1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Force between Current Loops

  1. Apr 10, 2013 #1
    I am working on a problem involving force between two loops of current. The problem is to prove the for any arbitrary loops of current, Newtons third law holds true.

    I understand the basics of the approach but I am having trouble seeing why a term goes to zero. The basic setup is to use the Biot-Savart law to predict the field of a segment of the current loop then to use the Lorentz force law to predict the force acting on a segment of the second loop.

    This has the form,
    $$d\mathbf{F}_{12}=\frac{\mu_0 I_1 I_2}{4\pi s^3}(d\mathbf{l_1}\times(d\mathbf{l_2}\times \mathbf{s}))$$

    Which when simplified by the triple-product and integrated gives the following form,
    $$ \mathbf{F}_{12}=\frac{\mu_0 I_1 I_2}{4\pi s^2}(\oint\oint d\mathbf{l_2}(d\mathbf{l_1}\cdot \hat{s}) -\oint\oint\hat{s}(d\mathbf{l_1}\cdot d\mathbf{l_2}))$$

    Its easy enough to see right from here that F12=-F21 just by the fact that you pick up a minus sign by switching the separation vector. So in that sense the problem is solved, at least from what I see right now.

    However I was looking in the Griffiths EM and apparently the first term on the left cancels out somehow and I can't figure why. (Problem 5.49) Any suggestions?
  2. jcsd
  3. Apr 10, 2013 #2
    a couple of things.

    1. s can't be pulled out of the integrals.

    2. look at this term:
    $$ \mathbf{F}_{12}=\frac{\mu_0 I_1 I_2}{4\pi}\oint\oint d\mathbf{l_2}\left(
    d\mathbf{l_1}\cdot \frac{\hat{s}}{s^2}\right) $$
    then realize that if we do the integral over loop 1 first, we have
    $$ \oint d\mathbf{l_1}\cdot \frac{\hat{s}}{s^2} = \oint d_1\frac{1}{s} = 0. $$
    where the subscript on 1 means i'm treating all the terms associated
    with loop 1 as variables and freezing all terms associated with loop 2.

    Just remember [itex] \nabla f \cdot d\mathbf{r} = df [/itex] and you're integrating
    over a loop that starts and end at the same place.
  4. Apr 10, 2013 #3
    Ah nice, I see now. Thanks for the trick.
  5. Jan 10, 2014 #4
    please could you explain why the integral over loop 1 is zero? I'm struggling to see your method.. :/
  6. Jan 10, 2014 #5

    Theres an important identity involving one of the terms. Once you use that identity, you can deduce by divergence theorem that the entire term for loop 1 vanishes.
  7. Jan 11, 2014 #6
    So if there is a loop current in a start say 10 light years away and I switch on a loop current in the earth now the Newtons third law holds true.

    Please, prove it with integrals and divergence theorem.

    By the way I did not see any t (time) variable in your formulae.
  8. Jan 11, 2014 #7


    Staff: Mentor

    Alva, the Biot Savart law uses the magnetostatic approximation. It doesn't apply in the far field and there is no time. This is a standard approximation, but it is an approximation.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook