MISCELLANEOUS PAPER S-72-29 COMPARISONS OF VIBRATED DENSITY AND STANDARD COMPACTION TESTS ON SANDS WITH FINES

by

F. C. Townsend

R5“ B> IDS II

TA June I972 7 .W 34m Sponsored by Office, Chief of Engineers, U. S. Army S -7 2 -2 9 1972 Conducted by U. S. Army Engineer Waterways Experiment Station Soils and Pavements Laboratory Vicksburg, Mississippi

APPROVED FOR PUBLIC RELEASE; DISTRIBUTION UNLIMITED library

AUG 1 6 1972

Bureau of Reclamation Penver, Colorado

Destroy this report when no longer needed. Do not return it to the originator.

The findings in this report are not to be construed as an official Department of the Army position unless so designated by other authorized documents.

This program is furnished by the Government and is accepted and used by the recipient with the express understanding that the United States Government makes no warranties, expressed or implied, concerning the accuracy, completeness, reliability, usability, or suitability for any particular purpose of the information and data contained in this pro­ gram or furnished in connection therewith, and the United States shall be under no liability whatsoever to any person by reason of any use made thereof. The program belongs to the Government. Therefore, the recipient further agrees not to assert any proprietary rights therein or to represent this program to anyone as other than a Government program. BUREAU Of DC______t\ 92016120 c 'Cj ^

COMPARISONS OF VIBRATED DENSITY AND STANDARD COMPACTION TESTS ON SANDS WITH FINES

by

$ F. C. Townsend

t>A\V c0

June 1972

Sponsored by Office, Chief of Engineers, U. S. Army

Conducted by U. S. Army Engineer Waterways Experiment Station Soils and Pavements Laboratory Vicksburg, Mississippi

ARMY-MRC VICKSBURG. MISS.

APPROVED FOR PUBLIC RELEASE; DISTRIBUTION UNLIMITED

FOREWORD

The investigation reported herein is part of a continuing evaluation of laboratory testing procedures for the Office, Chief of Engineers (OCE), under Item ES 516 of the Engineering Studies Program. Authorization for the testing program was given by OCE letter dated 16 November 1970, subject: Engineering Study 516, stating that a study should be made of sandy soils with fines as to which test method gives the higher compacted density, the vibratory table method or the standard compaction test. Testing was conducted during the period January 1971 through September 1971. The tests were performed by Mr. F.G.A. Hess under the general supervision of Dr. F. C. Townsend, both members of the Laboratory Research Section, Embankment and Foundation Branch, Soils and Pave­ ments Laboratory. The report was prepared by Dr. Townsend under the general supervision of Mr. J. R. Compton, Chief, Embankment and Foundation Branch, and M essrs. J. P. Sale and R. G. Ahlvin, Chief and Assistant Chief, respectively, Soils and Pavements Laboratory. COL Ernest D. Peixotto, CE, was Director of WES during prepara­ tion of this report. Mr. F. R. Brown was Technical Director.

m

CONTENTS

P age FOREWORD ...... iii CONVERSION FACTORS, BRITISH TO METRIC UNITS OF MEASUREMENT...... vii SUMMARY...... ix PA R T I: INTRODUCTION...... 1 B ackground...... 1 Purpose and Scope...... 3 Previous Investigations...... 3 PA R T II: D ESCRIPTION OF THE STUDY...... 7 Investigative Procedures ...... 7 Test Results ...... 8 Discussion of Test Results ...... 8 PA R T III: SUMMARY AND CONCLUSIONS...... 15 LITERATURE CITED...... 16

Figures 1-16 Table 1

APPENDIX A. CORRELATIONS OF DENSITY VALUES WITH GRADATION PARAMETERS

APPENDIX B. ANALYSIS OF VARIANCE TABLES FOR DENSITY CORRELATION EQUATIONS

v

CONVERSION FACTORS, BRITISH TO METRIC UNITS OF MEASUREMENT

British units of measurement used in this report can be converted to metric units as follows:

Multiply By To Obtain inches 2. 54 centimeters pounds 0.45359237 k ilo gram s cubic feet 0.0283168 cubic meters pounds per square inch 0.6894757 newtons per square centim eter pounds per cubic foot 16. 0185 kilograms per cubic meter

SUMMARY

Generally, two laboratory test methods, relative density and standard compaction (impact), are used in establishing density require­ ments for the placement of embankment m aterials. The relative density test method is specified for cohesionless soils, generally when fines do not constitute more than 5-12 percent by weight, while the standard compaction test is for cohesive soils. However, for sandy soils contain­ ing varying amounts of fines, selection is often based upon the test method considered appropriate to the m aterial. This study was an in­ vestigation of various criteria for assisting in compaction test method selection for cohesionless soils with fines.

The effects of gradation, percentage and plasticity of fines, and moisture on vibratory and impact compaction of granular soils were evaluated by adding measured percentages (9, 16, and 23 percent) of low plasticity (ML) and medium plasticity (CL) fines to a poorly graded (SP) and a nearly well-graded (SW-SP) sand. Maximum density tests using a vibratory table were made on both oven-dry and saturated soil, minimum density tests were made on oven-dry soil, and standard compaction tests were performed on m aterial at various water contents.

Test results indicate that a uniform sand, due to its higher void space, can accommodate more fines and densify more effectively than a well- graded sand with fines. Plasticity of the fines and moisture were found to be interrelated factors affecting the compaction of sand with fines. For low plasticity mixtures, saturation facilitated vibratory compaction. Conversely, for more plastic mixtures, adhesion of the fines to the sand grains restricted vibratory shifting of the grains into a denser structure. The same densities are produced by impact and vibratory compaction at higher percent fines added to the well-graded sand compared to the per­ cent fines added to the uniform sand. Apparently, compaction of a well- graded sand with fines is more affected by water content than a uniform sand with fines.

Because moisture and plasticity of fines have such opposing effects on impact and vibratory compaction of sandy soils, guidance for com­ paction test selection is not clear cut. The current practice of basing compaction test selection on results of relative density tests on oven- dry m aterials and standard compaction densities may not be realistic

IX of field conditions and may lead to the untenable conclusion that vibratory compaction should be used for sands containing in excess of 20 percent fines. It is recommended that the use of the relative density method for compaction control be limited to granular soils with 12 percent or less fin e s.

x COMPARISONS OF VIBRATED DENSITY AND STANDARD COMPACTION TESTS ON SANDS WITH FINES

PA RT I: INTRODUCTION

Background

1. Proper compaction is an important criterion in the placement of embankment materials. Commonly, two laboratory test methods of compaction are utilized as standards for comparing placement densi­ ties. For cohesive materials, the standard compaction test (impact) is used and for cohesionless soils the relative density test is used. In the case of sandy soils containing fines (i. e. , particles < No. 200 sieve), current Corps of Engineers’ guidance is not clear cut. The following statements from Engineer Manuals relate to this matter: a.. EM 1110-2-1906, Laboratory Soils Testing (1) Appendix VIA. Compaction Test for Earth-Rock Mixtures, Paragraph 1, ”If less than 5 percent by weight of the total sample is less than the No. 200 sieve, maximum density should be determined by vibratory methods. ” (Appendix VI, Compaction Tests, which applies to material having not more than 10 percent larger than the 3/4-in. sieve, is silent on the subject. ) (2) Appendix XII. Relative Density, Paragraph 7. ’’The relative density is meaningful only for cohesionless materials; if a soil has any appreciable dry strength, the methods for determining the minimum and maximum densi­ ties described in this appendix are not applicable. ”

