Baseball field lights tripping breakers.

[11bgrunt]

10-2013 http://forums.mikeholt.com/showthread.php?t=156764&highlight=metal+halide+tripping+breakers Augie47 started a thread about something similar. I couldn't see that he ever reported a resolution.

In our case, the school built the new field three years ago. They started with a 100kVA POCO transformer serving a 600 amp, 240 volt panel. There are 9 poles and 12 circuits, all at 240 volts. The Metal Halide fixtures are 1500 watt. There are banks with 4 lights on some and 5 on others. Since the first day they have been dealing with random nuisance tripping of the circuit breakers. They started with 40 amp breakers and now there is a mix of 50 and 60 amp breakers serving those circuits. Those are the breakers that are randomly tripping after the lights have been on for just a few minutes. They say the breakers have not tripped as the light starts.
Original thought was that there was too much heat in the 600 amp panel so a 225 amp panel was subbed out of the 600 and 5 circuits were moved to the 225 and all breakers were spaced so none were next to another. There is a contactor station with 9 contactors in one cabinet. The new plan is to replace the 600 amp panel with a bigger, better, shinier GE. Existing equipment is GE. Breakers trip in the 600 and the 225. All equipment is mounted outdoor.
There has been an architect, engineer and electrician involved for the last three years since this was all new. This is warranty work so the school just wants something fixed. I am an innocent bystander asked to help. I will be there in a couple days and plan to start all the lights and thermal everything.
We have talked about power factor, overloads and voltage drop. The electrician reports the voltage at source and load are good. They did provide temperatures of enclosures and each breaker. The numbers suggest it was all done with a handy dandy spot gun. None of that information suggest a problem.
What else can we do to help the school?

 

[Jraef]

10-2013 http://forums.mikeholt.com/showthread.php?t=156764&highlight=metal+halide+tripping+breakers Augie47 started a thread about something similar. I couldn't see that he ever reported a resolution.

In our case, the school built the new field three years ago. They started with a 100kVA POCO transformer serving a 600 amp, 240 volt panel. There are 9 poles and 12 circuits, all at 240 volts. The Metal Halide fixtures are 1500 watt. There are banks with 4 lights on some and 5 on others. Since the first day they have been dealing with random nuisance tripping of the circuit breakers. They started with 40 amp breakers and now there is a mix of 50 and 60 amp breakers serving those circuits. Those are the breakers that are randomly tripping after the lights have been on for just a few minutes. They say the breakers have not tripped as the light starts.
Original thought was that there was too much heat in the 600 amp panel so a 225 amp panel was subbed out of the 600 and 5 circuits were moved to the 225 and all breakers were spaced so none were next to another. There is a contactor station with 9 contactors in one cabinet. The new plan is to replace the 600 amp panel with a bigger, better, shinier GE. Existing equipment is GE. Breakers trip in the 600 and the 225. All equipment is mounted outdoor.
There has been an architect, engineer and electrician involved for the last three years since this was all new. This is warranty work so the school just wants something fixed. I am an innocent bystander asked to help. I will be there in a couple days and plan to start all the lights and thermal everything.
We have talked about power factor, overloads and voltage drop. The electrician reports the voltage at source and load are good. They did provide temperatures of enclosures and each breaker. The numbers suggest it was all done with a handy dandy spot gun. None of that information suggest a problem.
What else can we do to help the school?

There are multiple TYPES of MH ballasts used out there, and some of them are much more sensitive to voltage drop than others. Read this web page to understand the differences. At 1500W per fixture it's unlikely that you have an Electronic ballast, but among the magnetic types out there, some don't do well with voltage variations and will spike current if there is a significant drop.

 

[Mouser]

Were the underground conductors megged?

 

[11bgrunt]

Were the underground conductors megged?

I doubt that anyone is thinking a short involving all circuits and that has probably not been done.

 

[GoldDigger]

Just out of the blue, an interchange of the hot and neutral along with a bond to earth at the poles can cause a more or less stable extra current on each circuit when energized.
Other possibility is mechanical damage where all the circuits run together
Has anyone put an amp clamp on the branch wires?

 

[zbang]

Some thoughts-
If you have one, thermal image pretty much everything you can.

Are the trips really random? That is, has anyone actually mapped out which ones trip and when? Might be a couple that don't; what's different about them?

Are the currents reasonable for the loads? If a 4-light * 1500w stand reads more than 25-ish amps, there is definitely a problem. Could be the wires were nicked during installation and are making a partial short in a water-filled pipe. The megger should also show this up.

If you have a recording amp meter, hang it on three separate lights and wait for the trip :lol:.

 

[templdl]

Interesdting. The OP overlooked advising if the breakers tripped thermally or magnetically. From the description and time derlay from ft orst energization you would guess thermally where the breasker's had to be left to cool off a bit before they could be lstcher/reset as nd closed. This would maybe give you some idea that there was and overload of some type. If they tripped instantaneously you we would be able to latch/reset and close the breaker without a delay.
As such what time/current event might have caused the breaker to trip?

