Ground Loop with Balanced only

[MingChops]

I have a ground loop (hum/ticking through speakers) when using the balanced connections between my NAD M66 and Audiolab 8300XP, but it is not there when using single ended.

To confirm I believe this is a ground issue, if I have just the balanced connected, and then connect a wire between the extra ground post on the M66 and touch a chassis screw on the amp, the noise goes away, and comes back when you remove it.

My old Pre, a Schiit Freya S, I had no such issues using balanced between that and the Audiolab.

However I don't really understand the engineering behind these things, so I don't know if this is normal or acceptable? I know that balanced is usually meant to solve these problems, so could it point to an issue with the M66?

TIA

 

[Sokel]

Are you sure your cables are ok?
Cause a proper XLR cable does exactly (or should do) what you did with the extra one,the shield of the XLR to be exact (the pin marked as "1" )

Did you try others?

 

[MingChops]

Yeah I tried different sets of XLR's from different manufactures.

I spoke to the retailer who ran it past their technical centre - they have come back to me and said it is likely because the 8300XP has a floating ground. And that this behaviour does not suggest a problem with the M66.

Either using an RCA between them, or the ground connection, is just what will need to be done with this power amp.
And that other amps with different grounding topologies likely will not exhibit this behaviour (such as the M23 which should work perfectly - but I'm too skint for that for a while!).

To someone like me who doesn't really understand the different ground topologies, I would have expected RCA to be the problematic one, so balanced is I guess not always the fix for everything?

 

[Sokel]

Yeah I tried different sets of XLR's from different manufactures.

I spoke to the retailer who ran it past their technical centre - they have come back to me and said it is likely because the 8300XP has a floating ground. And that this behaviour does not suggest a problem with the M66.

Either using an RCA between them, or the ground connection, is just what will need to be done with this power amp.
And that other amps with different grounding topologies likely will not exhibit this behaviour (such as the M23 which should work perfectly - but I'm too skint for that for a while!).

To someone like me who doesn't really understand the different ground topologies, I would have expected RCA to be the problematic one, so balanced is I guess not always the fix for everything?

Oh,the beloved floating ground,not even the 80 Ohm one that was used back in the day

In that case use RCA,if you don;t hear any noise or distortion you're more than ok.

 

[MingChops]

Oh,the beloved floating ground,not even the 80 Ohm one that was used back in the day

In that case use RCA,if you don;t hear any noise or distortion you're more than ok.

Is floating ground not a good thing/not used anymore?

My electrical engineering knowledge is not great, at all!!

 

[Sokel]

Is floating ground not a good thing/not used anymore?

My electrical engineering knowledge is not great, at all!!

Take it from from your very own experience.
There are cases that can come handy.

In general though a properly grounded device is the norm.

 

[Sokel]

The strange thing is that normally it should be the other way around,that the RCA would cause trouble or loud pops if connect or un-connect with the gear on.

 

[pjug]

It sounds like the same issue some are having with the Bluesound Icon so you might watch that thread to see if anything happens there that might help you.

Bluesound releases 2024 ICON streamer

So are you saying they all have this fault or are you just relunctant to return it for some reason ? It sounds like a ground loop. I don’t fully understand the why and how, but maybe @amirm or @Buckeye Amps can explain (because there is a YT video of someone with the buzz with Buckeye Amps...

www.audiosciencereview.com

 

[MingChops]

It sounds like the same issue some are having with the Bluesound Icon so you might watch that thread to see if anything happens there that might help you.

Bluesound releases 2024 ICON streamer

So are you saying they all have this fault or are you just relunctant to return it for some reason ? It sounds like a ground loop. I don’t fully understand the why and how, but maybe @amirm or @Buckeye Amps can explain (because there is a YT video of someone with the buzz with Buckeye Amps...

www.audiosciencereview.com

Thank you, yes exactly the same noise is coming from my speakers when using the M66 and the 8300XP.

I'm just working my way through the thread and the Bluesound pages - though I've seen it stated on the ICON (and therefore I assume the M66) that it uses a floating ground, as my amp is also using a floating ground then I have no stable reference. I know from doing some basic DC sensor stuff, you always need a common reference point for consistent readings - so with my two devices both being floating I can't see I have that.

