July 24, 2022 at 9:14 ammenkchSubscriber
Hello, I am having trouble setting up the FGM model for the combustion of hydrogen coming out of a tube into still air. I believe there is a problem with hybrid initialization. I have tried patching reaction progress to c=1 everywhere at first, but the result is H2 mole fraction is 1 everywhere and temperature is 400K everywhere (these are actually my tube inlet boundary conditions... it's as if they propagate to the entire domain). I do not have an air inlet, I would just like to set still air for the whole domain except the fuel inlet initially but I could not do that. I tried patching c=1 in the tube and c=0 elsewhere using cell registers but that gave the exact same result (H2 everywhere and constant 400K everywhere). Is anyone able to explain how to initialize?
July 26, 2022 at 12:19 pmDrAmineAnsys Employee
Are you experiencing the same issues when using Standard Initialization? FMG initialization is not comptabile with Premixed Model.
July 26, 2022 at 1:25 pmmenkchSubscriber
I am currently trying it. I used standard initialization with default settings (progress variable is 1), then patched mixture fraction to be 0 in the full domain. It is running now but given my report of volume average temperature which stays at 300K, I do not believe a flame is appearing, or at least not yet... (transient simulation). Does my initialization make sense though? Mixture fraction is 0 because there is only oxidizer initially and progress variable is 1 for ignition.
July 26, 2022 at 1:46 pmDrAmineAnsys Employee
Initialisation for transient should reflect state at zero. So for me makes a sense.
You are not using premixed model as you are referring to Mixture Fraction...
July 26, 2022 at 2:16 pmmenkchSubscriber
I am using Flamelet Generated Manifold with premixed flamelets more precisely
July 26, 2022 at 3:37 pmmenkchSubscriber
So this is what I was scared of : there is still no flame appearing. I have thought of creating a hot point so that it ignites, maybe doing a cell region/register near the tube and patching the temperature there at the auto-ignition temperature of H2 in air. Any thoughts or better ideas?
July 27, 2022 at 1:19 pmRobAnsys Employee
Patch sounds good. Also patch a small fraction of products over the domain. I don't do combustion modelling but I remember one of the models needing a tiny fraction of products to work correctly.
July 27, 2022 at 2:31 pmDrAmineAnsys Employee
Patching is almost all combustion cases I know is required.
July 27, 2022 at 2:39 pmmenkchSubscriber
I have tried patching and when I do and check the contour to see if it was taken into account before running the simulation, the contour still appears to be 300 K everywhere in the domain and not the patched temperature which is 1000 K. My supervisor does not understand this either; he will send a request to the Ansys Customer Portal but meanwhile if anyone has an idea how to solve this it would be appreciated. I am now trying with the spark ignition... I feel like the issue with patching is related to the FGM model.
July 29, 2022 at 10:55 amRobAnsys Employee
Please post some images. Are you patching to a cell zone or register?
July 29, 2022 at 11:03 ammenkchSubscriber
For information my whole domain is 0.2 meters in the x and y directions and 2 meters in the z direction starting from origin. I have tried creating this region (region_0 in the below picture) and patching this way (above picture) which did not work, and I also tried with a spherical cell register near my nozzle, but it did not work either.
July 29, 2022 at 11:09 ammenkchSubscriber
And I wonder if this might be due to the FGM model which is based on the progress variable. I tried patching the whole fluid domain using a different case file with Species Transport reactive model and it worked...
- You must be logged in to reply to this topic.
Simulation World 2022
Earth Rescue – An Ansys Online Series
- Suppress Fluent to open with GUI while performing in journal file
- Heat transfer coefficient
- What are the differences between CFX and Fluent?
- Floating point exception in Fluent
- Difference between K-epsilon and K-omega Turbulence Model
- Time Step Size and Courant Number
- The solver failed with a non-zero exit code of : 2
- Exporting Data Results
- error in cfd post