# Nonlinear simulation

Status
Not open for further replies.

#### abhijitrc

What is the basc difference between spice and harmonic balance simulation.

#### johnjoe

Hi,

Harmonic balance is done in the frequency domain to calculate the steady-state respone. Which type of Spice simulation do you mean? Transient, DC, AC?

Kind regards

#### abhijitrc

Dear guys,

I am trying to develop a full nonlinear model of HEMT, I had adopted the methodology as below:
1.Extracting small signal s-parameters , and based on these extarcted s-parameters a table based small signal parameters have been calculated(converting the s-parameters to Y-parameters). Multi-bias approach has been carried out for this purpose.
2. once this has been achieved, i am trying to devise a nonlinear model using these small signal parameters with DC IV characteristics, these parameters have to be fitted with nonlinear model equations to predict the device behavior at different bias points at different power levels( curve fitting using ICCAP/ MDLGRED).
All these have been carried out using, EESOF ICCAP software.
Kindly let me know whether my approach is correct ?? For extraction i had TRL claibration .

#### abhijitrc

plz reply to my thread ! i am waiting for the valuable suggestions.

##### Super Moderator
Staff member
2. once this has been achieved, i am trying to devise a nonlinear model using these small signal parameters with DC IV characteristics, these parameters have to be fitted with nonlinear model equations to predict the device behavior at different bias points at different power levels( curve fitting using ICCAP/ MDLGRED).

This looks as though it might have to do with calculating things such as the resistance of a PN junction... which may have a certain amount of resistance inline... with a certain voltage being applied to the ends of the wire.

The voltage has to be above a certain threshold, in order for the diode to conduct.

Current cannot be greater than the ohmic resistance will permit for a given applied volt level.

I do this in my simulator, by first calculating current through the ohmic resistance alone.

Then, based on this current, I calculate voltage drop across the diode. My formula is more detailed than the linear model which simply assumes a drop of .6 or .7 V for a silicon diode.

I subtract this voltage from the applied voltage at the ends of the wire.

I recalculate current through the ohmic resistance.

I go through this process once or twice. It yields approximations which reflect the typical operating volt range of .3 to .8 V, corresponding to an ampere range of 1 uA to 1A.

Status
Not open for further replies.