Neon light exploding and falling from fixture

Posted by Edwin Pawlowski on August 12, 2007, 9:35 pm

>> Neon fixtures use very high voltage. A short circuit in the fixture could
>> cause this. What is this being used for?

> As Noozer pointed out, I meant to say fluorescent. The fluorescent lights
> are used in the carport. They are under cover and I could not see any
> water or anything getting in there. They are exposed to hot and cold
> temperatures, mind you, because there is no door on the carport.

I've had it happen in our warehouse. There is some vibration from the floor
above though and it is still a rarity. I can think of maybe two times in 15
years with a couple hundred fixtures.

Posted by Bob M. on August 12, 2007, 9:43 pm

>> Neon fixtures use very high voltage. A short circuit in the fixture could
>> cause this. What is this being used for?

"carport" suggests it's not truly cold there but if you do get sub-freezing
temperatures there, you may want to put in regular screw-in light bulbs
since they work no matter what the outside temperature is. Flourescents
don't like cold weather. Yeah, there are flourescent fixtures designed for
cold-weather use, but not extreme cold like I get here in Montana.

If you want to hang on to the flourescent lights, maybe replace the ballasts
with electronic units. They don't buzz at all, at least not the kind you
get at the electrical supply houses (not Lowe's, Home Cheapo, etc) and they
run cooler, which means they're more efficient and use less electricity.

Posted by on August 13, 2007, 7:04 am

On Sun, 12 Aug 2007 14:01:30 -0700, "Lesley"

>> Neon fixtures use very high voltage. A short circuit in the fixture could
>> cause this. What is this being used for?

>As Noozer pointed out, I meant to say fluorescent. The fluorescent lights
>are used in the carport. They are under cover and I could not see any water
>or anything getting in there. They are exposed to hot and cold temperatures,
>mind you, because there is no door on the carport.
>Lesley

I thought you meant florescent. You cant change the tubes on a neon
(without special equipment anyhow).

Anyhow, when I was a young boy I was in a store with my father when a
florescent bulb exploded. Both of us were covered with glass and
florescent dust. The store wanted us to go to a hospital but my
father said we were ok. I will never forget that. As a youngster it
was quite frightening.

I'd suspect the ballast is bad. I'd replace the ballast if I were
you. (or the whole fixture). But all you really need to change is the
ballast. The rest of the fixture is just a metal shell and some
wires, plus the lamp "sockets". Unless the bulb fell out of the
socket, there is nothing else that could cause this except the
ballast. You said it exploded in the fixture, so that eliminates the
falling out of a bad socket.

I am glad this does not happen often considering how many places have
florescent lights. I think those old "transformer" type ballasts were
more likely to fail than the modern electronic ones. This dont happen
to be one with the old twist in starters, is it?

Jim

Posted by Jeff Wisnia on August 13, 2007, 3:07 pm

jamesrdens@ya_oo.com wrote:

> On Sun, 12 Aug 2007 14:01:30 -0700, "Lesley"
>
>

>>>Neon fixtures use very high voltage. A short circuit in the fixture could
>>>cause this. What is this being used for?

>>As Noozer pointed out, I meant to say fluorescent. The fluorescent lights
>>are used in the carport. They are under cover and I could not see any water
>>or anything getting in there. They are exposed to hot and cold temperatures,
>>mind you, because there is no door on the carport.
>>Lesley

>
>
> I thought you meant florescent. You cant change the tubes on a neon
> (without special equipment anyhow).
>
> Anyhow, when I was a young boy I was in a store with my father when a
> florescent bulb exploded. Both of us were covered with glass and
> florescent dust. The store wanted us to go to a hospital but my
> father said we were ok. I will never forget that. As a youngster it
> was quite frightening.
>
> I'd suspect the ballast is bad. I'd replace the ballast if I were
> you. (or the whole fixture). But all you really need to change is the
> ballast. The rest of the fixture is just a metal shell and some
> wires, plus the lamp "sockets". Unless the bulb fell out of the
> socket, there is nothing else that could cause this except the
> ballast. You said it exploded in the fixture, so that eliminates the
> falling out of a bad socket.
>
> I am glad this does not happen often considering how many places have
> florescent lights. I think those old "transformer" type ballasts were
> more likely to fail than the modern electronic ones. This dont happen
> to be one with the old twist in starters, is it?
>
> Jim

It would be helpful if Don Klipstein jumped in and educated us about
what sort of ballast failure causes a flourescent bulb to "explode".

