Built a Wheatstone bridge with two 10k resistors, one 5k rheostat, and a thermistor. Adjusted rheostat so that voltage across bridge reads < 0.013v, close enough.
Powered LM324N, pin 4 to 6.07v, pin 11 to 0v (just one 6v battery supply)
When I connect pin 3 of the LM324 (Vin1+) the voltage of that leg of the Wheatstone bridge changes to 5.4v. I assumed I had blown the opamp previously, and tried a fresh one. Same result. So I simplified the circuit. The only connection to the Wheatstone bridge is a single wire that goes to pin 3 of the op amp. Nothing else is attached, and yet connecting that wire changes the voltage between the two resistors to 5.4v. When I disconnect, they go back to 3.03v as they should be.
Here's a picture of the breadboard in case I am just out of my mind. I can't believe power is leaking out of the input of an LM324, I must be doing something boneheaded but can't see it. Picture attached. The perspective isn't 100% ideal, but the wire from the Wheatstone bridge is connected to pin 3, and pin 4 is Vcc.
This was my attempt to simplify the original problem. I built a differential amplifier to amplify the output of the Wheatstone bridge, and current was leaking out of both positive inputs back into the bridge. So I wanted to isolate the problem. I don't have a picture of the board implemented that way, but here's the schematic:
R1 was 10k, Rgain was 1k, R2 and R3 were all 1k