A Piper PA-30 Twin Comanche on a private VFR flight out of Jandakot Airport suffered a loss of control due to sufficient airspeed not being maintained.
At 0826 Western Standard Time on 28 March 2010, a Piper PA-30 Twin Comanche, registered VH-KDS, departed Jandakot Airport for a private flight under the visual flight rules (VFR). On board were two qualified pilots, both of whom were endorsed on the aircraft type. No details of the flight were submitted to Air Traffic Services nor left with any other person. At 1815, following the failure of the aircraft to return to Jandakot, the Australian Rescue Coordination Centre was notified and a search was initiated to locate the aircraft.
Following examination of radar data, the aircraft was located the following morning by the crew of a search and rescue (SAR) helicopter. Upon landing, the helicopter crew established that the two occupants had sustained fatal injuries. Analysis of data recorded by onboard GPS equipment identified that while maintaining about 3500 ft above mean sea level, the speed of the aircraft steadily decreased followed by a steep descent that continued to ground level. Examination of the aircraft identified that the propeller of the left engine was feathered prior to impact; however, no evidence of a defect or other circumstance that would have necessitated feathering of the propeller was identified.
The investigation identified that the circumstances of the accident were consistent with a loss of control due to sufficient airspeed not being maintained. In addition, the investigation found that the lack of flight details available for the search and rescue authorities and the non-activation of the portable emergency locator transmitter hampered the SAR response.
History of the flight Following the failure of the aircraft to return to Jandakot, the Australian Rescue Coordination Centre (RCC) was notified by an associate of the aircraft owner and a search was initiated to locate the aircraft. RCC enquiries established that no emergency satellite signals or radio calls had been received from the aircraft. An examination of the recorded radar information, established that, following departure from Jandakot, the aircraft proceeded north-east to the vicinity of a property owned by one of the pilots. The data indicated that the aircraft conducted operations between the property and Toodyay township prior to the loss of radar identification about 5 km north-east of the pilot’s property, as the aircraft descended below 2100 ft AMSL.
At 0110 on 29 March 2010, an SAR helicopter departed from Jandakot Airport to conduct a search in the vicinity of the last radar position. The helicopter crew completed a search of the area at about 0230 without locating the aircraft. At 0618 the RCC was notified that further analysis of the radar data had re-identified what was believed to be the aircraft proceeding south-west from the vicinity of Northam township. The radar data indicated that, at about 38 km from Northam, the aircraft commenced a steep descent before radar identification was again lost.
Information on the aircraft’s flight path was recovered from a GPS unit onboard the aircraft. That information, which included position, groundspeed and altitude, indicated that following departure from the aircraft tracked to the north via a VFR route to the east of Perth Airport. The GPS data also indicated that in the minute before the descent, the groundspeed steadily decreased while the aircraft maintained an approximately constant track and altitude.
On the morning of the accident a witness, who was located about 2 km from the accident site, recalled hearing an aircraft engine that was approaching from the north ‘splutter’ once or twice before resuming normal operation. A short time later the engine revolutions were heard to increase significantly followed by a ‘thud’ or ‘boom’. The witness routinely observed aircraft using the VFR route, including previous instances of apparently rough-running engines, and associated the final sound with blasting that routinely occurred at a nearby mine.
Personnel information Both occupants of the aircraft were qualified pilots and had completed initial multi-engine endorsement training in the aircraft in late 2008. That training, which consisted of familiarisation with the aircraft type and its associated systems, together with asymmetric control and performance aspects, satisfied the requirements of a biennial aircraft flight review. The investigation could not determine if any additional proficiency training was conducted by either pilot since completion of their PA-30 endorsement. Pilot A had held a PPL-A since October 1994 and had around 330 total flying hours. Pilot B had held a PPL-A since May 1996 and had accumulated 719 total flying hours. However, in the past six months they only had 1.2 and 3.7 hours, respectively, on PA-30 types.
Aircraft information The Twin Comanche, serial number 30-952, was manufactured in 1966 and the owner at the time of the crash had owned the aircraft since 1986. An examination of the available refuelling records identified that the aircraft was refuelled with 132 L immediately prior to the departure from Jandakot Airport. Discussion with the refueller established that this quantity of fuel was added to the aircraft’s main and wing auxiliary tanks, such that all four tanks were visibly full. Full main and wing auxiliary tanks provided 317 L of usable fuel which provided at least four hours endurance under normal operation. No fuel was reported as being added to either of the wingtip fuel tanks, which was consistent with advice from the aircraft owner that the wingtip tanks were not normally used for local flights.
