NTSB Factual Report

HISTORY OF FLIGHT

On May 16, 2009, about 1548 eastern daylight time, a Diamond Aircraft Industries GmbH, DA 40, N323JT, registered to and operated by Waltzing Matilda Training LLC, experienced separation of the right main landing gear during landing at Norwood Memorial Airport (OWD), Norwood, Massachusetts. Visual meteorological conditions prevailed at the time and no flight plan was filed for the 14 Code of Federal Regulations (CFR) Part 91 instructional flight. The airplane was substantially damaged and the certificated student pilot, the sole occupant, was not injured. The flight originated at OWD about 1543.

The student pilot stated that earlier that day he flew with his certificated flight instructor (CFI) and performed three touch-and-go and two full-stop landings; all were uneventful. His CFI signed him off for his first solo flight, and exited the airplane. The student departed and remained in the traffic pattern for runway 10, turning onto downwind, base and final. While on final approach with the flaps extended, he maintained 75 knots and reduced power to 15 inches manifold pressure when the airplane was over the threshold. He leveled the airplane, then applied aft elevator control input which resulted in a slow decrease in airspeed followed by activation of the stall warning horn. The airplane landed with an indicated airspeed of between 60 and 65 knots with no bouncing, and turned left. He applied right rudder pedal input and right brake to correct but was unsuccessful. The longitudinal axis of the airplane was not aligned with the runway, and while sliding, approaching the edge of the runway, he heard a noise and felt a bump which he thought was the landing gear breaking. After coming to a stop, he notified the tower and then exited the airplane. The student further reported that prior to the accident flight he had performed 39 landings.

The student’s CFI stated that he flew with the student earlier that day practicing five normal and abnormal landings which he described as "safe." The weather conditions were "prime" for the student's first solo flight; the wind was straight down the runway at less than 10 knots and the windsock at the approach end of runway indicated no wind. After receiving his endorsement for solo flight, the student started the engine and taxied to runway 10.

The CFI elected to stay on the ramp to observe the flight and was located approximately 400 yards away. He reported seeing the airplane on final approach with the flaps down and perceived that the ground speed was good and the pitch attitude was stable. The airplane's first ground contact was located within 500 to 600 feet from the displaced threshold, and there was no float or bounce. The longitudinal axis of the airplane appeared aligned with the runway at touchdown. Approximately 1 to 3 seconds later, he noted that the airplane pitched up and the right wing raised, followed by the sound of screeching tires. He reported the pitch up wasn't, "too steep" and the airplane bounced no more than 1 to 2 feet. The airplane veered off the north side of the runway, and the CFI reported a "very violent skidding turn (brakes alone would not cause such a violent turn."

PERSONNEL INFORMATION

The student pilot, age 31, was issued a Federal Aviation Administration third class student pilot medical certificate on March 31, 2009, with a limitation to wear corrective lenses. The National Transportation Safety Board Pilot/Operator Aircraft Accident/Incident Report submitted by the operator indicated the student pilot had 17 hours total time, all of which were in the accident make and model airplane. Within the last 90 days and 30 days, he accumulated 8 and 7 hours, respectively. Other than the accident flight, he did not have any logged solo flight time.

AIRCRAFT INFORMATION

The fixed tricycle gear, composite airplane was manufactured in 2006 by Diamond Aircraft Industries, Inc., as model DA 40, and was designated serial number 40.714. It was equipped by Design Change Advisory MAM 40-123/e with 18 mm thick landing gear struts. Design Change OAM 40-124, also incorporated onto the airplane, specified that smaller main landing gear tire 15x6.0-6 was approved for installation.

Review of the maintenance records revealed the airplane was last inspected in accordance with an annual inspection on October 31, 2008. The airplane total time in service at that time was 174.6 hours, and the airplane total time in service at the time of the accident was 226.7 hours.

Further review of the maintenance records revealed that, on August 11, 2008, at airplane total time 173.8 hours, a new right main landing gear tire 15X600X6 (actually 15X6.0-6) and new tube 6.00-6 were installed. On February 17, 2009, at airplane total time 272.3 hours, a new right main landing gear tube P/N 6.00-6 was installed.

According to the airplane manufacturer, the correct tube P/N based on the installed tire is 15X6.0-6.

METEOROLOGICAL INFORMATION

A surface observation weather report taken at OWD, at 1553, or approximately 5 minutes after the accident indicated the wind was calm, the visibility was 10 statute miles, and overcast clouds existed at 6,000 feet. The temperature and dew point were 19 and 12 degrees Celsius, respectively, and the altimeter setting was 30.05 inches of mercury.

