Shrinkage Testing: Available Tests · 2018. 7. 28. · NCC – Shrinkage Tests Slides Prepared by...

NCC – Shrinkage Tests Slides Prepared by Jason Weiss Slide 1 of 27

Shrinkage Testing: Available Tests

Jason Weiss, Edwards Distinguished Professor of Engineering

April 26th, 2017


A Bit of Background

• I was asked to discuss: • discuss available shrinkage tests • pro’s and con’s of shrinkage tests

• I am assuming this is driven in part by AASHTO PEM spec

• I will be glad to share some thoughts on this but I am so sure of what pros and cons we are after

Concrete Shrinkage: Is This the Key?


PEM Guide Developed

• Section 6.4 deals with reducing volume change (Options)

• Do Nothing

• Limit Paste

• Measure Volume Change

• Restrained Shrinkage Cracking

• Probabilistic Cracking

• Not Covered Plastic Cracking (Evaportation, Settlement, Autogenous)

• I will also state that I disagree that this is the most important part of warping or curling and there are ways to deal with that directly

Property Specified TestMixture

QualificationAcceptance Special Notes

6.3.1 Flexural Strength AASHTO T 97 4.1 MPa 600 psi Yes Yes

6.3.2 Compressive Strength AASHTO T 22 24 MPa 3500 psi Yes Yes

6.4.1.1 Volume of Paste 25% Yes No

6.4.1.2 Unrestrained Volume Change ASTM C157 420 me at 28 day Yes No Curing Conditions

6.4.2.1 Unrestrained Volume Change ASTM C157 360, 420, 480 me at 91 days Yes No

6.4.2.2 Restrained Shrinkage AASHTO T 334 crack free at 180 days Yes No

6.4.2.3 Restrained Shrinkage AASHTO TP XXX s < 60% f'r at 7 days Yes No Dual ring test is currently under consideration as an AASHTO Provisional Test Method

6.4.2.4 Probability of Cracking Appendix X1 5, 20, 50% as specified Yes No

Commentary Quality control check ~ ~ ~ No Yes Variation controlled with mixture proportion observation or F Factor and Porosity Measures

6.5.1.1 Water to Cementitious Ratio ~ 0.45 ~ Yes Yes Choose Either 6.5.1.1 or 6.5.2.1

6.5.1.2 Fresh Air Content AASHTO T 152, T196, TP 118 5 to 8 % Yes Yes

6.5.1.3 Fresh Air Content/SAM AASHTO T 152, T196, TP 118 ≥ 4% Air; SAM ≤ 0.2 %, psi Yes Yes

6.5.2.1 Time of Critical Saturation "Bucket Test" Specification 30 Years Yes No Note 1 Note 2 Variation controlled with mixture proportion observation or F Factor and Porosity Measures

6.5.3.1 Deicing Salt Damage ~ 35% SCM Yes Yes Are calcium or magnesium chloride used

6.5.3.2 Deicing Salt Damage AASHTO M 224 ~Topical

TreatmentYes Yes Are calcium or magnesium chloride used, use specified sealers

6.5.4.1 Calcium Oxychloride Limit Test sent to AASHTO Yes No Are calcium or magnesium chloride used

6.6 Transport Properties

6.6.1.1 Water to Cementitious Ratio ~ ≤ 0.45 or ≤ 0.50 ~ Yes Yes The required maximum water to cementitious ratio is selected based on freeze-thaw conditions.

6.6.1.2 Formation Factor Table 1 ≥ 500 or ≥ 1000 ~ Yes Yes Based on freeze-thaw conditions. Other criteria could be selected

6.6.2.1 Ionic Penetration, F Factor Appendix X2 Yes, F through r Determined using guidance provided in Appendix X2.

6.7.1 D Cracking AASHTO T 161, ASTM C 1646 ~ ~ Yes No

6.7.2 Alkali Aggregate Reactivity AASHTO PP 65 ~ ~ Yes No

6.8.1 Box Test Appendix X3 No

6.8.2 Modified V-Kelly Test Appendix X4 No

Note 1: Choose Either 6.5.1.1 or 6.5.2.1

Note 2: Choose either 6.5.1.2, 6.5.1.3, or 6.5.2.1

< 0.15g CaOXY/g paste

25 mm at 30 year


Option 0 – Do nothing


Option 1 – Limit Paste Volume

• L’Hermite, Robert G. "Volume changes of concrete." 4th Int. Symp. on the Chemistry of Cement. 1960.

