I was thinking the same 
..and although I should be promoting emonhub/emoncms, I was quite interested in @dynamicdave's post above, as the PZEM-004T modules are cheap and therefore scaleable.
Paul,
Do you use emoncms for your main power inlet only, or to measure the major consumers, or ...
Unfortunately I have very few time lately for this kind of stuff 
So I would appreciate if you could explain a bit more what you can do with it.
I assume you only show your energy consumption, but don't use it somehow to schedule devices?
Thanks!!
Hmm, isn't that an in-line tester though?
- AC 80-260V voltage test range, 80-260V working voltage, maximum current can reach to 100A
Can't say I'd want that in-line with my primary mains feed even if it were legal 
Ah, OK, you can get this using the same module but not in-line:
It's not in-line, it's a clamp (split CT transformer) that goes around the live cable.
I've edited my post above to make it clearer - that it clamps around the live wire.
These measure the current rather than the power, which is fine for resistive loads such as a heater, but if you have inductive loads such as motors they don't give an accurate measure of power. In practice, though, most of the load on things like washing machines will be from the heater so the error may well not be significant.
1 Like
I have an inline 2-way meter installed, when my solar array was fittend to the house. Has a modbus.tcp connection and directly works with my EV-Wallbox.
For a more flexible solution, with clamps but 3-phases, I'd recommend this: https://shelly.cloud/products/shelly-3em-smart-home-automation-energy-meter/ ...comes with mqtt enabled, without a chinese cloud connection required.
In addition, most shelly devices can be cross-flashed with tasmota and they are much higher quality than the chinese ones, like sonoffs.
....but that's what all grid meters do as well, don't they? ..and then use ohm's law to produce the power readings from voltage and current measured- For a 3 phase setup, which is common around here, they even use a cast algo/function accross the phases to create a single Energy, not Power, reading.
No, they measure true power, otherwise by using inductors/capacitors you could fiddle the power factor and get free electricity. How Electric Meter Works | ElectronicsBeliever
1 Like
In addition to that, you need to understand how the tariff works and how the grid-meter measures the consumption, like peak-power or peak-energy/power-over-time, which is typically measured in 15mins intervals. For the latter, one would need to consider that this is an average/mean value and this also means, that you are allowed to cross the "hot-line" for a certain amount of time until the meter casts the counts into its tariff-register.as above the line.
2 Likes
I use a emonpi version (which I built), to help me use solar energy efficiently (reducing electricity costs), and using a contrib node that I wrote I am able to process and visualise supply/usage in node-RED. So I have no need for emoncms (which is the visualising software).
I monitor realtime my main supply feed to find out how much power is being consumed by my household appliances, so for example let's assume 300W.
I also monitor realtime the feed from my solar panels to see what is being generated, example 1800W.
In that example it means that I have 1500W spare that I either let flow back to the supply grid (and give it away free), or using a Triac, I can divert 1500W to a water immersion heater, and heat my domestic water. So yes, it is sort of a schedule, but a dynamic one.
It has worked well for many years now, and it supplies our total hot water for about 6 months of the year.
As Julian has said, measurement is key, and once you know your usage & production, it becomes clearer how to maximise efficiency via node-RED calculations, and then control household devices using Shellys or other similar devices.
Have a look at the video in this article - Overview ā OpenEnergyMonitor 0.0.1 documentation for some inspiration!
1 Like
Establish that 'baseline' to be sure, I swear we covered that already... but bears (bares?) repeating as well. 
Solar subsystems have to qualify their internal consumption of power, as a function of the efficiency of the installation. If a given install is radically reporting differences in how it consumes 'power' that will often signal pending component failures, shorts, etc., etc. Some vendors don't itemize this in their reporting to the user, but in their implementation (software) it typically is there. The total efficiently of the system can hide this delta, because vendors want to present the greatest efficiency rating. So when the system states it is 90% efficient don't be surprised if that is true rating less the internal consumption, to baseline what is called 100%. Versus your true rating is 85% because the implementation is consuming 5%. Of course I skipped or ignored various variables, factors, in favor of a simpler example/explanation.
And if you want to monitor your whole house - i would thoroughly recommend the IOTAWatt - upto 13 monitoring channels in a small package - with CT clamps (provide your own or purchase with the unit.
Has a voltage reference (and if you are anally retentive you can use 3 voltage references for 3 phase or just derive the other two) it understand power factor and outputs to emoncms and influxdb and finally has a rest api so can be queried in realtime. Its a fantastic project that Bob has put together and he has about the 2nd best online help forum i have come across (after this one)
Craig
1 Like
I just asked my friends at the local solar installer, how they access the 'true' statistics, I was told I had to buy lunch to get into the details. So, I suspect it is possible, but not easy for a typical customer, for their system design, anyway.
I don't pay peak usage fees but generally those are billed as the highest 15 minutes of usage in a billing cycle. If I had a peak usage fee like that the goal is basically to make sure the sum of loads is minimized in any 15-minute window.
I'd start with the low-hanging fruit. Electric appliances like dryer, stove, air conditioner, heater, water heater. The goal would be to make sure none of these things are running at the same time. Some of those are easy to schedule, some are easy to modify or delay while something else is going on. Some require user training.
The water heater is pretty simple to put on a schedule where it is unlikely to be used while other appliances are on.
The air conditioner, heater and water heater are pretty easy to turn on and off depending on usage of the stove and dryer - when stove or dryer are using power, shut off HVAC and water heater. I would automate this.
The others probably require user training. Don't run the dryer while you are cooking, for example. If you are cooking the HVAC will turn off until you are done...
When talking about peak demand other things like dishwasher, clothes washer and refrigerator use a tiny fraction of appliances with a heating element. Turn the clothes washer to cold/cold and turn the dishwasher sanitary setting off. You are basically turning off the heating element in the dishwasher and reducing the hot water demand of the clothes washer.
If you have some money to spend, replace most of those appliances with gas. Get a modern refrigerator. They run almost constantly while using very low watts.
If you have a LOT of money to spend, get a solar system with battery backup and enable load-shaving. Combined with some of the techniques above you could be the master of peak demand. I like to arc weld occasionally, so I hope they never come here with demand charges!
Jon
1 Like
For the ARC welder... an array of 1,024 hamsters, in spin wheels geared to small generators, wired in serial... should work just fine. I use a few dachshunds myself... you would not believe the number of watts I get per dog treat!
...been there, done that....let me tell you that the WAF dropped below zero after the dachhounds ate the hamsters on their day off...I am now back to electric eels.
I find that replacing the wollen carpets with nylon creates plenty of sparky electrics 
1 Like
I find eels difficult to work with... too slippery for me, never mind the bad hair days. My dachshunds are very well behaved, but since I never implemented the hamster solution, I can't debate the potential for unexpected events? Worst issue I have had so far... A cat from next door did walk into the generator room once, and that really improved the total kilowatt yield for the few minutes it took for the cat the freak out and escape. Fortunately, the next day was quite sunny, so the dachshunds got a rest day... they needed it.
Just today... I got in the mail, that my utility is imposing a new tier 3 cost penalty... to encourage conservation of power... really meaning they do not want to increase capacity, because it costs them more, so they buy power from elsewhere that is also not cheap. Since I live in a very hot desert, during the peak of summer it is not always possible to avoid tier 3 use... this summer I passed into tier 3 only once, and only for a few kilowatts, just barely over the line. Of course, I have quite efficient appliances, house, etc. But this summer was a record setting year. Looks like I maybe doing some serious power analysis coming next summer, and local solar is looking likely in the near future.
