Lighting troubleshooting

[drbond24]

Before I state my problem, let me avoid sermons by explaining that I realize that this is the lightning forum (extra 'n') and I am asking a lighting (no extra 'n') question. I'm here because of the power quality part. :grin:

I have two problems that may actually be one problem. First problem:

We have a warehouse that is illuminated with 400 W metal halide fixtures. These fixtures go dim extremly quickly A brand new bulb will be down to about 50% in a week. These bulbs should not noticably dim for at year, even if ran continuously. What could be causing this? All I have done is check the voltage on the lighting circuits, and the highest voltage I found was 287 V on a 277 V circuit. Not high enough to be a concern.

Second problem: We have a manufacturing plant that sits right next to the warehouse, but each is fed with a separate service. They are supposed to be independant of one another. However, any time the manufacturing plant loses power for any reason, after about 4 or 5 hours the warehouse will go down and we won't be able to get it on again until the plant is on. Last time, the warehouse burned a ballast in one of the aforementioned lights which apparently happened right before the whole building went off.

Now, is that one problem or two? Does it sound like there is a connection between the buildings that shouldn't be there that is causing both problems? Any input will be appreciated. If more details are needed just ask and I will provide them.

 

[Lxnxjxhx]

The lifts in a ski resort worked perfectly for years. Another resort opened 9 miles away, and now the first resort was tripping breakers.
It was a power line resonance effect. Don't know how they fixed it.

Don't assume anything about how these places are really wired.

The first thing I would do is figure out what is causing a "time constant" of 4-5 hours. This might be the key to the whole puzzle.
What is happening in this interval? Something mildly overloaded and heating up? Make up a theory and try to prove/disprove it.

And you have a steady-state problem with the lamps. If it's not the environment it must (?) be power quality.

 

[76nemo]

Before I state my problem, let me avoid sermons by explaining that I realize that this is the lightning forum (extra 'n') and I am asking a lighting (no extra 'n') question. I'm here because of the power quality part. :grin:

I have two problems that may actually be one problem. First problem:

We have a warehouse that is illuminated with 400 W metal halide fixtures. These fixtures go dim extremly quickly A brand new bulb will be down to about 50% in a week. These bulbs should not noticably dim for at year, even if ran continuously. What could be causing this? All I have done is check the voltage on the lighting circuits, and the highest voltage I found was 287 V on a 277 V circuit. Not high enough to be a concern.

Second problem: We have a manufacturing plant that sits right next to the warehouse, but each is fed with a separate service. They are supposed to be independant of one another. However, any time the manufacturing plant loses power for any reason, after about 4 or 5 hours the warehouse will go down and we won't be able to get it on again until the plant is on. Last time, the warehouse burned a ballast in one of the aforementioned lights which apparently happened right before the whole building went off.

Now, is that one problem or two? Does it sound like there is a connection between the buildings that shouldn't be there that is causing both problems? Any input will be appreciated. If more details are needed just ask and I will provide them.

Do both buildings share the same cans?

 

[BackInTheHabit]

What's the voltage of the MH fixtures?

Try changing bulb brands. We use Sylvania and have had no problem with them.

 

[cadpoint]

Oh a whole lot of thoughts come to mind about the light. I'd go mid-point of service to light and ohm and meg, I'd also look for swapped netural and ground wires, then ohm & meg them all. Is there even a bonding wire or are you using the conduit ?

Is the whole panel running at stated voltage or only at point of use?

Haven't you changed the ballast or the whole fitting ?

How's the Record Drawings at the plant? Anyone open any cans lately, anyone ever verify anything?

 

[quogueelectric]

Why do you think the lightning dimms the bulbs??

 

[cadpoint]

Point Taken!

 

[quogueelectric]

I was just kidding. It sounds to me like you have an open or loose neutral somewhere in the branch circuitry. Did you test the voltage under load and where the problem is occuring?? Or did you test back at the panel ? If it was a multi wire branch circuit the other circuit may have been sent out where the fixtures were experiencing high voltage .

 

[don_resqcapt19]

Are the lamps the correct lamps for the installated position? (hortizontal, base up, base down)

 

[drbond24]

Sorry for taking so long to get back on this. I only go through the forum at work and I took a vacation yesterday. :smile:

Do both buildings share the same cans?

I'm unfamiliar with that term.

The two buildings have separate transformers and switchgear. The only thing they should have in common is the primary line from the POCO.

What's the voltage of the MH fixtures?

277 volts.

Is there even a bonding wire or are you using the conduit ?

Is the whole panel running at stated voltage or only at point of use?

Haven't you changed the ballast or the whole fitting ?

How's the Record Drawings at the plant? Anyone open any cans lately, anyone ever verify anything?

Not sure, but most likely it is just the conduit.

The whole panel.

No ballasts have been changed unless they fail completely. This isn't specific to one ballast or even one circuit. Lights all over the warehouse are suffering from this problem, whatever it is.

I don't have any drawings. Nothing has been verified; we just got started working on this.

Did you test the voltage under load and where the problem is occuring?? Or did you test back at the panel ?

The voltage has only been tested at the panel so far.

Are the lamps the correct lamps for the installated position? (hortizontal, base up, base down)

Yes.

 

[zbang]

How long as this been happening and how old are the MH fixtures? (Another wild -a$$ thought is whether a 480v MH fixture might behave this way on 277 or simply not start the lamp at all, but I would make sure that everything is expecting 277. OTOH, since ballasts have been replaced, I would think that someone has looked at voltage ratings somewhere along the line.)

 

[drbond24]

Here is the fixture:

I attached the info on the bulb.

I am only aware of this happening with the last year or so, but that's only as long as I've been employed here so it could have been happening before that.

Edit: The spec sheet for the bulb says that its operating voltage is 135 V. I've been assuming that was the output of the ballast and not the circuit voltage. Is that correct? If not, that could be the problem.

 

[wirenut1980]

It looks like you are using the correct bulb for the ballast, but that 135 V rating is strange? What's up with that?

Anyways, you probably should monitor voltage at a lighting panel over a period of at least a week to see if you are getting a little higher at times than 287 V. Is this a 24 hour operation? Is there ever down time where voltage can creep higher? Any caps or regulation automatic or manual?

When you say the warehouse goes down, what happens? Does the main breaker trip, or is there something else that happens?

If the factory load is influencing the power to the warehouse over the primary lines, then perhaps the factory load is huge compared to the primary system capacity...?

 

[drbond24]

It looks like you are using the correct bulb for the ballast, but that 135 V rating is strange? What's up with that?

Anyways, you probably should monitor voltage at a lighting panel over a period of at least a week to see if you are getting a little higher at times than 287 V. Is this a 24 hour operation? Is there ever down time where voltage can creep higher? Any caps or regulation automatic or manual?

When you say the warehouse goes down, what happens? Does the main breaker trip, or is there something else that happens?

If the factory load is influencing the power to the warehouse over the primary lines, then perhaps the factory load is huge compared to the primary system capacity...?

We operate 24 hours a day Monday-Friday normally, sometimes we work Saturday. The lights are on continuously during that time, and the load is fairly uniform across the board in both buildings.

The main breaker trips in the warehouse. Last time the cause was a bad ballast. I don't know why the main breaker tripped instead of the circuit breaker. Afterwards, they turned all sub-breakers off then energized the main breaker and turned each sub-breaker on one at a time until they found the circuit with the problem. After replacing the ballast, everything was fine.

The factory pulls around 2500 amps most of the time, but that is on the secondary side of our transformer. The primary is 34.5 kV, so we're only pulling ~35 A from the POCO.

 

[wirenut1980]

Could be that the main breaker was a gfi and is set very sensitive. The ballast was leaking to ground, not enough current to trip a breaker, but enough for the GFI...just a theory.

I would be interested to see what the voltage is running in the warehouse during downtime. Are the lights turned off during downtime?

 

[drbond24]

Could be that the main breaker was a gfi and is set very sensitive. The ballast was leaking to ground, not enough current to trip a breaker, but enough for the GFI...just a theory.

I would be interested to see what the voltage is running in the warehouse during downtime. Are the lights turned off during downtime?

I don't even know if the main breaker is GFI...I'll check later.

I intend to measure the voltage during downtime, but that is easier said than done. The lights would be turned off, yes. I would also like to measure the voltage with the warehouse attempting to run normally while the plant is offline, but usually when the plant is offline I'm scrambling to get it back online, not measuring voltages somewhere else.

 

[donselectric]

put a line monitor on the service and see whats happening
dont rule out poco. sometimes there switching things does funny stuff
i remember going into our sub station one nite
cause a main tripped. as i was looking the phone is ringing
its the poco wondering what we are doing over there ?
i said i'm the only one here and i aint doing nothing, what
are you doing ?

 

[quogueelectric]

I am still going with the open / loose neutral on the branch circuit theory. Did you check the neutrals in the panel main and branch ckts first. always pick the lowlying fruit first.

 

[drbond24]

I've checked everything my little brain can think of (that I have test equipment for) and can't find anything that looks like a problem. The voltage is a little high at the panel, about 287 V phase to neutral. The lighting circuits are a little long, but since the voltage starts a little high by the time it gets to the lights it is just right after voltage drop. There is no reason they shouldn't be working right.

I don't have any way to check power quality, which is retarded. My employer won't spend the money for a meter, so if that is the problem then I can't diagnose it.

I'm going to have to report a "I have no freaking clue" later today if I don't have a stroke of genius this morning.

 

[drbond24]

harmonics?

harmonics?

What effect would harmonics have on the lighting? Could that explain what is happening?

The are some motor loads in the building but nothing very large. The only other thing are the converters for charging forklift batteries. There are at least three large converters fed from the same panel that the lighting is fed from. What effect would that have on the circuit, if any?