My question is for the amount of rise. It's my understanding that the rating is based on a amount of rise that the manufacturer chooses when testing (i.e. 50C rise from ambient of 25C). Or am I mistaken? So if you wanted your heatsinks to run cooler than the spec'd rise, you'd need to take this into account.
As an example, this manufacturer specs at a 70c rise and provides a correction factor for lower rises.
As an example, this manufacturer specs at a 70c rise and provides a correction factor for lower rises.
Last edited:
You can push things as far as the device junction temperature will allow,
to minimize the derating factor, but a sink temperature of around 60C is as high as most here would go.
Also remember that the Rth is based on a uniform thermal loading of the entire useable sink surface.
Don't we still need the temp rise the heatsinks are spec'd at? If I make my calcs for a max temp of 60c and assume ambient of 25c, I need c/w spec'd for 35c rise. If it's spec'd at 70c rise, then I need to adjust. Without knowing this, how do I select a chassis and it's related heatsinks other than trial and error?
Aavid gives a correction factor, but it's still one of a number of variables, like the device thermal interface,
how flat the mounting surfaces are, torquing of the mounting bolts, usage factor of sink surface, etc.
The temperature will vary significantly over the sink surface, even with a long stabilization time.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.