SCENARIO:
A guest was leaving after visiting our pool.
He claimed to have "gotten a little jolt" from my mudroom
lightswitch. The next day I could not replicate the shock
on myself... perhaps he was wet (?)... so I perfromed a series
of tests with my digital multimeter. I recorded the results
on a single piece of paper (scan attached: test1.jpg).

BACKGROUND:
The switch is part of a 5-yr. old addition built by a
bankrupted contractor as new construction that I bought 4
years ago. Most wiring is now completely inaccessible to inspection.

TESTING:
What I know from the 16 possible combinations of switch positions:
1)Two double switches control three lights: 1 and 2 (garage), 3 and 4 (mudroom).
2)Switches 1 and 3 control a breezeway light "in between" them,
except that both have to be "on" to make the light come on.
3)Switches 2 and 4 control other lights each... they work fine.
4)1,2, and 3 are controlled by a newer single breaker in the garage;
4 is controlled by a 20-yr. old breaker in the house.
5)It would appear hot leaks at 120V to ground on the switch cover screws of 3
ONLY under the following conditions: "1-off/3-off", "1-off/3-on"; 3 also
leaks a hot ground at about .2V ONLY with "1-on/3-off".
6)The voltage climbs slowly between the screw and ground... for instance, if I
touch the probe and kneel with bare skin on the cement (I wore rubber shoes)
it climbs quickly to 60V+ (although I felt no shock).
7)I cannot verify that any of the six breaker boxes on the property are
grounded, but my home inspector tested everything with a GFCI tester
so I think that probably would've been discovered. I simply can't find
any grounding rods anywhere except one attached directly to my pool pump
far away from any of this, the water pipes are all PVC, and I'm not originally
from far south Texas (they seem to have their own method of doing things).

CONCLUSIONS:
The connections to the breakers are all sound... no rubbing/touching/crossing.
The wiring in the garage attic to this area contains a lot of splicing/dicing
I would not have done outside of junction boxes... on the other hand the caps
are tight and there are no missing connections.
My solution so far has been to remove the screws on three... it is a double
switch plate held on by the screws to 4. I can find no bad wiring in the breezeway
light or anyplace hot or neutral touch ground where I can get at this wiring in its
20 ft. enclosed breezeway. This could be totally harmless after 5 years (I could put
in some plastic screws), but since I can't verify anything I am more
inclined to cap off the wires in switches
1 and 3 and leave the breezeway light without function in case there is a
hot bare ground touching wood somewhere (seems to me this would only flip
a GFCI breaker which this is not). On the other hand
I may be overlooking a bigger problem such as no master ground to the breaker box.
I get no readings from the grounds at the 1 and 2 switch, but
I can get continuity BETWEEN the screws on 3 and 4 which are on totally
different circuits, and since 4 is probably grounded correctly it leads me to
conclude a ground is loose/rubbing up in the breezeway where I can't get.
I'm guessing my guest was wet and well grounded, or simply bridged the 3/4
screws with fingers and is lucky to have discovered
this as only a brief grope for lighting. I have wired and re-wired many many things in my
life, but there are a lot of variables here and any addition to these ideas would be greatly appreciated... like maybe the switch itself is shorting??

Since you have a pool, that creates lots of problems. I's might initially assume that nothing is wrong, however I would d the following:

1. Replace switchplate with a plastic one. Decora style switches might be an option. Really concerned replace the metal screws with nylon.

2. Power circuit from a GFCI breaker.

Now to address the ground problem.

1. Is there a separate neutral and ground bus in your main panel. If so, the ground may be at the transformer or at the meter disconnect, if you have one.

2. There has to be a ground somewhere.

1. Outside disconnect (ground an neutral will be separate)
2. Main panel (ground and neutral will be separate)
3. At transformer
3a. Ground and neutral may or may not be separate

Usually there is a ground at the transformer and the house. The pool should have it's own rod because it is detached.

Ground can be a water pipe, but you ruled that out.

First, thank-you for the response. It is appreciated. The bold is NOT shouting,
But just highlighting my answers...

