The paper investigates the effect of several design parameters of a dynamometer water absorber (brake). This type of absorber is widely used in dynamometer testing of various applications due to its relatively inexpensive and power-dense design. Such applications include the bench testing of motorcycle, automotive, heavy equipment and even marine engines. The drawback of water absorbers is their complicated control characteristics making them undesirable for use in precision tests. The aim of this study is to evaluate the change in efficiency and linearity of torque absorption with respect to design parameters. Steady state computational fluid dynamics analysis at constant inlet flow rate was carried out in SolidWorks Flow Simulation environment. Twenty-one different design parameters were tested at three different operation speeds and compared to the baseline test. The output absorption torque values were graphed and analyzed, coefficient of linearity and low speed absorption torque was compared. With the primary focus on low speed torque absorption efficiency, it was found that certain values of the tested parameters, such as rotor and stator blade angles, fillet and rotor-to-stator gap, have a substantial effect on both the linearity and efficiency of the absorber, but reversed values of the same parameters were found to degrade the efficiency in most cases. Without considering undesirable design parameters, the highest low speed torque absorption efficiency gain was achieved by filleting the rotor blades or slanting the stator blades -15 degrees around the axis of rotation.