Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

# Thermal resistance of a given copper area

RF Lurker

Thermal resistance of a given copper area | 3 June, 2003

Does anyone have a method to calculate the thermal resistance of a given copper area?

Thermal resistance of a given copper area | 3 June, 2003

It's always good to have the lurkers out in the open. Welcome.

Being uncertain about what you seek, lets take a couple of tacts: * For resistivity, look here: http://www.smtnet.com/Forums/Index.cfm?CFApp=1&Message_ID=24595

* For resistance to damage by heating, look here: http://www.ipc.org/4.0_Knowledge/4.1_Standards/test/2.4.13f.pdf

RF Lurker

Thermal resistance of a given copper area | 6 June, 2003

I understand that there may not be a lot of material on this, as most threads I've seen discourage doing it. It goes against DFM to do it. But I wish to use a large copper area instead of heat sink and thermal epoxy to cool a regulator. But I can't find a reference to calculate the thermal resistance of the heat sink (ie. the big copper pad). I suspect that there is conduction and radiation. I can get the thermal resistivity of air, but I can't even guess at the effective hight of the boundry layer. It would be great if there was a chart/graph to relate copper area to thermal resistance (in C/W). But I haven't found it. Well I found one, but it was linear, and I'm sure the relationship is quite nonlinear (thermal resistance goes to inf when area goes to 0). Know what I mean?

Thermal resistance of a given copper area | 6 June, 2003

No, no. It's koscher to have thermal planes. You just need to: * Work with your fabricator to keep the layup balanced. * Relieve the plane so that it doesn't cause assembly problems.

While probably not directly for your part, it gives a starting point. Goto http://www.national.com/dt/LM317MDT_Qualification_Package.pdf

Page 8, HEATSINKING TO-263, SOT-223 AND TO-252 PACKAGE PARTS. Look at Table 1 for top and bottom side heat sink areas, Figures 12 and 13 for heat sink patterns, and the dissipation graphs.

We'd much rather see you do this well than force a heat sink, like http://www.smtnet.com/Forums/Index.cfm?CFApp=1&Message_ID=24717)