Just to be clear on this. No one is suggesting using A/S without an insulator like mica or a Silpad.
Yep, very thin layer. We use a razor blade to scrape (gently) off the excess, it's only to fill in the micro pores of the heatsink / heat generating device.
Ron
Yep, very thin layer. We use a razor blade to scrape (gently) off the excess, it's only to fill in the micro pores of the heatsink / heat generating device.
Ron
Last edited:
Straight rip from A/S website:
............While it is not electrically conductive, the compound is very slightly capacitive and could potentially cause problems if it bridges two close-proximity electrical paths.)
Being capacitive means its conductive at very high frequency, no?
no, it may act as a dielectric and factor up the capacitance between two adjacent conductors.Being capacitive means its conductive at very high frequency, no?
I am not trying to keep this argument going, but I believe the point is some of these silvery looking pastes are electrically conductive and some are not.
How can we tell them apart?
I think it's far simpler and easier to avoid the silvery type where electrical conductivity will cause a problem.
no, it may act as a dielectric and factor up the capacitance between two adjacent conductors.
I am not trying to keep this argument going, but I believe the point is some of these silvery looking pastes are electrically conductive and some are not.
How can we tell them apart?
I think it's far simpler and easier to avoid the silvery type where electrical conductivity will cause a problem.
Andrew T,
I couldn't agree more.
I have NOT used it on my F5, I was just trying to find out the "source" of the earlier comments against using it. As it has performed flawlessly for 100s of CPUs. (personally)
Tinitus,
"Being capacitive means its conductive at very high frequency, no? "
that's way beyond my expertise............If I had any.
Ron
it has performed flawlessly for 100s of CPUs. (personally)
Ron
Ah, I see now
I think a CPU chip is a different matter, and arctic silver compound may be just perfect there
But its a different construction
With output devices you cannot be 100% certain that the compound wont "connect" with the mounting bolt/screw
My conversation with the folks from Arctic Silver.
I wrote:
I have a need to use a thermal compound with mica insulators between the transistors and the heatsink. Would Arctic Silver 5 be a good product to use?
Regards,
Paul Blossom
Their reply:
Hi Paul,
We recommend Ceramique for your application.
Long term stability should be good.
Best Regards,
Kirk
Arctic Silver Incorporated - Home
I wrote:
I have a need to use a thermal compound with mica insulators between the transistors and the heatsink. Would Arctic Silver 5 be a good product to use?
Regards,
Paul Blossom
Their reply:
Hi Paul,
We recommend Ceramique for your application.
Long term stability should be good.
Best Regards,
Kirk
Arctic Silver Incorporated - Home
CPUs are electrically insulated whithin their case and the silver (which we all know is a very good conductor of electricity) is also a very good conductor of heat and is used for this purpose. The same company, Arctic Silver, also makes a thermal greese without silver for power mosfets and other electrically sensitive computer components. It is advised to use the proper compound for its designed application or you risk serious problems. It may be safe when used carefully, it may blow your amp if it creates an electrical short. Doesn't make since to risk your amp over a couple dollars worth of thermal greese. My 5 cents worth...inflation ya know🙁
Tony
Tony
I use Ceramique for CPUs because it's a little thicker than traditional grease, so the heatsink won't slide around. However, I still use Thermalcote for almost everything else. Works fine for me. It costs less and is rated at higher temperature than all the Arctic Silver products. I think Nelson uses Thermalcote. So I'm guessing it's probably not necessary to use a boutique product for this particular application.
I use Thermalcote for the usual reason - I have a large case of
it. The heat sinks on FW products are bead blasted, and the
result is that the surface is not smooth enough for really good
heat conduction without grease.
Another interesting factoid, the Drain of the Jfet in the F3 does
not use an insulating pad - the part is insulated by the grease
and the anodizing.
😎
it. The heat sinks on FW products are bead blasted, and the
result is that the surface is not smooth enough for really good
heat conduction without grease.
Another interesting factoid, the Drain of the Jfet in the F3 does
not use an insulating pad - the part is insulated by the grease
and the anodizing.
😎
Just to be clear on this. No one is suggesting using A/S without an insulator like mica or a Silpad.
Yep, very thin layer. We use a razor blade to scrape (gently) off the excess, it's only to fill in the micro pores of the heatsink / heat generating device.