Quote:

---

Originally Posted by KeepItSimpleStupid

Since you have a pool, that creates lots of problems. I's might initially assume that nothing is wrong, however I would d the following:

THE POOL IS ON ITS OWN BOX AND METER AND SEPARATE FROM THIS. I MENTIONED
IT AS AN EXAMPLE OF WHAT I DON'T SEE. THERE IS NO TRANSFORMER GROUND. THE METER ON THE HOUSE IS IN SERIES WITH THE MAIN (ONLY) TRANSFORMER, AND THE ONLY GROUND I CAN GUESS AT IS COMING BACK INSIDE THE PANEL TO THE METER...BUT I can't OPEN THE METER TO SEE IF IF IT REALLY GOES INTO THE PVC PIPE TO THE GROUND.

1. Replace switchplate with a plastic one. Decora style switches might be an option. Really concerned replace the metal screws with nylon.

SWITCHPLATE IS PLASTIC ALREADY, CURRENT IS FROM SCREWS ONLY. WILL LOOK FOR NYLON.

2. Power circuit from a GFCI breaker.

I COULD BUY ONE FOR THE WHOLE GARAGE, BUT IT SEEMS CERTAIN IT WILL TRIP
IF GROUND IS ALREADY HOT (WHAT AM I MISSING?).

Now to address the ground problem.

1. Is there a separate neutral and ground bus in your main panel. If so, the ground may be at the transformer or at the meter disconnect, if you have one.

YES, IN FACT THEY ARE SEPARATE IN THE GARAGE MAIN PANEL AND 220V BREAKER GROUND WIRES ARE ON BOTH NEUTRAL AND GROUND BARS...BUT FROM WHAT I READ IF THERE IS A MAIN DISCONNECT IN THE PANEL (WHICH THIS GARAGE PANEL HAS) THEY ARE MARRIED AND THIS IS IRRELEVANT (PLEASE DISPUTE...I THOUGHT THIS WAS WRONG AT FIRST).

2. There has to be a ground somewhere.

AGREED. BUT IN MY MUDROOM WHY DOES IT SEEM TO BE FEET?

1. Outside disconnect (ground an neutral will be separate)

IT WOULD APPEAR THIS PANEL IS SECONDARY TO THE MAIN HOUSE PANEL WHICH FEEDS BOTH THE INSIDE HOUSE PANEL AND THE GARAGE PANEL.
NEITHER HOUSE PANEL HAS A MAIN DISCONNECT, BUT THE 200A PANEL IN
THE GARAGE DOES.

2. Main panel (ground and neutral will be separate)

THEY ARE SEPARATE IN THE GARAGE PANEL AND MIXED WITH 220V GROUND WIRES
ON THE NEUTRAL AND GROUND BAR WHILE THE 120V ARE ON THE GROUND BAR ONLY...BUT THE MAIN
DISCONNECT IN THIS PANEL IS SUPPOSED TO MAKE THIS NULL (?).

3. At transformer

I SEE NO GROUND THERE...WAS HIT BY LIGHTNING LAST YEAR WITHOUT ANY RAIN
WHILE I WEEDWHACKED A FEW FEET AWAY...LIKE A BOMB...THEY RESET IT AND REPLACED SOMETHING SMALL THAT TOOK THE CHARGE...NEVER SAW A DOG MOVE SO FAST IN MY LIFE.

3a. Ground and neutral may or may not be separate

RIGHT...DO I POWER UP THE PANEL WITH THE COVER OFF AND START FINDING OUT?
IS IT POSSIBLE THE BARS ARE NOT CONNECTED? WHY WOULD THE BREEZEWAY LIGHT ONLY LIGHT WITH BOTH SWITCHES ON INSTEAD OF EITHER?

Usually there is a ground at the transformer and the house. The pool should have it's own rod because it is detached.

I'M GUESSING THE GROUND AT THE POOL IS FROM A WIRE INSIDE THE POWER CORD TO
THE SPA HEATER OR PUMP...AND THAT EFFECTUALLY DOES THE WHOLE PANEL (?).

THE WATER PIPE ENTERING THE GROUND IS PVC.

Ground can be a water pipe, but you ruled that out.

