Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations GregLocock on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Board level thermal analysis

Status
Not open for further replies.

JordonMusser

Mechanical
Dec 22, 2006
40
Basically, I am modeling a few key components on a board and doing a thermal analysis in an enclosure. I need to model the thermal resistance between the junction and the board and junction and the ambient so that I can accurately predict junction temperature (for reliability calcs)

I have theta j-a and j-c for some components. I have the component modeled and the board (simple, just a block). I was going to use J-C, multiple by surface area and use that for contact resistance between the part and the board. Then do the same with J-A for a convection coeff on the other 5 surfaces on the part.



Is this the right way to simulate this? Or a better way? Are there good "rules of thumbs" or ways to do this? thanks
 
Replies continue below

Recommended for you

Did you get an answer to your question? The problem with theta j-a and theta j-c is that these are determined using either component on a standard board in still air or moving air. I think the best way is to model the part on the board the with a conduction element for the leads or pads, an estimate for the circuit board thermal conductivity and apply the heat load to the parts and solve. Floworks will take care of the convection term and conduction to the board and air. Once you have case temperatures determine the junction temperature using theta j-c and the device power.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor