# Torque on a moving dipole

1. Feb 8, 2010

### tavsaito

I am at a loss for how to start this problem.
t = r x F and F = q(E + v x B) i know but there is no external E or B field its just a dipole that moves with a speed v in the x direction the dipole is oriented with +q at (d,d,0) and -q @ (-d,-d,0)

how do i derive a torque?

2. Feb 8, 2010

### Staff: Mentor

If there is no external E or B field, there should not be any torque. Are you sure you've stated the problem correctly?

3. Feb 8, 2010

### tavsaito

yes there are no external E or B fields i think it has to do with relativity. the one particle creates a E feild in its own frame which becomes distorted when it moves allowing there to be a non parallel component.

4. Feb 8, 2010

### tavsaito

the problem is.

Consider two point charges +/- q embedded on a square dielectric in the x-y plane the dielectric is a perfect insulator and the charges cannot move or be neutralized. Assume that i) the origin of the coordinates is in the center of the square and the positive and negative charges are respectively at (d,d,0) and (-d,-d,0) and ii) the square dielectric is moving at velocity (V,0,0).
a) derive an expression for the torque on the system. Calculate E and B from maxwell's equations but without relativistic corrections
b) It was proposed to use a similar set up to measure the absolute velocity of a body in space. For example this could be used to detect the velocity of earth in its rotation around the sun. Do you think such an experiment would work? If necessary in your explanations, use the relativistically correct expressions for the E and B fields cause by each charge at the others position.

5. Feb 9, 2010

### tavsaito

Solution: Basically were asking to prove that it wasnt possible unless taken in a relativistic regime. See the Right-Angle (Lewis-Tolman) Paradox where you can observe a torque in the moving frame but it does not result in any actual rotation