Operating Amp Without Load

[AmpBuilder225]

Hi, I know its generally considered to be a big no as to whether an amplifier can be operated without a load (be it a resistive/reactive load like a hot plate or be it a speaker cabinet), but what problems would this cause and why?

Thanks for any help offered!

[rknize]

It is the transformer that is the issue. Without a load it will experience very high fly-back voltages in the windings from inductance which will can cause the insulation on the windings to break down and create shorts in the transformer. The tubes don't like those either as they get reflected back into the primary.

A short, accidental disconnection is likely not to be the end of the world, but don't ever do it on purpose or while testing at high volumes. It's generally OK to disconnect speakers while the amp is on as long as there is no input. Note that having no source connected at all is not the same thing, as the amp may pick up 60Hz and other noise and amplify that instead. When swapping speakers while listening, I'll always pause the music first, quickly do the swap, and then unpause.

[AmpBuilder225]

Thank you! That answers the question well and agrees with what i suspected.

[Johan Potgieter]

In well-designed amps there is usually a resistor across the load of some 20 - 50 x the output impedance. E.g. for 8 ohm out it could be 330 ohms (say >2W, just in case). This leaves the amp with some load in case of accidental speaker wire breakage, and the output loss is negligible.

[llwhtt]

Vox always used a 470 Ohm resistor across the 16 Ohm tap on their AC50 and AC100 heads. I assume it was for that reason, since they were heads only and not combos like the AC10/15/30
which always had speakers connected. Fender amps usually shorted the output to ground with no load connected, same reason I suppose. Hiwatt shorted to ground also I think.

Craig

[Wavebourn]

Ada Depot used high voltage diodes across tubes.

[Johan Potgieter]

Originally posted by llwhtt
Fender amps usually shorted the output to ground with no load connected ....

OUCH!

I am not familiar with Fender, but with any NFB that would have created a real horror show - huge gain, output tubes working into a short .... Even without NFB, output anodes having to remain at full voltage ...

But off-topic. [I occasionally made the feedback resistor 680 ohm with the rest corresponding to ratio - dual purpose then.]

[Wavebourn]

It is true, but anyway Fender amps were designed to be abused to give "a proper guitar overdrive".
Speaking of feedback resistors, if to choose them so low you have to care about distortions: heated by signal resistors show non-linearities.

[EC8010]

So long as you don't overdrive the amplifier, there is no problem with applying a signal to a valve amplifier without a load. How else do you think it would be safe to drive electrostatic headphones? The flyback insulation problem mentioned earlier only happens under overdrive.

[Johan Potgieter]

Originally posted by EC8010
So long as you don't overdrive the amplifier, there is no problem with applying a signal to a valve amplifier without a load. How else do you think it would be safe to drive electrostatic headphones? The flyback insulation problem mentioned earlier only happens under overdrive.

Ah EC, would that that was always true .....

Not so for some amplifiers, although as you say it should be true, and I myself like to design that way. In mitigation, one must realise that the OPT-output tube circuit capacitances will have a different pole than when loaded. But it should not be difficult to achieve such a situation, and will safeguard circuits when some funny loudspeaker's impedance goes all over the show.