Maintenance history The aircraft’s maintenance release authorised day and night VFR operation in the private category, and was valid until 12 June 2010 or 5722.8 hours in service. At departure, the aircraft had been operated for 5638.6 hours and there were no outstanding defects recorded. The most recent maintenance on the aircraft was a scheduled inspection that was conducted on 12 June 2009 and the replacement of both main tyres on 18 February 2010.
Weight and balance An estimation of the aircraft’s weight and balance indicated that at the time of the accident the aircraft was operating about 214 kg below the maximum allowable gross weight of 1690 kg and forward of the forward longitudinal centre of gravity limit by between 3 and 19 mm. The manufacturer’s performance data indicated that, based on the estimated weight at the time of the accident, the aircraft was capable of maintaining a cruising altitude of about 3,00 ft with one engine inoperative (OEI). The aircraft manufacturer advised that operation at the calculated weight and balance would increase the control force necessary to manoeuvre the aircraft and the nose-up trim required during takeoff and climb. In addition, the forward c.g would increase the download on the tailplane, increase the rudder effectiveness and may increase the aircraft’s aerodynamic stall airspeed.
Meteorological information The Bureau of Meteorology stated in after-flight analysis stated that the likely weather conditions at the incident site during the morning of 28 March 2010...were clear skies with southeast winds between 15 and 20 knots extending to at least 4000ft. With near-surface winds of around 20 knots it is possible that some light to moderate terrain-induced turbulence would have been encountered.
On-site examination Examination of the accident site identified that, while travelling in a westerly direction, the aircraft clipped the overhanging foliage of a tree before significantly impacting a second tree and the overhanging foliage of an adjacent third tree. The aircraft then travelled a further 9.5 m and collided with a fourth tree about 6 m above the ground before coming to rest at the base of that tree, facing back along the approach direction. The aircraft occupants were found outside of the cockpit, adjacent to the main wreckage.
A comparison of the aircraft’s 10.6 m wingspan with the relative height and position of the contacted and untouched trees in the vicinity identified that the aircraft was banked during the contact with the second and third trees. An examination of the recovered GPS data indicated that the aircraft descended from 3500 ft to the accident site at an average angle of about 38°. Both left propeller blades appeared to be in the feathered position with no rotational damage evident.
About 11 L of fuel was recovered from the intact left wing auxiliary fuel tank and was free of water and other contaminants. There was no evidence of fuel leakage from the fuel filler caps during the flight. A damaged, 406 MHz portable locator beacon (PLB) was recovered from the aircraft wreckage. It had been stowed in a black pouch and was not activated.
CAPTION: After impact, the right propeller remained attached to the engine with one blade bent rearwards about 90°. (ATSB)
Analysis The data from the onboard GPS and the physical evidence at the accident site indicated that, during the return to Jandakot Airport in good weather conditions, the aircraft’s airspeed decreased followed by a loss of control. Examination of the aircraft and associated systems did not identify any engineering defect or failure that contributed to the development of the accident. The results of the pilots’ post-mortem examinations and toxicological testing did not identify any evidence of incapacitation that may have affected either pilot. That included the potential for the non-completion of maintenance on the combustion heater to have resulted in carbon monoxide poisoning in the unlikely event that the heater was selected ON.
The following discussion examines a number of other factors with the potential to have contributed to the loss of control.
Development of the accident Technical examination of the left propeller and associated engine tachometer identified that the left propeller was feathered prior to impact. The derived airspeed profile in the minute before the descent was consistent with the operation of the aircraft with one engine inoperative (OEI). Although there was sufficient time to have feathered the propeller during the steep descent, any pilot recovery action is unlikely to have involved the intentional feathering of a propeller. It was therefore considered likely that the left propeller was feathered prior to the descent. An assessment of the right propeller and engine tachometer identified that the right propeller was operating at low RPM and low (fine) pitch prior to impact. Low RPM on the right propeller, although consistent with an attempt to recover from a loss of control, may have also reflected a problem with the right engine.
In the absence of any evidence of a technical defect or failure, the investigation considered the possible circumstances that would have required the shutdown of the left engine and subsequent feathering of the propeller. Based on the quantity of fuel onboard at the commencement of the flight, and the amount recovered from the intact auxiliary tank, fuel exhaustion or starvation were both considered unlikely at normal fuel consumption rates. Similarly, there was no evidence of a fuel leak or other fuel system defect that would have produced a fuel-related power loss. The investigation also considered the possibility of a power loss due to fuel selector mismanagement; however, the uncertainty associated with the pre-impact selector position precluded an assessment. Additionally, in the event of inadvertent fuel starvation, an engine restart should have been possible without recourse to propeller feathering, provided that the problem was correctly identified.
