I have two adjacent flourescent fixtures of two 40W bulbs each. Both
fixtures (all four lamps) went out at once. The fixtures are both getting
120v and there is a slight humming noise from one (or both?) of the
Well, I can understand one of the ballasts going out but not both at once.
The black and white leads are connected to the fixtures and the voltmeter
shows 110vac on them. The ground wire is attached to the fixture also and a
voltmeter shows 0v between the fixture and a good remote ground (plumbing)
when the power is applied.
The ground is a safety feature but don't see how it could prevent the lamps
from lighting even if the ground was defective. What is your thinking on
how a bad ground could cause the problem?
The voltage measurement between the fixture and a good ground does not
indicate the fixture is grounded. You need to measure resistance. Also
don't every assume plumbing is a good ground. In most newer homes it is not
and in many older homes it is not. It may not be grounded at all.
Trust me on this. Most units need a ground. Not all, many of the new
electronic ballast don't need it, but the magnetic ones do. It is need to
set up the magnetic field.
Sounds like you lost the earth ground connection to the lites.
Or, it could be the cold, but not real likely unless something
changed in your environment. Usually low temps make them get
dimmer, then you'll see spirals inside the bulbs and finally they
wont' come on when it gets cold enough depending on the temp, age
of the bulbs, xfmr, and ground connections.
KILL POWER BEFORE THOUCHING ANYTHING! You don't sound real
|I have two adjacent flourescent fixtures of two 40W bulbs each.
| fixtures (all four lamps) went out at once. The fixtures are
| 120v and there is a slight humming noise from one (or both?) of
| Well, I can understand one of the ballasts going out but not
both at once.
I investigated the missing ground theory in howthings work" and was
fascinated by what it said. Seems that the flourescent metal box should be
grounded to give reliable ionization of the tubes. that's a new one on me
and still puzzles me. E.g., I have a two tube flourescent fixture in my
garage with a two prong plug feeding it with no problem.
Well, I lashed an extra ground to the fixtures with no effect. Then I
bought one new fixture as that was as cheap as a ballast, my second suspect.
Bought two new tubes too, just in case. Just for grins I tried replacing
the tubes and you guessed it - lights came on. Problem is that I only
replaced TWO tubes and lost track of whether they both went into the same
fixture or one into each fixtures.
Also, all four of the old tubes had continuity through the filaments on each
I don't want any more problems like this!
Magnetic florescent fixtures have always been a nuisance. Especially 2 light
four footers, with all the connections on the ballast, four contacts on each
lamp and eight contacts between the four sockets, there's a lot of week
links. I too have had incidents where I'll change bulbs and ballasts and the
thing still doesn't work. What is nice is the current electronic ballasted
florescent fixtures. They use a 32 watt T8 lamp instead of the 34 -40 watt T
12. The ballasts are in like their fourth generation, and really work well.
Each lamp is in parallel as well so one lamp can burn out and the other work
fine. There's also no blink blink blink before they come on
Grounding of the metal "reflector" above the bulbs is not always
essential, but can be helpful if the ballast (or maybe the bulbs) are a
little wimpy. The symptoms of an ungrounded fixture can be bulbs glowing
just at their ends, but not "firing up" all along their length.
The grounding works by providing an electrode to create an electrostatic
field extending away from the electrodes at the ends of the bulbs.
That field helps get the ionization of the gas (mercury vapor)inside the
tube extend the full length of the bulb. The ionized gas is
electrically conductive and current keeps flowing through it from one
end of the bulb to the other, keeping it ionized. (At that point you
could probably remove the fixture ground wire and it'd stay lit fine
'till you switched it off.)
The ionized gas emits ultraviolet light which excites the phosphor on
the inside of the glass and makes it flouresce white, much like an
ultraviolet "black light" will make certain inks and stuff glow
brightly. That's why they call them "flourescent" lights BTW.
Doubtlessly Don Klipstein can do a more elegant job of presenting this
If the bulbs glow throughout their length, grounding is not the problem.
When grounding is the problem, the ill effects of lack of grounding may
be intermittent, and may vary with temperature and/or humidity.
Close enough and I would award a cigar!
The way I see it, grounding makes unionized gas in most of the length of
the bulb more alike in potential, and that increases the voltage gradient
(electric field) in the gas approaching the electrode at either end of the
bulb - which is where ionization of the gas normally starts. Once
ionization happens where it can be initiated, then the boundary between
ionized gas and unionized gas is where the voltage gradient is high -
resulting in ionization there, and thus the boundary between ionized and
unionized gas propagates (usually in quite a small fraction of a second if
True! But anyone trying to make the coating glow from a blacklight will
usually not get that to work - the phsophor coating in most fluorescent
bulbs works well from shortwave UV and not from the longwave UV from a
blacklight. And most "fluorescent style" blacklights are true fluorescent
lamps - they have a phosphor converting shortwave UV (UVC) to longwave UV
(UVA, and at wavelengths too long for suntanning). Any UVC passing
through the phosphor is well-blocked by the glass.
Actually in a fluorescent lamp, maybe about 20% of the light is visible
light (bluish) from the glowing low pressure mercury vapor and about 80%
is from the phosphor. In a compact fluorescent, it's more like 25-75 or
so. Very roughly!!!
Maybe not quite if I work 2 jobs and then get online after working 13
hours.... :) :)
- Don Klipstein ( email@example.com)
Grounding may not have been your problem, but it is a common one.
Having grounded sheet metal close to the bulbs affects the electric field
distribution within a bulb that is trying to start.
Sometimes cleaning the bulbs also affects this, if the bulbs get coated
with dust that is hygroscopic and the humidity is high (more likely in
Anyone who does this ought to replace the ballast with a good
"commercial grade" one suitable for the bulbs in question once things
start working! Cheap fixtures, especially for "T12" (1.5 inch diameter)
bulbs as opposed to "T8" (1 inch diameter ones), tend to have "residential
grade" balasts. "Residential grade" ballasts for T12 or F40 "lamps"
(bulbs) are all too often something that I like to refer to as "turds".
More politely, paperweights that have wires coming out.
2 tubes go out in a 2-tube fixture - chances are only one is dead. 2
tubes go out in a 4-tube fixture - chances are only 1 is dead.
Same story if 2 tubes are dim or dim-and-flickering (but not outright
blinking and doing so out of sync with each other), or if one is dim with
one completely out.
In this usual failure mode, the dead tube usually has some major
darkening at one end - about 3 inches long in a 1.5 inch diameter tube,
about 2 inches long in a 1 inch diameter one. Less likely you see such
discoloration at both ends. That bulbs's viable mate usually has much more
minor but easily visible discoloration at the ends, usually at least
somewhat similar from one end to the other. Much newer bulbs that
got mixed in should have much less end discoloration, and possibly
If you can identify the mate to the dead bulb, its life is probably
mostly expended and the bulb can be disposed of like the actually dead
- Don Klipstein ( firstname.lastname@example.org, http://www.misty.com/~don/ltrouble.html )
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.