BULETINUL INSTITUTULUI POLITEHNIC DIN IAŞI - Universitatea ...

BULETINUL 

INSTITUTULUI 

POLITEHNIC 

DIN IAŞI 

Publicat de 

UNIVERSITATEA TEHNICĂ „GHEORGHE ASACHI”, IAŞI 

Tomul LVI (LX) 

Fasc. 2 

Secţia 

CONSTRUCŢII DE MAŞINI 

2010

BULETINUL INSTITUTULUI POLITEHNIC DIN IAŞI 

PUBLISHED BY 

”GHEORGHE ASACHI” TECHNICAL UNIVERSITY OF IAŞI 

Editorial Office: Bd. D. Mangeron 63, 700050, Iaşi, ROMANIA 

Tel. 40-232-278683; Fax: 40-232-237666; e-mail: polytech@mail.tuiasi.ro 

Editorial Board 

President: Prof. dr. eng. Ion Giurmă, Member of the Academy of Agricultural 

Sciences and Forest, Rector of the“Gheorghe Asachi” Technical University of Iaşi 

Editor-in-Chief: Prof. dr. eng. Carmen Teodosiu, Vice-Rector of the 

“Gheorghe Asachi” Technical University of Iaşi 

Honorary Editors of the Bulletin: Prof. dr. eng. Alfred Braier, 

Prof. dr. eng. Hugo Rosman, 

Prof. dr. eng. Mihail Voicu, Corresponding Member of the Romanian Academy, 

President of the “Gheorghe Asachi” Technical University of Iaşi 

Editors in Chief of the MACHINE CONSTRUCTION Section 

Prof. dr. eng. Radu Ibănescu, Assoc. Prof. dr. eng. Aristotel Popescu 

Honorary Editors: Prof. dr. eng. Gheorghe Nagîţ, Prof. dr. eng. Cezar Oprişan 

Associated Editor: Prof. dr. eng. Eugen Axinte 

Editorial Advisory Board 

Prof. dr. eng. Nicuşor Amariei, „Gheorghe Asachi” Prof. dr. eng. Nouraş-Barbu Lupulescu, University 

Technical University of Iaşi, Romania 

Transilvania of Braşov, Romania 

Assoc.Prof.dr.eng. Aristomenis Antoniadis, Technical Prof. dr. eng. Francisco Javier Santos Martin, 

University of Crete, Greece 

University of Valladolid, Spain 

Prof. dr. eng. Virgil Atanasiu, „Gheorghe Asachi” 


Prof. dr. eng.Fabio Miani, University of Udine, Italy 

Prof. dr. eng. Manuel San Juan Blanco, University of Prof. dr. eng.Mircea Mihailide, „Gheorghe Asachi” 

Valladolid, Spain 


Prof. dr. eng. Petru Berce, Technical University of Cluj Prof. dr. eng. Sevasti Mitsi, Aristotle University of 

Napoca, Romania 

Thessaloniki Salonic, Greece 

Prof. dr. eng. Ion Bostan, Technical University of Prof. dr. eng. Gheorghe Nagîţ, „Gheorghe Asachi” 

Chişinău, Repablic of Moldova 


Prof. dr. eng. Walter Calles, Hochschule für Technik und Prof. dr. eng.Vasile Neculăiasa, „Gheorghe Asachi” 

Wirtschaft des Saarlandes, Saarbrücken, Germany 


Prof. dr. eng. Doru Călăraşu, „Gheorghe Asachi” Prof. dr. eng. Dumitru Olaru, „Gheorghe Asachi” 



Prof. dr. eng. Francisco Chinesta, Ecole Centrale de Prof. dr. eng. Cezar Oprişan, „Gheorghe Asachi” 

Nantes, France 


Assoc.Prof.dr.eng. Conçalves Coelho, University Nova of Prof. dr. eng. Juan Pablo Contreras Samper, 

Lisbon, Portugal 

University of Cadiz, Spain 

Assoc.Prof.dr.eng. Mircea Cozmîncă, „Gheorghe Asachi” Prof. dr. eng.Loredana Santo, University „Tor Vergata”, 


Rome, Italy 

Prof. dr. eng. Spiridon Creţu, „Gheorghe Asachi” Prof. dr. eng.Cristina Siligardi, University of Modena, 


Italy 

Prof. dr. eng. Gheorghe Dumitraşcu, „Gheorghe Asachi” Prof. dr. eng.Fernando José Neto da Silva, University 


of Aveiro, Portugal 

Prof. dr. eng. Cătălin Fetecău, University „Dunărea de 

Jos” of Galaţi, Romania 

Prof. dr. eng. Filipe Silva, University of Minho, Portugal 

Prof. dr. eng. Mihai Gafitanu, „Gheorghe Asachi” Prof. dr. eng. Laurenţiu Slătineanu, Technical 


University of Iaşi, Romania 

Prof. dr. eng. Radu Gaiginschi, „Gheorghe Asachi” Lecturer dr.eng. Birgit Kjærside Storm, Aalborg 


Universitet Esbjerg, Denmark 

Prof.dr.ir.Dirk Lefeber, Vrije Universiteit Brussels, Belgium Prof. dr. eng. Ezio Spessa, Politecnico di Torino, Italy 

Prof. dr. eng.Dorel Leon, „Gheorghe Asachi” Technical Prof. dr. eng. Alexei Toca, Technical University of 

University of Iaşi, Romania 

Chişinău, Repablic of Moldova 

Prof. dr. eng.James A. Liburdy, Oregon State University, Prof. dr. eng.Roberto Teti, University „Federico II”, 

Corvallis, Oregon, SUA 

Naples, Italy 

Prof. dr. eng. dr. H.C. Peter Lorenz, Hochschule für Prof. dr. eng.Hans-Bernhard Woyand, Bergische 

Technik und Wirtschaft, Saarbrücken, Germany 

University Wuppertal, Germany

Papers presented at 

THE INTERNATIONAL CONFERENCE on 

DESIGN, TECNOLOGIES & MANAGEMENT 

IN MANUFACTURING 

Iaşi, May 14 th – 16 th , 2010 

organized by the 

FACULTY OF MACHINE MANUFACTURING & 

INDUSTRIAL MANAGEMENT 

Papers published with the support of 

NATIONAL AUTHORITY for SCIENTIFIC RESEARCHERS 

EDITORIAL BOARD 

MACHINE CONSTRUCTION 

Fascicle 2 

Conf.univ.dr.ing. Irina Cozmîncă 

Prof.univ.dr.ing. Radu Ibănescu 

Conf.univ.dr.ing. Vasile V. Merticaru


BULLETIN OF THE POLYTECHNIC INSTITUTE OF IAŞI 

Publicat de 

UNIVERSITATEA TEHNICĂ „GHEORGHE ASACHI” DIN IAŞI 

Tomul LVI(LX), Fasc. 2 2010 

Secţia 


S U M A R 

ANTÓNIO M. GONÇALVES-COELHO, GABRIELA NEŞTIAN și 

ANTÓNIO MOURÃO, Modelul matricei de proiectare în cazul 

soluțiilor redundante de proiectare (engl., rez. rom.)……........................ 

TAXIARCHIS BELIS și ARISTOMENIS ANTONIADIS, Model pentru 

evaluarea uzurii dinților frezelor de danturat pe baza determinării 

așchiilor 3D (engl., rez. rom.). .................................................................. 

NIKOLAOS TAPOGLOU și ARISTOMENIS ANTONIADIS, Determinarea 

prin simulare CAD a componentelor forței de așchiere la frezarea 

danturilor (engl., rez. rom.)........................................................................ 

EUGEN STRĂJESCU, CONSTANTIN DOGARIU, OLIMPIA PAVLOV și 

DUMITRU DUMITRU, Contribuţii privind controlul informatizat al 

cuţitelor roată (engl., rez. rom.)................................................................. 

SILVIU BERBINSCHI, VIRGIL TEODOR, NICOLAE DUMITRAŞCU și 

NICOLAE OANCEA, Contribuţii la elaborarea unei metode grafice 

pentru profilarea sculelor care generează prin înfăşurare. I. Algoritm. 

(engl., rez. rom.) ....................................................................................... 

Pag. 

1 

9 

21 

31 

41

SILVIU BERBINSCHI, VIRGIL TEODOR, NICOLAE DUMITRAŞCU și 

NICOLAE OANCEA, Contribuţii la elaborarea unei metode grafice 

pentru profilarea sculelor care generează prin înfăşurare. II. Aplicație 

pentru profilarea sculei-cremalieră (engl., rez. rom.)................................ 

CĂTĂLIN FETECĂU, DANIEL-VIOREL VLAD și COSTEL MOCANU, 

Simularea procesului de strunjire folosind analiza cu element finit 

(engl., rez. rom.)........................................................................................ 

MIRCEA COZMÎNCĂ, CRISTIAN CROITORU și CĂTĂLIN 

UNGUREANU, Cercetări experimentale pentru validarea unei noi 

metode de evaluare a forțelor de așchiere (engl., rez. rom.) ..................... 

ANA-MARIA MATEI și MARIUS NICOLAE MILEA, Forțele de așchiere 

la frezarea frontală în funcție de forțele dezvoltate la nivelul unui dinte 

(engl., rez. rom.)........................................................................................ 

MARIUS-IONUŢ RÎPANU, GHEORGHE NAGÎŢ, IOLANDA-ELENA 

MANOLE și ANDREI WEINGOLD, Aspecte comparative privind 

croirea la operaţia de decupare-perforare pe prese clasice și centre de 

presare prevăzute cu comandă numerică (engl., rez. rom.)....................... 

IUSTINA ELENA ROTMAN, PETRU DUȘA și RADU ADRIAN BACIU 

LUPAȘCU, Considerații teoretice și experimentale privind 

determinarea efectului de divergență (engl., rez. rom.) ........................... 

CĂTĂLIN UNGUREANU, RADU IBĂNESCU și IRINA COZMÎNCĂ, 

Sistem de măsurare computerizat (engl., rez. rom.)................................. 

BIRGIT KJÆRSIDE STORM, Lipirea cu adezivi a aluminiului tratat 

superficial (engl., rez. rom.)...................................................................... 

IOANA PETRE, DAN PETRE, CRISTINA FILIP și LAVINIA NEAGOE, 

Aplicaţii industriale ale muşchilor pneumatici (engl., rez. rom.) ............. 

MIHĂIȚĂ HORODINCĂ, Noi resurse ale cercetării experimentale asistate 

de calculator a puterii electrice absorbite în sistemele de fabricație 

(engl., rez. rom.)........................................................................................ 

ION BOSTAN, VALERIU DULGHERU și ANATOL SOCHIREANU, 

Dezvoltarea integrată CAE a transmisiilor precesionale utilizând 

platforma Autodesk Inventor (engl., rez. rom.) ....................................... 

ILEANA FULGA și EUGEN STRĂJESCU, Propuneri de optimizare a 

funcționării morilor fluidice cu jeturi în spirală (engl., rez. rom.)............ 

DORU CĂLĂRAŞU, IRINA TIŢA, DAN SCURTU și BOGDAN 

CIOBANU, Analiza dinamică a mecanismului mecano-hidraulic de 

protecţie a turbinelor eoliene cu ax orizontal de mică putere prin 

basculare în plan vertical (engl., rez. rom.)............................................... 

CONSTANTIN CHIRIȚĂ, ADRIAN HANGANU și DANIEL CALFA, 

Cercetări privind sistemele hidraulice pentru deplasarea maselor mari 

pe distante mici, cu frecvenţă redusă (engl., rez. rom.)............................ 

49 

57 

65 

75 

83 

91 

97 

105 

117 

125 

135 

143 

153 

161

EROL MURAD, CĂTALIN DUMITRESCU, GEORGETA HARAGA și 

LILIANA DUMITRESCU, Sistem de măsurare pneumatică a masei de 

apă extrasă în procesele de uscare convectivă (engl., rez. rom.)............... 

MIHAI FLORIN MĂNESCU și VALERIU PANAITESCU, Tehnologii 

folosite pentru monitorizarea defecţiunilor structurale apărute în 

funcţionarea grupurilor eoliene (engl., rez. rom.)...................................... 

AURORA ALEXANDRESCU, ADINA SIMONA ALEXANDRESCU și 

ADRIAN CONSTANTIN ALEXANDRESCU, Reabilitarea stației de 

pompare pentru alimentare cu apă (engl., rez. rom.)................................. 

ILARE BORDEAȘU, MIRCEA OCTAVIAN POPOVICIU, DRAGOȘ 

NOVAC, LIVIU MARSAVINA, RADU NEGRU, MIRCEA VODĂ, 

VICTOR BĂLĂȘOIU și MARIAN BĂRAN, Contribuţii în evaluarea 

durabilităţii arborilor hidroagregatelor axiale orizontale (engl., rez. 

rom.) ......................................................................................................... 

TEODOR MILOŞ, MIRCEA BĂRGLĂZAN, EUGEN DOBÂNDĂ, 

ADRIANA MANEA, RODICA BĂDĂRĂU și DANIEL STROIŢĂ, 

Traseul optim al unei conducte de aducţiune utilizând algoritmul 

Bellman-Kalaba (engl., rez. rom.) ............................................................ 

DĂNUŢ ZAHARIEA și MIHAELA TUDORACHE, Analiza structurală a 

cuplajelor fixe de tip manşon cu ştifturi cilindrice (engl., rez. rom.) ....... 

DĂNUŢ ZAHARIEA și MARIUS STACHIE, Analiza structurală a arcurilor 

bimetalice lamelare (engl., rez. rom.) ....................................................... 

IRÈNE ALEXANDRESCU, HANS-JOACHIM FRANKE și THOMAS 

VIETOR, Managementul cunoașterii în configurarea produselor 

complexe personalizate în fazele incipiente ale dezvoltării acestora 

(engl., rez. rom.)........................................................................................ 

PETRU DUŞA și IULIANA LAURA TARANOVSCHI, Cercetări cu privire 

la activitatea de inovare din mediul tehnic (engl., rez. 

rom.).......................................................................................................... 

ANDREI MIHALACHE, GHEORGHE NAGÎŢ și MARIUS-IONUŢ 

RÎPANU, Avantaje şi puncte slabe ale diferitelor tehnici de inginerie 

inversă (engl., rez. rom.) ........................................................................... 

ROBERTO LOPEZ, MANUEL SAN JUAN, FRANCISCO SANTOS, 

OSCAR MARTÍN și FLORIN NEGOESCU, Termografie aplicată în 

frezarea osoasă (engl., rez. rom.)............................................................... 

OLGA MARINA MONTES și VASILE V. MERTICARU, Studiu asupra 

oportunităţii unei noi fabrici de reciclare a sticlei, pentru o dezvoltare 

regională durabilă (engl., rez. rom.).......................................................... 

CĂTĂLIN DUMITRAȘ, CARMEN LOGHIN, SULEYMAN YALDIZ, 

MEHMET SAHIN și LUMINIȚA CIOBANU, Modelarea 3D a 

suprafețelor textile cu destinația optimizării curgerii fluidelor (engl., 

rez. rom.).................................................................................................... 

167 

177 

189 

197 

205 

215 

221 

227 

237 

245 

251 

259 

267


BULLETIN OF THE POLYTECHNIC INSTITUTE OF IAŞI 

Published by the 

„GHEORGHE ASACHI” TECHNICAL UNIVERSITY OF IAŞI 

Tomul LVI(LX), Fasc. 2 2010 

Section 

MACHINE CONSTRUCTION 

C O N T E N T S 

ANTÓNIO M. GONÇALVES-COELHO, GABRIELA NEŞTIAN and 

ANTÓNIO MOURÃO, On the Pattern of the Design Matrix in 

Redundant Design Solutions (English, Romanian summary). …............. 

TAXIARCHIS BELIS and ARISTOMENIS ANTONIADIS, Hobbing Wear 

Prediction Model Based on 3D Chips Determination (English, 

Romanian summary)................................................................................. 

NIKOLAOS TAPOGLOU and ARISTOMENIS ANTONIADIS, CAD- 

Based Calculation of Cutting Force Components in Gear Hobbing 

(English, Romanian summary).................................................................. 

EUGEN STRĂJESCU, CONSTANTIN DOGARIU, OLIMPIA PAVLOV 

and DUMITRU DUMITRU, Contributions Concerning The Computer 

Aided Control of the Fellows' Cutter (English, Romanian 

summary)................................................................................................... 

SILVIU BERBINSCHI, VIRGIL TEODOR, NICOLAE DUMITRAŞCU 

and NICOLAE OANCEA, Contributions to the Elaborations of a 

Graphical Method for Profiling of Tools which Generate by 

Enveloping. I. Algorithms (English, Romanian summary)....................... 

Pag. 

1 

9 

21 

31 

41

SILVIU BERBINSCHI, VIRGIL TEODOR, NICOLAE DUMITRAŞCU 

and NICOLAE OANCEA, Contributions to the Elaborations of a 

Graphical Method for Profiling of Tools which Generate by 

Enveloping. II. Application for Rack-Gear Tool’s Profiling (English, 


CĂTĂLIN FETECĂU, DANIEL-VIOREL VLAD and COSTEL 

MOCANU, The Numerical Simulation of Turning Process using Finite 

Element Modeling (English, Romanian summary)................................... 

MIRCEA COZMÎNCĂ, CRISTIAN CROITORU and CĂTĂLIN 

UNGUREANU, Experimental Researches Regarding a New Method 

for Cutting Forces Evaluation (English, Romanian summary)................ 

ANA-MARIA MATEI and MARIUS NICOLAE MILEA, Face Milling 

Forces Depending on the Forces Developed on a Single-Tooth (English, 


MARIUS-IONUŢ RÎPANU, GHEORGHE NAGÎŢ, IOLANDA-ELENA 

MANOLE and ANDREI WEINGOLD, Comparative Aspects 

Regarding the Nesting for Blanking-Punching Operation on Classical 

Presses and Numerical Commanded Pressing Centers (English, 


IUSTINA ELENA ROTMAN, PETRU DUȘA and RADU ADRIAN BACIU 

LUPAȘCU, Theoretical and Experimental Considerations on 

Determining the Effect of Divergence (English, Romanian 

summary)................................................................................................... 

CĂTĂLIN UNGUREANU, RADU IBĂNESCU and IRINA COZMÎNCĂ, 

Computerized Measurement System (English, Romanian summary)..... 

BIRGIT KJÆRSIDE STORM, Adhesive Bonding of Surface Treated 

Aluminium (English, Romanian summary).............................................. 

IOANA PETRE, DAN PETRE, CRISTINA FILIP, and LAVINIA 

NEAGOE, Industrial Applications of the Pneumatic Muscles (English, 


MIHĂIȚĂ HORODINCĂ, Some New Resources on Computer Assisted 

Experimental Research of the Absorbed Electric Power in 

Manufacturing Systems (English, Romanian summary)........................... 

ION BOSTAN, VALERIU DULGHERU and ANATOL SOCHIREANU, 

Integrated CAE Development of Precessional Drives using Autodesk 

Inventor Platform (English, Romanian summary).................................... 

ILEANA FULGA and EUGEN STRĂJESCU, Proposals for the 

Improvement of the Fluidic Spiral Jetmills' Activity (English, 


DORU CĂLĂRAŞU, IRINA TIŢA, DAN SCURTU and BOGDAN 

CIOBANU, Dynamic Analysis of Mechanical-Hydraulic Protection 

Mechanisms of Low Power Horizontal Axis Wind Turbines through 

Vertical Tilting (English, Romanian summary)........................................ 

49 

57 

65 

75 

83 

91 

97 

105 

117 

125 

135 

143 

153

CONSTANTIN CHIRIȚĂ, ADRIAN HANGANU and DANIEL CALFA, 

Research on Hydraulically Systems which Move Heavy Masses on 

Small Distances with Lower Frequencies (English, Romanian 

summary)................................................................................................... 

EROL MURAD, CĂTALIN DUMITRESCU, GEORGETA HARAGA and 

LILIANA DUMITRESCU, Pneumatic Metering System for Amount of 

Water Extracted in Convective Drying Processes (English, Romanian 

summary)................................................................................................... 

MIHAI FLORIN MĂNESCU and VALERIU PANAITESCU, Technologies 

for Monitoring Structural Damages Arising in the Functioning of Wind 

Turbines (English, Romanian summary)................................................... 

AURORA ALEXANDRESCU, ADINA SIMONA ALEXANDRESCU and 

ADRIAN CONSTANTIN ALEXANDRESCU, Pumping Station 

Exoneration for Water Supply (English, Romanian summary)................. 

ILARE BORDEASU, MIRCEA OCTAVIAN POPOVICIU, DRAGOS 

NOVAC, LIVIU MARSAVINA, RADU NEGRU, MIRCEA VODA, 

VICTOR BALASOIU and MARIAN BĂRAN, Contributions regarding 

Durability Evaluation of Horizontal Axial Hydraulic Turbines Shafts 

(English, Romanian summary).................................................................. 


ADRIANA MANEA, RODICA BĂDĂRĂU and DANIEL STROIŢĂ, 

Optimal Routes of Pipeline Supply using the Bellman-Kalaba 

Algorithm (English, Romanian summary)................................................ 

DĂNUŢ ZAHARIEA and MIHAELA TUDORACHE, Structural Analysis 

of a Sleeve Rigid Coupling with Cylindrical Pins (English, Romanian 

summary)................................................................................................... 

DĂNUŢ ZAHARIEA and MARIUS STACHIE, Structural Analysis of a 

Bimetallic Strip Thermostat (English, Romanian summary).................... 

IRÈNE ALEXANDRESCU, HANS-JOACHIM FRANKE and THOMAS 

VIETOR, Knowledge Management for the Configuration in Early 

Phases of Complex Custom Products (English, Romanian summary)... 

PETRU DUŞA and IULIANA LAURA TARANOVSCHI, Researches 

Regarding to Innovative Activity in Technical Environment (English, 


ANDREI MIHALACHE, GHEORGHE NAGÎŢ and MARIUS-IONUŢ 

RÎPANU, Advantages and Weak Points of Different Reverse 

Engineering (RE) Techniques (English, Romanian summary)................. 

ROBERTO LOPEZ, MANUEL SAN JUAN, FRANCISCO SANTOS, 

OSCAR MARTÍN and FLORIN NEGOESCU, Thermography Applied 

to Bone Drilling (English, Romanian summary)....................................... 

OLGA MARINA MONTES and VASILE V. MERTICARU, Study on the 

Opportunity of a New Glass Recycling Factory for Regional 

Sustainable Development (English, Romanian summary)........................ 

161 

167 

177 

189 

197 

205 

215 

221 

227 

237 

245 

251 

259

CĂTĂLIN DUMITRAȘ, CARMEN LOGHIN, SULEYMAN YALDIZ, 

MEHMET SAHIN și LUMINIȚA CIOBANU, Modeling 3D Surface of 

Textile Structures for Fluid Flow Improvement (English, Romanian 

summary)................................................................................................... 

267


Publicat de 

Universitatea Tehnică „Gheorghe Asachi” din Iaşi 

Tomul LVI (LX), Fasc. 2, 2010 

Secţia 


ON THE PATTERN OF THE DESIGN MATRIX IN 

REDUNDANT DESIGN SOLUTIONS 

BY 

ANTÓNIO M. GONÇALVES-COELHO 1 , GABRIELA NEŞTIAN 2 , 

and ANTÓNIO MOURÃO 1 

Abstract. Axiomatic Design was created with the aim of building a 

systematic model for engineering education and practice, taking into 

account the initial hypothesis that there are fundamental principles that 

govern good design practice. According to this design theory, the design 

solutions can be classified as uncoupled, decoupled or coupled, depending 

on the way their design matrices are populated. Uncoupled solutions are the 

best, decoupled solutions are acceptable, and coupled solutions are poor 

design and should be avoided. Redundant designs make up a specific class 

of design solutions, in which the number of functional requirements is 

lesser than the number of design parameters. This paper discusses how the 

design matrix could be populated, so that a redundant design could be either 

uncoupled or decoupled. 

Keywords: Axiomatic Design, Design Matrix, Redundant Design, Ideal 

Design. 

1. Introduction 

Axiomatic Design (AD) was created in the late 1970’s by Nam P. Suh with 

the aim of building a systematic model for engineering education and practice 

under the initial hypothesis that there are fundamental principles that govern 

good design practice [1]. 

According to AD, any “design object” — being it a product, a process or any 

other technical system — can be described by a vector in each one of four 

design domains (see Fig. 1). The design process starts at the customer domain 

with the definition of the customer needs (CNs). Mapping between the customer

2 Antonio M. Gonçalves-Coelho et al. 

and the functional domains allows finding the functional requirements (FRs). 

Once this is done, another mapping translates the FRs into design parameters 

(DPs), i.e. the set of properties that physically describe the design object. 

Finally, mapping from the physical to the process domain leads to the process 

variables (PVs), which outline how to make the design object [1]. 

Fig. 1 – The Design Domains [1]. 

The left-to-right mapping between any two contiguous domains can be 

represented by a “design equation” of the form: 

(1) 

{} Y = [] A {} X ; Aij = 

∂Yi ∂X j 

; i = 1, ..., m; j = 1,...,n , 

where {Y} is a vector that represents the set of m requirements that should be 

accomplished, {X} is a vector representing the set of n parameters of the design 

object that is expected to fulfil the requirements, and [A] is the design matrix. 

Usually, any prospective design equation is bounded by constraints [1]. 

Eq. (1) is not unique, and different {X} vectors would represent different 

design solutions that are characterized by distinct design matrices, which 

patterns would make the difference between “good” and “poor” design. The 

good or the poor quality of any design solution is ruled by the Independence 

Axiom, which states that, in good design, the selected parameters {X} should be 

such that the requirements {Y} are fulfilled independently. As a result, the ideal 

design solution should have the same number of requirements and parameters 

(m = n) and the design matrix should be diagonal, case of which the design 

solution is called “uncoupled” [1]. A triangular design matrix is acceptable as 

well and corresponds to a “decoupled” design [1]. Any other pattern of square 

design matrix corresponds to a “coupled” design, which should be recognized as 

poor and as such should be avoided [1]. 

For any design solution where m > n, AD’s Theorem 1 states that either the 

design is coupled or some of its FRs can never be fulfilled [1]. An example of 

such a design can be found elsewhere [2]. In the case of m < n, AD’s Theorem 3 

states that the design is either redundant or coupled [1]. At last, the specific case 

of a design with a single requirement (m = 1) is worth to mention. In this case, 

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 

the decoupled and the coupled conditions are impossible to attain in any singlerequirement 

design because there is no different FRs to couple. 

The present work contains an analysis of the pattern of the design matrix of 

redundant designs with more than one requirement (m > 1), a matter to which 

the researchers have not paid sufficient attention so far. 

2. The Key Characteristics of Redundant Designs 

Fig. 2 depicts the design of a simple clamping device with one only 

customer need: the clamping action, which can be attained by adjusting one 

only FR — the distance d. 

Fig. 2 – Example of a simple redundant design. 

The adjustment of d can be achieved by suitably setting the values of two 

design parameters: the position of the cam in the end of the hand lever (see Fig. 

1), which is denoted by angle α, and the angular position of the threaded rod, 

which is represented by angle β. Thus, the design equation for this redundant 

design with one requirement and two parameters is: 

(2) d 

⎡ 

⎣ 

{}= A 11 A 12 

⎧ 

⎤ 

⎪ α ⎫⎪ 

⎦ ⎨ ⎬ . 

β 

⎩⎪ ⎭⎪ 

A more general case is the Eq. (3) that represents a design with arbitrary 

numbers of requirements m and parameters n, with m < n. The equation relates 

to a redundant design, and its quality depends on the pattern of the design 

matrix, as per Theorem 3. 

To better understand how the pattern of the design matrix of Eq. (3) could 

characterize a good or a poor design, let us consider the three coexisting designs


represented by Eq. (4), Eq. (5) and Eq. (6), where Akj denotes the possible nonzero 

elements of the related design matrices. 

(3) 

(4) 

(5) 

⎧ 

⎪ 

⎨ 

⎪ 

⎩⎪ 

⎧ 

⎪ 

⎨ 

⎪ 

⎩⎪ 

FR 1 

FR 2 

FR 3 

FR 1 

FR 2 

FR 3 

⎧ 

⎪ 

⎪ 

⎪ 

⎫ ⎡ 

⎪ ⎢ 

⎬ = ⎢ 

⎪ ⎢ 

⎭⎪ 

⎣⎢ 

A11 A21 A31 A12 A22 A32 A13 A23 A33 0 

0 

0 

0 

0 

0 

0 

0 

0 

0 

0 

0 

0 

0 

0 

⎤ ⎪ 

⎥ ⎪ 

⎥ ⎨ 

⎥ ⎪ 

⎦⎥ 

⎪ 

⎪ 

⎪ 

⎪ 

⎩ 

⎧ 

⎪ 

⎪ 

⎪ 

⎫ ⎡ 

⎪ ⎢ 

⎬ = ⎢ 

⎪ ⎢ 

⎭⎪ 

⎣⎢ 

0 

0 

0 

0 

0 

0 

0 

0 

0 

A14 A24 A34 A15 A25 A35 A16 A26 A36 0 

0 

0 

0 

0 

0 

⎤ ⎪ 

⎥ ⎪ 

⎥ ⎨ 

⎥ ⎪ 

⎦⎥ 

⎪ 

⎪ 

⎪ 

⎪ 

⎩ 

DP 1 

DP 2 

DP 3 

DP 4 

DP 5 

DP 6 

DP 7 

DP 8 

DP 1 

DP 2 

DP 3 

DP 4 

DP 5 

DP 6 

DP 7 

DP 8 

⎫ 

⎪ 

⎪ 

⎪ 

⎪ 

⎪ 

⎬ , 

⎪ 

⎪ 

⎪ 

⎪ 

⎪ 

⎭ 

⎫ 

⎪ 

⎪ 

⎪ 

⎪ 

⎪ 

⎬ , 

⎪ 

⎪ 

⎪ 

⎪ 

⎪ 

⎭

(6) 

⎧ 

⎪ 

⎨ 

⎪ 

⎩⎪ 

FR 1 

FR 2 

FR 3 


⎧ 

⎪ 

⎪ 

⎪ 

⎫ ⎡ 

⎪ ⎢ 

⎬ = ⎢ 

⎪ ⎢ 

⎭⎪ 

⎣⎢ 

0 

0 

0 

0 

0 

0 

0 

0 

0 

0 

0 

0 

0 

0 

0 

0 

0 

0 

A17 A27 A37 A18 A28 A38 ⎤ ⎪ 

⎥ ⎪ 

⎥ ⎨ 

⎥ ⎪ 

⎦⎥ 

⎪ 

⎪ 

⎪ 

⎪ 

⎩ 

One can see that DP1, DP2 and DP3 are the only possible contributing 

parameters in Eq. (4). As for the design of Eq. (5), just DP4, DP5 and DP6 are 

significant. At last, in what concerns to Eq. (6), only DP7 and DP8 contribute. In 

other words, each one of the above-defined coexisting designs can fulfil all the 

FRs of Eq. (3) using entirely different subsets of the DPs included in the latter 

equation, in such a manner that all the DPs are taken into account. 

Now, one can figure out that the design of Eq. (3) could be achieved by 

merging the designs of Eq. (4), Eq. (5) and Eq. (6). 

The design matrices of Eq. (4) and Eq. (5) could be reduced to block 

matrices of size m x m by eliminating the zero elements of their design matrices. 

Therefore, the surrogate “condensed design equations” could be obtained 

from those equations by eliminating the non-relevant DPs, and by using the all 

the existing FRs and the non-zero block matrices. For example, Eq. (7) is the 

condensed equation that is obtained from Eq. (5). 

The same treatment could be done to Eq. (6), but in this case we would 

obtain a non-square block matrix of size (n mod m) x m. 

⎧ FR ⎫ ⎡ 

1 A14 A15 A ⎤ ⎧ 

16 DP ⎫ 

4 

⎪ ⎪ ⎢ 

⎥ ⎪ ⎪ 

(7) 

⎨ FR2 ⎬ = ⎢ A24 A25 A26 ⎥ ⎨ DP5 ⎬ . 

⎪ FR ⎪ ⎢ 

⎩⎪ 

3 ⎭⎪ 

A34 A35 A 

⎥ ⎪ 

⎣⎢ 

36 ⎦⎥ 

DP ⎪ 

⎩⎪ 

6 ⎭⎪ 

As a result, the redundant design of Eq. (3) could be considered “suitable 

design” if the condensed equations obtained from Eq. (4), Eq. (5) and Eq. (6) 

are not coupled. The coupled condition is excluded in Eq. (4) and Eq. (5) if the 

relevant DPs are chosen so that their non-zero block matrices are either 

triangular or diagonal. 

As for Eq. (6), the coupled condition is excluded if its non-zero block 

matrix is populated in such a manner that the condensed equations correspond 

to uncoupled or decoupled designs. If this is not the case, one can selectively 

“freeze” as many DPs as required, so that the condensed design become 

uncoupled or decoupled, as exemplified elsewhere [3]. 

DP 1 

DP 2 

DP 3 

DP 4 

DP 5 

DP 6 

DP 7 

DP 8 

⎫ 

⎪ 

⎪ 

⎪ 

⎪ 

⎪ 

. 

⎬ 

⎪ 

⎪ 

⎪ 

⎪ 

⎪ 

⎭


4. Conclusion 

The key conclusion of this paper can be summarized as a new theorem: Let 

us suppose a design with m requirements and n parameters, with m < n. Its 

design matrix can be partitioned in (n div m) square block matrices of size 

m x m, and an extra non-square block matrix of size (n mod m) x m, in such a 

manner that the block matrices have not common elements. Such a design is 

acceptable if each block matrix describes either an uncoupled or a decoupled 

design. Otherwise, the design is of the coupled type. 

Received: January, 30, 2010 1 UNIDEMI, Universidade Nova de Lisboa, 

Faculdade de Ciências e Tecnologia 

Departamento de Engenharia Mecânica e Industrial 

Monte de Caparica, Portugal 

e-mail: goncalves.coelho@fct.unl.pt 

ajfm@fct.unl.pt 

2 Ministerul Comunicaţiilor şi Societăţii Informaţionale, 

Bucureşti, România 

e-mail: gabriela.nestian@mcsi.ro 

R E F E R E N C E S 

1. S u h N. P., The Principles of Design, Oxford Univ. Press, N. Y., 1990. 

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 

Meaning of the First Axiom. In: A. Toca, O. Pruteanu (Eds.), Tehnologii 

Moderne, Calitate, Restructurare: Culegere de Lucrări Ştiinţifice, 3, pp. 341-344, 

Universitatea Tehnică a Moldovei, Chişinău, 2003. 

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 

axiomatic design approach for the cost optimisation of industrial coupled 

designs: A case study. In: Gonçalves-Coelho A.M. (Ed.), Proc. 5 th International 

Conference on Axiomatic Design - ICAD 2009, pp. 201-207, Campus de 

Caparica, 2009. 

MODELUL MATRICEI DE PROIECTARE IN CAZUL SOLUTIILOR 

REDUNDANTE DE PROIECTARE 

(Rezumat) 

Proiectarea axiomatică a apărut din necesitatea creării unui model sistematic de 

educare şi practică inginerească, pornind de la principiile fundamentale de bune practici 

ale proiectării. Conform acestei teorii, în funcţie de modul de populare al matricelor de 

proiectare, soluţiile de proiectare pot fi clasificate în soluţii necuplate, decuplate sau 

cuplate. Soluţiile de tip necuplat sunt cele mai bune, soluţiile de tip decuplat sunt


acceptabile iar soluţiile cuplate nu sunt satisfăcătoare din punct de vedere tehnic şi ar 

trebui evitate. Matricele diagonale corespund soluţiilor necuplate, matricele 

triunghiulare corespund soluţiilor decuplate iar toate celelalte tipuri de matrice pătratice 

corespund soluţiilor cuplate. În cazul în care numărul cerinţelor funcţionale este mai 

mare decât numărul parametrilor de proiectare, unele cerinţe nu pot fi îndeplinite 

niciodată şi soluţiile sunt de tip cuplat. Proiectarea redundantă creează o clasă specifică 

de soluţii de proiectare în care numărul cerinţelor funcţionale este mai mic decât 

numărul de parametri de proiectare. 

Această lucrare prezintă modalităţi de populare ale matricei de proiectare, astfel 

încât un proiect de tip redundant să poată deveni de tip necuplat sau decuplat. 

In concluzie putem formula o nouă teoremă: Presupunând că avem un produs sau un 

proces a cărui matrice de proiectare are m cerinţe şi n parametri, unde m < n, atunci 

matricea de proiectare poate fi descompusă în (n div m) submatrici pătratice de ordinul 

m x m şi o submatrice nepătratică de ordinul (n mod m) x m, astfel încât submatricile să 

nu aibă elemente comune. O astfel de soluţie este acceptabilă dacă fiecare bloc descrie o 

soluţie necuplată sau decuplată, dacă nu, atunci soluţia este de tip cuplat.


Publicat de 



Secţia 


HOBBING WEAR PREDICTION MODEL 

BASED ON 3D CHIPS DETERMINATION 

BY 

TAXIARCHIS BELIS and ARISTOMENIS ANTONIADIS 

Abstract. Gear hobbing is a machining process widely used in the industry 

for massive production of external gears. In this process the variant chip formation 

on each generating position causes uneven wear distribution on the hob teeth. This 

paper presents a new method for the determination of wear parameters based on 

3D chips. The 3D chip characteristics are fed to an existing wear model in order to 

calculate the hob wear distribution more precisely. 

Key words: gear hobbing, tool wear, simulation 


Gear hobbing is one of the most efficient generating processes for cutting 

external cylindrical gears. Although, hobbing cutters are still quite expensive 

due to their complex geometry. Thus their extensive exploitation is very 

important for industry. In gear hobbing the chip formation varies for each 

cutting tooth due to the fact that every tooth always cuts the same generating 

position. As a result, different wear laws are developed leading to uneven wear 

distribution on the hob teeth. Thus, the need to adopt an effective wear 

prediction model in gear hobbing arises. 

In this paper 3D chip determination is studied in order to feed more accurate 

data to an existing wear prediction model. Moreover total wear distribution on 

the hob teeth can be calculated. This information can be used to optimize 

tangential shifting conditions in order to maximize tool life and prolong the 

time interval before hob cutter resharpening. 

Figure 1 presents the basic nomenclature of the hob cutter that has been 

used in the wear simulation program developed in the present work. As it can be 

seen in the upper part of the figure, three distinct motions are required by this 

cutting method those being the workpiece revolution, the tool rotation, and the

10 Taxiarchis Belis and Aristomenis Antodiadis 

tool tangential displacement. Considering the above kinematic, two different 

hobbing types may be applied according to the direction of the axial feed, the 

climb and the up-cut hobbing respectively. In the present paper the tooth’s 

profile generation complies with DIN 3972 [1]. 

Fig. 1 – Basic kinematics and essential parameters of gear hobbing. 

2. State of the Art 

As mentioned before, wear prediction in gear hobbing is a challenging task 

because of the large amount of wear influencing data and the complex 

kinematics of the process. Many studies have been published, introducing 

different methods for wear determination. These studies can fall under two 

categories. The first includes methods based on wear calculation in individual 

generating positions [2], [3].These methods are based on the simulated chip 

geometry and cutting conditions. Lately, FEM-based methods for wear 

determination have been introduced [4]. These methods take into account 

occurring stresses in the tool-chip contact areas. Nevertheless a FEM-based 

simulation needs a considerable amount of time before it can be realised. 

The wear model presented in [2] is used in the present paper,. The 

contribution of the proposed method is the precise 3D chips determination


because the chip characteristics can be calculated with the accuracy provided by 

a CAD system. 

3. Hob Wear Simulation Process 

In order to achieve uniform wear distribution in the majority of the cutting 

teeth, a hob tangential shifting is required. In the left part of Figure 2, simulated 

chips and the corresponding gaps formed at the indicated generating positions, 

are illustrated. These chips are categorized into groups according to their shape 

and the cutting direction of the process. In the right part of Figure 2 the wear 

laws of several teeth after tangential shifts can be observed. 

Fig. 2 –Determination of the flank wear at individual hob teeth 

considering the shift data.


In the first column the wear laws of teeth number 1, 2, 3 and 4 and the total 

wear distribution on the hob after a certain number of cuts are presented. The 

second column shows the wear laws after the first tangential displacement equal 

to the tool axial pitch ε. After this sifting the cutting tooth i is placed at the 

generating position of the tooth i+1. Thus the tooth number 1 that was cutting in 

the generating position number 1 quits, whereas the tooth number 5 cuts for the 

first time in the generating position 4. 

For example, tooth number 4 was cutting initially in the generating position 

4. After AS* number of cuts, tooth 4 reaches a certain flank wear depth. In the 

first tangential shift, tooth 4 cuts in the generating position 3, obeying the wear 

law that governs that generating position. In this position, flank wear increases 

with a different rate for another AS* number of cuts. Finally, in the last 

tangential shift, tooth 4 cuts in generating position number 2, obeying a 

different wear law. 

As a result, in the bottom side of figure 2, the normalized flank wear 

distribution on the hob after two tangential displacements is presented. 

Consequently the wear distribution becomes uniform and the tool exploitation is 

enhanced. 

Fig. 3 – The effect of chip geometry and shape on the tool wear development.


3.1. HOBWEAR Formulation 

Figure 3 presents wear laws for different chip groups. There are five chip 

groups concerning the chip flow obstruction intensity, introduced by [3]. The 

determination of the chip group involves prior knowledge of the cutting 

direction and the examined gear flank. As it can be easily noticed, the wear rate 

in group I is more intense. On the other hand, cutting with chip group 0 causes 

less flank tooth wear and the achieved number of cuts increases. 

This can be explained by the fact that in group I there is intense chip flow 

obstruction. In the region of the tooth head chip flow obstruction phenomena 

are the most intense due to collision of chip distinct sections. In the bottom side 

of the figure the included equivalent chip thickness and length are further 

explained. 

Figure 4 presents the general structure of the program developed. The left 

part of the figure shows the categorized data input. Data such as chip group, 

equivalent thickness and length are loaded into the program from an external 

source. The undeformed chip geometries are calculated in various generating 

and revolving positions with the aid of HOB3D [2], [3]. 

Fig. 4 – HOBWEAR simulation algorithm. 

The wear behaviour on the hob teeth is primarily influenced by two sets of 

parameters. The first set refers to machining data and the geometry of hob and


gear. The second set refers to the machine tool, the cutting and working material 

combination and the used cooling lubricant. Every possible combination of the 

above parameters results in different coefficients in the model. Wear 

coefficients are experimentally defined for a variety of materials and geometries 

widely used in gear industry. 

The middle column of the figure shows the flow chart for wear calculation 

and optimal tangential displacement determination. The program has the ability 

to calculate wear progress at all generating positions including transient areas. 

In order to shift the hob cutter, all the group gears have to be processed. At the 

bottom of the figure the outputs of the program are presented. The output 

includes wear progress graphs for all generating positions, and total wear 

distribution graphs for all tangential displacements. The developed program also 

calculates the maximum flank wear and the corresponding number of cuts. 

3.2. 3D Chips Determination 

As illustrated on figure 5, this paper introduces a novel way of calculating 

equivalent chip dimensions. The equivalent chip thickness calculation is more 

accurate due to the fact that is measured on the 3D chip’s cross-sections. Detail 

A in the upper right part of the figure, shows the chip cross-section in plane 1. 

The calculation of thickness in tooth’s head and in arc distance a f 

) is hereby 

determined. The equation predicting the wear progress in the individual 

generating positions and the equations of equivalent chip dimensions introduced 

in [2] are presented in the bottom side of the figure. 

Fig. 5 – Determination of equivalent chip dimensions.


The wear determination model includes two stages. The first stage involves 

the employment of hobbing data in order to calculate the chip thickness, length 

and group for every cutting position of all generating positions. Categorization 

of chips in groups is determined by the program FRSWEAR [3]. 

Fig. 6 – Calculations of the equivalent chip thickness and length based on 3D chips. 

The measurement of the total length of the chip and the length compressed 

on the head’s corner are quite precise. As shown in the left bottom side of figure 

6, the first step for measuring the chip’s length compressed on the tooth’s 

head’s corner, is to cut the section of the chip from the left flank to the tool head 

center H, as the hob’s profile is revolving. In this cross-section we can easily


and precisely measure the chip’s thickness and length in head hH and lH,TF in 

CAD environment. This data is fed to the equation illustrated in figure 5 in 

order to calculate the arc length a f 

) . Determination of a f 

) is very important due 

to the fact that maximum wear depth has been observed by researchers in this 

region [2], [3], [7]-[9]. The calculation method of chip thickness hf is similar to 

hH. After hs and lH are determined for every generating position, data can be fed 

to the wear prediction model. All the calculations above can be simplified by 

the use of parametric design in CAD environment. In the first step the arc a f 

) is 

set as parameter. In the next step the section of the chip from point K to F as the 

hob profile revolves is cut. In the chip produced hH and lH,TF can be easily 

measure and the arc a f 

) calculated precisely. The parameter is set at the 

calculated value and the chips regenerated automatically. Thus, chip thickness 

hf can be easily measured. The final step includes calculation of the equivalent 

thickness hs. The upper part of figure 6 presents the variation of the chip 

thickness for all the revolving planes in several generating positions. These 

diagrams have been generated from the sequence of the chip thickness results 

hH as mentioned above. 

3.3. Development of the Wear Prediction Program 

The gear flank wear determination program is implemented in MATLAB 

high-level matrix array language. 

Fig. 7 – The Graphical User Interface of the wear simulation program HOBWEAR.


The user has the ability to select gear and hob material from a variety of the 

most popular ones in industry. This selection changes the wear coefficients used 

to determine the flank wear. An automatic report generator has been 

incorporated for better organization of the program results. 

4. Simulation Results 

Wear graphs for each individual generating position is the first output of the 

program developed, as presented in figure 8. As it can be seen, teeth number 10, 

11 and 12 have almost reached the wear limit of 0,6 mm. For wear width below 

0,2 mm the relation between number of cuts and flank wear is linear. 

The normalization progress of total wear distribution on the hob cutter from 

tangential shifting number 5 to 8 is illustrated in Figure 9. Optimization of 

tangential shifting in gear hobbing has been studied by [10]. The credibility of 

the above wear prediction model has been verified with the aid of a wide variety 

of cutting experiments [2], [3]. 

Figure 10 presents the wear distribution on the cutting teeth of a hob cutter 

for two cases. In both cases, 18 gears of the same width have been cut. In the 

first case, 3 tangential shifts took place with 1 ε displacement per shift. In the 

second case, 2 tangential shifts took place with 2 ε displacement per shift. It is 

evident that in the second case the maximum wear depth on the hob cutter is 

less than in the first one. 

Fig. 8 – Wear progress in individual generating positions.


Fig. 9 – Total wear distribution on the hob cutter after shifting. 

As it can be seen in the second case, the wear distribution is more 

uniform due to the fact that more teeth are involved in the cut. Authors in [10] 

have studied the optimal selection of the shift displacement and number of 

shifts and introduced a nomogram for various shifting conditions. 

Fig. 10 – Cutting of 18 gears with different tangential displacement parameters.


5. Conclusion 

This paper suggests a model which simulates the wear progress on the 

hobbing cutter. The end-user has the ability to optimize tangential shifting 

conditions, owing to this model, in order to minimize the total gear 

manufacturing cost. 

Acknowledgements. The authors wish to thank the Research Committee of the 

Technical University of Crete for their financial support (via basic research project 

2009). 

Received:March 10, 2010 Technical University of Crete, 

Department of Production Engineering & Management 

Chania, Crete, Greece 

e-mail: tbelis@isc.tuc.gr 

antoniadis@dpem.tuc.gr 


1. D I N 3972, Bezugsprofile von Verzahnwerkzeugen fuer Evolventenverzahnungen 

Nach DIN 867“, 1981. 

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 

Hobbing Cutting Process Simulation and Tool Wear Prediction Models. ASME 

Journal of Manufacturing Science and Engineering, 124, 1, 42-51 (2002). 

3. B o u z a k i s K. D., Konzept und technologishe Grundlagen zur automatisierten 

Erstellung optimaler Bearbeitungsdaten beim Waelzfraesen, Habilitation, TH 

Aachen, 1980. 

4. F r i d e r i k o s O., Simulation of Chip Formation and Flow in Gear Hobbing Using 

the Finite Element Method, Ph.D. Thesis, Aristoteles University of Thessaloniki, 

Greece, 2008. 

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 

Design Simulation of Gear Hobbing by Means of 3-Dimensional Kinematics 

Modeling, ASME Journal of Manufacturing Science and Engineering, 129, 911- 

918., (2007). 

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 

Process for the Manufacturing of Spur and Helical Gears, International Journal 

of Advanced Manufacturing Technology, 41, 3-4, 347-357 (2008). 

7. A n t o n i a d i s A., Determination of the Impact Tool Stresses During Gear 

Hobbing and Determination of Cutting Forces During Hobbing of Hardened 

Gears, Ph.D. Thesis, Aristoteles University of Thessaloniki, 1989. 

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 

Cemented Carbide Tools in Gear Hobbing. Part 1: FEM Modeling of Fly


Hobbing and Computational Interpretation of Experimental Results. ASME 

Journal of Manufacturing Science and Engineering, 124, 4, 784-791, (2002). 

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 

of Cemented Carbide Tools in Gear Hobbing. Part 2: The Effect of Cutting 

Parameters on the Level of Tool Stresses – A Quantitative Parametric Analysis. 

ASME Journal of Manufacturing Science and Engineering, 124, 4, 792-798, 

(2002). 

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 

Hobbing, Annals of the CIRP, 44, 1 (1995). 

MODEL PENTRU EVALUAREA UZURII DINȚILOR FREZELOR DE 

DANTURAT PE BAZA DETERMINĂRII AȘCHIILOR 3D 

(Rezumat) 

Frezarea danturii este un procedeu de prelucrare larg răspândit în industrie, în 

special pentru producția de serie a danturilor exterioare. În acest proces, modul diferit 

de formare a așchiei în fiecare poziție de generare a profilului duce la uzura neuniformă 

a dinților frezei. Lucrarea prezintă o nouă metodă de determinare a parametrilor uzurii, 

frezelor pe baza așchiilor 3D. Caracteristicile așchiilor 3D au fost stabilite pornind de la 

un model existent de uzură, în vederea determinării cu mai multă acuratețe a distribuției 

uzurii la nivelul dinților frezei.


Publicat de 



Secţia 


CAD-BASED CALCULATION OF CUTTING FORCE 

COMPONENTS IN GEAR HOBBING 

BY 

NIKOLAOS TAPOGLOU and ARISTOMENIS ANTONIADIS 

Abstract. One of the most commonly used gear manufacturing process 

especially for external is gear hobbing. The optimisation of the process of gear 

hobbing is of great importance in the modern industry. This paper presents a novel 

simulation program called HOB3D that can simulate the cutting process in a 

commercial CAD environment, thus producing results with the optimal precision. 

The outputs of the program include the 3D chip and gap geometry as well as the 

developing cutting forces which were validated using experiments. 

Key words: gear hobbing, cutting forces, simulation. 


One of the key components of any torque transmission system is high 

precision involute gears. Gears can be constructed with a wide variety of 

methods; Gear hobbing is the one mainly used in the modern industry. Gear 

hobbing kinematics consists of three relative motions between the cutting tool 

and the workpiece. This makes it difficult to simulate with analytical models. 

HOB3D is a novel simulation program based on a commercial CAD 

environment which can simulate the cutting process and provide results 

including 3D solid chips and gaps as well as predicting the developing cutting 

forces. 


The research conducted in the area of gear hobbing can be divided into two 

categories: gear hobbing process simulation and wear prediction. In the first 

field a series of simulation models have been developed using CAD [1],[2], 

FEA [3] or analytical [4]-[7] based models. Experiment based models [8], [9]

22 Nikolaos Tapoglou and Aristomenis Antoniadis 

are used in the field of wear prediction order to calculate the wear of the cutting 

tool and optimize the cutting so as to obtain uniform wear along the cutting tool. 

Cutting forces prediction is an area of great interest also. Research conducted in 

this area is based on Kienzle-Victor’s equations and depend on geometry of 

chips. 

3. Gear Hobbing and HOB3D Simulation Process 

Gear hobbing kinematics is based on three relative motions between the 

cutting tool and the workgear. These motions must be synchronised in order to 

produce high quality helical and spur gears. As presented in Fig. 1 the work 

gear rotates round its axis while at the same time the hob rotates round its own 

axis and moves parallel to the gear axis. The hob is positioned in an angle 

relative to the gear. The magnitude of this angle is relative to the hob helix 

angle and the helix angle to the gear produced correspondingly. 

Fig. 1 – Gear hobbing. 

Gear hobbing process is affected by a series of parameters which be divided 

in three categories: hob, gear and process parameters. The first include module 

(m), external diameter (dh), number of origins (z1) and number of columns (ni) 

of the hob. Gear parameters are number of teeth (z2), helix angle (ha) and gear 

width. Finally, process parameters include axial feed (fa) and cutting speed (v). 

A series of parameters can be calculated from those above mentioned, those 

being distance e, the helix angle of the hob (γ), gear diameter (dg) and depth of 

cut (t). 

In order to simulate the process of gear hobbing, new software has been 

developed. The proposed simulation model has been embedded in a commercial 

CAD program thus taking advantage of its accuracy resulting in more detailed 

calculations.


The new simulation code called HOB3D uses three coordinate systems in 

order to calculate the results required. The first (1) is positioned on the 

examined cutting edge and has the x axis parallel to the hob’s axis, the y axis 

perpendicular to x axis and finally z axis perpendicular to the prior two. 

Coordinate systems (2) and (3) have axis z running through the workgear’s axis. 

In coordinate system (2), x axis is always rotating in order to point to the gap, 

while the axes of coordinate system (3) are fixed. 

Fig. 2 – HOB3D flowchart.


Fig. 2 presents the flowchart of this simulation model. As it is illustrated, 

after all the input data, shown on the left side of the figure, has been defined, the 

initial workpiece is developed in the CAD environment and the number of the 

effective teeth is determined accordingly. The gear hobbing simulation process 

is executed for all these teeth. The first step in the simulation process is the 

positioning of the cutting edge on the 3D space. 

This positioning is repeated along the path that is covered by the cutting 

edge. The profiles that arise are combined in order to form a 3D surface. The 

next step is the creation of an assembly that includes the workgear and the 3D 

surface. Afterwards, Boolean operations are used in order to generate the 3D 

chip and the 3D gap. The above described process is repeated for all active teeth 

in every rotation of the cutting tool. Finally, after the end of the simulation 

process the cutting forces on every one of the active teeth of the hob are 

calculated. These forces are added up and the total cutting forces are calculated. 

4. HOB3D Formulation 

4.1. Simulation Formulation 

In HOB3D all the movements involved in the process are transferred to the 

hob cutting tooth motion. Furthermore, the simulation process is carried out on 

one gap of the gear thus reducing the simulation time. In order to identify the 

cutting teeth, those are numbered. The tooth which has the local axis Y1 parallel 

to local axis X2 when it passes through the center of the gap is named Tooth 0, 

the tooth that passes after that is named Tooth 1 and the tooth previous to that is 

named Tooth-1 and so on. 

