Forced/precautionary landing

Circumstances:

During a flight from Orange to Camden, the pilot was unable to proceed beyond the Hartley Valley due to low cloud on the ranges. He attempted to return to Orange but found that route was also blocked by deteriorating weather conditions. In decreasing light, the pilot began a search and located a paddock where he conducted a precautionary landing, during which the main landing gear and wing struck a fence, causing substantial damage. The area forecast had indicated extensive areas of low cloud, with reduced visibility associated with drizzle, over the mountain ranges.

Circumstances:

The pilot reported that, soon after completing the pre-landing checks during the descent to land, the engine lost power. After checking that the mixture was full rich, he called Hamilton Tower to advise of his situation and was cleared for a straight-in approach. The pilot then changed the fuel selector from the left to the right tank for a short time, but did not actuate the electric fuel boost pump. When the engine did not respond, he re-selected the left tank and again attempted unsuccessfully to restart the engine. Moments later the aircraft made a forced landing into the sea approximately 2 km short of the runway. No fault was found with the aircraft which might have contributed to the accident. Only minute quantities of fuel were found in the fuel system and both tanks contained only sea water. The pilot stated that the aircraft departed Maroochydore with full tanks (337 L) with the left tank selected, and that he changed tanks every 30 min during the flight. He said he had not personally refuelled the aircraft but had noted the fuel level in both tanks prior to DEPARTURE to be at the bottom of the filler tubes. These tubes extend downward into each fuel tank from the filler caps a distance of approximately 10-12 cm. The pilot advised that much of the flight was conducted below 2000 ft with the mixture full rich or only slightly lean. Investigation determined that the actual fuel quantity on board at DEPARTURE was 270-280 L (135-140 L per tank). This meant that 55-65 L remained on board at the time of the accident. Judging from the duration of the flight and the tank selection procedure used by the pilot, the left tank should have been selected for 120 min (4x30 min periods) and the right tank for 90 min. Total usage from the left tank should therefore have been 110-120 L, still leaving 20-30 L in the tank when the power loss occurred. However, had the pilot missed a tank change and had the left tank been selected for 150 min instead of 120, the fuel quantity used from the tank would have been around 150 L, close to the calculated quantity in the tank of 135-140 L. It is likely, therefore, that the pilot missed a tank change, thereby draining the left tank, and that the engine lost power as a result of fuel starvation. Because the electric boost pump was not actuated when the pilot selected the right tank after the power loss, it would have taken more time for the fuel to flow from the right tank to the engine. The left tank was probably reselected before this had time to take place.

Significant Factors:

The following factors were considered relevant to the development of the accident

1. The pilot did not have adequate knowledge of the aircraft fuel system.

2. The aircraft fuel tanks were not filled to capacity prior to DEPARTURE.

3. The pilot probably missed a tank change, thereby draining the left fuel tank.

4. The pilot did not actuate the electric fuel boost pump when attempting to restart the engine.

Circumstances:

The pilot planned a short flight to Bunbury prior to boarding an international flight from Perth Airport that afternoon. He dipped the fuel tanks and checked the fuel gauge reading against the fuel calibration card, concluding that the aircraft contained about 70 litres of fuel, which was sufficient for the intended flight. Enroute to Bunbury, the pilot noticed that the fuel usage appeared higher than normal but did not take any action to remedy the situation, beyond leaning the mixture. The pilot noticed that both fuel tank indicators were approaching empty when he broadcast the inbound to Bunbury radio transmission. He was not concerned, as a check of the fuel gauge calibration card indicated that there were still 40 litres remaining. Eleven kilometres from Bunbury, the engine stopped and except for one brief burst of power would not restart. The pilot attempted to enter a forced landing pattern two or three times but had to change the selected landing area on each occasion due to a proliferation of power lines in the area. The aircraft eventually touched down on the bank of a river and overturned. A check of the aircraft found that eight and a half litres of fuel remained in the system, this was three and a half litres less than the published unusable fuel for the aircraft. Perusal of the aircraft records indicated that an error had been made during the latest fuel gauge calibration, and the fuel gauge calibration card was inaccurate. It is probable that the aircraft contained much less than 70 litres on DEPARTURE. The pilot sustained facial lacerations when the aircraft overturned on landing. He was not wearing the shoulder portion of the lap/sash seat belt. It is probable that there would have been no injuries if the complete belt had been worn.