---

Quote:

---

2. Main panel (ground and neutral will be separate)

THEY ARE SEPARATE IN THE GARAGE PANEL AND MIXED WITH 220V GROUND WIRES
ON THE NEUTRAL AND GROUND BAR WHILE THE 120V ARE ON THE GROUND BAR ONLY... BUT THE MAIN
DISCONNECT IN THIS PANEL IS SUPPOSED TO MAKE THIS NULL (?).

---

This is the one I picked to deal with. I'm either having trouble understanding or there is something wrong.

Let's go with the NORMAL case of ground and neutral are the same bus bar in the main panel and ground and neutral should be separate in the garage panel. If the garage is detached, then there should be a ground rod at the garage.

1. If the main panel is connected to a ground rod, then there should be a single large wire that seems to be out of place. Probably the largest bare wire. Similarly for the garage panel.

2. How about uploading a photo using "Go advanced" for the garage panel.

3. In the garage panel, only bare wires should be connected to the ground bus. Only white wires should be connected to the neutral bus.

For 240 ckts, white of a wht/blk cable can be connected to a 240 breaker.

For 240 ckts, red/black of a wht/blk/red cable will be connected to a 240 breaker and white to the neutral bar.

For 120 ckts, blk will be connected to the breaker and white to the neutral bar.

THERE SHOULD BE NO WHITES ON THE GROUND BAR. Bares only on the ground bar.

The garage panel may or may not contain a main disconnect.

BONDING SCREW

Look in the garage panel for instructions for a bonding screw on the neutral bar. It should not be in place. The screw is likely green.

Here is a source for Nylon screws:

Nylon Fasteners | Nylon Screws | Flat Head 82 degree Nylon 6/6 Machine Screws.Molded Nylon 6/6 Machine Screws offer excellant insulation in electrical and electronic appliactions. Lightweight and resistant to vibration and abrasions, nylon is nonmagn

In your drawing, you say both switches have to be on, I'm hoping you mean a 3 way(with 2 travellers) Shown it looks like neutral is gotten from 1 switch, and hot from the other switch. I would locate the problem of the shocks. Be sure the Neutral handles the load and the ground is intact between all boxes and fixtures. If something is shorted or leaking to something not grounded, it can become live.

First, thank everyone very muchfor their input(s)!

1)House panel has no main cutoff and 100A breaker to the garage panel.

There are no visible rods near the house or the garage, and no bare wires
On the way anywhere except on the transformer where one runs down the
Pole and into the dirt (not a rod or block). With binoculars I think I
Saw a second ground from the transformer to a support cable. This unit
Took a direct lightning strike and nothing of mine got hurt. The house
And garage are attached by a breezeway.

2)Panel pic attached (garagepanel.jpg). Yes, there appear to be extension
Cords and romex both. (3) 220V third wires on the left side (all bare) go
To the neutral bar. Some of them are long and I can't be 100% that
One of them doesn't get very close to the panel cover when on. There is
Also some wiring attached to nothing in this panel, such as for future
Connections and even a generator hook-ups that dangle bare.


3)220V is installed much as you described except for one thing.
On the right side of the panel there is one 110V and one 220V breaker
And for some reason the third bare wire of the 220V goes to the ground block
On that side instead of neutral (something I would have not thought about doing). 110V's all use breaker/neutral/ground on each side.
I do not see a bonding screw (neutral and ground appear to be isolated
From each other except for the 220V on the right side using ground instead
Of neutral... that is also where the breezeway breaker is grounded? ).

Yes, it would certainly appear we have a hot/neutral situation in the breezeway.
Could this be coming from the panel because ground has "gone neutral" on
The grounding block on the right side, or a bare neutral 220V third wire is
Touching the panel cover on the left? The switches on each end of the breezeway
Are simple 2-way switches wired normally with grounds which shock on
One end with switches off.

1. No disconnect. How did that happen? Is there a separate box below the meter? This is where the disconnect can be.

You probably won't see a rod, only a wire going into the ground. The rod and connection is buried.

3. You have some work to do:

a) move all bares to ground bars.
b) ideally, there should be no sheathed cables in the box. They should only be wires after they enter the box.
c) Check presence of stray voltages after changes made.

