Appalling Accident Reports
Aircraft accident reports are key to the aviation authority’s drive towards making the skies safer by allowing pilots to learn from others’ mistakes. Thus the pre-amble of each CAA accident report states “… this report was compiled in the interests of the promotion of aviation safety ...” This is simply not true.
I am informed that this critical job is being performed by junior staff – few, if any, of whom hold any sort of pilot licence and therefore cannot be expected to have any credible insight into the real causes of accidents. The result of this lack of skills and insight is that the CAA’s accident reports are an embarrassment, both to the CAA itself as the issuing body, and to the broader flying community at large. The system makes a mockery of flight safety.
We are told that the dismal state of the CAA’s accident reports is caused by a vacuum at the top of the Accident and Incident Investigation Division. There seems to be no greybeard to monitor and mentor the investigators. The CAA is consequently publishing reports without them having being vetted by anyone with a sound knowledge of accident investigation – and the necessary competence in written English.
If the CAA is to produce accident reports that make a meaningful contribution to flight safety, they need to recruit experienced specialists. The CAA should have at least one full-time, experienced, investigator who will oversee the apprentice investigators. Ideally, they should also have a reserve of specialist consultants whom can be called upon when needed. These could be retired pilots and engineers who don’t need to hold a current licence or medical. People whose years of experience could be used to guide the team on the road to becoming a cohesive and productive body.
Here are some examples of the CAA’s poor output:
Case 1: A Piper Arrow crashed at night two miles short of the runway at Lanseria, killing the 400-hour pilot.
The report says:
It was a training flight. It was not.
The aircraft had a total time of 2,200 hours.
It was actually 6,800 hours.
The prop had done 360 hours SMOH.
It was actually 252 hours.
The accident happened at 4,400 ft.
No, it happened at 4,735 ft.
The aircraft hit the ground “at a high rate of descent and high forward speed” 2 nm short of the runway.
This means that it lost 600 ft in a very short time, yet the report did not consider the possibility of an engine failure.
The PAPI and runway lights were at only 10%.
In his/her analysis, the investigator failed to consider this as a possible cause, or contributory cause, of the accident.
The CAA used a C172 to replicate the approach.
What did they do this for? I ask because they failed to mention their findings of this flight.
Cause of the accident: The pilot was low on the approach for landing and collided with high tension wires.
Come on CAA – that is not the cause of the accident – that’s a description of the final stage of the accident. The readers need to know WHY this happened. Was there an engine failure, or a problem with the lights being set at low intensity, or what?
NOTE: The CAA ‘lost’ the file and took two years to complete this disgraceful report.
Case 2: A Cessna 210 lost directional control and crashed into a tree to the left of the runway during takeoff from a farm strip near Cullinan (this is reviewed in this month’s accident report). The 110-hour pilot had not completed a conversion to the type. He, and one passenger, were killed in the ensuing fire and another was badly burned. The report says:
The pilot’s hours on type were unknown.
How can this be? He had done four training sessions at a registered ATO. Did they have no record of these hours in their autho-book, or their accounts department, or the aircraft logbooks, or the flight folio?
The majority of the pilot’s previous flying was on a weight-shift microlight.
This has the throttle where the Cessna’s right-hand toe-brake is, and both brakes where the Cessna’s left-hand toe-brake is. A seasoned investigator would have realised that, as the aircraft started pulling to the left, under ‘P’ effect, the pilot may well have released pressure on his right foot (thinking he was throttling back) and pressed on the left-hand toe brake (thinking he was braking both wheels). In fact, one witness saw the left wheel skidding while the aircraft was trying to accelerate. Unfortunately, no seasoned investigator was consulted.
The ATO converting the pilot had no paperwork on the training.
The ATO is not censured for its casual lack of conformity.
The pilot had never before flown an aircraft with fuel injection, cowl-flaps, retractable undercarriage, turbocharger or a constant-speed prop.
The investigator failed to consider this as a contributing factor to the accident.
The witness, a pilot and the front-seat passenger, said that when he saw them heading off the left-hand side of the runway he applied LEFT rudder.
Really? Would a qualified pilot have made that mistake, or is this another example of careless reporting?
Case 3: A Piper Aztec, which had no maintenance records for the past six years, crashed immediately after lift-off at Wonderboom. The 600-hour pilot, who was not rated on the Aztec, or any other multi-engine aircraft, survived but was injured. Both fuel selectors were on the auxiliary outboard tanks, which were empty. Immediately after getting airborne, the aircraft crashed to the right-hand side of runway. It then skidded along the ground for 33.3 metres before it came to rest. The report says:
The aircraft veered to the right of Runway 29, and entered into an uncontrollable spiral dive.
Did this really happen just after getting airborne? An experienced investigator would say the pilot lost directional control immediately after lift-off when the right engine stopped producing power. This is not a spiral dive – in a spiral dive the aircraft would have hit the ground near vertically, instantly killing all on board. A totally different thing.
… both fuel selectors were selected to the outboard main fuel tanks ….
This is rubbish – the Aztec does not have outboard main tanks. The inboard tanks are clearly marked as the mains, and the outboard tanks are the auxiliaries. The investigator makes exactly the same mistake as the pilot whose crash they are investigating.
The pilot was not rated on this aircraft, nor is there any evidence of him ever having flown the type before the accident. The pilot therefore did not understand the fuel system.
