Ok troubleshooting wizards

[hardworkingstiff]

Still think dropping a rod near the ramp will minimize or eliminate the problem. And code requires it as it is a seperate structure with more than one circuit.

I think dropping a rod near the ramp will minimize or eliminate the symptom, although I'm not happy with the new situation of current from the service neutral getting back to the substation through the ground rod I will drive near the ramp.

Can you go into a little more detail as to why the code requires this rod?

 

[hardworkingstiff]

From one of George's links.

Stray voltage. Good grounds can improve stray voltage but also can make it worse. Also, better grounding to resolve stray voltage issues may not be an effective solution.​

 

[hardworkingstiff]

I would think that the opened neutral on the side of the source of the stray voltage would produce a voltage reading to earth, while the other neutral would have no voltage-to-earth reading. This would, of course, require a high-impedance meter, and not a solenoid tester.

When the voltage went away at the ramp after the service neutral was lifted, I would think this would indicate the source of the stray current is the service neutral, no?

 

[77401]

I think dropping a rod near the ramp will minimize or eliminate the symptom, although I'm not happy with the new situation of current from the service neutral getting back to the substation through the ground rod I will drive near the ramp.

Now your guessing on what will fix the problem.
Trust me! the problem is in the neutral
A ground rod is just wasting your customers money & your time & won't solve the problem.
How do I know? Cuz when I was inexperienced I ran across the same situation, did the ground rod, got called back a week later, problem still there, then found the problem with a POCO neutral. All fixed! This forum can be a great resource for learning but you have to use our knowledge.

 

[hardworkingstiff]

77401,

Are you actually reading what I'm writing? Your last response is written like you think I believe the ground rod is the solution. If I am correct, then either I am not doing a good job of writing or you are not doing a good job of reading.

 

[e57]

Can you go into a little more detail as to why the code requires this rod?

For this specific reason: The earth has different impedance values as it is refferenced from place to place. Mainly due to soil make up, and a good many other reasons. For instance the earth right under your panel is brought to equal potential of the neutral by your grounding electrode that would be instslled there local to it. Thus bringing your grounded, and grounding conductors closer to the potential of the electrode installed at the transformer feeding it, that may be some distance away. Say you are running a bunch of circuits much further away, which you are. The earth, in this case water, will be at a slightly different potential from the earth near your main panel, and the transformer that are further away. This different potential can be measured as voltage between the earth, and any metalic object (like the ramp you bonded) that is connected to your electrode system that is far away from that point. Depending on soil conditions that voltage can vary widely. Thats the reason code requires an electrode at seperate structures fed from the same service. To bring the Earth near that panel, and the Grounding and possiblely the Grounded conductors as well closer to each other in respect to a voltage difference. And reduce this type of shock hazard. It is NOT "stray voltage", you brought it there! On your EGC! Which is why lifting the neutral at the meter made it go away. The voltage is on you EGC, not the other way around.

You should not be worried about current on the GEC that gets installed as you would as seperate structure on the panel closest to this ramp. It will be closer to the same voltage as the earth around it then. Technically any other remote free-standing panel between the ramp and the main is a seperate structure as well, and would be treated the same way. Even if not on a building... You may be driving a few rods????? After this the earth near any one of them will be as close as possible to the potential of the grounded conductive surfaces at each. Hope that helps.

 

[e57]

Now your guessing on what will fix the problem.
Trust me! the problem is in the neutral
A ground rod is just wasting your customers money & your time & won't solve the problem.

You may have had a different type of problem, "stray voltage" is a broad term.

 

[LarryFine]

It is NOT "stray voltage", you brought it there! On your EGC! Which is why lifting the neutral at the meter made it go away. The voltage is on you EGC, not the other way around.

This is why I suggested reading voltage to earth from both sides of the temporarily-opened neutral connection; to see whether the potential is between the source's neutral and ground, or betaeen the load's neutral and ground.

This is the kind of information I like to plug into my brain when trouble-shooting. Even if I can't explain why I want to know, I still want to, similarly to always checking voltage from line to neutral as well as line to line when checking 2-pole breakers.

 

[don_resqcapt19]

After this the earth near any one of them will be as close as possible to the potential of the grounded conductive surfaces at each. Hope that helps.

