The Electric Field Due to a Point Charge — Question 9

Section 22-3 · Physics pages 3528–3529

Question 9 A distant pair resembles one charge with a calculable finite-size correction Two equal positive point charges Q lie at x = ±a. Observe their field at x > a. A point approximation replaces them by charge 2Q at the origin. x Q Q P Schematic; labels, not drawing lengths, specify the data. (a) Derive the exact field and its ratio to the point approximation. (b) Evaluate the fractional excess over the point approximation at x = 10a and derive the leading far-field correction.
Show solutionHide solution
Question 9 - Solution A distant pair resembles one charge with a calculable finite-size correction x Q Q P (a) Method and calculation Eexact = kQ[(x −a)−2 + (x + a)−2] = 2kQ(x2 + a2) (x2 −a2)2 . Epoint = 2kQ/x2, Eexact Epoint = 1 + u2 (1 −u2)2 , u = a/x. (b) Interpretation and further calculation u = .1 : Eexact/Epoint −1 ≈.0305071. 1 + u2 (1 −u2)2 = 1 + 3u2 + O(u4). The leading error is quadratic in source size divided by distance because the symmetric first-order terms cancel. Physical check The fractional error here is explicitly normalized by the approximate field.

Question 9

A distant pair resembles one charge with a calculable finite-size correction

Two equal positive point charges Q lie at x=±ax=\pm a. Observe their field at x>ax>a. A point approximation replaces them by charge 2Q2Q at the origin.

Diagram shown in the original SVG worksheet.

  1. Derive the exact field and its ratio to the point approximation.

  2. Evaluate the fractional excess over the point approximation at x=10ax=10a and derive the leading far-field correction.

Question 9 - Solution

A distant pair resembles one charge with a calculable finite-size correction

Diagram shown in the original SVG worksheet.

(a) Method and calculation

Eexact=kQ[(x−a)−2+(x+a)−2]=2kQ(x2+a2)(x2−a2)2.E_{\mathrm{exact}}=kQ[(x-a)^{-2}+(x+a)^{-2}]=\frac{2kQ(x^2+a^2)}{(x^2-a^2)^2}.Epoint=2kQ/x2,EexactEpoint=1+u2(1−u2)2,u=a/x.E_{\mathrm{point}}=2kQ/x^2,\quad\frac{E_{\mathrm{exact}}}{E_{\mathrm{point}}}=\frac{1+u^2}{(1-u^2)^2},\quad u=a/x.

(b) Interpretation and further calculation

u=.1:Eexact/Epoint−1≈.0305071.u=.1:\quad E_{\mathrm{exact}}/E_{\mathrm{point}}-1\approx.0305071.1+u2(1−u2)2=1+3u2+O(u4).\frac{1+u^2}{(1-u^2)^2}=1+3u^2+O(u^4).The leading error is quadratic in source size divided by distance because the symmetric first-order terms cancel.

Physical check

The fractional error here is explicitly normalized by the approximate field.