This study develops a subsurface detection system based on a photoacoustic effect. The main parts of the system are a laser diode as an exciter, a standard microphone as an acoustic detector, and a software-based phase sensitive detection (PSD) method. Artificial defects of various depths were made inside an alumunium plate as the object which was painted with acrylic paint to cover the defects. The difference in the thermal diffusivity of alumunium and acrylic paint will cause different amplitudes of photoacoustic so they can be used to detect the subsurface defects. Photoacoustic signal amplitudes used in this study as measured by a microphone have an average amplitude of −57 dB with an SNR of 11.2. High noise level was found below 2.153 Hz, so that the photoacoustic signal amplitude cannot be detected. The accuracy of PSD method was tested by measuring amplitude of 400-600 Hz sinusoidal voltages from a signal generator. The range of the amplitudes was 0.73-6.27 µV with a resulting maximum error of 9%. The PSD method was succesfully implemented to detect the subsurface defects of a 300 µm thick alumunium plate at depths of 215 µm, 241 µm, and 254 µm with a resulting maximum error of 5.83%.