Axle fracture of an ICE3 high speed train
Highlights
► We analysed the causes for the axle fracture of the German high speed train ICE3 in 2008. ► Due to severe damage of the fracture surfaces NDT had to be used to find crack initiation. ► Fatigue crack initiated subsurface at non-metallic inclusions (NMIs) in high strength steel. ► New combination of non- and destructive methods for the detection of NMI is presented. ► We provide recommendations on improvement of railway axle safety.
Introduction
The German high speed train ICE3 is approved for a top speed of 330 km/h. First public service took place in late 2000. Deutsche Bahn (DB) operates a fleet of 63 ICE3 trains. On 9 July 2008 one wheelset of the powered bogie of a trailer vehicle, Fig. 1, derailed just before crossing the Rhein bridge at Cologne Central Station, Fig. 2. Since the train travelled at very low speed at this moment, fortunately there were no injury to persons and only minor damage to the rail track. The train previously came from the high speed line and changed travelling direction in Cologne Central Station as scheduled.
The cause for the derailment of the wheelset was the previous fatigue fracture of the hollow axle, Fig. 2, at the T-notch, Fig. 3.
Because of public interest, the public attorney’s office solicited BAM’s analysis of the root cause. The task for BAM was to look for deficiencies of the material, manufacturing and service of the wheelset, defects of the vehicle and compliance with standards.
The failure case was investigated by the authors of the present paper and documented in detail [1] and in extracts [21].
The in-service ultrasonic testing (UT) inspection interval for this type of driving axle was 300,000 km. Last non-destructive testing (NDT) using mechanized UT of the hollow axle took place in March 2008, 150,000 km before fracture: according to the operator DB no crack-type indications were recorded at this time. During UT in-service inspection on crack type defects at the outer surface the recording level was set to approx. 2 mm saw cut depth (SCD) [21].
A few days before the axle fracture, the wheelset and its axle were inspected visually by the operator DB, without any indications recorded.
The final fracture occurred when the axle was in service ∼3 million kilometers which refers to ∼109 revolutions. Regarding the load cycles, this belongs to the very high cycle fatigue (VHCF) range.
DB operates a large number of temperature sensor systems positioned at the track for detection of elevated temperatures of wheelset bearings and brakes of passing trains. The systems installed at the high speed line to Cologne are able to measure the temperatures in trains travelling at 300 km/h for each axle separately. This way it was possible to detect and record that only the axle concerned (cp. Fig. 1) showed unusual high temperatures. This was recorded first about 50 km before Cologne.
Section snippets
Investigation of components and axle material
Mechanical fracture of a component, in principle, occurs when the applied stress is higher than the strength of the material either for static or cyclic loading.
The first task is to check if the loading of the failed axle was in the range of all other axles (∼1200 driving axles in the ICE3 fleet [36], [37]) that survived the same in service time without fracture (∼1/3 of all driving axles), Section 2.1.
If this could be affirmed, the second task would be to check whether the strength of the
Origin, effect and detection of non-metallic inclusions
In some previous failure cases, investigators were successful in detecting and documenting non-metallic inclusions as fatigue crack starter on fracture surfaces [2], [3] e.g. of a big crankshaft made from a similar high strength tempering steel [4].
In order to understand the role of non-metallic inclusions and to verify their contribution to the root cause of the ICE3 axle fracture, their origin and effect was investigated in more detail. Within a literature review the influence of non-metallic
Root causes for axle fracture and failure sequence
Coming back to the initial failure analysis of the high speed train axle fracture, the results of the survey of components, Section 2.1, material testing, Section 2.2, and the investigation of the crack initiation area, Sections 2.3 Fractography, 2.4 Crack initiation: micro computer tomography and metallography, now need to be combined for identifying the root cause(s) of axle fracture [1], Section 4.1. The chronology of crack initiation, growth and time of final fracture of the axle is
Immediate measures
Immediately after the fracture, all ICE3 trains were stopped by the German Federal Railway Authority (EBA). The UT inspection interval for all hollow driving axles made from this and similar high strength steels was reduced from 300,000 km to 60,000 km and further to 30,000 km in October 2008 [31].
EBA 2008 also decided to significantly reduce the inspection intervals for German trains with axles from similar high strength steels as the 67 ICE-T trains with tilting technology and the S-Bahn Berlin,
Acknowledgments
The authors cordially thank S. Bohraus for his enduring microfracto- and topographic work, J. Goebbels, D. Gohlke and T. Heckel for their intense search for inclusions using μCT and UT immersion technique, B. Bogel and O. Paulinus for their careful metallographic target preparations and purity assessment, R. Häcker and B. Abbasi for preparation and testing of specimens as well as standards research, R. Saliwan Neumann for the EDX analysis of inclusions and the other colleagues of BAM failure
References (38)
- et al.
Effects of small defects on fatigue strength of metals
Int J Fatigue
(1980) - et al.
Quantitative evaluation of effects of non-metallic inclusions on fatigue strength of high strength steels. II: fatigue limit evaluation based on statistics for extreme values of inclusion size
Int J Fatigue
(1989) - et al.
Quantitative evaluation of effects of non-metallic inclusions on fatigue strength of high strength steels. I: basic fatigue mechanism and evaluation of correlation between the fatigue fracture stress and the size and location of non-metallic inclusions
Int J Fatigue
(1989) - et al.
Effects of defects, inclusions and inhomogeneities on fatigue strength
Int J Fatigue
(1994) - et al.
Effect of inclusion on subsurface crack initiation an gigacycle fatigue strength
Int J Fatigue
(2002) - et al.
Estimation of maximum inclusion size and fatigue strength in high-strength ADF1 steel
Mater Sci Eng A
(2005) - et al.
The fatigue behaviours of zero-inclusion and commercial 42CrMo steels in the super-long fatigue life regime
Acta Mater
(2004) - et al.
On the critical inclusion size of high strength steels under ultra-high cycle fatigue
Mater Sci Eng A
(2006) - et al.
Probabilistic fatigue S–N curves including the super-long life regime of a railway axle steel
Int J Fatigue
(2009) - et al.
Influence of inclusion size on fatigue behaviour of high strength steels in the gigacycle fatigue regime
Int J Fatigue
(2007)