The Effect of a Varying Moisture Content on the Settlement of Soil Subjected to a Constant Load Citation Leeper, Laverne Davidson (1932) The Effect of a Varying Moisture Content on the Settlement of Soil Subjected to a Constant Load. Master's thesis, California Institute of Technology. doi:10.7907/dn6n-z437. https://resolver.caltech.edu/CaltechTHESIS:09182024-183742361 Abstract No abstract. Item Type: Thesis (Master's thesis) Subject Keywords: (Civil Engineering) Degree Grantor: California Institute of Technology Division: Engineering and Applied Science Major Option: Civil Engineering Thesis Availability: Public (worldwide access) Research Advisor(s): Converse, Frederick James Thesis Committee: Unknown, Unknown Defense Date: 1 January 1932 Record Number: CaltechTHESIS:09182024-183742361 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:09182024-183742361 DOI: 10.7907/dn6n-z437 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 16742 Collection: CaltechTHESIS Deposited By: Benjamin Perez Deposited On: 24 Sep 2024 22:00 Last Modified: 24 Sep 2024 22:15 Thesis Files PDF - Updated Version See Usage Policy. 11MB Repository Staff Only: item control page

THE EFFECT OF A VARYI NG MOISTURE CONTENT ON THE SETrLEMENT OF SOIL SUBJECTED TO A CONSTA...'l\JT LO.AD Thesis by Laverne Leeper In Partial Fulfillment of the Requirerrents For the Degree of Master of Science California Institut e of Technology Pasadena, California 1932 TABLE OF CONTENTS ............................................. 1 ............................................... 2 Experimental Procedure ···•··•··••···••••••··•·••·••···••••• 6 Acknowledgment Introduction ................................ Soil Contai ner ........................................ General Considerations 6 Se curing a Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Iioisture Control ................................. •.... 10 Preparation of Sample for Test ·····•·•••·•·•·•··•··••• 12 Loading System ·•···•···••·••···•····•······•·····•·••· 13 ....................... 14 Device for Measuring Settlement Classification of Soil ................................ ........................................ Experimental Results ....................................... Conclusions ................................................ Miscellaneous Informa tion .................................. Test of Sample No. 1. ................................ Test Procedure Test of Sam.pile No. 2. ................................ 15 16 19 22 25 26 27 1 ACKNOWLEIXHvIENT The subject for this thesis was suggested by Mr . Fred J. Converse, and the work was carried out under his personal supervision. The writer wishes to express his sincere appreciation and gratitude for the valuable advice and assistance which Mr. Converse has so freely given. 2 INTRODUCTION It is surprising that while engineers have devoted so much time to the development and refinement of the theories of structures they have almost entirely neglected to consider the foundations upon which their structures must rest. Only recently has there been a general awakening to the importance of foundations in structural engineering. Some fifteen years ago Dr. Charles Terzaghi began experimenting with sands at the .American Robert College in Turkey. He later broadened his work to cover the field of soils in gene ral. The work which Dr. Terzaghi has done is the outstanding contribution so far in the field of soil mechanics. The results which he published in 1925(l) attracted the attention of the entire engineering profession and impressed it with the importance of such studies. Investigators have shown an increasing interest in the subject ever since. A special department for soil research was created at the Massachusetts Institute of Technology under the personal supervision of Dr. Terzaghi. Other schools, notably Iowa State College, have taken up the work on a large scale. The United States Bureau of Public Roads has undertaken extensive work with reference to highway subsoils. The engineering literature has begun to carry articles on soil mechanics and foundati ons. Apparently the profession is beginning to do justice to this heretofore neglected science. (1) A series of articles in Engineering News-Record, November 5 to December 31. 3 One factor which has to a great extent been responsible for the neglect of soil mechanics is the inherent complexity of soil itself. There s eems to be an almost unlimited number of variables that are involved in the behavior of soils. To make the science exact, all t he l aws governing these variables should be determined. On the surface this appears to be an almost hopeless task, not only because of the number of variables, but because of their inter-relation and effect on one another. However, some of the variables, such as density , perme- ability, cohesiveness, size of particles, coefficient of internal friction, void ratio, etc., a r e really functions of the type of soil. If soils are clas sified on the basis of clay, silt, and sand, it is found that for each type the above variables are constant - or at least nearly enough so for engineering purposes. If the study is to be the physics, chemistry, or geology of soils, the above mentioned variables and many more - will have to be considered. However, for engineering purposes, where the emphasis is on the me chanics of soils, the type of soil may be considered as a single variable which is the sum total of all the others. From the standpoint of foundations and footings t he variables have to do vdth soil mechanics. of these is the type of soil. Of course, the first and most important The othe r variabl es are: the unit load applied, the settement which occurs, the time of settlement, the shape of the loading plate, the diameter of the loading plate, (called d), the ratio of the depth at which load is a pplied to the diaraeter of the 4 loading plate (called t/d), and the moisture content of the soil. Since in structural work the practice is to use settleirent as a basis for determining allowable bearing values, it is desirable to know the re- lation of settle!l'Y:)nt to the other variables. For given soils the time-settlement relation under a constant unit load, with fixed-values of d and t/d, and constant moisture content has been quite well determined.(l) Similarly, for given soils, the unit load - final settlement relation with fixed values of d and t/d and constant moisture content has been studied.( 2 ) A variation of this relation is obtained by periodically removing the load during the test and observing the so-called ''rebound" ability of the soil. ( 3 ) ·The relations between final settlement and such factors as the shape of plate, d, and t/d have been observed and are definitely known.( 4 ) There is one relation, however, which is much less understood than those mentioned above. It is the effect of moisture content on the fi nal settlement of soil when all the other variables are constant. Terzaghi has observed t he effect of moisture content on the deformation of small unconfined cylinders of clay subjected to constant unit load.( 5 ) (1) Terzaghi in Engineering News-Record for November 12, 1925. in Bulletin 101 of Iowa State College, June 1931. (2) Williams and Wing in Transactions, .Am.Soc.C.E., 1929. ( 3) Griffith in Bulletin 101 of Iowa State College, June, 1931. ( 4) Terzaghi and Housel in Transactions, .Am.Soc.C.E., 1929. (5} Terzaghi in Public Roads for October, 1926. Briffith 5 His results, while not strictly applicable to foundation and footing condit ions, indicate that t he moisture content is an important consideration in any deforrnative test of soil. Apparently this fact has not been fully appreciated in connection with most tests for soil bearing values. An example of the neglect of moisture considerations in load tests can be found in the various building codes. In the past the problem of safe bearing loads on foundati ons has been taken care of in the codes by giving arbitrary beating values to different types of soils. The Los Angeles Code of 1923 and the Pasadena Code of 1926 have such provisions. Some of the newer and more advanced codes, such as the Uniform Building Code of the Pacific Coast and the proposed code for New York City, go a step further. 1bey state that where the bearing capacity of the soil is not definitely known or is in question, load test shall be made. The procedure for conducting these tests is given in detail, but nothing is said concerning the moisture content of the soil at the tL~e of test. The question immediately arises as to how much this important variable will affect the test. If a test is made in the spring the soil will be saturated from the winter rains, (in a California climate), and will probably not show the same results as one made in the fall after the summer drought. It was with the intent of determining the effect of moisture con tent on settlen:ent, particularly as it may relate to tests for allowable bearing values, that the problem of this thesis was undertaken. 6 EXPERIMENT.AL PROCEDURE General Considerations Most en gineeri ng soil research has been carried out in the field. For a test such as proposed in the Introduction to this thesis, work in the field is almost out of the question because of the enormous difficulty of realizing the main objective - moisture control. It was decided, therefore, to conduct the tests in the laboratory , while approximating as nearly as possible field conditions. The nature of the sample used is one of the most important conside r ations in any laboratory test of soil. If the r esults of t he test are to hav e any application to field con ditions the soil in the sample should have the same structure and physical properties as t hat in the field. This means that the sample should be removed from the ground without being in any way distunbed, since it is almost impossible to take loosened dirt and reproduce the original granular arrangement by artificial means. It becomes a requirement, therefore, that the sample used in the proposed laboratory tests be of undisturbed natural s oil. The particular problem of these tests was that of varying the moisture content of the soil. Reduction of the percent moisture would require either the application of heat or the effect of evaporation over a long time. Such procedure would be both inconvenient and difficult to control. The moisture content can be increased compara- ? tively easily , however, by merely introducing water to the sample. So for these tests it was decided to restrict the variation of moisture content to an increase. Generally the footings used for structures have a soil surcharge of some height, t, above their bearing