Causal Factors of the Piper Malibu Crash near Guernsey
Sometimes it seems like these posts get longer every year that I keep doing this. This collection of details from the AAIB report that I left out of the Piper Malibu crash near Guernsey is almost as long as the original post. But there are a number of aspects that are interesting to look at in more detail. I’m afraid this post won’t make much sense without having read the previous one but I promise the rest of the posts this year will be stand-alone, succinct and to the point (a promise I only ever dare make in December).
Unfortunately, as you can see from the comments last week, it is simply impossible to know what was actually wrong with the plane.
The pilot reported four separate issues that weekend:
- A bang accompanied by a low mist
- Loss of pressure in the left brake
- Spurious stall warning that continued during taxi
- Engine oil leak
I should clarify that last week I wrote “low-level mist” but looking again, the pilot said “low mist” not low-level. The left brake was looked into on the Monday and the engine oil leak was dismissed as not an issue. The pilot dealt with the stall warning by pulling the circuit breaker as they taxied to parking at Nantes. He never mentioned the issue again so it isn’t clear if resetting the circuit breaker fixed the issue or if he left the stall warner disabled. As the final manoeuvres didn’t include a stall, it wouldn’t have made much difference anyway. That leaves us with the bang and the mist.
The AAIB could not come up with a clear answer as to what might have caused the sound which startled the pilot, let alone the low mist. This is partially because the pilot wasn’t consistent in his descriptions.
After he arrived, he said that he’d heard a loud bang accompanied by a low mist halfway across the channel. However, when he spoke to the mechanic about it the following day, he described it as a thud and said that it had happened on the approach to Nantes, with no reference to the mist. Initially, he said he’d experienced the low mist before; however there was no record of this in the journey log and no one else who flew the plane reported having experienced this.
To notice a “low mist”, the pilot would have had to either look down into the footwell or back into the cabin. With temperatures at freezing outside, the cabin heat control would have been open to keep the cabin warm, including conditioned ram air blowing out of the foot warmers onto the pilot’s feet. One theory is that he never saw a mist but rather that he felt a coldness around his feet as the temperature of the heated air briefly dropped.
The investigation focused instead on the possibilities for the loud bang. No evidence was found of a bird strike or a mechanical failure of the engine. His description didn’t match a change of the undercarriage setting. If the exhaust system had failed and the mist was actually smoke, then the issue would have continued for the rest of the flight to Nantes. A cabin pressurisation issue is possible but the pilot never mentioned using the cabin pressure or losing it. The wreckage is such that it isn’t possible to know if the cabin was pressurised for the return journey, let alone verify whether it had failed on the outbound flight.
It could not have been a failure of a turbocharger turbine as the engine was operating normally on the ground at Nantes; however, an interesting possibility is that one of the turbochargers had become detached. If that were the case, then some of the exhaust gas would pass through the common duct and into the turbine section of the turbocharger, with the result that the engine wouldn’t function at maximum capacity in terms of boost and rpm. But the pilot may well have felt that the engine was still running normally and not considered that there there was an ongoing issue.
As he never reported finding any issues (for example, did the cabin pressurisation still work or whether he reset the circuit breaker for the stall warning), it is very difficult to work out what went wrong.
I wrote briefly about the engine in the accident aircraft: a Teledyne Continental TSIO-520-BE engine: fuel-injected, twin-turbocharged, air-cooled, horizontally-opposed, six-cylinder engine. The turbochargers, a combination turbine and air compressor, take in air at higher pressures which means they can burn more fuel and produce more power than a regular engine. The engine oil is cooled by ram air, external air which is “rammed” into the engine through a forward-facing inlet, so that as the airspeed increases, the ram air pressure is increased. Both turbochargers were replaced in 2017.
There was no technical log on the aircraft and, as an aircraft for private use only, there was no requirement to have one. The aircraft had a journey log which recorded the destinations and hours flown. The person who managed the operation of the aircraft said that he did not know of any faults on the aircraft other than those that the pilot reported after arriving in Nantes.
