Lighting burning out Capacitors

[MartyParsons]

Hello,
You guys seem to know lots. We have had an issue with 400W lights in our shop. We replaced the Capacitors in them about 6 months ago and now we are in the dark again. I removed the lights and the capacitors are open. What would cause them to fail? The information I seen on the net was a 60,000 hour fail rate. We have about 14 of them and 6 are not working.

Here is some information on them. 

OVP2405400
24MF
400 Volt.AC 50.60 hz

the other one I have is one we just replaced in June

SH Capacitor25 MFD
440 VAC

[pineywoods]

Open capacitors are a bit unusual, the normal failure mode is a direct short between the terminals. In either case, the cause is usually a high voltage spike that either arcs through the internal insulation resulting in a short circuit, or melts the foil where the wire leading to the external terminal is attached causing open. Lightening hits on the power line is usually to blame. The best defense against lightening is to fool it into hitting somewhere else.

[Magicman]

Chances are that your neutral is not properly or poorly grounded to earth.  First, look for loose or corroded connectors.  Have the power company check their neutrals before it hits your entrance.  Are they properly grounded at the service pole?  Is your meter base properly grounded?

If all is OK, then you may need to build a ground bed.  Our specifications call for three copper clad 8' ground rods spaced in a triangle 10' apart.

You have to realize that the sky is one and the earth is the other electrode with lightning.  Do all that you can to shorten this path without traveling through you.

[Gary_C]

First thing to do is call your electric utility and report the problem to them. Chances are very high it's their problem but knowing them, they will deny it. But hopefully they will quietly fix the problem.

Is your utility PECO?  ;D

[Qweaver]

Hi Marty,
This may seem too simple but have you measured line voltage? We had a transformer go bad once and the voltage just kept going higher.  Wierd things were happening with our appliance displays and computers.  When I measured my voltage it was @ 270v.  By the time the power company got there it was almost 300v but I had already flipped the power off to the house by then.  They changed the transformer and all was well again. 

[Woodwalker]

If it's an open neutral on the supply side or a transformer going bad , your going to have more than just a few lights burned out. With either situation your line to ground voltage is going to act like a seesaw, up on one leg down on the other.
With more than one light involved then is the problem common to all on the same circuit?. With around half involved, check connections\grounds at junctions boxes and the panel. Check the voltage at the panel and at the end of the circuit. When checking for an open neutral you have to have load on that circuit.
The goal of any grounding system is to provide a low impedance path for fault or transient
currents to the earth. It's not the number of ground rods your connected to, it's the impedance value.
Piney's right about  the causes of capacitor failure. Secondary lighting arresters may be something to look at should lighting be the culprit.

[MartyParsons]

Thanks for the information. I called the electrician about two weeks ago, he said he would be right up. You know the rest of the story. I removed the light and put a new T 5 light in it's place. The electrician was here and repaired the lights that were out. I kept asking why they burnt out. He did not know. We did not install any ground bank here. I have seen them in the past. I will also check with the electric company. We have Valley Rural Electric co op.
Secondary lighting arresters may be something to look at should lighting be the culprit.  Can you explain?

[Magicman]

Marty, we can't change lightning, we can only work around it and do the things that past experience has taught us.

A direct lightning hit is unavoidable, and nothing that we can do can change or stop it.  Lightning that has traveled through the sky at 186,000 miles per second is not going to suddenly change it's direction because of something that we do or not do.  Most lightning damage though is not a direct hit, and I imagine that your damage fits that category.

We can learn to live with it, design, build, and maintain our electrical systems to proven specifications.  Don't be too hard on your electrician.  I'm sure that he knows his stuff about wiring and changing out zapped electrical components.  Some soils are not very conductive and I have seen conditions where ground beds were necessary.  I doubt that to be your case, but it is a possibility.

Your power company should have an engineer that will come out and verify their grounding and advise you regarding your requirements.  Bottom line, if everything is properly grounded, you should not be having reoccurring failures.   

[Gary_C]

Your power company should have an engineer that will come out and verify their grounding and advise you regarding your requirements.  Bottom line, if everything is properly grounded, you should not be having reoccurring failures.   

Magicman is right. The big fault with the electric distribution system in this country is the lingering use of the earth as a conductor of electricity thru the grounding system. In some areas the soil conditions are just not able to provide a good ground. But I know all electric utilities are sensitive to these grounding problems and will do their best to eliminate them, if for nothing else but liability protection.

