Hello,
can I use ceramic insulators for the big transistors in the F5? Must cooling paste be used?
Many thanks, Rents
can I use ceramic insulators for the big transistors in the F5? Must cooling paste be used?
Many thanks, Rents
Any interface that can trap air must use some form of Thermal compound.
This includes all the hard surface insulators.
This includes all the hard surface insulators.
temperature coefficient of insulator pads
If what I found is correct (at least I hope so):
mica:---------------------0,40 K/W
silicon:----------------.-> 0,34 K/W
aluminiumoxide ceramic:--.-0,30 K/W
Kapton:-------------------0,15 K/W
. . . you certainly want Kapton pads. 😉
Obviously, there must be some slight differences depending on thickness.
If what I found is correct (at least I hope so):
mica:---------------------0,40 K/W
silicon:----------------.-> 0,34 K/W
aluminiumoxide ceramic:--.-0,30 K/W
Kapton:-------------------0,15 K/W
. . . you certainly want Kapton pads. 😉
Obviously, there must be some slight differences depending on thickness.
If what I found is correct (at least I hope so):
mica:---------------------0,40 K/W
silicon:----------------.-> 0,34 K/W
aluminiumoxide ceramic:--.-0,30 K/W
Kapton:-------------------0,15 K/W
. . . you certainly want Kapton pads. 😉
Obviously, there must be some slight differences depending on thickness.
Oh that's interesting! I always thought Mica to be the best and just not used anymore due to expense. I had no idea Kapton was so much better then silpads!
I am still confused with all of this insulator on chips.
One is suppose to use as little paste as possible, that I understand.
Then you have the option of using an insulator which seems to defeat the whole purpose
of the heat sink/paste combination.
If the chip's case is negative then why does it need to be insulated, won't the negative just follow the ground out of the case or will that cause a ground loop?
Thanks
One is suppose to use as little paste as possible, that I understand.
Then you have the option of using an insulator which seems to defeat the whole purpose
of the heat sink/paste combination.
If the chip's case is negative then why does it need to be insulated, won't the negative just follow the ground out of the case or will that cause a ground loop?
Thanks
You must use the specific data for the actual insulator you are using to understand the thermal resistance. Properties of these vary wildly - and using a simple table of representative data as presented above is misleading at best.
The typically available silicon insulators do not need paste. Mica does. Ceramic will.
Nobody has mentioned paste - materials vary wildly here too.
That said: in typical audio applications this just does not matter that much. At least in the home environment dissipation a are low enough that you are working far below the limits of devices.
Can you use ceramic? Yes, unless there is something really peculiar about them. You do need paste. You do not want to get wrapped about the axle comparing them to everything else in the market.
The typically available silicon insulators do not need paste. Mica does. Ceramic will.
Nobody has mentioned paste - materials vary wildly here too.
That said: in typical audio applications this just does not matter that much. At least in the home environment dissipation a are low enough that you are working far below the limits of devices.
Can you use ceramic? Yes, unless there is something really peculiar about them. You do need paste. You do not want to get wrapped about the axle comparing them to everything else in the market.
I agree, don't take such numbers too serious, as in practice the may vary significiently.
Several manufacturers' data will vary as well and as said, thickness is a major point.
Many insulators are much thicker than neccessary for audio applications (you don't need
2 KV safety). As always, of course, working carefully is of similar importance.
To my understanding a tight thermal coupling is best to get the most of the heatsink.
Ideally you want no thermal resistance between device and heatsink at all and in con-
sequence as little temperature as possible between them. You just want the neccessary
electrical insulation, nothing more than this.
However, I'd highly recommend to read this:
ESP - Heatsink design and transistor mounting
BTW, NP said in 2009:
Several manufacturers' data will vary as well and as said, thickness is a major point.
Many insulators are much thicker than neccessary for audio applications (you don't need
2 KV safety). As always, of course, working carefully is of similar importance.
To my understanding a tight thermal coupling is best to get the most of the heatsink.
Ideally you want no thermal resistance between device and heatsink at all and in con-
sequence as little temperature as possible between them. You just want the neccessary
electrical insulation, nothing more than this.
However, I'd highly recommend to read this:
ESP - Heatsink design and transistor mounting
BTW, NP said in 2009:
Last edited:
Interesting that Rod finds Mica "fragile". I guess in the sense that is can
be cleaved into thinner layers, but I don't find that to be a problem.
I find Mica to be very tough at resisting small metal burrs and sharp edges
around mounting holes, an issue which cost me money with Silpads.
Also, I have a lot of heat sinks which the manufacturer thoughtfully finished
with bead blasting, making a nice appearance, but creating additional thermal
resistance due to the uneven surface. So I use white goop regardless of the
insulating material.
(Actually I have been known to use white goop without an insulator, as the
anodizing on the sink is a pretty good insulator. I just don't do it at high
voltages.)
😎
be cleaved into thinner layers, but I don't find that to be a problem.
I find Mica to be very tough at resisting small metal burrs and sharp edges
around mounting holes, an issue which cost me money with Silpads.
Also, I have a lot of heat sinks which the manufacturer thoughtfully finished
with bead blasting, making a nice appearance, but creating additional thermal
resistance due to the uneven surface. So I use white goop regardless of the
insulating material.
(Actually I have been known to use white goop without an insulator, as the
anodizing on the sink is a pretty good insulator. I just don't do it at high
voltages.)
😎
- Status
- Not open for further replies.