DEFENSE DOCUMENTATION CENTER DEFENSE …

1 UNCLASSIFIEDAD 631 848 DEADMAN ANCHORAGES IN VARIOUS SOILMEDI Smith, et Civil Engineering LaboratoryPort Hueneme, CaliforniaApril 1966 Processed DOCUMENTATION CENTERDEFENSE SUPPLY AGENCYFOR FEDERAL SIENTIFIC AND TECHNIDL INFORMWAIONU. S. DEPARTMENT OF COMMERCE / NATIONAL BUREAU OF STANDARDS I INSTITUTE FOR APPLIED TECHNOLOUNCLASSIFIED*~ R434e''TechnicoI Report DEADMAN ANCHORAGES SOIL MEDIUMSA pril1 1966 BUREAU OF YARDS AND DOCKS* ;~~\I/, ~ U. S. NAVAL CIVIL ENGINEERING LABORATORYq Port Hueneme, CaliforniaDistribution c d" Aa- n is -cin -eFOR FEDERAL SCIENTIFIC ANDTECHNICMd INFORMATION ftrdeopy Wi~crof icheI3~flJfl~EE (ppIDEADMAN AtN'CHORAGES IN VARIOUS SOIL MEDIUMST echnical Report E.)

2 Smith and J. V. StalcupABSTRACTA test program was conducted to investigate deadman anchorage holdingcapacities under applied horizontal loads. Deadmen fabricated of concrete andranging in face area from 5 to 72 square feet were tested in depths of embedmentfrom ground level to 7 feet. The deadmen were pulled both singly and in groups ofthree, in sand and in two soils with cohesive characteristics. The test program alsoincluded tests on a model applied load versus horizontal displacement relationship exhibited abasic recognizable farm for cdl conditions of tests.

3 By graphic analysis, a seriesof reaction-pattern curves was developed relating deadman holding power in eachcohesive soil to thrse factors: deadman face area, depth of embedment, andwhether the deadmen were embedded singly or in a group. The results of the sandtests which were described in a previous report were converted from the previousanalysis to a compatible form and presented with the cohesive soil test curves provide an empirical means for determining deadman holding capac-ities at different amounts of displacement within the range of conditions investigation disclosed that multiple anchors develop a higher holdingcapacity per net area than a single deadman with the same total face area.

4 Theincrease in holding capacity ranging from 5 to 209/6 depends upon such factors asdepth of embedment, the type of soil, and the spacing between deadmen. Undermost test conditions, up to a 30% increase in holding capacities was attained incohesive soils as compared to sand, but 2 to 3 times the horizontal displacementwas required to achieve the maximum holding of this document is . available at the Clearinghouse (CFSTI) $ Laboratory invites comment on this report, perticlarly on theresults obtained by those who hove applied the NTRODUCTI ON .. ITEST PROGRAM.

5 1 History .. IFramework of Tests .. 2 Description of Deadmen .. 5 Soil Summary .. 5 EQUIPMENT AND PROCEDURES .. 11 Full-Scale Test Apparatus .. I IFull-Scale Test Procedures .. 11 Model Test Apparatus .. 19 Model Test Procedures .. 21 RESULTS .. 24 General .. 24 Analysis .. 24 DISCUSSION .. 30 Application of Results .. 30 Computer Program Potential .. 40 CONCLUSIONS .. 41 ACKNOWLEDGMENT .. 41 APPENDIXESA -Observed Data Points and Empirical Curves for theFull-Scale Tests .. 42B -Mathematical Analysis .. 58 REFERENCES .. 70 INTRODUCTIONThe use of deadman anchorages to provide lateral support for structures suchas quay walls is an accepted engineering practice.

6 High quay walls must remainverticai for ship-docking purposes while containing and supporting large quantitiesof fill material behind them on the dockside. Theoretical computations for theholding power of deadmen generally are based on equivalent fluid pressure or onan assumed failure of a soil mass along a shear plane. But the theories do notreadily account for the ioad- characteristics of U. S. Naval Bureau of Yards and Docks sponsored an investigation,Task No. Y-F015-15-01-010, conducted by the U. S. Naval Civil EngineeringLaboratory at Port Hueneme, California, to determine deadman reactions underapplied horizontal loads.

7 The investigation has encompassed three separate phasesof testing. In Phases I and 2, full-scale tests were conducted in sand in whichdeadmen were pulled singly and in rectilinear groups of three; the results werepreviously reported. 1,2 In Phase 3, full-scale tests were conducted in two cohesivesoils with test conditions similar to those in sand; model tests were included. Thisreport describes the overall investigatiun; the results obtained from the cohesivesoil tests are presented along with the results previously obtained in the sand PROGRAMH istoryThe three separate test phases of deadman anchorages were conducted atintermittent times over a period of years.

8 Phase I included tests of two sizes ofdeadmen in sand and in depths to 3 feet. Subsequently, Phase 2 included tests ofdeadmen of seven sizes in various depths to 7 feet. The results of the Phase 2 testswere combined with those of Phase 1. By means of computer programming, aregression analysis of the data was performed and an empirical equation was devel-oped relating holding power at specific horizontal displacements to three variables:depth, face area, and whether the deadmen were single or in a group of three. Bymeans of a nomograph based on this equation, it was possible to rapidly determinedeadman holding capacities for different amounts of horizontal deadman displacementsin sand.

9 2 The encouraging results obtained from the deadman tests in sand promptedPhase 3 of the program in which deadmen were tested in soils with cohesive charac-ieristics. The intent was to combine and analyze the results of all three phases ina manner that would enable the accurate determination of deadman holding powersat various displacements, while considering such variables as soil characteristics inaddition to the variables which were accounted for in the prior programs. To thisend, Phase 3 was designed so that a minimum number of tests could be integrated withthe previous tests to give a maximum amount of information.

10 In addition, model testswere incorporated into the third phase for two purposes: first, to investigate thevariables affecting deadman performance over a wider range of conditions than waspractical to investigate on a full scale; and second, to investigate the effect ofadditional variables not practical to investigate on a full of TestsFull-Scale Tests. Fifty-seven full-scale tests were included in the programcovered by this report. Each test represented a separate set of conditions relating tosuch factors as soil type, depth of embedment, face area of deadman, and whetherthe deadmen were pulled singly or in multiples of three.