See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/228932393

Symbolic equation using modified nodal analysis for linear electrical circuit
using Matlab

Article · January 2010

CITATIONS READS

0 1,598

5 authors, including:

Zoltan Erdei Olivian Chiver

Universitatea de Nord din Baia Mare Universitatea Tehnica Cluj-Napoca
28 PUBLICATIONS 61 CITATIONS 53 PUBLICATIONS 98 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Zoltan Erdei on 05 June 2014.

The user has requested enhancement of the downloaded file.

 Proceedings of the 4th EUROPEAN COMPUTING CONFERENCE

Symbolic equation using Modified Nodal Analysis for

linear electrical circuit using Matlab
Zoltan ERDEI, Luiza Alexandra DICSO, Liviu NEAMT, Oliver CHIVER
Department Electrical Engineering
North University of Baia Mare
Str. Victor Babes Nr. 62/A, 430083
ROMANIA
[email protected]

Abstract: - In this paper it is presented a program which generates the modified nodal equation for
electric analog circuits in a symbolic, partial symbolic and numerical mode. The program is an
application, made in the environment of the program MATLAB version 7.1, which has a powerful
symbolic math toolbox.

Key-Words: - analog circuits, circuit functions, symbolic equations, Matlab, netlist, modified nodal analysis.

1 Introduction numerical equations using modified nodal

The modified nodal methods have been analysis for linear electrical circuits;
implemented in a program on a Pentium Dual Core - generates symbolic form of the extensive matrix
computer compatible to obtain symbolic forms, of the system equations and determines the
partly symbolic or numerical equations for symbolic solution, partly symbolic or numerical
electronic linear analog circuits. Between the input to the circuit system;
gate and the output one specified by the user, any of
the four types of circuit functions (transfer Data entry is a set of lines describing nl circuit
impedance - Zei, admittance transfer function - Yei, branch. Each branch of the circuit is described by a
voltage transfer factor (amplifier) - Aei and current proper letter which indicates the type of circuit
transfer factor (amplifier) – Bei) for linear analog elements, the first node, final node, the nominal
electric circuits around a functionary point. Starting value and the value of the tolerance. In the case of
from the description of the circuit through to a file- controlled sources we should also introduce nodes of
type input netlist, according to the window shown in command branch. The netlist is similar to those
figure 1, the following capabilities: described in the program SPICE.
- generates symbolic, partly symbolic or

Fig.1 Application window

ISSN: 1790-5117 73 ISBN: 978-960-474-178-6

 Proceedings of the 4th EUROPEAN COMPUTING CONFERENCE

2 Modified Nodal Analysis (MNA) are used for reconstruction of RGB and component
To solve the circuit and obtain the desired circuit video (Y, Pb, and Pr) signals. They are placed
function we use the modified nodal analysis. following the video DAC to remove the higher
In the modified nodal analysis the circuit matrix is frequency replicas of the signals, as well as before the
obtained by the augmentation of the nodal ADC for anti-aliasing. Figure 2 shows a one-channel,
conductance (admittance) matrix corresponding to dual-supply configuration incorporating the
the NA- compatible circuit elements with additional
MAX4382. It is a three-pole, Sallen-Key Butterworth
rows and columns for non-NA-compatible circuit
elements. lowpass filter, in which the current DAC generates the
The circuit equations in operational corresponding video signal, and the resistor (RL) sets the amplitude.
to the modified nodal analysis method, have the With the MAX4382, the RL, R1, R2, C1, and C2 form
following form: a two-pole, Sallen-Key lowpass filter having a gain of
2. The driving load (75Ω) at the output, plus RT and
Yn−1,n−1 ( s) B n−1,m ( s ) Vn−1 ( s) I sc ,n−1 ( s) Cp, sets the real pole.
A ⋅ = , (1) In the Figure 2 circuit, the -3dB bandwidth is about
 m ,n−1 ( s ) Z m ,m ( s )   I m ( s)   E m ( s)  30MHz. The attenuation is approximately 14dB at
44.25MHz, and 28dB at 74.25MHz. The group delay
Where: - is the admittance matrix corresponding to
is roughly 6.5ns. If the current DAC load is different
the n-1 independent nodes; is an (n-1)xm matrix
and it contains the elements -1, 0, +1 and the current than 75Ω, just use the following relationship to set the
gains of the CCCS’s; is an mx(n-1) matrix value of R1: R1 + RL = 150Ω. For RL greater than
containing the elements -1, 0, +1 and the voltage 150Ω, C1, C2, R1, and R2 will need to be adjusted.
gains of the VCVS’s; is a mxm matrix having the
transfer impedances of the CCVS’s; is node
voltage vector corresponding to the n-1 independent
nodes. The vector represents the current vector
corresponding to the non-NA compatible circuit
branches and it has the following structure:

