Mystery voltage on ground

[Potato Cakes]

Hello, Everyone,

I recently finished successfully building a UA176, then used the knowledge gained from mistakes I had made and implemented them into building what should have been a successful second unit. After chasing down solder joints and hardware possibly shorted things to ground, I still get 2.3VAC between ground and either + or - on the output XLR. There is no voltage between + and - themselves. This voltage translates to buzzing and hum together on the output. I unassembled everything to see when the voltage would go away, but it never did. I now have completely disassembled the wiring from the two main PCB and even pulled the AC hot and neutral going into the unit, but even with just the ground connected while the AC cable is plugged into power I am getting about 0.7VAC between the same pins as before on the output XLR. There are no shorts or measurable resistance between any of the conductors on the AC cable. I am about to start the process of going through all of the parts again to see if anything new reveals itself (this will be my third time doing this process), but I wanted to know is how voltage could be transmitting to ground and increasing as components are being removed? Am I chasing false readings? Compared to the other unit that is working properly, under the same test and power conditions has 0.06VAC between pin + and ground on the output XLR upon power up and 0.02VAC with it powered off but with the AC cable still connected to power. I thought I had the rare occasion of having a bad power transformer, but all of voltages in the unit when assembled are correct and the issue is still present with it fully disconnected. This problem is also the same with all of the tubes disconnected. The 2.3VAC voltage I am getting when completely assembled is immediate upon power up.

I do not intend for this discussion to be how to troubleshoot a 176 but more for a general understanding to be applied to similar situations when building or repairing gear. However, if you do happen to have some insight regarding this circuit and the aforementioned issues I would gladly welcome that information.

Thanks!

Paul

 

[cyrano]

I can think of two things:

- a leaky power transformer. Got another one on hand?

- AC injection via the mains ground connection. Especially if you're in an area with sandy soil and if it's very dry out there. The ground rod(s) sit in a near perfect isolator and some part of your mains has a slight leak, going everywhere because it can't drain into the earth. Break the mains ground connection, of course just for the test.

I've seen a studio go broke because their gear went wild. Usually on a Wednesday or Thursday.

It wasn't until I checked Google maps to see what was in the area that might explain this strange interference. Behind the building was a railroad shunting station. There was only one shunting loc left and that ran on 550 VDC. It only got used once in a while and when it passed, just a few meters from the building, the needles on the meters of most gear moved. The only thing they could do was keeping the ground rod wet. That only lasted till one of the other tenants complained about water in the basement. By then it was too late. The studio's clients had left...  :

 

[scott2000]

under the same test and power conditions

Exactly the same set up? Same outlets,location, etc??

 

[Potato Cakes]

Exactly the same set up? Same outlets,location, etc??

Yes. Outlet power is not the issue. I get the same results with regular wall power in my garage and with transformer balanced power in the studio.

Thanks!

 

[scott2000]

Something must be different from your working unit obviously... Of course there are ghost voltages but, if your other unit behaves differently, then I'd guess it's something else??

I suppose the voltage you seem to be reading is in the area of say half of  5 volt??? ... Just making silly observations to bump this...

Please let us know how it goes....

 

[ruffrecords]

Pardon my ignorance but what is a UA176?

Cheers

Ian

 

[scott2000]

https://groupdiy.com/index.php?topic=45039.0


UA176

 

[ruffrecords]

OK, so this has a transformer balanced output. This means both + and - on the output XLR are floating with respect to ground. So the voltage you measure between either of these and ground is undefined because they are isolated from each other.

Cheers

Ian

 

[Potato Cakes]

OK, so this has a transformer balanced output. This means both + and - on the output XLR are floating with respect to ground. So the voltage you measure between either of these and ground is undefined because they are isolated from each other.

Cheers

Ian

I did think that was getting a false reading (or an irrelevant one), but the only reason I would bring this up was that the same measurement on an identical build (or at least as far as I can tell) that is working properly is yielding a far different result. This is what has led me to believe that the difference in said measurements was related to the fact that one is buzzing/humming and the other isn't. If one thing doesn't have any correlation with the other then I'll not reference it in my troubleshooting and look somewhere else. I do appreciate you shedding some light on this matter.

Thanks!

Paul

 

[Potato Cakes]

To put this inquiry to rest.

The buzz/hum and difference in behavior are separate issues. The characteristics was due the 6BC8 tubes on the input. Swapped those tubes between units and the compression behavior swapped.

I wound up installing a DC heater circuit in the unit and it is dead quiet. Scary quiet, especially for as many tubes are in this thing. I still don't know or won't ever know why two identically built units had such drastic levels in noise unless there was something going on in the power transformer that I don't have the tools to measure. So mystery not solved but the problem is, which I most always take the latter over the former.

Thanks!

Paul

 

[Audio1Man]

Hi Paul
Most meters are high input Z today. The input is 10 Meg Ohms. I guess that you are measuring the leakage currents.

Put a 1k resistor across the input of your meter and if the voltage goes to zero you are indeed, measuring the leakage current.

Duke