Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Transistor Insulator.

Status
Not open for further replies.

Toe Cutter

New Member
Hello. I am replacing some TO-247 transistors and I am trying to figure out what type of insulator to use. It looks like the old insulator was made of Kapton with a waxy coating. It measures 22x28 mm. I'm guessing the waxy coating is intended to melt when the transistor heats up and create a better thermal interface. A phase-change material?

Does anyone know who makes insulators like this? I've looked at mouser, avid, and others. The closest I could find was Mouser p/n: 532-43-77-20, but no mention of any coating.

If I buy Kapton-type insulators _without_ the waxy coating should I use conventional silicon thermal grease?

I attached a picture of a transistor and insulator I pulled. Notice how the white waxy stuff adheres to the transistor. Thanks.
 
Bergquist Sil Pads from Digikey

We use their relatively cost effective Sil Pad 900-S.

I've also experimented with their Kapton Sil Pad K10's. They offer better heat transfer and 20°C lower temperatures in our application but cost three times as much.

The manufacturer also offers these pads with an adhesive backing for mass production. I suspect that's what you're seeing left on the back of your device.
 
Last edited:
Thanks. I did some more searching and think I found the part. It might be Laird Technologies K52: "...insulator pad consisting of a ceramic filled phase change compound coated on MT Kapton film...". Unfortunatly, they don't seem to distribute this insulator to anyone.
 
Thanks. I did some more searching and think I found the part. It might be Laird Technologies K52: "...insulator pad consisting of a ceramic filled phase change compound coated on MT Kapton film...". Unfortunatly, they don't seem to distribute this insulator to anyone.

Who use them, then?
 
it's not uncommon for an equipment manufacturer to ask a part manufacturer for a part that meets certain specifications. if the part is successful, the OEM may contract with the part manufacturer to ONLY make it for them for a certain period of time. after the contract expires, the part manufacturer is free to market the part, but often it is a "niche" part and becomes discontinued, or remains in very limited production, since the uses of the part meet a very narrow range of requirements. sometimes the manufacturing processes are rather expensive, making the part too expensive for mass marketing.

the Kapton insulator in the picture looks like it was coated with regular heat sink grease, which feels waxy after it has dried out.
 
Last edited:
Bergquist also makes Hi-Flow. Call for samples. I've found them to be pretty accomodating for a cut to size sample if you represent an actual business.

Of course you'll then have to find something like a leather punch to make the hole but they're cheap enough and not particularly hard to find.

BTW, if you're up to anything serious and haven't started running your own business of sorts, I recommend it. The tax advantages are very often worth it.
 
Unless you're working with very high voltage I would have thought almost any very thin insulating material (e.g. plumbers' ptfe tape) would do, providing it's used in conjunction with a thermally-conducting grease. The important thing is to eliminate air spaces, air being such a poor thermal conductor. The variation in thermal conductivity between different solids is insignificant compared to the conductivity difference between air and a solid.
 
I don´t think teflon tape is much better at conducting heat than air, also the plumber´s tape is very thin so it might get ruptured easily.
I would use standard silicon or mica pads.
 
Plastics (I have no figures for teflon specifically) have a conductivity about half that of mica but 20 x higher than air.
 
True, but if that is indeed a phase change compound then it is likely high performance.

High performance may well be needed in this application.

My experience with electrical tape and PTFE plumbers tape has been, well, less than stellar. If it has good thermal properties like PTFE plumbers tape, then it doesn't isolate well enough to use above 24 volts or so. It it has good dielectric strength like electrical tape, then you are going to have to use what amounts to an infinite heat sink.

To be honest, I've only tried PTFE plumbers tape twice and electrical tape once. If anybody has any secret homebrew electrically insulating but thermally conducting miracle materal though, I'd love to hear about it.

What is your guesstimate of how many watts you need to dissipate, the voltage it's operating at, and the dimensions of the heat sink?
 
Last edited:
Good questions, but I don't know the answers. The igbt is out of an inverter, it might see 1000 volts peak on the collector, maybe 50+ watts, smallish heat sink. This is repair I am doing, but may do more in the future. Laird's web site says they make To-247 insulators as a standard part, but it seems they don't distribute anywhere. Although, Digikey has a 11"x17" sheet of the stuff. Here's the blurb off of their web site:

Tgard™ K52 is a high thermal and dielectric performance insulator pad consisting of a ceramic filled phase change compound coated on MT Kapton film.
Tgard™ K52 phase change coating all but eliminates contact thermal resistance. The phase change coating melts at 52ºC and replaces all contact areas that contain air. Tgard™ K52-1 is ideal for applications requiring the best thermal performing insulator material.

Interesting replies, thanks. I think I'll try what has been suggested or what I can find on Mouser with silicone grease. I might try monitoring the transistor temperature with a probe. If it doesn't get too hot, it should be a good enough insulator.
 
If it has good thermal properties like PTFE plumbers tape, then it doesn't isolate well enough to use above 24 volts or so.
Useful to have a figure for it's voltage limits.
 
hi,
Is this the item you are looking for.??

**broken link removed**
 
The phase change coating melts at 52ºC and replaces all contact areas that contain air.
But how? If the insulator is clamped between a transistor and a heatsink how is the air in the surface imperfections going to escape? Won't it just be trapped as bubbles in a gooey liquid?
 
some manufacturers seem to like using thin pieces of aluminum oxide as heat sink insulators. Klipsch uses them on their class D amp FETs. because AlO2 is not flexible, they seem to use quite a bit of heat sink grease too..., but AlO2 does conduct heat much better than mica, and is a better electrical insulator. BeO2 (beryllium oxide) has even better thermal characteristics, but is poisonous.
 
Useful to have a figure for it's voltage limits.

To be 100% honest, I don't really know what the limit of PTFE plumbers tape is. I've tried some at a bit over 24V in a homebrew bench top power supply and it worked. We tried it on a few controllers at 170V plus flyback from an inductive load and it didn't.
 
Ah well, somewhere between the two then :).
 
PTFE is listed as 1500V/mil breakdown voltage, but plumber's tape is very pliable, and easily punched through by any irregularities in the surface. so unless the surfaces are very smooth, and the mounting pressure of the transistor carefully controlled, the plumber's tape can punch through (or deform to a point where there's bound to be a spot much thinner than 1 mil)
 
I did a little more checking and I believe that's for sintered PTFE.

PTFE plumbers tape is generally unsintered and much more porous.

The number I stumbled upon was something like 6000V/mm = 150V/mil for unsintered.

Maybe if we had done a triple layer of plumbers tape it would have worked at 170V.
 
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top