oberkc

You stated that you were "able to successfully link". Do you not see this one on your admin panel?

1) what is the load connected to the switchlink? What other devices are plugged into the circuit that powers the PLM and switchlink? Yes, using the ISY-99 introduces the possibility that you are not compatible with some of the newer devices. 2) I have found that nearly all device configuration can be done throguh the ISY. What do you see when you select the device from the admin panel? Is there an option for device settings along the bottom of the admin panel?

Yes. That does seem more obvious now. That is embarassing.

Make sure, also, to confirm your instructions match the device. Older "access points" (now range extenders) did not have red or green indicators. If your instructions describe "bright white", they are likely for the older versions.

I thought this most likely, but some refer to "geofencing" more generically.

IOS or android?

I repeat, then, my suggestion to use an extension cord to temporarily plug in your PLM to a different circuit. If this helps, I suggest a filter for your UPS.

Yup. Me too. It has outlasted several cell phones, TVs, tablets, chumbys, amd TV streamers. It remains one of the more valuable and interesting pieces of technology I own.

There was a lot of news traffic many years ago about some military operations on the same frequencies as many garage doors of the day. I assume you are correct that later doors are less immune to any interference that may have existed. Unfortunately, I am unaware of the age of pyrorobert's opener. Or it certainly could have been other causes.

I am, by no means, fully knowledgeable about these logs, or your device naming conventions. What I can tell you, however, is that the relay will not transmit status...only the sensor does. I expect that your log was showing a change in sensor status. Based upon the name alone, do you believe this log is showing a relay command?

Make sure, also, that your power strip is absolutely passive...no surge suppression...no noise filters...no UPS. This can degrade insteon communication.

Another consideration is the apparent relationship between garage door and security system. If the security system can be disabled by nothing more than a garage door opening, I wonder if this is providing the level of security one I hoping for.

I offered the RF theory as one plausible explanation (I have no actual evidence to support the theory)...mainly to consider the possibility that, sometimes, things happen for reasons beyond the control of insteon or the ISY. My experience is that if something happens because of the presence of the ISY, there is evidence in the log. If no log evidence exists, most likely the problem is elsewhere.

Spurious radio signals from unknown sources that the opener mistook as a remote control? Garage door remotes, I understand, are on shared frequencies. It is not unheard of that a door opens for no apparent reason.

The extension cord test was meant for the PLM,by the way. I apologize for the typographical errors.

I would not expect lack of ground to be a factor in quality of communications. Metal box, however, could impact RF range, if applicable. Consider, also the devices plugged into heame outlet and circuit as the PLM. These devices could interfere with PLM ability to communicate.

As LeeG described, there are only two "legs" (aka "Phases") in a house. For these purposes, I would use the term "leg" and "phase" synonomously. I tend to use "circuit" when talking about all devices on a given breaker. I started to suspect we were using terminology differently, so I am glad you cleared it up.

I am certainly curious how you know they are on the same "leg"...and that others are not. What is the load attached to the troublesome switch? Sometimes it seems that there are enough performance variances between a given device that one may work and another not.

I would use the PLM to initiate the phase detection mode. Go to smarthome.com and find the quick start guide. Min the guide, find the section titled "Use PowerLinc Modem as a Phase Bridge". Follow those directions. Let us know the results.

If there is a mismatch between the isy status and actual device ststus, I would first suspect a comm problem, likely beteen the PLM and the keypad. Have you confirmed communication between the legs of your electrical system? How may dual-band devices do you have? Do you have any range extenders? To test for this theory, choose one of your button scenes from the admin panel. Turn the scene on and off. Does the button status reliably change? Try a scene test on the button scene. If not reliable or fails the scene test, get an extension cord and plug it into a different outlet on a different circuit. Minto the extension cord plug the PLM. Doe the scene respond reliably now?

What is blinking green? The device you used to initiate the test, or the responders? What other dual-band devices do you have? I, actually, don't see the four-tap method mentioned in the user manual for the switchlinc 2477D. A blinking green LED suggests that it is linking mode as near I an tell on the user manual. Is your PLM dual-band? I would focus on confirmation that you have communication across the legs of your electrical system. Identify your dual-band devices. Find one that can perform the test, and confirm that you have another dual-band device responding appropriately (depends on model and era). Once confirmed, we can deal with inability to find the device. It is very possible that you cannot find a device because there is no communication between the electrical legs (or phases, if you prefer).

3) have you tried the 4-tap test? Why are you unable to initiate? Is it possible you are having difficulty tapping four times rapifly enough?

Choose the baulky device on the admin panel. Right click. Show device links table. Compare when done. If there are any mismatches, restore the device.

Cycling power does NOT cause a factory reset. One needs to depress the button on the device for a short period to reset it.

This is only temporary...a troubleshooting technique...to confirm or eliminate the possibility that the load is causing the problem. If you remove the load, and the scene tests now pass and the switches stay synced up, then this is a good indication that the neon lights were causing the problem and a filter is needed.