Automated Noise Calibration System (VT-1000)

Abstract

This paper details an automated Noise Source calibration system in development at Jet Propulsion Laboratory, California Institute of Technology (JPL). The paper begins with a discussion on noise figure and excess-noise-ratio (ENR) theory, fundamentals and governing equations. As part of the fundamentals there is a discussion of the system’s use of the Y-factor method to obtain accurate measurements of the unit under test (UUT), and how these measurements are compared against a known ENR standard to obtain the UUT’s ENR values. There is also an in-depth discussion on uncertainty quantification for noise source system calibrations. The architecture of the automated calibration system is provided, which includes both the system’s hardware and software configuration. The software is written in Python 3, and provides the user detailed instruction on how to proceed, including step-by-step connection requirements. This system automates much of the measurement process, including real-time uncertainty quantification and report generation, as well as real-time feedback to the user to allow intervention. The system takes advantage of a database of results from previous measurements to compare calibration history of the ENR measurements. The automated system presented here operates over a frequency range from 10 MHz to 50 GHz, and has shown substantial time savings over traditional manual methods of performing this calibration