The age old debate of grounding the shield/shell on XLRs

[digitaldrummer]

So I just got some new XLR cables (instead of making my own this time) from Monoprice. I got a few of the older "premier" series 16AWG and then a couple of the newer 16AWG design too. (side note -- they seem pretty well built). So far all of the cables I have opened have both ends where the cable shield (pin 1) is also connected to the tab that goes to the connector's shell. I had always been told that connecting both ends (for an analog cable) is a potential source for ground loops (since you are connecting 2 devices -- say perhaps a tube mic, and your preamp) that may be connected through 2 different outlets/sources and possibly at 2 different chassis ground potentials, and thus the "ground loop" that send all kinds of wonderful buzz into your signal chain.

I am in the process of cutting the connection on the female side of the XLRs, since having only the male side connected (or sometimes no connection on either side) is typically how I have had the most success building and using analog XLR cables, and so I continue to do it that way. btw, I am only talking about the shell connection here -- pin 1 shield is always connected on both ends.

No need to provide links to debates or technical papers, etc. I've read a lot of them and have seen the religious wars. What I'm curious about is what do you do? I'm not posting this to disagree with your or tell you who's right and who's wrong. I am just curious what others do (or if they have even considered this...)? And I guess if you have not ever thought about this, then you can read along and choose a side (after all that's what the internet is all about right?)

[Scodiddly]

Yeah, you don't want that shell connection. None of mine do, neither does the pro sound company where I work.

[I'm Painting Again]

I've never connected the shell ever.

I go by the Rane wiring manual for everything.

I believe there are some situations that it might be correct to do so but I've never come across that situation.

[emrr]

I’ve never connected the shell.

[digitaldrummer]

yeah, I'm rethinking this... I looked at a bunch of other cables and realize most don't have the shell connected. Some I built, some are manufactured. So I'll likely be snipping both ends of these Monoprice cables. Honestly I don't think it makes a difference in most cases, but that one time that it might, I'll be glad if it doesn't... (also the official AES recommendation is not to connect them)

[emrr]

I forget right now what I’m missing - what would the problem be?

Shell connection should just extend shielding to the connection gap. Shield wire should connect through either way. You encounter poorly shielded/grounded equipment, it’s not the cable’s fault. Chassis xlr should ground immediately to the shell case, connecting to equipment chassis at point of entry. Should it go elsewhere inside the case first it might allow external pickup in where it shouldn’t be, again an equipment design fault. I keep a pair of ground lift barrels in case of problems, they’re not something I ever have to use. When there are problems with long lines over different power grid potentials, it usually takes both lift and transformer isolation.

So….yeah…. I never connect shell on anything I build. The gap in shielding has never presented me a problem. It’s maybe the safer bet over connecting shell and then plugging it into an improperly grounded equipment chassis.

Remind me what else I’m missing, not having the time to go look it all up again.

[Nick Sevilla]

I'm firmly in the "do not connect the shell to ground." camp.

I believe this thing began, like most dumb audio myths, in the heady 1970s, when everyone was on drugs (JK), and no one had a degree in electrical engineering, nor a buddy with one.

Back in those days, there were less companies adhering to proper electrical standards, unlike today. So, little mom and pop companies would design their gear themselves, sometimes with crappy grounding designs. And then some enterprising soul at 3 am in the studio, with a soldering iron, would try different things to try to reduce the ground hum. With mixed results. then the old "broken telephone" thing would happen,and VOILA! We have people soldering connector shells to the ground connection. XLR, TRS, and other sundry ones, I have seen them all with this "feature." I once encountered an entire patch bay build with tiny ass capacitors ON EVERY GROUND CONNECTION. Thousands of the tiny blue critters. Had to get rid of all of them. That had been done, unsuccessfully, to get rid of the local Latino radio station.

The best practice is indeed to not connect anything to the mechanical connections. Ever. And ensure each pin has no cross talk, indicating bad wires with crumbling sleeving around the copper wire.

If there is a ground potential difference between two devices, that gets solved by an iso box / DI or simply temporarily disconnecting the ground from the receiving end with a custom adapter. Instead of permanent wiring.

[ulriggribbons]

I’ve repaired a lot of ribbon mics over the years, and I find across many microphones, having the microphone body connected to ground, which is often provided by a grounded shell, results in a microphone that rejects noise better.

Some mics will connect pin 1 to the body, but many do not.

$.02

Jon

[emrr]

I’ve repaired a lot of ribbon mics over the years, and I find across many microphones, having the microphone body connected to ground, which is often provided by a grounded shell, results in a microphone that rejects noise better.

Some mics will connect pin 1 to the body, but many do not.

$.02

Jon

That's kinda crazy, so pin 1 floating in the mic, nothing to it?

[Scodiddly]

I’ve repaired a lot of ribbon mics over the years, and I find across many microphones, having the microphone body connected to ground, which is often provided by a grounded shell, results in a microphone that rejects noise better.

Some mics will connect pin 1 to the body, but many do not.

That's kinda crazy, so pin 1 floating in the mic, nothing to it?

I recall that things like microphones, at the end of a run, were the one exception to not connecting. In rack gear there's often a different grounding path for the shell vs. pin 1, and of course there are both inputs, outputs, and power supply grounding going on.

[digitaldrummer]

Remind me what else I’m missing, not having the time to go look it all up again.

My problem was that a few cables I bought had all of the shell's connected to audio ground (pin 1) and I'm in agreement that the "chassis ground" of the equipment on either side of the cable should ground the shell properly (and if it doesn't then something is faulty with that equipment). So I snipped them all (along with a few others I had with one or more ends connected) so all my cables are at least consistent. there's nothing worse than having cables that behave differently than others... unless its on purpose and you mark them as such.

[Nick Sevilla]

Found this interesting article, from 2002.

https://www.ranecommercial.com/legacy/note151.html

The one problem, is that there is still inconsistency from different audio manufacturers regarding as to how they ground their gear. Some have chassis grounds, some have disconnectable chassis grounds from the audio ground (usually a tab with screws on the back of the unit), and some have a hot mess of ???

That is why I do not connect the mechanical ground (the connectors themselves) to the audio ground pin 1 / sleeve.

In the now rare case where I do get hum loops, it is safer to use a DI box with a ground lift, breaking the ground loop there, and not having to make any permanent alterations to either the cables nor the equipment.

[Scodiddly]

In the now rare case where I do get hum loops, it is safer to use a DI box with a ground lift, breaking the ground loop there, and not having to make any permanent alterations to either the cables nor the equipment.

Grounding and shielding *is* a lot better these days, and there's been an even bigger improvement in household electrical wiring. Three-pin grounded outlets are code, and have been long enough for most of the horrors of the past to have been replaced.

There are still difficult bits of legacy gear, but in the live world it's almost always just something a muso brings and we can use a DI or mic to get that signal.

Don't ever cut off ground pins on power cables, that's one of those "horrors of the past" things. If you depend on something that's fussy about power, get it updated by a professional tech. They'll probably fix some other problem(s) in the process.