70s house, aluminium conductors

[Pretty Mouth]

I visited a client about a week ago, an elderly couple complaining that their lounge, hall and landing lights were all flickering simultaneously with a noticeable buzzing noise. When I arrived the lights appeared to be working ok (I asked them to switch them off until I could attend), so an intermittent fault. Opening up accessories revealed the likely culprit - aluminium conductors, the first time I have come across these.

Some observations:
House built 1973
1mm² (or so I believe, next time I'll measure w/ micrometer) single solid strands
Dull silver appearance, brighter when cut.
Not copper clad aluminium, not tinned copper.
Very soft and supple compared to copper.
Higher resistance values vs copper
Prone to breaking off where screwed at the terminations.
It was hard to get a clean reading by testing continuity with the probe placed directly on the bare conductors (CPCs were unsheathed), the reading would jump around constantly in the range of a couple of ohms. I presume this is due to the insulating oxide layer aluminium is said to form. Clipping a jump lead to the conductor and testing to that gave a steady reading, as did testing to the brass terminal where soundly terminated.
Other circuits in the house are, at a glance, aluminium also.

I did what I could in the time I had available - I went through the circuit (downstairs lights) checking and tightening everything, found several conductors broken or very nearly broken at the screws so reterminated those, and ended up with satisfactory R1+R2 vs RN+R2 results. Borrowed neutral on landing light as you would expect for a property of that age.

Anyway, received another call from them on Christmas eve, they had noticed a small flash and pop! from one of the bedroom wall lights when switched on. Also switching on say the main bedroom light would cause a momentary flicker on the landing lights. He also told me they had an intermittent problem in the past with the bathroom light flickering, and certain sockets sparking when switched on. Advised not to use dodgy lights and sockets until I can get to them in a day or 2.

The place is obviously due a rewire, and that's what I will recommend. In the meantime, I'll need to make good any problems. I understand standard accessories are not to be used with aluminium conductors. Apparently some Wagos can be used on aluminium conductors when used in conjunction with the Alu-plus paste:

Syringe (249-130) | WAGO GB

Syringe; Contents: 20 ml Alu-Plus contact paste (4017332133890) Online at WAGO GB!

www.wago.com

I was thinking of using this to connect a length of copper conductor to the ends of the aluminium conductors, then terminating the copper in the accessories. I have reservations though, I suspect the 1mm² aluminium will be too soft and bendy to force into a wago 773.

What have other members of the forum done in these circumstances and what would you recommend? As always, all information gratefully received

 

[Pretty Mouth]

Interesting I note that the metal back boxes have plastic threads on both sides?

Yes, all of the lighting back boxes I opened had them. Have see them a couple of times before, similar aged properties

 

[freddo]

Very common for installations with wiring or switch drops with no CPC. Insulates the plate screws from the back box.

 

[lozarus]

Very common for installations with wiring or switch drops with no CPC. Insulates the plate screws from the back box.

Yes i figured as much. In this example it could make for a nasty swap if someone were to fit a class 1 switch and not link the CPC currently terminated in the back box to new metal fitting. Interesting stuff

 

[Julie.]

Very common for installations with wiring or switch drops with no CPC. Insulates the plate screws from the back box.

Yeah, that installation looks very odd, 1mm2 was not normal, nor would be the use of twin & earth on lighting circuits in that period, also running the neutral to the switch was old-style - a sign of the use of singles just prior to using twin cable.

I would guess that it was an old electrician following the 'old' ways with the new cable of the time!

Btw, since there is a new consumer unit (mcbs or cartridge fuses) then you can up the cable rating by 133% as it's no longer course protection, but I guess you would know that

 

[Pete999]

The conductors should probably be coated with an anti-oxidant paste designed for aluminium electrical terminations. I use Ideal Noalox, though it's usually on much larger conductors.

View attachment 54887rank Mate it's not help

The conductors should probably be coated with an anti-oxidant paste designed for aluminium electrical terminations. I use Ideal Noalox, though it's usually on much larger conductors.

