Whats the maximum peak to peak voltage an amplifiers input can handle

[Atodovax]

Hello everyone i was wandering what is the maximum ac voltage or the greatest signal an amplifiers input can handle without having problems. I heard that some overdrives can output as much as 15volts peak to peak.

[PRR]

The classic tube guitar amps, about 0.5V peak (1V p-p).

Most solid-state amps have variable input gain, and "can" take larger inputs, but not at normal knob setting.

[Atodovax]

The classic tube guitar amps, about 0.5V peak (1V p-p).

Most solid-state amps have variable input gain, and "can" take larger inputs, but not at normal knob setting.

Thanks for replying but how is It that some boosters can have 5v peak to peak at there output?. Is that harnfull for the amp?

[GibsonGM]

What are the problems that can develop from applying too great a signal to a  1) tube amp   2) solid-state amp?  They are different creatures.

- Input cap....if your amp (typically solid state) has an input cap, you can absolutely pop it if you exceed its working voltage. 

- Solid state device limits...if you exceed the max input a BJT, JFET, MOSFET, Opamp...can handle, you will fry the device.

- Grid: 12AX7 says max 'negative DC grid voltage' is 50V...what happens 'down stream' if we apply say, 10V p-p to a 12AX7 input stage intended for a .5V p-p signal, which would put out 60V or something at the intended drive level?   We know it's going to clip like mad, and sound like crap....but would there be DAMAGE?   Could blow a speaker...could you overload stages enough to kill them, or damage the OT?  I'd 'guess' that the stage will hit grid current limiting and cutoff before damage would occur. It would square wave on you, and sound like total @ss.  The following stage would surely be well-overdriven, but shouldn't do more than the first, I am 'supposing'.     

Interesting question! We've all done it to some degree.  I don't know anyone who was so foolhardy as to apply a +/-15 V signal to their input, though!   I can assure you that I have personally hit mine with at least 5V p-p...it didn't sound nice (at all), and didn't kill it.  In fact, it was downright scary, and I wouldn't do it again...I think if you come close to doing damage, they do let you know, as long as it's not so high as to kill it in one shot.     

[Atodovax]

What are the problems that can develop from applying too great a signal to a  1) tube amp   2) solid-state amp?  They are different creatures.

- Input cap....if your amp (typically solid state) has an input cap, you can absolutely pop it if you exceed its working voltage. 

- Solid state device limits...if you exceed the max input a BJT, JFET, MOSFET, Opamp...can handle, you will fry the device.

- Grid: 12AX7 says max 'negative DC grid voltage' is 50V...what happens 'down stream' if we apply say, 10V p-p to a 12AX7 input stage intended for a .5V p-p signal, which would put out 60V or something at the intended drive level?   We know it's going to clip like mad, and sound like crap....but would there be DAMAGE?   Could blow a speaker...could you overload stages enough to kill them, or damage the OT?  I'd 'guess' that the stage will hit grid current limiting and cutoff before damage would occur. It would square wave on you, and sound like total @ss.  The following stage would surely be well-overdriven, but shouldn't do more than the first, I am 'supposing'.     

Interesting question! We've all done it to some degree.  I don't know anyone who was so foolhardy as to apply a +/-15 V signal to their input, though!   I can assure you that I have personally hit mine with at least 5V p-p...it didn't sound nice (at all), and didn't kill it.  In fact, it was downright scary, and I wouldn't do it again...I think if you come close to doing damage, they do let you know, as long as it's not so high as to kill it in one shot.   

Thanks you very much for replying. I started doubting when i meassured the output of a FSH1 from Tonepad reaching 5volts peak to peak sometimes on the sample and hold mode on my oscope. Then i meassured a jfet Booster outputing 3volts peak to peak . but both of them sound Great on my tube amp no square wave sounding or anything similar. Booster is driving the Preamp obviously but in a perfect sounding way and the FSH is sounding killer too so..... Is my osciloscope reading simething wrong or whats happening?

[PRR]

> Is that harnfull for the amp?

If it was, there would be a lot of amps dead within warranty, and factories going broke.

Most stage gear is easy to distort but hard to blow-up.

[Atodovax]

> Is that harnfull for the amp?

If it was, there would be a lot of amps dead within warranty, and factories going broke.

Most stage gear is easy to distort but hard to blow-up.

I still dont understand if there is any rule when desingning a pedals output. Is It ok to have 5volts peak to peak in a Booster? Could that ac voltage damage the amp somehow?

[ashcat_lt]

5V won't break anything.  It might sound like it's falling apart, but it probably won't unless there's already something else wrong with it.  The only good reason to give it that much is to push it to distort, though.

