Can high output voltage damage an effect or amp downstream?

[Nick C.]

I built a 32v modified Valvecaster and was wondering if the high ac output could cause damage downstream if the volume out was set too high? Let's assume that my amp is set at a reasonable volume as not to blow a speaker. Let's say the amp was solidstate or tube and the effect was true bypass or fet switched.

So would the signal just at the rail of the next devise?

Nick

[PRR]

Use the Output Volume knob; that's what it is for.

[Nick C.]

Use the Output Volume knob; that's what it is for.

That is the essence of the question, having only use 9v effects for years I wonder what possible harm may result from cranking that volume knob and potentially sending a signal 3 times what was possible before. What if it's a 16v signal? Would it just clip harmlessly or take out a cheap smd cap or chip?

[GibsonGM]

Another way is to use a voltage divider at the output to bring the level down to something 'normal', and limit the ultimate highest signal to something manageable.  You might need to treble bypass as well, there.
Then use an output volume knob.

The answer to "can too high an AC signal damage something downstream, like an effect or amp?" is YES...particularly by popping input caps, which may be rated fairly conservatively...if you have a 6.3VAC rated input cap, you can imagine what happens when you hit it with 20VAC    Not that anyone would use one rated so low, BUT --

Amps would use a higher rated cap, but you still could have problems...and if the input circuitry is solid state instead of tube, you can enter the realm of IC destruction...

[Nick C.]

Thanks Mike.

[ashcat_lt]

32V is pretty absurd unless even if you're driving studio grade line level gear.  My converters will theoretically handle almost 27V peak-to-peak, but it starts to get a bit curvy and crunchy well before that point.

What the heck are you trying to accomplish?  There's no reason to amplify a guitar that loud except to (directly) drive a speaker.

[GibsonGM]

YW, Nick.

^^ What Ashcat says....you just having trouble 'turning down' your cumulative gain?  For tube work, typically it's what I was talking about, a resistive divider with a bypass cap to keep the highs from being attenuated.  You can find out 'how much' and what each R should measure by temp. sticking a pot in there, then measuring each side when it sounds 'right'.   No way do you need those levels, LOL!!!

Now, a FEW extra volts of boost - that's another thing entirely  

[Nick C.]

I've got no trouble getting the vol turned down.

  "What the heck are you trying to accomplish?"
I was just experimenting with super saturation and compression, when I got concerned that I might blow up my Tube sound fuzz or $10 Fab echo or amp.

What, noone's seen Back to the Future;)

[R.G.]

All electronic instruments have limitations on both how far the outputs can swing and how far the inputs can go without distorting because the input can no longer make any (electronic) sense of what it's being fed.

For instance, the 12AX7 vacuum tube grid can only swing a volt or so before the input starts clipping the signal. When the grid reaches the cathode voltage, the grid turns from a near-infinite input impedance to a roughly 5K nonlinear resistor for all higher voltage. About 1.5-2V below that, the tube is cut off and nothing happens because no electrons flow.

Whether your high signal level will damage some input depends on the input *and* whatever is connected. Well designed electronics takes into account that it may be fed too-big signals and so it will make such cases non-damaging. I went into some of this at geofex, in "What are all those parts for?". Geofex isn't trendy, but it has answers for many of the questions that come up here, in articles written over a decade ago.

But I digress. You can't get a solid answer to that one, because you don't know what the circuit will drive. It could be anything with a 1/4" jack on it. There are absolutely things with 1/4" jacks that would be damaged. But you have to study the schematic and parts list of each one to know this without smoke-testing the setup. About the only thing you could do on your end is to make the output have high voltage, but also high-ish output impedance. If your signal widgie has an equivalent of a series 10K resistor on the output, it's hard to get much current to flow out. But the voltage appears undamaged if it drives another device with an input impedance of significantly over 100K.

Probably most pedals will be fine. But good, preventative circuit design is rare in pedals, very rare. So you either have to build protection into your widgit, know the circuit and sensitivities of everything with a 1/4" input jack, or just cross your fingers and hope that the problems are very rare. If you're selling this stuff, be prepared to have a very few, but very angry customers.   

[Nick C.]

R.G., nice article.

I've got a 470k plate resistor on the 2nd stage of the 12ax7, so there shouldn't be a lot of drive, right? Maybe that helps. Looks like there are a lot of variables here with various circuit designs and component limits of downstream boxes. At least I know there is a risk involved and should keep the output reasonable.

[R.G.]

Thanks.   

The output impedance of a 12AX7 plate is about 60K, which appears in parallel with the plate resistor, so you're about 53K or so. The way source impedance works is that the output appears to the following circuit to be the no-load voltage from a perfect voltage source, but through the source resistance. If you're using a 12AX7 plate, your output will be 37V swing that looks like there's a 53K resistor in series. So if you loaded the output with a 51K real resistor, the output voltage should drop by half. That's a useful test for you to make - measure the output voltage, stick a 51K (or 47K, etc.) on the output to ground and see if the voltage about cuts in half.

That will mean you're probably OK with bipolar input anythings after it, as the current is limited. JFETs, probably. MOSFETs are a distinct maybe. If a designer knew what he was doing, he would have protected a MOSFET input from voltages bigger than the MOSFET gate can stand. MOSFET gates break over at something like 12V to 20V, either polarity. When the gate is broken by voltage, the amount of current that flows through is almost immaterial. It's a tiny puncture in a glass layer that was only 20V thick to start with, so the device is toast. But not everyone who "designs" pedals knows what they're designing with, so you can't count on every device having the right input setup.

[PRR]

> 12AX7 plate is about 60K

And higher at low B+ voltage / low current.

32V out of a 100K resistor is at-most 0.32mA. As a 95% valid rule-of-risk, "most" stuff will survive over 1mA. JFET gates and most BJTs stand 10mA (though reverse breakdown could permanently raise hiss). A 12AX7 grid is not rated for "any" current, but will in fact take anything it can suck off the cathode (a few mA). Also the classic 12AX7 g-amps (but not all!) have 34K series resistors, so survive even 100V.

But you get into pedals, and especially Ge stuff, anything goes (or blows).

[Nick C.]

Ok, sounds like my Tube Sound Fuzz is my weakest link.

Hmm, I can add some LEDs to ground at the input of the TSF, adding more overdrive and protecting it at the same time...