Forced/precautionary landing

Cessna A185F, VH-SLC

Significant Factors

  1. The pilot made an inappropriate power reduction before terrain clearance was assured.
  2. Turbulence and downdrafts in the lee of the headland significantly degraded the aircraft's climb performance.



 

Analysis

The combination of near maximum take-off weight, and the reduction of engine power to 81% soon after take-off, meant that the aircraft had marginal climb performance when it encountered the turbulence and associated downdrafts. By not using the full take-off distance available the pilot placed the aircraft on a climb profile that reduced terrain clearance and increased the risk of exposure to strong downdrafts.

The pilot's judgement may have been influenced by previous flights where different wind directions and lower wind strengths combined to give more favourable take-off conditions. Additionally, the detrimental effect of an early power reduction would not have been as perceptible on training flights conducted at lower aircraft weights where the aircraft's climb performance would have been far greater. Although turbulence in the lee of the headland may have been present on previous occasions, the pilot had not encountered any significant downdraft activity. Consequently, he was unprepared for conditions of that severity.

Although the pilot turned left to avoid the elevated terrain when the aircraft descended in the turbulence, it was likely that the flight path placed the aircraft into even stronger downdraft activity. Without the immediate application of a higher power setting, the aircraft did not have sufficient performance margin to continue the climb or maintain altitude.

From the point where the pilot attempted to land the aircraft there was insufficient water distance remaining on which to land and stop the aircraft normally. The aircraft contacted the beach at a speed that was fast enough, when combined with the high centre of gravity of the aircraft type, to cause it to overturn.

Summary

The pilot of the Cessna 185 Floatplane, with five passengers on board, was making a water departure for a charter flight. The pilot positioned the floatplane for a take-off into a north easterly wind of 15 kts that was gusting to over 20 kts. The take-off path was over a sand spit, approximately 50 ft above the water level. To the north, and left of the take-off path, was a steep, rocky headland that rose to a height of approximately 300 ft above mean sea level.

The pilot reported that he had selected 20 degrees of flap and applied maximum power for take-off. The aircraft became airborne after a short run and the pilot climbed it at an indicated airspeed (IAS) of 70 kts. At about 200 ft the pilot reduced engine power to 25 inches of manifold pressure and 2,500 RPM. The pilot reported that just after he reduced power, the aircraft encountered turbulence and started to descend rapidly. He turned the aircraft left, away from the spit, with the intention of regaining altitude over the water before he attempted to cross the spit. However, the aircraft continued to descend, and the pilot decided to land straight ahead. The aircraft contacted the water and bounced, then ran aground on the beach and overturned.

The pilot reported that he exited through a window and instructed the passengers to evacuate quickly, as there was a possibility of fire. The passengers reported that they were entangled in their seat belts and had difficulty releasing the buckles. A small child was being held by a passenger and another passenger was temporarily restrained by clothing that became caught on the right control yoke.

The load chart for the flight showed that the aircraft was 31 kg below its maximum take-off weight. The pilot commenced the take-off with a take-off distance of approximately 1,100 m, which exceeded the minimum take-off distance of 1,000 m stipulated by the aircraft's flight manual. However, this take-off distance was less than the 1,300 m pilots were directed to use by the operator's Authorised Landing Area (ALA) register. The pilot reported that he had not used the full length available as previous take-offs that day, from the same point in lighter winds, had been uneventful. He considered that the increased headwind component would have improved the take-off performance and climb gradient of the aircraft.

The ALA register also stated that a north-easterly wind required a climb over the spit to avoid turbulence in the lee of the adjacent headland. Another warning in the operator's ALA survey report cautioned "Dumping will be encountered on the lee side of the headland especially in the north easterly winds".

A fact sheet on mountain wave turbulence that accompanied a recent ATSB report (Occurrence 200104092) involving mechanical turbulence stated, in part:

"Flowing air near the ground is forced up the windward side of any elevated barrier and then sinks down the leeward side. Air flowing at speeds greater than 20 kts produces seriously turbulent air and significant downdrafts on the leeward side."

That situation was referred to as "dumping" in the operator's ALA survey sheet. The pilot reported that he hadn't encountered severe "dumping" during any previous take-offs.

The fact sheet also stated, in part:

"In addition to generating turbulence that has demonstrated sufficient ferocity to significantly damage aircraft or lead to loss of aircraft control, the more prevailing danger to aircraft in the lower levels in Australia seems to be the effect on a aircraft's climb rate. General aviation aircraft rarely have performance capability sufficient to enable the pilot to overcome the effects of a severe downdraft generated by a mountain wave, or the turbulence or windshear generated by a rotor."

The Cessna 185 Pilot's Operating Handbook (POH) procedures for both normal take-off and short field take-off recommend that once clear of any obstacles, the pilot retract the wing flaps and select full throttle and 2,700 RPM. Operations manual data produced by the operator listed the climb power setting as 25 inches manifold pressure and 2,700 RPM, with a footnote that the information be used as a guide only and that the user refer to the POH and Flight Manual. The aircraft's flight manual did not provide guidance on take-off procedure or associated power settings. The climb power setting of 25 inches manifold pressure and 2,500 RPM, selected by the pilot when the aircraft reached approximately 200 ft, delivered only 81 per cent of the available power.

The pilot reported that he had been encouraged by the operator to reduce power as soon as possible after take-off as a noise reduction technique. The Chief Pilot stated that a power reduction early in the climb was demonstrated during training to reduce the noise impact and to reduce engine wear. The Chief Pilot also stated that, during training it was emphasised that power reductions should only be made when clear of obstacles and when terrain had been cleared. It was also stressed that when required, full power should be used, at the pilot's discretion.

Occurrence summary

Investigation number 200105926
Occurrence date 23/12/2001
Location Palm Beach, (ALA)
State New South Wales
Report release date 17/07/2002
Report status Final
Investigation type Occurrence Investigation
Investigation status Completed
Mode of transport Aviation
Aviation occurrence category Forced/precautionary landing
Occurrence class Accident
Highest injury level Minor

Aircraft details

Manufacturer Cessna Aircraft Company
Model 185
Registration VH-SLC
Serial number 18503231
Sector Piston
Operation type Charter
Departure point Palm Beach, NSW
Destination Rose Bay, NSW
Damage Substantial

Schweizer Aircraft 269CB, VH-AHV, on 2 November 2001

Analysis

The displaced circlip from the upper end on the accelerator pump plunger, discovered during the technical disassembly of the carburettor, would have resulted in insufficient fuel being available to the engine during rapid high power requirements.

The engine logbook documentation did not reveal any maintenance performed to the carburettor since it's last repair and reinstallation. It was unlikely that the circlip had become displaced during the repairs of the carburettor, as bench testing would have revealed the problem.

If the maintenance action of the reversing of the linkage plate had been carried out without removing the top plate of the carburettor, it would have been difficult to verify the continued connectivity of the accelerator linkage and the proper functioning of the accelerator pump mechanism within the carburettor.

The investigation could not conclusively establish the reason for the disconnection of the accelerator pump plunger mechanism.

Summary

The pilot of the Schweitzer 269CB helicopter reported that during cruise on a repositioning ferry flight, the helicopter's engine misfired and then stopped completely so he completed a power off autorotation to a nearby dam wall. The pilot, the sole occupant, received no injuries and the helicopter was not damaged.

Troubleshooting by maintenance personnel isolated the problem to the engine's carburettor. The carburettor was replaced and the engine test run. No other anomalies were noted and the helicopter was flown back to the maintenance base.

A technical disassembly of the Precision Airmotive carburettor, model HA-6, part number 10-6030, serial number 75060303 was initiated. The examination included fuel flow level checks. During the checks, the throttle linkage was exercised through its full range from idle to full power. Throttle operation was smooth and unhindered. Several rapid accelerations were then carried out to observe the operation of the accelerator pump. No fuel was observed to flow from the accelerator pump discharge tube, positioned in the carburettor venturi, during these tests. Further examination revealed that the accelerator pump plunger mechanism was disconnected at the upper circlip attachment point on the plunger.

In addition, the accelerator pump/economiser linkages were found to have severe wear on the cam lobe face corresponding to the full throttle position and on the linkage plate that the camshaft lobe acted upon. The linkage plate also had severe wear on the cam lobe-mating surface and on the opposite non-cam lobe contact surface. That indicated that the plate had been reversed at some point during its operational life.

Carburettor background

At the time of the occurrence, the carburettor had accumulated 1,668.2 hours time in service (TIS). The carburettor had been removed from the engine at 1,154.7 hours TIS for a discrepancy reported as suspected low engine power. The repair documentation noted an air-metering pin adjustment defect. Following readjustment, the carburettor was bench tested and returned to service.

At 794.4 hours TIS, the carburettor had been removed from the engine, for a discrepancy reported as leaking fuel when in the full lean/cut-off position. The repair documentation noted that a new float needle and seat were fitted. The carburettor was bench tested and returned to service.

Occurrence summary

Investigation number 200105273
Occurrence date 02/11/2001
Location 9 km N Mareeba Aero.
State Queensland
Report release date 04/02/2002
Report status Final
Investigation type Occurrence Investigation
Investigation status Completed
Mode of transport Aviation
Aviation occurrence category Forced/precautionary landing
Occurrence class Incident
Highest injury level None

Aircraft details

Manufacturer Schweizer Aircraft Corp
Model 269
Registration VH-AHV
Serial number 0063
Sector Helicopter
Operation type Aerial Work
Departure point Mareeba, QLD
Destination Mareeba, QLD
Damage Nil

Bell 206B(III), VH-NVH

Safety Action

Because of the deficiency with the manufacturer's maintenance requirements, identified during the investigation, the following safety recommendations have been issued.

R20000189

The Australian Transport safety Bureau recommends that Bell Helicopter Textron P/L revise the maintenance manual for the Bell 206B III series helicopter to require the inspection and lubrication of the hydraulic pump drive splines on a calendar basis, in addition to the hourly time in service inspection and lubrication requirements.

R20000190

The Australian Transport Safety Bureau recommends that the Civil Aviation Safety Authority advise Australian operators of Bell 206B III series helicopters of the finding of this accident and revise the calendar requirement for the lubrication of the hydraulic pump splines.

R20000191

The Australian Transport Safety Bureau recommends that the Federal Aviation Administration note the findings of this accident. It also recommends that the Federal Aviation administration alert all operators of bell 206B III series helicopters of the deficiency in the maintenance manual.

Summary

The pilot of the Bell Jet Ranger reported that while cruising at about 2,500 ft AGL, the rotor RPM indication decreased to zero. He initiated auto-rotation immediately by lowering the collective pitch. The controls were very stiff, however all engine parameters were normal. The helicopter descended at higher than normal speed and landed heavily. The right skid crosstube support failed and the aircraft rolled onto its side. Both occupants vacated the wreckage with minor injuries.

The pilot reported that he did not hear the low rotor alarm and did not notice the low rotor RPM caution light illuminate.

Investigation determined that the drive spline and coupling to the hydraulic pump and rotor tachometer generator were devoid of lubrication and worn to the degree that drive to both units had failed. This resulted in the loss of hydraulic power to the controls and the loss of the main rotor speed indication.

The low rotor RPM sensor which provides a signal to activate the low rotor alarm and the low rotor RPM caution light was tested and functioned normally. However, it is likely that the pilot did not notice the low rotor RPM warnings, due to his immediate initiation of an auto-rotation, and the ambient light conditions. The warning horn is cancelled by a cut-out switch on the collective pitch handle when the handle is lowered as in the case of an auto-rotation. As the helicopter was tracking south at the time, the bright sun from the northeast would have made illumination of the low rotor RPM caution light difficult to distinguish.