Piston engines produce high concentrations of carbon monoxide. Carbon monoxide is a serious hazard because it is an odourless gas which can quickly overwhelm the occupants, which is why some system of carbon monoxide detection is important to safe flight.
If the exhaust system has cracks, poor seals or poorly fitted components, the carbon monoxide can enter the cabin. Common failures of the exhaust system are caused by engine vibrations, thermal cycling, high temperatures and the corrosive effect of engine exhaust gases.
In 2004, the NTSB recommended that the FAA require the installation of carbon monoxide detectors in single-engine aircraft at risk of carbon monoxide entering the cockpit. However, the FAA does not require a detector but simply recommends that owners and operators use one. There is no rule that owners or operators install a carbon monoxide detector; instead, the FAA relies on education regarding the use of these and the importance of properly inspecting and maintaining the exhaust systems.
Similarly, the CAA argues that the potential for carbon monoxide poisoning in small aircraft is dealt with through regulation of the design, maintenance and operation. The CAA promotes the use of a current carbon monoxide detector but it is not mandatory.
A common inexpensive method of detecting carbon monoxide levels in the cabin of the aircraft is a CO strip detector or spot detector. They can be mounted onto a card and stuck to a highly visible location, with the strip turning black in the presence of carbon monoxide. A strip detector lasts between 3 and 18 months after being removed from their packaging.
The person who managed the aircraft believed that there was a strip detector fitted to the right side of the instrument panel, although it wasn’t fitted by him.
The various maintenance organisations who had done the annual maintenance in 2016, 2017 and 2018 had no record of installing a CO detector. There was no record of the Malibu ever having a CO detector installed.
The exhaust system on the aircraft consisted of right and left exhaust tailpipes made of stainless steel. Exhaust gas on the sides of the engine is directed through a wastegate on the left tailpipe. The wastegate directs excess exhaust gas away from the turbines.
The maintenance schedule for the aircraft included an inspection of the exhaust but did not require the tailpipe/heater muff to be removed or the exhaust system to be pressurised to check for leaks. The manual included the warning (in the introduction and repeated in the section on scheduled maintenance) that the vendor publications must be consulted when inspecting vendor equipment.
Continental Motors was the vendor for the engine fitted to the PA-46-310P. The engine manual had instructions and guidance for the continued airworthiness of the engine.
The guidance recommended that the exhaust system be pressure tested by applying air at a pressure of 5 psi to the exhaust tailpipe, using soapy water to check for leaks.
A service bulletin for the visual examination of the exhaust system and pressure-testing the exhaust system had been incorporated into the Standard Practice Manual for the engine.
Here’s the key point: If the aircraft were following Part 135 maintenance in order to be used for commercial flights, then this service bulletin would have been mandatory. Every year, a pressure test of the exhaust system would have been a requirement as a part of the examination of the exhaust system. For Part 91 (private flights), the requirements are much less strict. There was no requirement to pressure test the exhaust system to look for leaks and in the case of the accident aircraft, only a visual examination was made.
The year before, the aircraft’s annual maintenance was done by a different maintenance organisation. They carried out a detailed visual inspection of the exhaust system. Both maintenance organisations believed that the visual inspection of the exhaust system was sufficient to establish its condition and that there was no need for the exhaust system to be pressure tested.
This takes us to the heart of the issue. Grey charters, unlicensed commercial flights, are illegal but the regulations are very hard to enforce. It is up to the aviation authority to spot the flight and gather specific evidence that the pilot or broker was paid. Those using these grey charters are unlikely to understand the risk or even to be aware that the operation is illegal, let alone whether the aircraft is licensed and insured.
In the UK, a private pilot can operate a flight on a “cost-sharing” basis. This is a private arrangement which means that everyone on board shares the cost of the flight. The pilot must also pay his share and only direct costs can be shared, basically fuel and fees for landing or handling. In addition, a passenger in a cost-sharing flight must deal directly with the pilot; there can be no intermediaries.
However, if you are paying for the flight, then the operator of the aircraft must hold an Air Operator’s Certificate (AOC). The operator, whether that is a pilot or an organisation or an airline company, needs to meet a number of regulatory requirements to do with public safety and insurance. AOC holders are regularly inspected and audited.
