## Can I change the frequency of the A15TG?

FAQ #98 Updated March 19, 2015

## Question:

Can I change the frequency that my A15TG operates on?

It is not feasible to modify the frequency of the A15TG. The nature of circuit design in oscillators makes this process too complex.

If you are looking for a 1 kHz tone generator, try SESCOM

Oscillators can be thought of as unstable feedback systems. An unstable system is one in which an initially small excitation or disturbance produces an output that grows exponentially in time due to constructive, or positive, feedback. Typically, a relatively large feedback is necessary to get oscillation started in the first place - but once oscillation starts, the feedback must decrease so that the amplitude of the oscillation is limited at some finite value. The original large feedback is supplied by the battery. The feedback loop occurs at the emitter of the second BJT. BJT’s were invented to be used as amplifiers. If these amplifiers are put in an unstable feedback system, they oscillate. In order for this oscillation to occur, the loop gain of the system must equal one. The condition for oscillation is intuitively satisfying since when the loop gain is 1, an excitation presented to the input of the circuit will appear back at the input after some delay (after traversing the feedback loop) with identical amplitude. This “re-circulation” of the disturbance would proceed indefinitely and the circuit “oscillates” in a steady state - the fundamental period of the oscillation would be equal to the delay around the loop. This is what sets the frequency.

In most oscillators the loop gain is usually only equal to 1 at one particular frequency. At other frequencies, the amplitude may be less than or greater than 1, but the phase angle would be non-zero. This means that only one frequency component can travel around the loop with no phase shift. Only this frequency component will be amplified and grow to become a steady state oscillation. An initial disturbance is necessary in order for the oscillations to start - usually, the thermal noise voltage that is always present in electrical circuits is sufficient to start the oscillations. The feedback path is dependent on three lumped impedances Z1, Z2, Z3. Z1 contains C1, R1, C2, R2, and C3. Z2 contains Q2, R5, R4, C4, and the transformer. Z3 is R3. Even if we determined the correct value of these individual parts, the odds of the loop gain being 1 and the phase angle being 0 is about one in 10,000. So, in short, it can't be done easily.

Communication Circuits: Analysis and Design, Kenneth K. Clarke and Donald T. Hess, Addison-Wesley, 1978.
Modern Communications Circuits, Jack Smith, McGraw Hill, 1986.
Solid State Radio Engineering, H.L. Krauss, C.W. Bostian, F.H. Raab, John Wiley & Sons, New York, 1980.
Design of Crystal and other Harmonic Oscillators, Benjamin Parzen, John Wiley & Sons, New York, 1983.
Crystal Oscillator Circuits, Robert J. Matthys, John Wiley & Sons, New York, 1984.
Crystal Oscillator Design and Temperature Compensation, Marvin E. Frerking, Van Nostrand Reinhold, New York, 1978.