CSE245: Computer-Aided Circuit
Simulation and Verification
Spring 2006
Chung-Kuan Cheng
Administration
CK Cheng, CSE 2130, tel. 534-6184, ckcheng@[Link]
Lectures: 9:30am ~ 10:50am TTH U413A 2
Office Hours: 11:00am ~ 11:50am TTH CSE2130
Textbooks
Electronic Circuit and System Simulation Methods
T.L. Pillage, R.A. Rohrer, C. Visweswariah, McGraw-Hill
Interconnect Analysis and Synthesis
CK Cheng, J. Lillis, S. Lin, N. Chang, John Wiley & Sons
TA: Vincent Peng (hepeng@[Link]), Rui Shi
(rshi@[Link])
Outlines
1. Formulation (2-3 lectures)
2. Linear System (3-4 lectures)
3. Matrix Solver (3-4 lectures)
4. Integration (3-4 lectures)
5. Non-linear System (2-3 lectures)
6. Transmission Lines, S Parameters (2-3 lectures)
7. Sensitivity
8. Mechanical, Thermal, Bio Analysis
Grading
Homeworks and Projects: 60
Project Presentation: 20%
Final Report: 20%
Motivation
Why
Whole Circuit Analysis, Interconnect Dominance
What
Power, Clock, Interconnect Coupling
Where
Matrix Solvers, Integration Methods
RLC Reduction, Transmission Lines, S Parameters
Parallel Processing
Thermal, Mechanical, Biological Analysis
Circuit Simulation
Input and setup
Circuit
Simulator:
Solve CdX/dt=f(X) numerically
Output
Types of analysis:
CdX(t)/dt=GX(t)+BU(t)
Y=DX(t)+FU(t)
DC Analysis
DC Transfer curves
Transient Analysis
AC Analysis, Noise, Distortions, Sensitivity
Program Structure (a closer look)
Models
Input and setup
Numerical Techniques:
Formulation of circuit equations
Solution of ordinary differential equations
Solution of nonlinear equations
Solution of linear equations
Output
CSE245: Course Outline
Formulation
RLC Linear, Nonlinear Components,Transistors, Diodes
Incident Matrix
Nodal Analysis, Modified Nodal Analysis
K Matrix
Linear System
S domain analysis, Impulse Response
Taylors expansion
Moments, Passivity, Stability, Realizability
Symbolic analysis, Y-Delta, BDD analysis
Matrix Solver
LU, KLU, reordering
Mutigrid, PCG, GMRES
CSE245: Course Outline (Cont)
Integration
Forward Euler, Backward Euler, Trapezoidal Rule
Explicit and Implicit Method, Prediction and Correction
Equivalent Circuit
Errors: Local error, Local Truncation Error, Global Error
A-Stable
Alternating Direction Implicit Method
Nonlinear System
Newton Raphson, Line Search
Transmission Line, S-Parameter
FDTD: equivalent circuit, convolution
Frequency dependent components
Sensitivity
Mechanical, Thermal, Bio Analysis
Lecture 1: Formulation
KCL/KVL
Sparse Tableau Analysis
Nodal Analysis, Modified Nodal Analysis
*some slides borrowed from Berkeley EE219 Course
Formulation of Circuit Equations
Unknowns
B branch currents (i)
N node voltages
(e)
B branch voltages (v)
Equations
N+B Conservation Laws
B Constitutive Equations
Branch Constitutive Equations
(BCE)
Ideal elements
Element
Branch Eqn
Resistor
v = Ri
Capacitor
i = Cdv/dt
Inductor
v = Ldi/dt
Voltage Source
v = vs, i = ?
Current Source
i = is, v = ?
VCVS
vs = AV vc, i = ?
VCCS
is = GT vc, v = ?
CCVS
vs = RT ic, i = ?
CCCS
i =A i,v=?
Conservation Laws
Determined by the topology of the circuit
Kirchhoffs Voltage Law (KVL): Every circuit
node has a unique voltage with respect to the reference node.
The voltage across a branch eb is equal to the difference
between the positive and negative referenced voltages of the
nodes on which it is incident
No voltage source loop
Kirchhoffs Current Law (KCL): The algebraic sum
of all the currents flowing out of (or into) any circuit node is
zero.