View attachment 54888

Are you measuring the csa….or the diameter?

Diameter I should imagine, with a micrometre as the test instrument

 

[freddo]

Your links don't work!

 

[davesparks]

Yeah, that installation looks very odd, 1mm2 was not normal, nor would be the use of twin & earth on lighting circuits in that period, also running the neutral to the switch was old-style - a sign of the use of singles just prior to using twin cable.

I would guess that it was an old electrician following the 'old' ways with the new cable of the time!

Btw, since there is a new consumer unit (mcbs or cartridge fuses) then you can up the cable rating by 133% as it's no longer course protection, but I guess you would know that

Twin cable was in use for many years before PVC insulated cable came along, TRS or lead twin was widely used much earlier than that aluminium cable was installed.

In all the old books I have read or collected on electrical installation I think I've only seen neutrals at the switches as a relatively recent method.
[automerge]1577481702[/automerge]

[automerge]1577482107[/automerge]

Are you measuring the csa….or the diameter?

Diameter I should imagine, with a micrometre as the test instrument

The diameter is being measured and the CSA calculated from that.

The diameter appears to be 0.210mm on the micrometer (if I'm reading it right) which gives a CSA of 1.03mmsq which is within a sensible tolerance of a 1.0mmsq nominal size.

 

[Julie.]

Double post for some reason - don't know how that happened
[automerge]1577491080[/automerge]

Twin cable was in use for many years before PVC insulated cable came along, TRS or lead twin was widely used much earlier than that aluminium cable was installed.

In all the old books I have read or collected on electrical installation I think I've only seen neutrals at the switches as a relatively recent method.
[automerge]1577481702[/automerge]

[automerge]1577482107[/automerge]



The diameter is being measured and the CSA calculated from that.

The diameter appears to be 0.210mm on the micrometer (if I'm reading it right) which gives a CSA of 1.03mmsq which is within a sensible tolerance of a 1.0mmsq nominal size.

Ok I will try again, I appeared to have double posted, then when I removed one, both disappeared???!!!

The diameter measured is 1.1mm which calculates to 0.95mm2 - so it is 1mm2

Twin and twin and earth only came into use in the late 40's, prior to this, it was single cables (double cotton /rubber covered), the practice was the live would run to the switch, then on to the next switch, with only the switch wire going to the ceiling rose, the neutral would just go rose to rose.

I think it was an old spark, continuing this idea, but with t&e ; and not someone trained with t&e, because almost straight away the three plate style of wiring came along with the new wire (rubber, and rubber/lead)

It was well before my time, but we ended up fixing/fault finding and replacing the old type wiring which was commonplace.

I think an old spark in the early 70's rewired the house following his old practices, but with the newest cable of that time - without following the associated wiring practice - He should of used twin - not twin and earth, with standard (of recent years) drop/3-plate method in 1.5mm2 (or 3/029 if really early 70's)

 

[Pretty Mouth]

Just to complicate matters, the hall and landing lights are 3 plate/ loop in at the rose (with obligatory borrowed neutral), and the bathroom is loop in at a junction box.

@Julie. , you also have a copy of the 14th ed of the regs, am.1976? I'd not heard of rewireables as being referred to as course protection before, until I saw it mentioned in the tables in it a couple of days ago. Also highly frustrating that table 24M is missing from it.

 

[Julie.]

Just to complicate matters, the hall and landing lights are 3 plate/ loop in at the rose (with obligatory borrowed neutral), and the bathroom is loop in at a junction box.

@Julie. , you also have a copy of the 14th ed of the regs, am.1976? I'd not heard of rewireables as being referred to as course protection before, until I saw it mentioned in the tables in it a couple of days ago. Also highly frustrating that table 24M is missing from it.

Yes, I have my old copy of the 14th, it is the one I started with!

Most of the guys at the time had older versions - although very few actually had the regs themselves, most had guide books - similar to the on-site guide, essentially a simpler version.

To be fair, things didn’t seam to change as much as today.

