Hahaha, T117! I haven't got time to look at this right now but your "closer to reality" did make me laugh.
We're trying to look at the output stage in isolation. What it is driven by is another question. I'll drive it with an op amp with 0.01 ohms output impedance if needed. And we know from Self that a buffer here and there does wonders for the distortion. A fet on the way in might work too. (Well, it would.) The point is comparison and most people I have read seem to think a CFP is potentially better, especially in a simple follower, than just an EF. Here I am giving the driver a bit more work to do and that seems to be paying off. Catch you perhaps tonight or maybe tomorrow.
We're trying to look at the output stage in isolation. What it is driven by is another question. I'll drive it with an op amp with 0.01 ohms output impedance if needed. And we know from Self that a buffer here and there does wonders for the distortion. A fet on the way in might work too. (Well, it would.) The point is comparison and most people I have read seem to think a CFP is potentially better, especially in a simple follower, than just an EF. Here I am giving the driver a bit more work to do and that seems to be paying off. Catch you perhaps tonight or maybe tomorrow.
😀
There are no miracles. OpAmp has the same cascades as PowerAmp. LTP, VAS and output follower. OpAmp is subject to the same laws of physics as PowerAmp. If built on discretes OS contributes to the signal distortion, the same distortion he will give in a monolithic chip. Just a point of occurrence of distortion will be unavailable for devices, but the mind we will know that they are there.
It is interesting that in both the Crimson and the Quad triple the lower half has the
last two transistors as a Darlington. This configuration allows the use of a
single package power Darlington. Does anyone know if either of these amps
had any stability issues?
last two transistors as a Darlington. This configuration allows the use of a
single package power Darlington. Does anyone know if either of these amps
had any stability issues?
Last edited:
Also, the Bryston 3B provides probably the most complex example, hard to believe
that it is stable. It is interesting that you can get different examples by deleting either
the NPN or PNP power device on each half.
that it is stable. It is interesting that you can get different examples by deleting either
the NPN or PNP power device on each half.
Yup, the Crimson was known to be a little unstable. On the top half, IIRC.
I think it varied from build to build and which power transistors you used. The designer did put up quite a lot of stuff on the net a while ago, if you are interested.
I think it varied from build to build and which power transistors you used. The designer did put up quite a lot of stuff on the net a while ago, if you are interested.
It must be two years since I was looking at this and I'd have to reread the whole thread to be sure of exactly what was going being discussed. I remember there was some good stuff and that I came to some conclusions that I will use at some point in the future (which should be embodied in the LTspice circuits I did at the time - but even those will be hard to find) but I don't know how I can replay all my internet activity from that time. I think you'll have to search for Crimson CFPs and instability. That's what I did.
- Status
- Not open for further replies.