 

[11bgrunt]

Just out of the blue, an interchange of the hot and neutral along with a bond to earth at the poles can cause a more or less stable extra current on each circuit when energized.
Other possibility is mechanical damage where all the circuits run together
Has anyone put an amp clamp on the branch wires?

Thanks to all,
GoldDigger, I don't understand what you are trying to teach me about an interchange.
I have not been to the site. Contractor asked for a clearance next week and some of our folks went to visit with our customer for the first time today. I have to hope with all that has been done and the fact that the contractor is willing to eat a 600 amp RB panel because it might maybe hopefully fix something, that checking amps has been thought of sometime in the last three years. My original thought was a design issue. i would have expected a 480 volt service. As zbang stated, what we do know is second hand info and they are random trips because somebody reported they were random with no ryme or reason.

 

[GoldDigger]

I was suggesting that perhaps they wired everything with black wire and at some point thought a hot wire was a neutral and treated it accordingly and vice versa with the actual neutral.
And did this many times or did it in a feeder.
Bonding a hot wire to a de facto grounded pole or to a driven ground rod would make a substantial shirt that stiil would not, by itself, trio a breaker.

 

[kwired]

You have never mentioned what these are drawing for current.

That is first thing I would have checked and then see if it is in expected range of the connected load.

Don't know actual current rating of these luminaires, 1500 watts @ 240 volts is 6.25A x 5 = 31.25A, but there is ballast losses on top of that.

Has anyone verified the ballasts are connected to the correct supply tap? 240 supply is about right in the middle of 277 and 208 and they may actually operate but not at proper current level if connected to wrong supply taps.

 

[ggunn]

...a substantial shirt that stiil would not, by itself, trio a breaker.

Your fingers are fat this morning?

 

[GoldDigger]

Your fingers are fat this morning?

Fat fingers and out of tune ought-to-correct.

 

[11bgrunt]

You have never mentioned what these are drawing for current.

That is first thing I would have checked and then see if it is in expected range of the connected load.

Don't know actual current rating of these luminaires, 1500 watts @ 240 volts is 6.25A x 5 = 31.25A, but there is ballast losses on top of that.

Has anyone verified the ballasts are connected to the correct supply tap? 240 supply is about right in the middle of 277 and 208 and they may actually operate but not at proper current level if connected to wrong supply taps.

On site with the engineering firm last week. All lights started and none tripped while we were there. Volts and currents measured at the panels were what I would expect with 4 gang and 5 gang 1500 watt fixtures on the 240 tap. The 600 amp panel and 225 panel are GE. The breakers are 240 volt bolt in. We found three breakers above the 40C rating. 80, 100 and 120C measured with a thermal camera on a 25C overcast day. We reviewed the history and no one could say if some of the breakers were changed along the way in the last three years. The first thing I noticed before pulling a cover was the smell of overheated bakelite. Nothing was out of the ordinary in the routine visual inspection other than the slight odor. The thermal was the key. Engineer will have EC change breakers and maybe even change panels to another brand.
Is there anything unique to HID lighting that would cause these breakers to fail when sized correctly? Planned to check harmonics on that visit but did not.

 

[zbang]

. The breakers are 240 volt bolt in. We found three breakers above the 40C rating. 80, 100 and 120C measured with a thermal camera on a 25C overcast day.

Lotsa loose bolts? Or the install over-torqued the bus bolt and stripped it. Either way, add in a few usage cycles and some vibration and it'll really start to overheat. Ask the EC to check the hole threads before reusing a slot in the panel.

 

[kwired]

To impact multiple breakers you would expect to see heating on the bus(es). Probably smart to replace breakers and bus whether that means swap with new or replace with something else.

 

[11bgrunt]

To impact multiple breakers you would expect to see heating on the bus(es). Probably smart to replace breakers and bus whether that means swap with new or replace with something else.

Attached is one thermal picture. I saw the heat on the breaker and not the buss.

 

[kwired]

Attached is one thermal picture. I saw the heat on the breaker and not the buss.

How much run time at that point?

Add: breaker 40C rating is ambient temp that they are tested/calibrated for, they can and will run at higher temps depending on load conditions. I can't tell you what is an acceptable reading though, but should take some time to reach the upper limits, if this is after only a short time there is a problem within the breaker I would think.

 

[11bgrunt]

How much run time at that point?

Add: breaker 40C rating is ambient temp that they are tested/calibrated for, they can and will run at higher temps depending on load conditions. I can't tell you what is an acceptable reading though, but should take some time to reach the upper limits, if this is after only a short time there is a problem within the breaker I would think.

The 52 lights were on for 30 minutes.

 

[zbang]

Can't completely discount a bad batch of breakers, although I think that's a stretch.

Make sure the EC notes or labels the replacements with date and slot number (in case they get moved later).

 

[mopowr steve]

Are the breakers HID listed?