Residual becomes unstable when second order upwind is set
 

Hello,

I was trying to simulate a 2D symmetric problem in fluent. However, I found a problem when I set second order upwind, I started the simulation with first order upwind, then when I achieve 1e-5 in residuals I set to second order upwind but I have the problem showed in the following figure. Can anyboy help me with this?

Thank you in advance.

hi
the trend is converging, so i think it is ok

Yes, it seems like the trend is convergin but it will need many more iterations and I do not know if the accuracy of the solution will be affected by this unstability. Does anybody know how could I alleviate this affect?

The problem is a supersonic problem, with air as ideal gas using a density based approach.

Thank you in advance and regards.