I have a domain which is seperated into almost 10 subdomains, each subdomain has different momentum source, and each subdomain may or maynot be continuous (i.e. all elements are linked). i realised that my simulation is slowed down considerably as compared to 3 subdomains.

With abaqus elements with similiar mechanical properties can be grouped together and we have grouped together over 100 differeny properties with success. Each elemenet is just given a different stiffness matrix.

unfortunately, in CFX it is not so easy. can anyone explain why? and how can i get around my problem?>

Dear Jemteo,

Would you mind explaining what you are trying to do? You have not said what your problem (physics) is, and your approach seems to have problems. More information will guide others on how to help you..

When comparing with Abaqus, are you using meshes as similar sizes?

Thanks,

Opaque

im simulating a flow into a chamber which has different permeability values at different regions. so i created the whole chamber and each region of different permeability values are a different sub-domain and assigned their permeability accordingly.

opaque: thanks for your reply, could u tell me why is my approach wrong?

Dear jemteo,

I am not sure what you are observing.. I just added 8 subdomains to a previous case, and compare the Discretization time per iteration of the Momemtum and Mass system and I could not see any major difference.. That is, though the time increased it was not something to be concerned about..

what times are comparing? slower than xxx?

Regards,

Opaque