Mind the bend

Over the weekend I was reading up on the story of the de Havilland Comet. The reason for this was that it came up in a conversation with my dad the other day: I had read about it before, but had forgotten the story since…

The Comet was the first commercial jet airliner. The maiden flight of the first prototype, the Comet 1, flew in 1947, with the debut for regular flights following in 1952. The aircraft boasted an aerodynamically clean design with four de Havilland Ghost turbojet engines buried in the wings, a pressurised fuselage, and large square windows. At that time, the Comet was the leader of the pack, de Havilland was well ahead of its competitors’ offerings.

Then, in 1954 two disastrous hull losses occurred to Comets belonging to BOAC and South African Airlines that broke up in mid-air, resulting of the loss of all passengers and crew on board. The inquiries looking into the causes of these incidents examined various scenarios, from sabotage to clear air turbulence, but in the end came to the conclusion that the principal cause of the accidents were much simpler: square windows.

Stress around the window corners was found to be much higher than expected, while stresses on the aircraft skin were generally more than previously expected or tested. This was due to stress concentration, a consequence of the windows’ square shape, which generated levels of stress two or three times greater than across the rest of the fuselage.

As a consequence of the inquiry, this issue was addressed by modifying the square window design and replacing it with round ones – yet for de Havilland the damage was done: competitors had learned from the Comet’s design flaws and had drawn ahead in the race for the skies.

Rather sad actually and certainly a cautionary tale, as it illustrates the risks innovators run and how these may – or may not – pay off. Things that in retrospect seem blindingly obvious were not so at the time. One thing is for sure though: next time I fly, the Comet tale will make me look at the round aircraft windows with fresh sense of appreciation.

All this brought to my mind discussions regarding bend radiuses for synthetic lines, as this also involves the deflection of forces, with failure occurring at the point of maximum deflection. This is one of the reasons for knots reducing the breaking strain of lines and also why we consider the ratio of the width of pulleys to the lines we run over them.

The other aspect it highlights is the need for high-cycle testing. I have mentioned this before, but I will say it again:

It is simply insufficient to only test the maximum force that an assembly or component can withstand, it is crucial that the fashion in which it responds to cyclical loading has also been assessed. One of the things that the Cohen Committee did when looking into the reasons for the Comet crashes was to build a dedicated water tank in RAE Farnham that enabled them to repeatedly re- and over-pressurise the hull of the aircraft. After 3’057 pressurisations the hull burst open.

Regardless of whether it is metal or synthetic components being considered, cyclical failure needs to be considered and tested for.