Bit of puzzler 3way.... any ideas please

[sythai]

Hi Guys

Recently swapped over few faceplates. Some of them being a 3way lighting circuit.

Customer mentioned today have slight issue, looks like I've got a wire the wrong way round !

• when intermediate switch is in position 1, they 2x other switches don't activate the light.
• when intermediate switch is in position 2, all works fine.

The ways it's been wired must back to a JB somewhere (which assuming will have live/ s-live in there also) which I'm trying to avoid having to hunt down if poss.

All I have at switches:

• switch 1 : 3c TE
• switch inter : 2 x TE
• switch 2 : 3c TE

Odd one one really, sure its something really simple.
Have had it before where strappers at inter are wrong way round and you cant get lights to turn off.

Any ideas please, what I am not seeing ?

Even just had a go trying re-create the fault on my garage rig with no luck.....

Cheers Sy (hot and bothered today ?)

 

[buzzlightyear]

Draw it out on some paper and you will
Find what you are missing.

 

[sythai]

Draw it out on some paper and you will
Find what you are missing.

Believe me I've tried !! Been a long day though ?

 

[snowhead]

Did you witness it working before you started?

Try swopping A and B on this drawing and see what happens

Drawing nicked from Light wiring diagram - http://www.lightwiring.co.uk and modified

 

[buzzlightyear]

Believe me I've tried !! Been a long day though

When the sun goes down then its a long day
You have stiil a few hours yet. Lol.

 

[sythai]

Did you witness it working before you started?

Try swopping A and B on this drawing and see what happens

View attachment 58584

Drawing nicked from Light wiring diagram - http://www.lightwiring.co.uk and modified

Should taken photo of inter before swapping over.... but so basic what could possibly go wrong !!

Didn't prove functioning fully before swapped but not the sort of customer on this one that would be trying it on (they're sound)

Have just tried swapping A and B and on my rig and all I can get is light staying on ?Cant be that many combinations that can be done at and inter switch with x4 cores surely ?

 

[buzzlightyear]

You could go back and try and take a switches with you and start from scratch.
Ones the horse has bolted you need a long rope to drag it back in lol.

 

[Lucien Nunes]

If both 'on' and both 'off' permutations of the 2-ways work correctly with the intermediate in position 2, we know that

• Both strappers must have continuity all the way through
• There can't be any shorts
• Both the 2-way switches must be working correctly

Therefore the problem must be at the intermediate. I reckon Snowhead has the answer; intermediate in position 2, L1 & L2 straight through as normal, all works. Switch it over and nothing is connected through. The other possibility would be a faulty intermediate switch that is not making contact to one terminal in one position, e.g. in position 1 either L1 or L2 has no continuity.

If the wiring looks correct, you could remove the switch and test it for continuity L1-L1 and L2-L2 in one position, and L1-L2 and L2-L1 in the other.

Have just tried swapping A and B and on my rig and all I can get is light staying on

Snowhead's diagram shows 'conventional' 2-way, with PL and SL to the two switch commons. If you swap 'A' and 'B' the light won't stay on, it will behave exactly like the customer's light, because with the intermediate in the crossover position, L1 and L2 will be broken and looped back to the switch they came from.

If your rig is wired as 'conversion' 2-way with PL and SL to L1 and L2 at one switch, then putting the intermediate in the crossover position will cause the light to get stuck on, because when L1 loops back to L2 at the end that has the switch drop, it will bridge PL and SL bypassing the 2-way switching completely.

I.e the results of 'A' and 'B' being swapped will depend on whether the 2-ways are wired 'conventional' or 'conversion' style and Snowhead's drawing shows a likely valid explanation for your symptoms with 'conventional' wiring.

edited for clarity

 

[HappyHippyDad]

Not the sort of problem you want at the end of the day.

When you go back you'll be far more clear headed and I expect you'll sort it out within the hour. I tend to draw everything out, gaining the information bit by bit through continuity testing and detective work. It doesn't help when hidden junction boxes have been used which join different coloured conductors, otherwise it's usually straight forward (as long as it's not the end of the day when your brain struggles more to solve problems).

 

[littlespark]

If you're certain the wiring is correct, then maybe you've got a faulty switch?
I see Lucien already suggested this....

Look for a simple loose connection first, then go in with the continuity tester.

 

[Lucien Nunes]

The funny thing is that I don't think it can be a loose connection as such. If the 2-ways work 100% with the intermediate in one position, then every connection and cable must have continuity. Granted there could be less-than-perfect connections, but at the moment they are making contact and are not responsible for the described fault.

I can't even see how there could have been a wiring error all along that had prevented it ever working correctly. If the wrong cores of the 3C+E had been routed to the intermediate, the fault could not occur exactly as described. Operating the intermediate is definitely breaking both strappers if it goes from fully working 2-ways in one position to no light at all in the other. Therefore both strappers must enter and leave the intermediate and there are only four conductors present so it must be them.

Not the sort of problem you want at the end of the day.

