comparison of rutting and cracking performance estimates ... 2015/hveem_sp_trb2015.pdf ·...

COMPARISON OF RUTTING AND CRACKING PERFORMANCE ESTIMATES FOR

HVEEM AND SUPERPAVE MIX DESIGNS UTILIZING CALME

Erdem Coleri Assistant Professor

Oregon State University

94th Annual TRB Meeting, Washington D.C. January 13th, 2015

Prof. John Harvey, University of California, Davis (PI) Dr. James M. Signore, Nichols Consulting Engineers Venkata Lakshman Mandapaka, Caltrans Khushminder Singh Sahasi, Caltrans

Co-authors:

• Hypothesis tested: Since Superpave mix design produces higher binder

contents than Hveem mix design, Hveem mixes provide better rutting performance and worse cracking performance than Superpave mixes

• Use CalME to investigate the pros and cons of using

Hveem and Superpave mix design

RESEARCH OBJECTIVES

Caltrans is shifting the HMA design method from Hveem to Superpave.

OUTLINE

Background Materials Testing

CalME simulations

Conclusions

• Mix A, dense graded HMA (NMAS=3/4’’), conventional

PG 64-16 binder (Refinery 1), crushed alluvial aggregate

• Mix B, dense graded HMA (NMAS=3/4’’), conventional

PG 64-16 binder (Refinery 2), crushed basalt aggregate

• Mix I, gap-graded HMA (NMAS=1/2’’), rubberized binder

with PG64-16 base binder (Refinery 2), crushed basalt

aggregate

BACKGROUND Mix types

• Mix A, Hveem binder content = 5.0 %

Superpave binder content = 5.5 %

• Mix B, Hveem binder content = 5.2 %


• Mix I, Hveem binder content = 8.0 %


BACKGROUND Mix types – Binder contents

Testing Program

Rutting Repeated Simple Shear at Constant Height

100 kPa shear stress Temperature: 45 and 55oC

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0 5,000 10,000 15,000 20,000 25,000 30,000

Perm

anen

t She

ar S

trai

n

Cycles

5% PSS Failure Line

Test Results Rutting Repeated Simple Shear at Constant Height

Mix A

Mix B

Mix I

Testing Program

Cracking Stiffness Flexural fatigue Flexural frequency sweep

200 and 400 micro strain Temperature: 20oC

Temperature: 10, 20 and 30oC Various frequencies

Test Results Flexural fatigue

1,000

2,000

3,000

4,000

5,000

6,000

7,000

1,000 10,000 100,000 1,000,000 10,000,000

Flex

ural

Fat

igue

Stif

fnes

s (M

pa)

Cycles

200 MicrostrainMix A

Mix B

Mix I

Rutting performance evaluation

Factorial to evaluate Hveem vs. Superpave mix design methods

2 structures x 2 traffic levels x 3 climates x 3 mixes

Structures: Structure 1: 125mm thick AC layer on top of 300mm thick AB (Eab=250MPa) and a subgrade with 150MPa stiffness.

Structure 2: 175mm thick AC layer on top of 300mm thick AB (Eab=250MPa) and a subgrade with 150MPa stiffness.

Traffic: Traffic 1: Axles 1st year = 3M with a 5% growth rate

Traffic 2: Axles 1st year = 6M with a 5% growth rate

Climates: Inland Valley

High Desert

Low mountain

Mixes: Mix A; Mix B; Mix I

Rutting comparison Hveem vs. Superpave

0

2

4

6

8

10

12

14

16

18

20

0 2 4 6 8 10 12 14 16 18 20

Surface rut - Hveem

Surfa

ce ru

t - S

uper

pave

Surface rut (m

m)

(mm)


Welch modified two sample t-test for surface rut predicted for Hveem and Superpave mixes

F1 and F2 are two distributions, the possible hypotheses and alternatives concerning these distributions are:

H0: F1(x) = F2(x) HA: F1(x) ≠ F2(x) Decision rule: Reject H0 if p-value < 0.10; accept H0 if

p-value ≥ 0.10

P-value = 0.09 for Hveem vs.Superpave

Predicted surface rutting for Hveem and Superpave mix design method are not equal. Superpave is higher.

