[julien91]

Hello to all,

I have been working for 4 months on the DSMC software from Bird, DS3V.
As I understood, there is only little help on internet about this software, so here I am:

The set-up studied in my case is a thruster firing in vacuum close on a plate tilted at 15° from the thruster axis and placed 30mm next to it. The result I am looking at is the net heat flux along the plate.

In DS3V, there is an option where you can set your initial flow in the whole domain either as:
-a reference stream (constant temperature, constant number density (linked to the density and molar mass), constant velocity, constant species composition)
-a vacuum (so in theory, no particle at all)

However, when a vacuum is set, the user need nonetheless to specify a positive temperature, species fraction composition, and number density in order for DS3V to initialize correctly the particle cell sizes, otherwise DS3V crashes.

The weird thing is that the change in these options should not impact results (physically i mean) but it does change the numerical results (see the two top curves in the picture).
So one of my question being: which temperature and which number density should be set with respect to my problem so that I get the "correct" simulation?

However, when a reference stream is set, the net heat flux on the plate should be lower than in vacuum, because the particles coming from the thruster lose energy by colliding with the particles present in the reference stream, so that they can't exchange as much energy with the plate than in the vacuum case. This trend is respected (see lower curves in picture), but I was expecting that the higher the pressure, the smaller the net heat flux on the plate (because of the phenomenon just described) but I get the inverted trend (100Pa gives more heat flux than 1Pa).

This is weird, because it means that if the pressure keep decreasing towards vacuum, the curve will jump on the top???

Here is the plot: (http://imgur.com/DimZwH8)

Does someone has an explication for that? Thanks!

[julien91]

No one has any idea about this?

[flotus1]

You seem to be searching for a steady-state solution of the flow, am I right? Have you made sure that the sampling process starts AFTER the steady-state flow has established?
I am not particularly familiar with this program. Does it adapt the collision cell sizes once they were initialized? Did you check if the final cell size is the same for all simulations and small enough (significantly smaller than the mean free path)?

[julien91]

Hi Alex, Indeed I am looking at the steady state regime. When you talk about the sampling process, do you mean the generation of the outputs? If so, then yes I do generate them when the steady state conditions are established.
There is no automatic adaptation of the cell size, only when the flow has "converged" enough, the user can adapt the cell. However it seems that the adaptation does not change the flow significantly, so for these simulations, I did not adapt the cells, so the cells are the same for all of them. Besides, since the user has little control over the adaptation process, I don't know if the size is smaller than the mean free path.
Thanks for your help

[liaofengmao]

hello~
have you solved your problem? i am new in DS3V program, and meet much problem, please help me, thanks! here are some problem that I have meet this days and nobody used this program before around me.
1.I want to simulate the molecular contamination in a vaccum chamber to know the molecular diffuse in this environment. the question is that how to set the initial status of vaccum chamber?
2.when i finish the set of the DS3V program, there are not plots in my program's window. there is nothing in "Pause Program for Detail Plot". But when I run the samples that doneload from the internet, the problem miss in this samples.
3.I want to supply some gas to the vaccum chamber and how can I set in this program?
sorry for my poor english.
I am looking forward to hearing from you, thanks a lot!

[harikaushik]

Originally Posted by julien91

Hi Alex, Indeed I am looking at the steady-state regime. When you talk about the sampling process, do you mean the generation of the outputs? If so, then yes I do generate them when the steady state conditions are established.
There is no automatic adaptation of the cell size, only when the flow has "converged" enough, the user can adapt the cell. However it seems that the adaptation does not change the flow significantly, so for these simulations, I did not adapt the cells, so the cells are the same for all of them. Besides, since the user has little control over the adaptation process, I don't know if the size is smaller than the mean free path.
Thanks for your help

Hi, I am new to DS3V and I am looking for a little help for generating output files during/after the simulation. I have gone through the manual but was did not find anything helpful. Kindly guide me in the right direction. Thanks