1 b. EM 1110-2-2300, Earth and Rock-Fill Dams, General Design and Construction Considerations, Paragraph 5-6b(l). "The average in-place relative density of zones containing cohesionless soils* should be at least 85 percent, and no portion of the fill should have a relative density less than 80 percent. This requirement applies to drainage and filter layers as well as to larger zones of pervious m aterials, but not to bedding layers beneath dumped riprap slope protection. The requirement also applies to filter layers and pervious backfill beneath and/or behind spillway structures. The relative density test is generally satis­ factory for pervious m aterials containing only a few percent finer than the No. 200 sieve. For some m aterials, however, field compaction results equal to 100 percent or more of the standard compaction test maximum density can be readily obtained and may be higher than 85 percent relative density. If 98 percent of the maximum density from the standard compaction test is higher than 85 percent relative density, the standard compaction test should be used. The design should provide that clean free-draining pervious m aterials be compacted in as nearly a saturated condition as possible. Otherwise compaction at bulking water contents might result in settlement upon saturation. " 2. The American Society for Testing and Materials (ASTM)

"* The standard compaction test may be more applicable than the relative density test for cohesionless soils having more than about 5 percent by weight finer than the No. 200 sieve, depending upon the particle size distribution."

2 Specification D 2049-64T suggests that 12 percent fines be considered as a basis of selecting between standard Proctor and relative density tests. Guidelines, specified by the Bureau of Reclamation, classify soils suitable for vibratory compaction into two groups: (a) suitable and (b) borderline. Borderline soils may contain up to 12 percent fines, but control is based upon 95 percent of Proctor maximum density or 70 per­ cent relative density, whichever produces the greatest unit weight.

Purpose and Scope

3. The objectives of this study were to investigate the effects of various factors (gradation, percentage and plasticity of fines, and moisture) on densities obtained in relative density and standard compac­ tion tests on sands with various amounts of fines. The test results would then be analyzed to establish criteria for selecting which test method, vibratory or impact, should be used for compaction control of cohesionless materials. 4. The testing program included testing of a uniform (SP) and a nearly well-graded (SW-SP) sand to which were added various per­ centages of low plasticity (ML) or medium plasticity (CL) fines. Maxi­ mum density tests using a vibratory table were performed on oven-dry soils, and standard compaction tests were performed on soils at various water contents.

Previous Investigations

1 2 5. Previous investigations, ASTM and USBR, have indicated that free draining soils respond more efficiently to vibratory compaction

3 than to impact compaction. Hence, it appears that the concept of free draining soil type according to the Unified Soil Classification System is the basis for the current guidelines used for selecting Impact or vibra­ tory compaction methods. For example, SW and SP sands are con­ sidered as being free draining and by definition contain less than 5 per­ cent fines, while SC and SM sands are somewhat impervious and are defined as containing in excess of 12 percent fines. Although free drainage would seem to be an important criterion in selecting the com- 2 paction test method, tests by the USBR have shown that there is poor correlation between permeability and effectiveness of vibratory compaction. 6. The influence of gradation, which involves both percentage of fines and grain-size distribution, on maximum and minimum densities 3 of sands has been emphasized by several investigators (Burmister, 4 5 Hutchinson and Townsend, Shockley and Garber, and USBR). Burmister has shown that the maximum and minimum densities of co­ hesionless soils can be correlated with the grading parameters, D , DU , and C^ (effective grain size range) and that reasonable estimates of relative density based on these correlations are feasible. However, he does not discuss any correlations for maximum density obtained by impact methods. Hutchinson and Townsend and Shockley and Garber have correlated Bagnold’s^ grading parameters by regression analysis with maximum density achieved by vibration and impact procedures. Hutchinson and Townsend’s results indicated that minimum density did not accurately correlate with Bagnold’s parameters. All the above correlations, however, were developed for fairly clean sands with only small percentages of fines. 7. Investigations by the USBR have shown that the percentage of

4 fines present in a material is a possible criterion for selecting whether impact or vibratory compaction yields the greater density. Their test results showed that compaction test method selection based upon the percentage of fines was confined to a relatively narrow range for the broadly different soils tested and that some sands (SM) with as much as 16 percent fines could be efficiently compacted by vibration. They con­ cluded that the two prim ary factors affecting the ability of a soil to be densified by vibration were (a) gradation and (b) plasticity of fines. Gradation affects the void space available for the fines present. For example, a uniform sand would contain more voids than a well-graded sand in which the voids are filled by the finer fractions. The greater void space of the uniform sand would thereby allow a higher percentage of fines to be present and still vibrate satisfactorily. Plasticity of fines also contributes to the effectiveness of vibratory compaction in that the more plastic fines tend to restrict particle movement into denser configurations by adhering to the sand grains and creating ’’bridging11 between grains. Obviously, plasticity would be an insignificant factor if the relative density tests were conducted on oven-dry material. 8. The 1970 edition of EM 1110-2-1906, Laboratory Soils Testing, has eliminated the laboratory procedure once permitted of determining maximum density by vibrating sands in a saturated state. Nevertheless, .7 Kolbuszewski demonstrated that vibratory densification of sand was more effective when the material was compacted under water rather thani-i.. m the air-dried condition. . Hutchinson . and Townsend 4 confirmed this observation for uniform sands, but showed that well-graded sands produced higher densities in the air-dried state. Felt 8 and Pettibone 9 and Hardin indicated that for sands of varying gradation and percentage of fines (0 to 15%), the difference between maximum density produced

5 by vibrating wet versus dry sand was insignificant. However, Felt points out that for coarse-grained (gravelly) m aterials, maximum vibrated density is achieved in the saturated condition. It was hypoth­ esized that for these coarse-grained m aterials, water aids in prevent­ ing segregation and in "lubrication" of the particles. At moisture contents between the wet and dry extremes, lower densities were ob­ tained. Pettibone and Hardin indicated that for the sands they tested, no reliable correlation was available between gradation and densities produced by vibrating saturated or dry m aterials. 3 9. In addition to effects of gradation, Burmister stated that grain shape (i.e. , rounded versus angular) has some influence on the maxi­ mum and minimum densities of cohesionless soils. Al-Hussaini showed that for a very uniform gradation (No. 3 to No. 30 sieve), little difference existed between the maximum vibrated densities of angular crushed stone and rounded gravel (however, there was considerable difference in minimum densities (92.4 versus 112.0 pcf*), the angular m aterial having the lower minimum density). 10. In summary, previous investigators have reported that the maximum and minimum densities of cohesionless soils are greatly in­ fluenced by gradation and percentage of fines, and to a lesser extent by plasticity of fines, moisture, and particle shape.

* A table of factors for converting British units of measurements to metric units is presented on page vii.

6 PA RT II: DESCRIPTIO N OF THE STUDY

Investigative Procedures

11. The natural sands utilized in this study were (a) subangular to subrounded concrete mortar sand (nearly well graded, SW-SP), and (b) a local subangular to angular sand termed Campbell Swamp sand (uniform, SP). These sands were processed by sieving to remove all natural fines. Standard soil samples of ML and CL material passing the No. 200 sieve were used as fines to be added to the sands. The proper­ ties of these fines and the grain-size distribution curves of the sand plus fines mixtures are presented in figs. 1 and 2. 12. Standard test procedures outlined in EM 1110-2-1906, Appendixes VI and XII, were followed for standard compaction and relative density determinations, respectively. The standard compaction procedure involves impact compaction of three layers of soil into a 4. 0-in. mold by 25 blows on each layer of a 5-1/2-lb hammer falling 12 in. Water content determinations were made after compaction. The maximum density test procedure for use in relative density determina- 3 tions involves vibrating oven-dried soil in a 0. 1 ft (6-in. diam) mold on a vibratory table at a frequency of 3600 vibrations per min and amplitude of 0. 019 in. for 8 min under a surcharge of 2 psi. Maximum vibrated density was also determined on saturated material using the procedure described in the 1965 edition of EM 1110-2-1906. In this procedure, wet sand was placed in the mold as it was vibrating at greatly reduced amplitude, maintaining a small amount of free water above the soil. After the mold was filled, the surcharge was placed and the mold vibrated for 8 min. In addition to the above tests,

7 specimens of (SW-SP) sand plus ML and CL fines were vibrated after being brought to water contents corresponding to the optimum water contents obtained in the standard compaction tests on the same mixtures. Minimum density of oven-dry material was determined by pouring the material through a funnel device into the mold.

