Experimental study on electrical signatures of common faults for packaged DX rooftop units

Fault detection and diagnosis (FDD) is an effective way to automatically detect the existing faults of a rooftop unit (RTU) and increase the reliability and availability of system. Most existing FDD methods use thermodynamic signals as fault indicators. Recently, however, major concerns have arisen in regards to whether refrigerant-related temperature sensors and intrusive pressure measurements are reliable. Electrical signal measurements provide plenty of information about a system but require the use of an electrical meter. One way to solve this issue is to install variable frequency drives (VFDs) on the HVAC system, since VFDs not only optimize the system performance, but also provide electrical parameter readings. In this study, a series of lab experiments were conducted to investigate the relationship between the electrical signals and common faults. Since environmental influences have been correlated with the system performance, the impact of several driving conditions on the electrical signals was also investigated. This study identified an electrical signature for each fault and individually separated the common faults based on three parameters: the fan power, compressor power, and the supply air temperature. The results of the study are of use to the development and implementation of a new FDD method based on combining the electrical and temperature parameter measurements.