Significant Factors:

The following factors were considered relevant to the development of the accident:

1. Inaccurate fuel gauge calibration chart.

2. Fuel exhaustion resulting in a forced landing on unsuitable terrain.

3. The pilot did not ensure that the complete seat belt was fastened.

The pilot reported that the helicopter was cruising normally at 2500 feet when there was a loud noise from the engine. He entered autorotation and landed but the helicopter rolled on its side on landing. Specialist examination of the engine revealed that the number four piston connecting rod failed as a result of fatigue cracking initiated by galling between the connecting rod and the crankshaft bearing. Galling is the deposition of metal from one surface on another when the two surfaces come into forceful contact. The galling was induced by continual over speeding of the engine. Other connecting rods from the engine also showed evidence of galling.

Significant Factors:

The following factors were considered relevant to the development of the accident

1. The engine had previously been subjected to over speeding.

2. The connecting rod suffered a fatigue failure as a result of galling.

3. Uncontained mechanical failure of the engine.

4. The pilot was forced to land the helicopter on unsuitable terrain. This accident was not the subject of an on-scene investigation.

Circumstances:

The aircraft was being ferried to a strip in the Hunter Valley to carry out parachute dropping. En-route, although performance appeared normal, the pilot sensed a possible problem with the turbine engine and landed at Somersby NSW. After landing, a visual inspection of the engine was undertaken and several ground runs carried out, without any defect being found. The pilot elected to return to Camden where further checks could be made. The aircraft took off towards the south and climbed to a height of about 1000 feet, at which point the pilot heard a bang from the engine compartment which was followed soon after by a very loud bang associated with noises described by the pilot as like "forcing a crowbar into the engine".

During the ensuing forced landing in fine, calm conditions, the pilot misjudged the final stage of the landing approach, resulting in the aircraft landing very heavily and sustaining considerable damage. A subsequent engineering investigation found the engine had suffered a catastrophic malfunction resulting from fatigue failure of a helical thrust washer from the high-speed pinion assembly. The fatigue crack growth resulted from applied alternating loads, probably due to excessive wear of the helical gear shaft bearing. About one week prior to the flight, engine oil sample and filter element had been removed in accordance with Airworthiness Directive AD/TPE 331/9.

Although the samples had not been immediately sent for laboratory analysis, the owner was allowed to operate the engine for a further 50 hrs before satisfactory test results were received. When the samples were examined during the investigation, articles were found which indicated bearing deterioration prior to the engine failure. Earlier in the year, the aircraft had sustained a propeller ground strike, resulting in the engine being removed and returned to an overhaul facility for inspection. It has been reported that when the engine was re-installed in the aircraft, a torque meter calibration check was not carried out in accordance with the procedures published in the manufacturer's maintenance manual. Although not positively established, the mode of engine failure was consistent with an excessive torque setting.

Significant Factors:

The following factors were considered relevant to the development of the accident:

1. Probable excessive wear of the helical gear shaft bearing.

2. Torsional loading of the helical thrust washer from the high-speed pinion gear assembly.

3. Fatigue failure of the helical thrust washer.

4. Pilot-in-command misjudged distance, speed and altitude during forced landing approach.

The ATSB has completed its technical analysis report of the flight recorder data from an Avions de Transport Régional ATR-42-300F aircraft, registered PK-YRP, on behalf of the Indonesian National Transportation Safety Committee (NTSC). The aircraft carried out a forced landing in a rice field, approximately 33 km (18 NM) north-east of Balikpapan-Sepinggan Airport, Indonesia, after reportedly losing power from both engines.  The aircraft was operating a scheduled passenger service from Berau-Kalimaru Airport to Temindung Airport, Indonesia. Two of the 51 passengers on-board sustained serious injuries.

The NTSC is responsible for investigating this occurrence. The NTSC requested assistance from the Australian Transport Safety Bureau (ATSB) in the recovery and analysis of information from the aircraft's flight data recorder and cockpit voice recorder.

To protect the information supplied by the NTSC to the ATSB and investigative work undertaken to assist the NTSC, the ATSB initiated an investigation under the Transport Safety Investigation Act 2003.

The ATSB's technical analysis report has now been provided to the NTSC, who is responsible for releasing a final investigation report on this occurrence.

