Z-Wave reliability?

[arjanm83]

My current z-wave setup is not as reliable as I hoped for when choosing for z-wave.

4 of the highlighted nodes in the attached screenshot are Fibaro Dimmer 2 modules.
All of these are located within 2 meters of the Aeon Labs z-stick gen 5, but there are thin walls in between.

The transmitter is located in the cupboard with the electrical meter and the gas / water meter.
The dimmer modules are located in the hall, toilet, kitchen, kitchen table.
What happens a lot is for example that when people go to the toilet, the light doesn't turn on immediately but after 5 seconds, and the log shows a timeout.

The topology shows all of these devices to be directly connected. Polling is disabled for all devices.

Looking at these statistics it shows that roughly 5-10% of all actions towards these dimmer modules result in a failure or retry.
My question is, is this normal for z-wave in general? Or are these Fibaro Dimmer 2 modules all of bad quality? Is it the thin walls? Can I do anything to improve this? I hope someone can shine more light on this.

zwave.JPG (149.53 KiB) Viewed 1046 times

[Egregius]

Do you have polling enabled? If yes, try disable it.
I only see higher retries on 2 nodes where I have polling enabled. All other nodes have retries < 5.

[arjanm83]

Do you have polling enabled? If yes, try disable it.
I only see higher retries on 2 nodes where I have polling enabled. All other nodes have retries < 5.

No, I don't have polling enabled. Added this information to the first post.

[Domosapiens]

arjanm83,
imagine that you are z-wave transceiver and you have to live in a cupboard.
Surrounded by all type of metal stuff ... you would become crazy of all the receiving echo's and
every time you transmit... again echo's.
Your router is in the same cupboard? Harmonics of other transmitters interfere also.
Move out of the way of all metal reflection and other transmitting sources.

Multipath causes multipath interference including constructive and destructive interference, and phase shifting of the signal:




Multipath is often a bigger problem then range!
The figure above happens every 17 cm (=related to the wave length of Z-Wave).

Try to put our Z-Wave stick on a short USB cable.
Try within a range of 17cm to find an optimum.