Came across loads of different ways of doing things, at college it was "this exact way or nothing" but every site we visited, you could see the evolution and mixture of practices over time

 

[Pete999]

Good idea, ferrules would give them a bit of back-bone wouldn't they.

Yes, I'm wondering if they ever complied. I have a copy of 14th ed amended to 1976, there's a table of copper clad aluminium down to 1.5mm² but for plain aluminium, minimum is 16mm². I suppose it could have been different in previous amendments though.

I seem to recall that there was a sort of rule of thumb in the days of Al cables that you went up a size from CU ie 2.5mm2 for lighting 4 mm2 for RFCs, certainly remember the complaining that went on w ith regards to using 2.5mm2 for lights, trouble connecting them, same with Sockets, all very well using Al but it took some time for manufacturers to realise that accessories needed to be redesigned to allow for the larger cables. Surface clipping was a pain as well.

 

[Lucien Nunes]

I agree with Davesparks that twin cable, and in some cases twin and earth, was in general use right from the beginning of sheathed wiring cables.

Obviously, with conduit installations, which were popular for domestic wiring in the 1930s, single-core VIR was universal and would loop in as appropriate, line at the switches and neutral at the lights. With sheathed cables, whether TRS or lead-sheathed, joint boxes were popular, with a twin run, twin switchdrop and twin to the rose, or alternatively a 3-plate rose.

With lead-sheathed, the earth goes everywhere a cable goes, provided the sheaths are interconnected at the fittings. This was done with (now confusingly named) bonding bars and bonding clamps, while junction boxes for lead-sheathed had these clamps built-in, designed for flat twin or flat triple cable. Even during the 1930s, some manufacturers were offering lead-sheathed cable with a tinned copper CPC, to reduce reliance on an effective connection to the lead (which was known to be troublesome) and to make it possible to terminate the CPC to a fitting e.g. a 3-pin socket without need for bonding clamps. With TRS, ordinary twin was the most common; when an earth was required it was sometimes run separately.

One odd single-core variation from the 20s / 30s where the line loop was always done at the switch, and the neutral not at all, was earthed-concentric wiring such as the Stannos system. This was true TNC with a combined neutral / earth throughout, via the metallic sheath. With earthed-concentric, the neutral goes everywhere a cable goes. At a switch you have three single-core cables - the feed, the loop and the switched feed to the light. At the light, just one single-core, with the neutral picked up from the sheath via the box or casing. Threaded sweating thimbles and threaded spouts were used to ensure a reliable CNE connection.

 

[davesparks]

The diameter measured is 1.1mm which calculates to 0.95mm2 - so it is 1mm2

Twin and twin and earth only came into use in the late 40's, prior to this, it was single cables (double cotton /rubber covered), the practice was the live would run to the switch, then on to the next switch, with only the switch wire going to the ceiling rose, the neutral would just go rose to rose.

Yes, I didnt think I'd drunk that much last night but apparently my ability to read a micrometer and do simple maths had vanished.

 

[ipf]

Copper plated aluminium and aluminium cables were all around the same, plated was ever so slightly lower, I think 1.5mm2 was 11A, and 2.5mm2 was 15A - clipped direct twin and T&E hence all the rings had to be in 4mm2 which was 19A

Only MK made fittings that were suitable - they clamped the cable and could take 2x4mm2, everything else was either too small or just used a screw - the aluminium flowed, oxidised and lost proper contact.

These were removed from the 14th in the 1976 ammendment, I think it was table 24M

But the smallest used was 1.5mm2 - I think the imperial 14th which was pre '70/72 may have had smaller cables!

As has been said, it needs a rewire, sounds like the cables have reached their end of life, once they start breaking, you will be forever repairing them!

You're not far off Julie. Table 24M was removed in the amendment. Below is table 26M.
Your current ratings are for 'bunched and enclosed' copperclad. As you can see (below), they're a bit higher for clipped direct.
Also, they're for 'coarse' excess current protection. Multiply by 1.33 for 'close'.

