I don't know if my amp is "stable" or not even though I've build it. But it's been running reliably without a fault so I guess that makes it stable for me 🙂 specially since nothing happened to the OT.
On the other hand I have another amp, this one has big old 80 rectifier tube. Once I turned the amp on without speakers by mistake while working on it and the rectifier turned in to plasma ball for about 15 seconds. Rectifier survived, but it was weird.
On the other hand I have another amp, this one has big old 80 rectifier tube. Once I turned the amp on without speakers by mistake while working on it and the rectifier turned in to plasma ball for about 15 seconds. Rectifier survived, but it was weird.
@6V6dude:
As you might have concluded by now, it is very much dependent on the topology of the amp if it will survive being used with no load or not. Does it employ global negative feedback (GNFB)? And is the output tube tetrode or triode connected? I haven't seen an answer to those questions yet.
Single ended triode amplifiers without feedback will be fine without a load, even when driven. Single ended pentode amplifiers without GNFB will also be fine without a load, as long as they are not driven. Things go wrong when the output stage is a tetrode/pentode one and GNFB is present. Without load, the gain of the output stage will be one to two orders of magnitude higher than with the rated load, and the feedback loop can become unstable when the amp has no designed-in provision to keep it stable in this situation. Like Johan said, an amplifier with pentodes and GNFB can be designed to be robust in this situation, and some designs even accidentally are, but don't count on it when you're not sure.
Signs of damage to look for are arcing inside tubes, (traces of) arcing at the output tube sockets, and insulation damage in the output transformer. If no voltage-related damage has occurred (arcing, insulation damage) then the amp is fine. Problem is that, voltage related damage is not always visible. If a flashover has occurred inside the OPT, it may still appear to work normally, but the broken down part of the insulation gradually deteriorates until it becomes noticeable, either auditory or olfactory. The only way to be sure that the output transformer has not been damaged is to measure its primary inductance and equivalent parallel resistance at operating voltage, and to compare the numbers with those from a known good specimen. Also, you can use an insulation tester to test the insulation between primary and secondary windings. In both measurements, increased leakage current is a sign of imminent failure.
So in true consultant style, the answer is "it depends". But chances are good that your amp is fine.
As you might have concluded by now, it is very much dependent on the topology of the amp if it will survive being used with no load or not. Does it employ global negative feedback (GNFB)? And is the output tube tetrode or triode connected? I haven't seen an answer to those questions yet.
Single ended triode amplifiers without feedback will be fine without a load, even when driven. Single ended pentode amplifiers without GNFB will also be fine without a load, as long as they are not driven. Things go wrong when the output stage is a tetrode/pentode one and GNFB is present. Without load, the gain of the output stage will be one to two orders of magnitude higher than with the rated load, and the feedback loop can become unstable when the amp has no designed-in provision to keep it stable in this situation. Like Johan said, an amplifier with pentodes and GNFB can be designed to be robust in this situation, and some designs even accidentally are, but don't count on it when you're not sure.
Signs of damage to look for are arcing inside tubes, (traces of) arcing at the output tube sockets, and insulation damage in the output transformer. If no voltage-related damage has occurred (arcing, insulation damage) then the amp is fine. Problem is that, voltage related damage is not always visible. If a flashover has occurred inside the OPT, it may still appear to work normally, but the broken down part of the insulation gradually deteriorates until it becomes noticeable, either auditory or olfactory. The only way to be sure that the output transformer has not been damaged is to measure its primary inductance and equivalent parallel resistance at operating voltage, and to compare the numbers with those from a known good specimen. Also, you can use an insulation tester to test the insulation between primary and secondary windings. In both measurements, increased leakage current is a sign of imminent failure.
So in true consultant style, the answer is "it depends". But chances are good that your amp is fine.
There can be several points of phase shift in a multistage amp. Some may 'lag' and some may 'lead'.
Some amp negative feedback networks use a capacitor across the feedback resistor.
This causes the negative feedback signal to 'lead'.
Output transformers have leakage reactance. It causes the signal to 'lag'.
The combination of all the amplifier 'leads' and 'lags' needs to be such that the negative feedback does not become positive feedback.
If you unload the output transformer, the leakage reactance does not cause as much 'lag' as it does in the loaded condition. If the negative feedback 'lead' and the reduced leakage reactance 'lag' cause the feedback to go positive, you get an oscillation.
