Difficulties to handle with waveform

[jc-barreto]

Hello,

Due to a lack of documentation, I'm facing some difficulties in understanding the waveform

First, I have tried to reproduce the results from this example that Heagy went through in this video to compare with the results from SimpegEM1D but they are not similar.

Second, I have tried to understand why in some cases the time zero in the waveform is setting when the current is one, and in other situations, like in this example not. This is the point that makes me more confused

Finally, to achieve my goal that is to have a good forward that returns the Jacobian and the predicted data, I tried to compare the 3 waveforms: VTEM, TEMFAST, and trapezoidal.The real case I'm studying is a square loop performed with TEMFAST equipment and I don't understand why I can't predict well the data with the original waveform

Last question: why to extract the Jacobian I need to multiply by a vector?

J = []
for i in range(len(model)):
    ei = np.zeros_like(model)
    ei[i] = 1.0
    J.append(simulation.Jvec(model, ei, f=fields))

J = np.vstack(J)

Is this the Jacobian that is called several times during the inversion?