Q: purpose of output cap on MuTron

[loauc]

What's the purpose of the non-polarized output cap in a MuTron III? As I see it, it can be used for decoupling the DC signal coming from the filter, but it's value is large and it's non-polarized. That's making me wonder what is its purpose.

Thanks,
Loauc.

[R.G.]

The Mutron has a small DC offset, the result of its opamps not being perfect. The signal can be bigger than the DC offset. It is bad practice to ever let an electrolytic capacitor be reverse biased, even momentarily by signal voltages. Since the  signal can be bigger than the offset, the capacitor is necessary to prevent the DC from coming out and it's necessarily non polar because the signal can be bigger than the offset. It has a secondary function of protecting the opamps from high currents if the output of the unit is connected to an amplifier that has a DC voltage on its input. That does happen sometimes.

[loauc]

Thanks R.G.

I've still don't understand why such a large capacitor is used. Why a smaller cap can't stop that DC offset? In multi-stage amplifiers, each stage is coupled via a ceramic cap around 0.1uF

What the new MuTron sound-alikes use? Q-tron / Q-tron+ for example.

Regards,
Loauc.

[R.G.]

Well, first off  you have to recognize that only the person who originally chose that value really knows why it's there. I'm only making guesses based on my experience with both circuits and other circuit designers' mental processes, what little they've explained to me over the years. I personally might have done it differently.

As to a smaller capacitor stopping DC, sure, it will. Let's make it 1pF. What's wrong with that? Stops DC just fine.

The problem is, of course, that a 1pF cap will only pass the audio spectrum into a 7.9 GIGAohm resistor. The smaller the cap is, the more a following input impedance must be.  A 1nF cap will pass the audio spectrum into a 7.9 megohm resistor A 100nF cap will pass the audio spectrum into a 790K resistor. 1uF is good into 79K, 10uF is good into 7.9K. And this gets us to the real question - what's the input impedance of the following equipment?

We have no clue. Could be a 1M guitar amp input. Could be a 1K Fuzz Face input. So a designer drops in whatever they think is good enough for whatever they think the equipment will encounter in most cases, and whatever the bean counters will let them use. Sometimes they flip in something from way out in left field for fun.

And that's as close to why as we can get without knowing exactly what follows the circuit.

[markphaser]

I think that output cap. does 3 things it sets the frequency response to allow a range of frequencys through the output , sets the output impedance and Blocks DC offset to give a AC output.

[R.G.]

it sets the frequency response to allow a range of frequencys through the output , sets the output impedance and Blocks DC offset to give a AC output.

As I noted, it does set the output frequency low end rolloff; but it does it only when you can specify the input impedance that it supplies. Without knowing whether it's driving a 1K, 10K, 100K, 1M or 100M input resistance, you can't say anything about the frequency rolloff. All you can say is that if the following input impedance is bigger than X then the low frequency rolloff will be no higher than Y.

And it cannot set the output impedance. In fact, the capacitance is chosen so that it is insignificant compared to the real output impedance of the circuit, so it doesn't get in the way and reduce signal level.

But yes, it is definitely there to block DC - if any.

[markphaser]

Thanks for the information

output frequency low end rolloff; but it does it only when you can specify the input impedance that it supplies. Without knowing whether it's driving a 1K, 10K, 100K, 1M or 100M input resistance


Most people use a Buffer to match the impendance on a output circuit to another input circuit
1K, 10K, 100K, 1M or 100M input resistance mostly reactance is impedance

But what about matching the output frequency to another input frequency? matching the frequency caps on the outputs and inputs
Because every guitar pedal have a output cap and a input cap that rolloff the frequency but how to match the input and output frequencys

[R.G.]

Most people use a Buffer to match the impendance on a output circuit to another input circuit

True.

However, do you know what the guitarist is going to hook up next? What did that designer do - thirty or more years ago? That's the idea behind making it relatively big. How big is big enough? No way to tell. As a designer, you guess. Betting on what most customers will do is what gets an engineer a transfer to system test.

1K, 10K, 100K, 1M or 100M input resistance mostly reactance is impedance

I'm sorry - I could not decipher that. All reactances are impedances. Impedance is the more general term, encompassing combinations of resistance and both inductive and capacitive reactances. But I can't figure out what you meant.

But what about matching the output frequency to another input frequency? matching the frequency caps on the outputs and inputs

Matching to what? The fundamental problem with defining output cap rolloffs is you have **NO IDEA** what will be plugged into that output. So you guess. You say to yourself "I guess most effects have an input impedance of over 10K. I'll make the output cap work good into all impedances of 10K or more." And you set the rolloff of the output cap to 10K by C = 1/(2*pi*F*R) where F is the lowest frequency you want to pass (guitars only go down to around 75 Hz, basses to maybe 32 even in drop tuning) and R is the input impedance that you guessed at. No matching involved.

The designer of the input device that you don't know about and *can't* know about has to make his own guesses about what YOUR output impedance is. And he chooses his input cap for passing all the frequencies that he wants to pass. That may or may not be the same as you decided for your design. Remember, the designer of whatever input you're hypothesizing driving may have done his design decades ago and be dead now.

There's no matching involved. It's nice if everyone plays fair and lets at least low E (82Hz) through, but not all of them do. What matching?

Because every guitar pedal have a output cap and a input cap that rolloff the frequency but how to match the input and output frequencys

On the input you have the knowledge to control what input rolloff happens, because you know (or should know!) your own input impedance, so you can set that one as you like. You have no idea what your output will drive, but you typically set that one to drive quite low impedances and quite low frequencies so you don't lose any of the pearls your circuit has made. You may not want the input and output low frequency rolloffs matched. They did not want it so in the Tube Screamer, for good and present design reasons. Again - what matching? Common US Engelish usage of the term "matching" implies that there is some advantage to there being a match. There may or may not be. Whomever is designing the circuit had better have it pretty clear in his head what he wants the frequency pass bands to be.

Or just be lucky if he's clueless, I guess.

[Bernardduur]

Imagine I did not even see it was non-polarised; unit works great though on guitar; on bass I thought it needed some work. Now I changed the polarity to non-polar (as in two polarised 10 µf combined at the (-)) it works like heaven on bass.....