I think it is exactly the sort of problem you want at the end of the day, because it can be understood and solved by logic and a few simple continuity tests. The kind of problem you do not want is one that involves ripping apart something you spent the whole day putting in (and thought you had finished) because the material seems to be defective and won't pass tests. That is soul-destroying.

 

[Highresdis]

Can't you just disconnect the 3 switches and check for continuity between the conductors at the 2 2-way switches? The one that is present at both goes into the common terminal... The others will be the strappers, disconnected at the intermediate.

 

[Lucien Nunes]

It's a good suggestion for double-checking for miswiring and makes perfect sense if it is wired as 'conversion' 2-way, but I'm pretty sure the wiring is correct from the symptoms of the fault. If you look at the possibilities; crossed connections, disconnects and shorts, individually or in combination, I don't think you can produce the symptom exactly as described, other than by a faulty intermediate switch, or the A/B mistake shown by Snowhead (although if the switch is faulty, there could also be a wiring fault that it is concealing, but the combination is unlikely.)

If it is wired as 'conventional', then naturally there won't be continuity between the commons as they are PL and SL.

 

[Mark Wright]

What snowhead posted would be uncommon in T&E.
T&E would be 3 core throughout with the live strapper bridged at the intermediate.

If you did it as the image above you would end up with a single core running back to/from the light at some point etc.

 

[sythai]

Thanks Guys.... really appreciate all your help and hopefully me not going bonkers !! ?

I'm going along the lines of dodgy switch, best I take a spare one along along (Hamilton hartland grid it.)

Not back until next week so will report back then

I've attached a schematic of what I've got..... conversion method which all must end up a JB somewhere.

 

[UNG]

I had an odd one a while ago and don't remember what make where the intermediate switch was configured side to side not top to bottom threw me a bit at the time and needed a check with the multimeter to verify how it was working
The other possibility is if your intermediate is actually a double pole switch that has found the wrong packaging

 

[Lucien Nunes]

Two other good possibilities there that I had not called out specifically, falling into my 'faulty switch' and 'faulty switch conceals wiring errors' categories. A DP switch would give the reported symptom with either 'conventional' or 'conversion' wiring. A side-to-side terminal layout would give it with 'conventional' only; on 'conversion' wiring the light would be permanently on in the 'straight through' position.

 

[littlespark]

Can I be the first to criticise the terminology?

Its 3c&e... not 3c t&e..... ie, a "3 core and earth"... not a "3 core twin and earth"

 

[7029 dave]

Thanks Guys.... really appreciate all your help and hopefully me not going bonkers !! ?

I'm going along the lines of dodgy switch, best I take a spare one along along (Hamilton hartland grid it.)

Not back until next week so will report back then

I've attached a schematic of what I've got..... conversion method which all must end up a JB somewhere.

View attachment 58597

Lets hope it is a faulty switch, with that wiring setup its hunt the JB good luck.
IMO, 2way or intermediate through a JB is bad practice, all connections should be at switch points.
Have you tested all commons/ strappers for continuity.?

 

[littlespark]

Lets hope it is a faulty switch, with that wiring setup its hunt the JB good luck.
IMO, 2way or intermediate through a JB is bad practice, all connections should be at switch points.
Have you tested all commons/ strappers for continuity.?

It wasn’t bad practice back in the 70’S or 80’s It’s only now we’ve realised fitted carpets and laminate flooring causes the problems of locating them.
Just next to the hatch in the attic, or on the landing, near the top of the stairs. The floorboards are possibly cut as access to the jb. And if the boards have ever been up in the past, they’ll squeak or sound loose when walked on.

Replace the jb with an mf equivalent, and happy days

 

[Mark Wright]

Presumably if it goes from 3core to T&E ? there is a junction box somewhere in the middle ? and it might be that the corresponding colours are not connected

also if it goes from 3core to T&E Red/Black how do you know If Red or Black is L1 or L2 ?

 

[sythai]

Lets hope it is a faulty switch, with that wiring setup its hunt the JB good luck.
IMO, 2way or intermediate through a JB is bad practice, all connections should be at switch points.
Have you tested all commons/ strappers for continuity.?

Was only a passing visit yesterday.... just had a go at swapping strappers around in every combination at the inter. Calling back next week with spare switch and ready to buzz out fully if so
[automerge]1591192532[/automerge]

It wasn’t bad practice back in the 70’S or 80’s It’s only now we’ve realised fitted carpets and laminate flooring causes the problems of locating them.
Just next to the hatch in the attic, or on the landing, near the top of the stairs. The floorboards are possibly cut as access to the jb. And if the boards have ever been up in the past, they’ll squeak or sound loose when walked on.

Replace the jb with an mf equivalent, and happy days

Luckily landing carpet is only loosely down (being replaced) and have spied a few screwed boards already ?️‍
[automerge]1591192621[/automerge]

Can I be the first to criticise the terminology?

Its 3c&e... not 3c t&e..... ie, a "3 core and earth"... not a "3 core twin and earth"

My bad ?

 

[Lucien Nunes]

also if it goes from 3core to T&E Red/Black how do you know If Red or Black is L1 or L2 ?

