PERMFILE108040 - Laserfiche WebLink

Home Browse Search

- 7- • strain. I*_ has long been known that a slope stability failure is a plane strain problem and that ideally the strength of ttre material in plane strain shouid be used in analysis. Because of the lack of equipment and convenience, most data on strength is obtained frc~n triaxial testing. blork done oy Lee (9) has mown That there is a significant difference be- t~~:eon the strength obtained from a triaxial test and a plane strain test an granular materials. This ratifier extensive wore indicates that the angle of internal friction obtained from plane strain testing can be up to 3 degrees higher than that obtained from triaxial tests. ble agree with the commentor that "ideally, the appropriate soil testing... shouid be conducted to decrease the necessity for assumption...". However, we Mould like to add that inappropriate soils testing will not only be wasteful tut also misleading. As indicated by Sowers (13), "A realistic test of a rockfill must meet two reouirements: (1) the material tested should be the same as that of the prototype embankment; and (2) the stresses imposed by the testing should simulate these in the dam." The large scale triaxial tests do not meet the first requirement and meet the second re- quirement only partially. Lops (10) indicates that the data available are sufficient only to indicate trends. hlarachi (11) states "the engineer in charge of the design of a rock-'ill dam s•rould have no logical and conclusive means to assess the strength and compressibility of actual rockfill materials on the basis of laboratory test results which are usually performed on the materials consisting of much smaller particles." These comments and our experience indicate the use of standard testing techniques is hardly accept- • able ror spoil material. The heterogeneous character of the spoil as com- pared with the uniform characteristics of the rockfill materials tested ;hake it e~!en more dlffi~Ult to obtain strength data from standard testing. The letter mention:, the tests and results of the r•rork done in Kentucky by riuang ;;) as an example of data. in our opinion, there are several pitfalls that must be examined carefully prior io using such data. A fe';I OT =;;252 CG.^,SiGc:"c~i 0ii~ InCIUdc: The samples used in the laboratory testing were all obtained approximately "...1 foot belov; the surface...". It is hardly an acceptable procedure to model the overall strength of the spoil pile r`rcm tests taken at a depth of i fool. .45 a matter of fact, materials which are very near the surface are normally the most weathered and least representative of the fill. 2. Tire use of standard test data is not acceptable from an anaineer- ing viewpoint. The character of different materials, as shoti•m by she test da*_a presented can provide widely aifferir.g results. Tile data presented shows variations in cohesion of 0 to 760 psf and friction angle variations of 15.5 to 36.6 degrees. lJ