How long to pre-heat filaments on 6SN7s?

[cbutterworth]

I want to idiot-proof my system. I have an Aikido preamp using 6SN7's, which feeds a solid-state power-amp.

Right now, I am using an Amperite 30S delay tube to pre-warm the filaments. before B+. Then I switch-on the power-amp. I shut things down in reverse, with the power-amp first, followed by the preamp. I can switch both off at the same time without any thumps, but if I turn on both at the same time, I get a very audible thump when the B+ switches on.

I just ordered a Belkin PF60 power conditioner, which can switch on different banks of plugs with a delay of up to 15S. Hence the following question:

QUESTION: If I replace my Amperite 30S delay with an Amperite 10S delay, then my 6SN7's will only get 10S of filament warm-up before the B+ switches on. Will this be detrimental to the tubes?

I vagualy remember reading that as little as 5S heating is sufficient (maybe this was in the Morgan Jones book).

Thanks,
Charlie

[Brian Beck]

I know that the 6SN7-GTB had a controlled heater warm-up time specified at 10.5 seconds. But this had more to do with heater current settling down than it did with cathode temperature and emission. Still, I'll bet that any version of 6SN7 will work fine with 10 seconds of delay between heater power and B+.

[cbutterworth]

Brian,

That was my guess. I suppose that the only risk is cathode-stripping and as far as I can tell, this is only a real issue at very high voltages, while I am only running mine at 280VDC (less per cathode).

In fact, when trying to trace hum, Bruce Anderson had the preamp running without any heater on at all (using cheap tubes of course).

Regards,
Charlie

[Brian Beck]

You'll be fine. The other risk is running the heater for long periods of time without B+, where cathode interface or poisoning can occur.

[El Zombre]

This is only a tangentially related question, so excuse my thread jacking and also my ignorance:

Lots of guitar amps have a standby switch, which I've gleamed to serve the same function as this delay of which you speak. So why don't more hifi amps just use a standby switch?

[cbutterworth]

El Zombre,

Good question. I expect that in tubed guitar amps, the standby switch serves to switch-off the B+ voltage, while keeping the heaters running at either 100% or lower. There have been threads on this.

It certainly makes sense with guitar amps, where a guitarist may play for while then take a break. However, I guess that most guitarists would fully power-down the amp for longer periods.

For audio use, I would hazard a guess that the tubed amp is either left on for a couple of hours playing or is fully switched-off. Considering that some of the popular tubes such as 6SN7's are getting expensive, fully switching off could extend tube-life.

In other words, I don't know what the answer is, and I am just guessing (or BS'ing).

Charlie

[jarthel]

Originally posted by Brian Beck
You'll be fine. The other risk is running the heater for long periods of time without B+, where cathode interface or poisoning can occur.

how long is "long periods of time"? what's ideal delay?

[Brian Beck]

That’s a good question, for which I have no crisp answer. Of course, it will depend on the tube type, and on the frequency and duration of “abuse”. All oxide-coated cathodes will eventually develop some interface; it’s part of aging – like wrinkles. But I think you can hasten this aspect of aging by frequently holding the heaters on without B+. I would guess (only a guess) that keeping the heaters on for a minute or two without B+, if done every day for months, might degrade the tube faster in the “out years” than if it had gotten B+ after 10 seconds or so each time. Perhaps someone can cite a study on this. BTW, computer triodes were designed to be switched “high” (no current) for long periods of time with heaters cooking. They used special cathode materials to reduce interface formation.

Perhaps this is one reason that many people like the sound of computer tubes like the 7044, 7119, 6350, 5963, 5965, etc. I know I do. The 5687, especially the 5687WA, is spec’d as an amplifier but some of the data sheets comment on its special design for low interface formation, too.

[kmtang]

I have built one Aikido with 12AU7 and 5687 as buffer.

As 5687 takes large amount of filament current, it takes sometimes for the amplifier to stabilise.

I would suggest you could measure the anode voltage of the two 6SN7's at the bottom to see how long it takes to stable. Then, you can add some more margin to it for the B+ delay circuit.


Johnny

[SY]

Well, Brian, I think you argue for the way I do it, just switch on everything at the same time. The voltages in small-signal service are way below anything that will strip the cathodes so I think that those fears are groundless- I think we could both name lots of pieces of classic equipment where the B+ was switched on at turn-on yet tube life was normal-to-outstanding.

Cathode poisoning is a hazard with a lot more solid data behind it.