This study aimed to find the effect of variations in cooling load on the performance of the prototype trainer refrigerator, which in this refrigerator prototype had never been tested performance. In this study, the experimental method was used. The values sought in this study were the rate of room cooling and the Coefficient of Performance (COP). In this study, the variation of the cooling load used started from temperatures of 300C, 350C, 400C 450C, with 15 repetitions, and the temperature performance used was 1000C. The study results showed that the highest room cooling rate occurred at 300C cooling load variation, namely 0.00265, and the smallest room cooling rate occurred at 450C cooling load variation, namely 0.00174. The highest Coefficient of Performance (COP) results occurred at a cooling load variation of 450C, namely 34.09, and the lowest Coefficient of Performance (COP) occurred at a cooling load variation of 300C, namely 29.60. From these results, it can be concluded that the cooling rate of the room was inversely proportional to the increase in cooling load because the more significant the cooling load, the longer the time obtained, resulting in a small room cooling rate. And for the results of the Coefficient of Performance (COP), it was directly proportional to the increase in the cooling load because the more significant the cooling load given, the higher the temperature at T1 and T2, resulting in a significant Coefficient of Performance (COP).