For sure call them to check everything.

[beenthere]

It is lightning (electicity bolts from the sky) here and not lighting...right?   ;)

[PineNut]

Just because a capacitor test as open does not mean that it is not shorted. Some capacitors have devices in them to open the circuit in the event of a short. Not a fuse but a device that detects the swelling or overpressure of the can and opens the circuit.

[Kevin]

You can test the ground and make sure it's 25 ohms or greater, some soil conditions aren't good ground.
Make sure everything grounded is bonded together, water, power, cable etc. so there is no difference in potentional.

[submarinesailor]

You can test the ground and make sure it's 25 ohms or greater, some soil conditions aren't good ground.
Make sure everything grounded is bonded together, water, power, cable etc. so there is no difference in potentional.

Kevin - can you give me a better explanation on the 25 ohms or greater.  Knowing a little bit about electricity, shouldn't that be 25 OHMS or less?  Knowing that point of reference is very important in electricity, could I be thinking of the wrong point.?

Bruce

[Gary_C]

Testing the ground is a very slippery slope. The problem is what is your reference? If your reference is the ground, then you are in reality testing two ground connections and the conductivity of the ground between the two grounds.

This has been a huge problem in the dairy industry. If you look under the topic "stray voltage" you will find many references to all kinds of studies and rules for combating stray voltage which is a misnomer and really deals with faulty grounds in the electrical distribution system. 

There have been countless lawsuits and even at least one electrocution of a dairy farmer that had such a stray voltage problem and resulting loss of milk production that he choose to disconnect the neutral and operate with a floating ground. Most public utilities in this country have been in denial of the problem, mainly for liability concerns but they have also had to deal with some large losses as a result of their denials.

Now all electric utilities have a greater awareness of the problem of poor grounds in the distribution system and will rapidly investigate questions and complaints. They now have established proceedures for checking grounding of the system and ways to deal with problems. And in many cases the problem is not right at the location that is having the problem. So that's why it is important to get them involved right away.

[Larry]

When I started my house I decided to hire an electrician as I wanted to change the meter base at the power pole.  While he was there I also let him install the 4/0 and conduit to the house.  Watched with amusement as he put in the ground rod...knew he would hit bedrock at 4' depth.  Sure nuff he cut the ground rod off when he hit bedrock and called it good.  He also drove the rod to close to the house where the roof line/gutters would keep the ground dry.

He did inform me the latest NEC required a fourth wire from the house to the power pole to connect the grounds together.  Something new for me.

I mentioned to the electrician that I had installed a Ufer ground system in my slab but that term went straight over his head.  So much for electricians...I completed the rest of the house wiring myself.

I suppose the point of this story is your ground system is always suspect until verified.  If inadequate it is easy to upgrade using several cures.

[Kevin]

Bruce, depends on how you look at it.
Greater is better with zero being the best.
You need a known ground to measure from.
At home in the city I would use the copper water line but that could go back to plastic in the rurals.
We have areas here that are gravel beds and have little or no grounding effect.
Capacitors usually blow on a voltage spike, they can only handle what they are rated for.

[Woodwalker]

Not anything to do with capacitors but on "stray" voltage's and Ufer systems;
Due to widening of a highway, we were building a new three phase line parallel to the existing one. New line was about 30' over and this section was about 3,000' long. It was sagged, tied in to the insulators and not connected to anything on either end. There was three sets of working grounds (temporary 4\0 cu ground sets that mechanically tie all three phase conductors to the neutral).  
The only thing is the road\line passes in front of a 60,000 watt AM radio transmitter. AM use the earth as one of the components to transmit the signal. With the new line tied to the earth through the ground sets we had a situation where "grounded" means it is "hot".  

NOTE FROM ADMIN. BE FOREWARNED THE FOLLOWING VIDEO LINK CONTAINS ROUGH LANGUAGE. VIEW WITH THAT KNOWLEDGE
This http://www.youtube.com/watch?v=cEAKQ8xoFFQ  link is to a You-tube video shot by the crew. The language is very rough.
About four minutes into the video they have a tail of the wire connected to the new construction, grounded on the first pole and they are drawing a arc from the wire to an anchor and to the ground. If you listen you can make out some music notes.

[DouginUtah]

When I built this house the county inspectors required that the electrical ground be connected to the rebar in the footings.

[Larry]

The reason Ufer's are not mandated countrywide is because a lot of places don't require steel in the footings/slabs.