Learning Forum

[RobinC]

Hi everyone, 

For my study I want to study the temperature evolution inside a wire of copper du to an electric current. So the case is quite simple, I have :

   Created a cylinder

   Imported that cylinder in the "thermoelectric" module

   Assigned copper as material 

   Choosen a 0 V voltage for a face and an electric current of 350 A at the opposite. 

   Choosen radiation for the "temperature" condition. The environment temperature is set at 10°C. 

350 A is a high value, one would expect an important rise of the temperature. Que nenni, the max temperature is lower than 11°C! 

I've tried different things to fix this issue, as :

Changing the radiation model for a temperature model (doesn't work).

Changing current for voltage, with only 1 V the max temperature reached 2080°C. Obviously a wrong result too. 

So I was wondering if one of you already had this kind of issue and if a solution has been found I'm curious too know what I'm doing wrong. 

Thanks in advance!

[jj77]

Simple test here. 0.5 m long steel wire with 1E-3 V on one and 0 V on the opposite face end. With tiny convection BC on all 3 faces of 1E-4 and bulk temp of 22 for the conv.

 

Voltage is 1E-3 on one end and 0 on other.

Dir. electric file int. Ez = -dV/dz = .001V/0.5 m = -.002 V/m, that is what ansys gives.

Dir current density: Jz=sigmaz*Ez = 1/(1.7E-7)*-.002= -1.1765E4 A/m2, that is what ansys gives

Joule heating = Jz^2*resistivity = (-1.1765E4)^2*1.7E-7 = 23.53 W/m2, that is what ansys gives.

Temp. rise is from 22 ambient to 314, that is correct answer (just add the internal heat on a pure steady state heat transfer run, and one gets same)

 

 

 

 

[RobinC]

Thanks a lot for your quick and accurate answer! 

The main issue is to understand why you've picked 1 mV. I mean I'm supposed to have a electric current of 350 A so this value seems realy low in comparison. 

Moreover I've the same values (in order of magnitude) for Ez, Jz and the joule heating but the temperature remains too high, 1e5 °C. Maybe I plot the wrong thing, but I don"t think so. 

If you want to check I can share a link with the project, I don't know if I can share it directly here. 

Thank you one more time 

[jj77]

Simple reason, linear regime, where hopefully all the material properties are not varying to much and perhaps are equal to the constant value provided in eng. data. But again I do not know that since the temperature increase was quite large (if convection is changed to normal values, then the temp rise is lower of course)

 

If temperatures gets higher, all material properties (resistivity, ….) will change, things are then nonlinear and thus more complex. I am not an expert in that.

But in any case if you want to do that, I would suggest that you find some experiments/data to compare with, or if you find some other published data on this.
A good ansys verification benchmark is vm119

 

Good luck with everything. That's it from me