Learning Forum

[Vratislav Košek]

Hello everyone,

I would like to ask for advice. I have problem simulating an explosion using the LBE coupling to the MM-ALE method. I can t get the shock wave to pass throught the ambient layer to the ALE domain (air). I have no idea what the problem could be. I would appreciate any advice  or link to working keyword deck for this method.

Thank you

[Jim Day]

I ran a similar test case and the pressure in the ambient ale elements does transmit to the rest of the ALE domain. Below are some guidelines. It appears you've correctly followed nos. 1 and 2. Check nos. 3 and 4. If that doesn't help, tell me which version of LS-DYNA you're running. 1) A one element thick layer of the ALE air domain which faces the blast source is given a new part ID. This new part is assigned material properties identical to those used in the rest of the air domain. In the *SECTION_SOLID definition of this new part the ambient type is set as AET=5. 2) The segments of the part described above which face the blast source are treated with *LOAD_BLAST_SEGMENT(_SET) and in that keyword IALEPID is set equal to that part ID number. 3) Currently the feature works only for air modeled using the ideal gas law with *EOS_LINEAR_POLYNOMIAL. The equation of state should be set up so that its initial pressure is equal to atmospheric pressure. 4) PREF in *CONTROL_ALE should be set to atmospheric pressure.

[Vratislav Košek]

Hello,

thank you for your reply. I am using the version ANSYS 2020R2 LS-DYNA Student 12.0.0 As for your advice. I think everything is fine. I am attaching images of all keywords used in the model. Maybe I´m just overlooking a basic thing... I can send the whole k-file if it helps.

Thank you

[Jim Day]

Your simulation runs to Normal Termination? To confirm whether pressure transmits into the non-ambient air, overlay a pressure time history of a blue, ambient element with a pressure time history of a nearby red, non-ambient air element. Post that image.

[Vratislav Košek]

Yes, the simulation ends with the normal termination. Here is a comparsion.

Thank you.

[Jim Day]

So you're concerned about the difference between the pressure loading on the surface of the ambient layer and the pressure seen in the ambient element. As an experiment, remove the *LOAD_BLAST commands and apply the pressure from blstfor directly using *LOAD_SEGMENT_SET. Do you get the same response?

[Jim Day]

I think what you've plotted from blstfor is "P=(Pi,Pr,theta)". Note that blstfor also includes "Pi" or "incident pressure". I think that incident pressure is what is applied to the ambient air elements in the coupled ALE approach. A number of papers on the subject of *LOAD_BLAST_ENHANCED may be found through dynalook.com or simply by Googling "*LOAD_BLAST_ENHANCED".