I recently installed this device (there are several alternatives out there, but this is a relatively inexpensive one). It installs in your breaker box, and measures full house power, as well as several (8 to 16) channels for individual circuits (you don’t necessarily need a sensor for every single circuit, just the important ones). You monitor it with a phone app (it connects to your Wi-Fi). You can watch real-time graphs and all that.
Anyhow, I’ve found it rather useful so far. It helped me discover a battery charger that just burned 40 watts constantly, as well as a rogue computer in my attic (No, I’m not being spied on. I put it there. It was supposed to be powered down, but must have powered itself up after an outage.). I’ve also found that some lights are using way more power than expected, and found a couple lights that I’d forgotten to switch to LED.
So, very useful, and I expect it to pay for itself in about two months. I had a Power Angel (plug-in power meter, similar to the Kill-A-Watt), which is also useful, but requires that you know about the device. You also have to plug the device into it, which is annoying if it’s a computer or something that you don’t want to power down. Or if it doesn’t actually plug into anything, like lights, or something big like a refrigerator.
Anybody else have one of these? Find any stupid power hogs?
Hey, thanks for that. I’m gonna order one of those. I have a plug-in monitor that I’ve used for measuring individual devices, but I just started looking for something like this, and it looks like just what I need.
I mentioned before that I’ve been told we need a service upgrade ($20k+) to install, well, anything else. Hot tub, freezer, level 2 charger. And yet, we have a 100A service and just tye standard list of large electrical appliances (stove, dryer, fridge, small A/C) with gas heat and water. Apparently lots of other people with similar appliances have gotten licenses for charges or hot tubs or whatever, so I want to start tracking actual power used and find out just how much we are using,
Does that device show peak current? The key for me isn’t how much total energy we use, but how close the peaks are getting to tripping the house breaker.
Note that there’s a 16-channel model for about $160, if you think that’ll be useful.
It has two 200 A sensors for the service lines. The box actually supports 3 phases, but only comes with two sensors, so if you have 3-phase you’ll have to buy an extra.
The individual sensors (8 with the box I linked, up to 16 total, and can upgrade later if desired) support 50 A each. The app has a “multiplier” setting, so for your EV charger, you’d just put a sensor on one line and then set the multiplier to 2.0. Probably the same for your hot tub and any other 240 V appliances (though some may be imbalanced).
The app shows a “balance” figure, which is just the whole-house power minus the individual sensors. Will show zero if you have all breakers hooked up, but more likely you’ll have a few unmonitored ones.
I’m still exploring the app, but it shows a “monthly peak 15 minute demand”. That appears to be the 15-minute-averaged peak over the last month. That should correlate pretty well to how close you are to your breaker limit (since they have some amount of leeway built in). You can also download the CSV data and probably gather the 1-second peaks that way (it samples on 1-second intervals).
The device was pretty easy to install if you’re able to install outlets, etc. The only tricky part is that if the house breaker is in the same box (or inconveniently located), then you’ll be installing the clamp-on meters on live circuits. I did mine live, and if you’re careful it’s not a big deal, IMO. It’s all well insulated so there’s no chance of accidentally shorting anything.
The biggest problem for me is stuffing all the wires into the box. I only have a small, 90 A box for my condo, and there wasn’t a huge amount of room. Enough, though.
Yeah, fitting it all in my electrical cabinet might be a challenge. I’m also concerned that this only works with their cloud servers. I’ve seen too much cloud based stuff turn into a subscription service to keep it working. I had originally planned to make my own monitoring system with some arduino, some inductive clamps and wifi. But that’s a lot of work for a one-off.
We bought a little ‘vector’ robot a couple of years ago for Christmas. Very cool little thing with a lot of AI enabled intelligence that used the cloud. It was also expensive, at $349 retail. A year later the company was sold to someone else who decided to ‘monetize’ all the owners and announced that it would now cost $10/mo to keep a little toy running. We passed. Now vector is pretty stupid.
