Notch Noise Loading Date on Pre-Detection Recording of FM Carriers

Abstract

In a previous paper (1) the authors reported notch noise loading tests of baseband (post-detection) tape recording showing dependency of the notch noise power ratio, NPR, on record level. The purpose of this paper is to report notch noise loading tests on pre-detection tape recording showing NPR dependence on record level, bias level, and output equalization. Both the baseband and pre-detection recording tests were made with the Hewlett-Packard Model 3950 recorder/reproducer. Generally, the pre-detection NPR were 5db or more improved over the baseband NPR. The extent of the improvement depends, in part, on the quality of the vco and discriminators utilized for the pre-detection recording as well as the carrier deviation. As the linearity of the vco and discriminator and the phase linearity of the circuits that pass the pre-detection signal are improved, the limiting factor in NPR is the noise added by the reproduce amplifier. Increasing the record level above IRIG standard produces an improvement in the output signal-to-noise ratio, NPRO, of 3 to 5 db using IRIG standard bias. Eliminating the bias and recording at the level which results in maximum output of the fundamental produced up to 8 db improvement in NPRO over IRIG. In the course of the tests it was discovered that if harmonics are present in the recorded pre-detection signal and/or are generated in the record/reproduce process, serious intermodulation can occur unless a low-pass filter ahead of the discriminator is used.