If you load the output transformer as normal, plus you add a capacitor across the load, the capacitor and the leakage reactance causes more 'lag'. If the negative feedback 'lead' is not enough to make up for the increased leakage reactance 'lag', it can cause the feedback to go positive, and you get an oscillation.
Of course this is an oversimplification, since the transformer has many causes of lead and lag, depending on the frequency, and it has more than one resonance.
Many amplifier tests just use a load resistor. Most loudspeakers are not a resistor.
That is why you see so many mentions of added capacitor loads, RC networks, and loudspeaker impedance simulators in this and other threads. They help to make sure the amp is stable and low distortion, etc., when used on real world (loudspeaker and speaker cable) loads.
Some amp negative feedback networks use a capacitor across the feedback resistor.
This causes the negative feedback signal to 'lead'.
Output transformers have leakage reactance. It causes the signal to 'lag'.
The combination of all the amplifier 'leads' and 'lags' needs to be such that the negative feedback does not become positive feedback.
If you unload the output transformer, the leakage reactance does not cause as much 'lag' as it does in the loaded condition. If the negative feedback 'lead' and the reduced leakage reactance 'lag' cause the feedback to go positive, you get an oscillation.
If you load the output transformer as normal, plus you add a capacitor across the load, the capacitor and the leakage reactance causes more 'lag'. If the negative feedback 'lead' is not enough to make up for the increased leakage reactance 'lag', it can cause the feedback to go positive, and you get an oscillation.
Of course this is an oversimplification, since the transformer has many causes of lead and lag, depending on the frequency, and it has more than one resonance.
Many amplifier tests just use a load resistor. Most loudspeakers are not a resistor.
That is why you see so many mentions of added capacitor loads, RC networks, and loudspeaker impedance simulators in this and other threads. They help to make sure the amp is stable and low distortion, etc., when used on real world (loudspeaker and speaker cable) loads.
Dave was doing his unloaded pre-mod and post-mod testing on Heathkit W-2M and 3M amps. Pentode, 20dB GFB, tapped screens. He swapped FB to all the taps for comparisons. No mention of safety loads.
Coincidentally, just a few weeks back, there was someone who reported arcing at the tube pins when running his amp without a load. Basically his power tubes were toast. I don't recall the rest of the details, including whether his OT was salvageable..
Sent from my phone. Please excuse any typpos.
Output tubes that are wired in Ultra Linear mode (tapped screens) have 'effective' mu (u) of about 2 to 3 times that of the same tube that is wired in triode mode.
And the output tube plate resistance of a Ultra linear stage is about 2 to 3 times that of the same tube that is connected in triode mode.
The open loop gain change from loaded to unloaded in Ultra Linear mode is more than the same tube that is triode wired.
The Ultra Linear open-loop gain to closed-loop gain ratio changes versus load variations, but is not as drastic as it would be for the same tube that is wired in pentode mode; but the Ultra Linear gain of open loop to closed loop ratio changes more drastically than the same tube that is wired in triode mode.
Negative feedback requires attention to open loop gain changes as the load is varied (be it resistive, capacitive, inductive, resonant load, or combination of them). As the frequency changes, so does the load.
And the output tube plate resistance of a Ultra linear stage is about 2 to 3 times that of the same tube that is connected in triode mode.
The open loop gain change from loaded to unloaded in Ultra Linear mode is more than the same tube that is triode wired.
The Ultra Linear open-loop gain to closed-loop gain ratio changes versus load variations, but is not as drastic as it would be for the same tube that is wired in pentode mode; but the Ultra Linear gain of open loop to closed loop ratio changes more drastically than the same tube that is wired in triode mode.
Negative feedback requires attention to open loop gain changes as the load is varied (be it resistive, capacitive, inductive, resonant load, or combination of them). As the frequency changes, so does the load.
Last edited:
Good summarising by 6A3sUMMER in posts #23 and #26.
Also as noted by DF96.
Yes, I 'chicken out' by always having at least a 1K resistor or as low as 680 ohms across the output internally. Or one can simply make the GNFB resistor small enough to act as an internal load, as in Quad II. One can risk having an adjustable resistor across the output and increasing it until instability is noted, but naturally then risky.
What to do with limited test gear?
An oscilloscope and signal generator going to several hundred KHz will be the minimum test equipment required. Start with the required load resistor, and check the frequency response up to say 200 KHz (low input signal). Any tendency to go unstable will manifest itself by a hump in the frequency response at the frequency of interest (usually in the 60K - 120 KHz region. If much lower than that - 'back to the drawing board'!). One can then proceed to increase the load resistor until the 'hump' shows an amplitude increase not exceeding 1,4 times (3 dB) over normal. If that can be achieved with a load of say 680 ohm - 1K.ohm you are probably safe - connect such a resistor permanently across the loudspeaker terminals, preferably a >2W one. A good design with correct phase compensation as outlined above should present no hump, just a smooth decline in response. (Likewise to 6A3sUMMER's explanation this is simplified but should suffice.)
