Learning Forum

[Morizio]

Hi All,

 

I'm using ANSYS fluent 19.1 to simulate a reactor in steady state condition with around 15 reactions in it. its a system of 15 reactions mechanism and I wrote a UDF (DEFINE_NET_REACTION_RATE...) for this purpose. the case is turbulent flow in a chamber with several tubes and I used realizable K-eps model for its hydrodynamics. the flow inside the chamber has two different behavior. there are some parts in the chamber that have very high velocity and high reaction rates (near some wall)  and some other parts with almost zero velocity but still high reaction rate! it should be noted that the flow behavior in the reactor is completely reasonable.

 

I chose volumetric reaction with Stiff Chemistry Solver and Finite-Rate/No TCI  in my first try. it is working well at initial iteration but after that, as the reaction rate is high in some dead zones (with very low velocity), the Isat abort error also appeared that time. I've changed the under-relaxation factor and isat error tolerance but still have the same problem. it caused divergence and some defects in these zones. it is clear that the high reaction rate and stagnant flow in that area cause that problem. 

 

what's your suggestion for my case?  is Finite-Rate/No TCI a good choice for me?

how can I get rid of this problem?

 

Many thanks for your consideration,

 

Morizio

[Amine Ben Hadj Ali]

No I would recommend using a turbulence chemistry interaction model. Example: EDC.

[Morizio]

Thanks for your reply Abenhadj. I reduce the isat error tolerance to 0.001 and the under-relaxation factor of species to 0.2. 

do you have any suggested procedure for using EDC? is there any special strategy for using this model (Initialization and etc)?

Thanks,

Morizio

[Amine Ben Hadj Ali]

Nothing to add what us already mentioned in the documentation. Sometimes starting from cold converged flow with reactions are off might be helpfull. If you can apply one of other model like non premixed or premixed is better before using it.

[Morizio]

Thank you Abenhadj for your valuable comments, 

the simulation is now working without any Isat abort error. Just one thing. the solving procedure is very slow. when I add the species and run it, this massage is appearing:

"ISAT cannot be used with two different gas-phase mechanisms.

switchhing to direct integration"

 and then between each iteration, it says:

Integrating chemistry on fluid zone 15 ...

 

Integrating chemistry on fluid zone 16 ...

 

is there any method to solve this problem? How can I do that? I couldn't find anything in the literature.

 

Many Thanks,

Morizio 

 

[Amine Ben Hadj Ali]

No as you have two different gas phase mechanism ISAT won't work. Check other chemistry agglomeration or other methods if they are compatible here. But why  do you have two mechanisms?

[Morizio]

I don't have two mechanisms! I don't know what does it mean...

I have around 15 species and wrote a UDF for their reactions. the reactions are all in water (not the gas phase). I just define the reactions in UDF and I didn't add any reaction to the mixture in material section. what does it mean? is there anything wrong in my UDF? or defining materials and reactions?

 

Thanks,

Morizio

[Rob]

If you're using the inbuilt Fluent models we have a reaction mechanism for each set of reactions: that's covered in the User's Guide. Without completely debugging the UDF I can't tell if anything is wrong, and I can't do that as it's an open forum. Is there a reason for using a UDF rather than the standard models? 

[Morizio]

Thanks for your comment rwoolhou,

I just define all the species in the material section and selected them in micture part and I haven't added any reaction in the material section (inbuilt fluent model). 

then I wrote a UDF containing the reaction rate for each species and hook it. I have to write UDF as I need to incorporate light irradiance along the device in the kinetics equations. 

what do you think about the following massage? is it related to the udf? should I define all the reactions in both the material section(I cant define those have light intensity in their equation)  and the UDF?

"ISAT cannot be used with two different gas-phase mechanisms.

switchhing to direct integration"

 and then between each iteration, it says:

Integrating chemistry on fluid zone 15 ...

 

Thanks,

Morizio

[Amine Ben Hadj Ali]

Make  a test without the UDF. The message is just informative that it has to integrate the reactions and is not using in situ tabulation.

[Morizio]

Thanks for your reply Amine. I've checked the simulation without UDF and I saw that it is switching to direct integration. I found that as I choose stiff chemistry solver (EDC model which considering stiff chemistry solver) in chemical species model (with or without UDF), it is switching to the direct integration. is it possible to have stiff reaction chemistry and solve the species model by ISAT?

Thanks,

Morizio

 

[Amine Ben Hadj Ali]

That is actually recommended whenever stiff chemistry solver is used.