 

[Lucien Nunes]

Also, thanks for the offer of the redundant cable based on the planned rewire. I can see the time is fast approaching where we have to actually pick out some fittings and start building these demo boards. It's nice to show both cable samples sectioned and framed and also the same types of cable 'installed' in mock setups with some real working fittings (albeit at 12V) and some points opened up to reveal the wiring. Sounds fun to make but actually quite a bit of work, beginning with finding the right wallpaper for each era.

 

[Julie.]

You're not far off Julie. Table 24M was removed in the amendment. Below is table 26M.
Your current ratings are for 'bunched and enclosed' copperclad. As you can see (below), they're a bit higher for clipped direct.
Also, they're for 'coarse' excess current protection. Multiply by 1.33 for 'close'.

View attachment 54909

Yeah, I would have been thinking of column 11 - twin core in plaster etc. Rarely did we really use anything else in practice, as everything would eventually get to be run in plaster at some point.

Different now with wall voids and insulation.

Not that I ever installed aluminium, merely fixed/replaced/added to it.

And yes, that was for rewireable fuses (is there anything else???? )

Bit before my time, honest gov'

 

[Chris1234]

I’ve not read the whole thread, but from memory aluminium has around half the conductivity as copper, so best to check MCB ratings.
In a previous life I used to use titanium conductors (with Wago’s) with is around half the conductivity of aluminium!

 

[Pretty Mouth]

Also, thanks for the offer of the redundant cable based on the planned rewire. I can see the time is fast approaching where we have to actually pick out some fittings and start building these demo boards. It's nice to show both cable samples sectioned and framed and also the same types of cable 'installed' in mock setups with some real working fittings (albeit at 12V) and some points opened up to reveal the wiring. Sounds fun to make but actually quite a bit of work, beginning with finding the right wallpaper for each era.

I've got a few more bits n bobs that may be of interest, an old wooden fuse box and some old switches and sockets.You're welcome to them if they can be of use to you. I'll send details when I've got the cable ready to send you.

I sometimes come across old steel conduit that uses couplers to clamp together, and has what I believe to be rubber singles (red and blue IIRC?). Usually found in lofts, and would have been used for the lighting. Was actually still in use on a job I worked on recently, although in terrible condition at the accessories. Any idea what era this would have been installed? I'm guessing 1930s to 50s judging by the age of the properties I find it in.

 

[Marvo]

Thanks for the follow up info, it's a very interesting thread.

The photos you posted are textbook examples of how aluminium wire and terminations are problematic.

Wagos sound like a sensible solution, I'd clean the wire thoroughly first and maybe even some inhibitor paste depending on what the Wago maunfacturers instructions have to say.

 

[Lucien Nunes]

Yes it's interesting to hear of the specific problems that we attribute to aluminium, actually all happening in a real installation (although not so much fun for the owners of the property).

There's an interesting aspect to the use of the porcelain Scruit connectors. With copper cable, the conductors make electrical contact wherever they touch, while the porcelain body of the Scruit only serves to apply a clamping force (by the tapered screw action) and insulate the joint. When correctly fitted (many weren't) and not overcrowded, Scruits were actually quite reliable and effective on tinned copper. But with dry Al cable, contact is only made where the oxide is scraped off or bitten through, e.g. by the tip of a screw or the serrations of a cage clamp. The wire-to-wire contact in a Scruit doesn't get its oxide scraped off, so is only likely to be punctured in tiny spots as the copper cold-flows at the high spots.

Wire nuts in the USA have a steel spiral spring inside that forms the screw thread, which bites better into the conductors and holds them more securely than plastic or porcelain. In the case of Al cable, it's a mixed blessing. It will definitely scrape off the oxide and make a connection to itself, but if the oxide stops the conductors contacting each another directly, this diverts the entire current through the spring of the wirenut. Being steel and of typically higher resistance, it can work like a tiny heating element even while the circuit appears to work normally as there is no loose connection as such. There have have been documented examples of wirenuts overheating dangerously through this effect, whereas with an old British porcelain Scruit, the circuit would probably just go dead.