|
[Sponsors] |
October 23, 2014, 22:15 |
Interface of rotor in a duct
|
#1 |
Member
Michelle
Join Date: Aug 2014
Posts: 92
Rep Power: 12 |
Hi I am trying to create an interface of a rotor in a duct see picture, but something is going wrong. Now it says just:
| ********* WARNING ********* | | The initial values file does not contain any field solution data. | | Please check the source file carefully. | +--------------------------------------------------------------------+ +================================================= ===================+ | ****** PROBLEM REPORT ****** | |--------------------------------------------------------------------| | Subsystem: Input and Output | | Subroutine name: ErrAction | | Severity level: Fatal Error | | Error message number: 001100279 | |--------------------------------------------------------------------| | Message: | | | | Stopped in routine CHECK_INTERP | | | | | | | | | | | +================================================= ===================+ +--------------------------------------------------------------------+ | An error has occurred in cfx5solve: | | | | Error interpolating results onto the new mesh: C:\Program | | Files\ANSYS Inc\v150\CFX\bin\winnt-amd64\solver-pcmpi.exe exited | | with return code 1. | +--------------------------------------------------------------------+ +--------------------------------------------------------------------+ | The following user files have been saved in the directory F:/ANSYS | | CFX/Thesis/Duct modelling of | | fan/fancylfrozenrotorver5_pending_tasks/dp0_CFX_Solution/CFX_023: | | | | job But before I feel there is something wrong with the interface between the frozen rotor and the fluid. I created the rotor as a solid and then interface that with the fluid of mesh around it. I created the cylinder around the rotor and rotor as a part, maybe thats wrong? Kind regards, Michelle |
|
October 26, 2014, 06:02 |
|
#2 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,819
Rep Power: 144 |
The error is quite clear, read it carefully. The problem is your initial conditions file. It does not contain data. Are you sure your initial condition file is valid?
|
|
October 26, 2014, 18:25 |
|
#3 | |
Member
Michelle
Join Date: Aug 2014
Posts: 92
Rep Power: 12 |
Quote:
I am sure that have an initial condition file. But I feel like I have a problem with the interface still. I trying to create an interface between the inner cylinder and the rotor blade as a frozen rotor. Or is it the cylinder (inner and outer parts) that should be modeled as a interface? A ceiling fan in a duct, is hard to find example of. But I created another simulation of the cylinder and a huge duct, cylinder, but without interface and a rotating frame a reference. But the results shows that the flow does not change at all. Is this model possible for using to model a rotating fan in a duct? Thanks, Michelle |
||
October 26, 2014, 18:32 |
|
#4 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,819
Rep Power: 144 |
Yes, you have specified an initial conditions file, but the error says "The initial values file does not contain any field solution data. Please check the source file carefully." So the initial conditions file does not contain data.
If you run this simulation without an initial conditions file (so specify an initial condition in the domain tab) it should start fine. If it does not then there is another problem in the setup, maybe your interfaces are incorrect. And by the way CFX can model a rotating fan in a duct just fine. |
|
October 26, 2014, 19:16 |
|
#5 |
Member
Michelle
Join Date: Aug 2014
Posts: 92
Rep Power: 12 |
Thanks for your answer, will check it as soon as I get my computer. But regarding my other question about the interface and rotating or frozen rotor. Is that possible any of those ways?
kind regards, Michelle |
|
October 26, 2014, 19:35 |
|
#6 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,819
Rep Power: 144 |
I do not understand your question about the interface and frozen rotor. Please show an image explaining what you mean.
|
|
October 26, 2014, 20:08 |
|
#7 |
Member
Michelle
Join Date: Aug 2014
Posts: 92
Rep Power: 12 |
So what I mean is from the pics I posted before, the inner cylinder and rotor, I am trying to interface them as a frozen rotor. But I was thinking that it might not work to do that due to the faces of the rotor blades are not touching the smaller, inner, cylinder.
Do I maybe have to put the inner cylinder and the outer cylinder surfaces as the frozen rotor instead? Kind reagards, Michelle |
|
October 26, 2014, 21:01 |
|
#8 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,819
Rep Power: 144 |
I do not know what the inner and outer cylinder faces are. I could guess but then I would get it wrong. Please attach an image which labels these features.
|
|
October 26, 2014, 21:15 |
|
#9 |
Member
Michelle
Join Date: Aug 2014
Posts: 92
Rep Power: 12 |
Hi, Sorry now I got a pics with what I denoted as inner/outer cylinder.
What I have done is put the rotor (green) with inner cylinder as a part. Then I also tried the outer and inner cylinder as a part. But nothing seemed to give me any streamlines or result. I then thought about taking away the inner cylinder and modeled the rotor as 'rotating' and the outer cylinder as stationary, but this made it worse, the streamlines when out from the cylinder and had one velocity. Do I need to make a domain around the rotor, inner cylinder, to model the problem? And in that case how should I define the interface? Thanks so much. Michelle |
|
October 26, 2014, 21:23 |
|
#10 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,819
Rep Power: 144 |
You do not need the stationary domain if there is nothing in it which requires a stationary mesh. The walls are stationary, so you can make them counter rotating in a rotating frame of reference.
