Simulating dependent sources

[sert]

I'm writing a circuit simulation program as an exercise and I'm
using the nodal method. The nodal method requires every branch
current to be a function of some branch voltage so the circuit
can be solved through the KCL.

What do we do with dependent voltage sources though? The only
way I can think of implementing them is by transforming them
into current sources, but that needs a resistor in series of the
voltage source.

 

[Kevin Aylward]

sert wrote:

I'm writing a circuit simulation program as an exercise and I'm
using the nodal method.

Some exercise...

The nodal method requires every branch
current to be a function of some branch voltage so the circuit
can be solved through the KCL.

What do we do with dependent voltage sources though? The only
way I can think of implementing them is by transforming them
into current sources, but that needs a resistor in series of the
voltage source.

This is a standard problem, and is solved in spice by using "Modified Nodal
Analysis" (MNA) as standard.

I would recomend the book "The Spice Book" by Andrei Vladimirescu ISBN
0-471-60926-9

Quoting bits and bobs from p.281-282

The Matrix equation

|G||V| = |I|

is replaced by

|G F| |V| = |C|
|B R| |I | |E|

To wit, "problem.. voltage sources...can not be included....this lead the
developers of spice to extend the set of nodal equations to include voltage
source equations represented by currents in the unknown vector and by
voltages in the RHS vector."

I will leave you to figure out the details of above..

Kevin Aylward
www.kevinaylward.co.uk
www.anasoft.co.uk - SuperSpice

 

sert wrote:

I'm writing a circuit simulation program as an exercise and I'm
using the nodal method. The nodal method requires every branch
current to be a function of some branch voltage so the circuit
can be solved through the KCL.

What do we do with dependent voltage sources though? The only
way I can think of implementing them is by transforming them
into current sources, but that needs a resistor in series of the
voltage source.

I know nothing of simulations, but all voltage sources have a series
reistance. A "good" voltage source has a low series reistance, as a
"good" current source has a large parallel resistance.

George Herold