The New NBCC - The Schulich School of Engineering

The new NBCC:Companion Loads,Wind/Snow Loadingand SeismicF. Michael BartlettF.CSCE, P. Eng.University ofWestern Ontariof.m.bartlett@uwo.ca

This Morning’s Agenda• Divisions A and B of NBCC 2005• Companion Action LoadCombinations• Importance Factors for Buildings• Changes to Specified Loads• Loose Ends• Summary

NBCC 2005 is Objective-basedDivision A (new!)• 4 objective categories• 46 functional statementsDivision B (familiar)• “Acceptable solutions”• transition mechanism

Code of HammurabiPhoto byC. R. Scollard

Wreckage1907

Finial asGraveMarker in St-RomualdCemetary17 Americans58 Canadians(33 CaughnawagaIron Workers)

Drop-in Span Collapse11 September 1916Death toll: 75 (1907) + 11 (1916) = 86

Canadian Code Milestones1922: CESA “Standard Specification forSteel Railway Bridges”1924: CESA “Standard Specification forSteel Structures for Buildings”1929: CESA “Standard Specification forConcrete and ReinforcedConcrete”1941: National Building Code of Canada

1974: Limit States DesignKey Players• D. J. L. Kennedy• J. G. MacGregor• D. E. AllenOthers• A. G. Davenport• N. C. Lind• D. A. Taylor

Safety Objectives in NBCC 05OS 1: Fire SafetyOS 2: Structural SafetyOS 3: Safety in UseOS 4: Resistance to Unwanted EntryOS 5: Safety at Construction andDemolition Sites

Structural Safety Objective“to limit the probability that aperson in or adjacent to thebuilding will be exposed to anunacceptable risk of injurydue to structural failure”

Structural Safety Sub-objectivesPrevent:• loads exceeding capacity of buildingelement or supporting medium• Damage/deterioration of elements• Vibration or deflection of elements• Instability of building or part of building• Collapse of excavations

Four Functional StatementsF20: To support and withstandexpected loads and forcesF21: To limit or accommodatedimensional changeF22: To limit movement underexpected loads and forcesF23: To maintain equipment in placeduring structural movement

Why this change?“Codes freeze technology”- Prof. Paul Gavreau, U of TShortcoming of Division ADesigner must define, for a particularstructure constructed using particularmaterials, “unacceptable risk”.

Hence Division B “AcceptableSolutions”• Designs in conformance with theprescriptive requirements ofDivision B are deemed to havemet the Objectives of Division A.• Division B still represents “theminimum a designer is legallyallowed to get away with”.

Questions?

Moving along…• Divisions A and B of NBCC 2005• Companion Action Load Combinations– Rationale– Quantitative– Serviceability• Importance Factors for Buildings• Changes to Specified Loads• Loose Ends• Summary

NBCC 1995 FormatφR > α D D + ψ γ {α L L + α Q Q + α T T}where ψ = load combination factor• NBCC 1995 load combinations:1.25 D + 1.5 L1.25 D + 1.5 Q (wind)1.25 D + 0.7 {1.5 L + 1.5 Q (wind)}= 1.25 D + 1.05 L + 1.05 Q (wind)

Time History of LoadingLoadMaximum Load,for DesignTransientLive/WindSnowSustainedLiveRenovationDeadtime

Turkstra’s Rule (early ’80s)Worst case of combined transientloads occurs when:• one load, the principal action, is itsextreme value• other loads, the companion actions,are the largest that would beexpected while the principal actionhas its extreme value

Companion Action FormatφR > α D D + α i S i + Σ α ik S k , i ≠ kwhere S i = principal actionS k = companion actionsTypical Load Combinations:1.25 D + 1.5 L + 0.5 W (wind)1.25 D + 1.5 W (wind) + 0.5 L

Designer can Envisage Hazards• Correlation of transient loadsexplicitly considered• Can you imagine a structure wheresimultaneous maximum values oftransient loads are:– unlikely?– expected?

