I think we have an issue with the neutral coming to our house (North America, split-phase power 240VAC total, two legs of 120V to neutral with opposite polarity.) Most of the time, everything is fine, but sometimes, our lights will pulse brighter, dimmer, brighter, dimmer with 1Hz frequency. At first I had no idea what the source was, but I later realized it coincides perfectly with PID heater cycles on my home espresso machine, which pulls about 5A.

The really interesting thing is that lights in the dining room get brighter at the same time that lights in the bathroom get dimmer, and vice versa. This got me thinking it was the neutral moving, as opposed to either line sagging. So, I've waited patiently for the rare moments when the problem appears and taken voltage measurements everywhere imaginable. It's not isolated to any portion of the house. All of the power in the house is doing this, all the way back to entrance lugs in the breaker panel. The line voltage from L1 to L2 is remarkably stable, maybe 0.1-0.2V fluctuations at any given time. But the line to neutral voltage during these events goes up and down by as much as 4V.

So for example, if I've got 244V total coming in, usually each line reads 122 to Neutral. During one of these events, L1 will fluctuate between about 122 to 126, while at the same time L2 is fluctuating between 118 to 122. The L1-L2 voltage always stays very stable, but the neutral moves around between them.

A variation of 4V due to a 5A load implies 0.8 ohm resistance on the neutral for the whole building, which seems abnormally high. What's far more upsetting is the intermittent nature of it. Sometimes the neutral is solid, and other times it dances around, anywhere from 1-4V in response to this 5A load (the 5A PID loop is ALWAYS cycling, on 24 hours/day, and usually its effect on mains voltages is only measurable at the outlet it's connected to - it's only during these rare "events" that the whole house has its neutral move in response to the 5A load.)

I've read horror stories all over the internet of floating neutrals and the terrible things that can happen to your appliances (or people) as a result of unpredictable household voltages when the neutral is floating... and I've also heard of "High Earthing Resistance of Neutral at Distribution Transformer" (item 4 about halfway through this article: https://electricalnotes.wordpress.com/2012/07/28/impact-of-floating-neutral-in-power-distribution/) which seems like a better fit for what I'm measuring. If it were actually floating, I'd expect voltage readings to be all over the place. As it is, it never moves too terribly far.

So, any advice on what I do next? The problem exists all the way back to the breaker box. I'm not comfortable trying to open up the meter, so I can't inspect connections from the panel to the meter, nor from the meter up to the drop. I've taken a few pics of the junctions at the mast where it drops down to the meter, and I can't see anything wrong - the wires look old and manky, but presumably bare wire always looks funky, and it's only the junctions I should be worried about? I also looked up at the power pole and the connections there - it's a rat's nest of junctions to get from the "real" neutral to the one feeding my house, so I imagine any of several joints up there could be the issue. If I can't prove the issue is outside of the house, do I need to get an electrician out first to inspect the few things I can't (won't/shouldn't) get into, or should I go straight to contacting the power company and hoping they can find an issue at the pole or mast?

I've attached a few pics which may or may not be helpful... In one version of the pole transformer pic I tried tracing each wire and color coding them to make it easier for me to follow.

 

I agree with your conclusion. It would be interesting knock a ground spike into the ground and monitor the voltage between that and neutral. As the path from the point where the centre tap of the supply transformer to the ground spike should have no significant current flowing through it should give a more accurate indication of the balance between the two legs of the supply at the transformer. The fault must be with the power company but convincing them of an intermittent fault could be difficult.

Les.

 

The symptoms you are describing do point to a neutral problem. While your images are sharp and clear we can't see below the tape on the utility pole end. Over years corrosion can manifest itself on the neutral line and over time start to deteriorate your neutral. Utility company neutral and your earth ground should be bonded at your circuit breaker panel.

At this point I would call your utility provider and describe the problem and request a service call. They can inspect your line to your residence for wear and any corrosion. You have pretty much exhausted what you can do. You may also want to give this a read. If this is a neutral issue from the utility company it is not going to cure itself and it will only grow worse. There should be no charge for the service call so you really have nothing to lose.

