Date: Aug 21, 2013 2:46 AM
Author: Torsten
Subject: Re: PDE - spherical coordinates question (polar)

"Saad" wrote in message <kuvtva$guv$>...
> Hi,
> I'm trying to solve the following PDE using (pdepe).
> Does anyone have any tips on how to input the variable coefficients for this equation.
> dH/dt + 1/sinx d/dx(H^3 (sinx)^2)=0
> Thanks,
> Saad

pdepe is suited for PDEs which contain a second-order spatial derivative (in your case, this would be an expression involving d^2H/dx^2).
Since your equation is a first-order PDE, you will have to solve it using a different tool.
Differentiating out, you get
dH/dt + 3*H^2*sin(x)*dH/dx = -2*H^3*cos(x).

Best wishes