Confederation Bridge: Wind + Live

Confederation Bridge: Wind + Ice

1995 NBCC Reliability Indices4Reliability index (50-year)321D + LD + WD + S00 1 2 3 4 5Nominal variable load/nominal dead load• reliability for snow load deficient?

2000/2001 Failures: Sarnia MallCollapseSource: Globe and Mail 09 December 2000

Snow Finds WeaknessesTop ChordEndDiagonalSupportBeam

Huge “Secondary” BendingActual load ~specified load

NBCC 2005 Calibration Process1. Reliability indices for 1995 NBCC2. Preliminary load combinations for 50-yr,500-yr loads by Bartlett, Hong & Zhou• review by Part 4 Task Group on Snow& Wind Loads• review by Part 4 Standing Committee3. Revised load combinations, 50-yr loads• review by Task Group and Part 4 cttee• public review

Calibration: Alberta FarmerWeighs Pig (MacGregor)Step 1:Put pig on plankStep 2:Add rocksuntil plank levelStep 3:Guess weight of rocks

2005 NBCC Combinations1.4 D1.25 D + 1.5 L + (0.4 W or 0.5 S)1.25 D + 1.4 W + (0.5 L or 0.5 S)1.25 D + 1.5 S + (0.5 L or 0.4 W)0.9 D + (1.5 L or 1.4 W or 1.5 S)Add to all combinations:P = prestressH = horizontal earth pressuresT = restrained deformations (safety)

Impact: Single Transient Load1.10Proposed/NBCC 19951.081.061.041.02D+SD+W1.000.98D+L0 1 2 3 4 5Transient-to-dead load ratio

Impact: D+L+S1.21.1Proposed/1995 NBCC1.00.90.8S/D = 0.250.51.02.04.00.70 1 2 3 4 5Live-to-dead load ratio, L/D

Impact: D+L+W1.2W/D = 4.01.1Proposed/1995 NBCC1.00.90.80.25 0.51.02.00.70 1 2 3 4 5Live-to-dead load ratio, L/D

Impact: D+W+S1.21.1Proposed/1995 NBCC1.00.90.250.51.02.0W/D = 4.00.80.70 1 2 3 4 5Snow-to-dead load ratio

Serviceability Limit States• Intention is not to greatly changeserviceability design criteria.• Consequence: specified snow loadincreases but for SLS check apply aSLS load factor of 0.9.• Simplification: eliminate 1-in-10 yrspecified wind load for SLS checkbut apply a SLS load factor of 0.75 tothe specified (1-in-50 yr) wind load.

Table A-4 in Commentary A:“Loads & Load Combinations for Serviceability”Limit States: Vibration serviceabilityOperation of moving equipmentDamage to non-structural comp.Damage to structural componentsGive:• Structural parameters to consider (stresses,accelerations, crack widths, deflections)• Loads + load combinations• References

Table A-5 in Commentary A“Load Combinations for Deflection Limit States”DifferentialSettlementLong TermDeflectionD + H + α L L+ α S SD + H + T P + P + α L L+ α S SShort Term (L + α S S) or (S + α L L)Deflectionor W or E

DrainageWaterrunsdownsurfaceWaterdripsawayfromsurface

Questions?• Far SideG. Larsen

Moving along…• Divisions A and B of NBCC 2005• Companion Action Load Combinations• Importance Factors for Buildings• Changes to Specified Loads• Loose Ends• Summary

Table 4.1.2.1 “ImportanceCategories for Buildings”Low: low direct or indirect hazard tohuman life.Normal: the kitchen-sink category.High: likely to be used as post-disastershelters or contain hazardoussubstances.Post-Disaster: essential to the provisionof services after a disaster.

Importance Factors for S, WImportance Ultimate ServiceabilityCategory (Snow or Wind) Snow WindLow 0.8 0.9 0.75Normal 1.0High 1.15Post Disaster 1.25 0.9 0.75

Questions?

