On April 27, 2008, during a pre-flight walk around inspection at Singapore Changi Airport (SIN), Singapore, hydraulic fluid was discovered on the ramp, adjacent to FedEx MD-11 Center Landing gear (CLG) Strut. Further investigation revealed that the gear had a vertical crack along the cylinder that measured approximately 33 inches. The CLG strut was sent to Hawker Pacific, Burbank, CA for teardown, and then to Boeing Southern California Materials Lab in Huntington Beach, CA. The CLG had accumulated 5,058 cycles since reinstallation on Feb 22, 2001, and had a lifetime total of 9,379 cycles. Use your browsers 'back' function to return to synopsisReturn to Query Page
The top portion of the strut, approximately 18 inches from the top, was completely cut circumferentially using a band saw at the Boeing Southern MP&P Laboratory. This cut was made to facilitate removing the inner components of the strut. During the performance of the cutting along the length of the cylinder, loud “bangs” were experienced, and the abrasion cuts extended in length by crack propagation, indicating that the cylinder had a high amount of residual stress.
Visual inspection of the inner components showed the piston to be in relative good condition except for some damaged/spalled chrome plating at the Aft/Left outer surface. There was no evidence of scraping along the chrome plated area or the substrate by spalled chrome plating particles, nor was there evidence of embedded spalled chrome plating particles. Except for this damaged chrome region of the piston, no other evidence of chrome plating damage was found and no mechanical damage was found.
The following examination pertained only to the cylinder.
Visual and Macroscopic Examination
Examination of the CLG Cylinder confirmed that there was a through-thickness, longitudinal crack along the aft side of the cylinder. The crack extended from the inner corner of the left hand beam to the upper edge of the cylinder, approximately 33 inches long and approximately 18.2-inches from the bottom of the cylinder.
Fracture traces indicated that the fracture originated from the outer surface at the inner corner of the beam to cylinder interface [the lower portion of the cylinder], approximately 18.2 inches from the bottom of the cylinder. The paint on the outer surface adjacent to the origin area had flaked off, revealing that there were at least two layers of white paint. Macroscopic examination in the failure origin region, along the outer surface, revealed that the first paint coating [adherent to the substrate] exhibited evidence of sanding marks and surface contamination. Under the paint, on the substrate surface, no sanding marks or contaminations were discovered. A repair work on the first coat of paint was evident, such that sanding paper and paint removal solution [paint thinner] may have been used to remove/repair the coating of paint in this area.
The fracture origin and adjacent region, to a depth of approximately 0.122 inch, had a roughly, teardrop shaped region with a facetted, coarse-grain texture, typical of intergranular rupture. There was also a band of facetted, coarse-grain texture, typical of intergranular rupture that was approximately 0.023 inch thick. This band was approximately 0.017-inch from the inner diameter surfaces. The remaining fracture surface had a coarse grain texture, typical of ductile overload.
Scanning Electron Microscope (SEM) Analysis
The white paint and underlying green primer were chemically stripped. SEM examination of the outer surface adjacent to the origin did not reveal any anomalous conditions. Examination determined that the origin and adjacent region consisted primarily of shallow dimples on intergranular facets, indicative of a brittle mode of fracture. Because intergranular is the dominant failure mode, this origin is considered an intergranular mode of failure. The depth of this Intergranular mode is approximately 0.086-inches. As the crack depth increased, the fracture dimpled rupture content also increased, indicative of a mixed mode of fracture. The band of facetted, coarse-grain texture also consisted of a combination of shallow dimples on intergranular facets and islands of dimpled rupture. The remaining fracture surface consisted primarily of dimpled rupture, typical of ductile overload, although there were several intermitted facets with shallow dimples. It was noted that the fracture surface had spots of oxidation that were concentrated at the regions with shallow dimples on intergranular facets. Fewer spots of oxidation were noted in the dimpled rupture regions.
A metallographic specimen was excised normal to the fracture surface at the origin and prepared for microscopic examination. The overall microstructure appeared normal for 300M steel in the hardened and tempered condition. Several secondary cracks were observed normal to the fracture surface and appeared jagged, typical for intergranular cracks.
Microhardness survey was conducted to determine if any evidence of decarburization or carburization existed on the surface of the cylinder. The results showed no evidence of decarburization or carburization.
Two sections excised from the CLG cylinder, in close proximity to the failure origin area were cleaned to expose the base metal [300M] along the outer and inner surface. The area was etched per DPS 4.715 and visually inspected. Visual examination of the outer and inner surfaces revealed no evidence of grinding burns. The etched surface appeared uniform and showed no surface conditions such as imperfections, segregation, inclusions etc., that would render this cylinder as unacceptable.
The wall thickness, in close proximity of the failure origin area, was measured to be 0.183-inch thick which met the Engineering drawing requirements of 0.169 – 0.186 inches. The inspection results of the cylinder is described in Appendix A
A section of the CLG cylinder was excised and submitted for hardness testing. The test results values are HRC 54.0-55.0 which met the hardness requirements of HRC 53.0-55.0 per DPS 1.05-4 for 300M in the hardened and tempered condition.
The failure origin for this failure was determined to be at the lower portion of the cylinder approximately 18.2 inches from the bottom of the cylinder. The Center Landing Gear appeared to have a large amount of residual stresses. This is evident by the wide separation gap between the fracture surfaces as the crack extended from the bottom to the top, and the experiences of cracking occurring during cutting with a hand held abrasive wheel. Cracking would develop at the tip of the saw cut and propagate several inches downward.
The failure of the Center Landing Gear Cylinder occurred by intergranular [IG] cracking at the lower portion of the cylinder. The IG crack extended some distance into the wall of the cylinder followed by an overload mode of cracking. The total crack depth at this point was approximately 0.122-inch. The overload crack stopped at this distance to be followed by another IG cracking that extended to approximately 0.017-inch. The final overload fracture followed this small region of IG crack. Evidence of contamination and sanding marks along the outer surface on the first coat of paint at the failure origin region suggest that some type of rework/repair was done on the first coat of paint in this area. The contamination could be from some chemical solution used to clean the area, and this chemical solution could have contributed hydrogen to this area. The hydrogen contribution, coupled with stresses sufficient for crack initiation, most likely initiated the first IG cracking. The second IG crack region [0.017-inch] was most likely caused by hydrogen assisted cracking which continued to the final overload failure that extended the crack into the inner wall and traveled toward the top of the cylinder.
The Center Landing Gear failed by intergranular [IG] cracking. Evidence of contamination and sanding marks along the first coat of paint on the outer surface of the cylinder at the failure origin region appears to indicate that some rework/repair was performed. The contamination could have resulted from the use of some type of cleaning solution which in turn may have caused embrittlement. The presence of hydrogen absorbed from the embrittlement solution coupled with stresses sufficient for crack initiation, initiated the first IG crack along the outer surface of the cylinder. No evidence of any manufacturing, processing or material discrepancies were found.