Ron
otherwise you get a direct short from drain to ground, like me😱
(measured and confirmed all your suspicions)
Last edited:
It might work safer with device clips
Its tempting
I dont know if black anodizing are made in different ways
The only one I have tried to grind smooth with sandpaper showed a white and VERY tough layer of something between the black anodizing and the alu profile
Im sure that white layer isolates very effectively
Somehow I expect silver anodizing to be different
But the mounting screw may change things
Doing the needed thread can cause various things
Like material rise around the thread hole, etc
Its tempting
I dont know if black anodizing are made in different ways
The only one I have tried to grind smooth with sandpaper showed a white and VERY tough layer of something between the black anodizing and the alu profile
Im sure that white layer isolates very effectively
Somehow I expect silver anodizing to be different
But the mounting screw may change things
Doing the needed thread can cause various things
Like material rise around the thread hole, etc
Pblossom,
Thank you for your input, yours is the only direct response from the "horses' mouth". They should know their products strengths and weaknesses.
Papa bought a case of Thermalcote for a reason.......Yes? Good enough for me!
Anodization is a great insulator, but fragile and prone to scratching.
I'll save the Arctic Silver for the CPUs, YMMV.
Ron
Thank you for your input, yours is the only direct response from the "horses' mouth". They should know their products strengths and weaknesses.
Papa bought a case of Thermalcote for a reason.......Yes? Good enough for me!
Anodization is a great insulator, but fragile and prone to scratching.
I'll save the Arctic Silver for the CPUs, YMMV.
Ron
A lifetime (and 1/2) supply on Fleabay.
One pound of Thermalcote. LOL
Ron
One pound of Thermalcote. LOL
Ron
Attachments
Last edited:
Papa bought a case of Thermalcote for a reason.......
I did not say I bought it.
😎
Hello Mr Pass,
I'm building an F5 amp, to feed a pair of 94db/2,83v speakers, do you think I'll need a preamp to get a realistic level ?
Thanks for sharing - this is my first pass diy project !
I'm building an F5 amp, to feed a pair of 94db/2,83v speakers, do you think I'll need a preamp to get a realistic level ?
Thanks for sharing - this is my first pass diy project !
The gain of the F5 is ~6times (~+15dB)
the max signal from a cdp is ~2.1Vac.
The max signal from the F5 will be ~12.6Vac
This is equivalent to ~20W into 8ohms. Not quite full power.
The F5 sensitivity is just slightly more than most CDP and considerably more than lower voltage sources. A pre-amp with gain may be required depending on the voltage from your sources.
Now let's look at what SPL you can get.
20W is 13dBW.
The speakers are 94dB/W@1m.
The maximum SPL is ~107dB @1m
If your average level is 20dB below your maximum then your average listening level with zero attenuation from a 210mVac signal (-20dB ref 2.1Vac) will be ~87dB @ 1m. That is getting pretty loud for most listeners. But that requires a high maximum voltage source. Maximum average level will depend on your source voltage.
the max signal from a cdp is ~2.1Vac.
The max signal from the F5 will be ~12.6Vac
This is equivalent to ~20W into 8ohms. Not quite full power.
The F5 sensitivity is just slightly more than most CDP and considerably more than lower voltage sources. A pre-amp with gain may be required depending on the voltage from your sources.
Now let's look at what SPL you can get.
20W is 13dBW.
The speakers are 94dB/W@1m.
The maximum SPL is ~107dB @1m
If your average level is 20dB below your maximum then your average listening level with zero attenuation from a 210mVac signal (-20dB ref 2.1Vac) will be ~87dB @ 1m. That is getting pretty loud for most listeners. But that requires a high maximum voltage source. Maximum average level will depend on your source voltage.
So,
for my domestic purpose, I guess It will be sufficent, with cdp or phono preamp, isnt'it ?
for my domestic purpose, I guess It will be sufficent, with cdp or phono preamp, isnt'it ?
Member
Joined 2009
Paid Member
Seems to me you don't need a preamp. But look on the bright side, if you find you need a little more gain you have a reason to build a nice preamp 😀
I did not say I bought it.
😎


Thanks for makin' us laugh Papa.
BTW, Love my F5!
Ron
- Home
- Amplifiers
- Pass Labs
- F5 power amplifier