I have been on vacation for the last week and a half or so and before I left I turned off/unplugged most everything in my home besides the WiFi Mesh system, a Raspberry PI3, an our 2020 purchased LG Refrigerator. No lights, no HVAC, no fans and all TV’s are unplugged too. Leaving a pretty clean “line” Sense successfully identified what I believe is the Defroster in this particular fridge but for daily operation, the linear compressor and always on nature of this fridge seems to have eluded Sense’s detection thus far. The good news is that I have a really good repeating signature of the Fridge with the home being so quiet power usage wise. At any rate, Here’s the very repeatable signature showing up at night. It is there during the day too but the scaling when viewing during the day is thrown off due to my solar production.
I have 2019 very simar model linear compressor and Sense has only discovered the defroster as well on it. What does your defroster power signature pattern look like?
Here is what Sense shows. Far as I can tell this is the defroster. Could also be an ice maker. This one has two a regular and a craft ice ball ice maker
Nice, this example shows exactly why humans can recognize this pattern, but it eludes Sense’s current detection mechanisms.
- The on ramps are long - maybe a rise of 100W in 4-5min. The fall looks much faster, but what good is an off-signature if there isn’t a matching on-signature. If you looked for that on-signature using a 1/2 second window, like Sense does, you wouldn’t see much of a rise or fall - that would be 0.2W over 1/2 second, maybe more like 1W/sec delta at the start.
- As you suggest, you have peeled your usage back to just this lone fridge pattern, with very little underlying noise. In a noisy household, I’m pretty sure that a human couldn’t discern the pattern, especially with the long on ramps blending into other switching activity. Sense is able to detect some slow ramps, like EV chargers, but that is aided by the sheer magnitude of the ramp. Easy to detect a slow ramp to 10kW, not as easy with a 100W ramp.
I think we have the same fridge, except mine is the 06 and yours the 16. I put a HS300 on it and here are the waveforms for the past week, day, and hour. Let me know if anyone wants further details or has questions. The ironic thing is this modern fridge is projected to cost $36 for the year, but my 20 year older, larger fridge, is project to cost just $17 for the year. Maybe is the two ice makers in the new fridge that set it far apart in energy usage?
Is your larger fridge in your garage or somewhere that’s typically colder than the rest of the house? The biggest variables that determine fridge consumption are typically the external temperature and the age of the unit.
Yes, but I think there is more to it. The older fridge is in the unheated park of the basement not in the sun, while the newer fridge is in the kitchen. The older fridge is a full sized unit though, and the new one is simply counter depth.
Sense found the older fridge right away, and I think it’s because the compressor turns on every roughly 45 minutes, and runs for 10 minutes. Otherwise it’s completely off. My newer LG fridge is always on using at least 5 watts, but it’s compressor turns on every 30 minutes and runs for 30 minutes. At their peak, the old fridge compressor uses 144w while the new LG peaks at 95w.
Do either of the fridges have a built in ice dispenser? Or a difference in the freezer location (top, side by side, bottom, etc)? Check out this Technology Connections video about freezers.
Also, yes ice makers will add to the consumption too because at the very end of their ice making cycle, they actually run a small heater for a few minutes to melt the surface of the ice a bit so it slides out of the ice maker. Mine runs about 180W for roughly 3 minutes just before dumping the ice. My ice maker seems to use about 3 - 4 kWh per month. That is about $6 - $7 per year in cost for just the ice maker, and that’s with our relatively cheap electric rates in FL compared to much of the county.
Exactly. The basement model has none, the newer model has two, one for the door ice dispenser and one regular one in the freezer. We also open and close the upstairs one multiple times a day, but the basement one could go a couple days without any openings.
I know there are blankets to put around water heaters to keep them warmer, but are there blankets to put around refrigerators to keep them colder longer? Is the most temperature loss just from the door seals, or from the body itself?
Hey @brian5. This blog does a deep-dive into refrigerator functionality, along with recommended tips (at the bottom) for upkeep. Changing the seals is recommended after a while for older units that are opened and closed frequently.
From what I remember in the blog I shared, this is a key contributor (along with the note you mentioned earlier about the older fridge being in an unheated area of the basement.)
Not that I’m aware of, but I did enjoy imagining my refrigerator nestled tightly in a blanket.
Biggest loses by far are due to doors being opened. In old units the door gaskets do get brittle and leaky, and that contributes to losses, but that shouldn’t be affecting your newer unit.
Very interesting thread. I have a Samsung Digital Inverter compressor (metal cooling) which seems pretty energy efficient. Here is what the typical cycle of the compressor looks like:
While I haven’t been able to identify a defrost cycle, I have noticed a pattern that Sense hasn’t detected that is coming from the fridge (determined by turning off breaker) Every 5 minutes, exactly, there is a small amount of electricity used for about 2 minutes. I’m guessing this has something to do with the metal cooling feature.
It isn’t much power.
As for the ice maker, Sense has detected both the ice maker and the motor that dispenses the ice from the fridge into my glass.
Above is the typical ice maker cycle when making ice.
And above is the motor that runs when ice is dispensed into a glass or other container.
I also have a small, deep freezer in the garage that runs routinely, but only uses a lot of power when it starts.
Been fun learning about my power consumption. Surprised to learn I really only have two heavy hitters, my electric hot water heater, and my electric clothes dryer. Since I am billed on usage and demand per hour, laundry day is a real challenge. I’ve put the hot water heater on a wireless switch so it won’t run at the same time as the dryer.
I was just playing around in the lg thin-q app (were you can controll certain LG smart devices and appliances). It turns out there is a built in energy monitor. Will be interesting to compare to sense data should it ever find my fridge.
They’re not far off:
I’d trust the Kasa more than the fridge. One device is built to monitor energy, the other is built to keep food cold.
That’s about a14% difference, which is pretty large. Also, the waveforms look quite different.
I’ve checked Sense, against a Kasa, against kill-a-watt, against my WELserver, against a lab grade wattmeter, and they all fall within about 1.4%. Kill-a-Watt was the closest at 0.4% with the Kasa a close second at 0.6%. Sense came in third at 0.7%. This was on a heater, not sure how they’d do on motors, etc.
So, I’d trust the Kasa readings. Remarkable accuracy for such an inexpensive device.
Thanks for the tip on the energy monitor in the fridge app. Sense is saying 1.5kWh per day if I add in the ice maker power, the app for the fridge says about 2kWh daily. But Sense hasn’t equated that 20w for about a minute, every 5 minutes, to the fridge yet.