Sciencemadness Discussion Board

electrical safety problem

Magpie - 19-2-2016 at 17:09

I couldn't use my lab today due an electrical safety problem that so far I have not been able to solve.

When I built my hood some 13 years ago I placed a 4' fluorescent light on top of a glass plate laying in an aluminum frame. The light has 2ea 32w bulbs and is powered by 120vac. Somehow this light has a weak short to my hood pan which is 304ss. The short is after the switch.

I discovered this today when I touched the bottom of my magnetic stirrer-hotplate on the lip of this pan. The hotplate was pugged in but not turned on. Testing with my volt-ohm meter indicated the voltage source was the pan and the hotplate had provided a route to ground. There was a spark and the 20 amp circuit breaker was tripped. I'm glad I wasn't using any ether at the time!

I took the lamp down, cleaned everything of dust, and inspected the internal wiring. There was nothing out of place. I put the lamp back in service, but the voltage is still there. (Definition of insanity, I know.) The pan is typically reading 2-3vac but I've seen it spike to 8vac.

This voltage may have been there since day one - I have no idea about that. But I would think I would have noticed it before in 13 years of use.

I don't know what to do next. Does any one have any suggestions? If you need more details please ask.

[Edited on 20-2-2016 by Magpie]

blogfast25 - 19-2-2016 at 17:40

Get a qualified electrician in. No, seriously, don't 'messy with the leccy'...

BromicAcid - 19-2-2016 at 17:52

So the voltage is not there without the lamp in place even with the switch on, is that the case. I mean, if it is you really already have it narrowed down to the lamp. Add additional insulation between the lamp and the hood itself or buy a new lamp.

JJay - 19-2-2016 at 18:03

You may have a bad ballast. You may also have some faulty wiring somewhere.

This guide will tell you if you have a bad ballast: http://code-elec.com/content/00/01/53/48/38/userimages/fluor...

I have done some work as an industrial electrician, and personally, I wouldn't recommend hiring a union electrician for anything unless you are doing new construction or major renovation and want to ensure code compliance.

m1tanker78 - 19-2-2016 at 18:08

I'm not sure what you mean by 'hood pan'. I assume that's where you set your hotplate, stands, etc. Do you have a water source plumbed to your hood?

I recommend buying an electrical tester (the type you plug into a socket -- has 3 neon bulbs and tells you if there's an open ground or switched neutral and hot). These help to quickly diagnose wiring problems.

If you have plumbing in the hood, it could be that a lazy/unskilled electrician wired the neutral line of another circuit to a nearby water pipe rather than splicing or adding extra wiring. 3V is certainly not a dead short but could be a voltage differential between 2 circuits in your house. Also, check that your earth ground wire(s) are making good contact with the ground rod outside the house.

If you unplug the light fixture, does the pan remain electrically hot?


JJay - 19-2-2016 at 18:27

Quote: Originally posted by m1tanker78  
I'm not sure what you mean by 'hood pan'. I assume that's where you set your hotplate, stands, etc. Do you have a water source plumbed to your hood?

I recommend buying an electrical tester (the type you plug into a socket -- has 3 neon bulbs and tells you if there's an open ground or switched neutral and hot). These help to quickly diagnose wiring problems.

If you have plumbing in the hood, it could be that a lazy/unskilled electrician wired the neutral line of another circuit to a nearby water pipe rather than splicing or adding extra wiring. 3V is certainly not a dead short but could be a voltage differential between 2 circuits in your house. Also, check that your earth ground wire(s) are making good contact with the ground rod outside the house.

If you unplug the light fixture, does the pan remain electrically hot?



Unfortunately, things like this do happen.

Oh and it is also a good idea to have a pen-type voltage detector if you do a lot of that sort of thing.

Magpie - 19-2-2016 at 18:42


@blogfast25:
I've been doing my own simple household wiring for 50 years and never had a problem. I will go to almost any other measure rather than invite a craftsman into the sanctum santorum of my lab. That's an invitation for a knock on the door, uninvited.


@BromicAcid:

Yes, when the lamp pigtail is disconnected at the wire nuts and the switch is turned on there is no voltage at the pan.

@JJay:

How could a bad ballast put a voltage on my hood and pan? magnetic induction?

Quote: Originally posted by m1tanker78  
I'm not sure what you mean by 'hood pan'. I assume that's where you set your hotplate, stands, etc. Do you have a water source plumbed to your hood?


The aluminum hood sets on a wood bench. The pan sets on the aluminum floor of the hood. The hood also carries this voltage.

Quote: Originally posted by m1tanker78  

If you have plumbing in the hood, it could be that a lazy/unskilled electrician wired the neutral line of another circuit to a nearby water pipe rather than splicing or adding extra wiring. 3V is certainly not a dead short but could be a voltage differential between 2 circuits in your house. Also, check that your earth ground wire(s) are making good contact with the ground rod outside the house.



I have no sink or plumbing in the hood. About 14 years ago I hired an electrician to put in a new Square D service entrance - I had a Zinsco which are notoriously bad. He put in a new ground rod at that time.