The simulation process is illustrated on Fig. 3. As it is presented, the hob 

profile according to DIN3972 [10] must be designed first. This profile is 

presented on the top left frame of Fig. 3 and positioned on the 3D space, as 

illustrated on the next frame of fig. 3. The positioning is repeated for a series of 

times until the tooth is out of the cut. 

After this step, there is a series of profiles correctly positioned on the 3D 

space. Every profile represents a revolving position of the hob and its 

positioning includes hob rotation, workpiece rotation and feed rate. All these 

profiles are combined in order to form a 3D surface like the one seen in the last 

frame of the first line of fig. 3. Then, the 3D surface is assembled with the 3D 

gap produced from the previous tooth. The final stage of the simulation process 

is the extraction of the final gap geometry and the non-deformed chip geometry. 

This is achieved with the aid of Boolean operations, as supported by the used 

CAD environment.


Fig. 3 – HOB3D simulation process. 

4.2. Force Component Calculation Formulation 

After the simulation is completed, the cutting forces can be calculated. The 

force calculation code is performed in accordance with Kienzle-Victor’s 

equations. The calculation process is based on sections made on every one of 

the solid chips. 

A series of sections are made for every chip. Each section is made in the 

plane of the hob’s cutting edge. The planes are visible on the top left frame of 

Fig. 4. After the creation of the section on the 3D chip, the crossection is 

discritised along the cutting edge. This discritisation is presented on the top 

right frame of fig. 4. 

The three force components are calculated according to the elementary chip 

width and thickness for every elementary chip section. The three force 

components are rotated in order to match the local coordinate system (1) and 

then added up in order to produce the total cutting forces on every crossection. 

After all the sections are calculated the total forces on the tooth are obtained, as 

presented on the bottom of Fig. 4.


Fig. 4 – Force calculation algorithm. 

4.3. Simulation Results 

HOB3D has been used in order to produce the 3D solid chips in the full cut 

phase of gear hobbing. These chips were analysed so that the cutting forces are 

calculated. The crossections made on the chip are presented on the right side of 

Fig. 5 whereas the chip thickness on six of the crossections is illustrated on the 

left side of the same figure.


Fig. 5 – Chip thickness of the solid chip. 

The cutting forces components for the above chip are presented in the next 

Fig. 6 which are calculated in accordance with the system 1. 

Fig. 6 – Cutting forces on one generating position.


5. Verification 

The verification of the force calculation module was conducted in two 

phases. First, the cutting forces simulated on specific teeth were compared to 

the ones measured by B o u z a k i s [4] while other cutting forces calculated by 

HOB3D were compared to the ones measured by G u t m a n n [5]. Fig. 7 

illustrates the results of the first phase of the verification. As it can be seen, each 

column of the figure illustrates the measured and calculated cutting forces on 

one generating position measured on the coordinate system of the 3D gap (2). In 

most of the cases the simulation code predicts not only the form but also the 

magnitude of the cutting forces. 

Fig. 7 – Comparison between calculated and measured cutting forces.


The second step of the verification is the comparison between measured and 

calculated forces for all the cutting teeth simultaneously. Fig. 8 presents the 

comparison between the measured and the calculated cutting forces in all three 

directions of the coordinate system, as illustrated on the top left side of the 

figure. 

Fig. 8 – Comparison between calculated and measured cutting forces. 

4. Conclusions 

1. A novel simulation model for gear hobbing, developed in a commercial 

CAD environment is presented in this paper. The model can simulate the 

manufacturing of helical as well as spur gears 

2. The developed model can predict the cutting forces components with 

great accuracy. The calculated forces have been verified with the aid of cutting 

experiments. 

Received: March 10, 2010 Technical University of Crete, 

Department of Production Engineering&Management, 

Chania, Crete, Greece, 

e-mail: ntapoglou@isc.tuc.gr



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 

Design Simulation of Gear Hobbing by Means of 3-Dimensional Kinematics 

Modeling, ASME J. of Manuf. Science and Eng., 129, 911-918., (2007). 

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 

Process for the Manufacturing of Spur and Helical Gears, Int. J. of Advanced 

Manuf. Technology, 41, 3-4, 347-357 (2008). 

3. F r i d e r i k o s O., Simulation of Chip Formation and Flow in Gear Hobbing Using 

the Finite Element Method, Ph.D. Thesis, Aristoteles University of Thessaloniki, 

Greece 2008. 

4. B o u z a k i s K. D., Konzept und technologishe Grundlagen zur automatisierten 

Erstellung optimaler Bearbeitungsdaten beim Waelzfraese, Habilitation, TH 

Aachen 1980. 

5. G u t m a n n P., Zerspankraftberechnung beim Waelzfraesen, Ph.d. thesis, TH 

Aachen, 1988. 

6. A n t o n i a d i s A., Determination of the Impact Tool Stresses During Gear 

Hobbing and Determination of Cutting Forces During Hobbing of Hardened 

Gears, Ph.d. thesis, Aristoteles University of Thessaloniki, 1989. 

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 

Cemented Carbide Tools in Gear Hobbing. Part 1: FEM Modeling of Fly 

Hobbing and Computational Interpretation of Experimental Results. ASME J. of 

Manuf. Science and Engineering, 124, 4, 784-791, 2002. 

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 

Cemented Carbide Tools in Gear Hobbing. Part 2: The Effect of Cutting 

Parameters on the Level of Tool Stresses – A Quantitative Parametric Analysis, 

ASME J. of Manufacturing Science and Engineering, 124, 4, 792-798, 2002. 

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 

Hobbing Cutting Process Simulation and Tool Wear Prediction Models. ASME 

J.of Manuf. Science and Engineering, 124, 1, 42-51 2002. 

10. D I N 3972, Bezugsprofile von Verzahnwerkzeugen fuer Evolventenverzahnungen 

Nach DIN 867“, 1981. 

DETERMINAREA PRIN SIMULARE CAD A COMPONENTELOR 

FORȚEI DE AȘCHIERE LA FREZAREA DANTURILOR 

(Rezumat) 

Unul dintre cele mai uzuale procedee de prelucrare a danturilor, în special a celor exterioare, 

este frezarea. Optimizarea acestui tip de proces de danturare prezintă la ora actuală o importanță 

deosebită în industrie. Lucrarea prezintă un nou program de simulare, numit HOB3D, care 

permite simularea procesului de așchiere într-un mediu CAD comercial, cu rezultate optime în 

ceea ce privește precizia de prelucrare. Elementele rezultante ale programului de simulare se 

referă atât la așchia 3D și la geometria golului dintre dinții frezei, cât și la nivelul forțelor de 

așchiere dezvoltate, elemente care au fost validate experimental.


Publicat de 



Secţia 

CONSTRUCȚII DE MAȘINI 

CONTRIBUTIONS CONCERNING THE COMPUTER 

AIDED CONTROL OF THE FELLOWS' CUTTER 

BY 

EUGEN STRĂJESCU 1 , CONSTANTIN DOGARIU 1 , 

OLIMPIA PAVLOV 2 and DUMITRU DUMITRU 3 

Abstract. In the paper are presented the bases of a methodology for the 

determination of all the geometric and constructive elements of the cutting tools 

for the fellows cutter, tools having curve edges, starting from the getting of a solid 

model 3D of the measured tool. There are shown the possibilities to obtain a 3D 

model and the mathematical model of the geometric parameters control. 

Key words: fellows' cutter, control, cutting tools, 3D model, geometry, 

curved edges. 


Preoccupations concerning the increase of the standard, standard type and 

special cutting tools’ production and preoccupations concerning new tools with 

superior constructive functional performances represent a major tendency in the 

manufacturing of machines. The software resources implicates the existence in 

the system of a collection of organized information in order to assure not only 

the complete design of the cutting tool but also the analysis, the syntheses, 

comparison, modeling, simulation optimization, visualization, the optional 

presentation of the partial results etc. So the modern design is made in order to 

obtain a solid 3D model that is later automatically detailed in sections and views 

2D. The existence of the solid model signify the spectacular increase of the 

information's’ number permitting practically the total knowledge of the 

positions of any point or surfaces from the element’s construction and of the 

angles between directions or planes. This demarche ameliorates the tackling of 

the cutting tools' design by means of the concurrent engineering.

32 Eugen Străjescu et al. 

The modern means for the assisted design (Catia, Solid Works, Ideas, Solid 

Edge) are capable to make automatically sections in any point and after any 

direction, pointing automatically distances between points or angles’ values. 

In this way we can avoid the situations in which we are choosing for the 

cutting tool a value for an angle at a peak, but in other sections or points the real 

values are outside the domain of availability. Starting from these observations, 

we want to develop a new method for the control of the manufactured cutting 

tool, starting from the idea of the obtention of a solid model. 

This demarche is extremely utile in the admission of the concurrent 

engineering. 

2. Using the Solid Model for the Cutting Tools' Control 

2.1. Possibilities to obtain the 3D Models of the Real Pieces 

2.1.1. The getting of the 3D model of the cutting tool by photography. The 

software D Sculptor presents a relative new in the large landscape of the CAD 

software and the software for treatment of the images and corps 2D and 3D. It 

permits the creation on the computer photorealistic models for a large game of 

objects using habitual photos, easy and fast. We do not need special hardware 

elements, but only a normal 

computer and a photo camera. 

A digital camera is required, 

but it is possible to use 

scanned photos. The main 

screen of the program is 

presented in the fig. 1. 

The basic processes for 

the model’s construction are 

presented down: 

• it is positioned the object to 

acquire the model on the 

calibration plane; 

• it is photographed the object 

from many angles; 

Fig. 1 – The main screen of D Sculptor software. 

• the photos are imported in 

the software D Sculptor; 

• D Sculptor detects automa-ticaly the model; 

• D Sculptor is calculating the three-dimensional model. 

• The obtained three-dimensional model (3D) (fig. 2) can be used with software. In 

time D Sculptor 2.0 was improved at a technical level and from the point of 

view of the interface.


It is possible to create models faster than before, and the last version of D 

Sculptor 2.0 Professional has a superior accuracy. 

2.1.2. Generation of the 3D model by scanning. A 3D scanner is a modern 

device, intense developed during the last years, device that analyses a real 

object or an average in order to collect data concerning the form, the texture and 

the color of the object. 

The acquired data can be used to digitally construct 3D models, utile for a 

large game of applications. In the most situations, a single scanning do not 

produces a complete model of the subject. Multiple scanning, some time 

hundreds, from many directions are necessary usually to obtain information 

about all the sides of the object. 

These scans must be introduced in a common reference system, processes 

named usually alignment or recording. The scans will fusion in order to create a 

complete model. 

Now the scanners are of two types: 

scanners 3D with contact and without 

contact. The scanners non contact can be 

active or passive. It exists a large gamut 

of technologies for the two categories. In 

techniques are imposed the non contact 

scanners, with laser. 

The essential problem of the getting 

of a 3D model by scanning is represented 

by the great number of necessaries scans 

and by the possible necessity to complete 

Fig. 2. – The solid model resulted 

after the images’ treatment. 

form of the scanned tool is more complex. 

the acquired model using programs for 

assisted design (Catia, SolidWorks etc.). 

This last activity is more necessary if the 

For these reasons we can affirm that for the simple tools the proposed 

method is not efficient, because the completion of the model need too much 

time for routine measurements, but becomes very efficient for the complex 

cutting tools, measurable with difficulties, at which we obtain the geometric 

parameters baffling to measure (and in the main cases with a low accuracy). 

3. The Control of the Geometry Based on the 3D Model 

After the getting of the 3D model using the software D Sculptor 2.0 or by 

scanning, the model is exported in the Solid Works software, resulting the up 

presented (fig. 2).


On the obtained model are choose points on the main edge and on the 

secondary one and there are given command for the construction of the planes 

in which are measured the requested geometric elements. 

3.1. The Determination of the Fellows' Cutters' Geometry. 

The fellows' cutters are tools destined to the manufacturing by mortising the 

internal and external gear of the spur gears with right, inclined of "in V" teeth, 

representing from that reason a high degree of universality, as consequence of 

the cutting edges' access in areas inaccessible for other kind of tools (for 

example, side mills, hobbing cutters, etc.). Also, the fellows' cutters are the 

single cutting tools that can machine cylindrical gear wheels with interior 

denture, by rolling. 

3.1.1. Geometric parameters. The clearance angle at the tooth' peak presents 

a big importance, because from that value depends the bigness of the profile 

correction in time after successive sharpening. 

Also, by the value of this angle depends the size of the lateral clearance 

angle on the two flanks of the tooth that result much more little. So, it is 

considered the lateral clearance angle at the level of the divided circ, 

respectively if the tooth is cut after the division cylinder, the intersection curve 

between this cylinder and the involute clearance angle of the tooth will be a 

helix. Because the disposition on the involute flanks of the points in which the 

values of the angles α, γ, κ and λ are determined, it results that in any point we 

obtain different values that must very between convenient limits. 

The actual methods do not permit the determination of these angles, but 

3 0 31' 

Fig. 3 – The determination of the angles αv and γv at the peak. 

2 0


only of the values from the peak of the tooth (Fig. 3). The minimal values (that 

must have a specified value) are determined on the base of mathematical 

relations, 

but the real angles' variations not controlled. 

If the intersection of the tooth with planes parallel with a medial plane of 

the tooth is realized (posterior planes, Fig. 4), we obtain different values for the 

angles αp and γp (the plane Pp 

- Pp) in the points of 

intersection between these 

planes and the cutting edge 

from the involute flank. The 

points in which the 

determination is made can be 

however dense (from example 

0.01 mm), but such frequency 

is not necessary, because the 

variation limits and the 

variation mode can be 

observed in only some points, 

and that fact simplify 

determination operations. 

the 

Fig. 4 – The intersection of the 

tooth 

with paralel planes. 

The Solid Works computer 

program, by the facilities 

that are offered, give directly 

under the form of a quote the 

values obtained for the angles 

determined in these planes. 

Using a program that permits 

the drawing of the mentioned 

planes at determined 

distances, with a pre 

established step, becomes 

Fig. 5 – Intersectio ns with orthogonal planes. 

possible to obtain the desired 

values in every point on the 

edge, and with these values it 

is possible 

to draw a chart 

(Fig. 6). 

Similarly, it is possible to 

construct normal planes at the 

cutting edge's tangent in the 

anterior determined points on 

that edge, or orthogonal planes (Fig. 5). Obviously, in these planes are 

determined the angles αn and γn, or the angles αO and γO. 

The determination 

conditions 

are similar with the before precise conditions.


3.5 

3 

2.5 

2 

1.5 

1 

0.5 

0 

4.5 

4 

3.5 

3 

2.5 

2 

1.5 

1 

0.5 

0 

3.31 3.31 3.31 3.31 

2 

2.11 

M1 M2 M3 M4 

α p 2 2.11 3.31 4.27 

γ p 3.31 3.31 3.31 3.31 

Fig. 6 – Variation chart of the angles αp and γp along the cutting edge 

on the involute flank of the fellows' cutter. 

3.31 

2 

2.91 

0.94 

Excepted the presented determinations, that are essential for a good service 

of the fellows' cutter, it is possible to determine another angles, e,g. in sections 

with quidam planes, as in Fig. 8, or with front planes (Pf - Pf) as in Fig. 10. 

In this case too, the angles are directly posted by the Solid Works computer 

program. The values for the angles for the determinations from 

the Fig. 8 and 

Fig. 10 are presented in the charts from the Fig. 9 and Fig. 11. 

2.52 

0.95 

4.27 

M1 M2 M3 M4 

α n 2 0.94 2.52 2.24 

γ n 3.31 2.91 0.95 0.97 

Fig. 7 – Variation chart of the angles αO, γO (noted in the figure with αn, γn). 

2.24 

0.97 

α p 

γ p 

α n 

γ n


Fig. 8 – Sections with certain parallel planes with the intersections' presentation 

7 

6 

5 

4 

3 

2 

1 

0 

3.36 3.38 3.31 

2.91 

2.02 

4.44 

M1 M2 M3 M4 

α 0 2.02 2.91 4.44 6.44 

γ 0 3.36 3.38 3.31 3.52 

We can observe the anomaly of the variation of the normal back rake angle 

γO. A lot of verification proves that the model is correct, so it is to research the 

real situation. 

Fig. 9 – The charts of the angles αO and γO 

A proposal for a complete control of cutting tools made on the basis on the 

getting of a 3D solid model is new and brings the first advantage that it is 

possible to obtain all the angles searched at any point of the cutting edges or 

active planes. 

6.44 

3.52 

α 0 

γ 0


2.5 

2 

1.5 

1 

0.5 

0 

Fig. 10 – Intersections with frontal planes in different points. 

2.32 

0.48 

2.05 2.01 

0.56 

0.62 

M1 M2 M3 M4 

αf 2.32 2.05 2.01 1.93 

γ f 0.48 0.56 0.62 0.74 

Fig. 11 – Values for the angles αf şi γf in frontal plans Pf - Pf. 

1.93 

0.74 

αf 

γ f


4. Conclusion 

The method is especially useful for complex tools, small, with the active 

surfaces and curved cutting edges, to which access control with current 

instruments is very slowness or impossible and where the definition of 

theoretical planes control angles is difficult to apply. The other method of 

obtaining the 3D model through photography has its limits. Further research 

will develop methodologies for control of complex tools for very large or very 

small, with curved surfaces, methodologies able to change the angles of the long 

edges or surfaces with steps as small. 

Received: March 25, 2010 1 "POLITEHNICA" University from Bucharest , 

Department of Machines and Production Systems 

e-mail: eugen_strajescu@yahoo.com 

2 S.C. MUNPLAST S.A. 

e-mail: ostefu@yahoo.com 

3 "VALAHIA" University from Targoviste 

e-mail: ddumitru@yahoo.com 


. 

1. Minciu C., Stră jescu E., ş.a., Scule aşchietoare, Îndrumar de proiectare. 

Editura Tehnică, Bucureşti, 1995. 

2. S t r ă jescu, E., Proiectarea sculelor aschietoare, Litografia I.P.Bucureşti, l984. 

3. E n a c h e , Ş t., Stră jescu, E., Minciu, C., Metode şi programe pentru 

proiectarea asistată a sculelor aşchietoare. Litografia I.P.Bucureşti, 1988. 

4. Strajescu E., Pavlov O., Metodologie de control asistat de calculator al 

sculelor aşchietoare pe baza unor metode neconvenionale de obţinere a 

modelelor 3D. Conferinţa ICEEMS, Braşov, 2005. 

5. S t r ă jescu E., Pavlov O., Dumitru, D., C ontributions Concerning the 

Informatic Control of the Cutting Tools, International Conference on 

Manufacturing Science and Education - MSE Sibiu, 2009. 

CONTRIBUŢII PRIVIND CONTROLUL INFORMATIZAT 

AL CUŢITELOR ROATĂ 

(Rezumat) 

În lucrare se prezintă bazele unei metodologii pentru controlul informatizat al tuturor 

elementelor geometrice şi constructive ale sculelor aşchietoare în general, şi al cuţitelor roată 

pentru mortezat roţi dinţate cilindricve în special. Metodologia are la bază obţinerea unui model 

solid, iar acest lucru se poate face fie prin fotografiere, folosindu-se programe potrivite, fie prin 

scanare 3D, fie în faza de proiectare. Metoda propusă permite controlul absolut al tuturor 

parametrilor geometrici şi constructivi, în orice punct şi în orice plan, precum şi trasarea 

graficelor de variaţie. Se poate imagina o metodă prin care un program de proiectare asistată de 

tipul Solid Concept, parametrizat, să modifice unghiurile sculei pornind de la valoarea admisibilă 

a unui anume unghi într-un anume plan.


Publicat de 



Secţia 


CONTRIBUTIONS TO THE ELABORATIONS OF A 

GRAPHICAL METHOD FOR PROFILING OF TOOLS 

WHICH GENERATE BY ENVELOPING 

I. ALGORITHMS 

SILVIU BERBINSCHI, VIRGIL TEODOR, 

NICOLAE DUMITRAŞCU and NICOLAE OANCEA 

Abstract. The generation of the ordered curls profile (surfaces) associated 

with the centrodes of the rack-gear tool represent an fundamental problem for the 

profiles reciprocally enveloping associated with a couple of rolling centrodes. The 

knowing, in principle, of the generating rack-gear allow the construction of the 

tool and, also, the determination of the axial profile of the worm mill for the 

ordered curl profile’s generation. They are known and used more solutions for the 

solving of this problem: analytical solutions, based on the fundamentals theorems 

of gearing; complementary theorems, obtained from the fundamentals theorem; 

graphical methods based on graphical environments capabilities. In this paper, is 

proposed an algorithm for the approach of issue of ordered curl profiles based on 

the CATIA design environment capabilities. It is proposed a general algorithm 

which allows making applications, ended with the numerical determination of the 

rack-gear reciprocally enveloping with ordered curl profiles. They are solved 

profiles known in discrete form, substituted by spline approximation. The method 

allows drawing the gearing lines as so as the interference trajectories. The 

obtained results are comparing with results obtained by analytical methods. 

Key words: enveloping surfaces, rack-gear tool, graphical design method. 

BY 


The profiling of tools which generated by enveloping by the rolling method 

— rack gear tool and gear shaped tool — may be make by some methods:

42 Silviu Berbinschi et al 

- analytical methods, based on fundamental methods of surfaces enwrapping 

– first Olivier theorem, Gohman theorem, normals method, W i l l i s [1], [2]; 

- complementary analytical methods – “minimum distance” method, the 

“substitutive circles family” method, the “in-plane generating trajectories” 

method [3]-[5]; 

- graphical-analytical methods [6]; 

- graphical methods, using the capabilities of CAD software [7]. 

We mention that the methods proposed and used for the study of 

reciprocally enveloping surfaces respect the enveloping fundamental theorem. 

The proposed solutions leads to comparable results, in most cases 

identically, for the crossing profile tool’s shape, which generated by rolling 

ordered curls profiles associated with a couple of rolling centrodes. 

Literature include multiples applications for various domains as methods of 

tool’s profiling, revolution surfaces, for generating helical surfaces [8]; the 

modelling of the cylindrical surfaces with disk tools [9]; generation of the 

helical compressors rotors [10]. 

In this paper, is proposed a method for profiling of the rack-gear tools 

reciprocally enveloping with ordered curl profiles, based on the capabilities of 

the CATIA design environment. 

2. Generation Kinematics 

In principle, the tool’s profiling determination problem for rack-gear tool 

reciprocally enveloping with an ordered curls profile, associated with a circular 

centrode, assume to respect the kinematics of the generating process, see figure 

1. 

Fig. 1 – Couple of rolling centrodes. 

The two rolling centrodes, C1 —circle, associated with the profiles ordered 

curls and C2 —straight line, associated with the rack-gear tool, are in rolling 

movement, so, is respected the condition:


λ = R ⋅ ϕ , 

(1) rp 1 

where λ is the linear velocity in the translation of C2 centrode; 

R ⋅ ϕ — the value of velocity in point O1—the gearing pole, from C1 

rp 

1 

centrode, in the rotation movement around z axis; 

ϕ1 — variable angular parameter. 

In the rotation of C1 centrode the translation movements along the C2 

centrode and the rotation around Z axis are evenness. 

They are defined the reference systems: 

xyz is the global reference system, with z axis overlapped to the rotation axis 

of the C1 centrode; 

XYZ — mobile reference system, initially overlapped to the global reference 

system, joined with the ordered curls profile Σ; 

ξηζ — mobile reference system, joined with C2 centrode of rack-gear tool, 

with axis parallel and of the same sense with the global reference system xyz. 

The kinematics of the rolling process of the two centrodes, C1 and C2, 

tangents in point O1 — gearing pole— presume that the velocities of points 

belongs to the two centrodes, temporally situated in point O1, to be equals. 

In this way, the global motion of the ξηζ reference system, joined with 

centrode C2, is described by the transformation, 

(2) x = ξ + a , 

where: 

(3) 

⎛ξ⎞ ⎛x⎞ ⎜ ⎟ ⎜ ⎟ 

ξ = 

⎜ 

η 

⎟ 

; x = 

⎜ 

y 

⎟ 

⎜ζ⎟ ⎜z⎟ ⎝ ⎠ ⎝ ⎠ 

represent the matrix of the current points in the space ξηζ, respectively xyz; 

(4) 

⎛ 0 ⎞ 

⎜ ⎟ 

a = ⎜ −λ⎟ 

⎜ R ⎟ 

⎝−rp ⎠ 

is the matrix formed with the coordinates of point O1, in the global reference 

system, with λ instantaneous velocity in the translation movement of C1 

centrode and Rrp the value of circular centrode C1 (rolling radius). 

Also, the revolution movement of C1 centrode is described by transformation


⎛x⎞ ⎛X⎞ ⎜ ⎟ ⎜ ⎟ 

⎜ 

y 

⎟ 

= 1 1 ⋅ 

⎜ 

Y 

⎟ 

⎜z⎟ ⎜Z ⎟ 

⎝ ⎠ ⎝ ⎠ 

T 

(5) ω ( ϕ ) 

⎛X⎞ ⎜ ⎟ 

where 

⎜ 

Y 

⎟ 

is the matrix of the current point in space XYZ, and 

⎜Z⎟ ⎝ ⎠ 

⎛1 0 0 ⎞ 

⎜ ⎟ 

= ⎜ ⎟ 

⎜ ⎟ 

(6) ω1( ϕ1) 0 cos( ϕ1) sin( 

ϕ1) 

0 −sin( 

ϕ ) cos( 

ϕ ) 

⎝ 1 1 ⎠ 

is the rotation transformation matrix, around X axis, with angle ϕ1 

(counterclockwise rotation). 

The assembly of equations (3) and (6), with the respect of rolling condition 

(1), determine the relative motion, 

⎛ξ⎞ ⎛1 0 0 ⎞ ⎛X⎞ ⎛ 0 ⎞ 

⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟ 

⎜ ⎟ 

= ⎜ ⎟⋅ 

⎜ 

Y 

⎟ 

−⎜ −λ 

⎟ 

⎜ ⎟ ⎜0 −sin 1 cos ⎟ ⎜ 

1 Z ⎟ ⎜−R⎟ ⎝ ⎠ ⎝ ⎠ ⎝ ⎠ ⎝ rp ⎠ 

(7) η 0 cos( ϕ1) sin( 

ϕ1) 

ζ ( ϕ ) ( ϕ ) 

while the Σ profile, belongs to the profile’s ordered curl, associated with the C1 

centrode, in form, 

⎛ 0 ⎞ 

⎜ ⎟ 

Σ= ⎜ ⎟ 

⎜ ⎟ 

⎝ ⎠ 

(8) Y( u) 

Z ( u) 

with u variable parameter, describe a profile’s family in the rack-gear space: 

(9) ( ) 

( u, 

1 ) = 0; 

( , 1) ( ) cos( 1) ( ) sin( 

1) 

( , ) ( ) sin( ) ( ) cos( 

) 

ξ ϕ 

u ϕ1 

Y u Z u Rrp 

1 

u Y u Z u R 

; 

Σ η ϕ = ϕ − ϕ + ⋅ ϕ 

ζ ϕ = ϕ + ϕ − . 

1 1 1 

The enveloping of the profile family (9) represents the rack-gear tool’s 

profile. 

rp


Often, the profile (8) may be replaced by the equations of one surface, 

cylindrical or cylindrical helical, of which belongs the profile (8), as crossing 

in-plane profile section of Σ surface (the section plane is a perpendicularly plane 

on X axis), surface which is described by equations on form: 

(10) 

( ) 

( , ) ; 

( , ) , 

X = X u, t ; 

Σ Y = Y u t 

Z = Z u t 

with u and t independent variable parameters. 

To the profile’s family (9) was associated the enveloping condition 

specifically for one of the enwrapping fundamental theorems (first Olivier 

theorem, Gohman theorem, “normals method” [1], [2]) or one of the 

complementary theorem (“minimum distance”, “substitutive circles family”, 

“in-plane generating trajectories”) in order to determine the envelope of profile 

family as generating rack-gear profile [3], [4]. 

Also, are known graphical solutions for profiling tools which generate by 

enwrapping method, ordered curls profile associated with a rolling centrodes 

couple (the “regions” method [6], [7]). 

3. Kinematics Method in the CATIA Design Environment 

Is proposed a new solution for the profiling of rack-gear tool reciprocally 

enveloping with an ordered curl profiles associated with a couple of rolling 

centrodes, using the capabilities of CATIA software, by making an kinematics 

entity which will simulate the rolling movement of centrodes: C1 — rolling 

circle with radius Rrp, associated with the profiles curl; C2 — straight line, 

associated with the rack-gear. 

The proposed solution has the advantages that use the capabilities of a very 

versatile product, which may offer rigorous numerical results. 

Also, being a graphical method, the rough errors, due firstly to the passing 

curves, which may be erroneously considered as profile’s zone, are easy to 

identify and deleted from analysis. 

The proposed solve is based on the capabilities of Part environment where 

are synthesized the elements of a mechanism able to simulate the enveloping 

condition, in this case, the normal condition. These elements, created in the Part 

environment, are inserted in a file of Assembly environment, assuring the 

position of the mechanism elements in the start position, in following in the 

DMU Kinematics environment are defined the kinematics couples. 

The mechanism motion is made by the command Simulation, establishing 

the intermediate position number (Shots) and creating with the command Replay


a movie of the mechanism successive positions. 

With command Trace is draw the trajectory of any point which belongs to 

one element of the mechanism regarding any another element, including the 

global reference system, determining the gearing line between the profile to be 

generate and the profile of the rack-gear tool. 

These trajectories represent spline curves, constructed by successive points 

obtained by the mechanism rolling. The coordinates of these points may be 

exported as text file or any spreadsheet, see figure 2. 

Fig. 2 – Generating algorithms in CATIA environment. 


1. The rolling process kinematics for two centrodes becomes possible to be 

described in the graphical design environment as CATIA. 

2. The precision for these profile description, in this graphical environment, 

the geometrical forms and the kinematics of these allow a rigorous analysis of 

the reciprocally enveloping surfaces. 

3. A software suit specifically for the graphical design environment was 

synthesised in the end of paper, for the tool’s profiling reciprocally of an 

ordered curl of surfaces (profiles). Specifically applications will allow 

establishing the proposed method quality.


Acknowledgements. The authors gratefully acknowledge the financial support of the 

Romanian Ministry of Education, Research and Innovation through grant 

PN_II_ID_791/2008. 

Received:January, 20, 2010 Dunărea de Jos University of Galaţi, 

Manufacturing Science and Engineering Department 

Galaţi, Romania, 

e-mail: nicolae.oancea@ugal.ro 


1. L i t v i n F.L., Theory of Gearing Reference Publication 1212, NASA. Scientific and 

Technical Information Division, Washington D.C., 1984. 

2. O a n c e a N., Generarea suprafeţelor prin înfăşurare, Vol. I, Teoreme 

fundamentale, Editura Fundaţiei Universitare “Dunărea de Jos”, Galaţi, 2004, 

ISBN 973-627-106-4. 

3. O a n c e a N., Generarea suprafeţelor prin înfăşurare, Vol. II, Teoreme 

complementare, Editura Fundaţiei Universitare “Dunărea de Jos”, Galaţi, 2004, 

ISBN 973-627-106-4; 

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, 

Editura Fundaţiei Universitare “Dunărea de Jos”, Galaţi, 2004, ISBN (10) 

973-627-333-4; 

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 

gear profiles, Proceedings of the International Conference on Manufacturing 

Systems, ICMaS 2006, Ed. Academiei Române, Bucuresti, ISSN 1842-3183, pp. 

311-314; 

6. B a i c u I., O a n c e a N., Profilarea sculelor prin modelare solidă, Ed. 

Tehnica-Info, Chişinău, 2002, ISBN 9975-63-172-X; 

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 

danturare, Ed. Cermi, Iaşi, 2007, ISBN 978-973-667-270-5; 

8. M o h a n, L. V., S h u n m u g a m, M. S., An orthogonal array based optimization 

algorithm for computer-aided measurement of worm surface, Int. J. Adv. Manuf. 

Technol. (2006) 30: 434–443; 

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 

Surface Approximation for Reverse Engineering, Computing, 60, 307-322 

(1998); 

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 

determination of profile helical surfaces, International Journal of Machine Tools 

& Manufacture”, Elsevier Science, Pergamon, 38, 8, pp.1001-1015, UK, (1998).


CONTRIBUŢII LA ELABORAREA UNEI METODE GRAFICE PENTRU 

PROFILAREA SCULELOR CARE GENEREAZĂ PRIN ÎNFĂŞURARE 

I. ALGORITM 

(Rezumat) 

Generarea vârtejurilor ordonate de profiluri (suprafeţe), asociate unor centroide, cu 

scule de tip cremalieră, reprezintă o problemă fundamentală pentru profiluri reciproc 

înfăşurătoare asociate unui cuplu de centroide în rulare. Cunoaşterea principială a 

formei cremalierei generatoare permite construcţia sculei-pieptene şi, de asemenea, 

determinarea profilului sculei melc pentru generarea vârtejurilor ordonate de profiluri. 

Sunt cunoscute şi utilizate multiple soluţii pentru abordarea unei astfel de probleme: 

soluţiile analitice, bazate pe teoremele fundamentale ale angrenării; teoremele 

complementare, derivate din metodele fundamentale, metode grafice, bazate pe 

posibilităţile mediilor de proiectare grafică — CAD. 

În lucrare se propune un algoritm general pentru abordarea problematicii 

profilurilor (suprafeţe) bazat pe facilităţile mediului de proiectare CATIA. Se propune 

un algoritm general, în baza căruia se realizează aplicaţia, finalizată cu determinarea 

numerică (grafică) a formei cremalierei reciproc înfăşurătoare vârtejului ordonat de 

profiluri. 

Sunt tratate profiluri cunoscute analitic sau sub formă de coordonate discrete, 

substituite prin forme de aproximare spline. 

Metodica permite tratarea liniilor de angrenare precum şi traiectoriile de 

interferenţă a profilurilor. Se compară rezultatele obţinute prin această metodă cu 

rezultatele obţinute prin metode analitice.


Publicat de 



Secţia 


CONTRIBUTIONS TO THE ELABORATIONS OF A 

GRAPHICAL METHOD FOR PROFILING OF TOOLS WHICH 

GENERATE BY ENVELOPING 

II. APLICATION FOR RACK-GEAR TOOL’S PROFILING 

SILVIU BERBINSCHI, VIRGIL TEODOR, 

NICOLAE DUMITRAŞCU and NICOLAE OANCEA 

Abstract. In this paper is presented an application of the general algorithm 

regarding the tool’s profiling which generate by enveloping, by rolling method, 

respectively the rack-gear tool generating an ordered curl of profiles. They are 

presented the successive steps of this process and, also, numerical examples 

regarding the crossing profile of the rack-gear tool. The method quality is proofed 

by comparing the obtained results with results obtained for the same problem, 

using one of the analytical methods, the Willis method, and software developed in 

Java. 

Key words: enveloping surfaces, rack-gear tool, graphical design method. 

BY 


It was present the general algorithm for profiling the tools generating by 

enveloping, by the rolling method, in the CATIA design environment. 

In following are presented an application of the proposed algorithm for 

rack-gear tool’s profiling generating ordered curl profiles, for a hexagonal shaft, 

as so as, the comparison for the profile determination of the rack-gear tool with 

results obtained by analytical methods, based on specialised software, realised 

in Java [6], [7]. 

In the paper entitled “Contributions to the Elaborations of a Graphical Method for 

Profiling of Tools Which Generate by Enveloping. I. Algorithm” is proposed a new


solution for the profiling of rack-gear tool reciprocally enveloping with an 

ordered curl profiles associated with a couple of rolling centrodes, using the 

capabilities of CATIA software, by making an kinematics entity which will 

simulate the rolling movement of the two centrodes [8]. 

2. Application for Rack-Gear Profiling 

In order to obtain the tool’s profile for various polygonal shafts, for various 

side lengths, is need to follow two stages. The first stage is to choose the side 

length and the diameter of the rolling circle as input data. This thing may be 

done introducing in an text or Excel file the value, in following, the CATIA 

software will automate modify the whole mechanism, with the new values. In 

the same stage is created the mechanism, the rolling of this and the tool’s profile 

determination. The coordinates of points belongs to profile will be exported in a 

text or Excel file. 

The second stage consist in the partial remake of the mechanism couples, 

the rolling of this for the new tool’s profile, for the previous modified values 

and the export of points in a new file. 

2.1. First stage 

2.1.1. Data input and mechanism elements construction. In this step are 

created a folder with all files needed for the mechanism. 

First of all for the input data is created a text of excel file with the names 

and the values of the input parameters, namely the piece shaft radius Ra, the 

polygonal side length L and the rolling circle radius of blank Rrp. These are the 

parameters which may be modified by user. This file will be linked with the 

parameters from the CATIA file. 

In the following example is considered a hexagonal shaft with L=50 mm 

and the including radius Ra=50 mm. In order to simplify the representation we 

consider the rolling radius Rrp=50 mm. The shaft sketch and the associated 

centrode are represented in figure 1. The mechanism elements are successive 

created, in the Part environment, are suggestive named and inserted in an 

assembly file, created in the Assembly environment, named GenCremaliera. 

The Part type files are: baza; piesa; cremaliera and tachet. 

The Baza file contains the standing part of the mechanism and represent 

more lines needed to mount others elements, see figure 2. 

From a point named OriginePiesa is construct a line named AxaArbore, 

representing the piece’s axis and next another line which will be as guide for the 

rolling line named GhidajDreaptaRulare. The segment having the rolling radius 

length and which contain the gearing pole is created in order to ease the


mechanism assembling. The point PolulAngrenarii is situated on the rolling 

circle with radius Rrp. 

Fig. 1 – The shaft sketch. 

This file contain a parameters table (Design Table) named DateIntrare 

which link the line and point’s parameters from the Part file (from CATIA) 

with the input data from the file DateIntrare.txt. 

The file Piesa contain, firstly, the geometrical elements of the profile for 

which is determined the tool’s profile, see figure 2. Another sketch from this 

file is CercRulare which contain a circle with radius Rrp. The two values Ra and 

Rrp are linked with the text file DateIntrare.txt. The file Cremaliera contain the 

rolling straight line (DreaptaRulare) representing a circular pitch of the curl to 

be generated and having the length as function of the rolling radius Rrp, 

controlled by file DateIntrare.txt. The file Tachet have the straight lines which 

are tangent and respectively normal at the piece’s profile in the contact points. 

Fig. 2 – Assembly file with kinematics couples and tool’s profile.


2.1.2. Mechanism assembling and the creation of kinematics couples. In the 

assembly 

file GenCremaliera are inserted all the above described elements, 

with the command Insert Existing Component. 

These have geometrical and dimensional constraints in order to be 

positioned in accordance with the mechanism functionality. After inserting the 

constraints, the GenCremaliera contain the whole mechanism in the initial 

position. In the Assembly environment the mechanism elements were positioned 

but doesn’t exist yet any couple between them. 

The kinematics couples are created in the DMU Kinematics environment. It 

is created the mechanism named Generare.Scula.Cremaliera and is established 

the standing element, Baza. 

After the kinematics couples creation the mechanism function may be 

simulated and look as in figure 2. 

2.1.3. Mechanism rolling and determination of tool’s profile. In order to 

rolling 

this mechanism is created a simulation with command Simulation and 

are established the intermediate position number, Shot, number which will have 

influence to the tool’s profile precision. For this example we considered 500 

steps, so it will result 500 points. With this simulation is determined the tool’s 

profile, see Figure 3. 

0 

1 

2 

3 

4 

5 

6 

7 

ζ 

η 

Fig. 3 – Tool’s profile in the rack-gear reference system. 

2.1.4. Points coordinates extraction. After the determination of the tool’s 

profile, these may be exported in a text file using the command Design Table, 

from the Part environment. 

This command will generate a text file PuncteProfil.txt, see Table 1.


Table 1 

Points coordinates on tool’s profile in CATIA program 

Nr. Crt. η [mm] ζ [mm] Nr. Crt. η [mm] ζ [mm] Nr. Crt. η [mm] ζ [mm] 

1 0 0 170 17.07 -5.96 340 36.18 -5.81 

17 1 .41 -0 .78 187 18.96 -6.24 357 38.04 -5.44 

34 2.98 -1.57 204 20.86 -6.45 374 39.88 -5 

51 4.59 -2.31 221 22.78 -6.6 391 41.69 - 4.51 

68 6.26 -3.01 238 24.71 -6.68 408 43.46 -3.96 

85 7.98 -3.64 255 26.63 -6.7 425 45.2 -3.35 

102 9.73 -4.23 272 28.56 -6.65 442 46.89 -2.69 

119 11.52 -4.75 289 30.48 -6.54 459 48.53 -1.97 

136 13.34 -5.21 306 32.39 -6.36 476 50.13 -1.21 

153 15.19 -5.62 323 34.29 -6.11 501 52.36 0 

2.2. Second stage 

In the second stage if the user modify the DateIntrare.txt 

file saving the 

modification of the input parameters, the CATIA software detect this and 

warning 

by an Update message. 

After this message the user have to update the assembly file and the 

kinematics file in order to remake 

the kinematics couples which may be 

modified. 

3. Method Quality Verification 

It is proposed in following the verification of this method 

by comparing the 

numerical results obtained with graphical method 

regarding the results obtained 

by one of the analytical methods: Gohman method; “normals” method; 

“minimum distance” method etc [3]. 

For the polygonal shaft is defined the analytical form of the profile to be 

generated, see figure 4: 

(1) 

X = 0; 

Σ Y = u; 

Z = 

a,


with u variable parameter, measured along the profile and a constant value 

depending the profile form. 

The Σ profile family is on form: 

(2) ( ) 

Fig. 4 – Straight lined profile. 

ξ = 0; 

Σ η = ucosϕ − asinϕ + R ϕ; 

ϕ 

ζ = usinϕ + acos ϕ −R 

, 

with ϕ variable parameter. 

The 

specifically enveloping condition, in the “in-plane generation trajectory” 

methods is: 

(3) 

or 

(4) 

cosϕ 

−usinϕ 

− acosϕ 

+ Rrp 

= , 

sinϕ ucosϕ − asinϕ 

u = R sinϕ 

. 

The equations 

assembly (2) and (4), for u variable between limits 

umin 

=−0.5⋅ Rrp ; umax = 0.5⋅Rrp , for a hexagonal shaft, represent the rack-gear 

tool’s profile reciprocally enwrapping with the shaft’s profile. 

In Table 2 are presented the tool’s profile coordinates determined by the 

analytical 

method [6], [7]. 

In Figure 5 are presented the profiles form and the errors obtained at 

profiling, regarding the coordinates 

transformation: 

rp 

rp 

rp

(5) 

Bul. Inst. P olit. Iaşi, t. LVI (LX), f. 2, 2010 

55 

ζ = −ξ 

CATIA JAVA 

η = η + 26.1799. 

CATIA JAVA 

Table 2 

Points coordinates on tool’s profile—normal method, in Java program 

Crt. 

no. ξ [mm] η [mm] Crt. no. ξ [mm] η [mm] Crt. no. ξ [mm] η [mm] 

1 1.40E-06 -26.1799 248 6.698025 -0.28286 496 0.199296 25.83104 

2 0.05004 -26.093 249 6.698477 -0.1695 497 0.149679 25.9186 

3 0.099941 -26.0059 250 6.698703 -0.05562 498 0.099923 26.00594 

4 0.149712 -25.9185 251 6.698701 0.057729 499 0.050028 

26.09306 

5 0.19934 -25.831 252 6.698471 0.171609 500 8.88E-07 26.17994 

M M M M M M M 

Fig. 5 – Errors obtained. 

4. Conclusion 

1. The proposed algorith m use the CATIA design environment capabilities. 

2. The Tachet mechanism used in the problem solving is universal, 

excepting the profiles with singular points. 

3. The profiling precision of rack-gear tool is comparable with results 

obt ained by analytical methods, based on which was created dedicated software, 

in present 

paper, in Java language. 

Acknowledgements. The authors gratefully acknowledge the financial support of the 

Romanian Ministry of Education, Research and Innovation through grant 

PN_II_ID_656/2007. 

Received:January, 

20, 2010 Dunărea de Jos University of Galaţi, 

Manufacturing Science and Engineering Department 

Galaţi, Romania, 

e-mail: nicolae.oancea@ugal.ro 

;



1. L i t v i n F. L., Theory of Gearing Reference Publication 1212, NASA. Scientific 

and Technical Information Division, Washington D.C., 1984; 

2. O a n c e a N., Generarea suprafeţelor prin înfăşurare, Vol. I, Teoreme 

fundamentale, Editura Fundaţiei Universitare “Dunărea 

de Jos”, Galaţi, 2004, 

ISBN 973-627-106-4; 

3. O a n c e a N., Generarea suprafeţelor prin înfăşurare, Vol. II, Teoreme 

complementare, Editura Fundaţiei Universitare “Dunărea de Jos”, Galaţi, 2004, 

ISBN 973-627-106-4; 

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, 

Editura Fundaţiei Universitare “Dunărea de Jos”, Galaţi, 2004, ISBN (10) 

973-627-333-4; 

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 

danturare, Editura Cermi, Iaşi, 2007, ISBN 978-973-667-270-5; 

6. C u c u M., O a n c e a N., T e o d o r V., Discretly Known Reciprocally Enwrapping 

Surfaces Representation Model - Software’s Descriptions, Buletinul Institutului 

Politehnic din Iaşi, Publicat de Universitatea Tehnica”Gh.Asachi” Iaşi, LII 

(LVI), Fasc.5A, Secţia Construcţii de Maşini (2006), pp. 229-232; 

7. C u c u M., O a n c e a N.,T e o d o r V., Metoda tangentelor – profilarea 

sculei-cremalieră pentru profiluri circulare, Tehnologii Moderne 

Calitate 

Restructurare (2007), Universitatea Tehnică a Moldovei, 2, Chişinău, pp. 86-89, 

ISBN 978-975-45-034-8, ISBN 978-9975-45-035-2. 

8. * * * CATIA Version 5.19. DS.- Documentation 

CONTRIBUŢII LA ELABORAREA UNEI METODE GRAFICE PENTRU 

PROFILAREA SCULELOR CARE GENEREAZĂ PRIN ÎNFĂŞURARE 

I. APLICATIE PENTRU PROFILAREA SCULEI-CREMALIERA 

(Rezumat) 

În lucrare este prezentată o aplicaţie a algoritmului general prezentat, privind 

profilarea sculelor care generează prin înfăşurare, prin metoda rulării, în speţă 

scula-cremalieră generatoare a unui vârtej ordonat de profiluri. Sunt prezentaţi 

paşii 

succesivi ai procesului de profilare a sculei cremalieră, de asemenea, un exemplu 

numeric privind forma profilului transversal al sculei 

cremalieră. 

Calitatea metodei propusă este dovedi tă prin comparaţia rezultatelor obţinute, în 

cadrul 

aceleiaşi probleme concrete de profilare, utilizând însă una dintre metodele 

analitice — metoda Willis, şi un produs soft dezvoltat în Java.


Publicat de 



Secţia 


THE NUMERICAL SIMULATION OF TURNING PROCESS 

USING FINITE ELEMENT MODELING 

BY 

CATALIN FETECAU, DANIEL-VIOREL VLAD 

and COSTEL MOCANU 

Abstract. In this paper we present a study on numerical simulation of turning 

process using finite element analysis program AdvantEdge TM for AISI 1060 

carbon steel material. The main objective is to determine the influence of different 

cutting parameters (depth of cut, feed rate and cutting speed) on cutting force, von 

Mises stress and temperature in the longitudinal turning process. The predict of 

cutting forces by numerical simulation is important for the optimization of the 

process. The simulations were conducted without coolant. 

Key words: turning, numerical simulation, cutting forces, temperature. 


The turning process is probably the most important process of all cutting 

processes. It is therefore very important to analyze the process for to 

optimization it. 

Cutting force has a value and direction of action that depend on the quality 

of processed material, on the cutting regime elements, geometric parameters of 

cutting tools, cooling-lubricating liquids [4]. 

The resultant cutting force is defined by the relation: 

(1) R = ( F + F + F ) ≈(1.1…1.18)·Fx, 

where: 

Fz=(0.2…0.3)·Fx, 

Fy=(0.25…0.4)·Fx. 

2 

x 

2 

y 

2 

z

58 Cătălin Fetecău et al. 

n 

piece 

Fz 

tool 

Fx 

Fy 

chip 

feed rate 

Fig. 1 – 3D cutting force components according AdvantEdge TM . 

2. Numerical Simulation of Cutting Process 

2.1. The Mathematical Model 

Computer simulation of cutting process was achieved with AdvantEdge TM 

program. Elasto-plastic behavior of material during the cutting process is 

characterized by a nonlinear relationship between stresses and strains. In the 

mathematical modeling of surface flow movement, in software, are 

implemented a series of mathematical models, where the most important being 

model Power Law and model Drucker Prager [5]. 

Model Power Law defined by the relation 

(2) p 

p 

σ( ε ε&, 

T ) = g( 

ε ) ⋅ Γ( 

ε& 

) ⋅ Θ( 

T ) 

p 

where: ( ) 

, , 

g ε is strain hardening, Γ ( ε& ) - strain rate sensitivity, ( T ) 

softening. 

Model Drucker-Prager defined by the relation 

(3) p 

σ ( ε 

p 

J1 

, & ε, 

T ) = G( 

ε , J1) 

⋅Γ( 

& ε ) ⋅Θ( 

T ) 

where: G( , J ) p 

sensitivity, ( T ) 

, , 

t 

Θ - thermal 

ε 1 is strain hardening plus hydrostatic pressure, Γ (ε& ) - strain rate 

p 

Θ - thermal softening, ε - the plastic strain. 

p 

The function g( ε ) for the Drucker-Prager model is defined as

p 

(4) g( 

) 

(5) p 

g( 

) 


1 n 

⎛ p ⎞ 

⎜ ε 

p p 

ε = σ 0 1 + ⎟ , if ε < ε cut ; 

⎜ p ⎟ 

⎝ ε 0 ⎠ 

1 n 

⎛ p ⎞ 

⎜ εcut 

0 1 ⎟ 

p p 

ε = σ + , if ε ≥ ε , 

⎜ p 

cut 

⎟ 

⎝ ε0 

⎠ 

p 

p 

where: σ 0 is the initial yield stress, ε - the plastic strain, ε 0 - the reference 

p 

plastic strain, ε cut - the cut-off strain, n - the strain hardening exponent. 

The rate sensitivity function Γ ( ε& ) for the Drucker-Prager model is defined 

as 

⎛ & ε ⎞ 

⎜ 

+ 

⎟ 

⎝ & ε 0 ⎠ 

1 

(6) Γ( 

& ε ) = ⎜1 

⎟ , if & ε ≤ & εcut 

; 

(7) 

1 

1 

m 

⎛ 1 1 ⎞ 

⎜ − 

⎟ 

⎝ m1 

m2 

⎠ 

⎛ & ε ⎞ m2 

⎛ & ε ⎞ 

Γ( 

& ) = 

⎜ + 

⎟ 

⎜ cut 

ε 1 1+ 

⎟ , if & ε > & εcut 

. 

⎝ & ε0 

⎠ ⎝ & ε0 

⎠ 

( ) 

The thermal function Θ T for the Drucker-Prager model is defined as 

2 3 4 5 

(8) ( ) 

Θ T = c + c T + c T + c T + c T + c T . if T < T , 

(9) ( T ) = Θ( 

T ) 

0 

1 

2 

3 

T −Tcut 

Θ cut − 

if T ≥ Tcut 

, 

T −T 

where: c0 through c5 

are coefficients for the polynomial fit, T - the 

temperature, 

temperature. 

Tcut - the linear cut-off temperature, T melt - the melting 

melt 

4 

cut 

2.2. Finite Element Machining Model 

The turning simulations were carried out on cylindrical extruded bar with 50 

mm diameter and 80mm length for AISI 1060. Length of cut L=20mm. The 

cutting tests were conducted without coolant. 

Order to study the simulation results, we set a plan of experiments as Table 

1. We set the tool’s parameters (tool’s angle and material), process parameters 

(feed rate, f, cutting speed, v and depth of cut, t). The tool’s material is carbide 

plates which had the parameters according to the Table 1. 

5 

cut


Table 1 

The plan of the experiments 

PROCESS PARAMETERS 

TOOL 

feed, f, speed, v, depth, t, 

PARAMETERS 

mm/rev rev/min mm 

Clearance angle, α, º 7 0.083 

Side rake angle, γ, º -5 0.208 

Lead angle, χ, º 45 0.416 

Nose radius, r, mm 0.4 0.5 

78.54 

157.08 

314.16 

The mechanical properties and chemical composition of work-piece 

material, AISI 1060, are show in Table 2. 

Table 2 

The chemical composition and mechanical properties of AISI 1060 

Chemical composition, % Yield Ultimate Tensile Hardness 

Material 

Cu Mn P S 

strenght, 

MPa 

Strenght 

MPa Bhn 

AISI 

1060 

0.605 0.75 0.04 0.05 370 625 179 

For meshing were used triangular elements with 3 nodes, both for piece and 

for the tool. AdvantEdge TM program automatically generates the finite element 

network after that specified a maximum length of finite element, mean that the 

smallest and most finite element of edge length [5]. 

Fig. 4 – The finite element network for the whole piece-tool-chip. 

The Druker-Prager model was used to simulate the process of chip breaking. 

1


Fig. 5 – The finite element network for the whole piece-tool-chip with continuous chips 

of plastic strain for AISI 1060, f=0.208 mm/rev; v=157.08 m/min. 

Fig. 6 – Von Mises Stress f=0.208 mm/rev; v=157.08 m/min. 

3. The results of numerical simulations 

In according with the plan of experiments, it was made numerical 

simulation for each cutting regime. For different values of the feed rate and the 

cutting speed the average values of tangential force Fx in according with 

AdvantEdge TM are presented in Table 3. 

Table 3 

The average values of tangential force, Fx, N 

Feed, f, Cutting speed, v, m/min 

mm/rev 314,16 157,08 78,54 

0.083 306.177 308.298 315.186 

0.208 544.703 557.041 544.703 

0.416 880.12 950.191 1073.24 

0.5 1015.5 1135.32 1205.5


The value of tangential force increases when feed rate increase as shows in Fig. 

no. 7. 

F [N] 

1400 

1200 

1000 

800 

600 

400 

200 

0 

0 0.1 0.2 0.3 0.4 0.5 0.6 

v1 

v2 

v3 

f [mm/rev] 

Fig. 7 – The effect of feed rate on cutting force. 

For the simulation made, the average values of temperature of cutting 

tool are presented in Table 4. 

Table 4 

The values of the temperature T, ºC 

Feed, f, 

Cutting speed, v, m/min 

mm/rev 314,16 157,08 78,54 

0.083 571.408 480.065 427.476 

0.208 699.709 589.149 496.942 

0.416 714.676 611.191 699.709 

0.5 729.953 622.045 729.953 

Fig. 8 – The temperature distribution in the cutting zone. 

