HEAT SINKS
If anyone is looking for F5 heatsinks, the Conrad heat sink group buy is closing soon. These black heat sinks will come milled and drilled for top and bottom covers, and with black bolts, cost about 62. USD each, delivered in the US. Further up in the thread, someone had indicated these would be a good choice for the F5.
See final info on the group buy here -
http://www.diyaudio.com/forums/showthread.php?postid=1589700#post1589700
Bob
If anyone is looking for F5 heatsinks, the Conrad heat sink group buy is closing soon. These black heat sinks will come milled and drilled for top and bottom covers, and with black bolts, cost about 62. USD each, delivered in the US. Further up in the thread, someone had indicated these would be a good choice for the F5.
See final info on the group buy here -
http://www.diyaudio.com/forums/showthread.php?postid=1589700#post1589700
Bob
Bias
In AudioXpress magazine, I've read Pass say at least twice that you need to overate power transistors by a factor of 6 if you want any kind of reliability. This just happens to be a bias of 1.3A as he designed it. Though, I'm sure there some flexibility here, with proper heat sinking. But not 3 amps IMO.
If I'm wrong I'm sure he'll chime in.
In AudioXpress magazine, I've read Pass say at least twice that you need to overate power transistors by a factor of 6 if you want any kind of reliability. This just happens to be a bias of 1.3A as he designed it. Though, I'm sure there some flexibility here, with proper heat sinking. But not 3 amps IMO.
If I'm wrong I'm sure he'll chime in.
Re: Bias
The key words here are "adequate" and "heatsinking"
You can push a mosfet real hard, if the temperature is kept down.
Magura
Johnloudb said:In AudioXpress magazine, I've read Pass say at least twice that you need to overate power transistors by a factor of 6 if you want any kind of reliability. This just happens to be a bias of 1.3A as he designed it. Though, I'm sure there some flexibility here, with proper heat sinking. But not 3 amps IMO.
If I'm wrong I'm sure he'll chime in.
The key words here are "adequate" and "heatsinking"
You can push a mosfet real hard, if the temperature is kept down.
Magura
The die dies?
40V, 3A
120W
Well, if you are going to cool down your mounting flange to 0°C, then OK. But if your sink is at 30°C (!!) and the thermal resistance between junction and sink is around 0.80- 1.0 K/W (on this value you don't have much influence), the poor MOSFET's die is above 120°C.
Tino
Magura said:The key words here are "adequate" and "heatsinking"
You can push a mosfet real hard, if the temperature is kept down.
40V, 3A
120WWell, if you are going to cool down your mounting flange to 0°C, then OK. But if your sink is at 30°C (!!) and the thermal resistance between junction and sink is around 0.80- 1.0 K/W (on this value you don't have much influence), the poor MOSFET's die is above 120°C.
Tino
Magura said:I have tested this sort of abuse for approx. 2.500 hrs. Heatsink running at 35C.
I for one don't care if I have to do some maintenance every 2.500 hrs., if it means that I can get better sound
Magura
Just keep in mind that you may also have to do some maintenance your loudspeaker or your home. I had an amp go DC at the output and ruined one of my Thiel CS1 speakers. We initially thought it was the house on fire and called the fire department. But later realized it was the speaker enclosure burning.
The voice coil of the tweeter was real close to the back of the cabinet. There was a big black burn mark behind the tweeter. A well known speaker designer once told me a tweeter looks like a short to DC. This makes no sense to me since tweeters are capacitively coupled to the amp through the crossover.
Maybe the amp was oscillating at a high freq and the tweeter broke and got hot? I don't know. The woofer was also melted.
that is a real danger in what we do, I have lost a few tweeters from r.f. oscillationsbecause I was to impatient to fully test. An RC network connected right at the output binding posts is good insurance, and has arguably no effect on the sound.
Years ago I repaired a Klipsch LaScala that was completely charred around the woofer, and the cone was ashes.
