My wife & I are involved in a major kitchen remodel in a home we just purchased. In relocating some outlets it became obvious the wire was heavy — like 10 & 12 ga and very stiff and oh no IT’S ALUMINUM!!!! How this got by me I’ll never know – and when my friends find out I’ll never live it down. The wire is copper clad aluminum. How much trouble am I in ???? When adding new outlets & fixtures can I use copper ?
The house is 28 yrs old, 4 story, on Puget Sound.
Looking forward to your input — John B
Replies
The biggest trouble with Alu wiring is that it will expand and contract more under heat from load. At connections, like recepticles, that can cause the screws to work loose from the movement. Once they work loose, the contact is weak so more resistence is encountered which produces more heat. Sometimes fire is the result.
There are special connections your electrician can substitute to prevent this buit the best insurance is to replace the wiring.
I have one word for you. Rewire.
Why: Aluminum versus Copper.
1) Copper conducts heat and electricity better than aluminum. An aluminum conductor must be larger in cross section to remain as cool as a copper conductor under the same load.
2) Metals exposes to air, except gold and platinum, oxidize. It happens in milliseconds. When cleaning aluminum for electrical work it is standard practice to wire brush any gross oxides, apply an anti-oxidant grease and wire brush through the grease. The grease prevents oxygen from coming into contact with the aluminum.
3) Why are oxides bad? Oxides don't conduct electricity as well as the bare metal. Copper oxides are less conductive than bare copper. Aluminum oxide is so much less conductive than bare aluminum that it is classed as an insulator.
4) Why is less conductivity bad? Lack of conductivity is termed resistance. When a current encounters resistance it is converted to heat. The greater the resistance the greater the heat.
5) Why is heat bad? Heat generation in a hair drier or space heater is fine. Excessive heat generation in wiring can melt insulation, wastes energy, can damage connections and causes materials to expand. Wasted energy costs you money. Damaged connections and melted insulation causes electrical failures and fires.
What about that expansion thing? With a few exceptions, rubber being one, materials expand as they get hot. For every degree of temperature rise aluminum expands more than twice as much as copper. Electrical connections use pressure to maintain contact. Loose contacts mean more resistance.
Heat = expansion = more pressure = better connections. Right? Right to a point except for plasticity. Given enough heat and pressure materials tend to flow away from the pressure. Aluminum, given the same amount of pressure and temperature, flows more than copper. The aluminum expands twice as much so the pressure will be greater than in a copper connection. So an aluminum connection as it heats up gets tighter than a similar one made of copper. It gets slightly better. The problem is what happens when the toaster clicks off and the connection begins to cool.
The aluminum has moved away from the pressure of the connection so as it cools small gaps open up. These gaps allow oxygen to reach bare aluminum and form oxides. The next time the toaster is run the oxides will impede the current flow and cause the connection to heat up more than it did last time. More heat causes more expansion, pressure and flow. Over time this cycle repeats and the connection gets worse until it fails. Copper connections are much more resistant to this cycle.
The use of aluminum wiring in branch circuits has been outlawed since the 70s. While aluminum can be a cost effective material for feeders if all connections are carefully made with the correct lugs aluminum should never be used in sizes smaller than #8. If cost, or weight was no object all of the electricians I know, even a few that I wouldn't trust to but batteries in a flashlight, agree that they would use copper for all wiring.
Rewire as much and as soon as possible. This stuff can only go down hill. If you can't afford to do it all at once do the higher use and load circuits first. Kitchen and bathroom circuits are usually a good start.
Getting to know your local fire department by having them chop you out of a burning building is not a good idea. If you want to meet firemen go to their annual barbecue.
Also check your local code. A major kitchen remodel may require rewiring when aluminum wiring exists....that's not a mistake, it's rustic
Not an electrician, and would certainly defer to one's advice. The wiring books I've seen always say that aluminum wiring should be left alone, and if there are any problems, either completely replace and/or contact a qualified electrician to deal with it. BUT, if I recall correctly from those wiring books that do mention it, copper clad aluminum is not as bad as straight aluminum, and you can connect/splice it as you would copper, maybe provided the devices are rated "CO/AL". It's still not as ideal as plain copper, but not a complete stop point. Is this true?
