Author Topic: 3PDT pop noise reduction  (Read 369 times)

helos

  • Diode Destroyer
  • *
  • Posts: 26
    • View Profile
3PDT pop noise reduction
« on: August 08, 2017, 03:04:13 AM »
This topic has been probably discussed many many times, I've read many articles and posts on different forums so far and I think I've figured out some things the builder should check to remove (or at least reduce) the upsetting pop noise of the 3PDT switch; I'd like to have your opinion on this (maybe I'm completely wrong):

1) Switch quality:
I've done some tests with different 3pdt switches (purchased from Aliexpress, Tayda, Alpha, Hong) and I can clearly notice that cheap switch = loud noise.
Probably due to the following reasons:
- High force required to perform the switching
- Bad bouncing performance
- Bad Internal spring
The only exception is that I've found the Alpha 3pdt to make a louder pop than the Hong 3pdt (with the same circuit), but the Hong is supposed to be a clone (and even cheaper) clone of the Alpha.

2) Pulldown resistors:
The classic 1M pulldown resistor at the input is helping but it doesn't really cancel the noise. A pulldown at the output is also good. Grounding the input in bypass mode improve the noise reduction.

3) Bypass capacitors:
Any (even small) fluttering in the supply voltage could be displayed at the output of the circuit as a little parasitic voltage, that we hear as pop when the switch suddenly change its state. Is this correct?
This is something really interesting that I've not yet tested. In all my builds I've always used 2 sided PCB (with ground plane at the bottom and power plane at the top). I want to try now to assigning Vref (half of supply voltage) to the top plane and manually route the power lines (20 mil width) to let them be as short as possible AND placing the film box type bypass capacitor (not the electrolytic ones) as close as possible to the supply pin of the active components.
This way the inductance should be reduced at minimum and the power supply filtering should be improved.

4) Optical bypass
This is another thing I do want to test: the idea of never really mechanically disconnect the signal from the circuit input could be great, and could make the switching quieter (a mechanical movement is needed only at the output)


These are just some of the consideration I'm thinking of... what's your opinion?
Thank you!  ;)

WormBoy

  • Stompbox Star
  • ***
  • Posts: 147
  • TJ
    • View Profile
Re: 3PDT pop noise reduction
« Reply #1 on: August 08, 2017, 03:27:37 AM »
I mainly use the standard 3PDT, with pulldown resistors at begin and end (if there is no path to ground already) of the circuit, grounding the input of the circuit when in bypass. Never have problems with popping (at least, not to a level that annoys me). The only problems with popping have been when using relay switching circuits ... especially those that are said to eliminate popping  ::) .... The optical systems also work very well, but they require some more wiring (they need a power supply as well) so it can make your build a bit more spaghetti-ish.

helos

  • Diode Destroyer
  • *
  • Posts: 26
    • View Profile
Re: 3PDT pop noise reduction
« Reply #2 on: August 08, 2017, 03:33:12 AM »
The only problems with popping have been when using relay switching circuits ... especially those that are said to eliminate popping  ::)

I've noticed that too... rather paradoxical isn't it?



I've also noticed a behavior that I cannot explain:
A (quite high gain) overdrive I've built pop ONLY the first time after I plug the power supply. The second time if there's a pop, it's really smaller than the first one. Why??  :o :o

WormBoy

  • Stompbox Star
  • ***
  • Posts: 147
  • TJ
    • View Profile
Re: 3PDT pop noise reduction
« Reply #3 on: August 08, 2017, 04:27:34 AM »
I've noticed that too... rather paradoxical isn't it?

The designers may not have considered/tested all possible ways that their switching system would be applied by users (without advanced electronic engineering degrees). I have a relay switching system that is really noiseless, but it mutes the signal for quite long (which probably annoys me more than a little pop). In the end, for me relay switching is a complex solution to an almost non-existent problem, but your mileage may vary.

I've also noticed a behavior that I cannot explain:
A (quite high gain) overdrive I've built pop ONLY the first time after I plug the power supply. The second time if there's a pop, it's really smaller than the first one. Why??  :o :o

I've had that too, with a pedal in one situation but could not reconstruct it later. I guess switch pop is a complex issue with a multitude of potential causes.

helos

  • Diode Destroyer
  • *
  • Posts: 26
    • View Profile
Re: 3PDT pop noise reduction
« Reply #4 on: August 08, 2017, 06:02:09 AM »
I've had that too, with a pedal in one situation but could not reconstruct it later. I guess switch pop is a complex issue with a multitude of potential causes.

Agreed. That's why I did this post, to list some of the causes (and their solutions) I've figured out.
The clickless true bypass relay system that mutes the signal (for like 40ms) is effective (you don't hear a pop) but actually I find it to be annoying and somewhat unnatural.
Muting the signal is like "taking a painkiller", you don't feel the pain, but you're not treating the desease...  ::) ::)

madbean

  • Administrator
  • Electron Doctor
  • *****
  • Posts: 7524
  • Brian
    • View Profile
I own madbeanpedals (duh). I am part owner of Function F(X).

Ralfg

  • Stompbox Star
  • ***
  • Posts: 160
  • Ralf
    • View Profile
Re: 3PDT pop noise reduction
« Reply #6 on: August 08, 2017, 08:24:31 AM »
Hmmm... a new 3pdt board option for sale at Madbean pedals?

