why didnt my Amp Blow up? Idled with no load

[Jerry2013]

My Marshall amp idled with no load for a few seconds 3-4 seconds probably, till I relaized and shut it off. Why didnt it blow up? Seems to be working fine.

[Colossal]

Why didnt it blow up?

Because you got lucky! You get one...

Don't feel too bad, we've all done it and you get that "oh sh*t!" moment when you realize your mistake.

[Reeltarded]

Because it was idling and because it takes more than that to kill it but don't test that for yourself just trust me.

Might go 30 seconds no load off idle, might go 60, might go longer. Hard to predict. Don't do that.

[Leo_Gnardo]

Many tube amps will let you get away with no load, as long as there's no signal. And some will even survive (for a little while, half a second to a minute) with a signal sent thru.

HOWEVER some won't. Some Mesa/Boogies go into a low frequency oscillation say 2 to 5 Hz and you can watch the output tubes blue glow blink on and off. Haven't seen one blow up yet due to this condition, but it's got to put some wear and tear on the tubes and OT. Other amps may start to self-oscillate at higher frequencies and may self-destruct quickly.

Some put a resistor, 10-20x the speaker ohms, across the output so that there's a load, even if non optimum, even if no speaker is plugged in. And others (Fender) have a shorting bar on the output jack so the amp sees a short if nothing's plugged into the output jack.

Only one rule of thumb, the bigger the amp, the less happy it will be without a load & more expensive to fix.

[Firestorm]

Everyone's first reaction is to shut the amp off. Probably the worst thing to do because of the back emf from the charged primary. Plug the speaker in if you can.

[LeftyStrat]

Maybe I'm just lucky, but I've always found tube circuits more resilient than solid state.

And those that prefer solid state won't be playing their amps after a nearby nuclear explosion results in an electromagnetic pulse that will fry all solid state, leaving us tube heads to play our remaining radiated hours at full volume. Assuming of course we still have power.

[tubeswell]

If the amp has NFB sourced from the OT secondary, it can self-oscillate and that creates a signal, which is bad is there's no load.

But if there's no signal, an OT can happily have no load, because the OT is not doing any 'work'. Its the AC going through the transformer that causes a problem if there's no load.

[es345]

There is no general rule like "if no load then the amp will blow up". The most critical situation is when at full output power you loose the connection to the load during a current peak at the secondary. Then you have induction peaks of up to several thousand volts at the primary side of the OT. This can kill the OT and the power tubes. If you loose connection when the current is just zero there is no induction spike and you might be lucky. So without any input signal nothing should be be damaged. If an amp in this condition (no input signal, no load) is already starting to oscillate then the design obviously has room for improvement.

I can confirm Leo_Gnardos mitigation method (10-20x the speaker ohms in parallel to the secondary). Myself I am using always 2 cables to the speaker cab(s) (redundancy approach).

[martin manning]

If the amp has NFB sourced from the OT secondary, it can self-oscillate and that creates a signal, which is bad is there's no load.

But if there's no signal, an OT can happily have no load, because the OT is not doing any 'work'. Its the AC going through the transformer that causes a problem if there's no load.

I can't imagine how the presence of negative feedback would increase the risk. Seems like it should be the opposite. Do you have a reference that discusses this?

[Reeltarded]

Wipe all controls to 0 and plug in the cabinet.

I had a catastrophe with three dimed 100s during a show when someone walking behind leaned into the middle one and they all went face down on the stage. The cabling ripped free from jacks at either end and all three output selectors flew into the dark corners of hell. Next day changed everything to twistlocks and installed the selectors @8R with white glue. heh

LD with a zipper tattoo over a massive scar from a leg surgery. Bret, are you out there? You dumbass!

[Leo_Gnardo]

I can confirm Leo_Gnardos mitigation method (10-20x the speaker ohms in parallel to the secondary).

Thankew, thankew! And to give credit I saw Dynaco doing it 50+ years ago, so steal good ideas, yes we should all do that.

Myself I am using always 2 cables to the speaker cab(s) (redundancy approach).

Wow I've never seen that done before. Score one for the Department of Redundancy Department. Sometimes - you get what you need.

[Zippy]

Next day changed everything to twistlocks and installed the selectors @8R with white glue. heh

LD with a zipper tattoo over a massive scar from a leg surgery. Bret, are you out there? You dumbass!

Yeah, please get Bret back to test your idiot-proofing. I'm betting he could still bust your gear. Let's see where your cords and cabs come apart!

I know from personal experience that no matter how hard you try to idiot-proof something, someone will come along with a better idiot.

Why do you think they invented the combo amp?

[Reeltarded]

Nternashional Stage Hands Union, local Lemmon 714.

Uhhh -drools-

[tubeswell]

If the amp has NFB sourced from the OT secondary, it can self-oscillate and that creates a signal, which is bad is there's no load.

But if there's no signal, an OT can happily have no load, because the OT is not doing any 'work'. Its the AC going through the transformer that causes a problem if there's no load.

I can't imagine how the presence of negative feedback would increase the risk. Seems like it should be the opposite. Do you have a reference that discusses this?

My thinking is that it depends on the amount of NFB and the design of the OT (incl the amount of iron in the core). Because there is a phase lag between Pr and Sec in the Ot, and the phase lag is different at different frequencies, it is feasible that if the amount of NFB is too great, it may trigger some of those frequencies into self oscillation if the phase lag is 'right'. Naturally I stand to be remonstrated with.

[img:414:220]http://labman.phys.utk.edu/phys222/modu ... s/iemf.gif[/img]

[img:336:286]http://labman.phys.utk.edu/phys222/modu ... former.gif[/img]

http://labman.phys.utk.edu/phys222/modu ... evices.htm

[martin manning]

Pete, the top picture you show is ok, but the bottom one showing 90 degrees shift from input to output is greatly exaggerated or just plain wrong, as an ideal transformer will have zero shift. If the output jack is shorted energy fed into the primary is dissipated in the secondary, and there would be no FB at all.

If one really wanted to be safe, a shorted jack and a resistor across the speaker output is best. A shorting jack won't protect against the other end of the cable being unconnected, and there will still be some very high voltages produced in the primary circuit with just the 10x-the-load resistor.

If you want more protection than that, abstinence the only solution.