In the FEM software AdvantEdge TM , the Johnson-Cook constitutive model 

were used to predict the thermal behaviour in longitudinal turning process [1]. 

This software allows the temperature distribution, cutting forces and von 

Misses stress to be predicted to appropriate measurements [2].


With AdvantEdge TM is possible to determine optimum machining 

parameters with lower costs and in short time without experimental processes 

[3]. 

4. Conclusion 

1. This paper presents a study on numerical simulation of turning process 

using finite element analysis for AISI 1060 material. Simulation was done using 

AdvantEdge TM program. Based on numerical simulations we determine the 

influence of cutting parameters (depth of cut, feed rate and cutting speed ) on 

cutting force, von Mises stress and temperature from cutting zone. 

2. Based on numerical simulations to determine the size variation of cutting 

force and temperature parameters based on cutting regime. 

3. It was also determined variation von Mises stress across-chip track-and to 

investigate the formation of the chips. 

4. Future work will be as based on results of the numerical simulation will 

to make the experiments and will compare the results with those of simulation. 

Acknowledgements. The work of Drd. Daniel-Viorel Vlad was supported by Project 

SOP HRD – EFICIENT 61445". 

Received: March 20, 2010 ”Dunarea de Jos” University of Galati, 

Department of Machine Manyfacturing 

Galati, Romania, 

e-mail: catalin.fetecau@ugal.ro 


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 

process using power law temperature dependent concept. Archives of Materials 

Science and Engineering, 29, 2 (2008), pp. 105-108. 

2. Ő z e l T., Modeling of hard part machining: effect of insert edge preparation in 

CBN cutting tool. Journal of Materials Processing Technology, 141, (2003) pp. 

284-293. 

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 

simulation of precision machining on AISI 1045, 2008, 5º Congresso Luso- 

Moçambicano de Engenharia, Maputo, Moçambique, pp.1-7(ref:25A003). 

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 

machining of steel AISI 1045 using FEM analysis, Material and Design., 30, 

2009, pp. 160-165.


5. *** Third Wave Systems AdvantEdge TM FEM. User’s manual, Minneapolis, 2008. 

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 

temperature distribution in the cutting zone in turning processes with differently 

coated tools, 13 th International Scientific Conference on Achievements in 

Mechanical and Materials Engineering, 2005. 

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 

models for the chip–tool interface using an Arbitrary Lagrangian–Eulerian finite 

element model, Wear. 2008, 265, pp. 542-460. 

SIMULAREA PROCESULUI DE STRUNJIRE FOLOSIND 

ANALIZA CU ELEMENT FINIT 

(Rezumat) 

In această lucrare se prezintă un studiu privind simularea numerică a procesului de 

strunjire folosind programul de analiză cu elemente finite AdvantEdge TM în cazul, 

oţelului OL60 (AISI 1060). Obiectivul principal este de a determina înfluenţa 

parametrilor regimului de aşchiere (adâncime de aşchiere, avans şi vizeza de aşchiere) 

asupra componentei tangenţiale a forţei de aşchiere, Fx, a tensiunii von Mises şi a 

temperaturii din zona de aşchiere. 

Comportarea elasto-plastică a materialului în timpul procesului de aşchiere este 

caracterizată de o relaţie neliniară dintre tensiuni şi deformaţii. 

În vederea modelării matematice a deplasării suprafeţei de curgere, în programul 

AdvantEdge TM sunt implementate o serie de modele matematice, cele mai importante 

fiind: Model Power Low şi Druker Prager Model. Pentru simulările efectuate , am 

folosit modelul Druker Prager. 

Forma aşchiei reprezintă un indicator al condiţiilor de aşchiere, arătând gradul de 

deformare plastică suferit de stratul de material detaşat. Forma aşchiei depinde de natura 

materialului prelucrat, geometria sculei, regimul de aşchiere etc. Se deosebesc aşchii de 

rupere şi aşchii de deforrnare plastică. 

Rezultatele obţinute vor sta la baza realizării experimentelor fizice, astfel încât să 

putem confrunta rezultatele obţinute din simulare cu cele din experiment.


Publicat de 



Secţia 


EXPERIMENTAL RESEARCHES REGARDING A NEW 

METHOD FOR CUTTING FORCES EVALUATION 

BY 

MIRCEA COZMÎNCĂ, CRISTIAN CROITORU 

and CĂTĂLIN UNGUREANU 

Abstract. The paper presents some preliminary experimental researches 

regarding the cutting forces level when cutting four types of steel. The aim of 

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 

theoretical cross section (txs), the compression yield point of the cut material and 

the chips contraction coefficient, as presented in our previous papers. The chips 

contraction coefficient includes all the working parameters’ influences and also 

the interdependencies between them. Beside those elements, the calculus models 

contain a theoretical component, depending on the constructive angles of the tool, 

γ, K and λ and the friction coefficient between chips and the cutting tool’s rake 

face. 

Key words: model, chips contraction, cutting force, experimental. 


The previous papers [1]-[4] were presented the new calculus models for 

cutting force evaluation (Eqs. 1), that could replace the traditional equations for 

Fz, Fx, Fy components. The theoretical quantities Cz, Cx, Cy in Eqs. (1) result 

using the Eqs.(2) and (3), if the chips deviation angle η=0 [5], and with Eqs.(4) 

if η≠0. 

(1) 

F 

F 

F 

z 

x 

y 

= C 

= C 

z 

= C 

x 

y 

σ 

σ 

σ 

o 

o 

o 

t s 

t s 

t s 

C 

C 

C 

1+ 

m 

d 

z 

1+ 

m 

d 

x 

1+ 

m 

d 

y 

, 

, 

.

66 Mircea Cozmîncă et al. 

(2) 

(3) 

(4) 

C 

C 

C 

z 

x 

y 

= cosγ 

cos 

k 

k 

k 

k 

C 

C 

C 

1 

2 

3 

4 

z 

x 

y 

= k cosλ, 

1 

= k 

2 

= k cos K − k k . 

3 

= cosγ 

= cosγ 

sin K + k k 

N0 

N0 

= sin λ cos K, 

= sin λ sin K. 

1 

1 

3 

4 

( 1+ 

μ tgγ 

N ) , 

0 

( μ − tgγ 

) , 

N0 

( λ + η)[ 

1+ 

μtgγ 

] , 

( λ + η) 

cos K[ 

1+ 

tgγ 

] + cosγ 

sin( 

K −η 

)[ μ − tgγ 

] , 

= cosγ 

sin 

= cosγ 

cos( 

K −η 

)[ μ − tgγ 

] − cos K sin( 

λ + η) 

sin K[ 

1+ 

μtgγ 

]. 

In Eqs.(3), the rake angle γ is measured in the plane normal to the active 

cutting edge and in Eqs.(4) this angle is measured in the plane including the 

chips’ real flow direction (η≠0). In the same equations, μ represents the friction 

coefficient between the chips and the tooth’s rake face, considering μ=0.5...0.6 

in designing activities [5]. 

Using the models in Eqs.(1) ask for the quantities mx, mz, my, that are 

amplifying the cutting forces level through the chips contraction coefficient. 

Therefore, experimental tests are necessary for the forces Fz, Fx, Fy and also for 

the chips contraction coefficient Cd, when cutting with a single tooth. 

2. Experimental Tests to Validate the Models for Fz, Fx, Fy 

In order to validate the models in Eqs.(1) for the force components Fz, Fx, Fy 

on the cutting tooth, a Kistler dynamometer is used to measure the main force 

component Fz when cutting four different types of steel. In the same time, the Cd 

values are determining by measurement of the chips lengths. These chips are 

resulted from certain cutting lengths limited by four longitudinal and equidistant 

channels executed on the workpiece and refilled with brass wedges, in order to 

ensure the continuity of the cutting process [5]. 

The experimental values for Cd result using the model in Eq.(5), where the 

cutting lenth L = 0.25πd–c and the chip’s length, La, result by measuring a 

number of n chips. The quantity c represents the channel width, measured on 

the workpiece circumference of diameter D. 

,

(5) 

Bul. Inst . Polit. Iaşi, t. LVI (LX), f. 2, 2010 

67 

C 

d 

L 

= 

L 

a 

, 

The experimental tests were conducted using five lathe tools (Co...C4) 

having different values of the main constructive angles, as presented in Table 1. 

This table includes also the calculated values for k1... k4 and Cz, Cx, Cy, based on 

Eqs.(2) and (3). 

Table 1 

Lathe tools’ constructive geometry 

Lathe 

tool 

Main constructive geometry 

γ λ K α r 

[°] [°] [°] [°] [mm] 

k1 Values calculated with Eqs. (2) and (3) 

k2 k3 k4 Cz Cx Cy Co 0 0 70 10 0 1 0.5 0 0 1 0.47 0.17 

C1 -10 0 70 10 0 0.898 0.67 0 0 0.898 0.63 0.23 

C2 0 -10 70 10 0 1 0.5 -0.06 -0.16 0.985 0.41 0.33 

C3 0 10 70 10 0 1 0.5 0.06 0.16 0.985 0.53 0.01 

C4 10 0 70 10 0 1.07 0.32 0 0 1.07 0.3 0.01 

The specific working conditions and the experimental results (Fz, Cd, mz) 

are presented in Tables 2 – 6. The values of mz were obtained using Eq.(6), 

where Cz is according to Table 1 and σo is a steel type characteristic. 

(6) 

F 

zexp 

z 

L 

a 

o 

= 

n 

∑ 

i= 

1 

L 

= C σ t s C 

ai 

1+ 

mz 

d 

Table 2 

OL44 – The influences of tool´s geometry and cutting feed 

Cut 

material/ 

Tool´s type 

σo [daN/ 

mm 2 ] 

t 

[mm] 

s 

[mm/ 

rev] 

n 

[rev/ 

min] 

L 

[mm] 

La 

[mm] 

D 

[mm] 

Cd 

Fz 

[daN] 

mz 

OL44/ Co 29 2 

0.1 

0.25 

630 

630 

42.5 

42.5 

16.52 

17.38 

58.5 

58.5 

2.57 

2.44 

54.9 

113.5 

1.38 

1.30 

OL44/ C1 29 2 

0.1 

0.25 

630 

630 

38.5 

38.5 

13.03 

14.18 

53.5 

53.5 

2.91 

2.72 

63.4 

125.8 

1.34 

1.27 

OL44/ C2 29 2 

0.1 

0.25 

630 

630 

41.5 

41.5 

14.25 

15.11 

57.3 

57.3 

2.91 

2.75 

58.16 

123.6 

1.17 

1.14 

OL44/ C3 29 2 

0.1 

0.25 

630 

630 

41.5 

41.5 

15.22 

16.18 

57.3 

57.3 

2.73 

2.57 

47.9 

118.0 

1.12 

1.24 

OL44/ C4 29 2 

0.1 

0.25 

630 

630 

41.5 

41.5 

16.47 

17.33 

57.3 

57.3 

2.52 

2.39 

47.65 

118.0 

1.21 

1.32 

n. 

.


Table 3 

40C10 – The influences of tool´s geometry and cutting feed 

Cut 

material/ 

Tool´s type 

σo [daN/ 

mm 2 ] 

t 

[mm] 

s 

[mm/ 

rev] 

n 

[rev/ 

min] 

L 

[mm] 

La [mm] 

D 

[mm] 

Cd Fz [daN] 

mz 

40C10/ Co 78 2 

0.1 

0.25 

630 

630 

36.56 

36.56 

19.86 

20.74 

51.0 

51.0 

1.84 

1.76 

52.5 

110.0 

0.98 

0.82 

40C10/ C1 78 2 

0.1 

0.25 

630 

630 

37.03 

37.03 

17.49 

19.38 

51.6 

51.6 

2.12 

1.91 

59.0 

117.3 

0.91 

0.86 

40C10/ C2 78 2 

0.1 

0.25 

630 

630 

36.56 

36.56 

20.32 

22.84 

51.0 

51.0 

1.80 

1.60 

55.0 

113.0 

1.16 

1.29 

40C10/ C3 78 2 

0.1 

0.25 

630 

630 

36.56 

36.56 

23.8 

26.36 

51.0 

51.0 

1.54 

1.39 

54.7 

104.7 

1.95 

2.05 

40C10/ C4 78 2 

0.1 

0.25 

630 

630 

36.56 

36.56 

21.14 

22.62 

51.0 

51.0 

1.73 

1.62 

40.4 

103.4 

0.60 

0.88 

Cut 

material/ 

Tool´s type 

MoCM14/ 

Co 

MoCM14/ 

C1 

MoCM14/ 

C 2 

MoCM14/ 

C3 

MoCM14/ 

C 4 

Table 4 

MoCM14 – The influences of tool´s geometry and cutting feed 

σo [daN/ 

mm 2 ] 

t 

[mm] 

s 

[mm/ 

rev] 

n 

[rev/ 

min] 

L 

[mm] 

La [mm] 

D 

[mm] 

Cd 

Fz 

[daN] 

mz 

74 1 

0.1 

0.25 

400 

400 

51.09 

51.09 

30.53 

32.16 

69.5 

69.5 

1.67 

1.59 

57.9 

116.4 

2.99 

2.96 

74 1 

0.1 

0.25 

400 

400 

50.54 

50.54 

23.47 

25.12 

68.8 

68.8 

2.15 

2.01 

60.9 

134.3 

1.88 

1.98 

74 1 

0.1 

0.25 

400 

400 

50.14 

50.14 

28.11 

29.96 

68.3 

68.3 

1.78 

1.67 

58.9 

119.5 

2.60 

2.64 

74 1 

0.1 

0.25 

400 

400 

50.14 

50.14 

18.08 

20.14 

68.3 

68.3 

2.77 

2.49 

57.1 

118.0 

1.01 

1.04 

74 1 

0.1 

0.25 

400 

400 

50.14 

50.14 

21.84 

24.13 

68.3 

68.3 

2.30 

2.10 

50.3 

102.6 

1.22 

1.24 

Table 5 

OSC10 – The influences of tool´s geometry and cutting feed 

Cut material/ 

Tool´s type 

σo [daN/ 

mm 2 ] 

t 

[mm] 

s 

[mm/ 

rev] 

n 

[rev/ 

min] 

L 

[mm] 

La [mm] 

D 

[mm] 

Cd 

Fz [daN] 

mz 

OSC10/ Co 52 2 

0.1 

0.25 

500 

500 

57.9 

57.9 

25.6 

28.3 

78.2 

78.2 

2.26 

2.05 

59.4 

119.2 

1.13 

1.12 

OSC10/ C1 52 2 

0.1 

0.25 

500 

500 

57.0 

57.0 

22.5 

24.3 

77.0 

77.0 

2.53 

2.35 

65.3 

137.2 

1.09 

1.07 

OSC10/ C2 52 2 

0.1 

0.25 

500 

500 

57.0 

57.0 

23.9 

25.9 

77.0 

77.0 

2.38 

2.20 

61.2 

128.6 

1.05 

1.04 

OSC10/ C3 52 2 

0.1 

0.25 

500 

500 

57.0 

57.0 

27.6 

29.9 

77.0 

77.0 

2.06 

1.90 

57.8 

113.9 

1.39 

1.32 

OSC10/ C4 52 2 

0.1 

0.25 

500 

500 

57.9 

57.9 

30.0 

32.8 

78.2 

78.2 

1.93 

1.76 

54.4 

106.4 

1.41 

1.36


Table 6 

The influence of main cutting speed 

Cut material/ 

Tool´s type 

σo [daN/ 

mm 2 ] 

t 

[mm] 

s 

[mm/ 

rev] 

v 

[m/ 

min] 

Cd Fz [daN] 

mz 

85 2.62 60 1.43 

OL44/ Co 29 2 0.1 115 2.57 55 1.38 

165 2.50 52 1.37 

80 2-35 68 1.19 

OSC10/ Co 52 2 0.1 122.5 2.26 63 1.21 

164 2.20 61 1.24 

87 1.67 30 1.72 

MoCM14/ Co 74 1 0.1 120 1.52 27 2.08 

160 1.48 25 2.09 

80 1.90 65 1.21 

40C10/ C3 78 2 0.1 102 1.86 63 1.24 

Constant parameters: 

150 1.80 58 1.22 

K=70°; α=10°; γ=0°; λ=0°; wear hα=0; cutting in air 

For the most ductile steel type (OL44) and the most resistant one (40C10), 

the experimental results regarding the influences of the constructive angles, 

cutting feed and main cutting speed are illustrated in the diagrams in Figs. 1-4. 

OL44 

λ=0° 

s=0.1 mm/rev 

Fig.1a – The influence of angle γ for OL44 steel.


OL44 

γ=0° 

s=0.1 mm/rev 

Fig.1b – The influence of angle λ for OL44 steel. 

40C10 

λ=0° 

s=0.1 mm/rev 

Fig.2a – The influence of angle γ for 40C10 steel. 

40C10 

γ=0° 

s=0.1 mm/rev 

Fig.2b – The influence of angle λ for 40C10 steel.


Fig.3a – The influence of feed values for OL44 steel. 

OL44 

γ=0° 

λ=0° 

OL44 

γ=0°; λ=0° 

s=0.1 mm/rev 

Fig.3b – The influence of speed values for OL44 steel. 

40C10 

γ=0° 

λ=0° 

Fig.4a – The influence of feed values for 40C10 steel.


40C10 

γ=0°; λ=0° 

s=0.1 mm/rev 

Fig.4b – The influence of speed values for 40C10 steel 

From the diagrams in Figs.1 – 4 result that Fz and Cd are decreasing when 

the angles γ, λ and cutting speed values are growing. When the cutting feed is 

increasing, the force Fz grows and the coefficient Cd decrease. For OL44 steel, 

the Fz values are smaller and the Cd are larger than those resulted for 40C10 

steel. 

For OL44 steel, the mz exponent is relatively constant in the 1.2 – 1.4 range. 

For 40C10 steel, the mz values are variable in a larger domain, 0.8 – 1.22. For 

λ=0° and γ=0° the mz exponent may achieve even the value of 1.95 (Fig.2b). 

This is the reason why mz it is necessary to be reached for every material type. 

The chemical composition, the nature and the structural micro constituents´ 

distribution of the cut material determine the plastic deformation capability. 

Consequently, the cutting forces level, the values of Cd and the mz values 

through which Cd amplify the cutting forces values are indirectly influenced. 


1. The experimental researches regarding the values of cutting force Fz and 

coefficient Cd for the four steel types were stressing the known dependencies 

from angles γ and λ, cutting feed and main cutting speed. 

2. The level of influence of the constructive angles γ and λ, of cutting feed 

and speed takes different values depending on the material being cut. 

3. Experimental tests are necessary for each material, to determine the 

values for mz, mx, my exponents in order to use the models in Eqs. (1)-(4) for 

the force components Fz, Fx, Fy on a cutting tooth.


Received: ”Gheorghe Asachi” Technical University, 

Department of Machine Tools 

Iasi, Romania 

e-mail: cozminca@tcm.tuiasi.ro 


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 

the Force Components Fz, Fx, Fy when Cutting Metals with Single Tooth Tools, 

Buletinul Inst. Polit. Iași, editat de Univ. Tehnică “Gheorghe Asachi”, LV(LIX), 

Construcții de mașini, 1, 2009, pp.1-7. 

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 

between Forces Fz, Fx, Fy and the Plastic Strain Force in Single Tooth Cutting. 

Buletinul Inst. Polit. Iași, editat de Univ. Tehnică “Gheorghe Asachi”, LV(LIX), 

Construcții de mașini, 1, 2009, pp.8-14. 

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 

Cutting with Single Tooth Tools. Buletinul Inst. Polit. Iași, editat de Univ. 

Tehnică “Gheorghe Asachi”, LV(LIX), Construcții de mașini, 2, 2009, pp.1-7. 

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”, 

Iaşi, 1995. 

CERCETĂRI EXPERIMENTALE PENTRU VALIDAREA UNEI 

NOI METODE DE EVALUARE A FORȚELOR DE AȘCHIERE 

(Rezumat) 

În lucrare se prezintă modul de realizare a unor cercetări experimentale preliminare 

în vederea stabilirii mărimii forțelor de așchiere, la strunjirea a patru mărci de oțel, în 

scopul definitivării modelelelor de calcul pentru componentele Fz, Fx, Fy ale forței de 

așchiere la nivelul uni dinte așchietor. Noile modele de calcul au în vedere aria secțiunii 

transversale a așchiei nedeformate (txs), limita de curgere a materialului așchiat și 

coeficientul de deformare plastică a așchiilor. Coeficientul de deformare plastică a 

așchiilor include toate influențele parametrilor de lucru, precum și interdependențele 

dintre aceste influențe. Pe lîngă aceste elemente, modelele de calcul propuse cuprind și 

o componentă teoretică, în funcție de unghiurile constructive ale dintelui, γ, K, λ și 

coeficientul de frecare dintre așchii și suprafața de degajare a dintelui așchietor.


Publicat de 



Secţia 


FACE MILLING FORCES DEPENDING ON THE 

FORCES DEVELOPED ON A SINGLE-TOOTH 

BY 

ANA-MARIA MATEI and MARIUS NICOLAE MILEA 

Abstract. This paper presents some theoretical models for the evaluation of 

cutting forces components in face milling, developed on a single−tooth level. The 

cutting forces at face milling depend on the every variant of face milling process 

(teeth number which simultaneously cut) and the relative position of cutting teeth 

and the material being cut (cut - down milling and cut - up milling). Regarding to 

this, the determination of cutting force components at face milling FZ, FX, FY is 

based on the forces for a single tooth, the cutting tooth position beside the XYZ 

coordinates system of the tool and the number of teeth that simultaneously cut. 

Key words: cutting, face milling, forces. 


The cutting forces developed during the milling process are affected by a 

series of parameters, including cutting feed, the milling depth and width, tooth 

number, tooth geometry, the relative position of cutting teeth and the material 

being cut, and the hardness of the material being cut. Therefore, the most 

calculus models of face milling forces are including these parameters, and also 

some specific elements of the tools’ circular motion [1]. 

This paper presents new theoretical models for the evaluation of cutting 

force components at face milling, considering besides the parameters described 

above, also the specific elements for each variant of face milling process: 

complete (full), symmetrical incomplete or unsymmetrical incomplete, and the 

relative position of cutting teeth and the material being cut: cut-down milling 

and cut-up milling. 

The face milling forces are also depending on the number of teeth that 

simultaneously cut (odd or even number of teeth) [2].

76 Ana-Maria Matei and Marius-Nicolae Milea 

2. Theoretical Models of Cutting Force Components in Face Milling 

depending on the Cutting Forces developed on a Single-Tooth 

The face milling forces components acting on the tool FZ , FX si FY are 

depending on the cutting forces components developed on a single tooth level 

Fz , Fx si Fy , which are influenced by the cutting tooth geometry (γ, k, λ), the 

standard compression yield point of the material being cut, the cross - sectional 

area of the chip, the chips’ friction coefficient μ, and the chips´ contraction 

coefficient Cd [3,4]. 

Considering the specific elements of the symmetrical or unsymmetrical 

milling, there are several possibilities of modelling the cutting force 

components in face milling. 

2.1. Theoretical Models of Cutting Force Components 

in Symmetrical Face Milling 

Fig. 1 shows the geometrical models of complete and incomplete face 

milling. The values of specific elements for each milling variant result as 

follows: the contact angle Ψ = 180°, t = D, the number of teeth that 

simultaneously cut zs = z/2 at complete face milling and Ψ < 180°, t < D, 

z t 

= arcsin at incomplete face milling [2]. 

z s 

π 

D 

The theoretical models for the evaluation of cutting force components in 

face milling are developed in the light of two simplifying assumptions, as 

follows: 

a) The cutting force components on a tooth, respectively Fz – the main cutting 

force (the tangential force), Fx – the radial force and Fy – the axial force, take 

values depending on the cross-sectional area of undeformed chip; 

b) The tooth z1 is the one to which the tooth coordinates system xyz 

corresponds to the mill coordinates system (FZ = Fz, FX = Fx, FY = Fy), and the 

other teeth that simultaneously cut are placed equidistant (φ = 2π/z). 

The values of cutting force components in symmetrical face milling depend 

on zs and Fz, Fx , Fy , as follows: 

a) For symmetrical milling with odd zs the Eq. (1) results. 

(1) 

F 

F 

Z 

X 

( zs 

/ 2) 

−1 

( zs 

/ 2) 

−1 

2π 

⎡ 2π 

⎤ 

= Fz 

+ 2 ∑ Fz 

cos( zi 

) = Fz 

⎢1 

+ 2 ∑ Fz 

cos( zi 

) ⎥ , 

1 z ⎣ 1 z ⎦ 

( zs 

/ 2) 

−1 

( zs 

/ 2) 

−1 

2π 

⎡ 2π 

⎤ 

= Fx 

+ 2 ∑ Fx 

cos( zi 

) = Fx 

⎢1 

+ 2 ∑ Fx 

cos( zi 

) ⎥ , 

1 z ⎣ 1 z ⎦ 

F = F ⋅ z . 

Y 

y 

s

(2) 


Fig.1 – Forces model in symmetrical face milling. 

b) For symmetrical milling with even zs the Eq. (2) results. 

F 

F 

Z 

X 

= 2 

= 2 

z / 2 

s 

∑ 

1 

z / 2 

s 

∑ 

1 

2π 

Fz 

cos( zi 

) = 2Fz 

z 

2π 

Fx 

cos( zi 

) = 2Fx 

z 

F = F ⋅ z . 

Y 

y 

s 

z / 2 

s 

∑ 

1 

z / 2 

s 

∑ 

1 

2π 

cos( zi 

) , 

z 

2π 

cos( zi 

) , 

z 

From Eqs. (1) and (2) result that in symmetrical face milling the forces FZ, 

FX and FY depend on the components Fz, Fx and Fy developed on a single-tooth 

level, the number of teeth that simultaneously cut (zs) and the cutting tooth 

position beside the XYZ coordinates system of the tool. The axial component FY 

doesn’t depend on the relative position of the cutting tooth. 

2.2. Theoretical Models of Cutting Force Components in 

Unsymmetrical Face Milling 

In this case, the values of cutting force components FZ, FX, FY are influenced 

by the relative position of the tool and the material being cut, resulting different 

equations for cut-up (conventional) milling and cut-down (climb) milling. 

Fig. 2 shows the geometrical models of unsymmetrical cut-up milling and its 

three possible variants: 

a) Unsymmetrical face milling with t' = D/2, Ψ = 90°, and z = z / 4 ; 

s 

1


b) Unsymmetrical face milling with t" < D/2, Ψ < 90° and 

z 2t 

zs = arccos( 1− 

) ; 

2 2π 

D 

c) Unsymmetrical face milling with t > D/2 , where t = t' + t"' and t"' = t -D/2, 

z 2t 

Ψ > 90° and z s = z , where 

3 s + z 1 s z = arcsin( −1) 

4 

s 

. 

4 2π 

D 

Fig. 2 – Forces model in unsymmetrical cut-up face milling. 

According to the geometrical model of face milling process shown in Fig. 2, 

the cutting force components FZ, FX and FY result using the Eqs.(3). 

a) zs = z / 4 

(3) 

1 

F 

F 

Z 

X 

= 

= 

z / 4 

∑ 

0 

z / 4 

∑ 

z 2t 

b) zs = arccos( 1− 

) 

2 2π 

D 

(4) 

F 

F 

Z 

X 

= 

= 

z 

0 

s2 

∑ 

z 

0 

s2 

∑ 

0 

2π 

Fz 

sin( zi) 

− 

z 

2π 

Fx 

sin( zi) 

+ 

z 

Y 

y 

z / 4 

∑ 

0 

z / 4 

∑ 

0 

F = F ⋅ z . 

2π 

Fz 

sin( zi) 

− 

z 

2π 

Fx 

sin( zi) 

+ 

z 

z 

s 

1 

s2 

∑ 

z 

0 

s2 

∑ 

0 

F = F ⋅ z . 

Y 

y 

s 

2 

2π 

Fx 

cos( zi) 

, 

z 

2π 

Fz 

cos( zi) 

, 

z 

2π 

Fx 

cos( zi) 

, 

z 

2π 

Fz 

cos( zi) 

, 

z


c) For the unsymmetrical milling with t > D/2, we suppose that for a 

number of teeth that simultaneously cut z = z + z , the values of cutting 

force components will be obtained as the sum of the values of cutting force 

components corresponding to a tooth z , Eqs. (3), and a tooth z , Eqs. (5). 

(5) 

F 

F 

Z 

X 

= 

= 

z 

−1 

s4 

∑ 

z 

1 

s4 

∑ 

s1 

2π 

Fz 

cos( zi) 

+ 

z 

−1 

1 

2π 

Fx 

cos( zi) 

− 

z 

Y 

y 

s 

3 

z 

s 

s4 

∑ 

z 

1 

−1 

1 

−1 

s4 

∑ 

s 

1 

F = F ⋅ z . 

4 

s 

4 

s4 

2π 

Fx 

sin( zi 

) , 

z 

2π 

Fz 

sin( zi 

) , 

z 

The values of cutting force components corresponding to the total number 

of teeth that simultaneously cut ( z ) result using the Eqs. (6). 

(6) 

F 

F 

Z 

X 

= 

= 

z / 4 

∑ 

0 

z / 4 

∑ 

0 

2π 

Fz 

sin( zi 

) + 

z 

2π 

Fx 

sin( zi 

) + 

z 

s 

3 

z 

−1 

s4 

∑ 

z 

1 

+ 

−1 

s4 

∑ 

2π 

Fz 

cos( zi 

) − 

z 

z 

−1 

s4 

∑ 

1 

x 

1 

z −1 

z / 4 

∑ 

2π 

Fx 

sin( zi 

) 

z 

2π 

F cos( zi 

) + 

z 

s4 

2π 

− ∑ Fz 

sin( zi 

) 

1 z 

F = F ⋅ z . 

Y 

y 

s 

3 

0 

z / 4 

∑ 

0 

2π 

Fx 

cos( zi 

) 

z 

, 

2π 

Fz 

cos( zi 

) 

z 

, 

It can be observed that in case of cut-up face milling with cu t' = D/2, and t'' 

< D/2, the same relations for the cutting force components FZ, FX, FY are 

obtained, distinguishing only by the number of teeth that simultaneously cut. 

Fig. 3 is illustrating the geometrical models of unsymmetrical cut-down 

milling and the same three possible variants:


Fig. 3 – Forces model in unsymmetrical cut-down face milling. 

For all variants of unsymmetrical cut-down face milling, the cutting force 

components FZ, FX, FY are given by the Eqs.(7) – (8). 

a) zs = z / 4 

1 

z / 4 

z / 4 2π 

2π 

FZ 

= ∑ Fz 

cos( zi 

) + ∑ Fx 

sin( zi) 

, 

z 

z 

(7) 

F 

X 

= 

1 

z / 4 

∑ 

z 2t 

b) zs = arccos( 1− 

) 

2 2π 

D 

F 

Z 

= 

z 

1 

s2 

∑ 

0 

2π 

Fx 

cos( zi) 

− 

z 

Y 

y 

1 

z / 4 

∑ 

s 

1 

F = F ⋅ z . 

2π 

Fz 

cos( zi 

) + 

z 

z 

1 

s2 

∑ 

0 

2π 

Fz 

sin( zi) 

, 

z 

2π 

Fx 

sin( zi 

) , 

z 

z 

(8) 

s 

z 

2 

s2 

2π 

2π 

FX 

= ∑ Fx 

cos( zi) 

− ∑ Fz 

sin( zi) 

, 

0 z 

0 z 

FY = Fy 

⋅ zs 

. 

2 

c) Similarly to cut-up milling, for the unsymmetrical cut-down milling 

with t > D/2, we use the same method for determining the values of components 

FZ , FX , FY . First, will be determined the values of cutting force components 

corresponding to a number of teeth that simultaneously cut z , Eqs. (9). 

s4

(9) 

F 

F 


Z 

X 

= 

= 

z 

−1 

s4 

∑ 

z 

0 

−1 

s4 

∑ 

0 

2π 

Fz 

sin( zi) 

− 

z 

2π 

Fx 

sin( zi 

) + 

z 

Y 

y 

z 

−1 

s4 

∑ 

z 

F = F ⋅ z 

0 

−1 

s4 

∑ 

s 

0 

4 

2π 

Fx 

cos( zi) 

z 

2π 

Fz 

cos( zi) 

z 

The values of cutting force components corresponding to a total number of 

teeth that simultaneously cut ( z ) result using the Eqs.(10). 

(10) 

F 

F 

Z 

X 

= 

= 

z / 4 

∑ 

1 

z / 4 

∑ 

1 

s3 

2π 

Fz 

cos( zi 

) + 

z 

− 

2π 

Fx 

cos( zi 

) + 

z 

z 

−1 

s4 

∑ 

z 

0 

z −1 

s4 

∑ 

z 

0 

−1 

s4 

∑ 

0 

z −1 

2π 

F sin( zi 

) + 

z 

2π 

Fx 

cos( zi 

) 

z 

2π 

Fx 

sin( zi 

) − 

z 

s4 

2π 

+ ∑ Fz 

cos( zi 

) 

0 z 

F = F ⋅ z . 

Y 

y 

s3 

z / 4 

∑ 

1 

z / 4 

∑ 

1 

2π 

Fx 

sin( zi 

) 

z 

, 

2π 

Fz 

sin( zi 

) 

z 

, 

Similarly to cut-up milling, it can be observed that in case of cut-down face 

milling with cu t' = D/2, and t'' < D/2, the same relations for the cutting force 

components FZ, FX , FY are obtained, distinguishing only by the number of teeth 

that simultaneously cut. 


1. The theoretical models presented in the paper are concentrating the main 

elements that influence the level of cutting force at face milling. These specific 

elements refer to the tooth geometry (γ, K, λ), standard compression yield point 

of the material being cut, the cross-sectional area of the theoretical chip, the 

milling width, the contraction chips coefficient, the constructive geometry of the 

tooth (γ, K, λ), the number of teeth that simultaneously cut, the cutter diameter 

and teeth number.


2. Using those models to evaluate the cutting force provides extra accuracy 

because we are considering the forces on a tooth, the number of teeth that 

simultaneously cut, and also the relative position of the tool and material being 

cut for all possible variants of the face milling process. 

Received: January 29, 2010 “Gheorghe Asachi” Technical University of Iaşi 


e-mail: anca_reea@yahoo.com 


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, 

Iaşi, 1995. 

2. C o z m î n c ă M. et al., About the cutting forces at face milling, Buletinul 

Institutului Politehnic din Iaşi, 2, Construcţii de Maşini, (2009). 

3. C o z m î n c ă M. et al., A new model for estimating the force components Fz, Fx and 

Fy when cutting metals with single tooth tools, Buletinul Institutului Politehnic 

din Iaşi, 1, Construcţii de Maşini, (2009). 

4. C o z m î n c ă M. et al., Variation of the ratios between forces Fz, Fx and Fy and the 

plastic strain force in single tooth cutting , Buletinul Institutului Politehnic din 

Iaşi, 1, Construcţii de Maşini, (2009). 

FORȚELE DE AȘCHIERE LA FREZAREA FRONTALĂ ÎN FUNCȚIE DE 

FORȚELE DEZVOLTATE LA NIVELUL UNUI DINTE 

(Rezumat) 

În lucrare sunt prezentate modele teoretice de evaluare a componentelor forţei 

de aşchiere la frezarea frontală în funcţie de forţele dezvoltate la nivelul unui dinte. 

Forţele la frezarea frontală depind atât de varianta de frezare, poziţia relativă a dinţilor 

aşchietori faţă de semifabricat. poziţia dintelui aşchietor în raport cu sistemul de 

coordonate XYZ al sculei cît şi numărul de dinţi care aşchiază simultan.


Publicat de 



Secţia 


COMPARATIVE ASPECTS REGARDING THE NESTING 

FOR BLANKING-PUNCHING OPERATION ON 

CLASSICAL PRESSES AND NUMERICAL 

COMANDED PRESSING CENTERS 

BY 

MARIUS - IONUŢ RÎPANU 1 GHEORGHE NAGÎŢ 1 , 

IOLANDA - ELENA MANOLE 1 and ANDREI WEINGOLD 2 

Abstract. The nesting operation represents an important chapter in the design 

of stamping technology of parts or blanks in the area of cold plastic forming 

processes, with a large applicability in industry. The type of nesting has a very 

important role in determining the material consumption, the number of blanks cut 

concomitant, in determining the type and the construction of the stamping device 

or of the punch. The goal of this paper is to elaborate a comparative study 

between nesting for the classical presses stamping and for the CNC pressing 

centres. In order to obtain this goal, in this paper a case study, along with the 

results obtained and the associated conclusions are presented. 

Key words: punching, blanking, stamping press, CNC, nesting. 


The nesting of the blanks is a problem of special importance in the area of 

cold plastic forming processes. By nesting the blanks we understand the optimal 

cutting of metal sheets into individual blanks or strips. We understand as well 

the judiciously positioning or placing on metal sheets or strips the plan 

components as structured technological forms used to detach the pieces in order 

to get a minimum amount of waste scrap and an optimum coefficient of material 

utilization [1].

84 Marius Ionuţ Rîpanu et al. 

The objective of this paper is to elaborate a comparative study upon the 

nesting for the classical presses and nesting for the numerical commanded 

centres in order to emphasize the advantages and the disadvantages of both 

operations and, not the last, based on the effectuated researches, to obtain high 

quality and high precision parts at the smallest cost possible. 

2. Particularities of the Nesting Operation 

The cold plastic forming researches and especially the nesting for numerical 

controlled pressing centres researches made by J a c k s o n and M i t t a l [2] are 

based on blanking-punching using a very well planned algorithm, method that 

leads to the automatical generation of the CNC program. R a g g e n b a s s and R 

e i s s n e r [2] have researched the connection between pressing and laser, with 

applicability on the processing centres with CNC. In Twente University was 

studied and developed the processing of the metal sheets domain through 

planning and managing the factors that influence the machinability of metal 

sheets using NC [2], [3]. Some of the most important CAD/CAM programs that 

are used to perform this technological process are: RADAN, WICAM, Nesting 

Software, EditCNC, CNCezPRO TM , and some of the pressing centre with NC 

are TRUMPF, AMADA ARIES, Finn-Power, Mazak, Bystronic, etc. 

The nesting of the blanks for classical presses takes into account certain 

coefficients that depend on the nature of the material, the placing of the part 

and the nesting style of it (with waste, without waste, with little waste). 

During the nesting of the parts for the NC pressing centres it must be taken 

into consideration the type of the material of the metal sheet. The software used 

for modelling, simulating and generation of the numerical control scheme will 

consider all the factors and parameters that may influence the process 

productivity. 

3. Comparative Study 

For this study was chosen a pinchbeck part made of CuZn15, SR EN 

1653:2003, having the configuration from Fig. 1 and the thickness of 1 mm. 

The processing of this part was followed both for the classical presses and the 

CNC pressing centres. For the classical nesting, three variants of placing were 

considered, as in Figure 2. For those variants were calculated the coefficients 

with values of 53.77%, 54.16 % and 30.75%. 

In case of using the NC cutting centres, CAD/CAM software were used in 

order to realise and integrate the designing, modelling, simulating, planning and 

development functions of the CNC program for processing the metal sheet, 

choosing the optimum cutting variant with minimum waste.


Fig. 1 – The processed part. 

a) b) 

c) 

Fig. 2 – Nesting variants. 

In the present case, for cutting the 6.6 m 2 of the metal sheet, it was chosen 

the RADAN software of designing, modeling and simulating with the AMADA 

ARIES 245 pressing center. Following there are presented some sequences of 

the program that calculated and cut the part from Fig. 1.


Fig. 3 – The choice of the pressing centre for the processing. 

. 

Fig.4 – Standard sizes of metal sheets with their coefficients 

of utilization and the number of cut parts per sheet.


Fig.5 – Position of the parts on 850 x 300 mm sheet metal. 

Fig.6 – The cutting path of the parts.


There are presented a part of the steps of the processing program. The 

generated program processes a number of 98 pieces from a 3000 x 1500 mm 

sheet metal. At the end, as it is noticeable, the program shows exactly the 

number of program blocks, the number of tool changes, the programmed stops 

and the run time. 

Line43: B2 

Line44: N43X – 2.5Y119.5T12 

Line45: N44Y40.5 

Line46: N45X202.5 

Line47: N46Y119.5 

Line48: N47X407.5 

Line49: N48Y40.5 

Line50: N49X612.5 

Line51: N50Y119.5 

Line52: N51X817.5 

Line53 N52Y40.5 

Line54: N53G98X17.5Y59.98I205.J0.P3K0 

Line56: N55G98X201.5Y162.52T1 called from Line 61 

Line57: N56G72X – 2.Y40.02 called from Line 61 

Line58: N57G66I204.J0.P – 85.Q – 5. called from Line 61 

Line59: N58M00 called from Line 61 

Line64: N63G98X17.5Y100.I0.J0.P0.K0 

Line65: N64G50 

Amada Aries 245(RA1) 

Verification Analysis 

Total number of program blocks 65 

Total number of program characters 25 544 

Number of sub programs 0 

Number of tool changes 3 

Number of programmed stops 4 

Number of repositions 0 

Run time 1813 secs = 30.21 min. 

In the figures below are presented the steps that must be covered in order to 

cut the part. Therefore, in Fig. 3, there are presented the processing centres that 

use the RADAN software. From these ones, it was chosen for the part to be cut 

using the AMADA ARIES 245, using as well the specified material with 1 mm 

thickness. 

In figure 4, based on the part geometry and dimensions, are presented four 

types of standard metal sheets that may be processed by the machine. There are 

also presented the usage coefficients of the material and the exact number of


parts that may result from the process and the number of metal sheets that are 

needed to cover the approximately 6.6 m 2 . 

It is noticeable that the coefficient of utilization of the material varies based 

on the dimensions of the sheet and the number of parts increases proportionally 

with the sheet dimensions. In Fig. 5 is presented the placing of the part on the 

metal sheet, the anchorage of the metal sheet on the pressing centre and the 

distances between the pieces that are about to be cut. In the last figure, Fig. 6 is 

presented the technological flow of cutting and the modality in which the tool 

approaches and clips step by step the part. 


1. The CNC pressing centres are much more efficient than the classical 

pressing methods, the evidence of this fact being the calculation made for the 

two types of processing of blanks and the coefficient of utilization of the 

material. 

2. For the CNC pressing centres the coefficient of utilization varies 

depending on the dimensions of the metal sheets while, for the classical 

pressing, if it would have been used a longer metal strip the usage coefficient 

would have been the same. I this case it is more appropriate to use a CNC 

pressing centre. 

3. The number of parts obtained from CNC pressing centres is much 

higher than the number of parts obtained with the classical presses. 

4. The CNC pressing centres are characterised by the easiness of 

calculation at changing the setting of a parameter that has direct influence over 

the processing of the blanks, due to the presence of the CAD/CAM system 

presented. 

5. An important advantage of the CNC pressing centres, unlike the 

classical presses, at the processing of simple configuration parts (like the one 

presented above) is the geometry and the fabrication costs of the tools used in 

production. 

Received: March, 22, 2010 1 “Gheorghe Asachi” Technical University, 

Department of Machine Manufacturing Technology, 

e-mail: ripanumariusionut@yahoo.com 

2 S.C. BMTech S.R.L of Vladeni, 

Iasi – Romania, 

e-mail: andrei.weingold@bmtech.ro



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 

Tehnica, Stiintifica si Didactica CERMI, Iasi, 2003, pp 138-163. 

2. P a n M., R a o Y., An integrated knowledge based system for sheet metal cutting– 

punching combination processing, Knowledge-Based Systems 22, 2009, 368– 

375. 

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 

manufacturing information system for mass sheet metal cutting, International 

Journal Advanced Manufacturing Technology 33, 2007, 436 – 448. 

4. E n d o J., O h b a S.,A n z a i T., Virtual manufacturing for sheet metal processing, 

Journal of Materials Processing Technology, 60 (1996, 191–196. 

5. C h a t u r v e d i S., A l l a d a V., Integrated Manufacturing System for Precision 

Press Tooling, International Journal Advanced Manufacturing Technology, 15, 

(1999) 356–365. 

ASPECTE COMPARATIVE PRIVIND CROIREA 

LA OPERAŢIA DE DECUPARE-PERFORARE PE PRESE CLASICE 

ŞI CENTRE DE PRESARE PREVĂZUTE CU COMANDĂ NUMERICĂ 

(Rezumat) 

Operaţia de croire reprezintă un capitol important în proiectarea tehnologiei de 

ştantare a pieselor sau semifabricatelor din domeniul deformării plastice la rece având o 

aplicabilitate foarte mare în industrie. Croirea joacă un rol important în determinarea 

consumului de material, numărului de semifabricate ştantate concomitent, în 

determinarea tipului şi construcţiei ştanţei sau matriţei. Scopul acestei lucrări este de a 

elabora un studiu comparativ între croirea pieselor pe presele clasice şi croirea acestora 

pe centrele de comandă prevăzute cu comandă numerică, de a scoate în evidenţa 

avantajele şi dezavantajele acestei operaţii şi nu în ultimul rând, pe baza cercetărilor 

făcute, să putem obţine piese de o calitate superiaoră şi precizie ridicată la un cost cat 

mai redus. In acest sens, in lucrare este prezentat un studiu de caz, impreuna cu 

rezultatele obtinute si concluziile aferente.


Publicat de 



Secţia 


THEORETICAL AND EXPERIMENTAL CONSIDERATIONS 

ON DETERMINING THE EFFECT OF DIVERGENCE 

BY 

IUSTINA ELENA ROTMAN, PETRU DUSA 

and RADU ADRIAN BACIU LUPASCU 

Abstract. This paper presents the experimental studies conducted on the basis 

of ultrasound examination of the parts with different geometries and structures. 

The main objective of this work is to eliminate disturbances in order to perform an 

examination and a correct interpretation in case of the A-scan representation. The 

experimental results and conclusions drawn follow the analysis of the factors 

involved in the process, which negatively influences the interpretation of signals 

from the investigation display device. 

Key words: transducer, device, back wall, divergence, discontinuity. 


The Ultrasonic Testing Method of materials is practiced by more than 50 

years [1], [3].Since the first examination in detecting defects using ultrasonic 

oscillations of different materials, it became a classic test method based on 

measurements. That method takes in consideration all those factors that have a 

certain impact [4]. It is expected that in our days the ultrasonic testing to give 

results with minimum tolerance, if that method is sustained by a large variety of 

instruments and by one advance technical execution [4]. 

The structure transformations of the material underlayers by manufacturing 

process lead to changes for the workpiece general properties [2]. After the type 

of defects are detecting, that will be compared with contractual standard. The 

contractual standard would decide if the detected defect could be accepted, not 

accepted, or remediable [2]. In many cases some of defects could be accepted if 

their values respect standard limits. Once detected the defects types, they must

92 Iustina Elena Rotman et al 

be confronted to the standards or to the contractual technical documentations to 

establish if they are admissible, inadmissible or remediable [2]. In many cases 

some influences can be neglected without to exceeding the measure limits 

prescribed in the technical documentation [2], [4]. This presume the exactly 

knowledge of the influence factors and of the results interpretation mode. 

2. Method of Investigation 

2.1. The Ultrasonic Testing Method 

Performing the non-destructive control with ultrasound effectuation consist 

principally in transmitting ultrasound signals (waves), produced by a waves 

generator, through the examined pieces [2], [3]. The ultrasounds signals are 

reflected by any surface and by any defect from the interior of the piece [1]-[5] 

that was exposed to testing, see fig.1. 

Fig. 1 – The ultrasonic testing method [5] Fig. 2 – The geometry examination [2] 

The ultrasonic equipment used in this research is presented in the table 

below: 

USM 

35X 

Table 1 

The necessary instruments of control US 

Device Transducer Piece Coupled 

USLT 

2000 

Reference 

model: B2S 

Aluminum bare 

Ø 90 x 318 mm 

Axle OL47 

Ø 900 x 1500mm 

2.1.1. The analysis of the investigation. Often in industrial practice there 

are problems of interpreting the image obtained on the screen investigation 

device. This paper presents an experimental study over the determination of the 

divergence marginal effect and the elimination of disturbances in order to obtain 

a correct examination and interpretation. Its objective is to highlight the journey 

Oil


of ultrasonic waves from transducer through the examined piece, the reflection 

modalities of the waves and the way displaying on the screen of the device. 

When the diameter of transducer is comparable to the width of the piece, due to 

the divergence and partial reflection of the beam from the lateral surfaces of the 

piece, along the back wall echo and additional signal is formed, more or less 

distinct, named divergence marginal echo [2]. The necessary condition for 

eliminated this effect is: 

(1) 

B−Dp ⎡ ⎛ λ ⎞⎤ 

≥ tg ⎢arcsin ⎜1, 22 ⎟⎥, 

2l 

⎣ ⎝ Dp⎠⎦ 

where: B – the diameter of the piece; Dp 

– the diameter of the transducer; l – 

the length of the piece; λ – the divergence beam of fascicle. Also the back wall 

echo can be confused with a fault echo. These confusions occur due to the piece 

geometry, to the transducer diameter which is comparable to the diameter of the 

piece and to the differences in length. To eliminate these interpretation errors of 

and disturbances we will present several different cases. 

3. Experimental Results 

The experimental studies were performed with the devices type 

Krautkramer (USM 35 X and USLT 2000) in order to check possible errors in 

measurement. The pieces subjected to ultrasonic testing are: an aluminum bar 

with dimensions Ø 90 x 318 mm and one axle in OL47, forged and heat treated. 

By conducting the experiments we achieved the following results. 

3.1. First case. In this case the back wall echo can be confused with an 

echo of a defect (See fig. 3). These echoes can be called divergent effects. 

Fig. 3 – Ultrasonic testing display. Fig. 4 – The divergent effect.


This measurement error is due to amplification (61 dB) that distorts the back 

wall echo base and to the transducer which is positioned near the edge of the 

sample. 

3.2. Second case. It may be noted that in this case the effect of divergence 

and misinterpretation can be eliminated by positioning the transducer on the 

axial direction of the sample. In this manner, the repetition rate of the flaw 

(PRF) does not influence the accuracy of the signal and the interpretation is 

objective. Bottom echo is very clear. (See fig. 5). 

Fig. 5 – Ultrasonic testing display. Fig. 6 – The ultrasonic beam shape . 

3.3. Third case. Analyzing the measurements made on both samples we can 

conclude that the divergence effect can be eliminated by reducing the repetition 

rate of the flaw. 

a) b) 

Fig. 7 –Ultrasonic testing display: 

a) – The parasite echo; b) – The additional echo disappears 

As we can see in Fig. 7a), parasites echoes are present and at a PRF9 (See 

Fig.7b)) these additional effects disappear. In this kind of situations it is 

recommended to get more measurements at a minimum level of PRF.


These cases study presented show how to determine the marginal effect of 

divergence and to eliminate stray echoes in order to obtain effective results. 


Behind the effectuated researches one can evidence the following 

conclusions: 

1. Considering the applicability of this method and its expansion 

possibilities, it is necessary to analyze the factors that influence ultrasonic 

nondestructive testing. 

2. The position of transducer directly influences the results of testing. It is 

preferable that measurements are effectuated in different points of the 

investigated surface. 

3. When the transducer is near the edge of the piece, the back wall echo 

becomes smaller and it can be interpreted as a defect. This leads to 

misinterpretation. 

4. To eliminate stray echoes that appear on the display device is 

necessary for the repetition rate of the flaw (PRF) is minimized 

Acknowledgements. This paper was realised with the support of BRAIN “Doctoral 

scholarships as an investment in intelligence” project, financed by the European Social 

Found and Romanian Government. 

Received:March 3, 2010 “Gheorghe Asachi”Technical University, 

Department of Machine Manufacturing Technology 

Iasi, Romania 

e-mail: yustinikrotman@yahoo.com 


1. B e r k e M., Nondestructive Material Testing with Ultrasonics, Introduction to the 

Basic Principles, The e-Journal of Nondestructive Testing & Ultrasonics, 5, 

2000. 

2. V o i c u I. S a f t a, Defectoscopie nedistructiva industriala, Ed. Sudura, Timisoara, 

2001. 

3. K r a u t k r a m e r J. & H., Ultrasonic testing of materials, Springer Verlag Berlin 

Heidelberg New York, 1983. 

4. P e t c u l e s c u P., Ultrasounds Fundamental. Aplications, Ed. Univ. Ovidius, 

Constanta, 2002.


5. * * * Ultrasound and Ultrasonic Testing. NDT Resource Center. Available from: 

http://www.ndt-ed.org/EducationResources/HighSchool/Sound/ultrasound.htm, 

Accessed: 07/02/2010. 

CONSIDERATII TEORETICE SI EXPERIMENTALE PRIVIND 

DETERMINAREA EFECTULUI DE DIVERGENTA 

(Rezumat) 

Aceasta lucrare prezinta un studiu experimental ce urmareste evidentierea 

parcursului ultrasunetelor de la traductor catre piesa de examinat, modul de reflexie al 

acestora si de afisare pe ecranul instrumentului de investigat cu scopul de a obtine 

rezultate eficiente. Obiectivul principal al acestei lucrari este determinarea efectului 

marginal de divergenta si eliminarea factorilor perturbatori ce pot influenta negativ 

rezultatele testarii cu ultrasunete. Studiile de caz prezentate si concluziile finale 

indeplinesc obiectivele propuse si pot contribui la dezvoltarea bazei de cunostinte.


Publicat de 



Secţia 


COMPUTERIZED MEASUREMENT SYSTEM 

BY 

CĂTĂLIN UNGUREANU 1 , RADU IBĂNESCU 2 

and IRINA COZMÎNCĂ 1 

Abstract. All technological machining systems parts made real surfaces 

certain deviations from the theoretical and nominal areas. The role of quality 

control is to decide whether or not these deviations fall within the allowable 

accuracy classes that imposed parts. The paper presents a computerized system for 

determining and recording deviations in shape of the real parts 

Key words: measurements, inductive transducers, data acquisition. 


Accuracy is the degree of form areas of agreement between actual shape of 

the areas resulting from the processing and documentation of performance 

prescribed form. The causes of form deviations are multiple of which can be 

mentioned: 

• elastic deformation of pieces during processing; 

• tighten the fixing the wrong parts in devices; 

• failure of machine tools guidance; 

• wear or deformation of system machine-tool - cutting tools elements. 

The effects of shape irregularities are of the least desired, with negative 

implications for economic and technical performance of pieces processed: 

• edit the sliding friction between the parts of joints, 

• reduce static and dynamic tightness of parts and joints, 

• edit character fits, 

• they increase wear and may lead to locking parts in contact 

• decrease the durability of parts in contact.

98 Cătălin Ungureanu et al. 

Deviations in shape are defined as deviations of actual surface shape of the 

surface adjacent to, or deviation from the actual profile shape as the adjacent 

profile. Profile is adjacent to the same profile as the profile geometry, the outer 

tangential to the actual profile, and placed so that the distance between it and 

the actual profile to be minimum. Maximum permissible deviation form as is 

tolerance. 

