAN SCURTU<br />
and BOGDAN CIOBANU<br />
Abstract. Generally, the control system a wind turbine is equipped with is<br />
designed to reduce dynamic load on its blades, to increase turbine reliability and<br />
operational safety. The paper present the dynamic analysis of a mechanicalhydraulic<br />
protection mechanism used for low power HAWT. The method consists<br />
of the vertical tilting of the wind turbine. The analysis of the system responses<br />
revealed a good dynamic performance when using the vertical tilting solution<br />
chosen to protect low power HAWT.<br />
Key words: HAWT, protection mechanism, vertical tilting.<br />
1. Introduction<br />
The control system a wind turbine is equipped with is designed to reduce<br />
dynamic load on its blades, to increase turbine reliability and operational safety.<br />
This system is one of the most important developments meant to consistently<br />
improve low power horizontal axis wind turbines.<br />
Further to the analysis of control and protection system design models used<br />
for low power horizontal axis wind turbines (HAWT), the best constructive<br />
solution was chosen, which consists of the vertical tilting of the wind turbine.<br />
The tilting begins when the wind speed exceeds the speed limit beyond which<br />
the turbine safety may be jeopardized. This solution meets the system<br />
requirements: it is reliable, it does not include electrical drive components, it<br />
has no built-in electronic devices and it does not require the existence of a<br />
hydraulic drive system.
154 Doru Calarasu et al.<br />
2. Constructive-Operational Description<br />
A wind power plant turns wind energy into mechanical power. The chosen<br />
solution refers to very low power (500 – 1000 W) wind turbines.<br />
When the wind speed does not exceed a certain speed limit, the rotor<br />
assembly does not tilt. The propeller rotates at an angular speed, which depends<br />
on the wind speed and on the mechanical output load.<br />
When the wind speed reaches or exceeds the speed limit, the rotor assembly<br />
tilts and the propeller changes its rotation plane.<br />
HAWT<br />
Damping<br />
device<br />
Protection<br />
mechanism<br />
Tilting<br />
device<br />
Fig. 1 – Constructive block diagram.<br />
Balancing<br />
device<br />
Thanks to its components (Fig. 1), this mechanical-hydraulic protection<br />
mechanism also ensures the operation of the wind turbine at wind speeds higher<br />
than the rating speed, having the following functions:<br />
a) the rotor assembly is directed towards the wind stream using the<br />
mechanism’s lower bearing, by means of which the assembly is placed on the<br />
support pillar;<br />
b) the tilting device becomes active when the wind speed exceeds the safety<br />
limit and it reduces the active turbine surface exposed to the wind;<br />
c) the balancing device ensures stable balance to the wind turbine rotor at wind<br />
speeds lower than the safety limit and unstable balance when the wind speed<br />
reaches the safety limit;<br />
d) the damping device protects the plant against wind gusts by delaying the<br />
propeller’s resuming its initial position.<br />
3. Digital Simulation Tests performed on the Conceptual Model of<br />
Mechanical-Hydraulic Protection Mechanisms<br />
The Matlab-Simulink programming environment was used for the<br />
performance of the digital simulations.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 155<br />
Fig. 2 shows the Simulink block diagram of the vertical tilting protection<br />
mechanism.<br />
Body1<br />
CS2 CS1<br />
Scope 2<br />
F<br />
Signal 1<br />
Signal 2<br />
Signal 3<br />
Signal Builder 1<br />
Revolute 1<br />
16<br />
lambda /\r<br />
B<br />
Body Actuator<br />
Joint Initial Condition<br />
Body<br />
CS1<br />
F<br />
Weld<br />
Product 3<br />
Product 4<br />
Product 5<br />
CS4<br />
CS1<br />
CS2<br />
B<br />
Link 2<br />
Revolute 3<br />
B<br />
F<br />
Ground 1<br />
Product 2<br />
Scope 1<br />
CS5<br />
CS3<br />
F<br />
Weld2<br />
Joint Spring & Damper Joint Sensor 2<br />
AxRo/2<br />
[1x3]<br />
B<br />
Env<br />
Machine<br />
Ground 2<br />
Environment<br />
[A]<br />
Goto<br />
Product 7<br />
Joint Actuator<br />
Body2<br />
CS1<br />
2<br />
Constant 1<br />
0.3<br />
Constant 4<br />
Integrator<br />
1<br />
s<br />
Product<br />
360<br />
Constant<br />
|u|<br />
Abs<br />
Constant 2<br />
0<br />
Body Actuator1<br />
sin<br />
Trigonometric<br />
Function 1<br />
-C-<br />
l<br />
-K-<br />
Gain<br />
[A]<br />
From<br />
Scope 4<br />
Scope 6<br />
Scope 5<br />
cos<br />
Trigonometric<br />
Function<br />
Scope<br />
Product 1<br />
-C-<br />
3r Constant 3<br />
65<br />
Scope 3<br />
1<br />
u<br />
Math<br />
Function<br />
Fig. 2 – The Simulink block diagram of the vertical tilting protection mechanism.<br />
Fig. 3 – The rotor’s surface exposed to wind depending<br />
on the turbine axis’ angle of inclination.<br />
Product 6
156 Doru Calarasu et al.<br />
The mechanical-hydraulic protection mechanism includes the tilting device,<br />
the balancing device of the wind turbine’s own weight and the two-way<br />
movement damping device. The initial position (meaning a normal vertical<br />
turbine operation) is triggered by a back stop. The model allows testing using a<br />
set of signals to simulate the wind speed evolution.<br />
Through vertical tilting, the rotor’s surface exposed to wind decreases as<br />
shown in Fig. 3.<br />
The arm of the component determining the tilting moment increases with<br />
the increase of the turbine axis’ tilting angle (Fig. 4). Fig. 5 shows the variation<br />
of the moment determined by the rotor’s tilting strength.<br />
Fig. 4 – Variation of the rotor tilting strength arm.<br />
Fig. 5 – Moment of the rotor tilting strength.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 157<br />
Fig. 6 shows the set of signals used to simulate wind speed variation.<br />
Fig. 6 – Signals for model testing.<br />
The second signal is employed to achieve static balance, where the wind<br />
speed is thought to have the same value as the maximum admissible rating<br />
speed.<br />
Using the ramp signal (signal 3), the system response is tested, that is the<br />
rotor tilting, when the wind speed exceeds the maximum admissible rating<br />
speed. Fig. 7 shows the axis inclination angle variation at the 3 rd signal.<br />
Fig. 7 – Axis inclination angle at wind Fig. 8 – Tilting strength at wind<br />
ramp speed. ramp speed.<br />
Figure 8 shows the tilting strength variation in the conditions described<br />
above.
158 Doru Calarasu et al.<br />
Fig. 9 – Inclination angle of the turbine Fig. 10 – Rotor tilting strength.<br />
rotor axis.<br />
If the system constructive-operational parameters are changed (the damping<br />
coefficient and the return strength decrease), figure 9 shows the evolution of the<br />
inclination angle of the wind turbine rotor axis, while figure 10 depicts the<br />
evolution of the tilting strength.<br />
4. Conclusions<br />
1. The simulation diagram allows both constructive and operational system<br />
parameters and operation testing signals to be altered.<br />
2. Digital simulation may be used to optimize constructive parameters<br />
(balancing components weight, rotor axis section lengths, balancing weight,<br />
damping system characteristics).<br />
3. The analysis of the system responses revealed a good dynamic<br />
performance when using the vertical tilting solution chosen to protect low<br />
power horizontal axis wind turbines.<br />
Acknowledgements. This work has been supported by the Romanian National Fund<br />
for Science and Research through National Centre for Programme Management under<br />
contract No. 21-047/2007.<br />
Received: April 16, 2010 “Gheorghe Asachi” Technical University,<br />
Department of Fluid Mechanics<br />
Iasi, Romania<br />
e-mail: dorucalarasu@yahoo.com
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 159<br />
R E F E R E N C E S<br />
1. B a n s a l R. C., B h a t t i T. S., K o t h a r i D. P., On some of the design aspects<br />
of wind energy conversion systems, Energy Conversion and Management, 43, 16,<br />
November 2002, pp. 2175-2187<br />
2. B i a n c h i F. D., D e B a t t i s t a H., M a n t z R. J., Wind Turbine Control<br />
Systems, Principles, Modelling and Gain Scheduling Design. Springer Verlag,<br />
Heidelberg, 2007.<br />
3. B i a n c h i F., M a n t z R., C h r i s t i a n s e n C., Power regulation in pitchcontrolled<br />
variable-speed WECS above rated wind speed. Renewable Energy<br />
29(11), 1911–1922, 2004.<br />
4. G r i m b l e M. J., J o h n s o n M. A., Advances in industrial control, Springer –<br />
Verlag London Limited<br />
5. M o l e n a a r D. P., Cost-effective design and operation of variable speed wind<br />
turbines, Delft University Press, 2003.<br />
6. N o v a c P., E k e l u n d T., J o v i k I., S c h m i d b a u e r B., Modeling and<br />
control of variable-speed wind-turbine drive-system dynamics. IEEE Control<br />
Systems, pp. 28 - 38, August, 1995.<br />
7. * * * 160 kW Wind turbine with Hydraulic Transmission at Schiedam Ref: 16013 or<br />
34127<br />
8. * * * Low-wind turbine with hydraulic blade control. Measuring program included<br />
Ref: 8700106<br />
9. * * * Wind turbine with hydraulic transmission, Patent number WO03098037.<br />
10. * * * Water current turbine, Patent number WO2005103484.<br />
11. * * * MATLAB Simulink http://www.mathworks.com/products/simulink/<br />
12. * * * Contract no. 21-047/2007 financed by National Centre for Programme<br />
Management from Romania.<br />
ANALIZA <strong>DIN</strong>AMICĂ A MECANISMULUI MECANO-HIDRAULIC DE<br />
PROTECŢIE A TURBINELOR EOLIENE CU AX ORIZONTAL DE MICĂ PUTERE<br />
PRIN BASCULARE ÎN PLAN VERTICAL<br />
(Rezumat)<br />
Sistemul de control cu care este prevăzută o turbină eoliană are rolul de a diminua<br />
încărcările dinamice pe pale, de a creşte fiabilitatea turbinei şi siguranţa în exploatare.<br />
Acest sistem reprezintă una dintre direcţiile prioritare de acţiune în vederea<br />
rentabilizării turbinelor eoliene cu ax orizontal de mică putere.<br />
Lucrarea prezintă analiza dinamică a unui mecanism mecano-hidraulic de protecţie<br />
a turbinelor eoliene cu ax orizontal (TEAO) de mică putere prin bascularea turbinei<br />
eoliene în plan vertical. Bascularea începe atunci când viteza vântului depăşeşte viteza
160 Doru Calarasu et al.<br />
considerată limită pentru siguranţa turbinei. Această soluţie întruneşte caracteristicile<br />
solicitate sistemului: este fiabil, nu conţine elemente de acţionare electrică, nu conţine<br />
electronică înglobată, nu presupune existenţa unei transmisii hidraulice.<br />
Pentru viteze ale vântului mai mici de o limită maximă impusă, ansamblul rotor nu<br />
basculează. Elicea se roteşte cu o viteză unghiulară care depinde de viteza vântului şi de<br />
încărcarea mecanică la ieşire. Pentru viteze mai mari sau egale cu cea maximă impusă,<br />
ansamblul rotor basculează iar elicea îşi modifică planul de rotaţie.<br />
Testarea prin simulări numerice a modelului conceptual de mecanism mecanohidraulic<br />
protecţie a TEAO de mică putere prin basculare în plan vertical s-a efectuat<br />
utilizând mediul de programare Matlab-Simulink.<br />
Din analiza răspunsurilor sistemului rezultă performanţe dinamice bune pentru<br />
soluţia basculării în plan vertical utilizată în scopul protecţiei turbinelor eoliene cu ax<br />
orizontal de mică putere.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
RESEARCH ON HYDRAULICALLY SYSTEMS WHICH<br />
MOVE HEAVY MASSES ON SMALL DISTANCES<br />
WITH LOWER FREQUENCIES<br />
BY<br />
CONSTANTIN CHIRIȚĂ, ADRIAN HANGANU<br />
and DANIEL CALFA<br />
Abstract. In this paper the authors try to present the mathematical models,<br />
static and dynamic, useful for designing of hydraulically systems which move<br />
heavy masses on small distances with lover frequencies. In the last time<br />
mathematical modeling represents an obligatory step in the designing process of<br />
the new products. In this paper are presented the results of the theoretical<br />
simulation in concordance with the experimental results. The aim is to show the<br />
realistic part of the theoretical simulation of one system designed to move a heavy<br />
owen, 200 t, on a short distances, 50 mm, with low frequency, one complete cycle<br />
in 24 hours.<br />
Key words: research, modeling, simulation, data acquisition.<br />
1. Introduction<br />
To move heavy masses on small distances with lover frequencies the author<br />
recommends one of the following hydraulic charts, see Fig. 1.<br />
In both variants the working tool is the linear hydraulic engine CH, with the<br />
active surface S and volume Vo, which moves the mass M.<br />
For the first variant the step of the move is programmed at the level of the<br />
pump, through the volume/course V. We will admit that for a complete course<br />
the pump send to the hydraulic engine the oil volume V. The oil rich the pump<br />
through the orifices A and B.<br />
In the case of the second variant the flow regulator RD controlled the oil<br />
flow, independent to the working pressure for a specific time, when the<br />
distributing valve is on. The quantity of oil which enters in to the linear<br />
hydraulic engine is controlled in the time by the time stepping system C (t).
162 Constantin Chirita et al.<br />
a) b)<br />
Fig. 1 – Hydraulic charts for lower frequencies and small flow generators.<br />
The mathematical models presented in the following rows are equal satisfied<br />
for both variants.<br />
2. Mathematical Models for the Stationary Regime<br />
For the engine CH which works on the strength F, when a small quantity of<br />
oil V, ideal liquid, incompressible, theoretical we obtain a theoretical movement<br />
of the mass, xT, with the mathematical expression:<br />
(1)<br />
V<br />
xT = .<br />
S<br />
For a real liquid, with the elasticity module E, we obtain a real movement, in<br />
static regime, xP.<br />
⎛ p ⎞<br />
(2) xR = xT<br />
⋅⎜1<br />
− ⎟ .<br />
⎝ E ⎠<br />
(3)<br />
In the relation (2) p is the balance pressure.<br />
F<br />
p = .<br />
S<br />
For V = 8 cm 3 , S = 80 cm 2 , F= 32.000 N, E = 1,5e9 N/m 2 , the results are:<br />
xT= 1 mm, xR= 0,999 mm.
(4)<br />
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 163<br />
3. Dynamic Mathematical Models<br />
For both systems from Fig. 1 we can use the following model:<br />
V0<br />
Q = S ⋅ v + a ⋅ p + ⋅<br />
E<br />
dp<br />
dt<br />
dv<br />
(5) M ⋅ + b ⋅ v + F = p ⋅ S .<br />
dt<br />
(6) V = ∫Q ⋅ dt .<br />
(7) ∫ ⋅ = dt v .<br />
x RD<br />
In the last relations we have the following supplementary parameters: v-<br />
instant speed of the CH engine, a- linear coefficient of the lost oil which are in<br />
direct ratio with pressure, t-time, xRD-real movement in dynamic regime.<br />
To find the real movement, in dynamic regime, we must simulate the<br />
equation on a computer program. We consider the same data like in the case of<br />
the static regime, in advance we know M = 200 kg and the values of a and b<br />
coefficients. When the V quantity of oil is inserted we presume that the system<br />
is in balance. The time which the system is in charge with the V oil quantity is<br />
established as well as we obtain a movement theoretical equal with that we have<br />
on the previous step (t = 2 s). The simulating chart in the simplified way is<br />
presented in Fig. 2.<br />
Fig. 2.a – The simulating chart.<br />
.
164 Constantin Chirita et al.<br />
The V part generates the oil flow which is introduced in the linear hydraulic<br />
oil CH. We can look on the evolution of the following parameters: pressure p,<br />
speed v, movement x.<br />
We can observe that the speed is stable around the value of 0,003 mm/ min<br />
and the final value of XRD is 0,995 mm.<br />
Fig. 3 – Data acquisition application in Lab View.<br />
The feedback time of the system is very short approximately 0,003 s, as we<br />
can see in Fig. 2. After the collecting data procedure from the experiment,<br />
where was used a LabVIEW interface with NI DAQ 6024, for the speed and<br />
movement we obtain the following characteristics, see Figs. 3, 4 and 5.<br />
Fig. 4 – Piston movement after a period of nr. of steps generated by the pump.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 165<br />
Fig. 5 – The real speed of the piston in mm/minutes.<br />
Regarding the system is permanent under the pressure his dynamics is very<br />
well. The system is a stabile one first order system.<br />
The dynamic simulation, are we expected, is the most precise one, regarding<br />
the fact that he take care of the specific oil gaps.<br />
4. Conclusions<br />
1. Regarding the system is permanent under the pressure his dynamics is<br />
very well. The system is a stabile one first order system.<br />
2. The aim is to show the realistic part of the theoretical simulation of one<br />
system designed to move a heavy owen, 200 t, on a short distances, 50 mm,<br />
with low frequency, one complete cycle in 24 hours for Mittal Steel - Galati.<br />
Received: February 25, 2010 ”Gheorghe Asachi” Technical University,<br />
DISAHP Department<br />
Iasi, Romania<br />
e-mail: disahp@gmail.com
166 Constantin Chirita et al.<br />
R E F E R E N C E S<br />
1. C h i r i ț ă C., D a m i a n L., H a n g a n u A.-C., C a l f a D., Împingător de<br />
translaţie pentru cuptor rotativ de calcinare. Propunere de brevet de invenţie, nr.<br />
A00317/ 10.05, 2007.<br />
2. C h i r i ţ ă, C., H a n g a n u, A.-C., Modelation and Simulation of Hydraulically<br />
Systems which move Heavy Masses on Small distances with Lower Frequencies,<br />
International Scientific-Technical Conference HYDRAULIC AND<br />
PNEUMATICS 2007, Wroclaw, Poland ,October, 10-12, 2007, pp. 404-408.<br />
3. P r o d a n D., Hidraulica maşinilor-unelte , Editura Printech, Bucureşti, 2004.<br />
4. P r. o d a n D., Maşini-unelte, modelarea şi simularea elementelor şi sistemelor<br />
hidrostatice, Editura Printech, București, 2006, pp. 210-218.<br />
5. * * * LabVIEW 7.2, 2006.<br />
6. * * * http://www.hydramold.com.<br />
7. * * * http://www.ni.com.<br />
CERCETĂRI PRIVIND SISTEMELE HIDRAULICE PENTRU DEPLASAREA<br />
MASELOR MARI PE DISTANTE MICI, CU FRECVENŢĂ REDUSĂ<br />
(Rezumat)<br />
În lucrare autorii prezintă cercetările efectuate în cadrul DISAHP pentru proiectarea<br />
sistemelor hidraulice care deplasează mase grele pe distante mici, cu frecvenţă redusă.<br />
Sunt evidențiate rezultatele simulării în concordanţă cu rezultatele experimentale la<br />
deplasarea unei sarcini grele de 200 t, pe distanţe scurtă 50 mm, cu frecvenţă joasă, întrun<br />
un ciclu complet de 24 de ore, pentru beneficiarul Mittal Steel din Galați.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
PNEUMATIC METERING SYSTEM FOR AMOUNT<br />
OF WATER EXTRACTED IN CONVECTIVE<br />
DRYING PROCESSES<br />
BY<br />
EROL MURAD 1 , CĂTALIN DUMITRESCU 2 ,<br />
GEORGETA HARAGA 1 and LILIANA DUMITRESCU 2<br />
Abstract. Increased decentralization of drying processes for agricultural<br />
products require reliable and easy handling installation at low costs. The<br />
economic efficiency of these plants depends on the use of optimal management<br />
for processes that require measurement of the mass of water extracted, in the<br />
thermal conditions of drying chambers. We developed a new scheme for<br />
measuring on-line the weight of tray loaded with dry material using a pneumatic<br />
force transducer working under sampled measurement. The solution presented is<br />
remarkable due to very low energy consumption and low cost automation feature.<br />
The measurement system is coupled to PLC for automatic management of driers.<br />
Dynamic behaviour and energy consumption were determined by simulation<br />
experiments performed with a model and a numerical simulation program,<br />
developed in simulation environment MEDSIMFP10. Simulation experiments<br />
have confirmed low energy consumption air and good measurement accuracy.<br />
Keywords: pneumatic transducer, force, drying, energy consumption, lowcost<br />
1. General considerations<br />
Drying is an effective, economic and ecological method for preserving<br />
agricultural products, plants, fruits and vegetables. An optimal drying process<br />
led to products with high nutritional potential, obtained with reduced specific<br />
energy consumption.<br />
For dryers with a capacity smaller than 20 m 2 surface of tray, the price of<br />
automatic driving system can double the cost of installation and thus leads to<br />
reduced automatic management functions, with impact on the quality of final<br />
products. These dryers can be easily transported to the place where is performed
168 Erol Murad et al.<br />
harvest of products which require drying; for this reason it is necessary a high<br />
level of energy independence and high reliability. [ 5] - [7], [9], [10].<br />
For optimizing the drying process is still necessary to measure the mass of<br />
water extracted during the drying process. In order to reduce production costs of<br />
the dryer, the measurement is performed for a sample represented by one or<br />
more boxes of dry suspended on a force transducer.<br />
To simplify design, transducer should be arranged in the drying chamber at<br />
high temperatures and high humidity. In papers [5], [6], [9] have been examined<br />
these issues and therefore was developed a pneumatic force transducer,<br />
unconventional, which is adapted to operating conditions of convective dryers<br />
and requires only 63mW electrical power and 5mW power air power. [12] - [14]<br />
Optimal management of drying processes requires online measurement of<br />
the variation of water extracted from the dried bodies during a drying.<br />
Constructive solution for measuring the change in weight of a column of tray is<br />
technically complicated and expensive. Hence, it is used a weight measurement<br />
of the change in middle column boxes, specifying that the drying process has a<br />
similar evolution throughout the column height. It should be noted that the<br />
temperature in the drying is in the range 50...80°C, therefore, was decided to<br />
use pneumatic sensors for measuring forces, because measured size, pressure, is<br />
not influenced by changes of temperature. [4] - [6]<br />
Convective dryers consume more thermal power and about 8% electrical<br />
power, required for operating the fan and automation system. [8], [10]<br />
This paper presents and analyzes the operation and performance of a<br />
scheme for measuring on-line weight of boxes loaded with dried material using<br />
a pneumatic force transducer working under sampled. The solution presented is<br />
remarkable due to very low energy consumption and a reduced price. The<br />
measurement system is coupled to dedicated leadership PLC automatic dryers.<br />
They use basic concepts of low cost automation, minimize energy<br />
consumption and not least are extremely efficient economically.<br />
2. Pneumatic Force Transducer<br />
Applying concepts of minimizing energy consumption, energy<br />
independence and low cost automation system was designed an unconventional<br />
pneumatic transducer which does not require special source of compressed air<br />
consumption is very low, which simplifies construction, and hence the cost<br />
transducer, and the total energy consumed is very small.<br />
It was chosen a pneumatic system because of the pressure signal, that is<br />
proportional to the measured weight, which is not influenced by environment<br />
temperature of measurement, and accuracy is high.<br />
The main element of the force transducer is pneumatic load cell, which<br />
converts measured force Fmas in a force proportional pressure pmas.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 169<br />
Figure 1 shows a functional diagram for pneumatic load cell for the drying<br />
process, which can be used for proving that force change occurs slowly and<br />
processes take place in a single direction.<br />
Fig. 1 – Functional diagram of pneumatic force transducer.<br />
Force measuring Fmas(t) is applied on top of a rod (4) attached to the rigid<br />
center (3) of an embossed flexible membrane (2) with effective diameter Def<br />
and constant effective area Sef, mounted on a body (1). On the rod (4) is<br />
attached a nozzle (5), with the diameter dd, which is based on a ball (6) with<br />
diameter db, closing the chamber as air access to the outside through holes in the<br />
membrane, rigid centre and rod.<br />
The measurement chamber can be connected in parallel with an air capacity<br />
Vad. Pneumatic circuit supply is done by RP variable air resistance and DP<br />
distributor type 2x2. Pressure source must be pal ≥ 1,5 pmas max.<br />
Pressure measured pmas (t) is applied to a converter p / U, which gives the<br />
output voltage yF∈[1, 3] Vdc.<br />
The steady, the balance of forces, pressure pmas (t) in the enclosure is:<br />
(1)<br />
p<br />
F<br />
( t)<br />
+ G<br />
mas em<br />
mas ( t)<br />
=<br />
,<br />
S ef<br />
where: Gem is the weight of mobile equipment, which is constant.<br />
For measuring a force with some variation, included in the measurement,<br />
the supply with compressed gas is sampled with an u1 order form, short pulse,<br />
which provides increased pressure pmas to Fmas balance; then the rigid centre,<br />
membrane and nozzle amount of the ball with h(t) and excess flow gas Dev(t) is<br />
discharged outside.<br />
For measuring a force with a change in one respect, specific processes of<br />
drying, the weight of bodies during drying decreases continuously, it is<br />
necessary to supply compressed gas only at the beginning of drying burden.<br />
During drying Fmas decreases continuously, pmas decreases continuously, and<br />
additional gas is discharged outside through the space between the gasket and<br />
valve, to maintain balance as it shows the Eq. (1).
170 Erol Murad et al.<br />
This method of measurement consumes a very small amount of compressed<br />
gas, which means little air power.<br />
Pneumatic output signal of the enclosure pmas is converted into unified<br />
electric signal using a converter P / U.<br />
As the outdoor temperature is constant and the variation of force Fmas(t) is<br />
slow, can be considered, in terms of thermodynamics, the measurement is an<br />
isothermal process at constant volume. Variation of the air mass in the<br />
enclosure depends on pressure variation pmas(t) produced by the variation of<br />
force Fmas(t):<br />
(2)<br />
V<br />
m<br />
ma<br />
( t)<br />
RaTa<br />
+ Vad<br />
= .<br />
p ( t)<br />
The position of dynamic equilibrium of cell phone equipment load, if the air<br />
distributor DP is closed, the volume of gas is constant Vm. Therefore,<br />
thermodynamic equilibrium equation is:<br />
(3) p mas ( t)<br />
( Vm<br />
+ Vad<br />
) = ma<br />
( t)<br />
RaTa<br />
,<br />
hence the differential equation of the dynamic behaviour of load cell:<br />
dp ( t)<br />
=<br />
dt<br />
mas<br />
a<br />
(4) ( Vm<br />
+ Vad<br />
) RaTa<br />
,<br />
and<br />
(5)<br />
mas<br />
a ( t)<br />
⎛ V V ⎞<br />
⎜ m + ad dF<br />
= ⎟<br />
dt ⎜ Sef<br />
RaT<br />
⎟<br />
⎝<br />
a ⎠ dt<br />
dm mas<br />
dm ( t)<br />
dt<br />
When the force produced by the membrane exceeds the loading:<br />
(6) ( mas ( t)<br />
⋅ Sef<br />
− Fmas<br />
( t)<br />
− Gem<br />
) > 0<br />
p ,<br />
is entering a phase characterized by acceleration a(t) that makes lifting the<br />
nozzle from the ball to h(t):<br />
(7) a( t)<br />
( pmas<br />
( t)<br />
⋅ Sef<br />
− Fmas<br />
( t)<br />
− Gem<br />
) / M red ( t)<br />
= .<br />
Mred low mass value (t) at the transducer rod is:<br />
.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 171<br />
1<br />
(8) M red ( t)<br />
= ( Fmas<br />
( t)<br />
+ Gem<br />
) ) .<br />
g<br />
Dmev exhaust mass flow (t) depends on the distance nozzle-ball h(t), the<br />
pmas(t) and external pressure patm. Sev exhaust section (h, db, dd) is a truncated<br />
cone with the tip in the centre of the ball and the base on the nozzle [3], where<br />
Rh is hydraulic radius [3]:<br />
⎛ 2 2 2 2 ⎞<br />
(9) R<br />
h<br />
= 0 . 5 ⋅ ⎜ h + h d<br />
b<br />
− d<br />
d<br />
+ 0.<br />
25 ⋅ d<br />
b<br />
− 0.<br />
5d<br />
b ⎟<br />
⎝<br />
⎠<br />
The kev rate of contraction of the jet exhaust is calculated from a<br />
relationship F(Re, Rh, Ra, Ta, pmas, patm) [3].<br />
If Fmas(t) decreases continuously, so in drying processes, change dma(t)/dt<br />
will be negative; over time, the gas is discharged outside for reaching a value<br />
pmas(t) with which is balanced Fmas(t).<br />
Under these conditions, mass balance for air in the enclosure is:<br />
dma<br />
( t)<br />
(10) Dmi<br />
( t)<br />
− Dmev<br />
( t)<br />
dt<br />
= .<br />
It was made a model and a simulation program of this type of transducer<br />
operation in MEDSIMFP.10 simulation environment. Figure 2 presents<br />
transducer operation for a sampling period starting sequence and in Figs 3 and 4<br />
are shown different working arrangements, with a simulation of 300 s.<br />
3. Measurement Algorithm<br />
For reducing energy consumption and air power, it was designed an<br />
algorithm to drive transducer operation under sampled with a period Tes ∈ (10<br />
... 100) s.<br />
Reading the transducer voltage output yg is made in PLC with a frequency<br />
fy ≥ 100 Hz.<br />
At the beginning of the sample, it is given a prompt ui = 1 which opens the<br />
DP and the through RP entry pass flow Dmi (t), that results in increased air mass<br />
ma (t) and default and pmas (t), see Figure 2.<br />
At the beginning of the sample, it gives a command ui = 1 which opens the<br />
DP and through the RP entry pass flow Dmi(t), which results in increased air<br />
mass ma(t) and hence the pmas (t), see Figure 2.
