The new one.
Hue Motion sensor could report lux without motion.
does this work as repeater
Don't know for sure, but it should since it's mains powered. I can buy one and verify it with my Xbee, but it would be the end of the month before I'm able to do that.
As other users have explained, both the Xiaomi and Aqara Temperature / Humidity Sensors send reports based on changes rather than based on time-intervals, with one exception (see below).
The change thresholds (for the "original" round Xiaomi model WSDCGQ01LM, anyhow) are any > 0.5C temp or > 6% humidity change, as reported in this zigbee2mqtt thread. However, from what I've seen, a change in those thresholds of either value (or for whatever the threshold requirement is for the atmospheric pressure change) will result in both (or all three) values being reported. All of this makes complete sense to me because it could be a waste of battery to force reporting of unchanged temperatures / humidity levels to follow a time schedule.
The exception to the change-based reporting is the 50-60 minute interval for the check-in message, which is when the battery voltage report is also sent by the sensor. That message also contains current temp / humidity (and for model WSDCGQ11LM, pressure) values, but I've not included code to parse those reports in the Hubitat driver (same for the SmartThings device handlers the Hubitat driver is based on).
If people are interested in having those regularly-timed temp/humidity/pressure events included by the driver, I can certainly add that in, but otherwise, these sensors don't allow any changes to how they determine when to send reports, which is quite different behavior from other brands of ZigBee temp-humidity sensors.
As for using Xioami / Aqara sensors for temperature tracking, I guess I'm not seeing how it would be a bad thing that they only report when there's been a change. If they were reporting on a time interval, and the temperature hadn't changed, there would be no new events anyhow. That would be expected with any temperature sensor because requests by device drivers to generate events with repeat values are normally ignored by the hub.
3 Likes
Thank you for the great explanation. I totally get what you're saying, and it makes perfect sense.
I was struggling with getting a maker webhook to write the data to a spreadsheet, so I never stopped to consider that the sensors didn't work the way I thought.
Personally I wouldn't ask you to change the driver. I just needed to better understand the data I was seeing.
Now I know. It's all working fine.
1 Like
Experience, yes, success, no...
So I got my 3 devices today. Temp, Water and contact.
Got the drives installed. Added them to HE without issues. but they seem to stop responding after about an hour or so.
I have 2 in-front of me (Contact and water) and would test them about every 10 to 15 mins. The temp one was in my freezer. for about 45 and then I took it out to see how quickly it would update the temps changes from being so cold.
first one to stop responding was the temp. Added at 1:39 last report was 2:27 (no update since)
Second was the water sensor. Added at 1:35 last working test was 2:04 (no update when I tried at 2:30)
Last to stop working was the contact. Added at 1:31 Last working 2:44 (nothing since)
I had made the temp hotter - no change.
Water crossed contacts noting
contact moved magnet away and still show closed.
Thoughts. (Even a reboot of HE does nothing to get them back) But if I remove and repair them they work again but for a limited time. (This was test 3 results above)
Do you have ANY other ‘smart’ devices such as outlets or lightbulbs in operation ?
If so they’re probably routing the signals.
The first post of this thread is quite comprehensive in explaining the potential reasons for the dropped connections of Xiaomi / Aqara devices:
From what you're describing, I would point to exploring the last reason of the above list, incompatible Zigbee repeater devices.
Zigbee repeater devices are generally any Zigbee device that is not battery-powered (with the notable exception of some Zigbee smart bulbs, such as Sengled brand smart bulbs).
What Zigbee devices do you have that are plugged into the wall (i.e., mains-powered)?
1 Like
I have no powered repeaters. I have door contacts and some z-wave. I am about 10 ft from my HE and no devices between us. I have moved the he to within 1 foot and I still get the same issue after some time. Just at a loss why it drops after an hour or so.
How many total Zigbee devices do you have? Hubitat can only handle 32 directly connected devices, so if you have more than that, they'll usually fail to pair at all, but I'm not sure what Xiaomi devices would do. In any case, adding repeaters would be the solution. (They have their limits for child devices, too, usually far less, but those children don't count against the hub limit.)
How are you pairing them? I usually press the button for 10 seconds or whatever it is to reset, then press once quickly again after that...and keep pressing every few seconds thereafter, though I'm not sure if that is totally necessary. Pairing in place is also best (staff has said if it doesn't pair in place, it probably won't work in place, either, secure Z-Wave devices like locks excepted).
Beyond that, it's pretty much got to be one of the reasons veeceeoh mentioned above.
How long did it take for them to initialize? I had no problem with my Xiaomi door sensors or motion detectors pairing up and only had one hiccup on initializing.
But I can't get the water sensors to initialize at all. They do pair pretty easily after figuring out that you have to do the reset cycle twice to actually reset the water sensor.
I was only 5 feet from my hub, and only have 4 other zigbee devices -- all xiaomi -- hooked up to my hub. No repeater devices.
So to find out if the newest Sylvania smart+ bulbs are any better with Xiaomi devices I setup a test.
I used a new smart+ bulb (new bulbs have a different zigbee chip indicated by "FCC ID DZO-AETHER") I connected the bulb to my ST hub and then joined the aqara flood sensor through the bulb. Both of these devices are in my old "coal cellar" with 12" concrete walls and foil insulation. The water sensor has stayed connected since 5/9/19.
I know it's not an HE hub, but I couldn't risk trashing my main mesh used for security.
I'm confident enough to start moving things to HE and hope to see the same stability.
Interesting you've had luck with bulbs as repeaters. Proceed with caution. The Trådfri outlets or the XBee are thoroughly tested as repeaters for Xiaomi and do an outstanding job. I simply don't even have to treat Xiaomi devices differently than any other now. They're just stable. Just ordered another Aqara motion sensor and going to add one of the dedicated Trådfri repeaters in my basement tonight to compliment my two Trådfri outlets.
I see the "SmartThings Zigbee Outlet" is on the list of known-incompatible repeaters, but there are several generations of this device. I'm assuming the Centralite 3-series one (similar to the Iris, if not identical on the Zigbee side) is not compatible. However, I have a the "2018" SmartThings Outlet (model GP-U999SJVLDAA) and seem to have paired a Xiaomi Aqara motion sensor through it, which has successfully stayed online for a couple days now. Has anyone else tried this as a repeater?
I am unfortunately unable to verify whether it is still routing through the ST 2018 outlet, as neither of my Xbees with XCTU nor my hub's "getChildAndRouteInfo" page shows this device--no matter how long I wait (or refresh). I did show there right after paring. And I've never seen a Xiaomi device switch routes (I've stated a few times that that all of mine prefer to drop off the network rather than find a new one), so I'm inclined to believe it's still there. (And in related news, if anyone has seen this behavior and has any ideas, let me know.)
So, tl;dr, I can't verify this anymore since it's not showing up in any of my map/route info (nor are many of my devices...), but I paired a Xiaomi Aqara motion sensor through a ST 2018 outlet and it's stayed online for a couple days now. Maybe it's a compatible repeater?
PS - While the Trådfri outlets are cheap and known to work, these are a bit prettier (albeit a bit more costly but discounted if you by mulitple direct from Samsung, an experience in itself but I digress). They also seem to have a stronger signal than the Trådfri outlets (which are reported to be on the weaker side).
I've added the ST 2018 Outlet to the compatible repeater list with a caveat that more reports are needed to confirm compatibility.
Thanks for reporting it!
1 Like
Any further info on the new 2018 Samsung outlet as a compatible repeater for xiaomi? I mean had it since been stable for you?
TIA
So, it was still hard to tell how the Xiaomi sensor was routing since it wasn't showing up in XCTU or Hubitat's child and route into page, but it's been working fine, so I did an experiment: I unplugged the 2018 ST outlet. As I suspected, the Xiaomi device stopped working. It worked again when I plugged the outlet back in (only had it out for a few minutes to test whether it would send an "active" event successfully). To me, this means it almost certainly was routing through the ST outlet (as I know it was when I paired it, the last time I could see it on a list). Of course, could also have been routing through something else that was routing through the outlet, but Hubitat lists all my repeaters as connecting directly to the hub (XCTU does show dashed lines between many), so that seems unlikely.
So: my conclusion is that the 2018 ST outlet works fine with Xiaomi devices. I've only tested it with the Aqara motion sensor and only one ST outlet, and as I said it's been hard to identify its route in Hubitat or XCTU, but I think all the clues I can put together indicate it works for me. More people testing more Xiaomi devices would make me more confident, though. 
1 Like
Several months ago, I set up my half dozen Xiaomi devices (5 'original' round buttons and a Cube) so they were repeating through one of Iharyadi's environment sensors; they have been working perfectly since. I like the device driver for the sensor since it shows the current count of child devices on its settings page. This was a help when I initially paired the devices, as the child count incremented with each successful addition.
Back in early March, I happened to look at the sensor's driver settings and noticed that the child device count had changed from 6 to 3. Strange, since all my Xiaomi's were still online and working perfectly. I dug out my Xstick and mapped the mesh to discover that 3 of the sensors had silently jumped ship from the environment sensor and chose Cree Connected bulbs as their new parents (I use a couple of Crees with circa 2016 firmware in my basement; one of these was parent of two Xiaomi's and another bulb repeated for one).
Out of curiosity I verified today (after another 3 months of trouble free operation) that these Xiaomi's are still repeating through the same Cree bulbs. Stranger still, when the Xiaomi's chose the Crees as their new parent, they did so in preference to several closer Iris V2 outlets (which in my experience normally grab new or orphaned Zigbee devices like a magnet-- but won't play well for long with Xiaomi's. I had to make sure these were powered down when I did my initial pairing).