[GibsonGM]

My reply was sort of wry, kind of...."here is what I know, what I have done, or heard about".  There IS no real hard and fast rule.  Clearly, your 5V p-p isn't hurting anything?   I think PRR's 1st post of .5V...is aimed at "what is normal, what won't distort the gear and cause noise" (?)

No, there is no rule.  If you put 5V P-P into a transistor input amp, it may not pop the cap, but may sound like CRAP.  Same pedal into a tube amp...might sound like "LOVE ITSELF"!    At some point, the transistor amp will suffer damage to its front end (cap, or device limitations).  Tubes are harder to kill....much harder.   Did I say...MUCH harder?

So, if you design a pedal for sale or wish to be nice....you state, "outputs 5V P-P" to cover any issues that may come up.    For your own use - you EXPERIMENT!

With YOUR amp (tube, I hope)...you can up the output and just hear where it sounds like garbage.  Where that extra output doesn't get you anything positive, so you cut back.    If I had a pedal putting out 5V P-P, personally?  I don't think I'd want to go much higher - nothing to gain, just in my opinion.    You 'do work' inside a pedal, and really don't have a need to *use* the output of it to *do something* other than act like a booster...that output is top end of boosting, LOL.

[DIY Bass]

Ooh, Ooh, Ooh, I know.  If you put too much signal into your amp input, then after the initial preamp stage has amplified it even more, then you will blow all the op amps on the eq board.  This will then kill the zener diode on the power input to this board.  The shorted zener will then overload the fuse.  Being a slo-blo, and only just over current, it will melt the fuse holder before it pops completely. 

[marcelomd]

Didn't the Meshuggah preamp/booster put out something like 30V?

It can even sound good if you like evil metal.

[tubegeek]

If you overdrive the grid of a tube so that it is positive relative to the cathode, the grid will allow current to flow through the grid - normally a high impedance node, it will become a low impedance node under that condition (large positive peaks.)

In other words, the driving circuit will be presented with a much more difficult load and be forced to supply current - perhaps beyond its capabilities. That 5V positive peak you see on your oscilloscope, loaded by the 10Mohm probe, will not necessarily exist when the driving circuit is presented with a 1K or 100 ohm load, which is what the grid will become on positive peaks.

Result: squared-off peaks. Not squared-off anywhere near as high as look into the 'scope, but squared-off somewhat abruptly as the grid goes into conduction, above ~ 0.5V positive. Asymmetrical clipping.

If the driving circuit can source excessive current for a sustained period, you could possibly damage the tube by overheating the grid (it's not really designed for a lot of current flowing through it.) But tubes are generally petty tough and it might not even be fazed by this, it depends on how hard and for how long.

[PRR]

Most classic tube amps have 34k in series with the first grid, which can take 10mA half the time (positive swings). This figures as 340V Volts peak. It is *really* hard to blow-up a tube amp which has a series grid resistor.

"Transistor" (and chip) designs are all over the map. The wise designer has some series resistance, at least a few kOhm, and most general-purpose audio devices will stand 10mA. So over 20V peak.

I would *hope* that every guitar input could eat the output of a 100 Watt amplifier. That type abuse HAPPENS. This is 28 Volts in 8 Ohms. (80V peak-peak.)

We do know some very ancient Heath transistor guitar amps where the input devices go crappy for no apparent reason. Speculation includes either accidental speaker level or random static electricity. And early devices less robust than most since 1970.

Beer and tailgate-drop are probably bigger risks.

[tubegeek]

Most classic tube amps have 34k in series with the first grid

Well, that's certainly true and important. And I certainly neglected to take that into consideration. I guess I was thinking about the simplest/worst-case condition.

[Atodovax]

Thanks everyone... I still dont understand why my filter is sounding good and with no distortion even when im meassuring 5v peak to peak on my ossciloscope .... I have a Hantek USB scope, maybe im reading something wrong? i have a similar output at two boosters reaching 4v peak to peak and the amp is not sounding bad or squared

[R.G.]

You got good advice - there is no rule.

Beyond that, you have to know what you are driving, and what you're driving it with. It's a lot like the song "The Gambler": you got to know when to hold 'em, know when to fold 'em. How you know is to know the circuits you're dealing with. Good (and competent, regardless of intent) circuit design says that the designer has to know the range of all reasonably possible drivers and receivers and make sure their circuit is not damaged by any of the conceivable ones, and even more, operates predictably with any of them.

Case in point: opamps. The first opamps had no input protection provisions, no output current limitations, and no thought for the available output power. So it was easy to push the inputs too far and damage the inputs and to load the outputs beyond what they could stand and blow the whole chip. Today's opamps, inside the IC package, almost universally limit both power and current out of the chip, prevent damage to the chip if the inputs are pulled too far apart, and make the chip behave predictably whenever either or both inputs are dragged too close to the positive or negative power supplies. It took about four generations of opamps for that to become something that could be expected from any opamp you pick up casually.