172 Erol Murad et al.<br />
When the condition (6) is accomplished, the nozzle starts to rise and there is<br />
an exhaust flow Dmev (t)> 0, which reduces ma(t) and pmas (t) begins to decrease.<br />
Once it finds a stabilizing amount pmas (t), driving algorithm keeps ui = 1 still<br />
400 ms, then ui = 0 and stops the air supply Dmi (t) = 0. In this period Dmev(t)> 0<br />
until h(t) = 0.<br />
Fig. 2 – Measurement scheme starting sequence.<br />
Fig. 3 – Experiment simulation pneumatic transducer Fmas (t) = const.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 173<br />
Fig. 4 – Experiment simulation pneumatic transducer dFmas(t) / dt
174 Erol Murad et al.<br />
Figure 4 presents a simulated experiment where the measured force<br />
decreases continuously and slowly. Transducer works well with an air<br />
consumption similar to the previous experiment.<br />
If it is considered that the opening between two orders of distributor DP,<br />
pressure may decrease too much due leaks and wear, hourly air consumption<br />
would be maxim 135.42 Ncm 3 / h, which is a small volume. For example in<br />
continuous operation for 24 hours would consume 3.25 Ndm 3 air, which would<br />
require a reservoir of 2 dm 3 uploaded once a day to 3 bar.<br />
5. Conclusions<br />
1. We designed a pneumatic force transducer for particular process<br />
consisting in a slow change of the measured force, typical for convective<br />
drying.<br />
2. Transducer operates with very low air energy consumption, is simple,<br />
affordable, sustainable, low-gauge and weight.<br />
3. They were applied mechatronics and automation principles at low cost<br />
for transferring as many functions to PLC management of the technological<br />
process.<br />
4. For exhaust valve nozzle-ball version has been used to ensure normal<br />
operation even in difficult assembly and operation, ensuring a good seal chosen<br />
option and a relatively large exhaust section to small movements.<br />
5. On the dynamic, by controlling the very low input flow , in a chamber<br />
measuring small volume, for reaching a maximum speed of less than 100 mm/s 2<br />
for the mobile part, and a maximum displacement of 0,3 mm. These values<br />
show that there are no shocks during the operation, and sustainability of nozzleball<br />
system is very high.<br />
6. Simulation experiments show that a transducer with head scale of 50 N<br />
can operate continuously 24 hours powered by a 2-liter tank loaded to 3 bar<br />
once a day.<br />
7. The results obtained in simulated experiments represent tests of the<br />
experimental model, which pursue to prove both design and functional<br />
optimized variants functional, through change values for pair ball diameter and<br />
nozzle diameter.<br />
Received: March 20, 2010<br />
1 <strong>POLITEHNIC</strong>A University of Bucharest,<br />
Faculty of Biotechnical Systems<br />
Bucharest, Romania<br />
e-mail: erolmurad@yahoo.com<br />
2 INOE 2000-IHP<br />
Bucharest, Romania<br />
e-mail: dumitrescu.ihp@fluidas.ro
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 175<br />
R E F E R E N C E S<br />
1. A v r a m M., Acţionări hidraulice şi pneumatice, Ed. Universitară, Bucureşti,<br />
2005.<br />
2. F e i d t M. L., Termodinamica şi optimizarea energetică a sistemelor şi<br />
proceselor, Ed. BREN, Bucureşti, 2001.<br />
3. D i m i t r i e v V. N., G r a d e ţ k i i B.G., Osnovâ pnevmoavtomatiki,<br />
Maşinostroienie, Moskva, 1973.<br />
4. M u r a d E., MEDSIMFP10, Mediu software simulare sisteme, Free Pascal-V2.1.4,<br />
U.P.B., 2008.<br />
5. M u r a d E., Măsurarea parametrilor procesului de uscare a produselor ceramice<br />
cu traductoare pneumatice neconvenţionale, HERVEX 2007, Călimăneşti 14-16<br />
noiembrie 2007.<br />
6. M u r a d E., C h e r c h e ş T., Traductoare pneumatice neconvenţionale cu<br />
consum redus de energie pentru măsurarea forţelor din instalaţii agricole şi în<br />
industria alimentară, HERVEX 2008, Călimăneşti 15-17 noiembrie 2008.<br />
7. M u r a d E., H a r a g a G., B u l e a r c ă M., B ă d i l e a n u M., Convection<br />
dryers under cogeneration, Conferinţa Internaţională, ENERGIE - MEDIU<br />
CIEM 2009, UPB, Bucureşti 12-14 noiembrie 2009.<br />
8. M u r a d E., P r e d e s c u C., R i z o i u G., S i m a C., H a r a g a G., Instalaţii de<br />
uscare mobile şi ecologice destinate zonelor montane, HERVEX 2009,<br />
Călimăneşti 18-20 noiembrie 2009.<br />
9. M u r a d E., M a i c a n E., M a r i n A., Du m i t r e s c u C., Dinamica<br />
traductoarelor pneumatice cu consum redus de energie pentru măsurarea<br />
forţelor în procese lente, HERVEX 2008, Călimăneşti 18-20 noiembrie 2009.<br />
10. M u r a d E., C h e re c h e s T., Optimizarea consumului de energie a sistemului de<br />
conducere automată a proceselor de uscare convectivă, HERVEX 2008,<br />
Călimăneşti 18-20 noiembrie 2009.<br />
11. R a d c e n c o V., A l e x a n d r e s c u N., I o n e s c u E., Calculul şi proiectarea<br />
elementelor şi schemelor pneumatice, Editura Tehnică, Bucureşti, 1985.<br />
12. * * * Catalog SMC, 2009.<br />
13. * * * Catalog FESTO, 2009.<br />
14. * * * Catalog Motorola 2008.<br />
SISTEM DE MĂSURARE PNEUMATICĂ A MASEI DE APĂ<br />
EXTRASĂ ÎN PROCESELE DE USCARE CONVECTIVĂ<br />
(Rezumat)<br />
Păstrarea produselor agricole vegetale, fructe şi legume, se poate face în condiţii<br />
foarte bune prin uscarea lor; această soluţie, beneficiind de un condus optimal,<br />
realizează produse cu potenţial nutritiv ridicat, cu consumuri energetice specifice<br />
reduse. În actualul context economic, derularea acestor procese în locaţii
176 Erol Murad et al.<br />
descentralizate, de dimensiuni mici şi medii, necesită instalaţii uşor de deplasat, fiabile<br />
şi cu un preţ redus. Eficienţa economică a acestor instalaţii depinde de utilizarea<br />
conducerii optimale a proceselor de uscare care necesită măsurarea, în condiţiile termice<br />
din camerele de uscare, a masei de apă extrase.<br />
S-a dezvoltat o nouă schemă de măsurare on-line a greutăţii unei casete încărcată cu<br />
material de uscat care utilizează un traductor pneumatic de forţă care lucrează în regim<br />
de măsurare eşantionat. Soluţia prezentată se remarcă prin consumuri foarte mici de<br />
energie şi este caracteristică pentru automatizarea cu cost redus. Sistemul de măsurare<br />
este cuplat la PLC-ul destinat conducerii automate a instalaţiei de uscare.<br />
Comportarea dinamică şi consumurile energetice s-au determinat prin experimente<br />
de simulare realizate cu un model şi un program de simulare numerică, dezvoltat în<br />
mediul de simulare MEDSIMFP10. Experimentele de simulare au confirmat consumuri<br />
reduse de energie pneumatică (un rezervor de 2 l încărcat la 3 bar odată pe zi), precum<br />
şi o precizie bună de măsurare.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
TECHNOLOGIES FOR MONITORING STRUCTURAL<br />
DAMAGES ARISING IN THE FUNCTIONING<br />
OF WIND TURBINES<br />
BY<br />
MIHAI FLORIN MĂNESCU and VALERIU PANAITESCU<br />
Abstract. The need for sustainable development requires the use of renewable<br />
energy, primarily to maintain the existing energy (fossil fuels) to a suitable level,<br />
but also for environmental conservation. Accordingly, renewable energy sources<br />
began to be intensively studied and developed, mainly in the last ten years. In<br />
Romania, the first wind groups were put into service in 2005 (Ploiesti Industrial<br />
Complex - a wind turbine Vestas V52 - 660 kW, relocated), now the installed power<br />
being around 14 MW. For the purposes of harnessing wind energy for electricity<br />
production, because at heights exceeding 100 m, the wind is not influenced by<br />
existing deficiencies ground (trees, buildings, hills, etc..) Wind groups have become<br />
increasingly high and access to review and repair is difficult. As a result of this it is<br />
necessary to monitor the time of damage that may occur during operation of a wind<br />
farm to reduce the time the wind group is stopped and production losses are as small<br />
as possible. To ensure proper function safely it is necessary to implement a<br />
monitoring system for structural resistance (MSR) of wind turbine. This system and<br />
some of the damage assessment methods will be presented in this paper. The MSR<br />
has two branches: a network of sensors that collect information and an algorithm /<br />
software to interpret data from the sensors. Some of the technologies that can be<br />
used, methods of ultrasonic, thermal scanning, x-rays, etc. will be presented in this<br />
paper. Likewise, the most frequent types of structural damage that can occur in a<br />
wind group will also be discussed in this paper.<br />
Key words: wind turbine, monitoring structure resistance, sensors and damage<br />
detection.<br />
1. Introduction<br />
In order to achieve sustainable development, which requires the use of<br />
renewable energy, primarily to maintain the existing energy (fossil fuels) to a<br />
suitable level, but also for environmental conservation, in the last 20 years has
178 Mihai Florin Mănescu and Valeriu Panaitescu<br />
emphasized the efficiency of plants used as feedstock in electricity generation,<br />
renewable sources. Unlike other renewable energy sources, wind energy is<br />
limited because of its technological maturity, good infrastructure and relative<br />
price competitiveness. To have an effective higher cost-benefit, wind generators<br />
groups have become increasingly high.<br />
Some of the problems arising from the operation of wind turbines are listed<br />
below:<br />
a) the areas where wind farms are developed wind regions and which are<br />
usually difficult to reach, such as high mountain areas or areas of Dobrogea, due<br />
to existing infrastructure;<br />
b) the height of pillars (about 100 m), which makes revisions and access<br />
for repairs difficult;<br />
c) there have been accidents that led to the collapse of the entire wind<br />
structure and consequently to enormous losses;<br />
d) the tendency to have a high cost-benefit led to large structures;<br />
e) in Romania there isn’t a mature experience in the production of<br />
electricity using wind power, the first wind turbine was put into service in 2005,<br />
with a pillar height of 54 m. The wind farms that are now developed are usually<br />
100 m high and, in case of damage, this fact can lead to serious accidents, since<br />
action plans may not be effective.<br />
Given the above, it becomes necessary to monitor the whole resistance<br />
structure of wind turbines using a network of sensors.<br />
The monitoring systems are safe and low-cost, and their implementation can<br />
prevent energy losses during revisions or damage. These systems can efficiently<br />
diagnose which parts of the wind turbine must be replaced and they can also<br />
work to reduce duration of a scheduled review. Furthermore, these systems can<br />
predict the life cycle of the plant whose working time averages10-30 years.<br />
According to F a r r a r and S o h n [1] a monitoring system is defined as a<br />
process that detects faults which occur in high buildings. A fault is defined here<br />
as a alteration of the materials and / or of the geometric properties of these<br />
systems, including alterations in bordering conditions and system connectivity,<br />
which adversely affect system performance. There can be many causes of<br />
structural damage such as moisture absorption, fatigue, wind gusts [2], thermal<br />
stress, corrosion, fire and even lightning [3].<br />
In general, the successful development of methods of MSR depends on two<br />
key factors: the technology employed for detection and signal analysis and the<br />
algorithm for interpretation [4]. MSR components consist of a data acquisition<br />
system, data filtering, pattern recognition and decision making. Each of these<br />
components is equally important in determining the safety status of a structure<br />
[5].<br />
The monitoring process involves well-timed observation, using data from a<br />
series of sensors and statistical analysis in determining the status of the<br />
structure. In the case of long-term process MSR, the sensors provide regularly
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 179<br />
updated information on the ability of the structure to perform its intended<br />
function, depending on the inevitable aging process and the degradation caused<br />
by the operation conditions. Ensuing extreme events such as earthquakes, MSR<br />
is used to obtain nearly real-time, reliable information on the integrity of the<br />
structure [1].<br />
The same as wind turbines, the MSR system can provide data on the state of<br />
integrity of the assembly. The damage caused by wind turbines which can be<br />
detected by MSR is listed below, depending on the functional characteristics of<br />
the system. Early warning, with the following advantages and benefits:<br />
a) Prevention of breakdown; better planning of the revisions; reducing the<br />
cost of repairs; minimization of the reparation time.<br />
Problem identification, with the following advantages and benefits:<br />
a) Servicing at the appropriate time; reduction of the number of replaced<br />
parts; solving problems while still in warranty; reduction of the costs; operation<br />
at normal parameters.<br />
Continuous monitoring, with the following advantages and benefits:<br />
a) Constant update on the working parameters; security, less stress.<br />
The benefits of having a detection system faults are: [6], [13].<br />
a) The prevention of premature breakdown: prevention of disasters, failure<br />
and secondary defects.<br />
b) Lower maintenance costs.<br />
c) Remote monitoring and remote diagnosis: turbines are usually built within<br />
a long distance from the premises of the beneficiary.<br />
d) Improvement of the capacity to warn against impending failure; the<br />
duration of the repairing process can be reduced in the windy season and<br />
consequently the production will not be affected.<br />
e) Support for the further development of a turbine: the data which has been<br />
gathered can be used to improve the models of the future turbine generation.<br />
With a reliable MSR, one can successfully plan an efficient maintenance<br />
and repair strategy of the wind turbines, especially offshore. Maintenance and<br />
repair can be performed on demand, when the weather conditions are<br />
favourable. Similarly, the mobilization to cover the costs of personnel, materials<br />
and equipment can be optimized [7].<br />
An ideal monitoring system consists of two main components: a network of<br />
sensors for data collection and an algorithm for data analysis, which is used to<br />
interpret the measurements and determine the physical condition of the<br />
structures.<br />
These methods, which are either already applicable or promise application<br />
to the wind turbine systems in the near future, are presented below.
180 Mihai Florin Mănescu and Valeriu Panaitescu<br />
2. Damage Detection Technologies<br />
2.1. Non-destructive Technologies for Monitoring<br />
2.1.1. Thermal scanning technology. This technology is used to detect<br />
anomalies, based on the differences in the temperature of the inspected areas,<br />
such as the blades of the wind turbine, with the help of infrared sensors or heat<br />
detecting cameras [6]. The differences in temperature are correlated with the<br />
diffusion of heat and, as a result, they indicate abnormalities or damage.<br />
Specialized equipment, called thermovision or thermographic camera and<br />
similar in size and appearance to the very familiar video camera of everyday<br />
life, is used to obtain thermal images. In this way, objects are viewed in terms<br />
of the infrared radiation (IR) emitted by them and not in that of the visible<br />
radiation, which can be effortlessly detected with the naked eye.<br />
The thermographic method is extremely useful in practice because it<br />
permits the achievement of the so-called "predictive maintenance", a term<br />
which is applicable to all industries. The first sign of a defect or an operating<br />
problem is often given by the increased heat in that particular area, therefore an<br />
increase in the emission of infrared radiation. In other cases, the unjustified<br />
lower temperatures of some areas or of some parts of the device may be a sign<br />
of negative phenomena occurring at their level. After having been generated,<br />
the termograma is digitally processed in order to locate the points where heat<br />
stress is present and the defects. It should also be noted that the assessment of<br />
the termograme obtained by scanning the device, system or facility during<br />
proper functioning, can provide extremely valuable information on the normal<br />
thermal map, which will serve as reference for future evaluation and will also<br />
help with the remediation of any potential failure.<br />
Fig. 1 – Image of wind turbine produced by thermography.<br />
Additionally, the thermographic method is a non-invasive technique for<br />
measuring the infrared radiation emitted by the object under examination and<br />
generates, after a very short thermal analysis performed in real time, a thermal<br />
paper which may be sometimes of vital importance to the wind turbines. This<br />
method enables the precise measurement of surface temperature; the high<br />
resolution is a valuable advantage in detecting differences in the temperature of
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 181<br />
those areas which are prone to failure. It can be stated that the method enables<br />
the so-called "defect management", the monitoring of the infrared radiation<br />
emitted by a technical system during normal operation often being the only<br />
imaging method that enables the actual visualization of the notions of "fault"<br />
and "inclination to failure".<br />
Thermography is very often the only quick on-site investigation of a<br />
facility. The areas which are prone to failure can be observed through a mere<br />
scanning of the frontage of the device.<br />
2.1.2. Ultrasonic technology. Ultrasound is applied differently, depending<br />
on the nature, geometry and destination of the product, taking into accounts the<br />
technical performance of the Probes Flaw system. Steel products are the ones<br />
which are most commonly examined by means of the ultrasound technique.<br />
Ultrasonic control can be differentiated and customized according to the<br />
methods specific to steel products: metal, forgings, castings, welded structures<br />
and products made of austenitic steel.<br />
The ultrasound check reveals all the types of internal defects of welded<br />
joints. The ultrasonic method can be used to determine wall thickness and the<br />
number of deposited layers. This technology is applicable to all metals and<br />
nonmetallic materials. Having a great penetrating force, the ultrasound enables<br />
the verification of large sections. The application is only limited by the rough<br />
structures which are highly heterogeneous. The devices are lightweight,<br />
portable and autonomous, and the technology will ensure good results even in<br />
on-site working conditions. The verification techniques, particularly the<br />
immersion control, lend themselves to mechanization and automation.<br />
The result is safe and immediate and it can accurately predict the location,<br />
size and depth of the defects. From an economic standpoint, the ultrasound<br />
method is much cheaper and more productive than the radiation technique in<br />
those cases where the number of defects exceeds a certain limit [14].<br />
2.1.3. X-ray technology. X-rays can penetrate a large number of different<br />
materials, including composites. Usually, the images are obtained as shadow<br />
variations along the X-ray propagation. Under normal operation conditions, the<br />
X-rays are not able to reveal the cracks which have a parallel orientation with<br />
respect to the rays.<br />
The energy requirement is low and the safety prerequisites can be fulfilled<br />
from a remote position. The X-ray sources can be small, air-cooled and easily<br />
implemented for various applications. This method is able to detect the lack of<br />
binder between the metal, as well as cracks and holes in the metal. Some<br />
sources enable the detection of defects at less than 10 micrometres. This method<br />
can be used during real-time inspection for quality control or shortly after the<br />
blade of the wind turbine is set in motion. [13].
182 Mihai Florin Mănescu and Valeriu Panaitescu<br />
2.2. Discussion<br />
The advantage of the thermographic method is that this technology is able<br />
to produce an overall picture of the whole wind farm. A quick assessment can<br />
be made by a user with some experience. The main problem with this method is<br />
that the thermal images are limited to some electrical components that produce<br />
excessive heat during operation. This can also be applied to defective<br />
components if they move excessively, causing friction and heat.<br />
The ultrasonic methods will not detect single fibre breaks or composite<br />
materials. These methods do not work for complex structures (only the blades,<br />
the tower and the nacelle can be monitored).<br />
An advantage of the X-ray technology is that the images are obtained in<br />
parallel, not through scan as is the case with the thermographic approach and,<br />
therefore it is faster. However, problems can appear when the X-ray image is<br />
interpreted. It should be noted that the sources emitting the X-rays are small and<br />
air-cooled, thus fulfilling the safety prerequisites and they can also be controlled<br />
remotely.<br />
3. The Damage of the Structural Integrity of the Wind Power<br />
Impairment can occur in any component or part of the turbine and it can<br />
take any form, ranging from a crack in the concrete foundation to a split blade.<br />
Various cases of structural damage are reported from time to time in several<br />
countries such as Wales, Scotland, Spain, Germany, France, Denmark, Japan<br />
and New Zealand [8]. In Germany, in 2002, a blade broke away, and parts were<br />
found scattered throughout the area [9]. In another case, a blade flew to a<br />
distance of 8 km and entered the window of a house. A detailed documentary of<br />
this accident is available http://www.caithnesswindfarms.co.uk. According to a<br />
survey performed in Germany [10] the frequency of damage is almost equal for<br />
all the mechanical systems and structures. The failure varies from 4% for the<br />
structural parts and the gear to 7% for the rotor blades.<br />
Although structural damage may occur in any component, the most<br />
common type of damage is encountered in rotor blades and in the tower [11].<br />
Special attention is given to the blades, as they are the key elements of a wind<br />
power generation system and as their cost can average 15-20% of the total cost<br />
of the turbine. It was noted that the damage to the blades is the most expensive<br />
type of repair and also the one which lasts the longest [12].<br />
Additionally, even minor damage on the blades can lead to serious<br />
secondary damage to the entire wind turbine system when action to repair it is<br />
not taken immediately, leading to the collapse of the entire set [8].
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 183<br />
The main construction elements of a turbine blade are shown in Fig. 2. The<br />
blades are made of a mixture of fiberglass and composite materials. They are<br />
designed to capture wind energy and transfer it to the turbine rotor and their<br />
profile is the result of complex aerodynamic studies, on which turbine<br />
efficiency is dependant. The development in size of the blades also caused an<br />
increase in the size of the rotor.<br />
Damage to blades may occur in different ways. The most common types of<br />
damage is listed below [5] and a sketch of these is available in Fig. 3.<br />
Typical damage description:<br />
1. Damage formation and growth in the adhesive layer joining the skin and<br />
the main spar flanges (skin/adhesive debonding and/or main<br />
spar/adhesive layer debonding);<br />
2. Damage formation and growth in the adhesive layer joining the up and<br />
downwind skins along leading and/or trailing edges (adhesive joint<br />
failure between skins);<br />
3. Damage formation and growth at the interface between face and core in<br />
sandwich panels in skins and main spar web (sandwich panel face/core<br />
debonding);<br />
4. Internal damage formation and growth in laminates in skin and/or main<br />
spar flanges, under a tensile or compression load (delamination driven<br />
by a tensional or a buckling load);<br />
5. Splitting and fracture of separate fibres in laminates of the skin and main<br />
spar (fibre failure in tension; laminate failure in compression);<br />
6. Buckling of the skin due to damage formation and growth in the bond<br />
between skin and main spar under compressive load (skin/adhesive<br />
debonding induced by buckling, a specific type 1 case);<br />
7. Formation and growth of cracks in the gel-coat; debonding of the gel-coat<br />
from the skin (gel-coat cracking and gel-coat/skin debonding).<br />
Fig. 2 – Components blade.
