Suggestions ???

Status
Not open for further replies.
I have a question, not trying to jump off the path here but just want to understand something in my head. Maybe I'm missing something. Since this is a 3 wire from the disconnect shouldn't the grounded conductor and the grounding conductors be seperated in the panel of the out building? Now I'm saying that based on it being bonded at the disconnect, if it's not shouldn't they still be split and a rod(s) be driven at the outbuilding panel? I'm a newbie at this but I wanted to make sure, as I mentioned maybe i'm missing something here or read to fast. :-? :-? I know he's planning on seperating them but I just want to make sure that i'm thinking correctly.
 
080614-2141 EST

AWC:

Think this thru.

What are typical resistances thru the ground from one ground rod to another spaced maybe 50 ft apart? You can do some experiments to get an idea for your area (use 28 V AC). You can look at the specs in the NEC for maximum ground rod resistance and judge from this.

Now isolate the neutral and EGC at the subpanel and connect the neutral to a ground rod at the subpanel. This means there is no path for current into the subpanel EGC except thru earth back to the center tap of the supply transformer.

Now assume the ground resistance is 10 ohms. Thru a short circuit from a hot wire at the subpanel to the EGC you apply 120 V to the EGC. How much current flows in the EGC. What is the resistance of 50 ft of 000 copper wire? With the hot wire at the subpanel shorted to the EGC what is the voltage drop on the 000 wire due to the short, and in turn what is the voltage on the EGC wire assuming 120 V at the home main panel?

Doing this analysis may answer your question.

.
 
Wow...........went out for dinner and missed alot here.

I never really thought about a problem with the frig. It appears fairly new. I'm going back next week and will check that first. I'm going to run an extension cord to the GFCI protected outlet for the pool pump and see if it holds. As stated previously, I doubt that it will.

I will also check all the connections in this panel along with checking other circuts for voltage on the EGC.

I am also going to drive a gound rod at the panel. I doubt if that will "fix" anything but I feel the panel needs supplmentry grounding.

What I'm not clear on is weather the EGC and the Grounded Conductors should be isolated in this panel.

The building has a concrete floor, btw.

Still looking for any suggestions.

Thanks again.......
 
gar said:
080614-2141 EST

AWC:

Think this thru.

What are typical resistances thru the ground from one ground rod to another spaced maybe 50 ft apart? You can do some experiments to get an idea for your area (use 28 V AC). You can look at the specs in the NEC for maximum ground rod resistance and judge from this.

Now isolate the neutral and EGC at the subpanel and connect the neutral to a ground rod at the subpanel. This means there is no path for current into the subpanel EGC except thru earth back to the center tap of the supply transformer.

Now assume the ground resistance is 10 ohms. Thru a short circuit from a hot wire at the subpanel to the EGC you apply 120 V to the EGC. How much current flows in the EGC. What is the resistance of 50 ft of 000 copper wire? With the hot wire at the subpanel shorted to the EGC what is the voltage drop on the 000 wire due to the short, and in turn what is the voltage on the EGC wire assuming 120 V at the home main panel?

Doing this analysis may answer your question.

.
Click to expand...


Ok, I think maybe i'm getting this but would the best way to have done this from the beginning been to have the fourth conductor for the EGC? Sorry for the delay but didn't get a chance to examine this until tonight. Would you want to treat it as a branch circuit? Sorry if you're ready to slap me but I just want to be straight on this. Thank you in advance for your help
 
awc said:
Ok, I think maybe i'm getting this but would the best way to have done this from the beginning been to have the fourth conductor for the EGC? Sorry for the delay but didn't get a chance to examine this until tonight. Would you want to treat it as a branch circuit? Sorry if you're ready to slap me but I just want to be straight on this. Thank you in advance for your help
Click to expand...

Slap me to........I had some extra time today so I dropped by to investigate further.

I ran the refrig. to a GFCI on the house and it did not trip under load. The voltage is present on the other 120V circuts, theres only 6 of them, so the problem is not the frig.

I checked all the connections in the panel and the disconnect on the house. I did find the neutral connection very loose on the load side of the main disconnect breaker (about 2-3 full turns) but tightening that didn't make a difference. I was alittle suprised I didn't see in voltage swing with as loose as it was.

