Huge Temperature fluctuation

Hi, all,

I tried to calculate the thermal conductance of a system. I applied nvt and changed to nve. Then I put heat source/sink at both sides.
As a result, the group temperature didn’t rise or fall, but the fluctuation in group temperature is huge (about 40K). This is a small system with only 120 in total and 30 atoms in each group. Is this why I got that big fluctuation? Is there some way I can reduce the fluctuation?
Thank you very much.

Yuan

There is a simple way to find out: try a larger system and see if your results get any better.