Positive expiratory pressure (PEP) is a technique used in respiratory physiotherapy to treat diseases related to the respiratory system through spontaneous breathing. This equipment consists of an oronasal mask coupled to a T connector with a unidirectional valve. Studies that evaluate whether the pressure level in the one-way valve corresponds to the actual pressure level provided are scarce in the scientific literature. In order to investigate the failures, bench tests were carried out on the spring-loaded valves, using a U-tube manometer. This pressure was exerted on the valve using a syringe that generated air flow inside the U-tube, allowing analysis numerical value of the measured pressure and the specified values of the valve, thus verifying the disparity of these measured values in relation to the PEP values operated by valves (0 to 20 cmH2O) from the three manufacturers under study. PEPs generated by spring-loaded valves from all three manufacturers were higher than pressures in the range of 2.5 to 20 cmH2O, with significant differences between manufacturers. This bench study showed inaccurate operation of all spring-loaded PEP valves of the three manufactures. The results obtained and the performance of the valves require a reevaluation of manufacturing procedures to preserve product quality and efficacy in clinical application.