what happens to tube amp if powered on without cabinet. how and why

[Enzo]

It isn't oscillation as in feedback. With an open secondary, the transformer becomes a step up coil and can generate spikes in excess of 1000v and arc over the transformer windings. A shorted output is sufficient to save the transformer. Adding more complexity doesn't protect the amp any better.


Arced over carbonized jacks is usually caused by pulling the plug or pushing it in while the system is operating. They spark. I don't like the hypothesis, of "after it has protected the output stage a few times." Arcs and sparks only flow across gaps, and a contact shorting across is just a wire to the circuit. If someone fires up their amp without a load so often that the shorting contact has worn out, I think there are greater problems than the jack.

[Tom Phillips]

I like the method of putting a permanent (relatively high value) resistor across the speaker output because it protects against open speaker cords, open speakers and things like forgetting to plug in the other end of the cord when using piggyback amps.

[Gingertube]

As a variation on Toms method, I prefer to put "voltage tether" resistors (say 330K to 470K) from each OT anode lead to 0V. This is a variation of using reverse biased diodes in this circuit possition (in my opinion a superior variation).
Many ways to "skin the cat".
Cheers,
Ian

[bob p]

I don't like the hypothesis...If someone fires up their amp without a load so often that the shorting contact has worn out, I think there are greater problems than the jack.

I don't like it either, but people do all sorts of stupid things with their amps. the problem with arcing and carbonization of the shorting jack is that the carbon buildup eventually disables the shorting jack's ability to protect the amp. The whole idea behind the safety resistor is to eliminate the arcing that takes place when a dumb user abuses the amp when only a shorting jack is present. Greater problems or not, I think that the shorting jack is a poor solution to the problem because it doesn't last very long when it actually gets used.

Hey Tom -- how high a value?

[Enzo]

WHAT arcing? The only time that thing matters is when the jack is empty, it won't spark then.

[Steve Conner]

Pentode and beam tetrode tubes look like constant current sinks. They hardly deliver any more current into a dead short than they would into their rated load.

So, shorting the secondary is perfectly harmless to the OT. The currents flowing in it hardly change. However, it increases the plate dissipation in the power tubes, because the plate voltage no longer sinks when plate current is flowing. All the power that would go to the speaker is now burnt up in the plates, so they are liable to red-plate.

On the other hand, if a pentode or beam tetrode output stage is cranked up into an open circuit, the screen grids will overheat drastically, maybe to the point of melting and shorting a tube. The 270 ohm resistor, flyback diodes etc., might save the OT from voltage spikes, but they won't save the screens.

I think overall the shorting jack would be a reasonable protection method, if it weren't so easily defeated by forgetting to plug in the other end of the speaker cord.

[soundmasterg]

This is only a problem when someone plugs in an amp and turns it on with no load while playing a signal through it right? Without a instrument signal, all you get is a noise signal going through the amp and that is likely not going to be high enough to cause a problem in the time it takes for you to realize you forgot to plug in a load....

Greg

[Steve Conner]

Every tube amp I ever tried has been happy with no load and no input signal, or indeed any input signal below clipping.

However, some tube amps might oscillate with no load, in which case they would drive themselves into clipping.

The classic case is that the guitarist turns on his amp, cranks it up and whacks a big power chord, then wonders why he's not getting any sound, just smoke.

[bob p]

> WHAT arcing?

Good question. I recently acquired a Super Twin Reverb whose shorting jack is totally carbonized / arc'd over. This is a combo-type amp, like a twin reverb, but with a sextet of 6L6GC. It's hard to imagine that anyone would have been plugging-unplugging that little flat speaker plug while the amp was running, as this is a combo-type amp, not a piggybacked head. Nonetheless, the shorting jack is wasted. Resistance measures the same across it, regardless of whether the jack is inserted or pulled.

I think this is testament that the shorting jack is good in theory, but in practice it sucks.

[Enzo]

I am willing to be shown wrong, but that sounds like bar dirt. If you take the wires off the jack, and the tip shunt measures open, then I believe you that the jack is bad. But measured in place, the almost zero ohms of the OT secondary will look pretty much like a dead short.

[bob p]

I was smart enough to remove the OT leads.

[g1]

Perhaps plugging the speaker into the "ext. spkr" jack and having nothing plugged into the proper "spkr" jack could cause this?
The switch should be a dead short, but if there is a little bit of dirt/resistance?

[bob p]

good idea. as it turns out, the super twin reverb doesn't have an "EXT SPKR" jack like you would find on a Twin Reverb. there is only one speaker output, and it's labelled "SPKR MIN LOAD 4 OHMS". there is another 1/4 jack next to it, but it's not a speaker output. there's a 8:1 voltage divider going to that jack, to provide a line output feed to a recording console. that jack is labelled " OUTPUT LINE/RECORDING" and it's a plain switchcraft mono jack.

before taking jack measurements I disconnected everything from both jacks -- i disconnected the green OT lead and I lifted the voltage divider. the switching jack contacts for the speaker output are definitely arc'd over so that inserting/removing a jack makes no difference. no two ways about it, the jack is going to have to be replaced. so now it's a question of whether it's better to replace it with the exact same type of jack, and let the same thing happen again, or to replace it with a regular 1/4 mono jack and strap a safety resistor across the OT secondary.

i'm leaning toward the safety resistor.

[pdf64]

' I prefer to put "voltage tether" resistors (say 330K to 470K) from each OT anode lead to 0V'
Ian, that's an interesting idea, what are the guidelines for the spec on such resistors, eg limiting element voltage and power rating?
Thanks - Pete

[pdf64]

My understanding is that the voltage at the power tube plate can be expected to swing from pretty much ground to twice VB+, so tether resistor element voltage rating will need to be 1000V.
A quick search through the data sheets (for stuff readily available in the UK) indicates that 2W CF look the best bet, eg 2 x 220k in series each side.
http://docs-europe.electrocomponents...6b80098b49.pdf
Carbon comp types seem to have better tolerance of high voltages though http://docs-europe.electrocomponents...6b80ab1e9d.pdf so it may be worth trying to find 1W or 2W versions of those (or use a series parallel arrangement of the 0.5W type).
Any thoughts?
Pete