Early Age Thermal Cracking of Mass Concrete Footings in Coastal

Environments

February 10, 2015

Dr. Christopher Bobko

Andrew Edwards Dr. Vahid Zanjani Zadeh Dr. Rudi Seracino Dr. Paul Zia

1

Outline

Objective Background Finite Element Model

Thermal Model Structural Model

Modified Schmidt Method Case Studies

Lab Experiment Oak Island and Sunset Beach Wilmington Bypass & Dimensional Study

Updated Special Provision 2

RP2012-09 - Objective

To create a finite element model to understand the temperature distribution and induced stresses created by large temperature differences in hydrating concrete

To verify the finite element model using data obtained through a mass concrete lab experiment

To use the verified finite element model to analyze current coastal bridge footings for early age thermal cracking probability

To assess the current NC DOT mass concrete specifications and make suggestions for recommended changes

3

Background: Large concrete structures have the tendency to develop large

thermal gradients due to the heat generated during the early age hydration process

High tensile stresses develop on the faces of the concrete due to excessive temperature differences between the inside and outside of the concrete

Cracking then occurs when these stresses exceed the concrete tensile strength which then creates durability issues, especially in coastal regions

Issues on multiple mass footings in the North Carolina coastal region have been identified with cracking problems that need to be assessed

4

Background: Common Practices

• Goal of current practices: to control the temperatures within the concrete to in turn reduce the maximum thermal gradients – Typical maximum temperature: 158-160°F – Typical maximum temperature difference: 35°F

• Accomplished through a combination of: – Mix design – Temperature Monitoring – Pre-cooling systems (lower the placement temperature) – Surface Insulation – Post cooling systems

5

Background: Common Practices

6

Placement

Size (ft)

Max Temp Difference

(°F)*

Max Temp of Concrete (°F)

Concrete Placement Temp

Range (°F)

Oak Island, NC ≥ 6 ≤ 35 ---- 40 - 75

Sunset Beach, NC ≥ 6 ≤ 35 ---- 40 - 75

US17 Bypass, NC ≥ 1.8 (m) or

6 ft. ≤ 20 (°C) or 36 °F

---- 4 - 24 (°C)

or 52 - 75 °F

Florida ---- ≤ 35 180 ----

Texas ≥ 5 ≤ 35 160 50 - 75

South Carolina ≥ 5 ≤ 35 ---- ≤ 80

Louisiana ≥ 4 ---- ---- ----

Maryland ≥ 6 ---- ---- ----

Virginia > 5 ≤ 35 160 ≤ 95

UFGS > 3 or 1

m ≤ 20 (°C) or 35 °F

158 Based on

thermal model

Background: Common Practices Concrete

Type Fly Ash

Content (%)* NOTES

Oak Island, NC Class AA 30 5% microsilica in addition to fly ash

Sunset Beach, NC Class AA 30 5% microsilica in addition to fly ash

US17 Bypass, NC Class AA 25 - 30 -------------

Florida Type II (MH)

18 - 50 -------------

Texas ---- ---- Keep form in place for at least 4 days.

South Carolina ---- ---- -------------

Louisiana Type II Type IP Type IS

---- Concrete mix shall generate a heat of hydration no more than 70 calories/gram (290 kJ/kg) at 7 days.

Maryland ---- ---- -------------

Virginia ---- 25 - 40 Or use 50 - 75 % of slag as cement replacement

UFGS ---- ---- Class C fly ash not permitted 7

Finite Element Model

• TNO DIANA was used as the finite element program for this project

• Uses a flow-stress analysis where DIANA combines two analyses – Thermal Flow Analysis: models the temperature evolution

in the concrete due to early age hydration – Nonlinear Stress Analysis: models the stresses over time

induced by the thermal gradients within the concrete

8

Finite Element Model: Thermal Model • Hydration of concrete modeled using an empirical model

developed by Schindler and Folliard and updated by Riding, et al. (2012):

𝑄ℎ 𝑡 = 𝐻𝑢 · 𝐶𝑐 · τ𝑡𝑒

β· β

𝑡𝑒· α𝑢 · 𝑒−

τ𝑡𝑒

β

· 𝑒𝐸𝑎𝑅

1𝑇𝑟

+ 1𝑇𝑐 – Model based on the chemical make up of the

cementitious material – Also takes into consideration the physical make-up of the

cementitious paste • Thermal properties for concrete, formwork, and

insulation explicitly modeled • External weather conditions were also modeled based

on historical local weather data

9

Finite Element Model: Structural Model • Mechanical Properties modeled using the equivalent age

method:

