Fluorescent troubleshooting

Page 1 of 2  
I have 3 identical fluorescent fixtures in the garage about 15 years old, 2 bulb 4' rapid start. A clue to their usage is that none of them had had any bulbs changed in their life.
One of the units showed intermittent flickering sometimes one bulb was good/bright, sometimes dim, the worse/dimmest bulb had a dark spot at the end.
So I changed the bulbs, with no improvement. Used another new bulb. Same thing. Took two working bulbs from the other fixture and put them in. No improvement. I began to suspect something was wrong with the ballast, but I wasn't smart enough to do any troubleshooting. All I have is a little volt/ohm meter I use to troubleshoot computer power supply problems.
So, I went to Home Depot and bought a ballast aided by an experienced sales person who understood what fixture and bulbs I had. The new ballast was not as 'fat' as the other but was the same length and color coding of the wires.
I disassembled the fixture/ballast and replaced the ballast using wire nuts acquired during the ballast trip and reconnected, reassembled, and resecured the ceiling fixture.
No improvement.
Now I don't know if my original ballast was good or bad and/or if the replacement ballast is good or bad. I would like to do some of the diagnostic testing I didn't do during my original guesstimation that it was/ must be/ the ballast.
How can I:
- benchtest my old ballast I removed - test the integrity of the new ballast installed - troubleshoot my malfunctioning fluorescent
If I could satisfactorily prove to myself that the new ballast isn't good, then I could return it; but under the present conditions I have no clue what is wrong.
--
Mike Easter

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Hate to go through this, but swap into your known 'good' fixtures. If it doesn't destroy it, it was ok.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 4/30/2012 4:27 PM, Robert Macy wrote:

Swapping bulbs that work from a working fixture and swapping bulbs that don't work from another fixture is a real good test.
A black end on a fluorescent tube generally indicates the filament is not heating. Might be burned out (try measuring resistance between pins). Could be bad connection at socket or wires connecting to socket. Could be bad ballast, but that is least likely.
--
bud--


Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Even with the same color wires, the wiring may be differant. The normal black and white wires should be the same as they go to the voltage comming in. The wires going to the bulbs may not go to the same place as the same color on the new ballast. There should be a wiring diagram on the ballast. Also double check that the words on the ballast match the words on the bulbs as they may be differant. There are many kinds of ballasts and bulbs that look similar.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 4/30/2012 5:30 PM, Ralph Mowery wrote:

...
True but when it functioned the same as the old, the probabilities went way up that the ballast wasn't/isn't the problem and the new one was installed correctly and is ok...
In general, the symptoms are symptomatic of quick-start fixture w/ bad/missing ground or missing/improperly installed reflectors that are needed for the req'd capacitive starting field...
I'd suggest the following page to the OP...
<http://nemesis.lonestar.org/reference/electricity/fluorescent/trouble.html
--
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

Why mess with it? A whole new complete 2 tube fixture is under US$20 (at least it is at my local Lowes). It's an easy installation, and only 2 wires to mess with. In my case I modernized with a nicer looking fixture and the new slimmer instant on tubes.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I thought fluorescent fixtures had to have a good ground to work properly. wouldn't that be 3 wires?
----------------------------------------------------------- Get free backup service using Drop Box Access your files from your smartphone or tablet Bonus space provided using this link: http://db.tt/RIONkfbZ
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Tue, 1 May 2012 05:46:17 -0700 (PDT), Limp Arbor

It will work just fine with just 2 wires (no ground). My last house was wired in the 50's (with no ground wires) so I just ignored the ground wire on the fixtures I replaced.

But you're right in modern houses it's wise to use the ground wire for safeties sake. So I stand corrected at 3 wires. But still that has to be easier than trying to match wires on a very old balun.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Instant-on is, on its own, reason enough to replace an old fixture instead of fixing it.
--
Tegger

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 5/1/2012 9:15 AM, Mike Easter wrote: ...