Traditional means of measuring the deviations in shape are instrumentality 

mechanical comparators, scale value of 0.01 mm or 0.005 mm. They have the 

advantages of simplicity and low cost, but recording and graphic representation 

of measurement data for a production series is difficult. These disadvantages are 

removed if for control is used a measuring system equipped with a displacement 

transducer, data acquisition card and computer [1], [2]. 

2. Experimentals 

Experimental facility is shown in Figure 1. 

Fig.1 – Experimental facility.


Transducer was chosen and the whole system was built base on 

recommendations from the literature [3] – [5]. 

Part of the measure is placed on two supports adjustable so that it can be 

defined by two points equal height verification. This determines a straight line, 

to which will determine the actual right - edge play. For measurement: using 

an inductive transducer type Hottinger Baldwin Messtechnik WA T-10mm, 

supplied with 20 V DC voltage from a source Hameg HM 8040-2. Output is 

taken from a data acquisition card National Instruments USB 6009. 

Measurement chain is completed with a PC. No other mode of signal 

conditioning is required. Virtual instrument was developed programming 

environment LabVIEW 8.2 Student Edition. 

The block diagram of the application is presented in Fig. 2 [6] and involves 

the DAQ Assistant Express VI, the Waveform Chart for displaying the 

acquired signal and also the Build Table and the Write to Measurement File 

blocks for writing the acquired signal values in a table and in a file for further 

analysis. 

Fig. 2 – The block diagram of the application [6]. 

The DAQ Assistant may be founded in the Express Functions palette of 

the block diagram and it is needed to create and configure a task that reads a 

voltage level in a specified range from a DAQ device. After the DAQ Assistant 

is placed on the block diagram, a dialog box window is launched in order to 

configure the data acquisition system. The dialog box displays a list of channels 

of the installed DAQ device and from this list the physical channel to which the 

instrument connects the signal must be selected. The others configurations are 

related to terminal configuration (differential or single ended), acquisition mode 

(continuous or not), signal input range and rate of acquisition. The Write to 

Measurement File Express VI is used for storing the information about the data


VI generates in a specific file (*.lvm) in the default LabVIEW Data folder 

installed by LabVIEW in the default file folder of the operating system. 

Experimental data were represented graphic with a Matlab application. 

Calibration curve was presented in [6]. Diagram thus obtained is shown in 

Fig.3. 

d [mm] 

0.02 

0.01 

0 

-0.01 

-0.02 

0 50 100 150 

l [mm] 

Fig. 3 – Experimental diagram. 

In these diagrams “l” represents the length of the measured piece in 

mm and “d” the form deviation also in mm. 

Trough numerous measurement data recorded it is possible that the 

chart should be influenced by measurement noise. To remove these 

noises, was made a filtration of data with a Matlab application using 

smooth function [7]. The default syntax of smooth function is 

yf = smooth (y,span), 

where y is a column vector containing the acquired data and yf is the 

result column vector after a moving average filter is applied. The smooth 

function implements a lowpass filter with filter coefficients equal to the 

reciprocal of the span. The span must be odd but the default span for the


moving average is 5 and in this case the first few elements of yf are 

calculated as follows: 

yf(1) = y(1) 

yf(2) = (y(1) + y(2) + y(3))/3 

yf(3) = (y(1) + y(2) + y(3) + y(4) + y(5))/5 

yf(4) = (y(2) + y(3) + y(4) + y(5) + y(6))/5 and so on. 

The diagram of the whole acquired data filtered for a span equal with 

3 is shown in Fig. 4. 

d [mm] 

0.02 

0.01 

0 

-0.01 

-0.02 

0 50 100 150 

l [mm] 

Fig. 4 – The filtered values for a span equal with 3. 

It is noted that the chart is much more precise and accurate 

assessment can be made on the actual shape of the part to be measured. 

The filtration effect is observed more clearly if the graph is made for 

filtered and the unfiltered data only for first portion of the controlled 

piece. In Fig. 5 is represented by the dotted line unfiltered data and 

filtered data with continuous line for a span equal with 3 and in Fig.6 for 

a span equal with 5.


d [mm] 

d [mm] 

0.02 

0.01 

0 

-0.01 

-0.02 

0 10 20 30 40 50 

l [mm] 

0.02 

0.01 

-0.01 

Fig. 5 – The filtered and unfiltered values for a span equal with 3. 

0 

-0.02 

0 10 20 30 40 50 

l [mm] 

Fig. 6 – The filtered and unfiltered values for a span equal with 5.



1. The computerized system for measuring deviations of form 

enables the rapid and effective control of the geometric quality of parts in 

machine building. 

2. It can measure the absolute value of deviation and may draw a 

diagram of the actual shape of the part to control. 

3. However, because of noise measurement is required filtering 

acquired data. You can use low-pass filter easily facilitated by MatLab 

software package. It remains to be studied in future what is the best level 

of filtering. 

Received: February 20, 2010 


1 Technical University “Gh. Asachi, 


Iaşi, Romania, 

e-mail: cungurea@yahoo.com 

2 Department of Theoretical Mechanics 

e-mail: ribanesc@yahoo.com 

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 

inductive proximity sensor for industrial applications, Sensors and Actuators A: 

Physical, 110, 1-3 (2004) pp. 93-97. 

2. M a r č i č M., A new inductive displacement transducer, Sensors and Actuators A: 

Physical, 70, 3, 30 (1998), pp. 223-237. 

3. P a r k J., M c k a y S., Practical data acquisition for instrumentation and control 

systems, Newnes, Elsevier 2003. 

4. R i p k a P., T i p e k A., Modern Sensors Handbook, ISTE Ltd, London, 2007 

5. S i n c l a i r I.R., Sensors and transducers, Newnes, Butterwork Heinemann, 

Oxford, 2001. 

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 

Linear Position Measurement, Buletinul Institutului Politehnic din Iasi, publicat 

de Universitatea Tehnica “Gh. Asachi” Iasi, LV(LIX), 4, 2009, pag.54-62. 

7. www.mathworks.com 

,


SISTEM DE MĂSURARE COMPUTERIZAT 

(Rezumat) 

În lucrare este prezentat un sistem de măsurare a abaterilor de formă geometrică. 

Piesa de măsurat este amplasată pe două reazeme reglabile, astfel încât dreapta 

adiacentă să fie paralelă cu placa de control pe care se efecuează măsurarea. Pentru 

măsurare se utilizează un traductor inductiv, alimentat de la o sursă de curent continuu. 

Se utilizează o cartelă de achiziţie de date, un PC şi mediul de programare LabVIEW. 

Cu datele astfel obţinute se pot trasa diagrame ale formei reale a piesei. Datorită 

zgomotelor de măsurare, forma reală a piesei este mai greu de observat. Pentru a 

înlătura influenţa zgomotelor de măsurare se recomandă o filtrare de tip ”trece jos” a 

profilului obţinut. Aceasta se poate face cu ajutorul funcţiei smooth a programului 

Matlab, cu nivele diferite de filtrare.


Publicat de 



Secţia 


ADHESIVE BONDING OF SURFACE TREATED ALUMINIUM 

BY 

BIRGIT KJÆRSIDE STORM 

Abstract. Adhesive Bonding of aluminium and aluminium alloys can be carried 

out so a strong bonding is the result. The perfect bonding between two aluminium 

parts will be a bonding, where the material will break in the aluminium part, because 

in that case the bonding is stronger than the aluminium. For making a strong bonding 

a structural adhesive shall be used. A structural adhesive can be made of epoxy, 

polyurethane, acrylate, cyanoacrylate and a few other types. In some cases an 

anaerobe adhesive can cure up with aluminium. There is interaction between the 

adhesive, the aluminium alloy and the surface treatment. Aluminium creates in 

atmospheric air immediately an oxide layer on the surface. Anodized aluminium 

reacts with contaminations and CO2, which gives a surface with a low surface tension, 

though aluminium will have a high surface tension. The low surface tension of the 

reacted surface cause a bad wetting of the surface with the adhesive, and an 

insufficient wetting causes bad adhesion between the aluminium and the adhesive. For 

making a good bonding a good wetting is necessary. The sufficient wetting can be 

made on the aluminium by making a surface treatment. The surface treatment can be 

an anodizing. For obtaining the best adhesion on anodized surface the adhesive shall 

be carried out before the sealing immediately after the anodizing. The fresh anodized 

aluminium has a bigger surface and open pores, which cause a mechanical bonding in 

addition to the adhesion. The non sealed aluminium has reactive groups, which in 

some cases can react with reactive groups in the adhesive and can cause a chemical 

bonding as well. The non-sealed anodized aluminium can be treated with a primer, 

which can react with the active groups and can run into the pores and fill out the 

porousity. Before the primer has cured up totally the adhesive shall be added to the 

system for making a chemical reaction between the primer and the adhesive. Using 

such a system can give a higher strength in the bonding. 

Key words: aluminium, anodizing, adhesive, adhesive bonding, surface 

treatment, wetting 


Adhesive Bonding of aluminium and aluminium alloys can be carried out so 

a strong bonding is the result. The perfect bonding between two aluminium

106 Birgit Kjærside Storm 

parts will be a bonding, where the material will break in the aluminium part, 

because in that case the bonding is stronger than the aluminium. 

Anodized aluminium can - as aluminium with other surface treatments - be 

bonded by an adhesive bonding. For obtaining a strong bonding, which means a 

bonding which has a least the same strength as the aluminium alloy, there need 

to be a good adhesion between the anodized aluminium surface and the 

adhesive. For adhesive bonding there are various theories. No single theory 

explains adhesion in general and in reality it is probably a combination between 

the different theoretical explanations. 

There are six theories of adhesion. They are physical adsorption theory, 

chemical bonding theory, diffusion theory, electrostatic theory, mechanical 

interlocking theory and weak boundary layer theory. 

For anodized aluminium the physical adsorption theory and mechanical 

interlocking are the main methods for making the bonding, but the chemical 

bonding theory can occur if the surface is treated in a special way. 

The physical adsorption need to be present, if a bonding shall be strong 

enough. For creating the physical adsorption it is necessary to have a sufficient 

wetting of the surface. 

The physical adsorption gives the adsorption forces in the interphase 

between the metal or the anodized layer and the adhesive. The larger interphase 

between the two interfaces the higher strength will be obtained in the adhesive 

bonding. 

The mechanical interlocking will always be possible and in many cases, it is 

the only connection between the metal/anodized layer and the adhesive. This is 

the case, if the adhesive is not able to wet the surface. 

The mechanical interlocking is good in many ways, because if the surface is 

larger the area over which the physical adsorption can occur is larger. If the 

surface is rough the forces created in the interphase will give forces in different 

directions, which also will give a better physical adhesion. 

The chemical bonding in the interphase is the absolute most attractive force 

to obtain, if it can be possible. In some cases it is possible to choose an 

adhesive, which can react chemical with the surface and especially this can be 

possible in a fresh made anodized aluminium surface, where reactive groups 

still are present. 

2. Wetting and HSP 

The surface shall be able to be wetted by the adhesive, if a strong bonding 

shall be created. If it is possible to obtain a good wetting, the possibility to 

increase the physical adsorption in the interphase is present. For obtaining a 

good wetting the metal/anodized metal surface must have a surface tension 

which is higher than the surface tension of the adhesive. In figure 1 the wetting


of a surface and the determination of the surface tension by the droplet method 

is sketched. The surface tension of as well the metal/anodized metal and the 

adhesives depends on the secondary bonding forces in the material. For 

aluminium and anodized aluminium the surface will react with CO2. This 

reaction causes a low surface tension of the surface, which gives a bad wetting 

with the adhesive. 

The surface tension can be measured or it can be calculated from knowledge 

about the solubility parameters. The solubility parameters can be measured 

indirectly or they can be calculated. 

Fig. 1 – Surface tension and determination of surface tension [5]. 

The solubility parameters can be a one-dimensional parameter or a three 

dimensional parameter. The solubility parameter is defined as 

(1) δtot 2 = ∆Evap/Vl, 

δtot is the total is the total solubility, ∆Evap the heat evaporation and Vl is the 

molar volume. 

The one dimensional solubility parameter gives not the total answer for the 

secondary bonding forces. Therefore Charles Hansen defined the Hansen 

Solubility Parameters, HSP, which are three dimensional solubility parameters. 

(2) δtot 2 = δd 2 + δp 2 + δh 2 , 

δd is the contribution from the dispersion forces, δp is the contribution from the 

polar forces and δh is the contribution from the hydrogen forces. 

The three contributions will be put into a three dimensional coordinate 

system, where the dispersion forces will be printed with the double value of the 

two others. Each material, chemical or solvent will besides a three dimensional 

solubility parameter, which will be the centre in a sphere, also have an action 

radius, R0. The action radius is the area around the centre where the 

material/chemical is able to be dissolve or in other way to interact with other 

materials. The HSP can be set up in a three dimensional diagram and the 

wetting possibility can in that way directly be observed by studying if there are 

overlap by the spheres. In Fig. 2, two spheres are set up besides each other.


They are not able to cover each other and the materials behind them are not able 

to wet each other. 

Fig. 2 – Two spheres in a HSP diagram. 

The two spheres are not touching each other. The two materials behind are not compatible. 

If wetting of a material with another material shall be possible, the solubility 

spheres for the two materials shall at least touch each other. To calculate if 

wetting of a material with another material for example an adhesive is possible 

the RED – relative energy density – value can be calculated. The RED value is 

defined as: 

(3) RED = Ra/R0, 

Ra is the distance between the centre of the two spheres and R0 is the centre of 

the material, which shall be compared with the other sphere. If the RED number 

is 1, there is no compatibility/solubility between the two 

compared materials. 

As mentioned above the surface tension, γ, can be measured or be calculated 

from the HSP. Different researchers have found the correlation between the 

surface tension and the HSP. The most used is 

(4) γ = 0,0688V 1/3 [δD 2 + k (δP 2 + δH 2 )] 

γ is the surface tension and k is constant depending on the liquids involved. 

Other similar correlations are used.


3. The surface of Anodized Aluminium 

Aluminium alloys and anodized aluminium will theoretically have a high 

surface tension, so wetting of the surface with an adhesive would not be a 

problem, because the adhesives will have a surface tension much lower. 

Aluminium and anodized aluminium will react with CO2 and possible 

contaminations in the air and will there fore in practical have a surface tension, 

which is so low that wetting with the adhesives is not possible, because the 

actually surface tension is lower for aluminium surface than for the adhesive. 

If the surface is new created the surface tension will still be high and wetting 

will not be a problem. For solving the problem with not anodized aluminium the 

aluminium surface is often chromatized or phosphatized to obtain a good wetting. 

The surface of anodized aluminium itself is possible to wet with an adhesive 

especially if it is new treated and that will say that as soon after the anodizing the 

wetting will take place the better adhesion will be obtained. The bigger surface 

the better wetting is possible and that means also that if the anodized surface is 

used for bonding before sealing. The fresh anodized aluminium surface is the best 

surface for adhesive bonding, because the surface there has reactive groups, 

which will give possibility for chemical bonding in the interphase. The reactive 

groups give a bigger contribution to the surface tension because of more hydrogen 

bonds and the physical adsorption in the interphase will increase. The mechanical 

interlocking will also increase because of the porousity in the anodized surface. 

After sealing the surface is like un-anodized aluminium and the advantages 

obtained by the anodizing will disappear for the bonding. 

4. Surface Treatments 

Before bonding the surface need to be clean. The surface shall as mentioned 

above be with so many reactive groups as possible and the surface tension shall 

be as big as possible. The possibility for mechanical interlocking shall be 

present if possible. 

If the surface is contaminated with oil or grease from production or cutting 

process the surface tension will be the same as the surface tension of oil and 

grease, which will say very low, because those materials will almost have a 

contribution form the dispersion forces, very few if any from the polar forces 

and no from the hydrogen forces. Such a surface is impossible to bond on. The 

grease and oil shall be removed with an organic solvent as f.ex. isopropyl 

alcohol or CO2. The surface can if it is non anodized aluminium or is a sealed 

anodized aluminium be treated with a Scotch Brite for obtaining a bigger 

surface for mechanical interlocking. The treatment with Scotch Brite shall take 

place before the cleaning for removing dust from the surface.


A fresh anodized surface does not need any treatment if it has not been 

touch before bonding. 

5. Adhesive Bonding 

An adhesive bonding shall be constructed so the forces on the bonding is the 

best possible for the bonding. An adhesive bonding like a shear stress and 

dislike a direct tension. In Fig. 3 the two types of forces are sketched. 

Fig. 3 – The figure shows to the left, the shear force and to the right, the tension [3]. 

The adhesive bonding dislikes a peeling, but in practice it is difficult to 

create a bonding, where peeling is not possible. The peeling is the weakest point 

for an adhesive bonding and is often the limitation for making the bonding. If it 

is possible the bonding shall be constructed so the forces are shear forces and 

peeling cannot occur. 

6. Structural Adhesives 

Adhesives used for aluminium will be a structural adhesive. With a 

structural adhesive means an adhesive with a high strength. The perfect bonding 

will be a bonding where the bonding will have a strength which is the same as 

the aluminium. The bonding shall not break in interphase between the 

aluminium and the adhesive but in the aluminium or in the adhesive layer. If the 

break will be in the interphase, the wetting has normally not been good enough, 

or the cleaning of the surface has not been sufficient, or the forces in the surface 

of either aluminium or the adhesive have not been sufficient. 

Structural adhesives can be many different polymers. All of them are 

polymers which will cure up under cross linking. All of them will have an 

opening time. The opening time – the time from the adhesive has been added to 

the product until it will start the cross linking - depends on the curing system 

and can be regulated to be short or long depending on the application of the 

product. There are of course limitations for how short and how long the opening 

time for the adhesive can be. The structural adhesives have no green strength, 

which means that for several applications it is necessary to use a tool to fix the 

products until the adhesive has obtained the strength. The structural adhesives 

are epoxy adhesives, polyurethane adhesives, acrylic adhesives or other 

adhesives which can cross link. In Fig. 4 the curing reaction of an epoxy


adhesive is shown. There are many different types of curing reactions for epoxy 

and the shown is only an example. In Fig. 5 the reaction of a polyurethane 

adhesive is shown. 

Fig. 4 – Curing of epoxy with a primary amine [4]. 

Fig. 5 – Formation of biuret from diisocyanat and urea [2]. 

A structural adhesive can also be a cyanoacrylate, but for working with a 

quick curing adhesive can give problems in a production. On the market there 

are many different adhesives of all the mentioned types. They are different in 

opening time and in strength and flexibility. They can have different colours as 

well, and often the colours are used as an indicator for seeing if the adhesive is 

placed all over where it shall be. 

The properties of the adhesive depends on the polymer and the cross linking 

system in the adhesive. Aromatic groups in the adhesive give higher strength 

and aliphatic groups give higher flexibility. The more aromatic groups and the 

higher density of the aromatic groups the higher strength will be obtained. The 

more aliphatic groups there are in the polymer, the better flexibility there will be 

possible to obtain. The density of the reactive groups in the adhesive will also 

give possibility for strength and flexibility. The more reactive groups for cross 

linking there are, the higher rate of cross linking can be obtained and the higher


rate of crosslinking the higher strength will be obtained. Opposite will a low 

density of reactive groups for crosslinking gives a lower strength and a higher 

flexibility of the adhesive. Bonding with a high strength and stiffness is 

necessary under a shear stress and especially if a peeling is a risk. If the load 

will be a dynamical load the bonding need to have some flexibility as well as 

strength and the adhesive need certain flexibility. 

The epoxy adhesives can general obtain the highest strength, because I most 

epoxy adhesives the density of aromatic groups is high. Most of the epoxy 

resins are made on basis of bis-phenol A, which contains two aromatic groups. 

In Fig. 6 the formation of the epoxy resion bisphenol-A-diglycidylether 

(DGEBA) is shown. Aliphatic groups can be added and they will change the 

strength and the flexibility of the cured adhesive. Adding aliphatic groups can 

also be used for changing the opening time of the adhesive, because a longer 

aliphatic chain will decrease the density of reactive epoxide groups and there 

fore change the reaction time. Epoxy adhesives are normal two component 

systems. With mixing the two reactants the curing process will start. Increasing 

temperature will decrease the reaction time. With changing in reaction 

temperature the curing process – the cross linking process – can change and the 

properties of the adhesive can change. 

Fig. 6 – Formation of DGEBA from bisphenol A and epichlorhydrin [4]. 

The polyurethane adhesives can as well obtain a good strength. Most of the 

polyurethane adhesives can obtain a good flexibility, especially if they are made 

on basis of an aliphatic isocyanate. The polyurethane adhesives can be one- or 

two component systems. Two component systems will start the reaction, when 

the isocyanate and the polyol – the two active components – are mixed with 

each other. Again increased temperature will decrease the reaction time and 

increased temperature can change the properties of the adhesives. One 

component polyurethane adhesives will start the reaction under influence of 

humidity in the atmosphere. The humidity will react with one of the 

components in the adhesive and make it active for reacting with the other 

component. If the humidity is low the reaction will not take place as expected.


The acrylic adhesives cure up with a radical reaction over a double bond. For 

initiating the reaction an initiator system shall be reacted. This can happen under 

influence of adding the initiator or under adding energy f.ex. with UV irradiation 

for starting the reaction. The acrylics can be one or two component adhesives. 

The rate of cross linking depends on the curing system. The rate of the cross 

linking depends also on the temperature and on the time in which the system will 

react. 

Fig. 7 – The HSP plot of DP460 from 3M. DP460 is an epoxy based adhesive. 

Fig. 8 – The HSP plot of Sikaflex-360HC. Sikaflex-360HC is a polyurethane based adhesive.


The HSP can be measured for the different adhesives. In Fig. 7 the HSP plot 

of the epoxy based adhesive DP 460 from 3M is shown and in Fig. 8 the HSP plot 

of the polyurethane adhesive Sikaflex from Sika Industries is shown. 

From comparing the HSP for an adhesive with the HSP for a surface – 

aluminium or anodized aluminium – it can be foreseen if it can be possible to 

wet the surface with the adhesive and if it can be possible to make a reaction 

between the adhesive and the surface. Fig. 9 shows the HSP plot for an unsealed 

anodized aluminium surface. 

Fig. 9 - The HSP plot of an unsealed anodized aluminium. 

The anodizing has been made in the laboratory 

For an unsealed anodized surface it can with some polyurethane adhesives 

and some epoxy adhesives be possible to make a chemical bonding in the 

interphase. Such an adhesive can be used as a primer if it can be possible to find 

another adhesive, which can give the strength of the bonding and can wet the 

primer. 

7. Results 

For several systems combined with anodized aluminium and a structural 

adhesive the strength and the flexibility has been measured. The adhesives used 

in the experiments have been commercial adhesives and the anodizing process 

has been carried out either in a laboratory or in a commercial system. For the 

laboratory made anodizing the process has been made with parameters and 

chemicals as near the commercial systems as possible.


The strength of the bonding depends of the adhesive and of the surfaces. In 

a group of experiments where the aluminium alloy 6063 with a tensile strength 

of 100 MPa was used the strength of the bonding on plate of 20 mm and the 

bonding area of 20 x 20 mm 2 have for anodized unsealed aluminium been 

measured to app. 7400 N with DP460 and to 3700 N for the flexible Sikaflex - 

360HC. The strength of bonding with DP460 (epoxy resin) can be up to 10000 

N on the area 20x20 mm 2 . 


It is possible to bond aluminium with an adhesive and obtain a good 

strength and flexibility. The adhesive bonding is best for a shear stress and 

sensitive for especially a peeling. The adhesive used will always be a structural 

adhesive. Epoxy adhesives, polyurethane adhesives and acrylics adhesives are 

the most common used adhesives for aluminium. 

For obtaining a good bonding the surface of the aluminium needs to be 

wetted by the adhesive. If a surface shall be wetted by an adhesive the surface 

tension of the surface need to be higher than the surface tension of the adhesive. 

For estimating the possibility of wetting of the surface with an adhesive the 

HSP can be used as a tool. 

Adhesive bonding give a better strength for anodized aluminium if the 

bonding is made on unsealed anodized aluminium. The unsealed anodized 

aluminium is porous and will give a better mechanical interlocking together 

with a physical adsorption in the interphase between the anodized layer and the 

adhesive. Use of an adhesive as a primer can increase the strength in the 

interphase. If it is possible a primer or an adhesive, which can give a chemical 

reaction between the unsealed anodized aluminium, shall be used for obtaining 

the highest strength of the bonding. 

Received:January, 30,l 2010 Aalborg University, 

Esbjerg Institute of Technology 

Esbjerg, Denmark 

e-mail: bks@bio.aau.dk 


1. H a n s e n Ch. M., Handbook of Solubility Parameters – A User´s Handbook, CRC 

Press, 1 st edition, 2000. 

2. O e r t e l G., Polyurethane Handbook, Hanser Publishers, 2 nd edition, 1994. 

3. P e t r i e E. D., Handbook of adhesives and sealants, McGraw-Hill, 2000.


4. P i z z i A. et al., Handbook of Adhesive technology, Marcel Dekker, Inc, 2 nd edition, 

2003. 

5. www.specialchem.adhesives.com 

LIPIREA CU ADEZIVI A ALUMINIULUI TRATAT SUPERFICIAL 

(Rezumat) 

Lipirea cu adezivi a aluminiului şi aliajelor de aluminium poate fi realizată astfel încât 

să se obţină îmbinări rezistente. O lipire perfectă a două piese din aluminiu va fi acea 

lipire la care ruperea îmbinării are loc în piesa de aluminiu, lipitura fiind mai rezistentă ca 

aluminiul. Pentru a obţine o lipitură de mare rezistenţă, se vor folosi adezivi structurali, 

care pot fi de tip epoxi, poliuretan, acrilat, cianoacrilat etc. În unele cazuri, un adeziv 

anaerob poate interacţiona cu aliajul de aluminiu şi cu tratamentul de suprafaţă. 

Aluminiul, în aer atmosferic, creează imediat un strat de oxid la suprafaţă. Aluminiul 

anodizat reacţionează cu contaminatorii şi cu CO2 rezultând o suprafaţă cu tensiuni 

superficiale reduse, deşi aluminiul are tensiuni superficiale ridicate. Asemenea tensiuni 

superficiale reduse vor determina o umectare slabă a suprafeţei cu adeziv, o umectare 

insuficientă determinând o adeziune necorespunzătoare între aluminiu şi adeziv. Pentru o 

lipire de calitate este necesară o bună umectare. O umectare suficientă poate fi obţinută 

prin aplicare de tratamente superficiale. Un asemenea tratament poate fi unul de 

anodizare. Pentru a obţine adeziune optimă pe suprafaţa anodizată, adezivul trebuie aplicat 

imediat după anodizare. Aluminiul proaspăt anodizat are suprafaţă de contact mai mare şi 

pori deschişi, rezultând o lipire mecanică suplimentară adeziunii. Aluminiul poate avea 

grupuri reactive care uneori pot reacţiona cu grupuri reactive din adeziv, rezultând şi o 

lipire chimică. Aluminiul anodizat poate fi tratat cu o amorsă, care poate reacţiona cu 

grupurile active şi poate intra în pori şi umple porozităţile. În asemenea cazuri, adezivul se 

aplică înainte de epuizarea amorsei, pentru a avea loc reacţie chimică între amorsă şi 

adeziv. Asemenea soluţii pot asigura rezistenţă sporită îmbinărilor cu adezivi.


Publicat de 



Secţia 


INDUSTRIAL APPLICATIONS OF 

THE PNEUMATIC MUSCLES 

BY 

IOANA PETRE 1 , DAN PETRE 2 , CRISTINA FILIP 1 , 

and LAVINIA NEAGOE 1 

Abstract. Pneumatic muscles were, lately, used in many applications 

especially in the field of industrial robots, because of the evolution of the 

technology. The characteristics of the pneumatic muscles make them being easy 

to use and with great performances. Pneumatic muscles have applications in 

robotics, biorobotics, biomechanics, artificial limb replacement and industry. In 

this paper I will present some industrial applications of the pneumatic muscles. 

The utilization of the pneumatic muscle is soon to become a good alternative for 

the present day electric or mechanically drives because of their characteristics and 

because of the evolution of the industrial development. 

Key words: pneumatic muscle, actuator, industrial application. 


Although the pneumatic muscles have been conceived since 1930 by S. 

Garasiev, a Russian inventor, only for a few years, they have been used in 

different applications. Lately, pneumatic muscles became a better choice than 

present day electric or other drives. They are usually used in factory floor 

automation and nowadays, in industrial robotics as a main motion power source. 

The current paper will focus on the pneumatic muscle with it’s applicability 

in the industry. These actuators are usually cylindrical in shape and they are 

composed by an interior inflatable tube typically made by neoprene rubber 

wrapped in a multilayer tissue typically made by nylon. The tube, under the 

action of compressed air, increases its diameter and decreases its length; the 

stroke resulted is in direct relation with the pressure of the compressed air 

passed into the muscle.

118 Ioana Petre et al. 

The utilization of the pneumatics offers many advantages, the most important 

being the low weight and the inherent compliant behavior of its actuators. 

Compliance ensures a soft touch and safe interaction. In contrast with 

pneumatic muscle actuator, hydraulic and electric drives have a very rigid 

behavior and can only be made to act in a compliant manner through the use of 

relatively complex feedback control strategies. [1] 

2. Pneumatic Muscle 

2.1. History 

Pneumatic muscle is an actuator system based on an inflatable and flexible 

membrane operated by pressurized air (Fig. 1). 

Fig. 1 – Pneumatic muscles and their function principle 

(source http://www.festo.com/hm2001/eng/1001.htm). 

Pneumatic muscles were first conceived in 1930 by G a r a s i e v, a Russian 

inventor [2]. According to B a l w in [3], J. L. McKibben introduced it to 

motorize pneumatic arm orthotics for helping control handicapped hands: due to 

the similarity in length-load curves between this artificial muscle and skeletal 

muscle, it seemed an ideal choice for this purpose [4], [5]. 

The artificial muscle, which construction is simple, was made of a rubber 

inner tube covered with a shell braided according to helical weaving. The 

muscle was closed by two ends, one being the air input and the other the force 

attachment point. When the inner tube was pressurized, the muscle inflated and 

contracted [6]. 

The Bridgestone rubber company (Japan) proposed a redesigned and more 

powerful version of the pneumatic muscle in the 1980s under the name of 

Rubbertuators and used them to power an industrial use robot arm, Soft Arm. 

At the present, McKibben-like muscles are being brought to the market by 

Festo Ag. & Co.


2.2. Generalities 

The pneumatic muscles have many characteristics that make them being 

easy to use and with great performances. Some of these characteristics are: 

shock - absorbing, adjustability, simulating capability, storage capable, 

safeness, lightweight, natural compliance and shock resistance. 

Favourable response to commands, known as compliance is in direct 

relation to air compressibility, and hence the pneumatic muscle can be 

influenced by controlling/adjusting the command pressure. 

The function principle consists in the fact that, under the action of 

compressed air, the pneumatic muscle, which is blocked at one end, shortens its 

lengths and expands its diameter. As the volume of the internal tube increases 

due to the increase in pressure, the actuator shortens with a certain stroke. 

Pneumatic muscle construction is based on an interior tube, made from 

neoprene rubber wrapped in braided sleeves made of nylon with strengthening 

and protecting role. The braided sleeves act to constrain the expansion for 

maintaining the cylindrical shape. As we can see in the figure 1, the angle of the 

enveloping tissue, denoted by α, is one in relaxed state and differs in contracted 

state. It has the value of 25.4º in the relaxed state of the muscle and of 54.7º at 

maximum contraction. [4] 

In Fig. 2 is presented the working principle of a pneumatic muscle. 

Fig. 2 – Working principle of a Fig. 3 – Dependence of the force on the 

pneumatic muscle envelope angle and on the working pressure [7] 

(1) 

The force F developed by pneumatic muscle is given by Eq. (1) [4]. 

π 

= ⋅ ⋅ 

4 d p F 

2 

⎡ 2 

3⋅ 

cos α − 1⎤ 

⋅ ⎢ 2 ⎥ 

⎢⎣ 

1 − cos α ⎥⎦ 

, 

where p is the working pressure and d the interior diameter of the pneumatic 

muscle. Upon completion of the maximum stroke the developed force is equal 

to zero. Equation (1) allows plotting of the graph featuring the force developed


by a pneumatic muscle versus the enveloping angle and feed pressure (Fig. 3) 

[5]. 

The advantages of the pneumatic muscles utilization are: power to weight 

ratios in excess of 1 kW/kg; a varying force-displacement relation at constant 

gas pressure, an adjustable compliance; the absence of friction and hysteresis; 

the ability to operate at a wide range of gas pressures, and thus to develop both 

very low and very high pulling forces; the possibility of direct connection to a 

robotic joint; cheaper to buy and install than other actuators and pneumatic 

cylinders; smooth and natural movement; fast -full contraction. 

Disadvantages are: the force which can be applied is only tensile in nature; 

its total displacement is only about 20% to 30% of its initial length; friction 

between the netting and the tube leads to a substantial hysteresis in the forcelength 

characteristics; rubber is often needed to avoid the tube from bursting; 

rubber deformation will lower the force output of this type of muscle up to 60% 

[8]. 

3. Industrial Applications of Pneumatic Muscles 

Pneumatic muscles were, lately, used in many applications especially in the 

field of industrial robots, because of the evolution of the technology. 

The major application of pneumatic muscles is in the field of industrial 

robots, usually in those that mimic human actions. Because of their analogy 

with human skeletal muscles, they are able to perform many functions specific 

to human hand. 

As example in this area we can specify the stepping robot WAP-1 is the first 

biped robot designed in 1969 by Ichiro Kato from Waseda University of Tokyo. 

(Fig. 4a) 

a b 

Fig. 4 a – WAP-1(source www.androidworld.com) ; b - Humanoid robot manufactured 

by Shadow robotic company (source www.shadowrobot.com)


The humanoid bipedal walking machine developed by Shadow Robot 

Company is a robot that can function in a human environment to do various 

repetitive things. (Fig. 4b) 

The humanoid muscle-robot developed by Festo AG & Co. in collaboration 

with the department of bionics and evolutionary technology of the Technical 

University of Berlin (Fig. 5). This robot is a feasible extension and place holder 

of men capable of operating in places either too dangerous or inaccessible for 

human beings. It can operate from terrestrial and deep sea operations to tasks 

carried out in outer space. 

Fig. 5 – Humanoid muscle-robot moving (source www.festo.com). 

Pneumatic muscles have been used in construction of artificial limbs. 

At the bio robotic lab of University of Washington the limb as shown figure 

6a was developed. The major requirements of their research team were: 

continuous operation, low weight, quieter operation, user satisfaction, no 

maintenance. 

a b 

Fig. 6a – Artificial limb developed at the bio robotics Lab, University of Washington. 

b – Shadow Leg (source www.shadowrobot.com)


Another application for the pneumatic muscle is the leg made for Shadow 

Robot Company by David Buckley from North Carolina A & T University. 

(Fig. 6b) 

The dexterous hand was developed by the Shadow robotic company. The 

hands operate just like human hands with five fingers. It contains an integrated 

bank of 40 Air Muscles which make it move. The figure 7 illustrates this fact. 

Fig. 7 – The Shadow Hand (source www.shadowrobot.com). 

Caldwell used 18 small McKibben Muscles to power a dexterous fourfingered 

manipulator. Hannaford built an anthropomorphic arm, having fifteen 

McKibben Muscles. The Soft Arm, developed by Bridgestone Co. has a 

shoulder, an upper arm, a lower arm and wrist, and a useful payload of 

maximum 3 kg. Yoshinada used hydraulically actuated McKibben Muscles to 

power an underwater manipulator [1]. 

New Application Areas are: Simulator Technology, High Speed cutting 

processes, Aeronautical Technology, Wood-working, Metal-working, Medical/ 

Biomedical, Mobile Applications, Building / Construction, Mining, Process/ 

Water, Amusement/ Recreation, Medical / Surgical / Hospital 


1. Even though pneumatic muscles are not capable of offering an 

extremely wide range of operations, in the case of humanoid robots they offer a 

wide range of possibilities. Their low assembly weight and high power-toweight 

ratio make the pneumatic muscles to be considered for use in mobile 

robotics [1].


2. The utilization of the pneumatic muscle is soon to become a good 

alternative for the present day electric or mechanically drives because of their 

characteristics and because of the evolution of the industrial development. 

Received: March 15, 2010 1 Transilvania University of Braşov, 

Technological Engineering Faculty, 

Economic Engineering and Production Systems Department, 

Braşov, România 

e-mail: ioana.petre@unitbv.ro 

2 Material Science and Engineering Faculty, 

Materials Engineering and Welding Department, 

e-mail: petredanbv@yahoo.com 


1. D a e r d e n F., L e f e b e r D., The concept and design of pleated pneumatic 

artificial muscles, International Journal of Fluid Power, 2, 3, pp. 41–50, 2001. 

2. M a r c i n č i n J., P a l k o A., Negative pressure artificial muscle—An 

unconventional drive of robotic and handling systems, Transactions of the 

University of Košice, pp. 350– 354, Riecansky Science Publishing Co, Slovak 

Republic, 1993. 

3. B a l d w i n H. A., Realizable models of muscle function, Proceedings of the First 

Rock Biomechanics Symposium, pp. 139–148, New York, 1969. 

4. S c h u l t e H. F., The characteristics of the McKibben Artificial Muscle, The 

Application of External Power in Prosthetics and Orthotics, pp. 94–115, 

National Academy of Sciences–National Research Council, Publication 874, 

Lake Arrowhead, 1961. 

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 

pneumatic muscle actuators, 4th International Conference on Climbing and 

Walking Robots CLAWAR, 2001. 

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 

Pneumatic Testing, 2006 

http://www.ee.cityu.edu.hk/~rtbrad/muscles%20riupeeec%202006.pdf 

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 

of Pneumatically Actuated Artificial Muscles. 6 th International Conference of 

DAAAM Baltic Industrial Engineering, Tallinn, Estonia, Vol. 1, pag. 215-220, 

2008 

8. * * * Air Muscles 2008 [Online] Available at: 

http:// www.techalone.com 23/01/2010] – [Accessed 27.12.2009] 

http:// www.festo.com – [Accessed 27.12.2009] 

http://www.shadowrobot.com – [Accessed 27.12.2009]


APLICAŢII INDUSTRIALE ALE MUŞCHILOR PNEUMATICI 

(Rezumat) 

Muşchiul pneumatic este un sistem bazat pe o membrană care se contractă, şi care, 

sub acţiunea aerului comprimat îşi măreşte diametrul şi îşi micşorează lungimea. 

Lungimea cursei depinde in mod direct de nivelul presiunii alimentate. Un muşchi 

pneumatic include un tub interior care este realizat dintr-un material elastic, de obicei 

neopren. Acest tub este acoperit de un ţesut cu mai multe straturi realizat din nylon, 

pentru a-i da rezistenţă şi pentru a-l proteja de influenţele din mediul de lucru. 

Caracteristici precum capacitatea de a absorbi şocurile, greutatea redusă, rezistenţa 

la şocuri, gabarit si masa redusa pe unitatea de putere (1KW/kg), elasticitate 

(comportare ca de arc) datorata pe de-o parte compresibilităţii aerului si pe de alta 

variaţiei forţei cu deplasarea, amortizarea şocurilor datorate impactului, posibilităţi de 

conectare uşoara, siguranţa (fără pericol de electrocutare sau incendiu) fac din muşchii 

pneumatici instrumente fezabile pentru utilizarea uşoară in domenii diverse şi cu 

performanţe deosebite. Domeniile de aplicare ale muşchilor pneumatici se referă la 

robotică, biorobotică, biomecanică, dispozitive de protezare si sustinere a scheletului 

osos şi industrie. 

În această lucrare se prezintă câteva aplicaţii industriale ale muşchilor pneumatici. Deşi 

prima apariţie a muşchilor pneumatici a fost în 1930 fiind inventaţi S. Garasiev, iar apoi 

în 1950 muşchii cu membrană împletită introduşi de J. L. McKibben aceştia au cunoscut 

o utilizare mai intensa abia in ultima perioadă. 

Câteva exemple in domeniul roboţilor industriali sunt: braţul acţionat pneumatic 

realizat de J.L.Mc Kibben, robotul WAP 1 realizat la Universitatea Waseda din Tokyo, 

robotul humanoid conceput de Festo in colaborare cu Universitatea Tehnică din Berlin, 

robotul humanoid realizat de Compania Shadow, proteza realizată de Universitatea din 

Washington, piciorul şi mâna realizate pentru Compania Shadow, care efectuează 

mişcări similare cu cele efectuate de corpul uman. Desigur că aplicaţii sunt multe si nu 

au fost toate menţionate in acest articol, însa trebuie reţinut că muşchii pneumatici, 

datorită caracteristicilor lor, sunt utilizaţi în special în domeniul roboţilor mobili. 

Aceştia înlocuiesc cu succes motoarele electrice sau hidraulice fiind mai uşor de folosit 

şi îndeplinind aceleaşi funcţii. 

Dezvoltarea continuă a tehnologiei a dus la apariţia unor noi arii de aplicaţie pentru 

muşchii pneumatici, cum ar fi: tehnologia simulării, procese de tăiere cu viteză mare, 

tehnologia aeronautică, lucrul cu lemnul, lucrul cu metalul, industria medicală, 

construcţii, industria minieră.


Publicat de 



Secţia 


SOME NEW RESOURCES ON COMPUTER ASSISTED 

EXPERIMENTAL RESEARCH OF THE ABSORBED ELECTRIC 

POWER IN MANUFACTURING SYSTEMS 

BY 

MIHĂIȚĂ HORODINCĂ 

Abstract. This paper presents some theoretical considerations and experimental 

results on computer aided research of manufacturing systems (e.g. machine-tools) 

using the evolution of the absorbed electric power (EP) of the driving motors. Based 

on an original computer aided data acquisition and processing method, the electric 

power evolution monitoring (and its components, especially the EP real part) should 

be now a very useful tool in experimental research (kinematic chains condition, 

working processes loading, and so on). In a new approach the numerical describing 

with high resolution of real EP evolution can be used also to evaluate the 

performances in dynamic conditions, especially using the evolution in frequency 

domain (by power spectral density analysis). Some phenomena in kinematic chains 

was observed with a certain scientific priority, using the real EP monitoring 

capabilities, such as the behaviour of driving belts and the behaviour of the elastic 

system of the rotor on the electric driving motor. The real EP monitoring can be 

used to supervise any other electric actuated equipment. 

Key words: manufacturing systems, experimental research, real electric power, 

computer aided monitoring, data processing. 


The working processes (WP) on manufacturing systems (MS) or any other 

working equipment needs mechanical energy, usually delivered by an electric 

driving motor (EDM) and transported by a kinematic chain. By mechanical 

loading point of view the behaviour of MS is mirrored in the real part of electric 

power (EP) evolution (active electric power, AEP). The EDM seems to be a 

very appropriate sensor of mechanical loading, useful in MS and WP 

monitoring and diagnosis or even in WP active control. The state of the art [2], 

[3] indicates that generally for monitoring it is used just the electric current

126 Mihăiță Horodincă 

absorbed by the EDM, but for MS driven with alternating voltage EDM, the 

best characterization of loading is done using AEP evolution. In some papers 

[4], [5] the AEP it is taken into account, but the research resources are not 

completely exploited. This paper presents some experimental research results 

based on a new approach on computer aided AEP monitoring for MS powered 

by EDM. 

2. EP Monitoring Principle on EDM and Experimental Conditions 

Fig. 1 describes a computer assisted monitoring bench test developed for an 

asynchronous EDM supplied on a three phase 50 Hz frequency sinusoidal 

voltage, symmetrical network (3×380V). A single phase is used for computer 

data acquisition (instantaneous current i(t) IC and voltage u(t) IV in time 

Fig. 1 – Experimental features of the EP bench test. 

evolutions [6] provided by two transformers CT and VT) via a numerical 

oscilloscope (ADC 212/50, Pico Technology Limited, UK). The acquired data 

are processed in a personal computer. The instantaneous electric power p(t) 

evolution (IEP) is given by: p(t)=3·u(t)·i(t). The AEP (one value on each period 

T) is given by: 

(1) 

T 

1 

3 

P = ∫ p( 

t) 

⋅ dt = ⋅ ∫u( 

t) 

⋅i( 

t) 

⋅ dt = 3⋅U 

⋅ I ⋅cosϕ 

, 

T T 

0 

T 

0 

where: T=20 ms is the period (reciprocal of the frequency f), U - the effective 

(root mean square, RMS value) voltage, I - the effective current, φ - the angle of


phase (AP) between current ant voltage. Using the numerical values u[tk] and 

i[tk] of the voltage and the current, the numerical evolution of the AEP can be 

described by calculus (based on Eq. (1), as average value on the period T) with: 

(2) 

n⋅( 

l+ 

1) 

n⋅( 

l+ 

1) 

3 

3 

P[ tl 

] ≈ ∑( 

u[ 

tk 

] ⋅i[ 

t k] 

⋅ Δ t) 

= ⋅ ∑( 

u[ 

tk 

] ⋅i[ 

t k]) 

, 

T 

n 

k= 

n⋅l+ 

1 

k= 

n⋅l+ 

1 

where: P[tl] is the numerical value of the AEP (tl=l·T, l=1,2….m, m values for a 

registration of m·T total duration), n=T/Δt, Δt is the reciprocal of the sampling 

rate for u[tk] and i[tk], tk=k·Δt. If Δt→0 then in Eq (2) the symbol ≈ becomes =. 

The sampling rate of AEP is equal with frequency f value (50 s -1 ). Using the 

effective values U[tl] and I[tl] (the amplitudes of the voltage and current divided 

by square root of 2), the AP and REP (reactive electric power Q) are given by: 

(3) 

P[ 

tl 

] 

ϕ [ tl 

] ≈ arccos 

and Q[ tl 

] ≈ 3⋅U 

[ tl 

] ⋅ I[ 

tl 

] ⋅sin( 

ϕ[ 

tl 

]) . 

3⋅U 

[ t ] ⋅ I[ 

t ] 

l 

l 

The experimental research was done on a Romanian lathe SNA 360 (used as 

MS, see Fig. 1) with the kinematics of the main shaft gearbox described in Fig. 

Fig. 2 – Overview on the kinematics of the SNA 360 lathe main shaft 

gearbox used as MS in experimental research. 

2. An asynchronous EDM with 5.5 KW and 1440 rpm (speed in rotations per 

minute) is used to drive it. Two different configurations of the gearbox were 

used (see Kd1 and Kd2), each one is able to actuate the main spindle (shaft) Ms 

on 1000 rpm or 1600 rpm.


3. Some Experimental Research Results on Monitoring 

Let be first a simple experiment. The belt drive transmission BT1 is 

switched-off (the belt was removed), no loading on EDM’s rotor anymore. 

Fig. 3 – The AEP evolution during EDM starting procedure 

(no loading on the EDM’s rotor). 

The evolution of AEP during the EDM starting procedure (star-triangle 

Fig. 4 – Gearbox start-stop clutching dynamics (on Kd1 

configuration) mirrored in AEP evolution. 

connection) is given in Fig. 3. A strong transitory regime (high power, short


time, sees Bs area) is observed, because of the dynamics of accelerated rotor 

motion, it needs a big amount of AEP stored as kinetic energy. In D the star 

connection is changed automatically in triangle connection, also a transitory 

regime (marked as Bt) occurs. 

Fig. 5 – Numerical filtered (with p=10) evolutions for AEP 

(given in Fig. 4) and REP. 

There is an interesting behavior of the EDM mirrored in AEP evolution, see the 

regions I and II. There are two free responses on 5.2 and 10.8 Hz frequency, due 

to the elastic system associated to the rotor (mechanical inertia, magnetic field 

stiffness and air friction damping). 

A second experiment describes (Fig. 4) the mirroring of the clutching 

dynamics of the gearbox (on Kd1). In A the MC3 clutch is engaged, the gearbox 

start the motion and the AEP consumption increases during a transitory regime 

B (with a 4.16 W·h average electrical energy stored as kinetic energy). The 

peak B1 indicates that the BT1’s driving belt start the sliding (the dynamic 

torque is bigger than the friction torque between the belt and the pulleys). In B2 

the sliding disappears and immediately after (in C) starts the steady-state regime 

on 1000 rpm. Here the absorbed AEP is used to cover the waste of mechanical 

energy (dry and viscous friction). 

The AEP evolution seems to be noisily, but it will be proved later (Fig.7) 

that it is very useful for gearbox diagnosis. In D the clutch is disengaged (by 

switching-off the electrical supply) so the AEP suddenly decreases. The level of 

power in E is for a short time lower than in A, the difference (183 W) describes 

the viscous friction in MC3 clutch.


For certain experimental purposes the AEP evolution can be numerical 

filtered using a sliding average low pass filter described with the equation: 

p xin[ 

i] 

xout[ 

j] 

= ∑ where: p is the filter parameter, xin[i] - the input signal, xout[j] 

i= 

1 p 

- the output signal, both in numerical format, j=i+p. Figure 5 shows the result 

of filtering for AEP evolution given in Fig.4 (with p=10). 

Also is presented the filtered evolution of REP in the same experimental 

conditions. As it is well known from the electrical network theory, the variation 

of REP (1,036 VAR peak to peak) is smaller than the AEP variation (4,805 W). 

The evolution of AEP during a third experiment is described in Fig. 6. 

There is a dynamic behaviour with negative absorbed AEP. Before A, the 

gearbox moves in steady-state idle running regime on 1600 rpm (using Kd2 

kinematic chain with MC2 clutch engaged). In A the clutch MC3 is engaged 

(and MC2 is automatically switched-off), so Kd2 becomes active and Kd1 

inactive. Here occurs a transitory regime B, with negative AEP (the angle φ > 

π/2, see Eq. (1)), for a short time the EDM works as a brake, it convert the 

Fig. 6 – Clutching dynamics with negative AEP. 

available kinetic energy from gearbox in electrical energy (with a total amount 

of 0.464 W·h) delivered on the electrical network. After that the AEP increases 

with an overshoot in C, the generator becomes again electric motor. In D the 

steady-state on 1000 rpm is installed. In E the MC3 clutch is switched-off, so 

just the shaft I is driven. The AEP consumption in steady-state idle regime is 

smaller than in Fig.4 because of the gearbox heating (the viscosity of the 

lubricant so the viscous friction decreases).


4. Gearbox Diagnosis using the AEP Processing in Frequency Domain 

There is also a very important experimental research resource useful in 

gearbox diagnosis. The AEP or IEP evolution in steady-state it is a mixture of 

many variable components. Each variable signal’s component describes the 

behavior of a gearbox’s part (belts, spindles, gears, et cetera). The identification 

of these components can be done [1] by computer assisted AEP or IEP power 

spectral density (PSD) analysis using Fast Fourier Transform (FFT). The 

correlation between signal’s components and gearbox’s parts can be done using 

the frequency and the speed of rotation. Figure 7 presents a zoom on the PSD of 

the AEP evolution in steady state on 1000 rpm (already described before; see on 

Fig. 4 the encircled area). 

On Fig. 7 there are a lot of different peaks, each one described by the frequency 

and the amplitude (the higher peak’s amplitude are not clearly indicated because 

Fig. 7 – The evolution of power spectral density of the 

AEP signal absorbed on 1000 rpm (steady-state). 

of zoom). The PSD amplitude is proportional with the amplitude of signal 

component. Each peak has a number, for each one there is a short description 

about the frequency and the correspondence with the gearbox parts (see Fig. 2). 

The peak 5 is associated with the spindle II behaviour, the peak 7 with the 

spindles III and V, the peak 9 with the spindle I and the peak 10 with the 

EDM’s rotor (24.84 Hz frequency, for 1490.4 rpm the real speed of rotation). 

The peak 9 is generated by the driving belt pulley run-out error motion (65 μm). 

There are also some harmonics of these peaks. There is a very interesting 

research result, with certain scientific priority, here is very well mirrored the


behaviour of the driving belt from BT1. The belt generates a very strong 

variable component of AEP, with the fundamental frequency on 5.4 Hz (first 

peak on Fig. 7) and 6 harmonics (see the peaks 4, 6, 9, 11, 14 and 18). It is 

generated by the belt’s stiffness variation (the belt is a little bit damaged, it has a 

small tear). The fundamental’s frequency fFB is given by: 

π ⋅ D1 

π ⋅125mm 

(4) 

fFB 

= f EDM ⋅ = 24. 

84Hz 

⋅ = 5. 

38Hz 

. 

L 

1812mm 

c 

Fig. 8 – The BT1’s belt behaviour mirroring in PSD 

analysis of the IEP signal evolution. 

where: fEDM is the rotation frequency of the EDM’s rotor, D1- the diameter of 

the belt pulley of the EDM’s rotor, Lc- the belts’ length. A new experiment was 

done, with PSD analysis on IEP evolution during the steady-state regime, with 

all the clutches disengaged, see Fig. 8. The BT1’s belt behaviour is better 

indicated with the harmonic F and the harmonics H1,H2…H6. Because the 

mechanical loading of the EDM is smaller, the speed of rotation is bigger so the 

frequency in Fig. 8 increases. The first harmonic is bigger than the fundamental 

because of the EDM’s rotor elastic system resonance. 


1. The paper proves that the computer-assisted monitoring of EP parameters 

on manufacturing systems (or any other electric actuated equipment) can be an 

available experimental research procedure. 

2. The computer assisted EP monitoring uses very simple procedures of data 

acquisition and processing. Using these research features, it is possible to detect 

and study a lot of new static and dynamic phenomena from kinematic chain.


Acknowledgements. The author would like to thank Prof. Costache D r u t u. For, he 

is the one who, long time ago, gave me the basic idea of some of my researches. 

Received: “Gheorghe Asachi” Technical University 

Department of Machine-Tools 

Iasi, Romania, 

e-mail: horodinca@tuiasi.ro 


1. B i r a n A., B r e i n e r M., Matlab for Engineers, Addison-Wesley Pub. (SD), May 

1995, ISBN-13: 978-0201565249. 

2. D i m l a E., D i m l a., Sensor signals for tool-wear monitoring in metal cutting 

operations – a review of methods, Int. J.of Machine Tools & Manufacture, 40, 8, 

(2000), pp. 1073-1098, ISSN:0890-6955. 

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 

monitoring research in Korea, Int. J. of Machine Tools & Manufacture, 39, 11, 

November, 1999, pp. 1697-1715. 

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 

using neural network based spindle motor power sensing, Int. J. of Machine 

Tools & Manufacture, 42, 10 (2002), pp 1113-1119. 

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 

end milling process, International Journal of Machine Tools & Manufacture, 42, 

14 (2002), pp. 1487-1496, ISSN:0890-6955. 

6. H o r o d i n c ă M., Utilizarea parametrilor energetici în monitorizarea, diagnoza şi 

conducerea sistemelor de prelucrare, Ed. Performantica, Iaşi, 2004. 