For some of this mind set, you might want to read http://www.geofex.com/circuits/what_are_all_those_parts_for.htm  and http://www.geofex.com/circuits/when_good_opamps_go_bad.htm for some ideas.

The general concept is to think about what possible voltages and currents MIGHT be imposed on your circuit's input and output. For most voltage-mode inputs, you worry about what voltages are possible and what the corresponding currents and power might be. For instance, the 12AX7 input mentioned: a 12AX7 grid only has an active range of about 0V to -2V for the entire range of linear control of the grid current. Much below -2V and no current flows at all in the grid or plate, so you can go fairly wild with negative voltages here, because nothing happens until you arc from the grid to another electrode. 50V? It's probably OK at -100V. Positive going is a different story. When the grid is more positive than the cathode, the grid stops looking like an open circuit and changes to about a 5K resistance to the grid. Now you can flow current in the grid. Whether this is destructive to the grid or not is questionable, as there isn't a lot of current the cathode can provide. It's probably OK...
IF your signal source isn't damaged or otherwise driven mad by the sudden change from infinite resistance to 5K. Most are fine. Some are not.
Try something similar with a bipolar and you find that unless the emitter is tied to the negative supply by something to limit current, the base always looks like a diode to the emitter, and it will happily conduct currents that will destroy the transistor. this is in addition to whatever currents the collector may be capable of conducting. You can kill the base-emitter all by itself. Drive the base way negative and at about 5V -7V negative, the emitter-base zeners and again you can have destructive currents. Much the same thing can happen with MOSFETs and JFETs, although with slightly different failure mechanisms.

To fix this, circuit designers have to KNOW how the inputs can die, and put in series resistances for the base/gate and emitter/source/cathode to limit currents under given overloads. It is possible to design a pedal circuit that you could connect 120Vac to without damaging the pedal circuit. But you have to know what you're doing and how the parts react. Different parts react differently in various combinations, and so there cannot be a rule about input voltages. You just have to go learn the circuits and what to do with them.

Outputs are another matter. Tube amps famously can die when they are open circuited. Solid state amps used to die if you connected too low an impedance to them. And then the designers started thinking about how to make these things not die on simple, forehead-slapper mistakes. Today, you're pretty safe shorting any solid state power amp because they all protect against overcurrent, over-power, and short circuits. Opamps are also (today...) pretty safe with output shorts.

Some of the circuit techniques in the two pages I referenced show how. Series resistances, current shunts (diodes to dump excess current) and current limiters can be used to prevent excess current and power from damaging devices.

You'll still damage capacitors with over voltage. A 10V input cap on a pedal powered from a 9V battery is OK - until someone puts +/- 50V on the input. Some commercial pedals had all 10V pedals in the past. Gotta know what's actually in that circuit to know if it's a problem.

Another sneaky way for damage to happen is particular to bipolar inputs. If you take a nice, high gain, linear bipolar transistor and reverse-break (zener) its base emitter >even once< and as long as current isn't excessive the transistor will still work. But it will be permanently a little noisier. This is one way that equipment gets hissier as it gets older. Put a reverse-biased diode across the base-emitter and it permanently prevents this damage as well as not interfering with normal function.

Sadly, you just have to know the parts and know the circuits. Rules of thumb are not a huge help.

[Atodovax]

You got good advice - there is no rule.

Beyond that, you have to know what you are driving, and what you're driving it with. It's a lot like the song "The Gambler": you got to know when to hold 'em, know when to fold 'em. How you know is to know the circuits you're dealing with. Good (and competent, regardless of intent) circuit design says that the designer has to know the range of all reasonably possible drivers and receivers and make sure their circuit is not damaged by any of the conceivable ones, and even more, operates predictably with any of them.

Case in point: opamps. The first opamps had no input protection provisions, no output current limitations, and no thought for the available output power. So it was easy to push the inputs too far and damage the inputs and to load the outputs beyond what they could stand and blow the whole chip. Today's opamps, inside the IC package, almost universally limit both power and current out of the chip, prevent damage to the chip if the inputs are pulled too far apart, and make the chip behave predictably whenever either or both inputs are dragged too close to the positive or negative power supplies. It took about four generations of opamps for that to become something that could be expected from any opamp you pick up casually.

For some of this mind set, you might want to read http://www.geofex.com/circuits/what_are_all_those_parts_for.htm  and http://www.geofex.com/circuits/when_good_opamps_go_bad.htm for some ideas.

