Hi Thanh1973,
I checked and the devices were flush and tight with the heat sink. The other channel with the fairchild mosfets biased at 1.3A the temps were 110C & 120CApparently they had been running so for weeks. I replaced the mica/artic silver with sil-pad K-10's and the fairchild now lives at 60C & 65C.
I think I've figured out what I was doing wrong. As per my days of building computers, I would use only a very thin even layer when mounting the CPU to the heat sink. A thin layer allows too much contact with the mica, which is a poor thermal conductor. The photo of the F5 in the 6moons article shows a lot of grease was used.
This all depends on two things surface roughness and flatness.
I am guessing flatness is not an issue from what you have explained.
So for good thermal transfer you would ideally have a very smooth heatsink surface (eg mirror finish) and use minimal thermal grease.
However this is rarely the case. There are often machining grooves and scratches on the surface. These scratches (or surface roughness) trap air (air is a poor thermal conductor) so the thermal grease or silpad fills these scratches etc and improves the thermal transfer (ie silpad has higher thermal conductivity than air)
In practice it is best to apply too much than not enough, since most of the excess will be squeezed out by the force of the mounting screws.
I am not sure that I am a fan of arctic silver. I have used this in the past with cpus etc. While arctic silver has quite good thermal properties it has poor flow properties (ie does not spread easy under pressure), while the regular thermal grease has very good flow properties but not as good thermal properties (catch 22 *).
Anyway, it is good to see you fixed the problem.
Off topic
* "Catch-22" is common idiomatic usage meaning "a no-win situation" or "a double bind" of any type. Within the book, "Catch-22" is a military rule, the self-contradictory circular logic that, for example, prevents anyone from avoiding combat missions. In Heller's own words:
There was only one catch and that was Catch-22, which specified that a concern for one's safety in the face of dangers that were real and immediate was the process of a rational mind. 'Orr' was crazy and could be grounded. All he had to do was ask to be grounded; and as soon as he did, he would no longer be crazy and would have to fly more missions. Orr would be crazy to fly more missions and sane if he didn't, but if he was sane he had to fly them. If he flew them he was crazy and didn't have to; but if he didn't want to he was sane and had to. Yossarian was moved very deeply by the absolute simplicity of this clause of Catch-22 and let out a respectful whistle.
"That's some catch, that Catch-22," Yossarian observed.
"It's the best there is," Doc Daneeka agreed.
Last edited:
The only time I use Arctic Silver is when I'm doing a CPU heatsink. And that's only because I don't want it to move around. Other than that I don't see an advantage. It won't run as hot as Thermalcote and it's expensive. And doesn't squeeze out so easily thus you run the risk of having too much between the surfaces.
I've used thermalcote for many years for TV repair ie large Hybrid ICs with only two screws and I haven't really had a problem with applying the proper amount. I rub a little (enough to cover) on both surfaces with my finger and screw it down. There's always some squeezout.
I've used thermalcote for many years for TV repair ie large Hybrid ICs with only two screws and I haven't really had a problem with applying the proper amount. I rub a little (enough to cover) on both surfaces with my finger and screw it down. There's always some squeezout.
I have yet to see a perfect flat and smooth heatsink
Flatness or smoothness is less significant with smaller width devices
But try and place one on a heatsink
Put a flashlight on it and try if you can see light coming through between the device and heatsink
I bet you can
So, actually the device is still only touching at its edges
You may have improved on this by using a softer isolator
btw, is there something like stressing a device if clamped/tightened too hard
I believe I occationally have heard amps sound slightly different depending of too hard tightening of the mounting screw
Last edited:
Really? I'll have to try a torque wrench and record the settings and the differences in sound.
"geez it sounded a little constipated at 26 inch pounds so I backed off a little and the soundstage opened right up"
Seriously, sounds like a discussion on tweaking in one of the hifi rags. But maybe you would crack the Mosfet into pieces if you 're not careful.
Last edited:
I vote fore that
if you know how to set it correctly, otherwise its more dangerousPersonally, my hands are my personal torgue wrench
Banned
Joined 2002
Arctic silver is good stuff, when i was building pc's for clients we were warming up the cpu heat sink a little then applying the silver paste, then moving it around a little with a applicator piece we had, it worked very well and most times kept the cpu cooler, it allowed us to spread it but not to thick.
When mounting Transistors, i now use ceramic isolation washers, and a washer that covers the whole transistor.
J'
So, actually the device is still only touching at its edges
You may have improved on this by using a softer isolator
I would rather be commented on this 100% mechanical issue, more than on being an insane tweaker
One reason I mentioned the screw tightening is on the issue when a device only touches at its edges
And it often does, really
So, you may be putting a lot of strain on the housing
Its designd fore an even pressure over the whole surface
Or maybe exactly NOT with pressure on the edges, which is mostly just the plastic housing
Last edited:
Hi thanh1973,
Great book!
In my early youth I pick it up reluctantly from a public library, thinking that it was a "war best-seller"...I was wrong and had a wonderful time: as with the best books I've read, I started it again after the unexpected end...
In my mind, Yossarian is a character á la hauteur de Raskolnikov.
End OT.
Cheers,
M
Great book!
In my early youth I pick it up reluctantly from a public library, thinking that it was a "war best-seller"...I was wrong and had a wonderful time: as with the best books I've read, I started it again after the unexpected end...
In my mind, Yossarian is a character á la hauteur de Raskolnikov.
End OT.
Cheers,
M
Yes great book.
One of my other favourite books is Hitch Hikers Guide to the Galaxy
The Hitchhiker's Guide to the Galaxy - Wikipedia, the free encyclopedia
OK it is time for me to get down to business. I am on holidays now, so I need to put insome serious time, working on finishing off all my projects
One of my other favourite books is Hitch Hikers Guide to the Galaxy
The Hitchhiker's Guide to the Galaxy - Wikipedia, the free encyclopedia
OK it is time for me to get down to business. I am on holidays now, so I need to put insome serious time, working on finishing off all my projects
doubling or tripling the number of output devices does not change the power output capability nor does it change the maximum ClassA output current.
The bias of the output stage determines the maximum ClassA output current.
The number of devices determines the capability of the amplifier to drive low impedance loads.
If you double the number of output devices and double the bias current to 2.6A then you have doubled the maximum ClassA output current to ~5.2Apk.
This would allow an 8ohm load to be driven to 27W (voltage limited) or it will allow a 4ohm load to be driven to 54W (both current and voltage limited).
Whereas a standard bias current of 1.3A would allow 8ohms to go to 27W (voltage and current limited) or a 4ohm load to go to 13.5W (current limited) in ClassA and 54W in ClassAB (voltage limited).
The bias of the output stage determines the maximum ClassA output current.
The number of devices determines the capability of the amplifier to drive low impedance loads.
If you double the number of output devices and double the bias current to 2.6A then you have doubled the maximum ClassA output current to ~5.2Apk.
This would allow an 8ohm load to be driven to 27W (voltage limited) or it will allow a 4ohm load to be driven to 54W (both current and voltage limited).
Whereas a standard bias current of 1.3A would allow 8ohms to go to 27W (voltage and current limited) or a 4ohm load to go to 13.5W (current limited) in ClassA and 54W in ClassAB (voltage limited).
That is a masterpiece. My other favorite (nonfiction) is
The End of Science by John Horgan.