The football agent who booked the flight on behalf of the customer probably had no idea of the regulatory issues. It wasn’t possible to get a direct scheduled flight from Cardiff to Nantes, so he contacted a man whom he knew was an experienced pilot; who had flown him and many of his players all over Europe on countless occasions. It didn’t occur to him that the man was running an illegal air taxi service or that he should ask if the pilot’s operation held an air operator’s certificate.
An air operator’s certificate would show that the operation followed much stricter safety standards. The company would have been required to follow more operational and engineering procedures. The pilots must be more highly qualified and their competence is tested more frequently. A pilot flying for an airline is assessed every six months, either in the air or in a simulator. To compare, a pilot with a private pilot’s licence generally only needs to renew their flying ratings every 24 months.
But who, in this case, was the operator? I have to be careful here because there is a lot of reading between the lines and the court case only actually covers the one flight, from Nantes to Cardiff, even though there is evidence of many other flights which were paid for by the passengers.
The flight was organised by a man who appeared to be generally operating the Malibu on behalf of the owner: an ex RAF officer who had hired a number of pilots in what was described in court as a “cowboy outfit” which did not follow regulations and ran undocumented operations. This appears to be the same person who is described as “responsible for the operation of the aircraft” in the report. He did not have an AOC and had never applied for one. He has since been convicted of endangering the safety of an aircraft and of arranging a flight for a passenger without permission or authorisation.
This man spoke to the football agent and then contacted the pilot to ask him to do the flights that weekend, as he was away on holiday. The pilot was clear that he would be paid for this, and in fact had been paid for similar flights on behalf of the operator in the past. The pilot agreed to fly the two daytime flights: Cardiff to Nantes on Saturday and returning from Nantes to Cardiff on Monday morning.
A representative of the trust, the owner of the aircraft, says that she told him that the pilot was not to fly the plane any more, after she received two letters from the CAA regarding airspace infringements. However, she is not on record as explaining whether she knew her aircraft was being used for illegal charter operations.
The man convicted, who initially denied being the operator, clearly understood the situation. He wrote an email to the pilot in which he made it clear that they needed to avoid CAA interference:
I have always said the flying we do is challenging and everyone has to be on the ball. It is a steep learning curve for someone new to the operation.
The prerequisite is a willingness to listen and learn. We both have an opportunity to make money out of the business model but not if we upset clients or draw the attention of the CAA… As self-employed sole traders we both have debtors and creditors and surely you understand that to remain legal we can’t take money in advance.
After the news broke that the aircraft had crashed, he messaged multiple people asking them to stay silent and said that the crash would “open a can of worms.”
[The pilot] has crashed the Malibu and killed himself and VIP pax! Bloody disaster. There will be an enquiry.
On the 15th of October last year, the director of the CAA announced that they were prosecuting the man for the following charges:
- On the 18th and 19th of January 2019, acted in a reckless/negligent manner likely to endanger N264DB (Articles 240, 256 and Part 4 of Schedule 13 of the Air Navigation Order 2016);
- On the 21st of January 2019, attempted to cause N264DB to discharge a passenger in the UK (Section 1(1) of the Criminal Attempts Act 1981, Articles 250, 256 and Part 3 of Schedule 13 of the Air Navigation Order 2016).
Just over a year later, the case is concluded and the man, who eventually conceded that he acted as an operator for that flight, was sentenced by the Cardiff Crown Court to eighteen months in prison. As a part of the court case, it was shown that the pilot had regularly flown for the operator and had clearly understood that he would be paid for this flight. Obviously, this operation, whatever the scale of it, was illegal, immoral and disregarding regulations that are put into place for the safety of the flying public.
It seems clear that the aircraft flew into the bad weather. Other pilots experienced little or no ice at the levels of the Piper Malibu that night but their reports and the radar show that there were intermittent heavy showers directly on the flight path. In addition to this, the layers of cloud obscured the moonlight and most likely the lights below, at least in the distance. The autopilot was known to be faulty and could have disconnected. This feels like a recipe for disaster: a pilot with no night rating and very little instrument experience was flying in instrument conditions in the dark with no visual references. It is no surprise that a pilot could lose control under these circumstances and plunge into the sea.