0246

81012

0 10 20 30 40 50Years

Surf

ace

rutt

ing

(mm

)

SuperpaveHveem

Rutting comparison Hveem vs. Superpave Surface rut for Mix A - Traffic level 1

02468

1012

0 10 20 30 40 50Years

Surf

ace

rutt

ing

(mm

)

SuperpaveHveem

02468

1012

0 10 20 30 40 50Years

Surf

ace

rutt

ing

(mm

)

SuperpaveHveem

High Desert – Thin AC High Desert – Thick AC Inland Valley – Thin AC

02468

1012

0 10 20 30 40 50Years

Surf

ace

rutt

ing

(mm

)

SuperpaveHveem

02468

1012

0 10 20 30 40 50Years

Surf

ace

rutt

ing

(mm

)

SuperpaveHveem

0246

81012

0 10 20 30 40 50Years

Surf

ace

rutt

ing

(mm

)

SuperpaveHveem

Inland Valley – Thick AC Low mountain – Thin AC Low mountain – Thick AC

Rutting comparison Hveem vs. Superpave AC rut for Mix A - Traffic level 1



01234567

0 10 20 30 40 50Years

AC r

uttin

g (m

m)

SuperpaveHveem

01234567

0 10 20 30 40 50Years

AC r

uttin

g (m

m)

SuperpaveHveem

01234567

0 10 20 30 40 50Years

AC r

uttin

g (m

m)

SuperpaveHveem

01234567

0 10 20 30 40 50Years

AC r

uttin

g (m

m)

SuperpaveHveem

01234567

0 10 20 30 40 50Years

AC r

uttin

g (m

m)

SuperpaveHveem

01234567

0 10 20 30 40 50Years

AC r

uttin

g (m

m)

SuperpaveHveem

Rutting comparison Hveem vs. Superpave Surface rut for Mix B - Traffic level 1



0

2

46

8

10

12

0 10 20 30 40 50YearsSu

rfac

e ru

ttin

g (m

m)

SuperpaveHveem

0

2

46

8

10

12

0 10 20 30 40 50Years

Surf

ace

rutt

ing

(mm

)

SuperpaveHveem

0

2

46

8

10

12

0 10 20 30 40 50Years

Surf

ace

rutt

ing

(mm

)

SuperpaveHveem

0

2

46

8

10

12

0 10 20 30 40 50Years

Surf

ace

rutt

ing

(mm

)

SuperpaveHveem

0

2

46

8

10

12

0 10 20 30 40 50Years

Surf

ace

rutt

ing

(mm

)

SuperpaveHveem

0

2

46

8

10

12

0 10 20 30 40 50Years

Surf

ace

rutt

ing

(mm

)

SuperpaveHveem

Cracking of thin layered structures is causing third stage rutting accumulation.

Cracking performance evaluation

0

5

10

15

20

25

30

35

40

0 5 10 15 20 25 30 35 40

Years to 0.5m/msq cracking - Hveem

Year

s to

0.5m

/msq

crac

king

- Sup

erpa

ve

Cracking comparison Hveem vs. Superpave Years to 0.5m/msq cracking – Mix A

3 CASES

Year 40 means no cracking larger than 0.5m/msq in 40 years

0

5

10

15

20

25

30

35

40

0 5 10 15 20 25 30 35 40


Year

s to

0.5m

/msq

crac

king

- Sup

erpa

ve

Cracking comparison Hveem vs. Superpave Years to 0.5m/msq cracking – Mix B


0

5

10

15

20

25

30

35

40

0 5 10 15 20 25 30 35 40


Year

s to

0.5m

/msq

crac

king

- Sup

erpa

ve

Cracking comparison Hveem vs. Superpave Years to 0.5m/msq cracking – Mix I


Two cases to compare Hveem and Superpave:

1. Evaluate the reflection cracking resistance of thin Mix I (Rubber) layer on top of cracked PCC

2. Do thickness design for Hveem and Superpave

mix designs and answer the question: What is the cost of using Hveem and Superpave mix designs to achieve equal performance?

Superpave uses more binder but higher cracking performance reduces the material volume used for construction

1. 3.2M ESALS - 20 years (1) 2. High Desert; Low Mountain (2) 3. .

4. Hveem and Superpave (2)

24 cases x 1 (with reflection cracking) = 24 Cases

60 mm

125 and 175 mm Cracked PCC - 35,000MPa 3 mixes – Mix A; Mix B; Mix I

100 mm AB-Class2 - 200MPa

SG – CL - 100MPa

(6)

Crack and seat –Crack spacing 1.2m

Case 1:

Low Mountain

Structure 1 – Thin PCC - Hveem Structure 2 – Thick PCC - Hveem

Structure 1 – Thin PCC - Superpave Structure 2 – Thick PCC - Superpave

0

1

2

3

4

5

0 10 20 30 40

Crac

king

, m /

msq

Years

P001-Mix AS311-Mix IS410-Mix B

0

1

2

3

4

5

0 10 20 30 40

Crac

king

, m /

msq

Years


0

1

2

3

4

5

0 10 20 30 40

Crac

king

, m /

msq

Years


0

1

2

3

4

5

0 10 20 30 40

Crac

king

, m /

msq

Years


1. 2.5M ESALS - 20 years: TI10 & 40M ESALS – 20 years: TI14 (2) 2. Inland Valley (1) 3. .


