This paper presented the device of determination of the changes of the cross-section area of the samples with non-lubricated end platens tested in triaxial tests. This device enabled also the determination of the average porosity of the samples changes and changes of these porosity in the zone of the biggest deformation. Investigations have been carried out on the saturated sand samples in the drained test. The measurement of the lateral strains of the samples (38 mm diameter) has been made with application of 5 celluloid tapes with the millimetre scale. They were placed around the perimeter sample’s in eagle intervals. This method gave in practice similar accuracy of the cross section area as the device of calculation of this area from the measurement of volume change. The average area of the same sample, calculated from the measurement of the perimeters in 1/3, 1/2 and 2/3 of the height sample’s, and in 1/3 and 2/3 of this height are smaller by 0.6%. This error of the results in inaccuracy of index void ratio smaller than 0.01. This is two time bigger, if the cross section area is measured in the middle sample’s height only. The changes of porosity in medium zone of sample, when h/D > 2.0 are comparable with the changes of porosity sample’s on the lubricated end platens. For the calculation of index void ratio changes in this zone, the end platens constant area has not bean considered.

27-36