@bangali I can confirm no other rules or apps connecting to these lights.
Rule lux set to 30
Fiabro lux before lights are on is 0
Fiabro lux after lights are on is 28
With lights on
Aqara motion 1 gets lux 54
Aqara motion 2 gets lux at 10
Aqara motion 3 gets lux at 24
When lights are off it is (0 + 54 + 10 + 24)/4= 21.25, in the app its saying 21.3, which is right due to rounding. Based on the last reported motion values, in this case I left the room
When lights are on it averaged at 30 because I walked through the door of the bright aqara sensor. But all other times the lux sensors didnāt really deviate.
However when the room went from vacant to engaged, I assume due to high footfall traffic and rooms lux stuck at 51, but not sure why as the lux sensors were (31 + 51 + 10 + 2)/ 4 = 23.5
Also when the room goes back to vacant by itself and lux value goes to 0.
Now it gets strange, during each of these tests I have not had disco lights, but these tests only had 2 motion sensors at the far end of the room. So in went the other 2 and so far again no disco lights.
This made no sense as disco lights are always happening, so I stead of staying still I started to walk around in circles triggering each sensor, I got my disco lights.
When in occupied and engaged mode when any one of the motion sensors went inactive the lights went out.
Running additional tests when in occupied mode and with the motion sensors still active the lights would also go out.
dont see anything incorrect in those settings. the only diff from what i tried is this is using holiday lights which i didnt try. if it still does not work tonight i will try that next.
if at all possible at a couple of mins before 11pm could you open up live logging and please check if anything is posted in the logs from room Outside at around 11pm.
to add to the question @TonyFleisher asked ā¦ if you change the settings to use last motion active instead of last motion inactive do the lights still go on/off?
how about if you change the occupied timeout to 75 seconds instead of 60 seconds?
for these tests leave all the motion sensors and lux sensors in please.
@TonyFleisher, @bangali fibro has a 15 sec blind time the aqara has 60 second blind time.
Occupied time already set to 70 in other tests , shall increase.
Shall change last motion setting.
Made change lights still going out when in occupied mode, ie they come on then back off again and the room stays in occupied status.
In terms of lux values the original theory may be true and itās a logic condition failing. With the lights off the room has gone to engaged and then checking, as I type (very gently ) , two of the aqara motion sensors are stuck on 59 and 32 as they have not been triggered since I stopped moving and they recorded their last value.
Iām beside the other aqara one. Room lux is recorded as 3 when in vacant state , Iāve lost one of my lights (so this makes sense), I trigger the aqara sensor that went to 10 and fibro went to 33, (59 + 32 + 33 + 10) / 4 = 33.4 and the room shows 33 which is 3 above the room lux rule value.
It looks like the room gets triggered to occupied and lights go on, lux is 33 and the rule kicks in to say thatās above the occupied lux level , turn off lights, which may be a bug in itself.
Its late here now, i can run some other tests tomorrow. The end result maybe the lux values can not be used with the aqara sensors and when it worked previously it was due to a bug which has since been terminated, which makes sense from the past until now.
I did not see the switches in the Rooms child app.
I was expecting to see them there. Just as a window sensor open/closed, so I expected to see lights (switches) on/off as I was building the sensors that define the room.
the only reason they are not in the room devices is because i wanted to be able to support different lights switches to trigger occupied and engaged states hence their placement in those respective settings.