Recently Released TSB Reports Recently Released TSB Reports


The following summaries are extracted from Final Reports issued by the Transportation Safety Board of Canada (TSB). They have been de-identified and include the TSB’s synopsis and selected findings. Some excerpts from the analysis section may be included, where needed, to better understand the findings. For more information, contact the TSB or visit their Web site at www.tsb.gc.ca. —Ed.

TSB Final Report A07W0138—Loss of Control and Collision with Terrain

On July 23, 2007, an Aerospatiale AS350BA helicopter was en route from a staging site at Johnson Lake to Fort McMurray, Alta., with the pilot and four heli-tack firefighters on board. About 20 minutes into the flight, as the helicopter was cruising at about 1 500 ft above ground level (AGL), the pilot initiated a rapid descent to just above the tree tops, and lost control of the helicopter when he attempted to level off. The helicopter rolled right, nosed down, struck the marshy terrain, and rolled over onto its left side. One passenger was fatally injured, and the other occupants were seriously injured. One of the passengers manually switched on the emergency locator transmitter, while another passenger contacted the forestry radio dispatcher on his radio. Rescue helicopters were dispatched immediately and arrived at the accident site within an hour. The time of occurrence was about 20:00 Mountain Daylight Time (MDT).

Loss of control and collision with terrain

Other factual information

The flight initially climbed to, and cruised at, an altitude of about 1 500 ft AGL. About 20 minutes later, the pilot descended to a lower altitude to observe wildlife. He did not notify the unit leader or consult with the passengers. Instead of lowering the collective to descend, the pilot pushed the cyclic forward to lower the nose of the helicopter and increase the airspeed. On reaching an altitude just above the treetops, the pilot attempted to level off by raising the collective slightly and pulling back on the cyclic. However, the cyclic control could not be moved. As the pilot continued to pull back on the cyclic with both hands, the helicopter rolled to the right, pitched up, then dove into the ground and came to rest on its left side. The passenger in the left rear seat was ejected from the helicopter when his inboard seat belt attachment failed and he became trapped under the fuselage.

Servo transparency

Awareness of servo transparency and recovery was part of the pilot’s initial and recurrent ground training on the AS350 series helicopters. It was reported that the pilot had previously flown in a similar manner on other flights when transiting between bases, with sudden climbs, descents, and pull-ups. Some of the passengers reportedly were discomforted by the manoeuvres; however, no complaints were submitted.

The terms servo transparency, servo reversibility, and jack stall all refer to the phenomenon whereby the aerodynamic forces on the rotor blades can exceed the opposing power of the hydraulic servos to control the blade pitch. This phenomenon can occur in any helicopter that has hydraulically actuated flight controls. Factors that affect servo transparency are as follows: high airspeed, high collective pitch, high gross weight, high g loads, and high density altitudes. The maximum force the servo actuators can produce is constant, and is a function of hydraulic pressure, the servo characteristics, and possibly the level of maintenance of the system. All components of the hydraulic system, with emphasis on the servos, were examined after the wreckage was recovered to Fort McMurray. No anomalies were found.

The manufacturer has stated that the transparency phenomenon in AS350s is non-violent and transitory, and normally lasts for a period of two to three seconds. The controls are fully operable throughout the event. However, the force required to move the controls increases significantly, to the extent that an unknowing pilot may think that the controls are jammed. On AS350s with a clockwise main rotor rotation (as viewed from above), the right servo receives the highest load; therefore, servo transparency will result in an uncommanded cyclic movement to the right and aft. This will cause the aircraft to roll to the right and pitch up. Normal recovery procedure is to decrease the aerodynamic load on the main rotor by lowering the collective. Depending on the aircraft weight, speed, and atmospheric conditions, the manufacturer has calculated that servo transparency can occur at g loads as low as 1.5 g.

On May 14, 2007, the Australian Civil Aviation Safety Authority issued Airworthiness Bulletin (AWB) 27-008, based on Federal Aviation Administration Special Airworthiness Bulletin (SAIB) SW-04-35 issued on December 19, 2002. These bulletins reference Eurocopter Service Letters 1648-29-03 for the Astar (AS350) family and 1649-29-03 for the Colibri (EC120) family, and provide detailed information on servo transparency, as well as recommendations to reduce the possibility of encountering the phenomenon.

