Dear Rahul Kumar, If we use the mass scaling, we are artificially increasing the density of the model under consideration. That eventually would give us wrong result. Is it right? if so how to decide an appropriate mass scaling scheme. can you suggest me any material on mass scaling..

You are saying with fixed increments the solution obtained may be inaccurate. but the fixed increment is calculated based on the minimum element length, density and young's modulus of the model right.. how could it be the solution inaccurate. Its a question with immense curiosity .. I would like to learn and know the consequences of using mass scaling in Abaqus explicit..

Mass scaling is given as a factor, so 2.35 corresponds to 235%. Depending on which regions it is in this could effect the physics of your simulation. Cheers.

Dear Rahul,

first of all, thank you for your proposal but can you explain why would fixed increment lead to inaccurate results because I don't have large deformations and my finite elements are not severely distorted so I don't see the advantage of using variable mass scaling.I used small mass scaling, less than 0.01% because I have few elements on different areas that I cannot mesh correctly.

Dear Harnoor, I need to disagree with you because I'm not talking about scaling factor but about DMASS(percentage of mass) so this is not a factor as I understand.

Hi Simun, I just checked the manual, you are right the value reported in the *.sta file is the percentage value...

Regarding the fixed-time increment... it could lead to non-physical results IF the STI drops below the fixed-time-increment (FTI). But I think abaqus will throw an error when the STI is "close to" the FTI and will abort if STI < FTI. On the other hand, you may be needlessly slowing down your simulation as Abaqus has a few "tricks" to scale up the STI. But I have also used FTI a few times, especially with VUMATs where auto time incrementation leads to stability issues.