DAMAGE TO AIRPLANE Use your browsers 'back' function to return to synopsisReturn to Query Page
On January 7th, 2010, Astar Air Cargo maintenance personnel observed hydraulic fluid leaking from a pin hole (crack) at the upper end of the outer cylinder of the nose landing gear (NLG) of airplane N803DH, a Boeing DC-8-73F, during a pre-flight walk around. The airplane was located at Cincinnati/Northern Kentucky International Airport (CVG) and loaded with cargo at the time of the discovery. On January 8th, 2010, in order to facilitate the removal of the NLG, the airplane was taxied from the cargo area to a hangar. While taxing, a loud bang was heard during a turn. The airplane continued to taxi for approximately one additional mile to the hangar. At the hangar ground personnel discovered that the NLG outer cylinder had fractured during taxi. There was no damage to the airplane other than the fractured NLG and there were no injuries to the maintenance personnel.
The incident NLG assembly was installed during an A-check by Commercial Jet in Miami, FL between January 3, 2010 and January 5, 2010. Prior to the incident the airplane had made one flight with no cargo from MIA to CVG. The NLG outer cylinder was last overhauled by AAR LGS in April 2007. Cycles since new (CSN) for the NLG outer cylinder are not known as the NLG is not life limited.
The NLG outer cylinder had fractured into three definable sections. The first measured approximately fourteen inches in length over 60% of the circumference. A small portion of the upper spline, measuring about 4 inches, remained attached to this first section. The second section was comprised of the remainder of the upper spline area with an approximate 8 inch triangularly spiked section. This section was attached to the orifice tube. The final section was comprised of the remaining outer cylinder.
The first fractured section was removed from the NLG for a visual inspection. A stained area was identified along the crack face measuring approximately one inch in length. A matching area was identified on the large remaining section of the NLG outer cylinder. The remainder of the crack faces were unblemished. A small ‘tear’ was found emanating from the stained area along the chrome plated I.D.
A review of the work order revealed that the working I.D. area not was repaired during overhaul. The initial condition report (ICR) stated that wear marks were found and only polishing was accomplished to the chrome plated I.D. The splines were also found to have minor wear marks which were removed by polishing.
The entire NLG assembly was shipped to AAR LGS in Miami for disassembly. The NLG was disassembled and the following discrepancies were noted:
1. The upper bearing was canted on the inner cylinder.
2. The I.D. of the split bearing had impact damage from contact with the inner cylinder.
3. The nut for the orifice tube was distorted and could not be removed using the specified tooling.
4. The orifice tube threads were damaged and a section of one of the threads had sheared off.
The following components were visually inspected and found with no discrepancies:
1. The torque link bushings in the inner cylinder and steering collar.
2. The torque links and steering cams.
3. The lower torque link attachment lugs were checked for parallelism – within 0.004 inches TIR.
The thickness of the chrome plating was measured on the section of the cylinder containing the stained area. In the stained area of the fracture face, the thickness was 0.0030/0.0035 inches. The thickness increased to .0065/0.0071 inches at approximately 120° from the stained area. The three sections of the outer cylinder were crated and sent to Boeing (Long Beach) for further analysis.
TESTS AND RESEARCH
Visual and Macroscopic Examination
Examination confirmed that the NLG Cylinder had several, complete through-thickness fractures that extended across the top third (approximately 13 inches) of the cylinder. The cylinder fractured into three pieces; top spline region, the "thumb nail" shaped half of the cylinder and the remaining cylinder. Fracture traces and discoloration determined that the fracture origin was a longitudinal crack region located approximately 6.5 inches from the top of the cylinder. Fracture traces indicated that the crack propagated from the aforementioned origin downwards and curving clockwise (looking down) while propagating around the cylinder and through the spline region. From the spline region, the fracture propagated downwards and intersected with the aforementioned origin.
The aforementioned origin had a secondary crack that ran diagonal to, and intersected the center of, the primary fracture. Fracture traces and discoloration indicated that the primary fracture originated from the center region of the crack. The discoloration ranged from rust colored, indicative of corrosion, to a blue-gray discoloration, indicative of high temperature oxidation. The primary fracture origin consisted of a highly oxidized, elliptically shaped pattern that extended through approximately three quarters of the thickness and was approximately 0.95 inch long. From the discolored region through the thickness, the fracture had a faceted appearance that was approximately 1.1 inches long. The remaining fracture surface had a dull, coarse grain texture typical of an over-stress rupture.