AIRPORT INFORMATION

The Norwood Memorial Airport is equipped with two asphalt runways designated 17/35 and 10/28. Runway 10/28 is 3,995 feet long and 75 feet wide.

WRECKAGE AND IMPACT INFORMATION

Examination of the runway by a representative of the airframe manufacturer with an FAA inspector revealed skid marks corresponding to the left and nose landing gear tires, left of runway centerline beginning approximately 1/3 down the runway. The skid marks from the left and nose landing gears depicted a left arc and were nearly continuous to the north edge of the runway. Two distinct skid marks associated with the right main landing gear were noted at the end of the skid marks near the north edge of the runway. A gouge in the runway surface associated with the right main landing gear wheel skid mark was noted; the skid marks associated with the right main landing gear were serpentine like. The total length of the skid marks were measured and found to be 247 feet. The airplane came to rest upright on grass off the north side of the runway with the right main landing gear nearly separated.

Examination of the airplane by a representative of the airframe manufacturer with FAA oversight revealed that the right main landing gear remained attached only by brake hoses. A spanwise crack was noted in the upper wing skin of the right wing from the wing root outboard. Inspection of the right main landing gear wheel assembly revealed the wheel was broken in large pieces almost to the center where the two halves join. The outboard rim was scarred and abraded by the runway surface. The right main landing gear tire (15x6.0-6) was flat and exhibited sidewall damage appearing as cuts. There were no wear or skid marks on the tread surface. The inner tube had a 5 inch long split along it’s inside center line. The wheel fairing was severely damaged.

Inspection of the right wing revealed the outer main landing gear rib part number (P/N) D41-5313-52-00_2-00 with attached main landing gear leg was torn out of the right wing center section. The aluminum retaining plate P/N D41-3213-01-51, which secured the right main landing gear in place, showed bending damage in the forward 1/3 of its length. The right main landing gear leg P/N D41-3213-12-50 was bent inward approximately 7 degrees at the lower radius, and approximately 2 degrees about the vertical axis. The right outer main landing gear rib was retained for further examination.

Inspection of the left main landing gear tire revealed uneven wear (flat-spotting) and reduction of sidewall to tread radius on ¼ of the circumference of the tire. Indications of high temperature shredding of rubber on the outboard radius was noted; fragments of rubber were noted on the sidewall/tread radius. Inspection of the nose landing gear tire revealed signs of skidding on the left tread surface only. The angle of abrasion pattern was 90 degrees to the roll line of the tire.

TESTS AND RESEARCH

The right landing gear outer rib was submitted to the NTSB Materials Laboratory for further examination. According to the factual report, the rib, approximately 7 inches tall, was adhesively bonded at the forward and aft surfaces of the rib to vertical webs of the front and rear spar carry-through structures within the cockpit area of the fuselage, and to the inside surface of the upper wing skin. No mechanical fasteners were used to attach the rib to the spars or wing skin. The main landing gear leg was clamped between the upper nylon block and the lower rubber-cushioned aluminum bar at the bottom of the rib.

The rib bonding agent consisted of epoxy matrix material mixed with cotton fibers and fumed silica particles. There were no indications of any improperly mixed or improperly cured adhesive. A Diamond representative reported that the paste would have been brushed onto both the rib and the spar surfaces, with expected bond thickness between the rib and the spar surfaces of 2 to 4 mm (0.08 to 0.16 inch); the bond thickness on the top of the rib was not strictly controlled. At the forward end, the adhesive layer remaining on the rib ranged from 0.09 to 0.21 inch thick. At the aft end, the adhesive layer remaining on the rib ranged from 0.08 to 0.13 inch thick. Along the top surface, the adhesive layer remaining on the rib ranged from 0.31 to 0.33 inch thick.

The aluminum bar at the bottom of the rib that clamped the right main landing gear leg was made up of 3.1354 T351 aluminum (comparable to 2024 T351), with a minimum specified yield stress of 290 MPa (42 ksi). The aluminum bar was bent in a smooth arc centered at a position approximately 1.2 inches back from its forward edge, with the forward end of the bar oriented at an angle of about 12 to degrees with respect to the rest of the bar. There was a localized linear indentation on the top surface of the bar approximately 0.95 inch back from its forward edge, consistent with downward force from the landing gear leg in this location. The forward bolt hole was somewhat elongated.

The forward bolt that held the aluminum clamping bar was also bent, at an angle of about 7 degrees, at a position approximately 1.1 inch up from its bottom end. The aft bolt was not visibly deformed.

