N86ABPiper PA-25-235 · 1965

On March 18, 2013, at 1405 central daylight time, a Piper PA-25-235, N86AB, was substantially damaged during a forced landing following a total loss of engine power near Panama City Florida. The airline transport pilot incurred minor injuries. Visual meteorological conditions prevailed, and no flight plan was filed for the local flight, which originated from Northwest Florida Beaches International Airport (ECP), Panama City, Florida, at 1355. The banner tow flight was conducted under the provisions of Title 14 Code of Federal Regulations Part 91. The pilot performed a preflight inspection of the airplane, including sampling fuel from the gascolator, and noted that the airplane had 75 gallons of fuel onboard. Prior to departing, the pilot performed a run-up check of the engine, which included a check of the carburetor heat, and noted no abnormalities. The pilot then departed the airport and picked up the banner. About 6 minutes later, about 2 miles west of ECP and at an altitude of 900 feet mean sea level (msl), the pilot reported that the engine suddenly experienced a total loss of power. The pilot responded by verifying that the mixture was set fully rich, that the magnetos were both on, and that the fuel valve was on. He then applied carburetor heat, which had no effect. After jettisoning the banner at an altitude of 500 feet msl, the pilot performed a forced landing to the pine forest below.

According to Federal Aviation Administration airworthiness records, the airplane was manufactured in 1965 and was originally equipped with a Lycoming O-540 engine. According to maintenance records, the accident engine was inspected and overhauled following a propeller strike. The maintenance log entry detailing the overhaul of the engine stated in-part, "Overhauled engine [in accordance with Lycoming Engine] overhaul manual." The entry noted that several individual engine components were "overhauled" including the crankshaft and crankshaft gear. The portion of the entry noting overhaul of the crankshaft gear was also followed by the statement, "SB 475 C excluding par 6." The entry also noted the installation of several new parts to the engine including a crankshaft gear bolt, though the installation of a new lockplate was not specifically called out. The entry further noted, "All ADs and SBs have been complied with thru 2010-03." The entry did not explicitly state the final crankshaft gear bolt torque, note and verify the installation and bending of the lockplate against the bolt head, or that the inspections and rework required by Lycoming Service Bulletin Number 475C had been accomplished.

An invoice detailing inspection and work performed to the crankshaft stated, "Inspected and/or repaired per Lyc. S.B. 475C excluding paragraph 6." In addition to maintenance log entries, the operator provided invoices and handwritten notes prepared at the time of the engine overhaul. One of the handwritten notes pages contained parts and part numbers, in an itemized list. The list stated in-part:

"SL STD-2209 Crank gear bolt"

"67514 Crank gear torqued to 204 inch"

"18638 Lock plate folded over on bolt"

During the overhaul, the engine was modified in accordance with a supplemental type certificate to produce 250 horsepower. The engine was installed onto the accident airplane in April 2010, and at that time it had accumulated 3,444 total hours of operation.

On February 22, 2013, the operator completed the airframe and engine's most recent 100-hour inspection. At that time the airframe had accumulated 8,100 total hours, and the engine had accumulated 1,371 hours since the most recent overhaul. Between the time of the 100-hour inspection and the accident, the airplane had accumulated 63 additional flight hours.

Following the accident, the airplane was recovered from the accident site to the operator's maintenance base. One of the fuel tanks was breached during the impact sequence and about 45 gallons of fuel was recovered from the remaining fuel tank. External examination of the engine revealed minimal impact damage and that 12 quarts of oil remained in the oil sump. The engine was subsequently disassembled. During the disassembly the crankshaft gear was found rotationally misaligned with its normal position relative to the crankshaft, and the crankshaft alignment dowel was sheared. The gap between the crankshaft gear and crankshaft mating surfaces measured less than 0.001-inch when utilizing a feeler gauge. The crankshaft, with the entirety of the crankshaft gear and its attaching hardware were forwarded to the NTSB Materials Laboratory for detailed examination.

According to the Materials Laboratory Factual Report, typically the crankshaft gear was mounted inside a counterbored pilot hole in the aft end of the crankshaft. The gear was pressed against the face of the recess by a bolt that engaged a threaded hole in the end of the crankshaft. A lockplate was installed between the bolt and the gear with ears that fold down against the side of the gear and up against the head of the bolt. The gear was rotationally aligned to the crankshaft by an alignment dowel. The alignment dowel from the accident engine was fractured, one end of the dowel retained inside the alignment hole in the gear, while the other end remained press-fit and retained in the end of the crankshaft. The previously noted rotational displacement of the gear relative to the crankshaft was measured to be about 45-degrees, and the lockplate was also rotationally displaced, though to a lesser degree.

The retaining bolt and the gear were removed. There were no visible discrepancies with the retaining bolt. One of the threads contained a soft dark material consistent with an anti-seize compound. The bolt was turned into the threaded hole until it nearly bottomed out. The bolt could be turned with fingers up to the thread containing the compound at which point the compound provided some resistance to further insertion (approximately 3.5 in-lbs. measured using a torque wrench). Over the area where the gear had been in contact with the face of the pilot hole recess, the surface exhibited circumferential scratch marks and had transferred metal adhered to it. The mating face of the gear exhibited corresponding scratches and gouges, consistent with galling.

The dowel fracture surface was examined on the piece of the dowel that was retained in the alignment hole. The fracture surface had a comparatively flat appearance and exhibited multiple crack arrest marks. The morphology of the fracture surface was consistent with a fatigue fracture. The curvature of the crack arrest marks was consistent with the fatigue fracture originating along the surface of the dowel in the area. The fracture features were compared with the mating dowel piece retained in the crankshaft. The comparison indicated the fracture initiated along the clockwise face of the dowel (leading face in the direction of rotation) and that the dowel fragment retained in the gear alignment hole had rotated by approximated 180 degrees.

Diameters on the gear, pilot hole, and alignment dowel were measured and compared with the dimensional limits in Lycoming Service Bulletin No. 475C. The diameter of the pilot flange on the gear was measured three times using a micrometer and the values were 2.1253 inch, 2.1254 inch, and 2.1257 inch. According to the service bulletin, the original dimension of the flange was 2.1250 inch – 2.1255 inch. The diameter of the counterbored pilot hole recess was measured three times using calipers and the values were 2.1270 inch, 2.1275 inch, and 2.1275 inch. According to the service bulletin, if the pilot hole diameter exceeded 2.1262 inch, the diameter should be plated and reground to 2.1250 inch – 2.1260 inch. The reduced diameter on the stepped dowel could only be measured at one location and it measured 0.2504 inch with a micrometer. According to the service bulletin, the original dimension of the dowel is 0.2505 inch – 0.2510 inch.

The hardness of the gear and the alignment dowel were each measured using a Rockwell hardness tester. The gear was placed directly on the stage and the hardness measurements were 38.2 HRC and 37.9 HRC. The piece of the alignment dowel that was retained in the gear was removed from the gear and mounted in a thermosetting compound. The flat end of the dowel was then ground using silicon carbide grinding papers through 600 grit prior to hardness testing. The measured hardness values were 29.7 HRC, 29.8 HRC, and 30.1 HRC. All of the hardness measurements met the material requirements.

Lycoming Engines provided guidance in the form of Service Bulletin No. 475C, which detailed procedure for attaching the crankshaft gear to the crankshaft during overhaul, following a propeller strike, or any other time when removal of the crankshaft gear was required. According to the bulletin, "Damage to the crankshaft gear and the counter-bored recess in the rear of the crankshaft, as well as badly worn or broken gear alignment dowels are the result of improper assembly techniques or the reuse of worn or damaged parts during reassembly. Since a failure of the gear or the gear attaching parts would result in complete engine stoppage, the proper inspection and reassembly of these parts is very important. The procedures described in the following steps are mandatory."

After aligning the crankshaft gear with the crankshaft utilizing the alignment dowel, paragraph 6 of the bulletin directed technicians to utilize a new bolt and lockplate, then tighten the bolt to 125 in.-lbs. of torque. The next step ensured proper seating of the gear by directing, "…with a hammer and brass drift, tap lightly around the pilot flange of the gear and listen for sharp solid sounds from the hammer blows that would indicate that the gear is seated against the crankshaft. As a check on the seating against the crankshaft, attempt to insert a pointed .001 inch thick feeler gage or shim stock between the gear and crankshaft at each of the three scallops. The .001 feeler gage, or any smaller feeler gage, must NOT fit between the two surfaces at any location. (.001 feeler gage is used as an indicator, however, there must be no clearance between crankshaft and gear.)" Finally, the 5/16-inch bolt would be tightened to 204 in.-lbs. of torque, and the clearance between the outer diameter of the gear flange and the counter-bored pilot (inner diameter) of the crankshaft should measure no greater than 0.0005-inch at any point.

Upon completion of the installation the bulletin specified, "A logbook entry, specifying the final bolt torque, verifying that the lockplate was properly bent in place against the bolt head and that the inspections and rework required by Lycoming Service Bulletin No. 475C were accomplished, should be made and signed by an authorized inspection representative."