GFCI Protected Fluroscent lighting

[imtxn]

We have several high output fluorescent fixtures, Hydra Farms #FTL-44 (125vac@ 2amps), supplying light in a wet lab which grows algae. They are in close proximity to each other mounted vertically next to clear cylynders which hold the algae. Due to the wet conditions the system was to be powered By 20 Amp GFCI breakers. When testing, the lighting the system worked and the GFI's opperated. After about and hour each breaker tripped. Suspecting an overload I clamped a circuit with a Fluke #36. Max hold was selected. Trying again, the circuit tripped at 14,1 amps. I wired one of the circuits directly to a GFCi outlet, it tripped too. I then tested each breaker with an ideal GFCI tester and the breakers all tripped at 7 ma. (the breakers are good). Next, I jumpered 6 fixtures with a #12-3 SO cable and wired it directly to a Breaker, it also tripped (the wiring is good). The system will operate constantly with non GFCI breakers. I am suspecting it takes awhile for the fixtures to turn on fully and when they do they get a fault.

Do any of you folks have another idea?

 

[480sparky]

Remove the GFI if the fixtures are hard-wired. Fixtures are not required to be protected.

 

[iwire]

Do any of you folks have another idea?

Remove the GFCI breakers and rock on. There is no NEC requirement to use GFCIs for florescent fixtures.

My guess is capacitance coupling between the lamps and the grounded reflectors or moisture on the sockets is causing your trips.

 

[imtxn]

GFI's protecting lighting

GFI's protecting lighting

Thanks for the fast replys guys. Let me describe the application further. The lab is 12'x24' with 48 fixtures. There are eight 14" wide x 5' tall clear fiberglass cylinders along the two long walls. Each cylinder is surounded by 3fixtures mounted vertically 1 or 2 inches from the cylinder. They will be filled to 4.5' high with water and algae. I need to protect my people if a cylinder breaks and floods the fixtures.

I realize the code doesn't require a protected circuit. What I don't know is what I would tell Kim's husband Mark if she was shocked and died.

 

[imtxn]

GFI's proteceing fluos'

GFI's proteceing fluos'

Further: Initally, the GFCI's operated perfectly. They trip after operating about an hour. Me

 

[480sparky]

With any HID fixture, there's a ballast involved. A GFI will not detect any shock or fault on the load side of the ballast. It can only protect the 120v primary side.

So even with a GFI, you're only partly protected no matter what you do.

 

[imtxn]

GFI Protection

GFI Protection

Thanks eh Sparky & Bob,
We are in the test stage at this point, no water in the area yet. No help from the manufacturer's techs yet, more than likely Bob's thought about cap coupling is the best possible. We will go with standard instantanious trip breakers as you suggested. The fixtures are mounted in boxs right now and we plan to install 1/16" clear acrilic plate in front of the fixtures to prevent any flooding potential from the cylinders. This should protect Kim and her guys and abate my insecurity.

Me

 

[iwire]

With any HID fixture, there's a ballast involved. A GFI will not detect any shock or fault on the load side of the ballast. It can only protect the 120v primary side.

So even with a GFI, you're only partly protected no matter what you do.

Hmmm, at first I was on board with that but aren't HID ballasts auto transformers so that are not SDS?

I do not know, I am asking.