Stanby switches

Paolo_M · 07-11-2008 5:00pm #1

A while ago there was a thread on an AC30 and the discussion came around the rectifier valve. At the time I mentioned I had read previously that a stand by switch was not actually required on a guitar amp. I just stumbled across the atricle that I was refering to and I thought some of you might have an interest in reading it.
Basically a standby is not required and may in fact cause damage to amps with valve rectifiers and could also cause preamp and output valve damage if left on for too long while powered up.

All of the information this guy provides is correct and can be cross referenced.
Anyway here goes:

Copyright 2006 © David B. Lamkins How to use a standby switch

There's a lot of information on the `net and in manufacturers' literature about how to use the standby switch on your tube amp. Not surprisingly, there are a lot of different procedures. In the words of at least one frustrated player, "there's not a lot of concensus among the experts."

Most of the experts cite the "cathode stripping" boogieman to convince guitarists to religiously follow a standby switch ritual. Their notion is that applying high voltage to the circuit before the cathode is fully up to working temperature will physically strip the delicate coating from the cathode and prematurely age the tube. You'll be left with a tube that doesn't work and has a bunch of white fluff (the stripped cathode coating) floating around inside the glass envelope. Expert opionions vary regarding how long you should wait before turning your standby switch to the "play" position; I've seen times ranging from thirty seconds to five minutes. And some experts even recommend that you put your amp on standby for some period of time to give your tubes a chance to "cool off" before turning off the power.

Let's look first at the range of "wait" times cited by the experts. Heaters come up to temperature in about 15 seconds. You can find out by powering-up your amp without using the standby switch. (I'll bet that makes you nervous, doesn't it? Read on...) Even if there was something to the threat of cathode stripping, the threat would be eliminated as soon as the cathodes came up to temperature and the tubes started functioning. So what's up with the recommendations to wait sixty seconds, or two minutes, or ... five whole minutes? If fifteen seconds is good, then three hundred seconds must be really, really good ?

Cathode stripping is real. (Sit down. I haven't misled you.) It happens at very high voltages and currents. If you go back to the manuals and engineering texts of the `50s and `60s you'll discover that no one ever wrote about cathode stripping with respect to "receiving tubes" (as compared to "transmitting tubes"). "Small" tubes like the 12A?7, 6L6, 6V6, EL84, EL34, EF86, ... are receiving tubes. With the exception of a few tube amps having very high (~700V) plate voltages (like the SVT and a few high-powered Marshalls), you don't really need a standby switch except as a convenience for muting the amp; I'd rather just pull the cord from the input jack or turn down the volume control on the guitar. There is no way that cathode stripping can occur in preamp tubes, even in an SVT; the available energy is limited to a very low value by resistors in the plate circuits.

OK, so what about that bad tube you found with tiny bits of white fluff rattling around inside the glass bulb? Yes, that can happen. It's a result of poor bonding of the cathode coating combined with temperature cycling and vibration. It's not caused by cathode stripping.

Then you have to wonder about those amps that don't have a standby switch. The experts will tell you two things about that:
You should probably have a qualified tech add one just to be on the safe side. Hey, techs have bills to pay just like the rest of us...
Some amps don't need standby switches because they have "controlled warmup" vacuum tube rectifiers which don't reach their working temperature until all the other tubes are ready to go. Nice try, but not quite right or consistent with other expert advice... Rectifiers start working before they come up to full temperature, so there's some DC on relatively cold tubes throughout the amp. And given that the rectifier heats up about as fast as the other tubes, what becomes of the advice to hold off operating power until the tubes have warmed up for some number of minutes? Hmmm?
Standby switches introduce their own problems. They're not rated to break DC at the voltages present in any tube amp, let alone the bigger ones. (Check the ratings on a switch some day... The DC rating is always much lower than the AC rating.) When you break high DC voltages, the switch arcs and will eventually fail. It's a lot tougher to swap in a new standby switch than a new tube...

Switches are most prone to arcing as they open a circuit. In the case of the standby switch this happens when you switch the amp to standby. However, switch contacts "bounce" when you close a circuit; arcing may also happen on each bounce before the contacts finally come to rest. Arcing in this case is much less severe because there's less stored energy (see below) in the circuit as you're switching the amp off standby.

Arcing depends somewhat upon the circuit design. In general, though, standby switches are exposed to DC voltages well in excess of their design center. The reason it's so hard to make a mechanical high-voltage DC switch is because an arc is formed as the contacts open. The arc tends to continue so long as there's enough voltage to maintain it as the contacts separate. In circuits where there's a big inductor (like a filter choke) attached to the switch, the collapsing magnetic field (created by breaking the circuit) creates a high voltage that's sufficient to cause arcing. The duration of the arc is related to the amount of energy stored in the inductor.