You did not mention, if this panel is in a detached garage?

On the garage panel pic, on the top of the left ground bar, there is a very large lug. This would typically be the place that the panel ground is connected.

When you get bored, post a picture of your main panel.

If the outside of your house near the meter schematically looks like this: http://www.ecmweb.com/mag/708ecmCBfig1.jpg then, the main box would not necessarily have a main breaker. Ground would be done in the disconnect AND the main panel would be wired like the sub-panel.

Do you have an outdoor disconnect?

As Kiss Mentioned, the Ground is missing from ground lug to Meter connection. At that point you should have a ground rod, Connecting to Meter Can Neutral, Your disconnect is your Main Breaker. Many Places require a Disconnect at Meter Can. With that you could work on your Panel with power switched off. Also like Kiss mentioned, remove sheath where it enters panel, bring all grounds together and connect to ground bar(s), neutrals all run together and to neutral bar, and all hots will go to each breaker. The Neutral on right side shoud be identified with white tape, above the lugs, I see 3 Wires? That just stop, and looks like bare copper conductors. If they are not used, they shoud be capped.
The gray romex that is connect to the bottom right breaker MAY be pulled tight against the neutral(could be the problem).

Strat:

Hey, your right, there should be a wire from the large ground lug all the way back to the main panel.

Good job catching the uncapped wires and the unmarked neutral.

The chiefengineer has to become the chief technician:

For now, move the bares but jump the neutral to ground bar until a ground is run back to the main panel. Be careful, because the grounds could spark when moving them.

The conductor size should be #1 copper with a #6 AWG ground back to the main panel for 100 A non-continuous. Meaning, breaker can be loaded at 80% forever.

Looks like you have some work to do.

Again, thanks. I am only an applied mathematician/computer guy... who can
Change starter brushes/program chips/wired whole additions/etc... so why this is such a puzzle stems from the fact
That I always had disconnects I could invoke/test/this has me afraid (none at the house panel... attached as requested... I am too chicken to remove the cover hot).
Also note the banana tree blocking the removal of the panel... not that I
Would do it without a disconnect.

Bummer. The box has almost 300 ft. of concrete all around it (no burials possible).
Why can't I just ground the garage panel in some silly fashion... like through an
Insulated wall pipe I would put in, and forget the breezeway light?

Do U mean "jumper" the neutral to ground or remove the neutrals and
Put them on the ground block?

The garage is semi-detached as in a breezeway roof only... the garage panel extends
To the house in the breezeway switch... the house's sub-panel in the attic controls the
Other part of the mud room.

One more thing... the pic is dark and there IS a disconnect in the garage panel at the top... just not the house. As you can see the config looks like your diagram:
http://www.ecmweb.com/mag/708ecmCBfig1.jpg without ground wire... unless
It is encased in plastic.

The resolution required here disallows detailed scaled analysis of the box but it is as I described. There are long bare wires which are normally intended to be used as 110V grounds that are being used as both 220V neutrals and grounds... without the neutral bar and the ground bar connected in some way why doesn't this create a hot ground somewhere?

Your main disconnect is in that panel next to the meter. This should be where the house ground is made.

Those cables that feed your garage panel have to be in conduit. Hopefully there is a ground in there UNLESS someone decided that the conduit IS GROUND which isn't right either.

Quote:

---

Do U mean "jumper" the neutral to ground or remove the neutrals and
Put them on the ground block?

---

Bares to ground bar in garage.
Neutrals to neutral bar in garage.

But, there has to be a ground from the house's reference point. ONE GROUND.

Since there isn't a ground, that we can tell, in the garage panel. A connection from the ground bus to neutral is safer than none at all. This jumper is only a TEMPORARY measure.

So, your "main panel" will then be wired like the sub-panel with separate grounds and neutrals.

Well, since your afraid to take a panel off live, it might be time to call a pro.

Your biggest problems:
a. ground and neutral isn't separate at the garage panel
b. There doesn't appear to be a ground from the main panel
c. Your afraid or can't open the panel next to the meter to turn off power.

Breezeway probably counts as non-detached, so a ground rod isn't necessary.

