When I try to fit my experimental data, why is the fitted curve a straight line

[couet]

I think the problem comes from your x limits:

root [1] 0.5*exp(-1280/0.5)
(double) 0.0000000