Standby bypass

[pompeiisneaks]

Also the easiest way to slow down the inrush is to just put balance resistors on the caps to slow the inrush a bit.

Really? My gut says nope. Happy to be proved wrong, tho.

I was under the impression the resistors to ground serve multiple functions:

1. if you have caps in series they help ensure they load up equally and controlled
2. it helps discharge the caps to ground on power off
3. if it's a single cap it helps slow inrush (just like on 1, but doesn't 'balance as well' since there's only one).

Am I mistaken?

~Phil

[sluckey]

You need series resistance (not parallel) to reduce inrush charging current.

[pompeiisneaks]

I am wrong, did some reading to figure out where I got that idea, and I can't find it.

Sorry, I'm confusing using a SERIES resistor for that, which you don't want in that type of setup. But basically the inrush current hasn't been a problem for tons of amps for a long time, unless you do a bad design

~Phil

[tictac]

Read Merlin's remarks about standby switches and / or R.G. Keen's if you want to put this to rest.... yes there are lots of myths out there...

TT

[dorrisant]

Phil, I don't mean to pile on but the data sheets for a GZ34 rectifier, (used in a Rocket) spec 60µF max for the first filter... Looking at the schematic for the Rocket, Ken used two 40µF caps in parallel for a total of 80µF as a reservoir. Try powering up without giving the rectifier a chance to get warmed up and fill the reservoir before supplying full load and you will see sparks after a short while. Not saying right away but I would bet money you will go through rectifier tubes much more often. There seems to be a rush to collect all of the well made 5AR4/GZ34 tubes so in order to preserve what you have, I would strongly suggest using a standby switch with this particular circuit. There are others where this seems important too, this is the most relevant that comes to mind.

I think the biggest myth about standby switches revolves around the cathode stripping/poisoning theory. This theory is absolutely true but only applies to transmission tubes, so we have little concern since most of the tubes we use in guitar amps are either receiving tubes or rectifiers.

[pompeiisneaks]

Phil, I don't mean to pile on but the data sheets for a GZ34 rectifier, (used in a Rocket) spec 60µF max for the first filter... Looking at the schematic for the Rocket, Ken used two 40µF caps in parallel for a total of 80µF as a reservoir. Try powering up without giving the rectifier a chance to get warmed up and fill the reservoir before supplying full load and you will see sparks after a short while. Not saying right away but I would bet money you will go through rectifier tubes much more often. There seems to be a rush to collect all of the well made 5AR4/GZ34 tubes so in order to preserve what you have, I would strongly suggest using a standby switch with this particular circuit. There are others where this seems important too, this is the most relevant that comes to mind.

I think the biggest myth about standby switches revolves around the cathode stripping/poisoning theory. This theory is absolutely true but only applies to transmission tubes, so we have little concern since most of the tubes we use in guitar amps are either receiving tubes or rectifiers.

That's because the capacitance is over spec for the tube. If you run tubes that have a max voltage rating of 300V at 500V they'll die sooner too. It may be why he uses a varistor or MOV not sure which, at the input of the AC side, to slow the roll at the first. Sounds to me like the rocket is designed to kill rectifiers. I don't thing having tubes that conduct will stop that inrush much from slamming the tube rectifier, though. You also can't count on guitar players to always remember to turn off the standby, so it's asking for trouble/dead rectifiers.



~Phil

[dorrisant]

Regardless of the rectifier being larger than design maximum, that is the way it was originally designed and it is the way that they are built. Not many people consider the data sheets when building a clone of a known good circuit. If you do not use a standby switch, you may be confused as to why this circuit seems hard on rectifier tubes. Knowing that the rectifier is at a higher risk would make you consider how you use the standby switch... Meaning, you may not want to omit it. Maybe you would redesign the filter setup or rectifier type to avoid using one... but if you do that, you start changing the character of the amp... Point being, the old standby switch can be handy in the right circuit. Maybe we shouldn't use them sometimes... when it really won't help anything. There are occasions where they are helpful. Thinking that you should never use them in any of your builds as a blanket statement seems to be burying one's head in the sand. I'm not saying you have said this, I have read statements from some who do though.

