I have been PM’ing with @bravenel and @chuck.schwer regarding this issue so I thought I would summarize additional testing that I did this morning for others to be aware of and maybe validate.
Its worth noting my house is two story and U shaped and my network closet is at the bottom of the U under the stairs on the first level. This is where my Hubitat hub, ST hub, and wireless router are located. All Hubitat devices are on the first level only as I haven’t migrated anything upstairs yet due to the physical event problem.
Bruce suggested turning off the ST hub to see if it may be causing interference so last night I shut down the SmartThings hub before I went to bed, left Hubitat in the same place and ran a zwave repair.
This morning nothing was different than before so the ST hub isn’t causing interference.
I moved Hubitat into my master bathroom, which is the center most point to the master suite (on the right size of the U) and where most of my Hubitat devices are located. I re-ran zwave repair.
All On/Off switches in the master suite now report physical on and off events
The GE Jasco 12724-2 Zwave dimmers reports physical on events but not off events.
GE Fan controls report both on and off physical events
Aeon minimote that I previously had issues with now works.
With hub location change:
Devices further away from hub that were previously reporting physical on and off events no longer do so. These are devices in center of house and on other side of the U in my kitchen.
These devices can still be controlled by digital events via web interface. No lag turning them on either.
To summarize something is definitely up with distance from hub and physical event reporting. I have GE switches of different generations that all can report physical events if close to the hub: old Iris clamshell packaging, green boxed v1/2, and zwave plus fan controllers.
Don’t know whether it helps but every time I have attempted to add the GE Z-Wave summer it detects only as device and I have to go into settings to change it to generic Z-Wave dimmer.
Hopefully this can be fixed though; older model GE devices could be a hardware limitation of GE switch and not Hubitat. Still waiting on a response from Hubitat on my findings.
Yes I reported this to support and the finger print has been added but will be fixed in a future release.
My Ge switches gave issues with physical in ST but something changed along the way and they started working with maybe a second delay on reporting (sometimes there was that delay sometimes not) so I hope this can be resolved
I just paired a GE Z-Wave paddle dimmer (12724) to Hubitat; also discovered as device. Not only are physical events not logged, no on/off action from the UI is logged either and no current state is displayed by the Hubitat UI (even when ‘Generic Z-Wave Dimmer’ is chosen for driver). Is this what everyone else is seeing with this device? Just wondering if it is worthwhile trying to redo the pairing process or just wait until this device is properly supported.
FWIW, when this device is operated from the SmartThings app, the event log shows:
2018-02-23 2:16:25.042 PM EST
moments ago DEVICE switch on Bedroom Lights switch is on true
2018-02-23 2:16:20.606 PM EST
moments ago COMMAND on on command was sent to Bedroom Lights true
The log shows about 5 sec. elapse from the app issued command to ST logging the new state of the device. That’s about what I see when looking for changes in the app to be reflected by my physical press of the switch.
Digital events are logged for me when I turn it on/off via web browser. If my hub is close then physical on events are logged. Sounds like you may have another issue happening if your digital events are missing.
Ah… things changed when I rebooted the Hubitat. i noticed that the UI wasn’t updating the current state for my other devices. Now it does show on/off under current states for the GE Z-Wave switch and digital events are logged.
If you find events not showing up, try opening a console (f12) and see if the browser is blocking a connection or throwing errors. This could be helpful to support.
I’m seeing one of my GE Z-wave switches not registering or logging On/Off events that physically happen. I can control the switch in the devices page for the device. The page/log will update the switches status to reflect the physical changes immediately if I press the Refresh button on the devices page for the device…
For example, I am glueing 2 switches together with Rule Machine. If switch 1 turns on, turn switch 2 on. If switch 1 turns off, turn switch 2 off and vice versa. They are not consistently doing this with physical switch changes and when they don’t, a device change is not logged or reflected in the devices page. If I go to Devices, select the device and then press refresh it wakes up, updates the devices status to the correct status (on/off) and runs the rule it missed making the other switch match the on/off status.
Are these the older model GE Z-Wave? The first version didn’t have the ability to report instant change events due to patents and not licensing it. Manually refreshing the device will report back current state.
What others have done is excessively poll these devices for status. This isn’t ideal but we are exploring options to get status updates faster. Obviously we can get it on refresh, but repeatedly sending refresh isn’t ideal either.
Patrick please see my post above of additional troubleshooting. The older GE devices do report status if the hub is close. These were the Iris labeled GE switches in clamshell packaging.
What I find strange is my older batch of GE Z-Wave Switches (ones with larger LED) report physical status events immediately. My newer ones (ones with tiny LED) seem to be the ones not reporting physical events consistently. NONE of mine are Z-Wave Plus.
That looks suspiciously like a line voltage switch installed in a low voltage data-ring. There is a full junction box behind it, right? Not just electrical wires spliced in open wood? O_o
Thanks, yeah. Its odd, but it appears that lack of repeaters or potentially older repeaters might be blocking something as well.
The more info we can gather the more we will be able to sort out and find the cause.
Building solid mesh networks are tricky and sometimes difficult. I have plug in outlets all over my house to strengthen my mesh and its still not perfect.
We will get to the bottom of it, just takes time and more data. Everyone’s mesh and device mix is different combined with environmental differences, there isn’t a perfect model.
And the just add more repeaters doesn’t mean things will always get better.
Just out of curiosity, do you see any difference in reporting after a z-wave repair and about 20 minutes?
