Learning Forum

[zain.qazi]

Hello. 

I am simulating a steady state Nitrogen High Aspect Ratio Cooling Channel using the Eulerian Model with Non Eq RPI boiling Model. The model was set up gradually bringing the wall temperature up from 100 K to 360 K. The target wall temp is 463 K. The model had initial convergence problems as well but these were resolved by using a combination of Interphase Force Models that allowed it to work smoothly. The model works very well with low residuals and converging monitors up until 360 K wall temperature. Raising the temperature further leads to high residuals and non converging monitors. It is worth noting that every ramp up in the wall temp raises the residuals but after 360 K (e.g 380 K) they're flat out high and pressure fluctuates showing no sign of convergence. 

Boundary Conditions: mass flow inlet  0.00425 (liquid) 80 K, pressure outlet 21.4 bar, constant temp outer wall target 463 K. 

This is a validation of an experiment and numerical case.

What could be the reason that the model loses its ability to converge after working fine between 100 to 360 K? 

[Prashanth]

Hello: 

The convergence difficulties with wall boiling models usually has to do with low y+ (with quenching correction used with it). Can you make the mesh coarser if that is the case..

[zain.qazi]

The y+ is already at 40 and rising as we move along the channel. The model stabilized when I switched the drag force from Ishii-Zuber to Schiller Naumann. The model worked upto 400 K and then the same problems. I added Enhanced Wall Treatment to my k-epsilon model and it allowed me to go all the way to my desired temp of 460 K. The problem is now the Lift force. I chose to take Fluent User Guide's tip and solved the case without Lift and Virtual Mass Force temporarily but adding any of the Lift Correlations now gives very non converging results. 

Physically, I am expecting to see a drop in wall temperature where the boiling starts to happen (at around 1/3 of the length of the channel) but there is no drop. And my guess is because the bubbles near the wall do not rise to make way for liquid to take its place and boil. This can be due to no lift force or the Wall Lubrication force not doing its job. Here is a contour of the gas vol frac

[Prashanth]

It looks all red on the right, which is mostly vapor. Maybe you need to use critical heat flux model to better capture this regime..

I agree that a wall temperature drop is expected in such cases. But also, there are too many interaction options to comment on.We already have two wall boiling verification cases (RPI - VMFL039 and critical heat flux - VMFL067) in the Fluent verifications manual with literature comparisons. You can check them for some pointers.

[zain.qazi]

If I am expecting a drop in wall temperature due to a rise in heat flux this shouldn't be a CHF case. But even plotting the Total surface heat flux or the Evaporation heat Flux on that heated interface shows that both of them are very low. Why could that be? Is it a body forces problem that is not capturing the evaporation happening? 
I see you have used growth layers in the RPI boiling verification case which I was suggested not to use by one of your moderators for wall boiling. How crucial are they? 
Secondly, you dont use wall lubrication at all. But plotting your wall temperature axially also shows that the temperature plateaus and the boiling does not really decrease the wall temperature. Was that the expected result?