SOL 153: Temperature dependent convection coefficients using fields

Valued Contributor
Valued Contributor

Hi everybody,


first of all: happy new year!


To my problem: I wanted to calculate the temperature distribution in a pin fin. There is of course an analytical solution. However, this solution requires a constant  convection coefficient. So I wanted to use NX NASTRAN 12 and defined a field in order to achieve a temperature dependent convection coefficient.


field_constant.pngField defining a temperature depending convection coefficient - values outside constant


I did not think about what should happen with values outside the table and left it on "Constant". The results calculated using NX 12 - SOL 153 - were wrong ... (I compared them with the results of a Simcenter Thermal / Flow simulation). So I just wanted to see what card was sent to NASTRAN and realized that a constant convection coefficient was sent to NASTRAN.


field_constant_synatx_preview.pngThis is a constant convection coefficient

I was confused and tried this and that. Switching to "Extrapolate" for values outisde table changed nothing ... Then I switched to "Undefined" and achieved a temperature dependent convection coefficient.


field_undefined.pngField defining a temperature depending convection coefficient - values outside undefined

field_undefined_syntax_preview.pngThis is a temperature dependent convection coefficient

 To be honest, I didn't understand this behavior at all. Nevertheless the obtained results were correct (compared to the Thermal / Flow simulation).


To understand the topic I continued to try this and that. Without any technical sense I changed the first temperature in my field from 200°C to 10°C. I didn't change "Undefined".

field_undefined_10.pngField with "Undefined" an 10°C as first temperature value


 To my complete confusion, I again received a constant convection coefficient.

field_undefined_syntax_10_preview.pngAgain a constant convection coefficient

To be honest again I'm completely confused. I tried another first temperatures and found out that one receives constant convection coefficients up to a temperature of 20°C. For higher temperatures I always got temperature dependent convection coefficients. To emphasize, I have changed nothing else, just the first temperature.


The behavior described above is absolutely reproducible.


Does anybody know this behavior? Why do "Constant" and "Extrapolate" not work? Why does "Undefined" work, provided that the first temperature is higher than 20 ° C? Why does "Undefined" not work for temperatures less than or equal to 20 ° C?


Am I doing anything wrong? If yes, what? Does anyone have an idea?
I would be very grateful for any hint.


I've attached my model.


Best regards