To facilitate examination, the secondary crack was excised and opened. Fracture traces and discoloration due to oxidation indicated that the fracture originated from the center region of the crack. The secondary fracture origins consisted of a heavily oxidized elliptically shaped pattern that extended through approximately three quarters of the thickness and was approximately 0.85 inch long. The remaining laboratory induced fracture surface had a dull, coarse-grain texture typical of over stress rupture.
The engineering drawing calls for deep electromechanical etch identification(P/N, S/N, and vendor) on one of the two cylinders protruding lugs located at the bottom of the cylinder. Stripping off the paint and primer to bare metal did not reveal any evidence of identification on either lug.
Scanning Electron Microscope (SEM) Analysis
The cylinder fracture was excised and cleaned by repeated applications of replicating tape followed by ultrasonic cleaning in a bath of acetone. The subsequent examination confirmed that the primary fracture originated from the aforementioned location. The surface directly adjacent to the primary origins consisted of light circumferential scratches.
The primary fracture origin consisted of cleavage rupture through the plating surface, typical of a fracture through chromium plating. The origins through the base metal consisted of intergranular rupture with general corrosion on the grain boundary facets. As the crack depth increased, within the discolored region, the amount of corrosion appeared to lighten. The non-discolored region to the outer diameter surface had an intergranular appearance with no appreciable corrosion products. At the crack ends, the non-discolored, slow growth region consisted of intergranular rupture with small islands of over stress rupture. The remaining rapid fracture region consisted primarily of over stress rupture with few facets.
SEM examination of the inner diameter surface directly adjacent to the aforementioned secondary fracture origins determined that it also consisted of light circumferential scratches.
The secondary fracture origin also consisted of cleavage rupture through the plating surface, typical of fracture through the chromium plating. the origins through the base metal also consisted of intergranular rupture with general corrosion on the grain boundary facets. As the crack depth increased, within the discolored region, consisted primarily of over stress rupture with a few intergranular facets.
SEM Energy Dispersive Spectrometry (EDS) analysis determined that the fracture origins and discolored slow-growth regions consisted primarily of iron, major amounts of oxygen, minor amounts of carbon, chromium, sodium, chlorine, manganese and nickel and trace amounts of silicon, sulfur and potassium, consistent with a low alloy steel corrosion products. Analysis determined that the non-discolored, slow-growth region was consistent with the base metal composition. Analysis of the inner diameter plating determined that it consisted of chromium and minor amounts of oxygen and carbon, typical of chromium plating on a steel substrate. Analysis determined that the base metal consisted primarily of iron with minor amounts of chromium, nickel manganese, and carbon, typical of a low alloy steel.
Microhardness (Knoop) Testing
Microhardness transverse testing from the inner surface edge, at close proximity of the failure origin area was conducted to determine if decarburization existed at the inner surface that could contribute to the failure. The tests showed no evidence of decarburization or carburization.
A specimen normal to the fracture surface at the fracture origin was excised and prepared for metallographic examination. Examination of the specimen revealed typical 4340 steel in the hardened and tempered condition with no evidence of abnormality. Observation at the origin area showed a crack through the chrome plating and coincidental to the crack in the base metal.
Sections of the cylinder in close proximity of the failure origin region were etch inspected. The results showed evidence of circumferential grinding burns along the inner surface of the cylinder. However, macroscopic examination did not show any cracking associated with these intermittent grinding burns.
A section of the cylinder in close proximity of the failure origin was excised and hardness tested. The measurement results were HRC 51-52 which met the HRC 50-54 for 4340 in the 260-280 ksi tensile range.
A sample from the cylinder was submitted to Durkee Testing Laboratories for chemical analysis. The results determined that the material and met the chemical requirements for the required material. The hydrogen content was reported to be 1 ppm, which was too low to cause hydrogen embrittlement.
The wall thickness along the failure origin region and the length of the secondary diagonal crack were measured. The measured dimensions were in where in the required range.
Chrome Plating Thickness
The chrome plating on the inner surface of the cylinder was measured in several locations and had an average thickness of 0.003 inches thick. No requirements are on the engineering drawing; however Boeing documents state unless otherwise specified, parts to be ground to finish size shall be plated to at least 0.002 inches thick.