Just My Take....
Ron

 

The symptoms you are describing do point to a neutral problem. While your images are sharp and clear we can't see below the tape on the utility pole end. Over years corrosion can manifest itself on the neutral line and over time start to deteriorate your neutral. Utility company neutral and your earth ground should be bonded at your circuit breaker panel.

At this point I would call your utility provider and describe the problem and request a service call. They can inspect your line to your residence for wear and any corrosion. You have pretty much exhausted what you can do. You may also want to give this a read. If this is a neutral issue from the utility company it is not going to cure itself and it will only grow worse. There should be no charge for the service call so you really have nothing to lose.

Just My Take....
Ron

Thanks!

Yes, I think I will contact the utility and see what they can do. That's where I was leaning already, and everything you've said reinforces that idea.

 

I agree with your conclusion. It would be interesting knock a ground spike into the ground and monitor the voltage between that and neutral. As the path from the point where the centre tap of the supply transformer to the ground spike should have no significant current flowing through it should give a more accurate indication of the balance between the two legs of the supply at the transformer. The fault must be with the power company but convincing them of an intermittent fault could be difficult.

Les.

Interesting thought with the ground spike. The house already has a grounding rod, and it is tied to the shared ground/neutral bus inside the breaker panel. Honestly, I've wondered if that ground rod might be carrying the load when intermittent neutral failures occur. I'm not sure how low the resistance of a ground rod setup can go, but I thought it might be serving as a backup of sorts in this case.

I don't have an extra ground rod to experiment with, nor many good places to drive one... but I'm going to take a closer look at the plumbing and see if the water line entering the house might be a useful point for measuring the difference between neutral and outside ground potential. The existing grounding rod doesn't look like it's in the best condition, and I wouldn't be surprised if a copper pipe running underground provided a better ground reference.

Either way, I'll contact the utility and have them check things out, but in the meantime I might try measuring voltages from plumbing pipes to neutral to see if I can find anything interesting or illuminating.

 

Hi ebeowulf17,
I was not shure how the grounding was done on the US power distribution system. I was assuming that the centere tap of the transformer was close to the transformer and that you did not have a local ground spike connected to the ground/neutral where it entered your house. That invalidates my suggestion.

Les..

 

I have nothing significant to add to this conversation except that I'm wondering if someone, some "Copper Thief" may have clipped some copper wire from the pole to ground. If you find a solid copper wire that appears to just dangle at some high point where someone might be able to quickly reach and clip, you may have found the problem.

Locally, copper thieves have caused a lot of problems. It's common to see power poles with steel wires replacing the copper that has been scavenged. One street in particular comes to mind where it's easy access to kype the copper.

In two of your pole images I see what appears to be a single solid wire, but it exits the frame before there's any evidence of piracy. Give that a closer look. If it's been removed then that may be an indication of where the problem lies.

I'm very fortunate to have a transformer on the pole in my back yard. Yeay! Love the look (not). There are at least three paths to ground. A copper wire from the transformer to the base of the pole and two guy wires that are also grounded to what is probably 10 foot rods driven to near ground level in my yard. Yet another thing to contend with when landscaping. Lucky me. Then there's the ground at the house. The mast is grounded too.

My garage has 240 service. Someone before me divided the 240 into two legs of 120 by adding a grounding rod and splitting the service. They used an aluminum rod. I have no idea what they were thinking. I accidentally kicked the rod and it broke off. It was that corroded at ground level. I promptly shut off the breaker, then went out and got a copper plated steel rod and drove it into the ground at the same location the aluminum rod was. I know that's not to code, but it works. Good grounding is important. If your neutral line, which is in fact grounded to earth, is weak - I can see getting the funny readings. My other thought was the transformer may be failing. Doesn't sound like that's the case though.

Like others have said, call the power company and report a problem. Especially if you can see wires that seem to go nowhere.

 

Well, after long delays on my part, I finally got someone from our local power company out to check things out. They said there were two tests they can perform, both of which involve pulling the meter. One is simple voltage measurements, and the other involves some sort of dummy load. They didn't have the gear for the second test, but they say everything looked good on the voltage test, and that they'll be back soon for the second one.