surface. Therefore the ratio t/d will usually have some value greater than zero. So if footing conditions are to be simulated in the laboratory tests there should be some method of providing a surcharge above the base of the bearing plate. Also, since the load applied to footings is due only to a stationary weight, there should be provisions for static loading of the soil in the proposed laboratory tests. Soil Container For the tests described in this thesis a container was built which made it possible to secure an undisturbed soil sample and furnished an easy means of creating a surcharge above the loaded surface. made from a piece of sixteen inch seamless steel tubing. It was The piece was cut into two sections, one to contain the soil sample and the other to accommodate the surcharge. The part which was made into the container proper was about twelve inches in length. welded on one end to make a bottom. A circular steel plate was In the center of this bottom a hole was bored and tapped for a half-inch pipe plug. was welded on the container for ease in lifting. A pair of handles FIG. I I FIG. 2 FIG. 3 FIG. 4 I -----~-- - - · TOP VIEW I 16" -----.I /4' __;___.__,. , ------,5 1 co ---4 5 ° BEVEL - - r - - --~ -_ - _ - _ - _ - _ - _ - _ - _ - _ - _ - - - - - , _ - _ - _ - _ - _ - _ - - - ' - - _ - __ - __ I ~ PLUG I/ I VERTICAL SECTION FIG. 5 - DETAIL OF CONTAINER 8 The remaining part of the tube, about eight inches, was made into a holder for the surcharge. Since this holder was to set upon the container proper, the contacting edges of both parts were machined to a forty-five degree bevel. This gave a nice fit, and rendered the two parts stable when set up. To f~cilitate setting the two parts together a few guides were welded on the surcharge holder. Figure 1 is a photograph showing the two parts separated, while in Figure 2 they are assembled. The drawing in Figure 5 shows the container in detail with all dimensions. Securing a Sample Figure 6 shows the method of securing samples which was used for these tests. A hole about five feet in diameter was dug in the field at the place where the soil was to be investigated. After firm, undisturbed soil had been reached a cylindrical column of earth was allowed to remain standing in the center of the hole as digging continued. The hole was carried down until the height of the column was six inches or so greater than that of t he conta iner. As the work progressed the column was carefully trimmed so its diameter would be an inch or two greater than the diameter of the container. When the excavation was finished the container was centered on the column in an inverted position. contact with the soil. This brought the sharp bevel edge in It was then a simple matter to drive the con- tainer dovm over the cylindrical column. A sledge hammer was used for / STEP I STEP 2 STEP STEP 4 FIG 6 3 METHOD OF SECURING SAMPLE 9 the driving, and the blows were cushioned by a wooden plank. As the container worked its way down the soil directly below the cutting edge was trimmed away with a trowel so as to reduce the column approximately to size. It was possible by means of the hole in the base - which also served as an outlet for the air - to see when the container was full. After the container had been completely driven down it was undercut, as shown in Figure 6, until it was completely detached. The re then remained only the tasks of hoisting the sample out of the hole and squaring off the exposed soil surface. There is considerable physical labor involved in getting a sample by the method outlined above. This is clearly shown by the fact that to secure a soil sample of approximately one cubic foot volume it is necessary to move a total of about three and a half cubic yards of dirt. Besides, the sample and container together are a good lift, since they weigh close to two hundred pounds. This method of securing the soil sample is satisfactory in that the sampe which goes to the laboratory is made up of soil in the identical state as that in the field. The only possible objection might be that the soil is inverted relative to what it would be in the field. However, soil is nearly enough homogeneous that inversion will probably not make an appreciable difference. Besides, a device for securing a sample which could be used without inversion would be considerably more complicated and expensive. 10 Moisture Control In these soil tests, as has already been said elsewhere, the variation in moisture cont ent v1as restricted to an increase which was secured by the direct addition of water. The moisture in soil can best be expressed as a percent of the dry weight. Therefore, in order to calculate the moisture content at any ti me during a test, it was necessary to know both the dry weight of the sample and the total weight of water present. The water added was measured by volume, so was easily express ed in pounds. The original weight of water in the sample was determined by making a special moisture test. A small sample for this purpose was t aken at the same ti me as the large one and immediately adjacent to it. This special sample was heated in an oven at 105° C. until it reached constant weight. By noting the loss in weight the ori ginal moisture content could be calculated readily. The assumption was maa.e that the same percentage existed originally in