During the past decades, Very High Cycle Fatigue (VHCF) research has developed into an active and prioritized research area. An increased interest in testing up to 107-1010 load cycles, realized within a reasonably short amount of time, has been enabled by the development of 20-30 kHz ultrasonic fatigue testing equipment. Here, a study is presented on fatigue crack propagation at 20 kHz of three different automotive steels tested at R=-1 and R=-0.24 load ratios. However, the high load rates provokes new challenges, as theoretically finding the best practice method to compute the stress intensity factor considering 20 kHz dynamic effects (inertia forces and damping). Calibrating the fatigue load system, monitoring, controlling and performing precise measurements of the growing crack during the tests are some examples of experimental challenges encountered. Here, a best practice method for computing the stress intensity factor is presented together with a complete 20 kHz fatigue crack growth testing procedure.Three different bar steel grades have been tested; a ferritic-pearlitic, a quenched and tempered martensitic and a carburizing steel grade. Crack propagation test results differentiated between the three steel grades and were depending on load ratio R. The obtained test results at 20 kHz were found to agree with results of the same steel grades tested at conventionally used load frequencies. The SEM fractography analysis revealed ductile transgranular crack propagation mechanisms, also this in agreement with the same steel grades tested at lower frequencies.

The 20 kHz load frequency enables fatigue tests for very high cycle fatigue life, 109-1013 cycles, within conveniently short time. In automotive applications, many components are subjected to flexural loading and hence bending fatigue is an important test mode. Ultrasound fatigue test instruments have been used successfully in several assessments of fatigue strength and more commonly in uniaxial loading. Here, a 3-point bending fatigue test rig operating in resonance at 20 kHz load frequency has been designed to test plane specimens at R=0.1 loading. The test rig design and stress calculations are presented. Testing for fatigue strength was conducted using the staircase method with 15 specimens of each steel grade, specimens reaching 108 cycles were considered run-outs giving fatigue strength at 108 cycles. Additional 15 specimens of each grade were tested for S-N curves with the upper limit above 109 cycles. Two different common automotive steels, 38MnSiV5, a micro-alloyed ferritic-pearlitic steel, and 16MnCr5, a carburizing martensitic steel, were tested. The fatigue strengths achieved from the staircase testing are 340 and 419 MPa stress amplitudes for the 38MnSiV5 and 16MnCr5 steels, respectively. The S-N curves of the steels appear to be quite flat in the tested life range 107 - 109.

Fatigue properties are evaluated in a large span of fatigue lives ranging from a few load cycles to more than 1013 load cycles. If the interest is focused on fatigue lives above 10(7) load cycles, we speak of the very high cycle fatigue (VHCF) range. For evaluation of properties in the VHCF range one often needs to use higher load frequencies to be able to perform testing within a reasonable time. Therefore, the influence of load frequency on fatigue strength and fatigue crack growth is an important issue, both from testing and design perspectives. Within an EU-RFCS research project on the frequency influence on high strength steel fatigue properties the present study has been conducted on fatigue crack growth testing to determine threshold values and crack growth material parameters. The testing was analyzed by FE-computation to determine geometry factors for AK-determination. The testing was performed in a 20 kHz ultrasound resonance instrument. In such a system the whole load train needs to be designed to run at a resonance frequency of 20 kHz, and it implies that the specimen needs to be designed and computations performed by dynamic computational methods. As the crack grows the dynamic response of the specimen will change, and hence calculation to obtain the geometry factor is made with a progressing crack length. A uniaxial tensile load at 20 kHz frequency is applied to a single edged notched side-grooved flat specimen. The specimen dimensions are calculated in order to have a resonance frequency of 20 kHz, which is the frequency used for the experiments. Dynamic FEM computation, with a 3D-model and a quarter symmetry was used with one of the symmetry planes parallel to and in the crack growth line. To avoid crack surface interpenetration during the simulations a rigid thin sheet was introduced and used as a counter-face to the crack surface. The solution obtained was then combined with the breathing crack model proposed by Chati et. al. (1997) in order to solve for the irregularities observed when crack surface interpenetration occurs. Finally, the whole load train was considered. Thus, also the computed frequencies were very close to frequencies observed in experiments. The computation of stress intensities was made for varying crack lengths in a series of simulations. The geometry factor relation was determined and used in 20 kHz crack growth testing to control the actual stress intensity at the advancing crack tip. Comparison of computations and experimental results were made.