184 Mihai Florin Mănescu and Valeriu Panaitescu<br />
There are numerous causes which can lead to damage in the wind power<br />
systems. Some of the reasons could be improper installation, operation under<br />
severe conditions or low quality components.<br />
Fig. 3 – A representations of some types damage in the blades.<br />
Lightning can cause severe structural damage and the destruction of several<br />
towers. Fire and strong wind can also damage a turbine. The most critical task is<br />
probably when the turbine is stopped due to high wind.<br />
The greatest danger occurs when, during a period of strong wind, the brake<br />
system fails, and the turbine cannot be controlled. The brake system of a turbine<br />
rotor is designed to halt if the wind is too strong.<br />
When the brakes do not work, the turbine gets out of hand. In Germany, on<br />
several occasions during 1999, 2000 and 2003, the wind turbine brakes failed to<br />
stop the rotor, and blades were loose, pieces of which are found more than<br />
500m away from the pillar.<br />
Fig. 4 – Examples of damaged turbines at 100%.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 185<br />
4. Conclusions<br />
1. Since the addition of the MSR monitoring system to a group of wind<br />
turbine can negatively affect performance, the number and location of the<br />
sensors is an important problem, which was addressed to a significant extent in<br />
the current literature. The methods to be implemented for wind turbines should<br />
demonstrate that they can perform well despite the small number of<br />
measurement points and under the constraint that these points will be located in<br />
areas prone to damage.<br />
2. The technologies for monitoring and damage detection can be selected to<br />
obtain comprehensive information. Detection methods are based on a network<br />
of sensors adapted for early warning at the onset of damage. The methods can<br />
detect the types of damages of less than 1 micron and the location of the<br />
damage can be accurately determined.<br />
3. Some of the limitations are the number of sensors they call for, as well as<br />
the noise produced during the operation of the turbine. In addition, the<br />
maintenance of a wired and wireless network between rotating parts, blades and<br />
nacelle is still difficult. The most promising methods are the ultrasound and the<br />
thermographic techniques. The X-ray methods, which were discussed here, can<br />
also be applied for non-destructive testing, where in situ results must be fully<br />
verified.<br />
4. Finally, since the environmental benefits are constantly emphasized, the<br />
generation of electric power with the help of wind systems is a necessary<br />
innovation. Therefore, related industries require monitoring systems that can<br />
provide effective maintenance programs, which provide the earliest possible<br />
detection of defects.<br />
5. The most important component of a wind turbine which requires<br />
monitoring is the blade, so we must implement a reliable system of low-cost<br />
sensors. This system should be integrated into the turbine blades in the<br />
manufacturing phase, which will signal any potential failure in the turbine.<br />
Therefore, the wind turbine should be built intelligently.<br />
Acknowledgements. The authors would like to thank HCI Environment SRL for the<br />
support and constructive comments.<br />
Received April 15, 2010 <strong>POLITEHNIC</strong>A University,<br />
Faculty of Energetics<br />
Bucuresti, Romania<br />
e-mail: mihai.manescu@expert-mediu.eu
186 Mihai Florin Mănescu and Valeriu Panaitescu<br />
R E F E R E N C E S<br />
1. F a r r a r C. R., S o h n H., Pattern recognition for structural health monitoring<br />
Workshop on Mitigation of Earthquake Disaster by Advanced Technologies,<br />
Las Vegas, NV, USA, 30 Nov.–1 Dec. 2000.<br />
2. G h o s h a l A., S u n d a r e s a n M. J., S c h u l z M. J., P a i P F., Structural health<br />
monitoring techniques for wind turbine blades. J. Wind Eng. Ind. Aerodyn., 85<br />
(2008) 309–24<br />
3. H a m e e d Z., H o n g Y. S., C h o Y. M., A h n S. H., S o n g C. K., Condition<br />
monitoring and fault detection of wind turbines and related algorithms: a review<br />
Renew. Sustain. Energy Rev. , 2007.<br />
4. K i m H., M e l h e m H., Damage detection of structures by wavelet analysis. Eng.<br />
Struct. 26, 347–62, 2004.<br />
5. S ø r e n s e n B. F., J ø r g e n s e n E., D e b e l C. P., J e n s e n F. M., J e n s e n H.<br />
M., J a c o b s e n T. K., H a l l i n g K. M., Improved design of large wind<br />
turbine blade of fibre composites based on studies of scale effects (Phase 1)<br />
Summary Report (Risø-R Report) Risø National Laboratory, Denmark, 2004.<br />
6. C a s el i t z P., G i e b h a r d t J., M e v e n k a m p M., On-line fault detection and<br />
prediction in wind energy converters Proc. EWEC, (Thessaloniki, Greece, 1994,<br />
pp 623–7<br />
7. G i e b h a r d t J., R o u v i l l a i n J., L y r n e r T., B u s s l e r C., G u t t S., H i n r i<br />
c h s H., G r a m-H a n s e s K., W o l t e r N., G i e b e l G., Predictive condition<br />
monitoring for offshore wind energy converters with respect to the IEC61400-25<br />
standard Germany Wind Energy Conf. DEWEK, Wilhelmshaven, Germany,<br />
2004.<br />
8. R o s e n b l o o m E., A Problem with Wind Power www.aweo.org, 2006.<br />
9. A s h l e y F., C i p r i a n o R. J., B r e c k e n r i d g e S., B r i g g s G. A., G r o s s<br />
L. E., H i n k s o n J., L e w i s P. A., Bethany Wind Turbine Study Committee<br />
Report www.townofbethany.com, 2007.<br />
10. H a h n B., D u r s t e w i t z M., R o h r i g K., Wind Energy (Reliability of Wind<br />
Turbines Experiences of 15 Years with 1,500 W) ed J Peinke et al, Berlin:<br />
Springer, 2007, pp 329–32.<br />
11. C a i t h n e s s W i n d f a r m Information Forum 2005 Wind Turbine Accident<br />
Data to December 31st 2005 http://www.caithnesswindfarms.co.uk/<br />
12. F l e m m i n g M. L., T r o e l s S., New lightning qualification test procedure for<br />
large wind turbine blades, Int. Conf. Lightning and Static Electricity,Blackpool,<br />
UK, 2003, pp 36.1–10<br />
13. C h i a C h e n C i a n g, J u n g-R y u l L e e, H y u n g-J o o n B a ng Structural<br />
health monitoring for a wind turbine system: a review of damage detection<br />
methods (IOP PUBLISHING) 2008.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 187<br />
14. L a d i n g L., M c G u g a n M., S e n d r u p P., R h e i n l a n d e r J., R u s b o r g<br />
J., Fundamentals for remote structural health monitoring of wind turbine<br />
blades—a preproject ANNEX B Risø-R-1341(EN) Report Risø National<br />
Laboratory, Denmark, 2002.<br />
TEHNOLOGII FOLOSITE PENTRU MONITORIZAREA<br />
DEFECŢIUNILOR STRUCTURALE APĂRUTE ÎN<br />
FUNCŢIONAREA GRUPURILOR EOLIENE<br />
(Rezumat)<br />
Necesitatea dezvoltării durabile impune utilizarea resurselor regenerabile de energie, în<br />
principal pentru menţinerea celor existente (combustibili fosili) la un nivel convenabil,<br />
dar şi pentru conservarea mediului înconjurător. În aceste condiţii sursele de energie<br />
regenerabila au început să fie intens studiate şi dezvoltate cu precădere în ultimii zece<br />
ani. În România primele grupuri eoliene au fost puse în funcţiune în anul 2005<br />
(Complexul industrial Ploieşti – o centrala Vestas V52 – 660 kW, relocată), în prezent<br />
existând o putere instalata de circa 14 MW.<br />
În sensul valorificării energiei eoliene pentru producerea de energie electrică,<br />
datorită faptului că la înălţimi de peste 100 m, vântul nu mai este influenţat de<br />
neregularităţile existente la sol (arbori, clădiri, coline, etc.) grupurile eoliene au devenit<br />
tot mai înalte, iar accesul pentru revizii şi reparaţii este dificil. Ca urmare a acestui fapt<br />
este necesară o monitorizare din timp a avariilor ce pot să apară în timpul funcţionării<br />
unui parc eolian, pentru a reduce timpul în care grupul eolian este oprit, iar pierderile de<br />
producţie să fie cât mai mici. Pentru a se asigura o bună funcţionare în condiţii de<br />
siguranţă este necesar a se implementa un sistem de monitorizare a structurii de<br />
rezistenţă (MSR) a ansamblului eolian. Acest sistem şi cateva metode de evaluare a<br />
avariilor vor fi prezentate în acest articol. MSR are în componennţă două ramuri: o reţea<br />
de senzori ce colectează informaţiile şi un algoritm/software pentru interpretarea datelor<br />
provenite de la senzori. Cateva dintre tehnologiile ce pot fi folosite, metode cu<br />
ultrasunete, scanare termica, raze X, etc. vor fi prezentate în articol. Tot în acest articol<br />
vor fi prezentate cele mai frecvente avarii de structură ce pot apărea la un grup eolian.<br />
Spre deosebire de alte surse regenerabile de energie, energia eoliană are o limită<br />
datorită maturităţii sale tehnologice, infrastructură bună şi o competitivitate relativă de<br />
preţ. Pentru a avea o eficienţă cost-beneficiu mai ridicată, grupurile generatoare eoliene<br />
au devenit tot mai înalte.<br />
Monitorizarea structurii de rezistenţă (MSR) este esenţială în raport cu celălalte<br />
sisteme de monitorizare pe care le are în componenţă un grup eolian, deoarece<br />
prăbuşirea întregii structurii are urmări grave asupra siguranţei populaţiei, a mediului şi<br />
din punct de vedere financiar al investitorului.<br />
Sistemele de monitorizare sunt sigure, au un cost redus, iar prin implementarea lor<br />
se pot reduce pierderile de energie din timpul reviziilor sau avariilor. Prin aceste sisteme<br />
se pot diagnostica ce piese ale grupului generator trebuie înlocuite şi reducerea timpului<br />
din cadrul unei revizii programate. Totodată aceste sisteme pot face o predicţie asupra<br />
ciclului de viaţă al centralei al cărui timp de lucru proiectat este de 10-30 ani.
188 Mihai Florin Mănescu and Valeriu Panaitescu<br />
Conform Farrar şi Sohn [1] sistemul de monitorizare este definit ca un proces care<br />
detectează defecţiunile apărute în construcţiile înalte. Defecţiunile sunt definite aici ca<br />
modificări ale materialelor şi / sau proprietăţile geometrice ale acestor sisteme, inclusiv<br />
schimbări în condiţiile limită şi conectivitatea sistemului, care afectează în mod negativ<br />
performanţa sistemului. Exista mai multe cauze de avariere a structurii cum ar fi<br />
absorbţia de umiditate, oboseala, rafale de vânt [2], stres termic, coroziune, foc şi chiar<br />
trăsnet [3]. Palele turbinelor care nu au un sistem de protecţie contra fulgerelor sunt<br />
adesea lovite de acestea.<br />
În general, dezvoltarea cu succes ale metodelor de MSR depinde de doi factoricheie:<br />
tehnologia de detectare şi analiză a semnalului şi algoritmul de interpretare [4].<br />
Componentele MSR sunt formate din sistemul de achiziţie de date, filtrare de date,<br />
recunoaşterea de model şi luarea deciziilor. Fiecare dintre aceste componente este la fel<br />
de importantă în determinarea stării siguranţă a unei structuri [5].<br />
Tehnologiile de monitorizare sau de detectare a avariilor pot fi selectate pentru a<br />
obţine informaţii globale. Metodele de detectare au la bază o reţea de senzori adaptată<br />
pentru o avertizare timpurie de la debutul avariei. Metodele pot detecta tipuri de daune<br />
de sub 1 micron şi se pot localiza deteriorările sau impactul destul de precis.<br />
Limitarea dintre aceste tehnoligii este dată de numărul de senzori solicitaţi, precum<br />
şi de zgomotul produs în timpul funcţionării turbinei. Metodele cele mai promiţătoare<br />
sunt cele cu ultrasunete şi termografia. Metodele cu raze X, discutate, pot de asemenea,<br />
să fie aplicate pentru control nedistructiv, atunci când rezultatele în situ trebuie să fie<br />
complet verificate.<br />
În cele din urmă, după cât de mult este pus accentul pe mediu, beneficiile sistemelor<br />
eoliene în generarea de energie electrică, sunt mari, dar şi costisitoare cu construcţia şi<br />
întreţinerea. Prin urmare, industriile conexe necesită sisteme de monitorizare care să<br />
poată oferi eficienţă programelor de întreţinere, care să ofere detectarea defectelor cât<br />
mai devreme posibil.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
PUMPING STATION EXONERATION FOR WATER SUPPLY<br />
BY<br />
AURORA ALEXANDRESCU, A<strong>DIN</strong>A SIMONA ALEXANDRESCU<br />
and ADRIAN CONSTANTIN ALEXANDRESCU<br />
Abstract. Profitability of water distribution activity depends largely on the<br />
relationships between operational capability and service costs, related to<br />
supplier’s performance, volume of distributed water and effective operating costs.<br />
The main variables that influence the total selling price are required investment<br />
value, specific consumption of electrical energy for pumping power, unit price of<br />
the electrical energy and total volume of monthly consumed water billed. The<br />
selection of rehabilitation and modernization measures must rely on market<br />
studies results that appropriately establish the quantities of water that may be<br />
distributed and billed. Present and future water requirements will be determined<br />
based on the analysis of actual operation data and on estimation of future trends in<br />
water consumption on national and international levels.<br />
Key words: adduction, conductivity, chlorine, pipe network, pumping station,<br />
tank.<br />
1. Introduction<br />
Many systems for which a centrifugal pump is otherwise suitable may,<br />
however, have a variable demand in which case, a certain loss of efficiency may<br />
have to be accepted from part of the head or part of the capacity used for control<br />
purposes, using either discharge throttling or bypass control. Both methods will<br />
inevitably result in power loss, so if economic regulation is of primary<br />
importance, discharge regulation by speed control should be investigated first<br />
since this is less wasteful of power and there is usually a considerably smaller<br />
loss of pump efficiency. Speed control is now a particularly attractive<br />
proposition with the increasing availability of variable frequency power units,<br />
[1]. The measures for the improvement of the economic performances are<br />
established from the system’s dispatcher. The criterions used are the maximum
190 Aurora Alexandrescu et al.<br />
output of the pumping equipages and the optimum diameter of the piping. The<br />
replacement of the existent equipment, that is obsolete from physical and<br />
technological point of view, must be done with new equipments with<br />
performances that will meet the requirements of an optimum operation from<br />
both energetic and economic perspectives, [2].<br />
2. Methodology<br />
This analysis is more likely to be of academic rather than practical interest,<br />
however, since the main requirements with regulated flow are:<br />
a) To achieve the required variation in delivery to meet demand variations.<br />
b) To realise good pumping efficiency at all demand levels, [3].<br />
The connections analyze between the functional parameters of pumping<br />
station and the quality parameters of drinkable water are:<br />
a) The connections estimation between the pump’s load, debit and the water’s<br />
turbulence.<br />
b) The values analyze of chlorine and ammonium concentration depending on<br />
debit.<br />
c) The water conductivity analyzes depending on debit.<br />
The best power and economical performances will correspond to the<br />
pumping solution which ensures the covering of the request area (Q, H) with the<br />
best output. The pumping efficiency is established by studying technical<br />
implications of modernization measures of the power station. It is imperative to<br />
use the automatic systems for the water treatment. Energy efficiency and<br />
economic efficiency for the pumping supply system are tightly connected to the<br />
proper choice of pumping device and appropriate operation of the hydraulic<br />
system, [4].<br />
The evaluation of connections between the functional parameters of<br />
hydraulic systems and the quality indicators of water, depending on time is<br />
obtained with the help of an original automatic calculation program, elaborated<br />
by authors. Allowing for the actual situation of the adduction and the<br />
effectuation’s necessity of the repair workings capital or the pipe’s replacement<br />
in some sectors, respective of the pumping aggregates, for to assure a powereconomical<br />
evolution favorable of the system, one are determined: the optimum<br />
nominal diameter of the pipes and the maximum theoretical outturn of the<br />
pumps what must be used and the change’s convenience of the existing<br />
equipment, [3].<br />
Depending for the installation’s total outturn and for the water volume W<br />
circulated in the period of reference, the consumption of power for the water’s<br />
pumping Eso is independent at stop’s duration „at top”:
(1)<br />
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 191<br />
Ntm. W KN. QH . t. W KN. QLJW . . .<br />
E so = = =<br />
3600 3600. η 3600. η<br />
p p<br />
In Eq (1) has used the following notations: Ntm (kW) - power average<br />
necessary for water’s transport:<br />
(2)<br />
KN. QH . t KN. QLJ . .<br />
Ntm( QDL , , ) = = .<br />
η η<br />
p p<br />
W (m 3 ) -volume of water brought in T (hours) period; J (-) - slope metric of the<br />
water transport; D (m) - nominal pipe’s diameter; Q (l/s) - debit of water; L (m)<br />
- pipe’s length; ηp (%) - pump’s outturn; KN , Kj - coefficients; Ht (m) - loss of<br />
charge:<br />
(3)<br />
Q<br />
Ht= L. J( Q, D)<br />
, J = K j.<br />
D<br />
γ<br />
β<br />
⎛ γ<br />
Q ⎞<br />
DQJ ( , ) = ⎜K j.<br />
,<br />
⎜<br />
⎟<br />
J ⎟<br />
⎝ ⎠<br />
It is calculated annual medium costs associated to the power consumption<br />
and the annual average cost associated to the investments:<br />
(4) CAE = pE.(1 + krE<br />
). E A= AIc + AP= ac. Ic + a . I<br />
.<br />
1/ β<br />
, p p<br />
The investment in the pipes network and the investment in the pumping<br />
station are calculated with the fallowing equations:<br />
(5)<br />
(6)<br />
α/ β<br />
p<br />
⎛ γ<br />
Q ⎞<br />
⎛1,15. KN. QL . . J ⎞<br />
Ic( D, L) = La . . ⎜K j. ; Ip( Nti) = Ipo.<br />
.<br />
⎜<br />
⎟<br />
⎜ ⎟<br />
J ⎟ ⎜ ⎟<br />
⎝ ⎠ ⎝<br />
η p ⎠<br />
It is calculated the annual medium total costs:<br />
( + )<br />
α/ β<br />
⎛ γ<br />
Q ⎞<br />
⎛1,15. KN. QL . . J ⎞<br />
CA = A+ CAI = ac. La . . ⎜K j. + ap. Ipo.<br />
⎜<br />
⎟<br />
⎜ ⎟<br />
J ⎟ ⎜ ⎟<br />
⎝ ⎠ ⎝<br />
ηp<br />
⎠<br />
pE.1 krE . KN. QL . . JW .<br />
+<br />
.<br />
3600. η<br />
p<br />
.<br />
α<br />
α p<br />
+
192 Aurora Alexandrescu et al.<br />
(7)<br />
(8)<br />
The annual medium marginal cost is calculated with the fallowing equation:<br />
( + )<br />
α/ β<br />
p<br />
C .<br />
A La . ⎛ γ<br />
Q ⎞ Ipo ap⎛1,15. KN. QLJ . . ⎞<br />
CM = = . ⎜Kj. + .<br />
+<br />
W W ⎜<br />
⎟<br />
⎜ ⎟<br />
J ⎟ W ⎜ ⎟<br />
⎝ ⎠ ⎝<br />
ηp<br />
⎠<br />
αγ / β<br />
pE.1 krE . KN. QLJ . . Q<br />
αp αp<br />
+ = A*. + BQ . . J + CQJ . . ;<br />
3600. η<br />
α/ β<br />
J<br />
p<br />
( )<br />
p<br />
La . α/ β Ipo. ap⎛1,15.<br />
KN. L⎞<br />
pE.1 + krE . KN. L<br />
A* = . ( K j ) ; B= . ⎜ ⎟ ; C =<br />
.<br />
W W ⎜ ⎟<br />
⎝<br />
ηp ⎠<br />
3600. ηp<br />
The minimum marginal cost has the following configuration:<br />
αγ<br />
∂CM α Q β<br />
αp αp−1 = 0 ⇒− A. . + B. α p.<br />
Q . J + C.<br />
Q = 0<br />
∂Jϕ α<br />
1+<br />
β<br />
J<br />
3. Results and Discussion<br />
The available data for choosing the proper pumping device require a careful<br />
analysis to determine the head characteristics of the pressurized supplies for<br />
each specific configuration and to evaluate the proper operational regimes of the<br />
pumping devices. The calculation method is applied in the exoneration case of<br />
pumping station CUG Iasi for drinkable water from Iasi city.<br />
Before of exoneration the pumping station CUG Iasi has been equipped<br />
with two 8 NDS pumps; after exoneration the pumping station CUG is equipped<br />
with two Wilo NPG 200-500-160/4/499 pumps.<br />
The measure and control system of drinkable water’s quality, of functional<br />
parameters of the pipes network from CUG pumping station has been complete<br />
automated in August – December 2008 with performance installations by<br />
Endress Hauser concern. All system has automated calculus program that<br />
permits of the dispatcher from S. C. APAVITAL S. A. Iasi industry to pursue<br />
the parameters’ network in 17 demurrages from the supply system with<br />
drinkable water of Iasi town, (Fig. 1). The registering is permanent, 24 hours.<br />
The water samples are analyzed automatic at each 2 ÷ 5 minutes. Drinkable<br />
water is treated with ammonium and chlorine in accordance with the properties<br />
of the analyzed water.<br />
α<br />
α
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 193<br />
The all measurement system of drinkable water’s conductivity from pipes<br />
network supplied of CUG Iasi pumping station includes the following elements:<br />
temperature integrated sensors inductive and conductive sensors. The turbidity<br />
and the solid suspensions quantity Tb from drinkable water system are<br />
measured with the Liquisys M CUM 223/253 type system. The turbidity sensors<br />
are Turbimax W CUS 31 type.<br />
Fig. 1 – The emplacement points of 17 automatic pursuit systems automate of the<br />
supply network with drinkable water from Iasi city.<br />
The conductivity, chlorine and turbidity concentration variations are<br />
measured in same moments of 1.06.2009 day, (Fig. 2) in CUG pumping station.<br />
The automized regulation system of chlorine and chlorine dioxide quantity from<br />
CUG pumping station is Liquisys M CCM 223/253 type. The CSC 140/141<br />
type sensors are used to visualize the automized measurement and processing of<br />
the Cl2, HOCl, OCl - concentrations, depending on the drinkable water pH. The<br />
measurement system of ammonium quantity from drinkable water is Stamolys<br />
CA 71 AM type. The ammonium concentration variation in same time in<br />
1.06.2009 day is presented in the CUG pumping station E1, in Fig. 3. The<br />
digital transmission system of the pressure p and the debit Q from CUG<br />
pumping station at the S. C. APAVITAL S. A. Iasi dispatcher is Cerebar M<br />
HART, respectively Priline Promag 50 types.
194 Aurora Alexandrescu et al.<br />
Fig. 2 – Diagram for the functional parameters of the CUG Iaşi pumping station:<br />
conductivity, chlorine, pressure, flow and turbulence for the perioad 5.07.2009 –<br />
6.07.2009.<br />
AIc ( D )<br />
AP ( k , Q , D )<br />
A( k, Q , D )<br />
1000<br />
900<br />
800<br />
700<br />
600<br />
500<br />
400<br />
300<br />
200<br />
100<br />
0.3 0.4 0.5 0.6 0.7 0.8<br />
Fig. 3 – The variation of annual cost associated to the investment in pipes AIc, annual<br />
total cost AP for pumping and the annual average cost associated to the investments A,<br />
depending on diameter D for the CUG pumping station.<br />
The results of the data working relative at either sector of the system,<br />
effectuated with MathCAD program, adequate to the supply with power under<br />
average and low tension are systematized in Fig. 3. It is present the variation of<br />
the following parameters: AIc – annual cost associated to the investment in<br />
D
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 195<br />
pipes, AP – annual total cost for pumping and A - the annual average cost<br />
associated to the investments in the usage’s conditions of the following<br />
parameters: ac = 0,075; a = 6000; α = 2; k – pipe’s equivalent absolute rugosity<br />
= 0,8 mm; Kj = 0,00145; γ = 2; β = 5,143+0,131.k; ηp = 72 %; T = 6200<br />
hours/year; αp = 0,46; Ipo = 65; Q = 0,15 m 3 /s; D = 0,3; 0,4; ...; 0,6 m.<br />
4. Conclusions<br />
1. It comes out that chlorine and ammonium concentrations grow up in the<br />
maximum consumption period of network: Cl = (1,1 ÷ 2,89) [mg/l] and Am =<br />
(42,55 ÷ 43,31) [mg/l]. In the same time, pressure and debit grow up because to<br />
the increase of the active consumers number in the network: p = (3,01 ÷ 4,09)<br />
[bar] and Q = (0 ÷ 1650,74) [m 3 /h].<br />
2. The water turbidity grows up at (0,49 ÷ 0,59) FNU values on starting<br />
and stop period for each pumps. In short time from the starting and stop<br />
moment, the turbidity declines at (0,26 ÷ 0,30) FNU values. The water’s<br />
conductivity is constant (0,53 ÷ 0,54) [mS/cm].<br />
3. The flow and load variation from hydraulic system influence the water<br />
quality and the substances quantity imperative for the drinkable water treatment:<br />
in the analyse period 1.05.2009, 00:00 hour and 15.05.2009, 23:55 hour,<br />
maximum chlorine is Clmax = 3,94 [mg/l] in 7.05.2009 day, 3:45 hour; at same<br />
moment the hydraulics system has the following parameters: Tb = 0,42 FNU<br />
turbidity, Q = 0,121 [m 3 /s] flow, H = 69,34 [m] load, Cv = 0,51 [mS/cm]<br />
conductivity. The minimum chlorine is Clmin = 1,1 [mg/l] in 3,05.2009 day, 3:35<br />
hour; the parameters of the CUG pumping station at same moment are: 3:35<br />
turbidity, Q = 0 [m 3 /s] flow, H = 64,344 [m] load, Cv = 0,52 [mS/cm]<br />
conductivity.<br />
Acknowledgements. This work has been supported by the National Centre of<br />
Management Programmers, Romania, under financial contract No. 21-041/2007.<br />
Received: February 25, 2010 “Gh. Asachi”Technical University,<br />
Department of Fluid Mechanics, Machines and Hydraulic Acting,<br />
Iasi, Romania,<br />
e-mail: auralexis@yahoo.com<br />
R E F E R E N C E S<br />
1. A l e x a n d r e s c u A., Statii de pompare, Ed. Politehnium, Iasi, 2008.<br />
2. A l e x a n d r e s c u A., Masini si echipamente hidraulice, Ed. Politehnium, Iasi,<br />
2008.
196 Aurora Alexandrescu et al.<br />
3. A l e x a n d r e s c u A., A l e x a n d r e s c u S. A., A l e x a n d r e s c u C. A.,<br />
Contributions concerning the power optimization of the pumping stations,<br />
ASME Fluids Engineering Division Summer Conference, ISI Thomson<br />
Proceedings of 11th International Symposium on Advances in Numerical<br />
Modelling of Aerodynamics and Hydrodynamics in Turbo machinery, August<br />
10-14, FEDSM2008-55007, Jacksonville, Florida, USA, 2008.<br />
4. A l e x a n d r e s c u A., B e r t e a A., Economic efficiency of the investment for<br />
pumping stations exoneration, ISI Thomson Proceedings in The 6th International<br />
Conference Management of Technological Changes, Centre for Continuing<br />
Education and Training – CETEX, MTC, Alexandropoulos, Greece, Vol. I,<br />
2009, pp. 437-441.<br />
REABILITAREA STATIEI DE POMPARE PENTRU ALIMENTARE CU APA<br />
(Rezumat)<br />
Sistemul de măsurare şi de control a calităţii apei potabile, a parametrilor<br />
funcţionali ai reţelelor de conducte trebuie complet automatizat în întregul municipiu<br />
Iaşi şi generalizat la nivelul întregii ţări. Deşi investiţia în sistemele moderne de<br />
automatizare, control şi măsurare din reţelele de alimentări cu apă potabilă din marile<br />
oraşe este mare, aceasta poate fi amortizată repede, prezentând multiple avantaje:<br />
- Posibilitatea optimizării funcţionarii ansamblului staţie de pompare<br />
- Obţinerea unei caţităţi foarte bune a apei potabile, dozarea fiind efectuată automat,<br />
în funcţie de calitatea apei ce curge prin reţelele de conducte.<br />
- Obţinerea unei concentraţii optime a bulelor de gaze conţinută în apa potabilă de<br />
(3 ÷ 10) % pentru evitarea fenomenului de vortex în instalaţii.<br />
Turbiditatea apei potabile creşte brusc la valori cuprinse între (0,49 ÷ 0,59) FNU pe<br />
toată perioada în care se pornesc şi se opresc fiecare din pompe. În scurt timp de la<br />
momentul închiderii, respectiv a deschiderii pompelor, turbiditatea scade la valori<br />
cuprinse între (0,26 ÷ 0,30) FNU. Conductivitatea apei rămâne aproximativ constantă,<br />
variind în intervalul (0,53 ÷ 0,54) [mS/cm]. Clorinarea variază în intervalul (1,1 ÷ 2,89)<br />
mg/l şi creşte odată cu creşterea turbidităţii apei potabile.<br />
Variatia debitului si a sarcinii din sistemul hidraulic studiat influenteaza puternic<br />
calitatea apei si implicit cantitatea de substante necesare tratarii apei pentru potabilizare:<br />
in perioada de analizată, cuprinsă în intervalul 1.05.2009, ora 00:00 şi 15.05.2009 ora<br />
23:55, clorinarea maximă Clmax = 3,94 [mg/l] s-a înregistrat în ziua de 7.05.2009, ora<br />
3:45; în acelaşi moment s-au înregistrat şi următorii parametri: turbiditatea Tb = 0,42<br />
FNU; debitul Q = 0,121 [m 3 /s]; sarcina H = 69,34 [m]; conductivitatea Cv = 0,51<br />
[mS/cm]. În aceeaşi perioadă, clorinarea minimă Clmin = 1,1 [mg/l] a fost înregistrată în<br />
ziua de 3,05.2009, ora 3:35. Ceilalţi parametri ai staţiei de pompare CUG în acelaşi<br />
moment au fost: Tb = 0,43 FNU; debitul Q = 0 [m 3 /s]; sarcina H = 64,344 [m];<br />
conductivitatea Cv = 0,52 [mS/cm].
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
CONTRIBUTIONS REGAR<strong>DIN</strong>G DURABILITY EVALUATION<br />
OF HORIZONTAL AXIAL HYDRAULIC TURBINES SHAFTS<br />
BY<br />
ILARE BORDEASU 1 , MIRCEA OCTAVIAN POPOVICIU 1 , DRAGOS<br />
NOVAC 2 , LIVIU MARSAVINA 1 , RADU NEGRU 1 , MIRCEA VODA 1 ,<br />
VICTOR BALASOIU 1 and MARIAN BĂRAN 1<br />
Abstract. Using previous researches regarding cracks initiation, the present<br />
paper analyzes the durability of horizontal bulb turbines shaft, especially in the<br />
joining zone between the shaft main body and the runner flange. For<br />
exemplifying, there were used the data of the turbines running in the Power Plants<br />
Iron Gates II and Gogosu. The obtained conclusions can be used to avoid serious<br />
failures of horizontal hydraulic turbines. The abstract of the paper is to be written<br />
here. It contains the main ideas and original contributions and conclusions of the<br />
authors’ research.<br />
Key words: axial hydraulic turbines, cracks, stress alternation cycles, fracture<br />
mechanics characteristics<br />
1. Introduction<br />
It is a well-known fact that during of the turbine running (rotational<br />
motion), under the effect of: hydraulic forces and moments developed on the<br />
blades, runner masses (with or without oil) and vibration, the turbine shaft is<br />
submitted to specific variable stresses (tension, bending and torsion). Taking<br />
into account the results obtained in [1], in the present work, the durability of the<br />
horizontal axial hydraulic turbines shaft was evaluated using the professional<br />
programs ANSYS and AFGROW. In order to obtain also numerical examples<br />
were used the bulb turbines in function at the Power Plants Iron Gates II and<br />
Gogosu, for which we know the necessary data. The obtained conclusions allow<br />
avoiding dangerous troubles during the turbine operation.
