[JohnnyOnTheSpot]

Hello,

I am hoping someone here can help me solve a convergence issue I am having with my flow simulation of an engine's intake port.

The simulation runs smoothly for approximately 1200 cycles until my energy residual diverges.

I have also noticed that the air temperatures do not seem to make sense, so I believe this is where my energy issue is coming from.

Any help appreciated!

[JohnnyOnTheSpot]

Well I spent my day troubleshooting the thermal issue I was seeing and was able to get to a more reasonable solution by changing my initial conditions.

However, I am still seeing my energy residual take off after about 1000 cycles while all other residuals hold steady. My mass-flow monitor also levels off, although the solution is a bit higher than what I observed experimentally (approx 20% error).

Can anyone explain what could cause my energy residual to increase after so many cycles? I am a relatively new CCM+ user and am not sure what to look for. I searched around on both this site and the Steve portal and wasn't able to get a clear answer.

Thanks!

[JohnnyOnTheSpot]

Well I spent my day troubleshooting the thermal issue I was seeing and was able to get to a more reasonable solution by changing my initial conditions.

However, I am still seeing my energy residual take off after about 1000 cycles while all other residuals hold steady. My mass-flow monitor also levels off, although the solution is a bit higher than what I observed experimentally (approx 20% error).

Can anyone explain what could cause my energy residual to increase after so many cycles? I am a relatively new CCM+ user and am not sure what to look for. I searched around on both this site and the Steve portal and wasn't able to get a clear answer.

Thanks!

[Dry Patrick]

It seems to be a mesh problem, I spot a too high energy residual in the firsts iterations (the residuals should be always under 1) what are you setting on prism layers? Whats the wall Y+ distribution?

Try to use this tool after 100/200 iteration in order to capture the problem before the divergence.

[JohnnyOnTheSpot]

Originally Posted by Dry Patrick

It seems to be a mesh problem, I spot a too high energy residual in the firsts iterations (the residuals should be always under 1) what are you setting on prism layers? Whats the wall Y+ distribution?

Try to use this tool after 100/200 iteration in order to capture the problem before the divergence.

Thanks for responding - I will take a look at those settings later this afternoon when I have a chance and will try that diagnostic tool.

I do know that I was able to get high mesh quality from my CAD model, but it required a very fine mesh (0.5 mm if I remember correctly).

I am relatively new to the CFD world - would you mind if I walked through my physics, mesh, and solver choices and settings to see if you can spot my mistake?

Thanks

[Dry Patrick]

Originally Posted by JohnnyOnTheSpot

Thanks for responding - I will take a look at those settings later this afternoon when I have a chance and will try that diagnostic tool.

I do know that I was able to get high mesh quality from my CAD model, but it required a very fine mesh (0.5 mm if I remember correctly).

I am relatively new to the CFD world - would you mind if I walked through my physics, mesh, and solver choices and settings to see if you can spot my mistake?

Thanks

A very fine mesh do not means much and doesn't ensure alone high quality results, for example if you have too many prisms layers with near to 1 y+ value you will need to activate the low y+ wall function, than 0.5 mm is the minimum or maximum size? Of the surface mesh or of the volume one??

Please report all.you setting on mesh and physics, and some screenshots of surface and volume mesh (a section will be preferred)