𝑡𝑒 = ∑ 𝑒−𝐸𝑎𝑅

1𝑇𝑐

− 1𝑇𝑟𝑡0 · Δ𝑡

which was input directly into DIANA from the lab test results • Thermal strains combined with shrinkage strains and

viscoelastic effects of the concrete to calculate the induced stresses within the concrete

• Cracking index was then calculated to find the cracking probability throughout the concrete model:

𝐼𝑐𝑐 𝑡 = 𝑓𝑡(𝑡)σ1(𝑡)

where cracking is likely to occur when: 𝐼𝑐𝑐 ˂ 1

10

Schmidt Method – ACI 207.02-R07

• Pros: – Simple numerical solution for heat transfer problem – Can be easily implemented in a spreadsheet – Good for structures with simple geometry – 1D idealization

• Cons: – Only gives temperatures (no stresses) – No consideration for insulation (only one material) – ACI provides a graph to get heat of hydration over time, but

very outdated, or suggests calorimetry.

• Suggested modifications: – Effective depth of non-hydrating concrete to model insulation – Incorporate Riding, et al. empirical model for heat of hydration

as a function of temperature

11

• A 5x5x5 foot concrete block was cast

• Was used to validate DIANA’s thermal models

• Typical mass concrete mix design used based on past NC DOT jobs

• Companion cylinders were used to gather mechanical properties for the structural model’s input

12

Lab Experiment

13

0

20

40

60

80

100

120

140

160

0 24 48 72 96 120 144 168 192

Tem

pera

ture

(F)

Time (hrs)

Sample Temperature Distribution: Vertical Slice Center (1)Center-Top (2)Between Center-Top (10)Center-Bottom (4)Ambient

• Temperatures were recorded over an 8 day period using thermocouple wire and automated data recorders

Lab Experiment: Thermal Results

152.6

21.5

-30

-20

-10

0

10

20

30

40

50

60

0 24 48 72 96 120 144 168 192

Tem

pera

ture

(F)

Time (hrs)

Temperature Differentials

Center-TopCenter-BottomCenter-Side

Typical Maximum Limit

14

Lab Experiment: Thermal Results

46.8

42.5

43.2

67.0

• Mechanical properties were measured at 1, 3, 7, 14, and 28 days according to ASTM standards

15

0

1000

2000

3000

4000

5000

6000

7000

8000

0 5 10 15 20 25 30Aver

age

Com

pres

sive

Str

engt

h (P

si)

Equivalent Age (days)

Compressive Strength

0

100

200

300

400

500

600

0 5 10 15 20 25 30

Aver

age

Tens

ile S

tren

gth

(Psi

) Equivalent Age (days)

Tensile Strength

Lab Experiment: Mechanical Properties

16

0500

1000150020002500300035004000

0 5 10 15 20 25 30

Aver

agee

Mod

ulus

(ksi

)

Equivalent Age (days)

Elastic Modulus

-1.40E-04

-1.20E-04

-1.00E-04

-8.00E-05

-6.00E-05

-4.00E-05

-2.00E-05

0.00E+000 20 40 60

Aver

age

Stra

in (i

n/in

)

Time (days)

Average Strain vs Time

Lab Experiment: Mechanical Properties

Autogenous Shrinkage

17

0

20

40

60

80

100

120

140

160

180

0 24 48 72 96 120 144 168

Tem

pera

ture

(F)

Time (Hours)

Sample Temperature Distribution: Vertical Slice CenterTopBottomNode CenterNode TopNode Bottom

Lab Experiment: Thermal Results

18

-20

-10

0

10

20

30

40

50

0 24 48 72 96 120 144 168

Tem

pera

ture

(F)

Time (Hours)

Temperature Differentials

Center-TopCenter-BottomCenter-SideMeasured Center-TopMeasured Center-BottomMeasured Center-Side

Lab Experiment: Thermal Results

46.8

42

19

Cracking Index at time of Max Temperature Difference

1.0

0.5

0.0

Lab Experiment: Structural Results

20

Lab Experiment: Modified Schmidt Method

0

20

40

60

80

100

120

140

160

180

0 24 48 72 96 120 144

Tem

pera

ture

(F)