Test Results

13. Results of relative density and standard compaction tests on the two sands and on the sand-plus-fines mixtures are summarized in table 1. Definitions of the notations used in this table and in subsequent figures to identify the various density determinations are given in the footnotes of table 1. 14. Standard compiaction curves for the sands and sand-plus-fines mixtures are shown in figs. 3 through 6. Densities obtained on oven-dry material are also plotted on these figures. Vibrated densities of oven- dry material and saturated material are plotted against percent fines in figs. 7 through 10, together with standard compaction densities of oven- dry material and of material at water contents producing the highest density. Figures 9 and 10 also show the densities designated Y^(VO) obtained in vibrating the SW-SP sand plus ML and CL fines, respective­ ly, with the mixtures at standard compaction optimum water contents.

Discussion of Test Results

Comparison of densities obtained on oven-dry versus wet material 15. Figures 11 and 12 graphically show the difference in dry

8 density obtained in vibrated tests on oven-dry versus saturated material, and in standard compaction tests on oven-dry material versus material at optimum water content. These plots indicate that higher vibrated densi­ ties are obtained on sands without fines when the materials are oven-dry. 4 Conversely, Hutchinson and Townsend observed that saturated uniform sands with less than 10 percent fines produced higher vibrated densities g than the same sands dry. However, data presented by Felt and Pettibone 9 and Hardin indicated differences in vibrated densities of saturated versus dry sands were not significant for sands with various percentages of fines. The current test results indicate that for the uniform sand, the vibrated density of oven-dry material exceeded that of saturated material by the greatest amount when there were no fines. Apparently, in a well-graded sand, the vibratory movement of the smaller particles into the voids be­ tween the larger particles is unhampered by saturation, while in a uni­ form sand the absence of smaller particles to fill the voids and the pres­ ence of water reduced the saturated density slightly. In standard compac­ tion of the sands containing no fines, the effect of "apparent cohesion" im­ parted by moisture is evidenced by the fact that the densities of oven-dry materials were greater than the maximum densities of moistened materials, differences being greater for the well-graded sand. Evidently, at opti­ mum moisture content, more "apparent cohesion" is developed between the various sized particles of the well-graded sand than for the uniform sand and a more open structure is created. 16. Obviously, plasticity and moisture are interrelated factors af­ fecting compaction of the sand-plus-fines mixtures since the plasticity characteristics of the fines would not be exhibited in compaction of oven- dry material. The test results in fig. II show that in the case of the sands with the more plastic fines (CL), the maximum vibrated density

9 of the dry material was higher than the saturated material for both the well-graded and uniform sands. Similarly for both the uniform and well- graded sands with small amounts of ML fines, the dry material initially produced the higher maximum vibrated density. However, as the per­ centage of fines increased, the saturated mixtures with ML fines ulti­ mately produced the higher maximum vibrated density. The vibrated density tests on saturated materials indicate that the more plastic fines adhere to the sand grains and thereby restrict the shifting of the sand particles into denser configurations. In the case of the sands with less plastic fines (ML), adhesion was not as great and the water present as­ sisted in filling the voids with these fines, thereby generating higher densities. Based upon these observations, it appears that the plasticity of the fine fraction rather than gradation is the primary factor in govern­ ing whether oven-dry or saturated sand with fines will produce the highest vibrated density. 17. Vibration of well-graded sand with fines at standard optimum water content produced the lowest values of vibrated densities. As shown in figs. 9 and 10, the density-percent fines curves were below and generally parallel to the curves for saturated material. 18. The effects of plasticity observed for the standard compaction densities, as shown in fig. 12, were somewhat different than for the vibrated densities. For the sands with less plastic fines (ML), the dry material densified more effectively than the same material at optimum moisture content throughout the range of percentages of fines tested. On the other hand, for sands with the more plastic fines (CL), the dry material with zero or small amounts of fines densified more than the same material at optimum moisture content but as the percentage of fines in­

creased, greater densities were obtained with the addition of moisture.

10 These results are as expected since moisture facilitates the compaction of cohesive materials, and the cohesive nature of the sand-plus-fines mixtures increases with greater percentages of fines. Comparison of vibratory versus impact compaction 19. Figure 13 graphically compares the differences between vibrated densities of oven-dry material and standard compaction densities for various percentages of fines. The plots show that for the uniform sand (SP) with fines, the difference in densities produced by the two compac­ tion methods reduced to 2 pcf with 9 percent CL fines or 12. 5 percent ML fines, and to zero with fines content of 14. 5 and 18. 5 percent. On the other hand, for well-graded sand (SW-SP) with fines, the differences in densities produced by the two methods remained large up to about 20 percent fines. At 12 percent fines, the vibrated densities of dry well- graded material exceeded the standard compaction density by 7. 4 to 8.7 pcf. These results and those presented in figs. 11 and 12 indicate that well-graded sands with fines are more affected by factors which lead to "bridging1' between particles; i. e. , "apparent cohesion” and plastic fines, than uniform sands with fines. Apparently, uniform sand contain­ ing more void space (i. e. , having higher void ratio) than well-graded sand can accommodate more fines in its structure, while the various sized particles of the well-graded sand restrict the movement of fines and are more conducive to "bridging. " 20. Figure 14 graphically indicates the difference between dry densities corresponding to 85 percent relative density of oven-dry material and 98 percent of maximum standard compaction densities at optimum water content for various percentages of fines. If guidance given by EM 1110-2-2300 (see para, lb) is followed, then compaction control by relative density rather than by standard impact compaction

11 would be used for sands containing as much as 14 to 21 percent fines. However, it appears to be inappropriate or artificial to compare vibrated densities obtained on oven-dry materials with impact compaction densi­ ties of moist materials, which because of substantial fines content, show definite moisture-density relations (i. e. , an optimum water content and a maximum density are defined). 21. Comparisons of standard effort density and vibratory density, when both tests are performed on oven-dry material, indicate generally lower fines contents at which standard effort densities exceed vibratory densities (6. 5 and 8 percent fines for the SP sand mixtures; 14 and plus 23 percent fines for the SW-SP sand mixtures). 22. Because moisture has such significant effects on the compac­ tion characteristics of the mixtures, it would seem preferable to compare maximum densities at optimum water content with those ob­ tained by vibratory tests on saturated materials. These comparisons are presented graphically in fig. 15. The test results indicate that materials containing up to 10. 5 to 15 percent of medium plastic fines (CL) will densify more by vibration when saturated than by impact methods when at optimum moisture content. However, in the case of lower plastic fines (ML) higher densities were produced by vibration in a saturated condition for fines contents in excess of 23 percent. Ap­ parently, the ML fines when saturated do not exhibit sufficient plasticity as do the CL fines to be restricted in their mobility when vibrated. As a result, the density increases with increasing fines (ML) until all the voids are filled by the fines. Beyond this condition, additional ML fines tend to segregate and move around the surcharge plate and out of the mold, as well as begin to alter the sand-fines structure. For the well- graded sand (SW-SP), it appears that due to the lower void ratio the

