There are a lot of good reasons to have a better understanding of one’s household power use, and that is especially true for those that do their own solar power collection. For example, [Frederick] determined that it would be more efficient to use large appliances (like a dishwasher or washing machine) when there was excess solar power available, but the challenge was in accessing the right data in a convenient way. His Raspberry Pi-based live energy monitor was the solution, because it uses an LED matrix to display live energy data that can be consulted at a glance.
Interestingly, this project isn’t about hacking the power meter. What this project is really about is conveniently accessing that data when and where it is best needed. [Frederick] has a digital power and gas meter with the ability to accept a small wireless dongle. That dongle allows a mobile phone app to monitor power usage, including whether power is being taken from or exported to the grid.
Since [Frederick] didn’t want to have to constantly consult his mobile phone, a Raspberry Pi using a Pimoroni Unicorn HAT HD acts as a glanceable display. His Python script polls the power meter directly over WiFi, then creates a live display of power usage: one LED for every 250 W of power, with the top half of the display being power used, and the bottom half representing power exported to the grid. Now the decision of when to turn on which appliances for maximum efficiency is much easier, not by automating the appliances themselves, but simply by displaying data where it needs to be seen. (This kind of thing, incidentally, is exactly the idea behind the Rethink Displays challenge of the 2021 Hackaday Prize.)
As for those of us without a digital power meter that makes it easy for residents to access power data? It turns out there is no reason a power meter’s wireless service interface can’t be sniffed with RTL-SDR.
8 thoughts on “Live Energy Monitor Helps Plan Power-Hungry Appliance Use”
I’ve objected to quite a few posts of late, as have others. It’s very nice to see a well written post about an interesting and non-trivial topic.
i’m sure there is some advantage to be had, but the idea of adding this complication to your daily chores seems expensive, and the reward quite small.
the appliances that use a lot of instantaneous current are heaters: various stoves / ovens, and the dryer, and so on. the appliances that use a total large number of amp-hours over a long period of time are the mostly climate control…a/c, dehumidifier, fridge, air circulation.
i guess you could get some advantage from timing your a/c to come on when the sun is strongest. that one has the benefit of being something that is generally controlled by an unattended thermostat already, so you could get a systematic improvement without babysitting the thing. but unless you have a lot of partly-cloudy days during the hot season, i doubt that would have much improvement over the fortunate coincidence that you need a/c more during the day.
i could be wrong but i think this situation is governed by amdahl’s law. the bulk of your electricity expense is incurred in things you don’t really want to time-displace, so playing with the ones that are easy to reschedule isn’t going to make a difference.
household energy usage is in stark contrast to things like giant arc furnaces for melting metal, which can often be timed to run when the cost of electricity is lowest.
Being in a household that actively manages our energy use in conjunction with solar production I can speak first hand of the coat benefits.
I’m am slowly adding controllers to big power devices to automate the process to make it a bit easier.
Just by putting out first energy monitor on paid for it self in the first bill (quarterly billing) – that was before adding solar production. Just by knowning what you are using at any point in time makes managing it a lot easier.
There are obvious things that you can do with out monitoring but it realy shines when kids leave lights on in a room that is out of sight – or I forget to turn the compressor off the AC is left on etc. having an centrally mounted energy display you get to learn what the base load is and different levels of consumption point to particular devices.
And it becomes a bit of a game – how low can we get the energy consumption and keep it – what do we realy need to leave on.
The energy savings really do outweigh the cost and energy of the monitoring system.
Home power monitoring based on an outlet, circuit, panel, or whole house seems like a cool idea at first. Perhaps because it involves induction coupling sensors at various points of wiring around the house and with sufficient data, producing pretty charts. After a while, you realize the obvious things are obvious as noted in the above post. As far as using the power monitoring to change habits, it doesn’t really work that way. You’ll learn that people self adjust those things they can and otherwise grow accustomed to things they can’t change. It already seems intuitive if you you’re on a tiered level price plan were running the washer and dryer at night is cheaper than running it during the day, you don’t need fancy graphs to tell you that. The most useful story I’ve seen regarding power monitoring is an individual who goes overboard in this endeavor and identified a freezer that needed repair based on its power usage fingerprint. Like so many projects, this is cool for a while once you build it and monitor it, but then it becomes pretty useless.
I’ve gone through this exercise a couple times. In my current set up I installed a couple years ago, I learned my nighttime minimal usage is about 130 W and a daily maximum is around 8000 W. Not hard to figure out what’s going on in between. These days I only fire up the pretty graphical interface every few months to see if the forgotten hardware is still running in the basement.
Home power monitoring is like a “hello world” program or blink an LED project. Fun the first few times but, meh afterwards.
But if you are not running on a time based price plan – as is the case if you have your own solar array this sort of thing is actually very useful so you can match your load to your own supply better (and thus pay much less). Also helps you catch when things are left on or going wrong – you know your baseline is around 130w, so if your freezer starts going wrong – perhaps as simple as its setting dial has been bumped having that easy visual display that says 200w lets you know something odd is happening which you can then go hunting for…
Some devices just have to go on when they are needed, but the dishwasher, washing machine are easy to shift the timing of. Cooking can be – we make stuff that keeps with the produce of our pear tree and berry bushes that is very energy intensive, but if you have such a monitor you can look at how much spare you have right now at a glance, perhaps decide that tomorrow is the better day for it as the forecast is sunnier, and even turn stuff that is usually on off for a bit if you are running out – like my computer if I’m going to be cooking for hours they can probably be off for that time and its no hardship, unless they are stuck processing something…
You are assuming all households can fall back on grid connected AC power.
When you are off grid, you know exactly how much power is needed for any given appliance, and will not use appliances like an electric kettle or dishwasher when it is cloudy, overcast or at night, if the battery bank is to last a reasonable amount of time.
We average around 2.9kWHrs a day of electricity use, helped a lot by demand management, passive solar design and good insulation.
A visual indicator of available power from sources other than one’s battery bank is indeed a useful thing.
Neat project! My only problem is that my electric cooperative recently replaced my SDR-friendly meter with a newer frequency hopper, and I haven’t yet taken the time to try to sniff it. I know [hash] has been doing incredible work decoding them, I just haven’t applied any effort to receive them yet.
Nice post, thanks
Been checking domestic power consumption on all major appliances at first then all the little stuff, can put up with taking out 25W LEDs for 3W units which have a slow start, get used to it quickly, saved about $35 AUD or so each month.
Having a real time display unit helpful as I do power systems testing too & exploring industrial oriented products, would be nice to somehow interleave system stats onto/into a modern (lab/office) TV – maybe a low end projector onto a narrow suitable white screen on top edge above the TV ie have TV on background docos & news etc when working at the bench, cheers
Please be kind and respectful to help make the comments section excellent. (Comment Policy)