The Sordid Story behind the Cork Fatal Accident : Manx2, Air Lada and Flightline

7 Feb 14 14 Comments

This is an in-depth look at the operations behind the crash that I wrote about last week: Manx 2 Fatal Accident at Cork: Below the Required Minima.

This might take a while, so get comfortable.

On the 10th of February in 2011, the Fairchild Metro III attempted to land at Cork in low visibility conditions.

The aircraft carried out two ILS approaches, both of which were continued beyond the OM equivalent point with conditions below required minima. On both of these approaches, descent was continued below [Decision Height], followed by a missed approach. The aircraft then entered a holding pattern following which a third ILS approach was made with conditions below required minima. This approach was continued below DH and a missed approach was initiated. Approaching the runway threshold, the aircraft rolled to the left, followed by a rapid roll to the right during which the right wingtip contacted the runway surface. The aircraft continued to roll and impacted the runway in an inverted position. The aircraft departed the runway surface to the right and came to rest in soft ground.

To understand the situation, we will have to untangle the complex relationships between the companies.

A Spanish bank owned the physical aircraft and leased it to Air Lada. In the accident report, Air Lada is referred to as the Owner. Air Lada subleased the flights to Flightline S.L., referred to in the report as the Operator. The service was sold by Isle of Man company Manx 2, referred to in the report as the Ticket Seller.

The Captain held a Commercial Pilot Licence issued in Spain and had a total flying time of 1,801 hours with 1,600 on type.

Records show that he had operated as a co-pilot (first officer) into Cork on 61 occasions and he flew in on seven occasions as Captain. There’s no records of any diversions on any of these flights. In addition, he’d never operated into Waterford or Kerry, which could have increased reluctance to divert.

He was promoted to Captain on the 6th of Feb 2011, four days before the accident.

Flightline’s Chief Instructor was involved in the training of both the Captain and the First Officer of the accident flight. He stated that although all the pilots were considered to be Flightline pilots, the pilots who flew the Metro III were paid by Air Lada.

He was the sole Class Rating Examiner at Flightline and it was his decision as to how many training sectors a candidate for captaincy would receive. The other Class Rating Instructor at Flightline had left the company shortly before. This meant that he both trained and checked the Captain. He agreed that there should be a separate Class Rating Instructor and Class Rating Examiner, but in the Captain’s case, this was not possible. He felt he would be able to objectively evaluate the Captain. He agreed that the command training of the Commander was very disrupted and thought it was possibly because they were tight in numbers and because the Captain had to travel from Belfast to Barcelona for training. He described the commander as enthusiastic with good crew resource management.

The First Officer held a JAA CPL issued in the United Kingdom and had a total flying time of 539 hours with 289 on type. He was employed as a co-pilot by the operator just three weeks before the crash. He went through a thirty-minute Operator’s Proficiency Check at that time. Flightline’s Chief Instructor described him as ‘was okay for his hours‘.

Flightline did not have any restrictions for newly qualified flight crew, so it was possible for newly qualified commanders and co-pilots to operate together.

The First Officer did not complete the line training and should only have flown with a training captain until he had completed his line training and passed a line check. The Flightline SL records show that all of First Officer’s flights were with line captains who were not instructors.

His CV indicated that he had studied Italian but there was no evidence of Spanish language competency.

It was quite clear that the duty and rest periods of the crew were not correctly documented.

The Co-pilot’s roster for February showed him ‘Libre’ (‘free’) between 8-12 February 2011. As another co-pilot requested a change of duty on 9 February, the Co-pilots duties were changed and he was required to operate the scheduled flights on 9 and 10 February 2011. The identity of the co-pilot was not noted in the flight paperwork of the short positioning flight between EGAA and EGAC on 9 February 2011. The Investigation is satisfied that the Co-pilot operated this flight and subsequently the two return flights to Cork. The two other possible candidates forwarded copies of their personal logbooks to the Investigation which showed they did not operate the sector.

The partner of the First Officer stated, “He did not have very much rest. He was working on a defined route incorporating some night-time flights carrying post.” She last spoke to him at 23:00 on 9th of February, the night before the crash.

Flightline, as the operator, was responsible for the flight crew, including ensuring staff competence and rest periods. The Quality Manager at Flightline stated that Air Lada produced a draft roster for flight crew, which Flightline would monitor and and amend as needed to ensure that flight time limitation rules were followed. Any roster change after the roster was published had to be approved.

However, none of the flight changes prior to the accident flight had been approved and Flightline were unaware that the co-pilot was on the flight until after the accident. The flight crew instruction regarding making a roster change request was in Spanish and no English translation was found. As a result of unofficial roster changes, neither the Captain nor the First Officer were fully rested at the time of the accident flight.

The aircraft had been involved in two previous significant events.

In the early mornings of 21 May 2004, EC-ITP was involved in a take-off incident at Palma de Mallorca Airport (LEPA). On take-off, the aircraft accelerated normally to 60 knots when the nose wheel steering (NWS) system was deactivated the aircraft veered to the right. The take-off was abandoned and reverse thrust applied but the aircraft departed the runway and incurred minor damage which was later repaired.

On 8 November 2009 EC-ITP suffered a heavy landing at Barcelona. As a result, the aircraft was ferried to an overhaul facility in Cologne for repair. This work was completed and the aircraft returned to service in October 2010, four months prior to the accident.

The repair for the heavy landing in November 2009 revealed that the left-hand engine installed in the engine was a loaner – meant for temporary use. It was removed and another loaner engine installed on 15 July 2010. The right-hand engine was removed from the aircraft on 27 April 2010 “for access to repair area” and re-installed on the right-hand side on 13 July 2010.

Despite the change of engine, a full Engine Ground Run check was not run. Instead, the checklist for engine adjustments was used. The engine ground runs did not include power lever split checks at Flight Idle.

On the last approach into Cork that day, the Captain, who was Pilot Not Flying, took control of the power on the last approach. At the decision height, he pulled back the power and called to continue, clearly hoping to spot the runway in the fog.

From last week’s blog post:

They once again descended below the decision height of 200 feet above the ground. However this time, the aircraft reduced power and at the same time experienced a roll to the left.

Terrain Awareness Warning System: ONE HUNDRED
Captain: Go around!
First Officer: Round.
Terrain Awareness Warning System: FIFTY
Terrain Awareness Warning System: FORTY

The Captain applied go-around power which is when they lost control of the aircraft. The aircraft rolled rapidly to the right and the right wingtip contacted the runway surface. The aircraft continued to roll.

It is likely that the First Officer, the pilot flying, did not realise what his Captain was doing with the power and made a control wheel input to the right to correct for the left roll.

The investigation cites three principle factors contributing to the loss of control:

The uncoordinated operation of the power levers and the flight controls, which were being operated by different Flight Crew members.

The retardation of the power levels below Flight Idle, an action prohibited in flight, and the subsequent application of power are likely to have induced an uncontrollable roll rate due to asymmetric thrust and drag.

A torque split between the powerplants caused by a defective Pt2/Tt2 sensor, became significant when the power levers were retarded below Flight Idle and the No. 1 powerplant entered a negative torque regime. Subsequently when the power levers were rapidly advanced during the attempted go-around, this probably further contributed to the roll behaviour as recorded on the FDR.

The Flight Data Recorder was recovered with 106 hours of data leading up to the accident. It showed that there as a mismatch between the torques being delivered by the two engines.

In general, the data showed that the torque being delivered by No. 2 engine exceeded that being delivered by No. 1 engine by up to 5%. It was also noticed that, as the power levers for both engines were being advanced prior to take-offs, the torque response for the No. 2 engine was faster than that for No. 1 engine.

The data showed that the pilots were adjusting the power levers to compensate for the engine torque differential. As the aircraft was descending towards the runway shortly before impact, the No. 1 engine was at 20-23% and the No. 2 engine was at 25-27%.

Then, 8 seconds before impact, a negative torque value was recorded. The next recorded values show the torques on both engines increasing but not in sync.

The next recorded values for this parameter [the No. 1 engine), following at intervals of one second, were +22%, +10%, +7% and +36%. Thereafter, recorded torque values for No. 1 engine rose rapidly. In a similar timeframe, the No. 2 engine torque values were recorded at +8%, 0%, +3%, +5% and +25%. Thereafter recorded torque values for No. 2 engine also rose rapidly to values in excess of 90%.

The negative torque the No. 1 engine caused the left roll. Then as both engines started to increase rapidly, the aircraft rolled to the right to 115° bank before the aircraft impacted.

The issue turned out to be a faulty sensor on the No. 2 engine. The defective engine intake air temperature and pressure sensor was caused by a crack in the side coil of the sensor bulb. The fracture surface was corroded, showing that the crack had been there for some time.

