Does voltage rating on elec cap matter?

[pereubu]

Hi again. im building a pedal kit, but one of the capacitors is missing. Its a 100uF cap, and looking at a pic of the pcb with everting installed on the website it looks to read 35V100uF. I just received an order for various capacitors etc, and I have a 16V 100uF. Will this work? Thank you.

[mcknib]

That'll do no problem with caps the voltage rating relates to it's physical size usually so anything from 16v to 63v for electros is ok, some films at 100v are usually ok but it's best to check the datasheet to get the actual size of the caps if they're over 100v just to make sure they're not to big for pcb work.

A general rule is to go for twice the power supply voltage so that it's tolerance to voltage is well covered i.e it won't heat up, burst or burn so if it's 9v, 16v will be alright it's greater than the circuit voltage albeit not quite twice the voltage

I posted a beginners guide to components PDF here recently I put together a wee while ago so here's the link, you can download it and read it at your leisure:

http://www.guitarpcb.com/PDF%20Files/A%20Beginners%20Guide%20to%20Effects%20Pedal%20Components.pdf

Mike posted whilst I was modifying my msg and looking for the link as he says voltage rating does matter especially if your building a high voltage circuit like an amp most BOM's (bill of materials) but not all will tell you the voltage ratings required

[GibsonGM]

Many times, YES, it does matter! 

But here - often we just use what is at hand, and the pedal designer may have done that (gone higher than required). 

Good rule of thumb in this ONE case...if it runs off a 9V battery or wall wart, it is likely to be fine as long as it is rated more than the power supply...not too many instances where *something* is going to connect to your stompbox and make that high a rating necessary.  With experience, you'll learn when it does or doesn't matter.

But, for safety's sake...what pedal is it?   If it's something that can connect to phantom power or something, it can be an issue.   If it's a normal stompbox, no, higher than 9 is fine (16V is very common).

[blackieNYC]

This would be a wonderful opportunity for someone to explain electrolytic capacitors used not in DC filtering but AC coupling. If the rating of a polarized electrolytic cap is 16v, that's plenty to pass a hot signal, clipper-wise, of 5volts or so - on ONE swing of the ac waveform. But that 16 volt rating is for only that swing. The other swing(we'll call it the negative) may well exceed the cap's limitations. It is exposed to reverse polarity. And it certainly can't handle 16 v.
Hopefully at this point I've hooked someone who knows more than I do, and they'll straighten us out. There are cap ratings that are indicative of how well a polarized cap will perform backwards. 105 degree caps may perform better.  Or simply a 50volt cap instead.
Tan theta. Could one of the upperclassmen take it from there?

[printer2]

This would be a wonderful opportunity for someone to explain electrolytic capacitors used not in DC filtering but AC coupling. If the rating of a polarized electrolytic cap is 16v, that's plenty to pass a hot signal, clipper-wise, of 5volts or so - on ONE swing of the ac waveform. But that 16 volt rating is for only that swing. The other swing(we'll call it the negative) may well exceed the cap's limitations. It is exposed to reverse polarity. And it certainly can't handle 16 v.
Hopefully at this point I've hooked someone who knows more than I do, and they'll straighten us out. There are cap ratings that are indicative of how well a polarized cap will perform backwards. 105 degree caps may perform better.  Or simply a 50volt cap instead.
Tan theta. Could one of the upperclassmen take it from there?

Please disregard above post and life will be easier for all of us. 

[blackieNYC]

Word.

[anotherjim]

For a coupling cap chosen to pass all frequencies used, it's impedance is very low, so there is little signal voltage developed across the cap and it's far, far less than the peak swing of the signal. I think it would be very difficult for signal to modulate the DC bias across the cap above its rating or reverse the polarity.

[blackieNYC]

Ah. The DC bias will (at 4.5) generally keep it out of reversal?  Makes sense. But I do believe that theoretically, a polarized electro has a much smaller voltage max when ac voltages (over the bias voltage?) are applied, no? Like perhaps in a bipolar supply situation?

[anotherjim]

I'm thinking that to exceed the caps breakdown voltage, that voltage has to develop across the plates.

I'll stick my neck out and try to size it...

Find the cap resistance at lowest likely frequency.
Fit that in series with the source and load impedance
Apply the highest likely peak voltage across the series string & ohms law the current.
Ohms law the voltage across the cap resistance.
Add it to the DC bias for the positive peak.
Subtract it from the bias for the negative peak.

10k source+100k load+(1uF cap @20Hz is 8k) = 118k
Signal Peak +/- 4.5v/118k=38uA
Cap V=8k*38uA=0.3V
Yes, I've ignored the ESR of the cap, but that should be <30ohm?
It barely wiggles the 4.5v bias across the cap. At 1Khz it gets too trivial to mention.

And yes, the cap remains properly polarized with the bias. I've read that sufficient bias helps against so called electro distortion that audiophiles complain about.

