20W cathode follower amplifier

[Marik]

I built a CF output amp once.
From memory, first stage 6sn7 (I think I might have had two stages), 6B4G with LL1677 driving PP 6550.
No global NFB.
I used Dynaco Mark IV iron. It sounded better than anything I ever heard.

[rdf]

It's not 20 watts but at the moment I'm listening to a CF parafeed headphone amp based around a single compactron that'll swing 11 volts p-p into a pair of 60 ohms headphones and sounds wonderful. The only gotcha is it takes a lot of current to keep HF distortion in the ballpark of LF distortion due to the CF's asymetrical slew rates.

[Wavebourn]

Did you gentlemen try such tubes like IRFP27N60K, or similar?
Sound awesome despite of huge and very variable capacitances!

[tubelab.com]

Did you gentlemen try such tubes like IRFP27N60K, or similar?

No, they don't fit any of my sockets! I can't see the glowing cathode too well either.

[Fuling]

IRFP27N60K

Sure, the transconductance would be hard to beat and there wouldn´t be any heater-cathode insulation issues...
A mosfet follower with a low ratio OPT and a tube input stage would be very interesting.

Anyway, cathode follower obviously has certain qualities as output stages.
I have a pair of 3,5k/8 15W OPTs that I bought for the 808 project, now I´m getting curious how these would perform wired as 1,75k/8 driven by PL519 followers.
The transformers sounded quite good when i tried them in a tiny spud amp last weekend and they would probably sound even better with properly sized output tubes.

Gah, so many projects, so little time...

[Jeb-D.]

The only gotcha is it takes a lot of current to keep HF distortion in the ballpark of LF distortion due to the CF's asymetrical slew rates

Interesting point

[Wavebourn]

Originally posted by Fuling
[B]

Sure, the transconductance would be hard to beat and there wouldn´t be any heater-cathode insulation issues...

...and there will be no need for an OPT!

I switched to them after experimenting with tube cathode followers. By the way, even ordinary power transformers sounded nice with cathode followers, auto-transformers even better, especially when the follower is loaded by CCS. A first, I used a MOSFET as a CCS. Then, I used it instead of a cathode follower. After that I replaced a CCS by a voltage to current converter made of pair of BJT's. That breadboard scared me when a water run on a record suddenly. I've jumped and turned around before realizing that there can't be a waterfall in my room...

Originally posted by Jeb-D.

Interesting point

...and absolutely valid one!

[Fuling]

I switched to them after experimenting with tube cathode followers. By the way, even ordinary power transformers sounded nice with cathode followers, auto-transformers even better, especially when the follower is loaded by CCS. A first, I used a MOSFET as a CCS.

A CF on top of a mosfet CCS, cap coupled to an autoformer is something I´ll try someday.
Until then, the idea of a PL519 CF amp with my existing 3,5k (miswired to 1,75k) OPTs seems more and more appealing.
If done right it could be very minimalistic:

Interesting point

Yes, but I imagine that to get these kind of voltage swings we need fairly "grunty" driver tubes anyway.

[Wavebourn]

Originally posted by Fuling


A CF on top of a mosfet CCS, cap coupled to an autoformer is something I´ll try someday.
Until then, the idea of a PL519 CF amp with my existing 3,5k (miswired to 1,75k) OPTs seems more and more appealing.

Good luck, you are on your own Road To The Rome!

[Fuling]

your own Road To The Rome

Considering the amounts of energy stored in the 320V 350mA PSU
this amp would require, I´m more likely on my own road to...hell.
I´ve been lucky with HV supplies so far, but I guess that when I finally get zapped it will be from a PSU capable of turning me into 300 pounds of roast beef in 3 seconds...