No Current Source Cut
Equation Formulation - KCL
R3
R1
Is5
R4
G2v3
0
i1
i
2
1 1 1 0 0
0
i
0 0 1 1 1 3
0
i
4
i5
Ai = 0
N equations
Kirchhoffs Current Law (KCL)
Equation Formulation - KVL
R3
R1
Is5
R4
G2v3
0
v1
v
2
v3
v4
v5
1
1
0
0
0
0
e1
1 1 0
e2
0 1
0
0
0 1
v - AT e = 0
B equations
Kirchhoffs Voltage Law (KVL)
Equation Formulation - BCE
R3
R1
R4
G2v3
Is5
R
1
0 G2
1
0
R3
0
0
0
1
R4
0
v
i
0
1 1
0 v
2 i2 0
0 v3 i3 0
v4 i4 0
0 v i i
5 5 s5
0
Kvv + i = is B equations
Equation Formulation
Node-Branch Incidence Matrix
branches
n
o 1
d 2
e
s i
1 2 3
(+1, -1, 0)
Aij =
+1 if node i is terminal + of branch j
-1 if node i is terminal - of branch j
0 if node i is not connected to branch j
Equation Assembly (Stamping
Procedures)
Different ways of combining Conservation
Laws and Constitutive Equations
Sparse Table Analysis (STA)
Modified Nodal Analysis (MNA)
Sparse Tableau Analysis (STA)
1. Write KCL:
2. Write KVL:
3. Write BCE:
A
0
K i
0
I
Kv
Sparse Tableau
Ai=0
v -ATe=0
Kii + Kvv=S
0
AT
i 0
v 0
0 e S
(N eqns)
(B eqns)
(B eqns)
N+2B eqns
N+2B unknowns
N = # nodes
B = # branches
Sparse Tableau Analysis (STA)
Advantages
It can be applied to any circuit
Eqns can be assembled directly from input data
Coefficient Matrix is very sparse
Problem
Sophisticated programming techniques and data
structures are required for time and memory
efficiency
Nodal Analysis (NA)
1. Write KCL
Ai=0
(N eqns, B unknowns)
2. Use BCE to relate branch currents to branch
voltages
i=f(v)
(B unknowns B unknowns)
3. Use KVL to relate branch voltages to node voltages
4. v=h(e)
(B unknowns N unknowns)
Yne=ins
Nodal Matrix
N eqns
N unknowns
N = # nodes
Nodal Analysis - Example
R3
R1
R4
G2v3
Is5
1. KCL:
2. BCE:
3. KVL:
Ai=0
Kvv + i = is i = is - Kvv A Kvv = A is
v = ATe A KvATe = A is
1
1
R G2 R
3
1
1
R3
1
G2
R3
1
1
R3 R4
e1 0
e i
2 s5
Yne = ins
Nodal Analysis
Example shows NA may be derived from STA
Better: Yn may be obtained by direct
inspection (stamping procedure)
Each element has an associated stamp
Yn is the composition of all the elements stamps
Nodal Analysis Resistor
Stamp
Spice input format: Rk
N+
Rk
N-
N+
N-
N+ N-
N+
N-
1
Rk
1
Rk
1
Rk
1
Rk
Rkvalue
What if a resistor is
connected to ground?
.
Only contributes to the
diagonal
1
iothers R eN eN is
k
KCL at node N+
1
iothers R eN eN is
k
KCL at node N-
Nodal Analysis VCCS Stamp
Spice input format: Gk
NC+
NC-
i
i
others
others
Gkvalue
N+
+
vc
N+ N- NC+ NC-
NC+
N+ Gk
G
k
N-
Gkvc
Gk eNC eNC is
Gk eNC eNC is
NKCL at node N+
KCL at node N-
NC-
Gk
Gk
Nodal Analysis Current source
Stamp
Spice input format: Ik
N+ N- Ikvalue
N+
N+ NN+
Ik
N-
N-
Ik
I
k
Nodal Analysis (NA)
Advantages
Yn is often diagonally dominant and symmetric
Eqns can be assembled directly from input data
Yn has non-zero diagonal entries
Yn is sparse (not as sparse as STA) and smaller than
STA: NxN compared to (N+2B)x(N+2B)
Limitations
Conserved quantity must be a function of node variable
Cannot handle floating voltage sources, VCVS, CCCS, CCVS
Modified Nodal Analysis (MNA)
How do we deal with independent voltage sources?
+
Ekl
k
ikl
1 ek
1 el
0 ikl Ekl
ikl cannot be explicitly expressed in terms of node
voltages it has to be added as unknown (new column)
ek and el are not independent variables anymore a
constraint has to be added (new row)
MNA Voltage Source Stamp
Spice input format: Vk
Ek
N+
N-
ik
N+ N-
Ekvalue
N+ N- ik
N+ 0
0 1
N- 0 0 -1
Branch k 1 -1 0
RHS
0
0
Ek
Modified Nodal Analysis (MNA)
How do we deal with independent voltage sources?
Augmented nodal matrix
Yn
C
B e
MS
0 i
Some branch currents
In general:
Yn
C
e
i MS
MNA General rules
A branch current is always introduced as and
additional variable for a voltage source or an
inductor
For current sources, resistors, conductors and
capacitors, the branch current is introduced
only if:
Any circuit element depends on that branch current
That branch current is requested as output
MNA CCCS and CCVS
Stamp
MNA An example
1
R1
+ v3 R3
R4
G2v3
0
Step 1: Write KCL
i1 + i2 + i3 = 0
-i3 + i4 - i5 - i6 = 0
i6 + i8 = 0
i7 i8 = 0
Is5
ES6
+
E7v3
+
R8
(1)
(2)
(3)
(4)
MNA An example
Step 2: Use branch equations to eliminate as many branch currents
as possible
1/R1v1 + G2 v3 + 1/R3v3 = 0
(1)
- 1/R3v3 + 1/R4v4 - i6 = is5
(2)
i6 + 1/R8v8 = 0
(3)
i7 1/R8v8 = 0
(4)
Step 3: Write down unused branch equations
v6 = ES6
v7 E7v3 = 0
(b6)
(b7)
MNA An example
Step 4: Use KVL to eliminate branch voltages from previous
equations
1/R1e1 + G2(e1-e2) + 1/R3(e1-e2) = 0
(1)
- 1/R3(e1-e2) + 1/R4e2 - i6 = is5
(2)
i6 + 1/R8(e3-e4) = 0
(3)
i7 1/R8(e3-e4) = 0
(4)
(e3-e2) = ES6
(b6)
e4 E7(e1-e2) = 0
(b7)
MNA An example
1
1
2
R
R3
1
1
R3
1
G2
R3
1
1
R3 R4
0
0
E7
1
E7
1
R8
1
R8
1
0
1
R8
1
R8
0
1
e 0
1
1 0 e
i
2
s
5
e 0
3
1 0
e4 0
0 1 i6 ES 6
i7 0
0 0
0 0
Yn
C
B
0
e
i MS
Modified Nodal Analysis (MNA)
Advantages
MNA can be applied to any circuit
Eqns can be assembled directly from input data
MNA matrix is close to Yn
Limitations
Sometimes we have zeros on the main diagonal
