I’ve been playing a bit with home monitor recently. What I’ve discovered is that some of my zigbee motion sensors (typically the tuya cheap ones) appear to generate motion events when there’s no one there. I’d never noticed this before as I typically only used them for lights and would not have noticed the lights were coming on in my absence.
I don’t think I’ve seen a spurious motion event from either of my two Aeotec motion sensors and possibly not the IKEA ones either.
Anyone else seen this behaviour ?
I’m using zigbee motion sensor mc driver, if that matters , which I suspect it doesn’t. Sensitivity setting is set at 2
My IKEA motion sensors (the Tradfri ones) could be relied upon for at least one spurious report a day each, and some of them would do several a night. I replaced them with Third Reality ones and a couple of those were giving false reports practically every minute on occasions (until relocated) while others were rock solid.
Moths, lights turning on and the sun coming out from behind clouds are just a sample of what may have been triggering them.
I have the original SmartThings Zigbee motion sensors that would create a lot of false positive motion detections when the coin battery was showing less than 50% in the SmartThings app. So, I replaced the 3V coin batteries with a 3V external dual AAA battery pack and a lot of the false positive motion detections went away. And some times changes in ambient light still triggers a false motion detection as well.
Also, the infra red night detection sensors from my cameras would to, so I just angled everything a little bit which resolved that from happening.
There are many home monitors fully suitable for intruder detection, but they’re not the kind you pick up sensors for for six dollars on Alibaba. Or anything made by Tuya. Or almost anything that works with SmartThngs.
these devices are designed and intended for home automation, even if they’re labeled security, and they just don’t meet the same quality standards as the purpose built security ones, which is why they don’t have UL listings for security or other independent third-party certifications, which would test reliability among other things. This has been true for more than 15 years. Sometimes you do get what you pay for.
There are many high-quality Zigbee security sensors out there, but it’s just a different device group. And much more expensive. (Sea control4, for example.)
Also, this time of year, we always see a lot more false alerts on PIR motion sensors because with the season change two things are happening. Insects tend to be more active in the home. And there are drafts throughout the house at quite significant temperature deltas which can trigger the alarms even when no one‘s in the room. Also, late day shadows can do this. It’s because it’s getting cool outside. It’s warm indoors, the heat systems come on, the shadows fall and you just get a lot of temperature throughout the house.
The traditional approach if you want to use cheap sensors is to set up a zone of three or more sensors and not trigger an actual security alert unless all three of them report an issue.
Here’s an FAQ discussion from 2016 that’s still valid in concept although the exact methods will vary with today’s architecture.
So the short answer is, if you want a good security system, buy a good security system certified by an independent third-party.
If you want to cobble together a decent cheap security system, using home automation devices, use the zone approach.
Yeah, I get that. I have already had a full, third party, security system for many years. I was experimenting with ST Home Monitor as a backup, to cover times when I have an outage.
I would definitely check the area for drafts of different temperatures. As I mentioned, that tends to be a huge issue in midfall and in some regions in mid spring. You can have two spots in a room which are 10°C different in temperature with the shift moving across the room, which is what sets off the motion sensor.
