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

[mhuss]

The KT88s have a lower maximum input resistance spec compared to EL34s (100k for high power operation per GEC data). 6550s are worse (50k per GE data). The 6146B (the original SVT tube) is also 100k.

[Helmholtz]

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.

As long as the LTP is able to drive the power tube grids up to 0V in the positive going half-cycle (in other words produces a peak-to-peak voltage of twice the bias) in the actual grid circuit without significant distortion, the LTP is not a power-limiting factor in normal class AB designs.

The 4xKT88/200W Marshalls used a normal LTP with an ECC82 (39k or 47k plate resistors) and 68k grid leaks.
The 6x6550 Marshall JCM800 Model2000 250W amp uses a standard ECC83 LTP with 82k/100k plate resistors at 150k grid leaks.

[Dave H]

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

I think Pete is correct. With a high value 'choke' resistor the screen supply is going to sag considerably when sine wave testing and there's a lot of crossover distortion for the level of clipping on the first scope picture. I suspect it's because the screen voltage is sagging. I'd check it with a lower value resistor there to see if it does reduce crossover distortion and increase power output.

[nevetslab]

This has turned into an EXCELLENT THREAD! I've got a lot of studying to do to really absorb this. I love R.G.'s discussion on the development of the Pentode, the Tetrode, with it's 'kink' in the low voltage/high current part of their plate curves because of electrons literally bouncing off the plate. Suppressor grids near the plate let electrons sail through but suppressed the "bounced" electrons back into the plate, eliminating the "kink". Beam forming plates did something similar, but with fields instead of actual grid wires. Hence the name "kinkless tetrode" for the KT series.

I never knew KT was the the acronym for 'kinkless tetrode' in the KT series tubes.

School is back in session! Love it!

[nsubulysses]

Messing around with all this makes me realize I do not have a grid drive issue. Raising grid leak resistors to 150K or 220K does increase the drive signal but it still does not make the amp do more power at clipping.

I now realize I am limited by my 650R 20W "choke" dropping resistor, but I think mainly I am limited by my 1.6K primary 200W OT. I don't know why I didn't realize this before. I went with 1.6K because that was what the transformer manufacturer recommended for 6x6CA7. I should have come on MEF and asked for a recommendation instead! I bet with 1.1K or 1.2K like Helmholtz suggests I could get closer to my expected triple power of 210W or so.

I have done a two power tube amp with same PS voltages 580V B+ 565V screen and a 1.2K 10W resistor in place of choke. With 4K primary it does 70W or so at clipping
4 power tube version with 1.9K primary and 1K 20W resistor in place of choke does 24V into 4 ohms for 144W at clipping
6 power tube version with 1.6K primary and 650R 20W resistor in place of choke does about 27V into 4 ohms for 182W at clipping

Lowering 650R to 350R I get 189W
Lowering to 150R I get 196W

But I think it will be too hard on the screens to lower it this much. My problem is the OT primary I guess.

An important thing I realized is that when this 6 power tube amp clips the power tubes are only drawing 110mA or so. I normally expect to see my power tubes clip at about 130mA. I checked back on the 100W amp I have here (485V B+, 500R in place of choke, 1.7K primary OT) and it clips at about 130mA. I don't have a 144W 4 power tube version with the higher plate and screen voltages and 1.9K OT on hand but I am pretty sure that one clips at ~130mA too.

So I guess this amp is easy on the plate dissipation which I guess is nice, but it does not allow me to get the power I was wanting/hoping for.

[nsubulysses]

Thanks for all the help along the way

Anyone know primary impedance for 6xEL34 sound city 120 OT or 6xEL34 DR201 OT by chance?

I guess my recollection of the older 6 power tube amp I made doing 196W with 6X6CA7 was a fluke or I am remembering it incorrectly (but then again isn't it common for amp builders to stretch the truth a bit with these types of things similar to how they say people who like to go fishing do !!??!?!?! whoops)

[pdf64]

...4 power tube version with 1.9K primary and 1K 20W resistor in place of choke does 24V into 4 ohms for 144W at clipping
6 power tube version with 1.6K primary and 650R 20W resistor in place of choke does about 27V into 4 ohms for 182W at clipping

Lowering 650R to 350R I get 189W
Lowering to 150R I get 196W

But I think it will be too hard on the screens to lower it this much...

