The University of British Columbia Interdepartmental

Faculty of Applied Science

Memorandum
Engineering Student Services
CEME 2053, 2324 Main Mall, UBC
Campus
Tel: (604) 822-6556
Fax: (604) 822-2021
email: students@apsc.ubc.ca
October 30, 2009
To: Dr. Hatzikiriakos, APSC 150 Coordinator
From: Ian Dusome (27603091)
Subject: De Havilland Comet 1 Failures

1. Introduction
This memo contains a failure report on the De Havilland

Comet 1 commercial aircraft. A series of incidents involving the De

Havilland comet, the most severe of which being the mid-flight

disintegration of three aircrafts, resulted in 111 deaths and the new

hope and pride of british aeronautical engineering to be striped of its

certificate of airworthiness. (Waterton & Hewat, 1955)(Withey, 1997)

This report contain the following sections: a description of the aircraft,

as well as the events leading up to and including the failure; an

explanation and analysis of why the failure occurred and finally the

lessons that can be learned from this failure.

2. Description
The De Havilland Comet 1 took its first commercial flight on

May 2, 1952, becoming the first ever jet engine propelled commercial

aircraft and promising a bright future for the british aircraft industry.

(Waterton & Hewat, 1955) (Withey, 1997) The aircraft was propelled by

Page 1 of 7
four Ghost Turbojet engines each producing 5000 pounds of thrust,

giving the plane a cruising speed of 450 mph at a altitude of 40 000

feet. This high altitude required the cabin to be pressurized to the

equivalent of 8000 feet. (Withey, 1997)(NASA) The majority of the

aircraft was constructed using an adhesive called “Redux” to glue the

22 gauge aluminum skin to the stringers, the exception being at some

locations on the wings and around cut outs such as windows where

standard riveting was employed. The pressure difference across this 22

gauge skin was 8.25 psi, twice that of any previous situation. (Withey,

1997) May 2, 1953, exactly one year after the comets inaugural flight

the comet G-ALYV (Yoke Victor) encounters a tropical storm and breaks

up mid-air over india. Next on January 8th, 1954, the comet G-ALYP

(Yoke Peter) was heard to be climbing to 27000 ft when contact was

lost and the aircraft could be seen to break up and crash into the sea

near Elba, this time in good weather conditions. (See Figure One) This

resulted in the comets being grounded but after only a few small

modifications the fleet was back in service and on April 8th 1954

another comet, G-ALYY ( Yoke Yoke) disintegrated as it reached the top

of its climb. Once again in good weather the aircraft plummeted into

the sea near Naples. (Withey, 1997) (Waterton & Hewat, 1955) After

the fatalities resulting from these unexplainable crashes the De

Havilland Comet was grounded for good, loosing its certificate of

Page 2 of 7
airworthiness. Extensive investigations began, including the recovery

and reconstruction of 70% of the Yoke Peter, in an attempt to uncover

the reasons behind these failures (See Figure one) (Waterton & Hewat,

1955)(UofP Faculty of Technology)

Figure 1: Recovered and Reconstructed Yoke Peter (UofP Faculty of

Technology) Reasons for the Failure

Overall the De Havilland Comet 1 can be said to be a failure

of engineering design for several reasons. The most significant

failures, the in-air disintegrations described, were a result of fatigue

crack growth due to repeated pressurization but a more in depth look

is required as there are several reasons behind why this occurred. On

one level the failures occurred due metal fatigue in high stress areas

the growth of small cracks around bolt holes created process of

riveting during manufacturing which were virtually invisible at the time

of manufacture. (See Figure 2 & 3) (Withey, 1997)(UofP Faculty of

Technology) The particular alloy of aluminum, DTD 546B, used for the

aircrafts skin has a high rate of crack growth and in areas around cut

Page 3 of 7
out where stress’s were concentrated, to an excess of 315 MPa, cracks

propagated to failure after relatively few pressurization cycles.

(Withey, 1997) On another level, the reason behind the use of riveting

around the square cut outs, despite the initial suggestions of the chief

designer, was that the tooling required to utilize redux in these areas

would be too difficult and expensive to produce. This would not have

been a problem had ovular cut outs been used as seen on revised

models of the Comet. (Withey, 1997) (UofP Faculty of Technology)

Finally the methods used in initial testing of the fuselage during

pressurization occurred in such an order that the faults were allowed to

pass unnoticed. (Withey, 1997) (Waterton & Hewat, 1955)

Figure 2: Evidence of Fatigue Failure in Aircraft Skin (UofP Faculty of

Technology)

Page 4 of 7
Figure 3: Main site of failure on Comet Yoke Peter (UofP Faculty of

Technology)

4. Lessons to be Learned

To begin with when dealing with any engineering project,

especially pioneering projects that could result in the highest of

consequences should they fail, we as engineers must be sure conduct

thorough investigations and testing in order to design the project such

that it minimize risk wherever possible. Thorough testing is also the

responsibilities of engineers to ensure that not even the smallest of

details is overlooked. In the case of the comet, had more thorough

Page 5 of 7
 testing reviled that small crack produced during manufacturing could

quickly propagate around high stress areas many lives could have

been saved. Finally nothing should take precedence over safety and no

decision should be made based on some concern at the cost of safety.

3. References

Waterton, W. A., & Hewat, T. (1955). The Comet Riddle. London,

England : Fredrick Muller LTD

University of Plymouth Faculty of technology. (n.d.). Failure as a

design criterion; Stucturanl Failures; Unforeseen Loads and

Consequences; Comet Airliner. Retrieved October 23, 2009,

http://www.tech.plym.ac.uk/sme/interactive_resources/tutorials/Failure

Cases/index.html

National Aeronautics and Space Administration (n.d.) Quest for

Performane: The Evolution of Modern Aircraft, Part 2 The Jet Age,

Chapter 13: Jet Transports. Retrieved October 23, 2009, from

http://www.hq.nasa.gov/office/pao/History/SP-468/ch13-1.htm

Withey, P. A. (1997) Fatigue failure of the de Havilland Comet 1

Engineering Failure Analysis, Volume 4, Issue 2, pg 147-154 Retrieved

Page 6 of 7
on October 23, 2009 from ScienceDirect Database via Compendex

Database

Page 7 of 7