Rotating Disc In Airflow

Summerhayes · March 13, 2014, 10:49

Hi,

I am new to Ansys and CFX and am not yet fully confident with the software.
I am attempting to model a bicycle brake disc with a heat generation that rotates in a moving airflow.

I have attempted this simulation in a number of ways, but seem to just be getting a multitude of errors each time.

If someone could outline the main steps I would need to take in layman's terms that would be great.

So far I have modelled a disc, put that disc inside of a small cylinder, and then enclosed that cylinder in a large 'environment cylinder. Then I performed a boolean subtract on the 'environment' using the small cylinder, and then a boolean subtract on the small cylinder using the disc. I created a new part from all parts and froze all parts to ensure the mesh matches at the boundaries.

The problem is within CFX I am having dificulty defining the rotation without errors, and dont have a large enough knowledge base to solve the issues.
I have given each part its own domain and set inlet outlet and wall conditions aswell as the heat generation boundary.

Yet I am still getting errors sometimes in CFX and sometimes in the solver.
Attached is a picture of my problem and the Workbench File.
Thanks in advance for any help

Spinning Disc In Air Flow.jpg

disc test 2.zip

ghorrocks · March 13, 2014, 16:39

My recommendations:
1) Start off by removing the disc solid model and just model the fluid domain. The disc surface can have a tangential velocity and a constant temperature. Then have a look at the heat fluxes over the disc. Is the rotation of the disc (which will only add convection of the heat in the disc due to rotation) going to be significant? I suspect not.
2) But if you insist on modelling the disc as a rotating body then have a look at the conjugate heat transfer examples in the CFX examples.
3) Note that convection of heat in a solid body due to rotation was only added in V14.5 I think. If your CFX is older than this you will be unable to model this anyway.

Summerhayes · March 14, 2014, 07:03

Hi Ghorrocks,
Thanks for the reply, I think I may have mis-informed on the aim of the simulation.
My project involves finding temperature profiles within discs themselves, the airflow is not the focus of the study. It has to be modelled however as the mojority of the heat is lost through convection. So the material and shape of the disc will affect how the heat transfers through the it.

You are correct that the disc spinning has a small affect on the heat loss, but it does affect the temperature profile, as you can see in the image (of a stationary disc in airflow) the temperature at the inlet side is cooler than the outlet. Obviously in practice this wouldnt happen so it needs to be modelled rotating.

The aim is to very design parameters of the disc to see which achieves the greatest thermal performance.

It is on Uni computers and we do have 14.5

I will have a look at the conjugate heat transfer tutorials though thank you

Cheers
Steve
Temperature Profile of Disc.jpg

Summerhayes · March 14, 2014, 09:30

After a quick attempt at performing the analysis without a disc, i think the problem is coming form the interface model, CFX is telling me that I must have a frame change option, I am unsure what this is and selecting any of them either gives another error or makes me provide a pitch change. Any ideas on what these are or how to go about selecting appropriate values?
It seems that when i attempt to put values in just to see if i can get it to run i get an exit code of 1 in the solver.
Hope this makes things a little clearer thanks.
Steve
No solid disc.jpg

ghorrocks · March 15, 2014, 04:19

Select automatic for the pitch ratio.

Which frame change option are you using?

Summerhayes · March 17, 2014, 04:49

I'm unsure as to which one i should use but all of them seem to give me errors.
I've attached some pictures of the issues.
Ive tried just running all of them with various options for pitch ratio but seem to just get exit codes 1 or 2 on the solver.
Frozen Rotor.jpg
No Frame Change.jpg
Stage.jpg

ghorrocks · March 17, 2014, 16:47

Then choose pitch change = none.

Summerhayes · March 21, 2014, 03:38

I couldn't get any of those options to work for me, but I managed to get the model working in fluent much more easily, and it is giving me positive results!!
Thanks for the help!!
Cheers
Steve

ghorrocks · March 21, 2014, 04:25

This should be a simple model to get working in CFX. But if it is working in Fluent then go with that.

