During this week we have been sent to replace a number of 250watt Metal Halide lamps,
Upon arriving we are confronted by an irate customer shouting that the `darn lights keep exploding`
we go into the factory and as we walk through the buiding a large bang was heard behind us and that area went dark, a bit scary.
We get up on the services area above the main factory floor to look at these lamps and they have litterally exploded within there envelope,

Having never come across this i am curious as to why ? has anyone got any ideas.?

John H

-no-name low-quality imported lamps

-defective name brand lamps from a bad factory run

-defective ballasts

????????

as far for exploing MH lights there is something wrong with that main thing is check the ballast to make sure it show what wattage it can run and check the capaitor to make sure it is not shorted ( shorted will dim the light ) and do not over tighten the bulb in socket and many time it can have small fracture and it will bulid alot of pressure i know 400 watts mh (probe type) arc tube can go high as 50 psi !!! that why becarefull when tighten it down and check the ballast voltage and current drawage and sometime i use "shorting test" that mean actally short the lamp lead not cord end just lamp and get amp meter to check the current drawage to make sure it is not too high

merci marc

p.s. if need more help just post it here i will try to help you with trobleshooting i work alot of this one

About 5% of MH lights will fail violently at the end of their life if they have been left on continuously. It is recommended that all MH fixtures be shut off for at least 15 minutes each week to help prevent this problem. The arc tube in the lamp runs at 1000°C or more so there is a potential for serious injury or fire when the lamp explodes.
Don

Here are links to more information on this problem. You may have to register to see the NEMA information.
OSHA Technical Information Bulletin
NEMA information on MH lamps

[This message has been edited by resqcapt19 (edited 04-11-2003).]

For US systems, I believe NEMA has a requirement that if the external envelope is broken, an internal fused link in the lamp must melt open within a particular interval. If the external envelope has not ruptured, then it is not considered a misoperation.

Without a log of voltage and current at the fixture terminals, it may be impossible to accurately attempt to decide who is at fault.

Killer and salient link, Don! “The small, but existing possibility of rupture is why all lamp manufacturers provide strongly worded warning statements with M-H lamps.”




[This message has been edited by Bjarney (edited 04-11-2003).]

Don,
That point about the "Down-time", required with M-H lighting, is often never abided by,
especially where lighting of non-stop factory processes are concerned, this drastically shortens the life of the bulbs and the control gear.
Regarding the safety of exploding lamps, over here all Gas Discharge light fittings must have a tempered glass cover over the bulb.

also i will like to add few more pointers here to get some matter straght here

1) all mh bulbs genrally best replace when get to 70 to 80 % of useable life but dont wait until end of life because more you leave on longer it more it risk to get arc tube inturpped ( expoled)

2)as far the metal hailde bulbs if useing wrong bulbs it can affect the performace some are base up some are horzontal type and some are universal type so please take a extra sec or so to read the suffex lettering what burning postion and safty requriement

3) i know most 250 watts mh are required to be inclosed unless they come with arc tube shroud they can run open most of them are enclose anyway expect 400 watts it is varis alot i cant say all the answer there

trumpy i do understand that fact most industral place dont really follow the light bulbs manfactures warning at all for 24/7 operation they should turn off the light at least once a week to make sure propley cycled

merci marc

Hi Marc,
How can you tell when a Metal-Halide lamp is nearing the end of it's life?.
Does the colour of the light change?, or can you tell by looking at the lamp?.
That's a very good point about using lamps that are not made for the particular mounting orientation(angle), I'd forgotten about this one, at least you know what you are talking about!.
That point about 24/7 operation, even in a Factory, the lights could be turned off during meal/smoko breaks, there are always ways around things like this.

Mike Trump.

CTwireman.

the lamps are sylvania (250w MH)

the control gear is thorn, in most but an unnamed brand in others, not all the exploding lamps have thorn, but some do.

Checked capacitors on all damaged fitings and all but one is ok, (that fitting just didnt work).

