[gautcho]

Hello,

I meanwhile really understand a lot of dsmcFoam, but I can't figure out where the dsmcFieldsCalc-routine is executed? I think, it is somewhere in the time.loop-method, but I cant get any further. Could anybody help me?

[gmacpherson]

There is no need to run the dsmcFieldsCalc utility if you have the dsmcFields and fieldAverage function objects from the dsmc tutorials in your controlDict.

All that dsmcFieldsCalc does is execute the dsmcFields function object.

Regards,

Graham

[gautcho]

Okay, perhaps I should think a little bit about this function-in-controlDict-thing. There also comes another question into my mind: Is the mean-function a standardmean over the last writeIntervall or are there any weights or values from the further writeIntervall (I think this depends on the "resetOnOutput"-param)?

Are you the MacPherson, who implemented MD? Than you could tell me, if I would be right, if Id say, that you implemented MD and Scanlon and other implemented dsmc on the basics you created. Because its not that easy to find this out .

[Francesca Scaramuzzino]

Hello, I know that this is not the right place to post this kind of question..anyway..I'm trying to understand the pitzDaily case, in the simpleFoam but in the XiFoam too (for the difference between RAS and LES simulations). So, I've found something about the experiment of Pitz and Daily and in every case I read that the mixture used is propane-air But in the directories I can't find this information. I mean, after the constant values of the variables where I could find the information about the mixture and the composition??

Thanks

Francesca

[gmacpherson]

Hi gautcho,

fieldAverage averages from the start of the simulation (and the averaging survives restarts), unless you have

resetOnOutput on;

in which case it averages between write intervals. This parameter is runtime modifiable, so I normally have it set to 'on' until a case gets to steady state, and switch it to 'off' after to collect and average fields.

Also, yes, I'm the Macpherson of the MD. I implemented the MD as my PhD, and the DSMC was implemented by me at OpenCFD with some discussions and testing by Scanlon et al. The MD and DSMC were both implemented on the basics that were already in OpenFOAM (i.e particle tracking).

Graham

[gautcho]

Oh,

thats nice. Thank you for the solver . I'm writing my diploma thesis about DSMC, dsmcFoam and some applications and I'm working my way through your code right now. If its okay for you, I would ask you some more questions.

1. Biggest at the moment: Where does the post-collision-velocity in the VHS-model come from? I found some similar formular in the Bird (page 36, formulas 2.22), but they are not the same, I think. And why do you use a randomly found angle?

2. Where does the 3rd-velocity-component after collision in MaxwellianThermal come from?
sqrt(-2.0*log(max(1 - rndGen.scalar01()

Its quote similar too the gaussNormal, but I cant get the real difference. Is it from Shen?

3. In which part of the code are particles able to disappear? As far as I can see only in the trackToFace-method, if they hit a patch and then they get deleted in the deleteParticle-method in Cloud.C. Problem: I made some screenprompts at any time deleteParticle is used, but the number of deleted particles in most of the times was higher than the number of screenprompts. If I switch of the possibility to disappear via "hit patch", than every particle stays in the volume. Where does this difference come from?

4. In variableHardSphere.C you used 4.79 from Bird, but I cant figure out what the exp in your implementation does (line 94)?

[Francesca Scaramuzzino]

Hello,
I'm trying to understand the pitzDaily case, in the simpleFoam but in the XiFoam too (for the difference between RAS and LES simulations). So, I've found something about the experiment of Pitz and Daily and in every case I read that the mixture used is propane-air. Anyway in the directories I can't find this information. I mean, after the constant values of the variables where I could find the information about the mixture and the composition?

Thanks

Francesca

[gautcho]

Originally Posted by gautcho

Oh,

thats nice. Thank you for the solver . I'm writing my diploma thesis about DSMC, dsmcFoam and some applications and I'm working my way through your code right now. If its okay for you, I would ask you some more questions.

1. Biggest at the moment: Where does the post-collision-velocity in the VHS-model come from? I found some similar formular in the Bird (page 36, formulas 2.22), but they are not the same, I think. And why do you use a randomly found angle?

2. Where does the 3rd-velocity-component after collision in MaxwellianThermal come from?
sqrt(-2.0*log(max(1 - rndGen.scalar01()

Its quote similar too the gaussNormal, but I cant get the real difference. Is it from Shen?

3. In which part of the code are particles able to disappear? As far as I can see only in the trackToFace-method, if they hit a patch and then they get deleted in the deleteParticle-method in Cloud.C. Problem: I made some screenprompts at any time deleteParticle is used, but the number of deleted particles in most of the times was higher than the number of screenprompts. If I switch of the possibility to disappear via "hit patch", than every particle stays in the volume. Where does this difference come from?

4. In variableHardSphere.C you used 4.79 from Bird, but I cant figure out what the exp in your implementation does (line 94)?

Does anybody else know some answers?

[rejirv]

Hello sir,
Previously there are more variable as results from dsmcfoam such as localMn, localMfp, localCf, localCh etc. But now it seems to be missing.

Any suggestions to obtain those variables ?
Thank you