I've been holding off getting any motions sensors to work with my C7 hub until a 700 series one came out. (Everything I have is Z-Wave plus so I don't want to start with just 1 or 2 Zigbee devices here or there). I saw this one today from Haozee, but the few reviews are pretty terrible based on 1) terrible battery life; 2) no Smarthings driver for humidity; and 3) price.
Taking the last one first, it seems it was $40+ but is now selling for $30 on Amazon
As to the batteries, I can't tell from the reviews if it is really going through batteries in a week or if they just shipped with bad batteries.
Of course, I don't really care about whether or not there is a SmartThings driver, but I'm assuming there isn't a Hubitat one either. I can program in VB, but that ain't going to help me here. :-).
Love to hear those more knowledgeable than me (almost everyone/anyone) chime in.
In a sense the accuracy is irrelevant as long as it is consistent. For example, myself and friends have had a few BMWs and they always report my speed as being 2 miles per hour faster than it really is. So, we know to compensate for it. Here, if I know I want Hubitat to turn on my hall light if the device detects motion and it reports illuminate of X, as long as it reports X consistency I'm fine - even if it reports X as being dark and and with an accurate device X would actually be too light.
A link to the device manual is in my driver thread above. It explains all of them.
Delta settings are exactly that - the delta value the sensor needs to read from the last reported value before it will report again. Evaluated at whatever sensor check interval that is set. Larger deltas will make it report less frequently, saving battery.
Blind - This parameter is configured the time interval between two motion events triggered, during this period the motion detector will not be triggered even there has someone move in front of motion
detector.
Setting it higher will make it use less battery, but motion events during the "blind" period won't be registered/reported.
Clear - This parameter is configured the time to clear motion event after a motion event detected. Time to motion clear, the device will send a clear event report to controller.
Setting it higher will make it use less battery, but the device will shows as status "motion" longer.
Have you found the accuracy of the various levels to vary by a percentage, a fixed amount or “random”? For example, if the temperature reading is always 2 degrees high (or low) the offset works perfectly to adjust for that. But if it is 10% off then a fixed number offset doesn’t correct it (but could come close). However, if the accuracy is “random” (i.e. sometimes it is correct, sometimes low and/or sometimes high) for the same temperature then the offset setting doesn’t help at all.
What has been your experience with the accuracy of the various sensors?
BTW, I looked at the manual and figured out that by “delta” you meant change. Once I did I remembered that from my college days, but that was many, many, many years ago. So saying the delta value is just that, the delta, wasn’t particularly helpful. I’m pointing that out to remind you that people up here vary greatly in their programming/science knowledge levels. But I do appreciate you taking the time to answer my questions.
I was looking at the temp and humidity readings the 1st month I had it, but I now have temp/humid/lux reporting turned off entirely to increase battery life.
When on I found the accuracy "as expected". While I didn't calibrate it with a known temperature in a box like I do for temps I use for automation, the temp and humidity seemed to track in a logical way that could be calibrated with an offset.
There are only 4 or 5 commonly used sensors in any of these devices, and they all have about the same accuracy. Usually 2-5% on humidity, but temp can be 0.1C to 1C. Not sure which specific sensor this one uses as I haven't ripped it apart yet (and might not since it is on my wall now).
I didn't spend much time with lux other than verifying it went up when lights on, and down with lights off. And that it was indeed reporting lux and not some other wacky illumination value.
Fair enough! I'm an engineer and only speak engineer most of the time. LOL
What I want to happen is for lights to come on in some hallways/rooms when someone walks into them but only if the light level is below a certain amount. So the simply wording of the rule would be like: If motion detected and illuminance is below X then turn on light A. And from there I might even make it more complicated to add a time value (which may be easier to create a Hubitat "mode" for those hours) so that
if motion is detected and illuminance is below X then
Select Case Mode
Case Late
Turn on the lights to 50%
Case DeadOfNight
Turn on the lights to 35%.
Case Else
Turn on the lights to 100%
End Select
Endif
So, knowing my use case would you recommend this sensor?
Eh, tough call. Depends on your expectations of how fast it should work and how fast it should respond to changes in lux.
Should work, but whether there will be more delay than you want is unknown. There will be some delay doing it this way - motion report, RM rule runs, light triggered. Should be less than 0.5s overall, bit some people get really annoyed by 0.25s delays in motion lighting. 🤷
Motion reporting is pretty fast on it, though, so at least has a chance of working.
Fair enough. I want it to respond like in public bathrooms where the instant you open the door it turns on. (As far as I can tell they use a motion sensor in the wall, not a door sensor). Since the motion turning it on won't happen that often I don't need a super fast cool down time, but I would want them to turn on virtually instantly.
There will be a small delay any time you are doing motion lighting controls in a hub versus all in the switch like commercial installs do. Whether it is close enough to instant for you or not is up to you.
Usually if the sensor isn't crappy (this one doesn't seem to be), your hub is healthy and not bogged down, and the light switch and bulb are fast then you should be ok.
But like I said, some people are very sensitive to lighting delays. Motion lighting done in the switch (no hub involved) usually turn on in <50 milliseconds.
Motion lighting done with Zwave and a hub is typically in the 250 millisecond range (150ms best case).
If not specifically using temperature or humidity you could slow those down (increase the delta) to extend battery. If you don't need rapid changes on lux you could increase the sensor check time too. But typically the radio is the main power user, so extending the sensor check time saves less battery than reducing event transmission does (although those two settings are interactive/interrelated).
Or if there is frequent motion (aka busy room) you could increase the motion clear and/or blind times to reduce the # of motion radio transmissions/events.
None of that is "necessary", but when done properly actually can make large differences in battery life in some situations.
Even in a busy room I would expect at least a year from the batteries with default settings though. I just like to go for 3-5 year when I can as I get tired of changing batteries.
