No. 1 October 22, 1920 APPENDIX A, PRELIMINARY REPORTS Information of greatest immediate value would be obtained from tests similar to those now being made by the Bureau of Public Roads. 1. Slabs of concrete 7 ft. by 7 ft. say 6 in. thick. 2. Placed directly on subgrade of adobe soil in various conditions of moisture content and adulteration. 3. Similar slabs on good gravel or crushed rock subgrade. 4. Test under static load, determine deflection of slab and fiber stresses at various positions in the slab during the progress of the test. Supplementing these tests the bearing power of adobe soil and other characteristic California subgrade materials could be determined as follows: 1. Pulverize sample, place in container about 14 inches in diameter and 8 inches deep. Let the soil absorb water from below, then dry the sample in air. Repeat this process until a constant volume is secured. (This method is far more satisfactory than mechanical tamping or other manipulation to secure uniform compactness of sample.) 2. Fill space between test sample and container with Plaster of Paris and place in testing machine. 3. Apply test load through flat-bottomed circular plunger (10 square inches in crosssectional area) to center of upper surface of sample. 4. Get data for load-deflection curve. I think that it would be impossible to prepare samples of adobe soil in this manner. The most practical and satisfactory method could only be determined by actual experiment but the test itself could be made as outlined. The adobe soil could be adulterated in any manner and the effect of such treatment on its bearing power readily determined. This is essentially the method used by the Bureau of Public Roads. I got the information verbally from a remote source, and it may not be correct, but there is no doubt that this method would be only our starting point if we attempted such tests. Adobe soil is in a class by itself and demands individual methods. If possible the grain size should be determined, see A. S. C. E. proceedings August, 1920, page 913. Change in volume under various moisture conditions should also be noted. (Signed) C. T. WISKOCIL. No. 2 TESTS ON ADOBE SOIL 1. Contraction and Expansion October 25, 1920. 1a1 Prepare samples, made of adobe soil paste, as large as can be conveniently manipulated. 1a2 Dry sample at room temperature for seven days. 1a4 Compute percentages of contraction (percentage reduction in volume on 1a6 When sample has absorbed all the water it will, determine its volume. 1a7 Compute expansion as percentage increase in volume on basis of volume in 1a3. Adobe soil mixed with foreign materials, such as sand, gravel, rock, or lime. (Same procedure as in 1a.) 2. Bearing Power Adobe soil in natural condition (as found). 2a1 Pulverize sample, place in container about 14 inches in diameter and 8 inches deep. Let the soil absorb water from below, then dry the sample in air. Repeat this process until a constant volume is secured. This method is far more satisfactory than mechanical tamping or other manipulation to secure uniform compactness of sample. 2a2 Fill space between test sample and container with Plaster of Paris and place in testing machine. 2a3 Apply test load through flat-bottomed circular plunger (10 square inches in cross sectional area) to center of upper surface of sample. 2a4 Get data for load-deformation curve. 3. Strength of Concrete Slabs on Adobe Soil Subgrade Grain Size Prepare slabs of cement concrete 7 feet by 7 feet by 6 inches. Place cured slabs on subgrade of adobe soil. 3a21 Natural adobe soil-vary moisture. 3a22 Adobe soil mixed with foreign materials vary moisture as in 3a21. Test slabs under static load, determine deflection and fiber stresses at various points in the slab during progress of test. 4a See A. S. C. E. Proceedings, August, 1920, page 913. No. 3 Dear Professor Derleth: (Signed) C. T. WISKOCIL. December 8, 1920. At our conference during which the letter of December 1 was written to Messrs. Lippincott and Brunnier you instructed me to prepare a statement of what we have been doing in the laboratory on adobe soil. The first revised program of October 25, also contained in your letter of October 29 to Messrs. Lippincott and Brunnier, has been adhered to insofar as possible. The results however while in a sense negative will be of great value in showing us what not to do. The laboratory work I am about to describe was done during the month of November. We attempted to follow the procedure recommended by Professor Shaw to the engineers of the Bureau of Public Roads working under Dr. Hewes. The actual work for this Bureau was done by Miss Edith Phillips, working in Professor Shaw's laboratory in Hilgard Hall. The work they were particularly interested in was the amount of shrinkage and expansion in adobe soil. This work was carried on as follows: The sample of soil was crushed by hand and then pulverized. A small quantity of this pulverized soil was mixed with water until it attained the consistency of rich cream. The mud in this condition was poured into small cups 2 inches in diameter and 12-inch deep, weighed and set aside to dry. After air-drying to a constant weight, the volume of the specimen was determined by micrometer measurements. In most cases a solid single piece of soil was obtained. Where the soil in drying cracked or adhered to the container the specimen. was rejected. This method of obtaining volumes was later abandoned because of its known. inaccuracy. The results as finally recorded were obtained by immersing the dry specimens in mercury. While this method is no doubt more