In the circuit, during 0-1s there is a constant voltage, and during 1-2s a sinusoidal voltage.
For the a Simulation Start Time at 0s I get a constant voltage, and at 1s a sinusoidal voltage, as expected.
However, when I change the Initial cycles to 2 or higher, with Simulation Start Time at 0s, I get the following figure:
I am surprised because when I set Initial cycles to 50, with a system period of 0.02s, I would expect the same plot as for the one of Simulation Start Time at 1 s. Why does it look different?
The reason I am asking this is because I want to apply this on a more complex circuit that requires an initialisation process, and I was planning to set the Simulation start at 0s and then the required initial cycles.
The important distinction is that the Initial cycles parameter does not advance the simulation time by N × T. Instead, the Steady-State Analysis repeatedly simulates the same system period while updating the initial conditions between iterations.
For example, with a system period of T = 0.02 s:
Simulation Start Time = 0 s: Each initial cycle runs from 0 to 0.02 s.
Simulation Start Time = 1 s: Each initial cycle runs from 1 to 1.02 s.
Initial cycles = 50: The same time interval is simulated 50 times, using the final states of one cycle as the initial states of the next.
Therefore, setting the Simulation Start Time to 0 s with 50 initial cycles is not equivalent to running a continuous simulation from 0 to 1 s.
This also explains the behavior you are observing. Since your voltage source changes its behavior at t = 1 s, increasing the number of initial cycles will not move the analysis into the sinusoidal portion of the source signal.
In your case, you should set the Simulation Start Time to a point where the system has entered its intended periodic operating mode.
