There are many factors affecting the permeability coefficient (k) of cohesive soil

and different test methods have great influence on the test results. To solve the problems existing in present test methods

a hollow cylindrical torsional shear apparatus (HCA) was used for seepage tests. The significance

feasibility and experimental procedures for HCA seepage test were demonstrated

and remolded kaolin was used to carry out a series of experiments to explore the standard rules for permeability tests. Finally

a preliminary study of the dynamic test was explored. The results show that: (1) If 3 600 s is taken as the unit of time

the difference in average permeability coefficient in the unit time is less than 2% in continuous 24 h. Therefore

it is accurate to take the average of permeability coefficient obtained in this stage as the final permeability. (2) The hydraulic gradient range in HCA permeability test for kaolin is suggested between 10 and 15 after comprehensive consideration of test efficiency

comparability

and the accuracy. (3) After 100 000 cycles of dynamic traffic load

the permeability coefficient of kaolin is 12.79% lower than that of the same axial strain in static compression test

indicating that it is important to study the influence of dynamic characteristics on permeability coefficient.