Previous in Forum: Subtransient Reactance   Next in Forum: Pressure Test on Transformer
Close
Close
Close
11 comments
Rate Comments: Nested
Guru
Technical Fields - Technical Writing - New Member Engineering Fields - Marine Engineering - New Member

Join Date: Sep 2008
Location: Vancleave, Ms about 30 miles inland from Biloxi and the coast
Posts: 3197
Good Answers: 106

Fluorescent Light Fixtures-Inconsistant Operation

09/28/2012 12:03 PM

My shop has 8 twin lamp tubes each 8' long (total 16 lamps). When the temperature is mild and humidity is low, all the lamps go on at once. When it is humid and hot, only some will light and others will come on after a few minutes. If I "flick" the switch off/on, I can sometimes get other lamps to light, but not all. What can I do to fix this problem?

__________________
Mr.Ron from South Ms.
Register to Reply
Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.

Good Answers:

These comments received enough positive votes to make them "good answers".

"Almost" Good Answers:

Check out these comments that don't yet have enough votes to be "official" good answers and, if you agree with them, vote them!
Guru

Join Date: Jun 2010
Location: New York
Posts: 972
Good Answers: 23
#1

Re: Fluorescent light fixtures-inconsistant operation

09/28/2012 12:05 PM

From askmehelpdesk.com

"I suspect the fixture is not grounded that is causing your problem, the humiditiy being the indicator.

The lamp starts by using the capacitance that exists between the lamp glass wall and a metal grounded surface.

Check the wiring feeding the fixture. If you have a green or bare ground wire, be sure it is connected the the metal frame of the fixture using a machine threaded screw.

Test the ground wire to be sure it is grounded by using a voltage tester and measure 120 volts across the black wire and the ground.

If you have BX cable, the metal sheathing is the grounding conductor, and the connector's locknut must be tight to dig into the paint on the fixture for proper ground connection. BX is popular for bad grounds, as the connections rely on every clamp and connector in the circuit to be perfectly tight.

If opening the fixture to work on the wiring, and possibly needing to open other outlet and junction boxes is beyond your capabilities, you may need to call in an electrician to do the troubleshooting and repairs."

__________________
The first law of thermodynamics is you do NOT talk about thermodynamics.
Register to Reply Score 1 for Good Answer
Guru

Join Date: Jun 2011
Location: Phnom Penh
Posts: 4019
Good Answers: 102
#5
In reply to #1

Re: Fluorescent light fixtures-inconsistant operation

09/29/2012 1:41 AM

The lamp starts by using the capacitance that exists between the lamp glass wall and a metal grounded surface.

No it doesn't. Flourolamps will still work without a ground connection. Polycarbonate enclosures seem to work just fine without a ground.

The ground is there for protection in case the iron core ballast has a meltdown episode and makes the metal enclosure live.

__________________
Difficulty is not an obstacle it is merely an attribute.
Register to Reply
3
Guru
Popular Science - Cosmology - New Member Technical Fields - Technical Writing - New Member Engineering Fields - Energy Engineering - New Member Engineering Fields - Electrical Engineering - New Member Engineering Fields - Control Engineering - New Member Engineering Fields - Electromechanical Engineering - Old Member, New Association

Join Date: Apr 2008
Location: Lexington, KY
Posts: 1639
Good Answers: 73
#2

Re: Fluorescent light fixtures-inconsistent operation

09/28/2012 12:43 PM

Great question! In general, the 8' lamp has no starting aid except for high voltage. If you still have magnetic ballasts, then the "flick" (when done near the peak of the AC sinusoid) will cause an inductive transient that boosts the voltage for a few microseconds. Unfortunately, that is difficult to do by hand because the incoming power is a moving target at 50 or 60 cycles per second.

The newer electronic ballasts don't provide this little boost.

There are two things you can try to help the situation. The first one is easier and cheaper than the second.

First, you can provide your own starting aid directly to the lamps. Some manufacturers have actually sold lamps with this applied and many people had no idea what it was all about. So, first I'll tell you what it is and then I'll explain it. You can read about starting aids in lots of technical literature if you like. It consist of a conductive (metallic) strip applied directly to the lamp from one end to the other (but not touching the metal base) in the form of a tape. This is the same tape used to detect broken window glass. Or you can get some copper tape from your local stained glass store.