Folks who want to know more can download one of several good treatises on NFB from the internet.
Timpert,
As you may have concluded by now we are talking of GNFB in all cases, with the likelyhood of problems being greater with UL than with triode output stages but still totally controlable. (Straigh pentodes - - grrrr ...) Those with access to very sophisticated test equipment may be able to display a Nyquist plot, which shows the exact situation. Such equipment is however probably only to be found in sophisticated research labs.
Also as noted by DF96.
Yes, I 'chicken out' by always having at least a 1K resistor or as low as 680 ohms across the output internally. Or one can simply make the GNFB resistor small enough to act as an internal load, as in Quad II. One can risk having an adjustable resistor across the output and increasing it until instability is noted, but naturally then risky.
What to do with limited test gear?
An oscilloscope and signal generator going to several hundred KHz will be the minimum test equipment required. Start with the required load resistor, and check the frequency response up to say 200 KHz (low input signal). Any tendency to go unstable will manifest itself by a hump in the frequency response at the frequency of interest (usually in the 60K - 120 KHz region. If much lower than that - 'back to the drawing board'!). One can then proceed to increase the load resistor until the 'hump' shows an amplitude increase not exceeding 1,4 times (3 dB) over normal. If that can be achieved with a load of say 680 ohm - 1K.ohm you are probably safe - connect such a resistor permanently across the loudspeaker terminals, preferably a >2W one. A good design with correct phase compensation as outlined above should present no hump, just a smooth decline in response. (Likewise to 6A3sUMMER's explanation this is simplified but should suffice.)
Folks who want to know more can download one of several good treatises on NFB from the internet.
Timpert,
As you may have concluded by now we are talking of GNFB in all cases, with the likelyhood of problems being greater with UL than with triode output stages but still totally controlable. (Straigh pentodes - - grrrr ...) Those with access to very sophisticated test equipment may be able to display a Nyquist plot, which shows the exact situation. Such equipment is however probably only to be found in sophisticated research labs.
1. My amps are un-conditionally stable;
2. in PP amps I add damper diodes across output tubes, even GU-50 with 800V B+
You can run them as many hours as want without load, on full throttle!
All depends on how the amp was designed.
2. in PP amps I add damper diodes across output tubes, even GU-50 with 800V B+
You can run them as many hours as want without load, on full throttle!
All depends on how the amp was designed.
I tend to disagree. A single trace oscilloscope and a simple oscillator (with adjustable frequency) lashed up on a breadboard will allow you to find phase and gain margin related issues, if you know what you're doing and are willing to do a lot of back-and-forth probing. Of course progress will be annoyingly slow to a professional, but it can be done. With a bit of thought and sufficient time, simple tools will get a hobbyist very far.
But anyhow, I think this thread is going way beyond the original question. To get that one answered, here's my attempt at a wrap up:
Question:
I have left my amp powered on for an extended period without a load. Did that hurt the amp in any way? (no extra information given beyond the output stage: a single ended 6V6)
Answer:
Check the output tubes and their sockets for signs of arcing. If they appear clean, plug the tubes back in and power the amp up. Do they bias OK? Does the amp work normally? If 2 x yes, your tubes are fine.
If the output transformer has not been singing or making other strange noises when it was left on without a load, it is probably fine as well. At the typical voltages at which a HiFi amp uses the 6V6, the output transformer's insulation won't be challenged, even in a situation like this. There is only a very slight chance that it may have suffered internal insulation damage. If you have no means to check for that, time will tell, but if the previous check passed OK, there's no need to worry.
Feedback loop stability often depends on stray capacitance. Put a scope probe on the circuit and the capacitance changes. Similar things can happen with parasitic oscillation within the output stage. Hence there is no way to guarantee stability with no load (internal and external) unless you are willing to risk the OPT by actually trying it.
Can you get away with it? Yes, sometimes. Should you try? No, not unless you feel lucky.
Can you get away with it? Yes, sometimes. Should you try? No, not unless you feel lucky.
