CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Steady/Unsteady (http://www.cfd-online.com/Forums/fluent/32128-steady-unsteady.html)

Pravesh Kumar September 13, 2003 04:58

Steady/Unsteady
 
Hi

I am doing multiphase simulations. I am using Eulerian model and patching the volume fraction in domain. But when I am doing steady state simulation the value of volume fraction is changing while for unsteady state it remains constant.

I don't know why it is changing in steady state as I patch it?

Regards Pravesh


ap September 14, 2003 08:26

Re: Steady/Unsteady
 
Probably I don't understand your problem: I don't know what system you're simulating.

But I don't see anything strange if your volume fraction changes in a steady simulation.

When you patch a value in your domain, you just set it's initial value, which is used as a starting point for calculation. If this value is different from the value in the steady state of the system (if this exists), it will change also during steady simulation.

Hi :)

ap

Pravesh Kumar September 15, 2003 04:19

Re: Steady/Unsteady
 
hi I want to keep the volume fraction(say 0.5)of solid phase fixed throughout the calculation. I have defined the velocity of solid phase as zero in domain as the phase is stationary. Now to fix the value of volume fraction I am patching the value in domain. When I am doing steady state it is changing(solid volume fraction) while for unsteady state it remains constant?

I don't know why this is happening? Regards Pravesh

ap September 15, 2003 05:54

Re: Steady/Unsteady
 
What are you trying to simulate?

Hi :)

ap

Pravesh Kumar September 15, 2003 23:35

Re: Steady/Unsteady
 
hi

I am trying to simulate multiphase packed bed.Air(primary phase) and water are coming from the top and flow is through the solid phase(not moving) with porosity value of 0.63. For this I have defined the three phases in materials panel (all fluid type).The phase which is not moving I have defined it as high density fluid and in the boundary -fluid-fixed values panel I have fixed the velocity of this phase as zero(constant). Then after initializing I am patching the value of solid phase as 0.63.I want this value to remain constant throughout the simulation.It is constant in unsteady state while in steady state it is changing?

If you want anything else feel free to ask. Regards Pravesh


ap September 16, 2003 07:20

Re: Steady/Unsteady
 
I'm not sure of why this happens, but I think the flow behaviour in your system is not steady, so it's possible that the steady simulation isn't able to represent it properly.

I had a similar problem in circulating fluidized beds, which are highly unsteady. When I tried to simulate them using the steady solver in order to see if I could obtain the "steady" configuration of the system in a shorter time, I just obtained an unstable behaviour and an unphysical result.

For example, in these reactors, the solid phase forms an annular layer on the wall, which is captured by the unsteady simulation, while steady calculations just show this layer appear and disappear.

Hi :)

ap

Pravesh Kumar September 17, 2003 00:53

Re: Steady/Unsteady
 
hi Thanaks!!Then I am going for unsteady state simulation. Then how I will come to know that the steady state solution is reached. Whether I should wait for flux balance or I should monitor some parameter.Then I am not sure which paramater I should monitor whether it is velocity(velocity magnitude or Y velocity, 2D geometry) or volume fraction and of which phase(liquid and gas)or if you can suggest any other parameter.

Regards Pravesh

Greg Perkins September 17, 2003 05:39

Re: Steady/Unsteady
 
First, as ap said patching only sets the inital value. To fix the volume fraction of your Eulerian granular phase initialise your solution and then turn off the solution of equations for this phase in the Solve/Control panel.

After you have done this rerun your steady simulation. You may find this converges and gives what you expect,

If not then try an unsteady simulation.

Greg

Pravesh Kumar September 17, 2003 07:10

Re: Steady/Unsteady
 
hi In the the Solve/Control panel we can deselect Flow and volume fraction. We can't deselect volume fraction as for liquid phase we want to select these equations. And fluent doesn't provide the facility to have a phase without solving that phase equation.

As whatever I patched are initial values but as I have fixed the velocity as zero for that phase then it should not move and the values should remain same. I don't know what is happening?

Regards Pravesh

Greg Perkins September 17, 2003 07:37

Re: Steady/Unsteady
 
I don't use the Eulerian Granular model so I'm not sure.

If you tern off the momentum equations for the granualar phase then I guess (?) the volume fraction can't change anyway. It makes sense that only N-1 volume fractions need be solved for an N phase system.

Greg

Pravesh Kumar September 17, 2003 23:47

Re: Steady/Unsteady
 
hi There is one primary phase and the remaining phase(N-1) (IF YOU SAY TOTAL ARE N PHASES)are secondary phases. How can I turn off the momentum equation for only one secondary phase as In solve/control panel all the secondary phases are to be selected/deselected.If I am desececting the volume fraction then all the secondary phases are deselected.I can't do like this I want to solve momentum equation for one secondary phase and don't solve it for other secondary phase. Is there a way?

Regards Pravesh

Greg Perkins September 18, 2003 00:21

Re: Steady/Unsteady
 
If there is not way in FLUENT then you can add source terms to the secondary phase you want to remain stationary so that the velocities are zero.

As I said I haven't used this model so I'm not sure, but I'm sure you can wdo a work-around to solve what you want .

Greg


All times are GMT -4. The time now is 23:07.