Ground loop issues in D-lator

[jlatrace]

Okay, I just found this tip on Brandon's "step by step" post that I am going to try and report back:

"I solder the backs of one of the pots to the ground bus because the individual pannels of the chassis are painted, and this connects them together, otherwise it makes a slight hum. This is the same reason that all the jack grounds are connected to the signal ground. "

[markr14850]

Okay, I just found this tip on Brandon's "step by step" post that I am going to try and report back:

"I solder the backs of one of the pots to the ground bus because the individual pannels of the chassis are painted, and this connects them together, otherwise it makes a slight hum. This is the same reason that all the jack grounds are connected to the signal ground. "

Before you go thru the work of soldering them, try just clipping a jumper from ground to each of them to see if it will help.

[odourboy]

If that doesn't work, could you post a schematic of your HV power supply and rectifier circuit and heater supply, noting where and how you've grounded center taps etc.? Might reveal an issue...

[markr14850]

I'm having trouble visualizing how to fit the "yin-tang diode breaker into the grounding scheme of the D-lator. For example, my gut is that I need to tie the power board grounds and not the signal grounds to the diode, is that correct? Or do I tie all grounds to one star connected to the DC side of the diode with the earth ground to teh AC side?

Do you have any pics of your install?

It's a simple idea, but a bit of a pain to try...

The mains/earth ground still goes to the chassis. That's for safety. What the yin-yang breaker does is to lightly separate the signal ground (0 volt line) from the mains ground. For this to work, the breaker must be the only connection between mains ground and signal ground. No chassis grounding for your signal grounds. No connections from signal ground to the backs of the pots. All connectors must be isolated.

What this does is to allow the d-lator to take it's definition of 0V signal ground from the amp, rather than from the mains ground.

There's still a small chance of a ground loop via the two paths of the send and return cables between the amp and d-lator. Keep these close together.

The next problem is that whatever effects you connect might try to impose their own definition of 0V on you. That standard fix to that problem would be transformers. (While you're debugging this, I'd advise to not have any effects connected to the d-lator.)

Makes sense?

[jlatrace]

Did you ground the back of your pots to the buss bar?

Thanks, soldering the ground buss to the back of the pots was the magic bullet!

Apparently the buss needs to be attached to the pots because the paint keeps the pots from making a good connection. In fact the back of two of my pots read to ground as an open before soldering them to the buss.

Thanks to everyone that helped me today. Maybe this will serve as a resource to others with D-lator grounding issues.

Les

[jlatrace]

Okay, I just found this tip on Brandon's "step by step" post that I am going to try and report back:

"I solder the backs of one of the pots to the ground bus because the individual pannels of the chassis are painted, and this connects them together, otherwise it makes a slight hum. This is the same reason that all the jack grounds are connected to the signal ground. "

Before you go thru the work of soldering them, try just clipping a jumper from ground to each of them to see if it will help.

Thanks, good tip. That is actually what I did after finding that two were reading infiniety ohms. Then I only soldered the two that needed grounding. Soldering pots is a pain, but it was worth it to solve this puzzle.

[martin manning]

Hang-on, were you using the backs of the pots for a signal ground or only as a ground for the shield on the coax cables?

[jlatrace]

Here are some pictures of the working d-lator. The screws by the jacks are where I tried mounting bypass switches. The last picture is a close-up of one of the jacks that I had to ground to the buss.

[jlatrace]

Hang-on, were you using the backs of the pots for a signal ground or only as a ground for the shield on the coax cables?

I wasn't using the backs of the pots for anything before. Instead, I ran a wire buss from the ground lugs on the pots, like Brandon.
Today I re-read Brandon's post after being told to ground the back of the pots to the wire buss.

Brandon's post said that he connects the back of the pot to the ground buss because, if not, there is a faint hum (paint keeps the pot from making a good connection).

Two weeks of troubleshooting and it was so simple.

[martin manning]

This kind of thing always seems mysterious. The pot cases do not need to be grounded for any reason other than to act as shields. The panel on my D'lator is bare metal and so makes good electrical contact with all of the pots and with the lower chassis of the enclosure, to which the circuit grounds are attached. If I remove the upper chassis cover and lift the front panel (un-grounding all the pot cases but not breaking the circuit or coax shield grounds) I get hum. It is evidently the recover-out pot that is most sensitive, since Brandon grounded only that pot and left the others (which probably are not well grounded) alone, and I guess you found the biggest improvement from grounding that same pot. So this does not seem to be a poor circuit ground or ground loop issue, but a shielding issue.

[jlatrace]

Thanks for the explanation. I really want to understand why issues occur. I am not an electrical/amp tech and over the last 10 years I have learned what little i know from folks like yourself. I started with a 5E3 kit from Bruce Collins at Mission Amps in 2001 and I am very proud of my #102, its really sweet and so quiet.

The time spent by you and others not only troubleshooting but helping me learn is deeply appreciated.
Thanks,
Les

[martin manning]

No problem... I'm trying to understand this too! The interesting thing is that you could stop the hum by lifting the ground on thr D'lator, even with power off, which would say that the earth ground was the source of the noise coupling into the return. Yet, grounding the pot case- connecting it to the noise source- solved the problem?

[jlatrace]

One more thing, the middle pot was the only one that had a good ground connection without being tied to the buss. I remember that the drive and the recover pots both seemed to add to the hum when turned up; while the middle pot did not seem to effect the hum much. Anyway, i suspect that both the pots contributed to the issue and I'll probably tie all the pots to the buss if I do this again.

[jlatrace]

No problem... I'm trying to understand this too! The interesting thing is that you could stop the hum by lifting the ground on thr D'lator, even with power off, which would say that the earth ground was the source of the noise coupling into the return. Yet, grounding the pot case- connecting it to the noise source- solved the problem?

Yes, you are right. I really cannot explain why but that is what I experienced.

Maybe someone else can explain?

[Structo]

Glad you got it going.

No offence intended but your solder joints aren't looking real great.

You aren't using non lead solder are you?

They should be smooth and shiny.

If the start looking chunky and dull, that is usually an indication that the iron is not hot enough.

Commonly called a "cold joint", it can have high enough resistance to cause a ground loop or just a poor connection.

Here is an example of what I am talking about.

I do realize that when you are trouble shooting or tweaking an amp, that things can get messy fast.
I just wanted to point out that if you improve all those solder joints so they are smooth and shiny, it may sound even better.