ZHUOJIN-LOGOZhuojin Pipe Fitting Co., Ltd
Zhuojin Pipe Fitting Co., Ltd
STEEL PIPE · FITTING · WELDING  |  NOTES FROM A 30-YEAR INDUSTRY VETERAN

Failure Analysis of a Fan Transmission Shaft After Surfacing Repair

Background and Failure Description

This paper documents a classic case that we see far too often in the field: a component that has been repaired by surfacing, returned to service, and then failed prematurely. The subject here is a fan transmission shaft, which had been in long-term service. The shaft journal and its sleeve had worn severely, and the journal surface was built up by surfacing welding before being put back into service. Not long after, the shaft fractured. From my own experience with rotating equipment, this is a textbook scenario where the repair itself introduces new problems that are not always obvious at the time of welding.

The investigators approached this systematically, which is exactly what I would expect from a competent failure analysis team. They performed macroscopic observation, chemical composition analysis, tensile testing, hardness testing, microstructural examination, and fractographic analysis. This is a good example of the full suite of investigative tools being brought to bear on a single failure, and it is worth noting that each technique contributed to the final conclusion.

Investigation Findings

The chemical composition of the shaft material was checked against the specification, and the tensile and hardness tests were performed on samples taken from the shaft. The results showed that the base metal properties were generally within acceptable ranges, which ruled out material degradation as the primary cause. The key findings came from the macroscopic and microscopic examination of the fracture surface and the surrounding microstructure.

Examination Method Key Finding Significance
Macroscopic observation Multi-source fatigue fracture with ratchet steps at the fracture edge Indicates multiple crack initiation sites under cyclic loading
Fractographic analysis Tire-like impressions (beach marks) visible on the fracture surface Confirms fatigue crack propagation over an extended period
Microstructural examination Abnormal microstructure in the surfacing layer at the shaft shoulder Suggests improper post-weld heat treatment or uncontrolled cooling
Hardness testing Hardness values in the repaired zone deviated from the base metal Indicates a hard, brittle layer that is prone to cracking
Chemical analysis Base metal composition within specification Rules out material substitution or severe alloy segregation

The most telling observation was the presence of tire-like impressions at the ratchet steps on the fracture edge. These are characteristic of a multi-source fatigue failure where several cracks initiate at different locations and then propagate simultaneously. The ratchet steps form where the individual crack fronts eventually merge, and the tire-like impressions are the marks left by the cyclic contact of the mating fracture surfaces. In a shaft that has been repaired by surfacing, the presence of these features is a strong indicator that the repair itself introduced stress concentrations that became the preferred sites for crack initiation.

Root Cause Analysis

The root cause analysis identified two distinct but related problems. First, the surfacing layer at the shaft shoulder showed abnormal microstructure and hardness. This is a well-known issue with surfacing repairs: the heat input from welding alters the local microstructure, and if the cooling rate is not controlled, hard martensitic or bainitic structures can form. These structures are brittle and have poor fatigue resistance. Second, the weld toe of the surfacing bead at the transition diameter was left unmachined. This is a critical detail. The weld toe is a natural stress raiser, and if it is left in the as-welded condition, the geometric discontinuity creates a severe local stress concentration.

Under cyclic loading, the keyway and the weld toe both acted as fatigue crack initiation sites. The keyway is a standard feature on transmission shafts, but it is also a known stress concentration. When the surfacing repair added a second stress raiser in the same region, the combined effect was sufficient to initiate multiple fatigue cracks. Once initiated, the cracks propagated rapidly under the working load until the shaft fractured.

From a practical standpoint, this case highlights several important lessons for anyone involved in surfacing repairs of rotating components. The first is that the weld toe must be machined flush after surfacing. Leaving it in the as-welded condition is an invitation for fatigue failure. The second is that the post-weld heat treatment must be carefully controlled to restore the microstructure and relieve residual stresses. The third is that the repair procedure must be designed with the same care as the original manufacturing process, including consideration of the stress state under service conditions.

Lessons for Repair Practice

This case has direct relevance to our work in the steel pipe and pipe fitting industry, particularly for components such as mandrel bars, piercing rolls, and other rotating equipment that are subject to surfacing repairs. The failure mechanism identified here is not unique to fan shafts. Any shaft or roll that is repaired by surfacing and then returned to service under cyclic loading is at risk if the repair is not performed correctly.

The key takeaways are straightforward. First, always machine the weld toe flush after surfacing. Second, ensure that the post-weld heat treatment is adequate to restore the microstructure and relieve residual stresses. Third, inspect the repaired zone using surface and volumetric NDT methods before returning the component to service. Fourth, consider the service stress state when designing the repair procedure. Fifth, document the repair parameters and results for future reference.

In my experience, the most common mistake is treating surfacing repair as a simple "build it back up and grind it" operation. The reality is that surfacing repair is a metallurgical process that requires careful control of heat input, interpass temperature, cooling rate, and post-weld treatment. When these factors are ignored, the result is a component that looks fine on the surface but is internally compromised. The failure in this case occurred not because the surfacing was poorly executed in terms of deposit quality, but because the stress concentration at the weld toe and the abnormal microstructure in the heat-affected zone created the conditions for fatigue crack initiation.

Conclusion

This failure analysis is a valuable reference for anyone involved in surfacing repair of rotating components. The multi-source fatigue failure mechanism, the tire-like impressions on the fracture surface, and the identification of the weld toe and keyway as crack initiation sites are all well-documented and clearly explained. The practical implications are equally clear: surfacing repairs must be treated as metallurgical processes, not just dimensional restoration operations. The weld toe must be machined, the microstructure must be restored, and the component must be inspected before being returned to service. If these steps are followed, surfacing repair can be a reliable and cost-effective solution for extending the life of worn components. If they are not, the result is a premature failure that costs far more than the original repair.