I've recently installed some Leviton DG6HD Zigbee dimmers. I use Motion Lighting to control some of them. One particular instance of ML is not working correctly. The logs are indicating the ML app believes the lights are already on, so it does not send the on command. This happens when the lights are in fact not on.
I've factory reset the dimmer and I've built a new ML app. That app then works for a while but then goes back to not working. In examining the app status page, the alreadyOn state is stuck on true.
The dimmer works fine from its page and from other control points including Alexa. The device page is correctly tracking the on/off state.
Below are the ML app pages
Bump. I realized that this is the only ML/Leviton DG6HD Zigbee dimmer combination that I have. As I stated above, the issue appears to be the ML app does not reliably change the "alreadyOn" state in the ML app to false when the lights do turn off. @bobbyD @bravenel @mike.maxwell
Can you show the device events when it is turned off? This is usually a sign that the 'off' event is not getting sent to the app.
Please describe the failure sequence. Is this failing when the light is turned off by ML, or by some other means?
The failure is the ML app does not turn the lights on. The ML app log message is "Not turning on, already on'". As I said, I recreated the ML app and it worked for a while. But eventually the "alreadyOn" state got stuck at "on". I don't know if when the app got stuck the app did not turn off the lights, I guess that is possible.
Events and logs from an on-off cycle from the device page.
Events and logs from an on-off cycle from the physical device.
It appears that changes in level during the ramp are being logged in both cases.
You didn't answer my question: How are the lights being turned off? By ML when motion stops or by something else?
Further to that, why are the lights being turned on immediately after being turned off? What is causing that?
ML won't turn off as I have the "don't turn off if already on" setting set to on. So currently ML does not turn on or turn off the lights.
The lights are off and ML will not turn them on. Here is the ML app log.
The lights are not being turned on after being off. Hubitat appears to be tracking the device level during the ramp. This is newer Hubitat device driver.
OK, I'm totally confused. How about getting rid of the option (for now) as to 'don't turn off if already on'.
Your device logs show this sequence: off (digital) at 11:14:14.166, followed by button 2 pushed at 11:14:14.176 followed by level 72% at 11:14:14.271. So that sequence, when I guess the lights were turned off, results in the lights being on as far as ML is concerned due to the level being set to 72%. ML interprets a level event as turning on the lights, and that would explain why it would say 'not turning on, already on'.
What driver is this?
Yes, changing the "don't turn off if already on" setting set to off gets the ML app back in sync and able to turn the device on and off.
The driver is a newer built-in one introduced in 2.2.4 - Leviton DG6HD
It's a driver problem. We will fix it...
There are a lot of parameters available to the dimmer that are not exposed by the driver (but not a crazy amount). It would be great to see them added. The Leviton end user documentation doesn't even show most of them. But this dimmer is a close cousin of the DZ6HD Z-Wave dimmer. The hardware can be configured in the same ways for both and the Z-Wave documents show the physical taps that are work for both devices. I don't know if the Zigbee dimmer allows those parameters to be set by a driver.
It doesn't, zigbee device options do not work the same way that zwave options do.
Options that are available via physical interaction don't always get implemented in software.
With zigbee devices, manufacturer options are implemented using manufacturer specific clusters of which this device has none.