How do you fix the "Vol Knob at 0 hum" with PNP Fuzz Pedals?

[GuitarMatt]

So I've built several PNP fuzz face style circuits and they have all had a loud hum when my GUITAR'S volume knob is at zero (not when the pedal's volume knob is at zero).  It's not a noise that simply increases inversely proportional to the output position, it's a loud hum when the pedal is ON and the guitar's output is grounded.

Has anyone else noticed this? Is there a way to filter out this phenomenon without severely changing the tone / freq response of the circuit?

Bonus if you can explain physically why this phenomenon happens. I'm assuming it has to do with the "positive ground" nature of PNP based circuits, but I can't explain it. 

[Rob Strand]

You need to put a large bypass cap across the supply, say 100uF to 220uF.

Next thing to try is an RC filter on the power rail.

*** don't forget to get the cap polarity right.

ckt1


if that doesn't display try, ckt2


Use 100uF to 220uF and perhaps 100 ohm.
Increasing the resistor value helps remove more hum but you get more DC voltage drop across it.
It's best to keep the DC drop across the resistor below 0.5V.

[Elektrojänis]

NPN fuzz I use on bass used to do that too. I use battery to power it an I have large cap across the power and ground.  (100µF if my notes are correct, the unit is at the rehearsal space.) No resistor for RC-filtering the power though. It also hummed a bit on maximum volume on the instrument (more than on mid settings). The circuit is based on Fuzz Face.

I solved it by putting a 2.2kOhm resistor in series with the input. The sound might have changed a tiny bit, but I actually liked it better. Any hum present stopped changing at all when changing the volume on the instrument.

Some of my earlier fuzzes not only had hum when the instrument volume was zero, but also recieved radio stations. Those were built with silicon transistors with highish gain and my circuit layouts on strip-board and soldering were definitely worse than my current stuff. I suspect that FF-style circuits tend to oscillate at frequencies over the audio band when you ground the input. I think grounding the input will also ground the feedback (providedby the typically 100k resistor coming from the emitter of Q2). Oscillation combined with nonlinearities could tune and demodulate all sorts of signals including radio. And if the grounding of the feedback doesn't make the circuit oscillate, it should make the gain higher and amplify any hum present. Well that's my quess anyway. The regulars on the forum with more EE-background may correct me if I'm way off.

[pinkjimiphoton]

ugh.
as suggested, a small resistance in series with the + rail. usually 220r-1k will do.

BIG ASS GIGANTIC CAP across the rails. i've had to use up to 2000uf with some q's.

a smallish cap across the rails... try 2.2n-4.7n. even bigger maybe.

make sure all your grounds are good, and fanned. in this case, i generally will ground the volume pot directly to the power inlet jack. that seems to help often.

be sure to use a scheme that will ground the input of the pcb when in bypass, cuz oscillating moose noises can leak thru the ground plain and sneak into your bypassed tone.

not a big fan of having to deal with this crap.

i remember my first fuzzface. it sounded great... until you turned your guitar down and it became a mating buffallo simultator.
the last 10th of a turn seemed to be a synthesizer, the off would just make it

MEWLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLL[/b][/size]

i had to use a 2200u cap to get that one to stfu

[zbt]

when breadboarding I used biscuit tin can

earth ground ?

https://sound-au.com/tcaas/jlhcapmult.htm



cap multiplier, works for me

last time I used inductor (4.4mH coil) in front of base Q1, series with C input
much better but mine still like morse code simulator, I guess is different channel from uncle Jimi 

[pinkjimiphoton]

brilliance!

i am but an egg

a dismally slow monkey

with breadboard

heavy stupidity

and little knowledge

of a 10 toed freak

lol

i think thats worser haiku than it is poetry, but fairly accurate in my case.

i godda watch da video cuz i is and even am kinda confused how/why this would work? but it looks coo

please hip unca jimi cuz.... umm... see above   

the wire jumper trick works great thats why i suggest making the earth ground for the pedal the ground on the power jack.

i disremember many years ago playing this bigger juke here called the palace.
i was getting fried. i mean electronically, not the more usual for me way.
i tell my brother jaybird the sound guy i'm getting zapped. he's like there is no way i am getting zapped.

"give me the guitar cord" saith he.

so i rendered it unto him. he went to grab a mic stand.

suddenly "turn off the guitar amp"...

why?

i can't let go... lol.

so to fix it, we took a g string outta my gig bag, wrapped one end around the input jack to my amp, and the other to one of the microphones.

i expected a minor explosion, but it did work, no more electrocution, and yes, less noise.. at the time i was running one of them 12 watt marshall combos into a randall rg120 double stack, with a boss me5 for fx. so it was a ....uummmmmm rather loud, noisy but TOAN-FULL system, back in them early 90's.... so it was a win-win in that respect

[Rob Strand]

earth ground ?

