Hi all, 1st post so please excuse if the answer to my question is buried in a post elsewhere, but searches have turned up nothing.
Problem: We have a new kitten and two kids, 10 and 8. Cat will be indoor only. Kids tend to leave doors open occasionally because…kids. We have a mudroom that separates the kitchen from outside, so we have a two door redundancy, which is good. I want to use ST multisensors at each of those two doors to trigger an immediate (local, non-cloud based, optimally) continuous “Doors open!” or similar voice warning from Alexa or Google Home when BOTH doors are open rather than either one.
Thoughts behind my proposed required solution:
Individual 90-120 db door alarms are clearly not the solution, but it is sometimes worth stating the obvious.
A voice warning that triggers when either (rather than both) of the doors are open will be extremely annoying and will be disregarded over a very short time due to continuous warnings. If the continuous voice warning is immediate (local) and triggered only by both doors being open, the warning will not likely be ignored over time.
Voice is more pleasant that alarm, without loss of urgency.
Advanced thanks to anyone and everyone who can help with this one!
Core is probably your best bet to do what you described, but at present, all the voice recognition services that I know of, including echo, are themselves cloud-based. Even Siri on your phone. That’s because voice recognition is extremely complex and requires very high-powered processing. So that gets done in the cloud.
On the other hand, if you just want text to speech announcements, those can be pre-recorded and played locally. But then SmartThings isn’t going to run local either, so I don’t think your idea of local processing is going to work if you’re looking for a SmartThings – based solution. (For example, core runs in the cloud, not locally.)
There are some sensor-based systems designed for people with Alzheimer’s which can do what you want, but they wouldn’t be part of SmartThings.
I also wanted to mention that Mike Maxwell has an excellent smartapp which can combine multiple SmartThings sensors into one zone. He wrote it to solve the problem of too many false alarms from motion sensors, but it can also be used for the kind of use case you’re describing, where you want two doors open to trigger an event, but not just one.
Once you use his zone manager, you now have a zone device available it smartthings which you can use with any other routines are smart lighting or core or whatever.
Again, though, it’s not going to run local. It’s going to run in the smartthings cloud. ( at present, all custom code runs in the cloud)
If you haven’t used custom code with SmartThings before, it’s pretty straightforward. Basically you copy the author’s code and paste it into your own account. Here’s the FAQ for that:
Then if you decide that something you’d like to try, here’s the other thread for the zone manager
So zone manager solves the first part of the puzzle: how to get two sensors to be treated as one zone in SmartThings.
What You do next depends on exactly what you’re trying to accomplish.
You could definitely use one of the regular door chime methods to have a voice alert play once. I’m not sure how many times it would repeat, though. Anyway, here’s the information on that:
Also, just in case you haven’t read about core previously, it’s a very powerful community created rules engine for SmartThings. You can definitely do things like repeat and have stacked conditionals and all that. You also wouldn’t necessarily need the zone manager code because Core itself can look at the two different sensors. The only issue with it is it can be quite complicated to set up.
There are two threads on core that you would need if you want to use it.
The first is a peer assistance thread. That’s where you can get help setting up your rules. (Core calls and individual rule a “piston”)
The second is the authors thread where you would get the Core code:
Thanks for the phenomenal drill-down response, JD. Much appreciated. I didn’t expect all of my wish-list boxes to be checked off and not surprised that local Echo/Google response is not in the cards (although I’ll look into text-to-speech, which may get me where I want to be). The local response is highly preferred because a cat isn’t going to patiently wait the two and a half seconds for the cloud-based response to come through while the doors remain open, lol. But you and Brian introduced me to the apparent rabbit hole of Core. I look forward to jumping down it.
(Jason "The Enabler" as deemed so by @Smart)
Not sure if you found a solution, but this might help you out.
I have a slightly different situation but could the solved in the same way. I have a cheap garage door sensor that sometimes says the door is open when it isn’t. I bought a second sensor to replace it with, and it has the same issue. (Not sure if it’s because it’s cold, or too far from the hub etc.
Anyways, what I’m hoping for is a virtual device that groups the two sensors together. Similar to the Simulated Presence Sensor (for presence sensors) but for the door sensors. Does anyone know where I could find such a device type?