The purpose of this master thesis is to improve the current method of conducting fatigue bending tests of crankshafts at CTM. Currently they are using a servo hydraulic fatigue machine, however, they also have access to a resonance fatigue machine. The resonance fatigue machine can work at much higher frequencies and obtain result more rapidly at a lower cost. Thus, the main objective of this project has been to design and manufacture a tool for bending fatigue testing of crankshafts for CTM’s resonance fatigue machine. Secondary goal has been to design a method for preprocessing the crankshafts without affecting the results from the fatigue tests. All of this has been done according to the Volvo’s standard method for bending fatigue strength testing for crankshafts. At the start of the project a meticulous literature study including state of the art has been conducted for bending fatigue testing of crankshafts. From this, several concepts were generated, followed by both structural and modal analysis in order to ensure the function of the tool. From tests conducted in the resonance fatigue machine it could be concluded that the tool reached a resonance around 50 Hz. However, the maximum stress introduced in the sample was not enough to break the crankshaft. When increasing the load/stroke length in order to increase the stress, the power usage excided the capacity of the machine.When it was discovered that the desired result could not be achieved by the resonance fatigue machine, the tool was modified to fit the servo hydraulic fatigue machine. In order to see if the tool is able to work at a higher frequency than the current tool. It was possible to break the samples with a 50 % higher frequency and with a load ratio of -1, which was not possible with the previous tool. This represents a 33% cost reduction per test and the tests coincide better with the conditions of a crankshaft in an engine.