198 Ilare Bordeasu et al.<br />
Fig. 1 – The 3D shaft model, obtained with the INVENTOR program [initiation].<br />
2. Estimation of Fatigue Crack Propagation for Variable Stresses<br />
with Constant Amplitudes<br />
For the selected axial hydraulic turbine (Power plant Iron Gates II) the shaft<br />
presents the following characteristics [1]:<br />
a) the initial crack occurs in the joining zone between the shaft flange (for<br />
coupling with the hydraulic runner) and the shaft body, fig. 2;<br />
b) in order to obtain the needed data we choose the AISI 1020 steel [6] instead<br />
of the actual used 20ГC steel [7]; both steels have almost the same mechanical<br />
characteristics but for the AISI 1020 we found all the supplementary<br />
characteristics needed for fatigue tests;<br />
c) the shaft dimensions are: length 7572 mm, maximum diameter 2300 mm,<br />
the main shaft body, between flanges, has an external diameter of 1200 mm and<br />
an internal one of 600 mm;<br />
d) according with the ANSIS program (Fig. 3), till crack initiation, the<br />
minimum number of cycles was 3,0139 10 8 , corresponding to 80,370 running<br />
hours.<br />
a) Deep circumferential cracks b) Cracks network<br />
Fig.2 – Cracks in the joining zone between the runner flange and the main shaft body.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 199<br />
Fig. 3 – Number of cycles till crack initiation.<br />
Usually, together with the increase of stress level is enhanced also the<br />
fatigue crack propagation speed. As a result, the propagation of the crack speed<br />
can be correlated with the variation of the stress intensity factor<br />
da<br />
ΔK, = f ( ΔK)<br />
, Fig. 4.<br />
dN<br />
Fig. 4 – Variation of crack propagation speed as a function of stress intensity variation.
200 Ilare Bordeasu et al.<br />
In conformity with Fig. 4, after the crack initiation, the maximum danger is<br />
represented by the zone III, characterized by great propagation velocities<br />
leading to instable increases of the failure. In this zone, the velocity of crack<br />
propagation is correlated with the variation of the crack intensity factor, through<br />
the relation proposed by Forman [2]:<br />
(1)<br />
da<br />
dN<br />
n<br />
C(<br />
ΔK<br />
)<br />
( − R)<br />
K − ΔK<br />
= 1<br />
C<br />
= f (ΔK, R),<br />
where: C and n are constant values depending on the used material;<br />
ΔK = Kmax − Kmin<br />
- represents the variation of the stress intensity factor; KC<br />
- is the critical value of the stress intensity factor (breaking tenacity); R -<br />
represents the asymmetry ratio of the stress cycle.<br />
Expressed by Nr, the necessary number of cycles for the extension of the<br />
crack, the lifetime can be obtained by solving the equation for dN.<br />
Integrating both members it results:<br />
(2) ∫ dN = N − N = N = ∫<br />
N<br />
N<br />
f<br />
d<br />
f<br />
d<br />
r<br />
a<br />
a<br />
cr<br />
d<br />
da<br />
.<br />
f ( ΔK,<br />
R)<br />
With the equation (2) it is possible to calculate the number of cycles Nr<br />
necessary for the extension of the crack from the detectable length ad (for which<br />
corresponds the number of cycles Nd) till the critical length acr (which<br />
corresponds to the Nf number of cycles).<br />
For the lifetime computation (the final number of cycles Nr) it was used the<br />
specialized program AFGROW, developed by H a r t n e r at WRIGHT-<br />
PATTERSON AIR FORCE BASE [3] in order to estimate the lifetime of some<br />
components for fighter planes. It was taken into consideration a ring section,<br />
having an elliptical crack as can be seen in Fig. 5. For such geometry, the<br />
intensity factor was proposed by Raju and Newman [4] as being:<br />
a<br />
= t b , i e ,<br />
Q<br />
(3) K ( σ + H σ ) π F(<br />
a c,<br />
D , D , φ)<br />
I<br />
where σt - represents the axial stress applied to the shaft, σb - represents the<br />
bending stress, H and F depends on the crack geometry (crack depths and<br />
length), the shaft thickness and the frontal position of the crack, a is the depth<br />
and c the half-length of the crack,
(4)<br />
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 201<br />
1.<br />
65<br />
⎛ a ⎞ a<br />
Q = 1+ 1,<br />
464⎜<br />
⎟ for ≤ 1.<br />
⎝ c ⎠ c<br />
Fig. 5 – Computation model for crack propagation.<br />
As was previously mentioned, the mathematical simulation was done with<br />
the program AFGROW, version 4.12.15 from 10.07.2008. The entrance data<br />
were: the external diameter (Do): 1.20 m, the internal diameter (Di): 0.60 m, the<br />
crack depth (A): 0.001 m, crack half length (C): 0.002 m<br />
In conformity with the shaft disposition, for studying the crack propagation<br />
there were taken into consideration two load situations with constant amplitude:<br />
a) a bending loading with a symmetrical alternative cycle having σmax = - σmin<br />
= 42 MPa, with an asymmetry coefficient of R = -1.<br />
b) a composed load (bending plus elongation); considering that the static<br />
elongation stress is superposed over the bending stress it result an approximate<br />
pulsating cycle with σmax = 88,9 MPa, σmin = - 6,5 MPa , R = - 0,07 [1].<br />
In order to introduce the material constants, characterizing the crack<br />
propagation, the used steel was considered to be equivalent with the AISI 1020<br />
for which the characteristic data are included in the data base of the AFGROW<br />
program. So, the constants for the Walker law [5]:
202 Ilare Bordeasu et al.<br />
(5)<br />
[ ] 1 1 m n<br />
K ⋅ ( 1−<br />
R)<br />
da<br />
= C1<br />
⋅<br />
dN<br />
max<br />
−<br />
for R ≥ 0,<br />
[ ] 1 1<br />
da −m<br />
n<br />
= C1<br />
⋅ K max ⋅ ( 1−<br />
R)<br />
for R < 0<br />
dN<br />
are C1 = 1,447 x 10 -12 ; n1 = 3,6; m = 1; the values for the breaking tenacity<br />
corresponding to a plane state of stresses are KC = 110 MPa m 1/2 , respectively<br />
for a plane state of deformation KIC = 77 MPa m 1/2 , the yielding point σC = 262<br />
MPa, the elastic modulus E = 206843 MPa and Poisson’s ratio ν = 0.3, the limit<br />
value of the stress intensity factor under which the crack does not propagate ΔK<br />
= 1,5 MPa m 1/2 .<br />
In Fig. 6 is presented the crack evolution for some time intervals under a<br />
load having a symmetric alternative bending cycle.<br />
a) a = 1mm, c = 2mm b) a = 16 mm, c = 32 mm<br />
N = 80370 hours (initiation) N = 153243 hours (propagation)<br />
c) a=150 mm and 2c= 320 mm<br />
N = 159737 hours (fracture, breakdown)<br />
Fig. 6 – Propagation of circumferential crack in the hollow shaft.<br />
The simulation begun with the following initial value of the crack a = 1 mm<br />
and c = 2 mm, values which are reached after 80370 running hours. The number
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 203<br />
of hours, for the crack propagation is added to the initial time (80370). The<br />
obtained results show that after 15937 running hours the crack penetrates<br />
completely the annular wall of the shaft.<br />
3. Conclusions<br />
The studies of the fatigue crack propagation lead to the following results:<br />
1. In the joint zone between the flange and the shaft body, the crack occurs<br />
after approximately 80370 running hours.<br />
2. The propagation of the fatigue cracks were obtained using the Paris law<br />
with the completion of Walker, and show that the crack reach the dimensions a<br />
= 16 mm (depth) şi 2c = 64 mm (length) after 153243 running hours. After that<br />
the propagation speed increases to a high degree and after only 159737 running<br />
hours the breakdown occur (a=150 mm and 2c= 320 mm).<br />
3. The obtained results can be used by the supervising staff in order to<br />
prescribe the intervals of periodical inspections and to avoid major damages.<br />
Acknowledgements. The present work has been supported from the Contract No.<br />
RU 177/10.10.2008, BC 146/13.10.2008 (Analiză privind soluţia de fiabilizare a<br />
arborelui turbinelor aplicată cu ocazia retehnologizării hidroagregatelor din CHE Porţile<br />
de Fier II. Propuneri de metodologie de urmărire în timp a stării arborilor turbinelor din<br />
CHE Porţile de Fier II şi CHE Gogosu)<br />
Received:March 15, 2010 1 Politehnica” University of Timisoara,<br />
e-mail: ilarica59@gmail.com<br />
2 Hidroelectrica Iron Gates, Dr.Tr. Severin<br />
e-mail: dragos.novac@hidroelectrica.ro<br />
R E F E R E N C E S<br />
1. B o r d e a ș u I., P o p o v i c i u M.O., N o v a c D., Fatigue Studies Upon<br />
Horizontal Hydraulic Turbines Shaft and Estimation of Crack Initiation Machine<br />
Design, University of Novi Sad, Faculty of Technical Sciences, 2009, pp 183-<br />
186.<br />
2. F o r m a n R.G., H e a r n e y V.E., E n g l e R.M., Numerical analysis of crack<br />
propagation in cyclic-loaded structures, Journal of Basic Engineering, Trans.<br />
ASME, Vol. 89, (1967).<br />
3. H a r t n e r J. A., AFGROW Users Guide and Technical Manual, Wright-Patterson<br />
Air Force BASE, Ohio, 2008.
204 Ilare Bordeasu et al.<br />
4. R a j u I. S., N e w m a n J. C., Stress Intensity Factors Circumferential Surface<br />
Cracks in Pipes and Rods, Proc. of 17 th National Symposium on Fracture<br />
Mechanics, Albany, NY, 1984.<br />
5. W a l k e r K., The effect of stress ratio during crack propagation and fatigue for<br />
2024-T3 and 7075-T6, ASTM STP 462, ASTM, 1970.<br />
6. * * * Fatigue Design Handbook, Second Edition, SAE, Warrendale, 1988,<br />
7. * * * Analiză privind soluţia de fiabilizare a arborelui turbinelor aplicată cu ocazia<br />
retehnologizării hidroagregatelor din CHE Porţile de Fier II. Propuneri de<br />
metodologie de urmărire în timp a stării arborilor turbinelor din CHE Porţile de<br />
Fier II şi CHE Gogosu, Contract No. RU 177/10.10.2008, BC 146/13.10.2008<br />
CONTRIBUŢII ÎN EVALUAREA DURABILITĂŢII ARBORILOR<br />
HIDROAGREGATELOR AXIALE ORIZONTALE<br />
(Rezumat)<br />
În cadrul lucrării se evaluează durabilitatea arborilor turbinelor axiale, în zona de<br />
racord a flanşei ce-l cuplează cu rotorul turbinei, plecând de la rezultatele obţinute în<br />
lucrările anterioare privind iniţierea fisurii. Ca model este folosit arborele de turbină<br />
bulb, aflată în exploatare la CHE Porţile de Fier II şi CHE Gogoşu, pentru care<br />
dispunem de datele necesare. Concluziile rezultate permit evitarea unei eventuale avarii<br />
periculoase.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
OPTIMAL ROUTES OF PIPELINE SUPPLY USING<br />
THE BELLMAN-KALABA ALGORITHM<br />
BY<br />
TEODOR MILOŞ, MIRCEA BĂRGLĂZAN, EUGEN DOBÂNDĂ,<br />
ADRIANA MANEA, RODICA BĂDĂRĂU and DANIEL STROIŢĂ<br />
Abstract. In this paper are presented the application of graphs theory for<br />
determining the optimal route of a pipeline supply being at the great distance of<br />
the target consumer (pipeline network of a city). It applies when the distance from<br />
source to target, because the configuration of the land, there are several variants of<br />
the route passing through some mandatory points. In this way the route has n<br />
sections and on each section the total cost (investment plus operating for one year)<br />
has a certain value. If it can browse the route by more than two then the method<br />
becomes profitable. Implementation of the method is through a special program,<br />
using the Borland Pascal programming and Bellman-Kalaba algorithm.<br />
Mathematical resolving is by the matrix. Numbering the sections with 1 ... n, in<br />
order to obtain the final optimal browsing, it is the range of selected sections. In<br />
actual conditions when more and more sources of drinking water are becoming<br />
more polluted, the feeding is justified to be from remote mountain areas of the<br />
natural springs.<br />
Key words: graph theory, pipeline, optimal route, matrix, Bellman-Kalaba<br />
algorithm, and network.<br />
1. Introduction<br />
Through scope of investment and energy consumption, adductions have a<br />
significant share in the systems of water supply, and their rational design<br />
behaves more optimization processes, among which an important place held<br />
optimize their route.<br />
In current practice of design, setting the optimal solution is, usually, by<br />
analytical study of two or three variants selected from many possible intuitive
206 Teodor Milos et al<br />
decisions, whose error is inversely proportional to the degree of experience of<br />
the designer.<br />
Modern mathematical disciplines, by operational calculation, put at<br />
specialist’s disposition a vast apparatus of scientific analysis in determining the<br />
optimal decisions for the design of water supply. In this context, describes a<br />
deterministic mathematical model optimization of the route of the water supply<br />
adduction, based on the theory of graphs.<br />
2. Calculation Algorithm<br />
Modeling this problem is achieved through representation related directed<br />
graph G=(X, U) consisting of the source as the origin, route as required arcs and<br />
points as vertices. For each arc u U is associated a number<br />
i j ∈ ( ) 0 ≥<br />
i<br />
λ u j , in<br />
conventional units, depending on the optimization criterion adopted. The route<br />
is the best way to graph, the minimum value which is determined by applying<br />
the algorithm Bellman-Kalaba.<br />
Graph G= (X, U) is attached to a matrix M whose elements mij<br />
are:<br />
(1)<br />
(2)<br />
( u )<br />
⎧ i<br />
λ j − the arc value<br />
from xi<br />
to x j<br />
⎪<br />
= ⎨∞<br />
− if vertices x and x are not adjacent .<br />
i<br />
⎪<br />
0 − for i = j<br />
⎩<br />
mij j<br />
The optimal route is the best way of graph μ , with the total value:<br />
λ<br />
( μ)<br />
λ(<br />
) → min<br />
= ∑<br />
u ∈μ<br />
i j<br />
i j<br />
u .<br />
If the notes with Vi<br />
the minimum value of the road μn , ( i = 0,<br />
n )<br />
existing from the tip of to the tip of x :<br />
xi n<br />
i<br />
(3) V λ(<br />
μ )<br />
hence:<br />
i<br />
= , ( i = 0,<br />
n ).<br />
(4) V = 0 ,<br />
n<br />
n<br />
i
then, under the principle of optimality:<br />
(5) V i ( V j + mij<br />
)<br />
=<br />
j≠i<br />
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 207<br />
min , ( i = 0, n −1<br />
; j = 0,<br />
n ) and V = 0 .<br />
To solve system (5) shall be iterative, noting Vi the value of Vi<br />
obtained at<br />
iteration k, namely:<br />
(6) V i = min<br />
Calculate:<br />
0<br />
(7) V i ( V j + mij<br />
)<br />
and then:<br />
0<br />
1<br />
= min , ( = 0, n −1<br />
j≠<br />
i<br />
k<br />
k −1<br />
(8) V i = ( V j + mij<br />
)<br />
j≠<br />
i<br />
( i = 0, n −1<br />
); V 0 .<br />
k<br />
0<br />
n =<br />
i ; j = 0,<br />
n ) and V 0 .<br />
n<br />
1<br />
n =<br />
min , ( i = 0, n −1<br />
; j = 0,<br />
n ) and V = 0 .<br />
Ordinal iteration of k expressed by relations (8) gives values only for the<br />
finite length of roads at most k + 1 arriving at xn<br />
, choosing between them is the<br />
minimum.<br />
From iteration to the next:<br />
(9)<br />
k<br />
−1<br />
≤ k k<br />
i Vi<br />
V , ∀ j .<br />
Numbers Vi<br />
( i ≠ n ; k = 0,<br />
1,...<br />
) monotone decreasing pattern formed that<br />
reach to the minimum necessary, after a finite number of iterations which not<br />
exceeding n −1.<br />
So algorithm stops when it reaches an iteration k, such that<br />
+ 1<br />
= , (<br />
k k<br />
V V i = 0,<br />
n ), and the minimum between the peaks road and x is<br />
i i<br />
k k + 1<br />
0 = V0<br />
V<br />
. To identify which roads have minimum values founded, are<br />
derived from (8) that have them at the last iteration we have:<br />
k<br />
n<br />
x0 n
208 Teodor Milos et al<br />
(10)<br />
k<br />
i<br />
ij<br />
k −1<br />
j<br />
V = m + V = m + V<br />
Based on the algorithm described, computer program named BEL_KAL was<br />
designed in Borland PASCAL language.<br />
Were made following nomenclatures: N is the order of the graph, V ( I,<br />
J ) is<br />
the column vector built at each iteration k, X ( I ) is the sequence of minimum<br />
road; VAL - the minimum road graph, M ( I, J ) - matrix associated with graph,<br />
whose elements are:<br />
(11)<br />
⎧<br />
⎪<br />
⎨<br />
⎪<br />
⎪<br />
⎩<br />
ij<br />
i<br />
λ(<br />
u j ) − if arc ( xi,x<br />
j )<br />
( u ) − if arc ( x ,x )<br />
n<br />
i<br />
mij = ∑ λ j<br />
i j<br />
i,j=<br />
1<br />
0 −<br />
for i<br />
= j<br />
k<br />
j<br />
exist<br />
.<br />
not exist<br />
Are introduced as data entry, graph order and its associated matrix, on lines,<br />
i<br />
whose elements are considered equal to λ ( u j ) , if there arc ( i j ) x x , , or equal to<br />
0, otherwise (MATR_EX1.dat file).<br />
As the data output sequence result road peaks of minimum values and value<br />
of the road.<br />
3. Case Study<br />
Apply dynamic programming method to solve the problem of site selection<br />
and capture of a main route to supply water through a sequential optimization<br />
where deterministic, discrete.<br />
It is considered a line of supply for locality L, departing from two locations<br />
of S1 and S2 capture (fig. 1). Possible routes through the required A, B, C, D, E,<br />
forming three sectors.<br />
Putting the problem of determining the route for which the total cost is<br />
minimum is determined for each track investments and prepare partial sequence<br />
graph in Figure 2, where each arc is associated a cost, in conventional units.<br />
They noted with x1 , 2 and the decision variables for each sector. These<br />
variables will not take numeric values, but will be vertices of graph that you are<br />
on the same alignment.<br />
x 3 x<br />
.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 209<br />
Fig. 1 – Variants of the adduction route.<br />
It exemplifies the application of Bellman-Kalaba algorithm to determine the<br />
optimal bus route for L locality from the source S1 (fig. 1).<br />
Fig.2 – Graph of adduction routes.<br />
Modeling problem is achieved through representation related directed graph<br />
G = (X, U) of order n = 7, consisting of the source point of origin, route as<br />
required arcs and points as vertices. For each arc u U is assigned a cost in<br />
i j ∈<br />
conventional units (fig. 2) and matrix M attached to graph G = (X, U) are the
210 Teodor Milos et al<br />
elements mi,j defined by relations (1), where λ(<br />
u )<br />
edge u U .<br />
i j ∈<br />
(12)<br />
1<br />
2<br />
3<br />
M =<br />
4<br />
5<br />
6<br />
7<br />
1<br />
0<br />
∞<br />
∞<br />
∞<br />
∞<br />
∞<br />
∞<br />
2<br />
42<br />
0<br />
∞<br />
∞<br />
∞<br />
∞<br />
∞<br />
3<br />
53<br />
∞<br />
0<br />
∞<br />
∞<br />
∞<br />
∞<br />
4<br />
∞<br />
∞<br />
92<br />
0<br />
∞<br />
∞<br />
∞<br />
5<br />
∞<br />
73<br />
61<br />
∞<br />
0<br />
∞<br />
∞<br />
6<br />
∞<br />
73<br />
82<br />
∞<br />
∞<br />
0<br />
∞<br />
7<br />
∞<br />
∞<br />
∞<br />
61<br />
92<br />
82<br />
0<br />
i j<br />
is the value attributed to<br />
V<br />
61<br />
92<br />
82<br />
0<br />
0<br />
i<br />
∞<br />
∞<br />
∞<br />
V<br />
1<br />
i<br />
∞<br />
153<br />
153<br />
61<br />
92<br />
82<br />
0<br />
V<br />
2<br />
i<br />
197<br />
153<br />
153<br />
61<br />
92<br />
82<br />
0<br />
V<br />
3<br />
i<br />
197<br />
153<br />
153<br />
.<br />
61<br />
Route with minimum total cost is given by way of minimum value in this<br />
graph, which is determined using Bellman-Kalaba algorithm.<br />
For each V ( k = 0,<br />
1,...<br />
is added to previous matrix a column in which<br />
values are inserted properly. It follows successively:<br />
k<br />
i )<br />
0<br />
a) Calculate the values i mi7<br />
V = (i = 1, ..., 7), which pass into the column<br />
92<br />
82<br />
0<br />
0<br />
V i .<br />
b) Calculate values 1<br />
V i for i =1,...,7 (j =1,...,7), which is passing in that<br />
column:<br />
(13)<br />
⎧V<br />
⎪<br />
⎪<br />
⎪V<br />
⎪<br />
⎪V<br />
⎪<br />
⎪<br />
⎨V<br />
⎪<br />
⎪V<br />
⎪<br />
⎪<br />
⎪V<br />
⎪<br />
⎪V<br />
⎩<br />
1<br />
1<br />
1<br />
2<br />
1<br />
3<br />
1<br />
4<br />
1<br />
5<br />
1<br />
6<br />
1<br />
7<br />
= min<br />
= min<br />
= min<br />
= min<br />
= min<br />
= min<br />
= 0<br />
j≠1<br />
j≠<br />
2<br />
j≠3<br />
j≠<br />
4<br />
j≠5<br />
j≠<br />
6<br />
0<br />
0 0<br />
0<br />
( V j + m1<br />
j ) = min(<br />
V2<br />
+ m12,<br />
V3<br />
+ m13,...,<br />
V7<br />
+ m17<br />
)<br />
0<br />
0<br />
0<br />
0<br />
( V j + m2<br />
j ) = min(<br />
V1<br />
+ m21,<br />
V3<br />
+ m23,...,<br />
V7<br />
+ m27<br />
)<br />
0<br />
0 0<br />
0<br />
( V j + m3<br />
j ) = min(<br />
V1<br />
+ m31,<br />
V2<br />
+ m32,...,<br />
V7<br />
+ m37<br />
)<br />
0<br />
0<br />
0<br />
0<br />
( V j + m4<br />
j ) = min(<br />
V1<br />
+ m41,<br />
V2<br />
+ m41,...,<br />
V7<br />
+ m47<br />
)<br />
0<br />
0 0<br />
0<br />
( V j + m5<br />
j ) = min(<br />
V1<br />
+ m51,<br />
V2<br />
+ m51,...,<br />
V7<br />
+ m57<br />
)<br />
0<br />
0 0<br />
0<br />
( V + m ) = min(<br />
V + m , V + m ,..., V + m )<br />
j<br />
6 j<br />
1<br />
61<br />
2<br />
61<br />
7<br />
67<br />
= ∞<br />
= 153<br />
= 153<br />
= 61 .<br />
= 92<br />
= 82
c) Calculate values<br />
(14)<br />
⎧V<br />
⎪<br />
⎪<br />
⎪V<br />
⎪<br />
⎪V<br />
⎪<br />
⎪<br />
⎨V<br />
⎪<br />
⎪V<br />
⎪<br />
⎪<br />
⎪V<br />
⎪<br />
⎪V<br />
⎩<br />
2<br />
1<br />
2<br />
2<br />
2<br />
3<br />
2<br />
4<br />
2<br />
5<br />
2<br />
6<br />
2<br />
7<br />
= min<br />
= min<br />
= min<br />
= min<br />
= min<br />
= min<br />
= 0<br />
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 211<br />
2<br />
V i for i = 1, ..., 7 (j = 1, ..., 7)<br />
j≠1<br />
j≠<br />
2<br />
j≠<br />
3<br />
j≠<br />
4<br />
j≠<br />
5<br />
j≠<br />
6<br />
1<br />
1 1<br />
1<br />
( V j + m1<br />
j ) = min(<br />
V2<br />
+ m12,<br />
V3<br />
+ m13,...,<br />
V7<br />
+ m17<br />
)<br />
1<br />
1 1<br />
1<br />
( V j + m2<br />
j ) = min(<br />
V1<br />
+ m21,<br />
V3<br />
+ m23,...,<br />
V7<br />
+ m27<br />
)<br />
1<br />
1 1<br />
1<br />
( V j + m3<br />
j ) = min(<br />
V1<br />
+ m31,<br />
V2<br />
+ m12,...,<br />
V7<br />
+ m37<br />
)<br />
1<br />
1 1<br />
1<br />
( V j + m4<br />
j ) = min(<br />
V1<br />
+ m41,<br />
V2<br />
+ m42,...,<br />
V7<br />
+ m47<br />
)<br />
1<br />
1 1<br />
1<br />
( V j + m5<br />
j ) = min(<br />
V1<br />
+ m51,<br />
V2<br />
+ m52,...,<br />
V7<br />
+ m57<br />
)<br />
1<br />
1 1<br />
1<br />
( V + m ) = min(<br />
V + m , V + m ,..., V + m )<br />
j<br />
6 j<br />
d) To calculate the values corresponding to column<br />
..., 7), an analogue, finally are obtained (15).<br />
2<br />
2<br />
i<br />
3<br />
i<br />
1<br />
61<br />
2<br />
62<br />
7<br />
67<br />
= 197<br />
= 153<br />
= 153<br />
= 61 .<br />
= 92<br />
= 82<br />
3<br />
V i for i = 1, ..., 7 (j = 1,<br />
Since V = V (i = 1, ..., 7), algorithm stops and the road is the minimum of<br />
3<br />
V1<br />
=V1<br />
= 197 . This value is reached on the way (1, 2, 6, 7), thus resulting in<br />
optimal route of adduction as: S1, A, C, and L. To resolve this problem the<br />
computer program BEL_KAL was used.<br />
(15)<br />
⎧V<br />
⎪<br />
⎪<br />
⎪V<br />
⎪<br />
⎪V<br />
⎪<br />
⎪<br />
⎨V<br />
⎪<br />
⎪V<br />
⎪<br />
⎪<br />
⎪V<br />
⎪<br />
⎪V<br />
⎩<br />
3<br />
1<br />
3<br />
2<br />
3<br />
3<br />
3<br />
4<br />
3<br />
5<br />
3<br />
6<br />
3<br />
7<br />
= min<br />
= min<br />
= min<br />
= min<br />
= min<br />
= min<br />
= 0<br />
j≠1<br />
j≠<br />
2<br />
j≠3<br />
j≠<br />
4<br />
j≠5<br />
j≠<br />
6<br />
2 ( V j + m1<br />
j )<br />
2 ( V j + m2<br />
j )<br />
2 ( V j + m3<br />
j )<br />
2 ( V j + m4<br />
j )<br />
2 ( V j + m5<br />
j )<br />
2 ( V + m )<br />
j<br />
6 j<br />
= 197<br />
= 153<br />
= 153<br />
= 61 .<br />
= 92<br />
= 82
212 Teodor Milos et al<br />
5. Conclusions<br />
1. This study put highlights how the economic fact of a technical problem<br />
can be optimized using mathematic-informatics structure type graph.<br />
2. Mathematical model is easily programmable in an evolved language,<br />
obtaining immediate results. The only problem is populating the matrix attached<br />
graph.<br />
3. Bellman-Kalaba algorithm was originally designed for the economy, but<br />
the adaptation was possible because the optimal path the economy is a virtual<br />
way and here is a real way.<br />
Acknowledgements. The present work has been supported by the Romanian<br />
Government – Ministry of Education, Research and Innovation, The National Centre<br />
for Programs Management (CNMP) through, CNMP project no. 21-036/2007 and<br />
CNMP project no. 21-41/2007.<br />
Received:<br />
R E F E R E N C E S<br />
1” Politehnica” University of Timisoara,<br />
Department of Hydraulic Machinery<br />
Timisoara, Romania,<br />
e-mail: teodor.milos@gmail.com<br />
1. Alexandrescu, A. Concerning optimization’s working of the pumping station<br />
for water feedings, 18 th International Conference on Hydraulics and Pneumatics,<br />
Prague, Czech Rep., Sbornik, ISBN 80-02-01567-3, 2003, pp. 263-268.<br />
2. A n t o n L. E., B a y a A., M i l o s T., R e s i g a R., Experimental Fluid<br />
Mechanics, Vol. 1, Ed. Academic Horizons, Timisoara, Romania, ISBN 973-<br />
8391-72-5, 2002.<br />
3. B e l l m a n , R.; K a l a b a , R. Dynamic Programming and Modern Control<br />
Theory, McGraw-Hill, New York, 1965.<br />
4. Nayyar M.L. Piping Handbook, 7 th Edition, McGraw–Hill, ISBN: 978-0-07-<br />
047106-1, New York, San Francisco, Tokyo, Toronto, 2000.<br />
5. Sârbu, I. Energetically Optimization of Water Distribution Systems, Ed.<br />
Academiei, ISBN: 973-27-0575-2, Bucuresti, Romania, 1996.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 213<br />
TRASEUL OPTIM AL UNEI CONDUCTE DE ADUCŢIUNE<br />
UTILIZÂND ALGORITMUL BELLMAN-KALABA<br />
(Rezumat)<br />
Prin amploarea investiţiei şi a consumului de energie, aducţiunile au o pondere<br />
însemnată în cadrul sistemelor de alimentare cu apă, iar proiectarea raţională a acestora<br />
comportă mai multe procese de optimizare, în rândul cărora un loc important îl deţine<br />
optimizarea traseului acestora.<br />
În practica actuală de proiectare, stabilirea soluţiei optime se face, de obicei, prin<br />
studierea analitică a două sau trei variante selectate din mulţimea posibilă prin decizii<br />
intuite, a căror eroare este invers proporţională cu gradul de experienţă al proiectantului.<br />
Disciplinele matematice moderne prin calculul operaţional pun la îndemâna<br />
