Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Induced EMF and farady's law

  1. Aug 16, 2007 #1
    well iwas going through the standard example that is there is a uniform magnetic field in aregion going into the palne and ther is a rectangular 'loop' with one side which can slide on metallic rods with constant speed [tex]v[/tex] and the the relation

    well the example is (almost) here
    well i could intutively understand that from lorentz' force concept that when the side starts moving so do the electrons so they experience a force and a current is produced.

    well i wanted to go about it from farady's alw.well what i couldn't get is farady's law states when the "flux across a circuit...".but here our loop itself is changing so how do i utilize the law
  2. jcsd
  3. Aug 16, 2007 #2
  4. Aug 16, 2007 #3
    but here the 'loop' itself is not constant
  5. Aug 16, 2007 #4
    What do you mean by the loop is not constant ?

    Generally, the flux changes due to :
    1) change in the position of the loop with respect to the B-field (rotation or translation)

    2) the loop changes its form (rectangle becomes circle eg).

    The basic ingredient is the fact that the flux must change with respect to time ! If this does not happen, no potential is created.

  6. Aug 16, 2007 #5
    i mean here the rectangle grows in size it does have a change in area but this change is due to addition of 'more conducting element' as oppsoed to whenit changes toa circle in which the lopp is 'constant' but it's shape changes
  7. Aug 16, 2007 #6
    First of all, if it changes in surface, this means that the flux WILL change and a voltage will be created.

    But how does the rectangle increases its surface ? I don't get the "due to more conducting elements" reasoning.

  8. Aug 16, 2007 #7
    well waht i meant was consider the rectangular loop PQRS no suppose i start stretching it or change it into a circle in presence of a mganetic field then the flux changes and an 'emf' is induced across the loop PQRS.
    but here the loop PQRS not only schanges its' shape but also gets some additional new elements.
  9. Aug 17, 2007 #8
    How does the loop acquire additional new elemnts and what are these elements ? I really don't get it. Besides, what exactly do you want to know other than the consequence of the applicaton of Faraday's law which we covered here.

  10. Aug 17, 2007 #9
    well i got thsi ...i read that in griffith where it is explained nicely...well it is sheer coincidence he writes that the motional emf has the same formula as that for the other two cases...
  11. Aug 17, 2007 #10
    By motional emf you mean the moving loop right ? What same formula ? You mean Faraday's Law ?

    Actually, could you explain the solution to your problem to me because i don't get it.

  12. Aug 17, 2007 #11
    yes motional emf means moving coil but stationary fie;d(here).
    and the formula refers to faraday's law
    if u could get a copy of griffith then he very beutifully denies to accept that the emf induced due to motion of the cil as the one given by faraday's law.infact goes on to say that in some sense it is not the 'true' farady's law that holds here.
    and most probably for the reason which i was worrying about
    because if u go through faraday's law it starts witha line that "if through a loop..." so the loop is fixed and by loop i mean a collection some points.
    but while we consider the motional emf as in my case the rectangle is taken to be the loop whose one side is moving.so the 'points' which make up the loop are not the same.
    crudely speaking the loop itself is changing
    so there is no point in assuming 'farday's law' for granted...though he discovered for this case also.it is just sheer coincidence taht thsi happened
  13. Aug 17, 2007 #12
    That's incorrect. What matters is the variation of the flux with respect to time, inside the loop. The loop may vary in structure as well.

  14. Aug 17, 2007 #13
    flux through what
  15. Aug 17, 2007 #14


    User Avatar
    Staff Emeritus
    Science Advisor

    If you're hopelessly confused by the area changing (honestly, I don't see why), just perform a line integral of the electric field around some closed curve.


    You will find that the intergal over a curve of the electric field component pointing along the curve is non-zero.

    If you like, you can use a vector identity to convert this line integral into an area integral. This is known as stokes theorem. This is why people are talking about integrating the flux, this is justified by Stokes theorem.
    Last edited: Aug 17, 2007
  16. Aug 17, 2007 #15
    i am not worried about the changing area...and stokes theorem is fine
    also when u ask me to calculate the line integral it has to be about a fixed loop
    what i am worried about is when we say that flux through a loop changes ...then obviously the flux has to be through a fixed loop,obviously i f i vary my loop with time my flux has to change(in most cases) ,so what i am saying is though the faraday's law holds true in case of motional emf...but that's just because it coincidentally happened to be so...
    what i am opposing is 'direct' use of farady's law in getting the induced emf in case of motional emf..though it gives u the right answer but that won't be using faraday's laws in the right sense,as most books do (thankfully griffith didn't) and for the reason i stated above
  17. Aug 17, 2007 #16
    to clarify things further as in this case the we assume the loop to be the rectangle PQRS then calculate the flux through it at any time t to be [tex]\Phi=Blx[/tex]
    then we write [tex]E=\frac{d \Phi}{dt}=Blv[/tex] where [tex]v[/tex] is the speed of the movable side.
    but this makes no sense because the flux through the loop is not changing as the field is constant here.what actually is changing is the loop which we are considering so the calculated flux changes.though the flux through PQRS is not changing.
    also here the emf induced is best explained by lorentz force.
    it is just by sheere coincidence(and i repeat) that what we do is working out right.
  18. Aug 19, 2007 #17
    But can't you see the contradictio in terminis in the above two sentences ? First you say the flux is constant, then you say it changes. Again, the flux IS changing because the number of magnetic fieldlines in the loop changes with time. The reason of this change is the varying loop ! The flux can change due to 1) the B field or 2) the loop that encloses the field lines !

  19. Aug 19, 2007 #18
    yes what i am saying is the flux through any loop is not changing with time.what is changing is the loop which we have taken into consideration.
    say for example there is magnetic field uniform in a room.at time t i take my loop to be a square one inside the room.the next moment 'i' will take my loop to be a square one outside the room.so where will the induced emf be .

    again i say the thing is the proof by faraday' law is not all correct(at least to me).lorentz force proof seems right.
  20. Aug 19, 2007 #19
    Again, if the loop is changing the flux is also changing. I would advise you to carefully investigate the definition of FLUX. After having done that, you will understand that what you are saying is incorrect.
    Yes because the flux has changed. Again, make sure you look at the definition of flux.

  21. Aug 19, 2007 #20
    flux [tex]\phi = \int_{S} \vec B . \vec dS [/tex]
    by loop i mean any fixed loop .consider any loop u want to the flux through it doesn't change with time.what is changing is the loop u are considering .so just because u r changing the loop of your consideration why should an emf be developed.

    when u say the flux has changed can u tell me flux through what has changed?
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?

Similar Discussions: Induced EMF and farady's law
  1. Induced Emf (Replies: 25)

  2. Induced emf (Replies: 0)

  3. Induced EMF equation (Replies: 28)