App to open garage door when presence sensor (phone) approaches?

That’s good, but brings back my original problem – which garage door to open? Can’t determine that from iPhone presence. And, my experience with iPhone presence and geofence crossing is spotty at best.

How many times a week do you experience say, a 10 second lag in ST doing something it’s supposed to do? I’d say that’s pretty much a daily event. If that 10 seconds coincides with presence arrival, it’s too late.

The overall reliability of presence sensors is quite high. That is, it almost always correctly determines that a presence sensor has arrived. It’s just that the timing is highly variable.

At the end that is what I realized. Our brains are much smarter to trigger our finger to push button in car to open garage door!!! And it is very efficient and it works always. So I switched back to my own smart brain to make decision to open garage or not. :sunglasses::sunglasses::sunglasses::sunglasses:

1 Like

Ah yes, the ultimate SMART THING.

1 Like

exactly. as i mentioned above you’d need a pretense sensor in the car in addition to the phone presence.

I don’t see any of this at all. I have 215 devices, so its not a number of devices thing.

I do have 5 of the old STs motions sensors though that are plugged in and act as nodes.

One would never be able to close the garage door as long as the phone is there.

How so?


if phone presence = open garage door, how am I supposed to close the door?

Just close it when you go inside. Its not always keep door open when presence is detected… its when a presence event is detected and the person is home open the garage once.

Alright. I just seems to this non-code writer that, the next time the hub “pings” the presence and the presence is “yes” then the garage door would open, again.

Nope:) you’d be good

So, the “presence” would have to go to “no”, i.e. the person leaves the Geofence, and then the next “yes” report would cause the door to open.

you got it :smile:


There are a lot of DIY ways to set up phone app control of a garage door, but most bypass UL recommended safety standards by hot wiring the button inside the garage to a simple relay. Obviously, you can do that if you want to (and your homeowners association/insurance/township allows non UL solutions), but there is an alternative that does meet current UL standards and let’s you use presence detection and other ST features.

Of course all the presence detection issues people have been discussing still apply.

If you do want to follow UL standards when using unattended control:

  1. Assuming you already have a garage lift installed…Buy and install the Linear GD00Z zwave certified garage door controller in your garage. It should retail for less than $100. This adds the ability to send instructions to your existing lift using the zwave protocol.

http://www.linearcorp.com/docs/Linear-GD00Z-Earns-UL-325-Certification_PR.pdf

  1. Then use @garyd9 's free custom code smart device type.
  1. Finally, include the GD00Z in your SmartThings network.

  2. Now you should be able to schedule the garage door opener through SmartThings while retaining all the manufacturer-provided safety features.
    :blush: :blue_car:

1 Like

Thanks for the education! But I still come down on the side of “Just use your remote garage door opener.”

I actually park my car outside so this becomes a nuisance for me. Having to bring the remote inside all the time so the boogey man can’t just use it to get inside my house.

I am also a fan of “just pushing the button” but… i am a bigger fan of personal security :wink:

1 Like

Gotcha. I believe that is why most of us are doing this. Automation is one thing. Personal security is another. Sometimes they are complementary; sometimes not so much!

2 Likes

I just enjoy it because I park my motorcycle in the garage and I don’t have a remote on it. I suppose I could stick one under a fairing or something, but I don’t like the idea of a remote on my bike. It’s really nice for it to open while I come up the driveway. It seems to be working 80% of the time, and the other 20% of the time, I just have to wait about 10 seconds at the door and it opens.

Works for me. Stinks it doesn’t work for everyone.

I’m sure everyone has their reasons for not automating the door or for automating the door, but I’d rather talk about the later since this is a SmartThings forum after all. Let’s automate it and make it work better.

As a side note for the people who talk about using their clickers when the presence sensor fails, then having ST close the door. Seriously, wat? Do you not have a sensor on your door? The app has a check to see if the door is open and does not open the door on presence arrival if that’s the case. I’ve never once had this happen to me, and I have used the clicker when coming home. I can see if this is a problem where you don’t have a tilt sensor as the app wouldn’t be able to preform the check, and is just doing the open/close blind.

The problem is that ST doesn’t necessarily update the status of the door quickly enough to stop this. If the app checks first before it attempts to open the door, it may or may not accurately reflect the state of the door. Or, due to lag in sending commands to z-wave devices (which lags happen all of the time), the door may have been closed when the app ran, then start to open from clicker, then get closed when the command finally reaches the relay. Basically, ST cannot be relied on for super accurate state information in real time – in lagged time perhaps, but not in real time. ST admits this to be true. It’s not deterministic, so there is no algorithm that’s foolproof. My failure rate was high enough for WAF rejection after two or three days, and I don’t think she was being unreasonable.

I’m all for automation as you suggest. I’d really like for this to work, but it’s a crap shoot. Your 20% for 10 seconds means it’s not working very reliably. How long will you sit on your bike in the driveway without the door opening before you’ll do something else? ST was down an hour ago for about 20 minutes, for example, where no z-wave commands were executed.

The 10 seconds on my bike is better than reaching for my phone and opening the door with the app, or getting off of my motorcycle, entering my door code, and then getting back on the bike and riding in… so it’s definitely worth it to wait the 10 seconds… Maybe if it was 20+ seconds or more that would be an annoyance.

That stinks that it takes too long to update. I guess I’ve just been lucky.

Does the amount of devices determine at all the response times of update events? I have two hubs controlling only 27 devices while you said that you had over 120 devices. The hub that would be responsible for the garage door is only responsible for 6 devices while my home hub controls the other 21. I wonder if this helps my chances of a positive state change.

I really don’t think it has anything to do with the number of devices. The reason I say that is that I have zigbee motion sensors in the garage that fire z-wave light switches. Those fire reliably almost instantly, unless ST is in some sort of lagging condition (which happens often enough to be quite noticeable). When ST is in a lagging condition, everything lags. So we can distinguish “normal” “Things are working” behavior from the case where ST is just being slow.

And, even if ST is working, there is a huge variability in exactly where you are when that presence sensor sets things in motion. That’s why you see 20% failure rate for instant response, it’s obviously slow in either recognizing the presence, acting on it, or getting the z-wave command to the relay (or some combination of all three).

Of course with a bike I’d wait it out like you. But for a woman in a car, or even me, you just really have to wonder why you’re trying to crack this nut. Her question would be, “what good does it do me if I have to wait, when just pushing the clicker always works and I never have to wait?”