Audio Engineering

880 Chapter 30

Rx  Ry  2 Rz  R , the frequency of oscillation will be 1/2 πCR , as in the Wien bridge
oscillator.

If the parallel T network is connected in the negative feedback path of a high gain amplifi er
(A 1 ), oscillation will occur because there is an abrupt shift in the phase of the signal passing
through the ‘ T ’ network at frequencies close to the null, and this, and the inevitable phase
shift in the amplifi er ( A 1 ), converts the nominally negative feedback signal derived from the
output of the ‘ T ’ network into a positive feedback, oscillation-sustaining one.

A problem inherent in the parallel T design is that in order to alter the operating
frequency it is necessary to make simultaneous adjustments to either three separate
capacitors or three separate resistors. If fi xed capacitor values are used, then one of these
simultaneously variable resistors is required to have half the value of the other two.
Alternatively, if fi xed value resistors are used, then one of the three variable capacitors
must have a value that is, over its whole adjustment range, twice that of the other two.
This could be done by connecting two of the “ gangs ” in a four-gang capacitor in parallel,
although, for normally available values of capacitance for each gang, the resistance values
needed for the ‘ T ’ network will be in the megohm range. Also, it is necessary that the
drive shafts of Cx and Cy be isolated from that of Cz.





A 1

R 4

R 1

C 1

R 5

0V

C 2

(Cx)(Cy)

R 3

(Rz)

C 3

(Cz)

(Rx)(Ry)

R 2

Sinewave output

Figure 30.6 : Oscillator using parallel T.