The picture to the right shows my setup. The power supply is a standard Elenco Precision, and I'm using the variable voltage input. I don't use much voltage at all -- just a few volts gives me enough current in the wire to see a deflection.
The pictures below show the experiment itself. On the right you see the wire carrying no current... it's aligned (approximately) with the earth's magnetic field. The white end of the compass needle points north. In the left-hand picture, the wire is carrying a current from right to left. By the second right hand rule, this current produces a magnetic field beneath the wire pointing west. (Toward the bottom of the picture. Yeah, sorry, I should've aligned the compass markings with the earth's cardinal directions, but I didn't think of that.
So, why doesn't the compass needle point straight to the bottom of the picture, in the direction of the wire's magnetic field? Because the earth's field is still stronger. In fact, this setup could be used to calculate the earth's magnetic field. Use an ammeter to measure the current in the wire. Use the equation
to calculate the magnetic field due to the wire at the position of the compass needle -- that means the variable r will be equal to only a few millimeters, the distance from the wire nearly to the table. Then, use vector analysis -- the wire's field is perpendicular to the earth's field, and we know that the vector sum of the two magnetic fields points about 30 degrees down from the earth's field. That should be sufficient to get close to the 10-5 T that is the typical magnitude of the earth's magnetic field.