[RiccardoCFD]

Good morning,
I'm writing a thesis about spark ignition engines fuelled with natural gas using a partially stratified charge injection strategy. I've spent much time so far trying to use the available combustion models in CONVERGE in order to match some experimental data. The results have been rather disappointing, so I've decided to implement the PaSR combustion model via the definition of a UFD. However, I find it very difficult to manipulate all those *.c files, since there are no enough comments to understand how the variables are defined.

I'll try to explain briefly what I would like to do: (let kk and i be two indices referring respectively to the computational node and to the chemical species)
1) Calculate the turbulent mixing time scale:

tau_mix[kk] = CONST*(mu_eff[kk]/(rho[kk]*epsilon[kk])^0.5

where rho is the fluid density, mu_eff is the global viscosity and epsilon is the turbulent energy dissipation.

2) Calculate the chemical time of the reference limiting species:

1/tau_chem[kk] = max(-omega_dot[i,kk]/Y[i,kk])

where omega_dot and Y are respectively the chemical reaction rate (mol/m3*s) and the molar concentration of species i in the kk-th cell of the computational domain.

3) Calculate the parameter kappa_pasr, defined as follows:

kappa_pasr[kk] = tau_chem[kk]/(tau_chem[kk]+tau_mix[kk])

4) Calculate the effective reaction rate (after understanding how the SAGE model refers to the laminar one):

new_RR[i,kk] = kappa_pasr[kk]*RR[i,kk] (kg/m3*s)

5) Use this new reaction rate in the source term related to combustion in the species transport equation and in the energy equation.

I know that my request is probably to hard to be satisfied, but I wonder if someone can give me some suggestions for the implementation of this UDF.

[xieshengbai]

Hey,

I am sorry that our current combustion models did not serve your purpose well. But I have to say that there are many parameters that you can tune with as well as that the mechanism itself matters.

UDFs are tricky and requires some knowledge of our data structure. So it is hard to tell via this forum. The best way is to send an formal inquiry as well as your case to us at support@convergecfd.com, then we can have some experts working on it in particular.

Thank you.

Originally Posted by RiccardoCFD

Good morning,
I'm writing a thesis about spark ignition engines fuelled with natural gas using a partially stratified charge injection strategy. I've spent much time so far trying to use the available combustion models in CONVERGE in order to match some experimental data. The results have been rather disappointing, so I've decided to implement the PaSR combustion model via the definition of a UFD. However, I find it very difficult to manipulate all those *.c files, since there are no enough comments to understand how the variables are defined.

I'll try to explain briefly what I would like to do: (let kk and i be two indices referring respectively to the computational node and to the chemical species)
1) Calculate the turbulent mixing time scale:

tau_mix[kk] = CONST*(mu_eff[kk]/(rho[kk]*epsilon[kk])^0.5

where rho is the fluid density, mu_eff is the global viscosity and epsilon is the turbulent energy dissipation.

2) Calculate the chemical time of the reference limiting species:

1/tau_chem[kk] = max(-omega_dot[i,kk]/Y[i,kk])

where omega_dot and Y are respectively the chemical reaction rate (mol/m3*s) and the molar concentration of species i in the kk-th cell of the computational domain.

3) Calculate the parameter kappa_pasr, defined as follows:

kappa_pasr[kk] = tau_chem[kk]/(tau_chem[kk]+tau_mix[kk])

4) Calculate the effective reaction rate (after understanding how the SAGE model refers to the laminar one):

new_RR[i,kk] = kappa_pasr[kk]*RR[i,kk] (kg/m3*s)

5) Use this new reaction rate in the source term related to combustion in the species transport equation and in the energy equation.

I know that my request is probably to hard to be satisfied, but I wonder if someone can give me some suggestions for the implementation of this UDF.