The U.S. Bureau of Mines is completing development of a diagnostic maintenance system as part of its investigation into using expert system techniques to diagnose and predict hydraulic problems on a continuous mining machine. Machine breakdowns due to hydraulic system failures are well-known contributors to sometimes prolonged maintenance delays, resulting in lost production time and increased operating expenses. The Bureau's effort to apply sensor-based expert system techniques to mining machine diagnostics will result in the availability of an effective new type of maintenance tool. This tool will help reduce the frequency of equipment failures and repair times, in turn incresing productivity and decreasing costs. The Bureau has developed an expert knowledge base to diagnose hydraulic problems on a Joy 16cm continuous mining machine. This diagnostic system is interfaced to machine-based sensors that monitor various hydraulic systems parameters, such as pressures, flows, temperatures, fluid level, and ferrous debris present in the oil. The status of these parameters is updated periodically and transmitted via a distributed interface to the diagnostic knowledge base. All diagnostic decisions are made based upon the available sensor information. This paper describes this sensor-based diagnostic maintenance tool and its components. The testing and evaluation plans for this system on the 16cm will also be outlined.