the large sample. Therefore, by noting the total weight of the big sample when it was taken, its dry weight and the original weight of water were easily figured. Some difficulty was encountered in devising a method of introducing water to the sample. The first attempt was to pour water through funnel s which were inserted di:c~ ctly into the soil. But it was found that in the stems of the funnels there was not sufficient area of soil available to permit rapid percolation, so the idea was giYen up. 11 A little further experimenting indicated that water could be readily introduced to the sample through funnels and a layer of sand spread over the soil surface. A sand was selected which passed the No~ 8 Tyler Standard Sieve and was retained on the No. 10. Since some of the water introduced into the sand would undoubtedly remain, an experiment was conducted to ascertain the amount. A quantity of the sifted sand was dried in the oven to constant weight, and then placed in a large glass funnel on a piece of dampened filter paper. Water was poured through the funnel until volume measurements indicated that no more was being retai ned. Knowing the dry weight of the sand and the volume of water retained it was possible to figure out a "constant" for the sand. This constant gave the volume of water retained per unit weight of dry sand. So by dryihg and weighing the sand used in the test it was a simple matter to determine the excess water necessary to make up for that retained. There was some question as to whether water introduced at the surface would percolate sufficiently to produce a uniform moisture content throughout the specimen. this point. So an experiirent was performed to clear up A defi n ite quantity of water was added to a sample of soil through a layer of sand at the surface. Twenty-four hours later tests were made for moisture content at a number of points across a diameter and at different depths. It was found that the percentage moisture was practically uniform throughout the specimen. There was a slight difference at different depths, but nothing to indicate a definite 12 trend one way or the other. In fact, the variation with depth was no more than the variation across a diameter. So the assu.~ption of uni- formity of moisture content was made throughout these tests. Preparation of Sample for Test Figure 7 shows the sample as it was prepared for a test. The soil was removed down to a point about an inch below the top of the container. The new surface was smoothed and leveled. centered on this new surface. Then the bearing plate was A tin pipe which fit over the bearing plate was used to keep the sur•i.foarge in place. with the soil surface. The pipe came down flush It was 3-5/8 inches in diameter and about ten inches high. The special sifted sand was placed in a layer a:bout three quarters of an inch deep on the soil surface, after it had been dried and weighed. The stems of three glass funnels were inserted in the sand. were located as shown in Figure 7. The funnels Small vent pipes were provided to carry off the displaced air in the sand. It was necessary to provide a water tight seal between the sand layer and the surcharge above, ao there would be no loss of water through capillary action. This seal was furnished by a layer of paraffin which was applied in the molten state. Filter paper was placed on top of the sand to keep the hot paraffin from working into it. To keep the paraffin from leaking into the sand around the edges of the filter paper or through the fwmel holes, special putty seals were provided at these places. TOP VIEW LOADING COLUMN ----- PIPE FU~ ., .. ·.... · i .•. , . 0 _,,y--- .. ' AIR VENTS ------- . •. •. _· . .·o·; .. . 0 • . .. ' . ' . , .•. 0 : . . . ' . •. 0' • . ., . 0 ' • •. • '0 ' 0 • • '' .. . . . ..•. o· •. . ... . .. • ... ~- _-.-.~ :_:_. ·o· •. •.·.- -~ ... : .. 0 •. •.•. •.• : o· : • SAND SURCHARGE -- ·" .• • PARAFFIN PUTTY SEAL FILTER PAPER VERTICAL SECTION FIG. 7 - PREPARATION OF SAMPLE FOR TE ST 13 After the paraffin had hardened the surcharge holder was set in place and the surcharge applied. The effect of weight is all that the surcharge supplies, so it does not have to be the same material as the test specimen. In these tests ordinary sand was used. Loading System Since a static load was necessary for the test it was out of the question to use an ordinary testing machine for the loading device. lever system was finally adopted. A The drawing in Figure 8 shows the details of the system. The essentials of the system are: the lever, the weights, the loading column, and the bearing plate. The lever has an effective length of 36-3/4 inches, and is free to rotate about a knife edge at one end. Load was provided for the loading column, 12-1/4 inches from the knife edge, by means of weights suspended from the other end. Since the ratio of lever arms is 3 to 1, the effect of the weights is tripled. '11he load is carried through the column to the bearing plate, where it is distributed to the soil. The lever was made from a one inch square reinforcing bar. In the region of the contact with the loading column the bar is offset an inch upwards to give more stability under the existing forces. knife edge at one end bears against a rigid surface. V-notched to carry a holder for the weights. The The other end is r r ---- - - - - ---- - - - 12¼" - - - 7 -·-··- ¾" 36 - - -······--------- -- ------· --- -- I 0 11 LOADING COLUMN 11/a ___-/ BALL ? BEARINGS .. . .'