AES50 Cables and ESD

[RexBeckett]

The minimum specification for the AES50 cables used to connect stageboxes and consoles is shielded Cat5e. Higher specification cables (e.g. Cat6a, Cat7, Cat8) can also be used provided that they are terminated correctly. It has been found to be very important that the cables are fitted with etherCON shells at both ends and that these are solidly connected to the cable shield - usually through the shielded RJ45 connector. End-to-end continuity between shells can be checked on an unconnected cable with a simple multimeter. In the absence of this ground path, Electrostatic Discharge (ESD) to any part of the sound system can cause clicks, pops and drop-outs.

These short YouTube videos by Brian Wynne demonstrate the problem and its solution:

X32 ESD Problem
X32 ESD Problem Various Tests
X32 ESD with Audio Playing
X32 ESD Final Test with Solution

[DwayneA]

(Risking spinning this topic into the rhubarb...)

Continuous AES50 shield connection through the EtherCon connectors requires secure bonding (or ground) between the power receptacles for the equipment at either end.

Case in point, a previously undetected poor ground connection at a stage 120V receptacle resulted in the AES50 cable becoming the primary ground connection between our console and stage located DL32. This caused the AES50 to become unreliable after about 30min of operation. Troubleshooting uncovered high resistance between the ground pins on the 120V receptacles at the console and DL32. Fixing the ground at the stage 120V receptacle fixed the AES50 reliability issues.

I expect this issue may arise any time the 'dis-located' equipment is fed from different panels/transformers/generators.

Final advice - engage a professional electrician or engineer if you think this scenario may apply to your installation.

D

[pvannatto]

Very interesting discovery that adds to the AES50 shielding continuity issue. Thanks Dwayne.

[GaryH]

@DwayneA Not to head off into the rhubarb myself but I have a few questions. What type and maker of aes50 cable were you using Cat5e, 6 etc? Are you also suggesting that the bonding of the aes50 cable to both rj45's at both ends is part of the problem in THIS particular case? In other words do you think the problem would not have surfaced had the ground been lifted at one side of the cable? Related to that question IF ethercons were NOT used at all would the problem have surfaced in this particular case? (As you probably know, we learned years ago it wasn't just shielded cable and rj45's that stopped ESD related issues, and that ethercons were also required) Last but most curiously, why do you think it would take 30 minutes for an unreliable connection to suddenly show up? Why not right away?

[DwayneA]

@GaryH
More info as requested. This will end up being quite a tome....

What type and maker of aes50 cable were you using Cat5e, 6 etc?

Our cabling is permanent installed shielded Cat5 with Neutrik EtherCon sockets. The patch cables are Amazon purchased shielded Cat 6 (or 7 or 8, I don't recall) fitted into Neutrik EtherCon shells. When we assembled the system we spend significant effort ensuring continuity between the EtherCon shells in the permanent wiring and the patch cables. When the AES50 started to act up in the first weeks following installation our first suspicion was continuity between the EtherCon shells at the two ends of the M32-DL32 connection. (more below)

Are you also suggesting that the bonding of the aes50 cable to both rj45's at both ends is part of the problem in THIS particular case?

No. I am suggesting that the continuous connection between the EtherCon shells HIGHLIGHTED a previously undetected house 120V wiring issue.

In other words do you think the problem would not have surfaced had the ground been lifted at one side of the cable?

Potentially.... That is the standard method of industrial instrumentation wiring - connect the shield at one end only. (We were most focused on adherence to the EtherCon shell continuity spec and as such did not even consider lifting one of the shield connections.) In this case if we had lifted the shield connection at one end we would not have uncovered the faulty 120V receptacle ground connection.

Related to that question IF ethercons were NOT used at all would the problem have surfaced in this particular case?

I cannot comment as stated above we remained focused on adherence to the spec.

why do you think it would take 30 minutes for an unreliable connection to suddenly show up? Why not right away?

In the course of troubleshooting we tested the AES50 shell continuity SEVERAL times, first by using house wiring ground as a return, then by using a known GOOD return conductor strung across the building on the floor. This was when we discovered that there was intermittent continuity between the 120V receptacle ground pins - the ohm meter would jump all over the scale with all players standing perfectly still - very frustrating. Questioning our understanding of building wiring we proceeded to testing ground pin continuity between random adjacent 120V receptacles expanding out to farther flung receptacles (using the roll of building wire as a return), always finding <3ohms impedance between receptacles. At this point we disassembled the 120V receptacle for the DL32 and traced its ground to a questionable connection. Securing that connection resulted in acceptable low impedance between the ground pins on the 120V receptacles for both the M32 and the DL32.

Having told this tale I can further report that the 120V receptacle on the stage for the DL32 was rarely, if ever, used for anything other than an organist's lamp before deploying the DL32. My belief is that the faulty ground connection on that receptacle has existed since installation years ago. I can also report that the before this event and repair, the receptacle passed the 3-lamp-duplex-plug-in-tester inspection.

30 minutes is an over-simplification. Sometimes it took 2h, sometimes less than 30min, but the AES50 connection would eventually fail every time we used the system. (Failure indicated by the green AES50 LEDs/indicators turning red then back to green....) My belief is that the AES50 cable became the ground equalizing any residual static charge buildup between the M32 and the DL32, which depending on environmental conditions could take varying amounts of time, and that equalizing current, however small, was what caused the AES50 to drop out.

