Modern methods of monitoring the performance quality of combine harvesters during their testing are applied in a «swath laying» mode. The authors have shown the relevance of a similar assessment method used in the most common “crushing-and-spreading” mode and developed a simple and reliable method and a technical tool for assessing grain losses in the field during the combine harvester operation. (Research purpose) To develop a method for controlling grain loss of a combine harvester operating in a “chopping-and-spreading” mode using a two-chamber sampler prototype. (Materials and methods) The authors have proposed a new design of the prototype of a two-chamber sampler. They have proved that it ensures separate collection of the lost grain resulting from combine harvester units. A limited number of samplers were installed in the field and a method for sampling grain losses from working units of a combine harvester based on the developed sampler was applied. The applied method of control consists in evaluating the test results of a two-chamber sampler prototype during a combine harvester test in a «chopping-and-spreading» mode. (Results and discussion) The authors have developed the prototype of a two-chamber sampler of a fundamentally new design. They offer a method for determining the performance quality of a combine harvester. They have obtained graphs of grain losses distribution across the width of a threshing-and-separating unit and the spreading width equal to the operating width of a header. (Conclusion) It has been established that the use of a two-chamber sampler will ensure separate sequential sampling behind a header, a threshing unit, and a chopping spreader, while excluding sequential overlapping of errors. It has been revealed that the proposed scheme for sampling a grain losses resulting from the working units of a combine harvester will reduce the total number of samplers used - no more than 7 in each replication, regardless of the width of a threshing-and-separating unit or a header.