VARIABLE Mix A


SG – CL - 100MPa

(multiple)

Amount of binder required for Hveem and Superpave mix designs to achieve equal performance

Case 2:

0

1

2

3

4

5

6

7

8

9

10

156 158 160 162 164 166 168

Crac

king

at y

ear 2

0 (m

/msq

)

Thickness (mm)

TI10 - Hveem (103 mm)

0

1

2

3

4

5

6

7

8

9

10

98 100 102 104 106

Crac

king

at y

ear 2

0 (m

/msq

)

Thickness (mm)

Design criteria: Cracking at year 20 = 1.6m/msq

Rutting at year 20 = 4.8 mm

TI14 - Hveem (162 mm)


TI10 - Superpave (84 mm) TI14 - Superpave (122 mm)

0

1

2

3

4

5

6

7

8

9

80 82 84 86 88 90

Crac

king

at y

ear 2

0 (m

/msq

)

Thickness (mm)


0

1

2

3

4

5

6

7

8

118 120 122 124 126

Crac

king

at y

ear 2

0 (m

/msq

)

Thickness (mm)


Amount of binder used for Hveem and Superpave

TI 10 - Vmix (m3) TI 14 - Vmix (m3)

Hveem 555.30 873.39

Superpave 452.87 657.74

Design AC (%) TI 10 - Vbinder (m3) TI 14 - Vbinder (m3)

Hveem 5.00 27.77 43.67

Superpave 5.50 24.91 36.18

Volume of asphalt mix used for a 1 mile section for a 3.35m lane width

For TI 10 - The amount of binder required is 11.5 percent larger for the Hveem mix

For TI 14 - The amount of binder required is 20.7 percent larger for the Hveem mix

To get equal performance:

Conclusions Lower binder content for Hveem mix design results in

• better rutting performance • worse cracking performance

Difference in cracking resistance between Hveem and Superpave

mix designs is not significant for the rubber mix (Mix I)

For TI 10: The amount of binder required is 11.5 percent larger for the Hveem mix TI 14: The amount of binder required is 20.7 percent larger for the Hveem mix Superpave reduces cost for Mix A for the climate and traffic used in the simulations.

WHAT HAPPENS IF RUTTING IS THE CONTROLLING FACTOR FOR DESIGN?

FUTURE STUDY:

Thank you!

Q & A

http://www.ucprc.ucdavis.edu/

Results • Mix I (Rubber) – Hveem vs Superpave: Close performance • Mix A - Hveem vs Superpave: Superpave design~better perfor.

1,000

2,000

3,000

4,000

5,000

6,000

7,000

1,000 10,000 100,000 1,000,000 10,000,000

Flex

ural

Fat

igue

Stif

fnes

s (M

pa)

Cycles

200 Microstrain

1. 76M ESALS - 20 years (1) 2. High Desert; Inland Valley; Low Mountain (3)


12 cases x 2 (with and without reflection cracking) = 24 Cases

60 mm

125 and 175 mm Mix A Mix I


SG – CL - 100MPa

(2)

Case 1:

With reflection cracking

Structure 1 – Thin DGAC - Hveem Structure 2 – Thick DGAC - Hveem

Structure 1 – Thin DGAC - Superpave Structure 2 – Thick DGAC - Superpave

0

2

4

6

8

10

0 10 20 30 40

Crac

king

, m /

msq

Years

High DesertInland ValleyLow Mountain

0

2

4

6

8

10

0 10 20 30 40

Crac

king

, m /

msq

Years


0

2

4

6

8

10

0 10 20 30 40

Crac

king

, m /

msq

Years


0

2

4

6

8

10

0 10 20 30 40

Crac

king

, m /

msq

Years


Future research Superpave improved cracking performance and reduced the cost

WHEN CRACKING IS THE CONTROLLING DESIGN CRITERIA.

HVEEM mix design might be more cost effective WHEN RUTTING IS THE CONTROLLING DESIGN CRITERIA WARMER CLIMATES AND HEAVY LOADS


Shift factor between predicted surface rutting for Hveem and Superpave mix design methods

0

2

4

6

8

10

12

14

16

18

20

0 2 4 6 8 10 12 14 16 18 20

Surface rut - Hveem (shifted)

Surfa

ce ru

t - S

uper

pave

Surface rutting predicted for Superpave mixes is 20% larger than rutting for Hveem mixes.

