N827DOconnor Paul A ULTIMATE AERO 10-200 · 1988

On April 18, 2014, about 1210 eastern daylight time, an experimental amateur built, Ultimate Aero 10-200, N827D, was destroyed when it impacted terrain near Saint Albans, Vermont, after an in-flight separation of a propeller blade. The commercial pilot was not injured. Visual meteorological conditions prevailed, and no flight plan was filed for the local personal flight operated under 14 Code of Federal Regulations Part 91, which departed Franklin County State Airport (FSO), Highgate, Vermont, about 1200.

According to the pilot, he recently returned from an airshow in Florida, where he performed aerobatics with the airplane. On the day of the accident, he intended to conduct an evaluation flight, as was his common practice after a long cross-country flight, to ensure the "satisfactory condition" of the airplane. He departed FSO without incident and climbed to 3,000 feet above mean sea level (msl). Shortly thereafter, there was a sudden loud bang/shudder and the canopy shattered. The engine "stopped instantly" and the canopy "clam-shelled open and then slammed back down." The airplane could not be controlled via the flight controls, it lost a lot of forward airspeed and entered a spin that he could not arrest. The pilot then decided to egress via parachute. He could not remember what altitude he egressed from the airplane, but after exiting the airplane, his parachute deployed fully at 700 to 1,000 feet msl, and he came to rest in the top of a tree. The pilot added that he practiced his egress routine prior to every flight and that the egress went as planned; however, his headset jacks would not unplug easily, and had to be broken off, which presented an unexpected challenge.

According to Federal Aviation Administration (FAA) and pilot records, the pilot held a commercial pilot certificate with a rating for airplane single-engine land, a mechanic certificate with airframe and powerplant ratings, and a surface-level aerobatic waiver that he utilized when performing at airshows with the accident airplane. His most recent FAA second-class medical certificate was issued on May 13, 2013. He reported that he had accrued approximately 1,000 total hours of flight experience, 300 of which were in the accident airplane make and model.

The accident aircraft was an experimental amateur built, strut-braced biplane, with cabane struts, interplane struts, and flying wires. It was configured with a single-seat, enclosed cockpit under a bubble canopy, fixed conventional landing gear with wheel pants, and was powered by a fuel injected, air cooled, four-cylinder Lycoming IO-360-C1C engine, driving a Whirl Wind Aviation (WWA) 200C two- blade constant speed propeller.

The airplane was made from metal with its flying surfaces covered in doped aircraft fabric. Its four full span ailerons provided a 400 degree per second roll rate. It was stressed for 10 Gs positive and negative.

According to FAA and airplane maintenance records, the airplane was issued its special airworthiness certificate on April 19, 2010, and was certificated in the aerobatic/experimental category. The airplane's most recent condition inspection was completed on March 10, 2014. At the time of the accident, the airplane had accrued 360 total hours of operation.

The reported weather at FSO, at 1155, included: wind 160 degrees at 12 knots, gusting to 16 knots, 10 miles visibility, clear, temperature 13 degrees C, dew point -14 degrees C, and an altimeter setting of 30.37 inches of mercury.

The accident aircraft was an experimental amateur built, strut-braced biplane, with cabane struts, interplane struts, and flying wires. It was configured with a single-seat, enclosed cockpit under a bubble canopy, fixed conventional landing gear with wheel pants, and was powered by a fuel injected, air cooled, four-cylinder Lycoming IO-360-C1C engine, driving a Whirl Wind Aviation (WWA) 200C two- blade constant speed propeller.

The airplane was made from metal with its flying surfaces covered in doped aircraft fabric. Its four full span ailerons provided a 400 degree per second roll rate. It was stressed for 10 Gs positive and negative.

According to FAA and airplane maintenance records, the airplane was issued its special airworthiness certificate on April 19, 2010, and was certificated in the aerobatic/experimental category. The airplane's most recent condition inspection was completed on March 10, 2014. At the time of the accident, the airplane had accrued 360 total hours of operation.

The airplane was discovered on the shoulder of the north bound lane of Interstate 89 were it had impacted, and was subject to a postimpact fire which consumed the majority of the airplane.

Examination of the wreckage revealed that there was no evidence of any type of structural failure, and flight control continuity was established from the flight controls to the control stick and rudder pedals.

Examination of the engine revealed that it was heavily damaged during the impact sequence and postimpact fire.

Examination of the propeller revealed that one of the two propeller blades was missing, and separated from the two-blade constant-speed propeller's hub.

The propeller assembly and governor were retained and forwarded to the NTSB Materials Laboratory, Washington, DC, for further examination.

Propeller System Description

The WWA 200C/400C series propellers consisted of two-bladed (200C), and three-bladed (400C), composite blade, hydraulically controlled, constant-speed, counter-weighted aerobatic propellers, which were designed for airplanes that used the Lycoming IO-360 or IO-540 engine.

The propellers automatically adjusted blade pitch angle to maintain a pilot selected engine rpm setting. A single acting piston (inside the hub) powered by engine oil, changed the blade pitch. Oil pressure was adjusted and regulated by the engine driven governor. Oil pressure was required to decrease propeller blade pitch (increased rpm) and in the event of oil pressure loss, the counterweights would drive the propeller to coarse pitch, decreasing the possibility of engine over-speed.

Once the engine rpm was selected, it was held constant at all flight airspeeds and power settings automatically by the governor. The propeller control lever was used to set the desired engine rpm by the pilot.

Once the engine rpm was set, it was held constant by the governor, which maintained this preset rpm by varying the propeller oil pressure automatically. Mechanical stops for high and low pitch, limited the pitch change travel of the propeller.

