EL34/6L6GC switchable bias question

[Tooboob]

Got it. I was misunderstanding the post. I confused "doesn't get above 0V" with no grid conduction, and I though there was some "new to me" magic going on, and yes, the grid stops usually come after the bias resistors.

[g1]

The prince and the count always insist on tubes being healthy before they're broken

I admit it, you are better than I am.
Then why are you smiling?
Because I know something you don't know.
And what is that?
I am not single-ended.
You're amazing!
I ought to be after twenty years.
There is something I ought to tell you
Tell me.
I'm not single-ended either.

[NateS]

You can blame RG's immortal amp manifesto for that. I always think of the hunchback in the pit of despair when the conversation turns to soft start, current limiting, runing filaments undervoltage, crash test dummy levels of instrumentation and other means of pampering... only to jam enought current down the grids just short of melting them.

[NateS]

I just reread my AB2 comment and feel it needs ammending. Allowing grid current to flow actuall softens the clipping of the wave tops, resulting in lower order harmonics. It looks like mostly 3,5 and 7 and it sort of peters out after that.

[J M Fahey]

How does that work? a source into the grid stopper (if the grid isn't driven positive) only sees the grid stopper and the bias resistor to the bias supply, forming a divier at the grid.

No divider, since those are different paths.
Audio voltage coming through a coupling cap will see 2 paths:
1) through grid resistor (usually 220K) to audio ground so basically it´s the load it "sees".
2) through grid stopper ( a few K´s ) into a grid which if negative, can be seen as an "infinite impedance".

1. Is this always enough to keep the grid below the cathode

No, the grid can and is driven positive with respect to the cathode with large signals (typical of overdriven guitar amps).

(usually ground or near it for AB)?

Not sure what you´re asking.
You mean "the negative biased grid can be driven up to ground (cathode) voltage"?
If so, yes, and even beyond, into positive with respect to cathode levels.

2. Is the grid stopper usually chosen for this behavior as one of it's criteria or its main criterium?, Does it just work out that way?

Grid stoppers are originally used for stability, cutting supersonic gain combined with Miller capacitance (which power tubes also have).
A secondary benefit is that when grid is driven positive and the grid-cathode diode gets forward biased, which dynamically messes the bias, (easy to see on a scope), they help limit that by limiting the rectified current charging the coupling cap.
A little is fine, it´s part of the overdriven amp sound, a lot causes farting blocking distortion which is *ugly*.

3. Why are folks buffering the power tube grids with FETs if a large enough grid stopper solves the problem?

Contradicting statements
1) folks add FET buffers (same as SVT designers add triode buffers) to allow some extra positive voltage in the power tube grids and , yes, increase available power.
As Enzo always says, difference between 300 and 250W RMS is nil, but it sure looks good in ads and brochures .

4. ls there an advantage to driving the grid positive?

Well, you get some honest extra power

Does the power tube sink ia higher current past that point?

You bet, just look at tube curves in the datasheet.

You won´t be able to drive grids 10 or 20V positive like in a transformer driven class AB2 amp, (although FETs might) but a couple volts might be achievable.
A typical 12AX7/T7 PI will be able to supply 1 or 2 mA tops (whatever flows at idle through those 47K or 100K plate resistors), 12AU7 can supply way more since they can be used with plate resistors as low as 10/12K, 12AU7 class tubes used as cathode followers too, and with FETs the sky is the limit.

Contrary to popular belief, tube grids can be happily be driven positive, you just must be able to afford the current.

I have successfully made a true 2W RMS power amp with a 12AU7 , with direct coupled grids driven by TL072 and BC547 as current boosters.
Way too loud for a bedroom amp, whichwas the original intention

In case you wonder: yes, it drives grids positive seamlessly.
You have an ugly "jump" when you go from negative to positive grids ... if you don't have the current capability, .... but if you do, transfer curves are very linear and give no ugly distortion.

Remember, you need to have the necessary "balls" to supply voltage without worrying about current .... which an Op Amp does beautifully.

Here´s positive grid curves, for 12AU7 .
They can supply amazing 150mA !!! as long as you can feed their hungry positive grids.

[Tooboob]

Thanks for that ...

I think we're all on the same page regarding my original question. FET buffers don't just let you drive beyond 0Vs. They stop blocking distortion (and a little blocking distortion could be useful too...).

NateS's comment on smoothing power tube clipping by driving grids into conduction is intriguing. It looks worth building something so I can hear it. I like the sound of guitars more than hard clips, and I lean on delay and chorus a lot, but a compressive amp is a blast to play.

JM - a 2W 12AU7 amp - too loud. You're definitely out there. A 2W tube amp that sounds good has lots of applications.

I thought the power stage stuff was simple ... There's much to hear. So much to explore.

[Tooboob]

Power Scaling implications ...

I've studied O'Connor on Power Scaling. He does a good job with the power tubes, but leaves the question somewhat open on what to do with the PI and earlier circuit. Most folks Power Scale the PI too. This keeps the door open for PI distortion, and keeps the drive from pegging to way too high a voltage when the power tubes clip, but it requires an adjustment to the input level to the PI. You wind up with interaction between the Power Scaling knob and some upstream volume control.

Exposing grid conduction capability makes the job even harder. Lots to think about.

Do FET buffers on the power tube grid drive change the power tube distortion beyond eliminating blocking? It looks like "yes".

Is 6L6 power tube distortion dominated by grid voltage clipping near 0V Vgc? Looks like it might be...

[g1]

JM - a 2W 12AU7 amp - too loud. You're definitely out there.

He did say for bedroom applications, and I think he meant cranked wide open for power tube distortion.
Guy's that build micro-amps for those purposes often say even half a watt is too loud. Probably for situations like late night in apartments to be able to crank wide open without waking the baby or disturbing the neighbours.
Also there are modern amps with built in attenuators (or PI limiters) with 2W settings that users claim are too loud for bedroom practice.

[NateS]

It's that first watt that's the doozy. After that, they're not so bad.

[Tooboob]

I can relate. 5W can be almost painfully loud in a 12x12 room. I have a large collection of amps, and most do a great clean clean at low volume, but if I want to practice high-gain, I use my old GNX4 modeller and headphones. I have a great after-market Engl model. I do think you're better off practicing clean, or acoustic, to highlight flaws in technique, but a high-gain setup is like another instrument due to the compression and sustain, and it demands some attention.

A bedroom solution needs to get down to 0.1W if you're married and play like me.