My understanding of ballast operation is that aside from heating the
filaments for starting their main purpose is to act as a "choke" which
provides an inductive impedance to drop line voltage down to the voltage
level which is correct for the particular bulbs they are rated for.

A fixed resistor could provide the same voltage drop, but would waste
too much power as heat. Back around 1949 my uncle Mo, who was well
versed in electricity and the electronics of that era made me a line
voltage operated "black light" from a small UV flourescent bulb and a
big wirewound resistor, all housed in one of those wooden Velveta cheese
boxes about 10 inches long. (Rememvber them?)

So, maybe if the "choke" in the ballast which provides that inductive
reactance in series with the bulb were to short out, the bulb would
receive full line voltage and the resulting arc would be so intense that
it could overheat the gas in the tube causing its pressure to increase
enough to burst the bulb. Or, maybe it would just run so hot that the
heat would precipitate a mechanical failure.

Just my .02, but I've got a feeling that if the OP didn't know the
difference between neon tube and a flourescent bulb then he/she may not
be able to judge the length of the bulbs "by eye" either. So let's not
get too worked up trying to figure out what's really installed there and
what went wrong.

Me, I'm down to telling the OP to get a qualified pro out there to check
things out, I wouldn't dare suggest that the OP tries working on it
him/herself.

Jeff

--
Jeffry Wisnia
(W1BSV + Brass Rat '57 EE)
The speed of light is 1.8*10^12 furlongs per fortnight.

Posted by Don Klipstein on August 13, 2007, 8:00 pm

>jamesrdens@ya_oo.com wrote:

>> On Sun, 12 Aug 2007 14:01:30 -0700, "Lesley"
>>
>> I thought you meant florescent. You cant change the tubes on a neon
>> (without special equipment anyhow).
>>
>> Anyhow, when I was a young boy I was in a store with my father when a
>> florescent bulb exploded. Both of us were covered with glass and
>> florescent dust. The store wanted us to go to a hospital but my
>> father said we were ok. I will never forget that. As a youngster it
>> was quite frightening.
>>
>> I'd suspect the ballast is bad. I'd replace the ballast if I were
>> you. (or the whole fixture). But all you really need to change is the
>> ballast. The rest of the fixture is just a metal shell and some
>> wires, plus the lamp "sockets". Unless the bulb fell out of the
>> socket, there is nothing else that could cause this except the
>> ballast. You said it exploded in the fixture, so that eliminates the
>> falling out of a bad socket.
>>
>> I am glad this does not happen often considering how many places have
>> florescent lights. I think those old "transformer" type ballasts were
>> more likely to fail than the modern electronic ones. This dont happen
>> to be one with the old twist in starters, is it?

>It would be helpful if Don Klipstein jumped in and educated us about
>what sort of ballast failure causes a flourescent bulb to "explode".
>My understanding of ballast operation is that aside from heating the
>filaments for starting their main purpose is to act as a "choke" which
>provides an inductive impedance to drop line voltage down to the voltage
>level which is correct for the particular bulbs they are rated for.
>A fixed resistor could provide the same voltage drop, but would waste
>too much power as heat. Back around 1949 my uncle Mo, who was well
>versed in electricity and the electronics of that era made me a line
>voltage operated "black light" from a small UV flourescent bulb and a
>big wirewound resistor, all housed in one of those wooden Velveta cheese
>boxes about 10 inches long. (Rememvber them?)
>So, maybe if the "choke" in the ballast which provides that inductive
>reactance in series with the bulb were to short out, the bulb would
>receive full line voltage and the resulting arc would be so intense that
>it could overheat the gas in the tube causing its pressure to increase
>enough to burst the bulb. Or, maybe it would just run so hot that the
>heat would precipitate a mechanical failure.

The gas in most fluorescents has a pressure at room temperature in the 2
mm Hg ballpark, roughly .2-.3% of atmospheric pressure. I don't see much
chance of one bursting from overpressure if the ballast shorted.

As for how I suspect a fluorescent lamp could explode? Or more likely
implode since the pressure in one is most of the way to a vacuum?

This gets a bit towards grasping at straws...

1. Very serious ballast failure sends a few amps through the bulbs, the
electrodes melt and the arc terminates on the lead wires, the leads melt
down and the arc terminates on lead stubs where the leads pass through the
glass, the glass overheats and cracks, then the bulb breaks and implodes.