 

[AnalogSteph]

Can you check with a multimeter which one of the devices is not connecting XLR shield (pin 1) to its respective chassis ground? Having an optional ground lift is one thing, it being like that permanently quite another.

Going by the "extra wire" experiment, your ticking must be present in the form of a large common-mode voltage between M66 and 8300XP that is testing the limits of input CMRR.

I would test whether an additional RCA cable can be used to establish the missing shield connection. Failing that, terminate the wire properly so this connection can be made permanent. Or just use the RCAs to begin with. If it works, it works.

 

[Speedskater]

On of Bill Whitlock's seminar papers has a section on troubleshooting this type of problem.
Let me go look.

 

[MingChops]

Can you check with a multimeter which one of the devices is not connecting XLR shield (pin 1) to its respective chassis ground? Having an optional ground lift is one thing, it being like that permanently quite another.

Going by the "extra wire" experiment, your ticking must be present in the form of a large common-mode voltage between M66 and 8300XP that is testing the limits of input CMRR.

I would test whether an additional RCA cable can be used to establish the missing shield connection. Failing that, terminate the wire properly so this connection can be made permanent. Or just use the RCAs to begin with. If it works, it works.

Would that be a continuity check? I only have a crappy multi meter, one that doesn't do the beep test for continuity - if you can tell me precisely what to do and the modes to use on the multi meter I can try, but otherwise sorry I'm just not confident enough to be poking probes into a my very (to me) expensive new toy!!

Yep when I had both the RCA's and XLR's connected, like with the extra ground the buzzing/ticking sound was gone too.
I have decided this is the neatest solution and cuts all the noise. Shouldn't be any downsides??

 

[Speedskater]

In Bill Whitlock's (retired Jensen Transformers) 2012 seminar paper:
"An Overview of Audio System Grounding & Interfacing"
The troubleshooting section starts at page 99.

https://indianaaes.org/wp-content/uploads/2012/09/indy-aes-2012-seminar-w-notes-v1-0.pdf

 

[HaveMeterWillTravel]

Yeah I tried different sets of XLR's from different manufactures.

I spoke to the retailer who ran it past their technical centre - they have come back to me and said it is likely because the 8300XP has a floating ground. And that this behaviour does not suggest a problem with the M66.

One possible interpretation is that instead of having the XLR Pin-1 signal return being tied/referenced to chassis (*), it (and most likely) pins 2 & 3 are isolated from chassis. For example, if the front end circuit is powered by an isolated power supply. There are some possible technical reasons to do this; but it would a similar design on the upstream "source". I am a bit perplexed at this design decision as looking at the 8300XP, it has a conventional 3-pin power supply.

I suspect what you are encountering is that one or both devices has significant common mode voltage, often due to transformer leakage current in the power supply.

That being said, I am more perplexed by the beneficial difference tying the two chassis together using the NADs ground terminal. Not that it is beneficial, but the implications. That seems to imply the 8300xp chassis is NOT tied to the power cords ground.

Either using an RCA between them, or the ground connection, is just what will need to be done with this power amp.
And that other amps with different grounding topologies likely will not exhibit this behaviour (such as the M23 which should work perfectly - but I'm too skint for that for a while!).

To someone like me who doesn't really understand the different ground topologies, I would have expected RCA to be the problematic one, so balanced is I guess not always the fix for everything?

If measured collectively, the noise floor of the RCA connection may be higher than with the XLR and grounded chassis. Just not enough to be noticable.

That said, an RCA cable/connection may be more susceptable to picking up radiated noise. The XLR would most likely be superior in that regard.

 

[HaveMeterWillTravel]

In Bill Whitlock's (retired Jensen Transformers) 2012 seminar paper:
"An Overview of Audio System Grounding & Interfacing"
The troubleshooting section starts at page 99.

https://indianaaes.org/wp-content/uploads/2012/09/indy-aes-2012-seminar-w-notes-v1-0.pdf

Bill Whitlock is always a great source!