The requirements to the quality of engines and vehicles have increased constantly and make testing more and more sophisticated. At the same time, there is a strong demand to reduce the periods and costs of development.Increasing importance is attached to dynamic engine test stands, that are capable of simulating steady-state and transient operating conditions as well as on-road driving.Loading units for this type of test stand must fulfill certain properties. Of special interest are the control response at continious and discontinious command value presetting, the parameters ‘mass ratio’ and speed gradient’. A comparison of the three common systems of dynamic loading devices, i.e. d.c. machine, asynchronous machine and hydrostatic unit, indicates the following:The d.c. machine is highly superior as regards limit frequency while the hydrostatic unit excells in the parameters of mass ratio and speed gradient. The asynchronous machine is superior owing to a simple mechanical design and good price/performance ratio.The superiority of the d.c. machine in limit frequency shall not be overrated. Tests on drivelines of common passenger cars have shown that it is only possible to simulate the first natural frequency with a range of about 3 to 9 Hz. This range is, however, reached by asynchronous machines and hydrostatic units without problem.