Investigation of the maintenance requirements for the helicopter found that, prior to January 1998, the hydraulic pump drive splines were required to be lubricated by hand every 1200 hours or twelve months, whichever came first, or under extreme operating conditions, every 300 hours. In January 1998 the manufacturer issued a revised manual with a changed requirement for the Bell 206B III model to lubricate the hydraulic pump drive splines every 300 hours with no calendar requirement.

The accident helicopter had only accumulated about 90 hours time in service since January 1998 and 217 hours since the initial issue of its Australian certificate of airworthiness in 1990.

Occurrence summary

Investigation number 200002899
Occurrence date 12/07/2000
Location 11 km S Aberdeen
State New South Wales
Report release date 13/06/2001
Report status Final
Investigation type Occurrence Investigation
Investigation status Completed
Mode of transport Aviation
Aviation occurrence category Forced/precautionary landing
Occurrence class Accident
Highest injury level Minor

Aircraft details

Manufacturer Bell Helicopter Co
Model 206
Registration VH-NVH
Serial number 1163
Sector Helicopter
Operation type Private
Departure point Tamworth NSW
Destination Bankstown NSW
Damage Destroyed

Beech Aircraft Corp 58, VH-NTG

Summary

The Beechcraft Baron aircraft was being operated on a freight charter flight from Groote Eylandt to Darwin.

The pilot said that as the aircraft descended through 5,000 ft the left engine fuel flow gauge reading decreased to zero and the left tachometer indication reduced to about 1,700 RPM. The cylinder head temperature and oil temperature readings also reduced. The pilot advanced the engine control levers to full power and tried to fly the aircraft at the best single engine rate of climb speed. The aircraft continued to lose altitude and the pilot realised there was not enough height remaining to reach an aerodrome. As the left engine tachometer was still indicating about 1,700 RPM, the pilot, believing the engine was still developing power, did not feather the left propeller.

The pilot landed on a two-lane highway. The aircraft was substantially damaged during the landing roll when it slid off the roadway and entered a ditch. The pilot, who was the sole occupant, was not injured.

The pilot had about 1450 hours aeronautical experience and 105 hours on the aircraft type.

Examination of the left engine found the fuel mixture control cable had failed near the fuel control unit control lever. This allowed the lever to move downward under its own weight into the fuel cut-off position. No other pre-existing damage or fault was found that may have contributed to the failure of the engine.

Both engines were only slightly damaged in the forced landing. They were fitted to an engine test cell where they operated in accordance with the manufacturer's standards.

The aircraft's maintenance records showed that the fuel mixture control cable was fitted to the aircraft about 230 hours previously. Specialist metallurgical examination of the cable, which was constructed of one central strand surrounded by six smaller diameter strands, showed it had broken because of fatigue cracking. The fatigue cracking showed significant alternating stresses had been applied to the cable during aircraft operation. The examination also found sliding contact wear close to the break, caused by abnormal alternating loads applied to the cable at the point where it was swaged to the fitting. Possible misalignment of the cable during installation may have led to the failure.

The pilot did not realise that the left engine had failed and the unfeathered propeller was driving the engine, because the RPM indication was higher than he had expected. He therefore was not aware that the unfeathered propeller was causing excessive drag preventing the aircraft from maintaining height.

Occurrence summary

Investigation number 200000624
Occurrence date 18/02/2000
Location Humpty Doo, 37 km SE Darwin, Aero.
State Northern Territory
Report release date 22/12/2000
Report status Final
Investigation type Occurrence Investigation
Investigation status Completed
Mode of transport Aviation
Aviation occurrence category Forced/precautionary landing
Occurrence class Accident
Highest injury level None

Aircraft details

Manufacturer Beech Aircraft Corp
Model 58
Registration VH-NTG
Sector Piston
Operation type Charter
Departure point Groote Eylandt, NT
Destination Darwin, NT
Damage Substantial

Robinson R22 Beta, VH-NLT

Safety Action

As a result of this occurrence, the Bureau of Air Safety Investigation is investigating perceived safety deficiencies involving helicopter performance, pilot licensing and experience requirements.

Any recommendation issued as a result of this deficiency analysis will be published in the Bureau's Quarterly Safety Deficiency Report.

Significant Factors

  1. The pilot was inexperienced.
  2. The density altitude and relative humidity were high.
  3. The wind conditions were gusty.
  4. There was no documented data or guidance available to the pilot to assist him in assessing the expected performance of the helicopter during the take-off.
  5. The pilot did not adequately plan the take-off to account for the weather conditions and helicopter landing site characteristics.
  6. The pilot used an inappropriate take-off technique.
  7. The helicopter was probably at or close to maximum all up weight and had inadequate performance to complete the take-off in nil wind.
  8. The weather conditions had changed since the first take-off and did not assist the helicopter during the second take-off.

Analysis

The helicopter weighed close to its maximum permissible all up weight. The pilot reported that the power required to hover the helicopter in ground effect was about 23 inches manifold air pressure, which was close to the placarded limit. The flight manual data also indicated that the helicopter had insufficient power to hover out of ground effect in nil wind. The reported high power being used to achieve an in-ground effect hover associated with the high-density altitude conditions and the available flight manual performance data, indicated that the helicopter's engine was unlikely to have been capable of providing appreciably more power than that already being used.

The reported site information indicated that the helicopter would have needed to achieve a climb profile of at least three degrees. However, the available performance data indicated that there was unlikely to have been sufficient power available to depart from the in-ground effect hover and achieve the required climb profile in nil wind. The helicopter's performance would have been adversely affected by the high relative humidity, which had the effect of further increasing the already high-density altitude. Had the pilot consulted the flight manual or conducted a power check prior to landing on the pad, he may have realised that there was unlikely to have been sufficient power available to attempt the departure, or at least he could have planned the departure with a rejected take-off in mind.

