Ground Loops

[JohnRodriguez]

The utility's company concern stops at the meter socket on you pole, they could care less about the problems you are having after the socket, as long as you don't knock customers off the grid. That's why they tell you to get a contractor.

Eric is correct on the static line on power poles, it is the common point for lightning to get to ground, it also is your ground reference, since it's tied to the ground grid at the sub stations.

3 phase Tranformers are not all the same, a delta's will kill you deader that hell if you have on phase grounded (which is not easy to tell find the source of if it is grounded) and happen to have a second leg go to ground, and you are in the wrong place at the wrong time. Thats why you have 3 phase indicating lights on a bank of delta transformers, if you have 2 burning brightly and 1 out, you better get out and investigate if you don't have a burned bulb or have something to ground. If a wye goes to ground then it will trip out instantly or blow the fuse, because it all ready has all three phases to ground.

This is the problem that I see that GG has:

1. triplex is run to the dock (2 hots, 1 neutral) run from panel at the house.

2. the neutral was tied to the frame (correct GG???) which esentially made the neutral the ground.

My solution:

1. run a quadraplex wire ( 2 hots, 1 insulated neutral, 1 bare ground "static line")

and bond the metal dock to the ground wire.

2. you need to put a ground rod by the bank. if you can't get a rod to go straight down then bend the rod at the end and dig a shallow trench about 8 inches deap and lay it in the ground, if you ask a inspector nicely, he'll tell you this is perfecty ok, all 8 feet are in the ground.

[George]

There is some good info here:

http://www.iaei.org/magazine/05_d/hamilton.htm

Interesting info concerning neutral currents showing up on ground paths and a utility based solution is in this paragraph at the end of the section on grounding:

"An additional action not required by the Code can also be taken to improve safety for workers and the general public who may be in the water. Owing to the absence of any equipotential bonding grid in these facilities, designers and installers should consider requesting the electric utility to install neutral separation (also called "neutral blocking") equipment at the utility transformer, acceptable under Section 097D2 of the National Electrical Safety Code ANSI C2, in order to mitigate propagation of stray ground currents into the water over the green wire via submerged, grounded metallic equipment enclosures and cases. Use of neutral separation equipment has been employed to this end successfully at pool facilities served by utility primary distribution systems operating at voltages up to and including 34.5 kV."

--

Regards,

[GuildSF4]

GG I understand your problem, I don't have a solution for sure. You can try unbonding the nuetral and ground (WIth power disconnected) and check for potential between the water and Dock that would tell you if George's idea is part or all of the issue. (If there is still a potential there I would definitely find a way to ground the Dock to the water or shore.)

I love your diagram with the person and the "capacitor" this capacitive effect is exactly how people have died on US Navy vessels that use an ungrounded three phase system.

Ground nets/grids, we use those on RF equipment since RF doesn't ground very well just pack to the substation or distrubution point (not dangerous currents, but a lot of noise which causes other equipment malfunctions.) And the lightening supression system is totally isolated from this grid. With computer controlled systems and RF systems gound loops suck.

[Gun Geek]

I love it!

Keep it coming guys...

This is the problem that I see that GG has:

1. triplex is run to the dock (2 hots, 1 neutral) run from panel at the house.

2. the neutral was tied to the frame (correct GG???) which esentially made the neutral  the ground.

<{POST_SNAPBACK}>

1) Yes - triplex terminates in a load center on the dock. The neutral goes to an unbonded (insulated) bus bar in the load center and all white wires gonnect to this bus. A ground was subsequently run down the hill and connected to a bonded (connected to the box which is connected to the frame) bar in the box - see below.

2) Yes

My solution:

1. run a quadraplex wire ( 2 hots, 1 insulated neutral, 1 bare ground "static line")

and bond the metal dock to the ground wire.

2. you need to put a ground rod by the bank.  if you can't get a rod to go straight down then bend the rod at the end and dig a shallow trench about 8 inches deap and lay it in the ground, if you ask a inspector nicely, he'll tell you this is perfecty ok, all 8 feet are in the ground.

<{POST_SNAPBACK}>

1) Sorta did this - in hope of a fix I ran an insulated ground wire from the ground rod up the hill. Disconnected the N from the frame, connected the G - still get the shock

2) Cain't do it. The bank is rock (limestone). Any soil there (and there is very little) will be only few inches thick and likely to wash away the next time the water comes up. The only soil that might be more than 3" deep is at the bottom the lake, 40' to 50' below the dock.

