Driver Adb Xiaomi
Magnhild Mongolo
Mate, literally my comment above yours.
Install the drivers from the first post. It is pretty easy. Just click the link and follow the instructions.
In the app, Add a new device, then scan. Then follow what I said in the previous post.
Here's my take on some drivers for a select few battery powered Xiaomi Aqara devices. There are a number of other excellent drivers out there with more complete support of the range, but these are for the devices I own and now use every day.
There's a little backstory below, but due to the faff I've had with Xiaomi products I can only recommend using Xiaomi devices on Hubitat via Zigbee2MQTT. I had partial success for a long time with them directly connected, but they were still troublesome. They are flawless using Zigbee2MQTT.
Search for the keyword "Xiaomi" on Hubitat Package Manager and you should see "Xiaomi Drivers from BirdsLikeWires". Requires HPM v1.8.7 or later. Or install manually from the links below.
Note!
If you have more than one driver installed for a particular device and pair it for the first time, there will be a "race condition" and whichever driver "wins" will perform the initial configuration.
I've encountered so many issues getting them to work reliably that keeping them on their own separate network and making them available using Hub Mesh is the only directly-connected method I can recommend. My setup was very simple:
I found that even some of Xiaomi's own wired Smart Wall Switches (namely models QBKG22LM and QBKG25LM) will cause mesh instability. Battery devices will become unreliable and then drop completely within a few hours, so you will never find drivers for those here. Instead I use very reliable Samotech modules fitted behind the Wireless Remote Switches.
The drivers do nothing special to try and keep devices alive and connected. The mesh is either good or it's not and devices should check in every 50-60 minutes with battery readings. These reporting intervals are a firmware default and cannot be changed.
Some of the devices send a unique transmission (sometimes with battery data) on a short press of the reset button. Where this could be useful it is mapped to a button press and referred to here as the "reset button" feature, to give an extra means of control for "free".
Note: Battery reports only come through every hour, so don't be surprised if you don't see that value appear when you've first paired a device. Give it a few hours and if it's paired correctly, it'll be there.
Provides temperature, relative humidity, calculated absolute humidity, pressure, pressure direction (rising or falling), reset button, device presence and battery reporting. Tapping reset is mapped as button one.
Supports push, double push, triple push, quadruple push, quintuple push, reset button, device presence and battery reporting. Tapping reset is the same transmission as the quintuple push, so is also reported as button five.
Supports push, double push, hold, release, reset button, acceleration (shake), level (calculated from duration held), device presence and battery reporting. Tapping reset is again reported as button five, while acceleration (shake) is reported as button six.
Supports push, double push, hold, long hold, device presence and battery reporting. Left key is button one, right key is button two, both keys together make button three, with each button supporting all four modes of press:
The "long hold" feature makes use of a message received from the device if a button is still being held down three seconds after the "held" transmission has been sent. A "hold" is reported after two seconds, therefore this "long hold" message is triggered after five seconds.
Given that there is no genuine "released" message supported by the device it would seem the best way to capture this functionality is to consider it an "autorelease" of the button. Therefore a long hold is reported as the held button being released.
I've been using "Xiaomi Aqara Mijia Sensors and Switches" which has done well - I only currently have one Aquara device (a temp/humidity sensor). I switched over to your driver and really like the results - a lot cleaner! Ty.
I recently installed a small he setup for a friend - installed was the Model WSDCGQ11LM manufactured by Aqara - I thought it was identical to the one I have here in my home. Her's in newer by about 3 months.
They came in what look like the same box - we paid the same price, I even forwarded the URL from Amazon - What could be different?
Untitled-11523720 111 KB
no! and that's the rub. I recently saw the post from @birdslikewires about his driver release and jumped on it. My HE and device took to it like a duck in water. The big thing was it reduced an error on 'battery' which was causing my personal battery management app angst.
I remoted my friends location, and applied the same driver to her's. and the result in the screen above describes the difference I see. The only difference I'm aware of is the age of purchase... a difference of 60-90 days. I think I previously mentioned, same price, same box... just ... not the same!