Neither the helicopter's flight manual, nor operator's operations manual provided climb performance data or guidance on expected power requirements for the helicopter to depart the hover and transition into forward climbing flight. Although the Civil Aviation Orders specify minimum performance requirements for aeroplanes, there are no similar requirements for helicopters. With no regulatory requirement for climb performance information to be provided by either the helicopter's manufacturer or the operator, the pilot had no documented guidance on the helicopter's expected performance in forward climbing flight. While it would have been prudent for the pilot to conduct a power check before commencing operations into or out of the helicopter landing site, there was no guidance provided by either the operator or helicopter manufacturer on required power margins. Therefore, a power check may have been of relevance to a pilot with experience in R22 limited power operations. However, the accident pilot's experience was not extensive, with a total flying time of about 173 hours. While he had accumulated 36 flying hours during the month prior to the accident he had only flown about 50 hours during the previous 16 months. It appears incongruous that, in accordance with the regulations and orders, the pilot had sufficient experience to transport passengers but insufficient flying experience to undertake mustering operations. While there was no documentary evidence detailing the pilot's initial training, there appeared to be some gaps in his knowledge with regard to operating the R22 helicopter close to its limits. The limited amount of flying he conducted in the 15 months between completing his course and joining the helicopter's operator may have also been a factor. He appeared to have been ill prepared to operate the helicopter in the high-density altitude conditions of Northern Australia.

Although the task of carrying passengers from the station was not authorised by the helicopter's operator, the pilot reported that he was advised that the on-site experienced pilot was "in charge". The helicopter's operator was unaware of the requirement to conduct passenger flights, and in using the term "in charge" in relation to the on-site pilot, probably contributed to the pilot's belief that the on-site pilot was a nominated person who could authorise all flights. If the operator had been aware of the passenger flights, then more appropriate supervision of the pilot may have been provided. There appeared to have been a misunderstanding between the operator and the on-site pilot as regards to the use of the helicopter.

The pilot reported that he had not experienced any problems during the first take-off from the site. He reported the wind as being gusty, averaging 15 kts and that there were thunderstorms in the area. With evidence indicating that there was probably little power available to depart the hover and achieve the required departure angle in nil wind, it was likely that the helicopter's performance was assisted during the previous take-off by the wind conditions existing at the time. This may have lulled the pilot into a belief that the helicopter had sufficient power to attempt further take-offs.

During the accident take-off, the helicopter's main rotor RPM decreased. No evidence of a mechanical fault was subsequently found. The pilot reported that the wind was gusty and the humidity was high and increasing as storms built-up in the area. It was possible that the relative humidity had increased sufficiently to decrease the main and tail rotor performance or decrease the amount of power available from the engine. Alternatively, or in combination with the high relative humidity, the wind may have changed direction or lost strength during the accident take-off. In either case, the pilot would have had to increase the collective pitch to maintain the departure profile. This may have resulted in the power required to drive the helicopter's rotor system exceeding the power available from the engine. As a result, the main rotor RPM drooped which reduced the lift being produced by the main rotor and the helicopter descended to the ground. Because the pilot had conducted the departure without considering the possible requirement to reject the take-off, the area into which the helicopter descended was unsuitable for landing.

CONCLUSIONS

Findings

While the pilot was conducting the take-off from Fossil Downs Station, the environmental conditions may have changed such that the power required to maintain the helicopter's departure profile exceeded the power available from the engine. As a result, the main rotor RPM decayed and the helicopter descended onto the ground. The pilot had not adequately assessed the power needed to conduct the takeoff and had used an inappropriate takeoff technique for the environmental conditions and helicopter weight. There was a misunderstanding between the operator and the customer as to the use of the helicopter. The pilot inadvertently believed he was authorised in accordance with the company operations manual to conduct the passenger flights. The helicopter's operator was unaware that the passenger flights were being conducted.

Summary

The Robinson R22 helicopter was sent to Fossil Downs Station by the operator to conduct a small mustering assignment. As the operator's pilot was relatively inexperienced and not qualified to conduct mustering operations, the helicopter was fitted with dual controls so that an experienced and qualified pilot who was on-site could conduct the flying while the operator's pilot occupied the other seat. However, after the aircraft arrived at the station, the on-site pilot requested that the operator's pilot transport two passengers from the station to Fitzroy Crossing. Although the operator's pilot had insufficient hours to conduct mustering, he held a commercial pilot's licence and was qualified to carry passengers.

The pilot transported the first passenger from the helicopter landing site without incident. However, during the second departure, at about 15 ft and just as the helicopter was achieving translational lift, it sank back towards the ground. When the pilot increased the collective pitch in an attempt to regain the required departure profile, the low rotor RPM warning horn sounded and the rate of descent increased. The pilot reported that he checked that the throttle was fully open but the main rotor RPM continued to decay. The helicopter landed heavily and the main rotor blades clipped a tree. The pilot reported that as soon as it landed, he shut the engine down. The helicopter was extensively damaged but neither occupant was injured. The pilot reported that he flew the second flight's take-off into wind along a similar path to that flown during the previous passenger flight.

The maintenance organisation that repaired the helicopter reported that no mechanical fault was found that would have contributed to the accident. The accident was not subject to an on-site investigation by the Bureau of Air Safety Investigation.

Weather conditions

The pilot reported that the ambient temperature was about 37 degrees Celsius. He also reported that the humidity was high and increasing as storms were developing in the area. The wind was averaging about 15 kts from the south-east and gusting. The density altitude at the site, without factoring the relative humidity, was calculated to be about 3,000 ft.

The helicopter landing site

The pilot reported that he was using a southerly departure from the site to align with the general wind direction. The pilot reported that in the southerly direction, the helicopter landing site had an available length of about 50 to 60 m from the departure point with about 3 m high bushes at the departure end. There was also a small fence running east to west about 40 m from the departure point.

Helicopter performance

The helicopter weighed close to its maximum all up weight of 622 kgs. The helicopter's flight manual indicated that the calculated maximum weight to hover out of ground effect in nil wind, was about 605 kgs. The pilot did not consult the helicopter's flight manual for likely power requirements and power availability; nor did he conduct a power check prior to arriving at the helicopter landing site in order to ascertain the actual power available. While the Robinson R22 Flight Manual provided hover performance data, it did not contain performance data related to the expected climb performance of the helicopter during take-off, or in forward flight. The operator's operations manual did not provide guidance regarding power margins. There was no requirement under the existing regulations for information or guidance related to required power margins for departure or climb performance to be provided by either the helicopter's manufacturer or operator.

