Understanding "grid drive" to drive, 2, 4, 6 power tubes

[Helmholtz]

I do not understand the point of this discussion. If you want to make a guitar amp with multiple tubes in parallel, you want it to sound as an amp with one tube per side does. So you keep the same ac coupled drive, but lower the impedance so that the drive per tube is the same.

I assumed that the discussion should include the question why some 6 bottle power amps actually use additional cathode followers to increase drive current capabilities. My answer is that such designs allow for class AB2 operation which pushes clean output.

[Mike Sulzer]

I assumed that the discussion should include the question why some 6 bottle power amps actually use additional cathode followers to increase drive current capabilities. My answer is that such designs allow for class AB2 operation which pushes clean output.

That makes sense. I think n... just wants to make a really powerful guitar amp with the expected overload characteristics.

I also worry about melting output tubes if you make a guitar amp with true AB2 dc buffered bias since you lose the protective feature of the bass shift with ac coupling.

[Justin Thomas]

Do you need more than 100W?
Sure, but for Bass, Keyboards, PA or Pedal Steel ...... all of which are happy with SS.

Oh, I'm just ribbing ya! You're one of like, 3 people whose SS guitar amps I'd play. As far as keys go, I will say this: every time I've plugged a keyboardist into my Concert or Bassman or Twin Reverb has said, "wow, that sounds so much better!" So they might be "happy" w. SS, but I can't fight the ears.

As far as bass goes, I have yet to hear anything better than that 73 SVT pushing a fridge. It was in a warehouse, we had it on 7, and it was definitely not a club amp, but it's worth having for the times you could use it... Too bad it literally did not fit in my friend's car - $700 would have been a steal.

And I would still like to dream about building one of those 1,000W PAs out if the GEC Applications Manual...

Justin

[Helmholtz]

Screen 560V

This seems very high. Spec max is 500V.

[nsubulysses]

Lowering the 100K grid leak resistors seems to help, although I have not gotten a chance to play it through speakers and listen yet, only view it on test equip. When i scope the output of the amp (blue) as well as the control grid of the power tube (purple) they get a weird peak in them with 100K grid leak resistors (R1, R2). When I lower them to 47K or 33K the peak goes away and the signal at the control grid looks way more similar to the signal at the speaker out.

100K grid leak resistor


47K grid leak resistor


Lower PI plate resistors to 47K or 33K seems like it has a minimal effect. Maybe it's a huge audible effect but on test equipment it doesn't seem to do much. Also, changing to 12AU7 or 12AT7 didn't seem much different as far as raising max power at clipping.

My B+ is 580V and at clipping it dips to about 515V.

I don't have a "high power" 4 power tube amp on hand right now that has the same high PS voltages as the 6 power tube version. All I have on hand now is a 4 power tube 100W version that has "normal" power supply voltages. So B+ is 485V. When that amp hits clipping the B+ drops to about 420V so it's pretty comparable to the 6 power tube amp. Scoping my 4 power tube amp I get the same looking images as posted above when I scope power tube grid and signal at the speaker jack. On my 4 power tube 100W amp if I lower the 100K grid leak resistor to 47K it does the same thing as well, it makes the signal at powr tube control grid look way more like the signal at the output. Maybe this has to do with what NickB mentioned in post #12.

I'm guessing this displays what G1 says which is extra current drive is not a necessity in a 6x6CA7/EL34 design? I get about 90-100V pk-pk on power tube grids when output at speaker jack starts to distort.

So that's how far I've gotten so far. Thanks for the suggestions. Still trying to figure out what all this means.

No matter what it does not seem to raise my total clean power at clipping. Maybe I am remembering wrong but I built a same 6 power tube amp about a year ago and I remember it doing 28V into 4 ohms for 196W at clipping with 6x6CA7 and 29V for 210W at clipping with 4xKT88. This current amp is stuck in the 26-27V at clipping range hmmmmm. The only difference is this amp has an FX loop in V3 position and the older amp had no FX loop

for FX loop version it is V1 gain stage 1, V2 gain stages 2 and 3, V3 cathode follower for FX loop, V4 cathode follower for FMV tonestack, V5 LTPI

The no FX loop version just has V3 omitted

not sure why that would matter but just mentioning it anyway in case

I always test amps at 122VAC from a variac to mimic 120VAC from the wall. So it's not like I am just measuring it at a moment when wall voltage is lower than average

I will be rereading these posts and some various authors chapters on phase inverter design in the meantime because I still don't really understand what's going on.

[nickb]

Lowering the 100K control grid resistors seems to help,

I think you may be misinterpreting the results. Lowering the 100K resistors made things worse. The Vpp drive is reduced so the clipping you see is caused by the PI not the final output stage.

[SoulFetish]

That's why I mentioned AB2 above. I think the SVT with cathode follower drive is designed for AB2. This operation mode requires low impedance drive to keep distortion at the non-linear grid input impedance down.

The SVT is not designed for class AB2. The 47k grid stoppers will prevent the grid from swinging positive once the grid begins to draw current

[pdf64]

With all the referencing of the Super Twin and its 33k grid leaks, it's worth noting that the 6CA7 tube nsubulysses is using here has a much higher limit of ~700k for its grid circuit resistance; hence the regular 220k may not be too far off the mark, and even playing it super safe with 150k would still make the drive requirement way easier.