NOI RESURSE ALE CERCETĂRII EXPERIMENTALE 

ASISTATE DE CALCULATOR A PUTERII ELECTRICE 

ABSORBITE ÎN SISTEMELE DE FABRICAȚIE 

(Rezumat) 

Lucrarea prezintă o serie de noi oportunităţi ale cercetării experimentale a 

sistemelor de fabricaţie (monitorizare şi diagnoză) pe baza analizei evoluţiei temporale 

şi în domeniul frecvenţă a parametrilor încărcării energetice (putere activă PA şi 

instantanee PI) ai electromotoarelor de acţionare. Se prezintă exemplificări legate de 

monitorizarea proceselor tranzitorii şi de utilizarea analizei spectrale a semnalelor PA şi 

PI, pentru cercetarea comportării lanţurilor cinematice ale mişcării principale.


Publicat de 



Secţia 


INTEGRATED CAE DEVELOPMENT OF PRECESSIONAL 

DRIVES USING AUTODESK INVENTOR PLATFORM 

BY 

ION BOSTAN, VALERIU DULGHERU 

and ANATOL SOCHIREANU 

Abstract: The paper presents the modelling and simulation of processional 

drives designed in two variants capable of high transmission ratio and torque for one 

stage compact construction. The constructions were designed in Inventor and also as 

multi body systems in MotionInventor. The simulations of the drives provide 

information concerning positions, velocities, accelerations, point trajectories, forces 

and moments, energies, as well as contact forces at the contact between gear teeth 

and satellite teeth and other data concerning the system. The bearings of the two 

drives were modelled as multi body systems in SolidDynamics for defining an 

approach for studying the noise emission during running through simulation with 

possibility of extension to the contact between teeth. 

Key words: precessional transmissions, Inventor, 3D multi body model 

SolidDynamics 


The engineering methods based on computer permitted to develop a new 

type of processional transmissions with multi-couple meshed teeth, which, from 

the technological point of view, can be manufactured by means of a new 

method of processing conical teeth with convex-concave profile. 

It appeared the necessity of elaboration of new profiles adequate to the 

spheroid-spatial motion of the gears, which would ensure high performances to 

the processional transmission. 

Considering the necessity of achieving the transfer function continuity and 

gear multiplicity some objectives were taken into account. One of them is the 

integrated method of design, modelling and simulation using powerful means of

136 Ion Bostan et al 

creation and management of parametrical models of the mechanical assemblies 

on the basis of CAD-CAE. 

2. General Information 

The ever-growing requirements, especially, concerning the bearing capacity, 

the kinematical accuracy and the kinematical possibilities, impose the necessity to 

develop a new type of planetary gearing with distinct performances. The gearing 

improving is one solution of the problem. Novicov - Wildhaber, Symarc and other 

gearings have increased considerably the bearing capacity of gear. Another 

direction of developing gears is the design of new types of mechanical gears. 

The creative search of designers has crowned with the elaboration of a new 

type of gearing – harmonic gear. W. Musser, an American engineer, patented the 

action principle of the harmonic gear in 1959. Starting that year W. Musser 

patented a big number of diverse constructive diagrams for harmonic gears (teeth, 

friction and tapped gears) and couplings, and demonstrated the possibilities of the 

new construction principle of mechanical gears. Thus, in 1961, the harmonic 

drive was produced at one of the American companies, for the first time at 

industrial scale. Harmonic drive are compact and possess increased bearing 

capacity; they provide high kinematical accuracy and possibility to transmit 

motion in sealed mediums, which is one of the basic advantages of harmonic 

drive. As their disadvantages we can mention reduced reliability of the flexible 

element (and, thus, of the gear, on the whole), reduced working order at high 

speeds, and also some technological difficulties. 

By the end of the 70s Prof. Ion Bostan designed a new type of gears, new in 

principle, - processional planetary gears with multiple gearing (B o s t a n, 1991). 

Over 20 years, research was carried out comprising the total range of problems 

from the idea to the implementation: fundamental theory of the precessional gearengineering 

calculation procedures-Know-How manufacturing technologiesapplications. 

The research results have been published in over 450 scientific 

papers, in about 150 patents, in 3 monographs and in one design guidebook. 

The absolute multiplicity of the processional gear (up to 100% of teeth pairs, 

geared simultaneously, compared to 5-7% in classical gears) provides increased 

bearing capacity and kinematical accuracy, small dimensions and mass. In 

addition to the above said, extended kinematical possibilities (±8÷3600 compared 

to 79÷300 in sinusoidal gears), reduced acoustic emission and solution of all 

technological problems, as advantages, open new perspectives for precessional 

planetary gears utilisation in various fields of mechanical engineering (B o s t a n, 

1991). On the whole, processional planetary gears can be divided in two basic 

groups: 

- power processional planetary gears; 

- kinematical processional planetary gears.

As the planetary processional 

transmission is the new transmission, in 

the beginning of effectuation the 

simulation and calculation with 

programs CAE it is necessary to 

calculate theoretically base parameters. 

On fig. 1 is shown kinematical scheme 

of planetary processional transmission, 

but on fig. 2 is shown the design of 

planetary processional transmission. The 

main elements of transmission make: 

crankshaft 1, block satellite 2, fixed 

wheel gear 3 and mobile wheel gear 4. 

The important point this that, having 

such design (single-stage) is possible to 

receive transfer rate up to 3600, using 


designs of double-stage it is possible to receive transfer rate up to 14 million. 

Obviously that, having such big transfer rate, there are big loadings on a tooth. 

This problem is solved by that, all teeth it is participate in the gearing, and 

loading is transferred by half from them. Meaning that, all teeth are in gearing at 

the big loadings, the kinematic error is small. 

Fig. 1 – Kinematical scheme of 

planetary processional transmission. 

Principle of functioning of the planetary processional transmission the 

following: crankshaft 1 with inclined section to specify a block satellite 2 

spatially spherical movement, block satellite by means of roller crown 

interaction with fixed wheel gear 3 and mobile wheel gear 4 (having a special 

structure generated by means of the equations), in turn a mobile wheel gear 4 it 

is rigidly connected with output shaft of a reducer, transfers the moment and 

speed of rotation. The direction of rotation of an input shaft and output shaft can 

be in one or in different directions. It is visible and in calculation of transfer rate 

if it has positive number we have an identical rotation. The transfer rate of a 

planetary processional transmission is defined by the relation (1). 

1 3 6 2 7 4 5 

Fig. 2 – Design of planetary processional transmission. 

(1) 

Zg Z 

1 a 

i =− 

, 

ZbZg − Zg Za 

2 1 

where: 1 , 2 are 

Z g Z g 

number of a roller the 

crown satellite g1 

and g2 

; 

Z a and Zb 

is represented 

number of a teethes the 

cog-wheels a and b .


3. Calculation by CAE Simulation of Kinetostatics 

Parameters of Processional Transmission 

3.1. 3D Model elaboration of Planetary Processional Transmission 

Calculation of planetary processional transmission by a simulation is carried 

out using the simplified 3D model created in program Motion Inventor 2004+ 

but (4) from which it is possible to determine and check up dynamic loadings in 

bearings which have been designed earlier. 

Dynamic processes in planetary processional transmission derive, to a great 

extent, from the interaction of conical rollers of the satellite crowns with 

generating surfaces of central wheel teeth. The bearing capacity defined by gear 

forces (static and dynamic), the noise emission and the transmission 

vibroactivity, depend on the gear dynamic processes on the whole. With 

account of these important factors in the elaboration of 3D model of the initial 

processional gearing, the linear contour of central wheel teeth profile (fig. 3, a) 

were designed applying parametric equations: 

(2) 

Where: 

X = k Z + d ; 

Y k Z d ; 

(k d 

= 

k d ) 

− 

m m m m 

1E 2 1E 2 

m 

1E = 

m 

1 

m 

1E − 

m 

1 

m 

Z1E 

m m m m 

1 1 − 2 2 

m2 m2 

k1 + k2 + 1 

(k d − k d ) + (k + k + 1) ⋅(R −d−d ) 

Z , 

m m m m 2 m2 m2 2 m2 m2 

m 

1E = 

1 1 2 2 1 2 

m2 m2 

k1 + k2 + 1 

D 1 2 

⎛ ⎞ 

• • • 

m m m m m m 2 m 

X1D⎜X1D X1D+ Y1DY1D⎟ + Z1D X1D 

+ 

m m 

1 = 

⎝ ⎠ 

2 = − 

• • 

m 

m⎛ m m m m ⎞ 

X1D 

Z1DX1DY1D−Y1D X1D 

k ; k 

⎜ ⎟ 

⎝ ⎠ 

2 m m 

( Rcos D β + dY 1 1D) 

• 

2 m 

m Rcos D β X1D 

1 = • • 

⎛ m m m m⎞ X1DY1D− X1DY1D m 

2 = 

m 

X1D 

d ; d . 

⎜ ⎟ 

⎝ ⎠ 

m m m ( kY 1 1D Z1D) 

The 3D model of the central wheel (fig. 3, b) was designed by using CAD 

Autodesk Inventor (5) (B o s t a n et al, 2007). 

;


a. b. 

Fig. 3 – 3D linear contour of the central wheel teeth (a), 

and the model 3D of the central wheel (b). 

The 3D model for calculation is shown on fig. 4, it includes a crankshaft 1, 

a fixed wheel gear 2, the block satellite 3, a mobile wheel gear 4 rigidly 

connected with output shaft 5. 

Fig. 4 – 3D Model of planetary processional transmission. 

The dynamic model has been created on the basis of rigid model. As the 

initial data has been specified speed on input shaft [deg/sec] and the moment of 

torsion on the output shaft [Nm]. 

Kinematic joints fig. 5, have been enclosed according to movements in 

gearing. The crankshaft and the case of a reducer by means of cylindrical roller 

bearings are connected by a cylindrical joint to an opportunity of a selfcentering. 

A crankshaft and the block satellite by means of two tapered bearings 

with pair rotation in points of the appendix of loading on bearings (point-line), 

pair rotation (revolution) rollers on an axis of the block satellite and 3D contact 

of rollers to a mobile and fixed wheel gear.


Fig. 5 – Kinematics joints in the planetary precessional transmission. 

3.2. Precessional Transmission Kinetostatic Simulation 

The simulation of model processional gearings has been created in some 

stage. At the first stage has been executed the kinematic analysis, from 

definition of the following parameters: transfer ratio, absolute angular speed of 

the block satellite, relative angular speed of the block satellite, angular speed on 

the output shaft. The received results are shown on Fig. 6. 

Fig. 6 – Kinematic calculation of planetary processional transmission. 

At the following stage has been executed the kinetostatics analysis with 

calculation and simulation of total loadings in gearings.


An important advantage of the processional planetary transmission consists 

in the insurance of absolute multiplicity (ε=100%) of the central wheel teeth 

and pinion simultaneous gearing. This advantage provides small dimensions 

and mass (reduced material consumption) and small cost of the final product. 

a. b. 

Fig. 7 – Normal forces among the teeth and 

their distribution among simultaneously geared teeth. 

The vertical line on the diagram (Fig. 7) shows the precession phase 

reported symbolically to the gearing time of the pinion tooth with the central 

wheel tooth (as reference the time was taken, because as CAE input parameters 

the angular ω velocity was introduced). On the diagrams one can see the 

distribution of load among the teeth which corresponds to the respective 

precession phase (blue colour – distribution of load on the active profile of 

central wheel tooth, red colour – per one precession cycle). It is possible to state 

from the diagrams that despite the precession phase, the load among teeth is 

uniform. 

Due to load transmission from the input shaft to the output shaft by a big 

number of teeth couples geared simultaneously ⎛ (Z − 

⎜ = 4 1) 

ε 

⎞ 

⎝ 2 ⎟ the normal 

⎠ 

forces at teeth contact are much smaller as in the classical transmissions with in 

volute transmission. 


1. Taking into account the fact that in processional planetary gearings the 

(Z4 − 1) 

teeth couples transmit the load simultaneously we can conclude that 

2 

the bearing capacity of the processional gear is much bigger than that of the 

classical in volute gear (in which only 5...7 % of wheel teeth gear 

simultaneously).


2. The elaboration of the planetary processional transmission dynamic 

sample (CAE) based on developed 3D model allows verification of kineticstatic 

parameters previously defined. 

Received: January, 15, 2010 Technical University of Moldova 

Department of Theory of Mechanisms and Machine Parts 

e-mail: dulgheru@mail.utm.md 


1. B o s t a n I. Precessionnye eredaci s Mnogoparnym Zacepleniem.(Ed.) Ştiinţa, 356p. 

1991. ISBN 5-376-01005-8, Chişinău. 

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 

harmonic transmissions, Bucureşti - Chişinău, 1997, 198 p. 

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 

Analysis of the Sphere - Spatial Mechanisms by using 3D-Models and - 

Simulations. Meridian Ingineresc, Revue of the Technical University of Moldova 

and Moldavian Society of Engineers, 2, ISSN 1683-853-X, (2004). pp. 59-61. 

4. http://www.solid-dynamics.com (Motion Inventor) 

5. www.autodesk.com (Autodesk Inventor). 

DEZVOLTAREA INTEGRATĂ CAE A TRANSMISIILOR PRECESIONALE 

UTILIZÂND PLATFORMA AUTODESK INVENTOR 

(Rezumat) 

În lucrare se prezintă modelarea şi simularea transmisiilor planetare precesionale 

proiectate în două variante cu raport de transmitere mare şi capacitate portantă ridicată 

într-o singură treaptă. Construcţiile transmisiilor precesionale au fost proiectate în softul 

Inventor şi simulate ca sisteme multicorp în MotionInventor. Simulările transmisiilor au 

permis obţinerea informaţiilor privind poziţia, vitezele, acceleraţiile, traiectoriile 

punctelor, forţe şi momente, energii, de asemenea, forţele de contact la contactul dintre 

dinţii roţilor dinţate şi ai satelitului şi alte date privind sistemul. Angrenajele a două 

transmisii au fost modelate ca sisteme multicorp în SolidDynamics pentru studiul 

emisiei de zgomot în timpul lucrului prin simulare, cu posibilitatea extinderii la 

contactul dintre dinţi.


Publicat de 



Secţia 


PROPOSALS FOR THE IMPROVEMENT OF 

THE FLUIDIC SPIRAL JETMILLS' ACTIVITY 

BY 

ILEANA FULGA 1 and EUGEN STRAJESCU 2 

Abstract. In this paper is presented a documentary study about the particle 

size reduction. There are presented the authors' realizations concerning the 

optimization of the spiral jetmills produced in the APTM Enterprises, based on a 

new mathematical and experimental model of the flows, developed by the authors. 

There are presented conclusions about the research directions to be attacked in the 

future. 

Key words: micronization, spiral jetmills, fluidic mills, optimization. 


The fluidic mills are the single equipments that permit the dry grinding and 

the get of an ultra fine granulometry (1-25 μm), indifferent by the product 

resistance and by the hardness of the product's crystals. 

The grinding is produced by the collision between the particles that are 

accelerated at very high speeds in flows and gas jets, or by their impact with 

solid surfaces (fixed or mobile) from the interior of the milling room. 

We consider important to signalize that the real or as a rough guide speeds of 

the entrainment and grinding particles' jets are considered secret and can not be 

found in the literature. At this moment exist in the world 7 fabricants that 

produce fluidic mills. In fact, there are many societies that commercialize 

fluidic mills, but they are procuring the equipments offered from one of the 

mentioned fabricants. 

All the fluidic mills fabricants consider strictly confidential any data 

concerning the sizing and the calculus of these equipments from the point of 

view of the fluidic mechanics, refusing to give any theoretical or experimental 

value concerning the geometry of the fluidic nozzle the jets' speeds and 

particles' speeds or values bound of the fluidic grinding process.

144 Ileana Fulga and Eugen Strajescu 

2. Clarifications about the Possibilities to improve the Spiral Jetmills' 

Activity, resulted from the Mathematical Model 

2.1. The Supersonic Jet's Protection 

The construction of the Venturi tubes by where the material is introduced 

and of the nozzles from where the compressed air is blowing off represents an 

important factor for a good activity of the fluidic spiral jetmill. The super sonic 

nozzle (the one by where is insufflated fluid under pressure, realizing important 

debits) must have a determined length. The prolongation of the active area of 

the nozzle behind the peripheral wall, or nozzles with inclined termination, and 

not perpendicular on the axis's direction of the nozzle could realize a 

compressibility effect manifested in an opposite sense, because the boost 

pressure decrease, the circulation is reduced and the jet is displaced in a 

direction more nigh to the tangent. 

Of course this kind of position must be considered as "off-set" beside the 

desired equilibrium position, considered as "ideal". It could be imposed in that 

case the "protection" of the protuberant peak of the nozzle by an aerodynamic 

modeling of the peripheral wall of the milling room, or by the practice of some 

clefts on the same wall. 

2.2. The Geometry of the Supersonic Jet 

It is proved that the employment of the convergent - divergent nozzles, at 

necessary and effective pressures do not present great advantages in the great 

majority of the practical applications. 

It is recommended for the reduction of the losing, the construction of jets as 

short as possible, with an alignment near the perfection, with the same radial 

angle. Because deviations from the alignment between different jets although 

minima, create important tourbillions, unequal, in the transversal flux, 

discontinuities in the peripheral circulation of the particles and tourbillions in 

the central classification vortex. 

3. Practical Applications 

3.1. The Optimization of the StarJet and CosmoJet Mills 

The APTM Enterprise realized a great number of researches in order to be 

able to optimize the activity of the produced spiral jetmills, presented in a great 

number of scientific papers [1] - [10] and in a doctoral dissertation.


Fig. 1 – The analyzis of the evolution of the APTM spiral jetmills, 

considering the application of the process' optimization. 

The results presented in this doctoral dissertation were transposed in


Fig. 2 – Example of the optimized technological flux for the fabrication of the 

cosmetic micronized powders (make-up, cheek powder). 

practice in the enterprise APTM, for a new jetmill family, named "StarJet- 

Millennium" (Fig. 1). An optimized technological process is shown in Fig. 3)


The first three typo dimension of this family (4" - 8" and 12") applied an 

important number of the optimization algorithms of the milling process that 

form the research object of this research. It results implicitly the fact that these 

algorithms (fig. 2) where already verified in practice for moreover micronized 

powders categories (that are partly confidential). 

Fig. 3 – The example of the spiral jetmills' optimization algorithm applied for 

the definition of the parameters recomended in industry and for the qualitative 

differentiation of the cosmethic powders.


The effect of the micronization about the dispersed pigments can be 

observed the besting the case of the sample of Prussian Blue (fig. 4): being in a 

non micronized stadium this pigment presents agglomerates at 100-150 μm, 

whiles after the micronization it is to observe a granulometry contained 

between 20-25 μm. 

Fig. 4 – Pigment (Prusia blue) nonmicronizes = 150 μm (left) and 

micronized = 20 μm (right). 

Fig. 5 – Example of cosmethic powders micronized with optimization 

algorithms; visualization on microscope with phase contrast. 

Even without the microscope or another magnifier, from the photos' 

observation it is to note the intensification of the chromatic coverage of the 

pigment (fig. 2), because after the micronization, the specific surface of the 

product is majorized for 2000 - 3000 times, and particles are very uniform from 

the point of view of the granulometry. 

Another object of the optimization is the getting of a micronization with a 

plain classification room, in order to not charge to frequent the mill (fig. 6).


Fig. 6 – Example for an optimized production for 

the filling up of the classificatory. 

We can observe the fact that after the micronization, all the materials 

are grinded at granulometry about 15-30 μm, with a remarkable 

uniformity. We make the assignation that in this paper where presented 

only the non confidential elements that show the optimization. 

We present (fig. 5) an example of a technological process for the getting of 

cosmetic powders micronized with optimization algorithms. In fig. 6 and 7 we 

present micronized powders realized with optimization algorithms 

(visualization on microscope with phase contrast). We can affirm that the


micronized powders are an especial good quality, representing a revolution in 

the cosmetic and pharmaceutical industries. 

Fig. 6 – Non micronized cosmethic pigments, 

with a granulometry of 120 - 350 μm. 

Fig. 7 – Micronized cosmethic pigments, with a granulometry of 15 - 30 μm. 

Mills from the family StarJet optimized with algorithms similar with the 

presented algorithms were delivered and perfectly work from the year 2002, in 

many enterprises for decorative cosmetic products of the societies Avon (USA), 

Natura (Brazilia), Belcorp (Columbia), EverBilena (Taiwan), Swan Cosmetics 

(Mexic), Cosmetica (Canada), GammaCroma (Italia), Hellenica (Grecia), 

GiPiccos (Italia), etc. 

The up presented algorithms, partially used in production, ask for verifications 

and ulterior confirmations, forming the object of some future studies. 


1. The research from what doctoral dissertation introduced an original threedimensional 

model of the fluidic running in the spiral jetmills that describes 

very veridical the complex jet in this kind of mills. 

2. The results are obtained as suit of a very large gamut of theoretical and 

experimental researches for the determination of the jets' speed. 

3. The optimization algorithms were verified on models and in realty on 37 

mills having different typo dimensions, in many fields (cosmetic, 

pharmaceutics, and metallurgies) in above 20 countries.


4. From the experimental researches it was obtained important conclusions 

that permit to optimize the spiral jetmills in at least two directions. 

5. The first one is the micronization operation conditions and the best 

geometry of the mill. 

6. A second one is the optimization of the operation conditions in order to 

realize the filling of the classification part of the mill. In this case, the 

productivity of the mill increases with important values. 

7. The researches for the optimization in the case of the spiral jet mills must 

go on,, because we appreciate that it exists a lot of unexplored ways and 

possibilities. 

Acknowledgements. The authors would like to thank APTM Enterprises for the 

support and constructive comments and also to the staff of the Laboratory of APTM. 

Received: March 20, 2010 1 APTM Enterprises, 

Lugano, Swiss 

e-mail: i@fulga.com 

2 POLITEHNICA University of Bucharest, 

Department of Machines and Production Systems 

Bucharest, Romania 

e-mail: eugen_strajescu@yahoo.com 


1. Fulga I., Huddleston K., Cosmetics cross an even finer line, Powder 

Reporter, vol. 1, UK, 1998. 

2. Fulga I., Stră jescu E., Sandu I. Gh., Experimental Model of the fluidic 

Mills with Spiral Jets, Academic Journal of Manufacturing Engineering, Editura 

Politehnica, Timişoara, România, 53 - 57, ISSN 1583 – 7904 (2006). 

3. Fulga I., Stră jescu E., Theoretical and Experimental Contributions 

Concerning the Logistics of the Research Systems of the Fluidic Mills with Spiral 

Jets, Buletinul Institutului Politehnic, Iaşi, România, Construcţii de Maşini, LII 

(LVII) , 5, 313 - 320, ISSN 1011 - 2855 (2007). 

4. S t r ă jescu E., Fulga I., The Visualization of the Fluidic Fluxes in the Mills 

with Jets, Industrial Engineering Journal RECENT, 8, 3b (21b), 361 - 366, ISSN 

1582 - 0246. Braşov, România (2007).


5. F u l g a I . , S t r ă jescu E., Contributions to the Fluid Dynamics in Jetmills by a 

New Aerodinamic Model of Supersonic Flow, Buletinul Ştiinţific U.P.B., Serie 

D, 74, 1, ISSN 1454 - 2358, România (2008). 

6. Fulga I., Gadonna J. P., Remolue M., Froidevaux L., Ajana 

A., Chayvialle L., Improvements in MDI technology with total process 

containment, European Aerosol Conference, EAC, Belgium, 2005. 

7. F u l g a I . , S t r ă jescu E., Probleme actuale şi de perspectivå în construcţia 

utilajelor pentru măcinare şi în tehno-logia măcinării, A XI-a Conferinţă 

ştiinţifică cu participare internaţională TEHNOMUS XIII, Universitatea "Ştefan 

cel Mare" Suceava, România, Facultatea de Inginerie Mecanică, pg. 94-103, 

ISBN 973-666-154-7, 2005. 

8. F u l g a I . , S t r ă jescu E. - The Present Stage and the Perspectives of Researches 

in the Area of Fluid Grinding Operation Equipment with Spiral Jets, 

International Conference on Manufacturing Systems, Buletinul Institutului 

Politehnic Iaşi, România, secţia Construcţii de maşini, LI (LV), 5, ISSN 1582- 

6392 (2005). 

9. F u l g a I . , S t r ă jescu E. - Contributions Concerning the Perfecting of the 

Crushing Technologies of the Cosmetic Particles and of the Metallic Carbides in 

the Wheels with Fluid Jets, International Conference “Technologies and Quality 

for Sustained Developement”, pg. 357 - 360, ISBN 973-720-035-7, 2006. 

10. F u l g a I . , S t r ă jescu E. - Experimental Model of the fluidic Mills with Spiral 

Jets. (Long Abstract), The 3rd International Conference on Manufacturing 

Science and Education, MSE, Sibiu, România, pg. 79 - 80, ISSN 1843 - 2522, 

2007. 

PROPUNERI DE OPTIMIZARE A FUNCȚIONĂRII 

MORILOR FLUIDICE CU JETURI ÎN SPIRALĂ 

(Rezumat) 

În lucrare se prezintă succint concluziile unui studiu anterior asupra realizării unui 

model matematic privind comportarea jeturilor supersonice ale morilor fluidice cu jeturi 

în spirală. Pe baza lor ,se enunţă unele posibilităţi de optimizare a funcţionării acestor 

tipuri de mori, precum şi rezultate comparative ale măcinării unor pulberi cosmetice. 

Rezultă din lucrare şi faptul că micronizarea produselor simplifică substanţial 

tehnologiile de măcinare, ceea ce micşorează costurile producerii pulberilor de orice fel.


Publicat de 



Secţia 


DYNAMIC ANALYSIS OF MECHANICAL-HYDRAULIC 

PROTECTION MECHANISMS OF LOW POWER HORIZONTAL 

AXIS WIND TURBINES THROUGH VERTICAL TILTING 

BY 

DORU CĂLĂRAŞU, IRINA TIŢA, DAN SCURTU 

and BOGDAN CIOBANU 

Abstract. Generally, the control system a wind turbine is equipped with is 

designed to reduce dynamic load on its blades, to increase turbine reliability and 

operational safety. The paper present the dynamic analysis of a mechanicalhydraulic 

protection mechanism used for low power HAWT. The method consists 

of the vertical tilting of the wind turbine. The analysis of the system responses 

revealed a good dynamic performance when using the vertical tilting solution 

chosen to protect low power HAWT. 

Key words: HAWT, protection mechanism, vertical tilting. 


The control system a wind turbine is equipped with is designed to reduce 

dynamic load on its blades, to increase turbine reliability and operational safety. 

This system is one of the most important developments meant to consistently 

improve low power horizontal axis wind turbines. 

Further to the analysis of control and protection system design models used 

for low power horizontal axis wind turbines (HAWT), the best constructive 

solution was chosen, which consists of the vertical tilting of the wind turbine. 

The tilting begins when the wind speed exceeds the speed limit beyond which 

the turbine safety may be jeopardized. This solution meets the system 

requirements: it is reliable, it does not include electrical drive components, it 

has no built-in electronic devices and it does not require the existence of a 

hydraulic drive system.

154 Doru Calarasu et al. 

2. Constructive-Operational Description 

A wind power plant turns wind energy into mechanical power. The chosen 

solution refers to very low power (500 – 1000 W) wind turbines. 

When the wind speed does not exceed a certain speed limit, the rotor 

assembly does not tilt. The propeller rotates at an angular speed, which depends 

on the wind speed and on the mechanical output load. 

When the wind speed reaches or exceeds the speed limit, the rotor assembly 

tilts and the propeller changes its rotation plane. 

HAWT 

Damping 

device 

Protection 

mechanism 

Tilting 

device 

Fig. 1 – Constructive block diagram. 

Balancing 

device 

Thanks to its components (Fig. 1), this mechanical-hydraulic protection 

mechanism also ensures the operation of the wind turbine at wind speeds higher 

than the rating speed, having the following functions: 

a) the rotor assembly is directed towards the wind stream using the 

mechanism’s lower bearing, by means of which the assembly is placed on the 

support pillar; 

b) the tilting device becomes active when the wind speed exceeds the safety 

limit and it reduces the active turbine surface exposed to the wind; 

c) the balancing device ensures stable balance to the wind turbine rotor at wind 

speeds lower than the safety limit and unstable balance when the wind speed 

reaches the safety limit; 

d) the damping device protects the plant against wind gusts by delaying the 

propeller’s resuming its initial position. 

3. Digital Simulation Tests performed on the Conceptual Model of 

Mechanical-Hydraulic Protection Mechanisms 

The Matlab-Simulink programming environment was used for the 

performance of the digital simulations.


Fig. 2 shows the Simulink block diagram of the vertical tilting protection 

mechanism. 

Body1 

CS2 CS1 

Scope 2 

F 

Signal 1 

Signal 2 

Signal 3 

Signal Builder 1 

Revolute 1 

16 

lambda /\r 

B 

Body Actuator 

Joint Initial Condition 

Body 

CS1 

F 

Weld 

Product 3 

Product 4 

Product 5 

CS4 

CS1 

CS2 

B 

Link 2 

Revolute 3 

B 

F 

Ground 1 

Product 2 

Scope 1 

CS5 

CS3 

F 

Weld2 

Joint Spring & Damper Joint Sensor 2 

AxRo/2 

[1x3] 

B 

Env 

Machine 

Ground 2 

Environment 

[A] 

Goto 

Product 7 

Joint Actuator 

Body2 

CS1 

2 

Constant 1 

0.3 

Constant 4 

Integrator 

1 

s 

Product 

360 

Constant 

|u| 

Abs 

Constant 2 

0 

Body Actuator1 

sin 

Trigonometric 

Function 1 

-C- 

l 

-K- 

Gain 

[A] 

From 

Scope 4 

Scope 6 

Scope 5 

cos 

Trigonometric 

Function 

Scope 

Product 1 

-C- 

3r Constant 3 

65 

Scope 3 

1 

u 

Math 

Function 

Fig. 2 – The Simulink block diagram of the vertical tilting protection mechanism. 

Fig. 3 – The rotor’s surface exposed to wind depending 

on the turbine axis’ angle of inclination. 

Product 6


The mechanical-hydraulic protection mechanism includes the tilting device, 

the balancing device of the wind turbine’s own weight and the two-way 

movement damping device. The initial position (meaning a normal vertical 

turbine operation) is triggered by a back stop. The model allows testing using a 

set of signals to simulate the wind speed evolution. 

Through vertical tilting, the rotor’s surface exposed to wind decreases as 

shown in Fig. 3. 

The arm of the component determining the tilting moment increases with 

the increase of the turbine axis’ tilting angle (Fig. 4). Fig. 5 shows the variation 

of the moment determined by the rotor’s tilting strength. 

Fig. 4 – Variation of the rotor tilting strength arm. 

Fig. 5 – Moment of the rotor tilting strength.


Fig. 6 shows the set of signals used to simulate wind speed variation. 

Fig. 6 – Signals for model testing. 

The second signal is employed to achieve static balance, where the wind 

speed is thought to have the same value as the maximum admissible rating 

speed. 

Using the ramp signal (signal 3), the system response is tested, that is the 

rotor tilting, when the wind speed exceeds the maximum admissible rating 

speed. Fig. 7 shows the axis inclination angle variation at the 3 rd signal. 

Fig. 7 – Axis inclination angle at wind Fig. 8 – Tilting strength at wind 

ramp speed. ramp speed. 

Figure 8 shows the tilting strength variation in the conditions described 

above.


Fig. 9 – Inclination angle of the turbine Fig. 10 – Rotor tilting strength. 

rotor axis. 

If the system constructive-operational parameters are changed (the damping 

coefficient and the return strength decrease), figure 9 shows the evolution of the 

inclination angle of the wind turbine rotor axis, while figure 10 depicts the 

evolution of the tilting strength. 


1. The simulation diagram allows both constructive and operational system 

parameters and operation testing signals to be altered. 

2. Digital simulation may be used to optimize constructive parameters 

(balancing components weight, rotor axis section lengths, balancing weight, 

damping system characteristics). 

3. The analysis of the system responses revealed a good dynamic 

performance when using the vertical tilting solution chosen to protect low 

power horizontal axis wind turbines. 

Acknowledgements. This work has been supported by the Romanian National Fund 

for Science and Research through National Centre for Programme Management under 

contract No. 21-047/2007. 

Received: April 16, 2010 “Gheorghe Asachi” Technical University, 

Department of Fluid Mechanics 

Iasi, Romania 

e-mail: dorucalarasu@yahoo.com



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 

of wind energy conversion systems, Energy Conversion and Management, 43, 16, 

November 2002, pp. 2175-2187 

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 

Systems, Principles, Modelling and Gain Scheduling Design. Springer Verlag, 

Heidelberg, 2007. 

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 

variable-speed WECS above rated wind speed. Renewable Energy 

29(11), 1911–1922, 2004. 

4. G r i m b l e M. J., J o h n s o n M. A., Advances in industrial control, Springer – 

Verlag London Limited 

5. M o l e n a a r D. P., Cost-effective design and operation of variable speed wind 

turbines, Delft University Press, 2003. 

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 

control of variable-speed wind-turbine drive-system dynamics. IEEE Control 

Systems, pp. 28 - 38, August, 1995. 

7. * * * 160 kW Wind turbine with Hydraulic Transmission at Schiedam Ref: 16013 or 

34127 

8. * * * Low-wind turbine with hydraulic blade control. Measuring program included 

Ref: 8700106 

9. * * * Wind turbine with hydraulic transmission, Patent number WO03098037. 

10. * * * Water current turbine, Patent number WO2005103484. 

11. * * * MATLAB Simulink http://www.mathworks.com/products/simulink/ 

12. * * * Contract no. 21-047/2007 financed by National Centre for Programme 

Management from Romania. 

ANALIZA DINAMICĂ A MECANISMULUI MECANO-HIDRAULIC DE 

PROTECŢIE A TURBINELOR EOLIENE CU AX ORIZONTAL DE MICĂ PUTERE 

PRIN BASCULARE ÎN PLAN VERTICAL 

(Rezumat) 

Sistemul de control cu care este prevăzută o turbină eoliană are rolul de a diminua 

încărcările dinamice pe pale, de a creşte fiabilitatea turbinei şi siguranţa în exploatare. 

Acest sistem reprezintă una dintre direcţiile prioritare de acţiune în vederea 

rentabilizării turbinelor eoliene cu ax orizontal de mică putere. 

Lucrarea prezintă analiza dinamică a unui mecanism mecano-hidraulic de protecţie 

a turbinelor eoliene cu ax orizontal (TEAO) de mică putere prin bascularea turbinei 

eoliene în plan vertical. Bascularea începe atunci când viteza vântului depăşeşte viteza


considerată limită pentru siguranţa turbinei. Această soluţie întruneşte caracteristicile 

solicitate sistemului: este fiabil, nu conţine elemente de acţionare electrică, nu conţine 

electronică înglobată, nu presupune existenţa unei transmisii hidraulice. 

Pentru viteze ale vântului mai mici de o limită maximă impusă, ansamblul rotor nu 

basculează. Elicea se roteşte cu o viteză unghiulară care depinde de viteza vântului şi de 

încărcarea mecanică la ieşire. Pentru viteze mai mari sau egale cu cea maximă impusă, 

ansamblul rotor basculează iar elicea îşi modifică planul de rotaţie. 

Testarea prin simulări numerice a modelului conceptual de mecanism mecanohidraulic 

protecţie a TEAO de mică putere prin basculare în plan vertical s-a efectuat 

utilizând mediul de programare Matlab-Simulink. 

Din analiza răspunsurilor sistemului rezultă performanţe dinamice bune pentru 

soluţia basculării în plan vertical utilizată în scopul protecţiei turbinelor eoliene cu ax 

orizontal de mică putere.


Publicat de 



Secţia 


RESEARCH ON HYDRAULICALLY SYSTEMS WHICH 

MOVE HEAVY MASSES ON SMALL DISTANCES 

WITH LOWER FREQUENCIES 

BY 

CONSTANTIN CHIRIȚĂ, ADRIAN HANGANU 

and DANIEL CALFA 

Abstract. In this paper the authors try to present the mathematical models, 

static and dynamic, useful for designing of hydraulically systems which move 

heavy masses on small distances with lover frequencies. In the last time 

mathematical modeling represents an obligatory step in the designing process of 

the new products. In this paper are presented the results of the theoretical 

simulation in concordance with the experimental results. The aim is to show the 

realistic part of the theoretical simulation of one system designed to move a heavy 

owen, 200 t, on a short distances, 50 mm, with low frequency, one complete cycle 

in 24 hours. 

Key words: research, modeling, simulation, data acquisition. 


To move heavy masses on small distances with lover frequencies the author 

recommends one of the following hydraulic charts, see Fig. 1. 

In both variants the working tool is the linear hydraulic engine CH, with the 

active surface S and volume Vo, which moves the mass M. 

For the first variant the step of the move is programmed at the level of the 

pump, through the volume/course V. We will admit that for a complete course 

the pump send to the hydraulic engine the oil volume V. The oil rich the pump 

through the orifices A and B. 

In the case of the second variant the flow regulator RD controlled the oil 

flow, independent to the working pressure for a specific time, when the 

distributing valve is on. The quantity of oil which enters in to the linear 

hydraulic engine is controlled in the time by the time stepping system C (t).

162 Constantin Chirita et al. 

a) b) 

Fig. 1 – Hydraulic charts for lower frequencies and small flow generators. 

The mathematical models presented in the following rows are equal satisfied 

for both variants. 

2. Mathematical Models for the Stationary Regime 

For the engine CH which works on the strength F, when a small quantity of 

oil V, ideal liquid, incompressible, theoretical we obtain a theoretical movement 

of the mass, xT, with the mathematical expression: 

(1) 

V 

xT = . 

S 

For a real liquid, with the elasticity module E, we obtain a real movement, in 

static regime, xP. 

⎛ p ⎞ 

(2) xR = xT 

⋅⎜1 

− ⎟ . 

⎝ E ⎠ 

(3) 

In the relation (2) p is the balance pressure. 

F 

p = . 

S 

For V = 8 cm 3 , S = 80 cm 2 , F= 32.000 N, E = 1,5e9 N/m 2 , the results are: 

xT= 1 mm, xR= 0,999 mm.

(4) 


3. Dynamic Mathematical Models 

For both systems from Fig. 1 we can use the following model: 

V0 

Q = S ⋅ v + a ⋅ p + ⋅ 

E 

dp 

dt 

dv 

(5) M ⋅ + b ⋅ v + F = p ⋅ S . 

dt 

(6) V = ∫Q ⋅ dt . 

(7) ∫ ⋅ = dt v . 

x RD 

In the last relations we have the following supplementary parameters: v- 

instant speed of the CH engine, a- linear coefficient of the lost oil which are in 

direct ratio with pressure, t-time, xRD-real movement in dynamic regime. 

To find the real movement, in dynamic regime, we must simulate the 

equation on a computer program. We consider the same data like in the case of 

the static regime, in advance we know M = 200 kg and the values of a and b 

coefficients. When the V quantity of oil is inserted we presume that the system 

is in balance. The time which the system is in charge with the V oil quantity is 

established as well as we obtain a movement theoretical equal with that we have 

on the previous step (t = 2 s). The simulating chart in the simplified way is 

presented in Fig. 2. 

Fig. 2.a – The simulating chart. 

.


The V part generates the oil flow which is introduced in the linear hydraulic 

oil CH. We can look on the evolution of the following parameters: pressure p, 

speed v, movement x. 

We can observe that the speed is stable around the value of 0,003 mm/ min 

and the final value of XRD is 0,995 mm. 

Fig. 3 – Data acquisition application in Lab View. 

The feedback time of the system is very short approximately 0,003 s, as we 

can see in Fig. 2. After the collecting data procedure from the experiment, 

where was used a LabVIEW interface with NI DAQ 6024, for the speed and 

movement we obtain the following characteristics, see Figs. 3, 4 and 5. 

Fig. 4 – Piston movement after a period of nr. of steps generated by the pump.


Fig. 5 – The real speed of the piston in mm/minutes. 

Regarding the system is permanent under the pressure his dynamics is very 

well. The system is a stabile one first order system. 

The dynamic simulation, are we expected, is the most precise one, regarding 

the fact that he take care of the specific oil gaps. 


1. Regarding the system is permanent under the pressure his dynamics is 

very well. The system is a stabile one first order system. 

2. The aim is to show the realistic part of the theoretical simulation of one 

system designed to move a heavy owen, 200 t, on a short distances, 50 mm, 

with low frequency, one complete cycle in 24 hours for Mittal Steel - Galati. 

Received: February 25, 2010 ”Gheorghe Asachi” Technical University, 

DISAHP Department 

Iasi, Romania 

e-mail: disahp@gmail.com



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 

translaţie pentru cuptor rotativ de calcinare. Propunere de brevet de invenţie, nr. 

A00317/ 10.05, 2007. 

2. C h i r i ţ ă, C., H a n g a n u, A.-C., Modelation and Simulation of Hydraulically 

Systems which move Heavy Masses on Small distances with Lower Frequencies, 

International Scientific-Technical Conference HYDRAULIC AND 

PNEUMATICS 2007, Wroclaw, Poland ,October, 10-12, 2007, pp. 404-408. 

3. P r o d a n D., Hidraulica maşinilor-unelte , Editura Printech, Bucureşti, 2004. 

4. P r. o d a n D., Maşini-unelte, modelarea şi simularea elementelor şi sistemelor 

hidrostatice, Editura Printech, București, 2006, pp. 210-218. 

5. * * * LabVIEW 7.2, 2006. 

6. * * * http://www.hydramold.com. 

7. * * * http://www.ni.com. 

CERCETĂRI PRIVIND SISTEMELE HIDRAULICE PENTRU DEPLASAREA 

MASELOR MARI PE DISTANTE MICI, CU FRECVENŢĂ REDUSĂ 

(Rezumat) 

În lucrare autorii prezintă cercetările efectuate în cadrul DISAHP pentru proiectarea 

sistemelor hidraulice care deplasează mase grele pe distante mici, cu frecvenţă redusă. 

Sunt evidențiate rezultatele simulării în concordanţă cu rezultatele experimentale la 

deplasarea unei sarcini grele de 200 t, pe distanţe scurtă 50 mm, cu frecvenţă joasă, întrun 

un ciclu complet de 24 de ore, pentru beneficiarul Mittal Steel din Galați.


Publicat de 



Secţia 


PNEUMATIC METERING SYSTEM FOR AMOUNT 

OF WATER EXTRACTED IN CONVECTIVE 

DRYING PROCESSES 

BY 

EROL MURAD 1 , CĂTALIN DUMITRESCU 2 , 

GEORGETA HARAGA 1 and LILIANA DUMITRESCU 2 

Abstract. Increased decentralization of drying processes for agricultural 

products require reliable and easy handling installation at low costs. The 

economic efficiency of these plants depends on the use of optimal management 

for processes that require measurement of the mass of water extracted, in the 

thermal conditions of drying chambers. We developed a new scheme for 

measuring on-line the weight of tray loaded with dry material using a pneumatic 

force transducer working under sampled measurement. The solution presented is 

remarkable due to very low energy consumption and low cost automation feature. 

The measurement system is coupled to PLC for automatic management of driers. 

Dynamic behaviour and energy consumption were determined by simulation 

experiments performed with a model and a numerical simulation program, 

developed in simulation environment MEDSIMFP10. Simulation experiments 

have confirmed low energy consumption air and good measurement accuracy. 

Keywords: pneumatic transducer, force, drying, energy consumption, lowcost 

1. General considerations 

Drying is an effective, economic and ecological method for preserving 

agricultural products, plants, fruits and vegetables. An optimal drying process 

led to products with high nutritional potential, obtained with reduced specific 

energy consumption. 

For dryers with a capacity smaller than 20 m 2 surface of tray, the price of 

automatic driving system can double the cost of installation and thus leads to 

reduced automatic management functions, with impact on the quality of final 

products. These dryers can be easily transported to the place where is performed

168 Erol Murad et al. 

harvest of products which require drying; for this reason it is necessary a high 

level of energy independence and high reliability. [ 5] - [7], [9], [10]. 

For optimizing the drying process is still necessary to measure the mass of 

water extracted during the drying process. In order to reduce production costs of 

the dryer, the measurement is performed for a sample represented by one or 

more boxes of dry suspended on a force transducer. 

To simplify design, transducer should be arranged in the drying chamber at 

high temperatures and high humidity. In papers [5], [6], [9] have been examined 

these issues and therefore was developed a pneumatic force transducer, 

unconventional, which is adapted to operating conditions of convective dryers 

and requires only 63mW electrical power and 5mW power air power. [12] - [14] 

Optimal management of drying processes requires online measurement of 

the variation of water extracted from the dried bodies during a drying. 

Constructive solution for measuring the change in weight of a column of tray is 

technically complicated and expensive. Hence, it is used a weight measurement 

of the change in middle column boxes, specifying that the drying process has a 

similar evolution throughout the column height. It should be noted that the 

temperature in the drying is in the range 50...80°C, therefore, was decided to 

use pneumatic sensors for measuring forces, because measured size, pressure, is 

not influenced by changes of temperature. [4] - [6] 

Convective dryers consume more thermal power and about 8% electrical 

power, required for operating the fan and automation system. [8], [10] 

This paper presents and analyzes the operation and performance of a 

scheme for measuring on-line weight of boxes loaded with dried material using 

a pneumatic force transducer working under sampled. The solution presented is 

remarkable due to very low energy consumption and a reduced price. The 

measurement system is coupled to dedicated leadership PLC automatic dryers. 

They use basic concepts of low cost automation, minimize energy 

consumption and not least are extremely efficient economically. 

2. Pneumatic Force Transducer 

Applying concepts of minimizing energy consumption, energy 

independence and low cost automation system was designed an unconventional 

pneumatic transducer which does not require special source of compressed air 

consumption is very low, which simplifies construction, and hence the cost 

transducer, and the total energy consumed is very small. 

It was chosen a pneumatic system because of the pressure signal, that is 

proportional to the measured weight, which is not influenced by environment 

temperature of measurement, and accuracy is high. 

The main element of the force transducer is pneumatic load cell, which 

converts measured force Fmas in a force proportional pressure pmas.


Figure 1 shows a functional diagram for pneumatic load cell for the drying 

process, which can be used for proving that force change occurs slowly and 

processes take place in a single direction. 

Fig. 1 – Functional diagram of pneumatic force transducer. 

Force measuring Fmas(t) is applied on top of a rod (4) attached to the rigid 

center (3) of an embossed flexible membrane (2) with effective diameter Def 

and constant effective area Sef, mounted on a body (1). On the rod (4) is 

attached a nozzle (5), with the diameter dd, which is based on a ball (6) with 

diameter db, closing the chamber as air access to the outside through holes in the 

membrane, rigid centre and rod. 

The measurement chamber can be connected in parallel with an air capacity 

Vad. Pneumatic circuit supply is done by RP variable air resistance and DP 

distributor type 2x2. Pressure source must be pal ≥ 1,5 pmas max. 

Pressure measured pmas (t) is applied to a converter p / U, which gives the 

output voltage yF∈[1, 3] Vdc. 

The steady, the balance of forces, pressure pmas (t) in the enclosure is: 

(1) 

p 

F 

( t) 

+ G 

mas em 

mas ( t) 

= 

, 

S ef 

where: Gem is the weight of mobile equipment, which is constant. 

For measuring a force with some variation, included in the measurement, 

the supply with compressed gas is sampled with an u1 order form, short pulse, 

which provides increased pressure pmas to Fmas balance; then the rigid centre, 

membrane and nozzle amount of the ball with h(t) and excess flow gas Dev(t) is 

discharged outside. 

For measuring a force with a change in one respect, specific processes of 

drying, the weight of bodies during drying decreases continuously, it is 

necessary to supply compressed gas only at the beginning of drying burden. 

During drying Fmas decreases continuously, pmas decreases continuously, and 

additional gas is discharged outside through the space between the gasket and 

valve, to maintain balance as it shows the Eq. (1).


This method of measurement consumes a very small amount of compressed 

gas, which means little air power. 

Pneumatic output signal of the enclosure pmas is converted into unified 

electric signal using a converter P / U. 

As the outdoor temperature is constant and the variation of force Fmas(t) is 

slow, can be considered, in terms of thermodynamics, the measurement is an 

isothermal process at constant volume. Variation of the air mass in the 

enclosure depends on pressure variation pmas(t) produced by the variation of 

force Fmas(t): 

(2) 

V 

m 

ma 

( t) 

RaTa 

+ Vad 

= . 

p ( t) 

The position of dynamic equilibrium of cell phone equipment load, if the air 

distributor DP is closed, the volume of gas is constant Vm. Therefore, 

thermodynamic equilibrium equation is: 

(3) p mas ( t) 

( Vm 

+ Vad 

) = ma 

( t) 

RaTa 

, 

hence the differential equation of the dynamic behaviour of load cell: 

dp ( t) 

= 

dt 

mas 

a 

(4) ( Vm 

+ Vad 

) RaTa 

, 

and 

(5) 

mas 

a ( t) 

⎛ V V ⎞ 

⎜ m + ad dF 

= ⎟ 

dt ⎜ Sef 

RaT 

⎟ 

⎝ 

a ⎠ dt 

dm mas 

dm ( t) 

dt 

When the force produced by the membrane exceeds the loading: 

(6) ( mas ( t) 

⋅ Sef 

− Fmas 

( t) 

− Gem 

) > 0 

p , 

is entering a phase characterized by acceleration a(t) that makes lifting the 

nozzle from the ball to h(t): 

(7) a( t) 

( pmas 

( t) 

⋅ Sef 

− Fmas 

( t) 

− Gem 

) / M red ( t) 

= . 

Mred low mass value (t) at the transducer rod is: 

.


1 

(8) M red ( t) 

= ( Fmas 

( t) 

+ Gem 

) ) . 

g 

Dmev exhaust mass flow (t) depends on the distance nozzle-ball h(t), the 

pmas(t) and external pressure patm. Sev exhaust section (h, db, dd) is a truncated 

cone with the tip in the centre of the ball and the base on the nozzle [3], where 

Rh is hydraulic radius [3]: 

⎛ 2 2 2 2 ⎞ 

(9) R 

h 

= 0 . 5 ⋅ ⎜ h + h d 

b 

− d 

d 

+ 0. 

25 ⋅ d 

b 

− 0. 

5d 

b ⎟ 

⎝ 

⎠ 

The kev rate of contraction of the jet exhaust is calculated from a 

relationship F(Re, Rh, Ra, Ta, pmas, patm) [3]. 

If Fmas(t) decreases continuously, so in drying processes, change dma(t)/dt 

will be negative; over time, the gas is discharged outside for reaching a value 

pmas(t) with which is balanced Fmas(t). 

Under these conditions, mass balance for air in the enclosure is: 

dma 

( t) 

(10) Dmi 

( t) 

− Dmev 

( t) 

dt 

= . 

It was made a model and a simulation program of this type of transducer 

operation in MEDSIMFP.10 simulation environment. Figure 2 presents 

transducer operation for a sampling period starting sequence and in Figs 3 and 4 

are shown different working arrangements, with a simulation of 300 s. 

3. Measurement Algorithm 

For reducing energy consumption and air power, it was designed an 

algorithm to drive transducer operation under sampled with a period Tes ∈ (10 

... 100) s. 

Reading the transducer voltage output yg is made in PLC with a frequency 

fy ≥ 100 Hz. 

At the beginning of the sample, it is given a prompt ui = 1 which opens the 

DP and the through RP entry pass flow Dmi (t), that results in increased air mass 

ma (t) and default and pmas (t), see Figure 2. 

At the beginning of the sample, it gives a command ui = 1 which opens the 

DP and through the RP entry pass flow Dmi(t), which results in increased air 

mass ma(t) and hence the pmas (t), see Figure 2.


When the condition (6) is accomplished, the nozzle starts to rise and there is 

an exhaust flow Dmev (t)> 0, which reduces ma(t) and pmas (t) begins to decrease. 

Once it finds a stabilizing amount pmas (t), driving algorithm keeps ui = 1 still 

400 ms, then ui = 0 and stops the air supply Dmi (t) = 0. In this period Dmev(t)> 0 

until h(t) = 0. 

Fig. 2 – Measurement scheme starting sequence. 

Fig. 3 – Experiment simulation pneumatic transducer Fmas (t) = const.


Fig. 4 – Experiment simulation pneumatic transducer dFmas(t) / dt


Figure 4 presents a simulated experiment where the measured force 

decreases continuously and slowly. Transducer works well with an air 

consumption similar to the previous experiment. 

If it is considered that the opening between two orders of distributor DP, 

pressure may decrease too much due leaks and wear, hourly air consumption 

would be maxim 135.42 Ncm 3 / h, which is a small volume. For example in 

continuous operation for 24 hours would consume 3.25 Ndm 3 air, which would 

require a reservoir of 2 dm 3 uploaded once a day to 3 bar. 


1. We designed a pneumatic force transducer for particular process 

consisting in a slow change of the measured force, typical for convective 

drying. 

2. Transducer operates with very low air energy consumption, is simple, 

affordable, sustainable, low-gauge and weight. 

3. They were applied mechatronics and automation principles at low cost 

for transferring as many functions to PLC management of the technological 

process. 

4. For exhaust valve nozzle-ball version has been used to ensure normal 

operation even in difficult assembly and operation, ensuring a good seal chosen 

option and a relatively large exhaust section to small movements. 

5. On the dynamic, by controlling the very low input flow , in a chamber 

measuring small volume, for reaching a maximum speed of less than 100 mm/s 2 

for the mobile part, and a maximum displacement of 0,3 mm. These values 

show that there are no shocks during the operation, and sustainability of nozzleball 

system is very high. 

6. Simulation experiments show that a transducer with head scale of 50 N 

can operate continuously 24 hours powered by a 2-liter tank loaded to 3 bar 

once a day. 

7. The results obtained in simulated experiments represent tests of the 

experimental model, which pursue to prove both design and functional 

optimized variants functional, through change values for pair ball diameter and 

nozzle diameter. 

Received: March 20, 2010 

1 POLITEHNICA University of Bucharest, 

Faculty of Biotechnical Systems 


e-mail: erolmurad@yahoo.com 

2 INOE 2000-IHP 


e-mail: dumitrescu.ihp@fluidas.ro



1. A v r a m M., Acţionări hidraulice şi pneumatice, Ed. Universitară, Bucureşti, 

2005. 

2. F e i d t M. L., Termodinamica şi optimizarea energetică a sistemelor şi 

proceselor, Ed. BREN, Bucureşti, 2001. 

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, 

Maşinostroienie, Moskva, 1973. 

4. M u r a d E., MEDSIMFP10, Mediu software simulare sisteme, Free Pascal-V2.1.4, 

U.P.B., 2008. 

5. M u r a d E., Măsurarea parametrilor procesului de uscare a produselor ceramice 

cu traductoare pneumatice neconvenţionale, HERVEX 2007, Călimăneşti 14-16 

noiembrie 2007. 

6. M u r a d E., C h e r c h e ş T., Traductoare pneumatice neconvenţionale cu 

consum redus de energie pentru măsurarea forţelor din instalaţii agricole şi în 

industria alimentară, HERVEX 2008, Călimăneşti 15-17 noiembrie 2008. 

