The detection of partial discharge (PD) is an effective way to evaluate the insulation of a power system for the early discovery of failures. It ensures the safety, reliability, and economic operation of equipment. In this study, an envelope sensor was designed for PD signals. It consisted of a high-pass filter circuit, a signal amplifier circuit, and an envelope detection circuit. PD oscillatory signals were converted into low-frequency impulse envelope signals to effectively identify the peak voltage and the phase of occurrence as important data for the PD diagnosis and identification of failure patterns. To test the validity of the envelope sensor, corona discharge and surface discharge signals were generated using various PD experimental models. The coupling of a high-frequency current transformer was introduced to acquire PD signals. Data acquisition and signal analysis were performed using an oscilloscope after the signals were processed by the detection system. The experiment conducted revealed that the proposed envelope sensor was effective for measuring the impulse envelope of PD oscillations. The peak voltage and the phase of occurrence were identified, and the output impulse envelope signals were of low frequency, which contributed to the subsequent signal acquisition and analysis.