Geotextile for reinforce weak soils

[ntuammm]

i am doing a penetration test with the machine shown in this website

http://fka.kuktem.edu.my/images/fka/articles/2323/CBR Machine.JPG

the cylinder metal mould where i put the subgrade material in has an internal diameter of 152 mm and an internal effective height of 127mm. the penetrator comes into the subgrade material has a diameter of 50 mm. the penetration has a speed 1 mm per minute and the capacity of the machine is 50KN. before start the penetration i close the mould with a load of 4363gr as shown in the figure with the cylinder plates.i am doing the test for six minutes taking reading every 15 sec so as to calculate the California bearing ratio for my material. in order to reinforce my subgrade material i use geotextiles( woven and non woven ). i am trying to find which is the best position to have the geotextile into the mould(which is the best depth to install o geotextile during the construction of a road in weak soils). after many tests having the geotextile in different positions( on top, 10mm from top, 25 mm from top, 50 mm from top, 75 mm from top, at bottom) i found that the highest CB ratio( more strength subgrade material) is when the geotextile is 10mm from top. i found that with high and low compaction and with woven or non woven geotextiles. have you got any idea why this happen and if this has to do with the diameter of the penetrator.

 

[msucog]

i am no pavement expert by any means, but i would suggest using caution with that data unless it is field proven. to me, it sounds that your data is testing more the strength of the geotextile rather than that its actual compliment to the material. why wouldn't the results be higher on top then you ask? perhaps because there is 10mm of material above the geotextile, it is providing additional restraining friction. intuitively, based on my experiences in the field, the geotextile does little good near the surface with unstable or poorly compacted material. it is most effective under these conditions at depth and preferrably with aggregate above and below the geotextile. also, the geotextile must extend several feet past the edge of the actual area of concern. for high compaction, your explanation of the results seems plausible.
perhaps one way of looking at the data is compare the high compaction and low compaction results with and without the geotextile. with the geotextile at 10mm, are the results remotely similar? in other words, with the geotextile near the top, are the results similar and then as the geotextile is placed deeper, do the low compaction results drop off more dramatically than for the high compaction results?
performing comparitive testing in the field to see if the higher test results actually provide "better" subgrades may be prudent (especially for the low compaction scenario).
i'm not sure my ramblings will help but maybe it will spark an idea that will answer your question.