Potential Difference Problem - setting up the integral

5 replies · 1K views · last reply
Taulant Sholla
Messages
96
Reaction score
5

Homework Statement


Diagram of a thin horizontal rod along the x-axis from the origin to length L, with a point P1 on the y-axis at distance d to the left of the rod's near end; the rod is labeled with dashed-line measurements d and L, and the text above gives L = 12.0 cm and linear charge density lambda = cx.


Homework Equations


V=kq/x​

The Attempt at a Solution


I know the correct solution. It's...​
Equation V = c/(4πε0) times the integral from 0 to L of x dx/(d + x), with c and ε0 in the prefactor and integration limits 0 and L.

On my first attempt, rather than use (d+x) in the denominator and integrate from 0 to L, I instead used (x) and integrated from (d) to (L+d).
Integral setup: V = integral from d to L+d of k(dq)/x = integral from d to L+d of k(lambda dx)/x = integral from d to L+d of k(c x dx)/x = c/(4 pi epsilon_0) times integral from d to L+d of x dx/x.

This produces the wrong answer, but why does it produce the wrong answer? Both approaches seem to capture the setup and handle the integration correctly. Obviously I'm wrong, but - again - why/how am I wrong? Thank you.
 
Attachments
  • Equation V = c/(4πε0) times the integral from 0 to L of x dx/(d + x), with c and ε0 in the prefactor and integration limits 0 and L.
    Capture.JPG
    3.5 KB · Views: 677
  • Diagram of a thin horizontal rod along the x-axis from the origin to length L, with a point P1 on the y-axis at distance d to the left of the rod's near end; the rod is labeled with dashed-line measurements d and L, and the text above gives L = 12.0 cm and linear charge density lambda = cx.
    Untitled.png
    7.5 KB · Views: 848
  • Integral setup dV = integral from d to (L+d) of k(cx)dx divided by x, with k(cx) dx over x shown as a fraction.
    der.JPG
    6.4 KB · Views: 430
  • Integral V = integral from d to L+d of k(cx)dx divided by x, with the integrand written as k(cx) over x.
    der.JPG
    3 KB · Views: 396
  • Integral setup: V = integral from d to L+d of k(dq)/x = integral from d to L+d of k(lambda dx)/x = integral from d to L+d of k(c x dx)/x = c/(4 pi epsilon_0) times integral from d to L+d of x dx/x.
    der.JPG
    10.1 KB · Views: 351
##\lambda## is no longer ##cdx## if you change variables. You can avoid this kind of oversight by using a different integration variable name, e.g. ##u ## or ##v##, but not the same name ##x##
 
Thank you! However, question: where did I "change variables" and/or why is λ no longer cdx?
 
You changed the integration variable ##x## that runs from 0 to L into a variable with name e.g. ##u## that runs from d to d + L. In other words ##u = x+d##. When ##x = 0##, ##\lambda = cx=0##, but when ##u=d##, then ##\lambda \ne cu##.
 
Ah, thank you so much. This is very helpful. I appreciate your explanation very much!
 
You're welcome. That's what PF is for... :biggrin: