Electrical Contacts

The fluorescent light in the closet failed last week. I replaced the bulbs with some spares I had, but it didn't help. Ballast is probably shot, so I go to the store and pick up a new one. The ballast is like $23, while a new fixture is only $30. Hardly seems worth it to futz with the ballast, just buy a new fixture and save myself a lot of grief. However. The new fixture requires the new, smaller T-8 bulbs, which will probably add $10 to this project. But if I do that, how am I ever going to use up these extra T-12 bulbs? Then there is the problem of getting rid of the old fixture. Space in the trash can is at a premium. Sure, the sheet steel frame can be recycled, but that big plastic lens will have to go in the trash. So there are certain advantages to fixing the old one.

Back home I open up the fixture and see the wires. I take a look at the new fixture and it has all the wires in the world. I was thinking I was going to have to cut, strip and splice the old wires to the new wires, but now I realize that if I can just unclip the old wires from the sockets, I will be able to just plug the new wires in and save myself a lot of grief. Except. The sockets are designed for one-time use only. There is no slot to release the wires. Bah, humbug. Go to bed.

New day, sunny day, I'll try it again. I take the fixture down and pop the sockets loose from the frame. I find that I can pull out the staple in the back that is holding the insulating panel on. When I do that, the whole thing falls apart and wires fall out. Put the clips back in, put the cover back on and secure it with the staple, which actually holds. That surprised me. Put the whole thing back together along with the new ballast, and presto! We have light!

The thing that got my attention was the contacts in the sockets. They are just little strips of brass, maybe a quarter of an inch wide and about as thick as a sheet of paper. When a wire is inserted in the socket, the end of the strip scrapes along the surface of the wire, and since it is at an angle, it locks it in place. The contact area between the strip and the wire is tiny, certainly no more than a few thousandths of an inch in either direction. Meanwhile the wires are like 60 thousandths of an inch. The wire has to be that thick in order to carry the load without overheating. But here we have a choke point that should provide considerable resistance to the current. It should overheat, melt the plastic, short out, start a fire and burn the house down. But it doesn't. This is standard wiring practice all over the house, all over all houses. Something funny going on here.

The only thing I can think of is that since the resistance of a wire is dependent on the material (and its' inherent resistivity), its' length and width, then these connections don't overheat and fail because they are very short, on the order of infinitesimal. Still, there is a limit to how much current you can push through that tiny little contact patch before it does overheat. It would be interesting to experiment with conductor size and contact point shape to see how much current they could carry. For instance, would two wires with carefully squared off ends and one tiny contact point between them be able to carry more current than a wire whose ends were tapered like a sharpened pencil?