Power supply noise when the effect is switched off

[pak420]

Hello all:

I'm brand new to this board.  I have tinkered with electronic projects over the years.  This is the first time that I built a design based on a schematic, made a few changes based on parts on hand, and tested on a test board to ensure everything worked as intended.  Then, I put my project (a boost pedal) in a box, added a DC jack and filtering, and plugged it in solely with a guitar and amp.  There is an incredible amount of buzz with the effect switched off (3PDT true bypass).  The input of the effect board is grounded when the switch is in the off position. 

I don't have room for a battery in the project box, but I connected one to the DC jack, and the only noise is from my pickups.  I'm using a brand new 1Spot for power.  I have used other pedals, including a Distortion Limited 001 kit that I built, and they are all dead quiet with this supply.  Even if I unplug the other effects from the daisy chain, once the DC goes into the power jack, I get the noise.

I have reflowed all solder joints, rewired grounding to ensure no loops and a single grounding point, pulled the transistor out of the socket, removed the LED, and I can't figure out how to stop the buzzy noise just from plugging in what should be a clean supply.  Any thoughts are greatly appreciated from this knowledgeable resource.

[antonis]

I built a design based on a schematic, made a few changes based on parts on hand, and tested on a test board

How remarkable lucidity..!! 

P.S.
Welcome..

[Rob Strand]

At first I suspected a wiring error in bypass mode or oscillations caused by the wiring.  However the fact the noise disappears with an externally connected battery is strong evidence the cause of the noise is the power supply.

The 1Spot  power supplies claim to be quiet.   They look like a switch-mode type  so you can't trust they are going to be quiet under all conditions.

One big problem with switch-modes is they misbehave with light loads.   Many DIY guitar pedals present light loads to the power supply, especially the vintage ones.   I would recommend trying to connect a 1k resistor across the power rails.   Ideally this should be done on the effect PCB but for the test it's best to place the resistor at the DC input Jack.   If that doesn't work you might try a lower resistor say 220 ohm.    You can go lower but the resistors are going to get hot.  If you hear a change in the character of the noise you might be on the right track.

There's many other possibilities but I'd try that first.

[pak420]

Thank you for the welcome, antonis!

Rob - I appreciate your response and guidance.  I will give the resistor a shot.  It should be easy to test....I'll report back the result.

[pak420]

I tried the 1k resistor across the power supply rails and it did not alter the buzz.  I then tried a 200R and, sadly, there was still no change in the noise. 

Are there other ideas?  I did use the debugging page already, and used the things that seemed to apply, and I just don't have enough experience to know where else to look. 

Thank you all.

[Rob Strand]

Are there other ideas?  I did use the debugging page already, and used the things that seemed to apply, and I just don't have enough experience to know where else to look.

OK cool.   These problems are just a process of elimination.   If one idea doesn't work you just keep moving down the list.

The next on the list would  be something relating to the grounds.

Are you sure that the grounds of both you input and output sockets are connecting through.    Sometimes touching a mains device onto a floating circuit will cause buzz.

Another ground test is to power the unit from battery then connect one lead of the power supply (the negative lead) to the circuit ground.   What this does is check the power supply isn't putting signal into the ground.

One question, when you use the power and are in effects mode, is there no noise at all?   and are you happy the unit is working correctly in effects mode?

It would help if you posted a schematic showing the filtering you added.

[pak420]

Thanks for the follow up, Rob!

This is just a boost pedal, and, as you might expect, noise that goes into a booster gets boosted, so the effect only works right when battery powered.

The filtering I put in line was a 100R in series with the 9V+, followed by two 22uF electro caps in parallel (I didn't have a 47UF on hand, so I did this for now) and a diode (1N4003 that I had on hand) between the rails.  The thing is that I added these after  the fact hoping they might change or remove the noise.

