Insteon after circuit changes?

[Mike]

This last weekend I had some circuits cleaned up (split 2 circuits into 5 to finish the upgrade to the panel done almost two years ago).

There are Insteon devices on all of these. I've been using the elk to control the outside lights (two switches linked together) and a few internal based on rules. This has been reliable (I saw them not turn off once and not turn on another time, but that was before additional switches were installed and the new firmware updated, after which I have seen it be reliable). The light that went on has only one link: to a control linc in the living room.

What happened today was the outside lights went off early (a few hours) when I went to turn it on, my wife commented that my office light upstairs was going on, I shut off the outside light and the office light went out. I then tried to repeat it but the 'wrong light' did not turn off or on again.

I know the lights were working properly last night (I've been checking everything as I've been moving all lighting and thermostat control to the elk).

I'm going to watch it over the next few days.

I was wondering if it made sense to relink the signallincs or anything like that. I was wondering if anyone heard anything like this, I am sure changing existing electrical circuits was not a typical test scenario. I am wondering if it is a temporary bug since the 'mesh network' has been reconfigured based on the circuit changes. Not sure if this makes sense at all (or is just a random bug).

I have since went into the web interface for the elk and tried out controlling the linked lighting, everything worked flawlessly (on/off). I have not double linked any lights yet for direct status reporting via the elk yet.

I'm going to watch it over the next few days. I'll post what I see. I'm guessing whatever it was will have been temporary (it is surprising that I did not get the issue yesterday if it is related to the circuit changes so perhaps it was random). This may show no further issues (and my assumption on the circuit change may not be accurate), but I am including the information in case it may be helpful.

 

[Mike]

Ok, I think the circuit change may not be related.

I came home today and my wife informed me my office light (the light that was coming on when I was flipping the switch but then couldn't reproduce it) was on.

I setup rules to trigger every minute to reproduce the same lighting sequence but it all worked properly.

The light going on improperly only has a link from a controllinc and nothing else.

Something else occurred to me though: I had an electrician in to put in a few of these switches as I needed a neutral in a few spots where there were three way switches. This was just in the last few weeks. In the case of the front-outside light (which is one being scheduled to turn on) there was originally an issue and when the traditional switch was replaced with an insteon one and was hooked up it would blow a GFCI outlet in my office (upstairs). He found out the neutral was not the 'same neutral', so the GFCI detected that and shut off the circuit.

His recommendation was to remove the GFI outlet, and noted there should be no sideeffects, and my computer equipment was on surge suppressors/ups equipment in that room.

I find it interesting that the neutral is shared from that foyer light that is being turned on to another circuit upstairs (my office I believe) and that light happens to come on unexpectedly (but repeatedly now when my wife comes home, so after a command to turn the foyer is given or at least that is a possible trigger).

Does this sound possible that it could be causing my issue? Should I be worried about what was done (it seems I might want to have a new circuit run if it will be an issue, at least if I want reliability)?

 

[Madcodger]

I'm not an electrician, but I'm almost certain that there should NEVER be a shared neutral between two separate circuits. It could easily lead to having a "hot" circuit even after you think you've shut it down via a breaker, creating a dangerous situation. Will it "work"? Yes. Is it safe and within code? Again, I'm not an electrician but I'm almost sure it's not.

I could see the electrician re-routing the entire light to a different circuit, but if his solution was to share just the neutral between circuits, PLEASE have someone re-check this guy's work.

There'a s similar thread to this on the Homeseer board right now, and I think you'd find the folks there in agreement. I don't have a solution for the light, except that you may need to have it rewired based on this situation.

 

[Guy Lavoie]

Shared neutrals are ok (and are often used for kitchen counter outlets where each half of a duplex outlet has its own breaker). The important thing is to use mechanically linked breakers in this case, so that both circuits sharing that neutral will always get turned off together whenever you work on it or it trips.

 

[Mike]

Madcodger > I thought the same originally, and spoke to someone else I knew, who didn't see it as a big deal.

Guy > That is a great point. I have him back in this weekend so I'll check on that.

If any one has any ideas on if that shared neutral could be the cause of my office light going on unexpectedly I'd appreciate it.

I'm wondering if it makes sense to look into having that issue corrected (although I suspect it means running a new circuit to the foyer and ripping the walls open).

 

[upstatemike]

Shared neutrals are ok (and are often used for kitchen counter outlets where each half of a duplex outlet has its own breaker). The important thing is to use mechanically linked breakers in this case, so that both circuits sharing that neutral will always get turned off together whenever you work on it or it trips.

The other reason to have them mechanically linked is to insure that each hot is on the opposite phase. Separate breakers could be re-arranged such that both hots end up on the same phase. This would cause return current from the two hots to add together instead of cancelling each other and you could easily exceed the current rating of the neutral wire.

 

[Mike]

If anyone is interested, I had a controllinc plugged into the downstairs circuit of the shared neutral. That controllinc had button 1 linked to the upstairs office light that was having issues. I'm trying to test to see if the controllinc could have anything to do with my unexpected office light issues.

I have unplugged it for now to see if the symptoms continue (the office light turning on). If they do not, I will plug the controllinc back in and try different links (to other lights) and see if I can reproduce the issue. The one link that was in there was just to test with (there was not a real reason to link it upstairs) so removing it is not a big deal. I will try to isolate the issue.

If the issue continues even without the controllinc plugged in, I plan on resetting the office switch. Only step after that I can think of is to swap it and see if the issue continues (to identify a possible faulty switch).

I will post my results once I get further information.

 

[Mike]

The issue continued with the controllinc removed. I had reset the switch that was turning on unexpectedly (factory reset). Right afterwards I had some additional electrical work done, among that was including another switchlinc dimmer and fixing some outdoor outlets.

I had forgotten about the additional switchlinc dimmer, but now things seem to be fine. No more office light turning on when it is not supposed to.

Perhaps the 'buy more switchlincs' does reach a critical mass.

That puts me at 16 switchlincs, 4 signallincs, 2 controllincs, and a lamplinc currently. Not sure if it was the switchlinc, but I have not had a reported symptom for a few days now.