The main objective of this study was to clarify the influence of an increase in production capacity of a pressure groundwood plant on the quality of pressure groundwood. The studied factors influencing production capacity were the lowering of shower water temperature in a grinder and the start-up of a new reject refiner. The drop of shower water temperature from 95 ºC to 70 ºC was carried out as a two week trial run. The main purpose was to examine the change in the PGW process and in the quality of pressure grounwood and paper. In the PGW70 trial a possibility to raise production rate of grinders was attained without burdening screening room and reject refining too much. The main character affecting the quality of pressure grounwood was the decrease in fiber length. It was observed, that by lowering shower water temperature it is possible to achieve better optical and surface properties while loosing some of the strength properties. In the start-up of the new reject refiner the purpose was to consider the possible change in quality of pressure groundwood. Furthermore, the influence of the reject refiner on the quality of refined pressure grounwood reject was studied with different levels of refiner's specific energy consumption (SEC trial run). Also the possibility to grind the pulp to higher freeness value and to defibrate it more in reject refining was considered. No change in the quality of pressure grounwood was observed in the start-up of the reject refiner. In the SEC trial run two trial series were ran by raising the specific energy consumption of the reject refiner. The first trial series was ran with lower pit pulp freeness and the other with higher. Unintentionally the trial series differed from each other also because of the consistency in refining. The best defibration level was achieved with lower pit pulp freeness (130-140 ml), lower refining consistency (less than 40 %) and with highest SEC achieved (0,80 MWh/t). Refining with high pit pulp freeness and high consistency proved to be too hard for the fiber length. A much higher SEC should be directed on refining coarse reject than was possible in this trial. In other words, the reject refiner couldn't be loaded enough. That should be improved for example by finding better refiner plates.