Hey @jshimota, looks like we all have the same device, though you've got some extras in the Device Details section which would have been added by other drivers. They'll make no difference to the one here, though.
It can take up to an hour before battery reports are received, but if you want to speed things up just short-press the reset button (that should generate a "pushed" message) and ask your friend to breathe into the sensor. The button press may send battery data and the increased temperature and humidity should generate the other messages.
hmm. I'm ultra cheap so this was a 'smart' buy. I've got to go onsite and rebuild her door lock and add a few things (she's so happy with the automation she wants an Alexa in the garage so she can listen to music while doing laundry), an add of a contact sensor to the garage door, etc.... She just found out her washer dryer are 'smart'. So we'll also be exploring that...
@birdslikewires Will do! excellent recommendations to cause a report from the device. She's at work, so I can't try anything for a few hours, which will in turn give the device time to report possibly.
Nice eyeball on the extra values! I was using Markus' ohlalalabs driver previously and didn't clear states/attributes - just switched driver and mashed config /refresh... The new driver is less cumbersome, and seems quicker - but thats just subjective opinion on my part. We have 2 more of these devices to install so it will be very interesting! (going to use them to control the bath fan for humidity). Thank you for returning my call!
Shu
Added support for the single-key Wireless Remote Switch WXKG06LM. The WXKG06LM (one key) and WXKG07LM (two keys) are largely identical in message format, so they've been integrated into the same driver.
Also added "autorelease" to the WXKG06LM / WXKG07LM driver. This is essentially "long hold" support, using a message which is always transmitted three seconds after a "hold" is detected. Seeing as these devices don't support real "released" messages, we use that "released" attribute to essentially report a long hold instead.
I usually use "hold" events to trigger a local group of devices to switch off. I'm now using this "long hold" to turn off related devices elsewhere in the house. Just remember that a "long hold / autorelease" always occurs in addition to the prior "hold" event.
Before I start down the path of pairing this device with a driver can someone confirm if I will be able to have an immediate changes in lux reported (specifically - over a certain threshold and even better if it could also require a certain number of milliseconds over that threshold).
Use Case: Think flashlight or headlights shining through a window into a pitch black space (and yeah, the "duration of the luminescence being above the threshold" was how I was thinking I'd thwart triggering during a lightening storm).
Same experience as @jlv. I have at least 4 of these. They are very responsive. Please look at the log file below - it indicates increases in a Lutron Caseta dimmer and the corresponding lux increases detected by this sensor. The starting lux value for the sensor was 34 (when the dimmer was off).
Thanks guys. Appreciate the input. Got one on reading some of your earlier dialogues elsewhere. Just wanting to make sure before I get knee deep in making it work that there isn't something I'll be banging my head against w/o realizing.
I have been using @ygerlovin edge drivers for Aquara contact sensors for a while without problems. So I ordered a bunch more and installed them around the house. However, they seem to malfunction with sliding doors. For some of them I get 3 readings for each open or close event. Since Smartthings truncates the timestamp for integrations, my Sharptools.io dashboard is getting messed up since it does not know which one is the latest reading and whether it is an open or close.
This device is not listed as yet on Amazon. Once it is available I will provide a product link. I also plan to do a video on it features. I has a loud beeping sound that is triggered by vibration. The beepin (three sets of three successive beeps), can be disabled via a circuit board switch. It also has two toggle (0/1) switches that can be used to set the vibration sensitivity levels:
It might be worth trying a new battery, when I first got it ir was fine, then I had to change the device handler (cant remember why) and it never worked right again,
It might be as simple as a low battery.
If you could allow a check against the Current State lastCheckin as and alternative to Last Activity that could be solved since this represents basically the same thing as Last Activity (in other drivers) but with some enhancements like Recovery etc.
You can see on the events for my Refrigerator Door that it received open, but not closed. On picture 2 there is a red line. That is when I realized that the door sensor was still listed as open. So I opened the door and closed it again to the get the closed status.
93ddb68554