accurate than the one first used it nevertheless was quite cumbersome and could be improved upon. Up to this time the results to my knowledge have not been computed and tabulated; but Miss B. F. Monroe, Professor Shaw's laboratory assistant, thinks that the average shrinkage was about 20% in volume, on the basis of the original mud paste. With this information in mind our first idea was to obtain a larger specimen than that used by the Bureau of Public Roads so as to reduce the experimental errors in determining the volumes. We therefore decided to use a cylindric shape. We adopted two sizes as follows: (a) 2.4 inches diameter by 41⁄2 inches long; (b) 1.3 inches diameter by 41⁄2 inches long. (These molds are in the background of Figure 8.) Soil No. 12, from Ventura County, was crushed and passed through a 100-mesh sieve. It was then mixed with a known amount of water to such a consistency that it could be poured from a container into a mold. Instead of air-drying these samples we attempted to accelerate this part of the test by more rapid drying, but without success. The specimens cracked and checked so that they were of no further use. Even approximate determinations of shrinkage could not be made. Another source of inaccuracy was the condition of the mud paste; but to be sure that our molds were of the proper size we tried a smaller set 1.3 inches in diameter and 2.7 inches long. (One of these molds is shown in Figure 8.) These molds were paraffined and placed on greased plates. The mud paste was poured into the molds and they were set aside to dry. After a week's time only the upper portion had dried. The lower part still filled the mold and was in its original wet condition. One of these smaller molds is in the background of Figure 8. It became obvious, therefore, that the entire surface of the specimen must be subjected to air instead of just the top as was the case in any of the molds used. In doing this it is clear that the mud must be of thicker consistency. Just yesterday we tried mixing the pulverized adobe soil with small quantities of water, thoroughly kneading the mixture with additional water until it had a uniform texture. The consistency was about that of modeling clay and very easily handled. We now propose to make up cylindric specimens of adobe soil of the consistency mentioned, remove the mold immediately and in that way we will be able to expose a large surface to dry and thus overcome one of the previous obstacles. The procedure will be as follows: 1. Weigh out pulverized adobe soil, add water until the proper consistency is secured. This will require very thorough kneading. Record the amount of water used and calculate the percentage of moisture on the basis of dry material. 2. Compact the adobe paste into the oiled mold. 3. 4. 5. 6. Remove specimen from mold. If it has any air bubbles or other holes visible, discard. If it appears to be a solid homogeneous mass, weigh and immerse in kerosene to determine its volume. Set aside to dry under room conditions. To check the percentage moisture in the material molded into specimens make a moisture determination in the usual way by drying to a constant weight at 100 degrees Centigrade. After the specimen has dried one week in air place in a cold oven which will gradually be brought to a maximum temperature of 70 degrees C. After they have dried. to a constant weight raise the temperature to 110 degrees C. 7. Weigh the dried specimens. 8. 9. Immerse the specimen in kerosene until it will absorb no more. Then determine the volume as was previously done with the wet specimen. Compute percentage reduction in volume on the basis of dry volume. We think, at this time, that this procedure will give us uniform, comparable results that will be of value. We realize that we are not duplicating field conditions in that the mud may be wetter than we use it and field drying is not 100 degrees C. The conditions to be used are, however, consistent with general laboratory practice and when obtained under known conditions, results can be modified according to the judgment of the person using them. We propose to mold specimens from each sample obtained. We will choose one sample which represents the average condition and to it add sand and other materials as outlined in 1b in letter of October 29, previously referred to. We have attempted to use compact lumps of adobe as obtained from field samples, but here we have also been unsuccessful. All the material so far examined contains pebbles or small hard particles which strike against the tool used in dressing the specimen and either tear large gashes or break it so that it can not be used. From conferences and searches in the library, we conclude at this time that it is impossible to secure valuable data on expansion. We will try out any new ideas if obtained, but we will concentrate on the program outlined which we expect will give reliable data on shrinkage. Very truly yours, (Signed) C. T. WISKOCIL. No. 4 ADOBE SOIL December 25, 1920. 1. Percentage moisture (air-dry, in laboratory about three months). 3. Expansion, air-dry to saturated condition (in porous mold). 3a Specimen cut from lump. 3b Specimen compacted from dust. 4. Force of expansion, produced by absorption of water by confined mass of adobe soil. Adobe passing 100-mesh sieve. 5. Specific weights (air-dry). 5a Lumps in lbs. per cu. ft. 5b Average material slightly tamped. 5c Material passing 100-mesh sieve. 6. Linear contraction. 6a Specimens made from material passing 100-mesh sieve. 6b Specimens made from soil as received. 7. Sieve anlyses. (Signed) C. T. WISKOCIL. |
