Aqara High Precision Motion Sensor (RTCGQ13LM) (NOT FP1)

AqarawithT1

Just bought one of these from AliExpress on the off-chance that it will work with CORE as @Markus has had one of these motion sensors since Nov 20 and gave a glowing review here as well.

Hopefully he won’t tell me I just wasted £40.00 !

Since then it looks like Zigbee2MQTT has released a driver for it recently.

Seems like their are 19 left at present (31 May 22) Aqara High Precision Motion Sensor (RTCGQ13LM)

A review by Markus can be found at [Review] Aqara High Precision Motion Sensor (RTCGQ13LM)

1 Like

At $50 a pop I’d need a very good reason to go with one of these but Markus’ review sure makes it look good. I look forward to your experience.

Then it will work with CORE.

I’m not sure why you posted this in the beta group. Maybe @april.brandt will move it to public exposure.

I assume this (along with ZWave2MQTT for ZW devices) is the only thing necessary to be “compatible” with CORE since they intentionally are not trying to reinvent the wheel.

Correct AFAIK

49 posts were split to a new topic: Aqara FP1 Motion Sensor (Splitfrom Aqara HIgh precision)

Split post as it looks like it flipped to FP1 in error. Aqara can be so confusing sometimes.
:grin:

I’m using the RTCGQ13LM in my kitchen, it has been great, though it is enormous, so unless you can mount it in-ceiling it does stand out. If you stand absolutely still it will stop updating just like any other PIR, but with the short timeout for checking for motion again it works really well in most scenarios (in my kitchen).

The one I received had a 12 volt US power brick. Some of the ceiling mount ones take 120-220 line voltage.

This is my item, and it works fantastic. Be careful to click on correct country version for included adapter
https://www.aliexpress.us/item/3256803919760969.html?spm=a2g0o.order_list.0.0.c8e71802xwvpLe&gatewayAdapt=glo2usa&_randl_shipto=US

Just arrived my tuya device.
Quite fast motion response, but the only motion detect attribute it exposes is “presence”. Quite fast to trigger but extremly long cooldown (>30 secs).
Is it a matter of z2m implementation or is the model? If I’m not in wrong the aqara model has also zones and directions or something like that.

EDIT1: to short the cooldown you must set fading time =0. Neverthless at a distance of about 4.5 m and sensitivity=max the sensor goes presence=false if I’ mjust sitting. It is not what I expected

EDIT2: the sensor is extremely sensitive to the position. After lots of tests it is workig overall quite well. Do as many trials as possible. The instructions does’nt help a lot: no mw frequency nor view angle.

EDIT3: it is orribly chatting: mqttexplorer receives > 1 msg per second. I was forced to restrict the “z2m in” node only to receive the exposed presence attribute and only when value changes. Is it a z2m problem or is a device problem?

1 Like

does anyone know how to decrease the number of messages sent by the Tuya mw sensor?
It is stressing z2m /mqttt with 2 msgs/sec !

Use throttle

In the node config I restricted the enabled attributes to presence only+“only if changed” option. That filtered a lot. But looking at mqttexplorer, that device is pulsing twice per second. The extra work of receive/filter such a huge amount of messages still is in charge to z2m/Nr I think.
What if I use many of them ?

Can you not change the reporting in z2m?

1 Like

It Is a Tuya device. If i understood well such devices have non standard cluster layout
I found this reference :

Reading the reference, the motion cluster should be IAS zone (0x0500), but I could’nt find it in z2m.
More over, I’m not sure what value I should assign to the minimum change attribute.
I’m worrying about changing report mintime values too, as I would’nt like to slowdown the device motion responsiveness .
This is a matter requiring a super @markus advice… :slight_smile:

2 Likes

New improved FP2 model

https://youtu.be/p1dVVJ7ug_c

2 Likes

Interesting that they went Wifi over ZigBee, maybe trying to release more Matter products in there range though.

once the device needs to be powered, wifi/ip is a reasonable choice…

1 Like

2023 will be an interesting year in smart home because of this. I hope for the best, but I have a feeling it will be a cluster and at the end of it consumers are going to lose.