Moving along…• Divisions A and B of NBCC 2005• Companion Action Load Combinations• Importance Factors for Buildings• Changes to Specified Loads– Uncouple L from S– 50 year return periods for W, S• Loose Ends• Summary

Uncoupling Snow & Live• Logical consequence of consideringLive and Snow as independent.• Similar format adopted in ASCE-7based on load combinations derivedin 1980.• Consequence: Members resistinglow D, high L (use + occupancy),high S require less resistance.

Is there a structure down there?Clearing Trans Canada Highwayin Newfoundland(G. Jin, photo)

Live Loads on Roofs• Table 4.1.5.3 specifies 1.0 kPa• Table 4.1.5.10 specifies 1.3 kNCommentary G “ Snow Loads”paragraph 51 states:“These are use and occupancy loads,intended to provide for maintenanceloadings, workmen, and so forth.”“They are not reduced as a functionof area or roof slope”

Return Period forEnvironmental Loads• NBCC 1995 specified:– 30 years for specified Snow, Wind– 10 years for Wind for deflections– 100 years for wind on ImportantStructures• Use 50 year or 500 year returnperiods (only) for 2005 NBCC?• Ratio n-yr/30-yr depends coefficientof variation of annual maximum load

Wind Speed DataCOV of Maximum Annual Wind Velocity0.400.300.200.100.00all data>10 yrs data0 20 40 60 80Longitude Difference (deg) from Alaska/Yukon Boundary

Specified Load Return Period2n-year pressure / 30-year pressure1.510.5mean90% of valuesin this range1-in-5001-in-1001-in-501-in-1000.00 0.05 0.10 0.15 0.20 0.25 0.30COV of Maximum Annual Wind Velocity

50-yr Wind & Snow Specified• typically ~10% greater than 30-yrvalues• snow load factor initially 1.7, impliesa 25% increase in factored load,deemed too big.• factored wind load unchanged,factored snow slightly greater• reduce for SLS checks

Questions?Far SideG. Larson

Moving along…• Divisions A and B of NBCC 2005• Companion Action Load Combinations• Importance Factors for Buildings• Changes to Specified Loads• Loose Ends– Dead load factors– Rain and ponding– Other examples• Summary

Proposed 1.2 D criticized• History: 1.3 proposed for 1975 NBCC.Reduced to 1.25 to maintain sameratio of dead/live load factor as in ACI318-71.• Proposed 1.2 D + 1.6 L not accepted• Specific concerns:– floor thickness variability– dead load of soil & landscaping– tributary area computation

2000 Survey:Concrete Floor Thickness• Marked variability for– Cast-in-place toppings on precast– Cover slabs in unshored compositeconstruction (tolerances?)• “Uncertain D” with load factor of 1.5not adopted• Make allowances for extra dead load• Consider deflections of supportingmembers

Tributary Areas in NBCCCommentary Fig F-1:0.55a0.5aa (typ)lines of zero shearhalfway betweencolumn lines• Safe for corners0.55b0.5b• Unsafe for 1 stinterior columns• Para 6: “For continuous construction,structural analysis is required to find the lines ofzero shear.”

Rain Loads and Ponding• Flat roofs deflect, intensifying load inworst place• Essential design consideration forCalgary and southern Alberta• NBCC Commentary H gives guidance– One-day rain load?– Residual stresses?– Multiple-span cases?• Western M.E.Sc. thesis by Praught

Regions where 1-day rain exceedssnow load on flat roofCalgary(103%)Port Hardy(131%)Tofino(147%)Ucluelet(132%)Leamington (107%)

1-d rain and 24-hr rain• 1-d rain recorded 12:00 to 12:00• 24-hr rain in any 24-hr periodRainIntensity24-hr captures whole storm1-day rain does notTime• 24-hr rain/1-d rain ~ 1.23

WhitehorseRegions where 24-hour rainexceeds snow loadIqaluitYellowknifePort HardyEdmontonStUcluelet/TolfinoVancouverVictoriaCalgaryReginaWinnipegQuebecCharlottetownFrederictonHalifaxMontrealOttawaTorontoLeamington• Similar to Nixon’s (1979) “Ponding Map ofCanada”

NBCC ignores residual stressesInitialEIMomentSecantEI atgiven MCurvatureUnderestimates ponding deflections

Animation by SchoutenEngineering Consultants, NL

Stepwise Analysis1. Load joists2. Calculate joist deflections, reactions3. Apply reactions to beams4. Calculate beam moments, deflections5. Calculate total deflected shape6. Do deflections converge?NO: recompute load and repeat analysisYES: stop, check capacities not exceeded

Gerber Beam Example Calculation• Joists at 1.5m span 10.5m betweenbeams• W530x66 cantilever/anchor beam• W460x52 drop-in beam• Design for Calgary snow load, checkfor 1-day (not 24-hr, not factored) rain

Deflected shape after 1 st iterationJoistDeflectionBeam Deflection

Beam Moments+26%Final (after 6 th iteration)InitialC sym L-20%Volume of water on drop-in & cantileverspans reduces 46%-26%

DeflectionsC LsymInitialFinal

Questions?

David’s Questions1. Composite prestressed concretemembers will have stress limitsbased on dead, live andenvironmental loads. In the past Ihave used working stress loadcombinations to verify the stresses.How is this handled now?

More Questions!2. Why use different load factors forwind loads than for occupancyloads? This does not follow theISO standards recommendationsand does not follow the trend beingfollowed in Europe (EuroCode 1990)and the US (ASCE-7-02 LRFD).

More Questions!3. In the US, ASCE-7-05 uses a returnperiod of 50 years but the C b set at0.70 versus 0.80 in the NationalBuilding Code of Canada. Why arewe increasing the snow loadrelative to the US?

A Loaded Question?4. What loads are temporarystructures to be designed for in thenew building code? Are thestructures to be designed for thesame likelihood of failure in anannual period or are they to bedesigned for the same likelihood offailure over the service life of thestructure?

Questions?

Summary1. Companion action loadcombination format proposed forNBCC 2005:• more realistic representation• permits logical decisions for unusualcases• little difference for many members• consistent with other internationalstandards (ACI 318, AISC LRFD, etc.)

2. Dead loads:• make allowance for extra thickness ofthin toppings• tributary areas for first interiorcolumns3. Snow loads are no longer classifiedwith live loads due to use andoccupancy.• less resistance needed for memberscarrying snow and live loads

4. Only 50-year environmental loadsspecified:• increases specified loads by ~ 10%• additional increases for important andpost-disaster buildings• load factors less than 1.0 reducespecified loads for serviceabilitychecks.5. New load combinations give similardemands to NBCC 1995:• less demand due to snow & live loads• more demand due to snow only

6. Watch out for ponding of rainwateron flat roofs:• 24-hr rain ~1.23 x 1-day rain• residual stresses reduce stiffness• multi-span members can sharewater between spans7. The National Building Code remainsa minimum standard:• consult with owner to confirmwhat really is necessary

“More than ever before,the challenge to theprofession is todevelop designers whohave sufficient intuitionto stand up to, andreject or modify, theresults of a computeraidedanalysis ordesign.”--- J. G. MacGregor,Professor Emeritus, University of Alberta

Additional References• Two papers by Bartlett, Hong &Zhou, Canadian Journal of CivilEngineering, April 2003.• Paper by Praught & Bartlett in 2005CSCE Annual Conference, Toronto.• More information:f.m.bartlett@uwo.ca

Acknowledgements• National Research Council of Canada• National Sciences & EngineeringResearch Council of Canada• NBC Part 4 Task Group on Snow andWind Loads (D. E. Allen, Chair)• Canadian Meteorological Centre• Steel Structures EducationFoundation• J. G. MacGregor