The pilot reported that the power setting required to hover the helicopter in ground effect was about 23 inches manifold air pressure, which was approximately the placarded limit manifold air pressure of 23.5 inches. The pilot could not recall the power indication during the accident take-off.

Ambient wind conditions can have significant and differing effects on a tail rotor equipped helicopter's performance. Engine power is delivered to a transmission system, which drives the main and tail rotors The power required to drive the transmission system is determined primarily by the amount of drag being produced by the rotors and the power available is determined by the power output of the engine(s). The difference between the power available and power required is known as the power margin. If the power required to drive transmission exceeds the amount of power available from the engine, then the main and tail rotor speed will decay, or droop. When the speed of the main rotor droops significantly, the main rotor loses lift and the helicopter descends. Wind blowing over a main rotor provides translational lift that can significantly reduce the power required to drive the helicopter's transmission system. Wind may also assist a helicopter to maintain heading, which also reduces the load demand on the transmission and therefore reduces the power required to drive the transmission. Conversely, a wind from an adverse direction may increase the load demand on the transmission and, in turn, the power required from the engine. Therefore, the wind may cause a net effect which, depending on its strength and direction, will reduce or increase the power required for a tail rotor equipped helicopter to maintain flight.

The density of air is affected by a number of factors including its moisture content. Relative humidity is the ratio of the amount of moisture in the air to the amount it is capable of absorbing at a given temperature. The greatest decrease in air density (increase in density altitude) due to moisture content will be at a high temperature. In general, as the density altitude increases, helicopter rotor and piston engine performance decrease. The performance data provided in the R22 helicopter's flight manual is only valid for nil-wind conditions and does not account for the adverse effects of high relative humidity.

Although the Civil Aviation Orders (CAOs) specify minimum performance requirements for single and multi-engine aeroplanes, there are no minimum performance criteria specified for helicopters.

Pilot Experience

The pilot had about 173 hours flying experience, of which about 83 hours were in command. All his flying experience had been gained on the Robinson R22 helicopter. Prior to joining the operator, he had flown about 14 hours during the 15 months since gaining his commercial (helicopter) pilot's licence in September 1997. He had been employed by the operator for less than one month and had accumulated just over 36 flying hours in that time. His initial training was conducted at several helicopter training schools in Queensland, and he reported that during the training, he had not experienced the helicopter being close to limits of power or practiced rejected departures and had not previously experienced main rotor RPM droop. He also reported that at the time of the accident, he was unaware of how to recover from a low rotor RPM condition. There were no available records related to his initial flying training, however, the pilot reported that he considered the training to be adequate. The operator conducted a proficiency check on the pilot about three weeks prior to the accident. The pilot's performance during the check was rated as satisfactory. About a month after the accident, the pilot attended a Robinson Safety Course where he flew with an experienced R22 helicopter instructor. The instructor reported that the pilot demonstrated an inappropriate take-off technique and that he required remedial instruction.

The Civil Aviation Orders impose a minimum requirement of 100 hrs as pilot-in-command before a pilot may conduct mustering operations. The Civil Aviation Regulations specify that a pilot may obtain a commercial (helicopter) pilot's licence, under certain circumstances, after a minimum total of 105 flying hours of which at least 35 hours are as pilot-in-command.

The company's operations manual required pilots to obtain authorisation from the Chief Pilot or a person nominated by the Chief Pilot before conducting any flights. The pilot reported that the Chief Pilot told him that the on-site pilot was "in charge". When he was requested to conduct the passenger flights, the pilot believed that the on-site pilot had the appropriate authority.

Occurrence summary

Investigation number 199900833
Occurrence date 03/02/1999
Location 9 km ENE Fossil Downs Station (ALA)
State Western Australia
Report release date 24/09/1999
Report status Final
Investigation type Occurrence Investigation
Investigation status Completed
Mode of transport Aviation
Aviation occurrence category Forced/precautionary landing
Occurrence class Accident
Highest injury level None

Aircraft details

Manufacturer Robinson Helicopter Co
Model R22
Registration VH-NLT
Serial number 2542
Sector Helicopter
Operation type Charter
Departure point Fossil Downs Station, WA
Destination Fitzroy Crossing, WA
Damage Substantial

Hughes Helicopters 369HS, VH-SHD

Summary

The Hughes 369HS helicopter was on a ferry flight from Western Australia to New South Wales. After landing at Ceduna to refuel, the pilot was advised by the local fuel agent that the Jet A1 fuel bowser was non-operable. The pilot reported that the operator's maintenance organisation advised him that the helicopter could be flown using avgas as an alternative fuel. The pilot reported that after the refuel, he completed a pre-flight inspection, started the helicopter and ran it for a short time before taking off. After take-off, he conducted a circuit to ensure that the engine was performing correctly. All the indications appeared normal so the pilot decided to depart the airfield. During the climb, at about 500 ft, the pilot turned off the fuel start pump switch. He reported that almost immediately after he turned the pump off, the engine failed. The pilot reported that he placed the helicopter into an auto-rotative descent, turned on the start pump and transmitted a Mayday that was acknowledged by flight service. The pilot also reported that although he noticed the engine-out warning system and the re-ignition systems operating during the descent, the engine did not relight. The helicopter's emergency locator beacon automatically operated following the firm auto-rotative landing. After landing, the pilot turned off the helicopter's fuel and electrical systems and vacated the cockpit. The helicopter was extensively damaged during the landing, but the pilot was not injured. The Bureau of Air Safety Investigation did not conduct an on-site investigation.

The version of the flight manual that was current when the accident occurred included a caution that stated, "When using alternate fuel mixtures or emergency fuels, the start pump should remain on until the engine is shutdown". Avgas has a higher vapour pressure than kerosene type fuel and is therefore susceptible to vapour lock in the vacuum type fuel system used in this helicopter. The pilot reported that the maintenance organisation did not advise him of the fuel pump requirement and that the caution did not appear in the flight manual available in the aircraft at the time. The helicopter's manufacturer reported that the amendment that included the caution relating to the use of the pump was issued in 1998. The operator was unable to provide a reason for the failure to incorporate the amendment relating to the use of alternative fuels in the helicopter's flight manual.

The investigation also found that the fuel filter element's outer fine mesh screen was substantially blocked with corrosion by-products from the stainless steel mesh filter element. Some of the screen pleats at the bottom of the filter element had cracked as a result of a corrosion-cracking mechanism. Analysis indicated that the corrosion of the fuel filter element was possibly caused by sulphur-bearing compounds found in Jet A1 fuel. It would be normally expected that such extensive blockage of the fuel filter would trigger the fuel filter caution light. However, the crack found in the filter element pleat may have been large enough to permit fuel to pass through and not create sufficient pressure differential within the fuel filter housing to trigger the caution light.

Following the refuel with avgas, the engine apparently lost power when the pilot turned off the fuel start pump. It is possible that when the fuel start pump was turned off, a vapour lock formed in the helicopter's fuel system, interrupting the fuel flow to the engine and causing the engine to flame out. The effect of the corroded and blocked fuel filter in contributing to the engine power loss was not determined.

Occurrence summary

Investigation number 199900820
Occurrence date 27/02/1999
Location 9 km E Ceduna, Aero
State South Australia
Report release date 19/10/1999
Report status Final
Investigation type Occurrence Investigation
Investigation status Completed
Mode of transport Aviation
Aviation occurrence category Forced/precautionary landing
Occurrence class Accident
Highest injury level None

Aircraft details

Manufacturer Hughes Helicopters
Model 369
Registration VH-SHD
Serial number 240564S
Sector Helicopter
Operation type Aerial Work
Departure point Ceduna, SA
Destination Port Augusta, SA
Damage Substantial

Piper PA-32R-300, VH-HUX

Factual Information

History of the flight

Two days later, on the day of the accident, the pilot again attempted to depart Moorabbin for Archerfield. The pilot stated that shortly after takeoff, as the aircraft climbed through approximately 1,600 ft, the engine appeared to cut out but then immediately operated normally again. Soon afterwards, the engine lost all power.

The pilot conducted a forced landing onto a stretch of road that was clear of vehicles. The aircraft was substantially damaged during the accident sequence when it collided with various obstacles, including powerlines, poles, a tree and a fence. The occupants of the aircraft were not injured, but were unable to open the doors. While exiting the aircraft through broken windows, both passengers sustained minor injuries.

Wreckage examination

Damage to the propeller indicated that the engine was rotating on impact, but not under power. A subsequent examination of the engine established that the magneto timing was between approximately 20 and 25 degrees after top dead centre. The correct timing for the engine was 25 degrees before top dead centre. The magneto gear was missing three teeth. The idler gear that drove the magneto gear and the engine-driven fuel pump, was no longer secured in either the accessory housing or the crankcase. At the accessory housing end, the mounting boss had cracked away from the housing and most of the fracture surfaces had worn smooth. At the crankcase end, the mounting bore had been substantially worn, and remnants of a bush remained in the bore. During normal operations, both idler gear mounts were subject to significant side loads from the engine-driven fuel pump drive cam.

Further examination of the idler gear shaft boss and bore established that the crankcase idler gear shaft bore had been repaired. This repair involved drilling out the bore and subsequent installation of a bush. The bush had not been secured to the crankcase bore and the metal used to manufacture the bush was commercially pure aluminium, a metal with low resistance to plastic deformation.

History of the aircraft

The aircraft was exported to Australia from the United States in early 1989. The Lycoming IO-540 engine installed in the aircraft at the time of the accident had been overhauled in 1988 in the USA, just prior to the aircraft being exported to Australia. According to the aircraft logbooks, in December 1989 the Australian company that imported the aircraft replaced the crankcase with an overhauled crankcase supplied by a USA crankcase repair company. There was no record of the reason it was replaced.

The USA company that supplied the overhauled crankcase had detailed its requirements for idler gear shaft bore repairs in its Federal Aviation Administration approved repair scheme. The repair involved drilling out the bore to a diameter of 0.813 inches, installing a bush to an interference fit and welding the bush to the crankcase. The repair scheme did not specify the material from which the bush should be manufactured.

The engine manufacturer issued a number of service instructions related to bushed repairs of idler gear shaft bores. The most recent of these, number 1417 issued 1 October 1982, required that the bore be drilled out to a diameter of 0.812 to 0.813 inches and that the outside diameter of the bush be machined to 0.814 to 0.815 inches. The instruction also stipulated that the bush be fixed to the crankcase by dowels and that the bush should be manufactured from AMS 4118 aluminium alloy. AMS 4118 referred to an alloy of aluminium and 3.5% magnesium. This alloy had a higher resistance to plastic deformation than commercially pure aluminium.

The crankcase idler gear shaft bore repair in the accident aircraft had not been conducted in accordance with either the Lycoming approved repair scheme or the US crankcase overhaul company repair scheme.

Maintenance

At the time of the accident, the engine had approximately 114 hours to run before it was due for overhaul. The aircraft was being maintained using the Civil Aviation Safety Authority Maintenance Schedule detailed in Schedule 5 to the Civil Aviation Regulations 1988. Section 2 (4)(c)(ii) of that schedule required that if a cartridge full-flow oil filter was fitted, that maintenance personnel should remove, open and inspect the filter at each periodic inspection. Because oil filters remove particulate contamination, such as metal, from engine oil, internal inspections of oil filters can provide an indication of the engine's condition.

The aircraft's engine was equipped with a cartridge full-flow oil filter, and periodic inspections were being conducted at 100 hourly intervals. The engine manufacturer recommended that the oil and oil filter be changed at 50 hourly intervals. The aircraft documentation indicated that the aircraft had been operating for approximately 83 hours since the last periodic inspection. There was no record that the oil and oil filter had been changed since the periodic inspection.

Personnel information

At the time of the accident, the pilot held a Private Pilot's Licence and had accumulated approximately 270 hours of flying experience. He had completed approximately 9 hours in the accident aircraft, his only experience in Piper PA32R-300 aircraft.

The pilot's training on the aircraft type involved two check flights, conducted a week before the accident. The aircraft engine ran roughly during taxi, however increasing the engine RPM and leaning the mixture cleared the problem. The pilot was told that the engine used a lot of oil and that the rough running and oil use related to the age of the engine.

Analysis

The investigation determined that an improper crankcase idler gear shaft bore repair resulted in increased vibration levels and excessive wearing of the accessory gears. These conditions led to various failures in the accessory drivetrain and the eventual failure of the engine. It was inappropriate to repair the crankcase idler gear shaft bore with a bush manufactured from a material with low resistance to plastic deformation, as the bush was subject to significant side loads.

It appeared that the personnel who operated the aircraft did not recognise that the problems they were experiencing were more than merely those of a worn engine. Had a 50-hourly oil filter inspection been carried out, as recommended by the manufacturer, it would have provided an opportunity for the problems in the accessory area to be identified prior to the engine failure.

Summary

The pilot, with two passengers, was conducting a trip in a Piper Lance from Archerfield to Moorabbin and return. During the engine run-up checks prior to departure from Archerfield, the pilot noted that the aircraft engine ran roughly, however increasing the engine RPM and leaning the mixture cleared the problem. During the flight to Moorabbin, he heard a slight miss in the engine note. On arrival at Moorabbin, the pilot noticed some oil on the outside of the engine cowl.

The pilot intended to depart from Moorabbin for the return flight to Archerfield two days before the day of the accident. On that day, the aircraft required the usual leaning to clear the engine roughness during taxi. Shortly after take-off, after the pilot had reduced the power settings to normal climb power, he noticed the vacuum gauge indicated zero. At about the same time, the engine began to run roughly, and the pilot elected to return to Moorabbin. After landing, he removed the engine cowling and noticed a significant amount of oil on the engine. The pilot then asked an engineer to investigate the problems. The engineer determined that there was a substantial oil leak from one of the top crank case bolts, and that the engine had lost approximately 3 L of oil. After resealing the bolt, refilling the engine with oil and replacing the vacuum pump, the engineer conducted a further engine run, during which there was no evidence of rough running.

Occurrence summary

Investigation number 199900252
Occurrence date 27/01/1999
Location 9 km N Moorabbin, Aero.
State Victoria
Report release date 30/04/2001
Report status Final
Investigation type Occurrence Investigation
Investigation status Completed
Mode of transport Aviation
Aviation occurrence category Forced/precautionary landing
Occurrence class Accident
Highest injury level Minor

Aircraft details

Manufacturer Piper Aircraft Corp
Model PA-32
Registration VH-HUX
Serial number 32R-7780546
Sector Piston
Operation type Private
Departure point Moorabbin, VIC
Destination Dubbo, NSW
Damage Destroyed

Cessna 210R, VH-MOK

Summary

While flying at 350 feet, during a final test line run for a geo-survey operation, the pilot of a Cessna 210 transmitted a distress message stating that the aircraft's engine had failed. The pilot conducted a forced landing in a nearby field. He later reported that he had selected an alternative fuel tank in an unsuccessful attempt to restart the engine prior to conducting the forced landing.

In consultation with BASI, an assessment of the aircraft fuel quantity and engine condition was carried out by a representative of the aircraft's insurer. This assessment indicated that the aircraft's fuel quantity had been approximately 70 L in the left tank and nil detectable fuel in the right. The examination of the engine revealed no identifiable problems that could have contributed to the engine failure.

The aircraft was fitted with long-range fuel tanks. The pilot reported that the refuelling operation had been carried out on a sloping surface and therefore, the aircraft may not have been laterally level. He was aware that there was a 'Caution' in the aircraft flight manual which stated that, 'indication errors of up to 14 gallons [53 L] per tank may result from a one-degree lateral deviation from level'.

The reason for the reported engine failure could not be determined.

Occurrence summary

Investigation number 199804254
Occurrence date 10/10/1998
Location 15 km W Cobar, Aero.
State New South Wales
Report release date 29/07/1999
Report status Final
Investigation type Occurrence Investigation
Investigation status Completed
Mode of transport Aviation
Aviation occurrence category Forced/precautionary landing
Occurrence class Accident
Highest injury level None

Aircraft details

Manufacturer Cessna Aircraft Company
Model 210
Registration VH-MOK
Serial number 21064933
Sector Piston
Operation type Aerial Work
Departure point Cobar, NSW
Destination Cobar, NSW
Damage Substantial

Beech Aircraft Corp F33A , VH-SIO

Summary

The student pilot was flying the Beechcraft Bonanza on a solo navigation exercise from Jandakot to Narrogin and Merredin before returning to Jandakot via Perth. The aircraft departed at about 1300 WST and was due to land at Jandakot about 2 hours later. The pilot reported that during the approach to Jandakot, he levelled the aircraft at 1,000 ft and set power to maintain the altitude. He then completed the pre-landing checks and lowered the flaps to slow the aircraft. Soon after lowering the flaps, he noted that the engine rpm was rapidly dropping. He reported that he opened the throttle and conducted the appropriate checks but the engine did not respond. The pilot transmitted a Mayday and decided to attempt a forced landing in an open patch of ground in the Canning Vale area. He then turned off the aircraft's fuel and electrical systems.

Avoiding parked trucks and buildings, the pilot landed the aircraft in a parking area of the Caning Vale Markets. The aircraft slid 15 m before hitting an unoccupied temporary building and coming to rest. The building was destroyed, the aircraft was substantially damaged and the pilot received minor injuries.

The post-accident inspection of the aircraft revealed that there was a substantial amount of fuel in both wing tanks but no fuel could be found in lines beyond the fuel selector valve. No fault was found with the aircraft's fuel system or engine.

Evidence was consistent with the engine sustaining fuel starvation and losing power but the investigation could not determine the reason for the fuel starvation.

Occurrence summary

Investigation number 199803888
Occurrence date 19/09/1998
Location Canning Vale
State Western Australia
Report release date 04/08/1999
Report status Final
Investigation type Occurrence Investigation
Investigation status Completed
Mode of transport Aviation
Aviation occurrence category Forced/precautionary landing
Occurrence class Accident
Highest injury level Minor

Aircraft details

Manufacturer Beech Aircraft Corp
Model 33
Registration VH-SIO
Serial number CE-1329
Sector Piston
Operation type Flying Training
Departure point Jandakot, WA
Destination Jandakot, WA
Damage Substantial

Cessna 208B, VH-URT

Safety Action

Following the incident, the Civil Aviation Safety Authority issued the following directions to the operator:

  1. Conduct daily water washing of the compressor and compressor turbine, after the last flight of the day when operating in a salt laden environment,
  2. Conduct engine condition trend monitoring (ECTM) in accordance with the procedures detailed in CASA Airworthiness Advisory Circular 6-29 Amdt-1, and
  3. Conduct boroscope inspections of the compressor turbine at intervals not to exceed 110 hours time in service, or alternatively conduct detailed hot end inspections at intervals not exceeding 750 hours time-in-service.

In addition to these directions, CASA issued Airworthiness Directive AD/PT6A/28 applicable to all PT6A series engines installed in single-engine aircraft. This airworthiness directive detailed the requirements listed above as well as requiring all operators to carry out a visual inspection of the compressor turbine blades for evidence of sulphidation, and to continue to carry out engine compressor turbine washing in accordance with the procedures detailed in the applicable Pratt and Whitney Canada maintenance manual.

CASA also published an article in the November 1998 issue of the CASA magazine Flight Safety Australia, which outlined the circumstances of the incident, and explained sulphidation, its causes, and preventative measures.

Summary

The pilot of the Cessna 208B reported that, after a normal departure from Badu Island, he established the aircraft in cruise at about 5,000 ft for the flight to Saibai Island. The engine had performed normally up to that time. Shortly after becoming established in the cruise, the pilot heard a muffled bang from the engine compartment. On checking the engine instruments, he saw that the gas generator RPM (Ng) had stabilised at about 52% and the turbine temperature was at about 700 degrees C. (The maximum allowable temperature was 740 degrees C.) The pilot initially thought that these symptoms may have indicated an engine fire, so he shut down the engine and feathered the propeller. However, appropriate checks indicated that there was no fire. During these observations, the pilot had turned the aircraft towards Badu Island.

The pilot then attempted two engine starts, neither of which was successful. On each occasion, Ng stabilised at 17%, fuel flow was 110 lb/h, and engine light-off occurred at 850 degrees C. Eventually, Ng stagnated at 42-43% and the propeller unfeathered, but there was no indication of engine torque.

The pilot transmitted a distress call after the first start attempt. After the second attempt, he decided to concentrate on flying the aircraft and realised that his best option was to try to land on a narrow beach on the northern side of Badu Island. He was able to achieve this without any damage to the aircraft.

The aircraft was powered by a Pratt and Whitney Canada PT6A turboprop engine. Examination of the engine revealed that a compressor turbine (CT) blade had failed at about mid-span. The liberated section of blade had struck two adjacent blades, causing them to break. Metallurgical examination of the remaining section of the failed CT blade indicated that the fracture line passed through the centre of a deep sulphidation pocket. Impact damage was observed on all the remaining CT blades. Detailed examination of these blades revealed a total of seven blades that exhibited some degree of cratering due to sulphidation. Inspection of the compressor section of the engine also revealed significant corrosion consistent with operation in a salt-laden atmosphere.

"Sulphidation" refers to the reaction of sulphur containing compounds with metallic components that have been exposed to a hot gaseous environment. Components of gas-turbine engines located in the hot gas path, such as blades and vanes, are exposed to sulphidation during normal operation. Corrosive sulphates are formed during the combustion process from sulphur in the fuel and sodium and potassium salts present in the fuel and air, in particular the air in marine environments. If the accumulations of sulphur-containing salts are not removed from the surfaces of the turbine blades and vanes, the protective oxide coating will be attacked and the underlying alloy rapidly corroded.

The PT6A engine maintenance manual required desalination water-washes to be applied to both the compressor and the turbine after the last flight each day. The operator was conducting compressor washes prior to the first flight of each day, using the compressor wash ring installed on the engine. A special wash tube assembly tool for installation into the gas generator igniter boss, to enable wash solution to be introduced directly to the first-stage turbine blades, was available from the engine manufacturer. The operator was not using this tool. As a result, effective washing of the turbine blades was not achieved, allowing salt deposits to build.

The engine was fitted with an Engine Condition Trend Monitoring (ECTM) system which recorded a number of key engine parameters such as fuel flow, inlet turbine temperature, torque, and Ng. Use of this system allowed operators to delete a fixed-time hot section inspection (HSI) requirement in favour of basing the HSI interval on the results of engine condition trend monitoring. Use of the ECTM system was acceptable to the Civil Aviation Safety Authority (CASA), provided the procedures were in accordance with CASA Airworthiness Advisory Circular 6-29 `PWC PT6A Series Engines HSI Policy' 5/98. Investigation revealed that the operator was not complying fully with all the requirements of the circular.

Occurrence summary

Investigation number 199803389
Occurrence date 21/08/1998
Location Badu Island
State Queensland
Report release date 27/08/1999
Report status Final
Investigation type Occurrence Investigation
Investigation status Completed
Mode of transport Aviation
Aviation occurrence category Forced/precautionary landing
Occurrence class Incident
Highest injury level None

Aircraft details

Manufacturer Cessna Aircraft Company
Model 208
Registration VH-URT
Serial number 208B0428
Sector Turboprop
Operation type Charter
Departure point Badu Island, QLD
Destination Sabai Island, QLD
Damage Nil