Bear in mind that regarding g2 dissipation, rather than the HT at idle, it's the fully loaded HT voltage to it that matters; so that may be down to <500V when the amp's pushing full output? So the situation may not be as harsh as you've thinking.

Here's the 60s 6CA7 info I meant to link to earlier http://frank.yueksel.org/sheets/127/6/6CA7.pdf
I was thinking that it may be best to use the 500k class B grid circuit resistance limit as that may make the operation a bit more robust.

[Helmholtz]

Bear in mind that regarding g2 dissipation, it's the fully loaded HT voltage to it that matters; so that may be down to <500V when the amp's pushing full output?

How this?

[pdf64]

I was thinking that at idle, g2 current per tube will be fairly low.

[Helmholtz]

I was thinking that at idle, g2 current per tube will be fairly low.

The manufacturers' screen voltage limit (500V for EL34/6CA7) holds for any situation with screen current. Like max. plate voltage it is obviously not dissipation based. Max. averaged screen dissipation (8W) is a separate limit.

[nsubulysses]

Bear in mind that regarding g2 dissipation, rather than the HT at idle, it's the fully loaded HT voltage to it that matters; so that may be down to <500V when the amp's pushing full output? So the situation may not be as harsh as you've thinking.

Here's the 60s 6CA7 info I meant to link to earlier http://frank.yueksel.org/sheets/127/6/6CA7.pdf
I was thinking that it may be best to use the 500k class B grid circuit resistance limit as that may make the operation a bit more robust.

I did not get to calculate or think too much about it it yet but I did make some notes while I was testing and adjusting the dropping resistor in place of the choke. all measurements are taken when the amp is brought up to the onset of clipping

650R resistor
Voltage drop over resistor: 84V
Plate voltage: 525
Screen voltage: 423
Power output: 182W

350R
Voltage drop: 53V
Plate voltage: 520
Screen voltage: 448
Power output: 189W

160R
Voltage drop: 25V
Plate voltage: 516V
Screen voltage: 465V
Power output: 196W

[Old Tele man]

Did you also check the screen dissipation wattage at that "onset of clipping" levels?

[nsubulysses]

I don't know how to do it the proper way. Nickb told me a good way to go about it sometime in the past but I remember it being pretty complicated.

I have crudely done it before by measuring voltage drop over 1K screen grid resistors at clip and at full blast (basically square wave) and multiplying it by what the screen voltage is.

How would you do it? I figured by measuring voltage drop over common 1K 20W resistor in series with plate supply you could estimate screen current if you divide by 6.

Is either of these ways a good way to do it?

[nsubulysses]

I have used a sine wave and have measured voltage drop over a screen grid resistor to gauge it. It is probably not truly a correct way to do it but it allows me to gauge how hard the job is of the screen when I am changing values adjusting for differentt B+s and different amount of power tubes scenarios. if I keep it within this range I am not too worried


4 power tube amp, 1K screen grid, 1K shared screen resistor, 585V B+, 560V screen

voltage drop over resistor -- screen voltage -- mA x screen voltage = screen disspation

at clipping
22V -- 410V -- .022 x 410V = 9 W

full power
30V -- 363V -- .03 x 361V = 10.4W

6 power tube amp, 1K screen, 650R shared screen resistor, 585V B+, 560V screen

at clipping
23V -- 408V -- .023 x 408 = 9.3W

full power
33V -- 326V -- .033 x 326 = 10.7W


Old story, but at first when I made the amp it was 4 power tube version 585V B+ and I used a choke, 1K screen grid resistor. it did 160W but after probably 20 hours of use it only measured 120W at clipping. Screens were getting crushed but no tubes ever shorted or fail. They just wore out. Using the same method as above I measured screen dissipation at clipping and full power and it was 14W and 17W.

[pdf64]

The manufacturers' screen voltage limit (500V for EL34/6CA7) holds for any situation with screen current. Like max. plate voltage it is obviously not dissipation based. Max. averaged screen dissipation (8W) is a separate limit.

I agree that we should endeavour to ensure new designs don't exceed the limits of the tubes used, but whilst not condoning the practice, my understanding is that the amp in question already tends to exceed that g2 limit, when idling / low signal levels at least.
The 6CA7 (vintage manufacture) g2 voltage limit looks to be lower than that of the EL34.
Note that many 6L6GC equipped amps breach the 450V g2 limit whilst at idle.