I never heard of this actually happening. Most ballasts can't do this
anyway:

a) The usual magnetic rapid start ballasts for 4-footers have current
limiting by leakage inductance caused by magnetic shunts in the core.
No short can defeat that. Shorting of part of the primary will increase
output, but not by a whole lot before the ballast overheats so badly as to
just burn out.

b) In any magnetic ballast, the windings will normally melt and work as a
fuse before such a high amount of current can crack the bulb.

c) I don't see much chance of an electronic ballast malfunctioning in a
way to put out that much current without conking out before the bulb can
break.

But if the ballast is an old lead-lag type that makes bulbs flicker out
of phase with each other, I am not as confident that it can't break a
bulb if it shorts, especially if a capacitor shorts.

2. The ballast blows up and spits out shrapnel that breaks the bulbs.

This sounds awfully improbable to me.

One day I was close to a ballast that did go "Pow" and spit stuff out.
This was one of the maybe 1,000 or at least a few hundred or so ballasts
for the fluorescent lights in the ceilings where I worked then. The bulbs
were 8-footers.
The ballast started humming, then a few aseconds later the hum changed
to a louder buzz, and after a second or two of that the ballast went POW!
There was a small hole blasted in the metal ballast case. I suspect a
localized small region of the tar poured into such ballasts boiled due to
a short.
If whatever blew out the hole hit a bulb, I suspect that could have
broken the bulb.

This is fairly rare, and normally in fluorescent fixtures there is metal
fixture material sorrounding the ballast.

3. Some instant start fluorescent fixtures have ballasts that put out
enough voltage to refuse to take "burned out" for an answer. I suspect
this is more likely with instant start ballasts rated for more than one
bulb length and running a shorter one, if the ballast is rated for cold
temperatures, or running a bulb that is shorter than the ballast is rated
for.
If the electrodes get worn out, the ballast provides enough voltage to
maintain the arc in the bulb anyway. Possibly the arc can melt down
electrode leads all the way to the glass and then crack the glass. Also,
sustaining of an arc on electrodes that are in no shape to do so my occur,
and the electrodes get much hotter than normal doing so, and the ends of
the bulb may overheat and crack.

I would give this increased chance if the bulbs are those old "slimline"
ones with only one pin on each end.

I have seen this occur with 32 watt 4-foot T8 (1 inch diameter) bulbs
with instant start type electronic ballasts, especially in fixtures with
the ballast running at least 2 bulbs (pseudoparallel arrangement in
ballast, where patrt of the current limiting is common to all bulbs and
part is applied individually to each bulb). When one bulb dies, the
remaining one(s) get(s) a little brighter. When the last bulb dies, the
ballast sometimes keeps it working by brute force.
I have seen a few bulbs where the glass cracks in the region of the bulb
around one filament. So far, I have only seen cracked bulbs and not
imploded ones. However, I have heard of lampholders (sockets) burning up
if they are not high temperature ones rated for this kind of heat.

Warning signs: Bulb flickers, one end (less frequently both ends) is
blackened, and the failing end may have purplish/pinkish flashes or
pinkish/orangish/orange-incandescentish glow. The failing bulb should be
removed unless you are confident that your fixture is fully tolerant of
whatever can happen there.

Bulb failure is much more of a non-problem with rapid start ballasts
than with instant start ballasts. Meanwhile, I'm sure that the
manufacturers have done things to mitigate bulb failure problems with
instant start ballasts for T8 fluorescents of 17-32 watts.

4. The ballast failed or malfunctioned and vibrated, or posibly just got
loose and vibrated, or external vibration shook the bulbs out, and bulbs
fell out of the fixture and imploded upon landing.

- Don Klipstein (don@misty.com)

Page 2 of 3 < 1 2 3 > last >>

Similar Threads	Posted
Strange Neon Night Light	June 25, 2006, 3:11 am
Broken gas light from falling tree limb	October 18, 2006, 1:30 pm
Light Fixture	November 8, 2005, 1:40 am
Light/Fan Fixture	November 12, 2006, 5:01 pm
Light Fixture	March 11, 2007, 10:24 pm
Light fixture	September 16, 2007, 10:51 pm
Light Fixture ?	September 22, 2009, 10:16 am
New Light Fixture / Old wiring	August 30, 2005, 11:53 pm
Fluorescent fixture won't light	September 6, 2005, 2:16 pm
Light fixture with outlet?	November 14, 2005, 12:45 pm

Neon light exploding and falling from fixture - Page 2