IT "simply doesn't add up".

That was the verdict of former RAF pilot and aviation lawyer, Jim Morris, after the two year wait for answers over the Clutha disaster left the pilot, inexplicably, in the frame.

Captain David Traill was "well-rested" on the night of the doomed flight. He was an experienced pilot with "above average" talents who had accumulated 646 flying hours on board a Eurocopter EC135, with years spent flying various helicopters as an RAF pilot in Afghanistan, Bosnia, Iraq, and Northern Ireland.

Yet on November 29 2013 he appears to have made a fateful error which cost himself and nine others their lives.

At some point while flying between Dalkeith, in Midlothian and Bothwell, South Lanarkshire two vital fuel switches - located in the cockpit ceiling to his right - were turned off.

They remained in the 'off' position for a "sustained period" prior to the crash, said investigators, for "reasons unknown".

As a result, fuel stopped pumping from the main tank into the helicopter's two supply tanks, which gradually ran dry and starved the engines of fuel.

Investigators believe that the first low fuel warning was triggered as the helicopter approached Bothwell around 10pm. Safety protocols dictate that the pilot should land within 10 minutes of a low fuel alert but, bizarrely, Captain Traill continued flying for another 20 minutes.

By 10.19, three minutes before the crash, low fuel warning lights had flashed four times, accompanied by alarm bells. Each time, the warnings were acknowledged and extinguished by Captain Traill.

The report stated: "It was not possible to discover what discussions took place between the crew in response to the warnings, of which all three crew members would have been aware. However, the evidence indicated that [guidance] to land within 10 minutes was not followed.

"On the limited evidence available, it could not be established why this did not happen. There was no indication of any concern in the radio transmissions that were received from the pilot or Police Observers throughout the flight."

There are only two operational circumstances where a pilot would turn off the fuel flow switches.

The first is where the main tank has run dry and the pilot turns off both pumps to prevent them overheating. This is irrelevant in the Clutha crash, where 76kg of fuel remained in the main tank.

Alternatively, the pilot may switch off one of the pumps if it becomes blocked. Again, investigators found no trace of blockages inside the Clutha helicopter.

Mr Morris, of Irwin Mitchell, said: “This was a highly qualified and experienced former RAF helicopter pilot who was assessed as ‘above average’. He should have been able to competently deal with fuel management, low fuel and emergency procedures.

"The report suggests that he had acknowledged low fuel warnings, but the AAIB could not establish why he did not complete the low fuel actions in the pilot's checklist, so something simply doesn’t add up."

There was no mayday and nothing to indicate that Captain Traill had attempted, an emergency landing, known as an auto-rotation, which is standard procedure in the event of double engine failure.

The lack of a black box has hampered investigators as this could have shed light on the conversations and events in the cockpit in the moments before the crash.

It is possible that the pilot believed the fuel sensors were malfunctioning.

In the weeks running up to the crash, the report notes than a series of repairs were made to faulty fuel sensors in the Clutha helicopter's tanks.

It is these thermal sensors which trigger low-fuel warnings to the pilot.

In July, the fuel sensor in the main tank was replaced as a result of “spurious” fuel indications, which were over- or under-estimating how much fuel remained in the tank.

In October, the fuel sensor in the left supply tank was also replaced after experiencing the same defect. Around the same time, the fuel sensor in the right supply tank was found to be malfunctioning and was “removed, cleaned, flushed with clean fuel and dried with warm air” to correct it.

These sensor faults have previously been reported in relation to EC135 helicopters. The cause has never been determined although water contamination is one theory.

When the manufacturer tested the fuel sensors following a worldwide recall in December 2013, sparked by faulty readings in an air ambulance in England, around 30 per cent required repairs.

However, AAIB investigators stress that “no defects” were recorded during any of the three flights conducted by the Police Scotland helicopter on the day of the crash.

A spokesman for manufacturer Airbus Helicopters said they took "no comfort from the fact our aircraft has been proven safe".

At least a technical fault can be fixed. Mystery "human factors" are harder to solve.