@user34722 So my implementation for the SDE is correct ?

I can't check all of the details and numerical factors, but I can at least say that the factor np.sqrt(dt) * eta[t] looks correct. Have you tried comparing the output of your code to any of the plots in the slide deck? If you can reproduce the plots on slide 35, then you can be pretty sure that your code is correct.

@user34722 Yes, well that plot I can reproduce. Btw isn't the noise factor that we have considered in this eqn due to the current source only ?

OK that's great! So that suggests that your code is correct and there is something going on in the experiment.

The stochastic term models current noise, but if you use the formula for Johnson noise, then you are assuming that the dominant noise source is the resistor in parallel with your Josephson junction (the R in the RCSJ circuit). The current source could have additional noise that is much larger than Johnson noise at 70 K.

Anyway, current source noise is my best guess. Can you guess any other noise sources that could explain the discrepancy?

Of course keep in mind that I have no details about your experimental setup, so I can't be sure if the RCSJ model is appropriate or if the parameters are correct.

@user34722 Okay I'll definitely check it out. I also wanted to ask you another thing, in these ideal graphs you can clearly see that for current > 10 * Ic it basically becomes a straight line. I had actually devised a different type of josephson junction where the graph was following something of the lines V = aI^2 + b for around 10-15* Ic (till where I have data available), do you have any idea whether combination of different josephson junctions (either series or parallel ) can lead to this ?