Improved Standby Switching

[Structo]

I like reading what the Wizard has to say a lot of times and saw this about standby switching.
I supposedly will make your tubes and caps last longer, especially if you are using a tube rectifier.

Anybody doing this in their amp?

<snippet>
Improved standby switching:
Most standby switches simply turn the HT on and off, allowing the amp to be muted while the valves are still idling hot. When the standby is switched to "on" all the power supply capacitor will instantly draw inrush current;- this 'hard switch on' and inevitably shortens their working life. However, if we are willing to forgo the complete muting facility we can instead redesign the standby switch for a 'soft start' to truly help extend component life (especially the rectifier). If we arrange the switch to allow the anode current to increase very gradually at the same time as the heaters warm up, then both inrush surge and cathode poisoning can be completely avoided. This can be done simply by placing a resistor in parallel with the switch. Obviously this works best if the switch is placed before the reservoir capacitor.

Finding a suitable resistor is relatively simple because the amount of current we allow to flow on standby isn't that important, and it is in fact easier to decide instead how much power we want it to dissipate. Because the amp might be left on standby for some time we don't want the resistor to get very hot, and equally, we would like to avoid using a large and expensive metal clad resistor.
A 2W resistor would be ideal. To avoid excessive heat we will allow it to dissipate no more than 1W. At the moment the mains switch is turned on, the resistor will have to drop the full HT voltage. If the standby switch is to be placed after the rectifier we can use the value of HT voltage. If the switch is placed before the rectifier then use the AC (rms) voltage of the transformer.
In this example the switch is to be placed after the rectifier and the HT is 300Vdc. For 1W dissipation:
P = V^2 / R
(Note, ^2 means "squared".)
R = (V^2)/P
R = (300^2)/1
= 90 000 ohms
The nearest standard is 100k, but since it only has to suffer the maximum power dissipation at the moment of switch on, half this value or 47k would be fine too

[David Root]

This is fairly common in higher voltage commercial designs, I've used it myself on occasion.

[greiswig]

I must be missing something. If you have resistance in series, wouldn't you either
a) have low enough R that the tubes are live, and not silent, or
b) have high enough R that the tubes are basically dead, but there would be inrush current when you shorted across the resistor

Naively yours,

[mlp-mx6]

I must be missing something. If you have resistance in series, wouldn't you...
a) have low enough R that the tubes are live, and not silent...

Note this sentence fragment from the original post:
"However, if we are willing to forgo the complete muting facility we can instead redesign the standby switch for a 'soft start'..."

Also note that the resistance will be high enough to slow the inrush.

[greiswig]

I must be missing something. If you have resistance in series, wouldn't you...
a) have low enough R that the tubes are live, and not silent...

Note this sentence fragment from the original post:
"However, if we are willing to forgo the complete muting facility we can instead redesign the standby switch for a 'soft start'..."

Also note that the resistance will be high enough to slow the inrush.

DOH! Missed that. Okay, so how much muting are we talking about? Any?

[Structo]

I must be missing something. If you have resistance in series, wouldn't you either
a) have low enough R that the tubes are live, and not silent, or
b) have high enough R that the tubes are basically dead, but there would be inrush current when you shorted across the resistor

Naively yours,

Merlin says that anymore the standby switch is not as important as it once was. It is a left over from the days of the big tube transmitters and receivers.

I think what he is saying that when the standby switch is turned on, the inrush current to the tubes and large filter caps is detrimental to them.
When the standby switch is flipped the caps appear as a short circuit for a few cycles until they charge up.

With the 2 watt resistor soldered across the standby switch it allows the caps to slowly charge while you wait for the heaters to warm up.

Then when you flip the standby switch to play there isn't a huge current put across it which will not only make the switch last longer but also the tubes and especially the filter caps.

Here is the rest of the article that may explain it better than I can.

