Help! Strange power supply circuit behavior.

[varialbender]

This makes absolutely no sense to me. Feels like bad voodoo, but maybe there's an explanation.
I've got a power supply circuit breadboarded, from 16VAC 1.1A wall wart to +/-15V out.
Right now, I'm trying to get things figured out without regulators, so it's more like +/-25V.
One wall wart wire goes to ground, the other goes to two diodes, each with a capacitor going to ground, and their outputs each have a 10k load to ground for testing.
I have it all mounted on a breadboard, and the positive side is on the top half of the board, and negative on bottom half. Each side gets a rail for either positive or negative voltage, and a ground line they share that I connected. The wire from my power supply goes to either the top or bottom ground line.
Here's the strange part:
When I connect the wire to the top half, the bottom half doesn't work, and vice versa. I read 0 volts from whichever side isn't seeing ground, even though I've got a wire connecting both ground lines. It seems the only side that works is whichever physically sees the wire from the wart first, but that doesn't make any sense to me.
I'm sure I've got everything wired properly. I've gone over it many many times. Each works alone, and I'm sure of the orientation of the caps and diodes, etc...
Anyone have any idea what could be going on?

Thanks a lot

[R.G.]

The interconnections under the surface of breadboards are tricky.

I'd bet that you are shorting one output while connecting the other.

Those lines of connections along the breadboard top and bottom are all connected together. Use your ohmmeter and some cut off resistor leads to ohm out what's connected and what's not. I bet you find at least one surprise.

[varialbender]

Yeah, I was going to try that, but wanted to see if anyone thought it might be something else, just in case I could avoide testing many many holes.

I still don't see how, if I have one line, it would matter where on that line I touch the ground, because there's no change along that line. The only thing I could think of is the resistance from wires, but that would be tiny and not affect anything. Unless there's some sort of bad connection making a huge resistance...?

Anyway, I'll try it and let you know what happens.

It's just so discouraging that such a simple circuit that I understand and know is correct won't work.

[R.G.]

Plugboards are great - until a quirk of the plugboard intrudes on the circuit.

If you wired your circuit up, it would eliminate the questions. It doesn't have to be on a circuit board. Hanging garden style with leads soldered together will work to prove the circuit out, especially something this simple.