Marc, this is a good idea, but unfortunately I don't think I can use it:
The switches are not installed in electric boxes, just frames generally used for low voltage wiring. While this is not insurmountable, more serious is that:
Doorbell transformers are way too large to fit in a single wall (or ceiling) box.
I also have one other concern; namely: is there any mechanism in place preventing high current from being sent over the low voltage wires (if a short should develop in the switch) and causing it to act like a heating coil? I'm not an electronics wiz by any stretch of the imagination, so forgive me if this is me just showing my ignorance.
Someone else also suggested going back to the rotary controllers and making all the connections there. Unfortunately, this won't work either: I did some testing yesterday, and it appears that most lights are NOT connected to these controllers (in fact, the only device that I know for a fact is connected to that controller is our house fan, which I'm planning to remove anyway)
So let me state a new question: is there any wireless system that provides a small inline switching module combined with a wireless single or dual switch? So far, all I'm seeing is these:
X10 Pro:
formatting link
Insteon:
formatting link
Neither system, however, appears to sell a single stick-anywhere switch. The X10 4-way would work in a pinch, I suppose, especially since one of the four is a dimmer and the other two could be programmed as macros...
Not actually a problem. These transformers are designed to mount _outside_ the junction box in the wall cavity with the AC leads inside the box and the low voltage outside. They take up virtually no room inside the box.
See my second post to the other thread (INSTEON over CAT5; was etc)
The National Electrical Code is your friend. This legitimate concern is part and parcel of what selecting a Class-2, current-limited, UL-listed transformer does for you. If you have (as suspect) 18 gauge or larger wire installed for the low voltage, shorting the transformer for too long may burn it out, but the purpose of the code is to protect against fires and electrocution and it does that very well.
1) Not suitable for permanent, only control of lighting. And 2) you have no good way of controlling it that does not depend on the X-10 protocol over the powerlines *AND* X-10 wireless protocol. People tolerate the many limitations/defects of X-10, because in general, one can resort to _manually_ turning lights on and off even if signal propagation over the X10 protocol fails. But going X10 switchless using the inline, AND wireless and would mean that you would be completely dependent on two problematic protocols _in series_ which no knowledgeable and responsible person is likely to recommend (IMO).
Has no wireless switch capability at present.
A disaster waiting (not very long) to happen. See above.
We still don't know what your budget is. From your hints, it seems pointless to discuss Zigbee, Zwave or anything other wireless because they all seem beyond your budget.
My suggestion would be to try the INSTEON dimming option at one or two locations where dimming would be most appreciated, and move the existing relay parts from them to the locations where the relays are problematic/worn-out.
That will cost about $65 per location (less labor) including a new switch box (yes ;-), two INSTEON ICON dimmers, two small transformers and four capacitors. See how that suits you in a couple of locations and take it from there incrementally. Makes sense to wait until INSTEON settles some of its early bugs in any case (but none that I am aware of should significantly affect the small installation described here).
There are less expensive analog control dimmers that I have used and like, but they are Vellemann (google this newsgroup and web) kits from Belgium and do not have UL-or any other US approvals as best I know. Also, they use unsealed/unprotected printed circuit board construction in contrast to the fully potted Solid State Relay Crydom power controllers that I recommended and that you found too expensive.
I use one stick-a-switch in my house, on an exterior wall that has absolutely no chance for having a wire pulled through it. The only way I have been able to get any reliable use of this wireless RF switch has been to put an X10 receiver in the outlet 3' from it. It does not work reliably (about 50/50 for on/off, even worse when holding down the DIM button) in the outlet (on the same circuit) 9' from it.
I've skimmed over the posts on this subject and wondered if you have one or more lighting control centers? These are generally a small metal can with a partition separating the low and high voltage sections. In the GE installations I am familiar with all the lighting loads were home run back to the control panel. The mechanical on/off relays (RR7 etc) mounted in the partition with the low voltage leads on one side and the high voltage connections on the other.
If your installation is similar perhaps you can remove all the relays and install some multi-circuit dimmers like the PCS Scenemaster LM4 (UL approved) which control 4 circuits (total 2000 watts) and is relatively small in size (4x9"). Unfortunately I believe these are out of production - so eBay may be your only resource for them. You might try contacting
formatting link
and see if they know of any units available.
If this is possible, you can then use your low voltage switch wires (existing switches) to signal inputs to an ADI Ocelot/Secu16I combination, or Homevision type controller that can interpret switch presses, double presses, hold, etc. and put x10 signals on your house wiring accordingly. Because the x10 signal only needs to reach the LM4 which is just a few feet away I would expect 100% reliability. The Homevision software has built-in functions to trap multiple keypresses making implementation simple. The Ocelot can be a little slow on reading the SECU16I inputs and is more tedious to program.
If you can make this work, it will be a clean and professional installation that would likely not scare off the next homeowner and, IMHO, not be in violation of any NEC or fire codes.
I don't know where you live, Steve, but the inspectors I'm familiar with would have a canary if you:
a) stripped out the existing low voltage controls for your (AFAIK) already inspected and approved GE lighting system and then:
b) replaced them with an assortment of electronic switches designed and labeled to operate at 110VAC but modified by you to operate at 20 to 24VAC, and then:
c) connected these switches to doorbell transformers as well as adding:
d) four 0.22 ufd 250VAC capacitors via the existing, apparently undocumented, GE low voltage multiconductor cabling.
My feelings about how the inspector would react is just a hunch, though, even though it's based on things like inspectors failing an installation just because the plug-in transformers don't say Class 2. He might not just have a canary, he might give birth to a whole cow!
Perhaps if an inspector has previous experience with your house and great faith in your skills, he will permit you to install the above equipment. Maybe. Of course, if the inspector thinks you're a PITA, he might approve your novel wiring scheme on the basis that some serious electrical shocks might cause a personality improvement! (-:
It's a legitimate question and one the inspector's bound to concern himself with. The National Electrical Code (NEC) abounds with rules concerning the proper isolation of low voltage control wires from the 110 or 220VAC devices they control. The issues in these cases are preventing electric shock as well as keeping undersized (and perhaps unfused) low voltage wires from carrying high voltage and starting a fire.
Connecting a device designed and labeled to operate on 110VAC as a low voltage device would make an inspector very unhappy. Even IF you explained you had wired it so the just the low voltage functions of the switch were in operation. He would be especially reluctant to approve that "off label" use if he got that idea that the switches you installed had already been found to be defective. As you know, Smarthome has admitted to issues with those Insteon switches and is modifying them before selling any more. Bearing that "quiet recall" in mind, he's even LESS likely to permit the use of such switches in a way substantially different than the manufacturer recommends.
That was me. I suggested splicing a connector out of the existing wiring harness going to the master switch as a means to interface with the house wiring from one place (and, restore it easily to its "premodification status").
Well, that sucks. It's hard to believe that they're not connected to anything except the house fan. I wonder whether it was always that way or whether a previous owner already performed some wiring surgery? I've cross-posted this to AHR as well as CHA because AHR has a much greater readership. Alt.Home.Repair also has a lot of electricians and very savvy DIY's. They may have already figured out a way of upgrading the old GE LV system that's in your house.
I really don't see any way of maintaining intellectual control over the upgrade without spending some serious time with a fox and hound tracing set and a multimeter. My experience in large cities and metro suburbs has been that the inspector is not going to approve a rewiring job if he has reason to believe you don't understand the existing wiring that you're modifying.
Yours, unfortunately, is not just a simple case of replacing existing switches or fixtures. Those types of upgrades often don't even require a permit in most jurisdictions, if done by the homeowner. However, that's not you. You're contemplating a fundamental change in the wiring of your house. The inspector will be very much on the lookout for modifications that have the potential to shock or cause fires - and the running of 110VAC through small gauge, low voltage wiring, even accidentally, is just asking for a fire in the walls.
If you search Google for "electrical inspection requirements" among other terms, you'll find that the more complex the rewiring and the less traditional your replacement solution, the more inspectors will demand of you in terms of wiring diagrams and demonstrations of proficiency in things electrical.
Out in the boonies where the chance of an electrical fire burning down your neighbor's house is small, they may look the other way and cheerfully let you immolate yourself. The risks rise as buildings get closer and closer to each other. Places like NYC and Chicago have *incredibly* stringent electrical rules (way above what the NEC requires) because urban fires can spread so quickly. In NYC, for example, you must use armored cable, not Romex. The requirements become even more stringent in multifamily and attached dwellings.
In most jurisdictions, the inspectors have final say. If you don't like it, you don't get approved. If they don't believe you're skilled enough to do what you propose or that your proposal is flawed, you won't get approved. That's why I'm encouraging you to select a replacement that uses approved components in an approved way (although it's something you already seem to know well!). You don't want the inspector to condemn your house while you're still living there!
I believe that once you get used to the four position switches, you'll not want to go back, especially if you program them in "walking order" so that you can turn off the lights behind you and turn on the lights ahead of you from the same switch. As you've already noted, you can also use the extra buttons for macros and my hunch is that they will become far more useful than you realize.
