Pedalboard Isolated PSU

[TelRay]

Hello all!
I have been using a TRUETONE 1-Spot (9V 1,700 mA) + Daisy chain for a while and “tolerating” some high pitched noise coming from a EH MEMORY MAN with HAZARAI. I say tolerating, because even if I knew the noise would go away connecting the MEMORY MAN to a dedicated PSU I kept on using the Daisy Chain for simplicity (I also did the math and knew the noise shouldn’t be a current draw issue).
I have recently acquired another DELAY PEDAL, and ECHOPLEX EP103 and (particularly at our practice studio) then noise became unbearable (even with the MEMORY MAN not in the pedalboard). NOTE: I double checked with an identical pedal from a friend and got the same noise.
Sample of the noise attached as it plays through the amp (I see the main peak at 3,000 Hz and lower ones at 6,000 and 9,000)
I have been trying different Low Budget Isolated PSUs with multiple 9V 100 mA, a couple of 9V 500 mA and extra 12V and 18 V outputs. Particularly: Donner DP1 ($41) , AMUZIK ($32) and HIREE ($40)
Long story short, none of them worked. For the AMUZIK and the HIREE, if I had only the ECHOPLEX Pedal connected, the moment I connect any other pedal to the PSU the noise starts.
The best result I got was with the DONNER DP1 to which I could have connected the ECHOPLEX (to the 500 mA output), a Behringer Tremolo (30 mA), Holy Grail Reverb (105 mA), TC Electronics Viscous Vibe (100 mA), Rat Distortion (? mA) and Selah Love V3 Overdrive (100 mA)… every single one of them to an isolated 9V 100 mA output. But whenever I would connect the EH Micro POG (180 mA) to either a 100 or 500 mA output the noise would be back.
I’ve opened the Donner and AMUZIK up (not the HIREE but the inside is shown on the AMAZON photos) and the DONNER seemed to have more electrolytic capacitors (22x 100 uF 25 V + 2x 220 uF 25 V) than the AMUZIK (14x 220 uF 25V) which I think would account for better filtering. Both have a 18V 1,000 mA PSU plugged to the wall.

AMUZIK



DONNER



I still can return all these PSUs and start thinking on spending the big bucks but I have the hope that maybe the solution is as simple as increasing the value and quality of the capacitors on the DONNER (and of course voiding the warranty and al chance of returning it if it doesn’t work)

What do you guys think?

Thanks in advance!

[nickb]

It sounds like you have two problem pedals viz Memory Man and the Echoplex. Let's focus on the Echoplex for now. I understand that that on its own there's no issue but when you use any other pedal there is. So, pick one of the other pedals are try these configurations:

1) Guitar -> Echoplex -> Other -> Amp
2) Guitar -> Other-> Echoplex -> Amp
3) Guitar -> Echoplex -> Amp. 'other' is still connected to the power supply
4) Guitar -> other-> Amp. 'Echoplex' is still connected to the power supply

Hopefully that will give us some clues. Also we don't know just how isolated these power supplies really are. They might be isolated at DC but yet have a low impedance at 3Khz.

[TelRay]

thank you Nickb!
i will pop by the studio later today, try the different configurations and report back

[galaxiex]

My experience with "some" digital pedals has been that they *absolutely require* their OWN dedicated power supply.

Nothing else works. Any other setup and they inject noise.

[TelRay]

let’s hope for a solution
a(n) (extremely) good isolated PSU should be “as” if the troubled pedal had a dedicated one

[TelRay]

as requested:

1) Guitar -> Echoplex -> Other -> Amp




2) Guitar -> Other-> Echoplex -> Amp



3) Guitar -> Echoplex -> Amp. 'other' is still connected to the power supply



4) Guitar -> other-> Amp. 'Echoplex' is still connected to the power supply

[TelRay]

to add to the previous post, this is the current configuration of pedals i have (GTR - POG - SCARLETT OD - RAT - VISCOUS VIBE - HOLY GRAIL - ECHOPLEX - TREMOLO) with 3 different PSUs:

1) 1-SPOT + Daisy Chain
the noisiest configuration



2) AMUZIK (each pedal to an individual 100 mA out and the ECHOPLEX to the 500 mA)
Reduces the 3KHz noise by 6 dB (also the 6 and 9 KHz)



3) DONNER DP1(each pedal to an individual 100 mA out and the ECHOPLEX to the 500 mA)
Maybe 1 dB noisier than the AMUZIK at 3 KHz
Same as AMUZIK at 6 KHz
Lower than AMUZIK at 9KHz
Much better at 18 KHz (which should not be audible in any case but maybe an indication of filtering efficiency)

[nickb]

Nice data. Thx.