specialistului un vast aparat de analiză ştiinţifică în stabilirea deciziilor optime pentru<br />
problemele proiectării sistemelor de alimentare cu apă.<br />
În această lucrare se prezintă modalitatea de aplicare a teoriei grafurilor din<br />
informatică pentru stabilirea traseului optim a unei conducte de alimentare aflată la<br />
distanţă mare de obiectivul consumator (reţea de conducte a unei localităţi).<br />
În acest context, se descrie un model matematic determinist de optimizare a traseului<br />
unei magistrate de aducţiune a apei, bazat pe teoria grafurilor. Traseul optim este dat de<br />
drumul de valoare minimă în graf, care se determină aplicând algoritmul Bellman-<br />
Kalaba.<br />
Se aplică în cazul când, pe distanţa de la sursă la obiectiv, datorită configuraţiei<br />
terenului, există mai multe variante de traseu care trec obligatoriu prin nişte puncte<br />
cheie. În acest fel traseul are n tronsoane, iar pe fiecare tronson costul total (investiţie<br />
plus exploatare timp de un an) are o anumită valoare. Dacă posibilităţile de parcurgere<br />
ale traseului sunt mai mult de două atunci aplicarea metodei devine rentabilă.<br />
Implementarea metodei se face prin intermediul unui program special, utilizând<br />
mediul de programare Borland Pascal, şi algoritmul Bellman-Kalaba. Rezolvarea<br />
propriu-zisă este matriceală. Numerotând tronsoanele cu 1…n, în final se obţine ordinea<br />
optimă de parcurgere ca șir al tronsoanelor selectate.<br />
În condiţiile actuale, când tot mai multe surse locale de apă potabilă sunt poluate,<br />
devine tot mai justificată alimentarea din surse îndepărtate, din zone montane cu izvoare<br />
naturale.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
STRUCTURAL ANALYSIS OF A SLEEVE RIGID<br />
COUPLING WITH CYLINDRICAL PINS<br />
BY<br />
DĂNUŢ ZAHARIEA and MIHAELA TUDORACHE<br />
Abstract. In this paper the displacements, as well as the von Misses stresses<br />
of a sleeve rigid coupling with cylindrical pins will be analyzed using CATIA<br />
Generative Structural Analysis workbench. The structural analysis procedure will<br />
be presented. The visual representations of the numerical results will be analyzed<br />
at both the assembly level and the part level. The assembly level analysis allows<br />
identifying the most critical section of the entire assembly. The part level analysis<br />
allows observing the results for the driving shaft, the driving pin, the sleeve<br />
coupling, the driven pin and the driven shaft.<br />
Key words: sleeve rigid coupling, cylindrical pins, structural analysis,<br />
CATIA.<br />
1. Introduction<br />
The transmission of the movement of rotation and torque between co-axial<br />
shafts can be achieved through couplings. From the category of permanent fixed<br />
couplings in this paper the sleeve rigid coupling with cylindrical pins will be<br />
analyzed. The geometric characteristics of the sleeve rigid coupling with<br />
cylindrical pins are shown in Fig.1. The cylindrical pins can be placed in the<br />
same plane (Fig. 1a), or more often in two different planes at 90 ° (Fig. 1b).<br />
The shafts diameter can be obtained by the preliminary calculation so as to<br />
ensure the transmission of a mechanical power P = 8 kW at the rotational speed<br />
of n = 200 rpm.<br />
With the calculated values for the angular rate ω = 2πn 60 = 20.944 rad/s;<br />
for the torque M t = P ω = 381.9719 Nm and for the safety torque<br />
M tc = kM<br />
M t = 800 Nm (the safety coefficient kM<br />
shall be adopted within
216 Dănuț Zahariea and Mihaela Tudorache<br />
1.5…3, the adopted value is 2.0944), the shafts diameter can be calculated using<br />
the relationship:<br />
(1)<br />
d<br />
3 16M<br />
tc = ,<br />
πτ at<br />
the value d ≅ 0.065m being thus obtained. The material used for all<br />
components is steel with yield strength σ = 2.5 N/m 2 8<br />
⋅10<br />
.<br />
a b<br />
Fig. 1 – Constructive types:<br />
a – with pins in the same plane; b – with pins in two different planes at 90 °.<br />
The geometric characteristics of the sleeve rigid coupling with cylindrical<br />
pins can be determined with the relationship: the outside diameter of the sleeve<br />
coupling D = kDd<br />
= 0.11 m (the coefficient kD<br />
shall be adopted within<br />
1.5…1.8, the adopted value is 1.6923); the length of the sleeve coupling<br />
L = kLd<br />
= 0.2 m (the coefficient kL<br />
shall be adopted within 2…4, the adopted<br />
value is 3.0769); the diameter of the pins ds = ksd<br />
= 0.028 m (the coefficient<br />
ks<br />
shall be adopted within 0.25…0.44, the adopted value is 0.4308) [1], [2].<br />
The sleeve coupling will be checked at the torsional strain with:<br />
(2) t =<br />
4 4<br />
π ( D − d )<br />
02<br />
16DM<br />
tc<br />
τ .<br />
The pins will be checked at the shearing strain with:
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 217<br />
4M<br />
τ = .<br />
tc<br />
(3) f 2<br />
πdds<br />
The obtained values are within the safety range of the torsional strength and<br />
the shearing strength of the material. Even if as a result of these calculations a<br />
good dimensioning of the coupling will be obtained, to answer a series of<br />
additional issues (the torque asymmetry on the coupling parts, the distribution<br />
of critical sections on each coupling part, etc.) should be further analyzed by<br />
advanced numerical methods.<br />
2. The Structural Analysis of the Sleeve Coupling<br />
with Pins in the same Plane<br />
The main steps of the procedure for structural analysis of sleeve couplings<br />
with cylindrical pins in the same plane using CATIA/Generative Structural<br />
Analysis environment are:<br />
1. Creating the 3D model. Shall be carried out separately the five elements<br />
of the coupling using the Part workbench. The appropriate materials are<br />
assigned to each element. The assembly of the five coupling elements is<br />
carried out using the Assembly workbench, Fig. 2a.<br />
2. Configuring the mesh. Shall be carried out separately for each element,<br />
by changing the characteristic parameters “Size” and “Sag”, Fig. 2b.<br />
3. Applying the restraints. On the free extremity of the driven shaft a<br />
clamp condition must be applied, Fig. 2c.<br />
4. Applying the loads. On the free extremity of the driving shaft an 800<br />
Nm torque around the shaft axis will be applied, Fig. 2d.<br />
5. Applying the conditions for the interaction between the five elements of<br />
the sleeve coupling. First, the interaction condition of General Analysis<br />
Connection type will be applied between the driving shaft and the first<br />
pin, between the first pin and the sleeve coupling, between the sleeve<br />
coupling and the second pin and finally, between the second pin and the<br />
driven shaft. Second, for all the active areas in terms of the torque<br />
transmission the shearing strain condition must be specified by setting<br />
the just property of interaction between the active coupling elements<br />
(Bolt Tightening Connection Property), Fig. 2e.<br />
6. Launching the solver, running the numerical analysis and results<br />
visualization. The von Misses stress will be presented for both the<br />
driving and the driven shafts in Fig. 2f (scale factor 500), for both the<br />
cylindrical pins in Fig. 2g (scale factor 700) and, finally for the sleeve<br />
coupling in Fig. 2h (scale factor 500).
218 Dănuț Zahariea and Mihaela Tudorache<br />
3. The Structural Analysis of the Sleeve Coupling<br />
with Pins in two different Planes at 90°<br />
For the sleeve coupling with pins in two different planes at 90 ° the analysis<br />
procedures takes the same steps: creating the 3D model (Fig. 3a), configuring<br />
the mesh (Fig. 3b), applying the restraints (Fig. 3c) and the loads (Fig. 3d),<br />
applying the interaction conditions (Fig. 3e), running the analysis and results<br />
visualization for both shafts (Fig. 3f – scale factor 500), for both pins (Fig. 3g –<br />
scale factor 700) and for the sleeve coupling (Fig. 3h – scale factor 500).<br />
a b<br />
c d<br />
e f<br />
g h<br />
Fig. 2 – Analysis procedure steps for sleeve couplings<br />
with cylindrical pins in the same plane:<br />
a – creating the 3D model; b – configuring the mesh; c – applying the restraints; d –<br />
applying the loads; e – applying the interaction conditions; f – von Misses stress for<br />
shafts; g – von Misses stress for pins; h – von Misses stress for sleeve coupling.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 219<br />
a b<br />
c d<br />
e f<br />
g h<br />
Fig. 3 – Analysis procedure steps for sleeve couplings with cylindrical pins<br />
in two different planes at 90 °:<br />
a – creating the 3D model; b – configuring the mesh; c – applying the restraints; d –<br />
applying the loads; e – applying the interaction conditions; f – von Misses stress for<br />
shafts; g – von Misses stress for pins; h – von Misses stress for sleeve coupling.
220 Dănuț Zahariea and Mihaela Tudorache<br />
4. Conclusions<br />
1. Both constructive solutions comply with the requirements of the<br />
resistance. Thus, the maximum equivalent stress is<br />
7<br />
2. 33⋅<br />
10 Pa for the<br />
coupling with pins in the same plane and 2. 97 ⋅ 10 Pa for the other case. Both<br />
the maximum stresses are in the shearing areas of the driving pin. For the two<br />
shafts, the maximum stress areas are placed near the holes. For the sleeve<br />
coupling the maximum stress areas are placed also near the holes, on the<br />
cylindrical inside surface of the sleeve.<br />
2. The two pins respond differently at the applied loads. Thus, the driving<br />
pin takes a larger effort having a bigger strain. Moreover, the strain profile<br />
along the pins longitudinal axis is different.<br />
3. The mesh configuration parameters have a great influence on both the<br />
hardware resource requirements and the numerical analysis errors.<br />
Received: March 20, 2010<br />
7<br />
“Gh. Asachi” Technical University<br />
Department of Fluid Mechanics, Hydraulic Machines and Drives<br />
Iaşi, România<br />
email: dzahariea@yahoo.com<br />
email: mihaela_tudorache10@yahoo.com<br />
R E F E R E N C E S<br />
1. D e m i a n T., Elemente constructive de mecanică fină. Ed. Didactică şi Pedagogică,<br />
Bucureşti, 1980.<br />
2. D r ă g h i c i I., et al., Calculul şi construcţia cuplajelor. Ed. Tehnică, Bucureşti,<br />
1978.<br />
ANALIZA STRUCTURALĂ A CUPLAJELOR FIXE DE TIP<br />
MANŞON CU ŞTIFTURI CILINDRICE<br />
(Rezumat)<br />
În lucrare se prezintă etapele analizei structurale şi rezultatele unui studiu cu privire<br />
la cuplajele fixe de tip manşon cu ştifturi cilindrice plasate în acelaşi plan, respectiv în<br />
plane decalate la 90°. Analiza efectuată în programul CATIA permite evidenţierea<br />
rapidă a unor aspecte particulare care ar putea fi puse în evidenţă cu dificultate printr-un<br />
calcul clasic de rezistenţa materialelor: asimetria distribuţiei momentului, asimetria<br />
stării de deformare, influenţa concentratorilor de tensiune, poziţia secţiunilor critice<br />
pentru fiecare element al cuplajului. Modificarea factorului de scală şi vizualizarea<br />
animată a tensiunilor echivalente şi a deformaţiilor permit interpretări interesante cu<br />
privire la evoluţia în timp a stării de solicitare şi a tendinţei reale de deformaţie a<br />
cuplajului fix de tip manşon cu ştifturi cilindrice.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
STRUCTURAL ANALYSIS OF A<br />
BIMETALLIC STRIP THERMOSTAT<br />
BY<br />
DĂNUŢ ZAHARIEA and MARIUS STACHIE<br />
Abstract. In this paper the structural analysis of a bimetallic strip will be<br />
analyzed using CATIA Generative Structural Analysis workbench. The structural<br />
analysis procedure will be presented. Six different bimetallic strips have been<br />
considered with titanium as active layer and zinc, aluminum, bronze, brass, copper<br />
and steel as passive layer. The numerical analysis has been performed under the<br />
same global conditions (geometrical characteristics, restraints and thermal load)<br />
using two variable parameters: the width and the length of the beams. For all these<br />
study cases only the thermal load generated by an imposed temperature field has<br />
been considered. Charts are presented for comparative analysis of the influence of<br />
the material, the width and the length of the beams on the von Misses stress and<br />
the maximum deflection.<br />
Key words: thermostat, bimetallic strip, structural analysis, CATIA.<br />
1. Introduction<br />
The bimetallic springs used as sensitive elements of the thermostats are<br />
made by two beams of rectangular cross section bounded together. The two<br />
beams are made of materials with different coefficients of thermal expansion,<br />
α a > α p . Under a uniform temperature field Δ T the bimetallic strip will curve<br />
toward the passive material (the material with lower coefficient of thermal<br />
expansion). The sensitivity of the bimetallic strip will be better as long the<br />
difference α a − α p will be higher. In this paper, the materials used are: titanium<br />
as passive layer and zinc, aluminum, bronze, brass, copper and steel as passive<br />
layer. Characteristics of materials relevant for the analysis are shown in Table 1.<br />
The main geometric characteristics of the bimetallic strip are shown in Fig. 1.
222 Dănuț Zahariea and Marius Stachie<br />
Passive<br />
layer<br />
Table 1<br />
Material characteristics<br />
Active layer<br />
Titanium Zinc Aluminum Bronze Brass Copper Steel<br />
E 10 11 , [Pa] 1.14 0.97 0.7 1.1 1.31 1.1 2<br />
α 10 -6 , [°C -1 ] 9.5 31.2 23.6 17.8 16.7 16.5 11.7<br />
σ ,[MPa] 825 140 95 520 350 290 250<br />
02<br />
Fig. 1 – Geometric characteristics.<br />
In the paper two comparative analyses will be performed. First one is<br />
studying the influence of the beams width b = { 1,<br />
5,<br />
7.<br />
5}<br />
mm for the same beams<br />
length L = 50 mm. The second one is studying the influence of the beams<br />
length L = { 50,<br />
100 , 150}<br />
mm for the same beams width b = 5 mm. The beams<br />
thickness will be the same for all analyzed cases h = h = 1 mm.<br />
2. The Structural Analysis Procedure<br />
The main steps of the procedure for structural analysis of the bimetallic strip<br />
using CATIA/Generative Structural Analysis environment are:<br />
1. Creating the 3D model. Shall be carried out separately the two<br />
rectangular cross section beams using the Part workbench. The<br />
appropriate materials are assigned to each element. The assembly of the<br />
two beams is carried out using the Assembly workbench, Fig. 2a.<br />
2. Configuring the mesh. Shall be carried out separately for each beam, by<br />
changing the characteristic parameters “Size” and “Sag”, Fig. 2b.<br />
3. Applying the restraints. At one side of the bimetallic strip a clamp<br />
condition must be applied, Fig. 2c.<br />
4. Applying the loads. On the both beams a thermal load with ΔT = 30 °C<br />
will be applied, Fig. 2d.<br />
5. Applying the conditions for the interaction between the two beams of<br />
the bimetallic strip. First, the interaction condition of General Analysis<br />
Connection type will be applied between the active layer and the<br />
passive layer. Second, for the active area in terms of the stress/strain<br />
a<br />
p
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 223<br />
transmission (the bounded surface) the stress/strain condition must be<br />
specified by setting the just property of interaction between the active<br />
and the passive layers (Fastened Connection), Fig. 2e<br />
6. Launching the solver, running the numerical analysis and results<br />
visualization. The maximum deflection will be presented for the<br />
bimetallic strip in Fig. 2f (scale factor 100). The von Misses stress will<br />
be shown in Fig. 2g (scale factor 100) for active layer and in Fig. 2h<br />
(scale factor 100) for passive layer.<br />
a b<br />
c d<br />
e f<br />
g h<br />
Fig. 2 – Steps of structural analysis procedure:<br />
a – creating the 3D model; b – configuring the mesh; c – applying the restraints; d –<br />
applying the loads; e – applying the interaction conditions; f – maximum deflection; g –<br />
von Misses stress for the active layer; h – von Misses stress for the passive layer.
224 Dănuț Zahariea and Marius Stachie<br />
3. Comparative Analysis<br />
After a preliminary analysis, the clamped end of the bimetallic strip has<br />
been identified as being critical in terms of stresses. In order to improve the<br />
computational process the local changing of the mesh parameters at the clamped<br />
end is required. Thus, for the critical area the “Size” mesh parameter will be<br />
0.25. The global mesh parameters are „Size”=1 and „Sag”=0.5.<br />
The first set of comparative analyses is determining how the width of the<br />
beams influences the behavior of the bimetallic strip. Thus, for every<br />
combination of materials changing the width of the beams and maintaining<br />
constant the other parameters were obtained the numerical values for maximum<br />
deflection (Fig. 3a) and for the von Misses stress on the active layer (Fig. 3b),<br />
as well as for the passive layer (Fig. 3c).<br />
a<br />
b c<br />
Fig. 3 – The influence of the width of the beams:<br />
a – maximum deflection; b – von Misses stress for active layer;<br />
c – von Misses stress for passive layer.<br />
The second set of comparative analyses is determining how the length of the<br />
beams influences the behavior of the bimetallic strip. Thus, for every<br />
combination of materials changing the length of the beams and maintaining<br />
constant the other parameters were obtained the numerical values for maximum
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 225<br />
deflection (Fig. 4a) and for the von Misses stress on the active layer (Fig. 4b),<br />
as well as for the passive layer (Fig. 4c)<br />
a<br />
b c<br />
Fig. 4 – The influence of the length of the beams:<br />
a – maximum deflection; b – von Misses stress for active layer;<br />
c – von Misses stress for passive layer.<br />
4. Conclusions<br />
1. The width of the beams has a negligible influence on the maximum<br />
deflection of the bimetallic strip. As far as the status of the stresses, it becomes<br />
apparent that with width increasing, the von Misses stresses grow as well in<br />
both the active and the passive layers. For the width of 7.5 mm, the von Misses<br />
stresses are very close to the admissible values, even overcoming for a small<br />
amount in the case of zinc material. Regardless of the value of the width of the<br />
beams, for all combination of materials, the stress in the active layer is greater<br />
than the passive layer.<br />
2. The length of the beams is definitely affects the values of the maximum<br />
deflection. The von Misses stresses are growing as well, but to a lesser extent<br />
than the growth of the maximum deflection. And in this case, the von Misses<br />
stress in the active layer is towards the passive layer.
226 Dănuț Zahariea and Marius Stachie<br />
3. The best combination of materials in terms of deflection is titan for<br />
passive layer and zinc for the active layer. Due to the smaller mechanical<br />
resistance of aluminum and zinc, increase the width of beams over a certain<br />
limit becomes dangerous, especially in the case of accidental temperature<br />
increases.<br />
4. For all cases analyzed the maximum von Misses stress shall be recorded<br />
in the clamped area. Another critical area is the interaction area between the two<br />
beams of the bimetallic strip.<br />
Received: March 15, 2010 “Gh. Asachi” Technical University<br />
Department of Fluid Mechanics, Hydraulic Machines and Drives<br />
Iaşi, România<br />
email: dzahariea@yahoo.com<br />
email: marius_stachie@yahoo.com<br />
R E F E R E N C E S<br />
1. D e m i a n T., Elemente constructive de mecanică fină. Ed. Didactică şi Pedagogică,<br />
Bucureşti, 1980.<br />
2. Z a m a n i N.G., CATIA V5 FEA Tutorials. SDC Publications, 2005.<br />
ANALIZA STRUCTURALĂ A ARCURILOR<br />
BIMETALICE LAMELARE<br />
(Rezumat)<br />
În lucrare se prezintă analiza structurală a unui arc bimetalic lamelar utilizând<br />
mediul de analiză CATIA Generative Structural Analysis. Sunt analizate şase arcuri<br />
bimetalice lamelare având ca material pasiv titanul, iar ca material activ zincul,<br />
aluminiul, bronzul, alama, cuprul şi oţelul. Analiza numerică a fost efectuată pentru<br />
aceleaşi caracteristici globale (dimensiuni geometrice, condiţii pe contur şi sarcini)<br />
folosind doi parametri de control: lăţimea şi grosimea arcurilor lamelare. Sunt<br />
prezentate diagrame comparative pentru analiza influenţei materialului, a lăţimii şi<br />
grosimii arcurilor lamelare asupra deformaţiei maxime şi a stării de solicitare a arcurilor<br />
bimetalice analizate. Principalele concluzii sunt: sensibilitatea maximă se înregistrează<br />
pentru combinaţia de materiale titan-zinc; zonele puternic solicitate sunt încastrarea şi<br />
interfaţa dintre cele două lamele; tensiunile echivalente mai mari se înregistrează în<br />
stratul activ; lăţimea lamelelor influenţează în principal starea de solicitare, în timp ce<br />
lungimea acestora influenţează cu precădere valorile săgeţii maxime a arcului bimetalic<br />
lamelar.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
KNOWLEDGE MANAGEMENT FOR THE CONFIGURATION<br />
IN EARLY PHASES OF COMPLEX CUSTOM PRODUCTS<br />
BY<br />
IRÈNE ALEXANDRESCU, HANS-JOACHIM FRANKE<br />
and THOMAS VIETOR<br />
Abstract. Nowadays, large and saturated markets in developed countries enforce<br />
the production of a broad palette of customized products, which satisfy the customers’<br />
individual needs. In addition, interesting emerging markets together with the constraints<br />
of the current economic crisis lead to further differentiation in terms of product<br />
functionality and allowable costs. Furthermore, global markets also imply many different<br />
standards and norms, lately more strict due to the worldwide environmental concerns. In<br />
order to be effective in producing custom products, companies in the field of mechanical<br />
engineering integrate in their product lifecycle technologies the newest parametric<br />
CAD/CAM systems, as well as different software for the sales and design departments,<br />
such as configuration or calculation tools. However, producing high quality products with<br />
relatively short delivery times requires that a common knowledge base exists between the<br />
enterprise departments and clear processes and interfaces are defined, integrating all the<br />
used systems. This is commonly not the case and therefore unnecessary iterations are<br />
done in order to produce a technically correct offering based on the specific customer<br />
request. This paper discusses the need and some ideas for a better methodology to faster<br />
and better design complex custom products in the field of mechanical engineering. It<br />
explains the need for a structured knowledge base prior to the beginning of the new<br />
custom product design process. This knowledge base offers the necessary data for the<br />
early design phases and connects the sales and design departments by ensuring a correct<br />
and transparent information transfer between the customer requirements, the configuration<br />
software, the used parametric CAD tools and the generated customer specific offers. The<br />
result is a faster, optimized and technically correct basic configuration of custom products<br />
in early design phases.<br />
Key words: complex custom products, early design phases, product configuration,<br />
knowledge management.<br />
1. Introduction<br />
The transfer from industrial to information and knowledge societies<br />
radically transformed the markets and increased customer expectations.
228 Irène Alexandrescu et al.<br />
Globalization and internationalization differentiated the spectrum of offered<br />
goods and increased the need for custom solutions. There are many different<br />
standards and norms coming from saturated markets or interesting emerging<br />
ones, which have to be satisfied with consideration to the massive constraints<br />
additionally imposed by the economic crisis and the environmental concerns.<br />
Product specific knowledge management nowadays is one of the main<br />
factors for the sustainable competitive advantage of a manufacturing company.<br />
It is an important differentiator for the unique core competences of the<br />
company, as well as the essence for developing successful, knowledge based,<br />
market oriented products and services.<br />
In order to achieve customer satisfaction, companies in the field of<br />
mechanical engineering producing custom solutions have to master their<br />
knowledge and generate fast, optimized and technically correct offers for their<br />
products. This paper describes how this can be achieved by using virtualization<br />
and automatization techniques on a methodically structured product knowledge<br />
base.<br />
2. Scattered Knowledge Base in Early Design Phases<br />
In order to deal with the increasing requirements and the competitive<br />
markets, companies use for the generation of the offering of complex custom<br />
products several software tools, such as: Computer Aided Design and<br />
Manufacturing (CAD/CAM) Software, Product Data/Lifecycle Management<br />
(PDM/PLM) solutions, Enterprise Resource Planning (ERP) systems,<br />
configuration tools and calculation programs. Common applications are in the<br />
aircraft [1] or automotive industry [2].<br />
However, through the variety of software, the complexity of the internal<br />
processes is increased. A good overview to the state of the art is shown in the<br />
CIRP Proceedings [3].<br />
Unfortunately, in praxis, the product knowledge base is scattered through<br />
these systems and normally in each development phase the product knowledge<br />
is documented in separate specific systems. The existing PLM/PDM systems<br />
are often not properly used and there is seldom a platform available to<br />
interconnect or translate the different pieces of information, for a global review<br />
or decision taking.<br />
Fig. 1 shows a sequence diagram for the standard processes that take place<br />
in the early development phases for the generation of a customized offer; the<br />
involved departments and their individual software are also displayed.
