Check out these new MQTT occupancy sensors. They look like they have a really cool feature set but the lack of invisibility really turns me off on the product.
Sticky tape to the door? Gaffer tape along the whole door and baseboard to the nearest plug? Hopefully they come up with a better solution than that.
If you read the FAQ they tried to build a battery powered one but it was too slow and didn’t last long enough. For something like this you can’t let the device go to sleep for long periods of time (100 msec polling period according to their site) and it takes a lot of power to drive 64 heat sensors (relatively speaking). I’m not sure what other options they have.
It’s a good idea. I remember seeing an academic paper a few years ago discussing the theory of how this would work. I wonder if this company is from those same folks.
Yes I wouldn’t expect this to work over battery - I was thinking more retrofit an old home security system using the cables to the door as power to the sensor - or some other new cable run. In my house I could do all the upstairs doors through the attic.
Like I said it is a very cool concept- just looks like pretty “clunky” hardware at this point rather than the sleek invisible stuff I try to install at my house.
Well, it’s standard USB power so if you have something that can generate 5v ?amps over a Cat5 or something like that I see no reason why you can’t use that to power it and run the wires in the walls. It’s a device marketed to people adding them to an existing home. Not many of those are going to be willing to run wires in the walls. I suspect having running wires as a requirement would turn off more “average joe” users than attract more hard core installers. And the hard core installers should know enough how to fashion their own power cables to run from a power source to the device.
I see no reason why you couldn’t run your own power lines for these things if that is what you wanted to do. At most I think you’d have to crimp your own ends, but if you are used to running wires that isn’t something that is foreign to you.
Just to add some color, I know one of the co-founders, and I’ve given him plenty of grief over the non-professionalness of the current system :). So he’s definitely aware, and I wouldn’t promise anything, but hopefully a more integrated install option becomes available in the future (both easier connection to your own power supply, and a form factor that I’m hoping I can recess into the door frame). I have 22/2 for reed switches I pre-wired to (most) my interior doors when I built that I’m going to repurpose for Hiome once that becomes available.
Also relevant: https://github.com/ktgeek/openhab-stuff/blob/master/HOWTOs/configuring_hiome_with_openhab.md
I think that’s what jumps out at me about their marketing the most:
“Hiome Door has a sleek, minimalist design that lets it blend in with any style. There’s no batteries to replace. Instead, we include a 20 foot (6 meter) microUSB cable, slim power adapter, and gaffer tape, the premium cotton mask that Hollywood uses to hide wires on movie sets. You’ll be surprised how invisible everything is.”
Putting aside that this is not sleek design (I’ll give them minimalist), they might be right about the surprising invisibility. You’ll look at everything you have to install and question if you really want to look at this every day. Then you’ll rarely ever notice it, because we don’t tend to look at our door frames as much as the doors in them. Surprise!
My suggestion would be for them to separate the device from the housing, so that you have more mounting options for different styles of trim. There are too many compromises when you try to make one mounting option that’s highly flexible, and in this case it shows.
Putting design aside…wow, that price tag.$159 for a core and one sensor (plus a roll of gaffer tape), and $79 for each additional sensor? Maybe I’m just cheap (I am), but that feels like a lot of money for a sensor with a very specific use case. The number-of-occupants detection is interesting, but I would guess that it will only work for rooms that have a single door.
I wish them luck, but feel that this will be an uphill battle to be successful.
For a room with multiple doors, you need a sensor on each door, and you define the connections between rooms in Hiome. What I’m concerned about is its functionality with non-standard sized opening between rooms, as you mentioned. Not just mounting, but will the sensor still detect things in an 4’ wide x 8’ tall opening? What about a 6’ wide, sensor mounted ~9’ high, on the wall instead of in the “door frame” (a rollup door)? How about “strange” things going under the sensor. Pets? Humans pushing other things (would carrying a platter of hot food count as another person)? Wet humans after running through sprinklers?
I’ve yet to receive my first one so I can test these uses cases, but I’ll be happy to report back as I do test them. I have another friend that has one, and he speaks very highly of its reliability in knowing actual occupancy of a room.
As per the price - yeah, I feel like it’s a bit on the high side, but it is significantly more advanced than other sensors. A ge zwave dimmer with occupancy sensor is ~$60. A dedicated zwave motion sensor is what, $35-50? A regular motion sensor for a security system can be as cheap as $10 (which I have a grundle of, but they’re insufficient for proper occupancy detection), but then you also need to add the security system panel, plus a means to integrate it into openhab, rapidly increasing the price (this is where I’m at for a lot of my interior doors. It was cheap to add a couple since I had some open zones in my security panel, but now I’m full up so I’d have to pay for an entire extra panel).
Well, that just makes sense.
You’re right, I’m just cheap. I wonder if automation newbies will see enough benefit to justify the cost of the Hiome core and sensors (e.g. “is it worth $159 to know how many people are in a room?”), but I can’t objectively comment on it since I’m not in that audience.
If I were going to design an occupancy sensor with this concept, I’d try something similar to the beam sensors used to stop garage doors when something’s in the way (IR? laser?). You could mount the emitter and reflector as high up the door frame as you want (to include/exclude pets), and have two side-by-side beams in each emitter to know which direction a person is going.