Again, as I type this, I'm not certain - I've never dealt with aluminum or copper-clad aluminum, nor do I really want that experience. This a vague rememberence from textbook reading, please DO NOT follow this as advice without further research from a proper authority. I'm just asking the question as part of my own selfish quest for knowledge.If everything seems to be going well, you've obviously overlooked something.
Aluminum wiring hit it's peak in about '70 to '73, I think.
Some banks and insurance companies require change from ALu when they know about it.
I hadn't thought about the whole remodel indicator. I believe that the electrical code requires that anytime you open up to 15%, you are required to re-wire all up to newest code. Not sure on interpretation of what the 15% is based on. Anyone?Excellence is its own reward!
Does "copper clad" change the picture. How about AWG pigtails ?
John
Oops - I meant AMP pigtails.
Sorry - John
We recently sold a house in New York, upstate, built in '67 that was wired in aluminum. We looked into the concerns with Alu several times over the years and again when this house sold.
There are special pigtails made for the purpose of separating the aluminum wire (which is perfectly fine and does not need to be replaced, as long as it was installed to code originally and 'gaged up' one size (12ga for 15A and 10ga for 20A)), to separate that wire from the devices (outlets, switches) where the thermal/corrosion issues cause the concern. Since there will be no joints anywhere in your wiring except in the device boxes you can have every one pigtailed with no need to re-wire the house.
The pigtails can only be installed by a liscenced electrician and are designed to deal with making a good, oxygen free connection between the alu wire and the cu pigtail. This is a good time to replace all the outlets and switches just because they are all pretty much worn out after 20-30 years of use anyway. Of course, they must add this pigtail device to all connections including ceiling light fixtures and so on.
There are lots of things in life that you can worry about, and there are those out there who can make money off of your worry. The challenge of modern living is to identify and select what really should matter to you and what is just effective advertising (this applies to radon gas, asbestous fibers, aluminum wiring, lead paint, mold, formaldehyde off-gassing from building products, etc, etc, etc).
You can choose to rewire your whole house, but really do not have to, because these pigtails exist.
What ever you do, DO NOT extend aluminium wires with copper. REPLACE REPLACE REPLACE.
This is what has happened in my home. The previous owners did a remodel of the kitchen and moved the stove over by 10'. They added to the alum. wire with copper. on inspection at purchase both myself and the building inspector missed it. Copper at the stove plug, alum. in the electrical panel.
On truly seeing the potential desaster after purchase we decided to go with a gas stove. Removed the stove wire from the electrical panel to replace 200 alum. with 110 copper. DW then tells me the dishwasher no longer works??????
It turns out that the local kitchen remodelers did their own electric job. Not only did they splice the alum, with copper, they took one side of the 220 circuit to power the dishwasher. To top it off they had twisted the ends together and taped. No box, No dissimilar metal grease. NO WIRE NUTS.
Ripped it out faster than you can say "Sonuvabi***"
An honest sparky wont touch this remodel unless it included a rewire. Youll have to include GCFI outlets and whatever code requires. Not so much to make money but thats pretty much a code no no and he would be liable.
Sidenote: Wasnt Alum used during ww2 and a bit after since the copper was going to the war effort? Or was alum wiring later on?
Aluminum was in very high demand in WWII for airplanes -- it was not going into wiring. During WWII some aluminum smelters even used silver buss bars made from silver from the US Mint. Copper was also in very short supply -- but still used for wiring.
Aluminum wire does have its place -- high voltage power transmission and in feeder lines -- but not in most house wiring applications. The aluminum industry tried using alumnium wiring in its own plants in the 1960's -- big mistake, we replaced it all with copper. I spent 30 plus years in that game.
Dave Smucker
In the early 70's Hydro Quebec and the leading aluminium companies convinced the builders in this area that it will be cheeper to use alum. wiring. As Hydro was selling the electricity to make the aluminium and the aluminium plants were making money selling the wire. No body really thought that the wire was going to be so bad. It is now illegal to install aluminium wire in new construction here and unless you go to the very large guage wire you can't even purchase it.
I now have GFI circuit breakers for the kitchen and all the high load circuits in the house have been replaced. Only the hard to reach low load circuits remain. Waiting for the next time the rooms need painted so I can rip into the walls and patch before painting.
Remember that a one sign of intelligence is the ability to hold your head when all those around you are losing theirs. Exactly what is the problem with aluminum wire? Why is it that you still find it in the high amp circuits? If it spontaneously combusted due to high resistance heat why would it still be OK to use it in the highest amp applications?
I owned a house with aluminum wire before the jackals set their sights on the product. Whole development had the stuff, they are all still standing and have not burned to the ground from the infamous high resistance connections. They do require some additional maintenance, when you hear any sort of buzzing when you turn on a light switch or plug in your toaster,plan to get in there and re-tightening the screws. No big deal, you have to do a lot more to maintain other systems in your house. Remember, not all the people on this site are nail bangers, some are electricians with payments to make on "The Lake House"
I'm sure that I just upset a lot of folks, but not all of us are making six figures and can aford to rewire our house and must make due with what we have.
Do you know why aluminum is used for pots, or to clad the bottom of high end stainless steel pots? Because it is the next best heat conductor when compared to copper.
If your wires are in conduit, you can use the existing aluminum to pull in new copper. If it's romex, you'll need an electrician who has the tools and training to pigtail it all.
-- J.S.
Some items not mentioned.
12-2 w/g cost $20/250 ft in '70 (1970$$), Cu clad Al was $12; I made $4/hr then. Cost for Cu now is only about $5/250 ft in '1970 dollars!
Al wire problems were 90% oxidation, not expansion.
Al expands 23E-6/C, Cu 17E-6/C, brass 19E-6/C. Hence, in a screw terminal, copper gets LOOSER at temp and Al gets tighter, but not enough to stretch the brass screws.
Soft Al has 9 ksi comp strength vs. 8 ksi for soft Cu, hence not much difference there.
I used Cu clad Al in my own previous house in '68-'70, Cu clad Al is a big difference from straight AL.
All comments on need for nickel loaded grease are 100% correct, especially on stranded wires - the nickel breaks thru the oxide layer on the Al, the grease prevents further oxidation. Any size Al without grease or chemical conversion coating (e.g alodine) will oxidize and elec joints fail. Biggest problem with Al was that one could not "just hook it up" like the forgiving Cu wire. A small arc on a loose Cu to Cu joint can weld itself together (have seen this often), where as a loose Al to brass or Cu joint will never weld, instead the increased heat accelerates the oxidation and results in a runaway (eg.fire) condition.
Problem with even Cu clad was that people still insisted on using the cheapo "push the wire into the slot back receptacles"! This cut the Cu cladding and allowed oxidation of the Al.
If you have Cu clad Al in your house and ALL are under screw terminals (no push ins) you are OK. Replalce any push ins.
Is there a safe way to splice copper to copper clad aluminum?
As Norm said above:
"The pigtails can only be installed by a liscenced electrician and are designed to deal with making a good, oxygen free connection between the alu wire and the cu pigtail.
The KEY words are "good, oxygen free connection", which usually means a tight crimp with correct sealant.
Don't necessarily agree with the "only be installed by a liscenced (sp)electrician" part, but you need to know what and what is not a "good" connection and have good crimper - (read good as not just pliers or $2 crimper/strippers, but something with compound advantage or even hydraulic). Used to crimp Al to Cu in missile silos, took about 10 long strokes on a big 2 ft hydraulic pump to make a "good" cirmp with the 10 lb crimp head.
I always thought the expansion problem with Al wiring envolved plastic deformation:
As the temp goes up, the wire wants to expand more than the screw (just as you pointed out). Since two sides of the wire are constrained by the screw and recepticle, while the other two sides are unconstrained, the aluminum wire tends to flatten out under the stress. Since (if?) that deformation is plastic rather than elastic, the wire will hold that shape when it cools. This results in a slight gap between the wire and recepticle contacts. That allows more oxidation. That increases the temperature under load. That produces greater thermal stress and consequent plastic deformation. And on, and on, etc.
It seems that Cu-clad Al would suffer a similar fate, albeit with the oxidation issues only arising if/when the aluminum is exposed due to the cyclic thermal stress.