Analysis

The atmospheric conditions, aircraft weight, and the pilot’s manoeuvres at the time of the occurrence were conducive to the onset of servo transparency, a phenomenon the pilot was aware of, and had been trained to recognize. He was not able to translate his training into a conditioned response: to lower the collective instead of fighting the cyclic, when the event occurred. The altitude and proximity to the trees at which the pull-up was initiated did not allow sufficient time for the pilot to correct his initial reaction. Servo transparency in AS350s is a well-known phenomenon and the recent service letters and airworthiness bulletins emphasize the need for operators and pilots to be more actively aware of the onset conditions and recovery procedures.

The passengers were not weighed, nor were the weights recorded or presented to the pilot, who did not complete an accurate weight and balance report before departure. These actions created the potential for the weight and balance to be outside allowable limits. This in turn introduces a risk that the helicopter performance could be affected. The gross weight, one of the factors affecting the onset of servo transparency, needs to be closely monitored by the pilot.

Findings as to causes and contributing factors

  1. The pilot initiated a sudden high-speed descent, and experienced a loss of control due to servo transparency when he attempted to level off at the bottom of the descent.
  2. The pilot did not initiate the correct recovery procedure when servo transparency was experienced and, due to the proximity to the trees, insufficient time remained for the pilot to correct his initial reaction.

Findings as to risk

  1. The pilot had previously initiated sudden climbs and high-speed descents that were not standard operating procedures. These manoeuvres had not been reported to Alberta Sustainable Resource Development (ASRD) or to the helicopter operator.
  2. The pilot did not complete a weight and balance report before departure. Therefore, the pilot could not confirm if the helicopter was being operated within allowable limits.

Safety action taken

Alberta Sustainable Resource Development (ASRD) amended its Representative Responsibility Standard Operating Procedure (SOP) with the addition of several detailed criteria regarding passenger and cargo weights (see final report on TSB Web site for more details).

TSB Final Report A07W0186—Engine Failure and Collision with Terrain

On October 26, 2007, a privately operated Piper Malibu PA46-310P was en route from Salem, OR, to Springbank, Alta., on an instrument flight rules flight plan. During the descent through 17 000 ft at approximately 55 NM southwest of Calgary, the pilot declared an emergency with the Edmonton Area Control Centre, indicating that the engine had failed. The pilot attempted an emergency landing at the Fairmont Hot Springs airport in B.C., but crashed at night at about 19:12 MDT, 11 NM east of Invermere, B.C., at approximately 3 633 ft ASL in wooded terrain in the Rocky Mountain ranges. The pilot and two passengers were fatally injured.

Engine failure and collision with terrain

The aircraft was certified, equipped, and maintained in accordance with existing regulations and approved procedures. In the summer preceding the accident, the engine developed a knocking sound that was audible when power was reduced for landing. This had not been entered into logbooks nor reported to any maintenance facility. All flights on the day of the accident were carried out without the oil filler cap in place, as it was found at the hangar where the aircraft was kept. The absence of the oil filler cap could have resulted in the loss of engine oil, but its absence did not result in any loss of oil through that opening. The crankcase oil breather has a tube running from the dipstick opening to the breather canister. There was no evidence of oil accumulation here or at the bottom of the cowling.

Two alternators generate electrical power, one belt-driven and one gear-driven. The gear-driven alternator derives its rotational power from a gear bolted to the crankshaft between the number four and five main journals. This in turn drives the alternator coupler. This coupler consists of a sleeve with an attached cup, locked to the alternator shaft. The cup is driven by a formed rubber ring on the inner surface of the cup outer wall, which is then attached to the gear on the alternator shaft. The alternator drive hub is designed to slip when abnormal torque is required to rotate the alternator shaft. This prevents engine damage or loss of power in the event of an alternator seizure.

In the months before the occurrence, a number of maintenance actions were performed on the gear-driven alternator as a result of alternator failure indication. The alternator drive coupling was replaced approximately five flight hours prior to the accident flight. The coupling that was removed had a substantial amount of rubber material missing from both the front and back surfaces. This allowed the slip joint to spin and not lock to the cup as is normally the case. The coupling also made use of a handmade unapproved flat washer inside the cup that had a number of very rough edges and markings. This type of alternator is not designed to be used with such a washer, nor is it approved as a repair for continued airworthiness of the engine.

This washer was found to have forced the rubber ring further out of the cup and engage the gear teeth of the crankshaft alternator drive gear. This resulted in the destruction of the rubber portion of the coupling. Rubber particles of various sizes found their way into the engine sump (see Photo 1).

The rubber particles found in the engine sump matched those of the old coupling. In addition, several of the lifters contained rubber debris, indicating that the oil filter had been in a bypass state, allowing debris to flow into the system. The oil filter was also found to contain a large amount of rubber and metal debris. When the coupling was changed, the engine oil and filter were not changed, nor was the engine oil system flushed. The engine maintenance manual recommends checking the oil filter for metal debris during oil changes, but does not specify checking for other types of debris during other forms of maintenance. The engine manufacturer issued Service Bulletin M84-5 in 1984 that addressed gear-driven alternator malfunctions on all of its 520 series engines. It specifies that if any contamination is found upon removal of the alternator, the oil sump must be removed, the pick up cleaned or replaced, and, if anything further is found, a Teledyne Continental service representative should be contacted. This service bulletin does not apply to the 550 series engines even though they are equipped with gear-driven alternators. Standard industry practice is to check oil systems when contamination of any kind is found or known about, to flush the system, and to ascertain the source before releasing the aircraft for flight.

Photo 1. Gear damage to rubber material
Photo 1. Gear damage to rubber material

The top of the engine at centreline had a large hole over the number two connecting rod. The crankshaft and the number two connecting rod had indications of extreme heat, which was localized to this area (see Photos 2 and 3). The number two main bearing on one side was broken due to low-cycle pounding stresses. The number two piston had been making contact over time with its cylinder head and valves.

Photo 2. Heat and contact damage to connecting rod 2
Photo 2. Heat and contact damage to connecting rod 2

Photo 3. Crankshaft showing heated area
Photo 3. Crankshaft showing heated area

Analysis

Examination of the airframe wreckage and its components found no indication of any mechanical malfunction that may have initiated or contributed to the accident.

Weather was also not considered to be a factor, though the darkness in the valley floor may have contributed to the pilot’s inability to find a better location to conduct the forced landing. The wreckage trail and the evidence of impact forces indicate that the aircraft crashed in a stalled flight condition.

An unapproved, shop-made washer that had been installed in the alternator drive coupling contributed to a quantity of rubber debris entering the engine. The presence of the washer in the coupling also caused the rubber disc to contact the alternator spur gear on the crankshaft, causing more debris to enter the sump. This debris then restricted oil flow in the failed area of the engine. The industry standard check of oil systems when contamination is found or known to exist was not carried out. The company performing the maintenance did not benefit from guidance developed in Service Bulletin M84-5 as the bulletin did not include this series of engine even though it had a gear-driven alternator.

It is highly probable that the engine failure was initiated by a partial blockage of oil flow, caused by debris in the oil, to the number two connecting rod journal and bearing. This resulted in a progressive loosening of the clearances at that location, which allowed a gradual increase in piston stroke and increasing contact between the piston and the cylinder/valves. This looseness caused repeated reaction forces on the number two main bearing, pounding it until fatigue cracking broke up the left bearing shell. The connecting rod journal continued to overheat and elongate the bearing area until the lower connecting rod end cap nut came apart. Total engine failure and seizure then occurred.

The engine knocking that occurred during the summer prior to the accident was not noted in the journey log book nor mentioned to maintenance personnel. Early detection of the loosening and overheating parts might have prompted preventative maintenance.

Findings as to causes and contributing factors

  1. An unapproved part was installed in the alternator coupling. This resulted in debris from the coupling causing a partial blockage of oil flow to the number two connecting rod bearing. This low oil flow caused overheating and failure of the bearings, connecting rod cap bolts and nuts, and the subsequent engine failure.
  2. The engine failure occurred after sunset and the low-lighting conditions in the valley would have made selecting a suitable landing area difficult.
  3. The engine knocking was not reported to maintenance personnel which prevented an opportunity to discover the deteriorating engine condition.

Finding as to risk

  1. All flights on the day of the accident were carried out without the oil filler cap in place. The absence of the oil filler cap could have resulted in the loss of engine oil.

Other findings

  1. There were no current instrument flight rules charts or approach plates on board the aircraft for the intended flight.
  2. The Teledyne Continental Motors Service Bulletin M84-5 addressed only the 520 series engines and did not include other gear-driven alternator equipped engines.

Safety action taken

Teledyne Continental Motors states that it will update Service Bulletin M84-5 to include the 550 series engines. The Teledyne Continental Motors Instructions for Continued Airworthiness will also be updated to reflect the content of Service Bulletin M84-5 as periodic updates to that document are performed.

TSB Final Report A08C0124—
Fuel Starvation/Forced Landing

On June 13, 2008, a Cessna 337D was returning to Buffalo Narrows, Sask., from Stony Rapids, Sask., after having dropped off a passenger. Approximately 14 mi. northeast of the airport, the pilot declared an emergency due to a double engine power loss. The pilot completed a forced landing in a swampy area on the east shore of Churchill Lake, Sask. The aircraft was substantially damaged. The pilot was transported to hospital in Île-à-la-Crosse, Sask., and was subsequently released with minor injuries. The accident occurred at about 11:40 Central Standard Time (CST).

Fuel Starvation/Forced Landing

Analysis

The pilot did not use any auxiliary tank fuel prior to completely exhausting the fuel in the main tanks. This procedure prevented a successful restart of either engine and is contrary to the procedures outlined in the C-337D owner’s manual. The pilot’s departure from the specified procedures and incorrect fuel estimate indicated that he did not fully understand the aircraft’s fuel system.

The fuel selectors are located overhead on the cockpit ceiling, which requires that the pilot divert his attention from monitoring primary flight information when changing fuel selections. In a high-workload situation such as dealing with a dual-engine power loss, this cockpit configuration could complicate the management of the fuel system. The overhead location and tandem layout of the fuel selectors, along with the nomenclature of the system, which includes “aux” pumps that do not pump fuel from the “aux” tanks, can make operation of the C-337’s fuel system confusing to pilots who are not totally familiar with its operation.

The higher rate of fuel consumption for the training flight as compared to cruise power fuel consumption contributed to the exhaustion of the fuel remaining in the main tanks.

Findings as to causes and contributing factors

  1. The pilot’s estimates of the fuel remaining in the main tanks and the amount required to complete a round trip to Stony Rapids were inaccurate. Consequently, both engines stopped operating when the fuel in the aircraft’s main tanks was depleted.
  2. The pilot did not have a full understanding of the aircraft’s fuel system and was unaware of the method and sequence for accessing the fuel in the aircraft’s auxiliary fuel tanks. As a result, the pilot’s operation of the fuel system rendered the fuel in the auxiliary tanks unusable after the fuel in the main tanks was depleted and the engines could not be restarted.
  3. The operator’s training program for the C-337D did not establish or test the pilot’s knowledge on how to operate the C-337D’s fuel system.

Finding as to risk

  1. The design and nomenclature of the C-337D fuel system complicates its operation during periods of high cockpit workload, thus increasing the risk of confusion.

Safety action taken

The operator has added questions to the C-337D training exam which test for knowledge of the operation of the fuel selectors, fuel management, and the auxiliary boost pumps.

TSB Final Report A09P0156—Engine Power
Loss-Forced Landing

On June 12, 2009, an amateur-built Glastar was on a recreational flight from Yellowknife, N.W.T., to Kelowna, B.C., with two pilots on board. At approximately 14:01 Pacific Daylight Time (PDT), shortly after passing Chetwynd, B.C., a severe powerplant vibration and loss of power was experienced. The engine power was reduced to 1 000 RPM and a forced landing into a field was attempted. On short final, the aircraft struck a power line and veered off course to the right, where it struck trees and rising terrain. The pilot in the left seat received non-life threatening injuries. The pilot in the right seat was fatally injured. There was no emergency locator transmitter (ELT) signal and no fire. The switch on the ELT was found in the OFF position.

Examination of the wreckage revealed that the No. 2 cylinder head had separated from the base (see Photo 1) and the crankshaft was severed at the propeller flange. The engine had 212 hr total time since new (TTSN) when the failure occurred.

Photo 1: No. 2 cylinder head
Photo 1: No. 2 cylinder head

The engine crankshaft sheared at the propeller hub and the propeller was found embedded in a tree at the accident site (see Photo 2). The propeller was nicked and scuffed in a manner consistent with striking the severed power line.

Photo 2: Propeller from the Glastar
Photo 2: Propeller from the Glastar

The Aero Sport Power O-360-A2A engine is assembled by Aero Sport Power based in Kamloops, B.C. These engines are built using parts purchased from various suppliers who hold parts manufacturing authority (PMA) granted by the U.S. Federal Aviation Administration (FAA). The Aero Sport Power O-360-A2A can be described as a non-certificated clone of the Avco Lycoming O-360-A2A engine, which holds type certificate number E-286 issued by the FAA.

PMA parts may be sold with certification for use on type-certificated engines. The Aero Sport Power O-360-A2A engine is sold to amateur builders as an experimental engine, and no certification is required for this category by Transport Canada.

Aero Sport Power used cylinders manufactured by Engine Components Inc. (ECi). At the time of the build-up, Aero Sport Power installed pistons that increased the engine compression ratio from 8.5:1 to 9.2:1.

The failed cylinder assembly was sent to the TSB laboratory for examination. A fracture surface analysis of the No. 2 cylinder and sheared crankshaft was completed. The crankshaft propeller flange showed signs of minor hydrogen embrittlement. The sheared crankshaft was determined to be from overload fracturing due to impact. The minor hydrogen embrittlement was not causal to the accident.

In the fall of 2008, the FAA issued Airworthiness Directive (AD) 2008-19-05 regarding ECi cylinder assemblies installed on the Lycoming engines. The AD addressed a manufacturing defect that caused the cylinder head to separate from the base. It did not, however, address engines with increased compression ratios. Engines that are not type-certificated, such as the Aero Sport Power engines, were not mentioned in the AD. The cylinder head failure fracture surfaces were typical of fatigue failures addressed by this AD.

Prior to the issuance of AD 2008-19-05, ECi issued Mandatory Service
Bulletin (MSB) 08-1. This MSB called for the inspection and replacement of the faulty cylinders by 350 hr total time in service.

On April 29, 2009, Aero Sport Power advised the owners of the occurrence aircraft by
e-mail that three of their cylinders were affected by the ECi MSB. The surviving owner/pilot was not aware of AD 2008-19-05 or the MSB regarding the faulty cylinders. He also was not involved with the building of the aircraft. The deceased owner/pilot had built the aircraft. Aircraft records do not indicate compliance with the AD or MSB. However, records do show that differential pressure checks, required by the AD, were carried out. None of the differential pressure checks resulted in values that would have required further inspection and action in accordance with the AD. These checks were carried out 22 hr prior to the failure, during the last annual inspection and service.

Analysis

The failure of the cylinder head occurred at 212 TTSN, well in advance of the 350-hr limit set out in the AD. Compression checks completed 22 hr prior to the accident failed to detect a problem. This is not uncommon because compression tests will not necessarily detect an impending failure. It is possible that the premature failure of the cylinder head was due to the increased compression ratio of the engine.

The surviving owner/pilot was unaware that there was a critical AD that affected their engine. The deceased partner was notified by e-mail of the MSB, which was referenced in the AD, and he performed the required differential checks. There was plenty of time remaining before reaching the 350-hr limit on the cylinder head.

Transport Canada does not issue ADs for non-type-certificated aircraft, propellers, engines, and equipment; nor do they notify the owners of these aircraft of ADs that could adversely affect them.

The number of amateur-built, owner-maintained, and ultralight aircraft is growing. The onus is on the owners of these aircraft to ensure airworthiness. Without additional system safeguards, there is a greater risk that these aircraft will not be properly built and maintained.

These aircraft often operate in the vicinity of populated areas, thereby increasing the risk to the public and to property.

Findings as to causes and contributing factors

  1. The failure of the No. 2 cylinder caused the engine to lose power.
  2. During the attempted forced landing, the aircraft struck power lines, control was lost, and the aircraft collided with trees and terrain.

Engine Power Loss-Forced Landing

Finding as to risk

  1. Non-type-certificated aircraft owners are not advised of, nor are the owners required to comply with, ADs that are potentially critical to aviation safety. This greatly increases the risk that important airworthiness issues may go unaddressed by the amateur-built community.

Other findings

  1. The population of amateur-built, owner-maintained, and ultralight aircraft is growing. This increases the risk to the public and to property if they are not properly designed, produced, and maintained.
  2. Evidence of minor hydrogen embrittlement was found in the crankshaft. While not causal to the accident, it increases the risk of material failure over time.

Safety action taken

Aero Sport Power
Aero Sport Power has notified all engine owners potentially affected by AD 2008-19-05 and those who have increased the compression ratio.

Recreational Aircraft Association of Canada
The Recreational Aircraft Association of Canada issued a notification to its members regarding AD 2008-19-05 and the possible effect of the increased compression ratio. This notification also reminded members how to search for ADs by aircraft registration using the Transport Canada Continuing Airworthiness Web Information System (CAWIS).

Danbury Aerospace
Danbury Aerospace, the parent company of Engine Components Inc. (ECi), has elected to limit the compression ratio of cylinders sold in engine kits. 

Coming soon in
ASL 4/2011!

Update on the newly-formed
Floatplane Operators Association of British Columbia!

In the meantime, check them out at:
www.floatplaneoperators.org.

Date modified: