DIY Modular Synth Problem - Voltage between power supply grounds

I have a technical question regarding a modular synthesizer I’m building. I’m using the Kosmo format for my modules, and I’m using two Frequency Central Power Supplies for +12/-12/Gnd. The two power supplies are the same model built with the same parts, both driven by the same external power converter (120VAC to 12VAC).

The problem I have is that when I have modules plugged into both power supplies, there’s a short circuit that fries one of the ground wires. From my own investigation I found that there is a potential difference of around +/-8mV (it fluctuates, but who really knows with my crappy multimeter) between the two power supplies’ grounds (note this is the same whether or not modules are plugged into the power supplies). So because each module’s face plate is grounded and the face plates of all the modules touch, the two grounds are connected and it creates the short circuit. My question is this: is there a way to make the ground from both power supplies equal, and if not, is there a way to make sure that things still work properly with this voltage across the grounds? (Also, when the modules’ faceplates are not touching and I tap them together, there’s a noticeable spark, which my intuition tells me means there’s more than 8mV, but I don’t really know)

My first idea was to place a very small resistor between the modules of one of the power supplies and the ground connection to the power supply, which would mean it doesn’t create a short circuit because there’s a load in between, and I think a 10R 1/4W resistor would work (P=V^2/R with V=0.008V and R = 10Ω means P = 0.0000064W, much less than the 0.25W that the resistor is rated for), but I think that would mess up the modules it’s connected to (discharging capacitors would take longer and stuff like that). But then again, if the two grounds are connected, couldn’t those capacitors discharge through the other ground?

Basically I’m out of my depth and would love some advice. Thanks!

Yeah, this doesn’t make sense to me. 8 mV is not going to create sparks, nor enough power to “fry a ground wire” (though I’m not quite sure what you mean by that). You sure the panels are connected to ground and not to a voltage rail? Have you directly measured voltage between two front panels?

You say both FC boards are powered by the same 12 VAC source. Are you sure they are connected in parallel, with each 12 VAC terminal on one connected to the corresponding one on the other? One of the 12 VAC terminals on each board is connected to ground, so cross connecting them would mean ground on one is connected to “hot” on the other which would be a bad thing. But if they’re not crossed, that means the grounds of both are connected and there should be no voltage difference between them. Hm, when you measured 8 mV difference was that on the DC voltage setting of your multimeter? Did you try the AC voltage setting? If they’re cross connected I can imagine you’d see little voltage difference on DC but a big difference on AC.

3 Likes

Thanks for the feedback! I was messing around with things and noticed something. At the time I thought it would be inconsequential, but I found the problem. I had flipped the wires into one of the power supplies, so it was getting an inverse sine wave. My thinking at the time was that this wouldn’t matter because they’re both outputting DC voltage, but I guess not. Anyway, I flipped the wires and now everything works perfect! So problem solved.

2 Likes

Now, if that fried the wire… are you sure they are thick enough ?
Sure it saved all your gear, but still… that’s not the job of wires !