...No matter what it does not seem to raise my total clean power at clipping. Maybe I am remembering wrong but I built a same 6 power tube amp about a year ago and I remember it doing 28V into 4 ohms for 196W at clipping with 6x6CA7 and 29V for 210W at clipping with 4xKT88. This current amp is stuck in the 26-27V at clipping range hmmmmm. The only difference is this amp has an FX loop in V3 position and the older amp had no FX loop...

My guess is that your 1k HT dropper for the g2 node may be significant in constraining your latest amp's power output; big power outputs require stiff HT nodes. I suggest to swap in a 100 ohm resistor or preferably choke with a suitably low resistance and sufficient current capability, and re-asses.
If still disappointing, check that the OT's primary resistance is suitably low; with big power amps, resistive losses there can be significant.

[Helmholtz]

The SVT is not designed for class AB2. The 47k grid stoppers will prevent the grid from swinging positive once the grid begins to draw current

Good point!
High value grid stoppers limit grid current necessary for real class AB2.

[Helmholtz]

When i scope the output of the amp (blue) as well as the control grid of the power tube (purple) they get a weird peak in them with 100K grid leak resistors (R1, R2).

Did you disconnect the NFB loop for these measurements? Feedback from a distorted ouput will produce artefacts in the drive signals.

[Dave H]

When i scope the output of the amp (blue) as well as the control grid of the power tube (purple) they get a weird peak in them with 100K grid leak resistors (R1, R2). When I lower them to 47K or 33K the peak goes away and the signal at the control grid looks way more similar to the signal at the speaker out.

100K grid leak resistor
[ATTACH=CONFIG]51729[/ATTACH]

47K grid leak resistor
[ATTACH=CONFIG]51730[/ATTACH]

Lower PI plate resistors to 47K or 33K seems like it has a minimal effect. Maybe it's a huge audible effect but on test equipment it doesn't seem to do much. Also, changing to 12AU7 or 12AT7 didn't seem much different as far as raising max power at clipping.

The top picture with the asymmetric grid signal looks 'right' to me. If you switch the scope to DC coupling on the power tube grid channel you should see the top peak of the signal clamped at 0V by power tube grid conduction and the bottom peak should be more negative than twice the bias voltage. If that's the case then I think it means the PI has enough voltage swing to drive the power tubes to full power but to be sure you could change the grid leak resistors to 150k as Pete suggested. If that doesn't produce more output power then I don't think the PI is the limitation.

The bottom picture with the symmetric grid signal looks like it's the PI clipping before the power tubes which will reduce output power.

[nsubulysses]

With all the referencing of the Super Twin and its 33k grid leaks, it's worth noting that the 6CA7 tube nsubulysses is using here has a much higher limit of ~700k for its grid circuit resistance; hence the regular 220k may not be too far off the mark, and even playing it super safe with 150k would still make the drive requirement way easier.


My guess is that your 1k HT dropper for the g2 node may be significant in constraining your latest amp's power output; big power outputs require stiff HT nodes. I suggest to swap in a 100 ohm resistor or preferably choke with a suitably low resistance and sufficient current capability, and re-asses.
If still disappointing, check that the OT's primary resistance is suitably low; with big power amps, resistive losses there can be significant.

Thank you this is kind of waht I was wondering. Everyone kept talking 33k but I am not using 6L6 so i didn't know if they were overlooking me saying "i am not using 6L6"

Also 1K 20W was kind of a mis statement. I use 1K 20w in the 4 power tube version with B+/screen at 580V/565V. in the 6 power tube version with same high voltages I use 650 ohm 20W because it seemed to give roughly the same measurements for screen dissipation that I found to be basically safe and not too punishing for hte power tubes. I can lower it some but it would probably be too hard on the screens since I have 565V at idle. I will check out this option too though

[nsubulysses]

Helmholtz, I did not disconnect NFB whooops. I will be going to my workspace today though and can do more tests

700K max grid circuit resistance seems to mean that I could even use a single 220K resistor since 700K / 3 is 233K?

Hiwatt DR201 schematic I posted earlier has 6 individual 220K resistors for each grid leak. So I guess that totals out to 73K with 6 power tubes. I think RG was hinting at this in an earlier post -- would there be much of an advantage to have 6 individual grid leak resistors vs just the one shared grid leak per side. I don't really wanna add 4 more resistors but I guess I would if it would make the amp better!

I will increase R1 and R2 100K grid leak resistors to 150K or 220K later today and report back

Thanks for all the help.

Im guessing the version I built before with no FX loop put out a bit more power because it has even MORE voltage swing in the PI compared to the FX loop version because the signal doesn't have to be knocked down to pedal level in V3 stage, so the version with no FX loop delivers a bigger pk-pk signal to grid of PI? I notice this phenomenon in the master volume. Both amps can achieve the same volume but I usually have to turn the volume up slighly more on the FX loop version vs the no FX version.

[mhuss]

The Hiwatt 6xEL34 DR201 has 220k grid resistors, so (reasonable) ~73 k per side, and it uses an ECC81/12AT7 as a PI tube. The 400-watt 6xKT88 DR405 does have a driver tube.

[nsubulysses]

With mhuss mentioning Hiwatt DR405 and g1 mentioning that Hiwatt STA400 both are 6xKT88 and use the driver tube arrangement (and SVT also) it seems to be a the preferred method with KT88 vs EL34. Why is this? Is it just in general a better method or is it for some reason a necessity with KT88 vs EL34.

If you have enough "grid drive" to achieve full power, would using a PI with cathode followers to provide more grid current rather than just the regular old LTPI sound better in some way when the amp is cranked to overdrive? I guess if I switch to 12AT7 instead of 12AX7 I can view this slightly