CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CONVERGE (https://www.cfd-online.com/Forums/converge/)
-   -   G-equation set up (https://www.cfd-online.com/Forums/converge/218683-g-equation-set-up.html)

mickaelmessias.silva June 30, 2019 03:23

G-equation set up
 
Hi all.
I have a couple of questions regarding G-equation for which I couldn't find the answer yet.

1 - In the example cases provided, the G-equation initialization temperature is set to 3500 K (in the spark tab). Why did they pick this temperature? Shouldn't we input some temperature close to the adiabatic flame (around 2300 K)? Does this value matter if I'm sourcing G directly?

2 - By using G-equation, should we expect the max temperature rises to 2400 (ish) K instantly? As if there was no ignition delay.

3 - For the source/sink modeling, I set the value of the passive to 8e+15 [1/m3s] since the manual says it can be as high as possible. However I left the max passive as default (0.001). Is it ok?

4 - Since I'm sourcing G directly, does the initial and end times of the (delta_t) of the source matter?

Kind regards,

Mickael

Yaju July 18, 2019 12:19

Hi Mickael,


1. The G-initialization temperature would have been used if temperature was sourced for ignition instead of G. With the G source you try to keep G-initialization temeprature relatively higher so that it does not interfere with the G source, unless you want to capture pre-ignition/knock in the unburnt region.



2. Yes, the G-equation model is primarily a flame propagation model. The G source is just trying to initiate a flame, it is not trying to model the ignition process/delay.


3. Yes, the max passive value of 0.001 is fine. Make sure to have embedding around the spark region similar to the SI8 example case.


4. ILonger source time would help with flame initiation for stratified cases. For most of the premixed cases it would not matter as much. But you should try to keep the source timing similar to how long the actual spark plug might be discharging.




Yaju.

luckyboy November 9, 2019 02:19

1 Attachment(s)
Quote:

Originally Posted by Yaju (Post 739340)
Hi Mickael,


1. The G-initialization temperature would have been used if temperature was sourced for ignition instead of G. With the G source you try to keep G-initialization temeprature relatively higher so that it does not interfere with the G source, unless you want to capture pre-ignition/knock in the unburnt region.



2. Yes, the G-equation model is primarily a flame propagation model. The G source is just trying to initiate a flame, it is not trying to model the ignition process/delay.


3. Yes, the max passive value of 0.001 is fine. Make sure to have embedding around the spark region similar to the SI8 example case.


4. ILonger source time would help with flame initiation for stratified cases. For most of the premixed cases it would not matter as much. But you should try to keep the source timing similar to how long the actual spark plug might be discharging.




Yaju.

I solved many questions after seeing your answer to Mickael's questions, but I also have some questions about the G-equation model.
I am doing the simulation of hydrogen combustion and knock prediction. And I also have some trouble with the G-equation model.
1. when the G source is used for ignition, the ignition always is failed. I don't what's wrong with it?
2. when the energy source is used for ignition, the mixture can be ignited only when the G initialization temperature is set at 1000K and above 4000K. what's more, the heat release rates for the two temperatures are quite different (the green line is 1000K and the yellow is 4000K). so I don't understand how to set the initialization temperature with the energy source for ignition.
Thanks for your help!

jlemoine November 14, 2019 14:27

luckyboy,
Based on the complexity of your case, it will be better if we keep talking through email. It is best if you send at support@convergecfd.com your inputs files so we can help you with your setup.

luckyboy November 14, 2019 21:55

Quote:

Originally Posted by jlemoine (Post 749799)
luckyboy,
Based on the complexity of your case, it will be better if we keep talking through email. It is best if you send at support@convergecfd.com your inputs files so we can help you with your setup.

Thank you very much and I have send my inputs files at email.


All times are GMT -4. The time now is 09:46.