Incorporation of Environmental Factors in Flexible PavementDesign · 2019-11-15 · – AASHTO T...

Incorporation of Environmental Factors in Flexible Pavement Design

Presented at theSoutheastern Pavement Management and Design

ConferenceNashville, TNJune 25, 2002

ByEric C. Drumm

The University of Tennessee

• Eric C. Drumm, University of Tennessee

• Ronald E. Yoder, University of Tennessee

• N. Randy Rainwater, University of Tennessee

• Roger W. Meier, The University of Memphis

Investigators

University of Tennessee

• Gang Zuo, University of Tennessee• Wesley Wright, University of Tennessee• P. Chris Marshall, The University of Memphis

Contributors

Acknowledgments• This work was performed as part of a

research contract supported by the Tennessee Department of Transportation, Nashville, TN. – Harris Scott– Jim Maxwell– Bill Trolinger– Len Oliver– Mike Presley

Introduction• Mechanics-based approach to flexible

pavement design – resilient modulus (Mr) used to define behavior

of each layer in pavement system– Mr is a function of stress level, loading

history, temperature, etc..• Mr of subgrade soil dependent upon

environmental factors– changes in moisture conditions– freeze/thaw state

• High or maximum value– the “best case” properties

may not be safe or reliable• Low or minimum value

– the “worst case” properties may not be economical

• Mean value– “average” properties may

not reflect the relative duration of the seasons during which the “best case” and “worst case” properties occur timetimetimetime

MMMMrrrr

What value(s) of Mr to use for design?

Introduction

Introduction• Laboratory studies investigated variation

of subgrade resilient modulus due to changes in – water content – degree of saturation– matric potential

• But, there is a little data on the seasonal variations of the moisture conditions

AASHTO Guide (1993)• Suggests a procedure to incorporate the

seasonal variation of subgrade Mr in the design process to determine MR– (MR = effective roadbed resilient modulus)

• MR is the single value of subgrade resilient modulus which reflects the cumulative damage from the entire year

• unfortunately, knowledge of the seasonal variation of modulus is required

Introduction

Objectives of Research

• Measure seasonal variations of water content and temperature in pavement systems

• Measure the effects of those variations on pavement structural capacity

• Develop methods to incorporate those effects into existing pavement design procedures used by TDOT

Objective 1

Measure seasonal variations of water content and temperature

inside pavement systems

University of Tennessee

Instrumented Pavement SitesTennessee Department

of Transportation

SUMNER COUNTY

215 mmAsphalt

255 mm Gravel

150 mm LSS

BLOUNT COUNTY

215 mmAsphalt

255 mm Gravel

OVERTON COUNTY

255 mmAsphalt

255 mm Gravel

Pavement SectionsMcNAIRY COUNTY

335 mmAsphalt

255 mm Gravel

LSS = Lime Stabilized Subgrade

Schematic of instrumentation system (section)Schematic of instrumentation system (section)Schematic of instrumentation system (section)Schematic of instrumentation system (section)

Pavement Instrumentation

roadwayroadwayroadwayroadwayshouldershouldershouldershoulder

Data Data Data Data collection collection collection collection vaultvaultvaultvault

Instruments under Instruments under Instruments under Instruments under outer wheel pathouter wheel pathouter wheel pathouter wheel path

Temporary trench for Temporary trench for Temporary trench for Temporary trench for instrument installationinstrument installationinstrument installationinstrument installation

Instrumentation installed under outer wheel path from trench in shoulder

• Time Domain Reflectometry (TDR)– Used to measure water content– Probes have 5 segments– Probes are 5 feet long

Pavement Instrumentation

Buried TDR Probes

Installing TDR ProbesPavement Instrumentation

Pavement Instrumentation

Asphalt Treated Base

Asphalt Treated Base

Surface LayerBinder Layer

Asphalt Concrete

Stone Base

Soil Subgrade

Pavement Edge

TDR Probes

AggregateUnderdrain

Paved Shoulder

TDR Probes

Pavement Instrumentation• Pan Lysimeters

– Measure infiltration through AC layers

– Some located in the wheel path

– Others located under a pavement joint

Pavement Instrumentation

Asphalt Treated Base

Asphalt Treated Base

Surface LayerBinder Layer

Asphalt Concrete

Stone Base

Soil Subgrade

Pavement Edge

TDR Probes

AggregateUnderdrainPan Lysimeter

Paved Shoulder

Pavement Instrumentation

• Thermistors– Measure pavement temperature

Installing Thermistors

– 1” below surface– 1” above stone base– AC mid-height– In soil subgrade

• Weather Station– Air temperature– Solar radiation– Rainfall– Relative humidity– Wind speed