You're a sharp guy. I'm sure you'll point out to me why this isn't the case. That's fine with me, since it means I've learned something.
I did run deformaion/elasticity numbers literally decades ago (when I used Cu clad Al in my own house) but don't have those electronically and don't have the time right now to rerun them. Probably the best simile to realization that plastic deformation is not the issue is a cold midwest or canadian garage in winter, where the opposite of what you describe could happen - copper "leaves a gap" when it warms back up, etc. Another practical reason is that there is more stretching deformation in the brass or steel screw (steel screws complicate the issue even more) since the cross section area of a wire wrapped under a screw is many times the area of the screw cross section, plus bending of the contact plate - mone factor in Cu/Al fixtures vs. Cu only, etc. Your reasoning however is right on when you consider one of the reasons a "good" crimp takes high power tools is that full plastic deformation is required to stay away from point contacts that more easily oxidize. Also, a common grounding/return practice in aircraft is cad plated copper to Al sheet with steel screws, high reliability, especially if Al is alodined.
I personally have a 35 year old sample of a sealed (polysulfide) aluminum alodine connector to alodine 1/4" plate attched with NAS screws (cad plated steel) that has decreased in resistance every time tested at 100 A (60 microohms start, about 56 microohms now).
I've been proven wrong thousands of times before (literally), so someone more knowledgable may correct my long held misunderstandings - AJ must have bailed at the prospero change and never came back?
PS: An urban myth related to "plastic flow" is the supposed flow of glass over centuries such as "found" in old cathedral windows - no such thing, 50% are thicker at the bottom, 50% narrower!. Millenia old glass from Egyptian tombs shows no such flow. Granted, this is totally different than deformation flow of metals.
Since this is getting long, might as well add another caveat to Cu clad concerns - rework, such as replacing an outlet with multiple tightenings and loosenings of a scres can cut the the Cu clad, can lead to breaking the Cu cladding and then the oxidation problem can occur.
All of the above taken together means that for Al or even Cu clad Al to be foisted on the general public can invite trouble and failures.
Edited 8/14/2002 11:38:35 PM ET by JUNKHOUND
Edited 8/14/2002 11:44:35 PM ET by JUNKHOUND
With all due respect, I'm not sure I buy most of your argument.
Certainly, under the proper conditions, copper will gap due to the difference in CTE. However, oxidated copper does not pose the same increase in resistance as does aluminum oxide. That increase in resistance is what provides increasing stress levels and (possibly) increased deformation of the wire.
Additionally, Cu gapping will occur at elevated temperature (ie under load), while the aluminum would gap will gap at reduced temperature (and, if my previous argument holds any water, under a no-load condition.) I suspect that most terminals spend far less time under load than they do cold. If that fragile sequence of assumptions is correct, then the Cu would have much less time to oxidize. I feel I'm on thin ice with this one, but find it just interesting enough not to delete it.
On more solid ground, your argument concerning cross-section areas, as they pertain to the stretching deformation, seems to ignore the fact that aluminum yields at a far lower load than does brass.
All of this is academic speculation, of course. I don't know what specific alloys are used, the temperature that develops at the joint, or even the reference temperature at which the recepticles are installed. Without these, I can't even WAG at the stress levels at the contact area, let alone justify a conclusion that the wire yields.
If you're telling me you've run the numbers and found that the aluminum was not stressed beyond yield, then I'm inclined to believe you.
Interesting assertion concerning the flow of glass. I'd always been taught that glass was a super-cooled liquid which flowed (nearly infinitesimally) over time. I never thought to challenge that.
Maybe those cathedral windows that are thinner at the bottom started off even more so. It's difficult to prove or disprove movement when you don't know what the initial state was. Just a desperate attempt to preserve the honor of my high school physics teacher.
Copper cladding only marginally improves the reliability of copper clad over aluminum. The NEC, does not differentiate between the two and bans the use of both in sizes smaller than #8 and for branch circuit wiring. This speaks volumes. The advantage of the copper cladding is so small that even though aluminum wiring is still good for feeders and extensively used by utilities, to my knowledge, no company still produces copper clad aluminum and it is not generally available.