It doesn't matter which is L1 or L2 or how many times they cross over. Provided all the strappers are connected in some way, it will work. Unless you can find a flaw in my reasoning above, the fact that the 2-ways work correctly with the intermediate in one position proves that all the strappers and junction box connections are present and correct, i.e. the problem can only be within the intermediate switch itself. If anything else were wrong, one 2-way switch would disable the other either on or off in one position.

Unless I've missed something... please try to prove me wrong

Update: It's a challenge. Draw me a possible circuit showing a fault other than wrong connections at the intermediate or open-circuit contacts inside it, in which the 2-ways work correctly with the intermediate in one position, but they cannot turn the light on at all in the other.

 

[7029 dave]

It wasn’t bad practice back in the 70’S or 80’s It’s only now we’ve realised fitted carpets and laminate flooring causes the problems of locating them.
Just next to the hatch in the attic, or on the landing, near the top of the stairs. The floorboards are possibly cut as access to the jb. And if the boards have ever been up in the past, they’ll squeak or sound loose when walked on.

Replace the jb with an mf equivalent, and happy days

Yes I was not talking so much about JB methods in general, wired many lighting circuits this way, especially with occupied rewires, but was talking about the wiring of strappers/ commons for 2 way etc, no real reason not to have connections in the switches.

 

[Mark Wright]

Still not sure how it could go from 3core to T&E can someone draw it out?

I think it would make a difference because you would end up with L1 and L2 the wrong way round on the final switch and that would produce a problem like you say.

 

[7029 dave]

Can I be the first to criticise the terminology?

Its 3c&e... not 3c t&e..... ie, a "3 core and earth"... not a "3 core twin and earth"

Or tripple and earth

 

[littlespark]

I think we're getting away from the original question here. We've all given advice, and the OP will investigate when he returns to the job.
Hopefully something we've said will be spot on.... no point rambling over and over

 

[Mike Johnson]

Bloody ramblers they always leave the gate unlatched.

 

[Timbo]

Thanks Guys.... really appreciate all your help and hopefully me not going bonkers !! ?

I'm going along the lines of dodgy switch, best I take a spare one along along (Hamilton hartland grid it.)

Not back until next week so will report back then

I've attached a schematic of what I've got..... conversion method which all must end up a JB somewhere.

View attachment 58597

Just a point on your diagram - red was generally used as common connection, yellow / blue to L1 / L2. Have you checked continuity of cores to intermediate from each 2 way? Sorry if I've missed it.

 

[7029 dave]

Just a point on your diagram - red was generally used as common connection, yellow / blue to L1 / L2. Have you checked continuity of cores to intermediate from each 2 way? Sorry if I've missed it.

Common was either Red or Yellow round here, back in the day.

 

[static zap]

A DP switch would give the reported symptom with either 'conventional' or 'conversion' wiring.

.. I would be removing Intermediate sw first for some simple
continuiity checks in both positions !.
..makeing sure the world has not gone sideways ,- DP only -or- half a cross ! ..

 

[Mike Johnson]

Worst nightmare is a defective new intermediate switch, not once but twice on two separate installations.

 

[telectrix]

Can I be the first to criticise the terminology?

Its 3c&e... not 3c t&e..... ie, a "3 core and earth"... not a "3 core twin and earth"

Pedant.

 

[andyb]

One of the conductors at one of the the 2 way switches will be live when all the concuctors are disconnected. Surely this will give you the colours of the strappers and common on the other 2 way swtch.
Providng this is true, it should be easy to show whether the problem is in a switch or the jb.
I suspect the problem is in the wiring of the switches.

 

[sythai]

Well thanks Chaps for all your feedback and pointers along the way...... at least I'm not going fully bonkers after doing this for the last 20 yrs !!?

Went and sorted yesterday morning, couldn't handle the suspense anymore.

..... it was a dodgy intermediate, first time for me ?

 

[Mike Johnson]

As Post 32, happens more than we think.

 

[Lucien Nunes]

Glad it was an easy fix (and that I was the first to suggest it). As I mentioned in that post, either the intermediate switch itself or its connections were the only suspects if there was only one fault location, which is probable on something that was thought to have worked originally and did not go bang.

This fault is a good demonstration of a specific troubleshooting tactic: By all means consider the possible impact of suspicious circumstances (in this case, hidden JB, change in cable colours) in parallel with logical analysis, but if the two conflict, give priority to the logic. It's reassuring to potter about testing strapper continuity while muttering how the numpty used the wrong colour for the common and how the junction box should have been more accessible, but it was never on the troubleshooting pathway because the original observation (if accurate) proved conclusively that all connections were correct and functional everywhere but in the intermediate.

If that inference doesn't leap out at you intuitively, draw out the two possible circuits (conventional, conversion), simulate the plausible single wiring errors / faults / switch malfunctions (not all of which apply to both circuits) and see the pattern of results, mentally filling-in the 8-line truth table for the light according to the switch positions if necessary. There are about twenty realistic different faults / mistakes, not counting the mirror images of those (e.g a single broken strapper could exist in four different places, but due to symmetry the effect would be the same in each case and only one need be considered.)