Pavement Instrumentation

Pavement Instrumentation

Cable Trench

Pavement Instrumentation

UndergroundVault

• Data collected once each minute

• Hourly averages are stored on data logger

• Data sent to UT by cell phone once a day

Underground Instrumentation Panel

Pavement Instrumentation

Pavement Instrumentation

Field Verification Tests

Drilling & sampling pavement and subgrade

Field Verification Tests

Pan lysimeter flush test

Material Characterization

• Index tests– Atterberg limits,

Proctor tests, etc.• Resilient modulus tests

– AASHTO Standard Test Method

• Resilient modulus – AASHTO T 307-99 Determining

the Resilient Modulus of Soils and Aggregate Materials

Material Characterization

020406080

100120140160180200

0 10 20 30 40 50 60 70Deviator Stress (kPa)

Res

ilien

t Mod

ulus

(MPa

)

Material Characterization

IncreasingConfiningPressure

Observations: Pavement Temperature

10

15

20

25

30

35

40

10/01/00 10/06/00 10/11/00 10/16/00 10/21/00 10/26/00 10/31/00

Mid

-Dep

th T

empe

ratu

re o

f AC

Lay

er (o C

)

Hourly Temperature

Daily Average

-10

0

10

20

30

40

50

1-Jul-97 1-Jul-98 1-Jul-99 1-Jul-00 1-Jul-01 1-Jul-02Date

Tem

pera

ture

(C)

0.0

0.5

1.0

1.5

2.0

Sola

r Rad

iatio

n (J

/hr)

Top Middle Bottom Air Solar Radiation

Observations: Pavement Temperature

0

10

20

30

40

50

1-Jun-97 1-Jun-98 1-Jun-99 1-Jun-00 1-Jun-01 1-Jun-02Date

Tem

pera

ture

(C

)

Regression Model Measured Data

Observations: Pavement Temperature

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Oct

-97

Nov

-97

Dec

-97

Jan-

98Fe

b-98

Mar

-98

Apr

-98

May

-98

Jun-

98Ju

l-98

Aug

-98

Sep-

98O

ct-9

8

Nov

-98

Dec

-98

Jan-

99Fe

b-99

Mar

-99

Apr

-99

May

-99

Jun-

99

Infil

tratio

n (m

m/m

2 )

McNairy County before binder and surface courses added

Observations: Infiltration

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Oct

-97

Nov

-97

Dec

-97

Jan-

98Fe

b-98

Mar

-98

Apr

-98

May

-98

Jun-

98Ju

l-98

Aug

-98

Sep-

98O

ct-9

8

Nov

-98

Dec

-98

Jan-

99Fe

b-99

Mar

-99

Apr

-99

May

-99

Jun-

99

Infil

tratio

n (m

m/m

2 ) Predicted

Measured

McNairy County before binder and surface courses added

Observations: Infiltration

• Since binder and surface course was placed, infiltration of water has essentially ceased (only small measurable amounts of infiltration).

• Function of lysimeters verified by field flush tests

• Will infiltration increase as pavement ages?

Observations: Infiltration

Observations: Subgrade Water Content

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0 200 400 600 800 1000 1200 1400 1600Time Since 01/01/1997 (Day)

Vol

umet

ric W

ater

Con

tent

0102030405060708090100110120

Tem

pera

ture

(o C) o

r Rai

nfal

l (m

m)

TDR Probe (0.15 m below the Top of the Subgrade)TDR Probe (0.45 m below the Top of the Subgrade)RainfallTemp. 3

Days Since January 1,1997

TDR Probe 0.15 m below top of subgradeTDR Probe 0.45 m below top of subgradeRainfallSoil Temperature

Objective 2

Measure the effects of moisture and temperature variations on pavement structural capacity

Pavement Response• Measured with

Falling WeightDeflectometer

FWD

Pavement Response

Courtesy of Dynatest, Inc.

Pavement Response

ET / E20 = 3.31e-0.060T

0.0

0.5

1.0

1.5

2.0

2.5

0 10 20 30 40 50

Measured Mid-Depth AC Temperature, T (C)

Mo

du

lar

Rat

io, E

T / E

20

Present Study

Pavement Response

ET / E20 = 3.31e-0.060T

ET / E20 = 3.55e-0.063T

0.0

0.5

1.0

1.5

2.0

2.5

0 10 20 30 40 50

Measured Mid-Depth AC Temperature, T (C)

Mo

du

lar

Rat

io, E

T / E

20

Kim et al.