A switch can and does arc with AC across it. However, the duration of the arc is limited because the AC voltage goes to zero 120 times per second (100 times per second in Europe). As the voltage approaches zero it very quickly becomes too low to sustain an arc, and once the arc is quenched it tends not to recur because the contacts have moved far enough apart that the voltage can't break down the air between the contacts. This is the reason that a switch has a higher rating for AC voltages than for DC.

Here's one more thing for you to ponder: "cathode poisoning". Normally the heated cathode throws off a cloud of electrons that get immediately pulled away from the cathode by the electrostatic field created by the plate voltage. When your amp is on standby you don't have any voltage on the plate, and that electron cloud hugs the cathode. Some of those electrons bump into the cathode coating and cause an electrochemical reaction which reduces the efficiency of the cathode coating which in turn reduces the useful life of your tubes. Unlike cathode stripping, cathode poisoning is not dependent upon how high the plate voltages are; when the amp is on standby there is no plate voltage. So now you have one more thing to worry about: If you put your amp on standby, how long before cathode poisoning becomes an issue? Is it cumulative?

You could lose sleep over things like this. Just keep in mind that no item of consumer electronics gear (and for that matter it applies to the small sampling of industrial electronics gear I've seen) made in the heyday of vacuum-tube electronics had a standby switch, except for guitar amps (which don't really need one) and ham radio transmitters (which really must have one). When you want to use the gear, you turn it on. When you're done, you turn it off. No rituals needed...

Frankly, with the exception of the arcing issue (which can cause an amp to become scratchy, poppy and to lose power because of carbon build-up due to arcing on the standby switch contacts), standby switches on most guitar amps (remember what I wrote about amps with very high plate voltages...) are harmless.

By all means use the standby switch if it makes you feel good. We all need rituals. But don't lose sleep over having powered your amp on (or off) the "wrong" way. It should be quite obvious from this discussion that guitar players follow a lot of different standby-switch rituals and no one's really suffering from having used any of the techniques described...

I leave my standby switches in "play"; they never move.

In deference to prevailing wisdom, though, I use the standby switch when cycling power on other people's amps; otherwise they get a bit pissy...

Eoin Madsen · 10-11-2008 10:38am

Interesting article, thanks for the post.

Thought this a little odd though:

David B. Lamkins wrote:

Arcing depends somewhat upon the circuit design. In general, though, standby switches are exposed to DC voltages well in excess of their design center. The reason it's so hard to make a mechanical high-voltage DC switch is because an arc is formed as the contacts open. The arc tends to continue so long as there's enough voltage to maintain it as the contacts separate.

He's perfectly correct about the reasons why DC rating for contact switches is always much lower, but "in general" standby switches are not exposed to DC voltages at all. In every amp I've worked on, the standby has been a DPDT before the rectifier. I would guess that the practice of putting the standby onto the DC, which does appear on a lot of old schematics, hasn't commonly occured since before electrical safety standards obliged electrical manufacturers to actually follow the ratings on electrical components (decades ago). Afaik, 500 VDC switches only exist for seriously heavy industrial applications.

Paolo_M wrote:

All of the information this guy provides is correct and can be cross referenced.

David B. Lamkins wrote:

Cathode stripping is real. (Sit down. I haven't misled you.) It happens at very high voltages and currents. If you go back to the manuals and engineering texts of the `50s and `60s you'll discover that no one ever wrote about cathode stripping with respect to "receiving tubes" (as compared to "transmitting tubes"). "Small" tubes like the 12A?7, 6L6, 6V6, EL84, EL34, EF86, ... are receiving tubes. With the exception of a few tube amps having very high (~700V) plate voltages (like the SVT and a few high-powered Marshalls), you don't really need a standby switch except as a convenience for muting the amp; I'd rather just pull the cord from the input jack or turn down the volume control on the guitar. There is no way that cathode stripping can occur in preamp tubes, even in an SVT; the available energy is limited to a very low value by resistors in the plate circuits.

This is pretty much the crux of the whole thing. Hit me with a citation. :pac:

felim · 10-11-2008 11:23am

Thanks Paolo, that was an interesting read. A lot of it went over my head (as does a lot of technical valve amp info) but I enjoyed it anyway.

I've always used a standby when it's there, around 30 seconds on power up but only a few seconds on power down. I've never had any problems with this use.

