Main Article Content
In this paper, the development of an instrument for temperature and relative humidity measurement using a low cost pre-calibrated sensor with calibration technique based on least square regression method is described. The instrument consists of 5 main parts i.e. a microcontroller unit, a sensing unit, a real time clock unit, a display unit and a data storage unit. The PIC C compiler is used to develop the operating software with the standard coefficients provided by manufacturer of the sensor. The discrete data of the temperature measurement using original coefficients are used to calculate the new coefficients needed for calibration. The results show that temperature measurement using modified coefficients based on least square method can reduce the true percent relative error more than 10 times. The new coefficients used for temperature conversion is involved to compensate for the measurement of relative humidity. By combining the modified coefficients of temperature conversion with the calibration method for relative humidity measurement, the true percent relative error is reduced more than one-third.