National Transportation Safety Committee
Ministry Of Transportation Republic Of Indonesia
Transportation Building 3^rd Floor
Jalan Medan Merdeka Timur No. 5
Jakarta Pusat 10110
Indonesia

Phone  :  +62 21 384 7601
Email    :  knkt@dephub.go.id

Website: http://knkt.dephub.go.id/knkt/ntsc_home/ntsc.htm

______________

Released in accordance with section 25 of the Transport Safety Investigation Act 2003.

The Australian Transport Safety Bureau did not conduct an on-scene investigation of this occurrence. The report presented below was derived from information supplied to the Bureau.

At 1015 hours Eastern Standard Time (EST) on 28 July 2004, the Robinson Helicopter Co R22, VH-KHU, was being operated to conduct circuit training at Mangalore aerodrome, Victoria. The pilot, the sole occupant of the helicopter, held a current private pilot licence (helicopter). The pilot reported that shortly after reaching circuit height on the crosswind leg of the circuit, the helicopter began to yaw rapidly in alternating left and right directions. The pilot turned the helicopter towards the aerodrome for an immediate landing. He subsequently reported that during the descent, the main rotor low RPM horn sounded twice, accompanied by the illumination of the main rotor low RPM light. The pilot also reported that at about 200 ft above ground level the main rotor low RPM warnings were again triggered by `the full collapse of engine RPM'. The pilot performed an autorotation, but the helicopter was landed heavily. Impact forces destroyed the helicopter, and the pilot received minor injuries.

The helicopter was not recovered from the aerodrome until the following day. The weather at Mangalore aerodrome included some periods of rain after the accident, and the operator reported that some water may have entered the helicopter's fuel tank, which was ruptured by the impact forces.

The helicopter's engine was removed and tested to determine its serviceability, but the engine operated normally, and no mechanical faults were detected. Some water was found in the engine's fuel system. The operator reported that testing was performed on a fuel sample taken from the Mangalore aerodrome fuel supply. The testing revealed that the fuel was not water contaminated. The operator also reported that an instructor and the pilot had each completed independent daily inspections of the helicopter before the accident flight. Both had conducted fuel drains, and both reported that the fuel samples contained no water.

It was subsequently reported that at the time of the occurrence, the cloud base at Mangalore aerodrome was about 1,400 ft. The 1000 EST Mangalore automatic weather station data revealed that the temperature was 8 degrees C, and the dewpoint temperature was 6.3 degrees C. The dew point depression was therefore 1.7 degrees C, which meant that there was a probability of serious carburettor-icing, as depicted at fig. 1. Other helicopters were operating in the Mangalore circuit at the time of the occurrence. Although the pilots of those helicopters reported that their helicopters had not been affected by carburettor-icing, the investigation was unable to discount that carburettor-icing may have been the factor that resulted in the abnormal operation of the helicopter's engine.

Figure 1: Carburettor icing-probability chart.

Source: Melting Moments: understanding carburettor icing, Asia Pacific Air Safety, June 1999, Issue 22.

The Australian Transport Safety Bureau did not conduct an on-scene investigation of this occurrence. The information presented below was obtained from information supplied to the Bureau.

At about 1130 Eastern Standard Time on 20 June 2004, the Cessna 150F departed runway 28 at Gladstone, Qld for a local flight. As the aircraft climbed through 200 ft above ground level, the engine began to lose power. The pilot selected the fuel shutoff valve to OFF for a forced landing on a nearby road. During the landing, the aircraft's right wing struck an embankment and the aircraft sustained substantial damage. The two occupants received minor injuries.

The pilot later reported that the aircraft had been washed about two hours before the flight, and had departed with 35 litres of fuel in the tanks. Meteorological information indicated that induction system icing was possible but it could not be verified. Similarly, the possibility of water contamination of the fuel system was considered, but could not be substantiated. The reason for the engine failure was not established.

The engine failure required the pilot to enter an autorotation while manoeuvring at low level and over inhospitable terrain. The investigation could not determine conclusively whether the single 'beep' reported by the survivor was a momentary warning of an impending engine power failure.

Once the main rotor blade struck the tree, further autorotation flight to landing was not possible and the helicopter impacted the ground in an out of control condition.

The survivor's seating position in the right rear of the helicopter shielded him from the majority of the impact. Although severely injured, he was able to escape from the wreckage. The rapid spread of the fire post impact did not allow the survivor to extricate any other occupants from the wreckage.