I’m replacing a Homey box by Home Assistant (because this box is driving me crazy). So I installed Home Assistant by following the tutorial with Etcher (All-In-One installation on a Raspberry Pi 4). For the ZigBee, I used a second Raspberry Pi with the DeConz platform (using RaspBee II).
Now I want to connect my Z-Wave devices. I only have Fibaro devices (Walli Outlet, Walli Rollershutter and Micromodules).
So I started by directly connecting a Z-Wave key (Aeotec Z-Stick Gen5) that has been recognized (by adding zwave in the configuration.yaml file) but I can’t add my Fibaro devices (either by pressing the key button or by adding a node from the interface).
As a result, I wonder. What’s the best, flawless and proper way to integrate my Z-Wave devices?
By continuing in this way
Going towards the same idea as for the Zigbee, namely to use another Raspberry Pi (Zwave2mqtt?).
The way forward in HA is to use qt-openzwave as a standalone application (either as an add-on if you are running a supervised HA-installaton or as an external container or installation. Then use a mqtt-broker (mosquitto is probably the easiest - also available both as an add-on if you are running supervised, or as a standalone application/container) and the new “OpenZwave (Beta)” integration in HA.
But as the name suggests, this is in “Beta” stage. It works mostly ok for most people already, but there might be bugs or issues you might run into.
This can also be set up with the qt-openzwave install on one box/VM/RPI, Mosquitto on another and HA on a third, if you would like to spread them out.
Please take note of the fact there is an issue with the Aeotec Z-Stick Gen5 on a RPi 4, you have to use a non-powered USB 2.0 hub or a non-RPi 4.
The built-in Z-Wave integration (OZW 1.4) is the most stable at this point, I am unsure if the OZW Beta (1.6) already has support for the Rollershutter. And it’s still in beta
It sounds complex but so far I’ve done pretty well with ZigBee so I’ll try to get by with the configuration you propose. I think I’m going to separate the responsibilities so probably have one pi for Z-Wave and another one for Mosquitto.
The Z-Wave part on Home Assistant is surprisingly much less documented than the ZigBee part (probably because the Z-Wave world is much larger with many different devices). It also seems that using deCONZ makes things much easier for ZigBee.
I use an extension cable for my Aeotec key and it seems to work because it is well recognized on Home Assistant.
Overall, I’m a bit confused about the Z-Wave part. It seems so complicated that it makes me dizzy
It’s been some time since I used the built-in Z-Wave, but it is/used to be under Configuration all the way at the bottom (where you probably found the node list) and then at the bottom of the Z-Wave page. Something like load full or tail X lines.
Let us know what happens with the USB hub (make sure it’s non-powered)
If the failed device still shows in HA (in HA the device would have a ‘is_failed: true’ attribute), you may be able to remove it using HA zwave commands. Otherwise (my nortek did the same thing) you will need to use software called ‘zensys tool’ to access the stick directly using a windows machine. This software will allow you to do a lot of things to a zwave stick, from removing zombie nodes, to bricking the stick altogether. So be careful with it. Still, Zensys is good to have on hand and know how to use for sticky situations just like this.
I’ve seen comments previously to NOT include devices securely, unless needed, such as f.ex. locks.
Now, I have repeaters that can repeat both secure and unsecure. If they are included un-securely, can they repeat packets for devices included securely ?
