error in vacuum

ghorrocks · December 1, 2016, 16:17

Some comments on your setup:

* Set the reference pressure to 1 torr and the inlet to 39 torr and atmosphere to 0 torr.
* Why have you got viscous work turned on? This is not going to have any viscous work done, so turn it off.
* I see you have free slip walls. I guess you are trying to reproduce the very low pressure behaviour. You should look into wall modelling for low pressure flows, there are some interesting physics which goes on for low pressures.
* You have a laminar flow model. Have you check the Reynolds numbers that the flow is in fact laminar?
* You you confirmed your time step and convergence criteria are OK?
* Your mesh is quite coarse, especially after the expansion. Have you checked you mesh density is OK?
* Note that the default ideal gas model might not be a very accurate material model for this case. You might want to think about a real gas model - but they are harder to converge, so make sure you can run ideal gas reliably before considering that.
* Have you had a look at the results file leading up to the crash? Save some backup files before it crashes. I bet that the crash happens when the jet first hits the outlet boundary or something like that.

You have lots of things to look at

kykylalac · December 2, 2016, 08:27

what can i set viscous in task?
about turbulense: i want to get simple decision

i know about real gas, but i want testing the task with ideal gas

if my mesh and other conditions (except wall condition temperature) with wall heat transfer coef. bad, why did my task solved?
how can i heat transfer turn on?

what boundary condition can i use for free gas outlet?
it's last image before crach ??????????3.jpg

ghorrocks · December 3, 2016, 04:17

My comments mainly are for accuracy. It is possible for an inaccurate simulation to converge.

You cannot tell anything from that image. You are going to have to look at the variable fields to work out what is wrong. A clue could be to look for negative absolute pressures, they will cause divergence. Your minimum density is 1e-7 but your ambient density is around 1e-4. Where is this minimum density? What is the pressure at that point? Is it realistic, or a numerical artefact?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Strange non vacuum behavior in DSMC DS3V	julien91	Main CFD Forum	5	October 8, 2021 04:40
Modeling a vacuum region in solar collector with fluent	amirokhovat	FLUENT	2	November 4, 2016 06:36
radiation modeling in vacuum	mirko	OpenFOAM Programming & Development	7	May 20, 2016 08:31
Assistance in Vacuum pump simulation	enr_venkat	CFX	5	November 20, 2012 11:50
Radiation in vacuum	nitin	CFX	0	July 6, 2009 01:49

December 1, 2016, 16:17		#21
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Some comments on your setup: * Set the reference pressure to 1 torr and the inlet to 39 torr and atmosphere to 0 torr. * Why have you got viscous work turned on? This is not going to have any viscous work done, so turn it off. * I see you have free slip walls. I guess you are trying to reproduce the very low pressure behaviour. You should look into wall modelling for low pressure flows, there are some interesting physics which goes on for low pressures. * You have a laminar flow model. Have you check the Reynolds numbers that the flow is in fact laminar? * You you confirmed your time step and convergence criteria are OK? * Your mesh is quite coarse, especially after the expansion. Have you checked you mesh density is OK? * Note that the default ideal gas model might not be a very accurate material model for this case. You might want to think about a real gas model - but they are harder to converge, so make sure you can run ideal gas reliably before considering that. * Have you had a look at the results file leading up to the crash? Save some backup files before it crashes. I bet that the crash happens when the jet first hits the outlet boundary or something like that. You have lots of things to look at

December 2, 2016, 08:27		#22
kykylalac New Member julie Join Date: Oct 2016 Posts: 16 Rep Power: 9	what can i set viscous in task? about turbulense: i want to get simple decision i know about real gas, but i want testing the task with ideal gas if my mesh and other conditions (except wall condition temperature) with wall heat transfer coef. bad, why did my task solved? how can i heat transfer turn on? what boundary condition can i use for free gas outlet? it's last image before crach ??????????3.jpg

December 3, 2016, 04:17		#23
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	My comments mainly are for accuracy. It is possible for an inaccurate simulation to converge. You cannot tell anything from that image. You are going to have to look at the variable fields to work out what is wrong. A clue could be to look for negative absolute pressures, they will cause divergence. Your minimum density is 1e-7 but your ambient density is around 1e-4. Where is this minimum density? What is the pressure at that point? Is it realistic, or a numerical artefact?