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 

dryers under cogeneration, Conferinţa Internaţională, ENERGIE - MEDIU 

CIEM 2009, UPB, Bucureşti 12-14 noiembrie 2009. 

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 

uscare mobile şi ecologice destinate zonelor montane, HERVEX 2009, 

Călimăneşti 18-20 noiembrie 2009. 

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 

traductoarelor pneumatice cu consum redus de energie pentru măsurarea 

forţelor în procese lente, HERVEX 2008, Călimăneşti 18-20 noiembrie 2009. 

10. M u r a d E., C h e re c h e s T., Optimizarea consumului de energie a sistemului de 

conducere automată a proceselor de uscare convectivă, HERVEX 2008, 

Călimăneşti 18-20 noiembrie 2009. 

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 

elementelor şi schemelor pneumatice, Editura Tehnică, Bucureşti, 1985. 

12. * * * Catalog SMC, 2009. 

13. * * * Catalog FESTO, 2009. 

14. * * * Catalog Motorola 2008. 

SISTEM DE MĂSURARE PNEUMATICĂ A MASEI DE APĂ 

EXTRASĂ ÎN PROCESELE DE USCARE CONVECTIVĂ 

(Rezumat) 

Păstrarea produselor agricole vegetale, fructe şi legume, se poate face în condiţii 

foarte bune prin uscarea lor; această soluţie, beneficiind de un condus optimal, 

realizează produse cu potenţial nutritiv ridicat, cu consumuri energetice specifice 

reduse. În actualul context economic, derularea acestor procese în locaţii


descentralizate, de dimensiuni mici şi medii, necesită instalaţii uşor de deplasat, fiabile 

şi cu un preţ redus. Eficienţa economică a acestor instalaţii depinde de utilizarea 

conducerii optimale a proceselor de uscare care necesită măsurarea, în condiţiile termice 

din camerele de uscare, a masei de apă extrase. 

S-a dezvoltat o nouă schemă de măsurare on-line a greutăţii unei casete încărcată cu 

material de uscat care utilizează un traductor pneumatic de forţă care lucrează în regim 

de măsurare eşantionat. Soluţia prezentată se remarcă prin consumuri foarte mici de 

energie şi este caracteristică pentru automatizarea cu cost redus. Sistemul de măsurare 

este cuplat la PLC-ul destinat conducerii automate a instalaţiei de uscare. 

Comportarea dinamică şi consumurile energetice s-au determinat prin experimente 

de simulare realizate cu un model şi un program de simulare numerică, dezvoltat în 

mediul de simulare MEDSIMFP10. Experimentele de simulare au confirmat consumuri 

reduse de energie pneumatică (un rezervor de 2 l încărcat la 3 bar odată pe zi), precum 

şi o precizie bună de măsurare.


Publicat de 



Secţia 


TECHNOLOGIES FOR MONITORING STRUCTURAL 

DAMAGES ARISING IN THE FUNCTIONING 

OF WIND TURBINES 

BY 

MIHAI FLORIN MĂNESCU and VALERIU PANAITESCU 

Abstract. The need for sustainable development requires the use of renewable 

energy, primarily to maintain the existing energy (fossil fuels) to a suitable level, 

but also for environmental conservation. Accordingly, renewable energy sources 

began to be intensively studied and developed, mainly in the last ten years. In 

Romania, the first wind groups were put into service in 2005 (Ploiesti Industrial 

Complex - a wind turbine Vestas V52 - 660 kW, relocated), now the installed power 

being around 14 MW. For the purposes of harnessing wind energy for electricity 

production, because at heights exceeding 100 m, the wind is not influenced by 

existing deficiencies ground (trees, buildings, hills, etc..) Wind groups have become 

increasingly high and access to review and repair is difficult. As a result of this it is 

necessary to monitor the time of damage that may occur during operation of a wind 

farm to reduce the time the wind group is stopped and production losses are as small 

as possible. To ensure proper function safely it is necessary to implement a 

monitoring system for structural resistance (MSR) of wind turbine. This system and 

some of the damage assessment methods will be presented in this paper. The MSR 

has two branches: a network of sensors that collect information and an algorithm / 

software to interpret data from the sensors. Some of the technologies that can be 

used, methods of ultrasonic, thermal scanning, x-rays, etc. will be presented in this 

paper. Likewise, the most frequent types of structural damage that can occur in a 

wind group will also be discussed in this paper. 

Key words: wind turbine, monitoring structure resistance, sensors and damage 

detection. 


In order to achieve sustainable development, which requires the use of 

renewable energy, primarily to maintain the existing energy (fossil fuels) to a 

suitable level, but also for environmental conservation, in the last 20 years has

178 Mihai Florin Mănescu and Valeriu Panaitescu 

emphasized the efficiency of plants used as feedstock in electricity generation, 

renewable sources. Unlike other renewable energy sources, wind energy is 

limited because of its technological maturity, good infrastructure and relative 

price competitiveness. To have an effective higher cost-benefit, wind generators 

groups have become increasingly high. 

Some of the problems arising from the operation of wind turbines are listed 

below: 

a) the areas where wind farms are developed wind regions and which are 

usually difficult to reach, such as high mountain areas or areas of Dobrogea, due 

to existing infrastructure; 

b) the height of pillars (about 100 m), which makes revisions and access 

for repairs difficult; 

c) there have been accidents that led to the collapse of the entire wind 

structure and consequently to enormous losses; 

d) the tendency to have a high cost-benefit led to large structures; 

e) in Romania there isn’t a mature experience in the production of 

electricity using wind power, the first wind turbine was put into service in 2005, 

with a pillar height of 54 m. The wind farms that are now developed are usually 

100 m high and, in case of damage, this fact can lead to serious accidents, since 

action plans may not be effective. 

Given the above, it becomes necessary to monitor the whole resistance 

structure of wind turbines using a network of sensors. 

The monitoring systems are safe and low-cost, and their implementation can 

prevent energy losses during revisions or damage. These systems can efficiently 

diagnose which parts of the wind turbine must be replaced and they can also 

work to reduce duration of a scheduled review. Furthermore, these systems can 

predict the life cycle of the plant whose working time averages10-30 years. 

According to F a r r a r and S o h n [1] a monitoring system is defined as a 

process that detects faults which occur in high buildings. A fault is defined here 

as a alteration of the materials and / or of the geometric properties of these 

systems, including alterations in bordering conditions and system connectivity, 

which adversely affect system performance. There can be many causes of 

structural damage such as moisture absorption, fatigue, wind gusts [2], thermal 

stress, corrosion, fire and even lightning [3]. 

In general, the successful development of methods of MSR depends on two 

key factors: the technology employed for detection and signal analysis and the 

algorithm for interpretation [4]. MSR components consist of a data acquisition 

system, data filtering, pattern recognition and decision making. Each of these 

components is equally important in determining the safety status of a structure 

[5]. 

The monitoring process involves well-timed observation, using data from a 

series of sensors and statistical analysis in determining the status of the 

structure. In the case of long-term process MSR, the sensors provide regularly


updated information on the ability of the structure to perform its intended 

function, depending on the inevitable aging process and the degradation caused 

by the operation conditions. Ensuing extreme events such as earthquakes, MSR 

is used to obtain nearly real-time, reliable information on the integrity of the 

structure [1]. 

The same as wind turbines, the MSR system can provide data on the state of 

integrity of the assembly. The damage caused by wind turbines which can be 

detected by MSR is listed below, depending on the functional characteristics of 

the system. Early warning, with the following advantages and benefits: 

a) Prevention of breakdown; better planning of the revisions; reducing the 

cost of repairs; minimization of the reparation time. 

Problem identification, with the following advantages and benefits: 

a) Servicing at the appropriate time; reduction of the number of replaced 

parts; solving problems while still in warranty; reduction of the costs; operation 

at normal parameters. 

Continuous monitoring, with the following advantages and benefits: 

a) Constant update on the working parameters; security, less stress. 

The benefits of having a detection system faults are: [6], [13]. 

a) The prevention of premature breakdown: prevention of disasters, failure 

and secondary defects. 

b) Lower maintenance costs. 

c) Remote monitoring and remote diagnosis: turbines are usually built within 

a long distance from the premises of the beneficiary. 

d) Improvement of the capacity to warn against impending failure; the 

duration of the repairing process can be reduced in the windy season and 

consequently the production will not be affected. 

e) Support for the further development of a turbine: the data which has been 

gathered can be used to improve the models of the future turbine generation. 

With a reliable MSR, one can successfully plan an efficient maintenance 

and repair strategy of the wind turbines, especially offshore. Maintenance and 

repair can be performed on demand, when the weather conditions are 

favourable. Similarly, the mobilization to cover the costs of personnel, materials 

and equipment can be optimized [7]. 

An ideal monitoring system consists of two main components: a network of 

sensors for data collection and an algorithm for data analysis, which is used to 

interpret the measurements and determine the physical condition of the 

structures. 

These methods, which are either already applicable or promise application 

to the wind turbine systems in the near future, are presented below.


2. Damage Detection Technologies 

2.1. Non-destructive Technologies for Monitoring 

2.1.1. Thermal scanning technology. This technology is used to detect 

anomalies, based on the differences in the temperature of the inspected areas, 

such as the blades of the wind turbine, with the help of infrared sensors or heat 

detecting cameras [6]. The differences in temperature are correlated with the 

diffusion of heat and, as a result, they indicate abnormalities or damage. 

Specialized equipment, called thermovision or thermographic camera and 

similar in size and appearance to the very familiar video camera of everyday 

life, is used to obtain thermal images. In this way, objects are viewed in terms 

of the infrared radiation (IR) emitted by them and not in that of the visible 

radiation, which can be effortlessly detected with the naked eye. 

The thermographic method is extremely useful in practice because it 

permits the achievement of the so-called "predictive maintenance", a term 

which is applicable to all industries. The first sign of a defect or an operating 

problem is often given by the increased heat in that particular area, therefore an 

increase in the emission of infrared radiation. In other cases, the unjustified 

lower temperatures of some areas or of some parts of the device may be a sign 

of negative phenomena occurring at their level. After having been generated, 

the termograma is digitally processed in order to locate the points where heat 

stress is present and the defects. It should also be noted that the assessment of 

the termograme obtained by scanning the device, system or facility during 

proper functioning, can provide extremely valuable information on the normal 

thermal map, which will serve as reference for future evaluation and will also 

help with the remediation of any potential failure. 

Fig. 1 – Image of wind turbine produced by thermography. 

Additionally, the thermographic method is a non-invasive technique for 

measuring the infrared radiation emitted by the object under examination and 

generates, after a very short thermal analysis performed in real time, a thermal 

paper which may be sometimes of vital importance to the wind turbines. This 

method enables the precise measurement of surface temperature; the high 

resolution is a valuable advantage in detecting differences in the temperature of


those areas which are prone to failure. It can be stated that the method enables 

the so-called "defect management", the monitoring of the infrared radiation 

emitted by a technical system during normal operation often being the only 

imaging method that enables the actual visualization of the notions of "fault" 

and "inclination to failure". 

Thermography is very often the only quick on-site investigation of a 

facility. The areas which are prone to failure can be observed through a mere 

scanning of the frontage of the device. 

2.1.2. Ultrasonic technology. Ultrasound is applied differently, depending 

on the nature, geometry and destination of the product, taking into accounts the 

technical performance of the Probes Flaw system. Steel products are the ones 

which are most commonly examined by means of the ultrasound technique. 

Ultrasonic control can be differentiated and customized according to the 

methods specific to steel products: metal, forgings, castings, welded structures 

and products made of austenitic steel. 

The ultrasound check reveals all the types of internal defects of welded 

joints. The ultrasonic method can be used to determine wall thickness and the 

number of deposited layers. This technology is applicable to all metals and 

nonmetallic materials. Having a great penetrating force, the ultrasound enables 

the verification of large sections. The application is only limited by the rough 

structures which are highly heterogeneous. The devices are lightweight, 

portable and autonomous, and the technology will ensure good results even in 

on-site working conditions. The verification techniques, particularly the 

immersion control, lend themselves to mechanization and automation. 

The result is safe and immediate and it can accurately predict the location, 

size and depth of the defects. From an economic standpoint, the ultrasound 

method is much cheaper and more productive than the radiation technique in 

those cases where the number of defects exceeds a certain limit [14]. 

2.1.3. X-ray technology. X-rays can penetrate a large number of different 

materials, including composites. Usually, the images are obtained as shadow 

variations along the X-ray propagation. Under normal operation conditions, the 

X-rays are not able to reveal the cracks which have a parallel orientation with 

respect to the rays. 

The energy requirement is low and the safety prerequisites can be fulfilled 

from a remote position. The X-ray sources can be small, air-cooled and easily 

implemented for various applications. This method is able to detect the lack of 

binder between the metal, as well as cracks and holes in the metal. Some 

sources enable the detection of defects at less than 10 micrometres. This method 

can be used during real-time inspection for quality control or shortly after the 

blade of the wind turbine is set in motion. [13].


2.2. Discussion 

The advantage of the thermographic method is that this technology is able 

to produce an overall picture of the whole wind farm. A quick assessment can 

be made by a user with some experience. The main problem with this method is 

that the thermal images are limited to some electrical components that produce 

excessive heat during operation. This can also be applied to defective 

components if they move excessively, causing friction and heat. 

The ultrasonic methods will not detect single fibre breaks or composite 

materials. These methods do not work for complex structures (only the blades, 

the tower and the nacelle can be monitored). 

An advantage of the X-ray technology is that the images are obtained in 

parallel, not through scan as is the case with the thermographic approach and, 

therefore it is faster. However, problems can appear when the X-ray image is 

interpreted. It should be noted that the sources emitting the X-rays are small and 

air-cooled, thus fulfilling the safety prerequisites and they can also be controlled 

remotely. 

3. The Damage of the Structural Integrity of the Wind Power 

Impairment can occur in any component or part of the turbine and it can 

take any form, ranging from a crack in the concrete foundation to a split blade. 

Various cases of structural damage are reported from time to time in several 

countries such as Wales, Scotland, Spain, Germany, France, Denmark, Japan 

and New Zealand [8]. In Germany, in 2002, a blade broke away, and parts were 

found scattered throughout the area [9]. In another case, a blade flew to a 

distance of 8 km and entered the window of a house. A detailed documentary of 

this accident is available http://www.caithnesswindfarms.co.uk. According to a 

survey performed in Germany [10] the frequency of damage is almost equal for 

all the mechanical systems and structures. The failure varies from 4% for the 

structural parts and the gear to 7% for the rotor blades. 

Although structural damage may occur in any component, the most 

common type of damage is encountered in rotor blades and in the tower [11]. 

Special attention is given to the blades, as they are the key elements of a wind 

power generation system and as their cost can average 15-20% of the total cost 

of the turbine. It was noted that the damage to the blades is the most expensive 

type of repair and also the one which lasts the longest [12]. 

Additionally, even minor damage on the blades can lead to serious 

secondary damage to the entire wind turbine system when action to repair it is 

not taken immediately, leading to the collapse of the entire set [8].


The main construction elements of a turbine blade are shown in Fig. 2. The 

blades are made of a mixture of fiberglass and composite materials. They are 

designed to capture wind energy and transfer it to the turbine rotor and their 

profile is the result of complex aerodynamic studies, on which turbine 

efficiency is dependant. The development in size of the blades also caused an 

increase in the size of the rotor. 

Damage to blades may occur in different ways. The most common types of 

damage is listed below [5] and a sketch of these is available in Fig. 3. 

Typical damage description: 

1. Damage formation and growth in the adhesive layer joining the skin and 

the main spar flanges (skin/adhesive debonding and/or main 

spar/adhesive layer debonding); 

2. Damage formation and growth in the adhesive layer joining the up and 

downwind skins along leading and/or trailing edges (adhesive joint 

failure between skins); 

3. Damage formation and growth at the interface between face and core in 

sandwich panels in skins and main spar web (sandwich panel face/core 

debonding); 

4. Internal damage formation and growth in laminates in skin and/or main 

spar flanges, under a tensile or compression load (delamination driven 

by a tensional or a buckling load); 

5. Splitting and fracture of separate fibres in laminates of the skin and main 

spar (fibre failure in tension; laminate failure in compression); 

6. Buckling of the skin due to damage formation and growth in the bond 

between skin and main spar under compressive load (skin/adhesive 

debonding induced by buckling, a specific type 1 case); 

7. Formation and growth of cracks in the gel-coat; debonding of the gel-coat 

from the skin (gel-coat cracking and gel-coat/skin debonding). 

Fig. 2 – Components blade.


There are numerous causes which can lead to damage in the wind power 

systems. Some of the reasons could be improper installation, operation under 

severe conditions or low quality components. 

Fig. 3 – A representations of some types damage in the blades. 

Lightning can cause severe structural damage and the destruction of several 

towers. Fire and strong wind can also damage a turbine. The most critical task is 

probably when the turbine is stopped due to high wind. 

The greatest danger occurs when, during a period of strong wind, the brake 

system fails, and the turbine cannot be controlled. The brake system of a turbine 

rotor is designed to halt if the wind is too strong. 

When the brakes do not work, the turbine gets out of hand. In Germany, on 

several occasions during 1999, 2000 and 2003, the wind turbine brakes failed to 

stop the rotor, and blades were loose, pieces of which are found more than 

500m away from the pillar. 

Fig. 4 – Examples of damaged turbines at 100%.



1. Since the addition of the MSR monitoring system to a group of wind 

turbine can negatively affect performance, the number and location of the 

sensors is an important problem, which was addressed to a significant extent in 

the current literature. The methods to be implemented for wind turbines should 

demonstrate that they can perform well despite the small number of 

measurement points and under the constraint that these points will be located in 

areas prone to damage. 

2. The technologies for monitoring and damage detection can be selected to 

obtain comprehensive information. Detection methods are based on a network 

of sensors adapted for early warning at the onset of damage. The methods can 

detect the types of damages of less than 1 micron and the location of the 

damage can be accurately determined. 

3. Some of the limitations are the number of sensors they call for, as well as 

the noise produced during the operation of the turbine. In addition, the 

maintenance of a wired and wireless network between rotating parts, blades and 

nacelle is still difficult. The most promising methods are the ultrasound and the 

thermographic techniques. The X-ray methods, which were discussed here, can 

also be applied for non-destructive testing, where in situ results must be fully 

verified. 

4. Finally, since the environmental benefits are constantly emphasized, the 

generation of electric power with the help of wind systems is a necessary 

innovation. Therefore, related industries require monitoring systems that can 

provide effective maintenance programs, which provide the earliest possible 

detection of defects. 

5. The most important component of a wind turbine which requires 

monitoring is the blade, so we must implement a reliable system of low-cost 

sensors. This system should be integrated into the turbine blades in the 

manufacturing phase, which will signal any potential failure in the turbine. 

Therefore, the wind turbine should be built intelligently. 

Acknowledgements. The authors would like to thank HCI Environment SRL for the 

support and constructive comments. 

Received April 15, 2010 POLITEHNICA University, 

Faculty of Energetics 

Bucuresti, Romania 

e-mail: mihai.manescu@expert-mediu.eu



1. F a r r a r C. R., S o h n H., Pattern recognition for structural health monitoring 

Workshop on Mitigation of Earthquake Disaster by Advanced Technologies, 

Las Vegas, NV, USA, 30 Nov.–1 Dec. 2000. 

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 

monitoring techniques for wind turbine blades. J. Wind Eng. Ind. Aerodyn., 85 

(2008) 309–24 

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 

monitoring and fault detection of wind turbines and related algorithms: a review 

Renew. Sustain. Energy Rev. , 2007. 

4. K i m H., M e l h e m H., Damage detection of structures by wavelet analysis. Eng. 

Struct. 26, 347–62, 2004. 

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. 

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 

turbine blade of fibre composites based on studies of scale effects (Phase 1) 

Summary Report (Risø-R Report) Risø National Laboratory, Denmark, 2004. 

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 

prediction in wind energy converters Proc. EWEC, (Thessaloniki, Greece, 1994, 

pp 623–7 

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 

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 

monitoring for offshore wind energy converters with respect to the IEC61400-25 

standard Germany Wind Energy Conf. DEWEK, Wilhelmshaven, Germany, 

2004. 

8. R o s e n b l o o m E., A Problem with Wind Power www.aweo.org, 2006. 

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 

L. E., H i n k s o n J., L e w i s P. A., Bethany Wind Turbine Study Committee 

Report www.townofbethany.com, 2007. 

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 

Turbines Experiences of 15 Years with 1,500 W) ed J Peinke et al, Berlin: 

Springer, 2007, pp 329–32. 

11. C a i t h n e s s W i n d f a r m Information Forum 2005 Wind Turbine Accident 

Data to December 31st 2005 http://www.caithnesswindfarms.co.uk/ 

12. F l e m m i n g M. L., T r o e l s S., New lightning qualification test procedure for 

large wind turbine blades, Int. Conf. Lightning and Static Electricity,Blackpool, 

UK, 2003, pp 36.1–10 

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 

health monitoring for a wind turbine system: a review of damage detection 

methods (IOP PUBLISHING) 2008.


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 

J., Fundamentals for remote structural health monitoring of wind turbine 

blades—a preproject ANNEX B Risø-R-1341(EN) Report Risø National 

Laboratory, Denmark, 2002. 

TEHNOLOGII FOLOSITE PENTRU MONITORIZAREA 

DEFECŢIUNILOR STRUCTURALE APĂRUTE ÎN 

FUNCŢIONAREA GRUPURILOR EOLIENE 

(Rezumat) 

Necesitatea dezvoltării durabile impune utilizarea resurselor regenerabile de energie, în 

principal pentru menţinerea celor existente (combustibili fosili) la un nivel convenabil, 

dar şi pentru conservarea mediului înconjurător. În aceste condiţii sursele de energie 

regenerabila au început să fie intens studiate şi dezvoltate cu precădere în ultimii zece 

ani. În România primele grupuri eoliene au fost puse în funcţiune în anul 2005 

(Complexul industrial Ploieşti – o centrala Vestas V52 – 660 kW, relocată), în prezent 

existând o putere instalata de circa 14 MW. 

În sensul valorificării energiei eoliene pentru producerea de energie electrică, 

datorită faptului că la înălţimi de peste 100 m, vântul nu mai este influenţat de 

neregularităţile existente la sol (arbori, clădiri, coline, etc.) grupurile eoliene au devenit 

tot mai înalte, iar accesul pentru revizii şi reparaţii este dificil. Ca urmare a acestui fapt 

este necesară o monitorizare din timp a avariilor ce pot să apară în timpul funcţionării 

unui parc eolian, pentru a reduce timpul în care grupul eolian este oprit, iar pierderile de 

producţie să fie cât mai mici. Pentru a se asigura o bună funcţionare în condiţii de 

siguranţă este necesar a se implementa un sistem de monitorizare a structurii de 

rezistenţă (MSR) a ansamblului eolian. Acest sistem şi cateva metode de evaluare a 

avariilor vor fi prezentate în acest articol. MSR are în componennţă două ramuri: o reţea 

de senzori ce colectează informaţiile şi un algoritm/software pentru interpretarea datelor 

provenite de la senzori. Cateva dintre tehnologiile ce pot fi folosite, metode cu 

ultrasunete, scanare termica, raze X, etc. vor fi prezentate în articol. Tot în acest articol 

vor fi prezentate cele mai frecvente avarii de structură ce pot apărea la un grup eolian. 

Spre deosebire de alte surse regenerabile de energie, energia eoliană are o limită 

datorită maturităţii sale tehnologice, infrastructură bună şi o competitivitate relativă de 

preţ. Pentru a avea o eficienţă cost-beneficiu mai ridicată, grupurile generatoare eoliene 

au devenit tot mai înalte. 

Monitorizarea structurii de rezistenţă (MSR) este esenţială în raport cu celălalte 

sisteme de monitorizare pe care le are în componenţă un grup eolian, deoarece 

prăbuşirea întregii structurii are urmări grave asupra siguranţei populaţiei, a mediului şi 

din punct de vedere financiar al investitorului. 

Sistemele de monitorizare sunt sigure, au un cost redus, iar prin implementarea lor 

se pot reduce pierderile de energie din timpul reviziilor sau avariilor. Prin aceste sisteme 

se pot diagnostica ce piese ale grupului generator trebuie înlocuite şi reducerea timpului 

din cadrul unei revizii programate. Totodată aceste sisteme pot face o predicţie asupra 

ciclului de viaţă al centralei al cărui timp de lucru proiectat este de 10-30 ani.


Conform Farrar şi Sohn [1] sistemul de monitorizare este definit ca un proces care 

detectează defecţiunile apărute în construcţiile înalte. Defecţiunile sunt definite aici ca 

modificări ale materialelor şi / sau proprietăţile geometrice ale acestor sisteme, inclusiv 

schimbări în condiţiile limită şi conectivitatea sistemului, care afectează în mod negativ 

performanţa sistemului. Exista mai multe cauze de avariere a structurii cum ar fi 

absorbţia de umiditate, oboseala, rafale de vânt [2], stres termic, coroziune, foc şi chiar 

trăsnet [3]. Palele turbinelor care nu au un sistem de protecţie contra fulgerelor sunt 

adesea lovite de acestea. 

În general, dezvoltarea cu succes ale metodelor de MSR depinde de doi factoricheie: 

tehnologia de detectare şi analiză a semnalului şi algoritmul de interpretare [4]. 

Componentele MSR sunt formate din sistemul de achiziţie de date, filtrare de date, 

recunoaşterea de model şi luarea deciziilor. Fiecare dintre aceste componente este la fel 

de importantă în determinarea stării siguranţă a unei structuri [5]. 

Tehnologiile de monitorizare sau de detectare a avariilor pot fi selectate pentru a 

obţine informaţii globale. Metodele de detectare au la bază o reţea de senzori adaptată 

pentru o avertizare timpurie de la debutul avariei. Metodele pot detecta tipuri de daune 

de sub 1 micron şi se pot localiza deteriorările sau impactul destul de precis. 

Limitarea dintre aceste tehnoligii este dată de numărul de senzori solicitaţi, precum 

şi de zgomotul produs în timpul funcţionării turbinei. Metodele cele mai promiţătoare 

sunt cele cu ultrasunete şi termografia. Metodele cu raze X, discutate, pot de asemenea, 

să fie aplicate pentru control nedistructiv, atunci când rezultatele în situ trebuie să fie 

complet verificate. 

În cele din urmă, după cât de mult este pus accentul pe mediu, beneficiile sistemelor 

eoliene în generarea de energie electrică, sunt mari, dar şi costisitoare cu construcţia şi 

întreţinerea. Prin urmare, industriile conexe necesită sisteme de monitorizare care să 

poată oferi eficienţă programelor de întreţinere, care să ofere detectarea defectelor cât 

mai devreme posibil.


Publicat de 



Secţia 


PUMPING STATION EXONERATION FOR WATER SUPPLY 

BY 

AURORA ALEXANDRESCU, ADINA SIMONA ALEXANDRESCU 

and ADRIAN CONSTANTIN ALEXANDRESCU 

Abstract. Profitability of water distribution activity depends largely on the 

relationships between operational capability and service costs, related to 

supplier’s performance, volume of distributed water and effective operating costs. 

The main variables that influence the total selling price are required investment 

value, specific consumption of electrical energy for pumping power, unit price of 

the electrical energy and total volume of monthly consumed water billed. The 

selection of rehabilitation and modernization measures must rely on market 

studies results that appropriately establish the quantities of water that may be 

distributed and billed. Present and future water requirements will be determined 

based on the analysis of actual operation data and on estimation of future trends in 

water consumption on national and international levels. 

Key words: adduction, conductivity, chlorine, pipe network, pumping station, 

tank. 


Many systems for which a centrifugal pump is otherwise suitable may, 

however, have a variable demand in which case, a certain loss of efficiency may 

have to be accepted from part of the head or part of the capacity used for control 

purposes, using either discharge throttling or bypass control. Both methods will 

inevitably result in power loss, so if economic regulation is of primary 

importance, discharge regulation by speed control should be investigated first 

since this is less wasteful of power and there is usually a considerably smaller 

loss of pump efficiency. Speed control is now a particularly attractive 

proposition with the increasing availability of variable frequency power units, 

[1]. The measures for the improvement of the economic performances are 

established from the system’s dispatcher. The criterions used are the maximum

190 Aurora Alexandrescu et al. 

output of the pumping equipages and the optimum diameter of the piping. The 

replacement of the existent equipment, that is obsolete from physical and 

technological point of view, must be done with new equipments with 

performances that will meet the requirements of an optimum operation from 

both energetic and economic perspectives, [2]. 

2. Methodology 

This analysis is more likely to be of academic rather than practical interest, 

however, since the main requirements with regulated flow are: 

a) To achieve the required variation in delivery to meet demand variations. 

b) To realise good pumping efficiency at all demand levels, [3]. 

The connections analyze between the functional parameters of pumping 

station and the quality parameters of drinkable water are: 

a) The connections estimation between the pump’s load, debit and the water’s 

turbulence. 

b) The values analyze of chlorine and ammonium concentration depending on 

debit. 

c) The water conductivity analyzes depending on debit. 

The best power and economical performances will correspond to the 

pumping solution which ensures the covering of the request area (Q, H) with the 

best output. The pumping efficiency is established by studying technical 

implications of modernization measures of the power station. It is imperative to 

use the automatic systems for the water treatment. Energy efficiency and 

economic efficiency for the pumping supply system are tightly connected to the 

proper choice of pumping device and appropriate operation of the hydraulic 

system, [4]. 

The evaluation of connections between the functional parameters of 

hydraulic systems and the quality indicators of water, depending on time is 

obtained with the help of an original automatic calculation program, elaborated 

by authors. Allowing for the actual situation of the adduction and the 

effectuation’s necessity of the repair workings capital or the pipe’s replacement 

in some sectors, respective of the pumping aggregates, for to assure a powereconomical 

evolution favorable of the system, one are determined: the optimum 

nominal diameter of the pipes and the maximum theoretical outturn of the 

pumps what must be used and the change’s convenience of the existing 

equipment, [3]. 

Depending for the installation’s total outturn and for the water volume W 

circulated in the period of reference, the consumption of power for the water’s 

pumping Eso is independent at stop’s duration „at top”:

(1) 


Ntm. W KN. QH . t. W KN. QLJW . . . 

E so = = = 

3600 3600. η 3600. η 

p p 

In Eq (1) has used the following notations: Ntm (kW) - power average 

necessary for water’s transport: 

(2) 

KN. QH . t KN. QLJ . . 

Ntm( QDL , , ) = = . 

η η 

p p 

W (m 3 ) -volume of water brought in T (hours) period; J (-) - slope metric of the 

water transport; D (m) - nominal pipe’s diameter; Q (l/s) - debit of water; L (m) 

- pipe’s length; ηp (%) - pump’s outturn; KN , Kj - coefficients; Ht (m) - loss of 

charge: 

(3) 

Q 

Ht= L. J( Q, D) 

, J = K j. 

D 

γ 

β 

⎛ γ 

Q ⎞ 

DQJ ( , ) = ⎜K j. 

, 

⎜ 

⎟ 

J ⎟ 

⎝ ⎠ 

It is calculated annual medium costs associated to the power consumption 

and the annual average cost associated to the investments: 

(4) CAE = pE.(1 + krE 

). E A= AIc + AP= ac. Ic + a . I 

. 

1/ β 

, p p 

The investment in the pipes network and the investment in the pumping 

station are calculated with the fallowing equations: 

(5) 

(6) 

α/ β 

p 

⎛ γ 

Q ⎞ 

⎛1,15. KN. QL . . J ⎞ 

Ic( D, L) = La . . ⎜K j. ; Ip( Nti) = Ipo. 

. 

⎜ 

⎟ 

⎜ ⎟ 

J ⎟ ⎜ ⎟ 

⎝ ⎠ ⎝ 

η p ⎠ 

It is calculated the annual medium total costs: 

( + ) 

α/ β 

⎛ γ 

Q ⎞ 

⎛1,15. KN. QL . . J ⎞ 

CA = A+ CAI = ac. La . . ⎜K j. + ap. Ipo. 

⎜ 

⎟ 

⎜ ⎟ 

J ⎟ ⎜ ⎟ 

⎝ ⎠ ⎝ 

ηp 

⎠ 

pE.1 krE . KN. QL . . JW . 

+ 

. 

3600. η 

p 

. 

α 

α p 

+


(7) 

(8) 

The annual medium marginal cost is calculated with the fallowing equation: 

( + ) 

α/ β 

p 

C . 

A La . ⎛ γ 

Q ⎞ Ipo ap⎛1,15. KN. QLJ . . ⎞ 

CM = = . ⎜Kj. + . 

+ 

W W ⎜ 

⎟ 

⎜ ⎟ 

J ⎟ W ⎜ ⎟ 

⎝ ⎠ ⎝ 

ηp 

⎠ 

αγ / β 

pE.1 krE . KN. QLJ . . Q 

αp αp 

+ = A*. + BQ . . J + CQJ . . ; 

3600. η 

α/ β 

J 

p 

( ) 

p 

La . α/ β Ipo. ap⎛1,15. 

KN. L⎞ 

pE.1 + krE . KN. L 

A* = . ( K j ) ; B= . ⎜ ⎟ ; C = 

. 

W W ⎜ ⎟ 

⎝ 

ηp ⎠ 

3600. ηp 

The minimum marginal cost has the following configuration: 

αγ 

∂CM α Q β 

αp αp−1 = 0 ⇒− A. . + B. α p. 

Q . J + C. 

Q = 0 

∂Jϕ α 

1+ 

β 

J 

3. Results and Discussion 

The available data for choosing the proper pumping device require a careful 

analysis to determine the head characteristics of the pressurized supplies for 

each specific configuration and to evaluate the proper operational regimes of the 

pumping devices. The calculation method is applied in the exoneration case of 

pumping station CUG Iasi for drinkable water from Iasi city. 

Before of exoneration the pumping station CUG Iasi has been equipped 

with two 8 NDS pumps; after exoneration the pumping station CUG is equipped 

with two Wilo NPG 200-500-160/4/499 pumps. 

The measure and control system of drinkable water’s quality, of functional 

parameters of the pipes network from CUG pumping station has been complete 

automated in August – December 2008 with performance installations by 

Endress Hauser concern. All system has automated calculus program that 

permits of the dispatcher from S. C. APAVITAL S. A. Iasi industry to pursue 

the parameters’ network in 17 demurrages from the supply system with 

drinkable water of Iasi town, (Fig. 1). The registering is permanent, 24 hours. 

The water samples are analyzed automatic at each 2 ÷ 5 minutes. Drinkable 

water is treated with ammonium and chlorine in accordance with the properties 

of the analyzed water. 

α 

α


The all measurement system of drinkable water’s conductivity from pipes 

network supplied of CUG Iasi pumping station includes the following elements: 

temperature integrated sensors inductive and conductive sensors. The turbidity 

and the solid suspensions quantity Tb from drinkable water system are 

measured with the Liquisys M CUM 223/253 type system. The turbidity sensors 

are Turbimax W CUS 31 type. 

Fig. 1 – The emplacement points of 17 automatic pursuit systems automate of the 

supply network with drinkable water from Iasi city. 

The conductivity, chlorine and turbidity concentration variations are 

measured in same moments of 1.06.2009 day, (Fig. 2) in CUG pumping station. 

The automized regulation system of chlorine and chlorine dioxide quantity from 

CUG pumping station is Liquisys M CCM 223/253 type. The CSC 140/141 

type sensors are used to visualize the automized measurement and processing of 

the Cl2, HOCl, OCl - concentrations, depending on the drinkable water pH. The 

measurement system of ammonium quantity from drinkable water is Stamolys 

CA 71 AM type. The ammonium concentration variation in same time in 

1.06.2009 day is presented in the CUG pumping station E1, in Fig. 3. The 

digital transmission system of the pressure p and the debit Q from CUG 

pumping station at the S. C. APAVITAL S. A. Iasi dispatcher is Cerebar M 

HART, respectively Priline Promag 50 types.


Fig. 2 – Diagram for the functional parameters of the CUG Iaşi pumping station: 

conductivity, chlorine, pressure, flow and turbulence for the perioad 5.07.2009 – 

6.07.2009. 

AIc ( D ) 

AP ( k , Q , D ) 

A( k, Q , D ) 

1000 

900 

800 

700 

600 

500 

400 

300 

200 

100 

0.3 0.4 0.5 0.6 0.7 0.8 

Fig. 3 – The variation of annual cost associated to the investment in pipes AIc, annual 

total cost AP for pumping and the annual average cost associated to the investments A, 

depending on diameter D for the CUG pumping station. 

The results of the data working relative at either sector of the system, 

effectuated with MathCAD program, adequate to the supply with power under 

average and low tension are systematized in Fig. 3. It is present the variation of 

the following parameters: AIc – annual cost associated to the investment in 

D


pipes, AP – annual total cost for pumping and A - the annual average cost 

associated to the investments in the usage’s conditions of the following 

parameters: ac = 0,075; a = 6000; α = 2; k – pipe’s equivalent absolute rugosity 

= 0,8 mm; Kj = 0,00145; γ = 2; β = 5,143+0,131.k; ηp = 72 %; T = 6200 

hours/year; αp = 0,46; Ipo = 65; Q = 0,15 m 3 /s; D = 0,3; 0,4; ...; 0,6 m. 


1. It comes out that chlorine and ammonium concentrations grow up in the 

maximum consumption period of network: Cl = (1,1 ÷ 2,89) [mg/l] and Am = 

(42,55 ÷ 43,31) [mg/l]. In the same time, pressure and debit grow up because to 

the increase of the active consumers number in the network: p = (3,01 ÷ 4,09) 

[bar] and Q = (0 ÷ 1650,74) [m 3 /h]. 

2. The water turbidity grows up at (0,49 ÷ 0,59) FNU values on starting 

and stop period for each pumps. In short time from the starting and stop 

moment, the turbidity declines at (0,26 ÷ 0,30) FNU values. The water’s 

conductivity is constant (0,53 ÷ 0,54) [mS/cm]. 

3. The flow and load variation from hydraulic system influence the water 

quality and the substances quantity imperative for the drinkable water treatment: 

in the analyse period 1.05.2009, 00:00 hour and 15.05.2009, 23:55 hour, 

maximum chlorine is Clmax = 3,94 [mg/l] in 7.05.2009 day, 3:45 hour; at same 

moment the hydraulics system has the following parameters: Tb = 0,42 FNU 

turbidity, Q = 0,121 [m 3 /s] flow, H = 69,34 [m] load, Cv = 0,51 [mS/cm] 

conductivity. The minimum chlorine is Clmin = 1,1 [mg/l] in 3,05.2009 day, 3:35 

hour; the parameters of the CUG pumping station at same moment are: 3:35 

turbidity, Q = 0 [m 3 /s] flow, H = 64,344 [m] load, Cv = 0,52 [mS/cm] 

conductivity. 

Acknowledgements. This work has been supported by the National Centre of 

Management Programmers, Romania, under financial contract No. 21-041/2007. 

Received: February 25, 2010 “Gh. Asachi”Technical University, 

Department of Fluid Mechanics, Machines and Hydraulic Acting, 


e-mail: auralexis@yahoo.com 


1. A l e x a n d r e s c u A., Statii de pompare, Ed. Politehnium, Iasi, 2008. 

2. A l e x a n d r e s c u A., Masini si echipamente hidraulice, Ed. Politehnium, Iasi, 

2008.


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., 

Contributions concerning the power optimization of the pumping stations, 

ASME Fluids Engineering Division Summer Conference, ISI Thomson 

Proceedings of 11th International Symposium on Advances in Numerical 

Modelling of Aerodynamics and Hydrodynamics in Turbo machinery, August 

10-14, FEDSM2008-55007, Jacksonville, Florida, USA, 2008. 

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 

pumping stations exoneration, ISI Thomson Proceedings in The 6th International 

Conference Management of Technological Changes, Centre for Continuing 

Education and Training – CETEX, MTC, Alexandropoulos, Greece, Vol. I, 

2009, pp. 437-441. 

REABILITAREA STATIEI DE POMPARE PENTRU ALIMENTARE CU APA 

(Rezumat) 

Sistemul de măsurare şi de control a calităţii apei potabile, a parametrilor 

funcţionali ai reţelelor de conducte trebuie complet automatizat în întregul municipiu 

Iaşi şi generalizat la nivelul întregii ţări. Deşi investiţia în sistemele moderne de 

automatizare, control şi măsurare din reţelele de alimentări cu apă potabilă din marile 

oraşe este mare, aceasta poate fi amortizată repede, prezentând multiple avantaje: 

- Posibilitatea optimizării funcţionarii ansamblului staţie de pompare 

- Obţinerea unei caţităţi foarte bune a apei potabile, dozarea fiind efectuată automat, 

în funcţie de calitatea apei ce curge prin reţelele de conducte. 

- Obţinerea unei concentraţii optime a bulelor de gaze conţinută în apa potabilă de 

(3 ÷ 10) % pentru evitarea fenomenului de vortex în instalaţii. 

Turbiditatea apei potabile creşte brusc la valori cuprinse între (0,49 ÷ 0,59) FNU pe 

toată perioada în care se pornesc şi se opresc fiecare din pompe. În scurt timp de la 

momentul închiderii, respectiv a deschiderii pompelor, turbiditatea scade la valori 

cuprinse între (0,26 ÷ 0,30) FNU. Conductivitatea apei rămâne aproximativ constantă, 

variind în intervalul (0,53 ÷ 0,54) [mS/cm]. Clorinarea variază în intervalul (1,1 ÷ 2,89) 

mg/l şi creşte odată cu creşterea turbidităţii apei potabile. 

Variatia debitului si a sarcinii din sistemul hidraulic studiat influenteaza puternic 

calitatea apei si implicit cantitatea de substante necesare tratarii apei pentru potabilizare: 

in perioada de analizată, cuprinsă în intervalul 1.05.2009, ora 00:00 şi 15.05.2009 ora 

23:55, clorinarea maximă Clmax = 3,94 [mg/l] s-a înregistrat în ziua de 7.05.2009, ora 

3:45; în acelaşi moment s-au înregistrat şi următorii parametri: turbiditatea Tb = 0,42 

FNU; debitul Q = 0,121 [m 3 /s]; sarcina H = 69,34 [m]; conductivitatea Cv = 0,51 

[mS/cm]. În aceeaşi perioadă, clorinarea minimă Clmin = 1,1 [mg/l] a fost înregistrată în 

ziua de 3,05.2009, ora 3:35. Ceilalţi parametri ai staţiei de pompare CUG în acelaşi 

moment au fost: Tb = 0,43 FNU; debitul Q = 0 [m 3 /s]; sarcina H = 64,344 [m]; 

conductivitatea Cv = 0,52 [mS/cm].


Publicat de 



Secţia 


CONTRIBUTIONS REGARDING DURABILITY EVALUATION 

OF HORIZONTAL AXIAL HYDRAULIC TURBINES SHAFTS 

BY 

ILARE BORDEASU 1 , MIRCEA OCTAVIAN POPOVICIU 1 , DRAGOS 

NOVAC 2 , LIVIU MARSAVINA 1 , RADU NEGRU 1 , MIRCEA VODA 1 , 

VICTOR BALASOIU 1 and MARIAN BĂRAN 1 

Abstract. Using previous researches regarding cracks initiation, the present 

paper analyzes the durability of horizontal bulb turbines shaft, especially in the 

joining zone between the shaft main body and the runner flange. For 

exemplifying, there were used the data of the turbines running in the Power Plants 

Iron Gates II and Gogosu. The obtained conclusions can be used to avoid serious 

failures of horizontal hydraulic turbines. The abstract of the paper is to be written 

here. It contains the main ideas and original contributions and conclusions of the 

authors’ research. 

Key words: axial hydraulic turbines, cracks, stress alternation cycles, fracture 

mechanics characteristics 


It is a well-known fact that during of the turbine running (rotational 

motion), under the effect of: hydraulic forces and moments developed on the 

blades, runner masses (with or without oil) and vibration, the turbine shaft is 

submitted to specific variable stresses (tension, bending and torsion). Taking 

into account the results obtained in [1], in the present work, the durability of the 

horizontal axial hydraulic turbines shaft was evaluated using the professional 

programs ANSYS and AFGROW. In order to obtain also numerical examples 

were used the bulb turbines in function at the Power Plants Iron Gates II and 

Gogosu, for which we know the necessary data. The obtained conclusions allow 

avoiding dangerous troubles during the turbine operation.

198 Ilare Bordeasu et al. 

Fig. 1 – The 3D shaft model, obtained with the INVENTOR program [initiation]. 

2. Estimation of Fatigue Crack Propagation for Variable Stresses 

with Constant Amplitudes 

For the selected axial hydraulic turbine (Power plant Iron Gates II) the shaft 

presents the following characteristics [1]: 

a) the initial crack occurs in the joining zone between the shaft flange (for 

coupling with the hydraulic runner) and the shaft body, fig. 2; 

b) in order to obtain the needed data we choose the AISI 1020 steel [6] instead 

of the actual used 20ГC steel [7]; both steels have almost the same mechanical 

characteristics but for the AISI 1020 we found all the supplementary 

characteristics needed for fatigue tests; 

c) the shaft dimensions are: length 7572 mm, maximum diameter 2300 mm, 

the main shaft body, between flanges, has an external diameter of 1200 mm and 

an internal one of 600 mm; 

d) according with the ANSIS program (Fig. 3), till crack initiation, the 

minimum number of cycles was 3,0139 10 8 , corresponding to 80,370 running 

hours. 

a) Deep circumferential cracks b) Cracks network 

Fig.2 – Cracks in the joining zone between the runner flange and the main shaft body.


Fig. 3 – Number of cycles till crack initiation. 

Usually, together with the increase of stress level is enhanced also the 

fatigue crack propagation speed. As a result, the propagation of the crack speed 

can be correlated with the variation of the stress intensity factor 

da 

ΔK, = f ( ΔK) 

, Fig. 4. 

dN 

Fig. 4 – Variation of crack propagation speed as a function of stress intensity variation.


In conformity with Fig. 4, after the crack initiation, the maximum danger is 

represented by the zone III, characterized by great propagation velocities 

leading to instable increases of the failure. In this zone, the velocity of crack 

propagation is correlated with the variation of the crack intensity factor, through 

the relation proposed by Forman [2]: 

(1) 

da 

dN 

n 

C( 

ΔK 

) 

( − R) 

K − ΔK 

= 1 

C 

= f (ΔK, R), 

where: C and n are constant values depending on the used material; 

ΔK = Kmax − Kmin 

- represents the variation of the stress intensity factor; KC 

- is the critical value of the stress intensity factor (breaking tenacity); R - 

represents the asymmetry ratio of the stress cycle. 

Expressed by Nr, the necessary number of cycles for the extension of the 

crack, the lifetime can be obtained by solving the equation for dN. 

Integrating both members it results: 

(2) ∫ dN = N − N = N = ∫ 

N 

N 

f 

d 

f 

d 

r 

a 

a 

cr 

d 

da 

. 

f ( ΔK, 

R) 

With the equation (2) it is possible to calculate the number of cycles Nr 

necessary for the extension of the crack from the detectable length ad (for which 

corresponds the number of cycles Nd) till the critical length acr (which 

corresponds to the Nf number of cycles). 

For the lifetime computation (the final number of cycles Nr) it was used the 

specialized program AFGROW, developed by H a r t n e r at WRIGHT- 

PATTERSON AIR FORCE BASE [3] in order to estimate the lifetime of some 

components for fighter planes. It was taken into consideration a ring section, 

having an elliptical crack as can be seen in Fig. 5. For such geometry, the 

intensity factor was proposed by Raju and Newman [4] as being: 

a 

= t b , i e , 

Q 

(3) K ( σ + H σ ) π F( 

a c, 

D , D , φ) 

I 

where σt - represents the axial stress applied to the shaft, σb - represents the 

bending stress, H and F depends on the crack geometry (crack depths and 

length), the shaft thickness and the frontal position of the crack, a is the depth 

and c the half-length of the crack,

(4) 


1. 

65 

⎛ a ⎞ a 

Q = 1+ 1, 

464⎜ 

⎟ for ≤ 1. 

⎝ c ⎠ c 

Fig. 5 – Computation model for crack propagation. 

As was previously mentioned, the mathematical simulation was done with 

the program AFGROW, version 4.12.15 from 10.07.2008. The entrance data 

were: the external diameter (Do): 1.20 m, the internal diameter (Di): 0.60 m, the 

crack depth (A): 0.001 m, crack half length (C): 0.002 m 

In conformity with the shaft disposition, for studying the crack propagation 

there were taken into consideration two load situations with constant amplitude: 

a) a bending loading with a symmetrical alternative cycle having σmax = - σmin 

= 42 MPa, with an asymmetry coefficient of R = -1. 

b) a composed load (bending plus elongation); considering that the static 

elongation stress is superposed over the bending stress it result an approximate 

pulsating cycle with σmax = 88,9 MPa, σmin = - 6,5 MPa , R = - 0,07 [1]. 

In order to introduce the material constants, characterizing the crack 

propagation, the used steel was considered to be equivalent with the AISI 1020 

for which the characteristic data are included in the data base of the AFGROW 

program. So, the constants for the Walker law [5]:


(5) 

[ ] 1 1 m n 

K ⋅ ( 1− 

R) 

da 

= C1 

⋅ 

dN 

max 

− 

for R ≥ 0, 

[ ] 1 1 

da −m 

n 

= C1 

⋅ K max ⋅ ( 1− 

R) 

for R < 0 

dN 

are C1 = 1,447 x 10 -12 ; n1 = 3,6; m = 1; the values for the breaking tenacity 

corresponding to a plane state of stresses are KC = 110 MPa m 1/2 , respectively 

for a plane state of deformation KIC = 77 MPa m 1/2 , the yielding point σC = 262 

MPa, the elastic modulus E = 206843 MPa and Poisson’s ratio ν = 0.3, the limit 

value of the stress intensity factor under which the crack does not propagate ΔK 

= 1,5 MPa m 1/2 . 

In Fig. 6 is presented the crack evolution for some time intervals under a 

load having a symmetric alternative bending cycle. 

a) a = 1mm, c = 2mm b) a = 16 mm, c = 32 mm 

N = 80370 hours (initiation) N = 153243 hours (propagation) 

c) a=150 mm and 2c= 320 mm 

N = 159737 hours (fracture, breakdown) 

Fig. 6 – Propagation of circumferential crack in the hollow shaft. 

The simulation begun with the following initial value of the crack a = 1 mm 

and c = 2 mm, values which are reached after 80370 running hours. The number


of hours, for the crack propagation is added to the initial time (80370). The 

obtained results show that after 15937 running hours the crack penetrates 

completely the annular wall of the shaft. 


The studies of the fatigue crack propagation lead to the following results: 

1. In the joint zone between the flange and the shaft body, the crack occurs 

after approximately 80370 running hours. 

2. The propagation of the fatigue cracks were obtained using the Paris law 

with the completion of Walker, and show that the crack reach the dimensions a 

= 16 mm (depth) şi 2c = 64 mm (length) after 153243 running hours. After that 

the propagation speed increases to a high degree and after only 159737 running 

hours the breakdown occur (a=150 mm and 2c= 320 mm). 

3. The obtained results can be used by the supervising staff in order to 

prescribe the intervals of periodical inspections and to avoid major damages. 

Acknowledgements. The present work has been supported from the Contract No. 

RU 177/10.10.2008, BC 146/13.10.2008 (Analiză privind soluţia de fiabilizare a 

arborelui turbinelor aplicată cu ocazia retehnologizării hidroagregatelor din CHE Porţile 

de Fier II. Propuneri de metodologie de urmărire în timp a stării arborilor turbinelor din 

CHE Porţile de Fier II şi CHE Gogosu) 

Received:March 15, 2010 1 Politehnica” University of Timisoara, 

e-mail: ilarica59@gmail.com 

2 Hidroelectrica Iron Gates, Dr.Tr. Severin 

e-mail: dragos.novac@hidroelectrica.ro 


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 

Horizontal Hydraulic Turbines Shaft and Estimation of Crack Initiation Machine 

Design, University of Novi Sad, Faculty of Technical Sciences, 2009, pp 183- 

186. 

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 

propagation in cyclic-loaded structures, Journal of Basic Engineering, Trans. 

ASME, Vol. 89, (1967). 

3. H a r t n e r J. A., AFGROW Users Guide and Technical Manual, Wright-Patterson 

Air Force BASE, Ohio, 2008.


4. R a j u I. S., N e w m a n J. C., Stress Intensity Factors Circumferential Surface 

Cracks in Pipes and Rods, Proc. of 17 th National Symposium on Fracture 

Mechanics, Albany, NY, 1984. 

5. W a l k e r K., The effect of stress ratio during crack propagation and fatigue for 

2024-T3 and 7075-T6, ASTM STP 462, ASTM, 1970. 

6. * * * Fatigue Design Handbook, Second Edition, SAE, Warrendale, 1988, 

7. * * * Analiză privind soluţia de fiabilizare a arborelui turbinelor aplicată cu ocazia 

retehnologizării hidroagregatelor din CHE Porţile de Fier II. Propuneri de 

metodologie de urmărire în timp a stării arborilor turbinelor din CHE Porţile de 

Fier II şi CHE Gogosu, Contract No. RU 177/10.10.2008, BC 146/13.10.2008 

CONTRIBUŢII ÎN EVALUAREA DURABILITĂŢII ARBORILOR 

HIDROAGREGATELOR AXIALE ORIZONTALE 

(Rezumat) 

În cadrul lucrării se evaluează durabilitatea arborilor turbinelor axiale, în zona de 

racord a flanşei ce-l cuplează cu rotorul turbinei, plecând de la rezultatele obţinute în 

lucrările anterioare privind iniţierea fisurii. Ca model este folosit arborele de turbină 

bulb, aflată în exploatare la CHE Porţile de Fier II şi CHE Gogoşu, pentru care 

dispunem de datele necesare. Concluziile rezultate permit evitarea unei eventuale avarii 

periculoase.


Publicat de 



Secţia 


OPTIMAL ROUTES OF PIPELINE SUPPLY USING 

THE BELLMAN-KALABA ALGORITHM 

BY 


ADRIANA MANEA, RODICA BĂDĂRĂU and DANIEL STROIŢĂ 

Abstract. In this paper are presented the application of graphs theory for 

determining the optimal route of a pipeline supply being at the great distance of 

the target consumer (pipeline network of a city). It applies when the distance from 

source to target, because the configuration of the land, there are several variants of 

the route passing through some mandatory points. In this way the route has n 

sections and on each section the total cost (investment plus operating for one year) 

has a certain value. If it can browse the route by more than two then the method 

becomes profitable. Implementation of the method is through a special program, 

using the Borland Pascal programming and Bellman-Kalaba algorithm. 

Mathematical resolving is by the matrix. Numbering the sections with 1 ... n, in 

order to obtain the final optimal browsing, it is the range of selected sections. In 

actual conditions when more and more sources of drinking water are becoming 

more polluted, the feeding is justified to be from remote mountain areas of the 

natural springs. 

Key words: graph theory, pipeline, optimal route, matrix, Bellman-Kalaba 

algorithm, and network. 


