Hi all,

Recently I have being working on the coupled Eulerian - Lagrangian modeling involving a thin shell tank filled with water under blast loading. The model set up is shown in the picture at the bottom.

The tank is empty or half filled with water, the idea is to try to simulate the contact between the air, water and the shell wall under the blast load P(t). The box is fixed at the bottom and the contact is empty or friction coefficient = 0.

I have created six cases to examine the interaction between the Eulerian and Lagrangian parts, as shown in the picture. In the first six cases, the models were all good and the simulations were fine. However, in the last case (case 6) when the Eulerian part outside the box was filled with air, the whole model just failed. 

From the previous five cases, I presume that the CEL is working fine when the Lagrangian part is in contact with one Eulerian material on two faces (case 2 and 3), or two materials on one face (case 5). It will not work when the Lagrangian shell is in contact with two materials at its two faces respectively (case 6). 

I am not very convinced about this and thought I might did something wrong in case 6. Does anyone have any ideas about the last case? I really want to finish the case 6. 

CAE manual 27.4 shows a very similar example to my model, but the Eulerian part outside of the tank is void. I was wondering if it suggests that the space outside the tank has to be void. 

Thanks a lot！