The effects of this fault were

  • Slower engine speed response when the speed lever was advanced
  • Faster engine torque response when the power lever was advanced
  • Higher torque for a given power level angle

Manx 2, based at Ronaldsway Airport on the Isle of Man, is referred to in the report as the “Ticket Seller”. They did not have an operating licence or an Air Operator’s Certificate, which is the approval required for an aircraft operator to use aircraft for commercial flights.

Instead, Manx 2 had contractual relationship with four AOC holders, including Flightline.

Flight crew on the flights wore a Manx 2 uniform and aircraft were painted with Manx 2 livery. However, the company did not want to have the “regulatory complexity and crewing problems associated with holding an AOC”. Manx 2 didn’t need an Air Travel Organisers Licence to sell the tickets, as none of the aircraft had in excess of 19 seats.

The UK Civil Aviation Authority was concerned that Manx 2 gave the impression that it was a licensed airline, at which point Manx 2 updated the website to state that it was a Marketing Group and acting as an agent for the four AOC holders.

In 2006, Manx 2 had made an arrangement with Air Lada (the owner of the two Metro IIIs) to use the aircraft using an Air Operator’s Certificate held by Eurocontinental Air. In 2009, the UK Department for Transport suspended Eurocontinental’s AOC following a series of safety incidents in UK airspace.

Agencia Estatal de Seguridad Aérea (AESA), the Spanish aviation safety and security agency, suspended Eurocontinental’s AoC and then revoked it completely in 2011.

Manx 2 stated that it was unaware of the number of safety occurrences reported because they were not included in the occurrence reporting list.

In November 2009, Air Lada and Flightline agreed to the operation of the two Metro IIIs under Flightline’s Air Operator’s Certificate. Operations and scheduled maintenance of the aircraft were to be conducted under Flightline’s AOC while Air Lada would arrange commercial arrangements and flight scheduling. All maintenance costs were met by Air Lada.

AESA accepted the transfer of the two Metro IIIs from a suspended AOC to a new AOC holder as Flightline had different procedures and management structure. The application provided no details regarding how the aircraft would be used. AESA had no knowledge of Air Lada or Manx 2’s history with the aircraft, nor that Air Lada were the owners of the aircraft. They were also unaware that the two former Eurocontinental Air pilots had moved with the aircraft to Flightline.

Thus, AESA approved the request to include the two aircraft on Flightline’s AOC. The AOC approval forms did not approve low visibility operations for Take-Off, Approach and Landing for either plane. This means that both aircraft were only permitted to operate in CAT I limits.

Once the transfer was complete, in 2010, the two Metro IIIs resumed flying for Manx 2 under Flightline’s AOC, offering the new service between Belfast City and Cork as well as night cargo flights for the Royal Mail. Manx 2 worked directly with Air Lada and there is no evidence of any direct contact between Manx 2 and Flightline. Although there were various documents which referred to the operator, the Operations Manager worked for Air Lada and the address and contacts used were all for Air Lada’s office in the Isle of Man.

The Isle of Man has its own Civil Aviation Administration and a flight to or from the Isle of Man and the UK by a non-UK AOC holder requires permits from the UK and the Isle of Man. The permits were applied for by the Operations Manager of Manx 2 on behalf of Flightline, with Flightline named as the airline.

A new operational procedures document still in draft had only contacts for Manx 2. The Operator’s Quality Manager named in the document was one of the pilots supplied by Air Lada. There was no evidence that Flightline appointed him to this position.

Flightline audited the operation on the Isle of Man two weeks after the new operation started in Ireland. The audit comments include the following:

4. Flight

The meteorology of the Isle of Man in particular, with strong winds and low minimum temperatures, and of England in general, necessitates a different approach to the operation. Both the commander and the co-pilot must be experienced and have a good level of English. Our company should guarantee this. Pilots who are currently operating do not have any problem in this sense.

5. Operational aspects

They must change the normal checklist and adapt it to our company, carry out the pre-flight inspection, sign it and apply the anti-icing system before entering the clouds. It is important to study the [Standard Operating Procedures] well, as well as clearly specifying in its list who is responsible for what and when…

The audit clearly states that experienced crews were required for the scheduled flights. Eight months later, the crew comprised of a Captain with questionable training who had been promoted four days earlier and a First Officer who had not completed his line check.

No further audits were done by Flightline.

Over the three months prior to the accident, there were no pilot reports, defects or maintenance entries made in the Technical Log. The Technical Log for the other Metro III included only two entries in its period of operation, both relating to an ignition problem.

Not a single normal maintenance issue over three months: no lights burned out, no oil top-ups, no defects at all. And no reference to the fact that the pilots were having to manually adjust the power levers to compensate for the engine torque differential. Scheduled maintenance was performed on the plane but, without a report, it’s very unlikely that the crack in the sensor coil could be spotted.

Flightline should have questioned the suspicious lack of defects in two of their aircraft. But for all intents and purposes, operations for the two aircraft were based on the Isle of Man and controlled by Air Lada and Manx 2, while using the operator’s certificate from Flightline in Spain.

Meanwhile, continuing oversight of the operator and its operation fell under the remit of AESA.

In the year before the accident, AESA performed eight audits and inspections were carried out on Flightlines flight operations. The evaluations included how the airline monitors and controls its operations.

However, they were unaware that the two aircraft were operating in the Isle of Man. There’s no obligation for an operator to inform the Authority regarding remote operations, although AESA did state that, had it known that the operation was remote and such a small number of people were involved, they would have taken a greater interest.

In the twelve months leading up to the accident, both Metro IIIs were subject to SAFA (Safety Assessment of Foreign Aircraft) ramp checks in Germany and Ireland. Both inspections included findings and, in one instance, the aircraft was not allowed to depart until the crew cleared the baggage/cargo which was blocking the emergency exits.

During the Eurocontinental Air operation, AESA had sent inspectors to the Isle of Man to carry out an extended ramp inspection. However, as they were unaware that the two aircraft had resumed operations there, no inspectors were sent.

AESA further informed the investigation that ‘in order to have better tools/procedures for proper oversight of a remote operation, EU regulation should require the operators to provide the certifying Authority with a formal declaration stating which are the organizations that ultimately decide the flight’s schedule, routes, crew roster, etc.’

After the accident, the EU Air Safety Committee met with AESA to clarify whether AESA’s surveillance activity of Flightline had provided the evidence that Flightline was capable of adequately supervising its remote operations. AESA stated that ‘they decided to limit the AOC of Flightline to prevent operation of the Fairchild Metro 3s, and that they had initiated the process to suspend the AOC.’

Commission Implementing Regulation (EU) No 390 of 2011 (establishing the European Community list of air carriers which are subject to an operating ban within the Community) stated the following:

Flightline explained that they had entered into a business arrangement with the company Air Lada, not a certified air carrier, to operate two Fairchild Metro 3 aircraft, registrations EC-GPS and EC-ITP, using pilots provided by Air Lada. The Commission pointed out to Flightline that the same aircraft had previously been operating within the AOC of Eurocontinental, another air carrier certified in Spain and that as a result of SAFA inspections and significant safety incidents with the operation of these aircraft, AESA had suspended Eurocontinental Air’s AOC.

The Commission invited the air carrier to make a presentation to the Air Safety Committee and noted that AESA had decided to limit the AOC of Flightline to prevent operation of the Fairchild Metro 3s, and that they had initiated the process to suspend the AOC.

At a meeting on the 19th of October 2011, AESA briefed the Commission on ‘the actions taken to date to address the identified safety issues with Spanish air carriers in a sustainable manner’. Flightline’s AOC was renewed, following corrective actions, but limited to exclude the aircraft of the type Metro III.

In 2010, Welsh ministers working with Manx 2 and FLM Aviation to provide a scheduled air service between Cardiff Airport and RAF Valley on Anglesey. However, the AOC of FLM Aviation, one of the four other operators working with Manx 2, was revoked by the German regulatory authority. Manx 2 continued to sell tickets for the route, replacing FLM Aviation with an air carrier operating under a UK AOC.

In late 2012, Manx 2 informed the investigation that its assets were being sold to a new company as a part of a management buy-out. The new company commenced operation on the 2nd of January 2013 and continued to sell tickets on the route.

The Manx 2 website was updated on 28th of January 2014 with a full statement regarding the final report. statement on AAIB Final Cork report

Manx2 contracted all the flying to EU airlines licensed and required, as was the Operator, to operate in compliance with the stringent standards and controls of the European Aviation Safety Agency (EASA), recognised to be among the most stringent in the world, under the oversight of their national aviation safety authorities. Unfortunately, the report is clear that the prime causes of the accident were decisions made by the Flightline crew in adverse weather conditions, compounded by inappropriate crew rostering by the Operator and a significant lack of oversight by the Spanish air safety authority.

