LED Lighting/Sports Venue NEC Violation/Safety Issue

[Jcdunson]

I’ve installed an LED lighting system for a major sports venue and the lighting manufacturer has a major safety issue and they refuse to provide data ie: internal fixture wiring schematic so that a solution can be found. The fixtures mount to a steel structure, are made of aluminum and the fixture to structure mate surface of fixture is painted(don’t know why). I’m recording a potential difference of 120 volts between stainless steel screws on exterior of fixture to steel structure. The manufacturer has installed #14awg bond “stingers” from fixture screw to steel structure in an attempt to “make issue” go away. Have noticed since this was done that insulation on #14awg “stingers” is starting to melt at termination(crimp) points. Insulation of 14awg is THHN. I think manufacturer is using equipment ground/ structure as neutral(current carrying conductor) which is a violation of the NEC and a major safety issue. Comments, thoughts are welcome. Thanks in advance. James D.

 

[don_resqcapt19]

Are the lights a listed product? If so take the issue to the listing agency...if not remove them as the NEC requires all lighting to be listed.

 

[zbang]

Should be really simple to meter from isolated fixture leads to the case and EGC; anything showing continuity from a supply lead to the EGC (or case) is obviously defective.

BTW, I have a very strong distaste for things that "make a problem go away" without actually solving the root problem.

 

[Todd0x1]

This is odd. You installed the fixtures, how are they wired? Have you thrown an amp meter on the 14ga bond wire, if so does the current match what the fixture neutral would see? With a fixture disconnected do you have continuity between any conductors and the housing?

 

[hbiss]

I assume the fixtures are 208 or 240 with no neutral and no separate EGC was pulled?

-Hal

 

[Jcdunson]

Fixtures are 277V w/ equipment ground pulled from panel to crossarms. Have not yet put amp meter on “bond stingers” from fixture to crossarm. Will be doing this 10-21-2020. Fixtures are ETL listed. The equipment ground from crossarm to panel is where I intend to measure current(approximately 109x fixtures on large towers and 86x on small towers) so should be substantial current flow if fixture component(s) looking for neutral path. Was told by mfg. that drivers operate at 52VDC, fixtures have internal cooling fan with variable speed control depending on temperature(internally controlled). Fixtures are computer controlled via a server/touchscreens and a fiber optic connection to a data box. Each fixture has a DMX cable to data box.

 

[zbang]

Do not rule out the possibility of a manufacturing defect, and if you have a spare or easily-detached unit, I'd test it on the bench (fully isolated, of course).

 

[Todd0x1]

Is the voltage present with the power on and lights off, or only when the lights are operating?

 

[Jcdunson]

Towers have 3phase 480V panel which is activated via a 400amp or 300 amp contactor(depending on tower) at service level. Fixtures can be on and dimmed to zero and voltage is still present from fixture to structure. With contactors off no voltage will be present. Have bench tested and have voltage to ground. Owner of our company reluctant to open fixture due to proprietary status of fixture internals and mfg. refuses to provide even a partial schematic/one line drawing. Think AHJ will have to force issue.

 

[Todd0x1]

I am very curious as to what this is. Probably something stupid inside the fixture like a missing insulator sheet or a metal screw used where a plastic one was supposed to be. Too bad you can't open one up. It is really surprising the mfg isn't being more helpful.

On your bench test, did you have the DMX cable attached? What is the wiring from the crossarm to the fixtures? Is the driver and electronics in the fixture or crossarm?

 

[zbang]

. Have bench tested and have voltage to ground. Owner of our company reluctant to open fixture due to proprietary status of fixture internals and mfg. refuses to provide even a partial schematic/one line drawing. Think AHJ will have to force issue.

If you're getting voltage to ground on the bench, IMNSHO they're defective (do you see a low resistance from power leads to the case?). Might be worth sacrificing one to open it and look around.

 

[winnie]

What instrument are you using the measure the 120V to ground? A high impedance meter, a low impedance meter, or something else?

It is _possible_ that the 120V to ground is the result of generally negligible leakage current which shows up as 120V when using a high impedance meter, but which will go away with a load applied. This is often termed 'phantom voltage' and seen for example in completely disconnected wires sharing the same conduit as conductors that are in use.

A ground connection is a perfectly acceptable method of eliminating this sort of phantom voltage.

The evidence of melting insulation suggests high current flow that is _not_ phantom voltage, however unless these are absolutely _huge_ lighting fixtures somehow leaking their entire supply current through the 14ga pigtail the current flow would be way too low to overheat that wire.

So I'd be looking for some sort of phantom voltage that isn't a real safety issue, and some other cause for the wire showing melting insulation.

-Jon

 

[Todd0x1]

What instrument are you using the measure the 120V to ground? A high impedance meter, a low impedance meter, or something else?

It is _possible_ that the 120V to ground is the result of generally negligible leakage current which shows up as 120V when using a high impedance meter, but which will go away with a load applied. This is often termed 'phantom voltage' and seen for example in completely disconnected wires sharing the same conduit as conductors that are in use.

A ground connection is a perfectly acceptable method of eliminating this sort of phantom voltage.

The evidence of melting insulation suggests high current flow that is _not_ phantom voltage, however unless these are absolutely _huge_ lighting fixtures somehow leaking their entire supply current through the 14ga pigtail the current flow would be way too low to overheat that wire.

So I'd be looking for some sort of phantom voltage that isn't a real safety issue, and some other cause for the wire showing melting insulation.

-Jon

Very good point. If the fixture housing is poorly bonded, and there are filter caps L-G as part of the power supply or driver that could explain it. I wonder if the melted insulation on the 14ga is something as simple as the mfgr that made up those jumpers soldered them together, or used a solder pot to tin the wires. Unlikely, but possible.

 

[zbang]

So.... anything happening with this? I'm curious.