. () : ~ . -~--- BEARING PLATE • _-i : -~<~:,: :<;): : .~:_: WEIGHTS RATiO OF LEVER ARMS 3 TO I DETAIL OF LOADING ~YSTEM >-1 14 The weights were of the type ordinarily used with platform scales. They weighed five pounds each and were slot t ed to fit around the holder. A short rod resting in the notch or the leve.r carried the holder. The loading column was made from a piece of 1-1/8 inch reinforcing steel. It is about 12 inches long. Its contact with the lever arm and the bearing plate is through ball bearings which seat in conical holes. The bearing plate is shown in detail in Figure 10. disk 3-1/2 inches in diameter. It is a steel The area is 0.066 sq. ft .. A unit load of 4000 lbs. per sq. ft. was desired on the soil so the total load transmitted to the bearing plate had to be 267 lbs~. Since the lever ratio is 3 to 1, 89 lbs. of weights were required at the end of the lever. Seventeen weights and the holder just ma.de the 89 pounds. A supplementary device used was the screw jack shown in Figure 11. This jack was paced under the weights, and whe:n~ver it was desired to relieve the load on the soil the weights were raised slightly. Device for Measuring Settlement The device consisted essentially of an Ames dial and a micrometer screw. The dial was rigidly attached to the loading column by means of brackets. The micrometer was located above the dial, and was carried from a fixed beam. Adjustments could be made so the piston of the micrometer contacted the buttom of the dial. Then as the bearing plate and loading column settled, the dial was carried down relative to the I ~ I d_ i ! ;> l I I t /l FIG. 9 AMES COLUMN "-- ~ LOADING DIAL I! ,.' ! 11; 1 ; ;/ / // // : ~> LEVER ~ WELDED ~ BRACKETS \ I 11.\ ® JI DEVICE FOR MEASURING SETTLEMENT --- - - - - - - LEVER l ____-/ I MICROMETER ! I ~- __, "1 l ~ F1XED BEAM ~ - - ·- - - - CONICAL HOLE TOP VIEW - ---- 3l/2" - - ~ 7 I 1/;• DIAM. ·-r : c:o ~ .____ _ _ _ _ _ _ _ _ _ _ ___, _L_ FRONT VIEW AREA= 9.6 SQ. IN.= 0.0665 FIG. 10 - DETAIL OF BEARING SQ. FT. PLATE SCREW ~ - - - - ROTATING COLL AR BEARING SURFACE -.. n= T '=tif .-_.. __..l.,:::------!==----,-,.i..__ --~ ,,@11 11~ 11 11 l=-c-J 11 I 11 11LJI • _IJ FIG. I I - DETAIL OF JACKING DEVICE 15 micrometer, which remained fixed. Therefore, the dial recorded the settle- ment. As the settlement neared the capacity of the dial, the button could be moved down by turning the micrometer, and the whole dial range again made available. This process could be repeated until the capacity of the micrometer was reached. The total range made available by this method was 1.3 inches. The drawings in Figure 9 show the system in detail. The device for securing vertical and horizontal adjustment of the micrometer is shown clearly. T'ne Ames dial read directly to thousandths of an inch and had a capacity of .3 inch. The micrometer also read to thousandths of an inch directly, and its capacity was one inch. Classification of Soil It was noted in the Inttoduction to this thesis that the type of soil is one of the most important factors influencing any engineering soil test. This means that a definite basis of classification must be adopted if the tests are to mean anything. The Bureau of Public Roads has evolved a method of classification based on mechanical analysis and rigid definition.(l) (1) "Highway Subsoils". The analysis is Transactions, Am.Soc.C.E., 1930. IOO%CLAY S ANDY CLAY SILTY C LAY FIG.12-TRILINEAR SOIL CLASSIFICATION CHART 16 made to determine the percentages by weight of sand, silt, and clay in the soil. These percentages are then plotted on a trilinear chart, such as shown in Figure 12. The plotted point will fall within one of ~he regions on the chart arbitrarily assigned to a certain kind of soil, and the classification is thereby made. The mechanical analysis of the soil was based on the A.S.T.M. Standards.(l) Clay was taken as that part of the soil which would not settle out through 8 centimeters of water in 8 minutes. Of the soil which did settle out, that passing the 200 mesh Tyler Standard Sieve was taken as silt, and that retained on the sieve was taken as sand. The percentage of each constituent was based on the dry weight of the soil. Test Procedure The samples were secured and preparations made for tests as outlined above. The particular values wanted ffom the tests were the readings of final settlement for each moisture content. Since it required very little additional effort, it was decided to observe also the time rate of settlement. Settlement was allowed to proceed and readings were taken only when the moisture content was uniform throughout the sample. (1) So the "Mechanical Analysis of Subgrade Soils". A.S.T.M., Serial Dl37-22T (1922). 17 tests had to be interrupted periodically while water was being added. During this period, and until uniform mo isture conditions again existed, the load on the soil was relieved. used to secure this effect. The jack previously described was The weights at the end of the lever were gradually raised by the jack until a backward movement of about one thousandth of an inch was recorded on the Ames dial. When this point was reached sufficient load had been removed so no settlement would occur, yet enough remained to prevent rebound. place for t wenty-four hours, The jack was left in which was sufficient time for percolation to produce a uniform distribution of moisture throughout the sample. For each moisture content settlement r eadings were taken with time. The time int erval was one minute at first, when the settlement was rapid, and. was gradually lengthened as the settlement becarne slower and slower. 