Also, I don't think I've ever seen an MOV used in the Rockets... Expresses and Liverpools yes, Rockets no. Go figure.

[pompeiisneaks]

Regardless of the rectifier being larger than design maximum, that is the way it was originally designed and it is the way that they are built. Not many people consider the data sheets when building a clone of a known good circuit. If you do not use a standby switch, you may be confused as to why this circuit seems hard on rectifier tubes. Knowing that the rectifier is at a higher risk would make you consider how you use the standby switch... Meaning, you may not want to omit it. Maybe you would redesign the filter setup or rectifier type to avoid using one... but if you do that, you start changing the character of the amp... Point being, the old standby switch can be handy in the right circuit. Maybe we shouldn't use them sometimes... when it really won't help anything. There are occasions where they are helpful. Thinking that you should never use them in any of your builds as a blanket statement seems to be burying one's head in the sand. I'm not saying you have said this, I have read statements from some who do though.

Also, I don't think I've ever seen an MOV used in the Rockets... Expresses and Liverpools yes, Rockets no. Go figure.

Standby switches are known to cause harm. They send a huge inrush into the circuit itself when opened up because the tubes are all ready to conduct and warm, and all other capacitors have to take that inrush as well. It's a debate for the ages. You can very easily remove the tube rectifier and still produce the sag with a resistor inline. Or so I've read many times, but I've never tested, to that could be a bit 'off.' But I also think rectifier sag would be almost completely eradicated with 80uF of filtering in that first stage no? or 'maybe' some sag at max volume slamming that poor guitar like a red headed step child . Also, since I've not built one, I definitely don't know.

~Phil

[pdf64]

Phil, I don't mean to pile on but the data sheets for a GZ34 rectifier, (used in a Rocket) spec 60µF max for the first filter...

Bear in mind that when tube info sheets were written, standard ecap value tolerance was as much as +100% to -50%.
The info notes 'input [ie reservoir] capacitor' not 'capacitance', so I suspect that the 60uF limit includes a bit of 'give' to account for the reservoir cap's actual value possibly being a bit above nominal.

Note that a very significant parameter mentioned in the info is the series / protecting resistance per plate, and for a given reservoir cap value, as the voltage increases (and hence the duration of time required to top up charge on the reservoir per half cycle), the value of the protecting resistance increases
http://www.mif.pg.gda.pl/homepages/fran ... g/GZ34.pdf
http://www.mif.pg.gda.pl/homepages/fran ... 5/5AR4.pdf
The protecting resistance is normally provided by the PT HT winding resistance, which means that the winding will necessarily have 'average to poor' regulation, and will get warm in operation.

I think an important consideration to take from that is if a 'cool running' PT with good regulation is used and no additional protecting resistors are added, then a smaller value of reservoir cap may be required to avoid breaching plate current limit.

...Ken used two 40µF caps in parallel for a total of 80µF as a reservoir. Try powering up without giving the rectifier a chance to get warmed up and fill the reservoir before supplying full load and you will see sparks after a short while. Not saying right away but I would bet money you will go through rectifier tubes much more often...

Rocket schematic http://ampgarage.com/forum/download/file.php?id=8003

Bear in mind that by design, the GZ34 has a slow warm up and its cathode voltage has a lovely smooth ramp up (my finding is that full HT is achieved about 45s after power up). That will naturally limit reservoir charging current, and by extension, the current that the power tubes can draw if powered up without standby, ie the power tubes can only draw 200mA at full HT.
At any lower point of the HT ramp up, the power tube current draw will be commensurately lower.
So I don't see how the use of standby takes any useful degree of stress off the GZ34 here? Rather the opposite, as the manufacturer's info emphasises the importance of limiting peak plate current.

To me, the important constraint with the Rocket and its 80uF reservoir, is that the PT has the intended degree of protecting resistance; I think that the standby is purely a user convenience.

