MathGroup Archive 2011

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

Search the Archive

NDSolve::ndsz

  • To: mathgroup at smc.vnet.net
  • Subject: [mg123856] NDSolve::ndsz
  • From: Gausstein <gausstein at gmail.com>
  • Date: Fri, 23 Dec 2011 07:13:56 -0500 (EST)
  • Delivered-to: l-mathgroup@mail-archive0.wolfram.com

Greetings,
I have a problem solving two coupled differential equations using
NDSolve.
The following message appears: "NDSolve::ndsz: At t ==
-0.008080592178665635`,
step size is effectively zero; singularity or stiff system suspected.
>>"

Can NDSolve actually solve these equations? or should I better try
another program?
I have tried everything!!!
I just need to know if it is possible to solve this equations with
Mathematica.
The code used is the following:
*********************************************************************************************************
H0 = 1/5000000;
m = 3/500000000;
A = 1/10^2;
V0 = 3 H0^2;
V0a = V0 -  1/2 A m^2 (2^(2 - 18 (1 - (7 Sqrt[51])/50)) 5^(2 - 21 (1 -
(7  Sqrt[51])/50)));

ti = -(11/10);
tf = -Exp[-10];
V[t_] :=  V0 + 1/2 m^2 phi[t]^2 + UnitStep[t + 1] (-V0 + V0a + 1/2 A
m^2 phi[t]^2)

value1 = 2^(1 - 21/2 (1 - (7 Sqrt[51])/50)) 5^(1 - 12 (1 - (7
Sqrt[51])/50)) 11^(3/2 (1 - (7 Sqrt[51])/50));
value2 = -3 2^(1 - 21/2 (1 - (7 Sqrt[51])/50)) 5^(2 - 12 (1 - (7
Sqrt[51])/50) ) 11^(3/2 (1 - (7 Sqrt[51])/50) -
     1) (1 - (7 Sqrt[51])/50);

temp = NDSolve[{

Derivative[1][a][t]/a[t]^2 ==Sqrt[1/3 (1/2 (Derivative[1][phi][t]/
a[t])^2 +V[t])],
  Derivative[2][phi][t] + 2 Derivative[1][a][t]/a[t] Derivative[1][phi]
[t] + m^2 a[t]^2 phi[t] (1 + A UnitStep[t + 1]) == 0,
  a[ti] == -1/(H0 ti),
phi[ti] == value1,
phi'[ti] == value2},
{a, phi}, {t, ti, tf}, MaxSteps -> \[Infinity], InterpolationOrder ->
All]
***************************************************************************************************************
I know it looks pretty messy, but once it is copied into the Notebook
(and if
it is converted into StandardForm)  it gets clearer.
I really need help!!!!!!!!!!!!!!!!!!
Thanks a lot!!!



  • Prev by Date: Re: Span Set nasty bug
  • Next by Date: Algorhitm needed
  • Previous by thread: Continued Fraction Form With Negative Numbers
  • Next by thread: Re: NDSolve::ndsz