SEVERE GSG Hum Problem

[cddeluca]

I've searched the forum but come up dry so far.
I've had a pretty major hum problem, so much so that I gave up on using the amp.

Construction is typical/stock, with the Edcor transformers others here have used. Chassis is by Tom McNally so layout is the same as dozens of others shown here.

Long story short, after trying several things I finally thought to just short the inputs and it's solid-state silent. With the inputs open, it's very loud. With the inputs connected to a passive pre it's quieter, but still loudly humming.

The wiring looks almot exactly like this amp of Tom's:

http://www.diytube.com/phpBB2/viewtopic.php?t=5150

The only difference I can see is that IEC inlet ground (not neutral) is grounded at the chassis and at J6 as someone here said it must be, even though Tom's isn't that way.

Can somebody suggest how to go about figuring out how to break the ground loop? Is it safe to ground the IEC connector to the chassis only and not J6? I don't want to change something and try it out only to empty a fire extinguisher on it.

Thanks for any help.

[Shannon Parks]

The IEC ground and J6-2or3 to chassis should be fine.

1) I would disconnect everything at J6 and then start ohming out the RCA inputs and speaker terminals. Everything *should* be floating at that point. If not, you can find your short to the chassis (and ground loop) there.

2) We could have a bad tube in the chain.

3) Verify with a diode checker on a DMM that all your bridge diodes are OK and with proper orientation.

Shannon

[cddeluca]

The RCAs are floating (no short to chassis). I didn't want to remove 200 screws to see/measure the diodes but I appear to have about 25 millivolts peak-to-peak at the heater pins with no tubes installed.

Could the driver tube actually cause the problem? It's not like it doesn't play music - it does - but there's just so much hum.

And as I wrote before, it's dead silent if the inputs are shorted. I would think a tube problem would be evident even with the inputs grounded?

[TomMcNally]

I just built a get*set*go this morning .... no hum!

Are your RCA jacks grounded to the chassis ?
They shouldn't be. Check the wiring to the input connectors
on the amp too.

Does your preamp have either a computer or cable TV connected
to it? That can cause hum.

If you don't have any hum with shorted inputs, your problem
is external to the amp.

Let's see some closeup pics.

BTW - I don't ground the board to the chassis.

[cddeluca]

The RCAs are clean/isolated. No short at the board end. No computer or cable TV connected. No preamp, really, totally passive.

I have a 300B sitting right next to the GSG and I don't have a hum problem with it just swapping the inputs from the GSG to the 300B.

Can you tell me how to post the pictures?

I know you don't tie the chassis ground to the board, but it doesn't make a difference one way or the other on my amp.

I do have a CL-90 between the HV center tap and J6, but I can't imagine that's a problem - should I remove it?

Thanks, Tom.

[TomMcNally]

Get that CL90 out if there! They go in the AC power line.

[cddeluca]

I'll try that next. I swapped the rectifier and driver tubes and that had no effect.

Not for nothin', but I and others have done it before to ramp up the HV after the heaters without a problem. The data sheet applications show it after the rectifier in power supplies to limit capacitor inrush. They're rated at 240VAC and in this case center to one leg is no higher than that so I didn't anticipate an issue there.

I'll report back...

[cddeluca]

Removed the CL-90; no change. Humming right along.

Sort of glad, in a way, because then I'd have to figure out why it's a problem this time and not previously.

HV seems to be well behind the heaters without it so it wasn't necessary anyway.

Thanks again.

[TomMcNally]

Post some pictures if you can ... maybe we can diagnose by following
the wiring.

How is the board mounted to the chassis? Did you use the spacers
on the parts list? Any chance a nut is touching something it
shouldn't?

The fact that it doesn't hum when the inputs are shorted points to
issues with the way the inputs are wired to the board, or the RCA jacks.

Are the output (speaker) leads floating? They should be.

Sorry I don't have a quick answer, I've built about 30 of these things
and never had any hum.

[cddeluca]

I didn't use the actual board spacers specified - I used drilled and tapped through ones and two screws, 'cuz I had them.

Guess I should ring out all the exposed screw heads to the input signal ground to make sure.

Were you able to see the pics I posted the last time?

http://www.facebook.com/media/set/?set= ... 91dcbb34bb

If you can, I'll post some new ones there.

Thank you.

[TomMcNally]

Yep, I can see the pics. Where is the choke?

Is the centertap from the high voltage winding now connected
directly to the terminal strip? It's important in any amp to
have that connected as close to the negative ends of the filter
caps as possible.

Otherwise, I don't see any obvious issues.

[cddeluca]

Choke is mounted to the back wall, as in your photo of the same amp. Yes, the HV CT is terminated on J6 - that's where the CL-90 was - between the CT and J6.

Thanks you for spending so much time. If I ever figure it out I'll be sure to let you know.

Best, Charlie

[Shannon Parks]

Re-reading this thread and all the clues, how about double-check your input wiring to J1 and J2? Make sure the ground lead goes to J1-2 and J2-2 (the outside screws) and the center post of the RCA connectors go to J1-1 and J2-2. If this was swapped, it would explain your symptoms perfectly.

Btw, here's my thread on getting a super quiet Clementine:
http://www.diytube.com/phpBB2/viewtopic.php?t=5236
The same things can be applied to the GSG, and you already have a lot of it done, except the speaker grounds run back to the PCB (which shouldn't be necessary). Ultimately, the GSG is one of the quietest amps I have - I think we can solve your mystery.

Shannon

[cddeluca]

That was one of the first things I verified, thinking the same thing. I also tried to find some high resistance short between the two + and GND terminals at the board but no luck there either.

Thanks for the suggestion.

[cddeluca]

This morning, out of ideas, I stripped out the signal wiring. I ohmed out the J1/2 connectors to the driver base pins (A-OK).

Step 1: I replaced the RCAs with the best ones I have that fit the chassis holes and went back with tightly twisted pair 20 ga instead of the mic cable I had in there previously.

No help

Step 2: Relocated the signal wiring from the side wall of the chassis to the center of the board/amp a la Tom McNally.

No help


Step 3: Grounded the binding posts.

No help - same story.

What would happen if a cathode bypass cap was bad - I'm thinking nothing but gain change. The hum's identical in both channels, though it could be modulating through the ground, I guess.