I love the idea of comparing the sensors between manufacturers! Great idea! Having said that, I'd love to see a base to compare for them. A professionally calibrated device which would be used as the comparison point. Lux and temperature should be fairly easy with the correct gear.
Did I miss the build video for the custom sensor that you displayed? I'm interested to see if you used capacitors with the AM312. I needed to use them to get rid of the false positives on motion.
FYI on that home built multi sensor, the temperature sensor works. I have the same problem with it running hotter. Noticed if I ran the thing outside the case like an angler fish (hanging off wires out the hole) it would match a regular thermometer in the room. The problem is likely the housing trapping heat produced from all the other electric components. Likely the temperature sensor should be isolated in another housing outside everything or the case should be redone walling it off from the rest.
That seems to be issue with many constant on sensors with the additional heat, even home thermostats have that issue with the added electronics. You can try to calibrate for it to get close enough but sometimes those battery powered sensors just work well by overcoming that heat issue. All down to the use case as I have done some sensors with the temperature sensor a few inches away.
Thanks for the infomative vid, i'm in the middle of making a d1 mini based motion/temp/lux sensor. I really don't want to do a DYI soultion because i would much perfer a profesional look. I don't understand why nobody makes one of these multi sensor boxes with a simple usb power port on it for those places where you have power available.
I have had that issue before where it just kept toggling back and forth over and over. I thought it was some sort interference thing but I swapped it with another and all was well. I have probably 4 or 5 around the house right and rely on them everyday.
Yes, PIR sensor gave lot of false positives in my tests. I added a doppler sensor (RCWL-0516) to the same board and created automation trigger when both PIR and Doppler sensor agreed on motion. There were very few false positive when both sensors were used.
It's illustrated very well in this video that all the environmental sensors (air quality, temperature, PM, etc.. ) are meant for the gadget consumer market and deliver fantasy values. As long as there is no way to calibrate them a bit, even if only by bringing them close to the values of a calibrated master, they will remain toys whether in a fancy package or not. Of course, for that cost, only toys can be bought. The spread of readings from side by side sensors such as the "lux" values doesn't really allow for an assessment other than guessing. Discouraging. The light sensor can be used to put the motion sensor out of action during the day for example. A temperature sensor that gives a deviation of 10% at 20°C is not worth the name unless you can adjust the offset or you are satisfied with an interpretation adjustment by your software or if necessary your brain.
Devil's in the details. I would love to see more calibrated testing. I've seen a huge discrepancy between my mut of sensors. I'll look for an Ikea motion for you.
Nice job with the follow up
The ikea motion sensors have been out of stock in Canada for ever too, i don't know what's up with ikea's supplier :/ !
I love the idea of comparing the sensors between manufacturers! Great idea!
Having said that, I'd love to see a base to compare for them. A professionally calibrated device which would be used as the comparison point. Lux and temperature should be fairly easy with the correct gear.
I didn't have a NWS certified unit around to borrow unfortunately so I had to stick with the others which worked out pretty well I think.
Did I miss the build video for the custom sensor that you displayed? I'm interested to see if you used capacitors with the AM312. I needed to use them to get rid of the false positives on motion.
I have done a few with the AM312. This was one ua-cam.com/video/9Yu57vjz7AY/v-deo.html
There's another newer I did in a bathroom shower setup.
FYI on that home built multi sensor, the temperature sensor works. I have the same problem with it running hotter. Noticed if I ran the thing outside the case like an angler fish (hanging off wires out the hole) it would match a regular thermometer in the room. The problem is likely the housing trapping heat produced from all the other electric components. Likely the temperature sensor should be isolated in another housing outside everything or the case should be redone walling it off from the rest.
That seems to be issue with many constant on sensors with the additional heat, even home thermostats have that issue with the added electronics. You can try to calibrate for it to get close enough but sometimes those battery powered sensors just work well by overcoming that heat issue. All down to the use case as I have done some sensors with the temperature sensor a few inches away.
How about Neo Coolcam sensor, i have several and they work great
I will check them out.
What about the ones on 433Mhz?, do you mind in include it in the next video?? Thanks!!
I will see if I can get one.
Thanks for the infomative vid, i'm in the middle of making a d1 mini based motion/temp/lux sensor. I really don't want to do a DYI soultion because i would much perfer a profesional look. I don't understand why nobody makes one of these multi sensor boxes with a simple usb power port on it for those places where you have power available.
Just a small product market I would assume. It's not too hard to make your own if you have a 3D printer.
I pried open one of my aqara motion sensors, i could not find a temp sensor of any kind on the pcb
It could be on the processor itself which would explain the higher value.
yeah! go shelly! go shelly! go shelly! :p and for the record: I don't mean mr cooper junior :) KUTGV
Awesome video as always! 👏🏻
Do you get false positives with the PIR AM312?
I have had that issue before where it just kept toggling back and forth over and over. I thought it was some sort interference thing but I swapped it with another and all was well. I have probably 4 or 5 around the house right and rely on them everyday.
Yes, PIR sensor gave lot of false positives in my tests. I added a doppler sensor (RCWL-0516) to the same board and created automation trigger when both PIR and Doppler sensor agreed on motion. There were very few false positive when both sensors were used.
Is there a way to calibrate the Aqara? I've got one and I've noticed it's high too. Maybe some trick to show an offset?
You could do some template sensor on it but then again is it linear?
It's illustrated very well in this video that all the environmental sensors (air quality, temperature, PM, etc.. ) are meant for the gadget consumer market and deliver fantasy values. As long as there is no way to calibrate them a bit, even if only by bringing them close to the values of a calibrated master, they will remain toys whether in a fancy package or not. Of course, for that cost, only toys can be bought. The spread of readings from side by side sensors such as the "lux" values doesn't really allow for an assessment other than guessing. Discouraging. The light sensor can be used to put the motion sensor out of action during the day for example.
A temperature sensor that gives a deviation of 10% at 20°C is not worth the name unless you can adjust the offset or you are satisfied with an interpretation adjustment by your software or if necessary your brain.
You can adjust them how you want with a templated sensor.
Thank you.
Devil's in the details. I would love to see more calibrated testing. I've seen a huge discrepancy between my mut of sensors. I'll look for an Ikea motion for you.
hue is trolling with that 69.420 temp
The odds... Lol! Nice catch!
Democratic sensor array - WiFi and Zb co-existing…and in peace and harmony!