It's almost certainly _not_ the ballast (prove it to yourself; put the old one you took from the troublesome fixture and put it in one of the working fixtures. Odds are the problem will stay w/ the old fixture.
Did you look at the troubleshooting guide at the link I posted? I still expect the most likely is the ground isn't what you think it is (as for how good or the reflector(s) aren't grounded or close enough to the tube in the bad one). Does it light or change characteristics if you touch the tube(s) while it's on? If so, that's pretty much conclusive.
--
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
dpb wrote:

I had actually looked at that guide in the beginning before I removed the fixture and replaced the ballast when I was trying to figure out whether or not to replace the ballast.

This is the same ground, big fat ground with the power, which is securely 'bolted' with a nut onto the fixture that has been working on the fixture for 15 years. We would have to imagine that something has 'gone wrong' with the ground wire's grounding condition. I don't even know how to test that.

No the light characteristics does not change.
Presently the light characteristics, new ballast, *old* (original) bulbs, are that the light is not flickering or variable, but that both bulbs are very 'dim' compared to the other lights.
The black ended bulb is dimmer than the other 15 y/o bulb.
With 2 brand new bulbs, the 'brightest' light is less than 1/3 as bright as the old lights and the other light is mostly very dim but a little brighter on one end, but still less than half as bright as the 1/3 bright bulb. There is no flickering and no variability when I 'fool with' the bulbs.
I'm puzzled about this bulb to bulb discrepancy with the new ballast.
--
Mike Easter

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 5/1/2012 11:01 AM, Mike Easter wrote:

...
...
It's not hard to conceive of the ground connection having developed some corrosion at the fixture end particularly in a garage or even a connection at the outlet where the fixtures are wired/plugged in.
Also, the ballast must make good contact for grounding as well. I'd try a little emery cloth to shine up the contacts and check all connections for being bright/tight first.
Are you sure you have the proper tube designations?
Also, it's possible w/ time the lamp contacts are dirty/loose and/or have some corrosion as well.
Have you checked the actual supply voltage to be sure there isn't something wrong w/ the supply--perhaps there's a loose or high-resistance connection in the circuit itself.
--
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
dpb wrote:

The appearance of the ground in the ceiling power connection is relatively shiny copper, no corrosion. The bolt and nut are not corroded. I can't imagine a discontinuity between the ceiling ground and its connection back to a breaker box ground, nor a breaker box ground problem. That wiring is off a subpanel inside the house. The location of these garage lights are a long way from 'outside' - the garage doors are always closed, no weather gets inside.

I would like to repeat that the two sides do not perform identically with the old or the new ballast. One side is almost not lit at all and what lighting there is, is on about 8" of bulb, while the other side is about 1/3 or 1/4 as bright as normal and uniformly distributed throughout the length of the bulb. If I switch the bulbs, the same side, not the same bulb, is as described.

Yes, except that the newer bulbs are 34W and the old bulbs are 40W.

The lamp contacts are clean to appearance and the problem does not change with twisting the contacts around.

I was going to work on/ measure/ the voltages (and resistance) at the ballast as soon I found some information beyond my knowing what the house voltage should be. The voltage at the ceiling fixture in question is bound to be the same as the other two, they are all on the same circuit.
I think the voltages that can be checked are those of the house current and also the output of the ballast. Also I think the resistance across the ballast (somewhere) is supposed to be zero. I'm trying to find some docs on that.
If there is some kind of ground problem here, I don't know how to find/prove it. The appearance and the physical security and overall 'beefiness' of the ground are all strong. It looks like a serious ground of good integrity and strong connection.
I have examined the specs in fine print on the installed ballast and it is like that of the removed one.
If I had to eat this ballast and disassemble a complete fixture which costs less than the ballast I bought and would be eating, it would be necessary for the fixture to have tombstones with slots which match the ones on my fixture. I don't really like that idea.
I also don't like the idea of swapping out a ballast from an existing working ceiling fixture. What I really want is some hard numbers on the installed ballast which satisfy the expected voltages input and output and resistance across somewhere on the ballast. I have no way to do frequency checking of the ballast output.
--
Mike Easter

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 5/3/2012 2:45 PM, Mike Easter wrote: ...