12 mobility of the ML fines is more restricted and alteration of the struc­ ture occurs rather than segregation, causing an increase in volume and lower densities. However for the uniform sand mixtures, evidently segregation occurs and the ML fines move out from beneath the surcharge plate, with results indicating the erroneous conclusion that vibratory compaction is better than impact compaction for unlimited quantities of fines. Although material consisting of 100 percent ML fines was not saturated and vibrated, it would be anticipated that the surcharge plate would merely settle to the bottom of the mold; this is based on the fact that in tests on sands with 23. 1 percent fines, it was noted that considerable fines were present on top of the surcharge plate at the conclusion of the test. 23. Analogously, for impact compaction, the density will increase correspondingly with the amount of fines up to a certain fines content; after this, the addition of fines will reduce the density to that cor­ responding to 100 percent fines (approximately 106 pcf for ML fines and 109 pcf for CL fines). The fines content required to achieve the maximum density is obviously quite different for impact and vibratory compaction due to the different sand-fines structure created by the two m eth ods. Correlations for compaction test selection 5 24. Previous investigators (Shockley and Garber ) have obtained 6 good correlations between maximum densities of sands and Bagnold’s grading parameters. Because vibratory techniques used by these investigators were different and the sands investigated contained little or no fines, these correlations were not applied directly in the present study. Therefore, independent correlations were made based on

13 Bagnold's grading parameters and percent fines, as presented in Appendix A. 25. Correlations were obtained relating standard maximum density at optimum water content and maximum density obtained by vibrating oven-dry material to percent fines and Bagnold’s distribution curve parameters . With these correlations, limiting values of Bagnold's parameter S and percent fines can be established to indicate which of c the two compaction methods would produce the higher density for a given sand with fines. Unfortunately, this is not a realistic criterion since the correlations, by ignoring the obvious effects of plasticity of fines and water content, lead to the untenable conclusion that vibratory compac­ tion should be used for sands containing as much as 25 percent fines. 26. It was thought that simple tests such as roughly checking dry strength by hand and noting whether or not moist material could be molded by hand into a coherent m ass might provide some guidance in selecting the appropriate compaction method. For both sands, it was found that material with 17 percent CL fines had considerable dry strength after being oven-dried, but 23 percent CL fines had to be added to produce a coherent mass of moist material. On the other hand, moist sands with 23 percent ML fines could not be molded except into a weakly coherent m ass, and the mixtures felt like wet sand. The sands, even with 23 percent ML fines, had no appreciable dry strength upon being oven-dried. It thus appears that dry strength and moldability provide no guidance in the selection of compaction test since both the well- graded and uniform sands behaved similarly, while compaction test data showed differences for the two sands.

14 PART III: SUMMARY AND CONCLUSIONS

27. The guidance provided by the testing of the two sands with fines of different plasticity characteristics is not clear cut, as there are dif­ ferent relations of vibrated versus standard compaction densities with increase in fines depending on sand gradation, moisture conditions, and plasticity of fines. Since it does not seem practicable to specify the method of compaction control on the basis of these param eters, and since field experience has shown that essentially cohesionless m aterials other than those containing large proportions of silts or rock flour are best compacted by vibratory equipment, it is recommended that the use of the relative density method be limited to sands containing 12 percent or less fines. It is believed that vibratory compaction of essentially saturated sands with fines in excess of this amount would be ineffective.

15 LITERATURE CITED

1. American Society for Testing and Materials, "Relative Density of Cohesionless Soils, M ASTM Test D 2049-64T, 1964. 2. U. S. Bureau of Reclamation, "Research Tests to Investigate Criteria for Selection Between Vibratory or Impact Compaction Methods," Earth Laboratory Report No. EM 441, 1955. 3. Burmister, D. M. , "Physical, Stress-Strain, and Strength Responses of Granular Soils, " American Society for Testing and Materials (ASTM STP 322, 1962, pp 67-97. 4. Hutchinson, B. and Townsend, D. , "Some Grading-Density Relationships for Sands," Proceedings, 5th International Conference on Soil Mechanics and Foundation Engineering, vol 1, 1961, pp 159-163. 5. Shockley, W. G. and Garber, P. K. , "Correlation of Some Physical Properties of Sands, " Proceedings, 3d International Conference on Soil Mechanics and Foundation Engineering, vol 1, 1953, p 21. 6. Bagnold, R. A. , The Physics of Blown Sand and Desert Dunes, 1943, Morrow and Co. , New York. 7. Kolbuszewski, J. J. , "An Experimental Study of the Maximum and Minimum Porosities of Sands, " Proceedings, 2d Inter­ national Conference on Soil Mechanics and Foundation Engineer­ ing, vol 1, 1948, p 158. 8. Felt, E. J. , "Laboratory Methods of Compacting Granular Soils," American Society for Testing and Materials (ASTM) STP 239, 1959, pp 89-108. 9. Pettibone, H. C. and Hardin, J. , "Research on Vibratory Maximum Density Test for Cohesionless Soils, " American Society for Testing and Materials (ASTM) STP 377, 1965, pp 3-19.

16 10. Al-Hussaini, M. M. , "Plane Strain and Triaxial Compression Tests on Painted Rock Dam Material, " Report 3, Technical Report No. S-71-2, 1971, U. S. Army Engineer Waterways Experiment Station, CE, Vicksburg, Miss.

11. Goodnight, J. H. , "Multiple Regression Analysis for the IBM System 360," 1967, Department of Experimental Statistics, North Carolina State University, Raleigh, N. C.

17 T able 1 Summary of Density Test Results MATERIAL M inim um Standard Compaction Maximum Vibrated Density D ensity Yd at °d Percent Fines Yd(dry) Yd(opt) w at Yd(opt) 95% Y^iopt) 98%Yd(opt) Yd(VD) Yd(VS) Yd

100 “n — r r _ "1 T— r r T "T ~ n P T 2 1 — 1I—r~ 1— r T

90 . 3 10 A I 80 \ 20 \

70 30 £ fi

% L Lt 40 * I 5 0 at Z < sz o -1 u u3 4 0 J 60 z V - U l ■ U -1 at d ____ I k . F i ne s 30 70 * S é * . . '(ti ;ì 3 . 1 20 80 ! -Gl 6 . 7

10 -* 9 • 1 90

0 100

GRAIN SIZE MILLIMETERS GRAVEL SANO COSBIES SILT OR CLAY COARSE

NAT W % LL PL p i SAMPLE NO. ELEV OR DEPTH CLASSIFICATION PROJECT CAMPPFLL SWAMP SAMP + q . i <£ 2 3 5 ML FIMES 2 8 2 2 ....12 16«?? and 2?01# fines. CL FIMES l h - . .

AREA S 0 .97 O Sand (SP) - .. = t ,R_ r 1 v ü BORING N O .

GRADATION CURVES DATE

ENG FORM 2 0 8 7 KEPI ACES WES FORM NO. 1241 SEP 1962. WHICH IS OBSOLETE. 1 MAY 63 Fig. 1. Grain size distribution of Campbell Swamp sand (SP) plus fines U.S. STANDARD SIEVE OPENING IN INCHES U.S. STANDARD SIEVE NUMBERS HYDROMETER 6 4 3 2 1% 1 xh % A A 6 8 10 14 10 20 30 40 50 70 100 140 200

100 50 1 0.5 0.01 0.005 0.001 500 GRAIN SIZE MILLIMETERS WR«»«. COBBLES SILT OR CLAY COARSE | PINE COARSE | MEDIUM — r ~ PINE

SAMPLE NO. ELEV OP DEPTH CLASSIFICATION NAT W% LL PL pi pro jec t QTA\rmnr) nnwr.fl f t f . m o p ^ a h Ml. 23 5 T? TMIT u 22 12 ______SAND - 0 t . o.lt. Ifi.7* an d ______

area ? 1 1 < f i n e s t P O F , V'r. : : _,f ;• _ , ~ * ' •* , = ( bo rin g n o .