'Ille settlement was considered to be complete for a particular moisture content when not more than a thousandth of an inch occurred in t welve hours. The first mois ture content was, of course, that originally present in the sample. After settlement was complete for this first condition, the load was jacked up and the calculated increment of water added to the sample. In addition, enough water was added to allow for that retained by the sand layer. Twenty-four hours later the jack was lowered, and time settlement data were taken for the second moisture content. Then when settlerrent was complete t he jack was inserted, t he water added, an d the whole procedure repeate d. The process was carried on 18 until t he limit of the micromet er and A~es di al was approached, or until time warrant ed the ces sation of the t es t. The wa t e r was adde d to the sampl e slowl y, usually in dos e s spaced about fift een mi nut es apart. 3ach dose consisted of approximat ely 50 c.c. per funnel. Time settlement dat a were usually taken st a rting at one o'clock in the afternoo n. By 6 :00 P.Ivi . t he se ttlement had usually slowed down sufficiently to permit of taking the next reading the f ollowing morning . 19 EJG>ERTMENTAL RESULTS Two complete tests were conducted , one on a sandy loam and the other on a clay loam. It was f ound that in general settlement took place more rapi dly with the former than with the latter. Fi gure 13 shows comparative time-set tlement curves for t h e two materials. It ca n be seen that for s andy loam t he great est part of th e s ettlement takes place during the first five minutes, and t hat after that time settlement is almost negligible. In the case of clay loam a l a r ge part of the settlement also t akes place during the first five minutes, but not such a large percentage as was true for s andy loam. It can also be s een that for clay material the s ettlement continues at an appreciable rate for an hour or so after the very f ast s ettlement of the first five minutes. At the higher moisture con tents the difference in s e ttlement rates for the t wo materials is more marked. Curves were plotted to show the r elati on of time, settlement, and moisture. The curve for sandy loam is shown in Figure 14, and that for clay loam in Fi gure 15. Due to the limitation of space, it was possible to plot only the first hour's settleme nt for the lower moisture contents. Of course in all cas es t he re was some s e ttlan ent aft er the first hour. It was neglecte d for th ese curves, howeve r, since t hey were only intended to aid in visualizing the time-s e ttlement rel a tion for progressively great er moisture contents. The curve for sandy loam, Figure 14, shows that a considerable settlement occurs when the load is applied at the first moisture con- TIME HOUR5 -- 0 2 Z 4 I I 5 LOAD= 4000 LBS. PER SQ. FT. CllO.O I w :I: u 3 - t/d=21/2 .02 MOISTURE= 6.5 % - UNIT WT (DRY)= 117.5 LBS. PER FT. 3 1- z .03 ~ w ~ w ...J I- .04 1-- w I --·- I '--- I (f) ~ - .05 .06 DENSE SANDY 0 MATERIAL TIME HOURS -2 3 4 I I 5 LOAD= 4000 LBS. PER SQ.FT 0.01 (/) - t/d = 21/z w 0 .02 MOISTURE= 8.6 °/, z - UNIT WT. (DRY)= 90.0 LBS. PER FT.~ I- .03 z w ~ I.LI ...J -- .04 II~ .05 ~ .06 --- ~ i .07 CLAY MATERIAL FIG. 13 TYPICAL TIME SETTLEMENT CURVE.S 20 tent. For the succeeding moisture contents the settlement increases in small but nearly unifonn steps. More settlerrent seems to take place after t he first f ew minutes for the higher moisture cont ents than for the lowe r. This is probably due to t he small amount of clay in the material being rendered plastic by t he moisture. Figure 15 shows that for clay material the amount of s e ttlement gets larger for each successive moisture content. ]'or th e last moisture content - 13% - the amount of settlement is much greater than for any other, being , in f a ct, nearly twice the ent ire previous settlement. The curve also s hows that as the moisture content increases more settlement tends to take place after t he first few minutes, until at 13% more than half the settlement takes place after the first ten minutes. This greater but less rapid settlement is due to t he increas ed plasticity of the clay at the higher moisture contents. The moisture-final settlement curves are shovm in Figures 16 and 17. For the sandy material it is seen that over the range of the test the settlement was proportional to the moisture, thereby giving a straight line. The clay material, on the other hand, shows an entirely different behavior. The curve is concave downward from the start, indicati ng that gr eater t han proportional settlerrent occurs for succeedingly hi gher moisture contents. At about 12% the curve begins to drop down rapidly, indicating a rather sudden softening of the 21 material. Terzaghi observed a similar phenomenon in his tests of clay cylinders.