The general concept is to think about what possible voltages and currents MIGHT be imposed on your circuit's input and output. For most voltage-mode inputs, you worry about what voltages are possible and what the corresponding currents and power might be. For instance, the 12AX7 input mentioned: a 12AX7 grid only has an active range of about 0V to -2V for the entire range of linear control of the grid current. Much below -2V and no current flows at all in the grid or plate, so you can go fairly wild with negative voltages here, because nothing happens until you arc from the grid to another electrode. 50V? It's probably OK at -100V. Positive going is a different story. When the grid is more positive than the cathode, the grid stops looking like an open circuit and changes to about a 5K resistance to the grid. Now you can flow current in the grid. Whether this is destructive to the grid or not is questionable, as there isn't a lot of current the cathode can provide. It's probably OK...
IF your signal source isn't damaged or otherwise driven mad by the sudden change from infinite resistance to 5K. Most are fine. Some are not.
Try something similar with a bipolar and you find that unless the emitter is tied to the negative supply by something to limit current, the base always looks like a diode to the emitter, and it will happily conduct currents that will destroy the transistor. this is in addition to whatever currents the collector may be capable of conducting. You can kill the base-emitter all by itself. Drive the base way negative and at about 5V -7V negative, the emitter-base zeners and again you can have destructive currents. Much the same thing can happen with MOSFETs and JFETs, although with slightly different failure mechanisms.

To fix this, circuit designers have to KNOW how the inputs can die, and put in series resistances for the base/gate and emitter/source/cathode to limit currents under given overloads. It is possible to design a pedal circuit that you could connect 120Vac to without damaging the pedal circuit. But you have to know what you're doing and how the parts react. Different parts react differently in various combinations, and so there cannot be a rule about input voltages. You just have to go learn the circuits and what to do with them.

Outputs are another matter. Tube amps famously can die when they are open circuited. Solid state amps used to die if you connected too low an impedance to them. And then the designers started thinking about how to make these things not die on simple, forehead-slapper mistakes. Today, you're pretty safe shorting any solid state power amp because they all protect against overcurrent, over-power, and short circuits. Opamps are also (today...) pretty safe with output shorts.

Some of the circuit techniques in the two pages I referenced show how. Series resistances, current shunts (diodes to dump excess current) and current limiters can be used to prevent excess current and power from damaging devices.

You'll still damage capacitors with over voltage. A 10V input cap on a pedal powered from a 9V battery is OK - until someone puts +/- 50V on the input. Some commercial pedals had all 10V pedals in the past. Gotta know what's actually in that circuit to know if it's a problem.

Another sneaky way for damage to happen is particular to bipolar inputs. If you take a nice, high gain, linear bipolar transistor and reverse-break (zener) its base emitter >even once< and as long as current isn't excessive the transistor will still work. But it will be permanently a little noisier. This is one way that equipment gets hissier as it gets older. Put a reverse-biased diode across the base-emitter and it permanently prevents this damage as well as not interfering with normal function.

Sadly, you just have to know the parts and know the circuits. Rules of thumb are not a huge help.

R.G thabk you so much for the explanation. I understand that different amps will have different requirements but i just dont understand how a Maxon 808 can output 35dbu and dont damage any amp

[R.G.]

The short answer is that it can't put out enough current. It might put out 35dbu with no load, but the amps' inputs can either withstand that much voltage or the 808 output voltage stops increasing when its max output current is exceeded.

It's been long enough since I messed with the Maxon 808 that I'd have to go look up the schematic, but I seem to remember a resistor in series with the output. This is effectively a voltage divider with any load following the circuit.

[Atodovax]

The short answer is that it can't put out enough current. It might put out 35dbu with no load, but the amps' inputs can either withstand that much voltage or the 808 output voltage stops increasing when its max output current is exceeded.

It's been long enough since I messed with the Maxon 808 that I'd have to go look up the schematic, but I seem to remember a resistor in series with the output. This is effectively a voltage divider with any load following the circuit.

Ok so maybe i was unexperienced enough to read the pedals output with no load at all. I was just reading the output jack with the osciloscope . should i plug the pedal into the amp and read that conection instead? Maybe with the amps load the voltages will decrease because i was reading more than 3 volts peak to peak on every single pedal that i tested

[PRR]

> long enough since I messed with the Maxon 808 that I'd have to go look up the schematic, but I seem to remember a resistor in series with the output.

Image I found shows MC1458 with 220 Ohms series.
https://lariklabs.files.wordpress.com/2016/07/schematic.png
The 220r does little to protect the world. The '1458 can deliver ~~25mA to a short. This is somewhat greater than many input devices are rated. The rating may be conservative. The signal may not clip 100% of the time.

I agree that self-protection has to be considered at every jack. (Not just "inputs", since all jacks look the same in the dark.)