[Gun Geek]

GG I understand your problem, I don't have a solution for sure. You can try unbonding the nuetral and ground (WIth power disconnected) and check for potential between the water and Dock that would tell you if George's idea is part or all of the issue. (If there is still a potential there I would definitely find a way to ground the Dock to the water or shore.) 

<{POST_SNAPBACK}>

I have disconnected the H (via breakers) and the N (took it apart) and left the ground connected. Shock is still there. I have not disconnected the U Co N from the ground rod to see if the shock still exists.

If it does not (ie. the dock ground wire does not touch the U Co N anywhere, but is still connected to the ground rod), the U Co might have ground loop problems of their own. This would dump a bunch of current into my ground point which would cause a voltage to develop.

I'll investigate this (though I'm not sure how to disconnect stuff - it will take the power down in the house when I do).

[MikeFoley]

How are you testing the shock? Least liked family member, unaccounted for illegal alien, fancy device, dead fish?

[JohnRodriguez]

ok, run a uninsulated ground along the walkway, underneath, and coil the rest of the wire under the ramp where you get on the walkway, maybe with enough coils touching the ground you'll get the results you need.

[GuildSF4]

I have disconnected the H (via breakers) and the N (took it apart) and left the ground connected.  Shock is still there.  I have not disconnected the U Co N from the ground rod to see if the shock still exists.

If it does not (ie. the dock ground wire does not touch the U Co N anywhere, but is still connected to the ground rod), the U Co might have ground loop problems of their own.  This would dump a bunch of current into my ground point which would cause a voltage to develop.

I'll investigate this (though I'm not sure how to disconnect stuff - it will take the power down in the house when I do).

<{POST_SNAPBACK}>

You may not have to disconnect it, if you shut off power to the house (breakers open) check with a clamp on amp meter the nuetral and ground with the dock connected and dis-connected. This may tell you how much current the UC is sending to your ground, or the current thru the ground loop to the dock.

[h4444]

Got any transmission lines running overhead????

Static factoring in anywhere???

Dad works for a utility company, they handle 69Kv up to 765Kv lines and I know from my 1 summer working there as an intern,,,,,it ain't fun picking up a static jolt from a transmission line when something metal has been under a line (truck) a while and a smart-ass lineman just stands back,,,,waits,,,,and chuckles while you grab the handle of a truck under a 765Kv line.

I know they've had situations where they've had guys get zapped by static when a bond/ground was removed at the wrong time,,,,,etc.

Other than that, I don't know squat about electricity,,,,,but I know static from transmission lines can give you a little jolt.

H4444

[Gun Geek]

How are you testing the shock?  Least liked family member, unaccounted for illegal alien, fancy device, dead fish?

<{POST_SNAPBACK}>

I invite my shooting buddies.

Strange - my number of friends seems to be dwindling. You think Clyde would be interested?

uhhh- that's me who gets shocked. Wouldn't ask anyone to do anything I wouldn't do

[MikeFoley]

Don't shock my buddy before the Nationals. I'll loan Keith to you though.

[George]

Found some interesting information about submerged metal objects and bonded G/N AC systems at this website:

http://www.boatus.com/boattech/casey/36.htm

The pertinent part:

In your breaker box at home the neutral wires and grounding wires all connect to the same terminal strip (or bus bar), but on a boat the AC grounding wire is connected to the DC ground. Also connecting the neutral wire to it makes underwater hardware a current-carrying path to ground, potentially lethal for anyone in the water nearby. On a boat the neutral (white) conductor and the grounding (green) conductor MUST NEVER BE DIRECTLY CONNECTED.

Hmmmmmm!

BTW, If anyone is interested, I have archived some of the pertinent items from this thread along with more info on AC power systems here:

http://www.glinder.com/files/PowerSystemsTech/

And here:

http://www.conceptusedgear.com/techlibrary/PowerTechnology/

--

Regards,

[Gun Geek]

Talked to an engineer at a local commercial contractor and got an earful about how much of a bitch ground loops are. He confirmed that it has been more of a problem in the dry weather (we did get some rain from the hurricane). Told me the story of this being a big problem on horse farms - horse standing on ground that is higher potential than the water - gets shocked and won't drink. We have big $$ horse farms around here.