The fatigue strength of two-duplex stainless steel grades, 2304 SRG and LDX 2101, with austenitic–ferritic microstructure is tested using ultrasonic fatigue testing equipment operating at 20 kHz. The testing is conducted in tension-compression mode with the load ratio R=-1. The fatigue strength is evaluated at 10⁷, 10⁸, and 10⁹ load cycles and the estimates of fatigue strength are higher for the LDX 2101 grade. The fatigue crack initiation mechanisms are analyzed using a scanning electron microscope. The fatigue cracks, in all cases, appear to initiate due to accumulation of plastic fatigue damage at the surface. In the 2304 SRG grade, accumulation of fatigue damage occurs at the external surface of fatigued specimens in the form of extrusions at the grain/phase boundaries and in the form of individual slip lines in the austenite phase. Meanwhile, in the LDX 2101 grade accumulation of plastic fatigue damage in the form of extrusions and intrusions occurs mainly within the ferrite grain. When the crack is microstructurally short, the crack growth appears to be crystallographic in nature and the crack appears to change its direction propagating from one grain into another.

The fatigue crack initiation mechanisms prevalent in high strength martensitic steel grades, hot rolled plate duplex stainless steels, cold rolled strip duplex stainless steel and a super alloy grade were compared. The fatigue testing of all the grades was conducted in the VHCF regime using an ultrasonic fatigue testing equipment operating at 20 kHz. Scanning electron microscope (SEM) observations of the fracture surfaces revealed the presence of a microstructure controlled initial growth of short fatigue cracks in all the tested grades. Fracture surfaces of the failed specimens of a high strength martensitic steel grade revealed the typical fine granular area (FGA) within the fish-eye area around the internal inclusions. Fatigue crack initiation in the cold rolled strip duplex stainless steel grade occurred at surface defects left over by the cold rolling process of this grade. However, the presence of FGA around the surface crack initiating defect was observed similar to the internal crack initiations in the high strength martensitic steels. By mapping the FGA size development during VHCF loading, as obtained from fracture surfaces, FGA growth results were obtained. A similar study on hot rolled plate duplex stainless steel grades, 2304 SRG and LDX 2101, revealed the presence of an initial crystallographic growth region (CGR) in which crack growth direction is changed by microstructural barriers such as phase and grain boundaries. The early plastic fatigue damage accumulation occurred predominantly in one phase or at the austenite-ferrite phase boundaries. On the other hand, an initial transcrystalline fatigue crack growth was observed in the Ni-based super alloy grade Inconel 718.

Two ductile iron grades, EN-GJS-600-3 a ferritic-pearlitic grade, and EN-GJS-600-10 a silicon strengthened ferritic nodular iron grade, are studied in the very high cycle fatigue range using a 20kHz ultrasonic test equipment. Fatigue strengths and SN-curves are achieved, and fracture surfaces and microstructures are investigated. The ferritic grade with higher ductility displays a lower fatigue strength at 10(8) load cycles than the ferritic-pearlitic grade, 142 and 167MPa, respectively. Examination of fracture surfaces shows that fatigue failures are controlled by micropores in both of the ductile iron grades, while the graphite nodule distributions do not seem to influence the difference in fatigue strengths. Prediction of the fatigue strengths, using a model for ductile iron proposed by Endo and Yanase, indicates a large potential for improvement in particular for the ferritic grade.

Fatigue studies of cold-rolled duplex stainless strip steel were performed in the very high cycle fatigue life region. The duplex austenitic-ferritic microstructure gives this grade a combination of high mechanical strength and high corrosion resistance. Fatigue properties of thin steel strips are particular due to cold rolling introducing a very fine microstructure. Crack initiation and fatigue strength are controlled by steel microstructure and alloying. The initiation and growth of the very short initial fatigue crack in very high cycle fatigue are unclear and subject to different descriptions. Fatigue test data of thin strip specimens at very high fatigue lives are scarce due to testing difficulties. For practical reasons testing must be performed at ultrasound test frequencies which involves fixturing problems. A test setup including the load chain ultrasonic horn, fixture and specimen was designed for resonance with a horse-shoe design of a screw fixture. The design of the horse-shoe fixture and the specimens along with FEM calculation of eigenfrequency are presented. Fatigue testing was performed at 20 kHz in R=-1 conditions up to fatigue life of 10⁷ to 5*10⁹ cycles. Fatigue strength was tested and crack initiation was studied on the fracture surface using FEG-SEM at the initiation site.