[Xtrchessreal]

While researching ignitors on gas arc tube lamps I read about how the magnetic field collapses when the circuit is basically opened as the AC input signal falls to zero and it causes a spike in the secondary as high as 2500-3000 Volt range for 150w to 400w lamps. I know the spike in a ballast is designed for the lamp to start. I should say the lamp is designed to take advantage of the spike.

SO:

A spike must happen to a tube push pull amp as well though perhaps not that high in voltage, I don't know.

I got to thinkin'. The Power Transformer, if opened while standby is closed, could do some real damage to the amp, right? If a spike like that is released across the rectifier and filter caps, the inductor, and the power tubes...I don't have testing experience in this so here is the question.

Q1 What happens to the amp as this spike works its way through a basic amp; e.g. http://drtube.com/schematics/marshall/2204u.gif

Part1: Assume a proper speaker is connected.
Part2: Assume without a speaker connected.

Slightly different note:
I've noticed that when I open the standby preparing the amp to be shut down after playing for a while that I can hear my guitar chord at lower volume and fades within a few moments through the speakers. This is a different discharging scenario since the standby is open no spike from the PT can pass..

Q2 What is contributing to the discharging of the amp, providing enough energy to the amp for the signal to still be heard for a few seconds? The heaters are still hot so is the amp dissipating the coils and caps stored energy only? Or is there stored energy in the hot tubes as well? Provided there is an input signal.

X

[Peter Wieck]

Please note the interpolations:

On Monday, June 26, 2017 at 5:04:33 AM UTC-4, Xtrchessreal wrote:
While researching ignitors on gas arc tube lamps I read about how the magnetic field collapses when the circuit is basically opened as the AC input signal falls to zero and it causes a spike in the secondary as high as 2500-3000 Volt range for 150w to 400w lamps. I know the spike in a ballast is designed for the lamp to start. I should say the lamp is designed to take advantage of the spike.

SO:

A spike must happen to a tube push pull amp as well though perhaps not that high in voltage, I don't know.

This is the fallacy of 'leaping to conclusions' - reasoning from the specific to the general. My cat is grey, therefore all cats are grey. The ballast in a gas-discharge lamp is wound and designed to deliver that spike at the collapse of the field. Note that a lamp ballast is essentially a massive choke with no secondary winding(s).

A power-transformer with multiple secondary windings, the 'hottest' of which is perhaps 4:1 (but relatively low current) is not wound to deliver such a spike. What *does* happen is that the current through the filaments is a bit high until the filament heats and resistance increases. But this is hardly what would be defined as a damaging spike.

This does not include the potential of an external spike. But, nothing generated within the 'system' will do as you suggest. Once that is understood, your first question is moot.

I got to thinkin'. The Power Transformer, if opened while standby is closed, could do some real damage to the amp, right? If a spike like that is released across the rectifier and filter caps, the inductor, and the power tubes...I don't have testing experience in this so here is the question.

Q1 What happens to the amp as this spike works its way through a basic amp; e.g. http://drtube.com/schematics/marshall/2204u.gif

Part1: Assume a proper speaker is connected.
Part2: Assume without a speaker connected.

Slightly different note:
I've noticed that when I open the standby preparing the amp to be shut down after playing for a while that I can hear my guitar chord at lower volume and fades within a few moments through the speakers. This is a different discharging scenario since the standby is open no spike from the PT can pass.

Q2 What is contributing to the discharging of the amp, providing enough energy to the amp for the signal to still be heard for a few seconds? The heaters are still hot so is the amp dissipating the coils and caps stored energy only? Or is there stored energy in the hot tubes as well? Provided there is an input signal.

The capacitors within the amp discharging through the residual conduction in of the tubes. When the filament shuts off, there is still some minimal conduction from the residual heat. A few seconds - perhaps a bit more depending on the design of the amp itself and the amount of capacitance in use.

