Design of a Compact Broadband Vector Network Analyzer to Diagnose a Semiconductor Process Chamber

Abstract

This paper presents a compact broadband vector network analyzer (VNA) designed for the cost-effective diagnosis of semiconductor process chambers operating in the frequency range of 3-13.6 GHz. The proposed system, optimized for compactness, comprises various monolithic microwave-integrated circuits (MMICs) under the digital control of a microcontroller unit (MCU). By employing a gain/phase detector, the disparity between the reference signal (Pa) and the reflected signal (Pb) can be efficiently measured through analog signal paths, resulting in a short measurement time. A comprehensive link budget analysis of each signal path, encompassing the transmitter, local oscillator (LO) path, and receiver, informed the design and implementation of the proposed system. Utilizing the implemented VNA, the broadband S11 parameters of a designed test bed chamber were measured under both normal and abnormal conditions. Applying a similarity algorithm such as Euclidean distance, differences in S11 between normal and abnormal conditions were analyzed according to frequency, providing a diagnostic tool for assessing the condition of the process chamber.