I Poisson's Equation: Solving for φ with ρ(x)

  • I
  • Thread starter Thread starter aaaa202
  • Start date Start date
AI Thread Summary
The discussion centers on the application of Poisson's equation when the charge density is a function of one coordinate, specifically ρ = ρ(x). It clarifies that simplifying the equation to d²φ/dx² = -ρ(x)/ε is not generally valid, as the potential φ is influenced by the entire charge distribution in space. The complexity of solving such problems is emphasized, noting that boundary and initial conditions are crucial for accurate solutions. Numerical methods, such as discretizing the charge density and using finite differences, may yield only a subset of solutions without proper boundary conditions. The conversation concludes that the problem is part of a script for solving Poisson's equation rather than a mere exercise.
aaaa202
Messages
1,144
Reaction score
2
Poissons equation states that:

2φ = -ρ/ε

Now suppose that the charge density is actually only a function of one coordinate ρ = ρ(x) but constant in y and z. Is the problem then equivalent to solving:

d2φ/dx^2 = -ρ(x)/ε

or what will the effect of the partial derivatives of y and z be in this case?
 
Physics news on Phys.org
No that's not generally true. Consider for example ##\rho(x) = x## then ##\varphi(\vec r) = -\frac{y^2}{2\epsilon}x## would be a solution.

To solve a problem like this can be quite complicated and depends on the boundary and initial conditions. You can learn how to solve boundary problems like this in a book/course about Fourier analysis.
 
Any charge, whatever its dependence on the coordinates may be, affects the potential at any point in space. The Poisson equation that you wrote is not a vector equation. So you cannot just take the x dependence on either side and say that they are equal. Each partial derivative on the left is related to the entire right hand side, and so is their sum.
 
Okay but I have a numerical problem where I am given ρ(x,y,z) = ρ(x). To solve for the electrostatic potential I then discretize ρ(x) on a grid of n points and approximate the second order derivative D = d2/dx2 as a matrix in the standard way using finite differences. I then calculate the electrostatic potential as:

φ = [d2/dx2]-1(-ρ/ε), where [d2/dx2]-1 is the inverted matrix of D written above.

Will this in general not give the correct solution? I guess not since, I am assuming that Poissons equation can be written as:

d2φ/dx2 = -ρ/ε
 
Is this an exercise? Maybe it should be in the exercise section then with all information provided. I don't know anything about numerical methods but if you assume the equation of that form doesn't mean you necessarily get a wrong solution. You simply only get a subset of all possible solutions. But imo the question doesn't make much sense without having boundary conditions.
 
The boundary condition is that the potential should vanish at infinity. It is not an exercise. Rather part of a script I am writing for solving Poissons equation.
 
Thread 'Gauss' law seems to imply instantaneous electric field'
Imagine a charged sphere at the origin connected through an open switch to a vertical grounded wire. We wish to find an expression for the horizontal component of the electric field at a distance ##\mathbf{r}## from the sphere as it discharges. By using the Lorenz gauge condition: $$\nabla \cdot \mathbf{A} + \frac{1}{c^2}\frac{\partial \phi}{\partial t}=0\tag{1}$$ we find the following retarded solutions to the Maxwell equations If we assume that...
I passed a motorcycle on the highway going the opposite direction. I know I was doing 125/km/h. I estimated that the frequency of his motor dropped by an entire octave, so that's a doubling of the wavelength. My intuition is telling me that's extremely unlikely. I can't actually calculate how fast he was going with just that information, can I? It seems to me, I have to know the absolute frequency of one of those tones, either shifted up or down or unshifted, yes? I tried to mimic the...

Similar threads

Back
Top