Variational Method of Analysis For Micro Strip Lines

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 15

Variational Method of Analysis for Microstrip Lines

Purti R. S.

Introduction
Microstrip is a type of electrical transmission line which can be fabricated using Printed Circuit Board technology, and is used to convey Microwave-frequency signals. It consists of a conducting strip separated from a ground plane by a dielectric layer known as the substrate.

Microstrip
The dielectric in between the strip and the ground plane does not fill the air region above the strip Hence most field lines in the dielectric region concentrate between the strip and the ground plane While some concentrate in the region above the substrate Therefore the medium is inhomogeneous

Quasi TEM fields


Fundamental mode of propagation in homogeneous medium pure TEM But in case of microstrip, phase velocity of the TEM fields in the dielectric is c/r whereas in air region it would be c . Due to the phase mismatch, a microstrip line cannot support a pure TEM mode and fields sustained are known as Quasi TEM fields.

Methods of Analysis
With this quasi TEM mode approximation, the calculation of propagation parameters or the analysis of the line essentially reduces to the solution of 2D Laplaces equation, subject to the boundary conditions determined by the geometry of the line. Two important methods to solve 2D Laplaces equation are, Conformal Transformation Method: Exact method used to find propagation parameters of homogeneous stripline complicated if used for inhomogeneous medium. Variational method: Approximate method simpler even for inhomogeneous medium

Basis of Variational Method


Consider a system of perfect conductors S1, S2, ... , SN with Q1, Q2, ... , QN charges on the conductors and held at potentials V1, V2, ... , VN. Let the potential function in the space surrounding the conductors be . Like all fundamental forces of nature, satisfies and is the solution to Laplaces Equation, 2 = 0 Boundary conditions being = Vi on Si with i = 1, 2, ... , N The stored electrostatic energy is given by, We = (/2) . dV Here the integration is carried over the entire volume containing the electric field E obtained from the gradient of .

Basis of Variational Method


Suppose the charges are slightly displaced, but Vi is kept constant There will be a change in potential function surrounding the conductors given by + where is the incremental change in potential. Change in We is given by, We = (/2) [(+). (+) dV - . dV] We = (/2) [ . dV + . dV] Considering a surface S enclosing the entire volume and then applying the divergence theorem, We = (/2) . dV

Basis of Variational Method


From this we conclude that if we insert a trial function for the potential distribution, which differs by a small quantity from its correct value, the resulting value of We will differ from its true value by an amount proportional to ()2. The energy function We is thus a stationary function for the equilibrium conditions. We is also seen to be a +ve quantity. Hence, the true value of We is a minimum since any change from the equilibrium increases the energy function We

Variational Expression for Upper Bound on Capacitance


Consider a two conductor transmission line as shown. Conductor S2 is at +ve potential V0 and
conductor S1 is at zero potential.

We per unit length along the line is given by, We = (/2)|t |2dxdy = (1/2) CV02 Where C is the capacitance per unit length of the line & V0 is the potential difference between the two conductors.

V0 = -

t . dl

This is the variational expression for the capacitance of C.

Variational Expression for Upper Bound on Capacitance


The variational method involves the determination of capacitance by employing a trial function containing several variable parameters. Let these parameters be i with i = 1, 2, 3, ... , N. So that C = C(1 , 2 , 3 , ... , N) The condition that the integral is stationary leads then to the simultaneous equations =0 Solution of N homogeneous equations yields the values of i and hence the minimum value of C

Variational Expression for Lower Bound on Capacitance


Poissons Equation: Usually used in electrostatics to find the electric potential for a charge distribution which is described by its given density function. 2 = - f / is the divergence operator f is the free charge density is the permittivity Consider a two conductor TEM Transmission line, with its cross sectional view as shown. Conductor S2 is at +ve potential V0 and conductor S1 is at zero potential. Thus in our case, the Poissons equation becomes 2 = - (x0, y0) / where (x0, y0) is the unknown charge distribution on S2

Variational Expression for Lower Bound on Capacitance


Aim: To find the unknown charge distribution on S2 using Greens function technique. Greens function: It is a type of function used to solve inhomogeneous differential equations which have initial or boundary conditions. To find the Greens function, we consider potential due to point charge at (x0,y0) Greens function should follow Poissons distribution Hence we have, 2 = - (x-x0)(y-y0) / (x-x0) and (y-y0) are Diracs Delta functions The solution of this equation is the charge distribution on S2 (x,y) = G( x,y | x0,y0 ) (x0,y0) dl0

Variational Expression for Lower Bound on Capacitance


On the surface of S2 the potential is nothing but V0 V0 = G( x,y | x0,y0 ) (x0,y0) dl0 This equation is used to obtain a variational expression for the Capacitance in which becomes a trial function. Multiply both sides by (x,y) and integrate over S2 V0 (x,y) dl = G( x,y | x0,y0 ) (x,y) (x0,y0) dldl0 As can be seen , the LHS gives total charge per unit length of conductor S2 and is equal to CV0 where C is the capacitance per unit length. (x,y) dl = Q = CV0 (1/C) = (1/Q2) G( x,y | x0,y0 ) (x,y) (x0,y0) dldl0 (1/C) = (1/Q2) (x,y) (x,y) dl This becomes the variational expression for lower bound on C

Variational Expression for Lower Bound on Capacitance


It is a stationary expression for C for arbitrary first order changes in the functional form of . Further when is changed by a small value there is a second order change in C. Squared change is always +ve. Thus , for any trial function, (x0,y0), the calculated value of (1/C) is always larger than the true value, and thus we know that the value obtained is the lower bound on the capacitance.

Thank You

You might also like