The problem only happens when both pedals are in use and therefore forming a ground loop at the frequencies of interest. Why it's 3HKz I have no idea as it seems too low for either a digital pedal or the power supply (they use switch mode). Besides if it were the power supply I'd expect it to be different between power supplies. I'll sleep on it and see I get any ideas.

In the meantime, I wonder if putting a common mode) choke in the power lead to the Echoplex in order to break the loop might help. I would use an old torioid core for a small power transformer and wind the power lead for the Echoplex ten times through it and see if it helps. Of course you probably won't have one so you're back to using a separate isolated power supply. It does seem these power supplies have poor isolation above a few hundred Hz.

[TelRay]

thank you for the reply!

"problem only happens when both pedals are in use"... yes, with the DONNER DP1 PSU, with any other PSU the problem with the ECHOPLEX starts as soon as any other pedal is connected

"toroid core for a small power transformer and wind the power lead"... isn't that what the DONNER guys are doing internally (having a toroid for every individual output)?



or maybe they are running both wires around the toroid (I see two colors)

[galaxiex]

On one of the "problem" digital pedals I have, I tried a toroid, and a homemade Pi filter to no avail.
Played around with cap values and different coils.

To be fair, it did cut the noise somewhat, but not enough to be acceptable.
Plus my homemade Pi filter was kinda bulky and unsightly, just more hassle than it was worth.

I caved and gave the pedal it's own supply.

Hope you find more success than I had....

[TelRay]

thank you galaxlex,
so far I am arriving to the same conclusion. As per the graphs on post #7 using and isolated PSU reduces the noise considerably but does not completely kill it.
let's keep digging to see how far can we get. previous experiences in this have proven to be very worthy

[nickb]

Those do look like common mode chokes ( see later*) but I expect they are not doing much at 3KHz judging by the size of them. By using the big toriod I was thinking of two things. One inductance at low frequencies and core losses at higher ones to kill the noise. I wonder is someone makes one off the shelf with a DC plug and socket on it.

*PS: Now that I look at it closer I think those are the output switching transformers as I don't see any others and there needs to be one per output. If that is the case then I cannot see anything other than capacitive filtering on the output. It's not unusual to have a small capacitor between the input and output to meet EMC requirements and that may the path that is causing the problem. The arrangement will work fine to suppress hum loops but will reduce in effectiveness as the frequency goes up. Still this is all rather speculative. I'd like to have one in front of me to see what is really there and to measure it.

[TelRay]

i think I’ve seen this “power cable wrapped around a toroid” thing in some power supply. I must have one somewhere. Though I’m not sure the size will allow me to do 10 wraps.
I’ll see if i can come
up with something

[nickb]

I suspect you might be thinking of those clip on ferrities? They won't do much at 3KHz. Think more like power transformer core. If you have a busted toroidal power transformer you can remove the copper and salvage the core, sort of 30VA kind of size. Put on as many turns as you can get.

Out of interest, I have a 1000VA one that I use in the power feed to my bench with the purpose of reducing the current in the safety earth loop that can get formed when making tests that involve my computer (audio card). It gives me 40dB attenuation at 50hz. It doesn't need to be 1000VA but it gave me lots of winding area.

Seems easier just to go with a real isolated PSU,

[TelRay]

i know those clip on ferrite things but no, i am sure i've seen a small toroidal bar with the PSU cable wrapped around it. maybe i'll find it somewhere
in the meantime i have disassembled a couple of old Laptop PSUs and found these in one of them... what do you think?