Any ideas as to what the resistor values would be?

helos

  • Diode Destroyer
  • *
  • Posts: 26
    • View Profile
Re: 3PDT pop noise reduction
« Reply #7 on: September 01, 2017, 04:04:49 AM »
This might help: https://www.mrblackpedals.com/blogs/straight-jive/6629778-what-really-causes-switch-pop

In your experience: do you think a pop noise could be also a consequence of the PCB design? Using the power and ground planes, if the traces fragment too much the continuity of these planes, a pop noise could be induced when switching on the pedal the first time. Could it be possible?

madbean

  • Administrator
  • Electron Doctor
  • *****
  • Posts: 7524
  • Brian
    • View Profile
Re: 3PDT pop noise reduction
« Reply #8 on: September 01, 2017, 05:55:32 AM »
In your experience: do you think a pop noise could be also a consequence of the PCB design? Using the power and ground planes, if the traces fragment too much the continuity of these planes, a pop noise could be induced when switching on the pedal the first time. Could it be possible?

Can't really speak from experience on this since I never use power planes (just ground planes top and bottom). Bad PCB design could cause several issues including noise and excessive parasitic capacitance. Maybe those could contribute to switch pop. But, designing a good PCB isn't terribly hard and I think most of the DIY'ers do an excellent job (I'm hardly the authority here). I guess there are other variables, too, like power supplies, other effects in the chain...maybe even cabling. But the likely culprit is usually a tiny amount of DC getting amplified through a circuit via mechanical switching.
I own madbeanpedals (duh). I am part owner of Function F(X).

helos

  • Diode Destroyer
  • *
  • Posts: 26
    • View Profile
Re: 3PDT pop noise reduction
« Reply #9 on: September 01, 2017, 06:54:23 AM »
In your experience: do you think a pop noise could be also a consequence of the PCB design? Using the power and ground planes, if the traces fragment too much the continuity of these planes, a pop noise could be induced when switching on the pedal the first time. Could it be possible?

Can't really speak from experience on this since I never use power planes (just ground planes top and bottom). Bad PCB design could cause several issues including noise and excessive parasitic capacitance. Maybe those could contribute to switch pop. But, designing a good PCB isn't terribly hard and I think most of the DIY'ers do an excellent job (I'm hardly the authority here). I guess there are other variables, too, like power supplies, other effects in the chain...maybe even cabling. But the likely culprit is usually a tiny amount of DC getting amplified through a circuit via mechanical switching.

Thanks.. Well, it would be awesome if you guys could do some sort of PCB design techniques tutorial. I don't mean how to design a PCB with some special software and to output the gerber, I mean how to place the components, where and why.

It'd be great...

gordo

  • Electron Doctor
  • *****
  • Posts: 1122
    • View Profile
    • Project Website
Re: 3PDT pop noise reduction
« Reply #10 on: September 01, 2017, 05:46:05 PM »
+1 on the layout methodology.

I've been quite taken with both opto and relay based switching, and while 3PDT switching is hardly a deal breaker for me, I've been going thru my live rig and replacing a lot of them with CODA and 1776 switching systems where practical.

I'm watching this thread though to see what comes up.

helos

  • Diode Destroyer
  • *
  • Posts: 26
    • View Profile
Re: 3PDT pop noise reduction
« Reply #11 on: September 02, 2017, 04:35:58 AM »
+1 on the layout methodology.

I've been quite taken with both opto and relay based switching, and while 3PDT switching is hardly a deal breaker for me, I've been going thru my live rig and replacing a lot of them with CODA and 1776 switching systems where practical.

I'm watching this thread though to see what comes up.

Have you noticed some particular improvement by replacing the 3pdt of your pedals with the optical bypass systems? if so, what kind of improvements?
Many times I read "3pdt are not reliable, and THEN I choose to use the optical bypass system"... but also a 2pdt has a (fairly comparable) failure rate of a 3pdt... so where's the deal?

gordo

  • Electron Doctor
  • *****
  • Posts: 1122
    • View Profile
    • Project Website
Re: 3PDT pop noise reduction
« Reply #12 on: September 02, 2017, 08:03:08 PM »
I don't notice any real difference between opto and relay. I like the soft touch switch but perhaps just because it's different. To be honest switch failures have been pretty rare for me so it's pretty much a none issue. When I was still doing repairs I replaced a lot of them but was more likely abuse than failure. Bar bands.

Stomptown

  • Electron Doctor
  • *****
  • Posts: 2075
  • Jon
    • View Profile
Re: 3PDT pop noise reduction
« Reply #13 on: September 02, 2017, 08:13:02 PM »
The best solution I have found for phantom switch pop issues is the TH Customs (no affiliation) Uber Switch which mutes the signal while bypassing.  I've had situations where nothing else worked, including optical bypass systems, and the Uber did the trick.

WormBoy

  • Stompbox Star
  • ***
  • Posts: 147
  • TJ
    • View Profile
Re: 3PDT pop noise reduction
« Reply #14 on: September 03, 2017, 01:31:59 AM »
The best solution I have found for phantom switch pop issues is the TH Customs (no affiliation) Uber Switch which mutes the signal while bypassing.  I've had situations where nothing else worked, including optical bypass systems, and the Uber did the trick.

I tried that one once (in a Britannia) ... and it popped ... very sharp edge-of-pop, just after the mute. Seems the mute was just a little too short. Or it could be that I messed something up  ;D. (Nobody else seems to have encountered this issue, so that is a viable possibility). In the end, I bent out the pin on the chip that triggers the mute. Now there's just a tiny bit of pop, but no longer the sharp noise. Seems there was a relation between the mute and the pop ... Also used the Molten Voltage one, which has a somewhat longer mute. Doesn't pop, but the mute is annoying when switching while playing.

In the end, I personally don't like relatively complex, current-slurping switching solutions, and I hardly have pops with the traditional 3pdts. I like the switching with optocouplers though. Your mileage WILL vary  8).