The investigation concluded that the the commercial model of a ticket seller providing an air service is not in the best interests of passenger safety, as the ticket seller has “an inappropriate and disproportionate role with no accountability regarding air safety”. It’s sad to note that in the end, they didn’t take responsibility and the new company that was formed uses the exact same model.

Category: Accident Reports,


  • Ironic to note that the operator was a company called “Flightline”.
    In the early 1980’s this name belonged to an Irish air taxi company which operated two Aero Commander turboprop aircraft. One of these aircraft was involved in a crash under circumstances that never really have been cleared. It quite literally fell out of the sky over the south of England during a night flight from Dublin to France. Aircraft of this class were allowed to be flown by one single crew member and were not required to be equipped with air data recorders. The most likely cause was distraction of the pilot by a passenger. Another passenger in the co-pilot seat may have inadvertently disconnected the autopilot (which in this type of aircraft was not fitted with a disconnect warning), possibly whilst the pilot was dealing with an unruly passenger behind him.
    Probably the pilot, whom I knew as a well trained professional, was distracted and became disorientated when the aircraft entered an “unusual attitude”.
    The aircraft broke up in mid-air and there were no survivors.
    So: there is a very unfortunate link between the name “Flightline”, Ireland and serious air accidents.

    The Manx2 story is different: the story of a flight that now emerges to have been executed by a number of companies that had been unable to meet even the most basic standards of operation, crew training, crew management and maintenance.

    It may even be suspected that the changes in AOC holders and various other entities involved were possibly a (deliberate ?) smoke screen to prevent close monitoring by the aviation authorities.

    In my previous comment I mentioned that, as a former professional pilot (Dutch, Irish and Swiss ATPL, Nigerian SCPL and US CPL with over 22.000 hrs TT) I hold the Irish Aviation Authority as well as the UK CAA in high professional regard. The operators seemed to have been flying out of sight from the Spanish authorities. I have not had any direct involvement with them but I have no doubt that they are equally professional.

    I flew the Metro II in the 1980’s out of Shannon. This aircraft had been cleared for single-crew operations and I have on numerous occasions flown it as such during night time cargo operations, even in winter time and on icy, partly snow-covered runways.
    The same aircraft was also used as a passenger carrying schedule between Shannon and Dublin. During passenger operations we flew with a co-pilot. This airplane was a new addition to the fleet and I flew it as captain from the beginning. The engines were problematic to start, prone to overheating (“cooking”)if the start cycle was not properly carried out.
    The Metro also had a relatively short wing span and as a result had a high wing loading. It could be a handful to fly if not properly handled. It was equipped with a stability augmentation system to assist control at low speed. But if flown properly, if was a lovely aircraft.

    A few general remarks:
    A complex ownership structure of an aircraft is nothing unusual. A bank or wealthy individual may have invested in financing it but then buries this in order to avoid being held responsible in case of an accident. It is a legal and accounting set-up and need not necessarily point to anything sinister.

    Determination of weather minima is a bit complex. In a properly run airline it will all be laid down in an “Operations Manual”, copy of which must be on board and within reach of the pilots.
    But the flight in general knows two phases: flight planning and flight execution.
    During the planning stage, all weather criteria must be met including cloud base and weather at alternative airports. Weather at the alternates, depending on the available approach aids, must be a certain increment above the minima.
    Once take-off roll has commenced the aircraft is in the “execution” stage and raised alternate minima and cloud base at the destination ceases to be a factor when it comes to continuing the flight. Which is not to say that the crew may ignore it.
    In most jurisdictions and also depending on the experience level of the crew, all minima must be met as long as the flight has not passed the “outer marker” or an approach fix.
    The visibility for category 1 ILS approaches is generally 550 meters RVR and this must be met at all times.
    Likewise the DH: Once past the marker the aircraft may continue even if the cloud base is reported to be below minima but still may not descent below this height UNLESS the crew has sufficient visual clues to continue. This is mandatory. The crew appeared to have broken this rule and paid dearly for it: with their life and that of some of their passengers.

    A “loaner” engine: an engine rented from usually a maintenance company to prevent down-time whilst the aircraft’s own engine is in overhaul or repair.
    It is relatively common and there is nothing wrong with it in principle.

    “Split” power: There should not be a large divergence between the position of the power levers. They are handles on the pedestal between the pilots and used to control the engines. If there is a large difference between them for the same amount of power, the aircraft can experience handling problems. Especially under conditions of high workload like during an approach to weather minima.

