One-Dimensional Swell Test

University of Texas at Arlington

Civil Engineering Department

Laboratory Test Report

One-Dimensional Swell Test of Soil from Austin, Texas

Written by:

Richard Benda

Joseph Muhirwa

Robert Sargent

July 10, 2012

Table of Contents                                                                                                          Page
Abstract...............................................................................................................................3
Introduction.........................................................................................................................4
Equipment and Materials....................................................................................................5
Procedure............................................................................................................................5
Results, Data, and Discussion.............................................................................................7
Conclusion...........................................................................................................................8
References...........................................................................................................................9
Appendices

            Appendix A: Dial Gauge Readings………………………………………………..10

            Appendix B: Sample Calculations………………………………………………...13

List of Tables and Figures                                                                               Page
Figure 1 – Sample being measured........................................................................4

Figure 2 – Sample being compacted……………………………………………….5

Figure 3 – 2, 4, and 6 percent samples…………………………………………….6

Figure 4 – Vertical Strain vs. Time……………………………………………......7

Table 1 – Vertical Swell Strain At 32 hours………………………………..……...7

Table 2 – 0% Lime Dial Gauge Readings…………………………………….…..11

Table 3 – 2% Lime Dial Gauge Readings…………………………………….…..11

Table 4 – 4% Lime Dial Gauge Readings………………………………………...12

Table 5 – 6% Lime Dial Gauge Readings…………………………………..…….12

Abstract

In this laboratory test, a one-dimensional swell test (1-D swell test) was performed on a set of disturbed soil samples from Austin, Texas in order to determine the heave potential of this soil. A control sample without any additives was compared to 3 samples with varying amounts of lime in order to observe the difference in the amount of resulting stabilization. The soil sample used in this experiment was of a tan color when dried, and it contained little organic material in the form of what appeared to be plant stems. Also, after performing Atterberg tests on the soil it was found to be of high plasticity with a plasticity index of 31 and a liquid limit of approximately 20. Another important property of the soil is that it contains a low amount of sulfate at roughly 261 ppm, which will result in very little (if any) sulfate heave. The procedure used followed ASTM standard D4546-96, Standard Test Methods for One-Dimensional Swell or Settlement Potential of Cohesive Soils. This test can be done on both disturbed or undisturbed samples, the former being used in this experiment. The first step in the process was preparing the sample to fit in the sample mold that has a diameter of 2.5 inches, a height of 1 inch, and a resulting volume of 4.91 cubic inches. Next, the amount of soil to fit in the mold had to be found using the soil’s dry density and optimum moisture content. Three additional samples with 2, 4, and 6 percent lime were prepared, each with additional moisture to compensate for the hydration of the lime. After compacting the soil slightly by hand to fit in each mold, each mold was placed under a triaxial loading machine to compact the soil to neatly fit in the mold. The next step is to place two porous stones on either side of each sample mold to allow the flow of water in and out of the sample. This apparatus is placed in a reservoir, and a ring with a dial gauge is placed adjacent to this setup. The dial indicator should be barely touching the specimen, and an initial reading should be taken. Once the initial reading is taken, the reservoir is filled with distilled water and simultaneously a timer is started. Readings are taken from the dial after varying amounts of time to determine how much the sample has swelled over time. This data can then be put on a graph and used to help determine the heave potential of the soil. From this data, the control sample can be compared to the samples with lime in order to determine the most effective reduction in expansive soil heave and swell pressure.

Introduction

The heave caused by swelling of soils is what causes damage to infrastructure. This is why it is important to find the swelling potential of the soil. Especially in an area like Austin, Texas which is known to have expansive soil. Also doing a swell test can tell what percentage of lime will reduce the amount of swelling the most. Therefore a 1-D swell test was performed on a soil sample which means the sample was only allowed to swell in the vertical direction. This test was performed on soil with no lime, a control sample, plus samples with 2, 4, and 6 percent lime. By comparing the three soils with different percentages of lime to the control soil, the percentage of lime reduced the swell the most was able to be observed. By knowing the swelling potential of the soil and the amount of lime that reduces the swell the most, economical choices can be made on how much the soil needs be stabilized and what is the minimum amount of treatment necessary to reach that goal.

Equipment and Materials

• Mass scale
• Distilled water
• Consolidometer (comprised of  a metallic reservoir, dial indicator support, dial indicator, 2 porous stones, and brass sample ring)
• 2 surcharge weights
• filter paper
• trimming knives

Procedure
The 1-D Swell test was conducted on Austin’s expansive soil in accordance with ASTM D4546-96.  The following procedure was used to conduct 1-D swell test.

