The Case of the ATV A-arm Weld Fracture
On a sunny, spring afternoon, during a casual ATV ride on a smooth road, one ATV A-arm failed unexpectedly, overturned the ATV and killed the adult driver.
The ATV was new with 23 hours operation. There was no evidence of unusual conditions or sudden, impact loading imposed on the failed A-arm.
The fractured region was a single-pass GMAW weld. The welded alloys were mild carbon steel. The weld regions of a typical, exemplar A-arm are displayed in the photos.
I assessed the fracture surfaces of a subject A-arm weld that joined the tube to the forged, ball-joint housing. I examined the tube fit-up to the housing and the failed weld regions visually, nondestructively and with Scanning Electron Microscopy (SEM). The broken weld region exhibited initiation sites, discontinuities, uneven terminations, stress risers and discontinuous weld terminations. Fatigued surfaces were located in the failed weld along with ductile fracture dimples. I reviewed sworn testimonies, welding codes, specifications and readily evident weld region defects. I wrote an expert report.
I determined that the failed welds were not as specified on the engineered drawings for the A-arm. The misaligned tube fit-up with the housing caused incomplete fusion of welds that quickly fractured during operation. The failed welds contained visible defects. Non-fused weld regions were initiation sites for progressive, low-cycle fatigue and cracking. The residue in gaps between the tube and housing contributed to unsound welds and were stress concentrators for the failure. At the factory, visual weld inspection of these welds did not detect insufficient fusion and defects.
The Lesson Learned:
Expert validation and reliable engineering of welds on vehicular support components was non-redundant; the ranges for proven and consistent welding procedures varied greatly; and visual quality assurance inspections were insufficient to confirm 100% fit-for-service welds.