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

[hesamadman]

I would like to understand more as to why the impedance selection on an amplifier is crucial and how a cabinet affects it. Thanks

[R.G.]

I would like to understand more as to why the impedance selection on an amplifier is crucial

The impedance selection on an amplifier is not crucial.

For minor mismatches (i.e. 2:1 or even 4:1) this has an effect on tone. Getting it just right may affect how much power the amplifier can put out, for instance a "50W" amplifier may only put out 40W with a mismatched load. This is almost impossible to hear as a change in volume, though.

and how a cabinet affects it.

The cabinet has a nominal impedance, generally 4, 8 or 16 ohms. I say "nominal" because a speaker is not a constant load. Its loading on an amp varies with frequency and with the acoustic loading on it by the enclosure.

If we ignore that, and say an "8 ohm" speaker really is the same as an 8 ohm resistor, then we can talk about the loading that this presents to the amp.

If we connect an 8 ohm resistor load to an amp, and set the impedance selector to 8 ohms, then the conditions the manufacturer envisioned are satisfied, and the amp **should** produce the specified power and response.

If we then change the selector to the 16 ohms setting, the actual load of 8 ohms is fed with 1.4 times as much voltage by the amplifier, and the output tubes see a reflected load resistance that's half what they expected. They will run out of current drive capability before they run out of voltage to feed the load. So the power they can produce goes down a bit. Otherwise, they don't much care. The output transformer may run a bit more current on some windings than it expects, but this is not damaging ***UNLESS*** the transformer design was already marginal or the transformer was already damaged in some way. There will be a loss of some bass response in most cases, I believe.

If we change the selector to the 4 ohms setting, the actual load of 8 ohms is fed with 0.7071 times as much voltage by the amplifier, and the output tubes see a reflected load resistance that's twice what they expected. They will run out of voltage drive capability before they run out of current to feed the load. So the power they can produce goes down a bit. Otherwise, they don't much care. The output transformer may run a bit less current on some windings than it expects, but this is not damaging unless the transformer design was already marginal or the transformer was already damaged in some way. There may be a bit more bass response.

The title of your post, "what happens to tube amp if powered on without cabinet. how and why" asks a different question.

If an amp is powered on without a cabinet attached, whatever load the amp expects by the impedance selector is still reflected as an open circuit. The output tubes then see only the loading of the primary side of the output transformer. This looks like a large inductor with a small inductor (the leakage inductance) in series with it.

What happens to the amp in this case depends on whether the amplifier is stable enough not to go into oscillation with a purely inductive load. If it is, nothing bad happens. If it is not, it can oscillate enough to let the primary and leakage inductances cause damaging voltage spikes. The stack of reverse biased diodes at the ends of the primary on may amplifiers are one way to prevent this in most cases. There are other techniques to prevent this situation from being fatal to the output transformer or tubes as well.

This scenario is why the common wisdom is to never power a tube amp without a load connected. Some of them survive this, some will be damaged by it.

[hesamadman]

I was always under the impression that running a cabinet with an ohm rating lower than what your amp is set at can damage the amp?

Also, how did you get the figures when you said running a head ar 16 on an 8 ohm cab gives you 1.4 the power?

[eschertron]

I was always under the impression that running a cabinet with an ohm rating lower than what your amp is set at can damage the amp?

Many amps that have a 1/4" jack for the speaker use jacks that are 'shorting'. That is, when there is no speaker cable plugged in the OT is shorted and the power amp sees a very low reflected impedance. The OT generates very little voltage across the windings (the tubes still dissipate their idle power rating), and thus 'shorting' the OT is a safety feature to protect the OT. Not every amp has this kind of plug, so it's a good idea to verify before powering any amp up. Also, the shorting jack does nothing to stop the open circuit that results from plugging a speaker cable in with no cab on the other side.

Short answer: No, running a cab with a lower Ohm rating will not hurt the amp.

Running a cab with a lower Wattage rating? First the speakers blow up, then the amp!

[hesamadman]

Running a cab with a lower Wattage rating? First the speakers blow up, then the amp!

That is the case only if you crank the amp. Correct? Like if you run a 1x12 30 watt speaker cabibet with a 75 watt head. You can do that at safe volumes right?

