Fluorescent light question

[Joe]

I've read where it's common for fluorescent lights not to work well when it's cold, but how about when it's hot out? I have 4 sets of twin 4' tubes that came on fine this winter (temperatures into low 30s), but when it got to 78 or so they got iffy. If turned on early in the morning when it's 65, they pop on as expected, but if I wait until later to get into the shop I get to play either the on-off-on or 'massage the lights' game. Curiously, I also have two sets of four 2' (?) tubes in ceiling-mounted fixtures on a different circuit inside the house that do the same thing. No buzz, hum, or flickering. This has gone on for years, by the way. I've shuffled bulbs between the fixtures and tried new bulbs. No change. Any ideas out there? I've learned a lot about fluorescents, but so far no answers by Googling.

[Leon]

On 4/25/2012 3:36 AM, Joe (temperatures into low 30s), but when it got to 78 or so they

Temperature in this case is not affecting the actual florescent lights in the same way. Basically even in the cold florescent lights come on but they simply do not come on bright until they warm up. Basically the warmer they are the better they work.

What you are experiencing is most likely a starter or ballast problem and especially if you have replaced the tubes and this has been going on for years. If these are old fixtures I would look for a starter to be replaced. This will be a small cylinder about 1.25" long and 3/4" in diameter and it will twist to unlock.

[Puckdropper]

Joe (temperatures into low 30s), but when it got to 78 or so they

Do you have the type that uses a starter? It might be worth changing that out (last I knew they were $.79 at the local hardware store, but that was 2005).

I've got to wonder if bulb type has anything to do with it. Are your fixtures rated for T12 bulbs and using the same? (Or T8 etc.)

Puckdropper

[HeyBub]

Florescent lights require a perfect ground.

Something to do with the galactic flux.

[Steve Barker]

not sure about the flux, but they do require a ground. I had a bunch strung up temporarily on two wire extension cord and had to go around each day and touch each one to get them to come on. Finally i strung a single strand of copper around to each one and grounded it and they never failed to start again.

[<jmohnike]

Check your switch. Solved my problem. Joe "Joe >" (temperatures into low 30s), but when it got to 78 or so they

[Joe]

Some good ideas, no winners, though.

No starters, just the ballasts. Used starters on the fish tanks for years so I knew what to look for. sorry I didn't mention it to start with.

3 of 4 are grounded using a 3-wire cord, the other has the ground wire attached to a screw. Connection is good. I could try a 12' ground rod through the floor, I guess. They have Sylvania WORKSHOP F40 bulbs that are always replaced in pairs except for a GE SHOPLIGHT I just stuck in - it does the same thing. Switch good - when turned on in warm weather they glow dimly, then one or the other may come on.

-J

[Scott Lurndal]

Joe mention it to start with.

The ballasts may be overheating. is there air space between the fixture and the ceiling?

Otherwise, the ballasts probably need replacement.

scott

[Artemus]

"Joe >" mention it to start with.

It's possible that your ground wire isn't actually connected to a good solid ground. Easy to test and risky not to. Art

[tiredofspam]

Run a ground wire to your pipes... maybe the ground is faulty.. You can check with a meter and see if hot and ground show voltage. If they do your good.

On 4/25/2012 4:56 PM, Joe mention it to start with.

[Sonny]

Buy a new ballast for one, install it and see if there is any change, for the good, in that light. If so, change the ballasts. If not, then it's probably a ground issue.

Replacing the fixtures, maybe? I bought seven 4-bulb fixtures (T-8 bulbs) for $35 a piece at Lowes. I didn't think that was a high price. I got one "damaged" fixture (broken bulb only) for $10. Check for damaged fixtures for discount purchasing.... at Home Depot, also.

Sonny

[Joe]

The shop lights hang at 8' with about 7' of space over them to the roof. There's about a 1/2" opening the whole length on both sides between the ballast and wiring and the lamp reflector for ventilation. When it's cool tomorrow I'll re-check the grounding. The outlets that they plug into show normal conditions when I plug in one of the LED testers that shows reversed, missing ground, etc. Probably time to present the concept of new lights to the 'finance committee' and see what she says. The lights on the inside of the house are what I would describe as kitchen lights - four 2' tubes in a square box mounted on the ceiling (not recessed) covered by a translucent panel. J

[clare]

Have you checked the grounds???

[HeyBub]

"3. Grounding of a ballast and the luminaire [complete lighting unit] can also aid in proper starting of the fuorescent lamp. This is especially important when magnetic fluorescent ballasts are utilized. The metal refector provides a capacitive path to ground through the wall of the fuorescent tube. This helps to ionize the gases inside the tube and initiate conduction in the tube. However, once current is fowing from end-to-end, the impedance in the ballast circuit is much lower than this capacitive path. Thus, the added capacitance is irrelevant once the tube has started."

[Joe]

Checked grounding this morning starting at the shop end and while I was at it went all the way through the house to the meter drop. All is solid.

When I bought the place 20+ years ago, I got a 'wiring fault' error on a UPS. Found just one wire serving many rooms along the outside wall of the house wasn't grounded. Easy fix, but surprising all the same. If the lights hadn't been strange the beginning, I'd suspect the ballasts were old and tired by now, With a set of inside ones of different style and age that do the same thing, and another set that works fine, the mystery deepens. Joe

[Keith Nuttle]

I can agree to that. I worked six months trying to solve a drain problem, and "Thought" I had checked everything. Then someone on this group suggested the sink overflow. It was plugged. When cleared everything worked normall

[Joe]

Thanks all. That's one of the best rules of thumb. Another is you gotta believe your meter.

ground. It is very consistent. Back to the recommendations that have already been made to you. The hardest part in troubleshooting this kind of thing is sometimes the ability to ignore what you think you know, and simply go back to the basics as if you didn't know... I have not yet tried wrapping some wire around the whole assembly, but it seems reasonable, and might provide the inductance needed similar to playing touchie-feelie with the bulbs.

also aid in proper starting of the fuorescent lamp. This is especially important when magnetic fluorescent ballasts are utilized. The metal refector provides a capacitive path to ground through the wall of the fuorescent tube. This helps to ionize the gases inside the tube and initiate conduction in the tube. However, once current is fowing from end-to-end, the impedance in the ballast circuit is much lower than this capacitive path. Thus, the added capacitance is irrelevant once the tube has started." :) :) :) Love it! If the voltage gets high enoough to where skin effect comes into play, I won't need the bulbs - there should just be a pleasant glow all around!

to ward off skin effect...

[G.W. Ross]

Glad to be of service. One of those little tidbits of experience that is useful maybe once in a lifetime.

[HeyBub]

Convince me. How did you "check" the grounds?

The only way I can think of - for sure - is to measure the resistance from the fixture to some outside dirt.

[Father Haskell]

Tube and the grounded fixture act like the plates in a capacitor. The tube won't fire reliably if one of those "plates" is disconnected.