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

Static electric fields

  1. Mar 18, 2010 #1
    Jus chek out careful when we have uniform electric charge distribution along wire or a sphere etc! Is there no any interaction between the charges in that uniform distribution!

    then look at the capacitor we only see straight field lines from positive charged plate to the negative charged plate are no more electric interaction that i expected ie. Between the adjacent charges on a plate or between two different charges in two different plates that are not connected by a straight horizontal field line!

    may any one help me out on what realy hapens?
     
  2. jcsd
  3. Mar 19, 2010 #2

    collinsmark

    User Avatar
    Homework Helper
    Gold Member

    The charges do interact. :smile: They repel each other.

    You may have heard that within a charged conductor, the (excess) static charges are always on the surface (and only on the surface). That's because they are repelling each other. In essence, they are "trying" to get as far away as they can from any other charge or group of charges. But since all the other charges are "trying" to the same thing, they just end up spreading out over the surface. Eventually, the charge distribution of the surface charges on a conductor align themselves in just such a way that the electric field in the conductor is 0.

    Your textbook might not show the electric field lines in between charges on the same plate, only because it's probably not relevant to the problem. But they are there.

    And the electric field lines in-between the plates do curve somewhat as you move toward the edge. Often this is ignored in a typical capacitor because the distance between the plates is so tiny compared to the length of the plates.
     
  4. Mar 19, 2010 #3
    Thanks man! But what i know is that every charge radiates electric fields radialy outwards

    but what i clearly see on a uniformly charged object is that only a single field among many seems to be shown

    also what hapens until a charge on one end of a linear charge distributed conductor cant cause electric flux on the plates of the cylindrical gausian surface drawn!

    may u tel me clearly on the interaction and at the same time gauss law says that the net sum of flux is the algebraic sum of the flux due to each charge while the diagram shows that some flux are neglected and i dont really understand what hapens to it?
     
  5. Mar 19, 2010 #4

    collinsmark

    User Avatar
    Homework Helper
    Gold Member

    Yes, but since there are many charges being modeled (an infinite amount of infinitesimal charges in a uniformly charged object), the electric field often cancels out in all components except for the component perpendicular to the surface (this is certainly true for a conductor).

    Imagine an infinitely long, charged wire laying horizontally on the x-axis. A given infinitesimal point charge will have electric field lines drawn spherically outward. So some of its electric field will be drawn in the general direction of the positive x direction. But there is also another infinitesimal charge right next to it, slightly further along the x-axis, who's field lines point in the negative x-axis direction (in-part), canceling out the field line components of the first charge, but only in the x direction. Now if you put an infinite amount of infinitesimal charges on the line all electric field lines cancel except for the field lines pointing radially away from the line.

    Yes, this is due to cancellation of the different components of the different charges. Everything sums up to zero, except for the component pointing radially outward. In a charged conductor, this direction is always perpendicular to the surface, when measured at the surface itself.

    Guass' law only works for an infinitely long charged wire or cylinder, when using a cylindrical Gaussian surface. If the charged wire/cylinder is less than infinitely long, Gauss' law is only an approximation. But it's a pretty good approximation if the place of interest is not near one of the edges, and if the distance to the cylinder/wire is small compared to the length of the cylinder/wire.
     
    Last edited: Mar 19, 2010
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook