SHRP R19BSHRP R19B BRIDGE FOR SERVICE LIFEBRIDGE …sp.bridges.transportation.org/Documents/SCOBS...BRIDGE FOR SERVICE LIFEBRIDGE FOR SERVICE LIFE BEYOND 100 YEARS: SERVICE LIMIT STATE

SHRP R19BSHRP R19B

BRIDGE FOR SERVICE LIFEBRIDGE FOR SERVICE LIFE BEYOND 100 YEARS:

SERVICE LIMIT STATE DESIGNPresentation to HSCOBS 2012

Research Team

M dj ki d M t I J h K li ki Ph D P E

Research Team

Modjeski and Masters, Inc.: John Kulicki, Ph.D., P.E.Wagdy Wassef, Ph.D., P.E.

University of Delaware: Dennis Mertz, Ph.D., P.E.University of Nebraska: Andy Nowak, Ph.D.NCS Consultants: Naresh Samtani, Ph.D., P.E.

TRB/SHRPJerry DiMaggioJerry DiMaggio

Formerly Mark Bush

Expected ProductsExpected Products• Framework for calibration• Bridge design procedures and proposed

specificationsspecifications • Tools required for future SLS improvement

using calibration frameworkusing calibration framework– Data bases team used

Modification procedures– Modification procedures

SHRP R 19B and NCHRP 12-83SHRP R 19B and NCHRP 12 83• Both projects have same core team – able to

take advantage of synergytake advantage of synergy• 12-83 deals with concrete SLS only and

R19B inherits results – modifies as neededR19B inherits results modifies as needed• R19B

Framework deterioration– Framework – deterioration– Most of Live Load development– General SLS– General SLS– Steel, foundations, bearings, joints

• Fatigue - joint• Fatigue - joint

Topics Covered Here Last YearTopics Covered Here Last Year

• Sources of WIM data and filteringSources of WIM data and filtering• Effect of removing only a few heavy trucks

A l i f lti l d• Analysis of multiple presence and conclusions for SLS

• Preliminary results for non-fatigue SLS live load

• Approach to Fatigue II

Progress Since Last Report toProgress Since Last Report to HSCOBS

Non-Fatigue SLS LL Model • Mean Bias and project LL model at mean• Mean, Bias and project LL model at mean

plus 1.5 standard deviations tabulated with and without DLA for parameters:and without DLA for parameters:– 5 ADTTs = 250, 1,000, 2500, 5000 and 10,000– 10 Time periods = 1 day 2 weeks 1 month 2 months 6– 10 Time periods = 1 day, 2 weeks, 1 month, 2 months, 6

months, 1 year, 5 years, 50 years, 75 years and 100 years

– 6 Spans = 30 ft, 60ft, 90ft,120ft, 200 ft & 300ft– With and w/o DLA

Progress in Last Year –S f S SStatistics of Non-Fatigue SLS Live Load

• Based on 95% limit:Based on 95% limit:– ADTT = I,000, Project Bias on HL 93 = 1.4– ADTT = 5,000, Project Bias on HL 93 = 1.45ADTT 5,000, Project Bias on HL 93 1.45

• COV = 12%• Based on 100 years:• Based on 100 years:

– Project Bias varies with time interval which will be reflected in calibrated load factorbe reflected in calibrated load factor

– Not strongly influenced by span length

Typical Results For SLS Live Load ModelTypical Results For SLS Live Load Model

Span 60 ft

1 20

1.40

1.60

0.80

1.00

1.20

Bias

ADTT 250

ADTT 1000

0.40

0.60

B

ADTT 2500

ADTT 5000

ADTT 10000

0.00

0.20

1 10 100 1000 10000 100 years

DaysDays

Basis for Current Service II For SteelBasis for Current Service II For Steel

~500,000 cycles of same load, y

WIM Input For Service II For SteelWIM Input For Service II For Steel

• After Filter #1 – includes “Permit Loads”• Prior elimination of several sites• 80 % of “Permit Loads” came from one80 % of Permit Loads came from one

site – excluded it• Ratios to HL93 (current=1 30)• Ratios to HL93 (current=1.30)

– Ratio =1.1 yields average = ~1 per monthR ti 1 2 i ld

Fatigue IFatigue I

• Moment based on “1/10 000” criteria inMoment based on 1/10,000 criteria in current spec.

Fatigue II - Damage Factor Compared to Current( ) 3/ r cF a t T rk nM M −

Current =0.75

( ) 3/ F a t T rke qA A S H T O

M Mn

Current  0.75

30 ft 60 ft 90 ft 120 ft 200 ft

0.52 0.71 0.66 0.68 0.73

0.57 0.74 0.71 .73 0.78

0 55 0 78 0 73 0 73 0 800.55 0.78 0.73 0.73 0.80

High = 0 87 or 116% of currentHigh = 0.87 or 116% of current

Fatigue II – Load SideFatigue II Load Side

• Fatigue I – 2 0 HL93 COV = 0 12Fatigue I 2.0 HL93, COV = 0.12• Fatigue II - 0.80 HL93, COV=0.07

R i d “ ”• Revised “n”

Longitudinal Members nSimple Span Girders 1.0

near interiorContinuous Girders 

near interior support 

1.5

elsewhere 1.0

Fatigue – Resistance Side and CalibrationFatigue Resistance Side and Calibration

COV BiasS C Probability COV, BiasS-n Curves ProbabilityPaper

β

With help from Drs. Keating, Fisher, Yen. Connor and Roy

Fatigue – ConcreteFatigue Concrete

• Underway at UDELUnderway at UDEL

Footing Settlement - Article 10.5.2Footing Settlement Article 10.5.2• Include the δ-0 Concept • Compute angular distortion, Adist = δ/L based

on consideration of construction points• Compute modified deformation δm, and

angular distortion, Adistm, as follows:g , distm,δm= γSE (δ) Adistm= γSE (Adist)

where γ is the load factor due to settlementwhere γSE is the load factor due to settlement• Perform bridge analysis using δm and Adist

17

Data for Immediate Settlement of Spread F ti (FHWA 1987)Footings (FHWA 1987)

ches

red, in

Measur

Calculated inches

M

18

Calculated, inches

Statistics of Various MethodsStatistics of Various Methods

P k & B l d &Statistic Schmertmann Hough D’Appolonia

Peck & Bazzara

Burland & Burbridge

Count 20 20 20 20 20Min 0.295 0.656 0.311 0.202 0.138Max 4.618 4.294 2.176 4.000 4.735M 1.381 1.971 1.031 0.779 0.829M 1.381 1.971 1.031 0.779 0.829SD 1.006 0.769 0.476 0.796 0.968COV 0.729 0.390 0.462 1.022 1.168

LegendLegend: M = MeanSD = Standard Deviation COV = Coefficient of Variation (=SD/M)

19

20

Section 3, New Table 3.4.1-4 for γSESection 3, New Table 3.4.1 4 for γSE

Deformation γDeformation γSEImmediate Settlement• Hough method 1.0g• Schmertmann method• XYZ method (e.g., a regional method)

1.21.X

Consolidation settlement 1.0Lateral deformation • P y or SWM soil structure interaction method 1 2• P‐y or SWM soil‐structure interaction method• XYZ method (e.g., a regional method)

1.21.X

21

Concrete-Related Limit StatesConcrete Related Limit States

LRFD Description Proposed SLS

articleDescription Proposed SLS

5.7.3.4Control of cracking bydistribution of reinforcement

Service I‐A:Crack control of R/C/

9.7.2.5Reinforcement requirementsfor concrete deck

Service I‐B:Crack control of R/C concrete deckService III‐A: Decompression

5.9.4.2Stresses check at service IIIlimit state after losses‐fully

t d t

Service III A: DecompressionService III‐B: Un‐cracked section (max tensile stress)S i III C C k d tiprestressed components Service III‐C: Cracked section (specified crack width)

Current Status of LL StudiesCurrent Status of LL Studies

• Service I: Deck Crack ControlAxle load, not truck load

Mean of Axle Load for Various ADTTsADTT 1 day 2 weeks

1 month

2 months

6 months

1 year

5 years

50 years

75 years

100025.60 30.50 32.00 33.40 34.90 36.30 38.20 39.70 39.90

250027.70 32.40 33.70 34.90 36.20 37.60 39.10 40.40 40.60

500028.80 33.80 34.80 35.90 37.20 38.00 39.60 40.60 40.90

1000029 90 34 70 35 90 36 70 38 00 38 90 40 10 41 30 41 40

Mean, Bias, Coefficient of Variation at Mean +1.5*

29.90 34.70 35.90 36.70 38.00 38.90 40.10 41.30 41.40

Reliability Indices for Existing Decks C kiCracking

Benign Exposure conditions(max. Crack width 0.016”)

Normal Exposure conditions(max. Crack width 0.012”)

Reliability index for decks designed for LL factor of 1

Severe Exposure conditions(max Crack width 0 008”)

for LL factor of 1.

Assuming ADTT 5000

(max. Crack width 0.008 )

Reliability Indices of Existing P/S Conc. B idBridges

Decompression Max. Allowable Tension

Reliability index of bridges designed for LL factor of 0.8, maximum concrete tensile strength of

Max. Allowable Crack Width (0.016 in., 1 year return period)

concrete tensile strength ofAssuming ADTT 5000

Framework For Calibration – Audience, Expertise and Timeline

• Audience/expertiseAudience/expertise– specification developers

researchers– researchers• Timeline

– 4 demo examples in Phase 1 report– Continued development thru Phase 2– After Academy review-----Spring 2014

Spec Product – Audience, Expertise and Timeline

• Initial Audience – HSCOBS (T5, T10, T14, T15)Initial Audience HSCOBS (T5, T10, T14, T15)• Ultimate Audience – All AASHTO LRFD

usersusers• Expertise – same as current AASHTO LRFD

usersusers• Timeline

– Concrete 2013 meeting– Rest 2014 meeting

Toolkit Audience - Expertise and Timeline

• Available to future Spec developers and p presearchers

• ExpertiseExpertise– Reliability theory– Bridge Design and evaluation– Bridge Design and evaluation– Bridge inspection helpful

• Timeline• Timeline –– Currently Spring 2014

Thank youThank you.