I know a couple guys on here don't understand how these things work so one of the ones I got was crap so I took it apart to show you guys how they worked. A positive and negative energize that big coil and it produces a magnet that pulls that contact in so it contacts the other terminal, then springs back when you turn off power to the coil.

Now to expand on ISX great video...

So now you look at the contacts and most are numbered...

85 & 86 are the eletromagnetic coil.

30 is the Common or the movable contact inside that ISX shows...

87a is normally OFF position where the eletromagnetic coil is NOT energized. So 87a and 30 are connected.

87 is normally ON position where the electromagnetic coil IS energized. 87 and 30 are connected.

More on relays..

http://en.wikipedia.org/wiki/Relay

Expanding on what Mike just said, here is a pic of the side of the relay. So now you can compare what he said to the picture and you can figure out how the schematic is layed out. You can see that they are showing it with the coil not energized. When it energizes, it switches to 87.For an example, to wire up my isolated grid heaters and be able to charge the battery when the grids are not on, I stuck a positive from the main batteries to 30 and the positive from the deep cycle battery that I use to power the grids is hooked to 87. I ran a negative to 85 and then I ran one of the grid heater wires back to 86, so when the grids turn on, the coil of the relay energizes and disconnects the deep cycle from the main batteries. I have nothing on 87a.

So what makes them go bad?

Well they really don't go bad very often but like that one was bad brand new because the internal part of the terminal was touching. The wire that comes off that coil is also hair thin so it can break easy. The terminals can get corroded inside (its not exactly water tight). So basically just being in the elements or having a rough (bumpy) life.

So what makes them go bad?

Being an electromechanical part there are many failure modes. Some stick open, some stick closed, sometimes the coil fails, sometimes the casing gets melted from an external source, over current, sometimes the terminals get corroded....you name it. However like ISX said they don't fail often. The ones that do are just cheaply manufactured. I.E. Mitsu fuel pump relays.

is there a way to test a relay with a digital multi-meter? when wiring aftermarket equipment, the fuse goes before the relay, correct? typically, will the relay say what it is rated for(amps)? if you were to rewind the coil, would copper clad steel work, or does it need to be 100% copper or silver?

You just see if there is continuity between the normally closed contacts when it isn't energized, and if there is continuity between the normally open contacts when it is energized. By continuity I mean less than 1 ohm. The last relay I got had the usual less than one ohm on the NC contact like it should but on the NO it had 30 ohms. So when I hooked a light up to the NO contact, the light was on but it was dim. The thing should either be less than 1 ohm or no ohms whatsoever.Yes, before the relay. They usually have an amp rating on them. No clue on the last bit.

is there a way to test a relay with a digital multi-meter? when wiring aftermarket equipment, the fuse goes before the relay, correct? typically, will the relay say what it is rated for(amps)? if you were to rewind the coil, would copper clad steel work, or does it need to be 100% copper or silver?

If you were to rewind it you would want to use magnet wire. You can find it at Radio Shack. Basically its copper wire (sometimes aluminum). It has a very thin layer of enamel insulation on it. Copper clad steel wire would have a significantly higher resistance than you want. Remember that electricity doesn't follow the path of least resistance per say rather it follows in proportion to the resistances it encounters.