Re: numerical solution of an ODE doesn't make sense

*To*: mathgroup at smc.vnet.net*Subject*: [mg49531] Re: numerical solution of an ODE doesn't make sense*From*: Alois Steindl <Alois.Steindl at tuwien.ac.at>*Date*: Fri, 23 Jul 2004 05:59:20 -0400 (EDT)*References*: <cdnqcj$l61$1@smc.vnet.net>*Sender*: owner-wri-mathgroup at wolfram.com

"Ita" <kitamar at actcom.net.il> writes: > Hello, and thanks in advance to any response. > I solved a 2nd degree ode with NDSolve for the unknown x(t) for 2 time > domains: 0<t<100 and 0<t<700, with the same boundary conditions. > than I ploted the two solutions for 0<t<20, and they came out different ! it > is like the solution in t<20 is influenced by the solution in t>100 - it > doesn't make sense. > thanks, > Itamar. Hello, this certainly depends on the differential equation and the boundary conditions! Consider the simple problem u'' = -1 u(0) = 0 u'(L) =0 (this might model the longitudinal deformation of an elastic rod under constant axial load). The solution strongly depends on L. Alois