Continue to Site

Welcome to MCAD Central

Join our MCAD Central community forums, the largest resource for MCAD (Mechanical Computer-Aided Design) professionals, including files, forums, jobs, articles, calendar, and more.

Heat Transfer

BobFrom Pa

New member
I am using WF4. I have a cylinder wraped in insulation. The temp on the metal ID is 500 F. I applied a film coeficient to the insulation surface. When I run the thermal analysis, there is a temp gradient in the insulation but not the cylinder wall. The 500F goes straight through to the insulation ID.


I modeled the geometry as an assembly because of the different materials. Do I need to specify something at the interface?


Thanks fro any help!
 
You do have three different types of interface options;
bonded, adiabatic, and thermal resistance. You may want to
test them to see if anything looks better if they meet your
criteria. I believe bonded is the default. Adiabatic is
probably not going to help you out.

You may want to double check your material properties if
you haven't already...
 
Hi BobFrom Pa,


Make sure the cylinder and insulation components are "bonded" in the assembly. This means in Pro/E the OD of the cylinder component must be exactly the same as the ID of the insulation component.


You can check this in WF 4: click Info, Model Connectivity to see the Connectivity Types (Bonded Interface or Thermal Resistance).


Or, in all versions, you can check element connectivity. Show the elements in the results window and make sure the elements are connected across the component boundaries. i.e. element corners and element edges match up across the component boundaries.


But, is it not unreasonable to expect that the cylinder have no temperature gradient compared with the insulation? Isn't that the purpose of "insulation"? To have a smaller thermal conductivity compared with the metal?


Hope that helps,
Ray.
---
Ray Miya
 
I would assume that cylinders thermal conductivity is fewdecades higher than in insulation - therefore the gradient in cylinder will be very small.
Edited by: mhinkkanen
 
Thanks for the responses. As it turns out, the geometry was OK. The problem was with the film coefficient units. Everything worked out OK after that.
 

Sponsor

Back
Top