We had a pretty good storm several days ago with high winds. The next day, I first noticed some flickering of ceiling LED lights I installed a month ago. This has continued since this first observation. Yesterday, I noticed flickering of ceiling LED lights installed a few days ago in another room on a different circuit. Then this morning, for several minutes, we lost power on at least 4 different circuits (including both circuits with ceiling LEDs installed recently), while at least 4 different circuits had power. So strange to see so many circuits without power while an energy hog like the coffeemaker was still running.

We didn't lose one of the hot legs since we had power on circuits on each of the hot legs.

Because the LED light flickering manifested before this morning's partial brownout(?), I thought the problem was either in the LEDs themselves or in my wiring. So I picked up a CAFI/GFCI breaker that I was planning to install on that circuit to try and see if there is a poor connection in the wiring, and if not, then I know it's the LEDs. But given this morning's partial outage, it seems there may be more at play, perhaps related to the storm. I'll have to check with my neighbors when I get a chance to see if they noticed any issues this morning. The drop from pole to my house looks fine.

Any suggestions on how to investigate/isolate the issue(s)?

 

Have you measured the voltage?

 

No, I should have at the time, but my first thought was to see which circuits were affected. I'm sure if I check now, it will be 120v on each leg to neutral.

I'll leave the panel cover off (no worries about anyone in house getting close to panel) so I can check voltage if this happens again in next couple of days.

 

I'm probably talkin' outta my butt...

A standard resistive-load light bulb, coffee maker, heater should still work on lower voltage albeit not very powerfully, whereas inductive-load LEDs, microwaves, motors may not function properly at all. Hence why traditional dimmers don't work well with LEDs.

If it's working well now and the voltage seems normal, maybe that was the issue?

I hope someone more knowledgeable than I will correct me.

 

It sounds like you did lose an incoming leg. Depending on the loads and if any 240v devices were active... it may not have appeared to be a problem leg.

You can't really see the condition of the connections on the drop but if the problem was wind related.... that's where the logical place for the problem to be.

 

You also need to measure hot to hot to check for 240 volts.

 

If you don't already own one, I'd recommend a kill-a-watt or something similar that you can leave plugged in to read voltages real-time at a glance. Whenever my lights flicker or my microwave seems to step-down in power, I try to glance at it to see if it's voltage related.

This is $10 at Amazon. I may get one to put on the other leg.

 

No 240v loads active (only the AC is 240v, so I'm certain no 240v loads were active).

After reading your post, I went through the list of devices and circuits, and most devices that were off were on the same hot leg, but there were other devices on this leg that were on. Here's the breakdown of circuits (all on same hot leg):
OUT (no power)
#7: recessed integrated LEDs, LED bulb, microwave
#11: fluorescent lights, LED bulb
#15: recessed integrated LEDs, bath heater/fan/light

ON (power)
#6: LED bulb, LED bulb
#10: coffeemaker, computer in sleep mode (no UPS, so computer exposed to line voltage)

Given no 240v loads, is it possible that I could have working devices listed above with the leg powering these devices in a low voltage condition?

 

I checked later, when everything seemed fine. I'm getting ~122v on each hot leg (to neutral) and ~245v between hot legs.

 

Thanks, I'll get a couple of those. That will be a lot quicker than trying to run down to the panel and measuring voltage every time I notice the lights flickering.

 

So I reported the issue to PoCo. Guy comes out and pulled the meter and he's saying this lug on the load side is burned and brittle and is the reason for the partial outage.

The line side is top half, load side is bottom half.




Wish I took a better pic while the meter was out. Looks almost like rust to me. But it does look a little darker than other rusty areas.

Here's a pic of the meter box from a few months ago. Here that lug looks more similar in rusty color to the other rusty spots. Seems darker now, so I guess that is a sign of scorching. Looks like the lower right slot that meter seats into is a little wider than the others. But I guess if that was the loose connection, he'd have noticed charring on the slot and on the meter.



He's saying this is a fire hazard and that I need an electrician to deal with it ASAP. Even asked me if I wanted for him to put a blade in to connect just the other leg and leave this leg disconnected. So I'll call an electrician in the morning. In the meantime, I'll shut off most of the breakers on that hot leg.

