PSI - Issue 13

Markus Berchtold et al. / Procedia Structural Integrity 13 (2018) 676–679 Markus Berchtold/ Structural Integrity Procedia 00 (2018) 000 – 000

677

2

basic research activities took place in the recent decades. Ultrasonic fatigue testing systems work on the resonant frequency at about 20kHz, and require relatively small specimens with a specific geometry. Ultrasonic fatigue studies showed that a fatigue limit in the traditional sense does not exist in the Gigacycle regime. Cracks may occur subsurface or on the surface, and may start for example from inclusions in the material [1]. Subsequent with higher testing frequency, an old question of fatigue testing is high-lighted and cannot be neglected: “What is the effect of the testing frequency on fa - tigue life?” Testing on very high testing frequency may lead to different damage mechanism than under real loading condition for example of an engine component. Since inclusions and imperfection play an important role in VHCF the manufacturing process has a significant effect on fatigue life in the Gigacycle regime. Particularly for relatively inhomogeneous materials the testing of material volumes that represents the scatter of the manufacturing process is a concern. In 2014 RUMUL could present a new resonant fatigue testing machine, with a testing frequency of 1000Hz. The dynamic load of maximum 50kN peak-peak is produced with an electromagnetic system, similar to established resonant fatigue testing systems which typically run on testing frequencies from about 40 up to 250Hz. The static portion of the load is provided by two mechanical spindles, the maximum load of the system is +/- 50kN. Any load ratio can be selected. Flat and round specimen types that are normally used in fatigue testing can be used. The new testing machine offers new possibilities for investigations of material properties in the very high cycle fatigue (VHCF) regime. Compared to other systems used in the field of VHCF testing the RUMUL GIGAFORTE provides several advantages. The size of the machine is smaller and energy consumption less compared to a servo hydraulic system. The actually tested material volume is larger than the material volume that is tested on ultrasonic systems. The testing frequency of 1000Hz allows normally continuous testing, without intermittently stopping the test for let the specimen cool down. In the past four years the new testing machine was intensively used for example at the laboratory of the Fraunhofer-Institut für Werkstoff- und Strahltechnik IWS in Dresden in Germany. It is used for testing material samples and small components as well. Some effects of the 1000Hz testing frequency on the fatigue behaviour of the material were observed [2]. Recently the IWS laboratory developed a small salt spray chamber and mounted it on the GIGAFORTE to preform fatigue testing under corrosive atmosphere.. What is the effect of the frequency of an alternating load on fatigue life and fatigue testing? This question is probably present since beginning of fatigue testing. And it is clear there are frequency effects. For lower frequencies the effects can be neglected very often, however the sometimes unknown magnitude of some effects led to a quite conservative approach of limiting the testing frequency in some areas of fatigue testing. The frequency effects can be divided in three areas: Temperature and environment as extrinsic factors and strain rate as intrinsic factor [2, 3]. The effects may superimpose, and affecting fatigue life in the same or opposite direction. 2. Effects of the loading frequency on fatigue life A higher material temperature lowers usually the fatigue life as the ultimate strength of a material is related to the temperature. Some materials show a temperature de-pending crystallographic transformation that affects the material properties and fatigue life. Maintaining the specified temperature range is therefore a basic requirement for fatigue testing. A material specific basic damping is always present when deforming a solid material. Microscopic plastic deformation during cyclic loading leads to additional damping and it is almost completely transferred to heat. The damping energy and correspond-ing heat that is produced per load cycle and volume is constant for an even axially loaded specimen. The produced heat per time is proportional to the frequency. The resulting material temperature depends as well on the present heat loss, for example the heat flow to the fixture and to the ambient atmosphere. Convectional cooling can be used to control the temperature during testing. Some material do not show the above described linear relation between temperature and testing frequency, with higher frequency the temperature does not increase as expected [4]. This finding may point to hardening (resp. softening) mechanism that belongs to the category “strain rate” in this context. 2.1. Temperature