HOW TO MAKE ODE14X RUN WITHOUT ERRORS?

Derivative input 24 of 'final3opamp/N-Channel MOSFET1' at time 3.0E-7 is Inf or NaN. Stopping simulation. There may be a singularity in the solution. If not, try reducing the step size (either by reducing the fixed step size or by tightening the error tolerances).

this is the error I get. I need a discret time solver in order to get the fft of some signals in my system. i use local solver(trapezoidal and backward) but it does nt produce quite good results- it is far too digital-many pulse-like oscillations ocure.

So I only need to run the fixed step ODE14X. any ideas?

here is a similar previous post http://www.mathworks.com/matlabcentral/answers/52196-residual-norm-and-unconverged-equation-error

thank you for helping.

[Andreas]

and when i changed the consistency tolerence to 1e-12 i get this kind of major error (which shows up in many occasions when i try to set the model up)

Steady state solve failed to converge. ...Nonlinear solver: failed to converge, residual norm too large. ......Here is the set of components with unconverged equations: 'final3opamp/N-Channel MOSFET' 'final3opamp/N-Channel MOSFET1'

[Andreas]

any ideas?

Answer 1

[Guy Rouleau]

Very likely this is because the dynamics in your model is too fast to be captured by the solver step size. You know... the Nyquist criteria.

This article should help moving from variable to fixed step:

http://www.mathworks.com/tagteam/61384_Real_Time_Simulation_Simscape_PDF.pdf

Answer 2

[ahmed]

Mr. Rouleau;

The PDF link was useful....thank you

But, still there is no direct solution to adapt the solver...! or even the system...!

My model is a Hybrid vehicle, including DC motor and Generic engine working together via PSD...the problem arises between the engine and the solver...any idea?

[doliph]

Hi ahmed,

I am doing a doing a similar simulation to you but for a fully electric vehicle. I am running into the same issues. Were you able to resolve the problem?

Cheers,

Dustin