The proof, as they say, is in the pudding. In the first month following installation the AES50 failed EVERY time we used the system, about 3times a week. Since repairing the ground on the suspect 120V receptacle in July of 2020, there have been NO AES50 dropouts. (In these COVID times the system is being used several times a week to record worship for web-broadcast.)

Finally, for the purpose of completeness, I'll state my qualifications as a certified Canada Red Seal Alberta Electrician, and a Professional Electrical Engineer (P.Eng) registered in Alberta.

I think we are now TRULY in the rhubarb.
D

[GaryH]

@DwayneA I'm officially out of the rhubarb now Thanks for all the the extra details. I was thinking there could have been a cable issue as well as the faulty box you discovered. However it's quite clear you were more than a little careful with the correct cable and wiring! Interesting issue for sure and good to know about!

[MJKlein]

@GaryH wrote: ↑Mon May 24, 2021 4:15 pm
In other words do you think the problem would not have surfaced had the ground been lifted at one side of the cable?

Potentially.... That is the standard method of industrial instrumentation wiring - connect the shield at one end only. (We were most focused on adherence to the EtherCon shell continuity spec and as such did not even consider lifting one of the shield connections.) In this case if we had lifted the shield connection at one end we would not have uncovered the faulty 120V receptacle ground connection.

In all the radio facilities under Grady's control (broadcast engineer in the Boston area) we always terminated the audio shields at one end. 115v is a nasty thing to have going along for the ride. When I read your comment about the Ethercon shield continuity standard, I was thinking "there's gonna be trouble with that."

[Dan Mortensen]

@GaryH
More info as requested. This will end up being quite a tome....

(Snipped lots of good stuff)

Finally, for the purpose of completeness, I'll state my qualifications as a certified Canada Red Seal Alberta Electrician, and a Professional Electrical Engineer (P.Eng) registered in Alberta.

I think we are now TRULY in the rhubarb.
D

Dwayne, thanks for that great post showing not only the solution and proof of solution to your problem, but also the process of identifying the problem which is valuable by itself.

Since this kind of problem is solved only by being deep in the rhubarb, given that it is a subtle and precise set of circumstances that have to be in place in order for the system to properly work, we all must be rhubarb farmers to some extent where knowing the proper techniques is important.

Just as Brian Wynne in the videos that are posted in the OP of this thread was the first one to fully identify and figure out the solution, you are the first AFAIK to confirm that outlet ground is very important to successful use of AES50 stage boxes, even though it would be easy to assume that as a given.

If there's no Ethercon ground to dissipate the interference, and there's no solid AC ground for the Ethercon ground to solidly connect to, then you'll have ESD problems if there is ESD in the local situation.

If you don't have ESD in the local situation, then you won't have a problem until you do.

Question: Did you identify a different ground-neutral potential at that faultily-wired plug when ground was open than when ground was connected, and vs. the one at the console end?

One thing that I believe has never been answered is what happens when the stage box is connected to an outlet with a significantly larger ground-neutral potential than the one at the console?

And that brings us to MF Klein's comment that having console solidly grounded through snake to stage box can be source of problems:

Hard experience has shown that more problems have occurred from the absence of that connection than from its presence. If your console is in a location that is warm and humid, then you may not have ESD problems, which are often as simple as shuffling feet on carpet and then touching a grounded object.

If that NEVER happens where you are, you might be able to avoid the continuous snake ground.

If that OFTEN happens where you are, especially in winter, then you are asking for mysterious problems if you don't follow Brian Wynne's videos.

In my experience.

Thanks again for sharing that with us, Dwayne.

[MJKlein]

And that brings us to MF Klein's comment that having console solidly grounded through snake to stage box can be source of problems:

Not ground - audio shield. Virtually all of my experience is in fixed facilities - from recording studios to a talk radio network control point. We always disabled the "Edison" ground (the AC mains system ground) and star grounded all equipment with it's own home run to the star point. We terminated the audio shields at the destination. IMO there should always be a separate ground system that is completely separate from the audio, designed to carry everything not associated with protecting the audio from RFI. I think that's the main issue. The snake is acting both as an audio shield and electrical ground.

[Dan Mortensen]

And that brings us to MF Klein's comment that having console solidly grounded through snake to stage box can be source of problems:

Not ground - audio shield. Virtually all of my experience is in fixed facilities - from recording studios to a talk radio network control point. We always disabled the "Edison" ground (the AC mains system ground) and star grounded all equipment with it's own home run to the star point. We terminated the audio shields at the destination. IMO there should always be a separate ground system that is completely separate from the audio, designed to carry everything not associated with protecting the audio from RFI. I think that's the main issue. The snake is acting both as an audio shield and electrical ground.

Both your first and last sentences are true, and the last sentence repeats what I said.

Whether or not it's best practice in normal audio world, it has been shown to be necessary for this console system we have adopted, or which has adopted us. That's my point.

And this is my 7th post on this forum, so that means it just gets posted?

Edit: Yay, it did! I'm a full-fledged member!