GRADATION CURVES DATE

U S. GOVCRNMtNT MINTING OFFICE . IM S OF —709-11« f"*?™ 2087 REPLACES WES FORM NO. 1241, SEP 1962, WHICH IS OBSOLETE Fig. 2. Grain size distribution of concrete mortar sand (SP-SW) plus fines Dry Density, pcf F ig . 3. M oisture-density relation sh ips for C am pbell Sw am p Sand Sand p am Sw pbell am C for ips sh relation oisture-density M 3. . ig F S) ls fne; tnad o paction com standard es; fin L M plus (SP) Dry Density, pcf F ig. 4. M oisture-density relationships for Campbell Swamp Sand Sand Swamp Campbell for relationships oisture-density M 4. ig. F S) ls L ie; tnad compaction standard fines; CL plus (SP) Fig. 5. Moisture-density relationships for concrete mortar sand (SW-SP) plus ML fines; standard compaction Dry Density, pcf F ig . 6. M oisture-density relatio n sh ip s for concrete m o rta r sand sand r rta o m concrete for s ip sh n relatio oisture-density M 6. . ig F S S) ls ie; tndr cmpaction com dard stan fines; L C plus -SP) (SW Dry Density, pcf 125 95 O g. . ions ewen brt nst n mai m u axim m and sity en d d rate ib v een betw s n riso a p m o C 7. . ig F tndad cmpcin de iy f mpbel wa sn (P) (SP sand p am Sw ell b p am C r fo sity en d paction com ard d stan 10

cent nes, % , s e in F t n e rc e P us fnes e fin L M s lu p 15

20 25

30 Dry Density, pcf 125 g. . mpaio bten irtd est ad xmu um axim m and density vibrated between s arison p om C 8. . ig F tnad o ato dniy o Ca bl S mp ad (SP) sand p am Sw pbell am C for density paction com standard ls fines L C plus A Dry Density, pcf 140 g. . mpaio bten irtd est ad xmum m u axim m and density vibrated between s arison p om C 9. . ig F standard com paction density for concrete m o rta r sand sand r rta o m concrete for density paction com standard S S) ls fines L M plus -SP) (SW c Fi , % s, e in F t n rce e P Dry Density, pcf 140 WO 0 g. 0 Co rs bten irtd est ad xmu um axim m and density vibrated between s n ariso p om C 10. . ig F standard com paction density for concrete m o rta r sand sand r rta o m concrete for density paction com standard 10 S S) ls fines L C plus -SP) (SW

re Fi , % s, e in F t ercen P 15

20 25

30 25 Percent Fines 20 10 1 O -10 5 ------F ig . 11. E ffect of fines on the differen ce between between ce differen the on fines of ffect E 11. . ig F

vibrated d en sities of oven-dry and and oven-dry of sities en d vibrated ML FJ NES FJ ML -5 d D-dV) pcf Yd( VD)-Yd(VS), aurtd sand rated satu □ L FINES- CL * S O O YMBOL SAND L O B M SY + 0 1 W-P + NES E IN F L C + -SP SW O W-P + FI S E IN F L M + -SP SW A S ML" NES E IN L "F M + SP O S CL NES E IN L F C + SP □ Percent Fines, F ig . 12. E ffect of fines on the d iffe ren c e betw een een betw e c ren iffe d the on fines of ffect E 12. . ig F sta n d a rd m a x im u m d e n sitie s of o v e n -d ry and and ry -d n e v o of s sitie n e d m u im x a m rd a d n sta i st e cnet sand content re tu is o m m u tim p o Y JD R Y )-Y ,(O p T ). pcf pcf ). T p ,(O )-Y Y R JD Y d d SYMBOL W-P C FNES E FIN CL + -SP SW O A P ML I ES FIN L M + SP o S + I ES FIN L C + SP □ W-P M FNES FIN ML + -SP SW S d n a Percent Fines, g. 3 Efect f i s n h difrnc bt e mai brtd rated ib v m u axim m een betw ce ifferen d the on es fin of t c ffe E 13. . ig F e iy f vn-r sn ad xmum sa r o ato nsity en d paction com ard d stan m u axim m and sand -dry oven of sity den d(D - dot, pcf Yd(opt), - (VD) Yd YMB L BO M SY O □ SP + M L F in e s s e in F L M + SP W-P CL Fi s e in F L C s e + in F -SP L M SW + s e in -SP F SW L C + SP SAND g 1. fc o ie o te ifrne ewen 5 ret f xmu um axim m of ercent p 85 een betw difference the on fines of ffect E 14. ig. F Percent Fines irtd est o oe-r sn ad 8 re o mai m um axim m of t ercen p 98 and sand oven-dry of density vibrated standard com paction density paction com standard SYMBOL O A □ W-P C Fines F CL + ines F -SP ML SW + -SP ines F SW CL + ines SP F ML + SP SAND Percent Fines, g. 5 Efect f i o te fer e bt e mai brtd rated ib v m u axim m een betw ce n re iffe d the on s e fin of t c ffe E 15. . ig F nst f aur ed sn ad xmum sa r o ato e sity den paction com ard d stan m u axim m and sand d te ra satu of sity en d YMB L BO M SY O □ W-P CL Fi s e in F L C + -SP SW SW -SP + M L F in e s s e in F L M + s e in -SP F SW L C s e + in F SP L M + SP SAND APPENDIX A CORRELATIONS OF DENSITY VALUES WITH GRADATION PARAMETERS

1. The Bagnold^* grading parameters were obtained as illustrated on the distribution plot shown in fig. A-1. The log of the percent retained on a given sieve is plotted against the log of the mean particle diameter of the sieve interval (assuming that the mean diameter of the material retained on a given sieve is the average diameter of that sieve and the next larger sieve). This special distribution identifies the three Bagnold grading parameters: the peak grain diameter (Dp), the slope of the distribution curve for the sizes coarser than the mode (S c ), and the slope of the curve for the sizes finer than the mode (Sf). A least squares regression analysis was used to establish the values of Sf and

S c . 2. Table A-1 lists values of Bagnold's grading parameters, per­ centage of fines, standard compaction density, and maximum and mini­ mum densities for relative density purposes from tests performed in this study and from tests on other soils. These values were analyzed assuming the following regression model:

Yd = + Pj(%F) + P2 (Sc) + P3 (Sf) + P4 (Dp) + e5[log(10 Sc)] + 06[log(lO Sf)].+ P7[l0g(10 D )] (Al)

A1

/ where

- dry density, pcf

(3 . . . 0^ = beta coefficients for given variables

%F = percent fines, %

S = slope of frequency curve for coarse sizes (Bagnold's distribution)

= slope of frequency curve for fine sizes (Bagnold’s distribution)

D = peak grain diameter, mm (Bagnold’s distribution) P A statistical analysis system (SAS) developed at North Carolina State

University by Goodnight*** was utilized for performing the multiple regression analyses. Analysis of variance tables of the correlations are presented in Appendix B. The analyses indicated that several of the selected variables were insignificant. Therefore, variables which were common to both the maximum standard compaction density and maximum vibrated density were selected for the following regression equations:

Y t (opt) = 112.78 + 0.95 (%F) - 2.70 (S ) (A2) d c

Yd(VD) = 124.89 + 0.47 (%F) - 3.80 (Sc) (A3)

2 2 The coefficients of multiple regression are r = 0.921 and r = 0.831, respectively, which indicated very good correlations. Standard errors

* Refers to similarly numbered item in list of literature cited, which follows the main text. A2 of estimate are _+ 3. 0 pcf and ± 4 . 2 pcf, respectively.

