By Bjorn Fehrm
June 5, 2026, ©. Leeham News: We do a series on aircraft structures and how they have shaped the way our airliners transport us around the world today.
We looked at how stressed skin aircraft structures are made last week, and also described how work to increase the strength of new aluminum alloys produced an alloy series today called the 7000 series, which had fatigue problems.
Fatigue has been a major problem for aircraft structures, especially since the introduction of pressurized cabins after the Second World War. A major step in understanding metal fatigue, how it creates fatigue cracks, and how these develop, was done during the investigation of the De Havilland Comet crashes during the mid/late 1950s.
Figure 1. The De Havilland Comet 1, whose crashes in 1954 led to an investigation that increased the understanding of fatigue. Source: Wikipedia.
The De Havilland Comet
The Comet was developed as the world’s first jet airliner in the late 1940s, with its first flight in July 1949. It was powered by four De Havilland Ghost jet engines, each producing 5,000 lbf. The Comet, due to its jet engines that worked well at altitude, had a cruise speed of 490mph and a maximum operating ceiling of 42,000ft. The high cruise altitudes, regularly over 35,000ft, necessitated a maximum cabin pressure of 8.3 PSI to prevent passengers from experiencing oxygen starvation.
The competing piston-based airliners, like the Douglas DC-6 and Lockheed Constellation, stayed below 25,000ft and had pressurization of 2.5 PSI for the DC-6 and 4.8 PSI for the Constellation. All had fuselage skins made of fatigue-resistant Duralumin-derived alloys, for Comet D.T. 564 and the US airplanes Alclad 24-ST, which today would be called 2024-T3.
De Havilland knew that the Comet would be a great technological advancement. They were competing to be the first company to offer pressurized jet service to the public. Since there was little experience in the design and production of pressurized commercial airliners at the time, De Havilland placed special emphasis on structural testing, including pressure testing of the fuselage.
As the Comet was flying at unprecedented altitudes, the authorities and De Havilland agreed on extensive tests of the fuselage with its pressure cabin. It was tested both in ultimate strength to a factor of 2 instead of the normal 1.5, and under pressure to 2*8.3 PSI instead of just 1.33 times as demanded by the regulations. The test of a forward and a mid fuselage was then cycled to 18,000 and 16,000 cycles, at which fatigue cracks appeared at the corner of cabin window cutouts. The Comet’s planned Safe Life was 10,000 cycles, so cracks at 16,000 were not a concern.
A problem with the testing was that the same fuselage parts were used twice, both for the two times the highest flight pressure test and then for the cyclic pressure test with operational pressures. The high-pressure test cold-worked the aluminum alloy in the highest-stressed areas, delaying the development of fatigue cracks in the test parts during subsequent cyclic tests.
The production fuselages did not pass a high-pressure test; thus, their stress concentration area were not cold worked before their operational pressure cycles during each flight. At the time, due to an insufficient understanding of fatigue, De Havilland and the authorities thought that a static pressure test at twice the normal pressure would reveal any high-stress area that could create a fatigue problem. The opposite happened; the tests disguised an operational fatigue life of around 1000 cycles.
The Crashes
The first fatigue-related crash occurred on the first production Comet while departing Rome, Italy, in January 1954. It then had 1290 flight cycles.
The Comet was grounded until March, but with no obvious cause of the accident, only the first pieces of the aircraft had been recovered from the Mediterranean, so flights resumed. On the 8th of April, the second Comet fell from the Sky near Naples on a flight from Rome to Johannesburg, having conducted 900 flights.
The Comet fleet was immediately grounded once again, and a large investigation board was formed under the direction of the UK Royal Aircraft Establishment (RAE).
RAE was given a BOAC Comet with 1121 flight cycles. It submerged the fuselage in a water tank and cycled it using water pressure (Figure 2).
Figure 2. The RAE water tank used to pressure and cover the test plane’s fuselage. Source: RAE report.
This was to use the less compressible water (as opposed to air) to show any cracks without blowing the parts to pieces in a violent decompression. The fuselage failed after a further 1830 tank cycles due to cracks that ruptured it at the escape hatch window on the port side (Figure 3).
Figure 3. Rupture of the tank tested fuselage around the port escape hatch. Source: Wikipedia.
This was the evidence needed to focus the investigation on a detailed understanding of fatigue. Special test parts from the cracked areas were built and extensively strain-gauged.
The Findings
It was discovered that the stress levels at the window and antenna cutouts were 50% higher than the hoop stress level in the fuselage skin away from the windows. The investigation could also, in detail, follow the development of the cracks, their initial rate of lengthening, and the rate once they became critically long (when the remaining material could no longer take the stress).
The extensive investigation and reporting served the aeronautical community, which now had detailed forensic data to better understand fatigue cracks and their development in aeronautical structures.
In 1960 and 1961, researchers could present scientific work and formulas to predict how long a structure would last when subject to cyclic stress, assuming the occurrence of specified notch defects in the material after it had been fabricated and assembled.
The understanding of fatigue was greatly improved by the Comet crashes and the subsequent investigation. One of the findings was that the Comet design principle of Safe Life needed reliable tools to estimate that safe life. The alternative was to design a Fail Safe structure, meaning that in case of a part failure in the structure, there would be alternate load paths that would stop the structure from hazardous failure. We will look into these subjects more in the next Corner.
Related
