CFD Online Discussion Forums

CFD Online Discussion Forums (
-   CFX (
-   -   linear layer thickness evolution (

keeper July 26, 2013 03:25

linear layer thickness evolution
Hi all.

I want to include thin material as an thermal resistance but I would like to define linear dependence by the z coordinate. But CFX warns that the value of thin material is defined to be single value.

Is there any option how to set it up properly to have linear dependence for thermal resistance?

Thank you for all your replies.

Best Regards

ghorrocks July 27, 2013 07:18

You will probably have to do this manually using source terms. Write a source term to handle the temperature difference over the interface, then you should be able to make it a function of spacial directions.

keeper July 29, 2013 04:51

Could you please explain little more what you mean? I donīt have any information about temperatures. I have only data for isolation layer so I would like to add it into the model...

Thank you for reply


ghorrocks July 29, 2013 07:55

If you model the thin surface as an interface you can put a thermal resistance on the interface. Have you tried this? Hopefully you can specify the thermal resistance as a function of position.

keeper July 29, 2013 07:59

that is exactly what I did. I defined material for isolation and to define thermal resistance, I used thin surface option.
I used linear dependence for definition of layer thickness and I put it as an expression to layer thickness determination.
And then I get message that it is supposed to be single value...
That is why I am wondering that I am able to use expression but not any spatial dependence..?

ghorrocks July 29, 2013 08:14

Can you post your CEL and the error message you are getting?

keeper July 29, 2013 08:56

1 Attachment(s)
sure, here it is..I just have so many domains and interfaces in the model that I attached relevant information only. Please note that I get this error message in CFX Pre.

ghorrocks July 29, 2013 21:53

Rather than use "Z Coordinate" just use "z".

keeper July 30, 2013 02:46

I used just "z" and it is the same. I was thinking about any expert parameter that changes the "Single Value" definiton that is mentioned in error message...

ghorrocks July 30, 2013 06:34

In that case I think you will find spatially varying thermal resistance has not been implemented so that is why you are being stopped from using it.

keeper July 30, 2013 06:57

that is exactly what I was afraid of :) but anyway thank you for your help

ghorrocks July 30, 2013 07:37

I do not understand why, there are not any high order terms created with spatially varying thermal resistance to my knowledge so I would have thought it easy to implement. It might be worth speaking to ANSYS support on this as there might be a simple work-around.

You might also be able to implement a thermal resistance as a source term. Then you will have more control over it and might be able to implement the spatially varying term.

keeper July 30, 2013 07:46

yes I had the same idea, it should be simple to define by some expression. So I will try to define it by thermal resistance source term option as you advised. thank you very much

All times are GMT -4. The time now is 16:15.