I was wondering if nobody here has ever heard of Kapton insulators. This is a sort of plastics often used for voice coil formers on speakers. I just collected some thermal data from the German company "Fischer Elektronik", they manufacture insulators of a lot of materials. Given a TO-247 footprint, these are the thermal resistances I have found:
Mica: 0,4 K/W
Silicone: 0,4 - 0,96 K/W
Aluminum oxide: 0,3 K/W
Kapton: 0,07K/W
I use Kapton insulators for many years now, and I have a lot less problems then before.
Mica: 0,4 K/W
Silicone: 0,4 - 0,96 K/W
Aluminum oxide: 0,3 K/W
Kapton: 0,07K/W
I use Kapton insulators for many years now, and I have a lot less problems then before.
Yup, that's the yellow film commonly used in car amps. It seems to work fine as well. I seem to remember it rips easily, but is otherwise resistant to damage. It is also used with heatsink compound (grease).
With any film you want to make sure nothing sharp "pokes" through to ruin your day.
-Chris
With any film you want to make sure nothing sharp "pokes" through to ruin your day.
-Chris
Nelson Pass said:
"I like to think of bead blasting as creating a fractal surface
which in fact has a larger surface area than described by
the square of it's boundary dimensions. Kind of like the coast
of England or the fabric of space at the sub-atomic scale."
No argument on that theory... I've just found by experience that better intimate contact is more efficient. Perhaps you can tell me why???
Damon Hill said:
"Another possibility is simply increasing the thickness of the
heatsinking anodizing and mounting the device directly"
This can easily be done by having the sinks hard coat anodized instead of a simple anodize. Any standard hard coating process will penetrate the metal a ten thousandth as well as build up another tenth on the surface, there are some other hard coat process that build up even more thickness up to .003 or so. Hard coating is also far more fade resistant , scratch resistant and very color stable over long time periods. Standard anodize even when properly sealed will eventually fade out to a bluish or bronze or some other color, partly on its own and partly from exposure to light. Hard coating anodize is one of the toughest coatings known to man and can be literally diamond hard. Normal dielectric strength of standard hard coating is about 2kv. So..... I have yet to have my Aleph 2's done and am seriously considering just hard coating the sinks and chassis parts black and throwing the insulators away. I haven't yet thought of a reason not to do it this way.
Mark
"I like to think of bead blasting as creating a fractal surface
which in fact has a larger surface area than described by
the square of it's boundary dimensions. Kind of like the coast
of England or the fabric of space at the sub-atomic scale."
No argument on that theory... I've just found by experience that better intimate contact is more efficient. Perhaps you can tell me why???
Damon Hill said:
"Another possibility is simply increasing the thickness of the
heatsinking anodizing and mounting the device directly"
This can easily be done by having the sinks hard coat anodized instead of a simple anodize. Any standard hard coating process will penetrate the metal a ten thousandth as well as build up another tenth on the surface, there are some other hard coat process that build up even more thickness up to .003 or so. Hard coating is also far more fade resistant , scratch resistant and very color stable over long time periods. Standard anodize even when properly sealed will eventually fade out to a bluish or bronze or some other color, partly on its own and partly from exposure to light. Hard coating anodize is one of the toughest coatings known to man and can be literally diamond hard. Normal dielectric strength of standard hard coating is about 2kv. So..... I have yet to have my Aleph 2's done and am seriously considering just hard coating the sinks and chassis parts black and throwing the insulators away. I haven't yet thought of a reason not to do it this way.
Mark
Netlist said:
I recall reading something about thermal isolation that had the best of both worlds. Must have been an old Elector magazine.
It appeared to be some kind of silpad that had the ability to become softer when heat was applied. The principle was to burn in the transistor or mosfet or whatever had to be isolated and after a period of time turn on the screws to tighten the contact with the heatsink.
/Hugo
You would be talking about What Laird sell as Thermaphase. It is the as far as I know best solution, as its easy to deal with and gives you the goo effect. Laird offers an very interesting product that Ive been using for a while with great results, Thermaphase on Kapton film, It have all the properties one could ever ask for, as its easy to handle, dosnt flow all over the place, got low thermal resistance (better than any of the other easy solutions) and is cheap.
Magura
You can buy Kapton tape from Digikey.
If you can find the page in their catalog where the SilPads are, the same company also makes Kapton based pad. The product name is somethinh like "K-10", anyway, it has a "K" in it. They are not cheap comparred to SilPads but since the tape doesn't deform much, the Kapton pads are probablt better far all but the flaest surfaces.