    “Negative torque”: the power has been retarded to a point where at a certain airspeed the propeller no longer provides power but drag.
    When there is a significant “split”, one engine can deliver power whilst the other provides drag.
    At low speed this can lead to a situation where the aircraft will start a roll which, especially close to the ground, may well become uncontrollable.

    The Manx2 operation would seem to have been executed far below any acceptable standards.
    In the report I read that the captain had a total time of 1800 hours of which 1600 on the Metro. This seems to indicate that this pilot, after gaining a CPL, had flown only for Flightline. He probably had no experience outside that company and may well have been unaware that he was in fact part of a substandard operation.

    No pilot (leave 9-11 out of it) departs with the intention to crash the aircraft and kill himself and his passengers.
    The crew that were involved in this sad accident were victims, as well as their passengers, of an “airline” operating at such a low, shoddy level that they had no right even to be involved in a push-bike rental scheme.

  • Very useful information, Rudy. Thank you for taking the time.

    I wasn’t sure about the loaner engines to be honest – the report makes a bit of a fuss about them but by far the bigger issue to me is the lack of thorough ground checks (which would have caught the variance) after replacing an engine.

    I concede that the primary cause was to continue descending in an aircraft not rated for CAT II below decision height with the airfield in sight for three approaches (most airlines only allow two attempts in low vis, as far as I know). But the beta mode followed by full power would still have taken the Pilot Flying by surprise and the right roll could equally have taken them down from 200 feet for more. And as you say, lack of training and lack of operational oversight was the real killer here.

  • Sylvia,
    I kept the technical stuff a bit simple because I do not know if everyone is familiar with technical terms like “beta mode”.
    Basically this is a condition where the propeller remains in a very fine pitch for ground operations. It is also known as “ground fine” pitch and, as the term says, intended for ground operations.
    Some engines, like the Garretts on the Metro, are single-spool. The compressor, turbine and propeller – through a reduction gear – are all mounted on the same axle.
    If the aircraft moves at low speed like during taxi and the propeller remains in flight mode, this will cause a lot of air drag on the rotating blades and the engine will very quickly burn itself out.
    “Beta” mode must be selected during taxi.
    The Rolls-Royce Dart engine was single-spool and ground fine (Beta) was so critical that at the end of roll-out after landing the propellers were forced into ground fine by the captain applying the gust lock e.g. in the Fokker F27.
    Other engines are not so critical: e.g. the P&W PT6 which has a free turbine driving the propeller.
    In the context of the Cork accident the above is important: One engine apparently was in “beta” mode when the aircraft had made an (illegal) descent below DH. For Cat 1 operations this is usually 200 ft above the runway.
    Indeed, lo-vis approaches are discouraged and in the Operations Manual of many carriers not even permitted more than twice in succession.
    Cat 2 operations may descend usually to 100 feet.
    But a crew, even if Cat 2 qualified, may NOT PLAN a flight under cat 2. In other words, the weather forecast (TAF) must, during the flight planning stage, indicate weather conditions at least allowing a cat 1 approach. Only when the flight has commenced (in the operational stage) may the flight continue to cat 2 minima.
    There are a lot of “strings” attached to cat 2: the carrier must be approved, the aircraft must conform to a lot of operational equipment including a sophisticated autopilot system and the crew must not only have undergone special training but also must have kept it up, either by a certain prescribed and logged number of practice approaches in the actual aircraft or in an approved simulator.
    The radar altimeter may never be used to determine a DH during Cat 1 operations, but does come into play during Cat 2.
    This also means that special approach charts must be used.
    It will be clear that the crew really were unable to adhere to a recogniseable standard of operating an aircraft under IFR in a scheduled passenger-carrying operation.
    The mis-adjusted propeller in my considered opinion played a major role in this accident:
    The captain, who seemed to be playing the role of “autothrottle” for the pilot-flying (who was probably not up to this role and probably already under severe stress after two missed approaches) did not ascertain that the PF knew what he was doing.
    At low altitude one propeller went into “beta” mode, intended for low-speed ground operations only. This propeller started to produce drag.
    When, now at a very and critically low altitude, the captain finally decided to “go around”, one propeller produced full power, the other was perhaps just coming out of “beta” but still produced drag.
    The result was predictable: asymmetric power beyond not only the co-pilot’s ability to handle but also outside the aerodynamic capability of the aircraft.
    Manx2’s comments as reproduced here? If I give my opinion on that outfit I will probably be sued for libel !
    So let me restrict myself by stating only that most of the passengers were probably of the opinion that this was an airline, not just a ticketing office. This impression reinforced by their website, the aircraft livery and uniform.
    Manx2, in my opinion, reneged on a statutory obligation to their passengers to at least audit their “vendors” on procedures, licensing and standards.