Having just got home from site, (over hear its now 1930 hrs) i have found a small problem that might be the route cause of the premature lamp failures,

Its a Power Factor Correction capacitor set up, fitted over three phase supply and it seams to be leaking,

i have to go back in the morning with a couple of lads to check out this PFC capacitor set up and replace the failed lamps/fittings before opening time of the yard on monday morning.

I think that the failure of one or more of the interconnected PFC capacitors has caused a fault with the load side of the lighting.

Previously the lighting was very old lead/lag 125 watt twin fluorescents at high level, the only thing the plants electrician (i use the term loosly) did was remove the fluorecents ans put in these metal halides.

I have not looked, yet, at the cabling, switchgear etc to see if the additional loading would be detrimental to the installation but we will, the company CEO was screaming blue murder about shutdowns etc money lost, money cost, but not a word about staff safety, funny that isnt it.

will keep you posted on what i find.

John H

John,
I think you may have just found the problem.
I would say that it sounds like the PFC Capacitor is too small (capacitance and voltage-wise) to handle the surge currents of the M-H lights, this should have been uprated when the fittings were changed over.

trumpy the bulb life determed by brightness between new and old one and colour shifting yes ideed if the bulbs are clear one if you look closely to the arc tube ( of course the light will be off anyway for safty sake ) near the end of life the arc tube will deform where the path runs like sagging pattern and also if you look at the manufactures catalog book it will listed the bulb life i know 250 w mh are listed about 15,000 hours or so depend on which type but one probem most factory dont like to cycle the switch for the light they claim it take too long to come back i say balony it is better do that way because you can catch bad one before actally exploding

john i did read that part about the power factoring capatitor that is instering to know but i am not sure about it but before it get out of hand what voltage you are running i know i belive you have 250 v 50HZ if i am correct ?? if so if the phase correction capatior going bad i wondering why they should have a fuse there to protect it from somekind of surge

IMHO : the capatoir do responde diffent between flourscent and hids they are differnt due current spike is diffrent i think that what keep overloading PPC unit

merci marc

Having further checked the PFC capacitors, all three are faulty,

Marc,
the supply running the MH is 3 phase 440v
split via two panel boards side by side.

The supply characteristics are as follows.
440V ac 50hz 3 phase at 600 amp per phase.

thsi oes via the local pocos fusable link and to the meter then to main disco (1000 amp)
from the disco it goes to a Bus Bar chamber then to local discos to sub panels.

the lighting boards are fed accross the PFC capacitors via there own Discos and then to panel boards.

Having now checked the installation we have removed the PFC and i am just in the process of checking loading capacity/lamp start/run currents to be able to re specify a new PFC.
the CEO was in again and had a few words to say to me, Again, the plant electrician has not been seen for a few days and we are all wondering what the hXXl he thought he was doing.

when i get the new PFC capacitor i will let you all know the outcome, i think it will be a good one.

By for now. John H

John H,
Man, is that a supply or what!
What sort of factory is on the end of that Bad Boy?.
If I was the Plant Electrician, I would be seriously looking for another job, he obviously doesn't understand the Basics of Electricity and Power Factor.

Trumpy,

A further update on this little fiasco, the supply is quit normal for the type of building, it is a heavy plant repair shop housed in 35,000 square feet (not all plant works some offices etc,

so you can see the reason for the supply !!!

its fun to work on, as you never can tell what the hxxl is goin to happen.

Also the `electrician` submitted his resignation today and left!!!!!

the new PFC capacitors are on order, the new fittings are on order, the new lamps are on order, etc etc etc, and the CEO wants the whole of the plant repaire and offices tested and inspected and put right, he has signed a proforma works order and away we go.

not a bad job in the end, if you discount the exploding MH lamps, it should bring into the company about $24,000 for all the repairs etc that are needed to make this eelctrical nightmare safe.

I will try to get some photos of the old stuff before we take it out.

Bye for now

John H

One other good thing, my quaterly bonus is safe.!!

440V??? Phase-to-phase or phase-to-neutral? Then they must have a standard 400V service as well?