Age (hr)

FE - Center FE - TopSchmidt - New - Center FE - BottomSchmidt - ACI - Center Schmidt - New - 1 node deep

Schmidt - ACI - 1 node deep

21

Dimensions: 60’-6” x 20’-0” x 6’-0” Oak Island: Overview

22

Dimensions: 42’-0” x 27’-0” x 6’-0” Sunset Beach: Overview

Primary Crack

0

20

40

60

80

100

120

140

160

180

0 24 48 72 96 120 144 168

Tem

pera

ture

(F)

Time (Hours)

Sample Temperature Distribution

Thermo 1Thermo 2Thermo 3Node CenterNode TopNode BottomNode Side

23

Oak Island: Thermal Results

24

Oak Island: Thermal Results

-10.00

0.00

10.00

20.00

30.00

40.00

50.00

0 24 48 72 96 120 144 168

Tem

pera

ture

(F)

Time (Hours)

Temperature Differentials

Center-TopCenter-BottomCenter-SideMax DifferenceMeasured 1-2Measured 3-2

25

Cracking Index at time of Max Temperature Difference

1.0

0.5

0.0

Oak Island: Structural Results

26

Oak Island: Modified Schmidt Method

0

20

40

60

80

100

120

140

160

0 24 48 72 96 120 144

Tem

pera

tute

(F)

Age (hr)

FE Center

Schmidt - New - Center

Schmidt - ACI - Center

FE Top

Schmidt - New - 1 node deep

Schmidt - ACI - 1 node deep

27

Dimensions: 28.4m x 18.6m x 4.5m [93.2’ x 61’ x 14’-9”]

Wilmington Bypass: Overview

0

20

40

60

80

100

120

140

160

180

0 24 48 72 96 120 144 168

Tem

pera

ture

(F)

Time (Hours)

Sample Temperature Distribution

Node CenterNode TopNode BottomAmbientTyp Max Temperature

Formwork/Insulation Removal

28

Wilmington Bypass: Thermal Results

70

-10.00

0.00

10.00

20.00

30.00

40.00

50.00

60.00

70.00

80.00

90.00

100.00

0 24 48 72 96 120 144 168

Tem

pera

ture

(F)

Time (Hours)

Temperature Differential

Center-TopCenter-BottomCenter-SideMax Difference

29

Wilmington Bypass: Thermal Results

156

30

Cracking Index at time of Max Temperature Difference

1.0

0.5

0.0

Wilmington Bypass: Structural Results

vertical

0

20

40

60

80

100

120

140

160

180

0 24 48 72 96 120 144 168

Tem

pera

ture

(F)

Time (hrs)

Sample Temperature Distribution

Node CenterNode TopNode BottomAmbientTyp Max Temperature

31

Wilmington with Insulation: Thermal Results

-10.00

0.00

10.00

20.00

30.00

40.00

50.00

60.00

0 24 48 72 96 120 144 168

Tem

pera

ture

(F)

Time (hrs)

Temperature Differential

Center-TopCenter-BottomCenter-SideMax Difference

32

Wilmington with Insulation: Thermal Results

160

52.0

33

Cracking Index at 156 hours

1.0

0.5

0.0

Wilmington with Insulation: Structural Results

vertical

34

Wilmington Insulated: Mod. Schmidt Method

0

20

40

60

80

100

120

140

160

180

200

0 24 48 72 96 120 144 168

Tem

pera

tute

(F)

Age (hr)

FE - Center

Schmidt - New - Center

Schmidt - ACI - Center

FE - Top

Schmidt - New - 1 node deep

Schmidt - ACI - 1 node deep

Dimensional Study – Wilmington Insulated

14.75 ft tall (As built)

X-axis

Z-axis

X-axis

Z-axis

X-axis

Z-axis

X-axis

Z-axis

X-axis

Z-axis

12.5 ft tall 10 ft tall

7.5 ft tall 5 ft tall

36

Dimensional Study – Wilmington Insulated

0

5.55

11.1

16.65

22.2

27.75

33.3

0

10

20

30

40

50

60

0 2 4 6 8 10 12 14 16

Max

Tem

pera

ture

Diff

eren

ce (˚

C)

Max

Tem

pera

ture

Diff

eren

ce (˚

F)

Minimum Dimension (ft)

Typical Maximum Limit

1

1

2

2

3

3

4

4

as built

center - side

center - top

Results/Conclusions The finite element model created in DIANA accurately

predicts the heat production and subsequent temperature changes during early age concrete hydration.

Reasonably sized mass concrete that follows the typical control plans do not have a high likelihood of cracking.

More care should be taken with the more “massive” mass concrete footings as they have a much higher likelihood of cracking.

When the temperature control plans are not followed there is a high likelihood for thermal cracks to occur

Modified Schmidt Method does a good job of predicting the temperatures during early-age hydration of mass concrete

37

Excerpt: NCDOT New Special Provision - 1 This special provision applies to substructure

components (footings, columns or caps) when the smallest dimension of that component is between six feet and eight feet.

The mass concrete temperature after placement shall

not exceed 158°F and the temperature difference between the core and exterior surfaces shall not exceed 35°F. Mass concrete should remain covered and monitored until the difference between the core temperature and the average daily ambient temperature is below 35°F. All mass concrete pours shall remain covered and protected a minimum of 7 days unless otherwise directed by the Engineer.

38

Excerpt: NCDOT New Special Provision - 2 Submit an analysis, for review and approval, of the

anticipated thermal developments in the mass concrete based on the proposed mix design, materials and casting procedures. At a minimum the analysis shall provide: an anticipated range of peak temperatures, temperature gradients, time to peak temperature and recommended cure time. The submittal shall also describe the measures and procedures that will be taken to limit the temperature differential to 35°F or less between the core and exterior surfaces.

39

Excerpt: NCDOT New Special Provision - 3 Methods for reducing thermal differential may involve

but are not limited to a combination of the following: A. Selecting materials that minimize the heat generated by hydration of the cement. B. Cooling materials to reduce the temperature of the concrete in its plastic state. C. Controlling the rate of concrete placement. D. Insulating the concrete surface to prevent heat loss. E. Providing supplemental heat at the concrete surface to prevent heat loss. F. Other acceptable methods which may be developed by the Contractor.

40

Excerpt: NCDOT New Special Provision - 4 Mass concrete shall contain an approved set-retarding,

water-reducing admixture, and flyash or ground granulated blast furnace slag in the amount of 25% by weight of the total cementitious material (portland cement plus fly ash) …

The Contractor shall provide and install a minimum of six temperature sensing devices in each mass concrete pour to monitor temperature differentials between the core and exterior surfaces …. One temperature sensing probe shall be placed near the core of the pour, and the remaining temperature sensing probes shall be placed at approximately two inches clear from the surface of the concrete furthest from the core …

41

Excerpt: NCDOT New Special Provision - 5 … record readings from temperature sensing devices at

two-hour intervals until consecutive readings indicated the temperature difference between the core and all exterior surfaces is less than 35°F.

If monitoring indicates the 35°F differential has been exceeded, the Contractor shall take immediate action to reduce the temperature differential to less than 35°F and revise the thermal plan to ensure future mass concrete pours meet the temperature limits. All revisions to the approved plan must be approved by the Engineer prior to implementation.

42

Excerpt: NCDOT New Special Provision - 6 At the discretion of the Engineer, all temperature

monitoring requirements may be waived provided the Contractor has proven to the satisfaction of the Engineer that the temperature after placement will not exceed 158°F and the temperature difference between the core and all exterior surfaces will not exceed 35°F.

Placement of mass concrete shall be continuous resulting in a footing, column or cap that is monolithic and homogeneous.

The entire cost of this work shall be included in the unit contract price bid for the class of concrete associated with the mass concrete.

43

Acknowledgements The research was sponsored by NCDOT under RP2012-

09, “Crack Free Mass Concrete Footings on Bridges in Coastal Environments”

In particular from Brian Hanks, Greg Perfetti, Kevin Bowen, and Owen Cordle, NCDOT

44

References Bobko, C., Edwards, A.J., Seracino, R., and Zia, P. (2014)

“Thermal cracking of mass concrete bridge footings in coastal environments.” ASCE Journal of Performance of Constructed Facilities, in press.

Bobko, C., Zanjani Zadeh, V., and Seracino, R. (2014) “Improved Schmidt Method for predicted temperature development in mass concrete.” ACI Materials Journal, in press.

Edwards, A.J., Bobko, C., and Seracino, R. (2015) “Modeling early age thermal cracking of mass concrete bridge footings.” Submitted to Construction and Building Materials.

45