I know someone who has that. Maybe not the same exact model, but basically the same kind of system. When I saw it, he was getting the 1-second data in real time from the thing; it broadcasted it but not via wifi and no app (or cloud server or internet) necessary.
How was it broadcasting it? Bluetooth? And you are talking about the energy monitor and not the Vector robot?
How was the data received if no app required? Serving uo a web page without wifi somehow? I’m confused.
There are a number of data transmission standards for local broadcast (Zigbee, LORA, for example), but you would need dedicated hardware on both ends.
One other option for house monitoring is to tap the Zigbee signal from your smart meter, if you have one. But that would just give you real time total use, not per-circuit.
I would LOVE to hear how this will enable you to save $50/month on your electric bill.
I’m sure I have more phantom loads that the average guy, but… my electricity usage is dominated by our HVAC. Everything else is basically noise.
My last electric bill was $169 for 604 kWh. That’s $0.28/kWh average. Going with 720 hours/mo, a 1 W constant load comes to $0.20/mo.
That’s actually conservative, since I’m already pretty diligent about scheduling my high usage (EV, clothes drying, etc.) to off-peak and generally am more active outside peak hours. But I’d have to run more numbers to know the adjustment factor (it’s not trivial because our bill is dumb). Ballpark, I’d guess the real cost of a 1 W load is around $0.25/mo.
The evil battery charger was 40 W by itself, and the rogue computer was maybe 180 W. I noticed that my main server was still using more power than I hoped, and after finding that the clock throttling wasn’t working properly, I clamped them manually to save around 15 W. My bathroom lights aren’t on all the time, but I do tend to leave them on longer than I should, and after seeing that they’re close to 400 W (I’d thought they were more like 200 W), I ordered some LED replacements. I actually don’t know yet if these will be bright enough, but if nothing else it’ll convince me to be more diligent about turning them off. There were a few other small things. I’m not counting finally emptying/unplugging the garage fridge, since I was already planning on doing that, but going through the process did finally push me to go and do it.
So, let’s say 250 W time-averaged load. At $0.20/W-mo, that’s $50.
Our last bill for electricity was $497. Only $200 of it was the actual electricity. The rest was taxes, ‘service fees’, ‘delivery fees’ etc.
We used a megawatt of power. That’s too much, but we have an electric range, electric dryer, and my son uses an electric heater in his room because the HVAC peoole screwed up the calculations when they built the house (a bedroom over an unheated garage has different heating needs than one over an interior heated space).
Also, all of us are heavy computer users, having three home offices going during the day, and my son does gaming at night. Most people don’t think about how much energy they are burning if they are gaming on a good gaming PC. Five hours per day of gaming can easily be 2.0-2.5 kWh, so 60-90 kWh per month per machine. Add to that our office computer use, a media/backup server running 24/7 and the (less amount) gaming and photoshop and stuff that I do, and probably 20-25% of our energy consumption is due to our computers.
Presumably, you mean a megawatt-hour . That is a pretty crazy high bill; about $0.50/kWh. Rooting out some unknown power users could be a big savings, even if they’re relatively small.
Oh course, megawatt-hours. Yeah, we get hit hard by service fees and taxes.
Carbon taxes hit us and Saskatchewn harder than many other provinces, because we really have no source of hydro, and 91% of our energy comes from fossil fuels. We have very poor capacity for wind and solar as well - wind is better than solar, but neither are great, and even combined to their maximum effectiveness can only make up maybe 20% of our power needs.
Saskatchewan has announced that they plan to build three nuclear reactors. Alberta should join them. If fossil fuels go away we will have no source of baseload power at all if we don’t. Alberta would essentially be impoverished and see massive capital and human flight out of the province. Canada is a high energy country. Quebec has huge amounts of hydro, Ontario has hydro and nuclear. They can provide baseload to their neighbors if necessary. Alberta has nothing but fossil fuel.
