Thanks for all the updates. Ironically, my Aqara sensor hasn't fallen off since my last post here. I guess my threatening it worked 
I'm still on lucky channel 13. Have been since the beginning, and my test hub is on channel 20. Interesting note about the preferred channels for Konke. Same boat as you @aaiyar. Way too invested in my current channel 13 to rock that boat. It's been working so far, so I wouldn't want to change it for one sensor.
It would be somewhat difficult for them to design it to only do 15, 20, 23 - and very little return for that work. So I'm skeptical.
But I guess we'll see when mine get here!
Pertinent to your question that made me try my "Peanut" Hubitat, according to getChildAndRouteInfo, Konke Temp/Humidity sensors can route through Peanut plugs.
Edit - also, without your having urged to try both Hubitats, I would have only tried the Tradfri Hubitat, gotten frustrated, and tossed them in the garage catch-all junk box. So thank you!
Funnily I tried the Peanut Hubitat first; they pair so quickly and in place. I tried all kind of crap with the Tradfri Hubitat. Right next to the Hubitat. Reboot the Hubitat. Next to a repeater - nothing worked, so I googled .....
So odd, since the Trådfri are 3.0 I would have thought they'd be fast friends.
That GitHub link I found has three people on it stating they had to change to a supported channel (15, 20, or 25) to get these things to work. I'll bet if my Tradfri Hub was on Channel 15, instead of 23, it would have worked.
The real test of the claims on GitHub will be when @JasonJoel gets his Konkes.
I would be too. Except I found two more links indicating that Konke zigbee 3.0 devices only work on channels 15, 20, an 25.
Here's information about the Konke zigbee 3.0 motion sensor from the zigbee2mqtt developer.
And then there's a post from Robert (@bertabcd1234) on the Konke zigbee 3.0 motion sensor on community.hubitat.com reiterating the same information.
Yup, we'll see. That will be disappointing if true, but that's life. I knew there were risks ordering Chinese devices sight unseen.
And not like it was a major investment or something. If you end up not using yours, and mine stay connected, I'd be interested in buying a couple off you.
The temperature readings are very accurate. Don't know about RH yet.
Edit - @SmartHomePrimer this might interest you:
CR-1, CR-2, MBR-1, MBR-2 are four sensors downstairs in conditioned space.
LR1, LR2, LR3 are all upstairs in conditioned space.
GBR-1 is upstairs, but the damper is 75% closed because there are no guests visiting.
The ecobee is set at 70F using its sensors.
BTW, sensors aren't next to each other - they're mounted on different walls.
Thanks. It is RH that I’m interested in. I have temp sensors all over and don’t use those values from any of them. My need is modest: A sensor that measures RH, communicates with HE, stays connected, and preferably is small.
The RH readings look ok, but I don't know if they are accurate. I will find out by Friday. I've got a psychrometer/thermometer/hygrometer but it's loaned out to a friend.
The battery readings are off. Only two of them are reporting battery levels. One is at 67% battery and the other is at -67% battery.
I don’t worry about battery readings. They are never that accurate. It’s the reason I spent over $100 on a tester. 
Yeah - but a range of +67% to -67% is atrocious 
It’s stuff like that. One of my LeakSmart sensors reads a plus value, and the other reads -215%. Same firmware, same driver. It’s just nonsense.
Well, I would think that is more likely than not a DRIVER problem, and not a device problem... But I guess anything is possible.
What is the event saying the raw volts are? If something crazy negative, yeah that would be weird and could be a device issue.
If it were a driver problem, then you’d expect either the same results on each device, which are identical, or you would expect the radically different results wouldn’t be consistent off between devices, but they are. The driver probably is to blame actually, I don’t disagree, but the compensation value doesn’t work because the value jumps all over the place within the same sensor. Both do that, but at different ends of the scale.
Those sensors have built-in monitoring and beep when low. Good enough for me. I know roughly their battery service length and I test leak sensors once a month or so.
I am travelling today so had some time to Google. And you're right - every reference says channels 15 20 25 only.
Oh well, I'll try it on my current 17 and 18 anyway. If it doesn't work, maybe I'll move one to 15. We'll see.
Genuine wtf design decision.
It kinda makes sense to me. Konke's hub handles both Zigbee and WiFi protocols. Maybe they are trying to avoid interference with the most popular 2.4 channels within in their own ecosystem.
Makes sense. That's a logical explanation. If I were doing it, I would restrict their hub to those three channels. So the devices could use the entire range of possible channels (with other hubs).
I also understand why they would recommend those channels. But making devices ONLY work on them is silly.
Anyway, as I'm a glutton for punishment, if/when it doesn't work on 17 or 18, I may try 25 again (and take better notes on which of my devices don't work on 25).
Download the Hubitat app
