Whine with power supply

[frokost]

I just built a ROG Flipster, and spent too much time debugging a horrible whine. I checked everything (as everybody always does  ) but couldn't locate the problem. So I hooked up a battery instead of my Boss power supply - poof! Whine is gone. The whine also disappears when I daisychain the Boss power from a TU-2.

I think this is strange, none of the other pedals I have built behave this way. Anybody care to explain?

[frokost]

I might add that I built it as per the perf layout on runoffgroove.com, biased the drains between 5 and 7 volts, no essential changes, replaced the volume pot with a 100K resistor to ground, added polarity protection and that's it. No bypass switch. J201 all through. When audio probing, the whine is present already at the drain of the first transistor.

[Paul Marossy]

What kind of power supply are you using? Is it one of those switching types?

[R.G.]

Boss has just released a new replacement for their old wall-wart power supply, and it's a switching type.

The one I tested whined on most pedals. This is not to say anything about them in general, just that the one I tested whined on many pedals.

[ayayay!]

Did you build it according to the layout?  Me thinks the 100uf should have been first in line, before the trimpots. 

[frokost]

It's an old one, so it's not a switching type as far as I know. This is the only pedal it happens to.

Yes, I built it according to the layout, and yes, I agree that the 100 uF should've been first. I'll try to move it and see what happens.

[frokost]

No change.

I used one ceramic capacitor, the 470pF. This seems to be somewhat microphonic. Could that be it?

[Paul Marossy]

Boss has just released a new replacement for their old wall-wart power supply, and it's a switching type.

The one I tested whined on most pedals. This is not to say anything about them in general, just that the one I tested whined on many pedals.

Hmm.... that confirms what I thought might be a possibility.

It's an old one, so it's not a switching type as far as I know. This is the only pedal it happens to.

Yes, I built it according to the layout, and yes, I agree that the 100 uF should've been first. I'll try to move it and see what happens.

It sounds like you have an oscillation happening in your circuit somewhere. What happens if you mess with the trimpot settings a little bit?  Does it affect it at all?

I had that happen to me once on a build, but it would only happen when it was in bypass mode. I changed it to a grounded circuit input scheme and the problem went away. Anyhow, it was a whining noise that would superimpose itself on the rest of the signal chain when the circuit was oscillating in bypass mode. I'm still stumped as to how that was happening. 

[frokost]

Something is definitely oscillating.. It's rather high in pitch. When I mess with the trimpots, only the volume varies. Same thing with the gain - more gain means more whine. The treble and bass pots are a little bit different. On zero, it whines, when I turn them, the whine gets a little bit less, and then it goes up again.

Still strange that this doesn't happen with a battery or with the same power supply in a daisy chain.

[Paul Marossy]

Something is definitely oscillating.. It's rather high in pitch. When I mess with the trimpots, only the volume varies. Same thing with the gain - more gain means more whine. The treble and bass pots are a little bit different. On zero, it whines, when I turn them, the whine gets a little bit less, and then it goes up again.

Still strange that this doesn't happen with a battery or with the same power supply in a daisy chain.

Hmm... Yeah, that is kind of weird. Not sure what to recommend looking at at this point.

I suppose if you keep it on a pedalboard and powered by a daisy chain, it's kind of a non-issue at that point, right? 

[earthtonesaudio]

The daisy chain makes it go away?  So, what did the daisy chain add?  Some tiny resistance in series with the + and ground power leads, and if the TU-2 is like other Boss pedals, there's probably 100uF of capacitance, about 10k of resistance, and a reverse-biased diode from + to GND.
I would recreate that, starting with the cap, inside your pedal.

[Sir H C]

Yep, some ceramic caps can be microphonic.  Loud little buggers too.

[R.G.]

The daisy chain makes it go away?  So, what did the daisy chain add?  Some tiny resistance in series with the + and ground power leads, and if the TU-2 is like other Boss pedals, there's probably 100uF of capacitance, about 10k of resistance, and a reverse-biased diode from + to GND.
I would recreate that, starting with the cap, inside your pedal.

The daisy chain adds some capacitance if it's not plugged into anything. It adds some more loading and potentially (!?) a lot more filter capacitance if it's plugged into other pedals.

Switching power supplies may be noisier under low loads.

[Sir H C]

The daisy chain makes it go away?  So, what did the daisy chain add?  Some tiny resistance in series with the + and ground power leads, and if the TU-2 is like other Boss pedals, there's probably 100uF of capacitance, about 10k of resistance, and a reverse-biased diode from + to GND.
I would recreate that, starting with the cap, inside your pedal.

The daisy chain adds some capacitance if it's not plugged into anything. It adds some more loading and potentially (!?) a lot more filter capacitance if it's plugged into other pedals.

Switching power supplies may be noisier under low loads.

Definitely if they go into pulse skipping mode dropping down the frequency to audiable.

[frokost]

I suppose if you keep it on a pedalboard and powered by a daisy chain, it's kind of a non-issue at that point, right? 

R.G.: It seems strange that I've never experienced this with any other pedal or build. Anyway, I thought about what you and earthtonesaudio said, went back to filtering and tried to tack on a 1000 uF capacitor between + and -. The whine was replaced with a much lower hum. I only tried with 100 uF earlier. By the way, the adapter is a Boss PSA-230P, about four or five years old.

Guess I'll try to add some series resistance and post back.

[frokost]

Whaddya know...

I changed the power input section to this: Power in -> diode between ground and +, 100 ohm in series and then 100 uF to ground. No noise.

Somehow I assumed that with a Boss adapter, I wouldn't need so much power filtering, especially when I haven't had any problems with power supply noise, ever. That 100 ohm is from now on a standard feature on all my future builds.

Thank you all for your help!

EDIT: By the way - I like the sound of the Flipster a lot!

[Paul Marossy]

Whaddya know...

I changed the power input section to this: Power in -> diode between ground and +, 100 ohm in series and then 100 uF to ground. No noise.

Somehow I assumed that with a Boss adapter, I wouldn't need so much power filtering, especially when I haven't had any problems with power supply noise, ever. That 100 ohm is from now on a standard feature on all my future builds.

Thank you all for your help!

EDIT: By the way - I like the sound of the Flipster a lot!

I thought about mentioning that earlier, but didn't. Yeah, the 100uF cap across the power supply with a 100 ohm series resistor does the trick. You can also do that to quiet down really noisy wall warts. In that case, I use a 1000uF filter cap with the 100 ohm series resistor.

In any case, I would have also assumed that the Boss power supply would have already been well filtered. I guess that should teach us not to asume, eh? You know what they say about assuming...