Quote: Originally posted by m1tanker78  

If you unplug the light fixture, does the pan remain electrically hot?


Yes, see the response to Bromic above.

BTW the lamp is working fine.

[Edited on 20-2-2016 by Magpie]

[Edited on 20-2-2016 by Magpie]

edit: major correction: hood and pan are metal to metal contact so of course the hood is hot too.

[Edited on 20-2-2016 by Magpie]

blogfast25 - 19-2-2016 at 19:12

Quote: Originally posted by Magpie  

@blogfast25:
I've been doing my own simple household wiring for 50 years and never had a problem. I will go to almost any other measure rather than invite a craftsman into the sanctum santorum of my lab. That's an invitation for a knock on the door, uninvited.


G-d, it must be really crap living in a police state. :( Become a member of the Stasi yourself, so you can keep an eye on them, from the inside! :D


[Edited on 20-2-2016 by blogfast25]

JJay - 19-2-2016 at 19:32

Quote: Originally posted by Magpie  

@JJay:

How could a bad ballast put a voltage on my hood and pan? magnetic induction?



A short to the ground.... But if the voltage is there with the fixture disconnected, it must be something else.

Magpie - 19-2-2016 at 19:40

JJay's link gave me a clue that a poor ground might be the problem. I just checked the ground wire with the switch on: 50,000 ohms on a continuity check - a very poor ground! I hope that is the problem.

edit: the lamp was on when I made that check. I should check with the lamp disconnected.

[Edited on 20-2-2016 by Magpie]

Magpie - 19-2-2016 at 19:52

Yes, there's no ground. It must have become disconnected in the switch. I'll look into this tomorrow. I have to go to a Stasi meeting now. ;)

Thanks for all the help.

gsd - 20-2-2016 at 04:03

@ Magpie

This certainly looks like an earthing fault.

To be on a safer side, inter-connect all the metal parts of your hood (including the fan) and provide a separate ground to the hood (independent of your house earthing connection). Better to use PVC sheathed copper or GI strip.

gsd

Magpie - 20-2-2016 at 14:11

Upon opening up the switch I found that the incoming and outgoing ground wires had not been connected (chagrin). I don't know why I did not do this. They are now connected to each other and the ground lug on the switch.

There is no longer 3-8vac on the pan/hood system and touching this to ground does not cause a spark, ground fault, or breaker trip. However, there still is a slight voltage ~0.5vac on the pan/hood. I don't know the source of this voltage.

Quote: Originally posted by gsd  
@ Magpie
To be on a safer side, inter-connect all the metal parts of your hood (including the fan) and provide a separate ground to the hood (independent of your house earthing connection).


Yes, I am going to ground this system as you suggest. How do I get a separate ground? Will I have to drive a copper rod into the earth and run a wire from it?

Quote: Originally posted by gsd  

Better to use PVC sheathed copper or GI strip.


Is the GI (galvanized iron) strip also buried, or what? I don't know about these items and will have learn about this.

[Edited on 20-2-2016 by Magpie]

IrC - 20-2-2016 at 15:57

Does the light use a ballast transformer? Some newer types do not they have a ceramic resistor heat sinked to the lamp housing. If so as mentioned earlier the fault could be there. Run the light for a while and see if it feels hot. Very possible enameled turns have burned through the coating and Cu wire is in contact with the core. Going by the voltage you mentioned it would be at a point of lower potential in the circuit, causing unwanted current to flow to ground. The cure besides doing the grounding you have done (these type lights do not always start quickly unless the ground to frame exists anyway - excessive filament heating times will tend to blacken one or both ends of the lamp(s)), and of course replace the ballast transformer. I have seen them get so hot that over years potting oozes out of the ballast.

arkoma - 20-2-2016 at 16:29

Be careful with grounds. Subpanels, for instance, don't have separate grounds. All earthing/grounding to the physical earth is done at the service entrance. There can be a difference of potential between different grounding points connected to the same mains, leading to shock/spark hazards also.

m1tanker78 - 20-2-2016 at 17:01

You should already have a earth ground source on the circuit you plug the hood accessories into. You may be able to 'borrow' a ground from the lamp, fan, whatever. If the hood is Al, make sure you grind a spot and tighten the crap out of the bolt/screw you use to connect the hood to the ground wire lug.

The .5V reading you get is probably characteristic ambient AC (will show a higher reading if you connect one of the leads to a larger antenna -- the hood).

First make sure that the earth ground is good to go on your lab circuit. Again, the plug-in circuit checker will answer that for you in 2 seconds. Good to keep one in the toolbox. JJay's suggestion of the pen type tester will alert you if there's a dangerous live AC voltage present without physically contacting the wire or whatever you're checking. Test it on a live circuit to make sure it works.

And yes, I've also seen ballasts that run too hot dump their pitch. After 13 years, I'd say you got your money's worth on that fluorescent lamp. No shame in replacing it if need be. :D

Magpie - 20-2-2016 at 18:51

Quote: Originally posted by IrC  
Does the light use a ballast transformer?


