aluminum wiring..
As far as aluminum wire- it has been used in houses since WWII, because they needed all the copper they could get, so they used aluminum. They didn't have too many problems until people started overloading the circuits and heating the wires, which became brittle and caused the insulation to bake. They went in and added to the circuits and replaced receptacles, which broke the insulation, caused arcing and shorts, fires and all kinds of other problems. The recommendation I heard was that if Al wire is used, it needs to be one size larger than copper to avoid problems. The lines to our houses have been aluminum for a long time, too. I really don't like it but it's lighter than copper and the electric company thinks it works well, so it would be hard to get them to replace it with copper.
Actually, the case of aluminum wiring is a tad different. I actually do not know what is out there on the web, so I'll explain from memory.
1. There is a difference in the thermal expansion coefficient (TCE)between copper and aluminum. Aluminum has a TCE of 25.5 ppm/degree C, and copper has a TCE of 16 ppm per degree C. ALL alloys of copper are within 1ppm of pure copper. Stainless also runs 16ppm, so all mounting hardware and screws match the expansion of copper. For aluminum wiring, when the wiring warms up as a result of temperature increase (note A), it will want to expand faster than the screw which is compressing it. Because aluminum wire (copper as well) is not hardened, but is soft to allow it to be bent without cracking, the aluminum will be compressed much more, and will cold flow (note B). Through repeated cycles of this, the connection will loosen up. Subsequent heating at the connection will result from the loosening, and this will speed up the heating/compression/flow/heating cycle. Eventually, the heat will be sufficient for a fire.
2. There is a significant difference in the conductivity of the oxides of aluminum and copper. Aluminum oxide does not conduct electricity, but copper oxide does. As a joint ages, aluminum wire will oxidize, and this will begin to compromise the joint integrity. As that starts, the contacts will begin to heat, starting the death spiral noted in 1.
3. The National Electric Code specifies exactly what each guage ampacity is for both aluminum and copper. Aluminum does indeed require a larger cross section for the same current. edit: tables 310.16 through 310.19 of the NEC specify the ampacity of copper, aluminum, and copper clad aluminum as well as correction factors for ambient temperature. 310.15B(2)(a) derates for the number of current carrying conductors. These tables cover wire rated for 60, 75, 90, 150, 200, and 250 degrees C.
4. There are a few ways to mitigate the problems associated with aluminum wiring, things like pigtails, paste, and end use hardware specifically designed and marked as suitable for copper/aluminum, or just aluminum. ALWAYS follow these recommendations, they are in the code for VERY good reasons. Remember, the National Electric Code is maintained by the NFPA, the National Fire Protection Association, so all of the code is based in historical evidence of fires and fire related problems.
Note A: ALL wiring is rated for current capacity (ampacity) based on the guage of the conductor, the conductivity of the conductor, the heat environment of the conductor,
and the temperature rating of the insulation. To assume that any wire remains cool in operation is folly.
Note B: Solder also cold flows. This is creep of the metal as a result of stress, either compressive, tensile, or shear. Never rely on the solder for the physical strength of a joint. And never pre-tin stranded wire which will be placed in compression by a screw or other mechanism. I've heard of some instances where stranded copper wire was used in wall for AC, and then tinning the end so that the wire can be captured by the screw head more easily. This method WILL result in long term problems, fire being one of them. I personally have a piece of equipment which cost 1.5 e6 dollars which the tech had pre-tinned all the signal wires which go into screw blocks. Every year, we have a tech tighten up every single screw on every block so as to avoid intermittent connections. This is very important when motion control hardware is involved..
My apologies for the long and off topic rant, but this subject matter is very important.
Cheers, John
Followup edit note:
The wiki entry for aluminum wiring is consistent with my blather in all but one case. They claim that studies have shown that it is not the aluminum oxide growth that continues to compromise the connection, but rather, the formation of aluminum-steel intermetallics. I cannot confirm nor deny the veracity of that statement, so I will retain both possibilities as plausible until I can find any peer reviewed research papers supporting the intermetallic claim..
