Strange Electricity Spikes

Hello, just today I started seeing huge spikes of power consumption on my Sense. I have 2 HVAC units that run throughout the day and I’m suspecting an issue there but I wanted to show see if anyone else had seen this kind of behavior before.

Screenshot_20190923-120301|250x500

Very interesting spikes… Could indicate AC capacitors gone bad. That some of the spikes are coincident with changes in power usage indicates they are likely linked. Can you zoom in on one or two of the spikes just to see how long they are.

Here are a few of the spikes. They are all very brief, 1-2 seconds.

Here are a few more

Agree with @kevin1 from the look.

In-rush current measurement time … perhaps make Support aware, they may be helpful with this stuff now.

https://community.sense.com/t/graphs-of-utility-voltage-for-detecting-issues/3698/44?u=ixu

Are you ok with the Data Science team taking a look? We’ve been doing some work to detect AC issues and may be able to offer some insights.

One thing about it being the possibility of a capacitor, it would be a the start of the compressor cycle and this appears to be somewhere mid cycle

Definitely Ryan. Can you help get the ball rolling?

1 Like

Good catch Sam. I’ll keep monitoring.

I have a 60 bottle Haier wine refrigerator. When the compressor cycles on, there is a very brief spike about 5x the magnitude of the running draw. I have no reason to believe this is anything other than the inrush current associated with starting the compressor. I’ve assumed it’s either due to this being a 16 year old inexpensive Chinese made device and/or the relatively small compressor size (about 170W when running.) I haven’t seen this with any other compressors (either identified devices, such as my A/Cs or unidentified devices, such as my Liebherr refrigerators) . In fact, the wine fridge was the very first device Sense identified, presumably due to the more unique signature.

That’s normal behavior for motors and especially compressors. It has to do with getting a motor going with single phase service.

it could possibly be the run capacitor is beginning to fail if it has one. how old is the unit? are the fans working properly? could be a lot of different reasons for the spike, sadly these kind of mysteries are difficult to track down and if something is failing unfortunately until it fails you may never find it. if it started recently the spikes are your warning, 50+ years in the HVAC/Refrigeration business has taught me that

You can’t really assume that very short spikes are necessarily represented precisely in the Sense display. The in-rush current spikes are often sub-second (of the order of 4 x 60Hz cycles) so the translation to the display is not really showing you what’s going on.

Reposting link from above regarding measurement of in-rush:

https://community.sense.com/t/graphs-of-utility-voltage-for-detecting-issues/3698/44?u=ixu

Zoom all the way in to the waveform and establish the duration. You can pretty much ignore any “line” spike.

1 Like

I only found it notable because, of my 7 compressor devices (5 of which are identified), this is the only one where sense records the in-rush current with such stark visibility.

It looks like this example has 2-3 seconds of high current draw. Agree that it looks like there’s a problem here. This kind of problem ought to be audible if you’re near the device that’s causing it when it happens.

@senseinaz I shot you a PM with some possible answers from the DS team :slight_smile:

In short, this is likely to be some sort of AC failure.

As usual in these threads I think we got a little tangled @pswired . I was referring not to the OT but this one:

I agree on the OT determination but was drilling down for @dudleynyc

1 Like

Got it. that makes perfect sense to me.

I don’t believe you have a problem, looks as it should. Here is my heat pump that is less than a year old and a refrigerator months old.
Notice the spikes.
You may not see the spikes with inverter driven motors as they are setup like three phase.
Those spikes are normal but I don’t know what is causing all the continuous choppiness on your graph.

“Choppiness” at that scale is pretty normal for the mains PM … I think you might be assuming that was a Device waveform?