Have you actually taken the ground connection loose and checked it or simply looked at it? Just because it looks good on the surface doesn't mean there's not a resistance layer between the mating surfaces or a layer of paint or whatever.
Doesn't say there is, but unless you check you can't rule it out for absolute certainty. Same is true back to the point of the connections to the good fixture(s).

...
That still sounds very much like the reflectors or a bad ground or connection or the contacts at the faulty. That with a new ballast the same symptoms appear (iiuc) pretty much eliminates the ballast altho I still note the need for it to be well grounded and if there's a problem there on the fixture side it could well remain for both. Did you make sure there's not paint or dirt or whatever preventing a good contact or just toss the new one in where the other one was?

Well, when you won't do the simplest of easy tests to confirm where the problem goes, it gets pretty hard to diagnose when you reject both the chances that there might be a problem w/o actually checking and the chance to see if the problem follows the ballast or is at the fixture itself.
--
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 5/3/2012 4:25 PM, Mike Easter wrote: ...

Still really haven't answered the question or indicated you took the action(s) suggested.
Make sure there's bare metal between the two mating pieces not just the edge of a hole and use some emery and/or find sandpaper and polish off any possible surface layers.
You indicated in another posting you're using very old tubes; have you tried a newer tube perchance? Have you tried the above cleaning on the pins of the tubes to ensure they've not oxidized over the years of storage? (I have no real clue on aging of fl tubes but given what appears abnormal situation I'd look for possibly unusual fixes as well altho it still seems should be relatively straightforward.)
I don't know that you can test the ballast w/ just a VOM; I surely don't know any spec's to tell you. I judge that if the new and the old act the same the likelihood it's the ballast itself is quite low -- unless it is the connection or the power in; I think that's barking up the wrong tree meself.
W/ all fixtures I've had, I've never had to take one down to swap ballasts so I don't quite follow why that's such a problem.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
dpb wrote:

The original design of the ceiling fixture was to affix the ballast to the fixture with a screw/bolt on the top side against the ceiling and a smooth round 'nut' (I don't know the name of that kind of fastener) on the down side exposed after the enclosure for the wires is removed. I didn't change that design/fastener (yet), so one has to remove the ballast by removing a bolt from the side of the fixture against the ceiling.
In addition to that inconvenience, the screws for the toggle nuts securing the fixture to the ceiling are on the other side of an enclosure for the tombstones, so both of those enclosures need to be removed to remove the fixture to remove the ballast. The enclosures are snap-in, so the sides of the fixture need to be spread apart to snap them out; which would be easy in the middle of the fixture's length, but of course this prying is on the ends where the metal is joined, so it doesn't want to spread there.
This fixture does not have a reflector.
In any case, I have more toggles and the fixture is off the ceiling again so that I can not only bench test it down instead of on the ceiling, but I can also remove the new ballast and replace it (see below/later post) and I also now have the ability to connect the ceiling fixture on the bench to test it for working condition.
--
Mike Easter

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 5/4/2012 5:42 PM, Mike Easter wrote: ...

,,,
[tales of woe re: sorry-ass design of fixture elided for brevity]
Dang, sounds like a reason to just replace the fixture for that reason alone... :) (sorta')

That is good...
--
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 5/4/2012 3:48 PM, dpb wrote:

move it over a few inches and screw it into a rafter?
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

I think I'd save on toggles and time and just replace the whole fixture with the toggles in an accessible position and be done with it.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Related Threads

    HomeOwnersHub.com is a website for homeowners and building and maintenance pros. It is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.