The space between the top of the main landing gear leg and the bottom recess in the rib was filled by a nylon block machined to fit in the gap. At the aft end of the rectangular slot in the block, the nylon above the main landing gear leg retained some permanent deformation, with a small but perceptible reduction in thickness and an increase in width of about 0.05 inch, consistent with an upward force from the landing gear leg in this location.

The upper flange of the rib was cracked just aft of the upper forward corner; the crack extended transversely approximately 1.1 inch from the inboard edge of the flange. A triangular region of the flange behind the crack retained delaminations and upward permanent deformation, consistent with an impact on the interior of the flange aft of the crack.

The fractured adhesive bond at the forward end of the rib generally occurred near the interface between the adhesive and the rib, near the interface between the adhesive and the front spar, or through a void. The forward adhesive bond included a number of relatively large voids, estimated to occupy approximately 30 percent of the bonded surface area at the forward end of the rib. A Diamond representative indicated that voids up to a size of 8 mm (0.31 inch) are generally accepted without repair. At the top and bottom of the forward surface of the rib, the fracture was near the interface between the adhesive and the rib, with the lighter-colored areas being exposed glass fabric, indicating fracture at the interface between the glass and the matrix inside the matrix-rich region that was formed at the exterior of the rib in the molding process. The white areas away from the void are areas where fracture occurred near the interface between the adhesive and the front spar, as the surface of the adhesive paste layer revealed imprints of a 2 x 2 twill fabric similar to that making up the rib, presumably from the surface of the spar.

The fractured adhesive bond at the aft end of the rib generally occurred near the interface between the adhesive and the rib or near the interface between the adhesive and the rear spar. Voids were also observed in the adhesive at the fracture surfaces, but the voids were separated and appeared consistent with the expected size of 8 mm (0.31 inch) or less. On the lower three-quarters of the surface, fracture was near the interface between the adhesive and the rib, and the fracture appeared to form a fan pattern of light and dark regions emanating from the inboard edge. The light and dark areas appeared to occupy approximately equal amounts of the surface. The lighter-colored areas were exposed glass fabric, indicating fracture at the interface between the glass and the matrix inside the matrix-rich region that was formed at the exterior of the rib in the molding process. The darker-colored areas were predominantly associated with fracture in the matrix material of the rib slightly away from the surface of the glass fabric, but some of the darker-colored areas indicated separation at voids. The orientation of the yarns in the glass fabric varied somewhat across the surface of the rib. The white area on the upper one-quarter of the aft end of the rib is where fracture occurred near the interface between the adhesive and the rear spar, with imprints of a 2 x 2 twill fabric; this area of the fracture retained material that appeared to be crushed and powdered adhesive.

The fractured adhesive bond at the top of the rib generally occurred near the interface between the adhesive and the rib, with the remaining fracture occurring near the interface between the adhesive and the wing skin. Beginning about 3 inches aft of the front of the rib, and extending 1.5 inches aft on the inboard edge and 3.5 inches aft on the outboard edge, the adhesive was fractured at both the top and bottom interfaces, with the adhesive held in place by the foam strips along both edges. Voids were also observed in the adhesive at the fracture surfaces, but the voids were separated and appeared consistent with the expected size of 8 mm (0.31 inch) or less. The fracture near the interface between the adhesive and the rib also showed lighter and darker regions, with the lighter areas being exposed glass fabric and the darker areas predominantly associated with fracture in the matrix material of the rib slightly away from the surface of the glass fabric. About 1.75 inches aft of the front edge of the rib, there was a 0.5-inch-wide strip across the rib of accumulated and interconnected voids.

The fracture at the bottom forward lip of the rib occurred near the interface between the adhesive and the rib. Parts of the fracture occurred at voids, including one void that was a continuation of the large void on the forward surface. The voids were estimated to occupy approximately 35 percent of the bonded surface area of the bottom forward lip of the rib. Other areas showed exposed glass fabric indicating interfacial fracture between the glass and the matrix inside the matrix-rich region at the exterior of the rib.

The fracture at the bottom aft lip of the rib occurred near the interface between the adhesive and the rib. The fracture showed light and dark areas, with the light areas being exposed glass fabric and the darker areas predominantly associated with fracture in the matrix material of the rib slightly away from the surface of the glass fabric. Some areas of the fracture occurred at voids, which were estimated to occupy approximately 10 percent of the bonded surface area of the bottom aft lip of the rib.

A Safety Board Materials Laboratory study indicates that the maximum bending moment did not occur at the position of the forward attachment bolt, but occurred closer to t

NTSB Probable Cause

The student pilot's failure to maintain directional control during the landing rollout, which resulted in excessive side loads to the right main landing gear and the separation of the outer main landing gear rib from the right wing.