(mm

)

(mm)

0

2

4

6

8

0 2 4 6 8

AC layer rut - Hveem (shifted)

AC

laye

r rut

- Su

perp

ave


Shift factor between predicted AC rutting for Hveem and Superpave mix design methods

AC layer rutting predicted for Superpave mixes is 56% larger than rutting for Hveem mixes.

(mm)

(mm

)

Rutting comparison Hveem vs. Superpave Surface rut for S410 – Mix B

0

2

4

6

8

10

12

14

16

18

0 2 4 6 8 10 12 14 16 18

Surface rut - Hveem

Surfa

ce ru

t - S

uper

pave

(mm

)

(mm)


Shift factor between predicted surface rutting for Hveem and Superpave mix design methods

Surface rutting predicted for Superpave mixes is 42% larger than rutting for Hveem mixes.

0

2

4

6

8

10

12

14

16

18

0 2 4 6 8 10 12 14 16 18

Surface rut - Hveem (shifted)

Surfa

ce ru

t - S

uper

pave

(mm

)

(mm)

Cracking comparison Hveem vs. Superpave Surface cracking for all mixes

0

2

4

6

8

10

12

14

16

18

20

0 2 4 6 8 10 12 14 16 18 20

Cracking (m/msq) - Hveem

Crac

king

(m/m

sq)-

Supe

rpav

e

3 CASES

High Desert

Structure 1 – Thin PCC - Hveem Structure 2 – Thick PCC - Hveem

Structure 1 – Thin PCC - Superpave Structure 2 – Thick PCC - Superpave

0

1

2

3

4

5

0 10 20 30 40

Crac

king

, m /

msq

Years


0

1

2

3

4

5

0 10 20 30 40

Crac

king

, m /

msq

Years


0

1

2

3

4

5

0 10 20 30 40

Crac

king

, m /

msq

Years


0

1

2

3

4

5

0 10 20 30 40

Crac

king

, m /

msq

Years


Rutting comparison Hveem vs. Superpave AC rut for Mix A

0

2

4

6

8

0 2 4 6 8

AC layer rut - Hveem

AC

laye

r rut

- Su

perp

ave

(mm

)

(mm)


Welch modified two sample t-test for AC rut predicted for Hveem and Superpave mixes

F1 and F2 are two distributions, the possible hypotheses and alternatives concerning these distributions are:

H0: F1(x) = F2(x) HA: F1(x) ≠ F2(x) Decision rule: Reject H0 if p-value < 0.10; accept H0 if

p-value ≥ 0.10

P-value = 0.05 for Hveem vs.Superpave

Predicted AC rutting for Hveem and Superpave mix design method are not equal. Superpave rutting is higher.

Without reflection cracking

0

2

4

6

8

10

0 10 20 30 40

Crac

king

, m /

msq

Years


Structure 1 – Thin DGAC - Hveem

-2

0

2

4

6

8

10

0 10 20 30 40

Crac

king

, m /

msq

Years


Structure 2 – Thick DGAC - Hveem

-2

0

2

4

6

8

10

0 10 20 30 40

Crac

king

, m /

msq

Years


Structure 1 – Thin DGAC - Superpave

-2

0

2

4

6

8

10

0 10 20 30 40

Crac

king

, m /

msq

Years


Structure 2 – Thick DGAC - Superpave

Rutting comparison Hveem vs. Superpave Surface rut for Mix I - Traffic level 1



0

1

2

3

4

5

0 10 20 30 40 50Years

Surf

ace

rutt

ing

(mm

)

SuperpaveHveem

0

2

46

8

10

12

0 10 20 30 40 50YearsSu

rfac

e ru

ttin

g (m

m)

SuperpaveHveem

0

2

46

8

10

12

0 10 20 30 40 50Years

Surf

ace

rutt

ing

(mm

)

SuperpaveHveem

0

2

46

8

10

12

0 10 20 30 40 50Years

Surf

ace

rutt

ing

(mm

)

SuperpaveHveem

0

2

46

8

10

12

0 10 20 30 40 50Years

Surf

ace

rutt

ing

(mm

)

SuperpaveHveem

0

2

46

8

10

12

0 10 20 30 40 50Years

Surf

ace

rutt

ing

(mm

)

SuperpaveHveem

AC layer does not rut. All rutting is accumulated in the unbound layers.

comparison of rutting and cracking performance estimates ... 2015/hveem_sp_trb2015.pdf ·...

Documents