I have two of these. They were working for a while. The other day I decided to change the batteries. Once I did this, I couldn’t get them to work again. So, I excluded them, did a factory reset. I had a terrible time getting them included again. Most of the time, they seemed to get stuck “interviewing”. I read that maybe this is because they don’t stay awake long enough to complete. But, I didn’t figure out how to wake them up again.
I finally managed to get them included but it was a huge pain in the butt.
Is there a better option? Is the zooz just not good or is there some compatibility issue with HA?
I have 7 of these sensors and find that they work really well. I’ve changed all their batteries in 2.5 years without issue, and test them every few months to make sure they’re working. The manual says how to wake them (four quick button presses) and that they stay awake for one minute, which should be more than enough time to complete an interview. If it’s taking longer, it suggests there may be a problem with the network mesh/signal, which could have gone unnoticed until now.
If you have a repeater nearby, do a network heal on it to make sure it has a short path to the controller. If you don’t, maybe consider getting one. Also keep an eye on the “last updated” time of your node_status entities, they should never be “asleep” longer than 12 hours (the default wake up time). I setup an automation to alert me if a sensor hasn’t checked in for >24 hours (usually because the battery has died); if the battery isn’t low it could indicate network issues.
If you still want to get new sensors, I found this video helpful:
Thanks so much for the detailed response. I kicked off a “rebuild network routes”. That should clean help if paths are not optimal, right? As for a repeater, I don’t have one. But, I have several z-wave switches in a room that’s only 20 ft from where the sensors are. This shouldn’t require me to have a repeater, should it?
Also, could you share the yaml for the automation you did. That sounds like a good idea.
Update: I excluded and re-included them again after the rebuild. It went much more smoothly this time. Perhaps I did have some network issues.
Glad it’s working better. Any Z-wave device with a hardwired power connection acts as a repeater, so you don’t need a dedicated device if another is within range. Note you also have the option of healing/rebuilding individual nodes by clicking the 3-dot menu beside the CONFIGURE
button on their device page (also note sleeping nodes won’t heal until they wake up and receive the command).
Here is a basic yaml for notification when a device stops waking up and checking in:
alias: Notify on late Z-Wave wake-ups
trigger:
- platform: state
entity_id:
- sensor.watersensor1_node_status
- sensor.watersensor2_node_status
# etc, or maybe a template of battery *_node_status entities?
for:
hours: 25
action:
- service: notify.mobile_app_iphone
data:
title: Z-Wave Node Warning
message: "{{ trigger.entity_id }} no status change for past day"
mode: parallel
max: 10
Thanks very much!