I also turned off the breakers one at a time and the voltage didn't change. I also disconnected all of the EGC conductors and that didn't effect anything which lead me to believe that the problem was upstream of this panel.

I did not have a ground rod on the truck but I question weather that will fix anything other than its needs to be done to ground the service.

No one was home so I didn't check the service inside or did I open the Meter Socket.

I wonder if maybe they have the same problem in the house and it has not been noticed.

I also saw some cooper plumbing in this building, the house is all plastic for what that worth.

Please excuse my ignorance, I would like to remedy this problem and learn in the process.
 
080617-0740 EST

AMC: and ElectricianJeff:

You need to ask iwire about any specific code requirements.

If you have 3 wires from the house and one is neutral and the neutral is properly connected to earth at the service entrance or main panel, and there are no currents in the earth, then the earth anywhere around this yard should should be at the same potential as at the main panel and this will be the potential of the neutral at the main panel.

If here is no current flowing thru this neutral to the outbuilding and no earth currents, then at the outbuilding the voltage difference between the cement floor and the neutral should be zero.

If either, but not both, the neutral or earth has a current flowing thru it, then there will be a voltage difference. If there is current in both, then there is probably a voltage difference.

Now suppose we have load current flowing in the neutral and no stray currents in the earth, then there is a potential difference between the cement floor and the neutral. 50 ft of 000 copper is 0.062*50/1000 = 0.0031 ohms. Further, suppose that 100 A is flowing in the neutral, then the voltage drop on the neutral is 0.3 V. Probably not enough to cause a tingle, but I have no specific knowledge on the threshold voltage for a tingle if you had perfect conductivity into the body. If there are poor connections in the neutral then even at lower currents there could be a much higher difference voltage.

In my post #9 above I outlined measuring voltages relative to the ground rod at the home. ElectricianJeff -- I suggest you make some of these measurements with and without any current flowing to the outbuilding. Use a high impedance meter like a Fluke. A screwdriver can be your test probe.

If you find voltages in the earth around the outbuilding relative to the home ground rod, then you have ground current from somewhere.

At
http://forums.mikeholt.com/showthread.php?t=99809&highlight=ground+voltage+measurements
post #37
I discuss measurements I made in my yard. Because the voltages are in the millivolt range and less than 1 V it is clear that ground currents around me are quite low.

I used the oscilloscope in post #37 to prove that the voltages were a result of the power distribution system. Generally a high input impedance meter will be adequate.

.
 
the tingle means there is current flowing ,.. You said there is voltage on all six EGC when removed and leads are from conductor to neutral bus,.. do me a favor and kill the power to house and out building,,.. repeat test.,.I think you will still have current flowing ,..If not check all neutral connections
 
Thanks for all the good explanations and suggestions.

I think this is starting to sink through this thick skull. I will try and slip out there Sunday when the owners are there and see what I can find.

Jeff
 
There was a lot to read to catch up on here and I missed where it was stated if the fridge was in the outbuilding? Assuming it is then as Gar and I think others pointed out there is a parallel path for current to return to the house service when the kids touch the fridge and stand on the concrete.

I would expect to read a voltage from the frame of the fridge to neutral when you isolate the EGC, possibly due to internal capacitance leakage to the frame when ungrounded.

As Gar suggested -reading the potential from the metal frame of the fridge to the concrete the kids stand on (all EGC connections still made).

Is the refrigerator isolated from the concrete floor? If so then the kids become a parallel path from the fridge frame (at the same potential as the neutral/egc in the remote panel) via the concrete floor back to the main service.

It might be as simple as making sure the fridge frame is not isolated from the concrete. If a very low impedance path were added from the fridge frame to the concrete (perhaps the ground rods mentioned) current would take that path rather than through the kids?
 
ElectricianJeff said:
I think this is starting to sink through this thick skull.
Jeff
Click to expand...


Ditto, now it's starting to make sense. Sometimes when i'm trying so hard to figure something out I tend to overlook a small important detail. :grin: Good Luck Jeff..........and thanks GAR!!
 
080621-1356 EST

M.D.

Your referenced article was interesting. I would like to see some terms with a more clear cut definition, and several figures were missing. For someone that lacks a good understanding of electrical circuit theory this article may not yet be very useful.