Does your breaker box look like a typical unit? Are there circuit breakers that represent areas of the house or is there a breaker or output for each separate load?
That would be great for what he wants since he could then do PLC system that was wired like a Centralite system but was controllable through standard PLC devices. It really doesn't matter whether the X-10 switch (actually, a self-contained PLC triggered relay) is at the load or at the panel as long as the load is the only one on that branch. I got the impression from following some of the links, that there were several different incarnations of this system. It would be nice to know more about the specifics or see photos of Steve's installation. (Hint, hint, Steve!)
If there is a central load controller, he will still be in good shape because it might be possible to install fairly traditional X-10 inline modules without worrying too much about size. Once again, I think he's got to de-energize the system and spend some time tracing wires with a fox and hound. I can't imagine completing such a refitting without a wiring tracer. Until we have a better idea of what's there and how it connects, we could be giving less-than-useful advice.
That was my thought as well. The low voltage switches would stay at low voltage but the Ocelot changes those signals into X-10 commands for dimming and controlling the various loads. I'm of the mind that the wiring could be as simple as insulation piercing beanies attached to the existing wires and "teaching" the Ocelot to interpret the output of a switch's state. Leave the system controls as they are and have dimming accomplished by X-10 controllers.
It may be only me, but on the one light I do dim often, I'd rather doing it sitting at my desk than at the wall switch. Who knows, his wife might even tolerate a Stickaswitch stuck next to the original LV controls. (-: Especially if you cross three or four big items off of your "honeydew" list before you suggest it!
If Steve moves or tampers with the original switches too much, he could run afoul of other electrical codes. I recall dad installing an oil burner cutoff switch too low for an inspector's taste. Dad wanted to be able to switch the unit while looking inside, the inspector said it had to be at a standard height. Guess who won?
I'm not familiar with Homevision but agree about the Ocelot. It would make the most sense if you can indeed interface to the main GE controller easily. I'd be tempted to use two Ocelots. One to handle the "autonomous" functions of low-level operability of the "translator" and the other to run timed macros, interpreted events, etc. If I had to characterize the most common plea for help in ADI's CMAX forums, it always seems to go like this: "I added something to my program and now everything's gone haywire."
From what I read of Steve's previous comments, he values the "clean and simple" approach in dealing with his upgrade. So, at least IME, do inspectors.
and play with. I think you would be impressed. Then, you can DL this
3rd party software for the HV at:
formatting link
Tinker with the software and I'm sure you'll find some features you've wished for with other systems.
Before the tornado took my house apart I was running HomevisionXL and was exceedingly thrilled with it and the Homevision controller. I had used an Ocelot for several years and always enjoyed excellent reliabilty. My biggest reservation about recommending the Ocelot for this application is the latency in response to inputs to the SECU16. Through experience I can tell you the Homevision unit responds quickly to it's inputs - it is an event driven controller. However, limiting the existing switches to simple on/off, and using something like a maxi-controller for dimming, the Ocelot could easily handle the task.
I have never actually "seen" one of the GE residential installations, but I worked on one in a commercial installation many years ago. They controlled the 277v lighting with the control centers - cans about 16" x 16" - with the typical momentary dpdt switches. The latching relays made it easy to install multi-way configurations to control any load from many locations. Worked great on the fluorescent bay lighting we installed, but lacking dimming I wouldn't care for it in my home.
I always enjoy reading the posts by Marc Hult - he's really a smart fellow - but I couldn't help but grin when I read his elaborate solution involving chime transformers in the walls. I can just imagine explaining to a prospective purchaser how the lighting works ;-) If I ever found myself marooned on a deserted island I would hope Marc would be close by ;-)
I think the simplicity of the GE system would make the modifications easy. A tone tracer and/or continuity checker could quickly identify switch wires. Assuming adequate space in the existing cabinets there should be nothing to it.
Unfotunately, it's probably going to be a couple of weeks before I can do a complete trace of all wiring. Here's what I know so far:
1) Wall switches are installed in frames embedded in plaster rather than full electric boxes, and are horizontal (The frames are horizontal, but the switches are vertical. This configuration is no longer sold by GE). Most plates only have one switch installed, but some have two or three.
2) Latching relays are installed inside ceiling boxes. Low voltage wires do not enter the boxes directly, but instead enter the relays through an entry in the side drilled through the box. There is no central controller. Unswitched 120V wiring runs to each ceiling box.
3) There is three conductor low voltage wiring going from each switch to the ceiling box that it controls. Some relays have more than one switch controlling them.
