Best PIR for motion detection?

Hopefully a simple enough question; What do people consider to be the “best” battery powered PIR for UI7/Edge. I’ve a couple of the Fibaro units and whilst they are good looking and have plenty of bells and whistles, the basic job of motion detection is a huge let-down. Even the basic units from Screwfix do a better job! The only limitation is that it must be battery powered - no requirements for temp/humidity or lux.

Regards, Pat

From what I have found, Z-Wave Motion Detectors are all designed for security applications rather than for occupancy or presence detection applications.

There is a big difference between these two applications: With security, you don’t want the detector to trigger except when it is obvious that something is there and moving. The major design consideration is avoiding false alarms.

With occupancy applications, you usually want the detector to trigger on small movements like someone writing at their desk. False alarms are no big deal and you prefer that the detector triggers with very little motion. With occupancy sensors, I think the PIR detection fields are smaller and more numerous. Some of these also trigger on sound or other conditions.

The best occupancy sensors that I’ve been exposed to are made by Sensor Switch (see http://www.sensorswitch.com). But I am not aware of Sensor Switch making anything with Z-Wave capability.

I have considered building a Fibaro Universal Binary Sensor module into a Sensor Switch CMP 9 PDT, but that seems like a lot of trouble and expense. I am surprised that Z-Wave occupancy sensors are not more common.

Ken

To add to that battery powered/wireless is another issue. Wireless/battery powered has to conserve power in order to not drain batteries to fast. This means sensors are slow to wake up, sleep after it reports a few motions or both.

More on that is like what was stated above they are designed with security not presence in mind. So if it see’s a motion keep tripping over and over (your in the room) it stops reporting to conserve battery. It’s focused on tripping only the first time when an intruder first steps in front of it.

This is why even with an alarm panel integrated into vera we always try to push for a wired motion vs. battery motion. (plus cost is cheap, no batteries to replace and several other features). Usually people think running a wire is impossible when in fact several years back that was the only way it was done. I guess it only became impossible recently when wireless started.

Camera’s are currently hitting this same block. POE cameras are way better then wireless but in a short time a wired camera will be impossible for most to install… So they think.

Thanks to both integlikewhoa and KMorley; all the points add up (unfortualtey!).

So, given the limitations, who’s appears to have the “best” balance of sensitivity vs battery life? The Aeotec Mitlisensor 6 feels like it has the edge over the Fibaro on range at least - 7m vs 2.3m.

Pat

I currently have both motion sensors in testing and I don’t notice any dramatic sensitivity difference between the two. According to the manufacturer’s manuals, when both sensors are mounted 2.5m above the floor, the Aeotec has 5m range and the Fibaro has 7m range. Both have an 120 degree field of view. Those specifications are for using the sensors as a security device where they require large motions to trip.

I have the Fibaro sensor mounted at chest level about 2m from my desk. When I am sitting at the desk and typing, it does reliably trigger on hand movements as I type. But there is no way that one mounted in my office would function acceptably as an occupancy detector for automatic lighting purposes. I think it would take at least four to get sufficient coverage for that application.

For that reason, I have decided to use a ceiling-mounted occupancy detector specifically designed for that application.

Good luck!

Ken

Has anyone seen this occupancy device on indiegogo:

http://makezine.com/2015/12/04/use-radio-tomography-monitor-home/

[quote=“KMorley”]I currently have both motion sensors in testing and I don’t notice any dramatic sensitivity difference between the two. According to the manufacturer’s manuals, when both sensors are mounted 2.5m above the floor, the Aeotec has 5m range and the Fibaro has 7m range. Both have an 120 degree field of view. Those specifications are for using the sensors as a security device where they require large motions to trip.

I have the Fibaro sensor mounted at chest level about 2m from my desk. When I am sitting at the desk and typing, it does reliably trigger on hand movements as I type. But there is no way that one mounted in my office would function acceptably as an occupancy detector for automatic lighting purposes. I think it would take at least four to get sufficient coverage for that application.

For that reason, I have decided to use a ceiling-mounted occupancy detector specifically designed for that application.

Good luck!

Ken[/quote]

I have been using Schlage

  • identical to most others, except Fibaro & Aeotec.
  • VERY sensitive for motion, but if you are very still (reading a book, sleeping, etc) they will not trigger.
  • VERY long range, don’t have an exact but WELL over 7 meters.
  • I use them in ‘test’ mode, which resets in 4 seconds. This is great for Entry alerts - when people are at your front door (trying to break in), etc

I just purchased Aeotec Gen5 hoping I can add some features like light sensor for Dusk/Dawn scene accuracy.

What do you use to take the output from the ceiling sensor into z-wave. Mains powered with instant update etc.
No point having a device with multiple inputs as it would have to be mounted next to the sensor which will only have 1 output.

I haven’t built anything yet because I’m not sure it’s really worth the trouble and expense. It would be fun though…

The SensorSwitch can be sourced with 12V to 24V AC/DC. Whatever you use for the source is outputted on the trigger line when the room is occupied. For example, if you source the SensorSwitch with 12VDC, then 0VDC is on the trigger line when the room is unoccupied and 12VDC is on the trigger line if the room is occupied.

Since the Sensor Switch outputs either AC or DC, the design probably uses a reed switch rather than a more expensive component like a triac. It probably wouldn’t be too much trouble to cut the trace connecting the relay one switch contact to the source and rewire both contacts to trigger an Aeon Micro Switch or Micro Dimmer.

As an alternative, you could also use an Arduino with a Ethernet Shield and skip the Z-Wave: just plug the Arduino’s Ethernet into the same network as the Vera. If your network is equipped with a PoE switch, you could even use PoE to power the SensorSwitch and the Arduino since they consume so little current.

But again, all of this seems like a lot more effort than it’s worth. It’s just surprising to me that no one makes a Z-Wave occupancy sensor.

Ken

I think you are correct, it does seem more trouble than its worth. I certainly don’t want to start putting in dc supplies. 240V is already in the ceiling so everything would have to run off that.
Overall I will probably stick with just turning the lights on and off by a switch

Yes, wired is certainly best. However wireless has really improved a lot. For wireless PIRs, I like:

http://www.amazon.com/Schlage-Z-Wave-Motion-Sensor-Intelligence/dp/B008Q5CTTG and
http://www.amazon.com/Enerwave-ZWN-BPC-Ceiling-Mounted-Occupancy-Association/dp/B00PV0AXWE

Both great for hands free lighting since they have fairly short resets which makes them more versatile.

Thanks for all the input. Unfortunately, neither of those devices are available in the UK…

For the stepped down DC SensorSwitch project, surely the Fibaro Binary sensor would do the trick to get it into Z Wave?

Pat

I’m sure there are several ways to do it and the Fibaro Universal Sensor is probably one of them. It depends upon how quickly the Fibaro wakes up and reports. If it wakes up on status change, then it should work. If it only wakes up every 15 minutes, then that might be a problem.