1

FDOT MECHANISTIC EMPIRICALFDOT MECHANISTIC-EMPIRICAL PAVEMENT DESIGN UPDATE

Project Manager Bruce Dietrich, P.E., FDOT

ResearchersEmmanuel Fernando, TTIJeongho Oh, TTI

Overview of ME-PDG Program

2

M-E Pavement Design Program

• Available from the web at• Available from the web at http://www.trb.org/mepdg/ for evaluation

• Program is a tool for pavement analysis

• M difi ti i i f th• Modifications arising from other research are expected

• Consider need for calibration during implementation

General Information- Site/Project Identification

Analysis Status

Inputs-Traffic-Climatic-Structure

Input & OutputSummary

3

Model Verification/Calibration

• M E PDG models calibrated using a• M-E PDG models calibrated using a national data base of LTPP sections

• Calibration to local conditions is important

• Calibration factors input to programCalibration factors input to program for specific distresses

Model Calibration Factors

4

Map of Calibration Sections

B df d (1 0)

Gadsden (1,1)

Polk (2 ,1)

St.Johns (0,1)

Bradford (1,0)

Alachua (1,0)Santa Rosa (1,0)

Palm Beach(2 0)

Lake (0,1)Seminole (1,0)

Volusia (1,0)

Hillsborough (2,3)

Pinellas (0,4)

15 HMAC sections

Charlotte (0,1)

Polk (2 ,1)

Broward (1,0)

(2,0)

Monroe (1,0) Dade (1,4)

16 PCC Sections

FIELD AND LABORATORY TESTS

5

Field Tests on Concrete Sections

Measurements of joint spacing & dowel diameter

Verification of M-E PDG Performance Predictions onPerformance Predictions on Calibration Sections

6

Transverse Cracking on Calibration Sections

100

40

60

80

Tran

sver

se C

rack

ing

(%)

0

20

0 20 40 60 80 100

Measured Transverse Cracking (%)

Pred

icte

d T

IRIs on Calibration Sections

7

Faulting on Calibration Sections

Failure Criteria for Generating Design Tables

Distress Criterion

Transverse cracking 10 percent slabs cracked

Faulting 0.12 inch

IRI 180 in per mile

8

Locations Considered inLocations Considered in Comparisons of Rigid Pavement Cross-Sections

Findings from Comparisons of Rigid Pavement Cross-Sections

• Cross sections 2 and 3• Cross-sections 2 and 3 gave same slab thicknesses

• These cross-sections gave up to 0.5-inch thicker slabs than the otherslabs than the other cross-sections

• Rigid design tables based on cross-section 3

9

10

Effect of Slab Width• Performance predictions compared• Performance predictions compared

for 12, 13 and 14 ft slab widths• M-E PDG predicts better performance

for 13-ft wide slab with tied shoulders

• Design tables for 13 ft wide slabs developed

Consideration of Regional Conditions• Used M E PDG data base to• Used M-E PDG data base to

characterize climatic conditions in each county

• Used soil-water characteristic curves established from county soil survey y yreports and soil suction tests done in the laboratory

11

95% reliability and 50 M Design ESALs

Map of Design Slab Thicknesses

SENSITIVITY ANALYSIS

12

Sensitivity to Base Modulus100

40

60

80

cent

Cra

cked

Sla

b

0

20

0 2 4 6 8 10 12 14 16 18 20

Year

Per

40 ksi 50 ksi 60 ksi

Sensitivity to Subgrade Modulus

100

40

60

80

cent

Cra

cked

Sla

b

0

20

0 2 4 6 8 10 12 14 16 18 20

Year

Perc

15 ksi 25 ksi 35 ksi

13

Effect of k-value

Effect of CTE on Crack Predictions

30

15

20

25

ckin

g (%

) at 9

5% re

liabi

lity

0

5

10

0 50 100 150 200 250 300

Month

Tran

sver

se c

ra

5.5 per F*10-66.0 per F*10-66.5 per F*10-6

14

Effect of CTE on Ride Predictions

Coefficient of thermal expansion

Effect of Concrete Compressive Strength

40

20

25

30

35

acki

ng (%

) at 9

5% re

liabi

lity

0

5

10

15

0 50 100 150 200 250 300

Month

Tran

sver

se c

ra

3000 psi4000 psi5000 psi

15

Effect of Slab Width30

15

20

25

acki

ng (%

) at 9

5% re

liabi

lity

0

5

10

0 50 100 150 200 250 300

Month

Tran

sver

se c

ra

12 ft13 ft14 ft

Effect of Slab Length

16

M-E PDG Rigid PavementM E PDG Rigid Pavement Design Tables

Design Assumptions

• CTE of 6 microstrain per °F• CTE of 6 microstrain per °F• 28-day compressive strength: 4000

psi• Slab widths: 13 and 14 ft• Joint spacing: 15 ft• Joint spacing: 15 ft• Subgrade k-value: 200 lb per in3