The key word here is near. If you are 3' away from the rod, you will have ~57% of the voltage that you had without the rod.
Don

 

[e57]

On second thought - Drop two rods at the ramp.... And as close to the water as possible, maybe one on each side.

Don wouldn't that 57% @ 3' depend on the soil make up? And its water content? Is this one of those 25 ohms or less items?

 

[ramsy]

The earth, in this case water, will be at a slightly different potential from the earth near your main panel.. This different potential can be measured.. Thats the reason code requires an electrode at seperate structures fed from the same service.

I thought ground rods were required primarily to direct lightning bolts to earth, at the seperate structure, rather that toasting the feeders back to the service electrode?

(electrodes are) ..To bring the Earth near that panel, and the (EGC)'s as well closer to each other in respect to a voltage difference. And reduce this type of shock hazard.

If different earth potentials are measured between two isolated ground rods why can't that voltage provide 1.5vdc of free energy to run at least a digital wrist watch? If earth potential, relative to itself, is the same everwhere that may explain why it can't provide free energy.

It is NOT "stray voltage", you brought it there! On your EGC! Which is why lifting the neutral at the meter made it go away. The voltage is on you EGC, not the other way around.

If the different earth potentials you refer used an external power source, such as the "fall-of-potential" method or other injection test system for earth electrodes, that may also explain why lifting the neutral (aka injection source) removed the 1.9vac.

Take away the injection source and electrode rods show nothing between each other. Bonding the injection source to all electrodes also shows nothing between electrode rods, but DVM's typically get 0-4vac common mode voltage (CMV) between branch EGC's and energized neutrals.

If the EGC is relatively dead, CMV values increase directly proportional to neutral loading. Therefore, either the EGC/aluminum dock/water is not dead, or since the open disconnect ruled out neutrals energized by a phase, that main bonding point (service neutral) may provide an injection (potential difference) from somewhere else.

I don't see an isolated earth potential between the water and a clean service neutral (free energy). So, if the PoCo cable test comes back OK -- no leakage (injection) -- and the problem will be the customer's, the only remaining idea for me is an offending vessel leaking inverted AC directly to the aluminum dock.

 

[hardworkingstiff]

I don't see an isolated earth potential between the water and a clean service neutral (free energy). So, if the PoCo cable test comes back OK -- no leakage (injection) -- and the problem will be the customer's, the only remaining idea for me is an offending vessel leaking inverted AC directly to the aluminum dock.

I guess I'm not communicating very well. The dock is a wooden dock. Only the ramp is aluminum. The voltage is betweent the ramp and the ICW (Intra Coastal Waterway).

There are no receptacles for shore power on the dock. Only some GFCI receptacles and 23-watt flourescent lights both of which are in UL listed Marina fixtures (made with fiberglass).

With the service main off, the voltage remains the same as when the main is on. When the PoCo feeder is lifted from the meterbase, the voltage is gone. I agree with the post that I'm bringing the voltage to the ramp with the ground.

If the power company determines there is nothing wrong with their feeder, I see no option other than driving ground rods behind the bulkhead and bringing a conductor from the ground rod to the neutral bar in the 1st means of disconnect.

 

[hardworkingstiff]

If the EGC is relatively dead, CMV values increase directly proportional to neutral loading. Therefore, either the EGC/aluminum dock/water is not dead, or since the open disconnect ruled out neutrals energized by a phase, that main bonding point (service neutral) may provide an injection (potential difference) from somewhere else.

Could it be as simple as the waterway is a good enough path for current back to the substation that we will always have current flowing from the PoCo grounding system through the waterway?

 

[George Stolz]

Could it be as simple as the waterway is a good enough path for current back to the substation that we will always have current flowing from the PoCo grounding system through the waterway?

I think you're exactly right. Does this picture look like the setup?

Your big chunk of aluminum in the water makes a low resistance path for neutral current from the utility neutral to the earth. Since that resistance is so low, the utility's neutral is getting less appealing all the time. So, a proportional amount of neutral current (120V from the neighbors, or 7200 from the primary, I have no idea) is getting back to it's source via your "electrode."

I'd recommend buying that "blocker" contraption that was mentioned before, this is exactly what the thing was designed for, as far as I can tell.

 

[e57]

Oooooo, thats what we need, a picture says a thousand words....

That blocker thing seemed to be installed at the transformer by POCO, something about seperating pirmary and secondary neutrals???? I'm not sure how that thing works, never even heard of one before.

Could it be as simple as the waterway is a good enough path for current back to the substation that we will always have current flowing from the PoCo grounding system through the waterway?

I aggree, this could very well be the case.

If the power company determines there is nothing wrong with their feeder, I see no option other than driving ground rods behind the bulkhead and bringing a conductor from the ground rod to the neutral bar in the 1st means of disconnect.

You may or may not be able to go to the neutral bar, due to the possibility of other parralel metalic paths to the structure you have two choices per 250.32. You may have to go to the ground bar, depending on what, if any other metalic paths exist.

 

[don_resqcapt19]

e57,

Don wouldn't that 57% @ 3' depend on the soil make up? And its water content?

Actually it doesn't, assuming that the soil in the area of the rod is somewhat homogeneous. You will have the same percentage of the total resistance of the earth around the ground rod at 3', no matter what the soil resistivity is. If you have the same percentage of the total resistance, the you will also have the same percentage of voltage drop.

 

[winnie]

Your big chunk of aluminum in the water makes a low resistance path for neutral current from the utility neutral to the earth. Since that resistance is so low, the utility's neutral is getting less appealing all the time. So, a proportional amount of neutral current (120V from the neighbors, or 7200 from the primary, I have no idea) is getting back to it's source via your "electrode."

Hmm. I would think that if the big chunk of aluminium were making a low resistance path to the water and earth, then there would have to be very good contact between the aluminium and the water, and thus very little voltage difference between the two. My _guess_ is that the aluminium is _insulated_ from the water, say by plastic floats, and that someone touching both the ramp and the water effectively completes the circuit. In this case, the solution would be to somehow _bond_ the ramp to the _local_ water.

An additional question: did you ever attempt to measure _DC_ voltage between the dock and the water? Is it possible that this voltage is the result of galvanic action between the various metals and electrolytes in the soil-water-ramp-egc-grounding electrode circuit? Did you attempt to measure the _current_ that would flow from the dock to a 'grounding electrode' suspended in the water?

-Jon

 

[George Stolz]

Jon, from the original post:

Ramp is aluminum. I ran 4 #6?s (cu.) in PVC and non-metallic flex from the existing panel to the dock and loop fed the power centers with ? of the pedestals on one circuit and the other ? on the other (used 30-amp breakers). I also bonded the aluminum ramp to the ground wire I ran for the power centers.

Here is what I found. With one test lead in the water and one on the ramp, I found 1.9 volts AC. Switched to DC and found less than .01-volts (discounted it as meter sensitivity). I turned off the 200-amp MB in the panel and found the 1.9-volts still present.

Mark, the only option I can see (if I'm now seeing the problem correctly) is either isolating the ramp from the Grounding Electrode System, or by installing the blocker (which I assume forces the service neutral conductor to be a one-way street, allowing neutral current out but not allowing it in).

I'm wondering if the ramp is required to be bonded at all. The provisions of 250.6(B) might be applicable to this problem. I looked briefly in Article 555, but didn't see anything definitive about this ramp needing to be bonded. (I don't remember the particulars of the electrical installation, though, or even if I have the picture right. It's just what I envision upon re-reading the thread.)

 

[LarryFine]

It was mentioned that the neutral lift proved that the POCO was the source of the offending voltage. Isn't it also possible that the water is somehow the source? In other words, is the voltage being measured occurring between the local earth and the source neutral, or is it occurring between the water and the source neutral?

It seems possible that another source is somehow energizing the water, and the POCO neutral is actually representing a genuine 0-volts-to-ground reference. Not likely, perhaps, but possible? That's what I was thinking when I suggested measuring the voltage-to-local-earth from both sides of the lifted neutral connection.

 

[George Stolz]

Larry, that could be. However, the 1.9V shouldn't have disappeared when lifting the utility neutral if that were the case.

Minus the "genuine utility neutral", the local grounding electrode system would (likely) have still been connected to the EGC's running out to the panel, which in turn were still bonded to the aluminum ramp. So, if the 1.9V were coming from the water, then breaking the utility neutral should not have zeroed the meter.

(If I'm picturing this right. Now that I have at least half a clue what's going on, I'm interested for Lou's corrections on my assumptions. )