Simple Flow Problem

prwl120 · March 18, 2015, 12:16

Actually boundary condition are assumed. But situation is practical.

Its Ball bearing, i am studying the pressure variation on lubricant between Balls and bearing wall.

So what should you suggest i would do, decrease the velocity or increase the pressure.

ghost82 · March 18, 2015, 12:31

Your simulation cannot be real with that boundary condition.
What do you want to study?
What results can you expect if you set assumed boundary conditions?

Fluent is a solver and you should expect divergence if you set wrong boundary conditions!

You cannot ask us what boundary conditions to set, it depends on your application and you must know your boundary conditions to simulate your domain.

prwl120 · March 18, 2015, 13:04

Thanks for your help...

But which condition i must have to remember while solving the question.... like is there any upper or lower limit of pressure or velocity in FLUENT.

One more question : is it ok if my mesh quality is just above 0 like 0.003. Or i have to work on my mesh too. I know low mesh quality will bring poor solution but does it also bring divergence in solution???

And does Fluent only accept unstructured mesh??? and what will happen if i provide it a structured mesh???

ghost82 · March 18, 2015, 13:18

As I know there are no limits, apart that you set in solve->controls->limits.
However, I repeat, that if you set wrong (non physical) boundary conditions, fluent will probably diverge.
In your case, with your boundary conditions, the pressure at inlet should be near 500000 atm and velocity at the restriction near 10000 m/s.
These values are both not physical.

If you just want to do a test, set a velocity at the restriction between 25-30 m/s, so to have an inlet velocity of about 0,0025 m/s.
You should enable cavitation model also, otherwise you will have negative absolute pressure at the restriction, which, again, is not physical.

I don't know how fluent reports quality of mesh: usually I evaluate different parameters directly in preprocessing.
Low quality meshes can cause not accurate results, difficult in converging and divergence.
As far as I know hexa meshes are better than tetra, however in some cases (complex geometries) tetra mesh is the only choice.

prwl120 · March 19, 2015, 00:16

Quote:

Originally Posted by ghost82

You should enable cavitation model also, otherwise you will have negative absolute pressure at the restriction, which, again, is not physical.

How do i enable cavitation model. Sorry if i am silly to ask this but i am new to ANSYS.

And Thanks for all the help so far.

edoan · March 19, 2015, 09:03

Just to be clear: Structured vs unstructured mesh deals with the way each element is named and ordered within the domain. like muthu said before just convert your premesh to unstructured mesh while in ICEM and you can easily import your hexa ("structured" appearance but technically unstructured) mesh into fluent you will no loss of quality. .003 quality is really bad.

Guruprasath S · July 29, 2016, 08:10

hai all..i am solving turbulance model sst k-omega with MRF method for blower.my problem is reversed flow in pressure outlets and during hybrid initialization pressure information is not available at boundaries.

If i am giving inlet velocity values,reversed flow occur and divergence happend...

How can i solve this problem and how can i fix this issues...and how can i achive convergence.....please help me

March 18, 2015, 12:31		#22
ghost82 Senior Member Rick Join Date: Oct 2010 Posts: 1,016 Rep Power: 27	Your simulation cannot be real with that boundary condition. What do you want to study? What results can you expect if you set assumed boundary conditions? Fluent is a solver and you should expect divergence if you set wrong boundary conditions! You cannot ask us what boundary conditions to set, it depends on your application and you must know your boundary conditions to simulate your domain. __________________ Google is your friend and the same for the search button!

March 18, 2015, 13:18		#24
ghost82 Senior Member Rick Join Date: Oct 2010 Posts: 1,016 Rep Power: 27	As I know there are no limits, apart that you set in solve->controls->limits. However, I repeat, that if you set wrong (non physical) boundary conditions, fluent will probably diverge. In your case, with your boundary conditions, the pressure at inlet should be near 500000 atm and velocity at the restriction near 10000 m/s. These values are both not physical. If you just want to do a test, set a velocity at the restriction between 25-30 m/s, so to have an inlet velocity of about 0,0025 m/s. You should enable cavitation model also, otherwise you will have negative absolute pressure at the restriction, which, again, is not physical. I don't know how fluent reports quality of mesh: usually I evaluate different parameters directly in preprocessing. Low quality meshes can cause not accurate results, difficult in converging and divergence. As far as I know hexa meshes are better than tetra, however in some cases (complex geometries) tetra mesh is the only choice. __________________ Google is your friend and the same for the search button! Last edited by ghost82; March 19, 2015 at 16:10.

March 19, 2015, 09:03		#26
edoan Member Ethan Doan Join Date: Oct 2012 Location: Canada Posts: 90 Rep Power: 14	Just to be clear: Structured vs unstructured mesh deals with the way each element is named and ordered within the domain. like muthu said before just convert your premesh to unstructured mesh while in ICEM and you can easily import your hexa ("structured" appearance but technically unstructured) mesh into fluent you will no loss of quality. .003 quality is really bad. ghost82 likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Domain format problem on airfoil flow simulation	andrenonaka	CFX	14	December 7, 2015 01:42
Problem with changing mass flow rate	roo	FLUENT	3	June 4, 2013 15:53
Problem of mass conservation in multispecies flow in porous media and T° gradient	pilou	FLUENT	0	January 31, 2011 09:19
Can 'shock waves' occur in viscous fluid flows?	diaw	Main CFD Forum	104	February 16, 2006 06:44
Multiphase flow problem	icedou	FLUENT	6	July 10, 2005 03:52

March 18, 2015, 12:16		#21
prwl120 New Member Shailesh Join Date: Mar 2015 Posts: 16 Rep Power: 11	Actually boundary condition are assumed. But situation is practical. Its Ball bearing, i am studying the pressure variation on lubricant between Balls and bearing wall. So what should you suggest i would do, decrease the velocity or increase the pressure.

March 18, 2015, 13:04		#23
prwl120 New Member Shailesh Join Date: Mar 2015 Posts: 16 Rep Power: 11	Thanks for your help... But which condition i must have to remember while solving the question.... like is there any upper or lower limit of pressure or velocity in FLUENT. One more question : is it ok if my mesh quality is just above 0 like 0.003. Or i have to work on my mesh too. I know low mesh quality will bring poor solution but does it also bring divergence in solution??? And does Fluent only accept unstructured mesh??? and what will happen if i provide it a structured mesh???

July 29, 2016, 08:10		#27
Guruprasath S New Member Guruprasath S Join Date: Jul 2016 Posts: 1 Rep Power: 0	hai all..i am solving turbulance model sst k-omega with MRF method for blower.my problem is reversed flow in pressure outlets and during hybrid initialization pressure information is not available at boundaries. If i am giving inlet velocity values,reversed flow occur and divergence happend... How can i solve this problem and how can i fix this issues...and how can i achive convergence.....please help me