Peter Wieck
Melrose Park, PA

[Xtrchessreal]

On Monday, June 26, 2017 at 5:12:34 AM UTC-6, Peter Wieck wrote:
Please note the interpolations:

On Monday, June 26, 2017 at 5:04:33 AM UTC-4, Xtrchessreal wrote:
While researching ignitors on gas arc tube lamps I read about how the magnetic field collapses when the circuit is basically opened as the AC input signal falls to zero and it causes a spike in the secondary as high as 2500-3000 Volt range for 150w to 400w lamps. I know the spike in a ballast is designed for the lamp to start. I should say the lamp is designed to take advantage of the spike.

SO:

A spike must happen to a tube push pull amp as well though perhaps not that high in voltage, I don't know.

This is the fallacy of 'leaping to conclusions' - reasoning from the specific to the general. My cat is grey, therefore all cats are grey. The ballast in a gas-discharge lamp is wound and designed to deliver that spike at the collapse of the field. Note that a lamp ballast is essentially a massive choke with no secondary winding(s).

A power-transformer with multiple secondary windings, the 'hottest' of which is perhaps 4:1 (but relatively low current) is not wound to deliver such a spike. What *does* happen is that the current through the filaments is a bit high until the filament heats and resistance increases. But this is hardly what would be defined as a damaging spike.

This does not include the potential of an external spike. But, nothing generated within the 'system' will do as you suggest. Once that is understood, your first question is moot.

I got to thinkin'. The Power Transformer, if opened while standby is closed, could do some real damage to the amp, right? If a spike like that is released across the rectifier and filter caps, the inductor, and the power tubes...I don't have testing experience in this so here is the question.

Q1 What happens to the amp as this spike works its way through a basic amp; e.g. http://drtube.com/schematics/marshall/2204u.gif

Part1: Assume a proper speaker is connected.
Part2: Assume without a speaker connected.

Slightly different note:
I've noticed that when I open the standby preparing the amp to be shut down after playing for a while that I can hear my guitar chord at lower volume and fades within a few moments through the speakers. This is a different discharging scenario since the standby is open no spike from the PT can pass.

Q2 What is contributing to the discharging of the amp, providing enough energy to the amp for the signal to still be heard for a few seconds? The heaters are still hot so is the amp dissipating the coils and caps stored energy only? Or is there stored energy in the hot tubes as well? Provided there is an input signal.

The capacitors within the amp discharging through the residual conduction in of the tubes. When the filament shuts off, there is still some minimal conduction from the residual heat. A few seconds - perhaps a bit more depending on the design of the amp itself and the amount of capacitance in use.

Peter Wieck
Melrose Park, PA

I agree with you. I went back to study the circuit of the Constant Wattage Autotransformer Ballast I was working with: http://cr4.globalspec.com/PostImages...797CED15AB.bmp

The Cap across the primary and secondary only charges/discharges at higher open circuit voltage and the starter closes during arc lamp ignition, once at operating voltage the cap stops charging and discharging because the operating voltage is lower and the starter stays open. At that point the Ballast is just an Inductor. The spike I was thinking about is only on the secondary at arc lamp startup. After operation for several hours and then switched off the starter is still at a lower voltage, the arc tube is still hot, the Cap is still not charged and stays open, no spike should occur at that point either.

The PT on an amp would not deliver any different peak voltage than normal.

I was confused because of the diagram, looks similar to a PT but the Cap in series with secondary is very different.

Thanks for your input

X

[Peter Wieck]

Not to worry. I have chased down more than a few rabbit-holes in my time!

It just makes SO much sense that it should be this way, why isn't it???

Peter Wieck
Melrose Park, PA

[Big Bad Bob]

On 06/26/17 02:04, Xtrchessreal wrote:
While researching ignitors on gas arc tube lamps I read about how the magnetic field collapses when the circuit is basically opened as the AC input signal falls to zero and it causes a spike in the secondary as high as 2500-3000 Volt range for 150w to 400w lamps. I know the spike in a ballast is designed for the lamp to start. I should say the lamp is designed to take advantage of the spike.