Through scope of investment and energy consumption, adductions have a 

significant share in the systems of water supply, and their rational design 

behaves more optimization processes, among which an important place held 

optimize their route. 

In current practice of design, setting the optimal solution is, usually, by 

analytical study of two or three variants selected from many possible intuitive

206 Teodor Milos et al 

decisions, whose error is inversely proportional to the degree of experience of 

the designer. 

Modern mathematical disciplines, by operational calculation, put at 

specialist’s disposition a vast apparatus of scientific analysis in determining the 

optimal decisions for the design of water supply. In this context, describes a 

deterministic mathematical model optimization of the route of the water supply 

adduction, based on the theory of graphs. 

2. Calculation Algorithm 

Modeling this problem is achieved through representation related directed 

graph G=(X, U) consisting of the source as the origin, route as required arcs and 

points as vertices. For each arc u U is associated a number 

i j ∈ ( ) 0 ≥ 

i 

λ u j , in 

conventional units, depending on the optimization criterion adopted. The route 

is the best way to graph, the minimum value which is determined by applying 

the algorithm Bellman-Kalaba. 

Graph G= (X, U) is attached to a matrix M whose elements mij 

are: 

(1) 

(2) 

( u ) 

⎧ i 

λ j − the arc value 

from xi 

to x j 

⎪ 

= ⎨∞ 

− if vertices x and x are not adjacent . 

i 

⎪ 

0 − for i = j 

⎩ 

mij j 

The optimal route is the best way of graph μ , with the total value: 

λ 

( μ) 

λ( 

) → min 

= ∑ 

u ∈μ 

i j 

i j 

u . 

If the notes with Vi 

the minimum value of the road μn , ( i = 0, 

n ) 

existing from the tip of to the tip of x : 

xi n 

i 

(3) V λ( 

μ ) 

hence: 

i 

= , ( i = 0, 

n ). 

(4) V = 0 , 

n 

n 

i

then, under the principle of optimality: 

(5) V i ( V j + mij 

) 

= 

j≠i 


min , ( i = 0, n −1 

; j = 0, 

n ) and V = 0 . 

To solve system (5) shall be iterative, noting Vi the value of Vi 

obtained at 

iteration k, namely: 

(6) V i = min 

Calculate: 

0 

(7) V i ( V j + mij 

) 

and then: 

0 

1 

= min , ( = 0, n −1 

j≠ 

i 

k 

k −1 

(8) V i = ( V j + mij 

) 

j≠ 

i 

( i = 0, n −1 

); V 0 . 

k 

0 

n = 

i ; j = 0, 

n ) and V 0 . 

n 

1 

n = 

min , ( i = 0, n −1 

; j = 0, 

n ) and V = 0 . 

Ordinal iteration of k expressed by relations (8) gives values only for the 

finite length of roads at most k + 1 arriving at xn 

, choosing between them is the 

minimum. 

From iteration to the next: 

(9) 

k 

−1 

≤ k k 

i Vi 

V , ∀ j . 

Numbers Vi 

( i ≠ n ; k = 0, 

1,... 

) monotone decreasing pattern formed that 

reach to the minimum necessary, after a finite number of iterations which not 

exceeding n −1. 

So algorithm stops when it reaches an iteration k, such that 

+ 1 

= , ( 

k k 

V V i = 0, 

n ), and the minimum between the peaks road and x is 

i i 

k k + 1 

0 = V0 

V 

. To identify which roads have minimum values founded, are 

derived from (8) that have them at the last iteration we have: 

k 

n 

x0 n


(10) 

k 

i 

ij 

k −1 

j 

V = m + V = m + V 

Based on the algorithm described, computer program named BEL_KAL was 

designed in Borland PASCAL language. 

Were made following nomenclatures: N is the order of the graph, V ( I, 

J ) is 

the column vector built at each iteration k, X ( I ) is the sequence of minimum 

road; VAL - the minimum road graph, M ( I, J ) - matrix associated with graph, 

whose elements are: 

(11) 

⎧ 

⎪ 

⎨ 

⎪ 

⎪ 

⎩ 

ij 

i 

λ( 

u j ) − if arc ( xi,x 

j ) 

( u ) − if arc ( x ,x ) 

n 

i 

mij = ∑ λ j 

i j 

i,j= 

1 

0 − 

for i 

= j 

k 

j 

exist 

. 

not exist 

Are introduced as data entry, graph order and its associated matrix, on lines, 

i 

whose elements are considered equal to λ ( u j ) , if there arc ( i j ) x x , , or equal to 

0, otherwise (MATR_EX1.dat file). 

As the data output sequence result road peaks of minimum values and value 

of the road. 

3. Case Study 

Apply dynamic programming method to solve the problem of site selection 

and capture of a main route to supply water through a sequential optimization 

where deterministic, discrete. 

It is considered a line of supply for locality L, departing from two locations 

of S1 and S2 capture (fig. 1). Possible routes through the required A, B, C, D, E, 

forming three sectors. 

Putting the problem of determining the route for which the total cost is 

minimum is determined for each track investments and prepare partial sequence 

graph in Figure 2, where each arc is associated a cost, in conventional units. 

They noted with x1 , 2 and the decision variables for each sector. These 

variables will not take numeric values, but will be vertices of graph that you are 

on the same alignment. 

x 3 x 

.


Fig. 1 – Variants of the adduction route. 

It exemplifies the application of Bellman-Kalaba algorithm to determine the 

optimal bus route for L locality from the source S1 (fig. 1). 

Fig.2 – Graph of adduction routes. 

Modeling problem is achieved through representation related directed graph 

G = (X, U) of order n = 7, consisting of the source point of origin, route as 

required arcs and points as vertices. For each arc u U is assigned a cost in 

i j ∈ 

conventional units (fig. 2) and matrix M attached to graph G = (X, U) are the


elements mi,j defined by relations (1), where λ( 

u ) 

edge u U . 

i j ∈ 

(12) 

1 

2 

3 

M = 

4 

5 

6 

7 

1 

0 

∞ 

∞ 

∞ 

∞ 

∞ 

∞ 

2 

42 

0 

∞ 

∞ 

∞ 

∞ 

∞ 

3 

53 

∞ 

0 

∞ 

∞ 

∞ 

∞ 

4 

∞ 

∞ 

92 

0 

∞ 

∞ 

∞ 

5 

∞ 

73 

61 

∞ 

0 

∞ 

∞ 

6 

∞ 

73 

82 

∞ 

∞ 

0 

∞ 

7 

∞ 

∞ 

∞ 

61 

92 

82 

0 

i j 

is the value attributed to 

V 

61 

92 

82 

0 

0 

i 

∞ 

∞ 

∞ 

V 

1 

i 

∞ 

153 

153 

61 

92 

82 

0 

V 

2 

i 

197 

153 

153 

61 

92 

82 

0 

V 

3 

i 

197 

153 

153 

. 

61 

Route with minimum total cost is given by way of minimum value in this 

graph, which is determined using Bellman-Kalaba algorithm. 

For each V ( k = 0, 

1,... 

is added to previous matrix a column in which 

values are inserted properly. It follows successively: 

k 

i ) 

0 

a) Calculate the values i mi7 

V = (i = 1, ..., 7), which pass into the column 

92 

82 

0 

0 

V i . 

b) Calculate values 1 

V i for i =1,...,7 (j =1,...,7), which is passing in that 

column: 

(13) 

⎧V 

⎪ 

⎪ 

⎪V 

⎪ 

⎪V 

⎪ 

⎪ 

⎨V 

⎪ 

⎪V 

⎪ 

⎪ 

⎪V 

⎪ 

⎪V 

⎩ 

1 

1 

1 

2 

1 

3 

1 

4 

1 

5 

1 

6 

1 

7 

= min 

= min 

= min 

= min 

= min 

= min 

= 0 

j≠1 

j≠ 

2 

j≠3 

j≠ 

4 

j≠5 

j≠ 

6 

0 

0 0 

0 

( V j + m1 

j ) = min( 

V2 

+ m12, 

V3 

+ m13,..., 

V7 

+ m17 

) 

0 

0 

0 

0 

( V j + m2 

j ) = min( 

V1 

+ m21, 

V3 

+ m23,..., 

V7 

+ m27 

) 

0 

0 0 

0 

( V j + m3 

j ) = min( 

V1 

+ m31, 

V2 

+ m32,..., 

V7 

+ m37 

) 

0 

0 

0 

0 

( V j + m4 

j ) = min( 

V1 

+ m41, 

V2 

+ m41,..., 

V7 

+ m47 

) 

0 

0 0 

0 

( V j + m5 

j ) = min( 

V1 

+ m51, 

V2 

+ m51,..., 

V7 

+ m57 

) 

0 

0 0 

0 

( V + m ) = min( 

V + m , V + m ,..., V + m ) 

j 

6 j 

1 

61 

2 

61 

7 

67 

= ∞ 

= 153 

= 153 

= 61 . 

= 92 

= 82

c) Calculate values 

(14) 

⎧V 

⎪ 

⎪ 

⎪V 

⎪ 

⎪V 

⎪ 

⎪ 

⎨V 

⎪ 

⎪V 

⎪ 

⎪ 

⎪V 

⎪ 

⎪V 

⎩ 

2 

1 

2 

2 

2 

3 

2 

4 

2 

5 

2 

6 

2 

7 

= min 

= min 

= min 

= min 

= min 

= min 

= 0 


2 

V i for i = 1, ..., 7 (j = 1, ..., 7) 

j≠1 

j≠ 

2 

j≠ 

3 

j≠ 

4 

j≠ 

5 

j≠ 

6 

1 

1 1 

1 

( V j + m1 

j ) = min( 

V2 

+ m12, 

V3 

+ m13,..., 

V7 

+ m17 

) 

1 

1 1 

1 

( V j + m2 

j ) = min( 

V1 

+ m21, 

V3 

+ m23,..., 

V7 

+ m27 

) 

1 

1 1 

1 

( V j + m3 

j ) = min( 

V1 

+ m31, 

V2 

+ m12,..., 

V7 

+ m37 

) 

1 

1 1 

1 

( V j + m4 

j ) = min( 

V1 

+ m41, 

V2 

+ m42,..., 

V7 

+ m47 

) 

1 

1 1 

1 

( V j + m5 

j ) = min( 

V1 

+ m51, 

V2 

+ m52,..., 

V7 

+ m57 

) 

1 

1 1 

1 

( V + m ) = min( 

V + m , V + m ,..., V + m ) 

j 

6 j 

d) To calculate the values corresponding to column 

..., 7), an analogue, finally are obtained (15). 

2 

2 

i 

3 

i 

1 

61 

2 

62 

7 

67 

= 197 

= 153 

= 153 

= 61 . 

= 92 

= 82 

3 

V i for i = 1, ..., 7 (j = 1, 

Since V = V (i = 1, ..., 7), algorithm stops and the road is the minimum of 

3 

V1 

=V1 

= 197 . This value is reached on the way (1, 2, 6, 7), thus resulting in 

optimal route of adduction as: S1, A, C, and L. To resolve this problem the 

computer program BEL_KAL was used. 

(15) 

⎧V 

⎪ 

⎪ 

⎪V 

⎪ 

⎪V 

⎪ 

⎪ 

⎨V 

⎪ 

⎪V 

⎪ 

⎪ 

⎪V 

⎪ 

⎪V 

⎩ 

3 

1 

3 

2 

3 

3 

3 

4 

3 

5 

3 

6 

3 

7 

= min 

= min 

= min 

= min 

= min 

= min 

= 0 

j≠1 

j≠ 

2 

j≠3 

j≠ 

4 

j≠5 

j≠ 

6 

2 ( V j + m1 

j ) 

2 ( V j + m2 

j ) 

2 ( V j + m3 

j ) 

2 ( V j + m4 

j ) 

2 ( V j + m5 

j ) 

2 ( V + m ) 

j 

6 j 

= 197 

= 153 

= 153 

= 61 . 

= 92 

= 82



1. This study put highlights how the economic fact of a technical problem 

can be optimized using mathematic-informatics structure type graph. 

2. Mathematical model is easily programmable in an evolved language, 

obtaining immediate results. The only problem is populating the matrix attached 

graph. 

3. Bellman-Kalaba algorithm was originally designed for the economy, but 

the adaptation was possible because the optimal path the economy is a virtual 

way and here is a real way. 

Acknowledgements. The present work has been supported by the Romanian 

Government – Ministry of Education, Research and Innovation, The National Centre 

for Programs Management (CNMP) through, CNMP project no. 21-036/2007 and 

CNMP project no. 21-41/2007. 

Received: 


1” Politehnica” University of Timisoara, 

Department of Hydraulic Machinery 

Timisoara, Romania, 

e-mail: teodor.milos@gmail.com 

1. Alexandrescu, A. Concerning optimization’s working of the pumping station 

for water feedings, 18 th International Conference on Hydraulics and Pneumatics, 

Prague, Czech Rep., Sbornik, ISBN 80-02-01567-3, 2003, pp. 263-268. 

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 

Mechanics, Vol. 1, Ed. Academic Horizons, Timisoara, Romania, ISBN 973- 

8391-72-5, 2002. 

3. B e l l m a n , R.; K a l a b a , R. Dynamic Programming and Modern Control 

Theory, McGraw-Hill, New York, 1965. 

4. Nayyar M.L. Piping Handbook, 7 th Edition, McGraw–Hill, ISBN: 978-0-07- 

047106-1, New York, San Francisco, Tokyo, Toronto, 2000. 

5. Sârbu, I. Energetically Optimization of Water Distribution Systems, Ed. 

Academiei, ISBN: 973-27-0575-2, Bucuresti, Romania, 1996.


TRASEUL OPTIM AL UNEI CONDUCTE DE ADUCŢIUNE 

UTILIZÂND ALGORITMUL BELLMAN-KALABA 

(Rezumat) 

Prin amploarea investiţiei şi a consumului de energie, aducţiunile au o pondere 

însemnată în cadrul sistemelor de alimentare cu apă, iar proiectarea raţională a acestora 

comportă mai multe procese de optimizare, în rândul cărora un loc important îl deţine 

optimizarea traseului acestora. 

În practica actuală de proiectare, stabilirea soluţiei optime se face, de obicei, prin 

studierea analitică a două sau trei variante selectate din mulţimea posibilă prin decizii 

intuite, a căror eroare este invers proporţională cu gradul de experienţă al proiectantului. 

Disciplinele matematice moderne prin calculul operaţional pun la îndemâna 

specialistului un vast aparat de analiză ştiinţifică în stabilirea deciziilor optime pentru 

problemele proiectării sistemelor de alimentare cu apă. 

În această lucrare se prezintă modalitatea de aplicare a teoriei grafurilor din 

informatică pentru stabilirea traseului optim a unei conducte de alimentare aflată la 

distanţă mare de obiectivul consumator (reţea de conducte a unei localităţi). 

În acest context, se descrie un model matematic determinist de optimizare a traseului 

unei magistrate de aducţiune a apei, bazat pe teoria grafurilor. Traseul optim este dat de 

drumul de valoare minimă în graf, care se determină aplicând algoritmul Bellman- 

Kalaba. 

Se aplică în cazul când, pe distanţa de la sursă la obiectiv, datorită configuraţiei 

terenului, există mai multe variante de traseu care trec obligatoriu prin nişte puncte 

cheie. În acest fel traseul are n tronsoane, iar pe fiecare tronson costul total (investiţie 

plus exploatare timp de un an) are o anumită valoare. Dacă posibilităţile de parcurgere 

ale traseului sunt mai mult de două atunci aplicarea metodei devine rentabilă. 

Implementarea metodei se face prin intermediul unui program special, utilizând 

mediul de programare Borland Pascal, şi algoritmul Bellman-Kalaba. Rezolvarea 

propriu-zisă este matriceală. Numerotând tronsoanele cu 1…n, în final se obţine ordinea 

optimă de parcurgere ca șir al tronsoanelor selectate. 

În condiţiile actuale, când tot mai multe surse locale de apă potabilă sunt poluate, 

devine tot mai justificată alimentarea din surse îndepărtate, din zone montane cu izvoare 

naturale.


Publicat de 



Secţia 


STRUCTURAL ANALYSIS OF A SLEEVE RIGID 

COUPLING WITH CYLINDRICAL PINS 

BY 

DĂNUŢ ZAHARIEA and MIHAELA TUDORACHE 

Abstract. In this paper the displacements, as well as the von Misses stresses 

of a sleeve rigid coupling with cylindrical pins will be analyzed using CATIA 

Generative Structural Analysis workbench. The structural analysis procedure will 

be presented. The visual representations of the numerical results will be analyzed 

at both the assembly level and the part level. The assembly level analysis allows 

identifying the most critical section of the entire assembly. The part level analysis 

allows observing the results for the driving shaft, the driving pin, the sleeve 

coupling, the driven pin and the driven shaft. 

Key words: sleeve rigid coupling, cylindrical pins, structural analysis, 

CATIA. 


The transmission of the movement of rotation and torque between co-axial 

shafts can be achieved through couplings. From the category of permanent fixed 

couplings in this paper the sleeve rigid coupling with cylindrical pins will be 

analyzed. The geometric characteristics of the sleeve rigid coupling with 

cylindrical pins are shown in Fig.1. The cylindrical pins can be placed in the 

same plane (Fig. 1a), or more often in two different planes at 90 ° (Fig. 1b). 

The shafts diameter can be obtained by the preliminary calculation so as to 

ensure the transmission of a mechanical power P = 8 kW at the rotational speed 

of n = 200 rpm. 

With the calculated values for the angular rate ω = 2πn 60 = 20.944 rad/s; 

for the torque M t = P ω = 381.9719 Nm and for the safety torque 

M tc = kM 

M t = 800 Nm (the safety coefficient kM 

shall be adopted within

216 Dănuț Zahariea and Mihaela Tudorache 

1.5…3, the adopted value is 2.0944), the shafts diameter can be calculated using 

the relationship: 

(1) 

d 

3 16M 

tc = , 

πτ at 

the value d ≅ 0.065m being thus obtained. The material used for all 

components is steel with yield strength σ = 2.5 N/m 2 8 

⋅10 

. 

a b 

Fig. 1 – Constructive types: 

a – with pins in the same plane; b – with pins in two different planes at 90 °. 

The geometric characteristics of the sleeve rigid coupling with cylindrical 

pins can be determined with the relationship: the outside diameter of the sleeve 

coupling D = kDd 

= 0.11 m (the coefficient kD 

shall be adopted within 

1.5…1.8, the adopted value is 1.6923); the length of the sleeve coupling 

L = kLd 

= 0.2 m (the coefficient kL 

shall be adopted within 2…4, the adopted 

value is 3.0769); the diameter of the pins ds = ksd 

= 0.028 m (the coefficient 

ks 

shall be adopted within 0.25…0.44, the adopted value is 0.4308) [1], [2]. 

The sleeve coupling will be checked at the torsional strain with: 

(2) t = 

4 4 

π ( D − d ) 

02 

16DM 

tc 

τ . 

The pins will be checked at the shearing strain with:


4M 

τ = . 

tc 

(3) f 2 

πdds 

The obtained values are within the safety range of the torsional strength and 

the shearing strength of the material. Even if as a result of these calculations a 

good dimensioning of the coupling will be obtained, to answer a series of 

additional issues (the torque asymmetry on the coupling parts, the distribution 

of critical sections on each coupling part, etc.) should be further analyzed by 

advanced numerical methods. 

2. The Structural Analysis of the Sleeve Coupling 

with Pins in the same Plane 

The main steps of the procedure for structural analysis of sleeve couplings 

with cylindrical pins in the same plane using CATIA/Generative Structural 

Analysis environment are: 

1. Creating the 3D model. Shall be carried out separately the five elements 

of the coupling using the Part workbench. The appropriate materials are 

assigned to each element. The assembly of the five coupling elements is 

carried out using the Assembly workbench, Fig. 2a. 

2. Configuring the mesh. Shall be carried out separately for each element, 

by changing the characteristic parameters “Size” and “Sag”, Fig. 2b. 

3. Applying the restraints. On the free extremity of the driven shaft a 

clamp condition must be applied, Fig. 2c. 

4. Applying the loads. On the free extremity of the driving shaft an 800 

Nm torque around the shaft axis will be applied, Fig. 2d. 

5. Applying the conditions for the interaction between the five elements of 

the sleeve coupling. First, the interaction condition of General Analysis 

Connection type will be applied between the driving shaft and the first 

pin, between the first pin and the sleeve coupling, between the sleeve 

coupling and the second pin and finally, between the second pin and the 

driven shaft. Second, for all the active areas in terms of the torque 

transmission the shearing strain condition must be specified by setting 

the just property of interaction between the active coupling elements 

(Bolt Tightening Connection Property), Fig. 2e. 

6. Launching the solver, running the numerical analysis and results 

visualization. The von Misses stress will be presented for both the 

driving and the driven shafts in Fig. 2f (scale factor 500), for both the 

cylindrical pins in Fig. 2g (scale factor 700) and, finally for the sleeve 

coupling in Fig. 2h (scale factor 500).


3. The Structural Analysis of the Sleeve Coupling 

with Pins in two different Planes at 90° 

For the sleeve coupling with pins in two different planes at 90 ° the analysis 

procedures takes the same steps: creating the 3D model (Fig. 3a), configuring 

the mesh (Fig. 3b), applying the restraints (Fig. 3c) and the loads (Fig. 3d), 

applying the interaction conditions (Fig. 3e), running the analysis and results 

visualization for both shafts (Fig. 3f – scale factor 500), for both pins (Fig. 3g – 

scale factor 700) and for the sleeve coupling (Fig. 3h – scale factor 500). 

a b 

c d 

e f 

g h 

Fig. 2 – Analysis procedure steps for sleeve couplings 

with cylindrical pins in the same plane: 

a – creating the 3D model; b – configuring the mesh; c – applying the restraints; d – 

applying the loads; e – applying the interaction conditions; f – von Misses stress for 

shafts; g – von Misses stress for pins; h – von Misses stress for sleeve coupling.


a b 

c d 

e f 

g h 

Fig. 3 – Analysis procedure steps for sleeve couplings with cylindrical pins 

in two different planes at 90 °: 


applying the loads; e – applying the interaction conditions; f – von Misses stress for 

shafts; g – von Misses stress for pins; h – von Misses stress for sleeve coupling.



1. Both constructive solutions comply with the requirements of the 

resistance. Thus, the maximum equivalent stress is 

7 

2. 33⋅ 

10 Pa for the 

coupling with pins in the same plane and 2. 97 ⋅ 10 Pa for the other case. Both 

the maximum stresses are in the shearing areas of the driving pin. For the two 

shafts, the maximum stress areas are placed near the holes. For the sleeve 

coupling the maximum stress areas are placed also near the holes, on the 

cylindrical inside surface of the sleeve. 

2. The two pins respond differently at the applied loads. Thus, the driving 

pin takes a larger effort having a bigger strain. Moreover, the strain profile 

along the pins longitudinal axis is different. 

3. The mesh configuration parameters have a great influence on both the 

hardware resource requirements and the numerical analysis errors. 

Received: March 20, 2010 

7 

“Gh. Asachi” Technical University 

Department of Fluid Mechanics, Hydraulic Machines and Drives 

Iaşi, România 

email: dzahariea@yahoo.com 

email: mihaela_tudorache10@yahoo.com 


1. D e m i a n T., Elemente constructive de mecanică fină. Ed. Didactică şi Pedagogică, 

Bucureşti, 1980. 

2. D r ă g h i c i I., et al., Calculul şi construcţia cuplajelor. Ed. Tehnică, Bucureşti, 

1978. 

ANALIZA STRUCTURALĂ A CUPLAJELOR FIXE DE TIP 

MANŞON CU ŞTIFTURI CILINDRICE 

(Rezumat) 

În lucrare se prezintă etapele analizei structurale şi rezultatele unui studiu cu privire 

la cuplajele fixe de tip manşon cu ştifturi cilindrice plasate în acelaşi plan, respectiv în 

plane decalate la 90°. Analiza efectuată în programul CATIA permite evidenţierea 

rapidă a unor aspecte particulare care ar putea fi puse în evidenţă cu dificultate printr-un 

calcul clasic de rezistenţa materialelor: asimetria distribuţiei momentului, asimetria 

stării de deformare, influenţa concentratorilor de tensiune, poziţia secţiunilor critice 

pentru fiecare element al cuplajului. Modificarea factorului de scală şi vizualizarea 

animată a tensiunilor echivalente şi a deformaţiilor permit interpretări interesante cu 

privire la evoluţia în timp a stării de solicitare şi a tendinţei reale de deformaţie a 

cuplajului fix de tip manşon cu ştifturi cilindrice.


Publicat de 



Secţia 


STRUCTURAL ANALYSIS OF A 

BIMETALLIC STRIP THERMOSTAT 

BY 

DĂNUŢ ZAHARIEA and MARIUS STACHIE 

Abstract. In this paper the structural analysis of a bimetallic strip will be 

analyzed using CATIA Generative Structural Analysis workbench. The structural 

analysis procedure will be presented. Six different bimetallic strips have been 

considered with titanium as active layer and zinc, aluminum, bronze, brass, copper 

and steel as passive layer. The numerical analysis has been performed under the 

same global conditions (geometrical characteristics, restraints and thermal load) 

using two variable parameters: the width and the length of the beams. For all these 

study cases only the thermal load generated by an imposed temperature field has 

been considered. Charts are presented for comparative analysis of the influence of 

the material, the width and the length of the beams on the von Misses stress and 

the maximum deflection. 

Key words: thermostat, bimetallic strip, structural analysis, CATIA. 


The bimetallic springs used as sensitive elements of the thermostats are 

made by two beams of rectangular cross section bounded together. The two 

beams are made of materials with different coefficients of thermal expansion, 

α a > α p . Under a uniform temperature field Δ T the bimetallic strip will curve 

toward the passive material (the material with lower coefficient of thermal 

expansion). The sensitivity of the bimetallic strip will be better as long the 

difference α a − α p will be higher. In this paper, the materials used are: titanium 

as passive layer and zinc, aluminum, bronze, brass, copper and steel as passive 

layer. Characteristics of materials relevant for the analysis are shown in Table 1. 

The main geometric characteristics of the bimetallic strip are shown in Fig. 1.

222 Dănuț Zahariea and Marius Stachie 

Passive 

layer 

Table 1 

Material characteristics 

Active layer 

Titanium Zinc Aluminum Bronze Brass Copper Steel 

E 10 11 , [Pa] 1.14 0.97 0.7 1.1 1.31 1.1 2 

α 10 -6 , [°C -1 ] 9.5 31.2 23.6 17.8 16.7 16.5 11.7 

σ ,[MPa] 825 140 95 520 350 290 250 

02 

Fig. 1 – Geometric characteristics. 

In the paper two comparative analyses will be performed. First one is 

studying the influence of the beams width b = { 1, 

5, 

7. 

5} 

mm for the same beams 

length L = 50 mm. The second one is studying the influence of the beams 

length L = { 50, 

100 , 150} 

mm for the same beams width b = 5 mm. The beams 

thickness will be the same for all analyzed cases h = h = 1 mm. 

2. The Structural Analysis Procedure 

The main steps of the procedure for structural analysis of the bimetallic strip 

using CATIA/Generative Structural Analysis environment are: 

1. Creating the 3D model. Shall be carried out separately the two 

rectangular cross section beams using the Part workbench. The 

appropriate materials are assigned to each element. The assembly of the 

two beams is carried out using the Assembly workbench, Fig. 2a. 

2. Configuring the mesh. Shall be carried out separately for each beam, by 

changing the characteristic parameters “Size” and “Sag”, Fig. 2b. 

3. Applying the restraints. At one side of the bimetallic strip a clamp 

condition must be applied, Fig. 2c. 

4. Applying the loads. On the both beams a thermal load with ΔT = 30 °C 

will be applied, Fig. 2d. 

5. Applying the conditions for the interaction between the two beams of 

the bimetallic strip. First, the interaction condition of General Analysis 

Connection type will be applied between the active layer and the 

passive layer. Second, for the active area in terms of the stress/strain 

a 

p


transmission (the bounded surface) the stress/strain condition must be 

specified by setting the just property of interaction between the active 

and the passive layers (Fastened Connection), Fig. 2e 

6. Launching the solver, running the numerical analysis and results 

visualization. The maximum deflection will be presented for the 

bimetallic strip in Fig. 2f (scale factor 100). The von Misses stress will 

be shown in Fig. 2g (scale factor 100) for active layer and in Fig. 2h 

(scale factor 100) for passive layer. 

a b 

c d 

e f 

g h 

Fig. 2 – Steps of structural analysis procedure: 


applying the loads; e – applying the interaction conditions; f – maximum deflection; g – 

von Misses stress for the active layer; h – von Misses stress for the passive layer.


3. Comparative Analysis 

After a preliminary analysis, the clamped end of the bimetallic strip has 

been identified as being critical in terms of stresses. In order to improve the 

computational process the local changing of the mesh parameters at the clamped 

end is required. Thus, for the critical area the “Size” mesh parameter will be 

0.25. The global mesh parameters are „Size”=1 and „Sag”=0.5. 

The first set of comparative analyses is determining how the width of the 

beams influences the behavior of the bimetallic strip. Thus, for every 

combination of materials changing the width of the beams and maintaining 

constant the other parameters were obtained the numerical values for maximum 

deflection (Fig. 3a) and for the von Misses stress on the active layer (Fig. 3b), 

as well as for the passive layer (Fig. 3c). 

a 

b c 

Fig. 3 – The influence of the width of the beams: 

a – maximum deflection; b – von Misses stress for active layer; 

c – von Misses stress for passive layer. 

The second set of comparative analyses is determining how the length of the 

beams influences the behavior of the bimetallic strip. Thus, for every 

combination of materials changing the length of the beams and maintaining 

constant the other parameters were obtained the numerical values for maximum


deflection (Fig. 4a) and for the von Misses stress on the active layer (Fig. 4b), 

as well as for the passive layer (Fig. 4c) 

a 

b c 

Fig. 4 – The influence of the length of the beams: 

a – maximum deflection; b – von Misses stress for active layer; 

c – von Misses stress for passive layer. 


1. The width of the beams has a negligible influence on the maximum 

deflection of the bimetallic strip. As far as the status of the stresses, it becomes 

apparent that with width increasing, the von Misses stresses grow as well in 

both the active and the passive layers. For the width of 7.5 mm, the von Misses 

stresses are very close to the admissible values, even overcoming for a small 

amount in the case of zinc material. Regardless of the value of the width of the 

beams, for all combination of materials, the stress in the active layer is greater 

than the passive layer. 

2. The length of the beams is definitely affects the values of the maximum 

deflection. The von Misses stresses are growing as well, but to a lesser extent 

than the growth of the maximum deflection. And in this case, the von Misses 

stress in the active layer is towards the passive layer.


3. The best combination of materials in terms of deflection is titan for 

passive layer and zinc for the active layer. Due to the smaller mechanical 

resistance of aluminum and zinc, increase the width of beams over a certain 

limit becomes dangerous, especially in the case of accidental temperature 

increases. 

4. For all cases analyzed the maximum von Misses stress shall be recorded 

in the clamped area. Another critical area is the interaction area between the two 

beams of the bimetallic strip. 

Received: March 15, 2010 “Gh. Asachi” Technical University 

Department of Fluid Mechanics, Hydraulic Machines and Drives 

Iaşi, România 

email: dzahariea@yahoo.com 

email: marius_stachie@yahoo.com 


1. D e m i a n T., Elemente constructive de mecanică fină. Ed. Didactică şi Pedagogică, 

Bucureşti, 1980. 

2. Z a m a n i N.G., CATIA V5 FEA Tutorials. SDC Publications, 2005. 

ANALIZA STRUCTURALĂ A ARCURILOR 

BIMETALICE LAMELARE 

(Rezumat) 

În lucrare se prezintă analiza structurală a unui arc bimetalic lamelar utilizând 

mediul de analiză CATIA Generative Structural Analysis. Sunt analizate şase arcuri 

bimetalice lamelare având ca material pasiv titanul, iar ca material activ zincul, 

aluminiul, bronzul, alama, cuprul şi oţelul. Analiza numerică a fost efectuată pentru 

aceleaşi caracteristici globale (dimensiuni geometrice, condiţii pe contur şi sarcini) 

folosind doi parametri de control: lăţimea şi grosimea arcurilor lamelare. Sunt 

prezentate diagrame comparative pentru analiza influenţei materialului, a lăţimii şi 

grosimii arcurilor lamelare asupra deformaţiei maxime şi a stării de solicitare a arcurilor 

bimetalice analizate. Principalele concluzii sunt: sensibilitatea maximă se înregistrează 

pentru combinaţia de materiale titan-zinc; zonele puternic solicitate sunt încastrarea şi 

interfaţa dintre cele două lamele; tensiunile echivalente mai mari se înregistrează în 

stratul activ; lăţimea lamelelor influenţează în principal starea de solicitare, în timp ce 

lungimea acestora influenţează cu precădere valorile săgeţii maxime a arcului bimetalic 

lamelar.


Publicat de 



Secţia 


KNOWLEDGE MANAGEMENT FOR THE CONFIGURATION 

IN EARLY PHASES OF COMPLEX CUSTOM PRODUCTS 

BY 

IRÈNE ALEXANDRESCU, HANS-JOACHIM FRANKE 

and THOMAS VIETOR 

Abstract. Nowadays, large and saturated markets in developed countries enforce 

the production of a broad palette of customized products, which satisfy the customers’ 

individual needs. In addition, interesting emerging markets together with the constraints 

of the current economic crisis lead to further differentiation in terms of product 

functionality and allowable costs. Furthermore, global markets also imply many different 

standards and norms, lately more strict due to the worldwide environmental concerns. In 

order to be effective in producing custom products, companies in the field of mechanical 

engineering integrate in their product lifecycle technologies the newest parametric 

CAD/CAM systems, as well as different software for the sales and design departments, 

such as configuration or calculation tools. However, producing high quality products with 

relatively short delivery times requires that a common knowledge base exists between the 

enterprise departments and clear processes and interfaces are defined, integrating all the 

used systems. This is commonly not the case and therefore unnecessary iterations are 

done in order to produce a technically correct offering based on the specific customer 

request. This paper discusses the need and some ideas for a better methodology to faster 

and better design complex custom products in the field of mechanical engineering. It 

explains the need for a structured knowledge base prior to the beginning of the new 

custom product design process. This knowledge base offers the necessary data for the 

early design phases and connects the sales and design departments by ensuring a correct 

and transparent information transfer between the customer requirements, the configuration 

software, the used parametric CAD tools and the generated customer specific offers. The 

result is a faster, optimized and technically correct basic configuration of custom products 

in early design phases. 

Key words: complex custom products, early design phases, product configuration, 

knowledge management. 


The transfer from industrial to information and knowledge societies 

radically transformed the markets and increased customer expectations.

228 Irène Alexandrescu et al. 

Globalization and internationalization differentiated the spectrum of offered 

goods and increased the need for custom solutions. There are many different 

standards and norms coming from saturated markets or interesting emerging 

ones, which have to be satisfied with consideration to the massive constraints 

additionally imposed by the economic crisis and the environmental concerns. 

Product specific knowledge management nowadays is one of the main 

factors for the sustainable competitive advantage of a manufacturing company. 

It is an important differentiator for the unique core competences of the 

company, as well as the essence for developing successful, knowledge based, 

market oriented products and services. 

In order to achieve customer satisfaction, companies in the field of 

mechanical engineering producing custom solutions have to master their 

knowledge and generate fast, optimized and technically correct offers for their 

products. This paper describes how this can be achieved by using virtualization 

and automatization techniques on a methodically structured product knowledge 

base. 

2. Scattered Knowledge Base in Early Design Phases 

In order to deal with the increasing requirements and the competitive 

markets, companies use for the generation of the offering of complex custom 

products several software tools, such as: Computer Aided Design and 

Manufacturing (CAD/CAM) Software, Product Data/Lifecycle Management 

(PDM/PLM) solutions, Enterprise Resource Planning (ERP) systems, 

configuration tools and calculation programs. Common applications are in the 

aircraft [1] or automotive industry [2]. 

However, through the variety of software, the complexity of the internal 

processes is increased. A good overview to the state of the art is shown in the 

CIRP Proceedings [3]. 

Unfortunately, in praxis, the product knowledge base is scattered through 

these systems and normally in each development phase the product knowledge 

is documented in separate specific systems. The existing PLM/PDM systems 

are often not properly used and there is seldom a platform available to 

interconnect or translate the different pieces of information, for a global review 

or decision taking. 

Fig. 1 shows a sequence diagram for the standard processes that take place 

in the early development phases for the generation of a customized offer; the 

involved departments and their individual software are also displayed.

Customer 

(online configurator) 

Sales 

(configurator, ERP,...) 

custom 

request 

Design 

(CAD, FEM, ERP, PDM, MBS,...) 

Part Design (e.g. Hydraulic) 

(CAD, FEM, calculation, PDM,...) 

Purchasing 

(ERP,...) 

Production 

(CAM, NX-Daten,...) 

record 

request 

Legend: 


request 

response 

analyze 

characteristic 

request 

curves 

request response 

design 

part 

consultation 

request response 

Department 

(used software) 

clarification 

timeline 

preliminary offering 

schedule 

delivery 

date 

document action 

Fig. 1 – Sequence diagram of the standard processes 

for the generation of a customized offer. 

offering 

order 

placed 

decide 

Currently the customer describes to the sales department the requirements 

for the desired product. The sales department interprets the customer 

requirements and communicates them further to the design department. There is 

normally no sufficient common platform for communication and therefore 

unnecessary iteration cycles are done. This can be improved through direct 

interfaces between the customer and the two departments, by using a common 

configuration tool, which generates department specific product information 

and documentation. This solution will be discussed further in the next section. 

Another common problem that occurs in the processes described in Fig. 1 is 

that generally the project information and awareness remain only in a small 

group of specialists. They are not reused for later following projects and are 

unavailable for new employees. The company specific knowledge base would 

help in solving this problem, by documenting existent intellectual property of 

the employees into the pre-structured product specific knowledge. 

A good example of a holistic knowledge framework to design and 

implement a knowledge based organization is presented by B i n n e r in [4]. This 

paper concentrates on the early design phases and can be seen as a methodology 

for the previous steps taken in knowledge frameworks systems. 

It is therefore not necessary to consider the entire company’s knowledge 

base, but rather only the product specific knowledge used in early design phases 

to generate a correct first offer. Such comprehensive and integrated knowledge 

management solutions are currently almost absent in practice.


3. Suggested Product Configuration Chain 

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 

faster and better generation of offers is a complete and integrated system that 

sustains the whole configuration chain, from the client request analysis to the 

configured model of the product. The suggested product configuration chain is 

shown in Fig. 2. 

Fig. 2 – Methodology for faster and better design of high quality 

complex customized products. 

In order to shorten the offering generation time, both the client request 

clarification and the product design process have to be accelerated and the 

quality of the transferred information has to be improved. The solution is a 

software system that integrates the product data with existing virtual methods in 

order to partially automate the configuration. If a customized product is needed, 

a completely automated configuration system is impossible. The 

algorithmization of the design process is elaborately discussed by F r a n k e in 

[6]. However, it is possible to accelerate and optimize the configuration chain 

through better requirement definition, product parameter visualization and goal 

conflict analysis. 

Fig. 3 presents the implementation solution – the main software tool is the 

configuration module, which is used by both sales and engineering design 

departments in order to configure a custom solution. The output of both 

departments is the customized offer information (such as prices, conditions, 

related product pictures and site plans) and the customer specific technical 

information (such as exact technical description and parametric 3D-CADmodels 

as starting point for a complete and precise product configuration. The 

product specific knowledge base is the fundamental information source for the 

whole configuration process. Its requirements and structure are presented in the 

next section.


Fig. 3 – The knowledge based configuration module, as 

a main software component of the solution. 

4. Knowledge Base Structure 

Before defining the company specific knowledge base structure for the 

configuration in early phases of a customized product, the following three 

concepts need to be defined: data, information and knowledge. The common 

interpretation is: 

- Data is unprocessed facts and figures without any added interpretation 

or analysis. 

(e.g. "") 

- Information is data that has been interpreted so that it has meaning for 

the user. 

Information = data + semantic + context 

(e.g.: "The document A was updated.") 

- Knowledge is a combination of information, experience and insight that 

may benefit the individual or the organization. 

Knowledge = information + linking 

(e.g.: "B was updated after A, so B contains the latest information.") 

The above mentioned three different concepts are presented in Table 1, 

according to the description made by B o d e n d o r f in [7]: information is just 

a formal step to transforming the data into in the knowledge. 

Table 1 

The different concepts of data, information and knowledge according to [7] 

Data Information Knowledge 

structured ⇔ mixed 

isolated ⇔ connected 

context independent ⇔ context dependent 

symbols ⇔ cognitive pattern of action


The context for the interpretation of the data has a strategic role in the 

knowledge output. The discussed company specific knowledge base for the 

product configuration should therefore consider both involved sales and design 

departments context and semantic. It is therefore important to recognize the 

different views, as shown in Table 2. Nevertheless, the common goal is to 

configure as fast as possible a high quality customized product, with minimum 

resources, minimum costs and minimum losses. 

Keeping this in mind, the following structure for a common knowledge base 

is proposed: 

Fig. 4 – Approach for the knowledge base.


Table 2 

Key aspects for the sales and design departments 

Departments: Sales Engineering design 

Specialization: Marketing Engineering 

Modules: Functional modules Product parts and assemblies 

Software: Configuration tool Virtualization and calculation tools 

Goals: Sell the product Design with minimum costs 

Degree of content: Sale probability Technical feasibility 

In the center of the knowledge base is the abstract representation of the 

product structure, from a neutral point of view. This representation is company 

and product specific and it represents the basis for all the department specific 

derived views. As exemplarily presented on the right side of the picture, the 

module Mi in the main structure will contain its technical (e.g. for the design 

department), as well as its functional (e.g. for the sales department) description. 

The module has additional parameters and logic, for other derived views, as for 

example for geometry calculation, validation of a module combination or 

visualization of dependencies systems. 

Further research is currently done to describe how at early stage goal 

conflicts can be recognized and solved based on this structure. 

5. Benefits of Using the Presented Knowledge Base 

The main benefits of the presented solution and its knowledge base are: 

- It reflects the company global view on its product structure definition; 

- The collection and preservation of knowledge can be carried out without 

strong efforts, as the configuration system is used on a daily basis; 

- The sales and design departments are interacting better; 

- The customized offers can be faster and technically more accurately 

generated; 

- There is a methodic, systematic and goal oriented approach towards a 

common context for the configuration. 


1. The need for customized products has increased due to market 

globalization, internationalization and customer awareness. For staying 

competitive, companies have to provide custom solutions for low prices, high 

product quality and short lead times. 

2. The knowledge necessary for configuring customized products in early 

design phases is often scattered through numerous software tools and separated


departments. Comprehensive and integrated knowledge base solutions rarely 

exist in praxis. 

3. A product configuration chain for an accelerated and technically correct 

generation of offers was presented. The main modules of the implementations 

are the configuration software and its knowledge base. 

4. The product specific knowledge base structure for the configuration in 

early phases was discussed. 

5. The benefits of the presented knowledge base usage were listed. 

Acknowledgements. The authors gratefully thank the German Federal Ministry of 

Education and Research (BMBF) for supporting the Project KOMSOLV - 'Fast Offer 

and Optimized Design for Complex Products with Conflicting Requirements'. 

Received: Technische Universität Braunschweig, 

Institute of Engineering Design 

Braunschweig, Germany 

e-mail: i.alexandrescu@tu-bs.de 


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 

Digital Design Data Availability in the Aerospace Industry. In Proceedings of 

the 19th CIRP Design Conference, p. 589-590, 2009. 

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 

Model Generation for Optimization and Robust Design of Passenger cars. 

Advances in Engineering Software, 38: 795-801, 2007. 

3. K r a u s e F. -L., The Future of Product Development, Proceedings of the 17 th CIRP 

Design Conference, Springer, 2007. 

4. B i n n e r H. F., Ganzheitiches Wissenskonzept - Wissensframework zur Gestaltung 

und Implementierung einer wissensbasierten Organisation. In ZWF Zeitschrift 

für wirtschaftlichen Fabrikbetrieb, Carl-Hanser Verlag, München, 103 (2008), 

p. 540-543. 

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 

Complex Custom Products. 1st Symposium on Multidisciplinary Studies of 

Design in Mechanical Engineering, p. 11-12, 2008. 

6. F r a n k e H. -J., Untersuchungen zur Algorithmisierbarkeit des Konstruktionsprozesses. 

VDI Verlag, 1976. 

7. B o d e n d o r f F., Daten- und Wissensmanagement. Springer, p. 2, 2003.


MANAGEMENTUL CUNOAȘTERII ÎN CONFIGURAREA 

PRODUSELOR COMPLEXE PERSONALIZATE ÎN FAZELE 

INCIPIENTE ALE DEZVOLTĂRII ACESTORA 

(Rezumat) 

În prezent, pieţele mari şi saturate ale ţărilor dezvoltate orientează fabricația spre o 

paletă largă de produse personalizate, în vederea satisfacerii nevoilor clienţilor 

individuali. Pieţele emergente interesate, confruntate cu constrângerile crizei economice 

actuale, conduc în plus la o diferenţiere a funcţionalității produselor şi a costurilor 

admisibile. Pieţele globale implică respectarea de standarde şi de norme diferite, însă 

acestea sunt din ce în ce mai stricte, datorită preocupărilor de mediu la nivel mondial. 

Pentru a fi eficiente în fabricația de produse personalizate, companiile din domeniul 

ingineriei mecanice integrează în tehnologiile ciclului de viaţă al produsului nou sisteme 

parametrice de tip CAD/CAM, precum şi multiple aplicații software pentru 

departamentele de vânzări şi de proiectare. 

Cu toate acestea, fabricația de produse de înaltă calitate cu termen de livrare relativ 

scurt impune existența unei baze de cunoştinţe comune tuturor departamentelor 

întreprinderii, proceduri clare şi interfeţe bine definite, care să integreze toate sistemele 

utilizate. Acest aspect nu se regăsește însă întotdeauna în practică, prin urmare apar 

iteraţii inutile ocazionate de elaborarea unei oferte corecte din punct de vedere tehnic, în 

baza unei cereri specifice a clientului. 

În lucrarea noastră abordăm problema unei noi metodologii, inovative și rapide, de 

design personalizat al produselor complexe din domeniul ingineriei mecanice. 

Abordarea noastră explică necesitatea structurării unei baze de cunoştinţe încă din 

fazele incipiente ale procesului de design al unui nou produs. Această bază de 

cunoştinţe oferă datele necesare pentru fazele incipiente de proiectare şi se conectează la 

departamentele de vânzări şi de proiectare prin asigurarea unui corect şi transparent 

transfer de informaţii între cerinţele clientului, software-ul de configurare, instrumentele 

parametrice utilizate în CAD şi oferta specifică generată de client. Rezultatul cercetării 

constă în creșterea rapidității, corectarea tehnică și optimizarea configuraţiei de bază a 

produselor personalizate încă din fazele incipiente ale dezvoltării acestora.


Publicat de 



Secţia 


RESEARCHES REGARDING TO INNOVATIVE 

ACTIVITY IN TECHNICAL ENVIRONMENT 

BY 

PETRU DUŞA and IULIANA LAURA TARANOVSCHI 

Abstract. The innovation concept can be concerned in terms of five basic 

elements: process, product, human factors, domain and socio-organizational 

environment. The study is conceived to stimulate the innovative potential and 

combine Bloom’s Cognitive Taxonomy with the creative model of self-managed 

team, following an anallogy with Contradictions Matrix for solving the innovative 

problems developed by Altschuller. The new model to stimulate the innovative 

potential it is described by the Innovative Acting Matrix and by the Innovative 

Support Matrix. 

Key words: innovation, creative self-managed team model, Bloom’s 

taxonomy, Contradictions Matrix, Innovative Acting Matrix, Innovative Support 

Matrix. 


In the literature, the word innovation is more favored than the concept of 

creativity [1], [6]. Many definitions show the value, the importance and the 

time intensity of innovation [8], [12]. 

Innovation is often described as the value-adding stage of the creativity 

process, suggesting a higher sense of value for innovation. Creativity is a precondition 

from which innovation develops, is a new realization in practice 

supported by creative thinking [8]. 

From the analysis of the information from literature, the innovation concept 

can be concerned in terms of five basic elements: process, product, human 

factors, domain and socio-organizational framework. If it could be demystified, 

described and modeled the creative process, it would be able to enhance 

individual innovative potential and facilitate the innovative process. When the

238 Petru Duşa and Iuliana Laura Taranovschi 

innovative process is facilitated, the innovative products are developed. These 

are, often, considered to be the result of collaborative teamwork. Increasing the 

organizational effectiveness in a specific domain is a primary goal for selfmanaged 

work teams. Self-managed teams are so intensively focused on high 

performance that often individual needs of the team members can be 

overlooked. As a result of the integration of viewpoints, self-managed teams 

offer multifunctional definitions and solutions to problems that generate 

innovative products or services. 

In this paper are explored the cognitive factors which influence the creative 

thinking of individuals who work in an organizational team and in an 

organization with high creative potential. 

1. Conceptual Determination 

In order to understand the manner in which the factors involved in the 

innovative process operate, it is used as starting point Bloom’s Taxonomy for 

educational objectives. 

In 1956, Benjamin Bloom, an 

educational psychologist, developed 

taxonomy for Educational Objectives. 

This taxonomy became a key tool in 

structuring and understanding the 

learning process. The taxonomy, 

examines the cognitive domain, 

which categorizes and orders thinking 

skills and objectives. It is a 

continuum of six domains: 

knowledge, comprehension, 

application, analysis, synthesis, 

evaluation. The domains are arranged 

Fig. 1 – Bloom’s Taxonomy [3]. 

in ascending order, each level is 

assimilated in the others’ presence, 

starting from the basic level - 

knowledge thought the higher one – evaluation, as it is represented in the Fig. 1. 

These depend one from another can not achieve one domain without other and 

are arranged at three levels: cognitive domain, corresponding verbs and 

cognitive activity support. Each cognitive domain, that is found in the center of 

the model presented in the Fig. 2, are in correspondence with a set of verbs 

located in the middle of the model and with a set of nouns situated at their 

external level in this model. The set of verbs express how can be acted in order 

to attain the corresponding domain. Nouns express the support that it is offered 

to perform the activities for each domain.


Fig. 2 – Bloom’s Taxonomy [4]. 

This paper aims to structure a model in which cognitive elements of 

Bloom’s taxonomy can be integrated with another model which is specific for 

activities of a team with high innovative potential. 

2. Case Analysis 

From previous studies in the area of the self-managed team concept, situated 

in a research environment from industrial engineering domain, resulted a model 

of creative self-managed team. To define this model, we used the following 

tools: General Innovation Skills Aptitude Test - GISAT [5], Belbin test to 

determine the role in a team [2] and Building Blocks focused on variables of 

effectiveness teams [13]. The model was developed on the most significant 

correlations between subjects’ scores to the three dimensions: innovative 

profile, team roles and team variables. 

The scores obtained for subject (S3) identified as Monitor-Evaluator and 

subject (S15) identified as Plant from the perspective of the roles assumption 

correlate with the scores obtained for first pillar - Generating and Assessing 

Ideas from the innovative profile and with the team variables: Openness and 

Confrontation, Support and Trust, Cooperation and Conflict (see on Fig. 3, the 

quadrant for S15, S3). The scores obtained for subject (S11) identified as 

Implementer and subject (S16) identified as Shaper from the perspective of the 

roles assumption correlate with the scores obtained for second pillar - Risk- 

Taking from the innovative profile and with the team variables: Judicious 

Procedures, Appropriate Leadership and Regular Review (see on Fig. 3, the


quadrant for S11, S16). 

The scores obtained for subject (S7) identified as Resource Investigator and 

subject (S10) identified as Coordinator from the perspective of the roles 

assumption correlate with the scores obtained for third pillar - Relationship 

Building from the innovative profile and with the team variables: Individual 

Development, Sound Inter-groups Relations and Good Communication (see on 

Fig. 3, the quadrant for S7, S10). The scores obtained for subject (S4) identified 

as Finisher and subject (S12) identified as Team Worker from the perspective of 

the roles assumption correlate with the scores obtained for forth pillar – 

Implementing and Turning Ideas into Products, Processes and Services from the 

innovative profile and with the team variables: Balanced Function, Clear 

Objectives, Organizational Support (see on Fig. 3, the quadrant for S4, S12). 

Fig. 3 – The model of self-managed team [9]. 

In the literature, can be identified a method to stimulate the creativity – 

TRIZ “The Theory of Inventing Problem Solving”, comes from Russian 

Language “Teoriya Resheniya Izobretatelskikh Zadatch”, developed by 

Genrich Altshuller, Russian author. The method was developed after analysis of 

almost two million of the world’s most successful patents. TRIZ is a method 

which provides resources to access the best solutions to solve the innovative 

problems. In this method can be identified four pillars that differentiate this 

strategy from other innovative strategies for solving problems [7]: 

• Pillar 1: Contradictions – many problem solvers try going directly 

from problem to solution through trial and error. Looking at an analogous 

standard problem and its associated standard solution is a more efficient


approach. It was determined that the most efficient innovative solutions grow up 

from contradictions. TRIZ Matrix operates with the contradictions principles. 

This is designed on the basis of 39 factors organized on rows and columns, 

represented in the Fig. 4, and at intersection between rows and columns can be 

identified the contradictions. For each contradiction it is presented a set of 

principles, from a total of 40, in order to solve it. If the problem is structured on 

the contradictions principle, the TRIZ matrix offers sets of innovative 

principles. The essential contribution of the user’s matrix is to match the 

standard principles to that innovative problem [10]. 

Fig. 4 – Extract from Contradiction Matrix [7]. 

• Pillar 2: Ideality – when TRIZ founder studied the patents database, he 

discovered that the systems always tend to evolve towards increasing “ideality”. 

The ideality part of TRIZ encourages problem solvers to break out of the 

traditional “start from the current situation” type of thinking, and start instead 

from what is described as the Ideal Final Result (IFR). The simple definition of 

IFR is that the solution contains all of the benefits and none of the costs or 

“harms” (environmental impact, adverse side-effects, etc) [7]. 

• Pillar 3: Functionality – this concept is developed by TRIZ through the 

fact that it integrates knowledge from physics, chemistry, mathematics, 

engineering, management and political science. It is successful largely because 

that knowledge is in a form accessible to all users. In this way, the functionality 

becomes the connection that makes possible sheering of knowledge between 

widely differing industries [11]. 

• Pillar 4: Use of resources – in TRIZ terms relates to the unprecedented 

emphasis placed on the maximization of use everything contained within a


system. Thereby a resource is anything in the system which is not being used to 

its maximum potential. Discovery of such resources then reveals opportunities 

through which the design of a system may be improved [7]. 

3. Case Analysis Results 

From the TRIZ analogy exploration, it can be combined the Bloom’s model 

presented in Fig. 2 with the self-managed presented in Fig. 3. The role of this 

analogical approach is to define a model to stimulate the innovative potential. 

