clean power struggle

[amptramp]

I took apart a 6 VDC 700 mA wall wart on one occasion and it was interesting: there was a diode bridge with an unidentified device which is either a small ceramic cap or a varistor across the input, and a 4700 uF 16V electrolytic in series with a 1.0 ohm resistor and the output shunted by 100 K.  The capacitor would be too much of a load for the diode bridge on start-up if the series resistor was not there.  But this causes the 120 Hz fullwave rectified signal to appear at the output.  Do not expect a DC wall wart to be particularly well filtered.  Most people do not use a wall wart with 700 mA capability to power a stompbox with a couple of op amp stages and a few transistors in it.  Most stompboxes only take a few milliamps (or they would not be able to run from a 9 volt battery), so the capacitor value would be decreased proportionally.  I would prefer 9-volt stompboxes to be powered from either an internal battery or a 12 VDC wall wart with regulation of the wall wart input in the stompbox.  Most 9 VDC units will run from 8 volts or less (or the battery would not last long), and most regulators will work with less than a 4 volt dropout voltage (input to output difference below which the regulator IC drops out of regulation), so the regulator could be set to 8 volts for most applications.

You could modify the wall wart to provide the power output directly across the capacitor, making the 1 ohm resistor appear in the supply from the bridge to the cap.  With a full load, this would drop 0.7 volts for a 700 mA device, but with only a few mA the loss would be negligible.  Of course, you would lose all UL/CSA/VDE approval by opening the case, so it's up to you.

[merlinb]

The enclosure of your PSU isn't connected to mains earth in any way is it? And if so, the circuit itself isn't connected to earth, right?

[Paul Marossy]

I took apart a 6 VDC 700 mA wall wart on one occasion and it was interesting: there was a diode bridge with an unidentified device which is either a small ceramic cap or a varistor across the input, and a 4700 uF 16V electrolytic in series with a 1.0 ohm resistor and the output shunted by 100 K.  The capacitor would be too much of a load for the diode bridge on start-up if the series resistor was not there.  But this causes the 120 Hz fullwave rectified signal to appear at the output.  Do not expect a DC wall wart to be particularly well filtered.  

Yeah, most of them suck as far as using them for audio purposes is concerned. They usually have too much 120Hz noise to be useful without doing something to help them out. I find that adding a 100 ohm resistor in series with one leg of the power supply and a 1000uF cap across it usually kills that noise from a wall wart.

[El Heisenberg]

The enclosure of your PSU isn't connected to mains earth in any way is it? And if so, the circuit itself isn't connected to earth, right?

correct.


Paul, 10000 posts!!!

[El Heisenberg]

hello? *bump*

[El Heisenberg]

I really wanna try this ground lift idea, but I need to know which way to do it? Lift the ground from the power jack or from the signal jack(s)?? Please please clear this up for me? Also what about the first pedal in my chain?

[petemoore]

  Draw just the ground connections, look for the difference between an "O'' [full loop] and a ''C'' open loop.
  Wherever you have a closed loop, opening it...opens it, it can no longer be a ground loop, or have ground loop problems.
  Once the loop is broken it's not a loop/groundloop.
  Isolation transformers with 'suitable' [?] frequency bandwidth have become less expensive...
  Breaking the loop at the power supply, then converting to DC...there's no frequency bandwidth to worry about, I'd call this the 'common' or 'traditional' way if not the least worrisome or expensive.

[El Heisenberg]

but I don't really know where my "loops" are. I'm assuming because I have a ground connection at the power jack and then one at the signal jacks which are both grounded to each other then grounded to the power jack. Should there onl be a ground at the power jack or at the signal grounds?? That's what I've been asking.

[El Heisenberg]

Also how could there be any loop at the power source and THEN convert to DC? My power supply isnt grounded to chassi ground. It's just floating.