Yes I do have some NFB, not much tho. Probably best I'll just upload the schematic that'll explain all.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
GNFB in normal operation is negligible with this amp, loop gain is (very roughly) about 0.1 times (yes, that's -20 dB) so this amp will be unconditionally stable when used with a load has an impedance of roughly the rated load (-100%, + 1000%). You might just as well go without GNFB altogether, it doesn't really influence anything at all here.
Without load, the gain of the output stage might go up by a factor of 50 or so, but probably only around the natural resonant frequency of the OPT (which is in the order of 100 Hz to 1 kHz, but exceptions are plentiful). This is the frequency at which the primary inductance resonates with the stray parallel capacitance of the windings. At lower and higher frequencies, the gain will drop due to the influence of primary inductance (1 kHz and below) or the stray capacitance. So at around 100 Hz (give or take a very wide margin) the loop gain will already have dropped well below 1, and above 10 kHz, this will also be the case. Between these extremes, the phase angle will be between +90 and -90 degrees, which is ample phase margin, no oscillation will occur. So is this amp stable at no load? I am almost certain about this, the only slight uncertainty I have is that the output transformer may be such a botch job that the above doesn't apply. But even with the simplest possible construction of the OPT (primary in a single layer, with a single layer secondary around it) you're in no danger.
Now if you drive this amp hard with no load, your most prominent risk is to blow up the screen grid of the output tube, but as that also didn't happen, I am 99.9% certain that the amplifier has remained damage free. Is the voltage at the cathode of the 6V6 still around 21.5 V? If yes, go have a beer and forget about the whole thing.
Without load, the gain of the output stage might go up by a factor of 50 or so, but probably only around the natural resonant frequency of the OPT (which is in the order of 100 Hz to 1 kHz, but exceptions are plentiful). This is the frequency at which the primary inductance resonates with the stray parallel capacitance of the windings. At lower and higher frequencies, the gain will drop due to the influence of primary inductance (1 kHz and below) or the stray capacitance. So at around 100 Hz (give or take a very wide margin) the loop gain will already have dropped well below 1, and above 10 kHz, this will also be the case. Between these extremes, the phase angle will be between +90 and -90 degrees, which is ample phase margin, no oscillation will occur. So is this amp stable at no load? I am almost certain about this, the only slight uncertainty I have is that the output transformer may be such a botch job that the above doesn't apply. But even with the simplest possible construction of the OPT (primary in a single layer, with a single layer secondary around it) you're in no danger.
Now if you drive this amp hard with no load, your most prominent risk is to blow up the screen grid of the output tube, but as that also didn't happen, I am 99.9% certain that the amplifier has remained damage free. Is the voltage at the cathode of the 6V6 still around 21.5 V? If yes, go have a beer and forget about the whole thing.
Last edited:
What actually can happen is the back EMF from the output transformer becomes very large with no load and the valves can arc internally. You would know if that had happened! Otherwise with low power, not any likely damage.
I have seen (and posted about) tube amps being totally fried, damaged, or totally unharmed, by running into an open circuit. There is a simple explanation for this....inductance, a random internet definition:
There is plate current flowing through the primary of the OPT. This plate current changes with the audio signal. Mutual induction couples most of this changing current into the OPT's secondary, which then flows through the speaker. There is a path for the current created by the counter EMF (opposing voltage) generated in opposition to the changing current in the primary through the OPT secondary through the speaker.
If the load is suddenly removed while there is a CHANGING current through the OPT, inductance will guarantee that there will be an opposing EMF (voltage) generated in an attempt to counteract the changing current through the primary. Removing the load breaks the path for current flow in the secondary, so the only remaining path for current to flow is in the primary. If there is no path for current to flow in the primary either, the voltage will RISE until there IS a path for current to flow in either the primary or secondary. This voltage rise will reach INFINITY in a theoretically perfect inductor. The ignition coil in your car operates in this manner. Tesla exploited this property and enhanced it by employing RESONANCE to further increase the voltage rise. An imperfect transformer (typical OPT) may see several kilovolts across the primary in this case.
As long as there IS a path for current to flow in the OPT primary the voltage rise will be minimal, and no damage should happen. If one or more output tubes are conducting, and not driven to cutoff by the counter EMF, there will be a path for current to flow.
In any amp with zero signal applied the current through the OPT should not be changing, therefore there will be no counter EMF, and no damage will occur. Some amps may go unstable and could possibly oscillate without a load. This will generate a changing current, and possible damage.
Your amp is a class A design which employs a small amount of GNFB to lower the pentode's output impedance. The open loop gain is low enough that instability shouldn't be a problem, and you say it still works, so it is fine.
