PDF Printable Version
of This Issue

The photograph shows a concrete girder being loaded with a 17,900 lb (8120 kg) concrete block.

Load test of a high strength lightweight concrete girder                     

High Strength Lightweight Concrete Properties of the I-85 Ramp over State Route 34
R. Brett Holland and Lawrence F. Kahn, Georgia Institute of Technology
The high strength lightweight concrete (HSLWC) girders for the center two spans of the I-85 Ramp over Georgia State Route 34 were fabricated by Standard Concrete Products in Atlanta, GA, in August 2008. Concrete material properties were measured using cylinders cast during girder construction and cured in a moist room in accordance with ASTM C31.(1) Cylinders from all concrete batches were tested at 56 days for compressive strength and modulus of elasticity. At other ages, cylinders from the concrete batch used for the center portion of each beam were tested. Additionally, vibrating wire strain gages (VWSG) were embedded in the girders to monitor strains and thermal profiles within the girders.

Concrete Mix Proportions

Materials Quantities
(per yd3)
(per m3)
Cement, Type III 740 lb 439 kg
Fly Ash, Class F 150 lb 89 kg
Silica Fume 100 lb 59 kg
Normal Weight
Fine Aggregate
932 lb 553 kg
Lightweight Coarse
980 lb 581 kg
Water 267 lb 158 kg
Water Reducing
30 fl oz 1.16 L
High-Range Water-Reducing Admixture 59 fl oz 2.28 L
Air Entrainment 2 fl oz 77 mL
Set Accelerator 148.5 fl oz 5.74 L
Wet Unit Weight 121 lb/ft3 1938 kg/m3

Compressive Strength
The compressive strength of the cylinders was measured in accordance with AASHTO T 22.(2) Figure 1 shows the average strength gain curve for the girders along with bars showing ± one standard deviation. A statistical analysis showed that all batches used for each girder, as well as for all the girders, were statistically equivalent within a 95% confidence interval. All girders met the specified strength of 10,000 psi (69 MPa) by 56 days.

The graph shows the variation of concrete compressive strength with concrete age.
                                    Fig. 1. Measured compressive strengths.

Modulus of Elasticity
The modulus of elasticity (Ec) of HSLWC was measured at an age of 56 days in accordance with ASTM C469.(3) In addition, five girders were loaded at an age of 56 days to measure their deflection, and thus determine Ec. Deflections of the girders were also measured during casting of the deck at an age of 14 months. The Ec values from cylinder tests and from the girder load tests were not in agreement, but the girder test modulus of elasticity matched previous work done with the same concrete mix design.(4) The results of the different methods for determining the modulus, as well as the values using the AASHTO LRFD(5) and Meyer(4) equations are shown in Figure 2. The values were calculated using the 56 day measured concrete compressive strength and an air-dry unit weight of 118 lb/ft3 (1890 kg/m3 ). Meyer’s equation, Ec = 44,000 [f 'c (wc /145)]0.5 where f 'c is compressive strength in psi and wc is unit weight in lb/ft3, gave the best estimate. It was developed specifically for HSLWC made using expanded slate aggregate.

The bar graph compares the different values of modulus of elasticity.
                          Fig. 2. Comparison of measured and predicted values of modulus of elasticity.

Transfer Length
Five HSLWC girders were instrumented with mechanical strain gage points to determine the transfer length using the concrete surface strain method. The average transfer length was 27.9 in. (710 mm) at strand release and 27.6 in. (700 mm) at 28 days. This value is less than the 36 in. (915 mm) calculated using the AASHTO LRFD Specifications.(5)

Prestress Losses
VWSGs were used to determine the prestress losses in the girders by monitoring the changes in strain throughout the depth of the girders. Creep tests were performed in accordance with ASTM C512(6) to compare the creep coefficient with the estimates provided in the prestress loss calculations. The creep tests were performed with a stress equal to 40% of the cylinder strength and were loaded at the time of strand release. The AASHTO LRFD(5) and Tadros(7) methods both predicted a creep coefficient of 0.89, which was slightly larger than the measured value of 0.82.

Sum (5)
Method (7)
Method (8)
Elastic Shortening27.55 26.95 26.95 27.59 26.95
Shrinkage of ConcreteN/A N/A 5.83 5.73 4.50
Creep of ConcreteN/A N/A 17.69 18.17 22.51
Steel Relaxation0.22 N/A 0.22 0.17 1.15
Total Time-Dependent28.87 20.83 23.74 24.07 28.16
Total Losses56.42 47.78 50.69 51.66 55.11
All losses are ksi units.

The above table shows a comparison between the measured values and losses predicted by four different methods. The measured time-dependent loss from creep and shrinkage was extrapolated to 100 years for comparison with the predicted values by fitting a logarithmic curve to the data. A calculated steel relaxation loss of 0.22 ksi (1.5 MPa) was included with the measured loss so that the total measured and calculated losses could be compared on the same basis. All methods underestimated the losses. The Tadros(7) and Shams(8) methods provided the closest estimates of the prestress losses. The current AASHTO LRFD refined prestress loss calculations are based on the Tadros method, but have a few minor differences that lead to a slightly smaller value for the predicted losses. The AASHTO LRFD lump sum method underestimated losses the most.(5)

The results of the ongoing research project suggest that the modulus of elasticity for design of HSLWC with expanded slate coarse aggregate should be calculated using the Meyer equation, the transfer length can be estimated using the AASHTO provisions, and prestress losses may be estimated using the Tadros or Shams method.

Further Information
The research was sponsored by the Georgia Department of Transportation (GDOT), Research Project 2041. The opinions expressed herein are those of the authors and do not represent the opinions, conclusions, policies, standards, or specifications of the GDOT. For further information about the I-85 Ramp over State Route 34, please contact Brett Holland at

1. Standard Practice for Making and Curing Concrete Test Specimens in the Field, ASTM C31, ASTM International, West Conshohocken, PA.

2. Standard Method of Test for Compressive Strength of Cylindrical Concrete Specimens, AASHTO T 22, American Association of State Highway and Transportation Officials, Washington DC.

3. Standard Test Method for Static Modulus of Elasticity and Poisson’s Ratio of Concrete in Compression, ASTM C469, ASTM International, West Conshohocken, PA.

4. Meyer, K. F., “Transfer Length and Development of 0.6-inch Diameter Prestressing Strand in High Strength Lightweight Concrete,” Doctoral Thesis, Georgia Institute of Technology, 2002, 616 pp.

5. AASHTO LRFD Bridge Design Specifications, 4th Edition, American Association of State Highway and Transportation Officials, Washington DC, 2007.

6. Standard Test Method for Creep in Concrete in Compression, ASTM C512, ASTM International, West Conshohocken, PA.

7. Tadros, M. K., Al-Omishi, N., Seguirant, S. J., and Gallt, J. G., “Prestress Losses in Pretensioned High-Strength Concrete Bridge Girders,” NCHRP Report 496, Transportation Research Board, Washington DC, 2003, 73 pp.

8. Shams, M. K., “Time-Dependent Behavior of High-Performance Concrete,” Doctoral Thesis, Georgia Institute of Technology, 2000, 611 pp.

HPC Bridge Views, Issue 61, May/June 2010