[martin manning]

The reservoir cap is typically before the standby (as it is in the linked Rocket schematic), so in that case the reservoir will charge slowly as the rectifier comes up to temperature with the standby on or off. The danger is in switching power off and then back on after the caps bleed down and the rectifier is still hot.

[dorrisant]

Thanks guys... I obviously hadn't thought of it that way... Off to my cave to contemplate and rearrange some synapses.

Seriously, I'm glad to learn this.

Sorry Phil, I didn't mean to step on your toes... Maybe I could buy you a drink?

[pdf64]

...The danger is in switching power off and then back on after the caps bleed down and the rectifier is still hot.

Thanks for pointing that out, I'd not previously considered that scenario.
Yes, 'hot switching', switching a capacitor on to a hot tube rectifier is the most stressful, worst case scenario for a tube rectifier, as it results (if only momentarily) in very high plate current.

[Stevem]

I guess the discussion over the need or not for a standby will go on forever, but one of my main reasons for having one is to cut down on thermal related mechanical failures of the output tubes.
Putting a amp in standby to allow the output tubes to cool back down closer to just there idle temp seems to me far less mechanically stressful then just turning the amp off!

I would love to see test results in terms of how long it takes for a output tube to develop a grid to heater short when they are cooled down slow as compared to running them near flat out and then just killing all power to the amp.
I would bet that shorts would take place at a rate of 50% greater!

Any of us who have played tube amps for a long time and worked on them for a long time have seen many cases where we take them off standby and the fuse blows right away due to a failed output tube.
In this type of failure the tube has already shorted out when the amp was turned off. The last time it was shut down , not when we just turned it on, and to me this a mechanical failure from the cycle extreme of hot and cold which a standby switch can help to ease!
Now let me take this one step more and say that if you kind of agree with me on what I posted here about the turn off cycle, then why no cut down on stress in the turn on cycle by allowing atleast the heaters to come up fully before the V+ voltage is applyed?

[pompeiisneaks]

Thanks guys... I obviously hadn't thought of it that way... Off to my cave to contemplate and rearrange some synapses.

Seriously, I'm glad to learn this.

Sorry Phil, I didn't mean to step on your toes... Maybe I could buy you a drink?

No worries, I'm wrong often, I learn all the time and don't mind finding things I didn't know. It's just one of the areas I wish would go away' the whole 'standby is needed' thing

~Phil

[pompeiisneaks]

I guess the discussion over the need or not for a standby will go on forever, but one of my main reasons for having one is to cut down on thermal related mechanical failures of the output tubes.
Putting a amp in standby to allow the output tubes to cool back down closer to just there idle temp seems to me far less mechanically stressful then just turning the amp off!

Not sure I know what that means? the tubes life is shortened by cooling too fast? and keeping the heaters going helps them cool slower? For how long? this is something I've never even heard of and makes me very curious.

I would love to see test results in terms of how long it takes for a output tube to develop a grid to heater short when they are cooled down slow as compared to running them near flat out and then just killing all power to the amp.
I would bet that shorts would take place at a rate of 50% greater!

Any of us who have played tube amps for a long time and worked on them for a long time have seen many cases where we take them off standby and the fuse blows right away due to a failed output tube.
In this type of failure the tube has already shorted out when the amp was turned off. The last time it was shut down , not when we just turned it on, and to me this a mechanical failure from the cycle extreme of hot and cold which a standby switch can help to ease!
Now let me take this one step more and say that if you kind of agree with me on what I posted here about the turn off cycle, then why no cut down on stress in the turn on cycle by allowing atleast the heaters to come up fully before the V+ voltage is applyed?

I'd heard that the life of the tube is increased by letting the heaters come up slowly and letting the tubes start conducting slowly, the 'heater warm current slam as they conduct instantly' was what I have read is a big tube life shorten-er, compared to a slow increase in current until they reach operating temperature. maybe that's a myth?