Hi Shashank,

maybe I did something wrong... I tried all of them but nothing really worked... also the individual time steps as you can see I tried...

Attached you find my last system and solve section

Dear Mike,

 welcome,

I will try to give a conceptual answer which may be suitable for use with any circuit analysis tool.You must at first define your excitation source precisely. In case of DC analysis you use a DC source or sources. In case of transient analysis you use normally square wave source or a unit step source or sources. In case of AC analysis you apply sinusoidal sources with specific amplitude and frequency and you are normally interested in the so called steady state solution in form of amplitudes and phase of the voltage or current in the circuit across the elements and through them.

For circuits containing electronic devices which are nonlinear normally the device is biased at certain DC operating point and then one can make the AC analysis by using the linearised AC equivalent circuit of the device.

In case of studying the switching of the devices one applies on them square wave excitation and study their transient performance assuming certain initial state.

However , one can study the transient performance of any device or circuit by applying a sinusoidal excitation. This is accomplishes by describing the source as 

v(t)= Vm sin (wt +phi) or the current source can be described in the same manner.

then one solves the circuit or component by following the change of the waveform with time, that is one makes transient analysis.

After the transient has elapsed the steady state solution is reached and the solution will be that of steady state.

You have to be sure that the physical and the mathematical models are able to tackle such problems.

In case of transient ac analysis, one solves the differential equations describing the components and their interactions. This must be clear in mind.

I hope that i shed some light on the type of analysis and their validity.

Best wishes

Dear Abdelhalim,

thank you for you response and the detailed explanation. That is exactly what I learned as electrical engineer for circuits.

As you wrote if you wanna analyse a single device in a circuit, you should ramp to DC and then apply AC as linearised equivalent circuit.

What I am interest now is, to simulate e.g. a capacitance and its time dependent behaviour for a pulse as input. In a first sense it does not matter, if it is sinusoidal or digital. The standard AC analysis as CV curve works fine and is no problem. Now instead of a continuous ramp from e.g. -10V to +10V a pulse you be applied and finally extracted over time. My expectation from analytic is a change of capacity over time and finally of charge.

I think the tooling is able to simulate. The question is how? It is a question of handling. Have you ever worked with Sentaurus?

Best

Mike

Dear Mike,

The problem becomes now more specific for me, you have a capacitor and you want to measure the change of its capacitance with pulse excitation and you expect to find that the capacitance of the capacitor changes with tome.

In order to make such simulation you have to apply on your capacitor a constant voltage source pulse v(t)  and sense the current i(t). Then c= i/ (dv(t)/dt).

Then what is remaining is to differentiate the pulse waveform to get  dv(t)/dt and calculates c.

May be the best waveform to the piece wise linear  waveform in ascending and descending manner. A sine wave also may be used with easy differentiation of the input voltage waveform.

It is not the matter of the simulator only it is the matter of precise definition of the quantities to be measured. If this is made then one can propose a measuring circuit for it. Here measuring is used as an alternative to simulation.

best wishes 

Dear Abdelhalim,

yes  that is excactly waht I wanna simulate and I try to setup. I agree, measurement for sure is the best you can do... but now I am in a situation, that the simulation shall help to define an experiment and concentrate on the materials to be studied... that is the reason why to simulate. Before simulating, for sure calibrate the simulation with existing experiments, to be predictable...

Thank you and best

Mike

Hi Shashank,

yes MOS cap... that is what I tried to do first, specify different timings to get the last step plot and then extract the caps manually... but it did not work at all... I saw no change of capacitance after different timings. Maybe the quasistationary command is not compatible in this way...

Hi Shashank,

I will try and let you know if it works...