(l) He calls the moisture content at which the material begins to deform rapidly the "plastic limi t r'. (1) Terzaghi in Public Roads, October, 1926. ,___ - - :' 6.5 - - ·-·-· >+----- - - - 7. 5 - - ·· ,...L.c .•' II l:QQ QI l l i 0.01 ! I i :- I I 2:QQ 3 : 00 ! 4'.00 TIME HOURS 6:00 - - + - -----+--- --- +--- ~ ________ - - --~ __ - -+, ---+1-_ I I t i ·- ---- - - I I ..___ ! •• I I 1 CURVE I I t I I TIME- SETTLEMENT- MOISTURE I I _j__ i ! i , I ' : ---, - 1 SANDY LOAM - 117.5 LBS. PER CUB. FT. i .04 - - - - - - - +i- -- s:oo 7:00 ....l _ _ _.._ :r _ __.... I _ _____ __ _ _ _I_ _--+-/_ __,__ _ _ _ _ _ _ _ _ __ ....i _ _-+-_ _ __ _ __ i- -- + - - - - + - - -~ - - - - ----r- ·- - --,..-- -- - - + -I- - - 4 - - - - - , . .!- - ! -- - ----------- - 12.5 - - - ·- ---__.: j ► s:oo ' ! .02 1 + I l i .03 ! ! MOISTURE - PERCENT - 10.5 ···-·· -· -+- ------ -------·- ·· 11.5 ---··-·· ---- -- >+- 8.5 - --·--;-·-- - 9.5 ! --+-- ----- --···-- .- - .. . - - L ___ - -'------,. -- --- - -- ·· -- -------- - _:.__ . I . • I ~ - - -- -----+- -• I I r LOAD= 4000 LBS. PER SQ. FT. ! I i I! I I I ! - - .-- --+-- - -- - ... - - --- -. -- --j- r _j_ ____ _j_ ______ _ I - • - '-- l I ' i i ~.07 t - - I - - - - -- - - --r ! a.08 ~ j ! i I I : -- -- r --- i_---_- _ ----.....;_ _ _ _ I ! i - - - - - - + - - ---+-- - -+ - - - - - - - < - - - - +I_ I .09 t - - - - + - - - + - - - - - + - - - + - -i- - - r - - - · _ __ ___ ---+---- - -- .I I .12 i ___ _ _ _ _ _ - -- ------~- _ __ _ l _ __ __ I i _ . 4 : i _,__ i ' 1 - - 11- - - - -- ~ - - ·- i - - - - -- - - -t·-I -- --•· -- i--- - - - - - - -- ~ I -- - - - - - + - - -~ - - - -: - + - - -- ------+:- - - - - - - - - ~ -- - - 1 - - - - + - - - + - - - - - t - - -- - --+--- - --- +------1 i !--1--r---~~-r=~~:i:I ~===i ~-- . • ---~ --j__! _____,____,.,.___-··-t ··-·- -· l ' _ .;:: 1 =:::jb:~:::±==dee==~-...:..,l....--l ! I 1 · i iI i ! i I - I __,I- I I I I __ · -+ - ~ I ~ -- - ! .13 ,~ I I ! ··r - - + -~- = = ~ t : : : : : = ~ - - 4 --- I !i - --- --- - -. i - J. _ _ __ - - + - - - - - - - - +- ! I"-'--- - · -- ______ : I I - ·- ------t-- - - - - - - • -· - ! IO >--- -- + - - - - - - - - + - - - + - - - - + - - - + - - - - - -- • ~ ! i - -+-- ---- --- _;_ __ . - - • - · -- .. _ !_ ! I I _ __t - -- __ J _ _·- +- I I FIG. 14 - -t-- - i ---+i- --+ ' -- - - 1 - - - - + - - - + - - - + - - - - t - - - - - - t I 1.. ► 1"" 8.6 1:00 0 MOISTURE- PERCENT 10.8 - - - . . J t - r - 1 ~ , ( - - I !. 9 ---,>~lt""'""~T-I_M_E___H_O_U_R_S____________ l 9.7 2:00 3:00 4:00 5:00 13.0 6:00 7:00 8:00 9:00 I !i I . f .· · ···· + - · · + - . I TIME- SETTLEMENT- MOISTURE I ! CURVE I I 1-------+-- -I - - t - - - - - - r - - - - r - -- . - - - - , - - - + - - - 1 - - - - + - - t - - - - t -- !i ~-t----- - ----;----------,- - - - - t -- CLAY ,! ·- LOAM - 90.0 LBS. PER CUB. FT. LOAD= 4,000 LBS. PER SQ. FT. .3 0 ..------r·- - ~ r - -- I i -·--+----+---+-------r--,-------i--' - .~_ ~ ~·t: ~ - - - t - - - ·-+--1- -1 ! I i I I- ·- -·- -·-~--- ....." ' ~ - - - --- -...--- - - - 11 1-~-=--~t-: : : ~ : .~ :.~ u- -- -+- - + - - - + - - + - - - - t - - - - t - - - - t - - - - - . - - - - - 1 - - - - r - - - - t - - - - -- -- - - - - - + L - - - - , i - - - - - + - - - i - - - - r - - - - t1-- - - - - t 1' +--tI .40 i-!, - - i - - - - - + - - - - - - - + - - - -- -;- - - - t - --- - -- - + - - - + -- -+---+--+---+---t----+ : _ _+i - - - t - - - - t - - - - + - - - - t - - - - t - - - - i i I 50 1---- - - - 4 - - + - ~ I- - ti- - - , -----r----r-----,---,--,-·J_· _ _ _- f - - _ t - - - - - ; - - - - 1 - - - - i - - - ~- - r - - · - - - - t - - - t L - --r-i ---i----;--~ · !_-~=_,·i,- ---i---· ---+-----+--- - --· ·- - ------ - .---·1: . 6 0 - ___ _ T ___ ·-·--+-1---1. I +-· t . . I r -- - .l--- - · -- ' -·--- --+----- -- ~ I l I .1 o 1-------t-- ' ! i -1·-- ' ·-~-I --r- I ·-~ - •• •• . " . I i l - ·- ·-+--·· ----+ --I i 11 -· . I I --~- -----~ ------ - - --+-----+ • . .• - - - - ·- ·-- . . - • . -----·- 7i ... I ·- ~ - -- - - I --L----·- "\ I . +- _L + - [ - _- r - . r- : • : ~ - .- : ·----- -t I -·-·- ·· ·--··· ~ ~- - --·---+-- - - - - - t - - - - · ··- i - ·- -·- - ~ --1 J i i i - + ----+---+., --+----+-! 1- - ~ - - + - - + - - - - - + - - +- - t - - - - t - - ~ c: t-~ - t - - - - t - - - t - - - t - - - t - - - i1- - - tj- - - t - - - - t:----t1- - r - :i- - i II i ,1 I I I I i I ! ! I l i I I I ; i 1 I l~r-------.... .80 ~--i----+--+---t--__j.--+---+-----t---t--+--j---t----t--t---j--t--~~-i---+----r--t---,--+---t--:-:-1 i i I I I i I I I i .9 0 l _ _ J .I __ I ! i _L_ I ' I I _L,_ I __,L._ _j__ I i ------II ! ! l 1 I i _ _ J_ _. J . . . _ _ . . . i , _ _ _ . . . . i . , _ _ ~_ I i i ! I I I _ _IJ _ , _ - - - - l , _ - : - - 1 . . - - - ~ - . . . . . _ _ _ . . . . . 1 . . . ._ _ _ _ _ _ _..........,_ FIG 15 I I _ _ . _ _ _ _ _ _ _ _ _ _ _ ~_ I II I _._.._ I ! i __.__ _ _ _ _ _ __ MOISTURE- PERCENT-8.0 9.0 I 0.0 7. 0 11.0 12.0 t- z w v,. 0 8 ~w t - - - - -- w:I: _JU t-Z.I 0 t-- w MOISTURE - SETTLEMENT CURVE (/) t . 12 SANDY LOAM ~QNSTANT LOAD .14 ,.___ ___.__ _ ___.___ _ _....___ _____._ _ ____.__ _ ___.___ __. FIG. 16 8.0 9.0 MOISTURE- PERCENT-10.0 I 1.0 12.0 ·-- - - - · - - 13.0 - - - ···- -·•- --+------I . 20 1 - - -- f - - (/) .30 w I ~ .4 0 , - - ---+--- - ------+--- I t- ~- 50 -- MOISTURE-SETTLEMENT CURVE ~ w _J t-. 1- 60 CLAY LOAM w (/} .70 -- --t t CONSTANT LOAD . 801--------1- - - --····-· ·-- - - .90 1.00 -- - - - l . . . - - - - - - ' - - - - - J . . . . . . . . __ __;___....