[ ( )( )(
Im(s) = (Ie (s)) , Iec (s) , IeC (s) , I jC (s) , (IL (s)) t, (2)
t t t
)
t t
]
where : (I e (s) ) is the ideal voltage source current
t

vector; (I ec (s ) ) t
the controlled (output) branch
current vector of all controlled voltage sources;
(I eC (s ) )
t
the controlling (input) branch current
vector of the CCVS’s; (I jC (s ) )
t
the controlling Fig.2 Lowpass Sallen-Key filter.
branch current vector of the CCCS’s. The vectors The input file for this example has the following
I sc ,n−1 ( s ) and E m (s ) represent the contributions of structure:
the excitation sources (independent current and
voltage sources). 6
R1 1 0 75 0.02
R2 1 2 75 0.02
R3 2 3 150 0.02
3 Description of the application R4 5 0 590 0.02
To explain and understand this application we R5 6 5 590 0.02
consider the following circuits (figure 2). This R6 6 4 75 0.02
application describes how to use a lowpass Sallen-Key Rf 3 5 100000 0.00001
C1 3 0 0.000000000022 0.1
filter with a dual supply voltage. The circuit provides a
C2 2 6 0.000000000022 0.1
Butterworth response with a 30MHz bandwidth, and is C3 4 0 0.000000000120 0.1
ideal for video-reconstruction filtering in HDTV E1 1 0 1 0.1
applications. In HDTV applications, lowpass filters G1 3 5 0 6 1000000

ISSN: 1790-5117 74 ISBN: 978-960-474-178-6

 Proceedings of the 4th EUROPEAN COMPUTING CONFERENCE

1/R1+ 1/R2 - 1/R2 0 0 0 0 1 0

 -1/R2 1/R2 + 1/R3 + s * C2 - 1/R3 0 0 - s * C2 0 0 

 0 - 1/R3 1/R3+ 1/Rf + s * C1 0 - 1/Rf 0 0 0
 
0 0 0 1/R6 + s * C3 0 - 1/R6 0 0
Aei = 
 0 0 - 1/Rf 0 1/R4 + 1/R5 + 1/Rf -1/R5 0 0
 
 0 - s * C2 0 -1/R6 - 1/R5 1/R5 + 1/R6 + s * C2 0 − 1
 1 0 0 0 0 0 0 0
 
 0 0 A 0 −A -1 0 0 
circuitelor electrice - Vol. II – Fundamentare
The A coefficient matrix obtained with the program teoretica, Aplicatii, Algoritmi si Programe de calcul,
for this example will be as bellows: Editura All Educational S.A., Bucuresti 2000.
The admittance transfer function for this example in a [6] M. Iordache, Lucia Dumitriu, Simularea asistata
symbolic form: de calculator a circuitelor analogice, Editura
POLITEHNICA Press, Bucuresti 2002.
Yei=GA*(R5+R4)/(1+s*C3*R6)/(R4*R5+R5*Rf+R4* [7] M. Iordache, Lucia Dumitriu, I. Matei, ASINOM
Rf+R5*R3+R5*s*C2*R2*R3- – Analiza SImbolica bazata pe Metoda Nodala
R2*R5*GA*Rf*s*C2+R2*R5*Rf*s*C2+R5*Rf*s*C1 Modificata, Manual de utilizare, Catedra de
*R3+ Electrotehnica, U.P.B., Bucuresti, 1999.
R2*R5*Rf*s*C1+R2*R5*Rf*s^2*C1*R3*C2+R2*R5 [8] M.Iordache, Lucia Dumitriu, PACER – Program
+R2*Rf*R4*s*C2+Rf*R4*s*C1*R3+R2*Rf*R4*s*C de Analiza a Circuitelor Electronice Rezistive.
1+R2*Rf*R4*s^2*C1*R3*C2+R2*R4*s*C2*R3+R4* [9] M.L. Piper and Z. Erdei FEM Analysis of Squirrel
R3+R2*R4+GA*Rf*R4+R2*R5*s*C2*R4+GA*Rf*s Cage Induction Motor with Broken Bars ATEE 2002
*C1*R3*R4+R5*s*C1*R3*R4+R2*GA*Rf*s*C1*R4 Bucureşti;
+R2*R5*s*C1*R4+R2*GA*Rf*s^2*C1*R3*C2*R4+ [10] Zoltan ERDEI Analiza toleranŃelor circuitelor
R2*R5*s^2*C1*R3*C2*R4)*Rf*E1 electrice – proiect de diplomă, Facultatea de
electrotehnică, U:P:B. 2002.
4 Conclusion [11] Zoltan ERDEI Fast Monte - Carlo Tolerance
Analysis SMACD 2004 Krakow, Polonia
The main advantage of the MNA method is that the
contributions of all individual circuit elements can be
inserted directly in appropriate places in (1),
independently of whether the primary variable of a
particular element is a voltage or a current.
MNA method associated with an efficient nod tearing
leads to an important reduction in computing time and
to an increase of the accuracy.
This is a new software environment that allows the
schematic representation, analysis and design for linear
analog circuits.

References:
[1] R. R. Boyd , Tolerance analysis of Electronic
Circuits using Matlab, CRC Press, Washington 2002.
[2] A. E. Schwarz, Computer-aided design of
microelectronic circuits and systems, Academic Press,
London, 1987.
[3] H. J. Orchard, G. C. Temes, “Filter design using
transformed variables”, CT, 15, Dec. 1968.
[4] Lucia Dumitriu, M. Iordache, Teoria moderna a
circuitelor electrice - Vol. I – Fundamentare teoretica,
Aplicatii, Algoritmi si Programe de calcul, Editura All
Educational S.A., Bucuresti 1998.
[5] M. Iordache, Lucia Dumitriu, Teoria moderna a

ISSN: 1790-5117 75 ISBN: 978-960-474-178-6

View publication stats