[eschertron]

That is the case only if you crank the amp. Correct? Like if you run a 1x12 30 watt speaker cabibet with a 75 watt head. You can do that at safe volumes right?

I was being somewhat tongue-in-cheek... yes you can avoid melting speaker voice coils if you take care. But what are the limits? A peak voltage excursion at the OT can damage the speakers even with your master at a 'safe' volume simply because the tubes, B+, and the OT have no physical limitation preventing the excursion. Only some sort of power soak will actually reduce the max voltage from the OT. It sounds like you are thinking about the limitations of the equipment, and that will go a long way to its safe use.

But if I had an expensive 30 watt speaker that I valued, I'd surely consider buying a cheap 100 - 150 watter to put in the cabinet while I was playing with a 75W head.

[olddawg]

It wasn't mentioned if the OP was asking about tube or solid state amps. It should be stated that with most SS amps you can run them all day long with no load at all. The amp does not care. But... You should never run an SS amp with a lower impedance load than it is rated. It will draw too much current from the output devices and they will fail. A shorted load will kill most SS amps immediately. If the amp is rated for an 8 ohm load, never go lower than 8. If f it is rated at 4 ohms never go lower than 4. Many SS amps will tolerate a 2 ohm load if they are designed and rated for it. Most are not and will not. But you can alway use a bigger load. You just lose efficiency and the amp runs cooler.

[R.G.]

I was always under the impression that running a cabinet with an ohm rating lower than what your amp is set at can damage the amp?

For solid state amps, perhaps. Most solid state amps are designed so they do not melt down with too-low a load or a short these days. Some old solid state amps will die.

For tube amps (which is what I thought you meant) they're fine into a short, it's the issues with voltage spikes from open loads that they have trouble with.

Also, how did you get the figures when you said running a head ar 16 on an 8 ohm cab gives you 1.4 the power?

I didn't. I said (in effect) that the 16 ohm tap provides 1.4 times the voltage. This is because there's a square-root-of-two in the math, and that's 1.414..., so the voltage on a 16 ohm tap is 1.4(and some fractions) times the voltage on an 8 ohm tap. The 4 ohm tap is 1/2 the voltage of the 16 ohm tap.

That is the case only if you crank the amp. Correct? Like if you run a 1x12 30 watt speaker cabibet with a 75 watt head. You can do that at safe volumes right?

Not correct. It ignores the human factor that even if a guitarist knows better, they will think all controls dimed is a "safe volume".

[GregS]

Not correct. It ignores the human factor that even if a guitarist knows better, they will think all controls dimed is a "safe volume".

To add to this there is also the possibility of unexpected large signals at the input such as when the power is initially connected to an effect pedal that's plugged into the amp. Even with the volume down the signal at the output can be very large resulting in a loud "blip" at the speaker.

[pdf64]

Also many classic guitar amps (eg with no master volume control) can be hitting power amp overdrive with a regular guitar and a channel volume setting of just 2 or 3; Marshall Super Lead being a prime example.
Pete

[hesamadman]

The impedance selection on an amplifier is not crucial.

For minor mismatches (i.e. 2:1 or even 4:1) this has an effect on tone. Getting it just right may affect how much power the amplifier can put out, for instance a "50W" amplifier may only put out 40W with a mismatched load. This is almost impossible to hear as a change in volume, though.

The cabinet has a nominal impedance, generally 4, 8 or 16 ohms. I say "nominal" because a speaker is not a constant load. Its loading on an amp varies with frequency and with the acoustic loading on it by the enclosure.

If we ignore that, and say an "8 ohm" speaker really is the same as an 8 ohm resistor, then we can talk about the loading that this presents to the amp.

If we connect an 8 ohm resistor load to an amp, and set the impedance selector to 8 ohms, then the conditions the manufacturer envisioned are satisfied, and the amp **should** produce the specified power and response.

If we then change the selector to the 16 ohms setting, the actual load of 8 ohms is fed with 1.4 times as much voltage by the amplifier, and the output tubes see a reflected load resistance that's half what they expected. They will run out of current drive capability before they run out of voltage to feed the load. So the power they can produce goes down a bit. Otherwise, they don't much care. The output transformer may run a bit more current on some windings than it expects, but this is not damaging ***UNLESS*** the transformer design was already marginal or the transformer was already damaged in some way. There will be a loss of some bass response in most cases, I believe.