The propeller blades twist was optimized for aerobatic performance and the blades were constructed of graphite composite using thermoset plastic polymer reinforcements. The propeller hubs were machined from solid aircraft quality aluminum alloy which was then shot peened and anodized.

The propeller could be operated in the rain and on unimproved airstrips. Each propeller blade was equipped with a nickel leading edge inlay to protect the blade leading edge from debris damage, increase blade protection, and maximize blade life.

The blades were prepared for finish using polyester based primer and then painted with a high-gloss urethane base coat / clear coat finish. The propeller system was completed with a composite spinner and bulkhead assembly in a ready-to-paint polyester primer coat.

Accident Airplane Propeller

The WWA 200C propeller installed on the accident airplane was installed in April of 2010, and had been operated for about 260 hours. The propeller was installed on a highly modified experimental engine which produced approximately 230 horsepower and was equipped with high compression pistons and a high performance camshaft. According to the propeller manufacturer, Lycoming IO-360's have very large power pulses and high torsional vibrations. In December of 2010, the propeller had been operated for about 8 months, and 100 hours, when WWA replaced one of the propeller blades under warranty due to premature paint cracking. In October 2011, as a result of WWA's concern of potential excessive torsional vibration, they recommended and sold to the pilot, a flywheel harmonic dampener to reduce the torsional vibration amplitude. The pilot never installed the harmonic dampener and subsequently returned it to WWA.

Accident Airplane Propeller Examination

The propeller hub, the root of one propeller blade (Blade 1), the remaining propeller blade (Blade 2), and the propeller governor were examined by the NTSB Materials Laboratory.

Blade 1 was fractured in the root within the hub and the portion outboard of the fracture was not recovered. Blade 2 was largely intact but showed damage consistent with exposure to a fire.

A drawing provided by WWA, showing the overall construction of the root end of a Whirl Wind 200C or 400C series propeller blade was examined. The drawing was adapted from WWA Mandatory Service Letter SL-200C/400C-002, dated February 4, 2003, which required the installation of pan-head screws into the root end face of the blade. According to a subsequent WWA mandatory service letter (SL-200C/400C-070214, dated July 8, 2014); the 200C series propellers affected by WWA SL-200C/400C-002 had serial numbers 200C-238 and lower. The propeller on the accident airplane was serial number 200C-280, and the propeller blades did not have the pan-head screws.

As installed, the propeller blades were retained in place during operation with a split ring insert that rested against the shoulder on the outer ferrule and against the bearing race half on the root side of the blade bearing (bearing that facilitated pitch changes). The bearing race half on the airfoil side of the bearing was retained by the propeller hub. At rest, the blade was prevented from moving inward by a split ring that sat in a groove in the outer ferrule and fitted against the exterior face of the propeller hub.

Examination of the remains of Blade 1 revealed that the outer ferrule was fractured around approximately 2/3 of the circumference with a fracture intersecting the outer surface in the fillet radius adjacent to the shoulder. The fracture surface had relatively flat features with curving crack arrest lines, features that were consistent with fatigue. The fatigue features emanated from an origin area at a bolt hole for one of the two bolts attaching the pitch change pin. The primary crack propagated around approximately 1/3 of the circumference. Fatigue features from secondary fatigue cracks were also observed propagating outward, and shear fractures were present where the outer ferrule fractured to the root face, and the remainder of the outer ferrule separated where it was bonded to the inner ferrule root flange.

Approximately four threads were observed between the threaded end of the bolt attaching the pitch change pin to the blade root and the fracture surface at the origin. Partially cut threads consistent with a location at or near the end of the tapped threads were observed in the bolt hole adjacent to the fracture surface.

The hole depth from the inner ferrule end flange face to the fracture surface was measured for each of the attachment bolt holes. At the fatigue origin, the hole was 0.806 inch deep, and the other hole was 0.796 inch deep. According to a representative from WWA, production procedure notes for the blade specified a hole depth up to 0.750 inch. The distance between the shoulder and the root end face of the blade was also measured. The shoulder was located 0.782 inch from the root face of the inner ferrule root flange in the area where the pitch change pin had been attached.

Examination of Blade 2 revealed that the depth of the pitch change pin attachment holes as measured from the root face of the inner ferrule root flange were also inconsistent. One bolt hole had a depth of 0.67 inches, and the other hole had a depth of 0.70 inch. The shoulder was located 0.782 inch from the root face of the inner ferrule root flange in the area where the pitch change pin had been attached. The depth of the threaded portion of the bolt hole was measured on the cross-section, and the threaded portion of the hole extended up to 0.65 inch from the root face of the inner ferrule root flange. Two partially-cut threads were present at the end of the threads.

Whirl Wind Aviation took the following postaccident actions:

Manufacturing Process Changes

As of May 2014, the pitch change pin attachment holes in newly-manufactured propeller blades have a specified depth of 0.670 inch ± 0.010 inch as measured from the surface of the inner ferrule root flange, and manufacturing and quality controls were instituted to prevent variations in the hole depths from specified levels. Additionally, the holes in blades manufactured since May 2014, are shaped with a flat bottom through a reaming operation, and the threads are formed with a roll-thread tap.

Mandatory Service Bulletin

On January 30, 2015, Whirl Wind Aviation issued Mandatory Service Letter SL-200/400C-070214, which included a requirement to measure the depth of the bolt holes for attaching the pitch change pin on model 200C (serial numbers 200C-385 and lower) and model 400C (serial numbers 400C-226 and lower) propellers. The stated reason for the service letter was that pitch change pin attachment bolt holes could vary in depth, which in some cases could compromise the integrity of the blade outer ferrule and lead to fatigue failure of the ferrule and loss of blade retention. Upon inspection, blades found with bolt holes greater than 0.760 inch in depth as measured from the root face of the internal ferrule root flange are to be removed from service.