Industrial development in Indonesia has increased and is supported by technology in the manufacturing sector, one of which is in the automotive industry which has large-sized machines which have a negative impact, namely causing noise effects that occur in the production process environment. The purpose of this study is to design a noise detection device using a microcontroller to determine normal and abnormal threshold values in the press machine area. The research method uses a quantitative approach with an experimental type, namely by taking data directly on the noise intensity level in the press machine area, measurement time and percentage error, the average error at night was 3.24% and an accuracy rate of 96.76%, and the average error during the day was 3.41% and an accuracy rate of 96.59%, at a measurement distance of 25 cm. This value is closest to the reading from the GM1352 sound level meter, there is a difference in noise measurements at night which is lower than during the day, this is because there are more activities during the day, causing greater noise. this is by further developing this tool in terms of function and the components of the tool used for maximum results obtained. It is better to choose a sound sensor that has a higher level of accuracy and has a more complete data sheet, so that the sensor specifications can be clearer.

Keywords: Noise, Industry, press machine, sound sensor