3. Unfortunately, the same variables, %F and S , did not correlate c well with the minimum density. The following regression equation 2 provided a good correlation (r = 0. 852 and standard error of estimate

3.87 pcf for estimating y . ). d(min)

Yd(min) = 96-26 ' 12-06(Sf)+ 9 .7 7 (D p)

26. 76 [ log( 10 S )] c (A4)

However, the following correlation based upon the maximum vibrated

density offers a somewhat more accurate means for estimating v 6 a(min) Yd(min) = 0-M3[Yd (VD)l - 6. 13

2 the coefficient of regression being r = 0.895 and the standard error

of estimate being + 3. 24 pcf.

4. Since the regression equations for y (opt) and y (VD) are d d functions of the same variables, by equating the two equations, a limiting

equation can be derived to indicate which of the two compaction test

methods would produce the higher density for a given sand with fines.

Equating equations A2 and A3:

112. 78 + 0. 95(%F) - 2. 70(Sc) = 124. 89 + 0 .47(%F) - 3. 80(Sc) yields

S - 11.01 - 0.44 (%F) c (A6)

A3 which is presented graphically in fig. A-2. The areas above and below equation A6 designate which test method gives the higher density.

A4 Table A -1 Values of Percent Fines, Bagnold's Grading Parameters, and Densities Utilized for Correlation D P Material %F Sc Sf mm Yd(VD) Yd(min) Yd(opt) Campbell Swamp 0 5.32345 -3.46551 0. 298 106. 1 90. 0 98.4 sand plus: 9. 1 5.34753 -3.46903 0. 300 no. o 94. 8 107. 2 ML Fines 16. 7 5.35825 -3.47732 0. 295 116.4 98.4 115. 3 23. 1 5.30821 -3. 47803 0. 290 115. 2 94. 6 116.2 CL Fines 9. 1 5.34753 -3. 46903 0. 300 109. 2 95. 2 107. 2 16. 7 5.35825 -3.47732 0. 295 115.4 98. 2 116. 1 23. 1 5.30821 -3.47803 0. 290 117. 6 96.4 122. 7 Concrete mortar 0 0.201905 -1.62283 0. 315 125. 4 111.8 114. 5 sand plus: 9. 1 0. 199205 -1.62241 0. 330 130. 3 113. 7 121.7 ML Fines 16. 7 0. 198900 -1.62183 0. 330 132. 7 112. 3 128. 3 23. 1 0. 198055 -1.62066 0. 330 129.0 108. 6 130. 5 CL Fines 9. 1 0. 199205 -1.62241 0. 330 133.9 113. 5 124. 2 16. 7 0. 198900 -1.62183 0. 330 136.0 112.9 130. 5 23. 1 0. 198055 -1.62066 0. 330 120. 7 108. 9 131. 1 Miss. River 0 2.31760 -3.41417 0. 410 110. 1 98. 2 106. 7 sand plus: 9. 1 2.33552 -3.41333 0. 410 123. 5 104. 6 116. 7 ML Fines 13. 1 2.31789 -3.41275 0.410 125. 5 107. 5 121. 3 16. 7 2.31802 -3.41454 0.410 126. 8 108. 6 125. 6 SD 7C* 11.5 1.17153 -1.02559 1.80 120.4 92. 3 - SD 13 & 14 2. 7 1.08367 -1.39540 1. 15 119.9 89.4 - SD 27-30 16. 5 1.97720 -0.62796 2.40 128. 5 99.2 - SD 13 17.4 5.53485 -2.27958 . 240 102. 1 81. 1 - Reid Bedford sand 2. 3 3.-73956 -3.18583 0. 186 108. 2 91.1 99. 6 Hopkinton- Everett sand 3.4 1.12252 -2.3572 0. 205 107« 2

* SD denotes Soil Dynamics; the numbers are the sample numbers Grain Size, mm 10 1.0 o. l o. 01 100

U. S. Sieve Nos.

Fig. A-1. Bagnold's grain-size distribution param eters for concrete mortar sand (SW-SP) and Campbell Swamp sand (SP) Bagnold’s Grading Parameter, F ig. A~2. Correlation between percent fines and and fines percent between Correlation A~2. ig. F anl' gaig aa tr S eter, param grading Bagnold's c A P P E N D IX B

ANALYSIS OF VARIANCE TABLES FOR DENSITY CORRELATION EQUATIONS .. . .-..... A.NAL_YJ¿JJ> 1;.F__ V A P 1 A N C E J.A l 'J t »... K F G P tS Sl uN COEF F ICI E NTb » A N u b 1 A T 1 SJ I p S UJ_ f I 1 F 0 H u EPE n DENT VARIABLE X 5 TOWN

SuUt'CL- üf SuM OF b U 0 A K E b MEAN bCUAKF F V A l. 1 1 t

R F 0 KtSSi 0,* 2 1767.6590 527i <íH3.H29826o6 19.38491869 V a r i a b l e R e p r e s e n t s -----__------X5 DL VIA ! IUNb ? n 357.9351120 0 1 7.89o765/P Maximum Relative Density XI Percent Fines, % í) N N S 22 23 25.69475/08 X 2 Bagnold’s Grading Param eter, S c

R-S;jUAPfc = 0 . 8ol60 7 0 7 SIGMA = 4 . 2 3 0 4 5 6 7 u

SOURCE Sb F0H a <1 ) AuJ bS IF X (1) LAST T FuR h ü ,;jh ( i ) = n tí V AluF b STD EkRuR B STD 0 VALUES

X U 124.88673782 X 1 ¿17.95797729 290.04768453 4 . U ? 57 56 18 0 . 4 7 2 9 0 6 0 2 U . 1 1 7 4 7 0 1 1 0 . 3 7 0 U 2 5 8 1 X 2 164 9 . / n 167542 15 49 .7 ij 16 75 42 -9.30543971 -3.80214116 ü . 4 0 8 5 9 3 3 9 -0.85530586

St T F XPF C TE J 0 H b F k V E1 ! ['IFF EK EN Ct ______1 ______104.64622974 1 O 6.1 0 u 0 0 0 3 8 - 1 . 4 6 3 7 7 ü 52 2 10 8. 85 81 1 90 l ÎIU.ÜOUOOOUO -I.l4l88u86 3 . 1 u8,8.5811 y o.i 109.19999981 - 0 . 3 4 1 ö 8 ü 6 ö 4 112.41144562 116.3999996? -3.98855376 5 112.41144562 115.39999902 - ? . 9 8 8 6 5 o 7 o 6 115.6280025/ 115.19999961 0.4 2 8 3 0 0 6 7 7 ______115.6280025/ 1 1 7.60 0 0 00 o 8 - 1 . 9 7 1 o 969U 6 124.1 i 9 il 6 6 ? 4 126.3999996? -1.28093272 ..... 9 128.432/7'/4 ¡J 130.29999924 -1.86722182 1Ú 128.4327774ü i3o.899999o? . -5.46722221 . ...‘...... il 13?.02802277 102.7 Ou P Ufi 76 # -0.67197794 12 1 3 2 . 0 2 8 U 227 / 136.OOOOOÜüO ’ - 3.9 71 9 7 / ? U _____ l A_. 135.06783275 ______1_? 9 . 0 ü UJ.JJ. Oj j 0 .... _ 6 . 0 5 7ô 34 Po 1 4 105.06783272 i 3 Ü . 7 0 u 0 u 0 / 6 4 . 3 5 / ü 3 o ?7 15 116.0 746958o llÜ.lÛuOuOok 5 . 9 7 4 6 9 5 4 6 l 6 120.31020646 120.6 0oGuOu 0 -3.18979371 17 .12?. 2.215709/ i 2 6 • 6 0 u 0 ü Ü LM1 -3.27842847 i o 12 3.9706290 1 1. ? 6.8 0 o 0 0 0 i 9 -2.829l7o64 1 9 1 6 6 . H 1 n n 3 4 3 •) 1?U.3999990? 5 . 4 7 0 8 3 6 3 9 2 G 1 2 2.0 4 3 O 1 7 7 9 119.899999o? ? . 1 4 3 o 1 ô 5U y .... - 21 1 2 5 . 1 / 2 U 9 0.3 9 128.5 0 UfiüOu P -3.32790601 2 2 111 .760 0 8636 108.19999961 3 . 6 5 6 0 8 7 2 9 23 111.8dl6588o 9.7 81 b 5 9 1 6 i 0 2.10 0 0 U 0 0 8 - .-.._......