If the amp was driven hard enough such that is was clipping, there will be periods of time when the output tube is not conducting, and therefore subject to damage. Note that an amp may have higher gain without a load (depending on feedback) so clipping could occur at a drive level that didn't cause clipping with a load.
A push pull amp operating in class AB, or class B will have one tube cutoff for a considerable part of an audio cycle. it's plate voltage is rising at the peak of the audio cycle, and will approach twice the B+ voltage at the crest of the cycle. The other tube will be conducting strongly and its plate will approach zero. If the load is suddenly removed at this point (speaker or load resistor blows) the counter EMF will drive the conducting plate's voltage negative ceasing conduction. A tube will not conduct when it's plate is negative WRT it's cathode. At this point the other tube's plate voltage will rise UNTIL A PATH FOR CURRENT FLOW IS CREATED!!!!! This path is usually an arc in the OPT, the output tube, or its socket. Once an arc starts it will be fed by the B+ supply until something blows. Typical audio tubes have their plate on pin 3, and the heater on pin 2. If the arc happens in a tube socket, B+ jumps into the heater supply. If the heater supply is grounded it is possible for nothing bad to happen except for a blown fuse. This is NOT usually the case.
The worse case I have seen was an Ampeg SVT bass guitar amp rated at about 400 watts. It was at full crank when the speakers blew. The arc turned into a fireball that took out the OPT, the power transformer, several tubes and a bunch of small parts.
I personally set an OPT on fire when I was cranking 150 watts out of a pair of TV sweep tubes when the load resistor I was using blew.
There is plate current flowing through the primary of the OPT. This plate current changes with the audio signal. Mutual induction couples most of this changing current into the OPT's secondary, which then flows through the speaker. There is a path for the current created by the counter EMF (opposing voltage) generated in opposition to the changing current in the primary through the OPT secondary through the speaker.
If the load is suddenly removed while there is a CHANGING current through the OPT, inductance will guarantee that there will be an opposing EMF (voltage) generated in an attempt to counteract the changing current through the primary. Removing the load breaks the path for current flow in the secondary, so the only remaining path for current to flow is in the primary. If there is no path for current to flow in the primary either, the voltage will RISE until there IS a path for current to flow in either the primary or secondary. This voltage rise will reach INFINITY in a theoretically perfect inductor. The ignition coil in your car operates in this manner. Tesla exploited this property and enhanced it by employing RESONANCE to further increase the voltage rise. An imperfect transformer (typical OPT) may see several kilovolts across the primary in this case.
As long as there IS a path for current to flow in the OPT primary the voltage rise will be minimal, and no damage should happen. If one or more output tubes are conducting, and not driven to cutoff by the counter EMF, there will be a path for current to flow.
In any amp with zero signal applied the current through the OPT should not be changing, therefore there will be no counter EMF, and no damage will occur. Some amps may go unstable and could possibly oscillate without a load. This will generate a changing current, and possible damage.
Your amp is a class A design which employs a small amount of GNFB to lower the pentode's output impedance. The open loop gain is low enough that instability shouldn't be a problem, and you say it still works, so it is fine.
If the amp was driven hard enough such that is was clipping, there will be periods of time when the output tube is not conducting, and therefore subject to damage. Note that an amp may have higher gain without a load (depending on feedback) so clipping could occur at a drive level that didn't cause clipping with a load.
A push pull amp operating in class AB, or class B will have one tube cutoff for a considerable part of an audio cycle. it's plate voltage is rising at the peak of the audio cycle, and will approach twice the B+ voltage at the crest of the cycle. The other tube will be conducting strongly and its plate will approach zero. If the load is suddenly removed at this point (speaker or load resistor blows) the counter EMF will drive the conducting plate's voltage negative ceasing conduction. A tube will not conduct when it's plate is negative WRT it's cathode. At this point the other tube's plate voltage will rise UNTIL A PATH FOR CURRENT FLOW IS CREATED!!!!! This path is usually an arc in the OPT, the output tube, or its socket. Once an arc starts it will be fed by the B+ supply until something blows. Typical audio tubes have their plate on pin 3, and the heater on pin 2. If the arc happens in a tube socket, B+ jumps into the heater supply. If the heater supply is grounded it is possible for nothing bad to happen except for a blown fuse. This is NOT usually the case.
The worse case I have seen was an Ampeg SVT bass guitar amp rated at about 400 watts. It was at full crank when the speakers blew. The arc turned into a fireball that took out the OPT, the power transformer, several tubes and a bunch of small parts.
