I'm looking for input and possible schematics, boards etc. to build a higher power class A amp. In particular though, I want to use the output transistors that I have an abundant supply of MJ15024/25. These are NOS and from the same lot code. I have about 150 of each. I currently have a Pass labs designed A-40 and F5. I would like to have something in the 100W/CH range and have the heatsinks from industrial drives that can support that. I can go with mono's or possible stereo chassis w/fan cooling. I love the sound of the A-40 and it would be great if that front end could be expanded to drive a different output stage to handle higher power. I realize that it is a darlington config but I did read somewhere where Pass was saying it would be simple to convert to a convential BJT with driver. I know that just about all the Pass stuff now is driven with MOSFETs and not the BJTs that I'm wanting to use. Any way interested in what people have to say on this.
Sounds like about ±50V DC supply rails and about 3 amps of DC bias current per channel. A push-pull class A output stage leaves class A when delivering 2xIbias to the load, and 6A peak is 4.2A RMS, giving > 100 watts into an 8 ohm load. The push half of the output stage dissipates 150W and the pull half dissipates another 150W. So, 6 pairs of output transistors per channel would mean each transistor dissipates 25 watts, which is a nice, safe, low number.
You'll probably want a 1000VA transformer with dual 40VAC RMS secondaries, perhaps like the AnTek AN-10440
You'll probably want a 1000VA transformer with dual 40VAC RMS secondaries, perhaps like the AnTek AN-10440
And a supplemental air conditioner. Really, evaluate why you actually want a high power class A. Do you have a belief it's lower distortion? Or is it just because? Have you considered a high bias AB? like 10W?
50V is a bit shy for a 100W amp. IMHO. Remember how much voltage is needed in the VAS to drive the outputs. More like +/- 70V. If you are thinking of building an amp, have you first studied Cordell and Self? Their books will answer most of your questions. Correctly.
25W per can STEADY as in a class A. is a LOT and I mean a LOT of heat. Do the math on the heat sink.
6 pairs of BJT's will but a heck of a strain on the VAS. You may have more distortion there than any class A output might gain you. ( Actually, it won't gain you anything over a correctly executed class AB, but that's another story. )
50V is a bit shy for a 100W amp. IMHO. Remember how much voltage is needed in the VAS to drive the outputs. More like +/- 70V. If you are thinking of building an amp, have you first studied Cordell and Self? Their books will answer most of your questions. Correctly.
25W per can STEADY as in a class A. is a LOT and I mean a LOT of heat. Do the math on the heat sink.
6 pairs of BJT's will but a heck of a strain on the VAS. You may have more distortion there than any class A output might gain you. ( Actually, it won't gain you anything over a correctly executed class AB, but that's another story. )
I was (still am, but on back burner atm) thinking something similar - was planning a high bias CFA on signal 80-25 power tranny in an old Bose 1801 chassis. Already have several of the trafos and a ton of MJ15004 and 2N3773. Thought about SE class A for a while, but my best speakers won’t like it (Z drops too low) and it might not be enough heatsink.
Strain on the VAS? Nah, this is a job for a triple, not an EF2. Don’t have to worry about the crossover distortion.
Strain on the VAS? Nah, this is a job for a triple, not an EF2. Don’t have to worry about the crossover distortion.
Not using BJT's but plain old IRFP240s and LU104D's to get 100wpc Class A.
https://www.diyaudio.com/community/threads/susylu-where-are-you.373533/
https://www.diyaudio.com/community/threads/susylu-where-are-you.373533/
I suggest crossover distortion is way over rated and mostly hype from magazine writers who have never built an amp.
Now, there is a lot to be said about twice as many outputs as might be done in an OEM amp as you stay in a more linear zone. There are many more details, but I have observed the "great" amps seem to have twice as many outputs. Another hint is enough power supply that the 4 Ohm rating is really close to the 8.
Now, there is a lot to be said about twice as many outputs as might be done in an OEM amp as you stay in a more linear zone. There are many more details, but I have observed the "great" amps seem to have twice as many outputs. Another hint is enough power supply that the 4 Ohm rating is really close to the 8.
I have BA-3 balanced mono blocks three deep (F4 output stage); nice for driving passive speakers but the best sound is in multi-amped speakers with smaller beasts
So thanks for all the input. Just remember that I have MANY ouput devices that are not being used. That's why I wasn't pursuing another mosfet design. Why 100W? BC my 40W class A runs out of steam when I push things. This is with my older Maggies and now dynamic speakers that are rated at 83-84db sensitivity. As far as HS and PS that is no problem. I have a 3-4kVA power transformer that will fit the bill. Currently working with industrial power electronics and can get any of the components needed as surplus. So again anyone that knows of a BJT design or how to go about modifying the A-40 frontend to handle 6-8 pairs of BJT's. What about this my current speakers are inefficient but never drop below 8 ohms most of the range are around 10. What if I ran the A-40 in bridge mode with differential input signal?? Surely it can handle minimum of 4-5 ohm load?? Also not against running high bias Class AB. I have a honeybadger that sounds very good. How hard can that be pushed. Really though Class A is more first choice. The A-40 was a real eye opener to me 15 years ago and believe that Pass has the right idea.
Thanks Mark. I think your being sincere. LOL. I think with my current speaker loading being pretty easy on the amp it’s worth try to bridge the A-40. I have a balanced out in pre and can make an adapter for input to amp. Worth a try.
Yes, N. Pass has the right idea:
''New Stasis front end''
https://www.diyaudio.com/community/threads/new-stasis-front-end.363701/
