The length of the footingwas made almost equal to the width of the tank in order to main-tain the plane strain conditions. The two ends of the footing platewere polished smooth to minimize the end friction effects. A roughbase conditionwas provided by using rough sandpaper on the baseof the model footing. The load is transferred to the footing throughthe ram and the ball bearings, located on the footing as shown inFig. 3. Such an arrangement produced a hinge, which allowed thefooting to rotate and move horizontally freely as it approachedfailure and omitted any potentialmoment transfer fromthe loadingram.2.2. Test materials2.2.1. ReinforcementsThe geogrids used in this study are shown in Fig. 4. Both geogridand grid-anchor were made of high-density polyethylene. Thegeogrid tested in this study was CE 131, the same geogrid tested byLee and Manjunath (2000). It has a mesh aperture size of27   27 mm and a maximum tensile strength of 5.80 kN/m. Grid-anchor has also a mesh aperture size of 8   6 mm and a maximumtensile strength of 5.8 kN/m.Width of reinforcement, b; number ofreinforcement layers, N; length of reinforcement, L; distance to thefirst layer of reinforcement, u; and distance between the rein-forcement layers, h used in model tests are illustrated in Table 1.Other relevant properties of the geogrids are given in Table 2.Basically, grid-anchor has great pullout strength than thecommon geogrid. The basic difference between common geogridsand grid-anchor is existence of short anchors attached to the geogridon one side which provides great pullout strength for grid-anchor.2.2.2. SandThe sand used in this research is medium to coarse, washed,dried and categorized by particle size. It is composed of rounded tosub-rounded particles. The maximum and the minimum dry unitweights of the sand were found to be 19 and 13.4 kN/m3. Theparticle size distribution was characterized using the dry sievingmethod and the results are shown in Fig. 5. The uniformity coeffi-cient (Cu) and coefficient of curvature (Cc) for the sand were 7.78and 1.24, respectively.The moisture content of the fill sand was kept at about 0%, drycondition, over the testing period. Model sand slope was con-structed by pouring sand and compacting in layers every 5 cm togain a uniform compaction inside the embankment.To achieve consistent soil densities and inclusion conditions inthe reinforced soil models, procedures of construction were care-fully controlled during model preparation. The relative densityachieved during the tests was monitored by collecting samples insmall cans of known volume placed at several locations in the testtank. The pouring-compacting technique adopted in this studyprovided a uniform relative density of approximately 70% witha unit weight of 16.9 kN/m3. A series of direct shear tests was carried out to assess the shearstrength properties of the sand using specimens prepared by drytamping. The estimated internal friction angle at the relativedensity of 70% was approximately 38 . Other parameters of the soilare illustrated in Table 4.2.3. The test setup and programsModel sand slopes were constructed 500 mm in height and1000 mm in length with a slope angle of 32  by pouring andcompacting of 50 mm of air-dried sand layers to cover the entirearea of the test tank. Using this method seemed to be more reliableto practice. The proposed testing geometry of the slope was firstmarked on the transparent glass walls for reference. The proce-dures for the construction of reinforced model slopes are differentto those of other researchers like Selvadurai and Gnanendran(1989), Yoo (2001), El Sawwaf, 2005 and Lee and Manjunath(2000). As shown in Fig. 6 the sand was deposited and compactedin layers up to a desired height. The geogrid reinforcement wasthen placed on the compacted level surface. The sand pouringprocess continued until the pointed height of the slope wasreached, leaving a fill thickness (u) over the reinforcement layer.The length of the reinforcement (L) was not varied and at any givenposition was located such that it extended to the face of the slope.The model footing was then placed at a specific location on thesurface of the compacted fill. In this method there is no need forexcavation, which had been doing by other researchers, Yoo (2001),and seems to be more reliable to practice.Many model tests in different test programs were carried out.First, the response of the model footing constructed on the unre-inforced case was determined. Then, two series of tests were con-ducted to study the inclusion effect of the geogrid and grid-anchorlayers on the footing behavior. Tests were conducted to find out thebest location, type and number and configuration of the geogridlayers that give the maximum influence in footing response.Each series was conducted to study the response of oneparameter while the other variables were remaining constant. Thevaried conditions include the number of geogrid layers (N), type ofgeogrid layers, vertical spacing between layers (u) and verticalspacing and depth to topmost layer of geogrid (h). Table 3 showsthe test program conducted for geogrid reinforced slope for thisstudy. The testing program for grid-anchor reinforced soil was alsosimilar to that for geogrid in addition one unreinforced test wasconducted. The sign ‘‘gg’’ refers to geogrid and ‘‘ga’’ refers to grid-anchor. 承载力的立足点在斜坡带砂与土工格栅英文文献和中文翻译(2):http://www.youerw.com/fanyi/lunwen_56226.html