That's the thing, they don't really correspond to what I see with my Xbee. I'll look at it some more, maybe the values need some converting.
Thanks
I do not, not yet anyway. Haven't really given it much though.
Ok, then it is clear, I will sort it out. This is plenty clear information.
Yes, some Aqara devices include extra, non-standard, information after the actual model name. My driver cleans that, but can get the cleaning wrong sometimes it looks like. It should have been cleaned to "lumi.sensor_switch.aq2" and not "lumi.sensor_switch". Once I have updated the driver a quick click on the reset button will set the model name correctly.
So something interesting. I have six Tradfri USB Signal Repeaters on my Xiaomi/Aqara Hubitat. And I routinely check them using getRouteTable(). For the last ~8 months, 4 of these had lots of entries in the route table, while the other two had basically empty route tables. This was consistent with the possibility that once Aqara devices pick a zigbee router, they stick to that router.
Today, the route table for all six have lots of entries. Do your drivers "encourage" Aqara devices to not stick to a single zigbee router? If true, this would make them much more stable - because it lets them use different routes to communicate with the coordinator.
Yes, indirectly, my driver sends commands back to all devices when they check in, plus some other commands for some devices that have those commands sent to them by the Aqara hub. It is part of what makes things "snappier" and to keep them online.
Excellent. Thank you. I've just switched my temp/humidity driver. It's been stable, but does have a tendency to fall off the network when the battery approaches 50%. I'll offer results in as we eventually get back to that level. It's at 85% right now.
I have the 6 button Opple on my Aqara hub, so I'll give that another shot on HE today. My first 3am pairing experience with that was an eyeopener when I triggered my alarm setup at the time.
At 50% chances are you have reached the point where voltage will drop very fast. Not sure if it is worth the effort, but maybe it would be worth adding a filter to take into account the actual voltage curve for the battery type. Another day maybe. At least now you will know within 3 hours of this happening.
Stay away from button 1 and 2 (those with the LED) during paring and you should be fine. Once the log reports that the pairing and configuration is done, then they are safe again
They're coming, I want to get them right and not have to make "breaking" changes after release. Basically the only devices I don't own myself of Aqaras Zigbee devices are the blinds, the B1 curtain (but that one I have driver for anyway) and the old wall switch models (I've bought the new D1 model, couldn't get the older models) . I'll probably sort them all out in the end.
Your WXKG11LM model is the one from 2015, I don't have one of those, so I had missed something regarding that one. It will be added soon.
Happy to hear that Always nice to hear when people have a use for the code I've written.
Let's see, maybe that brings over some more devs that can create some more fun stuff.
Yes, that is a nice creation, the first real use we have for the event history!
I had one go crazy last night at 54% using 3.1v max and 2.7v min. One of my exhaust fans turned on because of a -167F, 670% RH report.
14 months on that battery-I'll accept it!
I bought a 6 button Opple a few weeks back and your fine drivers now let me use it. But when I look at it under devices, it reports 30 buttons, so the built-in Button Controller app creates 30 buttons. If I use the first 6, it works fine, but it would be nice to clean up the 24 unused buttons
Ah, Of course. But we have 4 states (push, release, held, doubletapped) and 6 buttons, equals 24? What is then the fifth state? Just trying to understand.
This has been pushed to GitHub. There is now also the option to create virtual devices for use with the contact sensor. For the button devices a momentary switch has been added to the available virtual devices.
Oh this is great. I've been slowly getting rid of my xiaomi window/door contact sensors and replacing with SmartThings Multipurpose sensors as they kept dropping... But this has restored hope in them not requiring repairing at times.. Reason I have some confidence is that my aqara WXKG02LM switches respond heaps faster than the other drivers.. As in almost instant which gives me confidence you've made some good improvements. The xiaomi temp and humidity sensors have been reliable for me even in the old drivers, typically until the power is turned off and sometimes require repairing) but I've changed them over to your drivers as well.
As I said. The fact the xiaomi WXKG02LM works loads faster has given me some good hope in all the drivers. I'll report back once I've used these drivers for longer than 30 minutes...
Fun you noticed 