In a first stage it is built a matrix structure having as columns the pillars 

from the model of self-managed team and as rows the cognitive domains from 

Bloom’s Taxonomy. In the cells resulted at the intersection of rows and 

columns, are inserted the Bloom’s domain verbs from the relative line. So, it is 

obtained a matrix that offers the necessary stimulants for innovative activity. 

The matrix is presented in Figure 5 as the Innovative Acting Matrix. The 

importance of this matrix is that, those who access it, being in situations marked 

by one of the innovative pillars, have the opportunity to explore the innovative 

problem in an acting style. The exploration in an acting manner of the 

innovative problem is achieved gradually from the base level - knowledge to a 

superior level - evaluation. 

Fig. 5 – Innovative Acting Matrix. 

In the second stage, on the same matrix structure described above, in the 

cells from the intersection of rows and columns, are inserted the nouns of the 

Bloom’s domain from the relative line. It is obtain a matrix that provides the 

necessary support for innovative activity. The matrix is presented in Figure 6 as 

the Innovative Support Matrix. The importance of this matrix is that, those who 

access it, being in situations marked by one of the innovative pillars, have the 

opportunity to receive the necessary resources to explore the innovative 

problem.


Fig. 6 – Innovative Support Matrix . 

The exploration of innovative problem is realized, by this time, using the 

tools provided as support for each acting-innovative stage. 

5 Conclusions 

1. As a result of this scientific approach, it was designed a model to 

stimulate the innovative potential with the support of two innovative matrices: 

Innovative Acting Matrix; Innovative Support Matrix. 

2. Innovative Acting Matrix represents a matrix that offers the necessary 

stimulants for innovative acting (the verbs). 

3. Innovative Support Matrix represents a matrix that offers the necessary 

support for innovative results (the nouns). 

4. Next research will focus on the applicability of these matrices in 

situations that require the innovative potential. 

Received: February 25, 2010 “Gheorghe Asachi”Technical University, 



e-mail: itaranovschi@tcm.tuiasi.ro 


1. A m a b i l e T. M., From individual creativity to organizational innovation. In 

Innovation: A cross-disciplinary perspective. (Gronhaug K., Kaufmann G., 

Eds.), Scandinavian University Press, Oslo-Norway, pp. 139-166, 1988. 

2. B e l b i n R. M., Management Teams. Butterworth Heinemann Publisher, London- 

UK, 1981.


3. Bloom’s Taxonomy, Consulted on 11.12.2009 at: 

http://blogs.wsd1.org/etr/files/blooms_taxonomy.jpg. 

4. Bloom’s Taxonomy, Consulted on 15.12.2009 at: 

http://mysciencelessons.files.wordpress.com/2009/07/bloomwheel3.gif. 

5. C a m p b e l l A., W a t t D., Building innovation capacity one skill at a time. Proc. 

of The Conference Board, Montreal-Canada, 2004. 

6. C a v a l l u c c i D., TRIZ, the Altshullerian approach to solving innovation 

problems. In Engineering design synthesis: Understanding approaces and tools. 

(Chakrabarti A., Eds.), Springer Verlag, London, pp. 131-149, 2002. 

7. CREAX Innovation Suite, Consulted on 03.12.2009 at : 

http://www.creax.com/triztab/creaxsuite31/english/TRIZ/TRIZ.html. 

8. L a n d r y C., The creative city: A toolkit for urban innovators, Earthscan 

Publications, London, 2001. 

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 

environment. Proc. 14 th International Conference ModTech, Slănic Moldova- 

România, 2010. 

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 

Introduction to TRIZ. CRC Press LLC, USA, 1998. 

11. V i n c e n t J., M a n n D., Systematic technology transfer from biology to 

engineering. In Philosophical transactions - Royal Society. Mathematical, 

physical and engineering sciences. Royal Society Publisher, London, 2002. 

12. W e s t A. M., F a r r J. L., Innovation at work. In Innovation and creativity at 

work: Psychological and Organizational Strategies, John Willey & Sons, New 

York, pp. 3-13, 1990. 

13. W o o d c o c k M., Team Development Manual. Gower Publisher, Aldershot, UK, 

1989. 

CERCETĂRI CU PRIVIRE LA ACTIVITATEA DE 

INOVARE DIN MEDIUL TEHNIC 

(Rezumat) 

Din cercetările anterioare asupra elementului inovativ, am ajuns la concluzia că 

structura care facilitează dezvoltarea acestuia o regăsim la nivelul echipei autoconduse. 

Modelul de echipă autocondusă propus pentru stimularea potentialului inovativ este 

susţinut de patru piloni: Generare de Idei, Asumarea Riscului, Construire de Relaţii şi 

Implementare. În momentul în care am privit aceste elemente din perspectiva 

obiectivelor inovative şi am considerat că inovarea se dezvoltă cu ajutorul gandirii 

creative, am utilizat taxonomia definită de Bloom pentru a elabora un model cognitiv 

privind stimularea potenţialului inovativ. Elementul strategic care a stat la baza definirii 

acestui model este este reprezentat de Matricea Contradicţiilor TRIZ elaborată de 

Genrich Altshuller. În urma acestui demers au rezultat două matrici: Matricea Acţiunii 

Inovative şi Matricea Suportului Inovativ. Rolul acestora constă în faptul că potentialul 

inovativ poate fi stimulat daca sunt asimilate succesiv elementele taxonomice cognitive.


Publicat de 



Secţia 


ADVANTAGES AND WEAK POINTS OF DIFFERENT 

REVERSE ENGINEERING (RE) TECHNIQUES 

BY 

ANDREI MIHALACHE, GHEORGHE NAGÎŢ 

and MARIUS-IONUŢ RÎPANU 

Abstract. The concept design, as part of the engineering process and as 

support for CAD/CAE techniques, allows optimizing the product concept before 

manufacturing, with CAM assistance. For some product development processes, 

RE allows us to generate surface models by three-dimensional (3D) – scanning 

techniques, and as follows, this method allows different parts to be manufactured 

in relatively short development period of time. The aim of this paper is to look 

over RE techniques and to find possible advantages or weak points of different 

scanning systems. 

Keywords: Scanning (digitizing), reverse engineering, product design, rapid 

prototyping, CAD/CAE/CAM. 


One of the main domains of interest in rapid product development is 

represented by machine manufacturing, where each machine contains thousands 

of parts which have to be made as quick and cheap as possible, in order to 

achieve the prescribed quality [1]. 

But, by economical means and, of course, due to progress and innovation, 

the products do improve, and this leads to major or minor modifications brought 

to the machine, thus, many other internal parts have to be changed. The 

available time for this changes, has become increasable shorter due to

246 Andrei Mihalache et al. 

competiveness, and the requirements push everyone in rapid development. In 

such case, Reverse Engineering (RE) techniques are useful. The most important 

instruments found in the process, are different scanning systems, which 

provides, in a short amount of time, dimensional description in digital concept. 

RE has now been accepted as part of contemporary product design and 

manufacturing processes. The RE technique is easy defined as the process 

which results in creation of a mathematical model from a physical one. 

Different cases have different RE requirements, from recovering the 

mechanical design information to design based modifications. In case of 

extracting the mechanical design information, we will be interested in tolerance, 

as opposed to the case of design based extraction where the precedent will be 

that of extracting the design intent. 

So, the RE methods and techniques are essential because they allow to 

capture and digitize an object surface geometry for later use with 

CAD/CAE/CAM [1]. 

2. Product Development Approaches 

Nowadays the management of product design can be achieved based on two 

methods shown in (Fig. 1), a “conventional approach” and an “unconventional 

approach” [1]. 

The conventional approach in product development using CAD/CAE/CAM 

systems, normally starts with geometrical modeling using a CAD system. The 

geometric model can be represented by wire frame or as surface or as solid 

structure. Generated CAD data, can be exported to a standard format (IGES 

points/STL binary, ASCII data, DXF poly line, VDA points or IGES/STL 

surfaces) and then imported in the same data format to CAE systems (allowing 

numerical model simulation) and/or CAM systems (allowing tools trajectory to 

be generated). In a system with a unique database, the design information can 

be divided to each application automatically, without manual transfer of data, 

each time [3]. 

In the unconventional approach, we can see that the product development by 

the conventional approach is not always applicable when we aim to re-engineer, 

simulate or optimize parts/moulds/tools already existing, but without the CAD 

data format. So, it will be necessary to apply one technique that allows to 

capture part/mould/tool (or prototype) geometry and to generate a numerical 

model to be used in CAE and CAM systems. This is called reverse engineering.

Conventional 

approach 

Client specifications 

No 

3D – CAD 

system 

Bul. Inst. Polit. Iaşi, t. LVI (LX), f. 2, 2010 

Triangular model 

(file format *.stl) 

Conceptual model 

CAE system 

Design 

requirements 

satisfied? 

Yes 

CAM system 

Final product 

(part or tool) 

Physical model 

(sample or prototype) 

Digitizing system 

3D geometry 

Points cloud 

3D coordinates 

Reconstruction software for 

the model 

Unconventional 

approach 

Surface model 3D 

reconstruction 

Fig. 1 – Sequences in manufacturing engineered products 

(parts/moulds/tools)-adapted [1]. 

No 

247


CMM 

with laser 

system 

CMM 

with 

touching 

probe 

Physical model 

(sample or prototype) 

CNC – milling 

machine with 

touching probe 

Digitizing 

software 

Points cloud 

(3D coordinates) 

CNC – 

milling 

machine with 

laser system CT 

Computer 

tomography 

Fig. 2 – Digitizing techniques for 3D-geometries and generated data. 

3. Differences between Digitizing and Scanning Processes 

3D scanning (digitizing) is the process of gathering data from undefined 

three-dimensional surface. During the scanning process, an analogue scanning 

probe is commended to move back and forth (contact or non-contact) across the 

unknown surface. During the process, the system records information about the 

surface in form of numerical data’s – and generates a points cloud matrix (3D 

coordinates). The terms of digitizing and scanning are often used to describe the 

same process. Traditionally, digitizing refers to the process of taking discrete 

points using a touch-sensible probe. However, with the introduction of new 

technologies in data capture, such as laser or camera, digitizing term is now 

used as general description for the process of data acquisition from undefined 

surfaces [2] – [4]. The digital points cloud can be captured using different 

digitizing techniques (Fig. 2), classified into two major groups: 

‐ Mechanical techniques (by physical contact sensors); 

‐ Optical techniques (by non-contact with the object). 

Related to the first group, usually they use a coordinate measuring machine 

– CMM, or a CNC-milling machine equipped with physical touching probe 

sensors (ex. Retroscan or Renscan – Renishaw, UK). Related to the second


group, we can also use CMM’s or CNC-milling machines but equipped with 

laser beam probes (ex. Renishaw, Metris LC Mitutoyo), or associated optical 

sensors (ex. CCD cameras) [2]. 

3. Advantages and Weak Points of Different Scanning Equipments 

Scanning 

equipments 

Advantages Week points 

Laser Precise and fast scanning on Z-axis (0.001 mm) Equipment high price 

Non-contact method It is not possible to scan reflective 

materials 

It is possible to scan soft materials (or liquids) Scanning in X- and Y-axis is 

inaccurate 

(0.035 – 0.060 mm) 

It is not possible to scan notches 

areas or steep surfaces due to 

additional reflections. 

Sensibility to air dust 

CCD Fast 

High price of equipments 

cameras It is possible to use two or three cameras Accuracy decreases linear with 

simultaneously 

camera distance 

Insensible to part color 

Scanning angle is equal regardless 

the shape of part’s surface; in case 

of steep angles the measurement in 

inaccurate 

Non-contact method, it is possible to scan soft By camera scanning, a very sharp 

materials 

picture is required; simultaneous 

scanning of far and closer surfaces 

demonstrates linear deviation of 

results depending on focal distances 

In case of special coaxial lightning it is possible to In case of oily or wet parts the 

scan small diameters and high depths on Z-axis measurement in inaccurate 

It is possible to scan very small areas: 1 mm 2 (one 

scan only needed) – accuracy is above few 

micrometers 

Scanning on flat surfaces is very fast 

Dust causes bad scanning 

Contact Very precise on all axis (depends on scanning It’s inadequate for soft materials. 

(classic) equipment) 

Fast scanning of known geometrical parts Scanning unknown surfaces is either 

impossible 

inaccurate 

or very slow and 

Contact 

Precise scanning of coins or similar geometries 

Manual or automatic scanning possibility 

Hand-held scanning equipment is useful for 

scanning big products (airplanes, ships, big 

machines or devices) 

Precise on all axis (0.001 mm) Minimum diameter of stylus is 0.3 

digitizers 

mm, scanning surface roughness is 

(Renishaw) 

not possible 

Relatively low price It not possible to scan soft materials 

It is possible to use different styluses for different Though material of stylus is wear 

surfaces 

resistant, it’s surface wears in time 

Oil, liquid or dust do not interfere with the Scanning speed is lower (compared 

scanning process 

It is possible to scan unknown steep surfaces 

to non-contact systems)



1. Product development (parts/moulds/tools) via integrated reverse 

engineering represents a fairly new method that is in research and development 

phase. 

2. This paper aims to reveal some of the highlights and weak points of 

existing scanning equipments. We see that as far as accuracy goes, the best 

results are obtained by means of contact sensors. Nevertheless, in case of fragile 

or soft materials, laser or optical systems, without contact, provide better 

results. The ideal scanning device should be the combination of a camera, laser 

and contact probe, all, supported by appropriate controller. 

Received: March, 12, 2010 “Gh. Asachi”Technical University, 

Department of Machine Manufacturing 

Iasi, Romania 

e-mail: andrei.mihalache@yahoo.com 


1. H e r b e r t s o n T., Reverse engineering, in: Fourth International Conference on 

Industrial Tools ICIT 2003, April 8th–12th, Bled, Slovenia, pp. 419–422. 

2. N. N., Scanning Systems for Reverse Engineering, Renishaw Apply Innovation, H- 

2000-3120-04-B, Renishaw plc, UK, consulted at 18 November 2009, 

http://www.renishaw.com/en/cmm-probes-software-and-retrofits--6329 

3. H o n g w e i L., Adaptive patch-based mesh fitting for reverse engineering, State 

Key Lab of CAD&CG, Zhejiang University, Hangzhou 310027, China, Received 

13 May 2007; accepted 1 October 2007, pp. 3-10. 

4. Z e x i a o X., Complete 3D measurement in reverse engineering using a multi-probe 

system, Engineering college, Ocean University of China, Qingdao 266071, 

People’s Republic of China, Received 7 October 2004; accepted 20 January 

2005, Available online 7 March 2005, pp. 1135-1139. 

AVANTAJE ŞI PUNCTE SLABE ALE DIFERITELOR 

TEHNICI DE INGINERIE INVERSĂ 

(Rezumat) 

Dezvoltarea de produse (piese/ matrite/ scule) împreună cu ingineria inversă 

integrată, este o metodă recentă, aflată în faza de cercetare şi dezvoltare. 

Aceasta lucrare prezinta unele posibilitati de utilizare si beneficii ale metodei în 

procesul de productie, dar evidenţiază şi avantajele şi punctele slabe ale diferitelor 

echipamente de scanare utilizate în prezent.


Publicat de 



Secţia 


THERMOGRAPHY APPLIED TO BONE DRILLING 

BY 

ROBERTO LOPEZ 1 , MANUEL SAN JUAN 1 , FRANCISCO SANTOS 1 , 

OSCAR MARTÍN 1 and FLORIN NEGOESCU 2 

Abstract. Infrared radiation is a form of electromagnetic radiation. All matter 

above absolute zero (0 K, -273ºC) emits infrared energy. Thermographic cameras 

can be used for measure temperatures in a lot of applications. With this new tool 

we will be able to determine the local heating experienced by the bone during the 

process of cutting (drilling) for dental implants. The applicability of the temporal 

sequence of thermograms for dynamic study of the temperature in drilling 

operations to implant dentistry, is shown to be very high. While there are several 

aspects to consider when conducting research of this type. 

Key words: thermography, bone, drilling. 


There are many studies directed to determine the influence of the cutting 

parameters in bone drilling and the production of heat and the increase in the 

bone temperature due to this heat. In those studies the temperature was 

measured with thermocouples, the most usual way, or through other variables. 

There are less studies in the same area that use a more advanced method for the 

register of the temperature in bone cutting, the thermology. But there are even 

less studies that use a more dynamic method to obtain the temperature and be 

able to relate it directly to other variables such as pressure, speed of cutting, etc. 

Hereby our study tries to make a contribution in the use of the infrared 

thermology. Infrared radiation is a form of electromagnetic radiation like: radio 

waves, microwaves, ultraviolet rays, gamma rays, visible light, etc. All of them 

emit energy in the form of electromagnetic waves and they travel at the speed of 

light. All matter above absolute zero (0 K, -273ºC) emits infrared energy. 

Thermographic cameras detect invisible infrared radiation emitted by objects 

and they transform it in an image inside the visible spectrum in which the color

252 Roberto Lopez et al. 

scale (or greys) shows the different intensities. The intensity of the infrared 

radiation depends on the temperature and on the surface characteristics of the 

object, the color and the kind of material. Thermographic cameras give a value 

of the temperature for every single point, without taking into account that, for 

the same temperature, two different materials can emit infrared energy with 

very different intensities. 

With this new tool we will be able to determine the local heating 

experienced by the bone during the process of cutting (drilling) for dental 

implants. This way, we can see, frame by frame, the evolution of the 

temperature in every single point in the area of study, so every frame 

corresponds to a thermography with the data of the temperatures of every single 

point of the image. With this data we can study data from heating and cooling 

patterns, zone and/or specific peaks and lows. 


Due to the drilling process, the bone tissue heats up and if the temperature 

around the drilling area exceeds the critical limit of 50 ºC, could result in 

thermal necrosis, which is the death of the tissue exposed to that high 

temperatures. Therefore, the importance of temperature control in this type of 

operations results obvious. 

Most researchers use conventional temperature measure methods, such as 

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., 

2001), (M a t t e w s et al., 1984). In this way, the temperature record can be 

continuous; not on the contact area between the tool and the bone, but in the 

gaps set for the thermo couples away from the drilling area. 

The measurement of temperature (for example for fever) is very well known 

in old and modern medicine. The infrared thermology is a contactless method 

for measuring temperature and seems to be very useful in research and everyday 

medicine. Everyday there are more and more studies that use thermography for 

the measurement of temperature in different process. This way, (W i l d et al., 

2003), determined the temperature in bone drilling for facial human bones for 

inplants. They obtain results for temperature peaks although they don´t specify 

how they record the thermography at the highest temperature. 

Other researchers, like (U d i l j a c k et al., 2007) also use the thermographic 

camera to measure the temperature of drilling while using different drill bits. 

But neither in this case, nor in other cases like (W i l d et al., 2003), there is a 

continuous record of the temperature to be processed. So it is understood that 

the evaluation of the highest temperature is done objectively. 

The advantages of the infrared thermography in continuous record to 

measure the temperature are obvious:


• There is no delay between the generation of heat and the perception of 

the local heat. 

• It is possible to measure the temperature over a whole area. 

• Dynamic changes in temperature can be seen during the process. 

The dynamic register of thermographies allows making a complete study of 

the evolution of the temperature during the process. 

3. Experimental Set-Up 

In this study we have used infrared thermology to determine the cutting 

temperature in bone drilling for dental implants. Figure 1 shows the 

experimental set-up. 

Fig. 1. Experimental set-up 

Tools used during the tests are: 

• Thermographic camera 

A camera Infratec of high-speed thermography was used (ImageIR 5300, 

range from 700 to 1 µm; with a high frame rates of up to 250 Hz). A lens with a 

focal length of 100 mm was used with a macro lens of 500 mm. The infrared 

camera thermography enabled the estimation of temperature increase and 

temperature fields at bone drilling. The camera was set at an angle of 30º in 

relation to the hole axis (drilling direction), and the temperatures were 

determined (Fig. 1). 

The results of the thermographic camera are images where each temperature 

isotherm is denoted with a different colour, see Fig. 4. 

• Surgical motor 

Surgic XT Motor System (with a built-in irrigation system). Technical 

data: Power 210 W; motor speed: 200 – 40.000 rpm; torque: 50 Ncm. We used


a handpiece for the drilling tests to keep it as centred as possible from the load 

transducer from which it will hang. 

To hold the handpiece we built a device, made from aluminum, which could 

both, hold the handpiece and also attach it to the machine that generates the 

force, as seen in figure 2. 

Fig. 2 – Surgical motor attached to the force generating machine. 

• Testing Machine 

For the movement generating we used a Shimadzu Precision Universal 

Tester machine, model AG-100KNIS-MO. It is a floor model with two 

columns, able to work with speed tests of 0.5 µm at 1000 mm/min, according to 

the manufacturer. The setting up can be seen in Figure 1. 

• Drills 

Tool used for drilling the bone is a cobalt drill bit for metals. Such a drill bit 

is not very different in shape from the ones used in dental surgery and dental 

implants. Also the material it is made of is very similar to the materials used for 

the dental drill bits. The difference in temperatures between the drill bits we 

used and the dental drill bits should not be very different. The point of this test 

is to measure the temperature of bone drilling using thermology, so for that 

matter, the drill bits used for the tests should be as good as others. The drill bit 

used in the tests is 2.22 mm in diameter. It is cylindrical straight and has two 

lips. Drills can be seen in figure 3. 

Fig. 3 – Drill used in test. 

• Bone 

For the tests we have used cow bone, specifically the femur. It is a fresh 

bone bought the previous day in a local market. The cow bone characteristics


are very similar to the ones of the human bone, for that reason cow bones have 

been used for the scientific community in multiple tests of this type. 

Fig. 4 – Drilling of bone and thermography. 

4. Methodology 

For the test we used a 2.22 mm in diameter drill bit turning at a speed of 

2,000 rpm attached to a handpiece from the surgical motor. For the set up of the 

surgical motor we choose the maximum power to avoid a fall in the turning 

speed. The handpiece was attached to the tester that made it move down at a 

speed of 1.25 mm/second. The drill was made in a single movement until we 

reached a depth of 8 mm. This is not the usual way for implantology, which 

uses a discontinuous drilling of one or two seconds. This way we could get a 

higher temperature in the drill bit and check the efficiency of thermology in 

bone drilling. 

The thermographic camera was placed in front of the force generating 

machine, as seen in figure 1, which gave us a general view of the cutting area. 

Over the thermographic camera was placed a macro lens in order to increase the 

resolution of the camera in the area of contact between the drill bit and the bone 

sample. This setting allowed us to register the temperature in the drilling zone, 

which included the temperature of the drill bit and the superficial temperature of 

the bone in the critical area, beside the tool. We made several tests where we


register the temperatures from the beginning of the drilling till the exit of the 

drill bit once reached the 8 mm depth. 

Fig. 5. Thermography sequence. 


These are some of the most relevant results of the thermographies during the 

measurement of temperature in bone drilling. These results could be applicable 

to the analysis of temperature using a thermographic camera in other tests, for 

example, in the drilling of other materials (wood, metals, resins, alloys, new 

materials). During the tests it is important to keep the bone at body temperature 

(37 ºC) or adjusted to the temperature of the tool to isolate the increase of 

temperature during the drilling process. With the use of the thermographic 

camera, we can isolate the average temperature of the elements in the test, as 

well as the peaks and lows, the highest and lowest puntual temperature, the 

highest and lowest temperature of the area around the tool before the drilling, of 

the area of the material, etc. 

The continuous register of temperature allowed us study the cooling of the 

surface following the heating of the surface due to the drilling. We did a linear 

regression of the data according to the time getting this adjustment: 

(1) Tª = 14.1566 – 0.112166*Time, 

with a cooling rate of 0.11º/s. The cooling is due to the deepest layers in the 

bone which cool the outer layers heated during previous drillings. 

The P-value of the analysis of variance (p


Since the first contact between tool and bone, the recorded temperature starts to 

increase, and in a following study of data, a few points can be made. 

Fig. 6 – The fall of temperature. 

The starting temperature in the drilling point of the bone is 14.415 ºC. The 

temporal sequence of the temperature in the drilling point is shown in this chart: 

Table 1 

Temperature in the drilling point 

0 1 2 3 4 5 6 7 8 

14.42 16.96 18.40 21.81 27.61 26.49 32.11 46.18 51.25 

The drilling lasted 8.5 seconds. This is the chart for the increase of 

temperature according to the time. 

Fig. 7 – The increase of temperature. 

There is a tendency in the increase of temperature as the drill bit drills the 

bone. Some adjustments were made for the analysis of data, and the closest one 

was: 

(2) Tª = 16.5913 + 0.5518*Time^2. 

Against a linear regression, the quadratic function explained up to 95.87% of 

the total variability of data (5% more than the linear), making the adjustment


more precise. Finally, we can concluse that the applicability of the temporal 

sequence of thermograms for dynamic study of the temperature in drilling 

operations to implant dentistry, is shown to be very high. While there are 

several aspects to consider when conducting research like this. 

Received:23.02.2010 1 University of Valladolid-Spain, 

Department of Materials Science and 

Metallurgical Engineering-Mechanical Engineering 

Valladolid-Spain 

e-mail: roblop@eis.uva.es 

2 “Gheorghe Asachi”Technical University, 


Iaşi, Romania 

e-mail: negoescu@tcm.tuiasi.ro 

Acknowledgments. This work has been developed thanks to DPI2006-15502-C02-02 

Project from the National Program of Design and Industrial Production supported by 

MEC (Spain) and FEDER. 


1. A b o u z i a M. B., J a m e s D. F., Measurements of shaft speed while drilling 

through bone. J. Oral Maxillofac Surg, 53 (1995) 1308-1315. 

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 

force on cortical temperatures and their duration: and in vitro study. Medical 

Engineering & Physics, 22 (2000) 685-691. 

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 

fixation-pin insertion in bone. Journal of Bone & Joint Surgery, 66A, 7 (1984) 

1077-1083. 

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 

thermal bone necrosis. Advances in Production Engineering& Management, 2 

(2007) 103-112. 

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 

thermography as a non invasive application for medical diagnostic. Proceedings 

of XVII IMEKO World Congress, pp. 1744-1748, June 22–27, 2003, Dubrovnik, 

Croatia. 

TERMOGRAFIE APLICATĂ ÎN FREZAREA OSOASĂ 

(Rezumat) 

Radiaţiile în infraroşu sunt o formă a radiaţiei electromagnetice. Camere 

termografice pot fi utilizate pentru măsurarea temperaturii într-o mulţime de aplicaţii. 

Cu acest instrument, este posibil să se determine zonele locale de încălzire în timpul 

procesului de taiere pentru implanturi dentare. Aplicabilitatea termografiei pentru 

studiul dinamic al temperaturii în operaţiunile de tăiere a implanturilor stomatologice, se 

dovedeşte a fi foarte mare.


Publicat de 



Secţia 


STUDY ON THE OPPORTUNITY OF A NEW 

GLASS RECYCLING FACTORY FOR REGIONAL 

SUSTAINABLE DEVELOPMENT 

BY 

OLGA MARINA MONTES 1 and VASILE V. MERTICARU 2 

Abstract. The paper presents a theoretical study and business justification 

concerning the opportunity for the establishment of a new factory for glass 

recycling, respectively for turning waste glass into raw material for manufacturing 

new usable glass products. The presented study is part of a research approach 

developed in the direction of establishing the economical and technical conditions 

for establishing the discussed glass recycling factory in the province of 

Valladolid, autonomous community of Castilla y León from Spain, as possibility 

to enlarge the number of similar enterprises, for regional sustainable development. 

The state of development for the glass recycling industrial sector in Spain is 

discussed and the adequacy of the location is analyzed in the paper, together with 

the presentation of related conclusions. 

Key words: glass recycling, sustainable development, business justification. 


Sustainable Development permanently represents the crucial goal of 

mankind and under this imperative, all the human activities, both social and 

economical and of course the productive ones must able to assure the harmony 

between profit, people and Earth. It is widely accepted that from engineering 

point of view, industrial society in general will reach Sustainable Development 

only via Sustainable Manufacturing and via Sustainable Design. In this sense, 

industrial development activities starting from plant engineering and design 

must be compulsory considered in relation to advanced concepts such as Life 

Cycle Analysis (LCA) and Extended Producer Responsibility (EPR). The

260 Olga Marina Montes and Vasile V. Merticaru 

concepts of Reducing, Reusing, Recycling and Recovery of material resources 

are also very important in providing Sustainable Industrial Development. 

Of course, the “cradle-to-cradle” way in LCA is nowadays generally 

accepted as the only viable solution for providing environmental protection and 

avoiding the depletion of natural resources, as long as materials can be reused, 

no waste gets produced or otherwise it can be recycled so that no damaging 

impacts on the environment are generated within the closed loop of the product 

lifecycle, [3], [5], as is shown in Fig. 1. 

Off-site recycle 

Product 

Reuse 

or Recycle 

Product 

use 

On-site recycle 

“Cradle to cradle” 

LCA 

Product 

marketing 

Raw material 

extraction 

Product 

transportation 

and storage 

Raw material 

processing 

Product 

Packaging 

Fig. 1 – LCA based on “cradle-to-cradle” concept. 

Product 

manufacturing 

On the other hand, as a policy measure which comes to outstand the 

industrial producer’s crucial role in reducing the environmental impacts of their 

products throughout their entire lifecycle, Extended Producer Responsibility 

(EPR), encourages the reuse of products and packaging and, not less important, 

the Design for Environment practices [5]. 

In relation to the above mentioned considerations, the paper presents a 

theoretical study and business justification concerning the opportunity for the 

establishment of a new factory for glass recycling, respectively for turning 

waste glass into raw material for manufacturing new usable glass products. The 

presented study is part of a research approach developed in the direction of 

establishing the economical and technical conditions for establishing the 

discussed glass recycling factory in the province of Valladolid, autonomous 

community of Castilla y León from Spain, as possibility to enlarge the number 

of similar enterprises, for regional sustainable development.


Glass recycling, as industrial activity which comes to replace important 

quantities of traditional raw material in new glass products industry with 

processed waste glass, can be included under the umbrella of the Cleaner 

Production techniques, [4], [5], covering the areas of off-site recycling and 

process change, as it is shown in Fig. 2. 

Waste 

Reduction 

Better 

housekeeping 

Cleaner Production Techniques 

“GREEN PRODUCTION” 

Product 

Modification 

Process Change 

Process 

control 

OFF-SITE 

Recycling 

GLASS 

RECYCLING 

Technology 

change 

RECYCLING 

Raw material 

change 

On-site 

Recycling 

Fig. 2 – Area covered by glass recycling within “Green Production”. 

In a plant engineering approach, it is important to analyze the industrial 

sector in which the project will belong and to see if it makes sense to think 

about the fact of beginning the studies to start it. In order to this, the state of 

development for the glass recycling industrial sector in Spain is discussed and 

the adequacy of the location is analyzed in the paper, together with the 

presentation of related conclusions. 

2. The Glass Recycling Sector in Spain 

Since the end of the Second World War and particularly during the last 

decade, there has been registered an increasing ‘throw away’ tendency. 

Statistics show that in the European countries the amount of household waste 

has increased to more than 1 kg per person and per day, on average. The 

composition of domestic waste varies from country to country, but the tendency 

is the same everywhere: more paper, more plastic, more valuable metal and 

glass, and less ash. Because of that phenomenon, the number of recycling plants 

is also increasing, as response to balance the residuals proliferation.


According to the Spanish Ministry of Environment and Rural and Marine 

Environment, each year in Spain approximately 592 kilograms of waste per 

inhabitant and per year, [2], is produced. Much of this waste ends up in 

landfills, occupying large areas of land and necessary addition to avoid soil 

contaminating. 

At the present time, in Spain there are a total of 15 plants dedicated to the 

glass recycling and in the region of “Castilla y León” there exist only one other 

glass recycling plant. 

Since the entry into force of the Packaging Act in 1998, all containers are 

required to fund and implement a collection system being outlook openrecycling 

of packaging they put on the market. This legal obligation can be met 

in two ways: 

• Individually, by a deposit-refund. 

• Collectively, through an integrated management system. 

In this context and in order to provide an effective and economical solution 

to all employers who package their products mostly in glass, Ecovidrio, [2], 

arose as a nonprofit association whose main objective is to ensure the 

traceability of glass packaging. In this way, every packaging that enters in the 

recycling chain follows the appropriate process to become a completely new 

one. Being the only integrated management system that specializes in glass, is 

able to reduce costs and provide better service to all companies, large or small. 

The 2008 financial year ended with 2455 companies adhered to Ecovidrio, with 

107 more than in 2007, to contribute to sustainable development and 

environmental care. 

In Fig.3 and Fig. 4 there are summarized the most important data on glass 

recycling in Spain from 2001 until 2008 [2]. 

Fig. 3 – Statistic data on glass recycling in Spain from 2001 until 2008 [2].

Rate of recycling, [%] 


Fig. 4 – Evolution of glass recycling rate in Spain from 2001 until 2008. 

3. Location Analysis and Justification 

For making the decision to locate an industry, we need to know a number of 

parameters that help defining the characteristics of possible locations and see 

which one bestly fits the needs of industry to be implemented. The location is 

an individual choice in which each industry values and ponders the most 

appropriate option for its, [1]. 

The location parameters can be grouped into the following points: 

• Parameters concerning the nature of the industry and its classification; 

• Parameters concerning the analysis of the geographical; 

• Parameters analyzing the urban conditions; 

• Parameters relating to the study of environment; 

• Parameters relating to human capital and intellectual capital. 

From the analysis of the above mentioned points there has been determined 

to locate industrial facilities of the plant object of labor in the industrial area of 

San Miguel del Arroyo, in the province of Valladolid, autonomous 

community of Castilla y León from Spain. 

The following considerations, related with each of the above named 

sections, provide a synthesis of the selected location. 

3.1. Nature of the Industry and its Classification 

The council enacted in 1982 the European Community Directive 82/501 on 

the prevention of major accidents arising from certain industrial activities. This 

standard classifies the considered facility as industrial activity and performing 

dangerous quantities manufactured, processed or stored. Moreover, legalization 

itself exists in each country and / or region and there may be legislation also 

own by municipality. First, is taken into account that the legislation in force in 

San Miguel del Arroyo allows the implementation of this industry to be built.


3.2. Analysis of geographical conditions 

In this section it has been checked that the chosen area has all the natural 

security conditions necessary for any deployment. There have been key issues 

such as security of land, about flooding (unlikely), forest fires, erosion and 

deforestation. With respect to climate and the presence or absence of surface 

water, to implement the factory has no special requirements. With regard to the 

existence of communications, industrial estate where the plant will be located 

has good links. This is a site near Valladolid, capital of the Community of 

Castilla y Leon and also is close to Madrid, capital of Spain, and near the 

national road Nº1, the road which connects the center of the country to the 

north. Communications are important both for the supply of waste glass and to 

deliver the finished products. 

In the community Castilla y León there are important industries like 

Renault, Michelin, Iveco, Pascual, Firestone, and the glass waste from these 

industries should be the materials the plant will recycle. Also, Madrid is near of 

the location, and the plant can recycle the waste from the industries of these 

regions or from other regions of the country. 

3.3. Analysis of urban conditions 

Each plot of the considered area has planning conditions that define the 

amount and location of the intended buildings. The following data provide some 

of the characteristics of the industrial area of the town of San Miguel del 

Arroyo. 

General Information - Surfaces: 

• Total area: 86,810.27 m²; 

• Endowment Surface: 5201.32 m²; 

• Public open space (green areas): 40,672.98 m²; 

• Maximum Plot: 2868.74 m²; Minimum Plot: 177.18 m²; 

• Minimum strippable: 175 m². 

Other Data: 

• Number of plots: 44; Free Plots: 11; 

• Access roads: Yes, in concrete pavement; 

• Supplies: Yes, municipal network; 

• Sanitation: Yes, primary treatment; 

• Lighting: Yes; Electricity: Yes; Gas: Yes; 

• Telecommunications: Yes. 

• Rules planning: Municipal Town Planning Regulations; 

• Maximum building height: 7 m and 10 m ridge cornice; 

• Setback: 5.00 m and 3.00 m front edge - rear edge. 

The previous planning conditions have been of paramount importance when 

deciding the final location of the plant.


3.4. Environmental Study 

Within this point, there have been considered the political and social 

stability of the country or region where the possible location is. In addition, the 

security of people working in and for industry should be maximal. With respect, 

it must be said that Spain is a safe country without major political conflicts, 

social, ethnic or religious, belonging to the European Community, which would 

also facilitates free trade and unimpeded across Europe. 

The industrial estates in the town of San Miguel del Arroyo is promoted by 

an industrial Sodeva (Provincial Development Society of Valladolid), which 

brings many advantages and subventions. 

It should be noted that the town of San Miguel del Arroyo is close to the 

University of Valladolid, which can determine a dynamic and availability of 

their students as interns in research and development processes as well as in 

other departments. 

3.5. Human and Intellectual Capital 

There has been taken into account that the area of localization has enough 

human capital to meet the staffing needs of the company to implement. The 

designed industrial plant requires very few employees, and most workers do not 

require a very thorough qualification. Staff will be working in industry and must 

have an adequate quality of life for their needs. They have the possibility of 

living in a city, Valladolid, with all necessary services available to them, or live 

in the town of San Miguel del Arroyo, where housing prices are lower, besides 

having all the primary needs; schools, leisure areas, sports areas, shops. These 

basic parameters can be framed within what would be called family welfare. 

Besides human capital, there must be taken into account the intellectual 

capital, i.e. preparation. As already mentioned above, the University of 

Valladolid is relatively close to the point of location and can have enough well 

prepared people. 


1. The presented study and the obtained synthesis results have proved the 

adequacy of implementing a new glass recycling factory in the selected location 

and they are useful for efficient establishment that plant for glass recycling. 

2. A manufacturing process design, together with a study of the technical, 

human and economic necessary for the plant will complete the best design of 

the plan to get a finished product with the required quality and the lowest 

possible cost.


Received: February, 20, 2010 1 Universidad de Valladolid, 

Valladolid, Spain 

e-mail: olga.marina.montes@gmail.com 

2 ”Gheorghe Asachi” Technical University, 


Iasi, Romania 

e-mail: merticaru@tcm.tuiasi.ro 


1. C a s a l s-C a s a n o v a, M., Complejos Industriales. Editions UPC, Cataluña 

Polytechnic University, 2005. 

2. Ecovidrio - Informe anual 2008, Available from: - http://www.ecovidrio.es/app/ 

GeneraPaginas.asp?seccion=../app/WebEcovidrioNet/wEstadisiticasRecicladoTo 

talNacional.aspx, Accessed: 02/02/2010. 

3. E l-H a g g a r, S. M., Sustainable Industrial Design and Waste Management: Cradleto-cradle 

for Sustainable Development. ISBN-10: 0123736234, ISBN-13: 

9780123736239, 406 p, Publisher: Academic Press, 2007. Available from: 

http://www.engineeringvillage.com. Accessed: 2008-02-02. 

4. G a l i s M. et al., Digital product development for the entire product life cycle. 

Academic Journal of Manufacturing Engineering, 6, 3 (2008), Timisoara, 

Romania, p.55-60 

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, 

for Integrating Manufacturing with Sustainable Industrial Design, Proceedings 

of ICOVACS 2008: International Conference On Value Chain Sustainability, 

Izmir, Turkey, November 12-14, 2008, Izmir University of Economics 

Publication No: IEU-026, ISBN 978-975-8789-25-2, pp.109-118. 

STUDIU ASUPRA OPORTUNITĂŢII UNEI NOI FABRICI 

DE RECICLARE A STICLEI, PENTRU O DEZVOLTARE 

REGIONALĂ DURABILĂ 

(Rezumat) 

În lucrare este prezentat un studiu teoretic şi justificarea din punct de vedere 

antreprenorial privind oportunitatea înfiinţării unei fabrici noi de reciclare a sticlei, respectiv 

pentru transformarea deşeurilor de sticlă în materie primă pentru fabricaţia de produse noi 

din sticlă. Studiul prezentat este parte a unui demers de cercetare dezvoltat în direcţia 

stabilirii condiţiilor tehnice şi economice pentru înfiinţarea fabricii de reciclare a sticlei în 

discuţie, în provincia Valladolid din comunitatea autonomă Castilla şi Leon din Spania, ca 

posibilitate de creştere a numărului de astfel de întreprinderi similare, pentru o dezvoltare 

regională durabilă. În lucrare, se discută stadiul de dezvoltare a sectorului industrial de 

reciclare a sticlei în Spania, se analizează adecvanţa localizării investiţiei şi sunt prezentate, 

de asemenea, concluziile aferente.


Publicat de 



Secţia 


MODELING 3D SURFACE OF TEXTILE STRUCTURES 

FOR FLUID FLOW IMPROVEMENT 

BY 

CĂTĂLIN DUMITRAȘ 1 , CARMEN LOGHIN 1 , SULEYMAN YALDIZ 2 , 

MEHMET SAHIN 2 and LUMINIȚA CIOBANU 1 

Abstract. The efficiency of thermal insulation in a windy environment can be 

increased by using adequate protective equipment. Following a FEA analysis 

regarding the flow of fluids (air) at body surface, the presence of critical zones 

with turbulences was detected. The critical zones are the ones with a change in 

geometry, such as the shoulders, hips or hands. This generates the idea that 

something has to be done to improve the fluid flow at surface level of the exterior 

garment layer. The main problem is to ensure a certain path for the fluid 

movement, so that the flow will become laminar. A simple and feasible solution is 

creating a textile structure with an architecture that includes these paths. Different 

variants of 3D geometries are proposed. Their dynamic behaviour is studied in 

order to rank the proposed fabricse abstract of the paper is to be written here. It 

contains the main ideas and original contributions and conclusions of the authors’ 

research. 

Key words: laminar air flow, weft knitted fabrics with relief effects, sandwich 

fabrics, FEA, meshed models, fluid flow. 


The problem of thermal protection in hostile environments is certainly not 

new, but it does not loose its interest due to the fact that there are large range of 

activities requiring such protective equipment, from winter and extreme sports, 

cycling, motorcycling to working in harsh winter conditions. There are a lot of 

studies regarding developments of textile fabrics and materials to improve 

thermal insulation. Still, the fluid flow around the garments in dynamic 

conditions is less investigated and the problem of designing textile structures 

with improved behaviour in relation to air flow remains open to debate.

268 Cătălin Dumitraș et al. 

In order to formulate solutions for the fabric structure one must understand 

the nature of the problem. The conditions characterising a hostile environment 

include among others strong winds. The flow of these air currents is different, 

according to the type of surface they encounter. Larger surfaces, with a planar 

distribution present a laminar flow of the air currents, while complex 3D 

surfaces are characterised by the formation of pressure fields and turbulences 

that affect the thermal behaviour of the exterior layer in the garment by 

reducing its insulation. Areas like the arms, hips or shoulders, with smaller 3D 

surface present problems in relation to the laminar air flow. Previous studies 

regarding the air flow around a human body show that these are the critical 

zones in a protective garment. Due to its specific nature, the shape of these areas 

in a protective garment cannot be modified. Therefore, the improvement of air 

flow must be based on modifying the fabric surface in the critical areas so that 

the pressure fields are diminished. The fabric surface has to be designed so that 

the air currents are channeled toward the exterior, avoiding turbulences. From 

this point of view, the 3D knitted fabrics represent the best solution. Apart from 

the shape of their cross section, it is important to define the optimum paths for 

these channels and their position in the protective garment. 

2. Definition of Air Flow at Garment Level 

An initial analysis was carried out in order to identify the air flow critical 

areas on the human body. Its purpose was to determine the air pressure and 

velocity distribution within the fluid. The model represented therefore a fluid 

mesh and not one of the solid body. It was created with the ALGOR v12 FEA 

software package, using a number of 491 2D elements (558 nodes), defined in 

Fig. 1. The FEA model is presented in Fig. 2. The body is considered an 

obstacle, while the environment is meshed using bidimensional finite elements. 

Fig. 1 – The type of FEA element used. Fig. 2 – Finite element based model. 

It was presumed that a pressure field caused by winds up to 100 km/h is 

applied to the model, corresponding to all human body. The selected properties 

of the environment are illustrated in Fig. 3.


Fig. 3 – Physical properties of the environment. 

The analysis results are presented as pressure fields, velocity and 

turbulences. The output variables were calculated for different values of the 

input parameters. The wind was considered to blow with 90 km/h velocity for a 

period of 90 seconds. The results are presented in Figs. 4 to 6. 

t = 15 s t = 80 s t = 3 s t =80 s 

Fig. 4 – Pressure fields 

distribution 

t = 3 s t = 70 s 

Fig. 6 – Turbulences distribution 

Fig. 5 – Velocity fields distribution 

The images presented show that the turbulences in the air flow, as well as 

higher velocities are characteristic to some parts of human body – the shoulders,


the hips and the chest. Such critical areas are susceptible to generate discomfort 

in windy conditions. 

A solution must be found to diminish these turbulences and the pressure 

fields encountered. From the fabric point of view, the air flow at surface level 

can be controlled through: the type of raw material, fabric structure and 

structural parameters. The most important factor is the fabric structure, mainly 

the surface geometry. Weft knitted fabrics offer a large range of structural 

possibilities and are characterised by good elasticity and high formability. In 

comparison with other types of fabrics, especially woven, the weft knitted 

fabrics can be produced with complex 3D architectures that are specific only to 

them and do not require costly technological changes on the machines. 

3. Models For The Air Flow At Shoulder Level 

3.1. Definition of the Models 

Two models were created using AutoCAD 2007, considering the human 

shoulder. The first model represented the shoulder without a modified surface 

and is illustrated in Fig. 7. This model continues the 2D analysis presented 

above. The other model considers the surface modified by the presence of the 

fabric. The fabrics considered had a rectangular shaped cross section as 

discussed above and the resulting model is presented in Fig. 8. 

The section height is 10 mm. Lower height values proved to be inefficient. 

The rectangular channels are spaced also at 10 mm. The modelled shoulder was 

included in a space volume in order to simulate accurately the air flow. 

Fig.7 – Model without 3D surface effect 

Fig.8 – Model with channels with 

rectangular cross section 

The models were transferred to the ALGOR v20 FEA software package to 

be analyzed. The model was meshed automatically resulting a number of 96,584 

3D elements (bricks) and a number of 108,512 nodes. The fluid considered is 

the air, its flow velocity chosen to be 50 km/h. This simulates an extremely 

hostile environment.


The fluid properties were selected from the Material Library Manager and 

are presented in Fig. 9. The meshed model for the shoulder is illustrated in 

Fig.10, while the meshed model for the shoulder and 3D fabric is presented in 

Figure 11. 

Fig. 9 – Selected air properties Fig. 10 – Meshed model 

Fig. 11 – Meshed model for shoulder and fabric 

The models were analyzed using finite elements, based on steady fluid flow 

processor. The parameters considered are illustrated in Figs. 12 and 13. 

Fig. 12 – Parameters for steady fluid flow Fig. 13 – Considered load curve


3.2. Results and Discussion 

The following results were obtained based on the analysis presented above. 

The pressure fields at shoulder level were identified for the shoulder without 

fabric, as presented in Fig. 14. This constitutes a reference base for all models 

containing the 3D fabrics. From Fig. 14 it results that there is higher pressure 

field at the back of the shoulder, possibly caused by a suction effect. 

Fig. 14 – Distribution of pressure fields at shoulder level. 

The distribution of air velocity at shoulder level is presented in Fig. 15 and 

16. High velocity values are identified at the entire shoulder level, rest of the 

human body these values are lower. That indicates that the turbulences are 

higher in the shoulder area. 

Fig. 13 – Air velocity distribution, 

general view 

Fig. 14 – Air velocity distribution 

The use of a 3D fabric with rectangular channels changes the air flow. The 

velocity distribution is different in comparison to the situation described for the 

shoulder without fabric. The areas characterized by high air velocity are 

distanced from the shoulder areas suggested by Figs. 15, 16 and 17.

Fig. 15 – Air velocity distribution for 

shoulder and 3D fabric – general view 


Fig. 16 – Air velocity distribution for shoulder 

and 3D fabric – bottom view 

Fig. 17 – Air velocity distribution for shoulder and 3D fabric – top partial view 

From Figs. 15-17 it results that the air flow is improved, diminishing the air 

velocity at shoulder level and implicitly the air turbulences. 


1. FEA studies concerning protective equipment for cold environments have 

shown that the air flow at body level presents critical areas where the flow is 

characterized by pressure fields and turbulences, causing heat loss and 

discomfort. It is the case of hips, chest and shoulders. 

2. One solution for the improvement of the air flow is using fabrics with 3D 

structure or 3D surface effects. These knitted fabrics are produced on flat 

knitting machines without any adaptation required. The paper presents two 

structural variants with 3D effects and 3D architecture – rib fabric with relief 

effects and sandwich fabrics with complex shapes. The fabrics are characterized 

and their 3D structure is discussed, emphasizing their advantages in obtaining 

surface effects that have the potential to control the air flow.


3. Two FEA studies were carried out, one at 2D level, in order to identify 

the air flow critical areas. The 2D analysis does not ensure complete data 

regarding the in depth air flow so a further 3D finite element analysis was 

required. This analysis indicated the flow phenomenon dimensions at shoulder 

level and allowed a comparison between air flow at shoulder level and the air 

flow when a fabric is added. 

4. The study took into consideration two models of the shoulder: without 

and with fabric. The fabric selected was the sandwich fabric with rectangular 

channels with an uniform distribution. 

5. The study shown that the use of this fabric brought an improvement in air 

flow when used at shoulder level. The study requires additional work taking 

into account the followings: fabrics with other 3D effects (differently shaped 

channels – circular) and non-uniform distribution of these channels (patterned 

distribution of the 3D effect). One important target is to determine the shape of 

the channel path. These studies will lead to the optimization of garments used 

for protection against cold. 

Received: March 20, 2010 1 Technical University “Gheorghe Asachi”, 



e-mail: dumitrascatae@yahoo.com 

2 Selcuk University, 

Department of Mechanics 

Konya, Turkeyt 

e-mail: syaldiz@yahoo.com 


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, 

Braga, Portugal, 2000 

2. H o l m e r I. – Textiles for Protection Against Cold, in Textiles for Protection, 

edited by R.A. Scott, Woodhead Publishing Ltd, Cambridge, UK, 2005, p. 378 – 

397 

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 

functions analyze and intelligent functions synthesis for the protective equipments 

for aggressive environments‘ research report, CEEX 105/2006 

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 

chemical processes at the interface level between aggressive environment and 

protective equipment‘ research report, CEEX 105/2006 

5. Z i e n k i e w i c i O. C. – The Finite Element Method for Fluid Dynamics, 

Elsevier, Amsterdam Netherland, 2006


MODELAREA 3D A SUPRAFETELOR TEXTILE CU DESTINAŢIA 

OPTIMIZĂRII CURGERII FLUIDELOR 

(Rezumat) 

Eficienţa protecţiei termice într-un mediu cu curenţi puternici de aer (vânt) poate fi 

îmbunătăţită printr-un echipament de protecţie adecvat. Ca urmare a unei analize cu 

elemente finite a curgerii fluidelor (aerului) pe suprafaţa corpului detectarea yonelor 

critice turbionare poate fi realizată. Aceste zone au fost detectate în locurile în care se 

modifică geometria corpului cum ar fi umeri, brate, etc. Concluziile acestei analize au 

condus la ideea că se poate obţine o îmbunătăţire a coeficientului de frecare a aerului 

prin modificarea geometriei echipamentului textil. Se pune problema determinării unei 

geometrii convenabile a acestui echipament astfel încât săse micşoreye sau să dispară 

zonele de curgere turbionară. Se propun mai multe variante de structuri textile cu 

geometrie 3D. Se realizează apoi un studiu destinat determinării profilului optim al 

structurii textile.

A 

Alexandrescu Adina 189 

Alexandrescu Adrian 189 

Alexandrescu Irène 227 

Alexandrescu Aurora 189 

Antoniadis Aristomenis 9,21 

B 

Baciu Lupașcu Radu 91 

Bălășoiu Victor 197 

Băran Marian 197 

Bărglăzan Mircea 205 

Bădărău Rodica 205 

Belis Taxiarchis 9 

Berbinschi Silviu 41,49 

Bordeașu Ilare 197 

Bostan Ion 135 

C 

Calfa Daniel 161 

Ciobanu Luminița 267 

Călărașu Doru 153 

Chiriță Constantin 161 

Ciobanu Bogdan 153 

Cozmîncă Irina 97 

Cozmîncă Mircea 65 

Croitoru Cristian 65 

D 

Dobîndă Eugen 205 

Dogaru Constantin 31 

Dulgheru Valeriu 135 

Dumitraș Cătălin 267 

Dumitrașcu Nicolae 41,49 

Dumitrescu Cătălin 167 

Dumitrescu Liliana 167 

Dumitru Dumitru 31 

Dușa Petru 91,237 

F 

Fetecău Cătălin 57 

Filip Cristina 117 

Franke Hans Joachim 227 

Fulga Ileana 143 

G 

Gonçalves Coelho Antonio 1 

H 

Hanganu Adrian 161 

Haraga Georgeta 167 

Horodincă Mihăiță 125 

Index autori 

I 

Ibănescu Radu 97 

L 

Loghin Carmen 267 

Lopez Roberto 251 

M 

Manole Iolanda 83 

Murad Erol 167 

Mănescu Mihai 177 

Marsavina Liviu 197 

Miloș Teodor 205 

Manea Adriana 205 

Mihalache Andrei 245 

Martín Oscar 251 

Montes Marina Olga 259 

Merticaru V. Vasile 259 

Matei Ana Maria 75 

Milea Marius Nicolae 75 

Mocanu Costel 57 

Mourao Antonio 1 

N 

Nagîț Gheorghe 83,245 

Neagoe Lavinia 117 

Novac Dragoș 197 

Negru Radu 197 

Negoescu Florin 251 

Neștian Gabriela 1 

O 

Oancea Nicolae 41,49 

P 

Panaitescu Valeriu 167 

Pavlov Olimpia 31 

Petre Dan 117 

Petre Ioana 117 

Popoviciu Mircea 197 

R 

Rîpanu Marius Ionuț 83,245 

Rotman Iustina 91 

S 

Sahin Mehmet 267 

San Juan Manuel 251 

Santos Francisco 251 

Scurtu Dan 153 

Sochireanu Anatol 135

Stachie Marius 221 

Storm Kjærside Birgit 105 

Străjescu Eugen 31,143 

Stroiță Daniel 205 

T 

Tapoglou Nikolaos 21 

Teodor Virgil 41,49 

Tița Irina 153 

Tudorache Mihaela 215 

Taranovschi Iuliana 237 

U 

Ungureanu Cătălin 65,97 

V 

Vietor Thomas 227 

Vlad Daniel Viorel 57 

Vodă Mircea 197 

W 

Weingold Andrei 83 

Y 

Yaldîz Suleyman 267 

Z 

Zahariea Dănuț 215, 221

BULETINUL INSTITUTULUI POLITEHNIC DIN IAŞI - Universitatea ...

Create successful ePaper yourself

Delete template?

Save as template?