|
[Sponsors] |
Full Impeller Domain In CFX Pre after Turbogrid Messhing of a Single flow passage |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
February 10, 2023, 13:21 |
Full Impeller Domain In CFX Pre after Turbogrid Messhing of a Single flow passage
|
#1 |
New Member
Calvin Stephen
Join Date: Feb 2020
Posts: 4
Rep Power: 6 |
Is there by any chance that I can obtain the full Impeller in turbogrid instead of a single flow passage?
I am simulating flow through a pump when its operated in reverse and using a single blade wont be representative of the problem as the flow field might be different at any point of the volute given the changing shape of the volute and thats why the need to obtain a complete impeller model. Anyone with an idea of how to obtain that please your help is much appreciated.. Regards, |
|
February 10, 2023, 21:46 |
|
#2 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,870
Rep Power: 144 |
You are correct in saying that if you run a pump in reverse this means that it is likely that it will run with large separations and other difficult behaviours. This means a single blade passage is unlikely to be accurate.
To generate a full impeller mesh you can remesh, of course - but if you already have a single passage meshed you can use the transform mesh option in CFX-Pre to transform the existing mesh mesh around for the rest of the blade passages. Note that you will probably need to use GGI boundaries to connect up adjacent blade passages (unless you made periodic matching mesh on the sides of your single passage).
__________________
Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum. |
|
February 11, 2023, 12:54 |
|
#3 |
Senior Member
Join Date: Jun 2009
Posts: 1,880
Rep Power: 33 |
adding to Glenn's response, you do not want an impeller mesh in one go. You want a representative part (single passage, or set) mesh and then replicate in as many as you need to fit in 360.
Why? You want every single passage/set to be discretized identically, and not closely to be certain that any flow differences between passages are due to physics and not due to numerical errors. Because of this you want to be certain you have conformal periodic surface (as Glenn pointed out) and glue them when replicating to avoid more numerical errors.
__________________
Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum. |
|
Tags |
cfx, mesh 3d, turbogrid |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
[TurboGrid] How to best model centrifugal pump impeller and suction domains using TurboGrid | jgross | ANSYS Meshing & Geometry | 1 | December 28, 2021 18:10 |
CFX - specified domain intialization | qntldoql | CFX | 4 | September 28, 2020 10:28 |
Can I achieve better convergence? | sheaker | CFX | 12 | September 19, 2019 16:36 |
Out File does not show Imbalance in % | Mmaragann | CFX | 5 | January 20, 2017 11:20 |
CFX Solver Memory Error | mike | CFX | 1 | March 19, 2008 08:22 |