• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Grounding problem?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I have for a long time used Dave Slagle's autoformers as passive preamp.

But because the main listening source DAC has a passive output stage as well. It think it was a case of "too much passive" and the sound lacked dynamics.

Therefore I am planning to build a preamp again. And next ...convert my Raleigh Audio line stage to work as active I/V converter for the DAC.

The thing is ......the preamp that I built quick and dirty...is absolutely quiet when using my pc as source. When I take the preamp to my main system...all hell brakes loose...very loud humming and noise.

So plugged it out and re-did the grounding...put it back into my main system..but no joy.

So I thought...why not put the MQ B7 10k:10k input transformers between my DAC and amplifiers so they can "run-in".

Too my complete suprise...a hell of a racket came out of my speakers! The same type of racket that sometimes comes when you have no input to your poweramps...very loud squealing.

It seems that whatever I put between my DAC and poweramps (except autoformer but even then sometime at the maximum settings noise came) I'm having grounding problems?

What can it be?

PS..My poweramps don't have mains grounding (Only became compulsory in the Netherlands in the 90's I think.)
 
Hi Bas,

Grounding can be quite a headache, and for me, galvanic isolation helps a lot. But … it sounds like your power amp doesn’t like the isolation that an input transformer provides. Have you tried putting your B7s between the DAC and amplifier and also running a separate wire from the amplifier chassis back to a point where it can pick up mains ground?

Also, in cases like this, I draw a picture all of the ground wiring and look for loops. The best case is that everything is grounded once and only once so there are no loops.

Dave
 
Have you tried putting your B7s between the DAC and amplifier
Hi Dave,
Good speaking you again!

and also running a separate wire from the amplifier chassis back to a point where it can pick up mains ground?
No I haven' t. I would have to take a wire from the living room to the kitchen.

I draw a picture all of the ground wiring and look for loops.

That is what is so funny.

Nothing is grounded to mains ground.

Here goes my system:
Squeezebox 3 -> Co-Ax spdif - RAKK DAC MK2 passive -> SE 300b amp.
Volume is now controlled via the SB3.

So there is a galvanic isolation from SB3 to (spdif transformer on the dac) RAKK DAC....galvanic isolation from DAC to amp. Stick ANYTHING between DAC and amp and the fun starts.

That is what was so suprising. Galvanic isolation everywhere. And sticking more galvanic isolation in between OR break the galvanic isolation by adding a preamp creates HAVOC.

:eek:
 
I built an Opus dac this spring using Lundahl 1544A's for balanced to single ended conversion providing galvanic isolation as my output RCA's were isolated from the chassis. With an SB3 as input, the ground loop hum was fierce with a gainclone, and non-existent with a Bottlehead S.E.X. amp. I killed the hum by connecting the digital input ground in the dac to the output ground. Something through AC neutral or ground. (grounding or lifting the AC ground on the gainclone made no difference.

Skip
 
Okay on your mains ground. Remember, there is no ground anyway – just references.

If you add additional isolation, this isolation must be referenced back to system ground or you will have floating signals which will act as antennas and pick up noise. So, for example, if you have:

Source > transformer 1 > transformer 2 > amplifier

The connection between transformer 1 and transformer 2 must have a low-impedance path to a reference. The best would be a center tap but I don’t think that is very important so it could be one of the two signals that is connected to the system ground..

Dave
 
Source > transformer 1 > transformer 2 > amplifier

Transformer 1 is spdif transformer and transformer 2 is LL1674.

The connection between transformer 1 and transformer 2 must have a low-impedance path to a reference

If Skip_Pack had an spdif transformer as well. That is exactly what he did...wire the two up (low-impedance path to a reference).
 
Dave,

I didn't try connecting the input spdif ground to the balanced output center ground at the 1544a primary. That could well work while maintaining total isolation downstream.

Further information. I used the TPA first generation Wolfson spdif input card through a Metronome ASRC to the dac card. Enough intervening stuff to baffle me. The three cards share a digital supply, the dac card has a separate analog supply.

Finally, the hum with the gainclone was there only with the SB3. Not there with a DVD spdif out or with no input device.

