The E50 in the hardening soil model（HS）when using the "soiltest" in PLAXIS 2D is different from the input value

Question

Dear all, 
 When using &ldquo;Triaxial test&rdquo; of the &ldquo;soiltest&rdquo; option together with the hardening soil model in PLAXIS 2D, input the parameters as Figure 1 and get the parameter E50 from Figure 2 according to the definition in Figure 3, then it can be found that the calculated parameter(17.88MPa) is less than half of the input value(40MPa). However, in the light of the PLAXIS &ldquo;Material Models Manual&rdquo; [1] , two values should be the same. Why does the difference happen? Figure 1: Parameters setting and calculation results 
 
 Figure 2: stress-strain relation 
 
 Figure 3: Hyperbolic stress-strain relation in primary loading for a standard drained triaxial test [1]

Reference 
 [1] PLAXIS. Material models manual[R]. 2020,69-69.

Dennis Waterman · Answer

Hi, 
 The difference is caused by the very high cohesion you used. The E50 definition is in principle for cohesionless soils. However, if you check the equation for stress-dependent stiffness you will see that the cohesion is part of the equation, which is done to avoid negative stiffness if tensile stresses occur in cohesive soils. However, if the cohesion is significant compared to the stress state the E50 will deviate because of the large influence of the cohesion in the stress dependent stiffness equation. If you would for instance make the cohesion 1 kPa you will see that the E50 according to the graph is about 40 MPa. Usually one shouldn't use such high cohesion relative to the stress state, unless the soil is cemented. And for cemented soils, sand stones, etc. one would have to calibrate the parameters afterwards. So if your labtest gives an E50ref = 40MPa, you may have to use a higher value in the HS model to obtain a good match with the laboratory test results. 
 With kind regards, 
 Dennis Waterman

The E50 in the hardening soil model（HS）when using the "soiltest" in PLAXIS 2D is different from the input value

Top Replies