# NonlinearModelFit:cvmit: Failed to converge to the requested accuracy or Precision within 1000 iterations

I am trying to find some parameters from a model using NonlinearModelFit, but I always get the same error:

NonlinearModelFit:cvmit: Failed to converge to the requested accuracy or Precision within 1000 iterations.

My model is

model = (1/C1) Exp[-t/(R2 C2)] + (1/C3) Exp[-t/(R3 C3)] + R1 DiracDelta[t]


and the fit is

myfit = NonlinearModelFit[data, currentMax model, {C1, R2, C2, C3, R1}, t]


where data is a list of voltage values and currentMax is the maximum current.

-
Can you show a sample of your data ? –  b.gatessucks Jan 8 at 9:07
1. Try a non-default Method. 2. Try increasing MaxIterations. –  Oleksandr R. Jan 8 at 10:36
hey thanks you for your Answer it don't work. even wenn i make MaxIteration->1000 i become unwanted values.Method->"NMinimize" take a long time and Method->"LevenbergMarquardt" don't even work. any other suggestions plz? –  yassine karmoudi Jan 8 at 12:13
@b.gatessucks its a CSV list of 33000 values i dont know how can i upload it hier –  yassine karmoudi Jan 8 at 12:18
It was a real surprise to see DiracDelta appear on the right hand side, because this is not a function. Try it: its values are either $0$ or are given as DiracDelta[0], which is not even numeric. That's a show-stopper right there (but is easily fixable). What was unsurprising was to see the rest of the RHS: the parameters of the exponentials can be extremely difficult to identify because, for many ranges of their values, they give very similar graphs. Plot them and see! –  whuber Jan 8 at 18:11