At those rates, nuclear makes a lot of sense. I’m a little surprised that you aren’t just getting super cheap coal power. I’m paying pretty high rates in large part due to clean generation efforts. California could still do better, and PG&E has lots of problems, but overall our power generation does pretty good on the carbon emission front. Lots of renewables, a fair amount of hydro+nuclear, and the remainder natural case. Coal is tiny.
I worked out the math for a 1 W load and it comes to $0.274/kWh. So, actually no adjustment factor as I’d guessed–I’m already pretty close to the average rate. I guess the much higher peak rates make an outsized difference in the weighted sum. At any rate, my $50/mo guess was based on the smaller figure.
You are likely to see some increases soon - Diablo Canyon is going offline, and with it about 9% of your electricity.
We had lots of cheap coal, but our last government ordered the premature shutdown of coal power without any replacements, and paid billions in penalties to the coal operators to do it. This was to save the planet. Then they had a glut of coal we aren’t burning, and sold it to China. So… planet saved, I guess.
I’m going to have to check my bill again and see where all the money went. It does seem very high.
Nuclear makes a lot of sense not just for the rates, but because I don’t see that we have any other option, other than to buy our power from our neighbors if we can. In a world of increasingly unstable grids, I would like my province to have its own power.
I agree, this is a significant limitation, but unfortunately that’s the way pretty much everything is headed. I have many devices that will be bricks when the cloud services shut down.
The Emporia device is perfectly capable of running a local web server, but they claim that data analytics is part of their business model and they won’t be giving that up. That’s also a big trend in smart devices of all kinds.
I can only hope that someone (maybe me) will hack it at some point to get at the data.
Yeah, that would be the alternative. Plenty of reliable solar down south. We do need better grids. It’s not like the electrical grid is some ethereal thing that can’t be controlled. We just need to build more capacity and more robust control systems.
I am talking about the energy monitor (a number of clamp probes attached to a control box).
The data was sent by radio and received by a remote meter with a rudimentary LCD display which displayed the current power consumption, average/daily/total use, cost assuming you programmed in the rate, etc.
My friend was able to receive and parse the data packets using an RTL-SDR dongle, if that helps narrow down what it could be. I can also ask him.
Probably not Zigbee, then (2.4 GHz is too high for RTL-SDR). LoRa is a possibility, or any of a zillion other standards and non-standards in the ~900 MHz ISM band.
The unit I linked to uses an ESP32. Probably wouldn’t be an extreme challenge to just build a custom firmware.
I don’t have a power meter for the whole house. But I did build this:
The top three show the voltages present inside the breaker box. As can be seen, there is 121 V on one leg, 121 V on the other leg, and 245 V between them. (It should read 242 V. Am guessing it’s due to meter error.)
The two bottom displays show the voltage and current for one of the circuit breakers in the panel. It is the circuit breaker for my generator. It allows me to monitor the amount of power produced by the generator when I am powering the house with it. (I power the house with the generator when the electric goes out.)
In the lower right you will see a switch and buzzer. They are connected to the output of the electric meter but before the main circuit breaker panel. It alerts me to when the electric has come back on. When the buzzer sounds, I turn off the buzzer, disconnect the generator, and then close the main circuit breaker in the panel.
Finally, the box contains a bunch of 120 V and 240 V transient voltage suppressors. So it also functions as a whole-house surge protector.
$20k??? Last year I had my electrical upgraded from the 1960 60 amp service to 200 amp, so I could add a minisplit unit. The new main breaker outside, replacing the fuse box with a breaker inside, and the power upgrade from the electric company only cost $3K. I didn’t have any outlets installed, or much change to the wiring indoors. I’d be getting some other estimates.
Very cool! I wouldn’t mind having a control panel thing for my setup. A Raspberry Pi with a small display would do the trick. Although it is still cloud-based, it’s possible to extract the data from the device I linked to, so I could put together a real-time monitor.
How’s the frequency stability for the generator? I expect that many will have frequency sag under heavy load. Could be worth displaying that, too.
. That is a pretty crazy high bill; about $0.50/kWh. Rooting out some unknown power users could be a big savings, even if they’re relatively small.