Learning Forum

[MS10]

Hello everyone 

I am simulating a VOF-LES model for a gas-liquid system with species transport (hydrocarbon gases). When I use the simulation without species transport (mixing rule for gases), the solution converges, but when I use the species transport modelling with ideal gas and constant density. it is not converging. However, I have refined the mesh (mesh 1.5 million) and adjusted the URFs, but couldn't get convergence. 

Thanks and regards

Ali

[Rob]

Is it the species equation that's not converging, or something else? Is the species mixing in the gas phase, or all entering from a single inlet? Was the gas modelled as ideal in the previous case? 

[MS10]

 

Hi @Rob. Thank you so much for your quick response. 

Yes, it is the species equation. All species mix in the gas phase and enter from a single inlet. Actually, I am running several cases using different density functions such as density as a constant (volume-weighted mixing law) and density as an ideal EOS. In ideal gas conditions, it started with good convergence and then the solution stopped converging in 40 iterations. It is a transient solver with PISO and PRESTO.  The same model I used without species transport (just mixing rule for density calculation) is converging very well. 

 

Thanks & regards

Ali

 

[Rob]

Check what species and mixture density you're using. Depending on the temperature, pressure and molecular weights you could be seeing some rapid changes in value. You may also have missed a setting, or have an incorrect value if you set the mixture level density but don't check the species for additional inputs. 

[MS10]

 

I am using a constant density for species and then defining it as a “volume-weight mixing law” in the mixture. Can you please explain further about this "if you set the mixture level density but don't check the species for additional inputs"? However, I didn’t patch the species. Is it necessary to patch all species after the initialization? Whenever I try to setup up the selection of species in the mixture. It shows an error message (image attached below). 

 

Thanks 

Ali

 

[Rob]

If you set volume weighted each species works out it's own density and then the mixture density that's used in the flow equations is calculated from those. If you set some of the other mixture level options you'll find Fluent requires different inputs at the species level. 

The CAR:CDR error means something somewhere isn't set. So, I suspect during a species/mixture change you've managed to get a field with no value. When you initialise the mixture will be set based on the initial conditions. No need to patch unless what you chose isn't what you wanted.