John,
Well, I'm just glad that all it turned out OK in the end!.
You must be glad of all of the exta work that this has brought on.
Just goes to show what can happen when you have a "Thinking Electrician" on-site, as opposed to a mere "Wire-puller".
However, I'm sure that this has been a learning experience for all concerned, especially the old Plant Electrician!.
Some pics of the old Capacitors would be cool!.
C-H, it must be 440V Phase-Phase, 440V to Earth, sounds a bit dangerous!

[This message has been edited by Trumpy (edited 04-14-2003).]

440V was the highest nominal phase-to-phase voltage before standardization some 30 years ago. You'll still hear it referred to sometimes, even though the actual nominal levels are 240/415V (or officially now 230/400).

If you work it out, the actual phase-to-phase voltage was 433V, but it was often quoted as 250/440V.

wow now i can get the picture what happend to the plant electrican seem he cant grasp new morden stuff around here anyway as far the light is that wired for 250 volts or wired for 415 volts ??? just wondering about that and here in north americane system for the light most common is 120 or 277 volts but have odd ball too both 208 or 240 volts but cananda do not allow use 208 or 240 per cec ( i dont know why but i will dig it up )
but canada have 347 for hid or flourescet light now that is istering my self i havent see it but i heard it plenty of it so i will check my canandnie friend to find out


merci marc

Hello Marc,
IMHO, I don't think I would like to work on any lighting fitting that uses more than 230V, above that, you've got some real Health and Safety issues, for people that service these sorts of fittings, especially where these are a long way up from the ground, electric shocks and explosions take on a whole new perspective, when you are 7-8 metres up off the ground.

Most common commercial service voltage in the area of the US I work is 480 phase to phase 277 to ground, transformers used to get it down to 208/120 so we end up with 4 voltages in most commercial spaces.

Small commercial service about 600 amps, a typical grocery we wire 1600 amp, and larger office buildings 3000 amp.

Trumpy, 480 to fixtures is common for high bay fixtures and parking lot / site lights, and florescent office lighting runs 277 volt.

I will tell you I have never gotten hit with 480, but I have gotten a couple of good 277 volt shocks, It hurts a great deal more than 120 volt's.

Incidental references to North American lighting voltages are:
http://www.ruudlighting.com/literature/spec_sheets/td-2.pdf
Lists 120, 208, 240, 277, 347 and 480 volts
http://www.fisherpierce.com/PDFFiles/FP%20SERIES.PDF
RATED VOLTAGE
FP120 Series -- 105-130V, 50/60Hz AC
  (120V Nominal)
FP208-277 Series -- 200-300V, 50/60Hz AC
  (208, 240 or 277V Nominal)
FP240 Series -- 190-250V, 50/60Hz AC
  (220 or 240V Nominal)
FP347 Series -- 310-370V, 50/60Hz AC
  (347V Nominal)
FP480 Series -- 400-520V, 50/60Hz AC
  (480V Nominal)




[This message has been edited by Bjarney (edited 04-14-2003).]

Yes, 347v flor's are horrible to work on, especially at Public Works facilities, and even more so off a ladder.

Last week, due to security reasons (even tho were cleared 'secret') we were not allowed to do the fixture bouncing after hours, which meant power cldn't be interupted.

As an 'apprentice' I refuse to do anything outside of whats deemed 'safe and proper', so needless to say when asked to just go in a bounce a few fixtures I said sure, expecting a simple, relocate the fixture still connected in the ceiling grid and rechain them.

Wrong!!!! The 1st fixture I attempt gives off that scary hummmmmmmmmm when I moved it. Pulled my hands away as it gave off its own lil liteshow, called my supervisor and told him I refuse to touch any of the system live!!!

Well doesn't he come down to the site to show me I'm wrong, lol, and the same happens to him (also lucky not to get hit from the fixture).

So he decides to go to the JB and disconnect that area. All I can say is who-ever authorized the ceiling grid should have been shot for not requesting electrical alterations. EVERY JB was over a grid cross with only about 3" clearance below the open box, data wires all over the place, and venting obstructing safe mannouvering.

Not to forget... who-ever installed this electrical mayday shld also be shot... there isn't a single joint (120/208/347) in the entire building that when you remove the marrets that the wires don't spring apart.

So Mr Bigshot (supervisor) tackles it anyways as I stand below watching at the floor level. Needless to say, as he removed that 1st marret he got energized.

And as this was my first time seeing someone hit with more than 120 I can only say I hope its my last. The look of fear and pain as somebody is hung up would scare the bejesus out of a hardened soul.

I kicked the ladder out from under him, he separated and hit the ground in a curled up tearing fashion. I asked if he was ok, checked for burns, but found no entry nor exit anywhere. And for that, he was VERY lucky as the hot connected with his hand, and he was unknowling grouded at the back on his neck with a BX.

Once he settled down, and was able to get up, and seemed ok, I quietly packed my tools up, called a cab, told him he is a dumbass and too dangerous to work with and to call me when PW is willing to shut the floor lighting down at the panel and consider themselves lucky I don't call the authorities for putting me at jeopardy.

Next day got called in, appologized profusely, and told I wld never be put in or asked to perform in a hazardess manner again.

I'll tell you guys, especially the newbe's out there, watch out for yourselves and use YOUR OWN common sense because I think a lot of the seasoned journey men out there are so used to encountering these types of routines that they tend to fear less as they do the mondane tasks asked of them.

And I'm sorry, I never have trusted a person with a holier than thou additude when it comes to proceedures or safetly and more importantly when it deals with electricity and MY life.

Sorry for rambling or hyjacking people, it was just still fresh in my mind and wanted to put it on paper so to speak.

-Greg aka: don't kill me with stupidity

What causes Metal Halide lamps to explode?

Smokeumchevy,

Remember this, isolate before working on circuits, test twice before working on circuits,

DO NOT TAKE IT FOR GRANTED THAT THE SUPPLY IS OFF, EVEN IF YOU ISOLATED IT YOUR,SELF

CHECK IT !!!


In regard to Exploding metal halide lamps;

the following can cause premature explosion and or failure of lighting lamp.

Incorrect connection of fitting;
over run time ( on to long without a break);
Failing/failed ignitors
Failing/failed PFC capacitors
Failing/failed control gear
incorrct supply characteristics

dropping them on the floor
throwing them at your supervisor.

these are a few of the faults that can cause the above.

Had a bit more fun today;

Local POCO have disconnected the driveway street lighting from there supply, so we are having to route a new cable underground to feed them, lousy fittings though 90 watt low pressure sodium on 5 meter poles, yuk, hate em.!!

Smokumchevy,
A BIG welcome to ECN, mate!
Glad to have you along.
Don't you just love Supervisors?
One thing that does kill and seriously injure experienced Electrical workers, is Complacency.
However, I wouldn't personally kick a ladder out from under a person to free them from contact with live wires, this can lead to even worse injures than the shock.

[This message has been edited by Trumpy (edited 04-15-2003).]

bonsoir trumpy :


yeah indeed we have multi voltage in commercal and industrail system and sometime if not too carefull it will get you by suprised and yeah yeah i got shocked by 277 volts few time and last time i got shocked with 480volts when i working on big parking lot light i did double check the voltage make sure it is dengerized (dead) somehow without waring i got hit with 480 OUCHHHHHH that got me the fixure have unground system i did swear about it and got the inspecter check the rest of system and end up rewire whole parking lot btw the wire was bad and reconnet to 277 volts and add fuse each large fixure they were a 1 kw mh light bulbs each that was big and hevey sucker to boot... anyway main thing i use is common sense what is going on. i work on lighting system for many years and very few mh light did exploaed on me but kinda rare for me because i convented my comster to replace them early before they hit the end of life and last thing i want to deal in middle of wintertime to replace bad ballast that is no fun i allready did it with 20-C temp with wind going about 25 mph ( 40kmt) anyway engough of chat but will chat more later about it

merci marc

Welcome Greg,

Your experience just goes to show that we should never become complacent. I must confess that I've been hit with 240V to ground a few times, but fortunately I've never found myself across two phases.

A good approach to adopt when checking that a circuit is de-energized is to put the meter onto a known energized circuit first.