I've use all of these with no problem, but only in Class-B. Class-A people probably need to be more finicky.
If you can find the page in their catalog where the SilPads are, the same company also makes Kapton based pad. The product name is somethinh like "K-10", anyway, it has a "K" in it. They are not cheap comparred to SilPads but since the tape doesn't deform much, the Kapton pads are probablt better far all but the flaest surfaces.
I've use all of these with no problem, but only in Class-B. Class-A people probably need to be more finicky.
I am careful to deburr the heat sink when using silpads. A good sharp burr will even do a number on mica. I have also used aluminum oxide (alumina) washers in the past, but torque is really critical because they crack so easily. I wouldn't use them in a production environment without a precision torque wrench. I'm refurbishing one of my old homebrews right now, and I replaced all the alumina under the TO-220s with mica because the alumina were all cracked.
A good way to apply grease to an insulator or device is to smear some on a round toothpick (or dip it) and roll it on.A wooden Q-tip handle works well, too. This technique gives you just enough grease for a good thermal interface without having it squish out everywhere when you tighten down the hardware. BTW, I always called the stuff bird dung (insert other expletive as desired).
A good way to apply grease to an insulator or device is to smear some on a round toothpick (or dip it) and roll it on.A wooden Q-tip handle works well, too. This technique gives you just enough grease for a good thermal interface without having it squish out everywhere when you tighten down the hardware. BTW, I always called the stuff bird dung (insert other expletive as desired).
Andypairo said:
In Poland such pads are available without problems.
For example for transistor case TO218/TO247 ( IRFP044 for example) such pad which has 3mm thickness and has price about
0.9$ with tax for one unit if You take only one and 0.85$ with tax for one unit if You take 5 units. I utilized them in my Aleph3 in one transistor output version ( something like mini-Aleph) where IRFP044N issues over 40W.
With mica and silikon I had problems with case temperature of transistors. With mentioned pads none problems.
Regards
Jacek
HBarske said:
If we compare only thermal resistance there should be almost none differences between alumnium oxide and mica. But difference at least in my circuits is huge !. I think that very important is thermal conductivity. Alumnium oxide has about 24W(m*K) , best material as beryl oxide almost 201W(m*K), kapton and others significanty less.
Regards
Jacek
Kapton insulators need goop of some kind to fill the voids between insulator and heat sink, as kapton is not all that conformable and will not flow to fill any voids caused by surface roughness. It is not really very thermally conductive, (it's just a plastic after all, albeit a tough one) and its major virtue is that it is physically strong and resistant to puncture and can be used in extremely thin cross section, just like mica. I find it hard to believe the very low fiigure for thermal resistance shown in at a site mentioned in a previous post. Either this is for an extremely thin insulator or is a typo, as it seems better than a greased junction with no insulator, and is also better than alumina. I will look around and see if I can find any other data to support the figure given, but I am skeptical. I would rate kapton to be about the same as mica when greased.
Try this site:
http://www.bergquistcompany.com/thermal_materials.cfm
Data on SilPads and other thermal materials.
http://www.bergquistcompany.com/thermal_materials.cfm
Data on SilPads and other thermal materials.
Or if specifically interested in Kapton based insulator pads:
http://www.bergquistcompany.com/objects/data_sheets/PDS_Sil_Pad_K_10_PDF.pdf
http://www.bergquistcompany.com/objects/data_sheets/PDS_Sil_Pad_K_10_PDF.pdf
Has anyone ever seen insulators like these, and/or know their name ?
It looks like a very thin layer of Mica covered on both sides with something chalky, the outside has a plastic feeling.
I stumbled across some obsolete devices again, the insulators came with them in the bag with Japanese labelling.
Before my time ?
It looks like a very thin layer of Mica covered on both sides with something chalky, the outside has a plastic feeling.
I stumbled across some obsolete devices again, the insulators came with them in the bag with Japanese labelling.
Before my time ?
Attachments
Heat sink grease isn't supposed to dry out, but it does. I had to do some maintenance on a piece the other day and the heat sink grease came off in flakes. In that case, I suspect age. Heat can do it also. I was recently given two dead Audio Source amps which had cooked their heatsink grease to brittle flakes in about a year. The culprit here was a lethal combination of too-small heatsinks, playing the amps at high volume, and poor ventilation, with a side order of putting one amp on top of the other. Incidentally, it was easy to see which amp had been on top--its entire circuit board had cooked to a toasty brown; the bottom one was only brown around the output devices.
Grey
Grey
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.