Yes, it has the transformer type ballast. I just ran it for over an hour. The ballast, felt through metal housing, was just warm, not hot. The bulbs look good - no dark areas, and start quickly.

To reiterate for clarity: the 0.5vac on the hood only appears when the lamp is on. The lamp touches the glass plate only.

And yes, after 13 years I've got my money's worth. Actual hours on the lamp probably aren't that many, however.

That was quite a spark I got before, cutting a tiny piece of ss out of the pan lip. If that would have happened when the hood was full of ether fumes - well, it could have been ugly. I can't believe that pan has been hot for 13 years.

@tanker: Yes, I have grounds at all my household, garage, and lab outlets. These grounds are all led back to the service box. From there the bus wire goes to copper ground rods buried in the backyard, installed about 14 yrs ago by a qualified electrician. My grass covered dirt yard is probably as wet as it ever gets right now due to a lot of rain this winter.

[Edited on 21-2-2016 by Magpie]

[Edited on 21-2-2016 by Magpie]

IrC - 20-2-2016 at 20:57

Sounds like the ballast is not a problem but the current that tripped the breaker came from somewhere. Reading the OP again has me wondering why you quickly dismissed the hotplate as the trouble. Keep in mind there could be an intermittent fault that may not always exist. At the moment you tested it you may have seen no problem. Something to consider. A leakage within allowances may show as a voltage to ground, I would be more concerned with the leakage current. A current flow high enough to trip a breaker is a completely different concern however.


Magpie - 20-2-2016 at 21:58

Quote: Originally posted by IrC  
Sounds like the ballast is not a problem but the current that tripped the breaker came from somewhere. Reading the OP again has me wondering why you quickly dismissed the hotplate as the trouble. Keep in mind there could be an intermittent fault that may not always exist. At the moment you tested it you may have seen no problem. Something to consider. A leakage within allowances may show as a voltage to ground, I would be more concerned with the leakage current. A current flow high enough to trip a breaker is a completely different concern however.


Yes, troubling questions remain.

At first I did think it was the hotplate as it was recently bought used off eBay. But I was touching the bottom of same and not getting shocked. I quickly poked it in several places on the chassis bottom with my volt-ohm meter but there was no voltage. Then I tested the pan where it was reading 3-8vac.

IrC - 20-2-2016 at 22:16

You need someone skilled in wiring to take apart and visibly inspect everything while making measurements. Not sure how helpful anyone can be if unable to physically see everything.

gsd - 21-2-2016 at 07:15

Quote: Originally posted by Magpie  


Yes, I am going to ground this system as you suggest. How do I get a separate ground? Will I have to drive a copper rod into the earth and run a wire from it?


Is the GI (galvanized iron) strip also buried, or what? I don't know about these items and will have learn about this.



To get good earthing you need to dig about 2 X 2 feet ditch which is at least 2 foot deep (deeper the better) in you garden and burry a square copper plate (about 6" X 6") to which the copper wire or GI strip running from your hood will have to be bolted. Fill the ditch with charcoal and saw-dust and add generous quantity of salt and cover with mud. Make sure that the place remains wet by occasionally adding water on the top.

gsd

Magpie - 21-2-2016 at 14:55

This morning I measured the voltage on the hood with the light on. It was 2vac. :(

I can't speculate on how this could happen other than the light was generating an electric field picked up by the hood acting as an antenna.

At my wits end I decided to buy a new lamp. I prefer not to troubleshoot electrical problems by component replacement but I felt I had no other choice.

I decided to change light technology and buy an LED shop light drawing 35w but equivalent to a 64w fluorescent light. I am happy to say that I no longer have any voltage impressed on my hood! Also, the lighting is brighter. :D

Thanks to all for your suggestions and insight.

chemrox - 25-2-2016 at 14:51

Quote: Originally posted by Magpie  

@blogfast25:
I've been doing my own simple household wiring for 50 years and never had a problem. I will go to almost any other measure rather than invite a craftsman into the sanctum santorum of my lab. That's an invitation for a knock on the door, uninvited.


@Magpie
I think you're being paranoid. I've had the electrician in several times. He wired my hood. But- I don't know what else you have or how obvious your peregrinations with energetics or whatever are. Hide the fucking nitrates if you must. Maybe the electrician will go home singing..I do not mess around with current. As you have surmised there is a ground missing somewhere. Can you ground the misbehaving part? Separate ground wire to the frame? something like that? If you still worry about snitching servicemen get to know one in your area.. take him out for a drink...

Magpie - 25-2-2016 at 15:04

As stated above, after removing the 13yr old fluorescent light I no longer have an impressed or inducted voltage on my hood. I'm no longer worried about it.

I still find it amazing that my original discovery was initiated by a spark and the tripping of a 20a breaker. :o

"A little paranoia is healthy!" :D

I have had craftsmen in my garage. I just don't want them rooting around in my lab.