The first point is quite correct. However, a competent investigator would realise that a licensed pilot, who owned and operated his own AMO, would be very familiar with fuel systems. Besides, the Aztec’s fuel selectors are extremely simple and clearly labelled.
Case 4: A V35B Bonanza was on an IFR flight from Cato Ridge to New Tempe when it flew into the cloud-covered Drakensburg Mountains at 8,878 ft AMSL. That is almost 4,000 ft below the Minimum Safe Altitude. The pilot and two passengers were killed instantly. The report says:
The pilot had 1,494 hours and they were all on type.
No, this is wrong because the report goes on to say that in the past 90 days the pilot had flown 68 hours, of which 47 were on type. Further, he owned a number of different types of aircraft which he few regularly. So, during that brief period, he had flown 21 hours on other aircraft. This is just careless reporting and should have been picked up by even the most junior investigator.
… the Minimum Safe Altitude would at least have to have been 12,700 ft AGL.
Another sloppy mistake – they mean AMSL. This accident is all about altitude, so you would hope an inspector could get the basics right.
Case 5: A low-hour pilot flying a 180 Cherokee from Potchefstroom to Wonderboom became lost in visibility of less than 5,000 m. Eventually he was running short of fuel and daylight, so he did a precautionary landing in an open field. The aircraft was damaged when it hit two ant-hills. Neither the pilot, nor his passenger, was injured. The report says:
After flying for approximately 50 minutes … the pilot opted to perform a precautionary landing.
That makes no sense. He could not have been flying for 50 minutes because he covered around 180 nautical miles. That would give the Cherokee a cruising speed of 216 knots – roughly 100 knots faster than normal. Once again, a competent investigator would have spotted this immediately.
Probable Cause 1: Unsuccessful force [sic] landing.
The reported stated, just a few lines earlier, that it was a precautionary landing. And what was unsuccessful about it?
Contributing factors: Lack of situational awareness.
The investigator doesn’t appear to understand the concept of ‘situational awareness’ as used in aviation. A pilot could have great situational awareness, and still not know his exact position – for instance, while crossing a desert or an ocean. Equally, he could have very poor situational awareness while knowing his exact position on the map.
The pilot was uncertain of his position. At no stage during the flight did the pilot look, think and act upon the situation.
That is plain nonsense, because the report also says, “He then entered the VOR frequency for HBV but could not determine his position in relation to HBV. He then contacted Johannesburg Information to ask them to help him determine his position.” So the pilot did, indeed “look, think, and act upon the situation”.
It is unknown to the investigator as to how the pilot passed the Pretoria/Johannesburg area without making any positive reference to the ground.
What does this sentence mean? That the investigator did not ask the pilot how this happened? Or does this mean that the investigator doesn’t understand the concept of limited visibility?
The pilot depended entirely on the GPS for the flight, and only reverted to a map when he was already uncertain of his position.
The investigator should know that although this is not ideal, it is common practice for pilots to use GPS as their primary navigation tool, and use maps as a backup.
Case 6: A 190-hour Cessna Cardinal (C177) pilot lost elevator control while on final approach for Runway 29 at Bethlehem. The pilot used power to control pitch and the aircraft was only slightly damaged. All three occupants were unhurt.
The report says:
The right-hand control cable had failed …
What does that tell us? I had to do a lot of phoning around to establish that it was the ‘down elevator cable’ that failed. Nowhere does the report mention what ‘the right-hand control cable’ does, leaving one with the impression that the investigator did not understand its function.
The cable failed … due to corrosion caused by a leakage of battery acid.
No, the battery acid did not ‘leak’; it was routed there by an incorrectly installed battery-box drain.
In the interest of aviation safety, it is recommended that the AMO should be audited regarding maintenance practice in terms of scope of work and approval certificate.
So, the CAA’s report recommends that the CAA does its job. But the CAA had done audits and given the AMO a clean bill of health.
The AMO inspected the cable but failed to notice the problem.
They also failed to notice that the battery-box drain was immediately above the cable, instead of being routed out through the bottom of the fuselage. This is just rubbish. If the cable has been inspected the problem would have been noticed immediately. A competent investigator would have known this. It seems that the report failed to censure the AMO because such criticism would have reflected on their own inadequacy.
A full-page diagram is included in the report.
Unfortunately, this diagram has nothing to do with the accident. It does not show the battery box, the breather or the cable. It is totally irrelevant. One can only imagine that either the investigator did not know what he or she was doing, or they inserted it to pad out the report.
Case 7: During takeoff from Kitty Hawk, the Zlin’s [there is no mention of what type of Zlin] engine spluttered and started emitting dark smoke. Soon afterwards the aircraft started to turn back towards the field, then spiralled, or spun, into the ground. Witnesses heard no engine noise. The pilot, the sole occupant, was killed on impact.
The report says:
The initial investigation revealed that the cause of the accident could be attributed to the impact with the ground at a high angle of impact following a stall during a turn.
Impacting the ground is not the cause of the accident – that is the accident. Surely all aircraft that crash eventually ‘impact the ground’.
These are just a few samples of the Accident and Incident Investigation Division’s disgraceful output.
One can only hope that our Civil Aviation Authority will be motivated to immediately make arrangements for qualified and experienced oversight of the reports before they are distributed for public ridicule.