Here's the interesting part. He said if it truly is a ground loop, then there is likely nothing to be done that isn't major $$$ (10's of thousands). AND he said that it is common around swimming pools that there is some potential between the ladders and the water, and that code allows 4 - 6 v! I'm seeing 1.5v. He rattled off the measured voltages at the public pools around town. He knows it because he checks it. His point was, if it was shock from a ground loop, don't worry about it! He said to make sure the dock was properly grounded so that a stray hot would not light up the dock, and move on. EEEEK!

OK, here's the plan as of now (I tell you this because it will take some time and I don't want you to think I let this go to sleep):

1) Disconnect the service N from the ground rod (If I can do it safely) and see if the shock is still there. If it is, then I am satisfied that this is truly a ground loop problem and not a leakage (N to G) problem.

2) Finish cleaning up some details disconnecting the N to the frame of the dock.

3) I don't like the Triplex going to the dock with a live exposed N, so I'm going to look for a different kind of wire (got some in mind that is 4 insutlated conductors + bare ground in insulation rated for wet locations. It feels like a giant extension cord and I think it is big $$$).

OR

4) I may use 2 runs (gives me 4 insulated conductors) of Triplex and just have the metal wire for mechanical connection. Probably will ground it to make sure there's no stray voltages.

5) Drive a couple of ground rods up the hill to improve grounding. It is code here (as of just a little while ago) that you have 2 ground rods (or 1 big honker).

This will take a little time, so I'll repost when I get my list done.

Keep the ideas coming if you have more!

[EricW]

My contact at the local electrical utility gave the same answer that John did: bury the ground in a trench. No clue what you do sans trench.

[George]

You are going to want to use UF type quadplex, or triplex with ground even if you don't bury it. It is water resistant and available in up to 6/3 with a 4th exposed ground wire. I found a farm supply place selling it for $2.67/foot.

It would be better to bury over the long term as UV exposure is a big issue with PVC and other plastic jacket cables.

http://www.do-it-yourself-pumps.com/uf-dir...urial-cable.htm

--

Regards,

[cking]

Since all discussion, I say you have a ground loop problem because of the soil and limestone. So no matter what you do the grounded dock has got some potential. Why not try taking a piece of bare copper ground wire attach to dock and toss in the water. Since copper corrision is conductive it shouldn't lose it usefullness over time.

Out in country where we don't have three phase. Just the hot line and the ground line. Every pole is tied to ground line and every step down transformer is tied to the ground line, and my house neutral is tied to the ground line.

[Gun Geek]

Since all discussion, I say you have a ground loop problem because of the soil and limestone. So no matter what you do the grounded dock has got some potential. Why not try taking a piece of bare copper ground wire attach to dock and toss in the water. Since copper corrision is conductive it shouldn't lose it usefullness over time.

Out in country where we don't have three phase. Just the hot line and the ground line. Every pole is tied to ground line and every step down transformer is tied to the ground line, and my house neutral is tied to the ground line.

<{POST_SNAPBACK}>

Not enough surface area on the wire (unless it was 100' long, in which case it would foul on the bottom and rip off as the dock went up and down).

This is why EricW and I discussed metal plates suspended under the dock. It may be in the plan as well.

[cking]

That sounds good, how about copper flashing, lots of surface area, easy to come buy. Solder on a some lead for weight and let hang in the water.

[George]

Thought a review of some "facts' might help put this in proper safety perspective.

Some things we know as a fact

- Neutrals should NEVER go into water

- The dock should be grounded (if at all possible)

- Your service ground is mechanically bonded to the neutral at at least two points

- Your service is single phase so the neutral should always be at, or very close to ground potential without grounding it (typically .5-1 volt separation if all is right).

Something we should conclude from these facts

- You should NOT ground the dock to your service ground as long as it is strapped to the utility Neutral

IMO, an ISO transformer regenerating the Neutral at the dock (letting it float) and using a local staked ground (go up the bank a little, but not to the house point) NOT bonded to the house service ground point may be the only way to safely earth the dock without sticking a live neutral in the water like you are doing now if you ground the dock to your service ground.

As I mentioned in an earlier post, the utility can be asked to decouple your neutral as an alternative way to eliminate the G/N bond so the dock can be part of the true service ground withoiut soaking a neutral ;-)

--

Regards,

[Gun Geek]

That sounds good, how about copper flashing, lots of surface area, easy to come buy. Solder on a some lead for weight and let hang in the water.

<{POST_SNAPBACK}>

Good idea - see posts 16 & 17 - Eric and I discussed it and over engineered it! I like the copper flashing...