Idea and Buchanon both produced supposed solutions to the aluminum wire problem. Cu/Al rated wire nuts and a few crimps were approved.I don't know if they are still available. Most of these were marginal solutions at best and were very dependant on the situation, having enough wire and room available in the box to allow precision application of the tool or nut, and both skill and care of the electrician. In the field many of these solutions have not proved reliable even when the tools, training, care and time were used.
Given that a smallish house has something like 40 receptacles and 20 switches and that each of these will need at least 3 of these pigtails or crimps you rapidly start to involve a major fraction of the time and money it would take, depending largely on accessibility and the skill of the crew, to rewire the place. Especially if a few of these patches are not effective.
Without doing any research on the subject, something I ought to look into, I have been told that aluminum wiring was used when the copper supply was disrupted. Something about a cold war proxy fight in Africa. Congo? Copper prices went through the roof with the copper in a penny, remember real copper pennies, being worth something like 2 cents.
Don't freak out about the wiring. It hasn't burst into flames to date so odds are that it won't anytime soon. On the other hand this was considered such a threat that a few fire departments were, after a few people died and a few fireman were injured trying to perform a rescues, buying mobile homes with aluminum wiring and dismantling them to get them off the market. It was deemed cost effective considering the cost of an emergency call or loss of life. When it became common knowledge that trailers with aluminum wiring were a death trap people, even to this day ask about it, stopped buying them and the industry stopped making them.
Rewire. The only difference between the patch and a proper rewire job is the snaking of the wires. A crawl space and attic will make this easier as will routing cables behind chair rails, baseboards, crown molding and molding around doors. If plaster work is being done for other reasons the additional price for snaking the wires can be further diminished. You can do it in stages.
Better to do it completely and soundly than have to do it again. Ask around about contractors that do lots of rewire jobs and old work. They will be the most likely to have people experienced in the techniques needed. If it was my house it is what I would do.
John, more to read
http://www.rshengineering.com/whatisaluminumwiring.htm
http://www.homestead.com/siliconecity/fatal_litigation_1_2.html
http://www.coyle-inspect.com/Aluminum%20Wire.htm
Nice links. I agree with most of what they say. The exception being in reference to copper clad aluminum. They claim that clad aluminum conductors are not a problem. I do not agree. I would like to see the evidence used to come to this conclusion and compare it to the evidence used by the NEC board when aluminum and clad aluminum wiring was banned for branch circuits.
The first and third links are both siting the same source material. The second is an interesting read but is an opinion pushing a anti-litigation line not a presentation of factual material relating to wiring. Aluminum wiring is central to the story only in the precedent set for class action cases and not the material itself. I have no wish to debate tort reform, the real point of the second link, at this time.
I have personally witnessed the failure of connections made with this material. These failures are less common than all aluminum connections where a great number of the devices, receptacles and switches, when all are removed during a rewire, have evidence of major corrosion and heating but much more common than an all copper system.
The type of problems encountered were also different from the obvious and wide spread failures seen with unclad aluminum. While generally not as flagrant a failure, corrosion and heat damage were concentrated, but not limited to, where the copper had been damaged during termination and it was much more localized than a straight Al connection, it was, common to all of the cases I am citing. The sort of surface damage that I think triggered these failures is, IMHO, common to any type of termination and would not cause a problem with copper wires.
The physical manipulation of the wires during stripping, bending and termination under the screw causes small scratches, nicks and abrasions in the soft copper no matter the care of the electrician. With a copper conductor these are not a problem but with copper clad aluminum the aluminum core is easily exposed and once laid bare it shares all the weaknesses with its unclad cousin.
Pigtails and other mythology. The emphasis of the proposed patch methods on precision in technique performed by factory trained personnel gives a big clue as to the expected robustness of the solution. This is a sign that it is more a theoretical solution best carried out in a controlled environment. Success in the field where temperature, lighting, cleanliness, time pressures and worker dedication is more variable gives results that are much less assured and reliable.
IMHO, in most cases, for about the same money you could rewire with copper and have a robustly engineered and time honored solution that is much less dependant on kid glove handling and micro-management.