• Work in the 1910s in Germany suggested limiting paste

• Dutron (1956) developed an empirical expression based on observations of shrinkage

• L’Hermite reviewed the role of paste

• Pickett and Hobbs applied substantial math to this problem


• The models by Pickett, Hobbs, and others provide a substantial framework to earlier observations

• Picketts model shown below is the easiest to use and understand

• Hobbs works well with very low stiffness aggregate and air

Option 1 – Limit Paste Volume

n

fAggPasteConcrete

fPastePastefAggAggConcrete

V

VV

)1(

ee

eee

Typical

Concrete

n 1.2~1.7

0 20 40 60 80 100

Aggregate Volume (%)

0

20

40

60

80

100

Rela

tive S

hri

nkag

e (

%)


Option 2 – Set a Volume Change Limit

• Can we measure volume change – of course

• ASTM C 157 and C 341 are two test methods

• Some comments on ASTM C 157 • Where does the test start

• The adjective game – many call this a drying test however it is total shrinkage

• Misses autogenous shrinkage before the test starts

• Rate depends on sample geometry

• Depends on boundary conditions


Comparing Volume Change Due to Temperature and Due to Moisture Loss

• Temperature provides a coefficient that describes a range

• ASTM C157 provides a single value for a given environmental condition

-10

00

-80

0

-60

0

-40

0

-20

0 0

20

0

40

0

60

0

80

0

10

00

Longitudinal Strain (me)

-20

0

20

40

60

80

100

Te

mp

era

ture

Ch

an

ge

(C

)

1

a

TTEMP ae 31 RHNSH e


A Lesson from Erich Weiss

http://en.wikipedia.org/wiki/Image:Hhoudini.jpg


Option 3 – Restrained Shrinkage Initial Specimen

Shrinkage Effect

Restraint Effect

Initial Specimen

Shrinkage Effect

Restraint Effect

Initial Pavement, Bridge Deck

Disconnect it From

The Subgrade or Structure

Apply Force to

‘Simulate’ Subgrade

Maintain “Zero”

Displacement

(In real-life actually

not complete but…) Weiss et al. 1998, JEM


Option 3 – Restrained Shrinkage

Creep/Cracking Effect

Stress Relaxation

28

,,

E

tdd

E

dtd SHR

se

se

s

Stress

Relaxation

Weiss et al. 1998, JEM

e

se

s

SHRdE

dtd ,

Initial Specimen

Shrinkage Effect

Restraint Effect

0 7 14 21 28

Age of Specimen (Days)

0

4

8

12

Stress Based

On Hooke’s Law

Stress In

Specimen

Ca

lcu

late

d T

en

sile

Str

es

s (

MP

a )

0 7 14 21 28

Age of Specimen (Days)

0

4

8

12

Stress Based

On Hooke’s Law

Ca

lcu

late

d T

en

sile

Str

es

s (

MP

a )

Final Stress State

Stress In

Specimen


A Bit of History of the Ring • Carlson and co-workers (1930-50 pub. in 1988)

• Douglas, McHenry, Brewer, Burrows (1930-50) – paste

• Malhotra et al. (1950’s) used an active ring to assess strength Swamy and Stavarides (1976) apprx. eqn for stress

• Krenchel & Shah (1985) Grzybowski & Shah (1990) FRC

• Kovler (1994) - active ring to measure cracking

• Weiss (1997) - active ring to measure creep

• Weiss (1999) – discussed geometry and B.C.

