Twin Training Troubles – Twice

This discussion is to promote safety and not to establish liability.

CAA’s report contains padding and repetition, so in the interest of clarity, I have paraphrased extensively.

Aircraft registration:                   ZS-FET

Aircraft owner                           Cape Aero Club

Date and time of accident:         25 April 2006; 1200Z

Type of aircraft:                         Beech D95A

Type of operation:                     Training

PIC license type:                       Airline Transport

License valid:                            Yes

PIC age:                                   62

PIC total hours:                         9300

PIC hours on type                     Unknown

Last point of departure:              Cape Town (FACT)

Intended point of landing:          Cape Town (FACT)

Location of accident site:           Atlantis – near Cape Town 316 ft.

Meteorological information:        Fine

POB:                                        2 + 1

People injured:                          0

People killed:                            3

Pilot undergoing conversion training.

Nationality                    Sudanese

Gender                         Male

Age                              21

Licence Type                Commercial

Licence valid                 Yes

Type Endorsed             No

Ratings                        Night

Total Hours                   232

Total Past 90 Days        5.4

On Type 90 Days          5.4

Total on Type               5.4

Synopsis

The instructor and pilot accompanied by a passenger (also an instructor) took off on a training flight from Cape Town. The training to be conducted was asymmetric circuits and engine failure during a missed approach.

The flight was to be FACT – Delta 200(FADX) – Fisantekraal (FAFK) and back to FACT. The aircraft wreckage was found approximately 5.4 nm north-east of FADX. There was no post-impact fire. There were no eyewitnesses nor was there any emergency call.

Cape Town ATC radar recordings show the aircraft in and around the Atlantis/Philadelphia area until it disappeared from the radar when it went below radar altitude detection at about 3 800 ft.

The pilot’s training file showed he had completed all his conversion training exercises, and they were signed out by the instructor. The engine failure during the missed approach was the only exercise left to complete the type rating.

Examination of the wreckage revealed that the aircraft was rotating in an anti-clockwise direction on impact. The nose-down pitch angle at impact was approximately 15º. There was no horizontal movement of the aircraft, and the wreckage was not scattered.

The indication on the turn-and-bank [sic. – to be pedantic it is a turn-and-slip indicator] instrument showed that the aircraft was in a left turn.

The flap lever was set at 20º and the landing gear was extended. The control column was in the aft position with the yoke broken. The left-hand wing was destroyed, but the right-hand wing was still intact with wrinkles on the top surface.

The propellers were embedded in the ground. No sign of rotation on either propeller was evident.

The planned exercise was for the pilot to demonstrate the capability of maintaining asymmetric climb with one engine inoperative. During the performance of the missed approach procedure and during climb – with the aircraft in the landing configuration – the instructor would simulate an engine failure by feathering one engine – the pilot is supposed to correct and maintain the flight.

The onsite evidence suggested that the aircraft entered into an anti-clockwise spin before crashing in a nose-down attitude, fatally injuring all onboard. The intention of the flight was to provide training as to how to conduct and maintain asymmetric climb of the aircraft during a missed approach procedure.

The possibility exists that the instructor feathered the left engine. The effect of this would have been a yaw of the aircraft to the left. If the pilot under training does not initiate the appropriate action or is not yet proficient in maintaining control of the aircraft, a spin can result. This would require considerable height to affect a recovery to normal flight.

By the nature of the exercise being practiced, the aircraft could have been at a low altitude, which would have precluded the instructor from being able to recover the aircraft to normal flight.

Jim’s Comments

This is almost a carbon copy of another Cape Aero Club accident that happened ten years previously. The CFI, Ray Grinstead, was testing a candidate on this identical exercise in a King Air 90. They also spun into the ground with a third person in the back.

Interestingly, the families of the King Air crash sued the Department of Civil Aviation, claiming that Grinstead, as the testing officer, had caused the accident, while representing the DCA.

The Supreme Court of Appeal ruled that Grinstead may have erred in his judgement, but that did not amount to negligence and there was insufficient evidence to show that Grinstead had caused the accident. Further there was no merit in the argument that the Minister was vicariously liable for the damage, as there was no contractual relationship between Grinstead and the Department of Transport.

