Acceptable %mass imbalance for turbomachinery
 

I am running a steady state simulation for centrifugal compressor. The mass and momentum residuals have reached 10^-4 ,but there is about +5,-5% mass imbalance.Should I consider this as converged?

Do a sensitivity analysis and find out if it is significant in your case. So do a another simulation where you converge much tighter and see if there is any difference to things of importance to you.

In general, the acceptable mass imbalance is 0. (machine round-off). Unless you are having severe convergence issues, the mass imbalances in ANSYS CFX are usually the lowest of all the equations.

Are you running a single-row compressor or a multi-row machine? Where is the mass disappearing? at the domain interfaces?