apostolakisl

This is a good question. I have only had one all on event, so perhaps I am not most qualified to characterize it. However, if you use if control . . . terminology, an all on event should not be an issue. I can tell you that I have the following program If control kpl button is switched on Then arm elk night mode and my all on event did not trigger that. I also have several other if control programs that do things like turn on the house music and what not and none of those ran either. I would be more careful with "is status" programs since I suppose that could trigger. I do not recall if ISY registered the change in status of all the devices or not during the all on event. I want to say yes, but I will not swear to it.

If the lights pull 480 then they are using two phases of the three to get that. You would then need a relay. I have never heard of single phase 480.

You are probably wrong to say what you just said. If your maintenance guy says you are on "480v" he probably means you are on 480v three phase. This means you have 3 phases of 277v. Cross two of those phases and you get 480v. Cross one phase to neutral you get 277v. If the lamps run on a single phase of the 480 service then the switch I listed above is specifically made for that.

I would think that listing 277v means he is powering a single phase 277 device off of a 480v 3 phase wye service, not a 480v device and only interrupting one leg of it. The part about the load being inductive or resistive is good. Do they make 277v single phase motors though? Seems like anyone with 277v/480v available would want to take advantage of 3 phase since it would be better and cheaper. But we are making lots of assumptions here. Would be good to know the truth.

http://www.smarthome.com/in-linelinc-relay-insteon-2475sdb-remote-control-in-line-on-off-switch-dual-band.html This is what you need, 277v 20 amp.

Just an FYI. I am on my 3rd PLM. The first was the older model which I replaced about 4.5 years ago because. . . I don't remember why, the old PLM worked fine but it seems to me there was some issue with using the 2412 unit, maybe since it wasn't dual band. Anyway, the new 2413s that i bought 4.5 years ago lasted 2 years and 3 months. . . as you would expect from Smarthome. I replaced it with a new one and recapped the old one as a spare. Well the new one died at 2 years and 2 months just last week. In short, I have had 3 plms over the years and only one all on event. This was using my third plm, the one I bought 2 years and 2 months ago and while it still had its original caps. This happened about 2 years into its life (or in other words about 2 months ago). Could not quite dead yet caps be part of the problem? I restored that 4.5 year old plm with the new caps last week and it is working fine. I'll be recapping my just died plm as soon as they arrive from Mousser and keep it as a backup.

The trouble with the all on is that I don't really think anyone actually knows for sure why it happens. It may be that there are multiple things that can do it. Which means you may think you fixed it, but there is no way to test that. I have had exactly one all on event in my system in 7 years. Nothing changed for a long time before it happened and I have changed nothing since. So how do you trouble shoot this and how do you know when you fixed it. I just don't trust Insteon to anything significant, I use Elk for all that stuff.

If I were you I would get rid of the Insteon control of your GDO. Since you have Elk, use that, it is too important to trust your GDO to Insteon. I neglected to pull wires from my Elk to the GDO when the house was under construction, but there is a simple work around provided your Elk is within radio range of the GDO. I simply took one of my standard push button radio control GDO with 3 buttons on it and soldered leads to the 3 micro switches and then ran those to 3 relays on the elk. If you care to have ISY control your GDO and have the Elk module, this is now just a simple process of adding a program that turns the relay on for a second. ISY is not the problem, it is Insteon, and this takes Insteon out of the loop and replaces it with the far more reliable Elk.

I have one and I assume it works, though it is hard to know what you didn't have happen. The one thing to realize is that surge suppressors only suppress surges originating outside the house, they don't block surges that originate from within the house. Specifically, when motors (like your refrigerator compressor) lose power they produce a surge. Normally this is not a problem because normally the reason the motor shuts off is that power to it is disconnected from the house power, but when the power to the house is cut, the connection between the motor is still connected to the house when it shuts off.

Either that or you can put programs into folders that are activated/deactivated based on that variable. I also use the status of my elk. Some things are based on whether I am home, others on whether my wife is home, and others yet are based on if anyone is home.

Yes. Just don't get shocked. If indeed one of those wires that you connect at the switchlinc box is hot at the fixture, then use this wire to connect to the red (load) lead on the Insteon switch and the other end connects to the hot on the light. You stated you had neutrals already at both boxes so connect the fixture and switchlinc neutrals to the neutrals already there.

Tasker sets a variable on ISY for me. When my phone loses it's wifi connection to my home wifi it sets the variable. ISY runs a delay on that to ensure the lost wifi was me actually leaving and not just a momentary issue with the wifi.

I use my bunghole strictly for O, no I's.

Yes, put the black wire on the insteon switch to the blues. Well one trick is to do the ohming out to find a wire that would work for the load. The other thing you can do is connect the other wires to your hot one at a time and look for one of the wires at the fixture to be hot. When you find one, connect it to the load wire on the insteon switch and connect the end at the fixture to the hot on the light. Just cap the wires at the other box and any unused wires at the other two boxes. If there are multiple neutrals, keep them spliced together. Otherwise, cap the rest separately.

This seems the most simple http://www.ebay.com/itm/Ultrasonic-Range-Finder-Maxbotix-LV-EZ4/321499270140?_trksid=p2047675.c100005.m1851&_trkparms=aid%3D222007%26algo%3DSIC.MBE%26ao%3D1%26asc%3D36142%26meid%3D3dbab5dd4c6b442ea159c1ec52a39642%26pid%3D100005%26rk%3D4%26rkt%3D6%26sd%3D311139739778 An ultrasonic range finder mounted in an opening in the top projecting down onto the fluid. It will output 10mv per inch, so if you have an AD converter that you can interface with ISY you can post the depth of the fluid (total depth minus what this device outputs).

Are you proposing a float in the tank that transmits to a rod that exits the tank, pushes on the long arm of a lever that translates the full range of fluid in the tank into 5 inches?

OK, then no need to unhook the spliced blues. They obviously have daisy chained the hots through that box to get power to some other box for something else. If you did take them apart, you would need to resplice them since presumably something down stream would go dead if you didn't. If I understand correctly, your hot blues are in the box that you are putting the switchlinc. If so, all you need now is a wire between that box and the fixture. This is your load.

Yes, you could leave them together and add your switchlinc hot to the mix. Do your other boxes have any blue wires? Are they also hot? If not, then this means the hot has entered that box on one of those blues and then got spliced to the other blues to deliver power to something else unrelated, like a different light or outlet.

If you disconnected everything and now you get power on 1 wire at one box, you know where your hot is entering the system. If that hot is in the box where you want the switch, then that part is done. If not, you need to find a path to get the hot to the switchbox. This is where ohming out the other wires comes in. You find a wire that goes from hot location to switchbox, then you splice them. You might need to go through the fixture box to get there depending on how it was wired in the first place. Then you need a wire that goes from switchbox to fixture for the load. You already have neutrals at all locations as I understand. So at this point your done. Process to follow 1) Cap and mark your hot 2) At the switchbox, splice one of the wires (not the hot) to ground/neutral 3) Go to fixture box and test each wire to ground. If one of those wires has 0 resistance, then you can mark off that wire as the load wire. 4) If none of the wires ohm out to 0, then go to the other switch box and find the wire there. label each end the same for future reference 5) Repeat for the other wires in the primary switch box. At that point, you might be done provided you can get hot and a load wire to the switchbox. If not, you need to go to the other switchbox and do the same process. Once you have all the wires labeled, you connect what needs to be connected to get a load and a hot at the switchbox.

You could use a freescale pressure transducer tapped into a line that is at the same level as the bottom of the tank. http://www.digikey.com/product-detail/en/freescale-semiconductor-nxp/MPX5010GP/MPX5010GP-ND/464055 Connect that to an AD converter and post to a variable on ISY. The CAI webcontrol would do that. I would expect the 0 - 1.45 psi range to be adequate for the tank. 1.45 psi. is 3.3 feet of water and fuel oil I assume is lighter than water, so you should get a bit more range, maybe 5 feet. If your tank is a barrel shape on its side you'll need to do some math since the depth of the oil will not be linearly tracking to volume. Also if you have access on the top, you could use an ultrasonic device to measure the level of the oil. Again, a barrel on its side will need some math.

I'm sure conduit as a ground is electrically sound, but it isn't code around here. Grounded conduit or not, you are required to have a ground wire here. It really doesn't matter. You just need to figure out which wire at box a is the other end of the same wire at box b, and so on. While I'm sure that many of us who have screwed around with these a bunch could do this without unhooking everything and ohming it out, it is kind of hard to figure this out any other way when using a forum as the only means of communication. Keeping the eye on the ball, you need a hot, neutral, and load wire at the switch box and at the fixture you need the other end of the load wire and a neutral. Anything else is capped. The only trick is that it is possible that one of those wires may need to be spliced through at the unused switch box.

You have to unhook all of the wires. If you are testing to a screw, it is not unhooked. This includes the wires at the fixture. I have to say I have never seen residential use conduit to all of the boxes and I thought code called for grounds for a really really long time (even in the presence of grounded conduit). Just to be clear, your going to have to unhook everything anyway at the two switchboxes, you are only adding work by unhooking and then rehooking at the fixture. And there is a good chance that the best way to wire this would have you changing the wires at the fixture as well.

It is very doable to not put a switch at your second box and have a single Insteon switch at the primary box. First, breaker off. The easiest way to tell you how to do it is to unhook all the non-ground and non-neutral wires (leave the grounds spliced together and the white's spliced together . . . but not to each other of course). Test that your whites are actually neutral by ohming between the whites and the grounds. Should be ~ 0 ohms. Turn the breaker back on. With your volt meter, find the hot by touching each candidate wire with one probe and the bare ground wire with the other probe. The hot will only be one wire at one of the three locations. breaker off again. If your hot is at the box where you want your switchlinc, you can call that part quits. If not, you need to find a traveler wire that goes between the hot location and that box. Splice each non-ground wire one a time to your ground at the switchlinc box, then go to the other two boxes with an ohm meter and cross each candidate wire there with the ground wire in that box. When you find the wire with ~ 0 ohms of resistance, then you found the two ends of that wire. Mark it. Do this until you figure out enough wires to have the following 1) Hot at the switchlinc box 2) A wire that can get the load to the light It is quite possible that at least one of these wires will go through the other box to get there. It sounds like you already have neutrals at every box, so that part is easy. Cap off all wires at the unused box and put a blank plate on it. Maybe use a sharpie and write on the backside of the plate what you did for future owner of house.

Yes, that is the trick. Fortunately the first time I tried to replace a device it wasn't in a folder and I was very pleased at how simple it was. Later on I was like "where the hell is that option?". Only to figure out that you need to have it in the root folder. But without the first time success, I may have missed it all together.

This is just to find and replace one currently enrolled device with another currently enrolled device in programs that reference that device. I don't think there is any role in using this when replacing a dead device unless you were to manually program the new device identical to the old device first and then swap it with the old device. If you want to replace the name of a device with a different name, you do that in the main console tree by right clicking and renaming it. The new name propagates through to all the programs.