___ _ ___.___ _ _--.&.-_ FIG.17 __, 22 ( '\ CONCLUSIONS From the results of these tests the following conclusions can be drawn: 1. The moisture content of soil has an important effect on settlement of the soil under load. 2. An increased moisture content causes an increased settle- ment under a constant load. 3. For sandy soil the increase in settlement is approxi- mately proportional to the increase in moisture content. 4. For clay soil the settlement increases at a rate more rapid than the increase in moisture content. 5. A moisture content is finally reached for clay soils where settlement increases enormously. limit". This is called the "plastic A one percent increment in moisture content near this point will result in an increase in settlement which may be double the entire previous settlement. 6. Settlement in general takes place very rapidly in sandy soil. 7. Most of it occurs during the first five minutes. In clay soil settlement is less rapid than in sand. More settlement takes place after the first five minutes, and as the moisture content increases this effect becomes more pronounced. 23 The results of this thesis show that the moisture content of soil is of! considerable importance in soil testing. "I'ests performed in strict accordance with building codes could give widely different results for a particular soil, depending upon the moisture content. For instance, it was folllld that with clay soil a load of 4000 lbs. per sq. ft. would produce almost three times as much total settlement at 13% as at 12%. According to the proposed New York Code, which rules out a soil that settles more than an inch under the specified load, it would be entirely possible for a clay soil to be satisfactory at one moisture content and quite unsatisfactory at one a little higher. In the light of the above facts, it would seem wise to require moisture determinations in connection with all tests for allowable bearing values of soils. The moisture content at the time of test should then be compared to the maximum likely to occur before arriving at a bearing value. A foundation condition could easily occur under a completed building that approximates closely that of the tests described in this thesis. The footings of the building might easily be placed during the summer or fall months. · A certain initial settlement would take place corresponding to the existing moisture content. Then during the winter and spring months the moisture content of the soil under the exterior footings would be increased by the rains and additional settlement would occur. Since the soil under the interior footings would not be so affected, the additional settlement would not take place at such places. The final result would be differential settlement of interior and 24 exterior footings. This condition will cause considerable moments in the building frame above. Mr. M. P. White in a thesis(l) has found that differential settlement of 1/4 inch produces moments in the frame approximately of the same magnitude as dead or wind loads. So the effect of moisture in the soil is not negligible. The possibilities in the field of soil research have barely been touched up to the present. Under the leadership o~ such men as Terzaghi a start has been made, and it is to be hoped that in the future work will be continued. (1) nprimary and Secondary Moments in a Building Bent Due to Footing Settlement", C~lifornia Institute of Technology, 1932. \ 25 MISCELLANEOUS I11FORMATION General Weight of container = 73.1 lbs. Volume of container utilized = 1.11 cu. ft. Area of bearing plate = 9.6 sq. in.= 0.066 sq. ft. Unit load on soil = 4000 lbs. per sq. ft. Total load on soil = 267 lbs. Load at end of lever = 89 lbs. Ratio of lever arms = 3:1 Passes #8 sieve, Retained on #10. Sifted sand used Constant for snnd = 28.4 c.c. water per pound. Capacity of Ames dial = .3 inch. Smallest scale division of Ames dial = .001 inch. Capacity of micrometer = l inch. Smallest scale division of micrometer .001 inch. Oven used was thermostatically controlled to within one degree Centigrade. Temperature for all moisture reduction tests was 105° C. Constant weight was assumed when the loss of weight in one hour was less than 0.1%. Tyler Standard Sieves were used with Ro-Top Machine, which was operated ten minutes for a test. 26 Test of Sample No. 1. Analysis: Clay = 15% Silt = Sand = Classification 77% Sandy loam. Unit weight dry = Original weight of sample = 139.0 lbs. Original moisture content = 6 .5% Final moisture content = 12.5% Increment of moisture = 1% = 600 c.c. Weight of sifted sand used = Excess water added = 150 c.c. Unit load applied = 4000 lbs. per sq. in. Shape of bearing plate Diameter of bearing plate Value of t/d Total settlement 117.5 lbs. per cu. ft. 5.3 lbs. circular. = 3-1/2 inches. = 2-1/2 = .122 inches. 27 Test of Sample No. 2. Analysis: Clay = 29.2% Silt = 30.3% Sand = 40.5% Classification Clay loam. Unit weight dry = 90 lbs. per cu. ft. Ori ginal weight of sample = 108.6 lbs. Original moisture content = 8.6% Final moisture content = 13.0% Incremant of moisture = 1.1% = 500 c.c. Weight of sifted sand used = 5.3 lbs. Excess water added = 150 c.c. Unit load applied = 4000 lbs. per sq. in. Shape of bearing plate circular. Diameter of bearing plate = 3-1/2 inches. Valle of t/d = 2-1/2. Total settlement = .966 inches.