March 13, 2014, 10:49	Rotating Disc In Airflow	#1
Summerhayes New Member Stephen Summerhayes Join Date: Mar 2014 Posts: 7 Rep Power: 12	Hi, I am new to Ansys and CFX and am not yet fully confident with the software. I am attempting to model a bicycle brake disc with a heat generation that rotates in a moving airflow. I have attempted this simulation in a number of ways, but seem to just be getting a multitude of errors each time. If someone could outline the main steps I would need to take in layman's terms that would be great. So far I have modelled a disc, put that disc inside of a small cylinder, and then enclosed that cylinder in a large 'environment cylinder. Then I performed a boolean subtract on the 'environment' using the small cylinder, and then a boolean subtract on the small cylinder using the disc. I created a new part from all parts and froze all parts to ensure the mesh matches at the boundaries. The problem is within CFX I am having dificulty defining the rotation without errors, and dont have a large enough knowledge base to solve the issues. I have given each part its own domain and set inlet outlet and wall conditions aswell as the heat generation boundary. Yet I am still getting errors sometimes in CFX and sometimes in the solver. Attached is a picture of my problem and the Workbench File. Thanks in advance for any help Spinning Disc In Air Flow.jpg disc test 2.zip

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Problem with VOF and rotating disc in Fluent	Ludvig	FLUENT	1	March 13, 2014 19:57
Rotating Disc in a Volume of Air	alcesalces	OpenFOAM	15	March 18, 2011 08:08
Rotating Disc in a Volume of Air	alcesalces	FLOW-3D	5	August 28, 2009 17:04
Rotating Disc in a Volume of Air	alcesalces	Siemens	0	August 24, 2009 13:59
Solver for a Rotatinng Disc in a Volume of Air	alcesalces	Main CFD Forum	0	August 19, 2009 14:33

March 13, 2014, 16:39		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,700 Rep Power: 143	My recommendations: 1) Start off by removing the disc solid model and just model the fluid domain. The disc surface can have a tangential velocity and a constant temperature. Then have a look at the heat fluxes over the disc. Is the rotation of the disc (which will only add convection of the heat in the disc due to rotation) going to be significant? I suspect not. 2) But if you insist on modelling the disc as a rotating body then have a look at the conjugate heat transfer examples in the CFX examples. 3) Note that convection of heat in a solid body due to rotation was only added in V14.5 I think. If your CFX is older than this you will be unable to model this anyway.

March 14, 2014, 07:03		#3
Summerhayes New Member Stephen Summerhayes Join Date: Mar 2014 Posts: 7 Rep Power: 12	Hi Ghorrocks, Thanks for the reply, I think I may have mis-informed on the aim of the simulation. My project involves finding temperature profiles within discs themselves, the airflow is not the focus of the study. It has to be modelled however as the mojority of the heat is lost through convection. So the material and shape of the disc will affect how the heat transfers through the it. You are correct that the disc spinning has a small affect on the heat loss, but it does affect the temperature profile, as you can see in the image (of a stationary disc in airflow) the temperature at the inlet side is cooler than the outlet. Obviously in practice this wouldnt happen so it needs to be modelled rotating. The aim is to very design parameters of the disc to see which achieves the greatest thermal performance. It is on Uni computers and we do have 14.5 I will have a look at the conjugate heat transfer tutorials though thank you Cheers Steve Temperature Profile of Disc.jpg

March 14, 2014, 09:30		#4
Summerhayes New Member Stephen Summerhayes Join Date: Mar 2014 Posts: 7 Rep Power: 12	After a quick attempt at performing the analysis without a disc, i think the problem is coming form the interface model, CFX is telling me that I must have a frame change option, I am unsure what this is and selecting any of them either gives another error or makes me provide a pitch change. Any ideas on what these are or how to go about selecting appropriate values? It seems that when i attempt to put values in just to see if i can get it to run i get an exit code of 1 in the solver. Hope this makes things a little clearer thanks. Steve No solid disc.jpg

March 15, 2014, 04:19		#5
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,700 Rep Power: 143	Select automatic for the pitch ratio. Which frame change option are you using?

March 17, 2014, 04:49		#6
Summerhayes New Member Stephen Summerhayes Join Date: Mar 2014 Posts: 7 Rep Power: 12	I'm unsure as to which one i should use but all of them seem to give me errors. I've attached some pictures of the issues. Ive tried just running all of them with various options for pitch ratio but seem to just get exit codes 1 or 2 on the solver. Frozen Rotor.jpg No Frame Change.jpg Stage.jpg

March 17, 2014, 16:47		#7
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,700 Rep Power: 143	Then choose pitch change = none.

March 21, 2014, 03:38		#8
Summerhayes New Member Stephen Summerhayes Join Date: Mar 2014 Posts: 7 Rep Power: 12	I couldn't get any of those options to work for me, but I managed to get the model working in fluent much more easily, and it is giving me positive results!! Thanks for the help!! Cheers Steve

March 21, 2014, 04:25		#9
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,700 Rep Power: 143	This should be a simple model to get working in CFX. But if it is working in Fluent then go with that.