NTC oscillators permit what appears to be conflicting requirements (i.e., accuracy, stability, and cost) to be realized. In general, as the number of requirements increases, or the severity of the requirements increases, the cost will also increase -- sometimes being prohibitive. Conventional oscillator designs (colpitts-type configurations) may yield acceptable performance, accuracy, and stability, particularly when extensive temperature control or compensation is employed, but is costly. Also, by virtue of compensation techniques, settling time and hysteresis is compromised. Low-frequency phase noise is still present and this is the primary source of error for very short measurement periods. NTC's negative gain-one pole response oscillators, reflection oscillators, and modified positive gain-one pole response oscillators are all low cost and have fast initial accuracy, with a minimum short-term (1 sec) stability of one part in a billion (10 -9). Advanced circuitry permits these oscillators to rival rubidium oscillators. The final report for phase I will compare conventional oscillators and their detrimental factors to ntc oscillators, and demonstrate the superiority and feasilibity of ntc's approach to oscillator design.
