EL34/6L6, p-p, UL - What B+ is safe for driving those tubes?

[JWK]

I've done a pretty good search of the forum here and have found that the general consensus is that UL is much tougher on the screens when really driving the power amp compared to pentode mode. I saw one post where Dr.Z informed someone that you need to use a saggy power supply to help with that.

I'm wondering where you would need to be to not worry about driving the power tubes in UL. I'm not looking for a specific power level, so much lower voltages would be OK. I'm looking at the Weber PT here: https://taweber.powweb.com/store/022723sch.jpg

I figure this would give me around 410vdc when rectified with a 5AR4. With the 200mA ability of the PT and the 5AR4, it wouldn't really be that "saggy".

Does anyone know if either a 6L6 or EL34 are built with screens that can take this better than the other? I know Dr.Z has stated he doesn't like 6L6 with UL, but he said it was because of the way it sounds. This was probably ten years ago or more and he didn't go into any detail.

So I guess I'm wondering three things:
1. Considering a B+ of 410v, would UL be harder on the power tube screens than pentode. This would be a typical AB1 guitar amp scenario. Not even getting into source followers with mosfets for AB2 right now (although I've always been intrigued with that idea).
2. Does either EL34 or 6L6 tubes stand up better than the other under UL abuse?
3. Has anyone here really seriously messed with UL guitar amps? Maybe better would be to ask why UL isn't more popular here on ampage. There certainly is no technical difficulty involved. But that would be four questions, and I said I was only wondering about three.

TIA for any advice or opinion.

[ThermionicScott]

The conservative approach is to look up the max screen voltage in the tube book, and use that for your B+ at full power, when the power supply is sagging. Some post-1959 or so tube manuals give more specific advice when running UL.

Or you could just set up what you have and see if the screens glow or melt.

ETA: My dad and I built a 2x6V6 UL amp when i was in high school. With its B+ of 260V, the wattage is low, but it has a very deep and clean sound before breaking up.

-Scott

[octal]

You could also use screen resistors on the UL taps to limit the screen dissipation, in which case you could probably safely use 'standard' Fender/Marshall voltages. I imagine the presence of screen resistors might limit the effectiveness of UL operation at reducing distortion, but I don't honestly know for sure.

Supposedly Pentodes have a sharper increase in screen dissipation at max power than beam power tubes like 6L6es, so you're probably safer with 6L6es compared to EL34s.

Nathan

[nashvillebill]

Look at the Peavey Classic VTX amp, it appears to be running ultralinear with a B+ of 525V on 6L6's.

[TransLucid]

The maximum screen grid voltage for the el34 is listed as 425V, while for the 6L6 it's 450V, so, as long as you keep your B+ lower than that you should be ok.

More importantly, I believe UL operation is better for low-distortion applications (like bass or hi-fi amps) than for guitar amps. The Dynaco Stereo 70 used EL34s in UL, while the Fender Bassman 10 used 6L6s in UL with 470 ohm screen grid resistors.

I like octal's suggestion if you really want UL.
Bob

[Steve Conner]

6L6s are beam tubes designed for a fairly high screen voltage. So they should handle UL operation better than true pentodes.

The original Philips EL34 datasheet shows 1k screen resistors in the UL example circuit. So does the schematic for the Mullard 5-20.

UL is just another kind of negative feedback, it decreases distortion and increases damping of the speaker. Some people didn't like the sound of the UL Fender amps, but I read somewhere that removing the NFB loop makes them sound a lot better. I think the moral of the story is that guitar amps only need a modest amount of feedback, so if you're using UL, you don't need to bother with any other NFB.

The converse is probably true, if you use NFB, you don't need to bother with UL.

[pdf64]

Check out
Dr. Z Amplification | Bias Notes
I think it should provide some useful info for you, especially paragraph 7 onwards, re B+ / tubes.
As you can see, Dr Z seems to prefer KT66 for octal type power tubes, at up to 480v B+.

[JWK]

6L6s are beam tubes designed for a fairly high screen voltage. So they should handle UL operation better than true pentodes.

The original Philips EL34 datasheet shows 1k screen resistors in the UL example circuit. So does the schematic for the Mullard 5-20.

UL is just another kind of negative feedback, it decreases distortion and increases damping of the speaker. Some people didn't like the sound of the UL Fender amps, but I read somewhere that removing the NFB loop makes them sound a lot better. I think the moral of the story is that guitar amps only need a modest amount of feedback, so if you're using UL, you don't need to bother with any other NFB.

The converse is probably true, if you use NFB, you don't need to bother with UL.

Absolutely. I remember talking to KOC on the phone many years back, and he told me then the amount of local feedback provided to the power tubes from a typical 40% tap was about the same as a Marshall plexi's global feedback. I don't know how accurate that is, but I have messed with an old UL before and realized they work/sound great without any global. You lose the ability to adjust the feedback, but you gain design advantages.

[JWK]

Check out
Dr. Z Amplification | Bias Notes
I think it should provide some useful info for you, especially paragraph 7 onwards, re B+ / tubes.
As you can see, Dr Z seems to prefer KT66 for octal type power tubes, at up to 480v B+.

That is interesting. Thanks for the link. The KT-45 is an EL34 amp which has a B+ of 480v with a SS rectifier. I was always under the impression that the KT-45 was not a "saggy" amp at all. Hmmmm.

[TransLucid]

As JWK points out, UL feedback is local, while NFB from the output to the PI is more global, so the two will act differently.

[Enzo]

That PV Classic VTX does run with 525 on the plates, but the cathodes also sit at about +80, so there is only about 445v across the tube.

Why don't more mess with UL? I'd have to think that while there are tons and tons of regular old OTs, you have to hunt up a UL OT for your projects or have one made, when you want to go UL.