Triode / pentode switch: can it be switched "live"?

[Mostro]

I posted the question in TAG, but I figure the more answers the better.

Can you switch a pentode/triode switch with the amp on? I´ve read people saying no, it has to be off or in standby. My question is, why?

Is it because you will get a big POP? Or will you bust something? Assuming the switch is a break-before-make type, I think this: when pentode connected all is fine. When the switch breaks the contact, you basically put the valve into standby because the screen gets no voltage. When the switch makes contact with the plate supply, it will get a rush of voltage / current, but if there´s a screen resistor it will be current limited. In my case I have 1K R´s.

So, any technical reasons why I should or shouldn´t do it? Experiences?

[firemedic]

It can cause arcing on the switch, no bueno. This was my experience, and I got sick of changing out shorted switches, so I never switch with standby on any more.

[HotBluePlates]

Is it because you will get a big POP?

It can cause arcing on the switch, no bueno.

Stop and think, "Why would I get popping?"

Because the screen and plate could be at different dc voltages when in pentode mode, the act of switching means you'll get a d.c. voltage shift and likely a transient thump through the system. At least if you're in standby that big thump won't go to the speaker, because there should be no plate or screen voltage.

If the switch was experiencing arcing, removing that voltage during switching also prolongs switch life, as firemedic pointed out.

... Assuming the switch is a break-before-make type, I think this: when pentode connected all is fine. When the switch breaks the contact, you basically put the valve into standby because the screen gets no voltage. When the switch makes contact with the plate supply, it will get a rush of voltage / current, but if there´s a screen resistor it will be current limited. ...

A break-before-make seems like it would be the worst-case type of switch to use.

Here's what I'm thinking: If the problem is shifting d.c. voltage levels during switching, then when you switch from pentode mode to triode, instead of the screen rising maybe 10's-of-volts (if that much) from screen voltage to the plate voltage, it will first drop to zero then rise hundreds of volts from zero to plate voltage. This magnifies the problem of d.c. voltage-shift.

If you use a make-before-break, the tube is in triode mode at any instant that it is not 100% in pentode mode (i.e., during switching).

Anyway, it just seems easier to switch while in standby. Do you really have a reason to switch while the amp is playing? If you've not played much with pentode/triode mode you might think it is something you'll want to switch on the fly, but I think you'll find a preference for one or the other and stay there most of the time.

[Jack1962]

If this amp is a factory built amp I strongly recommend you read the manual , No you should not flip that switch while the amp is on , when you do it slams the screen grid , it will kill your tubes and it's not doing your output transformer any favors either.

[Mostro]

Guys, thanks so much for the answers.

The idea is to apply the switch to an AC30 I´m building, in wich I´m trying all sorts of things to lower volume. I have a VVR, a two valve disconnect switch, a cathode cap decoupling switch in it, and thought the triode/pentode switch could be a nice addition to the rest of the things.

Yes, all at the same time! Quite a populated back panel!

Since the amp is on the bench, not finished yet, i´m trying everything that comes to mind, eventually I´ll take out what doesn´t work and finish the schem and the build.

Since all the other volume reduction options work "live", I thought it would be nice that the t/p switch also did. I never used/had an amp with it, so I don´t know what to expect from it sound wise. Also, if someone ever borrows the amp, I don´t want a control in it that has the potential to blow up the amp or the power valves...

At TAG someone suggested a 1M or thereabouts R at the screen supply leg of the switch, bridging it to the screen R/switch common. what do you think of that?

[HotBluePlates]

At TAG someone suggested a 1M or thereabouts R at the screen supply leg of the switch, bridging it to the screen R/switch common. what do you think of that?

The pentode/triode switch (for a single tube) is a SPDT switch: The screen (and screen R) is the switch common (center lug), the screen supply node (for pentode operation) is one outer lug, and the plate/OT connection (for triode mode) is the other outer lug.

So I'm thinking they mean a 1MΩ bridging from one outer lug to the other, or between the two settings that the screen is switched. I don't think that would do anything.

Alternatively, they may mean a 1MΩ from each outer lug to the screen node, so that the switch is shorting a high-valued resistor. This might reduce popping to/from the pentode setting. I'm not sure it will do anything for the triode setting (but that may be irrelevant). The concept is that the high-valued resistor provides a charging path for the filter cap to minimize current in-rush.

I dunno if adding the resistors will help. But they won't hurt anything and are cheap and easy to try.

... Also, if someone ever borrows the amp, I don´t want a control in it that has the potential to blow up the amp or the power valves...

I don't believe you have to worry about that.

The switch arcing is internal, and causes an unnerving loud thump. With repeated arcing, there is an increasing chance that a carbon track is formed between the arcing contacts. Once that carbon track is sufficiently established, the switch will be shorted internally between those contacts, and will seem to stop working properly.

The net result should be a permanently shorted switch that needs replacing to regain functionality. I don't see an obvious failure mechanism that could harm the tubes or the OT.

[rzenc]

Experiences?

One of my personal amps has a 3/5 mode switch. I have flipped it while playing and no damage has ever happened. I run 6550's in it. Also, there is no pop while playing and even with the amp on and no playing I get a small click when switching between 3/5.

Bets Regards

R.

[HotBluePlates]

... One of my personal amps has a 3/5 mode switch. ...

3/5 mode switch = "Triode"/"Pentode" mode switch

[rzenc]

Can you provide the schematic or at least the name of the amp with the 3/5 switch?  Thanks.

It's one of my own amps. I built it in 2008. I'm far from home for a couple of days, but will send you the schem if you are interested.

Best Regards

R.

[Mostro]

OK guys. I tried the switch with a 1M R bridging the common with the B2 contact. A small sliding 2PDT. First turning the amp off and then, with some trepidation, with the amp on. No problem at all, no sparks, shorts, blown parts or elctrocuted people (me). Just a mild pop.

Sooo, I guess you can do it "live". Will the switch break with use? Maybe... I really don´t see it happening, after all the currents are low... is the R helping? Well, if it ain´t broke don´t fix it! Meaning, I won´t try it without the R.

Thanks all for the help.

[HotBluePlates]

Awesome! I never argue with experience.

Please do let us know if you do run into switch failures in the future. Maybe it's something to watch out for, or maybe it's isolated to some switch models.