The way this works is to allow a build up of ionized gas by capacitively linking from each coil (electrode) through the glass, and then through the tape down to the other end where it capacitively links back to the other coil. The wider the tape, the better the linkage. But I have also done this with a piece of wire taped on from one end to the other. A lot depends on the proximity of the coil to the conductive surface on the outside so you might want to shadow box the ends of the lamp to figure out where the ends of the coils are closest to the glass. Capacitive linkage depends on distance and surface area.

The second way to do this is more likely to work but it may cost a little more unless you just happen to have this stuff lying around. What you need to do is to temporarily or permanently boost the supply voltage to the ballast circuits. This will help even as the humidity rises (more on that in a second). If your line voltage is supposed to be 120 VAC there is a 10% tolerance band which the power company is supposed to keep within. That means that all your devices should operate anywhere from 108 VAC to 132 VAC. Your ballast is also supposed to be designed to operate without any problems within that range as well. But the lamps always start faster when you are in the higher end of the tolerance range. So if you increase the line voltage to the nominal value +10% they will always start. They may also be a little brighter and run a little warmer due to the increased supply. Once they start, you could drop back down to nominal. It all depends on what you want to do.

If the glass accumulates water vapor from high humidity it can offer a capacitive linkage (when attempting to start) with a high enough resistance from one end to the other so that it has the opposite effect of the starting aid mentioned above. One of the ways to combat this humidity problem was to apply a chemical similar to silicone (or car wax) that would cause the water to form droplets that were not connected, thereby preventing the starting problem.

Either solution mentioned above should help but the voltage boost will work best.

Before a lamp starts, the open circuit resistance in in the mega-ohm range. As it begins to light by ionizing gas inside, the resistance drops very rapidly to just a few hundred ohms. The trick is getting enough ionization to happen so that it creates a more conductive path from one end of the lamp to the other. I'll be watching to see how you do.

__________________
A great troubleshooting tip...."When you eliminate the impossible, whatever remains, however improbable, must be the truth." Sir Arthur Conan Doyle
Register to Reply Good Answer (Score 3)
Guru
Technical Fields - Technical Writing - New Member Engineering Fields - Marine Engineering - New Member

Join Date: Sep 2008
Location: Vancleave, Ms about 30 miles inland from Biloxi and the coast
Posts: 3197
Good Answers: 106
#3
In reply to #2

Re: Fluorescent light fixtures-inconsistent operation

09/28/2012 1:22 PM

Thank you for the information. I'll give it a try and let you know how it works out. BTW, the fixtures are grounded and all electrical is to code.

__________________
Mr.Ron from South Ms.
Register to Reply
Guru

Join Date: Aug 2012
Location: Land of Fruits and Nuts
Posts: 4481
Good Answers: 54
#4

Re: Fluorescent Light Fixtures-Inconsistant Operation

09/28/2012 11:03 PM

One thing you should check is to the capacitor. Swap the capacitor from the weak fixture to the strong one and see if it makes a difference. In a high humidity condition, some of the high voltage (needed to start the light) will leak. Think about this - the water vapor will allow some current to flow, which is more prevalent at high voltage. A 1.5V battery will experience no effect from humidity, because 1.5V isn't enough to "push" current through humid air.

If this works, you'll need to get new capacitors for all the fixtures.

Good luck!

__________________
Enjoy and be happy! Life is too short!
Register to Reply
2
Guru

Join Date: Jun 2011
Location: Phnom Penh
Posts: 4019
Good Answers: 102
#6

Re: Fluorescent Light Fixtures-Inconsistant Operation

09/29/2012 1:50 AM

You didn't mention whether your fittings had iron core ballasts and starters or electronic ballasts,,did you?

Assuming the former....

The lightswitch flick trick is the big clue here.

I reckon some of your starters are stuffed and will need replacing, or some of the tubes have open circuit heater elements which can be remedied by reversing the tube in the fitting so the heater element at the other end (unless both ends are expired) can do the starting preheat.