Customer<br />
(online configurator)<br />
Sales<br />
(configurator, ERP,...)<br />
custom<br />
request<br />
Design<br />
(CAD, FEM, ERP, PDM, MBS,...)<br />
Part Design (e.g. Hydraulic)<br />
(CAD, FEM, calculation, PDM,...)<br />
Purchasing<br />
(ERP,...)<br />
Production<br />
(CAM, NX-Daten,...)<br />
record<br />
request<br />
Legend:<br />
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 229<br />
request<br />
response<br />
analyze<br />
characteristic<br />
request<br />
curves<br />
request response<br />
design<br />
part<br />
consultation<br />
request response<br />
Department<br />
(used software)<br />
clarification<br />
timeline<br />
preliminary offering<br />
schedule<br />
delivery<br />
date<br />
document action<br />
Fig. 1 – Sequence diagram of the standard processes<br />
for the generation of a customized offer.<br />
offering<br />
order<br />
placed<br />
decide<br />
Currently the customer describes to the sales department the requirements<br />
for the desired product. The sales department interprets the customer<br />
requirements and communicates them further to the design department. There is<br />
normally no sufficient common platform for communication and therefore<br />
unnecessary iteration cycles are done. This can be improved through direct<br />
interfaces between the customer and the two departments, by using a common<br />
configuration tool, which generates department specific product information<br />
and documentation. This solution will be discussed further in the next section.<br />
Another common problem that occurs in the processes described in Fig. 1 is<br />
that generally the project information and awareness remain only in a small<br />
group of specialists. They are not reused for later following projects and are<br />
unavailable for new employees. The company specific knowledge base would<br />
help in solving this problem, by documenting existent intellectual property of<br />
the employees into the pre-structured product specific knowledge.<br />
A good example of a holistic knowledge framework to design and<br />
implement a knowledge based organization is presented by B i n n e r in [4]. This<br />
paper concentrates on the early design phases and can be seen as a methodology<br />
for the previous steps taken in knowledge frameworks systems.<br />
It is therefore not necessary to consider the entire company’s knowledge<br />
base, but rather only the product specific knowledge used in early design phases<br />
to generate a correct first offer. Such comprehensive and integrated knowledge<br />
management solutions are currently almost absent in practice.
230 Irène Alexandrescu et al.<br />
3. Suggested Product Configuration Chain<br />
As discussed in [5] by A le x a n d r e s c u and F r a n k e , the solution for a<br />
faster and better generation of offers is a complete and integrated system that<br />
sustains the whole configuration chain, from the client request analysis to the<br />
configured model of the product. The suggested product configuration chain is<br />
shown in Fig. 2.<br />
Fig. 2 – Methodology for faster and better design of high quality<br />
complex customized products.<br />
In order to shorten the offering generation time, both the client request<br />
clarification and the product design process have to be accelerated and the<br />
quality of the transferred information has to be improved. The solution is a<br />
software system that integrates the product data with existing virtual methods in<br />
order to partially automate the configuration. If a customized product is needed,<br />
a completely automated configuration system is impossible. The<br />
algorithmization of the design process is elaborately discussed by F r a n k e in<br />
[6]. However, it is possible to accelerate and optimize the configuration chain<br />
through better requirement definition, product parameter visualization and goal<br />
conflict analysis.<br />
Fig. 3 presents the implementation solution – the main software tool is the<br />
configuration module, which is used by both sales and engineering design<br />
departments in order to configure a custom solution. The output of both<br />
departments is the customized offer information (such as prices, conditions,<br />
related product pictures and site plans) and the customer specific technical<br />
information (such as exact technical description and parametric 3D-CADmodels<br />
as starting point for a complete and precise product configuration. The<br />
product specific knowledge base is the fundamental information source for the<br />
whole configuration process. Its requirements and structure are presented in the<br />
next section.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 231<br />
Fig. 3 – The knowledge based configuration module, as<br />
a main software component of the solution.<br />
4. Knowledge Base Structure<br />
Before defining the company specific knowledge base structure for the<br />
configuration in early phases of a customized product, the following three<br />
concepts need to be defined: data, information and knowledge. The common<br />
interpretation is:<br />
- Data is unprocessed facts and figures without any added interpretation<br />
or analysis.<br />
(e.g. "")<br />
- Information is data that has been interpreted so that it has meaning for<br />
the user.<br />
Information = data + semantic + context<br />
(e.g.: "The document A was updated.")<br />
- Knowledge is a combination of information, experience and insight that<br />
may benefit the individual or the organization.<br />
Knowledge = information + linking<br />
(e.g.: "B was updated after A, so B contains the latest information.")<br />
The above mentioned three different concepts are presented in Table 1,<br />
according to the description made by B o d e n d o r f in [7]: information is just<br />
a formal step to transforming the data into in the knowledge.<br />
Table 1<br />
The different concepts of data, information and knowledge according to [7]<br />
Data Information Knowledge<br />
structured ⇔ mixed<br />
isolated ⇔ connected<br />
context independent ⇔ context dependent<br />
symbols ⇔ cognitive pattern of action
232 Irène Alexandrescu et al.<br />
The context for the interpretation of the data has a strategic role in the<br />
knowledge output. The discussed company specific knowledge base for the<br />
product configuration should therefore consider both involved sales and design<br />
departments context and semantic. It is therefore important to recognize the<br />
different views, as shown in Table 2. Nevertheless, the common goal is to<br />
configure as fast as possible a high quality customized product, with minimum<br />
resources, minimum costs and minimum losses.<br />
Keeping this in mind, the following structure for a common knowledge base<br />
is proposed:<br />
Fig. 4 – Approach for the knowledge base.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 233<br />
Table 2<br />
Key aspects for the sales and design departments<br />
Departments: Sales Engineering design<br />
Specialization: Marketing Engineering<br />
Modules: Functional modules Product parts and assemblies<br />
Software: Configuration tool Virtualization and calculation tools<br />
Goals: Sell the product Design with minimum costs<br />
Degree of content: Sale probability Technical feasibility<br />
In the center of the knowledge base is the abstract representation of the<br />
product structure, from a neutral point of view. This representation is company<br />
and product specific and it represents the basis for all the department specific<br />
derived views. As exemplarily presented on the right side of the picture, the<br />
module Mi in the main structure will contain its technical (e.g. for the design<br />
department), as well as its functional (e.g. for the sales department) description.<br />
The module has additional parameters and logic, for other derived views, as for<br />
example for geometry calculation, validation of a module combination or<br />
visualization of dependencies systems.<br />
Further research is currently done to describe how at early stage goal<br />
conflicts can be recognized and solved based on this structure.<br />
5. Benefits of Using the Presented Knowledge Base<br />
The main benefits of the presented solution and its knowledge base are:<br />
- It reflects the company global view on its product structure definition;<br />
- The collection and preservation of knowledge can be carried out without<br />
strong efforts, as the configuration system is used on a daily basis;<br />
- The sales and design departments are interacting better;<br />
- The customized offers can be faster and technically more accurately<br />
generated;<br />
- There is a methodic, systematic and goal oriented approach towards a<br />
common context for the configuration.<br />
6. Conclusions<br />
1. The need for customized products has increased due to market<br />
globalization, internationalization and customer awareness. For staying<br />
competitive, companies have to provide custom solutions for low prices, high<br />
product quality and short lead times.<br />
2. The knowledge necessary for configuring customized products in early<br />
design phases is often scattered through numerous software tools and separated
234 Irène Alexandrescu et al.<br />
departments. Comprehensive and integrated knowledge base solutions rarely<br />
exist in praxis.<br />
3. A product configuration chain for an accelerated and technically correct<br />
generation of offers was presented. The main modules of the implementations<br />
are the configuration software and its knowledge base.<br />
4. The product specific knowledge base structure for the configuration in<br />
early phases was discussed.<br />
5. The benefits of the presented knowledge base usage were listed.<br />
Acknowledgements. The authors gratefully thank the German Federal Ministry of<br />
Education and Research (BMBF) for supporting the Project KOMSOLV - 'Fast Offer<br />
and Optimized Design for Complex Products with Conflicting Requirements'.<br />
Received: Technische Universität Braunschweig,<br />
Institute of Engineering Design<br />
Braunschweig, Germany<br />
e-mail: i.alexandrescu@tu-bs.de<br />
R E F E R E N C E S<br />
1. S h e h a b E., B o u i n - P o r t e t M., H o l e R., F o w l e r C., Enhancement of<br />
Digital Design Data Availability in the Aerospace Industry. In Proceedings of<br />
the 19th CIRP Design Conference, p. 589-590, 2009.<br />
2. H i l m a n J., P a a s M., H a e n s c h k e A., V i e t o r T., Automatic Concept<br />
Model Generation for Optimization and Robust Design of Passenger cars.<br />
Advances in Engineering Software, 38: 795-801, 2007.<br />
3. K r a u s e F. -L., The Future of Product Development, Proceedings of the 17 th CIRP<br />
Design Conference, Springer, 2007.<br />
4. B i n n e r H. F., Ganzheitiches Wissenskonzept - Wissensframework zur Gestaltung<br />
und Implementierung einer wissensbasierten Organisation. In ZWF Zeitschrift<br />
für wirtschaftlichen Fabrikbetrieb, Carl-Hanser Verlag, München, 103 (2008),<br />
p. 540-543.<br />
5. A l e x a n d r e s c u I., F r a n k e H. -J., Fast Offer and Optimized Design for<br />
Complex Custom Products. 1st Symposium on Multidisciplinary Studies of<br />
Design in Mechanical Engineering, p. 11-12, 2008.<br />
6. F r a n k e H. -J., Untersuchungen zur Algorithmisierbarkeit des Konstruktionsprozesses.<br />
VDI Verlag, 1976.<br />
7. B o d e n d o r f F., Daten- und Wissensmanagement. Springer, p. 2, 2003.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 235<br />
MANAGEMENTUL CUNOAȘTERII ÎN CONFIGURAREA<br />
PRODUSELOR COMPLEXE PERSONALIZATE ÎN FAZELE<br />
INCIPIENTE ALE DEZVOLTĂRII ACESTORA<br />
(Rezumat)<br />
În prezent, pieţele mari şi saturate ale ţărilor dezvoltate orientează fabricația spre o<br />
paletă largă de produse personalizate, în vederea satisfacerii nevoilor clienţilor<br />
individuali. Pieţele emergente interesate, confruntate cu constrângerile crizei economice<br />
actuale, conduc în plus la o diferenţiere a funcţionalității produselor şi a costurilor<br />
admisibile. Pieţele globale implică respectarea de standarde şi de norme diferite, însă<br />
acestea sunt din ce în ce mai stricte, datorită preocupărilor de mediu la nivel mondial.<br />
Pentru a fi eficiente în fabricația de produse personalizate, companiile din domeniul<br />
ingineriei mecanice integrează în tehnologiile ciclului de viaţă al produsului nou sisteme<br />
parametrice de tip CAD/CAM, precum şi multiple aplicații software pentru<br />
departamentele de vânzări şi de proiectare.<br />
Cu toate acestea, fabricația de produse de înaltă calitate cu termen de livrare relativ<br />
scurt impune existența unei baze de cunoştinţe comune tuturor departamentelor<br />
întreprinderii, proceduri clare şi interfeţe bine definite, care să integreze toate sistemele<br />
utilizate. Acest aspect nu se regăsește însă întotdeauna în practică, prin urmare apar<br />
iteraţii inutile ocazionate de elaborarea unei oferte corecte din punct de vedere tehnic, în<br />
baza unei cereri specifice a clientului.<br />
În lucrarea noastră abordăm problema unei noi metodologii, inovative și rapide, de<br />
design personalizat al produselor complexe din domeniul ingineriei mecanice.<br />
Abordarea noastră explică necesitatea structurării unei baze de cunoştinţe încă din<br />
fazele incipiente ale procesului de design al unui nou produs. Această bază de<br />
cunoştinţe oferă datele necesare pentru fazele incipiente de proiectare şi se conectează la<br />
departamentele de vânzări şi de proiectare prin asigurarea unui corect şi transparent<br />
transfer de informaţii între cerinţele clientului, software-ul de configurare, instrumentele<br />
parametrice utilizate în CAD şi oferta specifică generată de client. Rezultatul cercetării<br />
constă în creșterea rapidității, corectarea tehnică și optimizarea configuraţiei de bază a<br />
produselor personalizate încă din fazele incipiente ale dezvoltării acestora.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
RESEARCHES REGAR<strong>DIN</strong>G TO INNOVATIVE<br />
ACTIVITY IN TECHNICAL ENVIRONMENT<br />
BY<br />
PETRU DUŞA and IULIANA LAURA TARANOVSCHI<br />
Abstract. The innovation concept can be concerned in terms of five basic<br />
elements: process, product, human factors, domain and socio-organizational<br />
environment. The study is conceived to stimulate the innovative potential and<br />
combine Bloom’s Cognitive Taxonomy with the creative model of self-managed<br />
team, following an anallogy with Contradictions Matrix for solving the innovative<br />
problems developed by Altschuller. The new model to stimulate the innovative<br />
potential it is described by the Innovative Acting Matrix and by the Innovative<br />
Support Matrix.<br />
Key words: innovation, creative self-managed team model, Bloom’s<br />
taxonomy, Contradictions Matrix, Innovative Acting Matrix, Innovative Support<br />
Matrix.<br />
1. Introduction<br />
In the literature, the word innovation is more favored than the concept of<br />
creativity [1], [6]. Many definitions show the value, the importance and the<br />
time intensity of innovation [8], [12].<br />
Innovation is often described as the value-adding stage of the creativity<br />
process, suggesting a higher sense of value for innovation. Creativity is a precondition<br />
from which innovation develops, is a new realization in practice<br />
supported by creative thinking [8].<br />
From the analysis of the information from literature, the innovation concept<br />
can be concerned in terms of five basic elements: process, product, human<br />
factors, domain and socio-organizational framework. If it could be demystified,<br />
described and modeled the creative process, it would be able to enhance<br />
individual innovative potential and facilitate the innovative process. When the
238 Petru Duşa and Iuliana Laura Taranovschi<br />
innovative process is facilitated, the innovative products are developed. These<br />
are, often, considered to be the result of collaborative teamwork. Increasing the<br />
organizational effectiveness in a specific domain is a primary goal for selfmanaged<br />
work teams. Self-managed teams are so intensively focused on high<br />
performance that often individual needs of the team members can be<br />
overlooked. As a result of the integration of viewpoints, self-managed teams<br />
offer multifunctional definitions and solutions to problems that generate<br />
innovative products or services.<br />
In this paper are explored the cognitive factors which influence the creative<br />
thinking of individuals who work in an organizational team and in an<br />
organization with high creative potential.<br />
1. Conceptual Determination<br />
In order to understand the manner in which the factors involved in the<br />
innovative process operate, it is used as starting point Bloom’s Taxonomy for<br />
educational objectives.<br />
In 1956, Benjamin Bloom, an<br />
educational psychologist, developed<br />
taxonomy for Educational Objectives.<br />
This taxonomy became a key tool in<br />
structuring and understanding the<br />
learning process. The taxonomy,<br />
examines the cognitive domain,<br />
which categorizes and orders thinking<br />
skills and objectives. It is a<br />
continuum of six domains:<br />
knowledge, comprehension,<br />
application, analysis, synthesis,<br />
evaluation. The domains are arranged<br />
Fig. 1 – Bloom’s Taxonomy [3].<br />
in ascending order, each level is<br />
assimilated in the others’ presence,<br />
starting from the basic level -<br />
knowledge thought the higher one – evaluation, as it is represented in the Fig. 1.<br />
These depend one from another can not achieve one domain without other and<br />
are arranged at three levels: cognitive domain, corresponding verbs and<br />
cognitive activity support. Each cognitive domain, that is found in the center of<br />
the model presented in the Fig. 2, are in correspondence with a set of verbs<br />
located in the middle of the model and with a set of nouns situated at their<br />
external level in this model. The set of verbs express how can be acted in order<br />
to attain the corresponding domain. Nouns express the support that it is offered<br />
to perform the activities for each domain.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 239<br />
Fig. 2 – Bloom’s Taxonomy [4].<br />
This paper aims to structure a model in which cognitive elements of<br />
Bloom’s taxonomy can be integrated with another model which is specific for<br />
activities of a team with high innovative potential.<br />
2. Case Analysis<br />
From previous studies in the area of the self-managed team concept, situated<br />
in a research environment from industrial engineering domain, resulted a model<br />
of creative self-managed team. To define this model, we used the following<br />
tools: General Innovation Skills Aptitude Test - GISAT [5], Belbin test to<br />
determine the role in a team [2] and Building Blocks focused on variables of<br />
effectiveness teams [13]. The model was developed on the most significant<br />
correlations between subjects’ scores to the three dimensions: innovative<br />
profile, team roles and team variables.<br />
The scores obtained for subject (S3) identified as Monitor-Evaluator and<br />
subject (S15) identified as Plant from the perspective of the roles assumption<br />
correlate with the scores obtained for first pillar - Generating and Assessing<br />
Ideas from the innovative profile and with the team variables: Openness and<br />
Confrontation, Support and Trust, Cooperation and Conflict (see on Fig. 3, the<br />
quadrant for S15, S3). The scores obtained for subject (S11) identified as<br />
Implementer and subject (S16) identified as Shaper from the perspective of the<br />
roles assumption correlate with the scores obtained for second pillar - Risk-<br />
Taking from the innovative profile and with the team variables: Judicious<br />
Procedures, Appropriate Leadership and Regular Review (see on Fig. 3, the
240 Petru Duşa and Iuliana Laura Taranovschi<br />
quadrant for S11, S16).<br />
The scores obtained for subject (S7) identified as Resource Investigator and<br />
subject (S10) identified as Coordinator from the perspective of the roles<br />
assumption correlate with the scores obtained for third pillar - Relationship<br />
Building from the innovative profile and with the team variables: Individual<br />
Development, Sound Inter-groups Relations and Good Communication (see on<br />
Fig. 3, the quadrant for S7, S10). The scores obtained for subject (S4) identified<br />
as Finisher and subject (S12) identified as Team Worker from the perspective of<br />
the roles assumption correlate with the scores obtained for forth pillar –<br />
Implementing and Turning Ideas into Products, Processes and Services from the<br />
innovative profile and with the team variables: Balanced Function, Clear<br />
Objectives, Organizational Support (see on Fig. 3, the quadrant for S4, S12).<br />
Fig. 3 – The model of self-managed team [9].<br />
In the literature, can be identified a method to stimulate the creativity –<br />
TRIZ “The Theory of Inventing Problem Solving”, comes from Russian<br />
Language “Teoriya Resheniya Izobretatelskikh Zadatch”, developed by<br />
Genrich Altshuller, Russian author. The method was developed after analysis of<br />
almost two million of the world’s most successful patents. TRIZ is a method<br />
which provides resources to access the best solutions to solve the innovative<br />
problems. In this method can be identified four pillars that differentiate this<br />
strategy from other innovative strategies for solving problems [7]:<br />
• Pillar 1: Contradictions – many problem solvers try going directly<br />
from problem to solution through trial and error. Looking at an analogous<br />
standard problem and its associated standard solution is a more efficient
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 241<br />
approach. It was determined that the most efficient innovative solutions grow up<br />
from contradictions. TRIZ Matrix operates with the contradictions principles.<br />
This is designed on the basis of 39 factors organized on rows and columns,<br />
represented in the Fig. 4, and at intersection between rows and columns can be<br />
identified the contradictions. For each contradiction it is presented a set of<br />
principles, from a total of 40, in order to solve it. If the problem is structured on<br />
the contradictions principle, the TRIZ matrix offers sets of innovative<br />
principles. The essential contribution of the user’s matrix is to match the<br />
standard principles to that innovative problem [10].<br />
Fig. 4 – Extract from Contradiction Matrix [7].<br />
• Pillar 2: Ideality – when TRIZ founder studied the patents database, he<br />
discovered that the systems always tend to evolve towards increasing “ideality”.<br />
The ideality part of TRIZ encourages problem solvers to break out of the<br />
traditional “start from the current situation” type of thinking, and start instead<br />
from what is described as the Ideal Final Result (IFR). The simple definition of<br />
IFR is that the solution contains all of the benefits and none of the costs or<br />
“harms” (environmental impact, adverse side-effects, etc) [7].<br />
• Pillar 3: Functionality – this concept is developed by TRIZ through the<br />
fact that it integrates knowledge from physics, chemistry, mathematics,<br />
engineering, management and political science. It is successful largely because<br />
that knowledge is in a form accessible to all users. In this way, the functionality<br />
becomes the connection that makes possible sheering of knowledge between<br />
widely differing industries [11].<br />
• Pillar 4: Use of resources – in TRIZ terms relates to the unprecedented<br />
emphasis placed on the maximization of use everything contained within a
242 Petru Duşa and Iuliana Laura Taranovschi<br />
system. Thereby a resource is anything in the system which is not being used to<br />
its maximum potential. Discovery of such resources then reveals opportunities<br />
through which the design of a system may be improved [7].<br />
3. Case Analysis Results<br />
From the TRIZ analogy exploration, it can be combined the Bloom’s model<br />
presented in Fig. 2 with the self-managed presented in Fig. 3. The role of this<br />
analogical approach is to define a model to stimulate the innovative potential.<br />
In a first stage it is built a matrix structure having as columns the pillars<br />
from the model of self-managed team and as rows the cognitive domains from<br />
Bloom’s Taxonomy. In the cells resulted at the intersection of rows and<br />
columns, are inserted the Bloom’s domain verbs from the relative line. So, it is<br />
obtained a matrix that offers the necessary stimulants for innovative activity.<br />
The matrix is presented in Figure 5 as the Innovative Acting Matrix. The<br />
importance of this matrix is that, those who access it, being in situations marked<br />
by one of the innovative pillars, have the opportunity to explore the innovative<br />
problem in an acting style. The exploration in an acting manner of the<br />
innovative problem is achieved gradually from the base level - knowledge to a<br />
superior level - evaluation.<br />
Fig. 5 – Innovative Acting Matrix.<br />
In the second stage, on the same matrix structure described above, in the<br />
cells from the intersection of rows and columns, are inserted the nouns of the<br />
Bloom’s domain from the relative line. It is obtain a matrix that provides the<br />
necessary support for innovative activity. The matrix is presented in Figure 6 as<br />
the Innovative Support Matrix. The importance of this matrix is that, those who<br />
access it, being in situations marked by one of the innovative pillars, have the<br />
opportunity to receive the necessary resources to explore the innovative<br />
problem.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 243<br />
Fig. 6 – Innovative Support Matrix .<br />
The exploration of innovative problem is realized, by this time, using the<br />
tools provided as support for each acting-innovative stage.<br />
5 Conclusions<br />
1. As a result of this scientific approach, it was designed a model to<br />
stimulate the innovative potential with the support of two innovative matrices:<br />
Innovative Acting Matrix; Innovative Support Matrix.<br />
2. Innovative Acting Matrix represents a matrix that offers the necessary<br />
stimulants for innovative acting (the verbs).<br />
3. Innovative Support Matrix represents a matrix that offers the necessary<br />
support for innovative results (the nouns).<br />
4. Next research will focus on the applicability of these matrices in<br />
situations that require the innovative potential.<br />
Received: February 25, 2010 “Gheorghe Asachi”Technical University,<br />
Department of Machine Manufacturing Technology<br />
Iasi, Romania,<br />
e-mail: itaranovschi@tcm.tuiasi.ro<br />
R E F E R E N C E S<br />
1. A m a b i l e T. M., From individual creativity to organizational innovation. In<br />
Innovation: A cross-disciplinary perspective. (Gronhaug K., Kaufmann G.,<br />
Eds.), Scandinavian University Press, Oslo-Norway, pp. 139-166, 1988.<br />
2. B e l b i n R. M., Management Teams. Butterworth Heinemann Publisher, London-<br />
UK, 1981.
244 Petru Duşa and Iuliana Laura Taranovschi<br />
3. Bloom’s Taxonomy, Consulted on 11.12.2009 at:<br />
http://blogs.wsd1.org/etr/files/blooms_taxonomy.jpg.<br />
4. Bloom’s Taxonomy, Consulted on 15.12.2009 at:<br />
http://mysciencelessons.files.wordpress.com/2009/07/bloomwheel3.gif.<br />
5. C a m p b e l l A., W a t t D., Building innovation capacity one skill at a time. Proc.<br />
of The Conference Board, Montreal-Canada, 2004.<br />
6. C a v a l l u c c i D., TRIZ, the Altshullerian approach to solving innovation<br />
problems. In Engineering design synthesis: Understanding approaces and tools.<br />
(Chakrabarti A., Eds.), Springer Verlag, London, pp. 131-149, 2002.<br />
7. CREAX Innovation Suite, Consulted on 03.12.2009 at :<br />
http://www.creax.com/triztab/creaxsuite31/english/TRIZ/TRIZ.html.<br />
8. L a n d r y C., The creative city: A toolkit for urban innovators, Earthscan<br />
Publications, London, 2001.<br />
9. T a r a n o v s c h i I. L., D u ş a P., N a g î ţ Gh., Exploring creativity in a research<br />
environment. Proc. 14 th International Conference ModTech, Slănic Moldova-<br />
România, 2010.<br />
10. T e r n i n k o J., Z u s m a n A., Z l o t i n B., Systematic Innovation, An<br />
Introduction to TRIZ. CRC Press LLC, USA, 1998.<br />
11. V i n c e n t J., M a n n D., Systematic technology transfer from biology to<br />
engineering. In Philosophical transactions - Royal Society. Mathematical,<br />
physical and engineering sciences. Royal Society Publisher, London, 2002.<br />
12. W e s t A. M., F a r r J. L., Innovation at work. In Innovation and creativity at<br />
work: Psychological and Organizational Strategies, John Willey & Sons, New<br />
York, pp. 3-13, 1990.<br />
13. W o o d c o c k M., Team Development Manual. Gower Publisher, Aldershot, UK,<br />
1989.<br />
CERCETĂRI CU PRIVIRE LA ACTIVITATEA DE<br />
INOVARE <strong>DIN</strong> MEDIUL TEHNIC<br />
(Rezumat)<br />
Din cercetările anterioare asupra elementului inovativ, am ajuns la concluzia că<br />
structura care facilitează dezvoltarea acestuia o regăsim la nivelul echipei autoconduse.<br />
Modelul de echipă autocondusă propus pentru stimularea potentialului inovativ este<br />
susţinut de patru piloni: Generare de Idei, Asumarea Riscului, Construire de Relaţii şi<br />
Implementare. În momentul în care am privit aceste elemente din perspectiva<br />
obiectivelor inovative şi am considerat că inovarea se dezvoltă cu ajutorul gandirii<br />
creative, am utilizat taxonomia definită de Bloom pentru a elabora un model cognitiv<br />
privind stimularea potenţialului inovativ. Elementul strategic care a stat la baza definirii<br />
acestui model este este reprezentat de Matricea Contradicţiilor TRIZ elaborată de<br />
Genrich Altshuller. În urma acestui demers au rezultat două matrici: Matricea Acţiunii<br />
Inovative şi Matricea Suportului Inovativ. Rolul acestora constă în faptul că potentialul<br />
inovativ poate fi stimulat daca sunt asimilate succesiv elementele taxonomice cognitive.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
ADVANTAGES AND WEAK POINTS OF DIFFERENT<br />
REVERSE ENGINEERING (RE) TECHNIQUES<br />
BY<br />
ANDREI MIHALACHE, GHEORGHE NAGÎŢ<br />
and MARIUS-IONUŢ RÎPANU<br />
Abstract. The concept design, as part of the engineering process and as<br />
support for CAD/CAE techniques, allows optimizing the product concept before<br />
manufacturing, with CAM assistance. For some product development processes,<br />
RE allows us to generate surface models by three-dimensional (3D) – scanning<br />
techniques, and as follows, this method allows different parts to be manufactured<br />
in relatively short development period of time. The aim of this paper is to look<br />
over RE techniques and to find possible advantages or weak points of different<br />
scanning systems.<br />
Keywords: Scanning (digitizing), reverse engineering, product design, rapid<br />
prototyping, CAD/CAE/CAM.<br />
1. Introduction<br />
One of the main domains of interest in rapid product development is<br />
represented by machine manufacturing, where each machine contains thousands<br />
of parts which have to be made as quick and cheap as possible, in order to<br />
achieve the prescribed quality [1].<br />
But, by economical means and, of course, due to progress and innovation,<br />
the products do improve, and this leads to major or minor modifications brought<br />
to the machine, thus, many other internal parts have to be changed. The<br />
available time for this changes, has become increasable shorter due to
246 Andrei Mihalache et al.<br />
competiveness, and the requirements push everyone in rapid development. In<br />
such case, Reverse Engineering (RE) techniques are useful. The most important<br />
instruments found in the process, are different scanning systems, which<br />
provides, in a short amount of time, dimensional description in digital concept.<br />
RE has now been accepted as part of contemporary product design and<br />
manufacturing processes. The RE technique is easy defined as the process<br />
which results in creation of a mathematical model from a physical one.<br />
Different cases have different RE requirements, from recovering the<br />
mechanical design information to design based modifications. In case of<br />
extracting the mechanical design information, we will be interested in tolerance,<br />
as opposed to the case of design based extraction where the precedent will be<br />
that of extracting the design intent.<br />
So, the RE methods and techniques are essential because they allow to<br />
capture and digitize an object surface geometry for later use with<br />
CAD/CAE/CAM [1].<br />
2. Product Development Approaches<br />
Nowadays the management of product design can be achieved based on two<br />
methods shown in (Fig. 1), a “conventional approach” and an “unconventional<br />
approach” [1].<br />
The conventional approach in product development using CAD/CAE/CAM<br />
systems, normally starts with geometrical modeling using a CAD system. The<br />
geometric model can be represented by wire frame or as surface or as solid<br />
structure. Generated CAD data, can be exported to a standard format (IGES<br />
points/STL binary, ASCII data, DXF poly line, VDA points or IGES/STL<br />
surfaces) and then imported in the same data format to CAE systems (allowing<br />
numerical model simulation) and/or CAM systems (allowing tools trajectory to<br />
be generated). In a system with a unique database, the design information can<br />
be divided to each application automatically, without manual transfer of data,<br />
each time [3].<br />
In the unconventional approach, we can see that the product development by<br />
the conventional approach is not always applicable when we aim to re-engineer,<br />
simulate or optimize parts/moulds/tools already existing, but without the CAD<br />
data format. So, it will be necessary to apply one technique that allows to<br />
capture part/mould/tool (or prototype) geometry and to generate a numerical<br />
model to be used in CAE and CAM systems. This is called reverse engineering.