This may help explain your problem. It's a ground loop and it wrecks havoc.

Ground Loops (or "Let Me Hum a Few Bars…")

The bare wire or protective ground should only carry a current under a fault condition.
Neutral is a current carrying conductor tied to the 0 V reference of the house.

When ground carries a current the reference moves and outlet plate screws that are supposed to be at zero volts with respect to the earth aren't anymore and they depend on what the loads are. Furthermore CATV gets affected as well because the signal is ground referenced.

Quote:

---

Originally Posted by KeepItSimpleStupid

Your biggest problems:
a. ground and neutral isn't separate at the garage panel
b. There doesn't appear to be a ground from the main panel
c. Your afraid or can't open the panel next to the meter to turn off power.


When ground carries a current the reference moves and outlet plate screws that are supposed to be at zero volts with respect to the earth aren't anymore and they depend on what the loads are. Furthermore CATV gets affected as well because the signal is ground referenced.

---

Thanks again.

Sorry... am out of town for a few. A + b agreed; c: if I can remove the cover
To get at the screws I will have a go at it. I have no idea why the main cutoff
Should not have a switch that is accessible.

Next paragraph: If I put a multimeter on a regular 120V outlet and connect
Probes to the ground and one plug hole and get a 120V reading would that indicate your
"ground loop" is present?

Para a:

You put a banana tree in the way.

Para b:

A tester such as this one: AEMC Instruments AEMC 3711

Between ground and the narrow blade (Hot), you should have 120
Between narrow (Hot) and wide (neutral) you should have 120.

Between Neutral (wide) and ground (round), you should have a very low voltage. Since zero volts rarely exists.

Now in your case, never done it, but it should work. Plug in an extension cord somewhere in the house and go to various outlets/switchplates and measure from the ground of the extension cord and the ground of various other outlets and switchplates. There should be negligible difference.

e.g. Ground of the extension cord plugged into a garage outlet and the outside of the meter disconnect will probably be your largest voltage.

Ground currents is the only effective way, but you can't break the ground and insert a meter to measure it.

If poster has an Amprobe it may help in some instances to measure any ground current.
It looks like a ground to the Meter Can behind the Banana Leaf goes to ground there, I would inspect and verify it is intact. Be careful, if separated from ground rod, it may be live. Stay insulated/Isolated/Safe.

Quote:

---

Originally Posted by KeepItSimpleStupid

Para a:

You put a banana tree in the way.

Para b:

A tester such as this one: AEMC Instruments AEMC 3711

Between ground and the narrow blade (Hot), you should have 120
Between narrow (Hot) and wide (neutral) you should have 120.

---

Thanks, I'm back.

a) Planted one to the side and a water sucker grew 20 feet
In one year... would've thought myself stupid until I realized I had a contingency:
Screws on top of door appear to be removable. Trouble is: it has been
T-storming like crazy every day and it runs off a metal roof onto the box.

b)Tried this technique in the garage (only 3 outlets). There is ZERO voltage
Ground to ground but about 3.6 OHMS. Hot/neutral is 120V end to end and
Hot/ground is 98V. Does this indicate there is a non-visible ground present
And it is leaking somewhere? Hot/ground in the house is 120V. Panel is as
Depicted still.

Quote:

---

Originally Posted by Stratmando

If poster has an Amprobe it may help in some instances to measure any ground current.
It looks like a ground to the Meter Can behind the Banana Leaf goes to ground there, I would inspect and verify it is intact. Be carefull, if seperated from ground rod, it may be live. Stay insulated/Isolated/Safe.

---

First, thanks for the help. There are a lot of flavors of Amprobes when I googled them(?).
Yes, there may be a ground inside that tube somewhere, but there is no rod in the ground,
Just a solid entrance into the earth. It would seem the house is grounded but whether
That extends to the garage is questioanable.

98 volts indicates the ground is missing in the garage.

Ground to ground means Ground (panel) to ground (outlet)

3.6 seems a little high, but I don't know what you used for your connections. There may also be an ohm in the leads of the meter. Measuring resistances this low generally requires a 4-terminal meter.

We also know the neutral and grounds hould not be mixed in the garage panel.