  • As an unqualified aviation geek #avgeek, quite a bit of this report goes over my head. But one thing that surprised me was that during the incident the control of the aeroplane appeared to be shared.

    I presumed that you were either Pilot Flying or Pilot Not Flying – you were either in control of the aircraft, or not in control. The captain, here, seems to be half in control by moving the throttle levers. To me, that seems as odd as a car driver allowing a passenger to steer….a recipe for disaster?

  • maidbloke,
    You are quite right. The PNF moving the power levers is not common practice. But on the other hand, it is not entirely unheard of either. During a well-flown precision approach (assuming little or no wind as is common during lo-vis conditions), the power should not need much adjustment. When in the 1980’s Air France flew Fokker F27’s internally in France on night mail flights, they used to carry a third crew member in the jumpseat who manipulated the power. This allowed them even to make cat 3 (!) approaches. However, the crew were experienced and had received special training. Also, some modifications had been made to the aircraft flap warning (during cat 3 they were not allowed to select full flaps) and special autopilots were fitted.
    In the case which is the subject of the discussion, the crew members were, so it seems, not very well trained. The operator’s identity was even obscure. The co-pilot was not even very experienced and yet they elected to make an approach well below limits, not once but three times.
    And that in an aircraft that had mis-matched power levers. Reading the report, I am getting the impression that the controls for one engine might not even have been set to have met the requirements regarding engagement of “beta” power in the air.
    Or, in simpler terms, even with the captain setting the power for the co-pilot (who was PF), one engine could too easily have gone into beta (ground mode)during flight.
    An extremely dangerous mechanical fault, as proven by the outcome.

  • PS:
    Normally, especially during cat 3, of course most modern airliners use autpilots, autothrottle and autoland during the approach.

  • The question in my mind is this: imagine the pilots had made a legal and sensible approach and levelled out at 200 ft to the MAP, with the PNF controlling the throttles and the left engine in beta mode. When he advanced the throttles to TOGA (or full power), would the PF have been able to control the roll?

  • Cliff,
    Even from 200 ft I doubt that there would have been enough room or time to counteract the roll.
    The Metro has a short, stubby wing. It has a high wing loading. The engines are relatively powerful but at or a little bit above Vref, unless one propeller has been feathered, the asymmetric forces would have been strong with one engine at TOGA power and the other one delivering drag.
    This may have exceeded the ability of the rudder and ailerons to control the aircraft at low speed. Add to this the fact that his crew were in their third attempt with a relatively inexperienced FO as PF. They must have started to suffer from fatigue and were possibly even a bit disorientated.
    Anyway, insofar as I read the report, the crew flew an ILS approach down to below even CAT 2 minima before making an attempt to go around.
    An ILS approach is a precision approach and the approach may be continued down to DH. At that point, if no visual cues like runway or approach lights are in sight, the crew must follow an immediate “missed approach” procedure.
    A precision approach does not allow for a level segment to a MAP, that is for a non-precision approach. In the “old days” we used to do this: after the initial fix we would descend to the minimum approach altitude, sometimes with a step until past the FAF and level off until the MAP. Nowadays with more accurate approach aids even a non-precision approach is more often than not flown using a pre-calculated rate of descent at as accurate and constant an angle as possible.

  • Oh, btw:
    Studying the photos and judging from the marks on the runway that show the path of the aircraft in it’s final moments, it would seem that the aircraft had been perfectly aligned with the runway.
    Had they just continued blind, pulling the power all the way back at the 40ft altitude from which they tried to make a go-around, they possibly would not have suffered more than a “hard landing”. The aircraft might even have stayed on the runway and all on board would have been able to walk away from the incident.
    Cruelly sad !

  • Rudy, of course you’re right about the precision approach not allowing for a level segment to a MAP. I blame the whisky for my silly mistake.

    My point was to question whether the roll induced would have been recoverable from 200 ft. However, thinking about it sober, I don’t think they’d ever have throttled back that far if they’d been flying to a 200 ft DH as they’d have been flying the glide path. So the roll and thus the accident wouldn’t have happened from 200ft.

    I agree with your statement that, having pulled the power all the way back, they may well have ended up on the runway. But that would have required a decision to live through a survivable incident and I don’t believe they ever saw that as an option. They just couldn’t believe there was a 20 ft cloudbase.

  • Cliff,
    I’d say you are probably right about that.
    Assuming that the flight was settled and stable on the ILS the aircraft would have needed but very small power adjustments. And therefore they probably would have had positive torque on both engines, neither one would have been in beta mode. They should have been able to make a go-around.
    But on the other hand, in this sad case, they were probably tired and confused. I cannot remember having read anything about their previous duties that day. When did they start, how many segments had they flown already. Had they been on duty perhaps for a long time, waiting for an improvement in the weather?
    Read reports of other accidents. They need not even all have to be air accidents, but you will find that fatigue will adversely affect the thinking and decision making process.
    I have been very harsh in my comments about this operation because from what I read I gather that they did not seem to meet anything even close to acceptable standards.
    The way Manx2 in their comments distanced themselves and washed themselves clean of responsibility adds to my feeling of discomfort.
    I suspect that the crew were poorly trained and never really had been exposed to an airline that maintained any professional standard to speak of.
    But on the other hand, as professional pilots they would have known some basics. Some airlines that are big enough have a department that publishes dedicated route manuals tailor-made for their operation.
    Nearly everyone else uses Jeppesen.
    The minima are clearly published on their approach plates and nearly everything a pilot may need to prepare and execute a flight (under IFR and in IMC) can be found in these manuals.
    In modern aircraft they can even be displayed on the navigation equipment, but normally the Metro is not equipped with that.
    Pilots have to do an annual IF renewal.
    Their licences would have been issued after passing a test that normally is conducted by examiners who are so authorised by a state civil aviation authority or under the umbrella of the JAA. They would have known about weather minima.
    They also would have known that ignoring them constitutes an offence which ultimately could have led to fines and withdrawal of the privileges of their licences.
    So they had no excuse for their decision to – repeatedly – ignore the rules and regulations.
    If e.g. a coach driver decides to ignore the speed limits and drives at high speed into dense fog, he will be held responsible if this leads to an accident.
    Pilots who ignore the weather and fail to adhere to safe standards also are ultimately responsible for their actions, even if the operator is applying pressure to ignore safety.
    It gives me no pleasure to be so harsh.
    But reading the reports as published by Sylvia makes my hair stand on end !

    I think we have well and truly exhausted this subject.
    I am looking forward to read what this site will come up with next.

  • Hah, no pressure, right? ;)

    Rudy, you are right about fatigue. Flightline also had no oversight of the real roster, as opposed to the submitted one, and the First Officer was supposed to have a rest day the day of the flight. Neither pilot was well rested and although it was two short flights that morning, they were hand-flying, no autopilot. That’s enough to wipe anyone out, let alone after two difficult approaches in low vis.

  • “I don’t know what they call this mess, but it’s not an airline!” – reportedly what one of the investigators said, and I agree!

    Thanks for the heads up about the airline that replaced Manx2 I’ll make sure I give them as wide a berth as possible! 0.0

  • As an afterthought:
    It has happened to most pilots who started their career on “steam driven” aircraft or, to put it in a more correct way: aircraft of an older generation with conventional systems, The now mandatory innovations like the much increased VHF radio frequencies, FDR, CVR and GPS navigation systems often are a retrofit. Unlike the last generation aircraft, the autopilot is not always considered an essential part of the systems.
    A malfunctioning autopilot can simply be disconnected with a sticker saying U/S near the “engage” switch. And stay that way for a very long time.
    The old Air France Fokker F27 had a special autopilot for Cat 3 operations, not the standard one normally fitted. That system was no longer supported by the manufacturer and, like Windows XP on your computer, if it was broken it remained broken.
    Air France operated under special rules that were not transferable to other operators so, after the aircraft were withdrawn, replaced by jets and sold off to cargo operators they were only certified to Cat 1. For which a serviceable autopilot was not required.
    Flying (maximum) 4 sectors at night, without autopilot, was very demanding and we would roll into our bed (daytime !) when back.
    In all fairness, the company did not ever roster pilots with insufficient sleep. And it has happened that we were delayed because fog at our final stop. The operation was from Germany to Rennes with an intermediate stop at CDG. Rennes was in fog and we had to wait at CDG for the fog to lift. The company sent a Navajo with a relief crew from Rotterdam to Rennes. This is the way a professional company should operate. We may have been flying an old “banger of the skies” but there was no compromise when it came to adherence to the minimum standards. And the company had a very good safety record as a result.

Post a comment:

Your email address will not be published. Required fields are marked *