Author Topic: Electrically conductive thermal pads.  (Read 1372 times)

0 Members and 1 Guest are viewing this topic.

Offline paul8fTopic starter

  • Regular Contributor
  • *
  • Posts: 112
  • Country: ie
Electrically conductive thermal pads.
« on: February 02, 2021, 11:27:07 am »
Hi all,

couldn't figure out why a device was functioning fine, but was drawing more p/s current, and was working a little hotter than it should have been.

Turns out an electrically conductive thermal pad was used on the bank of TO-220 tabs, instead of insulating thermal pads. The pad had a measured resistance per inch of a few kOhm.

So my question is.... what scenario or what type of products call for a conductive (or semi-conductive) thermal pad like this? I knew these pads existed, but never encountered them before. Maybe the pad in question might be good for ESD purposes allowing a slow drain to gnd??!    :-DD
 

Offline penfold

  • Frequent Contributor
  • **
  • Posts: 675
  • Country: gb
Re: Electrically conductive thermal pads.
« Reply #1 on: February 02, 2021, 12:35:45 pm »
The non-electrically isolating gap pads can have much better thermal performance and can get away with being much thinner etc so are useful when you already have electrical isolation via the package, something like that fully encapsulated TO-220 variant I can't remember the name of or one where the IC is bonded to an aluminium oxide insulator like most IGBT modules tend to be, or even microcontrollers/processors and misc ICs etc

There is a variety of silicone based pad (I think) made by Bergquist/Henkel, intended for both electrical and thermal conductivity, the performance of both relies on pretty high mechanical pressures so it wouldn't measure particularly well when not compressed.
 

Offline paul8fTopic starter

  • Regular Contributor
  • *
  • Posts: 112
  • Country: ie
Re: Electrically conductive thermal pads.
« Reply #2 on: February 03, 2021, 11:03:46 am »
I think you just hit the nail on the head with your answer there!

The pad material was a lot thinner than the usual stuff, and when combined with high pressure mechanical fixing, I can see how thermal performance would be much higher. :-+

Electrical isolation via the packages would be one scenario. I suppose another scenario would be where all TO-220 devices in the bank were identical (e.g. all were transistors using their tabs as common collectors). All collectors would be bolted to a stand-alone heatsink (or cooling aggregate), so any parallel resistance (due to the conductive thermal pad) between the transistor tabs could be ignored? The conductive pad in this case might be a wise substitute for messy thermal-paste??


 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf