Hi, good people:

Must I transform the unit(inch or foot) of the GAMBIT into Meter when I read the case file into fluent? If it is, why do I find that some length varibles is to be given a value in INCH or FOOT in FLUENT? Thank u for your help.

Several unit system can be allowed for user's convenience. I think that it is only for convenience. Fluent uses SI units(meter, sec, kg ....) for all internal calculation.

Sincerely, Jinwook

Hi,Jinwook:

Very glad to see you again, I' Harry Qiu. Thank you for your perfect answer, I understand it now.But I have another question.

Now, I want to simulate a complex model of laminar flow, heat conduction, convective heat transfer, radiation, and premixed gaseous combustion. The geometry model is of 2-dimensions. Interior of the geometry is gas(or flue gas) area, out of which(normal to flow direction) is ceramic(solid),I want to calculate the temperature distribution of the flue gas in 2-dimension, and at the same time compute the temperature distribution of the ceramic(solid).

My questions are: (1)Can I simutaneously simulate the 2-dimension fluid flow and 2-dimension solid heat conduction(surrouding the flow area)?

(2)Must I connect the connective edge of fluid and solid area as one edge(two edge ->one common edge)?

(3)I want to take the heat transfer(convective and radiation) between fluid and solid into consideration. Then how can I specify the boundary condition at this common edge? I'm eagerly looking forward to your help, hank you! Sincerely yous, Harry Qiu

Dear Harry Qiu. I am happy, too.

Q1) Can I simutaneously simulate the 2-dimension fluid flow and 2-dimension solid heat conduction(surrouding the flow area)?

A1) Of course you can. I have similar experience by Fluent. Dont' worry about it.

(Q2) Must I connect the connective edge of fluid and solid area as one edge(two edge ->one common edge)?

A2) You should generate it by one edge in GAMBIT. Then Fluent(solver) separates it by two edges. One is interior adjacent to the fluid region for gas phase calculation and the other is exterior adjacent to the solid for heat conduction calculation.

(Q3) I want to take the heat transfer(convective and radiation) between fluid and solid into consideration. Then how can I specify the boundary condition at this common edge?

A3) You should give the internal emmisivity of the wall at the one of common edge which is adjacent to the fluid region.

That's all. I think that you can do above mentioned operation by couple of 'CLICKs' in the boundary condition panel.

Sincerely, Jinwook

Hi, Jinwook:

How are you recently?

Thank you very much for your perfect answer last time. It helps me so much. But, I have more questions. I'm calculating that laminar premixed gas combustion in a pore(porous burner). The questions are as follows:

(1)I employed the "finite rate reaction model" for simulating gaseous combustion and "DO radiation model" for modeling radiation heat transfer. When I only simulate cold flow, it is quite easy to converge. But it becomes very difficult(or impossible)to converge once I add the reaction and radiation, even though I do it according to the FLUENT 5 Documentaton(Two steps cold-flow-->hot-flow simulation, under-relaxation for density, patching an ignition). Can you tell me any other measures I can take for convergence?

(2)When I patch a "spark" in the fluid zone, the position(location) of reaction front is not the same as I expected preveously. So, does the position of the reaction front(laminar premixed combustion) vary with where the "spark" is patched?

(3)In Fluent5 Documentation, there is an item "patching an ignition in the selected cells". What does "the selected cells" mean? How to select the cells? In fluent or in Gambit?

Looking forward to your help! Thank you very much!

Sincerely yours, Harry Qiu.

Dear Harry Qiu

I have some experience for turbulent pre-mixed combustion. At that time, I used eddy-dissipation model.

I think that it is very difficult to converge finite rate reaction model for pre-mixed LAMINAR combustion. I do not have any experience for the simulation of laminar premixed combustion. But, I think that you had better try another method. Consider the situation: Because oxidant and fuel are already premixed, all fuel tends to burn at the first cell just after mixture inlet, so that it causes numerical divergence problem. That's why I've used eddy dissipation model for premixed turbulent combustion.

What I ONLY can recommend you is,

Try with lower calorific value, e.g., 1/10 of actual LHV and increase it step by step to the real value. But I am afraid that it might not work.

Sincerely, Jinwook

Hi, Jinwook: Thank you for your help. If I want to consider radiation of the flue gas and the wall, For the flue gas, can I use GRAY-discrete ordinate radiation model? Tank you. Sincerely yours, Harry Qiu

Hi Harry Qiu

Of course, you can. However, the selection of constant value for absorbtion coefficient is entirely up-to your decesion. The result is fairly sensitively dependent on the absorbtion coefficient.

Sincerely, Jinwook