Page 236 - Analog and Digital Filter Design
P. 236

Impedance Matching Networks  233




                       sidering the conversion from a series RL circuit into its parallel equivalent, and
                       vice versa. Also the conversion from series RC to parallel RC, and vice versa.
                       This transformation could provide a useful simplification to the mathematics of
                       an impedance matching circuit where, perhaps, the equivalent series reactance
                       of  a parallel RC load needs to be found.

                       First find the circuit Q. For series circuits this is the reactance divided by  the
                       resistance. For parallel circuits this is the resistance divided by  the reactance.
                       The  Q is equal to tan(@, where  8 is the phase angle of  the impedance. The
                       following equations summarize these statements.






                       The relationship between series and parallel resistance is given by the equation:


                                             ZS
                             Rp = R,(l+Q')  = -
                                            COS0'

                       The relationship between series and parallel reactance is given by the equation:






                       The equivalent parallel or series model is only valid at one particular frequency.
                       This is simply because the reactive element changes with frequency and, hence,
                       so does the circuit Q. However, impedance matching circuits are also only valid
                       for one particular frequency; therefore this is not an issue.



                       Matching Using L,  T,  and PI Networks
                       Networks that comprise two or three reactive components can be constructed
                       to provide narrowband matching. The networks are described by the shape of
                       the components, as drawn on a circuit diagram. Thus L networks use two reac-
                       tive components; the L represents a shunt branch followed by  a series branch,
                       or a series branch followed by a shunt branch (a little imagination may be needed
                       here, the L is upside-down!). Both T and PI (IT) networks require three com-
                       ponents and represent either series, shunt, series branches; or shunt, series, shunt
                       branches. With T and PI networks the physical component layout corresponds
                       closely to the symbol. The four configurations are given in Figure 8.8. Each con-
                       figuration can be highpass or lowpass, depending on whether the series elements
                       are capacitors or inductors (and hence whether the shunt elements are induc-
                       tors or capacitors).
   231   232   233   234   235   236   237   238   239   240   241