Mica versus Silicone pad insulators

yes

you have too much time on your hands :)

only joking honest

Ron

The *heatsink* temperature won't be affected by the choice of insulator one tiny bit.

You actually need to measure the *device* temperature to determine how effective the heat transfer washers are. I don't know how well calibrated your fingertips are !

Graham

Your fingertip response change is very steep in the area 32-33 C.

-- John

Silicone pads are terrible thermally.

My finger is pretty well calibrated from 50 to 60C. I can stand 50C forever, 60C for about one second, and interpolate between.

John

The cheap silicone pads are terrible. There are however some very good silicone pads too.

The white ones that the OP mentioned he had in another post are normally higher performance than mica+grease.

Graham

My favorite coffee temperature :)

D from BC

silicone

higher

Yes white ones I'm wondering if they can chemically change over time and/or excessive temperature , downgrading to be more of a thermal insuilator. They are not discoloured or hardened or anything different in the ex-compressed area by sight or flexing, just permanetly deformed , the ruffling is permanent.

-- Diverse Devices, Southampton, England electronic hints and repair briefs , schematics/manuals list on

Sounds normal to me.

Because they're quite soft, these silicone washers will flow slightly under pressure and conform to the heatsink and device surfaces very well. This aids the thermal transfer. It also explains the ruffled edges.

Graham

Is it possible that it's been caused by some kind of attack by a chemical substance, silicon grease? Maybe some clown has sprayed something like WD40 all over the inside. People really do that you know!

Ron(UK)

I didn't quite follow your entire post, but I would have used a IR thermometer pointed on the devices, then the heatsink, then the room. I never thought silicone was better than mica, except in the case of a poor fit, or warped pieces. I have also tried to sand down surfaces that were not flush, and even remove the oxide on aluminum, to optimize transfer.

grge

I prefer silicon pads in high power amps both RF and audio as do quite a few manufacturers.

When building stuff in the west, the reduction in labour with 'grease-free' assembly is a not inconsiderable cost saving. It should pay for the extra cost of a decent performance silicone pad.

Graham

The previous post, that said it was the transistor temperature that would be variable, was correct; remember, the whole heat output is dumped into the heatsink, and that isn't affected by the thermal resistance of your mica. Heatsink-to-air temperature difference is whatever it takes to drain that heat.

It is possible to fit a second heatsink onto the TOP of the transistor, if you really need better thermal transfer (using both bottom and top surfaces of the metal plate).

What the heck *WAS* TOP66 ? I'd have thought that was a plastic TO-66 size device.

I'm sure those devices are in some other package.such as TO247, TO264 (TO-3P) whatever

Graham

It's interesting that the tongue is the most temperature-sensitive part of the human body, and the most resistant to temperature damage.

John

...

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

That's only because it's the wettest part that ever sees the outside. ;-)

I heard some standup comic talking about hot food - "It's too hot to pick up with my hands, so I'm going to put it into my mouth?" ;-)

Ever blistered your palate with hot pizza? ;-)

Cheers! Rich

Yup; the silicone is too thick. And the vendors measure theta with insane clamping forces applied, then lie by another 2:1 for good measure.

John

The interface between the device and the heatsink is probably the most critical piece in the puzzle, and is the piece that a builder/repairer has most control over. The device temperature is what you're trying to control. You have no control over the internal thermal impedances of the device... they are determined by the manufacturing process and materials. Given a heatsink has adequate capacity for dissipating the heat (heatsink to air thermal impedance), then that only leaves the device to heatsink interface to manipulate. In comparison between greased mica and silpad insulators, the mica/grease combination has the better thermal characteristics. Mica can be problematic though, mainly because there seems to be no quality control on the manufacturing process. The thickness of a mica slab can vary from 0.010 to 0.050 or more. Mica can withstand about 1000 volts per mil, so it really doesn't need to be so thick. You can usually split a mica slab several times and still be safe in using it. The thermal impedance goes down somewhat with thinner slabs, too.

Overall, mica/grease can exhibit almost half the thermal impedance of a silpad. Given the same device power dissipation, lower thermal impedances mean lower device temperature.

And this beats mica+grease hands down.

Graham