Conventional<br />
approach<br />
Client specifications<br />
No<br />
3D – CAD<br />
system<br />
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010<br />
Triangular model<br />
(file format *.stl)<br />
Conceptual model<br />
CAE system<br />
Design<br />
requirements<br />
satisfied?<br />
Yes<br />
CAM system<br />
Final product<br />
(part or tool)<br />
Physical model<br />
(sample or prototype)<br />
Digitizing system<br />
3D geometry<br />
Points cloud<br />
3D coordinates<br />
Reconstruction software for<br />
the model<br />
Unconventional<br />
approach<br />
Surface model 3D<br />
reconstruction<br />
Fig. 1 – Sequences in manufacturing engineered products<br />
(parts/moulds/tools)-adapted [1].<br />
No<br />
247
248 Andrei Mihalache et al.<br />
CMM<br />
with laser<br />
system<br />
CMM<br />
with<br />
touching<br />
probe<br />
Physical model<br />
(sample or prototype)<br />
CNC – milling<br />
machine with<br />
touching probe<br />
Digitizing<br />
software<br />
Points cloud<br />
(3D coordinates)<br />
CNC –<br />
milling<br />
machine with<br />
laser system CT<br />
Computer<br />
tomography<br />
Fig. 2 – Digitizing techniques for 3D-geometries and generated data.<br />
3. Differences between Digitizing and Scanning Processes<br />
3D scanning (digitizing) is the process of gathering data from undefined<br />
three-dimensional surface. During the scanning process, an analogue scanning<br />
probe is commended to move back and forth (contact or non-contact) across the<br />
unknown surface. During the process, the system records information about the<br />
surface in form of numerical data’s – and generates a points cloud matrix (3D<br />
coordinates). The terms of digitizing and scanning are often used to describe the<br />
same process. Traditionally, digitizing refers to the process of taking discrete<br />
points using a touch-sensible probe. However, with the introduction of new<br />
technologies in data capture, such as laser or camera, digitizing term is now<br />
used as general description for the process of data acquisition from undefined<br />
surfaces [2] – [4]. The digital points cloud can be captured using different<br />
digitizing techniques (Fig. 2), classified into two major groups:<br />
‐ Mechanical techniques (by physical contact sensors);<br />
‐ Optical techniques (by non-contact with the object).<br />
Related to the first group, usually they use a coordinate measuring machine<br />
– CMM, or a CNC-milling machine equipped with physical touching probe<br />
sensors (ex. Retroscan or Renscan – Renishaw, UK). Related to the second
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 249<br />
group, we can also use CMM’s or CNC-milling machines but equipped with<br />
laser beam probes (ex. Renishaw, Metris LC Mitutoyo), or associated optical<br />
sensors (ex. CCD cameras) [2].<br />
3. Advantages and Weak Points of Different Scanning Equipments<br />
Scanning<br />
equipments<br />
Advantages Week points<br />
Laser Precise and fast scanning on Z-axis (0.001 mm) Equipment high price<br />
Non-contact method It is not possible to scan reflective<br />
materials<br />
It is possible to scan soft materials (or liquids) Scanning in X- and Y-axis is<br />
inaccurate<br />
(0.035 – 0.060 mm)<br />
It is not possible to scan notches<br />
areas or steep surfaces due to<br />
additional reflections.<br />
Sensibility to air dust<br />
CCD Fast<br />
High price of equipments<br />
cameras It is possible to use two or three cameras Accuracy decreases linear with<br />
simultaneously<br />
camera distance<br />
Insensible to part color<br />
Scanning angle is equal regardless<br />
the shape of part’s surface; in case<br />
of steep angles the measurement in<br />
inaccurate<br />
Non-contact method, it is possible to scan soft By camera scanning, a very sharp<br />
materials<br />
picture is required; simultaneous<br />
scanning of far and closer surfaces<br />
demonstrates linear deviation of<br />
results depending on focal distances<br />
In case of special coaxial lightning it is possible to In case of oily or wet parts the<br />
scan small diameters and high depths on Z-axis measurement in inaccurate<br />
It is possible to scan very small areas: 1 mm 2 (one<br />
scan only needed) – accuracy is above few<br />
micrometers<br />
Scanning on flat surfaces is very fast<br />
Dust causes bad scanning<br />
Contact Very precise on all axis (depends on scanning It’s inadequate for soft materials.<br />
(classic) equipment)<br />
Fast scanning of known geometrical parts Scanning unknown surfaces is either<br />
impossible<br />
inaccurate<br />
or very slow and<br />
Contact<br />
Precise scanning of coins or similar geometries<br />
Manual or automatic scanning possibility<br />
Hand-held scanning equipment is useful for<br />
scanning big products (airplanes, ships, big<br />
machines or devices)<br />
Precise on all axis (0.001 mm) Minimum diameter of stylus is 0.3<br />
digitizers<br />
mm, scanning surface roughness is<br />
(Renishaw)<br />
not possible<br />
Relatively low price It not possible to scan soft materials<br />
It is possible to use different styluses for different Though material of stylus is wear<br />
surfaces<br />
resistant, it’s surface wears in time<br />
Oil, liquid or dust do not interfere with the Scanning speed is lower (compared<br />
scanning process<br />
It is possible to scan unknown steep surfaces<br />
to non-contact systems)
250 Andrei Mihalache et al.<br />
5. Conclusions<br />
1. Product development (parts/moulds/tools) via integrated reverse<br />
engineering represents a fairly new method that is in research and development<br />
phase.<br />
2. This paper aims to reveal some of the highlights and weak points of<br />
existing scanning equipments. We see that as far as accuracy goes, the best<br />
results are obtained by means of contact sensors. Nevertheless, in case of fragile<br />
or soft materials, laser or optical systems, without contact, provide better<br />
results. The ideal scanning device should be the combination of a camera, laser<br />
and contact probe, all, supported by appropriate controller.<br />
Received: March, 12, 2010 “Gh. Asachi”Technical University,<br />
Department of Machine Manufacturing<br />
Iasi, Romania<br />
e-mail: andrei.mihalache@yahoo.com<br />
R E F E R E N C E S<br />
1. H e r b e r t s o n T., Reverse engineering, in: Fourth International Conference on<br />
Industrial Tools ICIT 2003, April 8th–12th, Bled, Slovenia, pp. 419–422.<br />
2. N. N., Scanning Systems for Reverse Engineering, Renishaw Apply Innovation, H-<br />
2000-3120-04-B, Renishaw plc, UK, consulted at 18 November 2009,<br />
http://www.renishaw.com/en/cmm-probes-software-and-retrofits--6329<br />
3. H o n g w e i L., Adaptive patch-based mesh fitting for reverse engineering, State<br />
Key Lab of CAD&CG, Zhejiang University, Hangzhou 310027, China, Received<br />
13 May 2007; accepted 1 October 2007, pp. 3-10.<br />
4. Z e x i a o X., Complete 3D measurement in reverse engineering using a multi-probe<br />
system, Engineering college, Ocean University of China, Qingdao 266071,<br />
People’s Republic of China, Received 7 October 2004; accepted 20 January<br />
2005, Available online 7 March 2005, pp. 1135-1139.<br />
AVANTAJE ŞI PUNCTE SLABE ALE DIFERITELOR<br />
TEHNICI DE INGINERIE INVERSĂ<br />
(Rezumat)<br />
Dezvoltarea de produse (piese/ matrite/ scule) împreună cu ingineria inversă<br />
integrată, este o metodă recentă, aflată în faza de cercetare şi dezvoltare.<br />
Aceasta lucrare prezinta unele posibilitati de utilizare si beneficii ale metodei în<br />
procesul de productie, dar evidenţiază şi avantajele şi punctele slabe ale diferitelor<br />
echipamente de scanare utilizate în prezent.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
THERMOGRAPHY APPLIED TO BONE DRILLING<br />
BY<br />
ROBERTO LOPEZ 1 , MANUEL SAN JUAN 1 , FRANCISCO SANTOS 1 ,<br />
OSCAR MARTÍN 1 and FLORIN NEGOESCU 2<br />
Abstract. Infrared radiation is a form of electromagnetic radiation. All matter<br />
above absolute zero (0 K, -273ºC) emits infrared energy. Thermographic cameras<br />
can be used for measure temperatures in a lot of applications. With this new tool<br />
we will be able to determine the local heating experienced by the bone during the<br />
process of cutting (drilling) for dental implants. The applicability of the temporal<br />
sequence of thermograms for dynamic study of the temperature in drilling<br />
operations to implant dentistry, is shown to be very high. While there are several<br />
aspects to consider when conducting research of this type.<br />
Key words: thermography, bone, drilling.<br />
1. Introduction<br />
There are many studies directed to determine the influence of the cutting<br />
parameters in bone drilling and the production of heat and the increase in the<br />
bone temperature due to this heat. In those studies the temperature was<br />
measured with thermocouples, the most usual way, or through other variables.<br />
There are less studies in the same area that use a more advanced method for the<br />
register of the temperature in bone cutting, the thermology. But there are even<br />
less studies that use a more dynamic method to obtain the temperature and be<br />
able to relate it directly to other variables such as pressure, speed of cutting, etc.<br />
Hereby our study tries to make a contribution in the use of the infrared<br />
thermology. Infrared radiation is a form of electromagnetic radiation like: radio<br />
waves, microwaves, ultraviolet rays, gamma rays, visible light, etc. All of them<br />
emit energy in the form of electromagnetic waves and they travel at the speed of<br />
light. All matter above absolute zero (0 K, -273ºC) emits infrared energy.<br />
Thermographic cameras detect invisible infrared radiation emitted by objects<br />
and they transform it in an image inside the visible spectrum in which the color
252 Roberto Lopez et al.<br />
scale (or greys) shows the different intensities. The intensity of the infrared<br />
radiation depends on the temperature and on the surface characteristics of the<br />
object, the color and the kind of material. Thermographic cameras give a value<br />
of the temperature for every single point, without taking into account that, for<br />
the same temperature, two different materials can emit infrared energy with<br />
very different intensities.<br />
With this new tool we will be able to determine the local heating<br />
experienced by the bone during the process of cutting (drilling) for dental<br />
implants. This way, we can see, frame by frame, the evolution of the<br />
temperature in every single point in the area of study, so every frame<br />
corresponds to a thermography with the data of the temperatures of every single<br />
point of the image. With this data we can study data from heating and cooling<br />
patterns, zone and/or specific peaks and lows.<br />
2. State of the Art<br />
Due to the drilling process, the bone tissue heats up and if the temperature<br />
around the drilling area exceeds the critical limit of 50 ºC, could result in<br />
thermal necrosis, which is the death of the tissue exposed to that high<br />
temperatures. Therefore, the importance of temperature control in this type of<br />
operations results obvious.<br />
Most researchers use conventional temperature measure methods, such as<br />
the placing of thermo couples (A b o u z g i a & J a m e s, 1995), (B a c h u s et al.,<br />
2001), (M a t t e w s et al., 1984). In this way, the temperature record can be<br />
continuous; not on the contact area between the tool and the bone, but in the<br />
gaps set for the thermo couples away from the drilling area.<br />
The measurement of temperature (for example for fever) is very well known<br />
in old and modern medicine. The infrared thermology is a contactless method<br />
for measuring temperature and seems to be very useful in research and everyday<br />
medicine. Everyday there are more and more studies that use thermography for<br />
the measurement of temperature in different process. This way, (W i l d et al.,<br />
2003), determined the temperature in bone drilling for facial human bones for<br />
inplants. They obtain results for temperature peaks although they don´t specify<br />
how they record the thermography at the highest temperature.<br />
Other researchers, like (U d i l j a c k et al., 2007) also use the thermographic<br />
camera to measure the temperature of drilling while using different drill bits.<br />
But neither in this case, nor in other cases like (W i l d et al., 2003), there is a<br />
continuous record of the temperature to be processed. So it is understood that<br />
the evaluation of the highest temperature is done objectively.<br />
The advantages of the infrared thermography in continuous record to<br />
measure the temperature are obvious:
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 253<br />
• There is no delay between the generation of heat and the perception of<br />
the local heat.<br />
• It is possible to measure the temperature over a whole area.<br />
• Dynamic changes in temperature can be seen during the process.<br />
The dynamic register of thermographies allows making a complete study of<br />
the evolution of the temperature during the process.<br />
3. Experimental Set-Up<br />
In this study we have used infrared thermology to determine the cutting<br />
temperature in bone drilling for dental implants. Figure 1 shows the<br />
experimental set-up.<br />
Fig. 1. Experimental set-up<br />
Tools used during the tests are:<br />
• Thermographic camera<br />
A camera Infratec of high-speed thermography was used (ImageIR 5300,<br />
range from 700 to 1 µm; with a high frame rates of up to 250 Hz). A lens with a<br />
focal length of 100 mm was used with a macro lens of 500 mm. The infrared<br />
camera thermography enabled the estimation of temperature increase and<br />
temperature fields at bone drilling. The camera was set at an angle of 30º in<br />
relation to the hole axis (drilling direction), and the temperatures were<br />
determined (Fig. 1).<br />
The results of the thermographic camera are images where each temperature<br />
isotherm is denoted with a different colour, see Fig. 4.<br />
• Surgical motor<br />
Surgic XT Motor System (with a built-in irrigation system). Technical<br />
data: Power 210 W; motor speed: 200 – 40.000 rpm; torque: 50 Ncm. We used
254 Roberto Lopez et al.<br />
a handpiece for the drilling tests to keep it as centred as possible from the load<br />
transducer from which it will hang.<br />
To hold the handpiece we built a device, made from aluminum, which could<br />
both, hold the handpiece and also attach it to the machine that generates the<br />
force, as seen in figure 2.<br />
Fig. 2 – Surgical motor attached to the force generating machine.<br />
• Testing Machine<br />
For the movement generating we used a Shimadzu Precision Universal<br />
Tester machine, model AG-100KNIS-MO. It is a floor model with two<br />
columns, able to work with speed tests of 0.5 µm at 1000 mm/min, according to<br />
the manufacturer. The setting up can be seen in Figure 1.<br />
• Drills<br />
Tool used for drilling the bone is a cobalt drill bit for metals. Such a drill bit<br />
is not very different in shape from the ones used in dental surgery and dental<br />
implants. Also the material it is made of is very similar to the materials used for<br />
the dental drill bits. The difference in temperatures between the drill bits we<br />
used and the dental drill bits should not be very different. The point of this test<br />
is to measure the temperature of bone drilling using thermology, so for that<br />
matter, the drill bits used for the tests should be as good as others. The drill bit<br />
used in the tests is 2.22 mm in diameter. It is cylindrical straight and has two<br />
lips. Drills can be seen in figure 3.<br />
Fig. 3 – Drill used in test.<br />
• Bone<br />
For the tests we have used cow bone, specifically the femur. It is a fresh<br />
bone bought the previous day in a local market. The cow bone characteristics
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 255<br />
are very similar to the ones of the human bone, for that reason cow bones have<br />
been used for the scientific community in multiple tests of this type.<br />
Fig. 4 – Drilling of bone and thermography.<br />
4. Methodology<br />
For the test we used a 2.22 mm in diameter drill bit turning at a speed of<br />
2,000 rpm attached to a handpiece from the surgical motor. For the set up of the<br />
surgical motor we choose the maximum power to avoid a fall in the turning<br />
speed. The handpiece was attached to the tester that made it move down at a<br />
speed of 1.25 mm/second. The drill was made in a single movement until we<br />
reached a depth of 8 mm. This is not the usual way for implantology, which<br />
uses a discontinuous drilling of one or two seconds. This way we could get a<br />
higher temperature in the drill bit and check the efficiency of thermology in<br />
bone drilling.<br />
The thermographic camera was placed in front of the force generating<br />
machine, as seen in figure 1, which gave us a general view of the cutting area.<br />
Over the thermographic camera was placed a macro lens in order to increase the<br />
resolution of the camera in the area of contact between the drill bit and the bone<br />
sample. This setting allowed us to register the temperature in the drilling zone,<br />
which included the temperature of the drill bit and the superficial temperature of<br />
the bone in the critical area, beside the tool. We made several tests where we
256 Roberto Lopez et al.<br />
register the temperatures from the beginning of the drilling till the exit of the<br />
drill bit once reached the 8 mm depth.<br />
Fig. 5. Thermography sequence.<br />
6. Conclusions<br />
These are some of the most relevant results of the thermographies during the<br />
measurement of temperature in bone drilling. These results could be applicable<br />
to the analysis of temperature using a thermographic camera in other tests, for<br />
example, in the drilling of other materials (wood, metals, resins, alloys, new<br />
materials). During the tests it is important to keep the bone at body temperature<br />
(37 ºC) or adjusted to the temperature of the tool to isolate the increase of<br />
temperature during the drilling process. With the use of the thermographic<br />
camera, we can isolate the average temperature of the elements in the test, as<br />
well as the peaks and lows, the highest and lowest puntual temperature, the<br />
highest and lowest temperature of the area around the tool before the drilling, of<br />
the area of the material, etc.<br />
The continuous register of temperature allowed us study the cooling of the<br />
surface following the heating of the surface due to the drilling. We did a linear<br />
regression of the data according to the time getting this adjustment:<br />
(1) Tª = 14.1566 – 0.112166*Time,<br />
with a cooling rate of 0.11º/s. The cooling is due to the deepest layers in the<br />
bone which cool the outer layers heated during previous drillings.<br />
The P-value of the analysis of variance (p
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 257<br />
Since the first contact between tool and bone, the recorded temperature starts to<br />
increase, and in a following study of data, a few points can be made.<br />
Fig. 6 – The fall of temperature.<br />
The starting temperature in the drilling point of the bone is 14.415 ºC. The<br />
temporal sequence of the temperature in the drilling point is shown in this chart:<br />
Table 1<br />
Temperature in the drilling point<br />
0 1 2 3 4 5 6 7 8<br />
14.42 16.96 18.40 21.81 27.61 26.49 32.11 46.18 51.25<br />
The drilling lasted 8.5 seconds. This is the chart for the increase of<br />
temperature according to the time.<br />
Fig. 7 – The increase of temperature.<br />
There is a tendency in the increase of temperature as the drill bit drills the<br />
bone. Some adjustments were made for the analysis of data, and the closest one<br />
was:<br />
(2) Tª = 16.5913 + 0.5518*Time^2.<br />
Against a linear regression, the quadratic function explained up to 95.87% of<br />
the total variability of data (5% more than the linear), making the adjustment
258 Roberto Lopez et al.<br />
more precise. Finally, we can concluse that the applicability of the temporal<br />
sequence of thermograms for dynamic study of the temperature in drilling<br />
operations to implant dentistry, is shown to be very high. While there are<br />
several aspects to consider when conducting research like this.<br />
Received:23.02.2010 1 University of Valladolid-Spain,<br />
Department of Materials Science and<br />
Metallurgical Engineering-Mechanical Engineering<br />
Valladolid-Spain<br />
e-mail: roblop@eis.uva.es<br />
2 “Gheorghe Asachi”Technical University,<br />
Department of Machine Manufacturing Technology<br />
Iaşi, Romania<br />
e-mail: negoescu@tcm.tuiasi.ro<br />
Acknowledgments. This work has been developed thanks to DPI2006-15502-C02-02<br />
Project from the National Program of Design and Industrial Production supported by<br />
MEC (Spain) and FEDER.<br />
R E F E R E N C E S<br />
1. A b o u z i a M. B., J a m e s D. F., Measurements of shaft speed while drilling<br />
through bone. J. Oral Maxillofac Surg, 53 (1995) 1308-1315.<br />
2. B a c h u s K. N.; R o n d i n a M. T., H u t c h i n s o n D.T., The effects of drilling<br />
force on cortical temperatures and their duration: and in vitro study. Medical<br />
Engineering & Physics, 22 (2000) 685-691.<br />
3. M a t t e w s L. S.; G r e e n C. A, G o l d s t e i n S. A., The thermal effects of skeletal<br />
fixation-pin insertion in bone. Journal of Bone & Joint Surgery, 66A, 7 (1984)<br />
1077-1083.<br />
4. U d i l l j a k T.; C i g l a r D. & S k o r i c S., Investigation into bone drilling and<br />
thermal bone necrosis. Advances in Production Engineering& Management, 2<br />
(2007) 103-112.<br />
5. W I l d W.; S c h ü t t e S. R.; P a u H. W.; K r a m p B., J u s t T., Infrared<br />
thermography as a non invasive application for medical diagnostic. Proceedings<br />
of XVII IMEKO World Congress, pp. 1744-1748, June 22–27, 2003, Dubrovnik,<br />
Croatia.<br />
TERMOGRAFIE APLICATĂ ÎN FREZAREA OSOASĂ<br />
(Rezumat)<br />
Radiaţiile în infraroşu sunt o formă a radiaţiei electromagnetice. Camere<br />
termografice pot fi utilizate pentru măsurarea temperaturii într-o mulţime de aplicaţii.<br />
Cu acest instrument, este posibil să se determine zonele locale de încălzire în timpul<br />
procesului de taiere pentru implanturi dentare. Aplicabilitatea termografiei pentru<br />
studiul dinamic al temperaturii în operaţiunile de tăiere a implanturilor stomatologice, se<br />
dovedeşte a fi foarte mare.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
STUDY ON THE OPPORTUNITY OF A NEW<br />
GLASS RECYCLING FACTORY FOR REGIONAL<br />
SUSTAINABLE DEVELOPMENT<br />
BY<br />
OLGA MARINA MONTES 1 and VASILE V. MERTICARU 2<br />
Abstract. The paper presents a theoretical study and business justification<br />
concerning the opportunity for the establishment of a new factory for glass<br />
recycling, respectively for turning waste glass into raw material for manufacturing<br />
new usable glass products. The presented study is part of a research approach<br />
developed in the direction of establishing the economical and technical conditions<br />
for establishing the discussed glass recycling factory in the province of<br />
Valladolid, autonomous community of Castilla y León from Spain, as possibility<br />
to enlarge the number of similar enterprises, for regional sustainable development.<br />
The state of development for the glass recycling industrial sector in Spain is<br />
discussed and the adequacy of the location is analyzed in the paper, together with<br />
the presentation of related conclusions.<br />
Key words: glass recycling, sustainable development, business justification.<br />
1. Introduction<br />
Sustainable Development permanently represents the crucial goal of<br />
mankind and under this imperative, all the human activities, both social and<br />
economical and of course the productive ones must able to assure the harmony<br />
between profit, people and Earth. It is widely accepted that from engineering<br />
point of view, industrial society in general will reach Sustainable Development<br />
only via Sustainable Manufacturing and via Sustainable Design. In this sense,<br />
industrial development activities starting from plant engineering and design<br />
must be compulsory considered in relation to advanced concepts such as Life<br />
Cycle Analysis (LCA) and Extended Producer Responsibility (EPR). The
260 Olga Marina Montes and Vasile V. Merticaru<br />
concepts of Reducing, Reusing, Recycling and Recovery of material resources<br />
are also very important in providing Sustainable Industrial Development.<br />
Of course, the “cradle-to-cradle” way in LCA is nowadays generally<br />
accepted as the only viable solution for providing environmental protection and<br />
avoiding the depletion of natural resources, as long as materials can be reused,<br />
no waste gets produced or otherwise it can be recycled so that no damaging<br />
impacts on the environment are generated within the closed loop of the product<br />
lifecycle, [3], [5], as is shown in Fig. 1.<br />
Off-site recycle<br />
Product<br />
Reuse<br />
or Recycle<br />
Product<br />
use<br />
On-site recycle<br />
“Cradle to cradle”<br />
LCA<br />
Product<br />
marketing<br />
Raw material<br />
extraction<br />
Product<br />
transportation<br />
and storage<br />
Raw material<br />
processing<br />
Product<br />
Packaging<br />
Fig. 1 – LCA based on “cradle-to-cradle” concept.<br />
Product<br />
manufacturing<br />
On the other hand, as a policy measure which comes to outstand the<br />
industrial producer’s crucial role in reducing the environmental impacts of their<br />
products throughout their entire lifecycle, Extended Producer Responsibility<br />
(EPR), encourages the reuse of products and packaging and, not less important,<br />
the Design for Environment practices [5].<br />
In relation to the above mentioned considerations, the paper presents a<br />
theoretical study and business justification concerning the opportunity for the<br />
establishment of a new factory for glass recycling, respectively for turning<br />
waste glass into raw material for manufacturing new usable glass products. The<br />
presented study is part of a research approach developed in the direction of<br />
establishing the economical and technical conditions for establishing the<br />
discussed glass recycling factory in the province of Valladolid, autonomous<br />
community of Castilla y León from Spain, as possibility to enlarge the number<br />
of similar enterprises, for regional sustainable development.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 261<br />
Glass recycling, as industrial activity which comes to replace important<br />
quantities of traditional raw material in new glass products industry with<br />
processed waste glass, can be included under the umbrella of the Cleaner<br />
Production techniques, [4], [5], covering the areas of off-site recycling and<br />
process change, as it is shown in Fig. 2.<br />
Waste<br />
Reduction<br />
Better<br />
housekeeping<br />
Cleaner Production Techniques<br />
“GREEN PRODUCTION”<br />
Product<br />
Modification<br />
Process Change<br />
Process<br />
control<br />
OFF-SITE<br />
Recycling<br />
GLASS<br />
RECYCLING<br />
Technology<br />
change<br />
RECYCLING<br />
Raw material<br />
change<br />
On-site<br />
Recycling<br />
Fig. 2 – Area covered by glass recycling within “Green Production”.<br />
In a plant engineering approach, it is important to analyze the industrial<br />
sector in which the project will belong and to see if it makes sense to think<br />
about the fact of beginning the studies to start it. In order to this, the state of<br />
development for the glass recycling industrial sector in Spain is discussed and<br />
the adequacy of the location is analyzed in the paper, together with the<br />
presentation of related conclusions.<br />
2. The Glass Recycling Sector in Spain<br />
Since the end of the Second World War and particularly during the last<br />
decade, there has been registered an increasing ‘throw away’ tendency.<br />
Statistics show that in the European countries the amount of household waste<br />
has increased to more than 1 kg per person and per day, on average. The<br />
composition of domestic waste varies from country to country, but the tendency<br />
is the same everywhere: more paper, more plastic, more valuable metal and<br />
glass, and less ash. Because of that phenomenon, the number of recycling plants<br />
is also increasing, as response to balance the residuals proliferation.
262 Olga Marina Montes and Vasile V. Merticaru<br />
According to the Spanish Ministry of Environment and Rural and Marine<br />
Environment, each year in Spain approximately 592 kilograms of waste per<br />
inhabitant and per year, [2], is produced. Much of this waste ends up in<br />
landfills, occupying large areas of land and necessary addition to avoid soil<br />
contaminating.<br />
At the present time, in Spain there are a total of 15 plants dedicated to the<br />
glass recycling and in the region of “Castilla y León” there exist only one other<br />
glass recycling plant.<br />
Since the entry into force of the Packaging Act in 1998, all containers are<br />
required to fund and implement a collection system being outlook openrecycling<br />
of packaging they put on the market. This legal obligation can be met<br />
in two ways:<br />
• Individually, by a deposit-refund.<br />
• Collectively, through an integrated management system.<br />
In this context and in order to provide an effective and economical solution<br />
to all employers who package their products mostly in glass, Ecovidrio, [2],<br />
arose as a nonprofit association whose main objective is to ensure the<br />
traceability of glass packaging. In this way, every packaging that enters in the<br />
recycling chain follows the appropriate process to become a completely new<br />
one. Being the only integrated management system that specializes in glass, is<br />
able to reduce costs and provide better service to all companies, large or small.<br />
The 2008 financial year ended with 2455 companies adhered to Ecovidrio, with<br />
107 more than in 2007, to contribute to sustainable development and<br />
environmental care.<br />
In Fig.3 and Fig. 4 there are summarized the most important data on glass<br />
recycling in Spain from 2001 until 2008 [2].<br />
Fig. 3 – Statistic data on glass recycling in Spain from 2001 until 2008 [2].
Rate of recycling, [%]<br />
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 263<br />
Fig. 4 – Evolution of glass recycling rate in Spain from 2001 until 2008.<br />
3. Location Analysis and Justification<br />
For making the decision to locate an industry, we need to know a number of<br />
parameters that help defining the characteristics of possible locations and see<br />
which one bestly fits the needs of industry to be implemented. The location is<br />
an individual choice in which each industry values and ponders the most<br />
appropriate option for its, [1].<br />
The location parameters can be grouped into the following points:<br />
• Parameters concerning the nature of the industry and its classification;<br />
• Parameters concerning the analysis of the geographical;<br />
• Parameters analyzing the urban conditions;<br />
• Parameters relating to the study of environment;<br />
• Parameters relating to human capital and intellectual capital.<br />
From the analysis of the above mentioned points there has been determined<br />
to locate industrial facilities of the plant object of labor in the industrial area of<br />
San Miguel del Arroyo, in the province of Valladolid, autonomous<br />
community of Castilla y León from Spain.<br />
The following considerations, related with each of the above named<br />
sections, provide a synthesis of the selected location.<br />
3.1. Nature of the Industry and its Classification<br />
The council enacted in 1982 the European Community Directive 82/501 on<br />
the prevention of major accidents arising from certain industrial activities. This<br />
standard classifies the considered facility as industrial activity and performing<br />
dangerous quantities manufactured, processed or stored. Moreover, legalization<br />
itself exists in each country and / or region and there may be legislation also<br />
own by municipality. First, is taken into account that the legislation in force in<br />
San Miguel del Arroyo allows the implementation of this industry to be built.
264 Olga Marina Montes and Vasile V. Merticaru<br />
3.2. Analysis of geographical conditions<br />
In this section it has been checked that the chosen area has all the natural<br />
security conditions necessary for any deployment. There have been key issues<br />
such as security of land, about flooding (unlikely), forest fires, erosion and<br />
deforestation. With respect to climate and the presence or absence of surface<br />
water, to implement the factory has no special requirements. With regard to the<br />
existence of communications, industrial estate where the plant will be located<br />
has good links. This is a site near Valladolid, capital of the Community of<br />
Castilla y Leon and also is close to Madrid, capital of Spain, and near the<br />
national road Nº1, the road which connects the center of the country to the<br />
north. Communications are important both for the supply of waste glass and to<br />
deliver the finished products.<br />
In the community Castilla y León there are important industries like<br />
Renault, Michelin, Iveco, Pascual, Firestone, and the glass waste from these<br />
industries should be the materials the plant will recycle. Also, Madrid is near of<br />
the location, and the plant can recycle the waste from the industries of these<br />
regions or from other regions of the country.<br />
3.3. Analysis of urban conditions<br />
Each plot of the considered area has planning conditions that define the<br />
amount and location of the intended buildings. The following data provide some<br />
of the characteristics of the industrial area of the town of San Miguel del<br />
Arroyo.<br />
General Information - Surfaces:<br />
• Total area: 86,810.27 m²;<br />
• Endowment Surface: 5201.32 m²;<br />
• Public open space (green areas): 40,672.98 m²;<br />
• Maximum Plot: 2868.74 m²; Minimum Plot: 177.18 m²;<br />
• Minimum strippable: 175 m².<br />
Other Data:<br />
• Number of plots: 44; Free Plots: 11;<br />
• Access roads: Yes, in concrete pavement;<br />
• Supplies: Yes, municipal network;<br />
• Sanitation: Yes, primary treatment;<br />
• Lighting: Yes; Electricity: Yes; Gas: Yes;<br />
• Telecommunications: Yes.<br />
• Rules planning: Municipal Town Planning Regulations;<br />
• Maximum building height: 7 m and 10 m ridge cornice;<br />
• Setback: 5.00 m and 3.00 m front edge - rear edge.<br />
The previous planning conditions have been of paramount importance when<br />
deciding the final location of the plant.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 265<br />
3.4. Environmental Study<br />
Within this point, there have been considered the political and social<br />
stability of the country or region where the possible location is. In addition, the<br />
security of people working in and for industry should be maximal. With respect,<br />
it must be said that Spain is a safe country without major political conflicts,<br />
social, ethnic or religious, belonging to the European Community, which would<br />
also facilitates free trade and unimpeded across Europe.<br />
The industrial estates in the town of San Miguel del Arroyo is promoted by<br />
an industrial Sodeva (Provincial Development Society of Valladolid), which<br />
brings many advantages and subventions.<br />
It should be noted that the town of San Miguel del Arroyo is close to the<br />
University of Valladolid, which can determine a dynamic and availability of<br />
their students as interns in research and development processes as well as in<br />
other departments.<br />
3.5. Human and Intellectual Capital<br />
There has been taken into account that the area of localization has enough<br />
human capital to meet the staffing needs of the company to implement. The<br />
designed industrial plant requires very few employees, and most workers do not<br />
require a very thorough qualification. Staff will be working in industry and must<br />
have an adequate quality of life for their needs. They have the possibility of<br />
living in a city, Valladolid, with all necessary services available to them, or live<br />
in the town of San Miguel del Arroyo, where housing prices are lower, besides<br />
having all the primary needs; schools, leisure areas, sports areas, shops. These<br />
basic parameters can be framed within what would be called family welfare.<br />
Besides human capital, there must be taken into account the intellectual<br />
capital, i.e. preparation. As already mentioned above, the University of<br />
Valladolid is relatively close to the point of location and can have enough well<br />
prepared people.<br />
4. Conclusions<br />
1. The presented study and the obtained synthesis results have proved the<br />
adequacy of implementing a new glass recycling factory in the selected location<br />
and they are useful for efficient establishment that plant for glass recycling.<br />
2. A manufacturing process design, together with a study of the technical,<br />
human and economic necessary for the plant will complete the best design of<br />
the plan to get a finished product with the required quality and the lowest<br />
possible cost.
266 Olga Marina Montes and Vasile V. Merticaru<br />
Received: February, 20, 2010 1 Universidad de Valladolid,<br />
Valladolid, Spain<br />
e-mail: olga.marina.montes@gmail.com<br />
2 ”Gheorghe Asachi” Technical University,<br />
Department of Machine Manufacturing Technology<br />
Iasi, Romania<br />
e-mail: merticaru@tcm.tuiasi.ro<br />
R E F E R E N C E S<br />
1. C a s a l s-C a s a n o v a, M., Complejos Industriales. Editions UPC, Cataluña<br />
Polytechnic University, 2005.<br />
2. Ecovidrio - Informe anual 2008, Available from: - http://www.ecovidrio.es/app/<br />
GeneraPaginas.asp?seccion=../app/WebEcovidrioNet/wEstadisiticasRecicladoTo<br />
talNacional.aspx, Accessed: 02/02/2010.<br />
3. E l-H a g g a r, S. M., Sustainable Industrial Design and Waste Management: Cradleto-cradle<br />
for Sustainable Development. ISBN-10: 0123736234, ISBN-13:<br />
9780123736239, 406 p, Publisher: Academic Press, 2007. Available from:<br />
http://www.engineeringvillage.com. Accessed: 2008-02-02.<br />
4. G a l i s M. et al., Digital product development for the entire product life cycle.<br />
Academic Journal of Manufacturing Engineering, 6, 3 (2008), Timisoara,<br />
Romania, p.55-60<br />
5. M e r t i c a r u V.jr., M u s c a G., A x i n t e E., PLM in Relation to SCM and CRM,<br />
for Integrating Manufacturing with Sustainable Industrial Design, Proceedings<br />
of ICOVACS 2008: International Conference On Value Chain Sustainability,<br />
Izmir, Turkey, November 12-14, 2008, Izmir University of Economics<br />
Publication No: IEU-026, ISBN 978-975-8789-25-2, pp.109-118.<br />
STUDIU ASUPRA OPORTUNITĂŢII UNEI NOI FABRICI<br />
DE RECICLARE A STICLEI, PENTRU O DEZVOLTARE<br />
REGIONALĂ DURABILĂ<br />
(Rezumat)<br />
În lucrare este prezentat un studiu teoretic şi justificarea din punct de vedere<br />
antreprenorial privind oportunitatea înfiinţării unei fabrici noi de reciclare a sticlei, respectiv<br />
pentru transformarea deşeurilor de sticlă în materie primă pentru fabricaţia de produse noi<br />
din sticlă. Studiul prezentat este parte a unui demers de cercetare dezvoltat în direcţia<br />
stabilirii condiţiilor tehnice şi economice pentru înfiinţarea fabricii de reciclare a sticlei în<br />
discuţie, în provincia Valladolid din comunitatea autonomă Castilla şi Leon din Spania, ca<br />
posibilitate de creştere a numărului de astfel de întreprinderi similare, pentru o dezvoltare<br />
regională durabilă. În lucrare, se discută stadiul de dezvoltare a sectorului industrial de<br />
reciclare a sticlei în Spania, se analizează adecvanţa localizării investiţiei şi sunt prezentate,<br />
de asemenea, concluziile aferente.
<strong>BULETINUL</strong> <strong>INSTITUTULUI</strong> <strong>POLITEHNIC</strong> <strong>DIN</strong> <strong>IAŞI</strong><br />
Publicat de<br />
<strong>Universitatea</strong> Tehnică „Gheorghe Asachi” din Iaşi<br />
Tomul LVI (LX), Fasc. 2, 2010<br />
Secţia<br />
CONSTRUCŢII DE MAŞINI<br />
MODELING 3D SURFACE OF TEXTILE STRUCTURES<br />
FOR FLUID FLOW IMPROVEMENT<br />
BY<br />
CĂTĂLIN DUMITRAȘ 1 , CARMEN LOGHIN 1 , SULEYMAN YALDIZ 2 ,<br />
MEHMET SAHIN 2 and LUMINIȚA CIOBANU 1<br />
Abstract. The efficiency of thermal insulation in a windy environment can be<br />
increased by using adequate protective equipment. Following a FEA analysis<br />
regarding the flow of fluids (air) at body surface, the presence of critical zones<br />
with turbulences was detected. The critical zones are the ones with a change in<br />
geometry, such as the shoulders, hips or hands. This generates the idea that<br />
something has to be done to improve the fluid flow at surface level of the exterior<br />
garment layer. The main problem is to ensure a certain path for the fluid<br />
movement, so that the flow will become laminar. A simple and feasible solution is<br />
creating a textile structure with an architecture that includes these paths. Different<br />
variants of 3D geometries are proposed. Their dynamic behaviour is studied in<br />
order to rank the proposed fabricse abstract of the paper is to be written here. It<br />
contains the main ideas and original contributions and conclusions of the authors’<br />
research.<br />
Key words: laminar air flow, weft knitted fabrics with relief effects, sandwich<br />
fabrics, FEA, meshed models, fluid flow.<br />
1. Introduction<br />
The problem of thermal protection in hostile environments is certainly not<br />
new, but it does not loose its interest due to the fact that there are large range of<br />
activities requiring such protective equipment, from winter and extreme sports,<br />
cycling, motorcycling to working in harsh winter conditions. There are a lot of<br />
studies regarding developments of textile fabrics and materials to improve<br />
thermal insulation. Still, the fluid flow around the garments in dynamic<br />
conditions is less investigated and the problem of designing textile structures<br />
with improved behaviour in relation to air flow remains open to debate.
268 Cătălin Dumitraș et al.<br />
In order to formulate solutions for the fabric structure one must understand<br />
the nature of the problem. The conditions characterising a hostile environment<br />
include among others strong winds. The flow of these air currents is different,<br />
according to the type of surface they encounter. Larger surfaces, with a planar<br />
distribution present a laminar flow of the air currents, while complex 3D<br />
surfaces are characterised by the formation of pressure fields and turbulences<br />
that affect the thermal behaviour of the exterior layer in the garment by<br />
reducing its insulation. Areas like the arms, hips or shoulders, with smaller 3D<br />
surface present problems in relation to the laminar air flow. Previous studies<br />
regarding the air flow around a human body show that these are the critical<br />
zones in a protective garment. Due to its specific nature, the shape of these areas<br />
in a protective garment cannot be modified. Therefore, the improvement of air<br />
flow must be based on modifying the fabric surface in the critical areas so that<br />
the pressure fields are diminished. The fabric surface has to be designed so that<br />
the air currents are channeled toward the exterior, avoiding turbulences. From<br />
this point of view, the 3D knitted fabrics represent the best solution. Apart from<br />
the shape of their cross section, it is important to define the optimum paths for<br />
these channels and their position in the protective garment.<br />
2. Definition of Air Flow at Garment Level<br />
An initial analysis was carried out in order to identify the air flow critical<br />
areas on the human body. Its purpose was to determine the air pressure and<br />
velocity distribution within the fluid. The model represented therefore a fluid<br />
mesh and not one of the solid body. It was created with the ALGOR v12 FEA<br />
software package, using a number of 491 2D elements (558 nodes), defined in<br />
Fig. 1. The FEA model is presented in Fig. 2. The body is considered an<br />
obstacle, while the environment is meshed using bidimensional finite elements.<br />
Fig. 1 – The type of FEA element used. Fig. 2 – Finite element based model.<br />
It was presumed that a pressure field caused by winds up to 100 km/h is<br />
applied to the model, corresponding to all human body. The selected properties<br />
of the environment are illustrated in Fig. 3.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 269<br />
Fig. 3 – Physical properties of the environment.<br />
The analysis results are presented as pressure fields, velocity and<br />
turbulences. The output variables were calculated for different values of the<br />
input parameters. The wind was considered to blow with 90 km/h velocity for a<br />
period of 90 seconds. The results are presented in Figs. 4 to 6.<br />
t = 15 s t = 80 s t = 3 s t =80 s<br />
Fig. 4 – Pressure fields<br />
distribution<br />
t = 3 s t = 70 s<br />
Fig. 6 – Turbulences distribution<br />
Fig. 5 – Velocity fields distribution<br />
The images presented show that the turbulences in the air flow, as well as<br />
higher velocities are characteristic to some parts of human body – the shoulders,
270 Cătălin Dumitraș et al.<br />
the hips and the chest. Such critical areas are susceptible to generate discomfort<br />
in windy conditions.<br />
A solution must be found to diminish these turbulences and the pressure<br />
fields encountered. From the fabric point of view, the air flow at surface level<br />
can be controlled through: the type of raw material, fabric structure and<br />
structural parameters. The most important factor is the fabric structure, mainly<br />
the surface geometry. Weft knitted fabrics offer a large range of structural<br />
possibilities and are characterised by good elasticity and high formability. In<br />
comparison with other types of fabrics, especially woven, the weft knitted<br />
fabrics can be produced with complex 3D architectures that are specific only to<br />
them and do not require costly technological changes on the machines.<br />
3. Models For The Air Flow At Shoulder Level<br />
3.1. Definition of the Models<br />
Two models were created using AutoCAD 2007, considering the human<br />
shoulder. The first model represented the shoulder without a modified surface<br />
and is illustrated in Fig. 7. This model continues the 2D analysis presented<br />
above. The other model considers the surface modified by the presence of the<br />
fabric. The fabrics considered had a rectangular shaped cross section as<br />
discussed above and the resulting model is presented in Fig. 8.<br />
The section height is 10 mm. Lower height values proved to be inefficient.<br />
The rectangular channels are spaced also at 10 mm. The modelled shoulder was<br />
included in a space volume in order to simulate accurately the air flow.<br />
Fig.7 – Model without 3D surface effect<br />
Fig.8 – Model with channels with<br />
rectangular cross section<br />
The models were transferred to the ALGOR v20 FEA software package to<br />
be analyzed. The model was meshed automatically resulting a number of 96,584<br />
3D elements (bricks) and a number of 108,512 nodes. The fluid considered is<br />
the air, its flow velocity chosen to be 50 km/h. This simulates an extremely<br />
hostile environment.
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 271<br />
The fluid properties were selected from the Material Library Manager and<br />
are presented in Fig. 9. The meshed model for the shoulder is illustrated in<br />
Fig.10, while the meshed model for the shoulder and 3D fabric is presented in<br />
Figure 11.<br />
Fig. 9 – Selected air properties Fig. 10 – Meshed model<br />
Fig. 11 – Meshed model for shoulder and fabric<br />
The models were analyzed using finite elements, based on steady fluid flow<br />
processor. The parameters considered are illustrated in Figs. 12 and 13.<br />
Fig. 12 – Parameters for steady fluid flow Fig. 13 – Considered load curve
272 Cătălin Dumitraș et al.<br />
3.2. Results and Discussion<br />
The following results were obtained based on the analysis presented above.<br />
The pressure fields at shoulder level were identified for the shoulder without<br />
fabric, as presented in Fig. 14. This constitutes a reference base for all models<br />
containing the 3D fabrics. From Fig. 14 it results that there is higher pressure<br />
field at the back of the shoulder, possibly caused by a suction effect.<br />
Fig. 14 – Distribution of pressure fields at shoulder level.<br />
The distribution of air velocity at shoulder level is presented in Fig. 15 and<br />
16. High velocity values are identified at the entire shoulder level, rest of the<br />
human body these values are lower. That indicates that the turbulences are<br />
higher in the shoulder area.<br />
Fig. 13 – Air velocity distribution,<br />
general view<br />
Fig. 14 – Air velocity distribution<br />
The use of a 3D fabric with rectangular channels changes the air flow. The<br />
velocity distribution is different in comparison to the situation described for the<br />
shoulder without fabric. The areas characterized by high air velocity are<br />
distanced from the shoulder areas suggested by Figs. 15, 16 and 17.
Fig. 15 – Air velocity distribution for<br />
shoulder and 3D fabric – general view<br />
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 273<br />
Fig. 16 – Air velocity distribution for shoulder<br />
and 3D fabric – bottom view<br />
Fig. 17 – Air velocity distribution for shoulder and 3D fabric – top partial view<br />
From Figs. 15-17 it results that the air flow is improved, diminishing the air<br />
velocity at shoulder level and implicitly the air turbulences.<br />
4. Conclusions<br />
1. FEA studies concerning protective equipment for cold environments have<br />
shown that the air flow at body level presents critical areas where the flow is<br />
characterized by pressure fields and turbulences, causing heat loss and<br />
discomfort. It is the case of hips, chest and shoulders.<br />
2. One solution for the improvement of the air flow is using fabrics with 3D<br />
structure or 3D surface effects. These knitted fabrics are produced on flat<br />
knitting machines without any adaptation required. The paper presents two<br />
structural variants with 3D effects and 3D architecture – rib fabric with relief<br />
effects and sandwich fabrics with complex shapes. The fabrics are characterized<br />
and their 3D structure is discussed, emphasizing their advantages in obtaining<br />
surface effects that have the potential to control the air flow.
274 Cătălin Dumitraș et al.<br />
3. Two FEA studies were carried out, one at 2D level, in order to identify<br />
the air flow critical areas. The 2D analysis does not ensure complete data<br />
regarding the in depth air flow so a further 3D finite element analysis was<br />
required. This analysis indicated the flow phenomenon dimensions at shoulder<br />
level and allowed a comparison between air flow at shoulder level and the air<br />
flow when a fabric is added.<br />
4. The study took into consideration two models of the shoulder: without<br />
and with fabric. The fabric selected was the sandwich fabric with rectangular<br />
channels with an uniform distribution.<br />
5. The study shown that the use of this fabric brought an improvement in air<br />
flow when used at shoulder level. The study requires additional work taking<br />
into account the followings: fabrics with other 3D effects (differently shaped<br />
channels – circular) and non-uniform distribution of these channels (patterned<br />
distribution of the 3D effect). One important target is to determine the shape of<br />
the channel path. These studies will lead to the optimization of garments used<br />
for protection against cold.<br />
Received: March 20, 2010 1 Technical University “Gheorghe Asachi”,<br />
Department of Machine Tools<br />
Iasi, Romania,<br />
e-mail: dumitrascatae@yahoo.com<br />
2 Selcuk University,<br />
Department of Mechanics<br />
Konya, Turkeyt<br />
e-mail: syaldiz@yahoo.com<br />
R E F E R E N C E S<br />
1. d e A r a u j o M., F a n g u e i r o R., H o n g H. – Texteis technicos, volume I,<br />
Braga, Portugal, 2000<br />
2. H o l m e r I. – Textiles for Protection Against Cold, in Textiles for Protection,<br />
edited by R.A. Scott, Woodhead Publishing Ltd, Cambridge, UK, 2005, p. 378 –<br />
397<br />
3. L o g h i n C., C i o b a n u L., D u m i t r a s C. – Study regarding the conventional<br />
functions analyze and intelligent functions synthesis for the protective equipments<br />
for aggressive environments‘ research report, CEEX 105/2006<br />
4. L o g h i n C., C i o b a n u L., D u m i t r a s C. – ‘Fundamentals of physical and<br />
chemical processes at the interface level between aggressive environment and<br />
protective equipment‘ research report, CEEX 105/2006<br />
5. Z i e n k i e w i c i O. C. – The Finite Element Method for Fluid Dynamics,<br />
Elsevier, Amsterdam Netherland, 2006
Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 275<br />
MODELAREA 3D A SUPRAFETELOR TEXTILE CU DESTINAŢIA<br />
OPTIMIZĂRII CURGERII FLUIDELOR<br />
(Rezumat)<br />
Eficienţa protecţiei termice într-un mediu cu curenţi puternici de aer (vânt) poate fi<br />
îmbunătăţită printr-un echipament de protecţie adecvat. Ca urmare a unei analize cu<br />
elemente finite a curgerii fluidelor (aerului) pe suprafaţa corpului detectarea yonelor<br />
critice turbionare poate fi realizată. Aceste zone au fost detectate în locurile în care se<br />
modifică geometria corpului cum ar fi umeri, brate, etc. Concluziile acestei analize au<br />
condus la ideea că se poate obţine o îmbunătăţire a coeficientului de frecare a aerului<br />
prin modificarea geometriei echipamentului textil. Se pune problema determinării unei<br />
geometrii convenabile a acestui echipament astfel încât săse micşoreye sau să dispară<br />
zonele de curgere turbionară. Se propun mai multe variante de structuri textile cu<br />
geometrie 3D. Se realizează apoi un studiu destinat determinării profilului optim al<br />
structurii textile.
A<br />
Alexandrescu Adina 189<br />
Alexandrescu Adrian 189<br />
Alexandrescu Irène 227<br />
Alexandrescu Aurora 189<br />
Antoniadis Aristomenis 9,21<br />
B<br />
Baciu Lupașcu Radu 91<br />
Bălășoiu Victor 197<br />
Băran Marian 197<br />
Bărglăzan Mircea 205<br />
Bădărău Rodica 205<br />
Belis Taxiarchis 9<br />
Berbinschi Silviu 41,49<br />
Bordeașu Ilare 197<br />
Bostan Ion 135<br />
C<br />
Calfa Daniel 161<br />
Ciobanu Luminița 267<br />
Călărașu Doru 153<br />
Chiriță Constantin 161<br />
Ciobanu Bogdan 153<br />
Cozmîncă Irina 97<br />
Cozmîncă Mircea 65<br />
Croitoru Cristian 65<br />
D<br />
Dobîndă Eugen 205<br />
Dogaru Constantin 31<br />
Dulgheru Valeriu 135<br />
Dumitraș Cătălin 267<br />
Dumitrașcu Nicolae 41,49<br />
Dumitrescu Cătălin 167<br />
Dumitrescu Liliana 167<br />
Dumitru Dumitru 31<br />
Dușa Petru 91,237<br />
F<br />
Fetecău Cătălin 57<br />
Filip Cristina 117<br />
Franke Hans Joachim 227<br />
Fulga Ileana 143<br />
G<br />
Gonçalves Coelho Antonio 1<br />
H<br />
Hanganu Adrian 161<br />
Haraga Georgeta 167<br />
Horodincă Mihăiță 125<br />
Index autori<br />
I<br />
Ibănescu Radu 97<br />
L<br />
Loghin Carmen 267<br />
Lopez Roberto 251<br />
M<br />
Manole Iolanda 83<br />
Murad Erol 167<br />
Mănescu Mihai 177<br />
Marsavina Liviu 197<br />
Miloș Teodor 205<br />
Manea Adriana 205<br />
Mihalache Andrei 245<br />
Martín Oscar 251<br />
Montes Marina Olga 259<br />
Merticaru V. Vasile 259<br />
Matei Ana Maria 75<br />
Milea Marius Nicolae 75<br />
Mocanu Costel 57<br />
Mourao Antonio 1<br />
N<br />
Nagîț Gheorghe 83,245<br />
Neagoe Lavinia 117<br />
Novac Dragoș 197<br />
Negru Radu 197<br />
Negoescu Florin 251<br />
Neștian Gabriela 1<br />
O<br />
Oancea Nicolae 41,49<br />
P<br />
Panaitescu Valeriu 167<br />
Pavlov Olimpia 31<br />
Petre Dan 117<br />
Petre Ioana 117<br />
Popoviciu Mircea 197<br />
R<br />
Rîpanu Marius Ionuț 83,245<br />
Rotman Iustina 91<br />
S<br />
Sahin Mehmet 267<br />
San Juan Manuel 251<br />
Santos Francisco 251<br />
Scurtu Dan 153<br />
Sochireanu Anatol 135
Stachie Marius 221<br />
Storm Kjærside Birgit 105<br />
Străjescu Eugen 31,143<br />
Stroiță Daniel 205<br />
T<br />
Tapoglou Nikolaos 21<br />
Teodor Virgil 41,49<br />
Tița Irina 153<br />
Tudorache Mihaela 215<br />
Taranovschi Iuliana 237<br />
U<br />
Ungureanu Cătălin 65,97<br />
V<br />
Vietor Thomas 227<br />
Vlad Daniel Viorel 57<br />
Vodă Mircea 197<br />
W<br />
Weingold Andrei 83<br />
Y<br />
Yaldîz Suleyman 267<br />
Z<br />
Zahariea Dănuț 215, 221