[El Heisenberg]

I've beern going through my pedal chain all day replacing input wires with shielded coaxial wire and making sure that there weren't any ground connections that weren't needed. The signal jacks grounds were connected together and then to the circuit. The circuit was then connected to the power jacks ground. This is how all my pedals were build. Two ground wires, but a third connecting both signal jacks. Anyway, the shielded wire didnt help.

[R.G.]

It's important to remember that ground loops are not the only way you can get hum.

We live in a literal electromagnetic ocean, filled with the shrieking of every device that bats an electron back and forth.

Hum can be:
- AC power line leakage through the power supply; batteries completely eliminate this, and really good AC adapters have so little it can't be heard.
- Power supply ripple; this is one octave up from the AC power line sound. Again, batteries completely eliminate this, and good adapters have inaudibly little.
- Electro-staticly coupled; This is the sound of an open input wire. No amount of tinkering with power supplies will fix this
- caused by two parts of the circuits being at different AC potentials; this is the sound of an unconnected ground wire.
- buzzy and trebly; this is the sound of AC power line coupled spikes happening one time per AC line half cycle. Fluorescent lights are infamous for this.
- magnetically coupled; this is the sound a wah makes when it's sitting on an amp over a transformer, acting like a transformer "pickup".
- magnetically coupled; this is the sound of a ground loop.

[El Heisenberg]

well, where my hum is comming from is definetly the power supply. So it's either AC power line leakage through the power supply or power supply ripple as R.G. said.



This is my set up...I was thinking of drawing pictures but nah.

It starts with a passive A/B switch. 'B' goes to the Fender tuner powered by the DC supply. 'A' goes to a rangemaster then an orange squeezer both battery powered. Then chopped Boss OC-2, Tubescreamer, Vox Wah (always on, stomp engages volume pedal mode), MXR noise gate (always on), RAT, Fat Boost, Fulltone Ultimate Octave, Big Muff, EA Tremolo, Phase 45, output buffer/splitter (always on).


Today I powered the Vox Wah/volume pedal and the MXR noise gate on batteries since they're always on. I then disconected all the other pedals from the power supply. Hum was gone of course. This is with the SPDT switch to disengage the noise gate off. So the noise gate is off. I then started plugging the pedals power in from last to first, one at a time, listening for when the hum comes back. It came back when I connected the Tubescreamer, right before the volume pedal. Since the volume pedal and the noise gate are always on, they amplify some noise that's getting to them from the power supply when pedals before are connected to it. I dunno how the noise is getting there tho. If I connect any ONE pedal to the supply before the ones that are on...I get hum...even the tuner before the A/B switch. I then powered the Wah and noise gate from the power supply, and got hum even when the pedals before them weren't connected to power supply. 


I dunno what to do. My setup is kinda useless with this hum. It's interfering with the notes. The hum is a little bit higher than a B note. I CAN'T use batteries. There's no way to turn them off and there are too many pedals. There's also no room inside my pedals. They were built for a power supply.

Please please help me figure this out! I've thought I did everything right! Shielded wire and filtering caps don't help at all!

[El Heisenberg]

I'm thinking that this IS some sort of ground loop thing. Someone said earlier that it's impossible since it's DC power only, but from what I tried just now it seems to make sense...to me at least. I don't see where I did anything wrong though. All my pedals have a connection from the circuit board to ground. The signal jacks are all grounded to the circuit boards. And that's that.  Is that a loop??

[stringsthings]

have you tried your pedal setup in other locations? ( could be a problem with the mains at your present location ) ..... it's a good excuse to visit a guitar store ....

[El Heisenberg]

There should be a guide or something out there for making sure noise doesn't happen. What's the point of building pedals if you can't use them?

[stringsthings]

There should be a guide or something out there for making sure noise doesn't happen. What's the point of building pedals if you can't use them?

are you serious?

[El Heisenberg]

Well I can't figure out what the heck it wrong! I've done everything right. There must be something I'm missing. They work fine with batteries.