Learning Forum

[Joerg]

Hello. I am trying to simulate a cyclic shear test, to get the material response in form of a hysteresis loop (shear stress vs. shear strain). I am using Ansys Workbench R24 LS Dyna, with the MAT_079 Soil Hysteretic material model on a single solid element. My main questions at this point are:

1) It appears that the available user defined results in the LS Dyna module are very limited compared to a Static Structural simulation in Ansys Mechanical. E.g. in LS Dyna, for strain the only available option to me is EPL, while Static Structural has a variety of strain output options. Is this really the only strain output I can get?

2) The backbone curve for MAT_079 applies for a given reference pressure that I specify, and the elastic moduli are then pressure sensitive. I am wondering how to best simulate this initial pressure condition. My approach was as follows:

• My elastic properties are based on shear and compression wave velocity (Vs & Vp) measurements at a given depth.
• I have created a Static Structural model of the same one element, and applied a vertical stress coressponding to the overburden pressure, and side pressures (on 2 sides, with nodes on opposite sides restrained) corresponding to the at rest earth pressure based on the poisson's ratio calculated from the Vs/Vp ratio.
• From this simulation, I have extracted the volume change due to this stress state, since the pressure in MAT_079 is calculated based on volume change (from the material manual).
• Then I have selected a reference pressure such that given the volume ratio (extracted element V above / initial element V) results in a calculated pressure equal (very close) to the reference pressure, using the formula given in the LS Dyna material model manual.
• This is the reference pressure I have entered in the MAT_079 material card, and I am using this Static Structural analysis as pre-stress condition. I can only use the 'displacements' mode,since 'material state' gives me an error.

Does this make any sense ? Or could I do this any easier ? My thought was that LS Dyna will consider the volume change in the element due to the displacements from the initial stress condition in calculating the pressure in the element... ?

3) If you have a detailed example on this topic I can follow would be appreciated. I have found many articles that do this and show hysteresis loops compared with lab testing, but I wasnt able to find an example with a description detailed enough that I can 100% recreate.

Any advise is much appreciated. Thank you.  

[Joerg]

 

I also wanted to mention that the data I used to create the input backbone curve is based on cyclic shear lab tests that were completed under the same vertical stress coressponding to the depth of overburden as the Vs/Vp measurements.

I was able to get the strain output in the worksheet (had to say 'yes' in the analysis settings 'output controls'). However it is unclear to me which strain & stress components I should plot for the shear stress vs. shear strain diagram (to compare with the MAT_079 input curve). E.g. what confuses me is that I do get results for EPTOXY, but EPELXY & EPPLXY both show zero. I have fixed the bottom nodes of the element, and applied a displacement in X-direction, holding both Y & Z direction constant.

Appreciate any help.