Having said that, I would not have been keen to sit in the back on either of these flights. They involved a training exercise that calls for sharp reactions – there is no room for errors and no back doors. Get it wrong and that’s the end of you.

But the worst of it is that the CAA approved of this exercise – in fact required it for the rating. To me this is criminal. Neither of the aeroplanes is capable of doing a go-around on one engine, in the landing configuration. The current Regs do not call for you to do it with the engine feathered. Even zero-thrust settings demand considerable finger-outness.

Certainly try it at altitude, at least 6000’ AGL, and note the rate of descent – not climb. This demonstrates that the manoeuvre is not within the climb parameters of that aircraft.

But beware, by doing asymmetric work at altitude you introduce another danger. The reduced engine power means that your VMCa (minimum control speed) is reduced, and this means that you may stall with full power on one side and none on the other. VMCa is normally faster than stall speed. So if VMCa is reduced to stall speed, then the two can happen simultaneously. This will almost certainly cause a spin.

Unfortunately, the report can’t tell us at what altitude they were practicing the exercise. Only that they lost contact at 3800 ft. We don’t know whether the aircraft was spinning at that stage, or only later.

I also have no idea what happened in the cockpit, but one, or perhaps both, of the crew made a critical mistake from which there was no room to recover. We all make mistakes – the trick is not to do so without a back door.

I have only flown the Travel Air a couple of times, and that was long ago. So I don’t have a detailed impression of what it was like. Briefly, I remember it as having Beech quality, but being underpowered and heavy.

This means that with three up and say half tanks, you would not expect it to climb away from a go-around, on one engine, with the gear down and some flap. But that’s what the exercise called for.

I don’t have the graphs in front of me, but you can be certain that pretty much all light aircraft will go downhill during such an exercise. If you try to prevent it from losing height, you will run out of airspeed, lose directional control and spin.

Sadly this is what happened.

A quick reminder about my sheep story. A CAA flying inspector once asked me what I would do if I were asymmetric on final approach in a light twin, when a herd of sheep taxi into the runway. I replied that I would take full power, retract the gear and flaps and do a go-around.

His reply was emphatic, “Wrong. The correct answer is, ‘Sir, I would start killing sheep.’”

Very wise words that all twin pilots should remember.

What Can We Learn?

  • No matter what you are doing with an aircraft, make sure you always have a back door. In this case, height – lots of it – could have saved the day.
  • Don’t ask aeroplanes to do things that are outside their published performance parameters.
  • Live Cowards’ Club members never shut down an engine, below 1000 ft agl. We use zero-thrust, and great caution.
  • Instructors: never be in a position where you can’t rescue a student’s mistake. Asymmetric work in a light twin leaves little room for either of you to stuff up. Make sure you have your finger well out before tackling any asymmetric training.
  • You need TWO very specific briefings – one before takeoff and the other immediately before each exercise.

Travel Air specs:

  • Crew: 1
  • Capacity: 4 passengers
  • Length: 25 ft 11 in
  • Wingspan: 37 ft 10 in
  • Height: 9 ft 6 in
  • Wing area: 199.2 sq ft
  • Aspect ratio: 7.16:1
  • Empty weight: 2,555 lb
  • Max takeoff weight: 4,200 lb
  • Fuel capacity: 112 US gal (93 imp gal)
  • Powerplant: 2 × Lycoming IO-360-B1B 180 hp
  • Propellers: 2-bladed Hartzell 8447-12 constant-speed, 6 ft 0 in diameter

Performance

  • Maximum speed: 180 kts at sea level
  • Cruise speed: 169 kts at 11,000 ft (65% power))
  • Never exceed speed: 210 kts
  • Range: 884 nm (1,017 mi) (50% power)
  • Service ceiling: 18,100 ft
  • Rate of climb: 1,250 ft/min
  • Takeoff distance to 50 ft: 1,280 ft
  • Landing distance from 50 ft: 1,590 ft
  • Rotate speed Vr: 74 KIAS
  • Minimum SE Control speed: 73 KIAS
  • Best rate of climb speed: 91 KIAS
  • Best SE rate of climb speed: 87 KIAS
  • Never exceed speed: 208 KIAS
  • Stall speed clean: 70 KIAS
  • Stall speed: (flaps 28º, gear down, zero thrust) 61 KIAS