I personally set an OPT on fire when I was cranking 150 watts out of a pair of TV sweep tubes when the load resistor I was using blew.
Yup, when the output tubes are driven to cutoff, you get a good inductive surge on the plate. Arc between pins 3 (plate) and 4 (screen) and/or 2 (heater) is most likely to occur, depending on your soldering.
Because the amount of GNFB is so low (it only becomes significant without load, and actually you might argue that its only real effect is to keep the amp intact when accidentally unloaded), and your amp will still have an ample stability margin without load, I expect the GNFB to fully prevent this surge from occurring, unless you drive the amp into clipping. This only applies to this particular amp, other amps may/will be less forgiving.
Certainly true, in this particular case I think that the overall gain will increase with a factor of about 10 or somewhat less. The output stage's gain will increase by a factor of 50 or so when unloaded, but this gain increase is compensated by the feedback suddenly becoming of significance.
The output tube will not cut off (and thus loose control over the OPT) unless G1 is driven beyond its normal drive level. The GNFB loop will work normally (it is stable), and reduce G1 drive voltage when the load is removed. The positive surge on the plate can be fully "caught" by the GNFB, because the output stage still amplifies normally (I am assuming the amp it is not grossly overdriven). When driven at a normal level, the positive surge will be limited to about 3 kV, which is a value that any good output transformer must be able to take without damage.
Trouble starts with the plate going negative and the grid to cathode voltage going towards zero. In that unloaded situation, the plate goes low very quickly, and screen grids take the full dissipation of the plate for half the cycle, which is more than they can take. In this case, the output stage does not amplify any more, and the GNFB will react by jacking up the grid drive, making the problem worse.
But, unless the amp was being driven at a level near or beyond clipping, there should be no damage from being used without a load. Again, this only applies to this particular amp, in this particular situation.
As Tubelab said, other amps are treated much worse, and are less forgiving by themselves to begin with. That can give some spectacular (and expensive) fireworks.
Because the amount of GNFB is so low (it only becomes significant without load, and actually you might argue that its only real effect is to keep the amp intact when accidentally unloaded), and your amp will still have an ample stability margin without load, I expect the GNFB to fully prevent this surge from occurring, unless you drive the amp into clipping. This only applies to this particular amp, other amps may/will be less forgiving.
Certainly true, in this particular case I think that the overall gain will increase with a factor of about 10 or somewhat less. The output stage's gain will increase by a factor of 50 or so when unloaded, but this gain increase is compensated by the feedback suddenly becoming of significance.
The output tube will not cut off (and thus loose control over the OPT) unless G1 is driven beyond its normal drive level. The GNFB loop will work normally (it is stable), and reduce G1 drive voltage when the load is removed. The positive surge on the plate can be fully "caught" by the GNFB, because the output stage still amplifies normally (I am assuming the amp it is not grossly overdriven). When driven at a normal level, the positive surge will be limited to about 3 kV, which is a value that any good output transformer must be able to take without damage.
Trouble starts with the plate going negative and the grid to cathode voltage going towards zero. In that unloaded situation, the plate goes low very quickly, and screen grids take the full dissipation of the plate for half the cycle, which is more than they can take. In this case, the output stage does not amplify any more, and the GNFB will react by jacking up the grid drive, making the problem worse.
But, unless the amp was being driven at a level near or beyond clipping, there should be no damage from being used without a load. Again, this only applies to this particular amp, in this particular situation.
As Tubelab said, other amps are treated much worse, and are less forgiving by themselves to begin with. That can give some spectacular (and expensive) fireworks.
Thanks for posting the schematic. And . . . it is clear and easily readable.
It really allows all of us to focus in on the probability that the amp was not damaged by
running it unloaded.
You got a bonus for posting the the schematic, timpert noticed the error in regards to the volume control. It will allow anyone to build a duplicate amp (and hopefully remember to wire the volume control different than it is in the schematic).
It really allows all of us to focus in on the probability that the amp was not damaged by
running it unloaded.
You got a bonus for posting the the schematic, timpert noticed the error in regards to the volume control. It will allow anyone to build a duplicate amp (and hopefully remember to wire the volume control different than it is in the schematic).
Yup, but in this case, there is probably no damage. We've established that the amp is only in danger of being damaged when driven at or near clipping. With increasing drive level, the screens of the output tubes are likely to go first. Only at a much higher drive level, the transformer's insulation is in danger.
- Status
- Not open for further replies.