Amps
[Top] [All Lists]

## Re: [Amps] 'heat spreader' sizing ??

 To: 'AMPS' Re: [Amps] 'heat spreader' sizing ?? "Dr. David Kirkby" Tue, 01 Feb 2011 23:32:06 +0000 mailto:amps@contesting.com>
 ```On 02/ 1/11 02:42 AM, Dan Sawyer wrote: > The project is to mount a 1KW, 2 transistor, board on a heat sink. The > transistor centers are about 2 inches. The question is: What should the > dimensions of the heat spreader be? I have seen dimensions of 1/4 x 2 x > 4 up to 1/2 x 4 x 7; that represents quite a range. The questions are: > > 1. What thickness is required. Is 1/4 sufficient? > > 2. What is the minimum effective distance from the edge of the devices? > 1/2 inch? 1 inch? > > Thanks, > > Dan If you can get a free trial of some finite element software for thermal analysis, this sort of thing would be relatively easy to model. You could then see the effects of changing the thickness and length/width. ANSYS is one bit of software I suspect could help with this. You can probably get a trial. You might find heat sink manufacturers could suggest solutions. I would think it quite difficult to guess this sort of thing, unless you have a lot of experience in the field. That's probably why you see widely different ranges quoted. My guess is there are diminishing returns for making this thicker and larger. But where those points are would be very difficult to calculate using anything other than finite element or finite difference software. I suspect you can make the copper too thick too. If very thick, the convection on the heatsink will be less effective. I can't believe this is rocket science, but I don't think it would be easy to determine by any method other than trying lots of different sizes/shapes (expensive) or computer modeling (cheap and relatively easy). -- A: Because it messes up the order in which people normally read text. Q: Why is top-posting such a bad thing? A: Top-posting. Q: What is the most annoying thing in e-mail? Dave _______________________________________________ Amps mailing list Amps@contesting.com http://lists.contesting.com/mailman/listinfo/amps ```