Having a Zigbee hub that I control has me much more inclined to have 'smart' things around the house. Local only, running FOSS, no required mobile app, no cloud service, no vendor lock, yet still using off the shelf consumer devices.

Tonight's project was replacing the can lights in the kitchen with some dimmable Phillips LED fixtures. Swapped out the wall switch with a Zigbee enabled dimmer switch.

@rho nice work. I've had zigbee2mqtt running for a while, but with Shelly BRI going out new no-neutral light switches, I might just put those in and WiFi everything ... :)

@gwmngilfen Ooo Shelly has some really cool stuff, and no neutral would make things a lot simpler.
That's what I love about the MQTT setup. Once the data can make it to the broker you don't really care where it's coming from (wifi, zigbee, 433mhz, lorawan, etc)

@rho yeah I'm using some IKEA bulbs and motion sensors now, but Shelly could replace that. I could use a good no-hassle temp sensor though - the Shelley H&T is great *if* you get the USB base, but a cheap ZigBee battery option would be good. Know of any?

@gwmngilfen I use a couple (Xiaomi) Aqara Zigbee temp sensors around the house. They run on coin cells, and pair nicely to Z2MQTT. I haven't had any issues with them. I think they were in the $20 range.

@rho cool thanks. How often do they check in, or what's the sensitivity if they do it on a threshold? I like responsive heating since I'm using a per-room heating solution

@gwmngilfen Seems to ping once a minute or so. Reports temps in Celsius to the tenths of a degree. They claim an accuracy of +/- 0.3deg.


@gwmngilfen Here's the last nearly 6 hours from that temp sensor. Annoyingly MQTT Explorer doesn't enumerate the time axis, but its pretty clear this thing is up to something. The largest interval is around 25 minutes, and the shortest 1minute, and seems to loosely cluster more frequent transmissions around temp changes.

@rho that would make sense - typically temp sensors wake up when the temp has moved more than a threshold amount, to save battery. The issue with the Shelly H&T is that the minimum precision is 1F, so roughly 0.5C, which is a bit too coarse for me. 0.3 sounds perfect, I'll be looking into those :)

Sign in to participate in the conversation

Fosstodon is an English speaking Mastodon instance that is open to anyone who is interested in technology; particularly free & open source software.