It seems the drummer From the Rock band Foghat decided to drum directly on a microphone totaly smoking a phase linear 700B amplifier. The amp was not in much better shape than the speaker. Believe it or not Klipsch covered the speaker under warranty. Even overnight expressing it to us.
My advice buy a scope, even an old limited bandwidth one .
Run your new projects into loads as long as you can stand, before chancing your hard earned speakers to it. It much easier to build a power R.F. oscillator than a stable high bandwidth amp . Of couse you can always connect a small wattage resistor , and run the amp with no input and see if the resistor becomes hot ( or becomes smoke)
Years ago I repaired a Klipsch LaScala that was completely charred around the woofer, and the cone was ashes.
It seems the drummer From the Rock band Foghat decided to drum directly on a microphone totaly smoking a phase linear 700B amplifier. The amp was not in much better shape than the speaker. Believe it or not Klipsch covered the speaker under warranty. Even overnight expressing it to us.
My advice buy a scope, even an old limited bandwidth one .
Run your new projects into loads as long as you can stand, before chancing your hard earned speakers to it. It much easier to build a power R.F. oscillator than a stable high bandwidth amp . Of couse you can always connect a small wattage resistor , and run the amp with no input and see if the resistor becomes hot ( or becomes smoke)
Heehee...
It's pretty cool when passive devices catch on fire from the inside out. When poly cones do it drips down the front of your speaker like syrup on a sundae. I have some pics of some older B&W's that did this...DC caused the adhesive bonding the voicecoil to the former to ignite.
Chris
It's pretty cool when passive devices catch on fire from the inside out. When poly cones do it drips down the front of your speaker like syrup on a sundae. I have some pics of some older B&W's that did this...DC caused the adhesive bonding the voicecoil to the former to ignite.
Chris
Johnloudb said:
When running mosfets this hard, you usually don't get DC at the output if they go south. They simply blow, and that's it.
A simple and nice way to make sure you don't get DC at rail voltage, to your speakers, is to use output caps.
I find it a bit strange, that people whom are pushing the limits of quality amps, somehow are reluctant to do a bit of maintenance.
I see this game, like with hefty sports cars. A hefty sports car, is usually not particular durable, and for a car that's perfectly acceptable......but not for an amp?
Magura
Oh, and BTW, all this talk about amps catching fire, has little to do with failure of the parts, but is rather a matter of building the chassis, using the right materials.
Whenever I build a power amp, there is close to nothing that possibly could burn in it, and the bit that can burn, is made of self extinguishing materials.
Magura
Whenever I build a power amp, there is close to nothing that possibly could burn in it, and the bit that can burn, is made of self extinguishing materials.
Magura
Magura said:Oh, and BTW, all this talk about amps catching fire, has little to do with failure of the parts, but is rather a matter of building the chassis, using the right materials.
Whenever I build a power amp, there is close to nothing that possibly could burn in it, and the bit that can burn, is made of self extinguishing materials.
Magura
Well, they do sell "flame-proof" resistors -- and Sony just recalled a batch of laptops (battery related again). (The accountants become too involved in the design of power supplies, batteries and other components.)
I use flame-proof 10R/2W resistor in my F5.
Magura said:
When running mosfets this hard, you usually don't get DC at the output if they go south. They simply blow, and that's it.
A simple and nice way to make sure you don't get DC at rail voltage, to your speakers, is to use output caps.
I find it a bit strange, that people whom are pushing the limits of quality amps, somehow are reluctant to do a bit of maintenance.
I see this game, like with hefty sports cars. A hefty sports car, is usually not particular durable, and for a car that's perfectly acceptable......but not for an amp?
Magura
I'm sure there are many scenarios here, but a push/pull amp will likely have some amount of DC at the output when one of the output transistors fail.
Yes, there are a number of ways to protect against DC including caps. The amp I mentioned was and old 150 Watt commercial design, and it didn't have DC protection unfortunately.
I do miss those speakers.Whatever you enjoy.