4) All boxes have a two conductor wire for power from a central transformer. The power wire is daisy chained.
5) Some relays have an additional wire running from them to a 9-way remote rotary controller.
I guess what I was hoping was that there would be something like an RS422 network that I could run on the existing wires. I may need to back to those controllers that Marc mentioned (but unfortunately those don't even provide for any automation).
It sounds like the switches are mounted in "mud rings" which are for low voltage wiring.
I don't think you will find any RS422 system but, if there is any possibility that you can daisychain the low voltage switch locations, you might want to investigate RS485 based systems like ONQ's ALC. There are a few other RS485 based systems, but I think all will require daisychained (or homerun) control wiring.
SOunds fine. Retrofitting a modern AC switch box if necessary would be straightforward assuming stud construction and not plaster lathe over brick or stone with negligible void behind the lather. (Chiseling out brick to create space for new installations can get old, fast.)
It might turn out to be useful that the National Electrical Code has changed so that low-voltage can enter box with AC. As you know, "your mission" is to determine is to whether each AC load is "home-runned" to the entrance box or not. If they are, you could install the very best hard-wired system for lighting automation for the cost of what others lay out for kludged stuff over power lines or RF.
Yes, That hasn't changed.
By "all" I assume you mean both the wall switch and the ceiling, right?
Key here to using RS-485 (assuming that power to the relays in not home-runned, because were that the case you would have other, better options) is the topology of the "two conductor wire for power from a central transformer .. .[which] ... is daisy chained."
If it is truly daisy-chained (connected serially, one after another, with no branches, or spurs) RS-485 control of dimmers in the ceiling is a good possibility.
Even if it isn't the perfect topology to begin with, it might still be coaxed into working for RS-485 (depending in part on baud rate) with a combination of judicious rearrangement (removal of spurs and other non-daisy-chain segments) and use of a electronics to buffer the RS-485 signal and split the line into segments.
Elk makes a ELK-M1DBHR M1 Data Bus Hub For Retrofit
formatting link
(bottom half of page) that is designed to retrofit non-twisted and star wiring to meet RS-485 twisted pair and daisy chain topology. Might be jist the ticket. You would install one where the transformer is now. Is the 2-conductor low-voltage power wire twisted (as in "twisted pair")?. Lots of old 3-conductor phone wire in my other houses has been twisted and would have worked great for RS-485.
If you can clean up the signals enough to use RS-485 (depending on baud rate too) there are numerous options including DMX512 (aka DMX-512 = the world's standard for dimmed lighting in theatres, stage, music venues etc). Or Leviton/NSI Luma-Net (I use both of these) or one of the several other struggling _proprietary_ systems that are "out there" that someone is sure to suggest (from the frying pan to the ... ;-)
I happened onto this thread while thinking about replacing my low-voltage l atching relay system with solid-state relays. I am less interested in home automation than having more reliable relays. My house was built in 1956 and there are two boxes in the attic full of the relays and wiring out. Our ho use also has a couple of the rotary switch, multiple location systems. One thing I am amazed at is that one of these switches has a lighted panel I th ink there's a small neon bulb in there. I know neon lamps last for a long t ime, but I think this is the original lamp and it still works.
You can get replacement relays for the system. They are General Electric RR
-series. The system I have uses RR-7 relays (single pole).
Here are some examples (note the varying prices for them) - I have no finan cial interest in any of these suppliers:
formatting link
itch/
formatting link
formatting link
ghting-control/
Just Google RR-7 relay and you will get quite a few hits.
One thing I have thought of is that each low-voltage switch is basically a momentary contact, single pole-double throw switch. Any low-resistance conn ection across the common and one of the wires will close the circuit to the relay and open or close it. I have wanted timers (yes, I know home automat ion can centralize this) on some of the outside lights, but these are on th e relays. If I wired in an SPDT timer, it should work fine. My thought was to put a jack on the wallplate that parallels the three conductors - I coul d plug in a timer module into that when needed.
Anyway, if you want to keep the original system, replacement relays (and mo unting panels) are available. Some of the new mounting panels (GE "Smart Sw eep") allow for home automation interfaces and still use the RR-series rela ys. These panels are fairly expensive, though. There are also replacements for the wall switches - some with lighting and some keyed. The suppliers wh o sell GE lighting products should have all these.
We now have a mix of direct wired and low-voltage relay switched outlets an d lighting. Very fortunately, the house was wired with copper grounded cabl ing, so all the outlets, even the original ones (though the ground went to the box with a two-prong outlet). I replaced those with three-prong, ground ed outlets and checked them all for proper hot-neutral wiring and grounding .
Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.