SO:

A spike must happen to a tube push pull amp as well though perhaps not that high in voltage, I don't know.

the best way to describe this is:

if you create a 5th harmonic while opening or closing the circuit to a
transformer, the current "rotates backwards" with respect to phase
rotation. Additionally, if you were to close (or open) a transformer
primary circuit at the peak voltage, you would theoretically have zero
(real) current if the load were purely inductive. And, if you opened or
closed the circuit at zero crossing, you would have maximum (reactive)
current.

This being due to reactive current 90 degrees lagging voltage.

I saw people discuss 'inrush current' once before, who then mercilously
FLAMED me, just for suggesting that experiments be done to show what the
peak amperage REALLY is (claims of 1000 amps, and so forth, pure
bull****). And that name 'inrush current' offers a poor description of
what is REALLY happening.

If you close the switch at "just the right point", or open the switch at
"just the right point", you might get unlucky and max out the current.
Zero crossing is considered one of the worst points, but 5th harmonic
(36 degrees) would be _THE_ worst.

I did some real calculations (which I don't think I ever posted, because
I was sick of the ASSHOLES that only wanted to make their own dicks look
bigger than mine) and determined that the biggest surge current would be
due to creating a 5th harmonic waveform. So if you were to close (or
open) the switch at the magic 36 degree point in the sine wave, assuming
no hysteresis on the core, and certain conditions for toroidal
transformers, you'd max out current, at an even higher level (as a
spike, not continuous) than you might expect based on the resistance of
the windings and impedence of the load, etc..

This is because phase rotation of the 5th harmonic OPPOSES the phase
rotation of the 1st harmonic. It's worse than a direct short, as you're
literally OPPOSING the current flow. This is known to burn up motors
and wiring etc. in power circuits with sufficient 5th harmonic in the
power circuit. 5th harmonic is bad. VERY bad.

https://www.allaboutcircuits.com/tex...ase-sequences/

Now the term 'inrush current' is (unfortunately) in use in the
electrical power generation industry. It's misleading as to what it
actually means. There's no "inrush" really. They just call it that.

https://en.wikipedia.org/wiki/Inrush_current -- misleading name

The wikipedia article also explains some of the issues with toroidal
transformers, which are moderately interesting - I'd just use a series
slow-blow fuse of an appropriate size, which would have a small series
resistance in the circuit, and would mitigate a lot of the problem.

Also, any 'spike' is basically in the primary of the transformer and (as
I understand it) does NOT induce problems in the secondary circuit(s).


In any case, energizing or de-energizing a transformer can cause issues
with the primary windings and the mains power you plug into. but the
secondary(s) shouldn't see any effects. Not in any significance, anyway.

[[email protected]]

On Monday, June 26, 2017 at 5:04:33 AM UTC-4, Xtrchessreal wrote:
While researching ignitors on gas arc tube lamps I read about how the magnetic field collapses when the circuit is basically opened as the AC input signal falls to zero and it causes a spike in the secondary as high as 2500-3000 Volt range for 150w to 400w lamps. I know the spike in a ballast is designed for the lamp to start. I should say the lamp is designed to take advantage of the spike.

SO:

A spike must happen to a tube push pull amp as well though perhaps not that high in voltage, I don't know.

I got to thinkin'. The Power Transformer, if opened while standby is closed, could do some real damage to the amp, right? If a spike like that is released across the rectifier and filter caps, the inductor, and the power tubes...I don't have testing experience in this so here is the question.

Q1 What happens to the amp as this spike works its way through a basic amp; e.g. http://drtube.com/schematics/marshall/2204u.gif

Part1: Assume a proper speaker is connected.
Part2: Assume without a speaker connected.

Slightly different note:
I've noticed that when I open the standby preparing the amp to be shut down after playing for a while that I can hear my guitar chord at lower volume and fades within a few moments through the speakers. This is a different discharging scenario since the standby is open no spike from the PT can pass.

Q2 What is contributing to the discharging of the amp, providing enough energy to the amp for the signal to still be heard for a few seconds? The heaters are still hot so is the amp dissipating the coils and caps stored energy only? Or is there stored energy in the hot tubes as well? Provided there is an input signal.

