Hello, I’m helping out a small local nonprofit and they recently got their first Home Assistant instance completely set up. It works perfectly for their IoT devices, but they’d like to also use it for data monitoring. The #1 priority for data monitoring is energy usage. Unfortunately, it seems like that’s not going to be very easy with this meter that the electrical company has installed. I’m going to outline what I’ve found out so far and if anyone has any suggestions or ideas on how we could possibly monitor the meter real-time, I’d greatly appreciate it!
Pulse Reading - There is no visible pulse on this meter.
ANSI C12.18 optical - This one is a bit more interesting. If you look at the meter in the PDF that I linked first, it has what looks to be an optical interface with the magnet built-in. Although it’s in the same model family, my meter looks like the one in the second link I sent. It has a plastic triangle around the sensor. From what I can tell, the standard plastic for these meters only includes the configuration port and not the C12.18 optical port I guess? If you take at page 6 of this PDF, it seems like Landis Gyr sells different plastic configurations, and the one the utility company went with seems to be the “Configuration Port Only” one. I used my phone to check for infrared signals, and I can definitely see a flash about 5 times per second. The flashing seems very consistent. If anyone has any experience with this type of port (or if it’s possible for me to integrate with it), that would be great!
Data from the Utility Company - The data that the utility company gives us lags behind by 2 days. This is great for seeing historical data, but ideally we’d be able to see almost real-time data.
At this point, are there any other options besides CT clamp sensors? If we were to go the CT clamp route, the issue is that the electrical service is 300 amps, so we’d essentially have to do a 150-200 amp clamp for every circuit that breaks off to get a somewhat-accurate reading of our overall usage.
I have a 100A (240V) single phase service. However I have never exceeded 30A in the two years I’ve been monitoring. So even though I have a 100A CT on the main supply I could drop this to a 50A CT for greater accuracy.
Thanks for the response! If I had to guess, they probably don’t ever use more than 120-150 amps. But, the issue is the wire size. As far as I can tell, there’s no way that those 120 amp CT clamps from Shelly would fit around the service wires. I did see the CircuitSetup 200 amp CT clamps, which would probably work since they have a 24mm opening. Do you have any experience with CircuitSetup? I probably would’ve preferred to go with a more well-known company like Shelly, but if CircuitSetup is the only option (and if I can’t get the data from the meter), then I guess that’s what we’ll do.
No I don’t have any experience with them sorry. I use an IotaWatt. I used to use a Shelly EM but found it to be inaccurate. It would report negative power on one CT when the other was measuring a load. Not as bad as my SMA energy meter though that thing was next to useless.
Iotawatt supports a wide range of CTs and a generic CT that you can supply specifications for.
Though the developer is no longer producing them the design is open source and third party’s are manufacturing them. He stopped manufacturing to work on development of a new ESP32 based unit.
Thanks for the suggestion. I did look at that before, but unfortunately I don’t think it’ll work. It specifically says “SML, DLMS, M-Bus and ANSI C12.18/19 standards are not supported by the software.” If our meter even supports an optical port reading, it would be with C12.18 (see the data sheet for the meter).
I guess my main question regarding the optical port is: theoretically, shouldn’t I still be able to read IR/optical signals from the meter even if the plastic cover is the “configuration only” one? It may be a little more difficult to read with the ambient light, but it seems like the optical port itself is still in the meter, even if the plastic cover isn’t meant to support it.
I have this same meter and I’m looking at solutions myself. If you see lights flashing you can probably hook up something to read those flashes since those are supposed to flash every X units of power used, however is that really still flashing for you? I found a PDF of the model and that is supposed to be the calibration port and is supposed to turn off after 24 hours of first getting power:
Maybe that PDF is for a older model?
I dont see my meter flashing but if it’s infrared light I won’t see it with my own eyes, I tried with my phone camera but didn’t see anything, maybe it doesn’t pick it up either for all I know, or it just does stop working after 24 hours
If you see lights flashing you can probably hook up something to read those flashes since those are supposed to flash every X units of power used.
Unfortunately they don’t seem to correspond to usage. They stay very consistent every time I check it.
however is that really still flashing for you? I found a PDF of the model and that is supposed to be the calibration port and is supposed to turn off after 24 hours of first getting power
Yeah I saw that too. I have a feeling the calibration lights are just always on for my utility company’s meters. I’ve checked 2 different meters in different locations (same utility company) and I can see the IR leds blinking (with my phone’s camera) and the power hasn’t gone out in a while. At this point, I assume the flashes are for calibration, but I wasn’t sure if maybe they were optical signals that just “seemed” consistent.
Unfortunately, at this point, it seems like there just isn’t a way to monitor these meters real-time. Maybe someone will make a discovery regarding the RF/ZigBee signals, but that’s probably the only promising thing at this point.
I just found this PDF and saw some interesting information in it. From page 47:
Whenever the meter is not communicating, the LED will pulse at a programmable rate proportional to the watt-hours flowing through the meter.
This PDF also mentions that the LED should stop flashing after 24 hours. If that’s the case, either I’m seeing some configuration data from the optical port or my utility company disabled that 24 hour timeout.
I’ll have to do some more testing in regards to this calibration LED. The pulses seemed consistent to me, but it’s possible that the loads were just consistent whenever I checked.
Can confirm that the flashing IR led doesn’t seem to correspond to usage (at least based on number of pulses). I checked 2 different meters in different locations (same utility company) and tried different loads on both meters. No matter what I did, there were always ~4 flashes per second (filmed at 24 fps). This leads me to assume assume it’s some kind of optical signal.