nFV|AIInWSQ9Q09 ... A ï i L jlS).ô 511 ? tUi______3.5 7 » 9 3.5.112 U.P 0 . Ü Q_0 U 0 u 0 u ANALYSIS ur VARlANCfc U hU , REGRESSION CuFFF IC i E H T ^ , a Nu STATlSIHjS Uf Ell I OK pERENDENI v A RI A BE E X 7

SuUKCL tjF SUM OF SQUARES MEAN SQUARE K v A l 1 i *- Variable Represents REGRESSION 2 I7b9.804o8933 «94 . 'M^344b7 99.0O48P655 X7 Maximum Standard Compaction Density DEVIA 1 IoNS i 7 153.5453109/ 9. o 3 2 0 7 7 u 7 XI Percent Fines, % X2 Bagnold's Grading Parameter, S total 19 1943.4300079s

R-S uUARl = 0.92099262 S l G M A = 3.00534144

SOURCE SS F Ok X( i ) AlJJ SS IF X( I ) l a s t T FOR HU#rj = 0 B Va l u e s S113 FhRtiR B s i d p Va l u e s

X 0 112.78045645 n . 9 512 R 918 u•u P515213 0 «7618246u X 1 X 1)84.9 3 7744x4 H27«25392l5i 1 1 * 1 71 64 3 14 /04*94£>9,j7l>6 7 ij 4 . 9 4 6 9 3 7 5 0 -8 .83454931 " 2 • 7 0 3 6 3 8 2 sj U • 30i>0u03 2 -0 *6 u?4 5 i «4 W * 2 ro St; T EXPtCTEu OBSERVED [UFFEKF'xrt 1 oft, x * i a j Oft 40909069 -0.01222339 2 1 0 6 .9/9 4 0 o5j 3 07.19999981 *• 0 • 22059556 3 1 ^ 6 ,9 / 0 4 n 3 r ,, in /.19009961 ______-0.22059556______4 114.100218 7.) 115.30000019 -1 . 11978118 N lif t . 1 n m n ii n 3 8 -1 .91978x3/ 6 120.4U3/590 J 116.19999961 4.203/5985 7 1 y n . A ii 3 / n 0 1 22.«9999901 -?.29624-j 12 h 112.2o458u 0 4 i 1 4.5 0 u 0 u 0 u n - 2.2 o 5 4 19 (1 4 9 1 9 n. P 9 8 1> 1 2 0 ? 121.69999981 - 0 . PO l38/4 o m 120.89861202 x24.19999961 - 3.3 (113 6 / 4 3 11 128•1 292343 L 128.29999924 -0.1/0/6444 12 128.1292343 L 1 3 0.5 C 0 0 0 0 U 0 -?■. 37 076521 l 3 13 4- 2 I 9 / 6 d 5 2 i 3 u. 5 0 ii n 0 0 u 0 3.71976968 14 134.21976652 i 31 . 1 0 0 0 0 0 « 8 3.1 19 7 6 9 6 u L5 lu 6 .51450634 l n 6.6 9 9 9 9 9 61 - 0.16 5 4 9 2 6 7 l o 115.122/8936 i 16.69999981 -1 . 5 / /¿I 0 ?o 17 118,88048172 1? 1.30 on no 1.9 -2.41951/70 18 122.39990044 125.60000038 - 3.2 0 fl U 9 9 3 5 19 112.9799537/ 10 / .19999961 5.7/995449 1 d . m n 0 « 4 A 9 9.6 n ;i n (1 n .s 8 ------8.2 5 H 11 n « VI------

D E V i A I I ij N s 9 9 9 9 o 1 5 3 .5 4 5 j 1 u97 I5jo4b3iof>7 o-ououounu .. A U A l Y„b F S ü F V a RIA n T l- iAil I-; , KjE bRL SSI n N An i e n t b , « n u b T A T I SJJ i; S_ OF F I 1 F OK U b P E N O b N 1 V Ah IABlF X 6 I0RN

SulHCb -jF SUM Of bQoA'XEb Mb AN bOuAKE F VALUE

RbGKFSSiON 3 1638.61099240 646.203666o9 36.48940420 Variable Represents M i n i m u m Keiative uensiry Mb. V I A 1 I uN!> iv 2 6 4 . 4 j 7 7 5 6 81 3 4.96682927 X6 X3 Bagnold's Grading Parameter, S. 'n.'j Nb .2 2 1923.01 07^o0t> X4 Bagnold ' s Grading Parameter, ET X9 Bagnold's Grading Parameter, S p expressed as Log (10 Sc)

K-SuUARb = (1.8 9 21 0 0 4 9 SIGMA = 3.80895/13

ÎOURüE Sb F OK X ( 1 ) AUJ bS IF X( i ) LAST T FUR H U S H ( I ) =0 B VALUEb S 1 1) bxRuR B STD B VALUEb

X Ü 96.2614116/ X 3 17 7.6.0 /244 49 410.21120634 -5.23491752 -12.00255221 2 .30424875 -1.31299862 X 4 371.28552246 167.36339356 3.34366947 9.76622033 2. 92080912 Ô.5F4 21916 X 9 1089.7162612Û 1069.71823120 -8.53224254 -26.75926566 3. 13625234 -1.74278162

_____ SL T EXPbCTEu UBbERVLD 0 IFFEHE^Ct 1 94.78265789 9 0 . 0 0 0 0 U 0 u 0 4.7d?85655 2 94.79252148 94.80000019 -n.UU747/90 3 94.7925214 « 95.19999961 -0.40747752 ...... ■■■ ■ ■ 4 9 4.82 0 416511 9b.39V9990? -3.5/958367 6 94.82041550 98.19999961 -3.37958387 ...... "...... ~... ______6 94.86918972 94.60üft00j8 0.28918970 7 94.86918972 96.39999962 -1.51080954 ______8____ .. 110.747/3121 111.80000019 -1.05226849 9 111. 0456162U 113.69999961 -2.65438414 ...... !U 3 11.0 45616 2U 113.600 0 001)0 -2.45438433 11 111.06642605 112.3 0 0 0 U 0 19 -1.24357377 ■ ____L2______11-1.05642605 ______112^69 99 996 2 -1.84357320 1 3 311.091/9020 106.6000U038 2.49179003 ____ 14 . I l l . 0 917.9 U 2 0 108.89999902 2.19179079 L5 104.9^173662 96.15999965 6.76173696 . . 1.6 104.92209301 104.60000038 0.22209328 i 7 1U4,90 315633 1 0 7.5 0 0 0 0 0 U 0 -2.59684384 __Lh______104.92409611 108.60000038 -3.67590424 19 97.612/8534 92.30000019 5.3127852/ ______2 0 96.63166605 89.35u0u038 7.2815656/ 21 92.593/7861 99.19999981 - 6 • 6 U 6 2 ? ü 4 8 . .. 22 ...... 94.41964722 91.100 OU 038 3.31964/64 23 79.46862484 81.10000038 -1.64137518

