NTSB Factual Report

HISTORY OF FLIGHT

On July 5, 2014, at 1513 Pacific daylight time, a Globe Swift GC-1B, N3773K, made a forced landing into a field following a loss of engine power during the approach to Pierce County Airport/Thun Field, Puyallup, Washington. The pilot, who was the registered owner, was operating the airplane under the provisions of 14 Code of Federal Regulations (CFR) Part 91. The private pilot, the sole occupant, was not injured; the airplane sustained substantial damage. The local flight departed from Vashon Municipal Airport, Vashon, Washington about 1455 with a planned destination of Puyallup. Visual meteorological conditions prevailed and no flight plan had been filed.

In a written statement the pilot reported that while on the downwind leg of the traffic pattern for runway 34, he configured the airplane for landing which included lowering the landing gear. He extended the downwind leg to allow for adequate spacing of other traffic and lowered the wing flaps as he turned onto the base leg. During the turn, the engine made a loud "bang" and began to experience a partial loss of power. The JPI engine monitor installed was indicating that the exhaust gas temperatures (EGTs) on the engines right side (cylinders nos.1, 3, and 5) were minimal. The oil and fuel pressure gauges remained in the green, showing normal indications.

The pilot further stated that gray smoke began emanating through the right side of the cockpit although he could not detect the source of any fire or flames. He activated the fuel boost pump to the "low boost" setting and the engine began operating worse. Thereafter he turned off the boost pump and the engine appeared to temporarily gain power, though the JPI now showed low EGT indications for the left cylinders. The airplane continued to descend and collided with an embankment that was located on the airport property prior to the runway. The airplane bounced and slid to a stop. As a result of the impact, the airplane incurred structural damage to the left wing.

A preliminary engine examination revealed that the no.3 cylinder's piston was not intact.

AIRCRAFT INFORMATION

The Globe GC-1B Swift, serial number (s/n) 1466, was manufactured in 1946. The engine's data plate indicated it was a Continental Motors IO-360-KB3B engine, s/n 288571-R.

The engine had accumulated 1,393 hours total time, which was determined by using the maintenance logbook and accounting for the tachometer time on the airplane at the examination (181.1 hours). The last annual inspection was dated October 20, 2013 at an airframe total time of 3,431.5 hours. The entry noted that the "engine will be airworthy concurrent with the airframe annual inspection, upon completion of repair to #3 cylinder for exhaust leak."

The engine underwent maintenance in December 2013, about 30 hours prior to the accident. The maintenance records and invoices indicated that the nos.3 and 5 cylinders underwent repairs due to the exhaust anomalies noted in the annual inspection. The invoice to the pilot detailed that the newly purchased (and presumably replaced) parts for the repairs were as follows (in pertinent part): two intake valves, two intake guide seals, one exhaust valve [the second exhaust valve was on back order], two exhaust guides, and eight valve keys. The labor detail paperwork that was filled out by the mechanics performing the work showed that the no.3 cylinder was repaired and the exhaust valve guides were replaced. The box was not initialed stating the work was completed for "assemble cylinder with valves and springs," but was initialed for "assemble cylinder complete with valves, springs, piston, and rings." That paperwork was signed off by the mechanic as completed on December 02, 2013. The same initialed boxes were found for the no.5 cylinder.

The logbook entry for this maintenance, dated December 12, 2013, stated that both the nos.3 and 5 cylinders were removed for repair, the exhaust guides and valves were replaced with new parts, the intake valves were replaced with new parts and the cylinders were installed using new gaskets and seals. The entry additionally noted that the oil filter was removed and normal deposits were found; a new oil filter was then installed.

A representative from the fixed-base operator (FBO) that performed the maintenance stated that although the pilot was not charged for two exhaust valves on the invoice, they both were installed as part of the repair. He stated that the labor detail indicated that both exhaust valves were rejected and the parts request sheet indicated that one valve was received while the other was back ordered. Their purchase order to the distributor and the inventory received indicated that the pilot did in fact receive two exhaust valves although was only charged for one.

The next maintenance entry in the engine's logbook (the last entry) indicated that on May 27, 2014 the pilot changed the oil and filter at a tachometer time of 170.8 hours, equating to 19.1 hours after the aforementioned maintenance. The entry stated that he inspected the oil filter and observed small metal flakes, which he took pictures of, the largest of which was 0.108 inches. The pilot sent a sample of the oil to a laboratory for analysis, which is a common practice for him at every oil change. A further notation in the logbook stated that he would run the engine for 25 hours and check the filter again.

The laboratory performing the oil analysis completed a report dated May 31, 2014. The results showed a high quantity of copper. According to the laboratory explanatory guide accompanying the report, the definition of "high" is a value that is "statistically in the high range for your engine, TIS [time in service] and cylinder type (if applicable)," it continued stating that "this wear metal value is in a range that is well above both the average and caution levels for that wear metal for your engine and therefore corrective action should be taken." It also listed that the most common sources of wear metal elements in oil were "cylinders, rotating shafts, valve train, and any steel part sharing the oil."

The pilot contacted an aviation expert mechanic asking for advice about how to precede with the oil analysis findings. He suggested sending the metal flakes found in for a scanning electron microscope (SEM) evaluation to determine the alloy of the metal in an effort to determine its origins. He further stated that, based on the quantity of metal found, an increase in the amount of metal that accumulated would be a determining factor as to whether or not to continue flying. He added that "as long as the amount of metal is relatively small and it seems to be getting better, it's generally okay to fly but on a short oil/filter-change interval until the problem is [resolved]."

