Someone who likes puzzles will probably come along in a minute and unravel this for you.
But personally, I would tackle it by disconnecting all the wires at the two roses and splay them out so that nothing is touching. On a good day, I might mark up the original connections, but then again I might not.
Then I would use a volt stick (or neon screwdriver or DVM) to locate the live incomer. You need the mains on for that. Mark that one with red tape, then turn the circuit off for everything else.
Figure out which rose wires go to which switches using a DVM in "ohms" mode.
There may be a pair of spare red and black wires which don't seem to be part of the circuit. They will be the ones which take unswitched live and neutral on to the next rose.
If, when you disconnect these roses, all your other lights still work then they are not feeding any other lights.
If you find some of the black wires have a red sleeve on them, God has been kind to you. The electrician is showing you where the switched lives are. (When I say red and black, I am using that term to include brown and blue, if the wiring is sufficiently recent).
With any luck you might find that the two way part has been wired up with three core plus earth cable, which also makes the diagnosis much easier.
Oh, and draw a diagram of everything as you begin to sort it out, you will get something wrong if you don't.
Your description of the effects applies irrespective of whether L1 abd L2 in S2 are swapped or not. Two single pole switches in parallel behave in the way you describe. When one switch is open the other one does the toggling. When one switch is closed the lamps are on and the other switch has no effect.
John Rumm wrote in news: snipped-for-privacy@brightview.co.uk:
Sure. I mean the switches should be in series. I left some of the old wiring in place and am trying to work out is wrong. I tested all cables coming from both switches. I thought I did it as shown on that page. I did this before in a garage without a problem. Something strange is going on with those old wires.
"Dave Plowman (News)" wrote in news: snipped-for-privacy@davenoise.co.uk:
I'm not sure I understand what you're getting at. Surely if you take the two cables that connect the switches the only effect of swapping L1 and L2 around at one switch in a properly working system should be that the orientation of that switch for a given orientation of the other switch and lamp status will change?
And next time you won't remember it of course! I had a lot of head scratching when I put on a timer on off switch in place of one of the 2 way switches once. I mean what could possibly go wrong I thought.. Ahem. Brian
Which is why you need to disconnect them at all accessible points. If there is a possibility of another junction box somewhere making connections other than a simple straight connection (for example a "Y" connection) you should eventually be able to infer it, although you might have to do some testing (and recording of results).
As an example, if you have *everything* disconnected then at one location, temporarily connect the live and neutral of one cable. Then you can positively identify the other end of that cable because it is the one where there is no resistance between live and earth. Don't stop testing when you think you have found it, if there is a Y connection (as there might be if a live is connected to two roses because it makes a more convenient cable run than a traditional loop-in) then a second cable will be found to be shorted.
It would seem that one of the conductors between S1 and S2 is connected to a 'live' feed from somewhere else which would give you the condition you describe.
I meant actually go to L1&2. Not one of them swapped with a common.
What sort of wiring is used? There are two common ways for two way switch wiring. One is to run a triple and earth back to where a single switch would be, ie with feed and load wiring there. The other is to run straps between the 'L' terminals, with possibly load and feed at different ends of the circuit. And if using TW&E, possibly some spares. ;-)
"dennis@home" wrote in news:599bfe3d$0$60096 $b1db1813$ snipped-for-privacy@news.astraweb.com:
OK. I have already tested what I thought might be an offending switch by putting a new switch in its place, but that didn't affect what is going wrong. There is probably something strange happening in a small section of old wiring I left (having fully tested it before doing anything!), which comprises two lengths of 3-core coming from that switch ({L1,L2,earth} and {common, unconnected, earth}). The second possibility is that there is something wrong with the similar wiring I left (again fully testing it) from the other switch. Those are the only sections I left in place other than the mains feed. The third possibility is that I failed to follow my own plan properly, so I will check everything I've done.
First I will disconnect the cabling to the first switch and replace it with a switch and test cabling. Hopefully that will solve the problem and I can then put the new cabling in place. If that doesn't work, I will recheck everything I've done. If I haven't made an error, then the problem can only possibly be the cabling to the other switch (that switch is new and works fine, and again I tested the cabling fully), which will be a nuisance to replace but it will have to be done.
The circuitry is a little more complicated than I showed. The two lights are on the ceiling of the garage and there is also an outside light in parallel, controlled by its own switch. That didn't work before, but now I have done what I've done it works fine! So the full circuit I want is this: .
HomeOwnersHub 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.