Heatsink compounds.
 

Hi Peeps.. I am sure this has been discussed before, but.....I have just built a HF SSB Linear Pa. Supposedly 180W....This module kit has 2 X IRFP250N devices, but as usual they do not supply heatsinks etc. I built an assembly.. see photo, and included a fan for good measure. However..to isolate the devices from ground I used those grey plastic/rubber/whatever pads.
Do I still use the white compound as a thermal interface or should these be left clean ?.
I have had to do a rebuild as in test a couple of days ago, the PA went S/C.. one Fet went S/c the other O/c. I was probably over driving them from my TX. I am going top reduce the supply from 13.5V to 12.5v in case it was an over volting problem,
I bought replacement devices from Farnell so they should be ok.
Your views would be most welcome.

Sorry Mod.. Maybe I should have put this in.. Homebrew.. Please move if you wish

Re: Heatsink compounds.
 

Hi Wendy,
As far as I was aware, those grey rubbery insulators were not designed to need heatsink compound. I suspect they were designed to speed up production. When bolted down, the device sort of sinks into them so I imagine the rubber provides sufficient thermal contact.

However, I can't imagine it would reduce their efficiency if you did put a touch of compound on both sides.

Cheers
Nick

Re: Heatsink compounds.
 

There is no advantage to be gained with compound on those grey pads.
The ISO 220 package goes on without a pad and does need compound.
See if there is such a thing as a IRFP250FI available to see what one looks like from the data sheet. Farnell often show a photo if they stock them.

Re: Heatsink compounds.
 

Hi Nick and Refugee. Set up the new FETS as per instructions.. i.e 20 m/A bias per device.. on 80M was getting 50W RF with a two tone generator.. Lovely linear RF output. @ 12.5A DC
Let the PA run for 5 mins..... marginal heating of heatsink. Changed band to 40M. Again reasonable drive waveforms until............switch off..... re applied TX control signal. PA current went to 6A then OFF scale and PSU went into current limit.. devices dead as before.
No sign of "self oscillation".
I am wondering if the 40M LPF has an issue. as it was on 40 previously that the PA died.
Getting expensive 4 FETS dead..dont want to kill anymore.

Re: Heatsink compounds.
 

A "round tuit" project on the list is to replace the rather puny LED light in the cooker-hood with this one, which has a warmer colour temperature and is much brighter.

I bought this little tube of heat sink compound which is also an adhesive, and intend to use a piece of the aluminium extrusion carpet/flooring off-cuts shown as the heatsink.

Can anyone tell me what the brass/gold surface is, and whether is is worthwhile removing it? I can tell you that it is electrically conductive.

Hope I'm not hijacking the thread too much.

Re: Heatsink compounds.
 

The brass/gold is anodising-with-a-suitable-gold-dye-in-the-anodising-bath.

No point removing it; most of the 'black' aluminium heatsinks you come across are anodized!