Upon this advice, the pilot sent in the metal flakes to the same laboratory for material analysis. That report, dated June 19, 2014, indicated that the magnetic flakes were carbon steel with a closest match of Aerospace Material Specification (AMS) #5040 and #7225. A comment following that AMS identification recommended contacting the engine manufacturer's service representative if further assistance was needed. Review of engineering drawings for the valve retainer keys (keepers) revealed that they are made from Society of Automotive Engineers (SAE) 1006, 1008, or 1010 materials (which are similar to standard specification of AMS 5040).

The pilot stated that after speaking with numerous mechanics, he decided to fly the airplane for an additional 10 hours before taking another oil sample. The intent of the accident flight was to land at the destination airport to purchase parts he needed for the upcoming oil change.

TEST AND RESEARCH

The engine and fuel system were examined following recovery of the wreckage; a detailed examination report with pictures is appended to the report in the public docket.

An external examination of the engine revealed that all cylinders were secured to the crankcase and there was no immediate evidence of mechanical malfunction. Removal of the top spark plugs revealed that the no.3 plug had sustained significant mechanical damage. Both the spark plugs for the nos. 1 and 4 cylinders contained a thin metallic splatter on the electrode face. According to the Continental Motor's representative, the nos. 2, 4 and 6 spark plugs revealed evidence of normal wear conditions and combustion deposits.

The induction system remained attached to the engine and the risers remained attached to their respective cylinders. Removal of the induction system revealed metal debris in the intake above the no.4 cylinder. No additional anomalies were noted with the induction system.

The cylinders' combustion chambers were examined through the spark plug holes using a lighted borescope. Examination of the no.3 cylinder revealed a large rectangular hole was mechanically formed on the top side of the piston, and the cylinder head sustained significant mechanical damage. The exhaust-valve head was separated from its stem and lodged in the exhaust valve seat perpendicular from its normal seated position. Removal of the rockerbox covers on the no. 3 cylinder revealed metal debris contained within an oil film. The no.3 cylinder's exhaust pushrod was bent and the cylinder's exhaust valve retainer key (keepers) was destroyed; the rocker toe sustained light deformation.

The cylinder was removed revealing that the piston skirt had been fractured, but the piston remained secured to the connecting rod by the piston pin and the remaining portions of the piston pin boss. The edges of the piston did not appear to be eroded or worn. A detailed description of the damage is contained in the metallurgy report in the public docket.

The propeller remained attached to the crankshaft's propeller flange and the two propeller blades remained attached to the propeller hub. The propeller blades were bent aft; one blade was bent aft approximately mid-span and the other blade was bent aft at its tip. Leading edge gouges were observed on the blade tips.

The oil sump remained secured to the bottom of the engine. Removal of the sump revealed that there were numerous pieces of fractured piston and piston rings along with other metal debris.

The oil filter was secured to the backside of the engine with the date May 27, 2014 written on the housing along with a tachometer time of 170.8 hours. The filter element was removed and cut open, disclosing metallic particles embedded in the pleats.

Metallurgy Information

Personnel from the National Transportation Safety Board Materials Laboratory examined the nos.3 and 5 cylinders. The complete report with pictures is appended to the report in the public docket.

The report noted that the top interior surface of the no.3 cylinder, which was opposite the piston crown, was heavily damaged. The adjacent portion of the cylinder exhibited batter, impact marks, and smearing. The intake valve was still present, though it sustained damage on the interior portions of the head facing the piston. The head of the exhaust valve detached from the valve stem and was wedged in the exhaust port perpendicular to its normal position. After removal of the exhaust valve head, more damage was observed on the interior portions of the exhaust port.

The interior damage to the no.3 cylinder and other components, along with the detached exhaust valve head, was consistent with the exhaust valve having fractured at the valve stem. Once liberated, the exhaust valve head was pommeled between the cylinder and piston crown. All fracture surface features were obliterated, consistent with repeated impact against other components in the no.3 cylinder while the piston was being actuated. The exterior exhaust components of the no.3 cylinder were examined. The stem of the exhaust valve was not present; no remnants of the stem material were found. The non-contact sides of the rocker arm exhibited material displacement consistent with abnormal wear of the rocker arm. No visible wear marks or scars were observed on the exterior faces of the retainer. The no. 3 exhaust rocker arm interior exhibited wear on the contact face with the stem, as well as gouging on the arm neck interior. The keepers from the no.3 exhaust assembly exhibited gouging and loss of material, deeper than the machined grooves on the interior faces. The exterior faces also exhibited chatter marks, material spalling, and wear marks. Most notably, one of the keepers was absent the entire upper half. The surface of this keeper facing the upper portion of the exhaust assembly exhibited crescent-shaped witness marks akin to chatter marks on the surface of the keeper fragment. These signatures or rocker wear and the keepers' deformation were consistent with the keeper misalignment during operation.

The debris submitted, along with debris found during disassembly of the no.3 cylinder, was consistent with material from the keepers, the valve, or the damaged cylinder. There were no indications of foreign material or contamination in or on the components of the no.3 cylinder assembly.

NTSB Probable Cause

The failure of the exhaust assembly in the No. 3 cylinder for reasons that could not be determined due to the damage incurred to the fractured parts. Also causal to the accident was maintenance personnel’s improper installation of the cylinder.