This is not specific to any hardware. I have been setting up various motion sensors in toilets so that they turn on the light when the user enters. The eternal problem is that the light will turn off after a set number of seconds (30, 40 seconds, whatever). This usually works, but often ends up in wild hand waving.
This is of course a problem and I have seen videos where people discuss this. I am wondering if there is a solution to this problem as I am sure it happens to everyone?
The only answer I can see if to use a Human Sensor and not a PIR, but the ones I have bought seem little better than PIRs. But then I have been getting only Tuya/Zigbee stuff, so perhaps I need to look at something a bit more up market?
I have thought to add in a door sensor, but we usually leave the doors either fully or partly open. Some are sliding doors and are hard to fully close anyway. Perhaps a pressure sensor on the toilet seat, but have not been able to find a Tuya pressure sensor. But that might be a good idea.
Dunno, which is why I am asking the good folks here. Any thoughts?
PIR’s are nice and fast to detect but require significant movement to maintain presence.
μWave sensors are a little slower to detect (barely perceptible) and can detect micro movements. However to get that sort of micro movement detection the sensitivity has to be cranked right up. In my case this leads to false triggering from trees outside the toilet window and triggering from movement in adjacent rooms.
Another solution I’ve been looking at implementing is a time of flight distance sensor, e.g.
The issue with these sensors is that they are only good for detection out to a couple of meters at best and 50cm at worst (depending on IR reflectivity of the detected object).
I suspect what I will end up doing is a PIR for turning the light on and a time of flight sensor mounted behind the throne for keeping the light on.
People learn quickly if they don’t want the light to go out. Close the darn door. :). If it weren’t a bathroom I’d have more sympathy for the user. In this case no. Closing the door helps me too.
HA! Just had a brilliant idea. One way is to sense the water level in the cistern. Put in a level sensor and when the water level drops, it means it has been flushed. Then wait 30 seconds and turn off the light. Might work.
I use multiple presence sensors in a single room because they’re so cheap. I use wasp in a box using all presence sensors. I use humidity in bathrooms and kitchens.
Finally I use a Bayesian sensor to aggregate the data with the prior obtained from history stats sensors.
I also feed in schedule and time of day sensors into the Bayes sensors as well as estimated distance sensors from the mobile app beacon sensors.
All in all they provide me with a fairly accurate room occupancy probability
I am currently using 2 of these and they work exactly as I want in a toilet and a bathroom.
I have them running since 6 months and there is no sign of battery drain yet, which is impressive.
Would buy them again.
Yeah, I have a couple of those in the PIR version also. They work fine in the hallway where you are only present for a short time and then pass onto other sections of the house. The toilet is different in that you can be sitting there for some time with very little movement. So far I have extended the time out to something like 2 minutes and have moved the sensor pretty close to the user (on a side wall). It works fairly well, but not perfectly. Bit of a work in progress.
