Blue Series Binding Problem -- WAF emergency!

I’ve installed 4 pairs of Blue Series (VZM31-SN) dimmer/switches, each pair in a 3-way configuration to control a corresponding dumb LED light. Binding doesn’t work properly in that the pairs lose sync.

Example: Initially, Load Dimmer (LD) is OFF and Non-Load Dimmer (NLD) is OFF. LD is turned ON manually. NLD does not show that it is also ON (which it should via binding sync). Manually turnng NLD ON does nothing (except change the indicator LED bar to appear to be ON). Turning NLD to OFF does nothing (the controlled dumb lights stay ON). Toggling NLD ON to OFF to ON once or sometimes twice more causes NLD to synce with LD and function properly – for awhile. Leave things alone for a some time (exact time not yet determined, but is likely over 1/2 hr) and the binding is “forgotten”, and NLD takes some toggling to get back into sync with LD.

Configuration:

  • Hubitat C-8 Pro with 2.3.9.197 firmware (latest generally available).
  • Driver: Inovelli Dimmer 2-in-1 Blue Series VZM31-SN
  • The Non-Load Dimmers are connected to line, neutral, and ground. The existing traveler wire is not used at all.
  • The Load Dimmers are connected to load, line, neutral, and ground.
  • App used for binding: “Zigbee Bindings”.

Here are the results in the Apps tab after using (and renaming) Zigbee Bindings 8 times (“Main” means load-connected dimmer, “Remote” means non-load dimmer):

  • Switch Bindings - Entryway Main to Remote
  • Switch Bindings - Entryway Remote to Main
  • Switch Bindings - Hallway Main to Remote
  • Switch Bindings - Hallway Remote to Main
  • Switch Bindings - Laundry Room Main to Remote
  • Switch Bindings - Laundry Room Remote to Main
  • Switch Bindings - Stairway Lights Main to Remote
  • Switch Bindings - Stairway Lights Remote to Main

In the App, here’s what the “Main to Remote” settings look like for each load-connected dimmer:

image

Here’s what the “Remote to Main” settings look like for each non-load dimmer:

image

NOTE: ALL of the dimmers show “numberOfBindings : 0”. Yet sometimes the pairs are bound.

So, what am I doing wrong??

Is there a reason you’re using binding instead of the travelers?

He is using pairs of 2-1 dimmer switches. You can’t use a traveler conductor in that configuration. The only solution is to bind. If you use the traveler, you would need to use an Aux instead of a 2-1 at the non-load location.

1 Like

My bad. I thought you could use multiple 2 - 1 with travelers.

So, further testing shows that in my setup,

  • “numberOfBindings : 0” never changes, even though 4 different binding methods were used (the manual process provided in the Inovelli driver, using the “Zigbee Bindings” app recommended by Inovelli, the “Zigbee Binding Tool” app (better UI than the “Zigbee Bindings” app), and the “Switch Bindings” app (also better UI than the “Zigbee Bindings” app AND has a polling feature described below).

  • The pairs of dimmers ARE bound, since they control each other (once sync is re-established if it has failed, by toggling the paddle up and down 1-4 times) even when the Hubitat hub is shutdown.

However, the non-load dimmers ALWAYS lose sync with the load dimmers after a few minutes (typically less than an hour, but usually more than about 5 minutes).

So I STRONGLY suspect a bug in the Inovelli firmware that times out – for whatever reason – and stops polling the state of each pair.

I did find a workaround: the “Switch Bindings” app has an Advanced Setting that allows a “master” switch to be selected (e.g., the load dimmer) and then the app periodically polls that device and synchronizes the bound devices (just a single non-load dimmer for each of my pairs set up in a 3-way light switch configuration). The default polling period is 5 minutes.

This is still not a perfect solution since the pairs can be out of sync for up to 5 minutes at a time (I could shorten the time but that increases the load on the hub). The out-of-sync issue is visually manifested in different LED states for the paired switches (a bit annoying) and is physically manifested in that the non-load dimmer must be toggled 1-4 times to restore sync (very annoying).

Hoping that someone from Inovelli can replicate the problem and provide a fix! But the workaround is decent enough that at least I don’t have to return 8 dimmers!