Grid independence

Hi all,

I am simulation hydrogen absorption in a metal hydride using heat transfer in porous media and laminar flow physics. The model is converging fine but when I run on normal and finer mesh I get the same results. However, when I run on fine mesh I get slightly different results. To be precise, the fine mesh size shows pockets of intense absorption whereas the other two do not. Although the average is the same for all three cases up to three significant digits. Any inputs on why it is happening and what can I do to correct it?


Thanks,
Mahvash