However, I don’t feel at all confident that the pilot’s skill level was the cause of that crash.
The pilot’s body was never found and we cannot know the medical state he was in before the crash. But the passenger was clearly suffering from extreme carbon monoxide poisoning with a carboxyhaemoglobin (COHb) level of 58%. Normally, COHb levels are 1-2% with smokers reaching up to 5%. Anything over 9% is considered to be from significant exposure to carbon monoxide from an outside source. The most common reason for elevated COHb levels is breathing air polluted with high carbon monoxide content. Carbon monoxide is believed to be the cause of over half of all fatal poisonings around the world. Common sources of carbon monoxide causing poisoning are fires, motor-vehicle exhaust and faulty domestic heating systems.
The most common symptoms of carbon monoxide poisoning are headache, dizziness and confusion, starting at 30% COHb. From 40-50% the passenger would have suffered from extreme headache, confusion and fainting. At 58%, he would have been unconscious with convulsions, respiratory failure and death to follow if the exposure continued.
In the Piper Malibu, there is no way that the pilot was protected from the contaminated air that the passenger was breathing. The truth is, with no evidence that he was suffering from carbon monoxide poisoning, it was only a matter of time before he, too, fell unconscious.
His voice sounded normal when he asked for an additional descent and when he explained that he was avoiding weather. But confusion and lack of judgement could easily take effect before any slurring of words.
The causal factors are listed as
1. The pilot lost control of the aircraft during a manually-flown turn, which was probably initiated to remain in or regain VMC.
2. The aircraft subsequently suffered an in-flight break-up while manoeuvring at an airspeed significantly in excess of its design manoeuvring speed.
3. The pilot was probably affected by CO poisoning.
Whatever happened that night as they crossed the English Channel, the pilot’s chance of survival was minimal. It didn’t matter that the pilot was struggling to maintain VMC. It didn’t matter that he lost control. It didn’t matter that he pulled up hard while the aircraft was diving.
If that flight had progressed with no issue, the fumes would have continued to affect the pilot, having already knocked out a young and very healthy footballer in his prime. The pilot would not have been competent to land the aircraft at Cardiff if, by some miracle, he was even still conscious.
The question as to whether the pilot was competent to fly in this weather is actually somewhat irrelevant. As the aircraft had no CO detector, there was no way for the pilot to know that the cockpit and cabin was filling with carbon monoxide.
There is no requirement for an aircraft to carry a CO detector, regardless of whether it was flying under Part 91 (private use) or Part 135 (commercial flights). The AAIB makes the point that passengers, pilots under training and individuals on cost-sharing flights are unlikely to understand the risk of carbon monoxide poisoning or that the pilot can choose not to have a CO detector installed. As a result, they recommend that the FAA, the CAA and EASA all require piston engine aircraft which may have a risk of carbon monoxide poisoning to have a CO detector with an active warning.
When I look at the factors of this accident, two issues jump out at me. The first is that, if the aircraft had been operating legally as a commercial flight, it would have been mandatory for the heater muff to have been pressure-tested which may have alerted maintenance that there was a carbon monoxide leak. But more importantly, a CO detector in the cockpit is the one thing that I can see that could have completely changed the course of events. In my opinion, the lack of a CO detector and the lack of a requirement for a CO detector is really a causal factor.
In the aftermath of this accident, the CAA released a campaign on Unlicensed Charters, explaining the requirements for a legal air taxi or air charter flights and encouraging passengers to verify the operator of their flight and check that they are legal by visiting https://caa.co.uk/aocholders, along with requesting the name and qualifications of the pilot. If it doesn’t look right or you are unsure, you can also email [email protected] and someone will look into it for you. If you think you are being offered an illegal flight, you should report it at [email protected].
- AAIB Accident Report: Piper PA-46-310P Malibu, N264DB
- BBC News: David Henderson jailed for organising flight
- Wales Online collection of articles about the crash and the case