DIYHiFi.org

As if cable talk was not enough divagation, now this?!
CG, stick around please.

_________________
Carlos Filipe

"Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius - and a lot of courage - to move in the opposite direction." Albert Einstein

Hi Andrew,

There are two camps in the pro audio world - those that connect the chassis together and those that do not. There are pros and cons to each and Bill Whitlock has an excellent article that discusses this here: http://www.edn.com/archives/1998/060498/12df_03.pdf

In either case, pin 1 of the XLR jack in both components should be connected to its chassis; and pin 1 of the cable connector should be connected to the cable shield in the send end of the cable. If the two chassis are not to be connected, pin 1 of the cable connector should be open on the receive end of the cable. If the chassis are to be connected, pin 1 of the cable connector should be connected to the shield at the receive end of the cable.

Considering the cable connector shell as part of the shield is a good idea and Neutrik makes several XLR chassis connectors to facilitate this. For example, they have ones that connect the shell to pin 1 and others that have a separate pin for the shell which can be externally connected to the chassis. Therefore, the chassis may be connected, with or without the connector shells, through the cable shield connected at each end to pin 1 of the jacks. A 4th wire is not necessary.

Dave

Neutrik actually make some connectors far more advanced than those. Look up their EMC XLR series.

Better still read Rane Note called "Pin 1 Revisited".

Apart from that, generally, the most robust method seems to be according to AES48 convention.

If you don't connect at the receiving end you can eliminate SCIN, which I think is what you hear in a cable. However that seems to invite common mode issues, modulation, rectification, which bloats or inflates the soundstage and moves it all around in ghostly fashion. When wired right it's boring and nailed in place and the imaging is that of the recording and system but not by ground potential.

Further there is PEC, which may be the "fourth wire", I wasn't sure if that's what Andrew was asking or not.

_________________
'I am... everything is... changed... they're calling... your face... interwoven... who is...

thanks gents/lassies.
I go back and re-read Whitlock.

The shell in all my XLRs is unconnected (or plastic). But it does make good contact with the mating scraper in the socket which in turn is bonded (bolted) to the chassis.

_________________
regards
Andrew T.

Andrew, I don't see any avantages in connecting the two chassis with yet another ground. It will create a ground loop.
It would be better to use a cable comprised of 3 wires and shield.
The shield would only be connected to the metal shell (chassis) at the sending device and left floating at the receiving end.
Now you have the cable shielded.

_________________
Carlos Filipe

"Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius - and a lot of courage - to move in the opposite direction." Albert Einstein

Hi Carlos,
I don't think so. 3wires and shield is equal to a 4wire cable. Two wires and shield is a 3wire cable.
The shield is the third wire and in all my cables is already connected at both ends to the Audio Ground, Pin1, not the chassis.
If the equipment at either or both ends connects Audio Ground to chassis then the third wire could/will pass earth to earth current. I don't think that would be good, even if the signal pins are isolated from the ground pin.

I am specifically referring to a separate shell connection which would NOT be connected to any of the Audio side pins in the cable.

As I said, I'll re-read Whitlock and if I am still uncertain I will come back.

_________________
regards
Andrew T.

Andrew, if you read my post again, I was suggesting the connection of the cable shield to chassis at one end.
Not to signal ground. I presumed that there is no direct connection between chassis (earth) and signal ground on the equipment.

_________________
Carlos Filipe

"Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius - and a lot of courage - to move in the opposite direction." Albert Einstein

I've seen it done both ways, from pro manufacturers. What you will also see is a ground lift switch, and some gear will have a chassis lift switch inside.

This allows the user to set up the configuration he wants.

I normally use a 10 ohm R between the shield and chassis.

The best to do is try the different methods and stick with the one that sounds best to you. It's easy enough.

Yeah, we should really not refer to the shield as the "third wire", though. Better we just call it what it is.

Read AES48 first, that's more about an extended shield whereby your chassis becomes the shield, which is why you do connect them..

Yes earth current flows but if you do it right it's immune, your signal ref doesn't see the potential created by it and neither do your inputs.

PEC, = parallel earth connector, can also reduce the effects of the loop current, worth looking up as well. Multiple channels (multi shields) and the mains earth all equate to PECs.

There may be some debate as to whether or not to apply ground lifts to signal ref within the extended shield chassis, and that may be worth experimenting with, but I think even Whitlock who advocated it said if there's any doubt and you have otherwise well designed equipment that includes a ground lift, set it to ground. Why that be is because AES48 convention= most robust method.

You're going to have earth current flowing parasitically even if it's not connected.

_________________
'I am... everything is... changed... they're calling... your face... interwoven... who is...

Hi Andrew,

Whether or not the two chassis are connected is a separate issue from what is connected to pin 1 in the XLR jack. A poor choice of what is connected to pin 1 is a major cause of ground loop problems; connecting pin 1 to audio signal ground rather than the chassis is called the "Pin 1 Problem." An overview of the Pin 1 Problem with references to further study can be found in my article on interconnecting and grounding, here:
http://www.raleighaudio.com/Audio%20Com ... ection.pdf

Dave

Though he may not be suffering hum it does invite other problems as well. I find it's a great way to make a low pass out of your interconnect's parasitics, and that greatly affects the sound! Not cool.

Your article at 63 pages seems thorough enough, I just skimmed it quick. To see things like "unplug all devices" (dvd, tv, fridge, cut neighbours power)... not realistic at all. Neither is "avoid SMPS"... we can't reshape the world to be how we like it, we have to work within the confines as things are today.

Admittedly that makes things more difficult, but it simply stresses the importance of adhering to the proper and robust techniques.

I'll give that article proper attention later, it does look deserving of it.

Regards

_________________
'I am... everything is... changed... they're calling... your face... interwoven... who is...

Thanks TD,

I'm sure that the article can stand improvement so comments are welcome.

Dave

Yes, we got to work it out "within the confines".

But the approach suggested by Dave D. is correct. I did try to UNunplug one by one the power cord of each component in my auido rig leaving only the tube phonostage & the tube power amp powered on & full volume on. I did find substantial improvement in hum level with all other components unplugged - from very low to virtually undetectable with my ears put against the woofer of my KEF bookselvers (13" x9.5" bass driver).

On plugging them back, the lowly hum came back again. So I tried to reverse the polarity of the 2-pin power plug of each component until I got the lowest hum level. But still I could NOT get back the same virtually undetectable hum level with all the other componets UNplugged.

So I have now rearranged to power the components by plugging them in small local on-off switchable power strips. Whatever components not playing are always have their local power strip swithced OFF. This does the tweak!

A thing is utmost crucial: - Analogue components, e.g. TT, tape deck, & amps do NOT share with the same powerline feeding digital components, e.g. CD-audio player & DVD-audio player.

I tested them with a wideband powerline & EMI nosie analyser & found out the noise level of the powerline where the CD/DVD players plugged in got noise level increased whenever the players were switched on !!!

This test prompted me to have had dedicated powelines installed & hooked up direct to my house main fuse panel exclusively for my audio rig, to power analogue & digital components with dedicated RFI inline filters. No mixing powerlines! NEVER use wallouts which are usually digitally polluted like hell.

Since I can't unplug my labtop, LCD TV & elctrical appliances at home while playing music, I
installed common-mode RFI suppressor to the power cords of ALL of them. This is a compromise, but whatelse I could do better?? This is a realworld, right?

c-J

_________________
"If the facts don't fit the theory, change the facts" quoted Hans Albert
Einstein.

Exactly where did you place all these filters on your dedicated lines and other components?

What exactly do the filters consist of??

Sorry to hear you still have hum after all that. BTW I can't consider running "dedicated lines" as working within the confines, as it is most likely a rather expensive placebo.

_________________
'I am... everything is... changed... they're calling... your face... interwoven... who is...

I think (not sure) that the susceptibility of your audio system to hum when powering other equipment is telling you that the audio equipment itself is not optimally grounded.

_________________
regards
Andrew T.