October 1, 2008
A New Method for Preparing Calibration Chamber Specimens
By Hsu, Huai-Houh
Preparation of Large-Sized Silty Sand Specimens Using a Water Spray Method In-situ tests are often used to determine the engineering properties of sand. The cone penetration test (CPT) is a widely used method. Interpretations of the CPT are mostly developed based on chamber calibration tests.
Early CPT calibration tests used clean, uniformly graded silica or quartz sand in most cases. Many empirical rules and analytical models have been established based on tests using these "academic sands." However, natural sand deposits are often layered and contain various amounts of fine particles (smaller than 0.075 millimeters). Research has indicated that gradation, grain shape, soil structure and mineral content can all affect the dilatancy, shear strength under monotonie and cyclic loading conditions and interpretation of in-situ test results for sand. Accordingly, it may not be appropriate to apply the empirical rules derived from academic sands to explain test results in silty sands. Systematic laboratory studies on the CPT in silty sand have been limited. The specimens involved in CPT calibration tests can be significantly larger than those of triaxial tests. The preparation of large, uniform silty sand specimens with desired soil structure in a repeatable manner can be a formidable challenge.
The concept of a calibration chamber test is to prepare a large sand specimen in the laboratory, consolidated to a desired stress level, and then perform the experiment under given boundary conditions.
Since the entire experiment is conducted in the laboratory, the test quality can be readily controlled. The large sand specimen, with uniform deposition and known engineering properties, provides reference values for the interpretation and thus calibration of the insitu test method.
To prepare uniformly graded, clean sand chamber specimens, the sand is "rained," or pluviated, into the calibration chamber under dry conditions. The density of the sand specimen is related to the drop height and volume of sand deposited per unit of time. However, when the sand contains fine particles, these dry pluviation methods can cause particle segregation.1
A pluvial tamping (PT) method was used to create silty sand chamber specimens.- The uniformly mixed soil was placed in a 500- kilogram bag initially, which was lifted by a crane. The sand was slowly released from the bottom of the bag immediately above the surface of deposition. A 30-millimeter-diameter, 500-millimeter- long steel rod was inserted in the sand and tamped it to the desired density of each layer. This sand deposition and tamping process was repeated four times to create a specimen.
Studies have shown that the PT method is capable of creating uniform and repeatable silty sand chamber specimens. Triaxial tests on reconstituted silty sand specimens have shown that the specimen preparation method can have a significant effect on the mechanical behavior of silty sand.
Specimens prepared by water sedimentation (WS) may most accurately duplicate natural alluvial silty sand in terms of soil structures and stress/strain and strength behavior. In the WS method, the specimen is prepared in a membrane-lined split mold. The moid is filled with water initially. The silt/sand mixture is slowly poured into the mold through a funnel at one to three millimeters above the water's surface. The soil then settles in the still water and forms a sediment. Creating chamber specimens using the WS method can be time consuming, and the WS specimens are inevitably layered. Since CPT results in layered silty sand specimens tend to be erratic, they are difficult to use for the purpose of calibration tests.
The slurry consolidation method is another way to prepare a chamber specimen of silty sand. The silty sand and water are mixed together by a concrete mixer and consolidated to the desired stress level. The slurry consolidation method is also time consuming. For a chamber specimen of 40 percent fine particle content (FC), 1.5 meters in diameter and 1.5 meters high, it took 18 to 20 days to complete the consolidation process and to reach a confining pressure of 70 kilopascals.3
Water Spray Method
The new method developed by the author will be referred to as the water spray method (WSp). The WSp method intends to create specimens similar to the WS method, but with much improved efficiency, and it avoids the possibilities of layering. The sand pluviation device of the WSp method consists of a sand tank, a porous plate and a water spray shell. The sand tank is used to store the sand before pluviation. The porous plate has a total of 33 perforated holes evenly distributed in a radial pattern and is fixed to the bottom of the sand tank. The shell has eight water spray nozzles pointing towards the center of the shell, distributed into two layers. The two nozzle layers are 400 millimeters apart and have four evenly distributed nozzles in each layer.
When preparing a specimen by WSp method, the soil is pluviated from the tank in dry condition first. A mist of fine water droplets is created by the water spray nozzles immediately below the tank. When passing through the mist, the soil grains inevitably encounter water droplets and become attached. In the case of silty sand, the relatively large clusters of sand/silt and water droplets formed in the mist are much more uniform in their overall dimensions, and thus, the method minimizes particle segregation during pluviation. The amount of water introduced through the mist is limited; there is usually just a thin film of water on top of the soil deposit created by the WSp method. Therefore, there is little room for the soil particles to settle in water or form a layered structure due to grain size differences.
