Wiring XLR connectors - advice needed

[ABB125]

I'm in the process of wiring up some install XLR cables. However, this is the first time I've made XLR, so some advice would be appreciated.

There is conflicting advice on the internet about whether the tab circled in the image below should be connected to the ground wire. 

The cables run from a 19" rack, which is the main patch panel for the venue, to a plastic patch box under the stage. The main use case will be providing signal for l'Accoustics LA8 amps for the few occasions when we use KARA.

Essentially, should I run the ground wire through the hole in the tab before soldering it to pin 1? 

Thanks

[kgallen]

I follow Rane:

https://www.ranecommercial.com/legacy/note110.html

in which case for a cable, the answer to your question is ‘no, you shouldn’t connect pin 1 to the connector barrel’.

This in part is probably because you don’t know if the equipment at each end is compliant with the same recommendations. The cable shield should be chassis ground not signal ground. By connecting pin 1 to the connector barrel you could cause a chassis ground to signal ground short. 

[ABB125]

27 minutes ago, kgallen said:

I follow Rane:

https://www.ranecommercial.com/legacy/note110.html

in which case for a cable, the answer to your question is ‘no, you shouldn’t connect pin 1 to the connector barrel’.

This in part is probably because you don’t know if the equipment at each end is compliant with the same recommendations. The cable shield should be chassis ground not signal ground. By connecting pin 1 to the connector barrel you could cause a chassis ground to signal ground short. 

Thanks. Having read that, I think the picture below is incorrect?

Am I right in thinking that the chassis ground (see image below) is the metal housing of whatever audio device I'm using, and the signal ground is the bare wire in the above image?

In which case, assuming these diagrams are for cables rather than devices, what is the chassis ground on the cable?

One thing I'm not sure about having read that page, should the metallic sleeve be connected to anything? I assume that it's connected to the bare ground wire so doesn't need to be.

[paulears]

It's so annoying when people produce diagrams that add to the confusion. None of my cables have the pin 1 connected to the shell, because I can never be sure that this hasn't also been done elsewhere and I could introduce hums into the system via multiple grounding, but worse, in a fault condition on the wiring, you have the possibility that the full mains current fault, designed to go to ground and trip breakers routes via your audio cables. Of course, some equipment is designed to have the pin 1 tied to ground at that equipment XLR. Doing it to cables is bad news - potentially (pun intended) and unnecessary. Swapping a cable that was luckily grounding a leaky piece of kit could leave the dangerous voltages floating on a mic cable that you connect to a mic and hold to your lips ...............

Chassis ground and signal grounds are totally different things. those tabs get used when they are designed to be used.

[DrV]

That picture is wrong in more than one way. Threading the drain wire through the loop then soldering it to pin 1 would create an intermittent connection to the signal ground. Arguably even worse than soldering the two together which would at least be solidly incorrect rather than intermittent. Then there's the soldering... stray strands and what look like cold joints.😱

[kgallen]

Your photo I would say is wrong. The screen of the cable should absolutely go to pin 1 - and only pin 1. And please put some insulating sleeving on it. And yes the soldering (sorry) is terrible in several ways as mentioned above.

Also I’m inclined to think that cable is not twisted pair either which is fundamental to the noise rejection of a balanced signal.

Any cable or cable-like structure should just connect 1-1, 2-2, 3-3. You’re not (I don’t think) ‘a piece of equipment’ so connecting the shell to chassis ground doesn’t come in to it. If you’re making a patch bay that uses metal components then the patch bay may be picking up a mains ground through other equipment. Even more reason not to connect the shell to pin 1 as Paul highlights.

Not good all around I’m afraid 🙁.

Edited March 11, 2023 by kgallen

[GR1]

Pin 1 is the cable screen with sleeving so it can't touch anything else.

Pin 2 is Signal/Hot/+ve

Pin 3 is Return/Cold/-ve

There is no connection to the lug you have put a red pen mark around in your opening message.

Please look at online demos of how to solder XLR plugs and practice to improve your technique if the photo in your message of 2 hours ago is the standard of your soldering.

Sorry if this comment is harsh but I want you to be successful in your wiring.

[Brian]

I'll just leave this here...

 

[KevinE]

There are certain lighting manuals that state something like.. 'in an emergency, mic cables can be substituted for DMX cables, but you must snip the link between pin 1 and the plug shell'.

I've seen many commercially sold xlr (audio) cables with pin 1 linked tothe shell, at least at one end,

And a few installations where horrendous hum on a long cable run (balanced) was only cured by adding an absent link at one end, for example where a double insulated DJ mixer was connected into a DLM unit and the entire signal ground was floating.

 

[Bazz339]

In 50years I rarely come across Pin 1 taken to chassis of the plug and have never done it. The only time I could see it might be advantageous is if the connectors were adjacent to something unscreened kicking out a lot of electrical or higher frequency interference. 

[Brian]

9 minutes ago, Bazz339 said:

In 50years I rarely come across Pin 1 taken to chassis of the plug and have never done it. The only time I could see it might be advantageous is if the connectors were adjacent to something unscreened kicking out a lot of electrical or higher frequency interference. 

That's why the tag should ONLY be connected on equipment, so that the ground continues onto the connector and surrounds the termination.

[sunray]

4 hours ago, Brian said:

I'll just leave this here...

 

I've never been able to work out why that format is used, the first point failure is usually the screen potentially putting mic body at 50v etc.

I believe it was standard at BBC but they banned it years ago.

The screen coming out of the back of the Hellerman sleeve makes more sense to me.

 

Edited March 11, 2023 by sunray

[sunray]

Years back, mid 90's, we got a contract at ROH for some of the infrastructure during a major refit, dressing room loudspeakers & patch panels etc.

The consultant spec was not only to link pin 1 to body at every location, it also required all patch panels be bonded to earth.

We tried hard to get them to change their mind but no dice.

I understand it caused lots of problems.

[Brian]

13 hours ago, sunray said:

I've never been able to work out why that format is used, the first point failure is usually the screen potentially putting mic body at 50v etc.

I believe it was standard at BBC but they banned it years ago.

The screen coming out of the back of the Hellerman sleeve makes more sense to me.

 

If you look carefully you will see that the cable is not a flexible 'mic' cable but an installation cable, in this case FST. As such, there is no strain on the screen as the whole thing will be laced within the rack.

I've probably been responsible for the termination of over 50k cables, so 100k connectors, in this fashion and can't remember a single failure.

[paulears]

I've always taken the view that the cables are universal conduits so should be transparent - no matter what is shoved down them. There are plenty of XLR cables out there still that don't even have a ground connection pin 1 - having been a speaker cable originally, and I have always smiled at how they actually work fine in many cases with absolutely no screening at all, but get found out when somebody swaps a dynamic for a condenser and it doesn't work!

The notion that the shells need grounding for lights never made much sense. If the chassis ground goes live, you'd think less grounding on the interconnects would be better?