I have confirmed that my grounding is sound and goes all the way through.  I'd like to try your suggestion of powering with a battery and connecting the negative of the power supply to the the circuit ground.  I have some questions about this to be sure I understand.  First, am I connecting the negative lead of the DC supply (which would not be plugged into the pedal), or tapping off of the negative lead of the battery to connect to ground of the circuit?  Also, what would expected outcomes be?  Am I looking to see if this stops the noise or alters it in some way as an indicator that power is going to ground? 

I can try to sketch out a legible copy of the overall circuit and add the filtering I've put in if it would be helpful.

Thank you for the help!!! 

[antonis]

First, am I connecting the negative lead of the DC supply (which would not be plugged into the pedal), or tapping off of the negative lead of the battery to connect to ground of the circuit?  Also, what would expected outcomes be?

Connect battery to circuit and leave it there..
Connect power supply GND (negative lead) to circuit GND..
If it's still "quiet", your PS is "clean" from noise injection..

[Rob Strand]

Still continue with antonis's clarifications.

Are you sure the power is getting through to the circuit?  Can you actually measure 9V at the PCB and get boost happening?   If not, a power supply short to ground might cause noise.

[deadastronaut]

'I can try to sketch out a legible copy of the overall circuit and add the filtering'

^^^ this will help. 

[antonis]

Maybe I'm confused a bit but didn't OP said that buzz occurs only on by-pass mode..??

[Rob Strand]

Maybe I'm confused a bit but didn't OP said that buzz occurs only on by-pass mode..??

I'm a little bit confused as well that's why I'm asking the question.   I don't think he means *only* in  bypass mode.
Later on he implies it gets worse in boost/effect mode.  If there's no sound with the PSU it's a different problem to
sound + noise in effects mode.   Obviously in the second case the effect is working and there's power
getting through.

[pak420]

I'm very appreciative of everyone's input.

I'm working right now (shhhh) so haven't had a chance to try the test initially described by Rob, but will do (hopefully) this evening.  I will also measure to be sure there are 9V+ getting to the effect board.

To clarify, the noise is there regardless of whether or not the effect is switched on.  It is louder with the effect switched on.

I'll also try to work up the schematic and upload a pic of it.

[pak420]

Here's my sketch of the schematic:

[antonis]

Take reverse polarity protection diode before RC filter..

P.S.
I hope that transistor ISN'T a BJT..

[iainpunk]

i bet its a bad ground connection, re-flow the solder on all the ground connections, that might help

Take reverse polarity protection diode before RC filter..

P.S.
I hope that transistor ISN'T a BJT..

i'd say he done it right, after the resistor so the resistor acts as a fuse.
it might work if its a Ge transistor with sufficient leakage.

cheers, Iain

[antonis]

i'd say he done it right, after the resistor so the resistor acts as a fuse.

I'd say you never heard the dictum: "Semiconductors are for fuse protection".. 

Iain, transistors are by far faster from resistors or fuses in blowing up..

[Rob Strand]

i'd say he done it right, after the resistor so the resistor acts as a fuse.

Agreed, if you are going to put in a resistor you might as well use it to limit the current.
It might not even fuse it could just get hot.

[duck_arse]

how come 500k pot for volume? why so large?

what we like to ask for is photos of what you've built, so we can see your parts layed out as in your schematic.

[pak420]

OK, so I was able to use battery power test as Rob suggested, and there was no change.  I measured and have just a hair over +9V on the power rail, and, the effect does switch on and amplify.

Other observations:  If I turn on the amp and there is a cable plugged into the input of the amp with nothing connected on the other end (pedal, guitar, etc.) then there is a buzz.  To my ear it is the same as the buzz I'm hearing with this issue.  When I plug the cable into the output of the pedal, the noise decreases considerably.  It is still audible, but barely.  There is no change when I plug the guitar cable into the input of the pedal (with the guitar on the other end of the cable).  When I add the DC supply to the pedal, the buzz gets louder, as though it is the same noise amplified.  I can't be sure, of course, and don't have a scope to measure.

duck_arse - The original schematic called for the 500K pot.  I'll work on getting a snap of the layout/internals for you all to see.  Thanks!