Thanks. That’s good to know the meter reader doesn’t care. My meter is not near any power and solar is challenging too, so I will need to try a battery pack and see how it goes. Even the Zigbee range might be too difficult.
The AA size 3.2V Feo4 rechargeable battery that I described in the initial post was recharged total of two times in more than an year. The Feo4 chemistry is stable in cold conditions for me is more than acceptable. As for the gas company they replaced the meeter and did not say anything about my sensor I had to retape the sensor in to position after.
2 Likes
Can anyone with the Landis+Gyr 750 tell me exactly where the magnet in the meter is?
Been using this for a couple months with no issues. I’d like to “calibrate” the number HA shows as I have sometimes used the gas with my HA instance powered off. How can I do that?
Go to developer tools in HA and use “Utility Meter: Calibrate” service.
If you scroll back up to my pics #87 I poke my reed sensore up through the hole in my 3d printed bracket and then bend it down to sit in the small 3d printed channel.
From that and the pics when it’s mounted you can extrapolate where it sits against the underside of my 750. Where it is exactly? It has to be close to right above that little alcove where my bracket sits (under the last two red wheels but I’m not 100% sure the magnet is in those wheels. It maybe another larger wheel that’s perpendicular and behind to the red wheels for all I know.
Hi, did anyone manage to get readings from this meter?
I also have it installed in Portugal, with no cable on the right hand side
Thanks,
António
A friend printed your bracket for me (thanks for the design!) and I have it working now. I found that the reed switch needed to be as far left as possible, but also raised from the bottom of the channel else it did not work. I just wrapped the cable around a bit inside the channel but I may go back and add some tape to pad it instead. Perhaps your reed switch is more sensitive than mine (Jaycar SM1002).
I have the same BK-G4M meter and attempted a similar setup but it seems like my D-Z73 reed switch isn’t sensitive enough to be triggered by the strength of magnetic field, or I haven’t found the correct location.
From playing around with the android sensors app on my phone, I’m seeing ~600microtesla be emitted from the meter on rotation, but d-z73 reed switch seems to need ~2500 microtesla before it will close.
Unsure if perhaps there’s something mis-aligned or broken about my meter that makes the field strength so weak or if I need a more sensitive reed switch.
Has anyone else attempted to measure the strength of magnetic field from their meter?
HEllo Mpetcu,
Could you please so kind to recommned any reed sensor that can be purchased and compatible with your solution?
Thanks for your support in advance!
I actually followed some advice elsewhere online and used my phone magnetometer sensors on and around the big blank bottom portion of the gas meter to detect the internal components moving around and had more success! Sobi went and swapped my reed sensor for a hall effect sensor (cat ZD1902) and some leads/wires. The next step is to figure out some code on the esp32 / arduino to read the sensor x y z values and pick values on the sine curves to convert to gas usage increments…
@djgrazzy firstly thanks for all the good work.
I have this all wired up and the reed switch works away from the meter however I can’t get it to register when in the case. Did the reed switch need to go into a particular end of the 3d case as there is plenty of room for movement. I had assumed as close to the last dial as possible but may be wrong. Any advice on this? Also has this stood the test of time and still running?
Just letting everyone know that this sonoff door sensor works. I used the reed sensor that was on there, just desoldered it and run a small cable.
https://itead.cc/product/sonoff-zigbee-door-window-sensor-snzb-04p/
For me it works without the need to unsolder anything. Just attached the board of a Sonoff SNZB-04 to the front of the gas meter using an elastic band.
Thank you for the great idea! It helped a lot. I was not able to figure out how to monitor the Gas usage for months and now it works. I went ahead with implementing it the original way, with exception that i was able to fit the sensor underneath after getting it out of the case and 3d printed a case that worked well for 10 days…
The problem was the battery. Out here at the moment is around -2 degrees celsius and with this battery I was able to send 5-6 000 signals before it died. Therefore, I had to take things in my hands 
I took two 1.5V D2 batteries with 16 500 mAh, printed two holders and a box for them, thus creating a battery that would be 150 times bigger than the original CR1632 :D.
Hope this one lasts for at least 2-3 years before recharge is needed.
Aqara door sensor takes almost 70k signals at one battery (700m3 of gas). Placed outside of house.
do you get the consumption correct with this method ?
Yes it’s reliable
Hi,
I just wanted to share This project It is a full Zigbee gas meter counter DIY, open source I’ve been working in the last weeks.
I started also modifying a door sensor for the same purpose but in the end I believe the full zigbee counter based on the seMetering cluster is a preferable solution.
Hope this information helps others.
Regards
Ignacio
1 Like