• Furgeson and Weiss (2001) – Geometrically correct eqn

• Weiss (2002) – dual ring test for expansion

• Moon and Weiss (2005) – Moisture Gradients


• Using an Instrumented

Ring

• Measure Strain that

Develops in Steel

• Determine the Pressure

Required to Obtain that

Strain

• Apply Pressure to

Concrete and Obtain

Tensile Stress

Original Ring

0 10 20 30

Time (Days)

-200

0

Ste

el

Str

ain

(m

e)

-100

Measured Strain

Pres

Determine Pressure

Pres

Obtain Stress

Ctt

RR

RR

R

RREtt

SteelRrConcrete

ICOC

ICOC

OS

ISOSSSteelRrConcrete

IC

IC

es

es

22

22

2

22

2

Furgeson and Weiss 2000



• Degree of Restraint from a Test

• Linear Test

• Ring Test

• A priori Ring


)1()1(

)(

)(

2

11

2

2

SS

OS

IS

SH

st

R

R

t

t

e

e

)1()1(

)1()1(

1

1

11

2

2

2

2

'

'

S

OS

ISS

C

OS

OCC

OS

OC

OS

IS

S

C

S

C

R

R

R

R

R

R

R

R

E

E

E

E

Moon et al. CCC 2006

If the restraint is too great, no strain is measured in the

restraining material

Shrinkage Free

Permitted1

1

1

e

e

RR

CC

EA

EA



Dimensions

[mm]

A 12.5 0.13

B 330 3

C 406 3

D 150 6

AASHTO PP 34-99 ASTM C 1581-04

Dimensions [mm]

A 12.7 0.4

B 305 5

C 457 5

D 152 5 B = 2RIC C = 2ROC

• AASHTO takes a long time to crack 3 to 6 months, DOR is more like what one

may expect in a bridge (60%) but circumferential drying

• ASTM cracks more rapidly but DOR is higher and the wall is thin resulting in

only tests for small aggregate MSA Radlinska and Weiss 2008


Option 3 – Drying Direction Matters

Circumferential

Drying

Top and Bottom

Drying

1/r2

Weiss et al. 2001

0 0.02 0.04 0.06 0.08

Distance from the outer surface (m)

-1

0

1

2

3

4

Str

es

s (

MP

a)

=0.002

InnerSurfaceof the Conc. Ring

Econ / Esteel =0.105

Steel

Ring

0 0.02 0.04 0.06 0.08


-1

0

1

2

3

4

Str

es

s (

MP

a)

=0.004



Steel

Ring

0 0.02 0.04 0.06 0.08


-1

0

1

2

3

4

Str

es

s (

MP

a)

=0.008



Steel

Ring

0 0.02 0.04 0.06 0.08


-1

0

1

2

3

4

Str

es

s (

MP

a)

=0.02



Steel

Ring

0 0.02 0.04 0.06 0.08


-1

0

1

2

3

4

Str

es

s (

MP

a)

=0.04



Steel

Ring

0 0.02 0.04 0.06 0.08


-1

0

1

2

3

4

Str

es

s (

MP

a)

=0.08



Steel

Ring

0 0.02 0.04 0.06 0.08


0

1

2

3

4

Str

es

s (

MP

a)

=0.2



Steel

Ring

0 0.02 0.04 0.06 0.08


0

1

2

3

4

Str

es

s (

MP

a)

=0.5



Steel

Ring

0 0.02 0.04 0.06 0.08


0

1

2

3

4

Str

es

s (

MP

a)

=100



Steel

Ring

0 0.02 0.04 0.06 0.08


-1

0

1

2

3

4

Str

es

s (

MP

a)

=0.002

=0.004

=0.008

=0.02

=0.04

=0.08

=0.2

=0.5

=100



Steel

Ring

Moon et al. Consec 2003


Can We Do More than Look for Age of Cracking

Strength

0 14 28

Time (Days)

0

2

4

6

Str

ess (

MP

a)

Residual

Stress

Visible

Cracking

9 mm Steel Wall Thickness with a 75 mm

Concrete Wall Thickness (W/C = 0.50)