My current amp (Marshall 2554) does act a little weird when using the standby on power down though. I'll switch the standby to the "non-play" position (I can never get if this is stanby on or off, I would think on) for a few seconds and then switch off the power. When I switch the mains off, the volume will swell for a second and then die away. It's been doing this for as long as I've had it (around 7 months) and hasn't caused any problems but I'm almost certain it's not the way it "should" behave.

Paolo_M · 10-11-2008 12:13pm

Yeah, he kinda let himself down a bit when he talked about DC on the stand-by switches. I've only seen one 1959 SL design like this and seen schematics for only a hand full more. Mostley, as you say, it's installed between the transformer secondary and the recifier where only AC is present.

Unfortunatley I have to admit to being a bit of a geek and have read RCA Receiving Tube Manual RC30 a couple of times. At no point in the book does it refer to the problem of cathode stripping or the use of a stand-by switch to protect against it. Infact it doesn't show the use of a stand by in any of the design references at all, even for muting purposes, though the book isn't written specifically for instrument amplifiers.
The original Marshall 1987s suffer badly from the use of a stand-by as the bias circuit is wired after the switch so when it's switched the power valves run free until the bias voltage builds up. This fries a lot of PTs. My own JMP has a replacement PT for this very reason. It's also a problem for many valve rectifiers as the switch can cause a sudden current surge that's beyond their max current rating. Repeating switching will cause valve failure sooner than it would otherwise have occured.
As early designers used these books as their design basis, his claim that they were originally installed to provide a muting function while keeping the amp warmed up and ready to play is a plausible statement IMO.
I have Radiotron Designers Handbook 4th edition and I haven't come across it yet (admittidly I haven't finished it yet, it's over 1,000 pages!!) in it either.

My girlfriends dad works in the Irish Lights and he used to work as a tech on the radio transmitting gear. He's talked about it a fair bit, I plague him constantly for information. Obvioulsy this gear uses transmitting valves, huge milk bottles of things, and DC voltages over 1,000Vdc is normal in these circuits. Funnily enough they operate a lot cooler than guitar amps as they're running in Class C mostly.

So, in short, I can't provide a specific citation but I can say his claim that it is not a problem in receiving valves and it is in transmitting valves is true. RCA didn't recommend it and they would have been concious of reliability as sales depended on it. As a result I think it's fair to say that a stand-by does not perform any protection function whatever but is simply a convenient method of muting an amp.

Paolo_M · 11-11-2008 1:02pm

Felim, I've not seen one acting this way either. I can see no reason from the schematic why it would do this. The only this I can think of is the stored emf in the power transformer secondary suddenly arcing across the open stand-by switch as the mains is switched off. This might cause a brief swell as the filter caps charge and the discharge again, but I'm really just guessing.

Eoin, any idea's?

Eoin Madsen · 12-11-2008 3:25pm

I'd put a probe on the plate and see if the voltage swells during the power down. But then, I get a buzz off the thrill of knowing I could die if my hand slips, so any excuse really. :pac: I don't know really. I think the collapsing magnetic field should generate some kind of momentary surge, but I've never heard anyone say it had an audible or measurable effect.

Btw, felim, it's normal for the sound to distort like crazy for a few seconds as the plate voltage drops and the volume dies.

felim · 13-11-2008 11:39am

Eoin Madsen wrote: »

Btw, felim, it's normal for the sound to distort like crazy for a few seconds as the plate voltage drops and the volume dies.

Yeah, I've noticed this with all the valve amps I've used, but the issue I described above is different. Switch to standby...the sound fades and distorts like you said. The amp is totally silent, then turn off the mains switch and the volume swells up and then fades and distorts again (taking 2 or 3 seconds max.), like you said.

It hasn't caused any problems but then again, I've only had it around 7 months. I'm not that worried about it....but should I be?

Eoin Madsen · 13-11-2008 1:05pm

felim wrote: »

The amp is totally silent, then turn off the mains switch and the volume swells up and then fades and distorts again (taking 2 or 3 seconds max.), like you said.

When you say the volume swells up, do you mean the amplification factor of the input (guitar) sound increases, or that random noise appears and swells in volume?

felim · 14-11-2008 10:46am

Eoin Madsen wrote: »

When you say the volume swells up, do you mean the amplification factor of the input (guitar) sound increases, or that random noise appears and swells in volume?

It's the actual guitar signal that swells. You don't hear anything if there's no guitar plugged in.

Eoin Madsen · 14-11-2008 10:57am

Yeah, that would seem to suggest that the plate voltage is surging a little.

felim · 14-11-2008 12:21pm

Eoin Madsen wrote: »

Yeah, that would seem to suggest that the plate voltage is surging a little.

I can check that next time I have the amp opened up. Maybe over the weekend if I can get someone to throw the switches for me....

Stanby switches

Comments