That Roland amp has a weird design regarding the grounds.

The amp uses a transformer but as far as I can see from the service manual there is no mains ground connected to the chassis.  I don't know how they got is through safety testing because that set-up normally requires double insulation (of everything related to mains). 

The chassis is connected to digital ground and analog ground but there's no path back to mains earth.   There's also a stack of caps from analog ground to digital ground and digital ground to chassis.   It's quite complicated and after all that it seems to promote noise getting into the signal lines.

Most amps have the chassis connecting to mains earth and that's pretty much the solution in the video for the Roland Cube 40 XL.    Incidentally, the amp is the place for the mains earth.  You don't need to add another mains earth to the pedals, that will cause a hum loop, the pedals get their mains earth through the amp.

So probably not a good example for pedals.    It's a good example of how hard it is to design floating ground devices which isn't noisy.

[anotherjim]

This is one example where there's a world of difference depending on how the guitar volume controls work. If I turn my Strats volume to zero, it grounds the output. If I turn my SG volumes to zero, it grounds the pickup and the output changes to just the pot resistance to ground (Norlins weird 300k) but my SG has independent volume wiring. This still gets noisy since the high resistance on the line picks up noise more than the 7k or so of the pickups. Not all SGs, LPs etc have independent volumes and placing those to zero will ground the output.

I like Elektrojänis' explanation of the problem. Makes sense to me.

[zbt]

Cheers Uncle, Your journey experience is very valuable
I would dare to try your old school way


@Rob Strand, Sir, I feel the way on the video is not safe, but it work,
and I though at least must connect to earth ground and one point only,
but where is the safe point?



Clean the environment

http://geofex.com/article_folders/oldspyder/oldspyder.htm
it solve by floating power supply for pedalboard

for single amp could be work at amp, what if we have stack of amp?
same as pedal, amp also float

http://www.geofex.com/FX_images/TransformerSplitter.pdf
using transformer (kind expensive)

http://www.geofex.com/FX_images/Onboard_Preamp.pdf
Maybe at first pedal like buffer, with bypass


Also curious if just earthing guitar string and body only would it be work?
add another cable or using stereo cable, but it make another problem
What if we count cable capacitance, would it be in series with C input
Adding cap to base to ground like tonebender MK2,
or miller cap in Q1 like one knob fuzz.


And as Sir Petri using battery which I belive the best hum free supply, also still make noise?
yup I agree Sir Jim, then the RF would be parallel with 2K2, if I am not mistaken



Now I see, as I experience biasing fuzz face, the backup RF could be work by adding 2K2 resistor.


I also overwhelming cause this circuit make me nuts, but that is part of the fun. 

Thank You, Thank You, Thank You.

[Rob Strand]

@Rob Strand, Sir, I feel the way on the video is not safe, but it work,
and I though at least must connect to earth ground and one point only,
but where is the safe point?

It's an *extremely* hazardous way to do it.  Imagine if the added wire gets loose and touches the mains active pin at the plug.  The whole amplifier and guitar is live.  It's a good recipe for how to die.

Having at least one earth connection in an audio system is generally a good way to prevent noise and buzz getting in.  Especially in single-ended audio systems like guitar equipment.    It's very rare for it not to help.   The problems start when you get multiple connections to the earth, for example hum loops (some of those connections aren't so easy to see with modern switch-mode power supplies).

What you have drawn in your pic will work fine.

A common way to add an earth connection is via a 10 ohm resistor in parallel with a 100n resistor.  What that does is connect the system to earth and low frequencies and at RF but tries to prevent hum loops if you later connect an earthed device.   See figure 5 in the link.     The network can be placed at any point in the system.  You can even have multiple points with these networks without creating a significant hum loop.

https://www.sound-au.com/earthing.htm

it solve by floating power supply for pedalboard

You use those in order to prevent hum loops.     (It also stops this: For old style pedal power supplies with just a rectifier and filter the current pulses on the filter cap can cause hum when connecting to multiple pedals.   It's not really a classical hum loop problem it's got an evil cause.)

Also curious if just earthing guitar string and body only would it be work?
add another cable or using stereo cable, but it make another problem

For hum and buzz it often doesn't matter where you put the single earth connection.   For RF related problems it can get more complicated.   An earth at the guitar is just another point.  It's inconvenient though.   Earthing the guitar body won't do much it needs to be a low impedance/wired connection to ground.

[zbt]

A good warning for everybody 

It's an *extremely* hazardous way to do it.  Imagine if the added wire gets loose and touches the mains active pin at the plug.  The whole amplifier and guitar is live.  It's a good recipe for how to die.

I guess It might be useful as probe, but also warning not point to live wire it can be short circuit, use Test light first.

Thank You Sir

[zbt]

hack with computer cable



cut L and N wire, and seal it, check again make sure the connection, plug to same outlet with amp.

hope it much safer