Hossain and Weiss, CCC, 2004

• Converts test results from limited

(it cracked or it did not crack) to

informative

• Analysis of the test method

enables an economic ‘QC/QA’

• Test is simple to perform and

complicated analysis can be

automated


Dual Ring Test – Addressing items not addressed in the standard ring test

• Expansive mixtures

• Temperature changes

• Unsatisfying to wait 6 months and see nothing (3 to 7 day answer)

• Error in thought ‘core of concrete’

0 12 24 36 48 60 72 84 96

Time (hr)

-2

0

2

4

6

Te

ns

ile

Str

es

s (

MP

a)

0

10

20

30

Te

mp

era

ture

(o

C)

STRESS

TEMPERATURE

s RIC eIN EINVARRIC

2RII

2

2 RIC2

ROC

2RIC

2

ROC2

RIC2

eOUT EINVARROI

2ROC

2

2 ROC2

2 ROC

2

ROC2

RIC2


A Comment on Restraint

• There has been a lot of speculation as to what the core material should be for a ring test and that ‘concrete or asphalt’ is better than a material like steel or invar

• Personally I do not understand this as the essential factors of the core is that it is volumetrically stable, linearly elastic and non absorptive

• Recall the discussion on the degree of restraint it is not just influenced by the modulus of the material but also the geometry


Option 4 – Simulation of Cracking Potential


Option 4 – Simulation of Cracking Potential

• Plotted the percentage of specimens cracked by a specific age

• Results of 10,000 simulations

• Can quantify risk or total probability

• Age of cracking is not the best measure of a mixture

0 14 28 42 56 70

Age of the Specimen (Days)

0

20

40

60

80

100

Spe

cim

ens

Cra

cked

(%)

Age of Cracking

Deterministic

Age of Cracking

5% Probability

PCRACK

Spec

ime

ns

Cra

cke

d (

%)


Cracking not Being Discussed

• Plastic Cracking (3 Main Components)

• Evaporation

• Differential Settlement

• Autogenous Shrinkage

• Kayir and Weiss 2002

• Qi and Weiss 2002

• Kwak et al 2008


Ease Easy Easy Moderate Moderate Hard

Initial Costs 0 Nearly 0 Low Moderate/

High

Moderate/

High

Time to Perform 0 Nearly 0 28 to 180 d 3 to 180 d Some time

Including Innovative Materials No No Yes, BC

Issue Yes Yes

Output None Paste Vol. me value Crack Age of

% Strength

Prob. of

Cracking

Do

Nothing

Limit

Paste

Linear

Strain

Rest.

Ring

Prob.

Model

An Attempt at a Pros and Cons Table


Conclusions

• Limit Paste Volume – Good First Step

• Volume Change Limit – OK but flawed initial steps, time consuming, only evaluates shrinkage, constant env RH/Temp (issue for innov)

• Restrained Ring – Discussed drying direction, wall thickness, degree of restraint and temperature sensitivity

• Dual Ring – Can be a faster test that can be taken to the field with less temperature sensitivity, enables more innovative mixtures, fallacy exists with core matching structural materials

• Probabilistic Model – Direction we could be heading with promise but requires more inputs


• We have lots of people sharing opinions about tests

• As an industry we spend a lot of time thinking about tests …… but not as much time thinking about how tests can be used in specifications

• I think we want tests that are simple to perform, and repeatable rather than indicies

• I think we want an ‘easy interpretation’ which can occur with simple apps and we should not be scared by an equation with three variables as standardization makes these ‘constants’

Just my opinion …..


Thank you Are There Any Questions

Jason Weiss

Edwards Distinguished Professor

[email protected]

https://www.researchgate.net/project/Restrained-Shrinkage-Testing

Shrinkage Testing: Available Tests · 2018. 7. 28. · NCC – Shrinkage Tests Slides Prepared by...

Documents

Transcript of Shrinkage Testing: Available Tests · 2018. 7. 28. · NCC – Shrinkage Tests Slides Prepared by...