Seems a bit odd that I had these circuits mentioned earlier that seemed to be operating fine during the partial outage:
ON (power)
#6: 2 different LED bulbs (no flickering observed)
#10: coffeemaker (running), computer in sleep mode (no UPS, so computer exposed to line voltage)

On a separate note, the guy said that most of the new meter enclosures have a disconnect. Would that count as the first disconnect as per the NEC? If so, that would make my main panel a subpanel, correct? How do people do generator interlock setups if the disconnect is on the meter enclosure? If it makes a difference, I'm in NY and most main panels are indoors.

 

Yes a disconnect at the meter makes the inside a sub panel. You would need a four wire feed from the meter to inside.
If the inside panel has a master disconnect, you can still have the interlock on the inside panel.

 

Are replacement lugs available for 20+yr old meter enclosures? Or is this going to cost me a new enclosure and new outside service?

 

Can you still get a meter panel without a disconnect?

 

They're still available on Home Depot (for ship to store or ship to home) in NY. And NY is on 2017 NEC, which I think does not require disconnect outside. I believe 2020 NEC does require disconnect on the exterior of the house, but even then that would probably only apply to new service and everyone with existing service without an exterior disconnect would be grandfathered in. So I don't need a meter socket with disconnect even if I had the entire exterior service replaced, but of course if I'm going down that road, why not get a meter socket with a disconnect.

I've seen a neighbor with a new service that has a disconnect at the meter, but he also has solar panels. Perhaps a disconnect at meter is required to connect solar panels to the grid?

 

How does that work? I was under the impression that a generator interlock isn't allowed on subpanels. I guess that was mistaken understanding?

I notice that NEC 230.85 (new for 2020) refers to the disconnect as an "emergency disconnect". As it pertains to disconnects on meter sockets:
The "NOT SERVICE EQUIPMENT" part makes me wonder - in cases where the main panel is inside, if the meter disconnect is "not service equipment", that means the main panel is the service equipment, so then it's still a main panel and not a subpanel?

Meter disconnects don't have breakers, right? So they can't be service equipment b/c no overcurrent protection, right?

 

Given what the PoCo said about the bad lug on the BLACK leg, I was going to swap a few breakers from the black leg to the red leg so I could safely have some power on these circuits while I get an electrician. So I shut off the main, pulled off a few breakers and immediately realized that my assumption of how the bus zigzaged was incorrect. The way the bus runs, circuits 1, 2, 5, 6, 9, 10, 13, 14, 17, 18 and so on, are on the black leg. So circuits 3, 4, 7, 8, 11, 12, 15, 16, and so on, are on the red leg.

So I went back to my notes of which circuits lost power during the partial outage. And it turns out that EVERY one of the 4 circuits that I saw lost power is on the RED leg. And EVERY one of the 4 circuits that I saw maintained power is on the BLACK leg. This clears up the confusion I had about why it seemed some circuits had power and others didn't on the same leg - I was mistaken about which legs the circuits were on because I was mistaken about exactly how the bus zigzags.

So yes, I did lose one leg during that partial outage, but it's not the black leg the PoCo said it was. It's the red leg. So while the lug on the black leg in the meter socket looks terrible, that's not the current problem. It's likely the red leg's connection in the meter socket, which could be line or load side, but per PoCo guy, if it was line side, the smart meter would have detected it and per his conversation with whoever he called, they didn't detect any issues, so assuming that's true, then it's the load side connection of the red leg.

I'm glad I didn't swap those breakers from black to red, and I'm glad I didn't let PoCo guy connect only the red leg as he offered to do.

The other interesting thing is I mentioned in OP that I had flickering lights beginning with the high winds several days ago, and I had flickering lights every single day until yesterday morning's partial outage. Since that several minutes long outage, we've had ZERO flickering. And I didn't touch anything before/after the partial outage until just a few minutes ago. Which makes me wonder if the outage on the red leg has anything to do with the connection in the meter socket or maybe it's further upstream and has been resolved?

If I see any flickering of lights, I'm going to assume it's a poor connection in the meter socket.