Overall, the conversation with the electrician who stopped by was pretty good, but there was one strange thing. He said that my measurements at the main breaker panel don't prove anything about where the problem is. He said that a neutral fault *anywhere* in the building could cause the floating neutral behavior I'm seeing. That doesn't seem right to me at all.

TLDR: Is there any way for a fault in the branch circuits of my house to cause the neutral at the breaker panel to move around by 3-4 volts?

Assuming there's a proper, low-impedance path from the utility neutral/ground to my house, I would expect all voltage readings at the entrance to my breaker box (L1-Neutral, L2-Neutral, L1-L2) to be stable and that I would only find weird voltage readings elsewhere in the house, on the other side of whatever neutral faults might exist there.

As one step in a sanity-check, I made a simulation with the mains supply, a load simulating the espresso machine heating element which reveals these faults, and two locations for neutral impedance. The two text boxes at bottom middle and bottom right show the test results in the two different scenarios. The graphic plots are from the second scenario, which I believe is the situation at my house:

• If I run the sim with very low impedance from the mains supply to the breaker panel, and higher impedance on a branch circuit after the breaker panel, I see stable, balanced voltages at the breaker panel, and I see errant voltage riding on the neutral past the fault in the branch circuit.
• If I run the sum with higher impedance between the transformer and the breaker panel, and low impedance on branch circuits, that's when I see the neutral voltage at the panel start moving around.

Am I somehow misunderstanding this? Is there a way for internal house wiring (downstream from the breaker) to cause inappropriate neutral voltage readings at the breaker? If so, can anyone here explain how this would happen?

 

How is your neutral re-referenced to earth ground at your panel?
Here where there is metallic water pipe, this is used, otherwise a pair of ground rods are typically driven.
Max.

 

Electrical wires strung up between poles and copper thieves stealing them?
In a civilized country?
I am glad the wires are underground in my neighbourhood.

 

Electrical wires strung up between poles and copper thieves stealing them?

Around my neighborhood everything up to my service entry is aluminum. Pole to pole and pole to service entry is all aluminum.

The power company is only responsible for how the power gets to my home. The meter is theirs but the mast head or weather head above the meter is my problem and everything below the meter is my problem. I can see where they would do a few voltage checks and maybe a few load test at your point of entry and that would pretty much cover their responsibility.

You really have some strange behavior and since the L1 to L2 voltage always remains stable that would tell me that yes, there is a neutral issue somewhere. Be it a loose bonding point or something along those lines I see it as neutral related. I can't explain why the simulation behaves like it does either.

Ron

 

How is your neutral re-referenced to earth ground at your panel?
Here where there is metallic water pipe, this is used, otherwise a pair of ground rods are typically driven.
Max.

There's a single ground rod, although it seems unusually close to the foundation. It's also under a covered deck, and I've wondered if that might keep the ground dry and reduce the effectiveness of the ground rod.

I freely admit I'm not well educated on best ground rod practices and the thoughts above are pure speculation.

 

Do you have a good earth ground on your panel? Has it been megged to test the grounding? You need a licensed electrician to open the panel and check it. The power company stops at the meter. Anything after their meter is your problem and not theirs is their attitude. IE "It's not my job!" Even with good wiring, without a proper ground, you have problems. Bad earth resistivity, badly corroded ground rods, bad ground wire to rod clamps, bad ground wire to panel connection, lots of ways to have ground problems which appear as a neutral problem.

 

Around my neighborhood everything up to my service entry is aluminum. Pole to pole and pole to service entry is all aluminum.

The power company is only responsible for how the power gets to my home. The meter is theirs but the mast head or weather head above the meter is my problem and everything below the meter is my problem. I can see where they would do a few voltage checks and maybe a few load test at your point of entry and that would pretty much cover their responsibility.

You really have some strange behavior and since the L1 to L2 voltage always remains stable that would tell me that yes, there is a neutral issue somewhere. Be it a loose bonding point or something along those lines I see it as neutral related. I can't explain why the simulation behaves like it does either.

Ron

FWIW, although I believe my tests and measurements indicate that there's no problem in the branch circuits, I freely admit that the problem could be in the mast or in the cable from the meter to the breaker panel entrance.

So, if the utility is convinced after their second test that everything is solid up to the meter, getting an electrician to check the meter to panel wiring seems like a logical next step.

As for the simulation behavior, what part seems inexplicable? The simulation behavior seems correct to me - the only thing confusing me right now is the electrician's claim that branch circuit wiring problems could cause bad neutral voltages at the breaker panel. If I've done something wrong in the simulation, or if I'm misinterpreting its results, please let me know.

You're one of the many people here who I trust and respect, and although I think I understand this scenario, I know it's possible that I'm misunderstanding things.

I never like to assume an electrician is wrong or that I know their trade better than them, so when the electrician told me things that didn't sound right, I wanted to get second opinions.

 

Here in Georgia USA, we build homes (and commercial) based on specifications of the "Southern Building Code" which in turn specifies the "National Electric Code" of the "National Fire Protection Association" for electrical construction. Georgia Power does not use the NEC requirements for power distribution. It up to their customers to abide by the code, not them.

 

There's a single ground rod, although it seems unusually close to the foundation. It's also under a covered deck, and I've wondered if that might keep the ground dry and reduce the effectiveness of the ground rod.

I freely admit I'm not well educated on best ground rod practices and the thoughts above are pure speculation.

Normally the minimum is two 8ft rods ~ 6ft apart, this is both NEC and CEC.
Check all neutral connection on the panel neutral bar.
Many weird conditions can occur when you have a loose N connection and both L1 and L2 happen to share the same neutral.
Do you happen to have a metallic water supply to the residence?
Max.

 

As for the simulation behavior, what part seems inexplicable? The simulation behavior seems correct to me - the only thing confusing me right now is the electrician's claim that branch circuit wiring problems could cause bad neutral voltages at the breaker panel. If I've done something wrong in the simulation, or if I'm misinterpreting its results, please let me know.

The simulation makes sense but having seen some really peculiar things over my years I can't explain what you have going on. I really like how Max summed this up:

Many weird conditions can occur when you have a loose N connection and both L1 and L2 happen to share the same neutral.

Coming from the load center each branch circuit of 120 VAC should have an individual neutral (and ground) run. The only place all neutral and ground circuits should be bonded is at the mains panel. 240 VAC appliances like a stove, drier, air conditioner or electric heaters should have their own L1, L2, Neutral and Ground at the load center. I have seen a few instances where on installation shortcuts where taken and later resulted in strange problems. Examples where a neutral for example was shared and should not have been. Things go along fine for years right till a bonded connection in a box somewhere gets overloaded or simple becomes loose with age. The trick to trouble shooting becomes trying to figure out which branch circuits go where which is a challenge since the wires are in walls.

My guess is the power company will show up again and run a load test at the point where their responsibility ends. Then the problem gets placed back in your lap. I don't see where the symptoms you have will be a result of a hot (L1 or L2) with a fault but think your symptoms point to a neutral. The problem lies in not actually seeing how everything is distributed in your residence.

Ron

 

Normally the minimum is two 8ft rods ~ 6ft apart, this is both NEC and CEC.
Check all neutral connection on the panel neutral bar.
Many weird conditions can occur when you have a loose N connection and both L1 and L2 happen to share the same neutral.
Do you happen to have a metallic water supply to the residence?
Max.

Yes, but there's a mix of copper and galvanized pipe in the house, so there are several dialectric unions and I'm not totally sure which, if any, pipes in the house might be meaningful ground references.

 

Likely the pipe on the inlet of your water meter is a good candidate.

 

The METAL water pipe as close to where it enters the building as you can (typically in basements). Not as good as a dedicated ground rod as close as possible to the panel, but was standard practice in rocky areas (or multiunit housing construction) where driving a ground rod was near impossible. Doesn't work worth a damn with PVC (yes it's a joke, but it has happened). Yes, I have had a jackleg electrician tell me he always shared the Neutral on separate circuits. IE using a single romex cable (BLK, WHT, GRN) for 2 separate breaker circuits (HOT WHT, HOT BLK, GRN NEU, no ground) and that the code said it was OK.