DfcV I A U uNb99999 264.407/5681 284.40775661 O.OOOUOUOJ ANALYSIS UF VARIANCE TAbLE , KEGRESSIuN CUEKFICIENTS » ANU STATISI ICS Of FI 1 KUR UfPFNDENI V A K I A Bl E X 6 1 O wn

SOURCE JF SUM Of SQUARES MEAN SQUARE F VALUE

V a r ia b le R e p r e s e n t s REGRESSION 2 173 3.5226440 3 856.76132202 81.79257/74 X6 Minimum Relative Density DEVIATIONS 2 0 209.49610329 1U. 47480512 X4 Bagnold's Grading Param eter, D X5 Maximum Relative Density ^ TOTAL 22 1923.01875008 1 — - ■ ■ -

R-SUUARE = 0.8910 587 3 SIGMA = 3.23648036

SOURCE SS FOR X

X 0 -3.96510375 \ __X- 4 63.306882^8 229.91239357 -4.68498510 -6.08513319 L . 2 9 8 8 5 8 6 0 -0.3515534O X 5 1650.21577454 1650.215/7454 12.551550/5 1) . 8 9 5 6 4 3 4 u U - 0 713733 ? 0.94184747

J set EXPtCTEO JBSERVtD DIFFERENCE l______89.2/081163 90.000QUOoO -0.72948807 2 92.75213051 94.80000019 -2.04786962 ------___3______9 2 • Q 3 5 4 5 4 7 5 95.19999961 -3.10454414 4 98.51595306 98.39999962 0.1159540/ . 5...... ______97.62010956 98.19999931 -0.5798895o 6 97.4 7136683 94.5000U038 2.8/136/07 7 99.6 213913,l 96.39V9V96? 3.22139251 8 106.45684147 111.80000019 -5.34315860 9 _____110.72519657 113.69999981 -2.9448 0 28 2 10 113.98023319 113.50U000UU 0.46023359 ------11 ____ 112.90522191 112.3000U019 0.6U522234 — - 12 115.86150452 112.89V99962 2.9635Ù544 1 s 109.59060097 in 8.60 JO 0 038 0.99060086 i 4 111.1 1353493 108.89v9y962 2.21363600 18. 92.1/234993 98.15999965 -5.98764926 — ------..-...... - ...... 18 1U4.1/665100 i 0 4.6 0 0 0 U 0 3 8 -0.42334651 17 ...... 1 l) 5.9 0 8 ...3 3 8 0 ) 107.50000000 -1.53166132 18 107.13293457 108.60000038 -1.4 6 7 U 6 511 1 9 ' 92.94120121 92.30JOU019 0.64120168 2'j 96.44861603 8 9 . 3 5 u 00038 - 7 . U 9 8 616 3 U 96.54645344 99.19 7999o 3 -2.65354601 • ------21 22 91 .8333168 o 91.10O0O03H 0.73331/OV 2 i ...... 06.0400753J 8l.l0o0u038 4.94007564

______DFV..1.A.I HiMS999.V9_____ ?09.4961Qu29 . ______2JL9 j _4 9_£>JLu 3_2_§__ O.OuOOOUflU Unclassified jìecurit^Cl^ DOCUMENT CONTROL DATA - R & D

(Security classification ot title, body of abstract and indexing annotation must be e ^ t 9 r e ^ i. ORIGINATING ACTIVITY (Corporate author) 24. REPORT SECURITY CLASSIFICATION Unclassified U. S. Army Engineer Waterways Experiment Station Vicksburg, Mississippi 2b. GROUP

3. REPORT TITLE

COMPARISONS OF VIBRATED DENSITY AND STANDARD COMPACTION TESTS ON SANDS WITH FINES

4. DESCRIPTIVE NOTES (1Vpe ot report and inclusive dates) Final report

5. AUTHOR(S) (First name, middle initial, last name)

F r ank C. Townsend

6 REPORT DATE 7«. TOTAL NO. OF PAGES 7b. NO. OF REFS

June 1 9 7 2 b 9 n 8a. CONTRACT OR GRANT NO. »4. ORIGINATOR’S REPORT NUMBER(S)

6. PROJEC T NO. Miscellaneous Paper S-72-29

c. Engineering Studies Item ES 516 9b. o t h e r REPORT NO(S) (Any other numbers that may be assigned this report)

d. 10. DISTRIBUTION STATEMENT

Approved for public release; distribution unlimited

11- SUPPLEMENTARY NOTES 12. SPONSORING MILITARY ACTIVITY

Office, Chief of Engineers, U. S. Army Washington, D. C.

a b s r a c t Qenera2.1y, two laboratory test methods, relative density and standard compaction (impact), are used in establishing density requirements for the placement of embankment materials. The relative density test method is specified for cohesionless soils, generally when fines do not constitute more than 5-12 per­ cent by weight, while the standard compaction test is for cohesive soils. However, for sandy soils contain­ ing varying amounts of fines, selection is often based upon the test method considered appropriate for the material. This study was an investigation of various criteria for assisting in compaction test method se­ lection for cohesionless soils with fines. The effects of gradation, percentage and plasticity of fines, and moisture on vibratory and impact compaction of granular soils were evaluated by adding measured per­ centages (9, 16, and 23 percent) of low plasticity (ML) and medium plasticity (CL) fines to a poorly graded (SP) and a nearly well-graded (SW-SP) sand. Maximum density tests using a vibratory table were made on both oven-dry and saturated soil, minimum density tests were made on oven-dry soil, and standard compaction tests were performed on material at various water contents. Test results indicate that a uniform sand, due to its higher void space, can accommodate more fines and densify more effectively than a well-graded sand with fines. Plasticity of the fines and moisture were found to be interrelated factors affecting the compaction of sand with fines. For low plasticity mixtures, saturation facilitated vibratory compaction. Conversely, for more plastic mixtures, adhesion of the fines to the sand grains restricted vibratory shifting of the grains into a denser structure. The same densities are produced by impact and vibratory compaction at higher percent fines added to the well-graded sand compared to the percent fines added to the uniform sand. Apparently, compaction of a well-graded sand with fines is more affected by water content than a uniform sand with fines. Because moisture and plasticity of fines have such opposing effects on impact and vibra­ tory compaction of sandy soils, guidance for compaction test selection is not clear cut. The current prac­ tice of basing -compaction test selection on results of relative density tests on oven-dry materials and standard compaction densities may not be realistic of field conditions and may lead to the untenable con­ clusion that vibratory compaction should be used for sands containing in excess of 20 percent fines. It is recommended that the use of the relative density method for compaction control be limited to granular soils with 12 percent or less fines.

POMI R fP LA C II DO FORM 147». 1 JAN 44. WHICH I» O t I O L K T I FON AWMV U IK . DD « MOV M 1473 Unclassified Security (Classification Unclassified Security Classification

1 4. LINK A LINK B LINK C

ROLE WT ROLE wr ROLE WT

Compaction tests (soils)

Fines

Sands

Soil density-

Unit weight determination

Unclassified Security Classification