What is the key technology for connecting aluminum?

A bit unrealistic.

Most people who hire an electrician don't know enough to determine that he/she is incompetant.

A normal pre-sale inspection will not include disassembly of the electrical system to determine wire size.

Ed

When I first posted this question. I was really trying to find out what happened in Al wiring that caused.

I know that pure copper (Cu) will cold weld when it flows with sufficient pressure applied. Is there welding taking place with Cu under binding posts? Certainly Cu oxidizes. Are Cu oxides conductive?

Does Al cold weld? How important is that in making connections? How do the anti-oxidizing pastes work?

Bill

On Mon, 14 Apr 2008 15:59:14 GMT ehsjr wrote: | snipped-for-privacy@ipal.net wrote: |> On Sun, 13 Apr 2008 14:54:40 -0400 Michael A. Terrell wrote: |> | |> | snipped-for-privacy@ipal.net wrote: |> |> |> |> On Sun, 13 Apr 2008 10:24:30 -0500 bud-- wrote: |> |> | |> |> | #14 Al wire has never been allowed for US branch circuits. |> |> |> |> But I was referring to an idiot's choice of wire. Likely he'll find that #14 |> |> Al is nearly impossible to find. And that would be sad because we'll lose the |> |> chance to remove that idiot gene from the pool. |> | |> | |> | Not if it's someone elses's home, or it lasts long enough to sell it. |> |> If you hire an idiot as an electrician, you get what you deserve. This is |> what inspectors are for (among other things). |> |> If you buy a home from an idiot (and don't inspect it to discover this error), |> you get what you deserve. |> | | A bit unrealistic. | | Most people who hire an electrician don't know enough to | determine that he/she is incompetant. | | A normal pre-sale inspection will not include disassembly | of the electrical system to determine wire size.

An electrician so dumb as to install wrong size aluminum wire is likely to be rather obvious in many aspects. Sure, some people are still too dumb to be able to recognize this. Back to my original statement.

The inspection of the work done by the electrician should determine wire size. If the work is of the type that requires new wire be installed, then it is likely to require inspection. Of course it depends on the local AHJ, and whether it is trying to avoid idiot electricians.

You can get good or bad pre-sales inspection services. Insurance requires a certain minimum. If you get what they require, and the house still burns down as a result, maybe you have a case against the insurance company. You might against the inspection service depending on the level of service contracted.

I think it was covered in the replies, but I'll do it here. Two factors cause the problem: 1) al and cu have different coefficients of expansion. 2) Aluminium oxide is quick forming and is an insulator. The combination is (slowly) deadly - with different expansion, when the splice heats due to current flow, some of the al that was in contact with the cu gets exposed to air and oxidizes where exposed. The joint cools when current stops and contracts, but with a bit less of the al in contact with the cu due to the oxide. The next application of current "sees" a higher resistance (because the contact area is smaller) so creates more heat. More heat means greater expansion, so more al gets exposed and thus more oxide is formed. Repeat the process enough times and you can get into real trouble.

The paste works to prevent oxidation. No oxidation = no problem. Of course, that assumes the joint was properly made in the first place. Paste won't "rescue" a poorly made splice.

As far as I know, cold welding does not occur in residential wiring with either material.

Ed

That is why CO/ALr devices us a screw that tracks aluminum's expansion rate.

The biggest change in the "new" wire is it is harder but the expansion rate is virtually the same

When aluminum heats, it expands faster than the screw it is constrained by. The aluminum may extrude and become smaller, which makes a looser connection, more heat, and progressive deterioration. I think that is a little different from what Ed said.

"New technology" wire and CO/ALR devices fix the expansion problem. But "old technology" wire (from before the change in UL standards) could still have problems.

I would add the obvious, that paste keeps oxygen away from the aluminum to prevent oxide. In general, for good connections you would like a 'gas tight' joint. Some pastes allegedly have metal particles to bite through the oxide.

For 15/20A circuits, the recommendation based on testing done for the CPSC is to apply antioxide paste, then abrade the wire, and use paste in the connection.

As an example of oxide problems, wires connected with a wirenut may have little metal-to-metal contact because of oxide. Large lugs deform the wire which breaks through any oxide (which is very thin). Wire nuts don't force the wires together hard enough. But the spring in the wire nut can cut through the oxide and make contact. The spring may be relatively high resistance and is not intended to be the current carrying element. If a couple turns make contact they can form a heater and can get red hot.

I don't think it occurs either.

For larger wire, Cu and Al, one splice method is to put the wires in the ends of a tube and compress the tube with high pressure. This forms a cold weld from wire to sleeve. I believe the Copalum splices, which gfretwell referred to, use high pressure compression and form a cold weld. But as he said they are quite expensive.

So one thought is why don't they make a device where the screw terminals are backed by some form of spring, like Belleville washers. A properly sized spring would just allow the Al to expand/contract but maintain a suitable compression for good contact.

Such 'lock washers' and 'precompression washers' have been around for years in mechanical fasteners, can't imagine why it wouldn't work with simple wire terminal screws.

Of course this doesn't address oxide or wire-nuts, but it seems like it would fix problems with screw-terminals.

I know the stab-holes on the back of many devices are *not* listed for use with Al, but it seems a properly designed, spring-loaded contact for Al wouldn't be impossible. The sliding action when inserting could scrape off the oxide and a paste would ensure further oxidation is prevented.

Again, I don't know of any such equipment on the market, but it seems perfectly possible to build such devices. Maybe when copper prices get too expensive, we'll all be relearning Al wiring again :-)

daestrom

Exposed Al oxidizes almost instantaneously. Does the paste remove the oxide thereby making a metal to metal contact possible? Why don't copper oxides (CuO and Cu2O) pose the same problem as Al2O3 does?

Bill

Cupric oxides have good conductivity. I understand there is one oxide that has poor conductivity, but it only forms in very high heat, like fire.

Pastes do not remove oxide. Pastes prevent further oxidation. Some pastes have?had metal particles that allegedly contact through the oxide. Some manufacturers, like Ilsco, want you to wire brush wire to remove oxides (large wire, not 15/20A stuff). As posted before, the recommended procedure for 15/20A wire, based on research, is to apply paste, abrade the wire, and use paste in the connection. See:

You make a good point. I wasn't addressing screw terminal connections, so it is different, but it needed to be said.

Ed