jcallaghan / home-assistant-config

My Home Assistant configuration & documentation.
https://www.jcallaghan.com/
MIT License
175 stars 8 forks source link

Aqara Motion Sensor (plus Aqara vs Hue motion sensor tests) #10

Closed jcallaghan closed 4 years ago

jcallaghan commented 4 years ago

Brought one to test the occupancy sensor alongside my Hue motion sensors after talking to @frenck. If they outperform the Hue motion sensor I will consider upgrading to them and selling my Hue sensors (11 indoor, 7 outside).

jcallaghan commented 4 years ago

Well it arrived. Much smaller than I expected. Now I plan to run some side-by-side tests with the Hue sensor.

First thoughts are is that is pretty damn quick at detecting motion, as well as detecting repeated motion. It also seems more accurate in detecting idle presence (room occupied but sitting still). I’ll continue with some tests and share my thoughts.

image

jcallaghan commented 4 years ago

Test Results

I wanted to give these two sensors a fair review. This isn't a particular scientific review but a review to compare the two devices side-by-side.

Sensors

These devices contain a number of sensors that are exposed and available in different places. Both devices include temperature and light level (LUX) sensors. Previously when I had the Hue sensor integrated via the Hue bridge I’m sure the humidity was also exposed.

Sensor deCONZ Hue deCONZ Aqara HA Hue HA Aqara
Motion Yes Yes Yes Yes
Temperature Yes Yes Yes No
Humidity No No No No
LUX Yes Yes Yes Yes
Battery Yes Yes Yes Yes

Motion refresh rate

To test the refresh rate I waved my arm continuously to see how often the device refreshed and if the state changed. For no motion, I placed the sensor in a cupboard. What was interesting is that the Aqara refreshes regularly whereas the Hue sensor doesn't update until no motion is detected. Furthermore, the Hue sensor becomes idle within 10 seconds but the Aqara takes 100 seconds to become idle after no motion is detected.

Device Continued Motion No motion
Aqara 5 seconds 100 seconds
Hue Hue remains active 10 seconds

Temperature variance

Device Temperature
Aqara 28°C
Hue 22°C
Aqara Temperature Sensor 23°C
Test thermometer 24.1°C

LUX variance

Both sensors reported close results when placed in the room but when placed in a closed cupboard with my Petzl Reactik + head torch (with three modes 110, 1350, and 2700 lux) the senors reported very different results. The torch was placed at close range within a cupboard. The Hue sensor appeared to be more accurate where at the Aqara seemed to max out at 851. The results in the room appear to be

Device Dark cupboard Torch (high) in cupboard Torch (low) in cupboard Room
Aqara 3 lux 851 lux 851 lux 192 lux
Hue 1 lux 14447 lux 6132 lux 137 lux

Note: I use a Conbee II zigbee gateway and deCONZ running in Docker.

Side-by-side testing from deCONZ. image

image

jcallaghan commented 4 years ago

Here is a basic history comparison between the Hue and Aqara in a frequently active room.

image

jcallaghan commented 4 years ago

I've continued to use Hue motion sensors as my primary room motion sensor. The Aqara would be great for projects in the future.