So I would make this entire model a single domain. Then you do not need interfaces with frame change. |
|
October 26, 2014, 21:38 |
|
#11 |
Member
Michelle
Join Date: Aug 2014
Posts: 92
Rep Power: 12 |
Hi, in the stationary domain, the outer cylinder is walls, like a duct. But it is air in it and also I want to simulate the rotor as a fan dragging the air through the duct. Which is why I simulate the rotor with a rpm.
What if I counter rotate the walls, would that not make the whole simulation not simulate the fan. The single domain would not simulate the rotor as a solid then, and the air as a fluid, which will make the whole simulation false. And what do you mean by interfaces with frame change? Thanks, Michelle |
|
October 26, 2014, 22:00 |
|
#12 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,819
Rep Power: 144 |
I do not think you understand what a rotating frame of reference is. The frame of reference just says whether the mesh in that domain rotates or not. It has nothing to do with what you talked about.
If you model the whole domain as a single rotating frame of reference then you define the outer wall to be counter-rotating and then it is stationary in the absolute frame of reference. You also need to define your inlet and outlet in a rotating frame. What do you mean "model the rotor as solid"? Are you trying to do heat transfer modelling the rotor? |
|
October 26, 2014, 22:40 |
|
#13 |
Member
Michelle
Join Date: Aug 2014
Posts: 92
Rep Power: 12 |
Hi,
What I do not get is if I model the walls of the outer cylinder as a rotating fram, mesh, when they are suppose to be stationary because it is not a mesh that suppose to change, its like a wall in a room. I defined the openings of the outer cylinder as two openings, which it is suppost to just be open to the atmosphere. My main point is if the inner cylinder is needed also? Kind regards, Michelle |
|
October 26, 2014, 22:46 |
|
#14 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,819
Rep Power: 144 |
Forget about the room analogy, it is not appropriate. The fluid can move, the mesh can move or both can move, they are all valid approaches.
You can model a stationary fluid with a moving mesh, you will just get the fluid moving in the opposite direction to the mesh motion (relative to the mesh). In the absolute frame of reference this still means the fluid is stationary. They thing which decides whether a rotating of stationary frame of reference is required is where the mesh is required to define something. So, to make your rotor turn around the mesh needs to rotate to make this happen. This means the rotor needs to go into a rotating frame of reference. But your duct is just a cylinder and that can be modelled as a stationary or a rotating domain (as it is axisymmetric). Then you can choose the easiest one to apply - and in your case if you make everything a rotating frame of reference then no interface reference frame changes is required. |
|
October 26, 2014, 22:55 |
|
#15 |
Member
Michelle
Join Date: Aug 2014
Posts: 92
Rep Power: 12 |
Hi,
So if i use the picture I posted, model as it looks. Just take away the inner cylinder ? I am still not sure how to model the mesh then with a motion. Is there a tutorial on this? I am sorry I have not done a rotating mesh before. Is the rotating frame of reference then applied to both the rotor and then also the walls of the 'duct'? But with different frame of reference (going different ways)? |
|
October 26, 2014, 22:57 |
|
#16 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,819
Rep Power: 144 |
Have you done the CFX tutorial examples on rotating machinery? There are quite a few.
|
|
October 26, 2014, 22:58 |
|
#17 |
Member
Michelle
Join Date: Aug 2014
Posts: 92
Rep Power: 12 |
Hi,
I have done a few, but not all of them. I get what they done it those but not in my case. michelle |
|
October 27, 2014, 17:23 |
|
#18 |
Member
Michelle
Join Date: Aug 2014
Posts: 92
Rep Power: 12 |
I was wondering if I make the mesh move, I need a rotating velocity, would that be the same as for the rotor?
Thanks, Michelle |
|
October 27, 2014, 17:29 |
|
#19 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,819
Rep Power: 144 |
If they rotate at different speeds then you still need an interface to connect them. The whole idea is you rotate the entire thing at the same speed so no interfaces are required.
|
|
October 27, 2014, 17:48 |
|
#20 |
Member
Michelle
Join Date: Aug 2014
Posts: 92
Rep Power: 12 |
Okey, thats what I thought. But if I would like to have an interface, because after this simulation I need to simulate the fan rotor in a room, would that then be best to do with an interface of a cylider domain around the rotor and an interface of that domain and the domain of the room as a stationary domain?
Thanks, michelle |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Wind turbine simulation | Saturn | CFX | 60 | July 17, 2024 05:45 |
sliding mesh problem in CFX | Saima | CFX | 46 | September 11, 2021 07:38 |
Question about heat transfer coefficient setting for CFX | Anna Tian | CFX | 1 | June 16, 2013 06:28 |
Error finding variable "THERMX" | sunilpatil | CFX | 8 | April 26, 2013 07:00 |
RPM in Wind Turbine | Pankaj | CFX | 9 | November 23, 2009 04:05 |