Skip
 
David Davenport said:
Source > transformer 1 > transformer 2 > amplifier

Hi Bas,

Sorry for the confusion. This is meant to be a generic example of two transformers directly connected together. The point is that the connection between the two transformers does not have a reference and this could be a problem.

So when you inserted the B7 between the passive output and the amplifier, perhaps the connection between the passive output and the B7 did not have a reference.

Or, if the secondary of the passive output was referenced to system ground it was providing a reference to the amplifier, and when you inserted the B7, this reference to the amplifier was lost.

When you add an additional transformer, you must add an additional ground reference somewhere.

Dave
 
Hi Dave,

No confusion. I just wanted to make really clear what the situation is/was. :) And even now I was not 100% clear....because as I mentioned earlier...even with the Slagle autoformers I used to get problems ...the last 3 or 4 volume positions where off limits...cause then the noise started. But only now have I come to realize that this was probably one and the same problem.

So when you inserted the B7 between the passive output and the amplifier, perhaps the connection between the passive output and the B7 did not have a reference.

Or, if the secondary of the passive output was referenced to system ground it was providing a reference to the amplifier, and when you inserted the B7, this reference to the amplifier was lost.

Well that seems to be the problem then. What is system ground? I suppose my definition of system ground used to be.....system ground = signal ground. And every piece of equipment is connected only at one point through signal ground.

I was obviously mistaken that with transformers you can get around each and every ground problem or at least that you need no system ground or reference anymore... because I thought everything was sort of floated now.
 
Bas Horneman said:
... because I thought everything was sort of floated now.


Hi Bas,

Well, at some level everything is floating – consider a circuit in an airplane. So, as I said before, there is no ground, only references. These references are low-impedance paths where the circuits can get “grounded.” Without these “grounds”, high-impedance points are prone to picking up noise. There can be several of these low-impedance references in a system, and they should be tied together. BUT, it is important that the paths where they are tied together are chosen such that no current flows in them under normal circumstances. For an example of this, take a look at http://www.raleighaudio.com/installation.htm

Dave
 
Hallo Dave,

Thanks.

Had a look at the diagrams again.

And I remembered that I've had to install the so called gnd jumpers. Otherwize I had mayhem and noise. But I do know that I've had them not connected for a long time. Something happened somewhere along the line...what I don't know...to cause the system to act in the way it did for me to have to connect those gnd jumpers.

I guess that means in my case I had to connect the external analog ground reference to the internal ground reference (in the case of my RAKK DAC).

Now lets assume I want to add a non-transformer coupled preamp (Which I have tried). In that case the ground reference of the power amplifier will be connected to the ground reference of the pre amplifier via signal ground. The noise is unbearable in this case.

Therefore I assumed I have introduced some kind of ground loop. Which is likely in such a situation.

If I introduce a signal transformer between DAC and poweramp I still get unbearable noise. The reason this time is that I have no " low impedance path" (is that a different name for a wire ;)) between the various ground references of the various parts of my audio system.

Thing is I can't find the problem in the preamp...i.e. no connection between chassis and rca input and output plugs. And otherwize my standard grounding scheme (buss).

With the signal transformer I could ofcourse try to connect the grounds of primary and secondary. If I have silence then I know that the poweramps would like a better system reference?
 
Bas Horneman said:
Therefore I assumed I have introduced some kind of ground loop. Which is likely in such a situation.

Thing is I can't find the problem in the preamp...i.e. no connection between chassis and rca input and output plugs. And otherwize my standard grounding scheme (buss).


Hi Bas,

What I would do is to draw a picture of your system wiring and internal paths and look for the loops. Remember that there are hidden paths, like capacitive leakage between the windings of the power transformers. Thus you will have a loop between two chassis connected to the mains even though there is no safety ground. This loop will include the shield of the unbalanced interconnect and this is where the noise is injected. A transformer will break this path. Without a transformer, the best you can do is to offer the noise current a lower impedance path between the two chassis, that is, a heavy wire between the two chassis.

You say that there is no connection between the chassis and the RCA jack. Thus, now your ground loop includes the signal ground in the preamp. It may actually be better to have a connection between the chassis and the jack instead of having the return noise current flowing through the internal signal ground in the preamp.

Dave
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.