Alarm.com, ADC-T2000 Thermostat & Smart Away Not Functioning Upon Return

Support
, , , , ,
1

Does anyone have experiences with Alarm.com, ADC-T2000 thermostat, and the Smart Away settings not functioning completely as advertised? When enabled, there are 2 options that can be selected here, one that sets the Away temperature set point when the panel is armed, and the other that sets the Away temperature set point when one or more (selected) devices leave a geo-fence.

I have used both options together for a period, as well as geo-fence option alone, with similar results.

In all cases that I have observed, Smart Away consistently sets the away temperature set point properly, as intended, per the configuration defined.

However, in no case that I have seen does it appear to successfully change the set point back to the schedule, when either the panel is disarmed or when a device enters the geo-fence.

Per the configuration description (with my emphasis added):

“Automatically save energy when you’re away from home. This option will set your thermostat to your Away temperature (pausing your regular schedule) whenever the system detects there’s no one home. Your thermostat schedule will resume as soon as the system is disarmed or any one of your mobile devices enters the geo-fence. (Make sure you: (1) Turn Schedules “On” using the website’s thermostat card or the mobile app and (2) Enable Geo-Services on each mobile device and set your preferences on the Geo-Services page.)”

In several recent cases, my wife has returned home (disarmed the panel, and carrying her cell phone well within the geo-fence), but even 30-90 minutes later, the Away temperature set point is still being used. She’s already very leery of the home automation stuff as it is, so experiences such as this don’t help when I am told that the thermostat is “broken”!

I am confident that geo-location in the Alarm.com app on her cell phone is configured and working properly, as other geo-fence-based triggers appear to be working just fine for her.

Has anyone experienced this or something similar? Conversely, has anyone seen Smart Away successfully reset itself back from Away to Home temperatures on its own, under the appropriate conditions?

Thanks in advance for any experiences you can share…

2

The be been successfully using the smart away “stuff” for almost two years now with a Qolsys panel and three of the T2000 thermostats. Once in a rare while, the geolocation fails to register that one of the devices has left or returned but in just about every instance it’s been due to cell data being turned off so ADC doesn’t actually know the device isn’t home. I’ve never had a failure of the temperature rules to reset when the panel had been armed or disarmed. I have had some intermittent programming challenges with schedules not taking or transferring after being updated on the website but thankfully that seems to have been addressed in firmware updates to the panel.

All I can offer is to make sure that your geofence is at least two miles in diameter per ADC recommendations and then make sure that the phone is actually on the cell network when you’re away.

3

Typically the first thing I would check in a case like this is the Z-wave diagnostics and dealer tools, but it appears this is not related to a suretyDIY subscription. Is that correct?

There are a few failure points to check. Based on your description I would hazard a guess that Geo-Fencing is the likely culprit though it could be Z-wave network strength. I think a good test would be to remove Geo-Fencing from the equation and just use arming status.

If you set Smart Away to kick in when armed away and revert when disarmed, does it follow this properly?

Geo-Fencing is a handy feature, but there are some caveats:

Low power battery save mode may in some cases reduce effectiveness.
A phone being powered off when returning won’t register having crossed fence.
Network issues with carrier affecting triangulation.
Carriers not allowing data during voice calls.
Geo-Fence should have a minimum of 1 mile radius. Any smaller is not effective.

All of these can hinder Geo-Fencing. Remember that Geo-Fencing is not GPS, but cell tower triangulation and vector mapping. It is determining when you’ve crossed a fence based on movement, not exact location. If an issue prevents accurate triangulation near the fence, the phone may enter or exit the fence without it being registered.

4

Thank you, J.P. and Jason, for your replies and advice!

Probably not relevant, but just to mention it, this is with a Concord 4 panel. I am using a couple RTS sensors, as well, but these have been working flawlessly, so I don’t think that they are relevant to the situation (at least not as far as I know).

Regarding geo-fences, I have separate ones for “security” and “automation”. I drive the thermostat off the “automation” one, which uses a pretty generous 8.5 mile radius. I have it this way to account for temp setback recovery times, as well as inherent delays with cell phone/Alarm.com app picking up on location/cell tower triangulation updates while cruising home on at highway speeds.

Also, when I configured it to use both geo-fencing AND arming status together, I had similar results. This is why it appeared to me to be more general than something relating to geo-fencing, specifically. I figured that this should have ruled out geo-fencing, since either condition (even alone) should be perceived as a “returned home” event.

I understand the limitations re: battery saving mode, data usage when phone call active on some carriers, etc. I don’t have reason to believe that these issues were factors here, but I will monitor for such conditions more closely in the future, so thank you for calling these out.

Jason, you are correct, this is not related to a suretyDIY subscription. Unfortunately, I don’t have access to any z-wave diagnostics tools and capabilities (would certainly love to have such capabilities, just don’t know if there’s anything I can do in this area, while tied in to my current plan with another provider).

However, I will say that I have 10 repeater-capable z-wave nodes (mostly light switches) throughout my house (covering all extreme outside walls by happenstance). I have also had a network rediscovery performed by my provider since the installation of the last of these devices, with everything in their final location. So, while I cannot rule out the presence of z-wave issues with hard data, things are otherwise working properly, to the extent that I can practically determine.

5
Also, when I configured it to use both geo-fencing AND arming status together, I had similar results.

It sounds like you’ve got most common concerns covered, so you would want to eliminate variables to test.

The best first test would be the suggestion above to completely remove Geo-Fencing and just use Arming Status. If you do this, does the system respond as expected?