X

Not to worry, do the following. Connect a 1000 PIV SI diode such as a 1N4007 so that its anode side is connected to the common lead (ground), the cathode lead to the load side of the B+. Any energy stored in the LCL filter network will be safely dissipated when the standby is set to OFF. Wish I could show you on a schematic. What make & Model is the amp? It might be in Aspens book.

Cheers to all, John L Stewart

[[email protected]]

In common power supplies the only example where there might be a problem would be that using choke input. The choke likes to keep the current flowing, so a switch between the rectifiers & the choke input could cause insulation failure. But a switch on the primary side as it normally would be is OK, the choke still has a way to dump stored energy thru the PS transformer wdgs & rectifer.

But choke input is now uncommon, most PS are cap input.

In my case I installed the fix I spoke of in a regulated PS with DC standby I built many years ago so that it can be run either in choke or cap input. The max volts or current can then be used to advantage.

Cheers to all, John L Stewart

[Xtrchessreal]

On Friday, August 4, 2017 at 11:37:24 AM UTC-6, wrote:
In common power supplies the only example where there might be a problem would be that using choke input. The choke likes to keep the current flowing, so a switch between the rectifiers & the choke input could cause insulation failure. But a switch on the primary side as it normally would be is OK, the choke still has a way to dump stored energy thru the PS transformer wdgs & rectifer.

But choke input is now uncommon, most PS are cap input.

In my case I installed the fix I spoke of in a regulated PS with DC standby I built many years ago so that it can be run either in choke or cap input. The max volts or current can then be used to advantage.

Cheers to all, John L Stewart

its a 2204 Marshall circuit that I built, I added an LED between the input filter and the standby, when switch closes and B+ hits the amp rail LED ON, such that when switch is later opened the Caps discharge through the LED to ground over a period of several minutes, for safety when opening the amp e.g. to bias new tubes or repairs, tweaks...

X

[Big Bad Bob]

On 09/23/17 00:05, Xtrchessreal wrote:
On Friday, August 4, 2017 at 11:37:24 AM UTC-6, wrote:
In common power supplies the only example where there might be a problem would be that using choke input. The choke likes to keep the current flowing, so a switch between the rectifiers & the choke input could cause insulation failure. But a switch on the primary side as it normally would be is OK, the choke still has a way to dump stored energy thru the PS transformer wdgs & rectifer.

But choke input is now uncommon, most PS are cap input.

In my case I installed the fix I spoke of in a regulated PS with DC standby I built many years ago so that it can be run either in choke or cap input. The max volts or current can then be used to advantage.

Cheers to all, John L Stewart

its a 2204 Marshall circuit that I built, I added an LED between the input filter and the standby, when switch closes and B+ hits the amp rail LED ON, such that when switch is later opened the Caps discharge through the LED to ground over a period of several minutes, for safety when opening the amp e.g. to bias new tubes or repairs, tweaks...

X


LEDs typically use up to 10ma, which is actually a LOT for a 400V power
supply (that'd need a 5W resistor in series with the LED).

A more interesting modification would do the following in standby:

a) short out the speaker leads on the output transformer (or use a very
very very low resistance) to prevent unusual oscillations, etc.
b) power the system with 50-100V and leave the fixed bias ON for the
output tubes

but if your system doesn't have fixed bias, then it probably won't work
right...

The reason why you want to apply at least SOME voltage in standby mode
is to prevent cathode damage. It's my understanding that tubes without
positive plate voltage have a tendency to form bizarre 'charge clouds'
around the cathode, damaging the coating among other things. I read
about things like this in an article on extending tube life, from the
1960's. I don't have a link handy, though...

So, in any case, you might consider at least doing 'b' but maybe
activate a relay to cause 'a' so you don't go into motorboating type of
oscillations in the power stage [with a big fat series resistor dropping
the voltage down like that]. Shorting the O.P.T. is generally bad, but
doing so while voltage is between 50V and 100V won't "cook" anything,
and would prevent oscillations. [merely opening the speaker leads won't
prevent oscillations - in fact it might make them WORSE]

You could also use a relay to make sure that the short is removed BEFORE
normal voltage is restored, a "break before make" condition using a
relay with multiple SPDT switches in it.