MathGroup Archive 2009

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Is this so hard?

  • To: mathgroup at smc.vnet.net
  • Subject: [mg95270] Re: Is this so hard?
  • From: "Steve Luttrell" <steve at _removemefirst_luttrell.org.uk>
  • Date: Wed, 14 Jan 2009 05:50:53 -0500 (EST)
  • References: <200901111138.GAA12346@smc.vnet.net> <gkgq3p$4a2$1@smc.vnet.net> <gkh0vt$6ia$1@smc.vnet.net>

Try solving it numerically.

t0 = 1.12;
sol = NDSolve[{x''[t] == 1/x[t]^2, x[0] == -1, x'[0] == 0},
  x, {t, 0, t0}]
Plot[(x /. sol[[1]])[t], {t, 0, t0}]

This gives the following error message:

"NDSolve::ndsz: At t == 1.1107203590726262`, step size is effectively zero; 
singularity or stiff system suspected."

There is a singularity where x[t] goes to zero, which makes the 1/x[t]^2 
term in the ODE blow up.

-- 
Stephen Luttrell
West Malvern, UK

"cool-RR" <ram.rachum at gmail.com> wrote in message 
news:gkh0vt$6ia$1 at smc.vnet.net...
>
>> You didn't want another initial (or boundary) condition for this 2nd
>> order ODE?
>>
>
> You're right.
>
> DSolve[{x''[t] == 1/x[t]^2, x[0] == -1, x'[0] == 0}, x, t]
>
> This still gives no answer.
>
> Any ideas?
> 



  • Prev by Date: Maximizing function which defines a routine
  • Next by Date: undefined function
  • Previous by thread: Re: Is this so hard?
  • Next by thread: Re: The use of Piecewise function