The fault diagnosis task in factories is technically demanding and time-consuming, especially for the green hands. In order to assist mechanicians to detect the fault parts, the fault diagnosis expert system is introduced to support decision-making and problem-solving processes. In this vision, how to enhance mechanicians’ willingness to use such a system becomes a key issue to the industry. The present study aims to find out the elements that affect the user experience of the fault diagnosis expert system. We carried out an experiment to investigate how system performance and empathetic design promotes user experience, and we also explored how those factors apply to different users with high and low expertise. Fourteen participants who majored in automatic control with a basic understanding of industrial automation devices took part in the study. During the experiment, the participants were instructed to repair a UV control cabinet with the help of the fault diagnosis expert system. The system has four levels of system performance (high, medium, low, and all-errors) * two levels of empathetic feature (standard design vs. empathetic design). After each task, the participants filled out a questionnaire to evaluate their experience. The results showed a beneficial effect of both system performance and empathetic design. Also, the interactive effect indicates that for the system with qualified performance, the empathetic design enhanced user experience; while for the system with all-error performance, the empathetic design no longer helped. In addition, the empathetic design played a better role among inexperienced users, while experienced users tended to be more sensitive to changes in system performance.