Cathode Follower? Yay or nay?

[fdegrove]

Hi,

You can't hear a CF anyway, remember?

Says who and what's the logic behind this super bright reasoning??

Cheers,

[SY]

Tim, one nice thing about CF drive is that you're not necessarily limited to the idle current of the CF. On positive swings, if the tube is capable, the output current of the CF can be considerably higher. On negative swings, well, who cares?

FWIW, I've had no trouble driving 2A3s with a 6SN7-based CF, though my preference is still for a mosfet source follower in this application. To paraphrase a bright young fellow I know, the best device is the one which suits the circuit electrically.

[Sch3mat1c]

Originally posted by SY
Tim, one nice thing about CF drive is that you're not necessarily limited to the idle current of the CF. On positive swings, if the tube is capable, the output current of the CF can be considerably higher.

Yup. Just using the thought process in his post. The point being, a CF can, and does, handle DC (such as rectified by grids) whereas coupling caps cannot.

Tim

[AKSA]

Huh?

Ah, a heavy hitter! Nice to meet you!

And you suggest driving grids with what??? Transformer coupling where there's NO corrective NFB applied? Maybe not...

I suggest driving grids via a cap from the plate. Not once did I mention transformer coupling. This is, of course, highly unusual; I can see why this confuses.

Howso? You need the 100 gain anyway. And now it can be provided by a single 6AV6 rather than a cascaded 6SN7. (If they had more gain (in both sections), a dissimilar dual triode would be perfect.)

A cathode bypassed 6SN7 has a gain around 13.5 with a load of 2.5rp. Followed by say a 6BX7, with gain around 7, the total gain is 95, about right. I prefer triodes. You may choose pentodes, no problem, your choice. And yes, a dissimilar triode like a 6EM7 would give you even more gain, probably around 450, which gives the option of feedback. If you want it, of course.

What's this matter? I'll assume you are talking in ZNFB terms so there's no loop to cause phase shift oscillations in.

Yes, of course.

Huh? Oh, more of that... Dude, the best operating point is the one which suits the circuit electrically. You can't hear a CF anyway, remember?

This is plain wrong. You must spend time finding the correct operating point. It is dictated by both electrical AND sonic requirements. This is often the difference between a good tube amp and a truly outstanding one. Attention to detail. Applies to everything.

Hmm, above you said grid current can easily be up in the 100mA range ... so um, 30mA plate... but... uh. . . hey, I think 70 miliamperes just disappeared. That must be it.

Actually, you assumed grid current would be 100mA. It is seldom more than 25% of total plate current; so if the output tube is going gangbusters at say 120mA, then you could see up to 30mA grid current in deep A2. For those first couple of volts of grid positive, it might be 5-15mA, no more. In this scenario a plate loaded triode running 30mA has a good chance of meeting short term overload, but an 8mA CF rather less.

You obviously believe that since you don't understand the point the writer is wrong. Foolish thinking. Categoric statements are the security blanket of the inexperienced.

If you mean while direct-coupled to an output tube, this is almost as bad as direct-coupling to the driver tube. With fixed bias, the driver's drift has NO effect, the CF's will shift 1V absolute maximum through its own drift and output tube current can be corrected as it ages.

Yes, 'waste' is probably fair enough, but have you considered the subjective sonic differences, or are you designing strictly for efficiency. If the latter, then you might as well ditch the triodes and run with pentodes, or better still solid state. I prefer cathode biasing because the sound is, to me, softer and more organic. But YMMV and no doubt does. And BTW, I'm no dude. I'm not even your age...

and the CF needs to be fixed at zero volts. But that's impossible, so either you use a negative supply for it (and if you're going to do that you might as well fixed-bias it) or you have to cap-couple it as in the original schematic above. But that defeats the purpose of using a CF in the very first place because you can't drive the output any better than a plate follower 6SN7.

You can if you use a large, grunty triode; my original proposition.

And what's the problem with cap coupling anyway? You can drive a loudspeaker surprisingly well via a cap, all the better if it's a good one. A cap coupled plate to output grid reduces the chance of biasing accidents as well, particularly if the output tube is cathode biased. A DC coupled tube circuit is nice in theory, but as the tube ages, operating points cascade down the line.....

