I’ve been using humidity sensors to detect when our showers are turned on, and while they have worked reliably, they are pretty slow to respond (minutes after the shower is turned on and off). So I’m interested to see what other options might be available that would respond more quickly.
I looked for flow sensors, but couldn’t find any great options for this purpose. I did find this:
It seems to fit my use case, but it’s unclear how to access the data. They mention having an open API, so I reached out to their support group for more info. We shall see…
I’m also considering using a leak sensor on the floor of the shower. I don’t know if I can get the placement right, but it might be worth a try. Has anyone done anything similar?
Final option is probably to combine the humidity sensor with a motion sensor. I think that’ll improve the “shower on” event, but I may have to get clever for detecting the “shower off” event.
I'm using NYCE motion sensors with humidity and they are extremely quick to respond. I would guess near instant honestly.
Heres what I do. Motion sensor plus door sensor. When door shuts, fan comes on. When door opens fan turns off UNLESS humidity is high, then it runs until humidity is low. Works great.
I'm off target as I didn't implement humidity sensor - I use motion sensor IN the shower. By using two motion sensors in the bathroom, I can identify (and adjust lighting, fans etc) and run scripts/rules based on that. My environment setup seems to work flawless for my needs fWIW.
My only bathroom weakness is fart smells I've been watching the dialogs about methane sensors hoping I can automate the bath fan better! As it stands, I now just use a weak scheduling implementation as I'm a pretty "Regular" kinda guy LOL
Ooo, now that’s a clever idea! I’m bummed that Aeotec hasn’t been able to get these in stock on the US Amazon yet, but I have a few on doors around the house that I might be able to borrow for a test drive. My shower heads aren’t very big, so I might not be able to get the placement right, but it’s worth a shot. Thanks for the idea!
I’ve had my eye on these, but haven’t pulled the trigger yet due to the cost (and Amazon’s struggle to keep them in stock). I may have to grab one to try the next time I can find it.
This might be the direction I end up going. I don’t love motion sensors in general, but I may have to bite the bullet on this one. Having two does sound like it would help with the reliability. I’m not convinced I move around enough in the shower, but if it works, this is probably the most practical option.
You might consider a temperature measurement of the show outlet pipe. You could sense the sharp increase in temperature start your fan and let the humidity sensor determine when it should go off.
I've not automated my shower fan but have thought about it. My thoughts are to shut the fan off sooner in the cold weather than in the summer. Goal is keeping as much heated moist air in the home during the winter as I can without condensing too much on the bath walls & fixtures.
I’m not really concerned about the bathroom fan itself, as the humidity sensor has worked well for that and I can live with a few minutes delay in that scenario. I’m more interested in determining occupancy, specifically when the room is vacant. Motion sensors suffer from the fact that they’re motion-based: did the person leave the room or did they just stop moving? Humidity kinda works, but is far too slow. My sensors typically do not report the humidity decreasing until 5-10 minutes after the shower has been turned off. Ideally, I want something that can detect that the shower is off within a few seconds of it actually happening. I can probably live with a delay of up to 30 seconds, but my current humidity sensors are not good enough for that.
I’m playing around with tracking room-level occupancy throughout my home by using every device I can. You open a door? It knows where you are. You turn on a TV? It knows where you are. A motion sensor is active? It knows where you are. And so on, and so on.
Instantaneous actions (door opening/closing, motion active, etc.) are helpful, but I’m finding that the most reliable devices are the ones where you can definitively say: as long as this device is active, then we KNOW that this room is occupied. For my family, so far, that’s when the TV is on, when certain fans are on, and when the shower is on. If I can more accurately track the shower status, that’ll help boost the accuracy of the entire system.
I’m strongly considering it. I’ve had this pulled up in my browser for several weeks now:
I’m think about having these in each room and using Locative on my phone as the receiver. But I think most people are doing it the other way: carry the beacon with you and keep a receiver in each room.
With a beacon in each room and GeoFency. The beacons seem cheaper than the sensors and more flexible because they don't use much power at all. And you can vary the range of the beacons by changing the power output. I've found tweaking power essential to prevent "room overlap." Locative looks like it uses the same basic principle... web hooks.
Right now I have two sensors set up on the front porch. I first look for presence via an aggregate presence sensor that uses geofencing from both Life360 and GeoFency to determine someone has actually arrived at the house. This prevents the door lock and alarm from reacting when I just go out onto the front porch for my daily cigar. Once I've left the geofence perimeter and returned it gives me 3 minutes to get to one of the two beacons.
(We might want to spin off this thread unless you want to talk about waterproof iBeacons and insisting everyone carry their iPhone into the shower. Hmmm... that's not such a bad idea!)
IF and only IF you can get to one or more of the water lines to the shower (optimally the actual line going to the shower head itself). You could use the temperature change in the pipe to indicate activity; with some lag after the water has been turned off which might invalidate this approach for your needs.
See the Senasys link in this thread for simple & reliable pipe mounted temperature switches. Then you'd need something connected to that switch monitored by HE per this discussion.
I've been watching the dialogs about methane sensors hoping I can automate the bath fan better! As it stands, I now just use a weak scheduling implementation as I'm a pretty "Regular" kinda guy LOL 
