New Amplifier Project

[subwo1]

Here are a couple of considerations. The MPSA8099 transistor supplying constant current to the emitter follower buffer for the VAS base drive may have a problem with +/-50 volt rails. Even though its base voltage keeps it on all the time, its high impedance effect at the collector is potentially compromised and could begin to behave more resistively or even more like a constant voltage source as it begins to suffer breakdown. Consider adding a cascode with the base referenced to ground or a single 2N5401 or MPSA92 or even a solitary resistor. ???

Also, consider checking to see if using phase compensation from the collector of one VAS transistor back to the base of the feedback transistor of the input pair is a better solution than slowing down the output drive with the 750pF caps. The effect could be to lessen the high frequency control of the last stages somewhat. The caveat is that I have never had the chance to actually try it yet in a real circuit, but you may like to give it a look-see.

[Eva]

For input stage I recommend BC546B [BC556B] since they are rated at 65V Vce, have betas around 400-500, are pretty fast and show decent noise figures. For VAS buffer I would recommend also BC550C [BC560C] since they are rated at 45Vce, are even faster and have betas arond 500-700

For the Vas I recommend BD139 [BD140] since in practice they show pretty fast turn-on and turn-off times, they are rated at 100Vce and have betas around 150

As drivers I recommend TIP41C and TIP42C since they have enough current capability for that circuit and I've found them empirically to turn on and turn off faster than BD911 and BD912 [test made with Samsung TIPs versus ST Microelectronic BDs]. There are also better alternatives like MJE15028/15029 with higher beta and much faster]

[The Peasant]

OK, here are the cases that I'm using for this project. I have two of these:

[The Peasant]

Inside view and one heatsink assembly:

[The Peasant]

Here is the prototype so far:

[The Peasant]

OK, I have the first amplifier up and running.

I still have some fine tuning to do, but initial listening tests have been quite positive. The biggest difference between it and the stock PS Audio model Two C plus is in the clarity and inner detailing. With the original amplifier you can, for example, hear the sound of a violin in a recording. With the new amplifier, you can actually hear the tone of the instrument, the sounds of the bow on the strings, and the resonance of the body. The difference is akin to what I'd expect from exchanging a cheap MM cartridge with a high quality MC unit. This is a much larger improvement than I had expected.

Another big improvement is with the bass and dynamics. I expected more authority with the new amplifier based on the higher power output, but the dynamics have completely taken me by surprise. Loud transients, like percussion, appear to almost jump out at you, and sound much more realistic. Even with casual listening this is very noticeable, and can be a bit scary if you're not expecting the sound of a live drummer in your listening room.

Other than that, the soundstage is a bit wider and better defined, with more space between instruments, and much more depth. As well, with the parallelled output devices and larger heatsinks, the amp runs very cool, unlike the original design.

Overall, the new amplifier is a whole lot more musical and listenable. Swapping back to the original amplifier, it now sounds like a cartoon image instead of the real thing. The difference is simply amazing.

Anyways, here is a pic of the completed amplifier:

[The Peasant]

Thanks to everyone who gave advice, I have sure learned a lot in a short time on this forum!

There are a few more pictures on my website if anyone is interested.

Here is another pic of the inside:

[tmblack]

Doug,

I'm really taken to this design.
How did it sound with the TIP41/42 drivers?
What accounts for the huge improvement?

I might consider better output transistors with less beta droop such as
MJ21193/MJ21194.

For inputs: 2SC2240 and 2SA970 have lower noise, higher beta.

Do you have the PCB layout available?

Take care

Tom

[Tekko]

How can we be sure this is not the original Bryston parts ?

Sorry, i couldent resist, it looks so well made

[The Peasant]

How did it sound with the TIP41/42 drivers?

I never had a chance to compare them directly, I only used them with some output transistors that turned out to be counterfeit and unusable. When I replaced the outputs with new ON Semiconductor devices I replaced the drivers as well.

What accounts for the huge improvement?

Better quality parts, solder, and power supply improvements.

Do you have the PCB layout available?

I have it, but it would not be of much use unless the boards were to be installed in a 4B chassis. I would design a generic pcb much differently.

How can we be sure this is not the original Bryston parts? Sorry, i couldent resist, it looks so well made

Thank you for the compliment! Actually, a real Bryston looks quite different than this on the inside!

Take care,
Doug