Evaluation of WSp Method
A batch of silty sand from Mai Liao, Taiwan, was used to make specimens for laboratory experiments. The Mai Liao sand had an FC of 15 percent, with an average grain diameter of 0.125 millimeters and a coefficient of uniformity of 2.15. The specific gravity was 2.69. The minimum and maximum dry unit weight was 12.8 and 16.6 kilonewtons per cubic meter, respectively.
For a CPT in a uniform, normally consolidated sand specimen, cone tip resistance (q^sub c^) under the boundary condition of constant vertical and horizontal stress should increase with depth initially and then reach a plateau where q^sub c^ remains a constant.4 The CPT is known for its capability of sensing thin layers of soil with different properties; thus, the stability of q^sub c^ values is a reflection of the specimen's uniformity.
In order to evaluate the methods' capabilities of producing uniform silty sand specimens, a CPT was performed in chamber specimens prepared by two different methods. The specimens were fully saturated and then consolidated by applying 98.1 kilopascals of confining stress to them before cone penetrating. The specimen of test No. 1 was prepared by the WSp method, and its relative density (Dr) was 35 percent. The test No. 2 specimen used the PT method. It had a Dr of 50 percent.
For test No. 1, the results show that a relatively stable q^sub c^ with an average value of 2.26 megapascals was reached from below a depth of 30 millimeters. The standard deviation (SD) of the q^sub c^ values for the depth range between 30 and 578 millimeters was approximately 0.182. The relatively low SD is an indication that the specimen was uniform.
The qc profile of test No. 2 reached its plateau at a depth of approximately 130 millimeters. The average qc value beyond the plateau was approximately 5.72 megapascals. The SD of the qc values within the depth range of 130 to 578 millimeters was 0.595. The pattern of qc fluctuation closely follows that of the four specimen layers associated with the PT preparation method.
A WSp method for preparing chamber specimens in silty sand has been developed.
Two chamber specimens were produced by different preparation methods, and then a CPT was performed to evaluate their quality. Results showed a rather uniform qc profile throughout the WSp specimen, much improved over the PT method. Without the sedimentation and slurry preparation/consolidation process, the WSp method is significantly more efficient, and it lacks the drawback of soil layering.
This study was funded by the National Science Council of the Republic of China under contract NSC 95-2211-E-270-010. Its support is greatly appreciated. The author would like to extend his gratitude to J.W. Chang, YJ. Huang, J.C. Hsu and RA Hong for their efforts and contributions to this endeavor.
(Above) Laboratory set up of the wet spray system. Major components include a sand tank, a porous plate and a water spray shell.
"Results showed a rather uniform qc profile throughout the WSp specimen, much improved over the PT method"
1. Lo Presti, D., S. Pedroni and V. Grippa, "Maximum Dry Density of Cohesionless Soils by Pluviation and by ASTM D 4253-83: A Comparative Study," Geotechnical Testing Journal, vol. 15, no. 2, pp. 180-189, 1992.
2. Huang, A.B., H.H. Hsu and ].W. Chang, "The Behavior of a Compressible Silty Fine Sand," Canadian Geotechnical Journal, vol. 36, no. 1, pp. 88-101, 1999.
3. Rahardjo, P.P., "Evaluation of Liquefaction Potential of Silty Sand Based on Cone Penetration Test," Virginia Polytechnic Institute and State University Ph.D. Thesis, 1989. 4. Parkin, A.K., and T. Lunne, "Boundary Effects in the Laboratory Calibration of a Cone Penetrometer for Sand," Proceedings of the second European Symposium on Penetration Testing, vol. 2, pp. 761768, 1982.
Visit our Web site at www.sea-technology.com, and click on the title of this article in the Table of Contents to be linked to the respective company's Web site.
By Huai-Houh Hsu
Department of Civil Engineering
Chienkuo Technology University
Changhua City, Taiwan
Dr. Huai-Houh Hsu is an assistant professor in the Department of Civil Engineering of Chienkuo Technology University in Taiwan. His research interests include physical modeling of geotechnical problems and calibration chamber tests.
Copyright Compass Publications, Inc. Sep 2008
(c) 2008 Sea Technology. Provided by ProQuest LLC. All rights Reserved.