• Initial IRI: 58 inches per mile

17

Design Assumptions

• Minimum slab thickness of 8 inches• Minimum slab thickness of 8 inches• Dowel diameters:

1 inch for 8 to 8.5-inch slabs1.25 inches for 9 to 10.5-inch slabs1.5 inches for 11-inch and thicker slabs1.5 inches for 11 inch and thicker slabs

Sensitivity to CTE80

20

30

40

50

60

70

erce

nt S

labs

Cra

cked

μ − 2σμμ + 2σ

0

10

20

Andes

iteBas

alt

Diabas

e

Granite

Schist

Chert

Dolomite

Limes

tone

Quartz

ite

Sands

tone

Aggregate Type

Pe

from Mallela et al.

18

Effect of Initial IRI120

60

80

100

mile

) at 9

5% re

liabi

lity

0

20

40

0 50 100 150 200 250Month

IRI (

in/m

50 in/mile58 in/mile65 in/mile

Implementation Considerations• Predicted performance sensitive to CTE• Predicted performance sensitive to CTE• Need to develop and implement specs to

measure and control CTEs for construction quality assurance purposes

• Predicted performance also sensitive to initial IRIinitial IRI

• Need ride specification to get desired initial smoothness

19

Establishing Preliminary Flexible Pavement Design Tablesg

Bruce Dietrich, P.E.E G Fernando Ph D P EE.G. Fernando, Ph.D., P.E.Jeongho Oh, Ph.D., P.E.

Sensitivity AnalysisVariables considered in New AC

AC Mix combination, AC material propertyModulus of Base, Subgrade and EmbankmentThickness of Base Cumulative ESALsEnvironmental RegionReliability

20

Climatic Effect on Alligator Crack of AC Pavements

40

15

20

25

30

35ac

king

(%) a

t 95%

relia

bilit

y

0

5

10

15

0 50 100 150 200 250 300

Month

Alli

gato

r cra

MiamiTallahassee

Base Modulus Effect on Alligator Crack of AC Pavements

60

30

40

50

acki

ng (%

) at 9

5% re

liabi

lity

0

10

20

0 50 100 150 200 250 300

Month

Alli

gato

r cra

30 ksi 45 ksi60 ksi

21

PG Grade Effect on Alligator Crack of AC Pavements

40

45

20

25

30

35

40

acki

ng (%

) at 9

5% re

liabi

lity

0

5

10

15

0 50 100 150 200 250 300

Month

Alli

gato

r cra

PG 64-22PG 70-22PG 76-22

Air Void Effect on Alligator Crack of AC Pavements

60

30

40

50

acki

ng (%

) at 9

5% re

liabi

lity

0

10

20

0 50 100 150 200 250 300

Month

Alli

gato

r cra

7%8%9%

22

Base Thickness Effect on Alligator Crack of AC Pavements

40

15

20

25

30

35ac

king

(%) a

t 95%

relia

bilit

y

0

5

10

15

0 50 100 150 200 250 300

Month

Alli

gato

r cra

10 inch

15 inch

Embankment Modulus Effect on Alligator Crack of AC Pavements

40

15

20

25

30

35

acki

ng (%

) at 9

5% re

liabi

lity

0

5

10

15

0 50 100 150 200 250 300

Month

Alli

gato

r cra

8 ksi10 ksi12 ksi

23

ESAL Effect on Alligator Crack of AC Pavements

40

15

20

25

30

35ac

king

(%) a

t 95%

relia

bilit

y

0

5

10

15

0 50 100 150 200 250 300

Month

Alli

gato

r cra

TTC 6 with AADTT=11250TTC 1 with AADTT=8600TTC 14 with AADTT = 16000

50 Million ESALs

AC Mix Effect on Alligator Crack of AC Pavements

Mi t C bi ti FC 5 FC 9 5 FC 12 5Mixture Combination FC 5 FC 9.5 FC 12.5

SP 9.5 42.27 41.64 41.81

SP 12.5 42.27 41.64 41.81

SP 19.0 41.99 41.34 41.64

Modified SP 41.27 40.64 40.91

24

Criteria of Preliminary Design Table for New Construction

1-inch FC 12.5

SP 12.5

10-inch LR Base

12-inch Stab.

Only alligator cracking criterion (35%, which is a default limit)FC 12.5 over SP 12.5 AC mixtureBase modulus = 30, 45, and 60 ksiSubgrade modulus = 16 ksiEmbankment modulus = 12 ksiMiami weather data used

Subgrade

Embankment

20 yrs design lifeFive reliability levels same as PCCCumulative ESALs considered

Sensitivity Analysis for Overlay

Variables considered in Overlay ACVariables considered in Overlay ACAC Mix combination, AC material propertyModulus of Base, Subgrade and EmbankmentThickness of Base Cumulative ESALsEnvironmental RegionReliabilityExisting Pavement ConditionExisting AC layer thickness after milling

25

Alligator cracking in New vs. Overlay Pavements with Identical Structures

35

40

10

15

20

25

30

35

Bot

tom

-up

Cra

ck (%

) Overlay

Existing

εWhere is from?

0

5

10

0 50 100 150 200 250

Age (Month)

Mean prediction_New Mean prediction_Resurfacing95% Reliability_New 95% Reliability_Resurfacing

εt Where is from?

Alternative Way to Model Overlay

Use new/reconstruction options inUse new/reconstruction options in M-E PDG to model existing and overlay lifts

Vary PG-Grade or Air Void to represent different existing pavement conditions

FDOT Practice

Good: No cracking, minor rutting

Fair: Crack rating ≥ 8

Poor: Crack or rutting rating ≤ 7

FDOT Practice

26

Effect of PG Grading vs. Air Void

40

4560

20

25

30

35

40ra

ckin

g (%

) at 9

5% re

liabi

lity

30

40

50

rack

ing

(%) a

t 95%

relia

bilit

y

0

5

10

15

0 50 100 150 200 250 300

Month

Alli

gato

r cr

PG 64-22PG 70-22PG 76-22

0

10

20

0 50 100 150 200 250 300

Month

Alli

gato

r cr

7%8%9%

Findings from Alternative Approach

We use a fixed PG-Grade for existing layer asWe use a fixed PG Grade for existing layer as PG 70-22

Varying air void of existing layer for different pavement conditions to have comparable overlay thickness to those obtained from FDOT design procedureg p

8.0 percent for poor condition

7.5 percent for fair condition

7.2 percent for good condition

27

Criteria of Preliminary Design Table for Overlay Construction

1 inch FC 12 5 Only alligator cracking criterion 1-inch FC 12.5

Existing SP 12.52 ~ 6 inch

10-inch LR Base

12 inch Stab

y g g(35%, which is a default limit)FC 12.5 over SP 12.5 AC mixtureBase modulus = 30, 45, and 60 ksiSubgrade modulus = 16 ksiEmbankment modulus = 12 ksiMiami weather data used20 yrs design lifeFi li bilit l l PCC

Overlay SP 12.5

12-inch Stab. Subgrade

Embankment

Five reliability levels same as PCCCumulative ESALs considered same as PCCPoor/Fair & Good condition

Comparison of AC Overlay Design Thickness based on FDOT Practice vs. M-E PDG

4.0

4.5

4.0

4.5

50 million ESALs

1.5

2.0

2.5

3.0

3.5

-12.

5) o

verla

y th

ickn

ess

(in) 25 million ESALs

with 2” existing AC

1.5

2.0

2.5

3.0

3.5

-12.

5) o

verla

y th

ickn

ess

(in)

with 4” existing AC

0.0

0.5

1.0

75 80 85 90 95

Reliability (%)

AC

(SP

FDOT_Poor M-E PDG_Poor FDOT_Good M-E PDG_Good

0.0

0.5

1.0

75 80 85 90 95

Reliability (%)

AC

(SP

FDOT_Poor M-E PDG_Poor FDOT_Good M-E PDG_Good

28

Questions?