If you're thinking of the case of interstage coupling where both source and load have similar DC bias? You get away with it! There's still only the small voltage across the plates calculated above to worry about (plus some difference in bias) -  the cap does get reversed, but that corrects every half cycle while >1k source & load impedances keep current in the cap at a harmless low level. Does leave it open to Electro Distortion though - simple cure is 2 caps of x2 value in series, neg to neg with a high value resistor to ground off the negs. Sanity and DC cap bias is restored.
Better though to have a high enough circuit impedance that you don't need electro interstage couplers? 100nF is plenty with 100k.

[pereubu]

Ok, you've lost me! I did instal that cap, and the pedal works ok. But... there are 2 potentiometers, one for vol, and one for texture. The vol works fine, but the texture pot actually seems to decrease the volume also! Could this be due to the different cap? Or a bad item on the board? I will post the schematic in a mo.

[pereubu]

[chuckd666]

In noob words, just make sure the voltage of the cap exceeds the max voltage you'll find in the circuit. Eg, it doesn't matter what voltage rating the cap is, as long as it's  generally above what you'll find in a stompbox circuit. 16V to 35V is probably most common for regular electrolytic caps.

[pereubu]

Thank you. IF the cap is ok, is there a reason the Texture control seems to be also acting as a volume control (at minimum it has good output, then goes very low and then increases gradually the more you turn it up. I need more testing for sure also though.

[ubersam]

The vol works fine, but the texture pot actually seems to decrease the volume also! Could this be due to the different cap?

Nah, the 16V cap you indicated is your dc supply filter and isn't affecting the function of the texture pot. It's the size of the texture pot that is most likely the culprit. At full CCW, you have a 500K/100K voltage divider (ignoring the 1M pulldown). That means that if you have a 1V output, you're only getting 167mV across the 100K vol. pot.

[pereubu]

Thanks. What is the solution so? A different pot? Sorry for being an idiot on this, but its a very steep learning curve at the moment!

[ubersam]

The texture pot looks like a clean(ish) blend, taking the clean from the output of the first bipolar stage. Kinda reminds me of the VL Sparkle drive. Anyway, I'd try a 10K~50K linear pot and see.

[mcknib]

Not that I'm an expert far from it

But thinking out loud is it some kinda fuzz? it looks to me but I could be wrong like the texture pot has a possible feedback / bypass / alternative path set up going back to Q2 depending on the pot rotation so as far as I can work out if you had it CW resistance would be greatest between pot lugs 1 & 2 and less between 2 & 3 or it could be the other way round - so nothing / very little would go up or come down from C7 following the path of least resistance or alternatively it'd go through all the stages and into texture pot lug 3 out of 2 to the volume pot.

Anyway long story short looking at the wee jumper / trace at Q1's collector would that along with the texture pot decide where the path of least resistance would be and if it went up and around to C7 and out would you not get a drop in output.

So you either get full on mental gain or very little depending on which path has the least resistance.

What is the circuit and where did you get the kit from I ask because there may be no error it might be how it is supposed to be some high gain circuits with for example different diode clipping options do have a noticeable drop in volume silicon diodes aren't as loud as LED's which aren't as loud as op amp clipping and this set up does look a little different with all the above

BTW how did we get here from is this caps rating ok to use? always interesting how things unfold around here

[Phoenix]

The texture control passively mixes an out-of-phase signal into the output. The behaviour you are describing, with lowest volume in the middle of the pots range, and maximum volume at either extreme, is exactly what is to be expected. There's nothing wrong with your build, it's just a quirk of the design.

EDIT:
The out-of-phase signal will be much lower amplitude, so the setting with the most cancellation (lowest volume) will be somewhere early in the rotation of the texture knob, not in exactly the middle position.

[PRR]

Put 100V across a big 50V cap for 20 minutes. BFWOOM! Vets hit the deck. Big cloud of smoke. Bits of cap-tissue all over the inside of the chassis.

In a 9-volt world, 16V caps are "usually" quite safe.

> Please disregard above post

Why?

> The other swing...It is exposed to reverse polarity.

As Jim said, "usually" an e-cap in audio passes "all" signal, which means it has nearly-no AC voltage *across* it, and voltage reversal doesn't happen.

Also: that 100V on 50V example was 100V from wall-power, transformer, rectifier, HIGH available current. In *audio* we generally come from (except power paths) a 4.7K resistor or a 30mA opamp. Even if the voltage does reverse, the short-circuit current is too small to heat the cap, much less blow it up.

All of this gets different again when you start Horsepower motors with large 230VAC power lines. Historically we used WET electrolytics because they could break-down and heal over and over again. Film-caps were more convenient where electrolyte slop would be a problem. Recent improved dry electros have been taking over this field. But this is far outside the usual topics of this forum.

[PRR]

> texture pot actually seems to decrease the volume also!

One end of TEX is boosted and somewhat distorted.

The other end is way-boosted and over-distorted, AND out-of-phase.

There will be a point on TEX where small signals will nearly cancel-out. When played HARD, the different distortions won't quite cancel, but will interact as the note decays.

So it may be working as designed.

C8 is certainly not involved here.

It is always possible some other resistors got in the wrong places, or solder joints didn't quite happen. Voltage readings (see Debugging) may give clues.