The initial current will mainly be swallowed by the reservoir capacitor so most of the HT will be dropped across our standby resistor, and the voltage on the valves' anodes will be negligible for the first few seconds while the heaters begin to warm. As the reservoir and smoothing capacitors charge the anode voltages will rise gradually and a small amount of anode current will be allowed to flow as the heaters reach full temperature. Once the capacitors are fully charged and the valves warmed up the amp will remain on standby, allowing a trickle current to flow at all times (so the amp will not be totally muted; there may be the faintest of strangled sounds if we tried to play). The current flowing will be small (and very difficult to calculate!) but sufficed to say that it will probably be about half the initial surge, and the anode voltages will all idle at somewhere in the region of half their normal value. The power now being dissipated in the standby resistor will be about 1 watt.

This is an enormously simple modification to any amp, and any value over about 47k (2W) or 150k (1W) should do. A secondary advantage of this method is that because the filter capacitors are allowed to partially charge up during standby there will be little delay between switching the standby switch to "on" and sound coming out of the amp!

[img:320:119]http://www.freewebs.com/valvewizard1/standby5.jpg[/img]

I think that this is a worth while mod on a D'Lite since it has the standby switch before any of the filter caps. So when you flip the standby the caps get slammed as well as the switch.

For only the cost of a 2 watt resistor if this can help I'm all for it.

Here is the link to the page.

[greiswig]

Tom,

It sure seems worth a try, particularly if the amp is still pretty muted. I can think of one way in which a standby switch is useful, at least to me. This is based on a real-world example:

I'm playing for our church, and we do one set then get off the stage to make room for the sermon. I just turn the amp down, not on standby, because I recently read that standby is passe.

Mid-sermon, one of my tubes decides it's had enough. Loud noises ensue, and I'm running up on stage to shut things down. If I'd had it in standby, at least the amp would have waited until the next set to start making noise, when I could have dealt with it a bit more gracefully.

[Structo]

Yeah, tube amps can be temperamental for sure.

It's kind of amusing in a way how these myths keep going on for years and years.

Merlin isn't the first amp guy that has published that the standby is not needed on today's amps.

It does seem to make sense that you would want the tubes to be heated before hitting them full force with the B+ voltage.

I have read that your typical standby switch ( Carling seems to be used a lot) does not like to switch DC voltage.
I have seen some schematics that have the standby switch on the HT center tap so that would be AC on the switch, right?

I'm probably over thinking the whole thing but it's fun at the same time.

[paulster]

I have read that your typical standby switch ( Carling seems to be used a lot) does not like to switch DC voltage.

If you switch DC then you'll get an arc as the contacts make or break (often heard as a switch-off pop). Over time this can gradually cause burn spots on the switch contacts which leads to resistance and therefore heat build-up in the contacts.

If you switch AC then it will still attempt to arc, but it is effectively self-quenching because the arc can only last for one half cycle at most because at the zero-crossing point there is no voltage to sustain it.

That's why switches will usually have an AC voltage rating far in excess of their DC voltage rating.

[Rick]

I use Merlin's website all the time and even bought his recent book on Preamp design "Designing Tube Preamps for Guitar and Bass". It's well worth the money. He has a lot of great stuff on his site. I like the idea of using a DPDT switch on the AC feeds to the rectifier. You are switching lower voltage (easier on the switch) and it's AC for the previously mentioned less arcing characteristic. I still encountered quite a pop with it so I bypassed both sides of the switch with 630V .1 uF caps and that stopped the popping about 95% of the time. It will still pop on the odd occasion, but it is far better and usually completely quiet. BTW, I always liked the idea of a standby switch merely because you can keep the bottles hot and ready without any rush noise or having to wait to kick thing in or mess with volumes. It's nice to have the amp ready to rock instantly without warming and cooling cycles. Not a big deal, just a convenience that is easy enough to implement.

[j-po]

Maybe a bit off-topic but anyway... What do you guys think, how large a voltage spike in the secondary does switching the mains cause? Isn't the transformer a bit like a car's ignition coil i.e. the fast transient in voltage in the primary causes a voltage spike in the secondary.

[Structo]

I'm not sure but I think you are right that when the power switch is thrown, there has to be a spike in the secondaries.
After all, the first radio transmitters were simply spark makers.

[j-po]

Maybe one could mitigate the spike by having a HV film cap right after the rectifier to provide a place for the coil to quickly discharge its energy.

[ampdoc1]

Here is a very good read on standby switching.

[Structo]

Thanks, that is a good read.