If we change the selector to the 4 ohms setting, the actual load of 8 ohms is fed with 0.7071 times as much voltage by the amplifier, and the output tubes see a reflected load resistance that's twice what they expected. They will run out of voltage drive capability before they run out of current to feed the load. So the power they can produce goes down a bit. Otherwise, they don't much care. The output transformer may run a bit less current on some windings than it expects, but this is not damaging unless the transformer design was already marginal or the transformer was already damaged in some way. There may be a bit more bass response.

The title of your post, "what happens to tube amp if powered on without cabinet. how and why" asks a different question.

If an amp is powered on without a cabinet attached, whatever load the amp expects by the impedance selector is still reflected as an open circuit. The output tubes then see only the loading of the primary side of the output transformer. This looks like a large inductor with a small inductor (the leakage inductance) in series with it.

What happens to the amp in this case depends on whether the amplifier is stable enough not to go into oscillation with a purely inductive load. If it is, nothing bad happens. If it is not, it can oscillate enough to let the primary and leakage inductances cause damaging voltage spikes. The stack of reverse biased diodes at the ends of the primary on may amplifiers are one way to prevent this in most cases. There are other techniques to prevent this situation from being fatal to the output transformer or tubes as well.

This scenario is why the common wisdom is to never power a tube amp without a load connected. Some of them survive this, some will be damaged by it.

Sorry to bug you again. I've read this thread multiple times. Just so I can understand as much as possible. I was just hoping you can explain to me how you got the figure of "1.4" when you said the amp feeds 1.4 Times the voltage in example 1 and .707 in ezample 2.

[Dave H]

Sorry to bug you again. I've read this thread multiple times. Just so I can understand as much as possible. I was just hoping you can explain to me how you got the figure of "1.4" when you said the amp feeds 1.4 Times the voltage in example 1 and .707 in ezample 2.

The OT is set up to produce the same power on each of its taps when they are correctly loaded and Power = V^2/R so the 16ohm tap must have 1.414 times the voltage of the 8ohm tap and the 4ohm tap must have 0.707 times the voltage of the 8ohm tap for the power to be the same. It comes out of the P = V^2/R equation.

[isaac]

Most amps with a standby switch won't have a problem with turning it on without a cab, just dont hit the standby on, as the output transformer has no power in it yet, and the tubes aren't trying to push anything through it. SO most of the time its alright, just don't hit the standby. Now some amps use a load resistor wired on the jack just so that when nothing is plugged in it will create enough load to not be bad for the amp, not many but some, but in all reality you wouldn't want to turn your amp up past 1-2 probably with that resistor as its probably not a very large wattage resistor and it could burn out. But its just good practice to not turn on your amp without plugging in a cab. I usually, just to save time, turn on the amp to warm up, NO standby yet, I then proceed to hook up whatever else cables, footswitch, speakers, guitar cord, etc. and then after making sure everything is plugged in, then I turn on the standby.

[Jazz P Bass]

The transformer 'works' on an ac signal.
With no signal, it is doing nothing.
So with the amp powered up & no cabinet hooked up, there is no harm.
Correct?

[bob p]

I want to nitpick for a minute, and ask you guys about the use of shorting jacks on the amplifier output. There are quite a few amps (Fenders come to mind) that use shorting jacks on the speaker output. The logic behind this is that if the user fails to plug in a speaker cabinet, the OT will see a dead short rather than an open circuit, which should protect the amp from oscillating, and the OT from being burdened with high voltage swings that could test the insulation on the windings.

I've never thought of a dead short as a partcularly kind way to protect the output transformer. I always thought that the amp manufacturers chose to use a shorting jack solely because the incremental cost of going from a regular jack to a shorting jack was less expensive than going from a regular jack to a regular jack plus a suitable power resistor. i've always thought that something like a 270R/5W resistor across the OT secondary would be a much kinder way to protect the amp, and that the use of shorting jacks was done for cost containment, not because it was a particularly good way to protect the amp.

having looked at a few shorting jacks, it seems all too common that they've got arc'd over contacts that are full of carbon, and the jack invariably seems to lose it's functionality after it's protected the output stage a few times. i;ve always thought that eliminating the jack as a failure point, and replacing it with a resistor, was a much safer option.