__________________
Difficulty is not an obstacle it is merely an attribute.
Register to Reply Good Answer (Score 2)
Commentator

Join Date: Jul 2010
Location: Scottish Borders
Posts: 66
Good Answers: 1
#8
In reply to #6

Re: Fluorescent Light Fixtures-Inconsistant Operation

09/29/2012 4:54 PM

Just tried what you suggested, by swapping the tubes end for end, have now given a new lease of life to 5 of my workshop lights. I will try the same with some other old tubes before chucking them. Thanks you have saved me a few quid.

Will give you a GA when I can get it to work.

Register to Reply
Guru

Join Date: Jun 2011
Location: Phnom Penh
Posts: 4019
Good Answers: 102
#10
In reply to #8

Re: Fluorescent Light Fixtures-Inconsistant Operation

09/29/2012 10:43 PM

I learnt that "trick" from my father (RIP).

Teach this to all your children.

__________________
Difficulty is not an obstacle it is merely an attribute.
Register to Reply
Commentator

Join Date: Jul 2010
Location: Scottish Borders
Posts: 66
Good Answers: 1
#11
In reply to #10

Re: Fluorescent Light Fixtures-Inconsistant Operation

09/30/2012 10:23 AM

I will do, assuming that flurescents are still around when they have grown up.

My 9 year old currently has designs on my Tig welder ( something about an off road buggy he wants to build ). If he wants to do it I won't stop him.

Register to Reply Off Topic (Score 5)
Guru
Technical Fields - Technical Writing - New Member Engineering Fields - Marine Engineering - New Member

Join Date: Sep 2008
Location: Vancleave, Ms about 30 miles inland from Biloxi and the coast
Posts: 3197
Good Answers: 106
#9
In reply to #6

Re: Fluorescent Light Fixtures-Inconsistant Operation

09/29/2012 5:59 PM

Iron core ballasts; no starters.

__________________
Mr.Ron from South Ms.
Register to Reply
Guru

Join Date: Jul 2006
Location: Eastern Kansas USA
Posts: 1503
Good Answers: 128
#7

Re: Fluorescent Light Fixtures-Inconsistant Operation

09/29/2012 1:23 PM

Hello, Ronseto,

New fluorescent tubes have a very thin dust and water repelling coating on them. I have found that when they get dirty the starting is harder. Try taking each tube out and cleaning it (no soap) then putting them back in. The manufacturers all state the need for a grounded surface typically within 1" (25mm) of the tube. In denying the need for this, Wal is unfortunately wrong this time. If your tubes have a single large pin exposed on each end, then there is no heater in the tube and switching it end-for-end is useless. High-output and very high-output (HO and VHO) tubes do have heaters. If you are getting power to the ballast (and its the older magnetic type) it will be warm but should not have tar leaking out of it.

Look at the ends of the tubes and see if there is any band of blackening in the first few inches (cm's) of length--if so, the tube is old and will be more difficult to start. Another possibility is that the ballast is getting near the end of its service life. I have also seen problems with the socket, so check these to make sure that there is no evidence of heat damage or loose wires.

In order, I suggest:

  • If the ballast is magnetic and cold, check for power failing to go past a disconnect lamp holder.
  • If the ballast has tar leaking out, replace it.
  • Clean the tubes and check the sockets.
  • Swap tubes from a working fixture to one that isn't working.
  • If the tubes light in their new location, replace the tubes that weren't working
  • If the tubes do not light, replace the ballast.
Register to Reply
Register to Reply 11 comments

Good Answers:

These comments received enough positive votes to make them "good answers".

"Almost" Good Answers:

Check out these comments that don't yet have enough votes to be "official" good answers and, if you agree with them, vote them!
Copy to Clipboard

Users who posted comments:

Autobroker (1); Brian Falconer (2); jmueller (1); Mizuti (1); NotUrOrdinaryJoe (1); ronseto (2); Wal (3)

Previous in Forum: Subtransient Reactance   Next in Forum: Pressure Test on Transformer

Advertisement