|
[Sponsors] |
July 18, 2016, 07:41 |
Suggest physics models
|
#1 |
Member
sriram
Join Date: Jun 2016
Posts: 48
Rep Power: 9 |
Hi evry1
Im working with solid particles suspended in water flow through pipes.Note:The solid doesnt dissolve in water... Under what category this comes?? What kind of physics model must be given? Can i change the water standards like density?? I am finding clueless...even the star ccm+ tutorial guide shows legrangian laden flow,Eulerian laden flow with rotating walls. I tried eulerian multi phase,eulerian segregated models.. But i kept gettin AMG solver divergd. Not very much familiar with CCM+ just startd using Thank u |
|
July 18, 2016, 10:47 |
|
#2 |
Member
Join Date: Aug 2014
Location: Germany
Posts: 77
Rep Power: 11 |
You can use any of the multiphase methods. If the number of particles are less, I would go for lagrangian.
You can change properties of water. It should be under Conitnua > Physics. Try reducing the relaxation factors to avoid divergence, atleast thats what help me mostly if my solution diverges from the very beginning..Also make sure your mesh size and time steps are appropriate. hope that helps |
|
July 19, 2016, 01:28 |
still diverging
|
#3 |
Member
sriram
Join Date: Jun 2016
Posts: 48
Rep Power: 9 |
thank u
but its still diverging...how to set time step...i've right clicked>continua i dont find casting physics option as given in the tutorial guide. |
|
July 19, 2016, 02:30 |
|
#4 |
Member
sriram
Join Date: Jun 2016
Posts: 48
Rep Power: 9 |
can u suggest me the list of physics models to be used...i think that s where im making mistake
|
|
July 19, 2016, 02:56 |
|
#5 |
Senior Member
kevin alun
Join Date: Sep 2011
Location: Germany
Posts: 106
Rep Power: 14 |
I was also thinking Lagrangian or DEM but maybe a better starting point is the tutorial for fluidized bed. Your particles laminar and liquid is turbulence, I think you wanted this also.
Keep things very simple in the beginning, use the minimum amount of models in the beginning, for example in the interaction use only drag and a constant interaction length scale. Perhaps steady-state, use constant properties, try to get a stable run using minimum amount of models. You are most likely going to have to reduce the relaxation factors or time step, Might even want to try laminar for the liquid phase as well. |
|
July 19, 2016, 05:40 |
|
#6 |
Member
sriram
Join Date: Jun 2016
Posts: 48
Rep Power: 9 |
thank u
i have changed the relaxation factors values as mentioned in the tutorial guide,now it converges. But the iteration doesn't stop once the maximum step is achieved.Why is that? |
|
July 19, 2016, 05:50 |
|
#7 |
Member
Join Date: Aug 2014
Location: Germany
Posts: 77
Rep Power: 11 |
Can you post a screenshot of your 'stopping criteria' tree ? (with the nodes expanded of course)
|
|
July 19, 2016, 06:37 |
image
|
#8 |
Member
sriram
Join Date: Jun 2016
Posts: 48
Rep Power: 9 |
i have attached the images
|
|
July 19, 2016, 08:48 |
|
#9 |
Member
Join Date: Aug 2014
Location: Germany
Posts: 77
Rep Power: 11 |
Are you running a steady state simulation ? I think no of time steps would matter only for transient simulations.
|
|
July 19, 2016, 09:31 |
|
#10 |
Member
sriram
Join Date: Jun 2016
Posts: 48
Rep Power: 9 |
yes it is a steady state.....i ran another simulation and it stopped correctly at the stopping criteria...so the relaxation factor change worked it seems...coz the simulation is not getting diverged
|
|
July 19, 2016, 09:53 |
changing of relaxation
|
#11 |
Member
sriram
Join Date: Jun 2016
Posts: 48
Rep Power: 9 |
i ran an earlier divergence occurred simulation...with changing the relaxation factor..but error occurs again....
|
|
July 20, 2016, 03:31 |
|
#12 |
Member
Join Date: Aug 2014
Location: Germany
Posts: 77
Rep Power: 11 |
Please be more clear, as I understand, you had a solution that wasn't diverging after changing relaxation factors. But then what did you change so that the solution started diverging ?
Keep your simulation as steady/unsteady and then change other simulation parameters one by one, if you change many things at once, it would be hard to identify what causes problems exactly. |
|
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Cross country Oil & Gas Pipelines - Accurate online models | sohail1 | Main CFD Forum | 2 | July 4, 2016 03:31 |
Physics Models | Roberto J. | STAR-CCM+ | 1 | March 8, 2016 18:41 |
Incompressible Turbulence models | achinta | OpenFOAM | 4 | May 27, 2010 11:35 |
comparing different models and grids | Philipp | Main CFD Forum | 4 | August 31, 2001 05:14 |
Reynolds Stress Models | Roued | Main CFD Forum | 20 | February 8, 2000 03:58 |