Present Study

Pavement Response

0

5

10

15

20

25

30

0 100 200 300Day of Year

Su

bg

rad

e M

od

ulu

s (k

si)

0.3

0.4

0.5

0.6

0.7

0.8

Vo

lum

etri

c W

ater

Co

nte

nt

Pavement Response

0

5

10

15

20

25

30

0 50 100 150 200 250 300 350Day of Year

Su

bg

rad

e M

od

ulu

s (k

si)

7000-lb Wheel Load10000-lb Wheel Load13000-lb Wheel Load

020406080

100120140160180200

0 10 20 30 40 50 60 70Deviator Stress (kPa)

Res

ilien

t Mod

ulus

(MPa

)

Material Characterization

IncreasingConfiningPressure

0 12 24 36 48 60 72

17294153657789

Radial Offset (in)

Dep

th (i

n)

Summer

0-0.5 0.5-1 1-1.5 1.5-22-2.5 2.5-3 3-3.5 3.5-4

0 12 24 36 48 60 72

17294153657789

Radial Offset (in)

Dep

th (i

n)

Winter

Pavement Response

Objective 3

Develop methods to incorporate environmental effects into TDOT

pavement design procedures

Temperature Averaging

• AC stiffness varies with temperature

• AC stiffness affects the stiffness of underlying stress-dependent materials

• Pavement life estimates are based on the pavement stiffness and so can vary widely depending on AC temperature used in the analysis

Temperature Averaging: Monthly Data

0%

5%

10%

15%

20%

0 5 10 15 20 25 30 35 40 45 50Temperature (C)

Freq

uenc

y

Temperature Averaging: Monthly and Daily Data

0%

5%

10%

15%

20%

0 5 10 15 20 25 30 35 40 45 50Temperature (C)

Freq

uenc

y

DailyMonthly

Temperature Averaging: Monthly, Daily, & Hourly Data

0%

5%

10%

15%

20%

0 5 10 15 20 25 30 35 40 45 50Temperature (C)

Freq

uenc

y

HourlyDailyMonthly

• What is the effect of the temperature averaging interval on computed design life?

• What is the effect of assuming a uniform distribution of traffic throughout the day?

Temperature Averaging

Temperature AveragingEffect of Traffic Distribution

0%1%2%3%4%5%6%7%8%

1 3 5 7 9 11 13 15 17 19 21 23Hour of the Day

Perc

enta

ge o

f Dai

ly T

rips Tractor-Trailer

Trip Distribution

Temperature Averaging

Monthly Average Temps

DailyAverage Temps

Hourly Average Temps

9%Stiff

10%Medium

10%Soft

11%Very soft

Pavement Life OverestimationUsing Uniform Traffic and ...Subgrade

Stiffness

Temperature Averaging

Monthly Average Temps

DailyAverage Temps

Hourly Average Temps

39%9%Stiff

47%10%Medium

54%10%Soft

58%11%Very soft

Pavement Life OverestimationUsing Uniform Traffic and ...Subgrade

Stiffness

Temperature Averaging

Monthly Average Temps

DailyAverage Temps

Hourly Average Temps

52%39%9%Stiff

62%47%10%Medium

71%54%10%Soft

76%58%11%Very soft

Pavement Life OverestimationUsing Uniform Traffic and ...Subgrade

Stiffness

Temperature Averaging

0.E+00

2.E+06

4.E+06

6.E+06

8.E+06

1.E+07

1.E+07

1.E+07

Very Soft Soft Medium Stiff

Subgrade Soil Type

ESA

Ls

Hourly (Nonuniform Traffic)

Hourly (Uniform Traffic)

Daily (Uniform Traffic)

Monthly (Uniform Traffic)

• To obtain the most reliable estimates of design life, use hourly temperature data whenever possible.

• If hourly truck traffic distribution is known, reliability can be increased even more.

Temperature Averaging

Summary and Conclusions

Summary

• Four years of data at 4 sites in TN– Climatic data– Pavement temperature– Base and subgrade moisture– Infiltration– Layer moduli

• This is a rare and valuable dataset

Conclusions

• TDOT surface courses are relatively impermeable (at least early on)

• TDOT designs limit the stresses in subgrade

• Water content of subgrade may not change much seasonally (at least in fill sections)

Conclusions

• FWD is sensitive enough to pick up stress-dependence of subgrade soils

• Using monthly average temperatures to estimate asphalt modulus can lead to under-designed pavements

Issues to Address

Issues to Address

• As instrumented pavements age, how will they respond to wheel loads?– AC properties will certainly change– Infiltration will probably increase– Drainage layers may stop working– Water content may change long-term

Issues to Address

• Do “cut” sections respond the same as “fill” sections now and over time– Measure FWD response in cut areas at

the same time as in fill areas– Measure water content (by sampling) and

compare to TDR data in fill sections

Questions?