Your post is confrontational. Tone it down.

Cheers,

Hugh

[fdegrove]

High,

This is, of course, highly unusual; I can see why this confuses.

Hi hugh,

If only you'd see that big on my face...

This topology isn't really all that unusual in the sense that it's quite accepted practice in Japan.
Kondo, Sakuma amongst others have been doing this for many years.

These designers, yourself included bien entendu, know what sounds good because they actually go out and listen to what they put to paper.

It doesn't stop to amaze me how much time passes by before people put aside what they've been taught at school and actually start thinking for themselves.
Sometimes travelling light, educationwise, isn't such a bad thing after all...Tabula rasa style...

No offense to solid engineering, of course, it's just that when it comes to audio, things aren't always as cut and dried as they're thought to be.

A DC coupled tube circuit is nice in theory, but as the tube ages, operating points cascade down the line.....

We touched upon this subject a little while ago...
Just imagine the worst of the worst, believe me I've seen it done, where two stages run from independent B+ supplies are direct coupled to each other...
They still wonder why it works fine one minute and not so good the next...
Not that it can't be done, technically spoken, provided you keep the lot at the same operating point, i.e. rock solid stable.

Ah well, Rome wasn't built in one day either...

Cheers,

[Sch3mat1c]

Originally posted by AKSA
I suggest driving grids via a cap from the plate. Not once did I mention transformer coupling. This is, of course, highly unusual; I can see why this confuses.
<SNIP>
Actually, you assumed grid current would be 100mA. It is seldom more than 25% of total plate current; so if the output tube is going gangbusters at say 120mA, then you could see up to 30mA grid current in deep A2. For those first couple of volts of grid positive, it might be 5-15mA, no more. In this scenario a plate loaded triode running 30mA has a good chance of meeting short term overload, but an 8mA CF rather less.

Many large triodes require significant amounts of grid current to get any useful power at relatively safe voltages (<1kV). 810 for instance. At most points, grid current tends to be a reasonable percentage of plate current, but when it's biting down hard, especially with plate voltage near grid voltage, it gets ugly. Even for a class A operating point.

Yes, 'waste' is probably fair enough, but have you considered the subjective sonic differences, or are you designing strictly for efficiency.

"Subjective" = "in the eye of the beholder" (or in this case, ear). I don't believe in there being significant differences between say, a Mullard and a Telefunken and I don't hear it either.

The sound isn't much different between these two amps here, Frankenhouse and Hept'AU7, as compared to my quad 6L6 amp. The former pair sound better of course because they're in stereo. Interesting, especially since Frankenhouse is a PP pentode amp with global NFB while Hept'AU7 is a ZNFB PPP triode amp. (Quad 6L6 is PPP pentode with global NFB and at least 60W available so it has far more dynamic range.)

I prefer cathode biasing because the sound is, to me, softer and more organic.

Electrically there is no reason for this (as long as the cathode is well bypassed). As such this is a psychoacoustic affect ("I believe it will sound different", and it does)..

You can if you use a large, grunty triode; my original proposition.

HMMM let's see.
6BX7 plate follower, cap coupled - plate resistance = 1.3k did you say?
6SN7 direct-coupled cathode follower - Zo = less than 300 ohms, far less with additional NFB applied locally
Now tell me, which one will go phut when subjected to a varying load? Which one will cause bias drift, and thus class AB2 operation when near full power?

And what's the problem with cap coupling anyway? You can drive a loudspeaker surprisingly well via a cap, all the better if it's a good one.

No duh, a loudspeaker draws as much current positive as negative. A grid only draws current when it's positive. Think about it a while.

A cap coupled plate to output grid reduces the chance of biasing accidents as well, particularly if the output tube is cathode biased. A DC coupled tube circuit is nice in theory, but as the tube ages, operating points cascade down the line.....

Which is why you move the cap between the driver and CF. What happens when the CF gets weak? What about the driver tube? Sounds pretty damned stable to me!

Tim