Based on the presence of water within the right fuel injector filter, the possibility that fuel contamination may have affected the one or both engines was also considered. The evidence of long-term corrosion and the advice of the owner concerning use of the fuel drains, indicated that the aeroplane had probably operated uneventfully with water present in the fuel selector drains on previous occasions. Given that the aircraft had flown for over an hour prior to the accident, it was considered probable that a change in fuel tank selection would have been required to facilitate a power loss due to fuel contamination. However, the in-flight fuel selector positions could not be confirmed.
Irrespective of the reason for the left engine shutdown, the manufacturer’s performance data indicated that the aircraft was controllable and probably capable of maintaining altitude with OEI. The deceleration of the aircraft towards VYSE evident from the GPS data, was consistent with the ideal response to a power loss; however, the continued airspeed reduction towards the nominal VMCA airspeed resulted in loss of control of the aircraft. The varying degree of possible asymmetry indicated by the propeller examinations, combined with the number of factors that affect the actual VMCA and stall airspeeds, meant that the loss of control may have been associated with asymmetric operation, a stall/spin or a combination of both. Irrespective of the actual mechanism, the accident was consistent with a loss of control due to insufficient airspeed.
The investigation was unable to determine the extent to which the out of limits forward centre of gravity may have contributed to the loss of control. The relative inexperience of both pilots and apparent lack of recency in the conduct of emergency procedures in multiengine aeroplanes increased the likelihood that a malfunction, especially a partial or fluctuating power loss, may not have been diagnosed and managed appropriately. The absence of a distress call indicated that the pilots probably experienced significant workload in responding to the situation. This may have resulted in the unintended deceleration of the aircraft below a safe airspeed. The indicative altitudes required for spin recovery and the recommended safety heights associated with the practice of stalling and VMCA demonstrations indicated that control recovery may not have been possible within the available height above the surrounding terrain.
Emergency response Since no detail of the intended flight was provided to either Air Traffic Services (ATS) or left with any other person, an emergency response relied on either a distress call or selection of an emergency transponder code by the pilot, observation of the accident or, as in this case, notification by a concerned third party that the aircraft had not returned to Jandakot Airport. The absence of an alerting distress call or selection of an emergency transponder code by the pilots, presumably due to the significance of the emergency, meant that the detection of the emergency by ATS relied on the observation of a radar trace that was outside of the controller’s responsibility and focus.
Recognising that there was no requirement for details of the flight to be provided to ATS or other agencies, the lack of such information hampered the search and rescue response. If information on the intended flight route had been available, it would have led to a focussed search effort that would probably have resulted in the rapid location of the aircraft and occupants. Additionally, although the carriage of a portable emergency locator transmitter (ELT) complied with the relevant regulations, an automatic impact activation feature may have provided the SAR agency with more timely advice that the aircraft was in distress. Although earlier location of the aircraft would not have reduced the severity of the outcome in this instance, the availability of accurate flight information will generally provide for a more timely emergency response.
CAPTION: What was left of the feathered left prop and engine. (ATSB)
Findings From the evidence available, the ATSB made the following findings. They should not be read as apportioning blame or liability to any particular organisation or individual.
Contributing safety factors: During cruise, the airspeed decreased to the extent that control of the aircraft was lost.
Other safety factors: The aircraft was operating outside of the longitudinal centre of gravity limits at the time of the accident.
The location of the aircraft was delayed due to the lack of flight details available for the search and rescue authorities and the non-activation of the portable emergency locator transmitter.
Other key findings: The crew complied with all requirements relating to flight notification and the carriage of emergency locator transmitters.
Sources of information - The owner of the aircraft. - The aircraft maintenance records. - The pilot’s endorsing instructor. - A hearing witness to the accident. - The aircraft refueller and refuelling documentation. - The aircraft manufacturer. - The United States National Transportation Safety Board (NTSB). - The Australian Maritime Safety Authority (AMSA). - The Bureau of Meteorology. - CASA. - Airservices Australia.
Piper Twin Comanche VH-KDS in better times. (ATSB)
After impact the right propeller remained attached to the engine with one blade bent rearwards. (ATSB)
What was left of the feathered left prop and engine. (ATSB)