1. On the first day, the soil sample to be tested was prepared. A control soil sample and three lime-treated samples (with 2%, 4% and 6% lime) were prepared. The Optimum Moisture Content (OMC) and the Maximum Dry Density of the control sample were known. These two properties were used to calculate the additional water for the mix design. An increase of 1 mL, 1.5 mL and 2 mL to the OMC was used for the mix design of lime-treated samples with 2%, 4% and 6% lime, respectively. All 4 samples were allowed to sit for a 24 hour mellowing time.
2. On the second day, the prepared soil samples were compacted using a Tritech machine (used for triaxial testing) and subsequently the height, weight and cross-sectional diameter of the four compacted specimens was measured.
3. Next, an air-dried porous stone was placed into the base of each consolidometer.  This was followed by the placement of the brass sample rings on top of the porous stones. The compacted specimens were inserted in each sample ring and topped by a porous stone and a loading plate.
4. The dial indicator supports were placed beside the consolidometer bases and the dial indicators were attached to the supports. The dial indicator supports were adjusted so that the stem of each dial indicator is centered with respect to the specimen. The dial indicators were also adjusted for the swell readings. Before the filling the reservoirs with water, each dial indicators was read to get the initial swell measurements.
5. Lastly, the four specimens were inundated by filling the reservoir with water and subsequently swell measurements were recorded at continuous and increasing times (30 sec, 1 min, 2min, 5 min, 10 min, 15 min, 20 min, 30 min, 1hr, 2hr, etc.).  The last reading was taken after 32 hours of swelling.

Results, Data, and Discussion

Figure 4 - Displacement vs. Time

Table 1 – Vertical Swell Strain At 32 hours

Soil	Vertical Swell Strain %
0 percent lime	7.36
2 percent lime	0.3
4 percent lime	0.3
6 percent lime	0.19

*See Appendix A for sample calculations

Upon observation of Figure 4 and Table 1, it is apparent that all three lime treatments (2, 4, and 6 percent) were very effective in reducing soil expansion, with even the least effective treatment reducing vertical swell to less than 1 percent in comparison to the 7.36% swell of the control sample.

Conclusion

After performing the one-dimensional swell test on the soil from Austin, Texas, it was determined that lime was a very effective additive for the purpose of reducing expansive soil heave. The 2, 4, and 6 percent lime mixtures all reduced vertical swell strain by over 7%, with the 6% lime mixture being the most effective at only 0.19% vertical swell strain. As the goal of this laboratory test was to determine the most effective amount of lime treatment on the soil, this test was successful in that respect. It is possible that some error may have arose, however. Due to the nature of the sample preparation, the 2% lime sample may have undergone some disturbance when removing it from the original sample mold. However, the top surface of the sample was undisturbed and it is unlikely that this attributed to erroneous measurements. The primary source of any error would likely be due to the disturbance of the 4% sample. In the middle of the process of taking dial gauge readings, it appears that the indicator may have been disturbed on accident. An abrupt increase in the swell measurements (by roughly .005 inches) was noted. However, due to the pattern in the 2 and 4 percent lime samples, this increase was blamed on a mistake from regular lab personnel.

References

ASTM. (n.d.). Standard Test Methods for One-Dimensional Swell or Settlement Potential of Cohesive Soils, ASTM D4546-96, West Conshohocken, PA.

Das, Braja M. (2009). Principles of Geotechnical Engineering, 25th ed., Cengage Learning, Stanford, CT.

Miller, D. J., Nelson, J.D. (1992). Expansive Soils: Problems and Practice in Foundation and Pavement Engineering, John Wiley & Sons, Inc., Toronto, Canada.

Appendix A

Dial Gauge Readings

Table 2 – 0% Lime Dial Gauge Readings

0 percent lime
displacement (in)	Time (hr)
0	0
0	0.00833
0	0.01667
0.002	0.0333
0.007	0.08333
0.012	0.16667
0.015	0.25
0.019	0.3333
0.024	0.5
0.034	1
0.047	2
0.064	4
0.076	24
0.077	32

Table 3 – 2% Lime Dial Gauge Readings

2 percent lime
displacement (in)	Time (hr)
0	0
0.001	0.00833
0.001	0.016667
0.001	0.0333
0.002	0.08333
0.002	0.16667
0.002	0.25
0.002	0.3333
0.002	0.5
0.002	1
0.002	2
0.003	4
0.003	8
0.003	24
0.003	32

Table 4 – 4% Lime Dial Gauge Readings

4 percent lime
displacement (in)	Time (hr)
0	0
0	0.00833
0	0.016667
0.001	0.0333
0.001	0.08333
0.001	0.16667
0.002	0.25
0.002	0.3333
0.002	0.5
0.002	1
0.002	2
0.003	4
0.003	8
0.003	24
0.003	32

Table 5 – 6% Lime Dial Gauge Readings

6 percent lime
displacement (in)	time (hr)
0	0
0	0.00833
0	0.016667
0	0.0333
0	0.08333
0	0.16667
0.001	0.25
0.001	0.3333
0.001	0.5
0.002	1
0.002	2
0.002	4
0.002	8
0.002	24
0.002	32

Appendix B

Sample Calculations

Vertical swell strain is determined by taking the change in sample height and dividing it by the original height of the sample. To convert it to percent, the resulting value is multiplied by 100 (Equation 1).

                                                                       (Eqn. 1)

A sample calculation for the control sample is demonstrated below using Equation 1: