[brentvw]

Yes, another pulsing lights thread! I've read all of them and am still at a loss, months and months of troubleshooting and can't figure it out.

Background: brand new house < 2 yrs old, 150A 240V service, any type of light will pulse, on/off a dimmer, LED or incandescent, and on any circuit. One leg 116 the other leg 120.

It took a while to figure out the culprit was a Ninja coffee maker. Every couple seconds it must be keeping some of the water hot so its always pulsing the lights where they dim and go back (I learned not to use the word flicker, there is no loss of power). So the lights go from a 10 to a 9 to a 10 to a 9 all day, it's like a mini strobe fest.

As I was about to write it off as a bad coffee maker (or as another thread so eloquently put it, cheap Chinese java pot), I found I'm able to re-create the same problem using one of those electric tea kettles that heat water rapidly (pulls about 10A when you turn it on). So I turn on the kettle and the lights go from say a 10 to a 9 and it holds the loss of light while the kettle is on, then when I turn the kettle off it goes back to a 10. If the lights are on a dimmer and dimmed low forget about it, the pulsing is much worse and rapid like the light is struggling to stay alive while the kettle is on.

So it's not isolated to a specific appliance or a circuit (same behavior on a dedicated kitchen GFCI 20A or a standard 15A), I can plug the tea kettle in just about anywhere and visibly see the light impact to surrounding circuits. Strange it doesn't seem to happen with microwaves, a/c, hairdryers, or garbage disposal, these have more normal behavior, where you turn a/c or disposal on it will pulse once and everything is normal while it runs, so turning it off has no impact.

Electricians have been out and tightened all the neutrals, POCO came out and did a triple deep freezer simulation on the 240 without any voltage drop. This was all before I figured out what caused the issue.

So I had electrician out and we would flip the kettle on and off and watch the voltage drop 3 volts on one leg and rise 1-2 volts on the other. He said something about harmonics? We went outside where an exterior outlet is next to service meter and we watched the same behavior there. He thinks I should call POCO out again and show them the voltage fluctuation at the meter and suggest thicker gauge wire be run to transformer.

I'm reading nobody will care about this level of voltage drop and it's normal. I'm out about $300 between 3 sparky visits and getting the right person at DTE that actually cares to look at this stuff takes a LOT of phone calls. They just see a clean 240 on their end and walk.

I suppose I could show DTE how the kettle behaves at the service entry point but I'm trying to minimize the continued back and forth.

Anything else I can try or communicate to move this along? I've never seen this level of sensitivity in a house so it's a bit surprising, especially a fairly new house, it just seems odd to write this issue off, I do have some other pulsing issues upstairs that I suspect are related to this load sensitivity? voltage drop? harmonics?

Any thoughts on where to go from here? Much appreciated!

 

[chiraldude]

Harmonics would have nothing to do with the kettle. It is a pure resistive load.
That said, the voltage drop on one side is simple but the rise on the other is strange.
The only thing I can think of would be a poor connection on the street side of the meter.
If you could pull 120 Amps for a minute or so and look at the wire connections for a hot spot with a thermal camera you might find something?

 

[joed]

voltage drop 3 volts on one leg and rise 1-2 volts on the other

Perfect symptom of a loose neutral.

 

[Domo]

First thing I'd do is disconnect every dimmer and fan speed control in the house and then test.

I've seen things similar to this a few times. Once, a dimmer had been installed as a fan speed control. When I got called in, it took a while to find and replace it.

Second time, it was a solid state light dimmer feeding an improper load (might have been an incandescent-only feeding an LED load - I forget, it's been a few years - but, there was a mismatch).

If all the dimmers/speed controls being disconnected does not solve the problem, I'd disconnect one leg of the incoming power to try and isolate to either leg and troubleshoot down from there.

Good luck.

 

[brentvw]

If you could pull 120 Amps for a minute or so and look at the wire connections for a hot spot with a thermal camera you might find something?

How can I do this?

This did happen before the dimmers were installed, so there were no dimmers when this happened.

The sparkys and poco tightened all the neutrals inside and out, I was hoping it would be something simple, but its not all devices all the time, definitely these heating elements causing some type of resistance

 

[HenryMac]

We had this very same issue at our last home.

One of the 200 Amp main breaker lugs was stripped out.

 

[CodeMatters]

Is the neutral same size as hots or is it reduced?

Regardless, it's a neutral issue.

 

[brentvw]

Ok, I got a couple multimeters and have been doing more testing on this. I can recreate this on a single circuit with all other breakers off, turn on tea kettle and lights will dim on same circuit, same behavior with hair dryer or coffee maker, they will gradually dim more as hair dryer is turned from low to high. I moved the breaker to other leg, same behavior.

Turning each of these appliances on voltage drops from 122 to 118 to 114 to 110 at receptacle. So this is a nice load for 1 20A kitchen circuit and maybe that voltage drop is acceptable but maybe it isn't?

I have a multimeter plugged into LEG2 receptacle watching as this load fires up, and it drops to 118.

So now I turn on microwave on another circuit on LEG1 and now LEG1 is 107 at receptacle and LEG2 is 114n at receptacle. I check service legs at entrance before main breaker and they are reading 114 LEG1 and LEG2 is 124.

So I guess what is normal here? Any ideas if this should drop this much? Is this pointing to supply side issue, maybe supply side neutral? How do I get poco to care about this type of issue when they just test 240 to house and not see this as just "normal"

It does seem a bit ridiculous that the lights are pulsing to all this activity relative to the voltage drop, but I'm not sure what is acceptable.

My only thought at this point is to go to a friends house and see what the behavior is there.

Only thing I have to go on with poco is rise in voltage of other leg but it's only a volt or two , why would they care? What would I say to them?

 

[Bud9051]

I didn't go back and read all of the thread but have a question.
In every electrical box you have 1 hot for 120v or 2 hots for 340v plus a neutral and a ground. For each of your readings what are you referencing to, copper ground or white neutral?

When you have this unbalanced situation have you checked from neutral to ground?

Bud

 

[curiousB]

If you are trying to sort out a loose neutral at the incoming service all your test loads need to be on the same phase (leg). The minute you put them on opposing phases the neutral only sees the difference current of the two loads. If they are closely matched, say two 1500W loads, then the neutral current will cancel and not be very revealing.


So I would get a few heavy wattage test loads and run them on different branch circuits that are all on the same phase. An iron, 120V space heater, toaster, microwave, .... As you go down the panel every second breaker is on the same phase (leg).



Try to avoid many or any 240 v loads (oven, electric furnace, electric dryer) being on for the test. Then measure each phase voltage, at the panel, while the 120V loads are on.



With loads on:



If your see 240V when you add the readings together it means the incoming phase lines and connections are good. If the neutral is bad you'll still see the phases totaling 240V but one will be larger than the other. The voltage split will vary with the load.



If you are seeing relatively stable voltage at the panel service lug point but see much less voltage at the receptacle you have a loose connection(s) in the branch circuit wiring.


If you see the phase to phase 240V at the panel decline under heavy loads it means poor lug connection(s) on the incoming lugs (panel, meter base) or that the service drop cable gauge is too small for your residence. The POCO should upgrade the service drop.

 

[Wiredindallas]

This is for sure a neutral situation. Very possibly at the power pole which only the power company can check and fix. They can be stubborn to deal with it. I had the same problem as did my neighbor. I am an electrician and he hired an electrician for his house. I had Oncor ( my power company) out and they said no problem. He had them out and told him the same. My Landlord had Oncor out when I wasn't here, then we all 3 had a conference call and they said that they did this and that sand there was no problem. After that, all was ok. Oncor never admitted they found the problem. But they did.

 

[brentvw]

Ok thank you for all the replies! I forgot about the A-B alternating phases vertically so when I tested other phase I moved the breaker horizontally which was same phase.

So I did test both phases by turning off all the alternating breakers and loading up each phase with heavy wattage appliances, space heaters, hair dryers, microwave, while the other phase breakers were off.

Both phases started at 120 give or take a tenth of a volt, and phase A loaded up dropped to 109-112 and phase B with no load would rise to 124-126, same happened on inverse so if I loaded up phase B and left phase A with no load the same behavior would happen, this was tested at entrance to main breaker. Same results from hot to neutral and hot to ground.

So definitely happening on the incoming supply side and not just in the house, so I will alert poco to this variation.

One last question, what to make of 109 + 125? That is only 234V. Is there another issue going on that they would not equal 240? Or is that normal?

 

[joed]

Did you measure the incoming voltage? It was probably 234 at the time of the test. Or just slight meter inaccuracy.

 

[MTN REMODEL LLC]

CAVEAT.... I AM NOT AN ELECTRICIAN... just a GC sort of guessing....

And I'm getting a little mixed up exactly on all your dfferent tests, which leg you were on, was the other leg loaded... and what exactly you were testing.... but... just food for thought...

I'm inclined to think the problem is focused around your service neutral.. either size or somewhat corrupted, or overly long also.

Basically, isnt' the service basically a three wire (MWBC... except it's not a branch circuit).

Assuming a corrupted/undersize service neutral:

If just one leg has a load on, and the greater the load, the more voltage drop would be expected....Yes

But, if you then put a load on your second leg, you should see a voltage rise, as the voltage drop due to the neutral service is negated by the split phase coming into play (at your main panel) and you no longer have the length of your service neutral causing voltage drop. (Or only effected by, and carrying, the differential current of each leg.)

Would this senario be consistant with your testing experience.

(this is sort of strange, but think about it. Are you sure a light is dimming per se, or is it just not running as bright as when you have a load on your other leg....)

Hope we discover the exact issue, as it seems interesting.

Good luck

 

[joed]

The voltage rise and drop is not related to the length of the cables. it is related to the two loads being in series with each other since there is no neutral. If the loads were the same size (no matter how large or small) The voltage across each load would be the same.

Here is a good demo.

https://www.youtube.com/watch?time_continue=3&v=HsSlQnGCvdg

 

[MTN REMODEL LLC]

The voltage rise and drop is not related to the length of the cables. it is related to the two loads being in series with each other since there is no neutral. If the loads were the same size (no matter how large or small) The voltage across each load would be the same.

Here is a good demo.

https://www.youtube.com/watch?time_continue=3&v=HsSlQnGCvdg

I would agree with what you said...... for a broken neutral with/or balanced loads...

but what occurs with a corrupted (higher resistance neutral) and unbalanced loads...

 

[brentvw]

If you see the phase to phase 240V at the panel decline under heavy loads it means poor lug connection(s) on the incoming lugs (panel, meter base) or that the service drop cable gauge is too small for your residence. The POCO should upgrade the service drop.

Yes, this was a knock down/rebuild electrician pointed out original poco wiring, as to whether or not 234 is decline worth worrying about or if it should always be 240? I’m not sure

If the neutral is bad you'll still see the phases totaling 240V but one will be larger than the other. The voltage split will vary with the load.

Yes, seeing 109/125

Did you measure the incoming voltage? It was probably 234 at the time of the test. Or just slight meter inaccuracy.

Phase A was 120 and Phase B 120 to start so 240 even. Phase A dropped to 109 when loaded, phase B rose to 125, so 234 total.

The test was to load up one phase and see what happens to other phase. I loaded up Phase A, voltage dropped from 120 to 109 and phase B which had no load rose from 120 to 125.

The same results happened for the inverse. So if I loaded up phase B, voltage dropped form 120 to 109, and and phase A which had no load rose from 120 to 125.

So from what I’ve gathered so far, this voltage rise should not be happening on the opposing leg to 125, and I should advise poco of this.

The other possible concern is the amount of voltage drop, for phase to drop to 109 volts may indicate problem with incoming service wire gauge. Recall the lights are pulsing everywhere, on the same circuit, on circuits on the same phase, and circuits on the other phase. I’m assuming they are reacting to the vairations in voltage or some type of arc. I can’t leave the coffee maker on without the lights changing brightness intermittently all day as coffee maker alternates the heating element on and off constantly.

So if it’s a neutral issue causing voltage variations (109/125) or incoming service line gauge needing upgrading (240 is reduced to 234) I don’t know, I suppose I could probably bring up both issues to poco and see what happens?

 

[brentvw]

Well, some good news, had poco out today, explained issue in detail, tech did a line side test after pulling off meter saw the 112/125 under load and diagnosed as open/loose neutral. Still said I was technically "in range" but would open an overhead ticket to address.

So, good news and will see what happens.

It's surprising how difficult this case is to keep open and actively working with poco, all the tech cases automatically close as the computer tests 240 clean to meter, so I have to call back to open a 2nd case or do override, then the overhead cases close as the linemen test 240 clean, it's like they forget that they provide two phases.

Extremely frustrating.

 

[brentvw]

Plot twist. Lineman came out and said to ignore everything I've been told by techs or sparkys and that he is the alpha and the omega of power.

He did 80 amp load on single phase at my meter and showed 105/125 and said absolutely not a neutral problem as 125 would go to 140+ under that load. He also thought 105 wasn't a big deal given that size load. He came in house and saw the small loads flashing the lights and a couple appliances droping me to 113.

He suggested pursing this as a load problem -- my house is at end of transformer 12 houses down from an open transformer serving 20-30 houses (open apparently meaning it was not connected to the next transformer down to help balance the power)

While pulling 80 amp test on my leg he walked over to neighbors house and clamped their meter, with my 80A going neighbors leg dropped to 113, stopping the load neighbor leg went back to 120.

I thought neighbor dropping to 113 when I pull 80A a concern (I have 150A service) 113 seems to be their barometer (5% variation)

We also discovered my house is at the end of transformer secondary lead using extended midspan service with extra long wire to pole a couple houses down, i.e. my wire doesn't tap in it actually clamps to the line to hold it, then runs to pole where it connects, probably a good 100 feet or more.

So he came up with 3 possible solutions:

1. Connect the two existing transformers half way down block together to provide more balanced power
   
2. Run a new thicker wire and tap into the wire at the mid-span, reducing the extension
   
3. Install additional transformer at my house since 20-30 houses already on one and seemed like too many

He was leaning toward #1 and #2 as he can do those himself and #3 would go into the utility vortex and potentially take months.

In the end he thought both of those are band-aids as I'd still be at the end of the transformer line. He also can't get his bucket to the corner of my back yard so he'd have to run a new line across the center of the yard over the pool area, which I don't like.

So, pursuing option #3 and very curious how poco will handle.

 

[joed]

said absolutely not a neutral problem as 125 would go to 140+ under that load

True.
Everything he is telling you sounds correct to me.