At my home I have what can be considered as an ungrounded floating delta. I do not know what the source is. It may be a grounded Y, but I only have three primary wires. Within my yard the maximum difference in ground potential is about 200 MV at the worst spot. My center tapped pole secondary is grounded at the pole. There is no grounding of the primary anywhere near me.

Generally my ground resistance is in the 15 ohm range over a distance of 50 ft between 8 ft vertical rods.

Nobody within 500 to 1000 ft of me will have a primary service different than mine. I do not know the location of the substation.

If we assumed the ground resistance from one side of my lot to the other was 1 ohm, probably is not, then I have 200 ma spread over a cross-sectional area of at least 3000 sq-ft (150 ft long x 20 ft deep). Assuming a uniform distribution this is less than 200 microamps per sq-ft.

My primary service was probably installed in the early 1960s, at least before 64.

All my water lines are copper, my supply line is 150 ft of copper to the street. All drains are copper to cast iron and under the basement comingled with the copper water supply.

My water is highly conductive so it should have a potential of the copper pipes. This I will measure.

If all plumbing is highly conductive and bonded together and you do not have 100s of amperes flowing into this plumbing, then how do you get a large potential differences along this piping, and thus a large voltage difference between the water supply and the faucet? It is a function of the relative impedances and currents.

.
 
Gar I'm not an EE All I want to do is get Jeff thinking that the problem might just be stray current due to the multi grounded neutral system.. He has voltage on all six grounding conductors it seems to me stray current is trying to get back home.. If he opens all disconnects and there is still voltage/ current , a very long and ,most likely ,frustrating conversation will need to take place between him and the utility.
 
080621-2113 EST

M.D.

To get a 10 V reading on a 10 megohm input resistance meter only requires 1 microampere. This is way below the threshold of human tingle detection. But we do not know the resistance of the meter used because it was not defined. When the EGC from the refrigerator was disconnected from the building EGC wire and the voltmeter was placed between the two, then we were in effect using the meter as a microampere meter instead of a voltmeter.

If this measurement was made with a low resistance ammeter then the information would be more useful. We would want to be making measurements in the fractional milliampere range, like 100 to 10,000 microamperes (0.1 to 10 MA).

Now if I take that same meter and use it to measure earth potentials the 10 megohms is advantageous because it is a high resistance relative to the earth resistance and thus does not much effect the actual ground potential.

.
 
I think you have a bad ground from the main panel to the fridge receptacle.
you could use a simple cube receptacle tester to check this. Many times the lighting circuits in a fridge go bad and get shorted to the metal of a fridge from sockets and or door switches rusting out.
The voltage is either coming up from the concrete and flowing to the grounded refridgerator door.
Or the refridgerator door is energized and flowing down to the concrete which is the more likely scenareo.
I have never seen a full sized fridge without a 3 prong plug and I would think the frame of the fridge would be grounded.
In your test with the gfci on the fridge you should have put a jumper wire from the door to the wet floor or something with a known good ground and seen if the gfci would trip.
Then you know the objectional current is coming from the fridge door and not the floor.
If the door became energized and it did not have a path to flow to the gfci experiment would not trip the device.
 
quogueelectric said:
I think you have a bad ground from the main panel to the fridge receptacle.
you could use a simple cube receptacle tester to check this. Many times the lighting circuits in a fridge go bad and get shorted to the metal of a fridge from sockets and or door switches rusting out.
The voltage is either coming up from the concrete and flowing to the grounded refridgerator door.
Or the refridgerator door is energized and flowing down to the concrete which is the more likely scenareo.
I have never seen a full sized fridge without a 3 prong plug and I would think the frame of the fridge would be grounded.
In your test with the gfci on the fridge you should have put a jumper wire from the door to the wet floor or something with a known good ground and seen if the gfci would trip.
Then you know the objectional current is coming from the fridge door and not the floor.
If the door became energized and it did not have a path to flow to the gfci experiment would not trip the device.
Click to expand...

Thanks for these tips, dang I'm getting some great trouble shooting ideas from this thread.

The fridge is less than 2 years old. It is on its own circut which is all of 2' long. Actually you have to pull out the frig. to get to the panel. I know clearence issues and all but I didn't design the kitchen area.

Thanks again....I'm still hoping to make it out there today.
 
Status
Not open for further replies.
Top