[SHIKHA BHUYAN]

Hello Friends,

I am working on nozzle simulation (inviscid flow). The flow inlet is subsonic while the outlet is supersonic. For this, I drew 2D nozzle and sliced the nozzle into two zones. I went for hybrid initialisation and wanted to patch the variables of the two zones (i.e subsonic and supersonic).
In the fluent settings I followed:
Hybrid initialization -> Initialise -> patch
When I select 1 zone and give the inputs and then after selecting the 2nd zone and then giving the inputs there update the previous values. I am not sure whether I am in a right path or not. If anyone can suggest me any solution for this matter or give me some alternate idea. I shall be highly obliged.

[LuckyTran]

Sounds like you're doing it right. Select the zone you want, select the variable, enter the new value and hit patch. Do this for each zone and you're good to go.

It doesn't matter what you initialize with since you're going to patch it anyway. Patching overwrites the current values with the new value that you set.

[SHIKHA BHUYAN]

Thank you so much for the reply.
It means it does not matter whether the new value overwrites the old value!! I will anyway initialise??

[LuckyTran]

Originally Posted by SHIKHA BHUYAN

Thank you so much for the reply.
It means it does not matter whether the new value overwrites the old value!! I will anyway initialise??

Your response is confusing me so just in case:

Initializing and patching are two separate processes (even if they are mechanically similar). Initializing overwrites all of the existing variables with the settings from the initialize tab. If standard initialization you manually input all the initial conditions. If hybrid, they values are calculated automatically.

Patching is similar to initialize but gives you more control. You can patch specific variables on specific zones whereas initialize is all or nothing.

If you initialize and then patch, the patched variables will overwrite the initialized variables. If you patch then initialize, the values from initialization will overwrite the patch. Order does matter. Question is which values do you want?

[SHIKHA BHUYAN]

Hello Sir,
I apologize if I am asking you silly question as I am nascent in this field.In the present work the nozzle inlet is subsonic field and the outlet is supersonic. Hence, I have initialized both the zones with different values (simultaneously, slicing the surface into two zones).
For this I hit the initialize button which activated the patch button.
According to your suggestion, I entered the values and hitted patch tab simultaneously for each zones.
Now the value space is displaying the latest values I entered(overwritting the previous value given for subsonic region). Sir, my question is that,whether the system is having both the values (subsonic and supersonic region, as the values are overwritten) and if not then how it will get to know the values of subsonic region.
How does the simulation work then??

[Tosbaga]

Originally Posted by SHIKHA BHUYAN

Hello Friends,

I am working on nozzle simulation (inviscid flow). The flow inlet is subsonic while the outlet is supersonic. For this, I drew 2D nozzle and sliced the nozzle into two zones. I went for hybrid initialisation and wanted to patch the variables of the two zones (i.e subsonic and supersonic).
In the fluent settings I followed:
Hybrid initialization -> Initialise -> patch
When I select 1 zone and give the inputs and then after selecting the 2nd zone and then giving the inputs there update the previous values. I am not sure whether I am in a right path or not. If anyone can suggest me any solution for this matter or give me some alternate idea. I shall be highly obliged.

Hello, I am currently working on the same type of problem with C/D nozzles, I have come to notice that when hybrid initialization is on solution tends to diverge from expected results and in some cases solution is terminated by solver due to high rates of convergence. On the other hand, using standard initialization with Inlet B/C, I managed to have stable and reasonable solution. I have looked it up online, however couldnt find any satisfying explanation.

I am pretty sure of the fact that when Hybrid initialization is used, all B/C's regarding the fluid domain are set (if you are not patching), and solver tries to solve between two B/C which are constant. That is the reason my solution diverged when I used Hybrid init. I humbly advise that if you are not sure the outlet B/C (example : Pressure outlet of CD nozzle) Use standard initialization for it will set only the inlet B/C as constant and will calculate the rest accordingly.

I hope it gives anyone who came across with the same problem good ideas.

Regards,