Physics 329: Solution of ODEs
Please report all errors/typos. etc to 
   matt@infeld.ph.utexas.edu
- General Reference
   
   
-  Source code covered in class can be found on
        the phy329 account on einstein in
        ~phy329/ode/ex1 and ~phy329/ode/ex2
   
- Lectures 1 and 2 (Nov 18 and 20)
   
      - lsoda.f ODEPACK
      routine for solving general systems of ODEs.
      
- tlsoda.f Sample
      driver routine which uses lsoda to integrate
      u''(x) = -u(x).  Sample
      usage on SGIs.
      Plot of computed
      and exact solution on x = [0 .. 15] with all error tolerances
      set to 1.0d-6. Plot of
      error in computed solution.
   
 
- Lecture 3 (Nov 22)
   
      - integral.f:
          Driver routine illustrating use of LSODA (ODE integrator)
          to compute definite integral.
          fcn.f:
          User function which defines integrand.
          Makefile
      
- 2body.f:
          Driver routine illustrating use of LSODA to solve restricted
          2-body gravitational problem (one body non-gravitating).
          fcn.f:
          User function which implements equations of motion.
          Makefile
   
 
- Lecture 4 (Nov 25)
   
      - Equations of motion
          for orbiting dumbbell problem.
      
- dumb.f:
          Driver routine for solution of orbiting-dumbbell problem.
          fcn.f:
          User function which implements equations of motion.
          Makefile
      
- Plots of omega (angular frequency of dumbbell about its
          center of mass) versus time for a
          circular orbit
          and for an
          elliptical orbit.
          Note the "chaotic" nature of omega in the latter case.
          Detail
          of omega for the elliptical orbit.
      
- Plots illustrating energy conversation for elliptical
          orbit (chaotic case).
          Energy quantities
          computed with LSODA tolerance 1.0e-6.
          Top to bottom: Translational
          kinetic energy, rotational kinetic energy, total energy,
          gravitational potential energy.  In this case, "non-conversation"
          of total mechanical energy is a good sign that we need to
          make the error tolerance more stringent.
          Energy quantities
          computed with LSODA tolerance = 1.0e-12.   Note that
          energy conservation is improved in this case.
          Plot of
          rotational kinetic energy.