5 Disasters Caused by Material Fatigue and What We Learned From Them
Versailles Train Crash, 1842
On the afternoon of May 8, 1842, birthday celebrations for King Louis Philippe I had just concluded in Versailles. Hundreds of spectators piled into countless train cars, so many that it required two locomotives to pull. As the train was making its way back to Paris, the leading locomotive broke an axle, causing the engine to derail. The chain reactions saw many of the trains cars pile onto each other and catch fire, caused by the scattering of the engines’ fire boxes. Most estimates attribute 55 deaths to the accident and subsequent fire. The common practice of locking passenger doors at that time may have also been a contributing factor.
The Versailles train crash was the first of its kind in France, and made world headlines. The crash also occurred during a period of time where metal fatigue and general degradation over time were not well understood, which led to fear and confusion among the general populous. Rail companies, government agencies, and scholars all set out to scrutinize and learn from the incident in order to prevent future catastrophes, as well as to restore public trust in the young railroad system as a safe and reliable means of transportation.
William Rankine and August Wöhler were just a few of the many researchers who dedicated years to the advancement of design, testing, and maintenance of train axles. Because of this, the Versailles accident is considered by historians to mark the beginning of serious human interest and research into the area of fatigue and fracture mechanics, which has allowed for the design and manufacture of safer, more durable goods and components.
4 stages of fatigue and fracture mechanics1. Crack nucleation 2. Stage I crack-growth 3. Stage II crack-growth 4. Ultimage ductile failure
The Boston Molasses Disaster, 1919
On January 15, 1919 a 2.3 million gallon tank filled with molasses collapsed in Boston’s North End neighborhood. Witnesses reported hearing what sounded like gunshots, as the rivets shot out of the 50 foot tall tank. The collapse created a molasses wave up to 25 feet high, traveling at up to 35 mph at its peak. The forceful wave damaged steel girders on an elevated railway track, swept multiple buildings off of their foundations, and flooded countless city blocks.
A full investigation ensued, which brought many contributing factors to the surface. One of the most critical factors was the neglect and general state of disrepair that the tank was in when the collapse occurred. Reports stated that basic leak and pressure tests were neglected to be performed prior to putting the tank in service. Reports also stated that the tank, when filled, leaked so bad that it had to be painted brown to hide the imperfections.
Despite this, the tank remained in service. Observations of the post-collapse evidence showed that the root cause originated near a manhole cover at the base of the cylindrical tank, where hoop stress concentrations are highest. It is believed that a fatigue crack initiated near the manhole cover and grew to a critical length prior to failure. Other contributing factors included fermentation within the tank and a sharp rise in temperatures, both of which would have caused the internal pressure of the tank to rise considerably.
De Havilland Comet Plane Crashes, 1954
The De Havilland Comet was the world’s first production commercial jetliner, produced by De Havilland of Great Britain. The Comet was the crowning achievement for Britain at the time, and further advanced their aviation superiority worldwide up until the first of several fatal accidents eventually attributed to metal fatigue.
In January 1954, BOAC Flight 781 experienced explosive decompression over the Mediterranean Sea in route to London from Rome. All 35 passengers and crew were killed and all Comet aircraft were immediately grounded. After an extensive search and recovery mission, officials began examining the recovered aircraft. It became clear that the aircraft broke up in mid-air, and officials initially believed that an engine turbine explosion had caused the accident. Turbine modifications were made to all Comets and the planes were once again allowed to fly.
Just weeks after being cleared for flight, another Comet aircraft, South African Airways Flight 201, experienced explosive decompression over the Mediterranean in route from Rome to Johannesburg. Again, all 21 passengers and crew were killed. This incident caused investigators to question their hypothesis of a turbine explosion as the main culprit of the decompression.
After an extensive, multi-year investigation into both flights, it was determined that metal fatigue caused by design defects ultimately led to the explosive decompression in both instances. The metal fatigue originated near a forward window used for navigation. Several contributing factors were observed. First, that the squared window design caused an extremely high stress concentration at the window’s corners. In fact, calculations revealed that up to 70% of the aircraft's ultimate stress under pressure was concentrated on the corners of the aircraft's windows. Secondly, that the supports around the window were riveted instead of glued, as originally specified, and that the rivet holes caused fatigue cracks to initiate after repeated pressurization cycles.
The findings from these accidents were used to overhaul aviation requirements for passenger cabin strength. In addition, sharp points and edges were eliminated in aircraft designs, in an effort to reduce stress concentrations.
Alexander L. Kielland Oil Platform, 1980
On March 27, 1980 the Alexander L. Kielland, an oil drilling rig, was stationed in Norwegian waters on the North Sea. More than 200 workers were off duty in the accommodations of the vessel when a ‘sharp crack’ was reported. The rig suddenly heeled over at a 30° angle. Five of the rig’s six anchor cables had snapped, leaving the final cable to support the massive stress levels. The rig remained relatively stable in this position for a short period of time until the final cable fractured and the rig capsized into the sea completely. More than 120 workers were killed during the capsizing, which stands as the worst disaster in Norwegian waters since World War II.
The investigation that followed was able to piece together the events from that evening and determined that the origin of the collapse was caused by fatigue cracking in one of the structural bracings of the rig. The crack was then traced to a small 6 mm fillet weld which connected a non-load bearing flange plate to the bracing. The fillet weld had a poor profile and significant cold cracking, which caused a significant reduction in fatigue strength. The flange plate was also weakened by significant lamellar tearing, which increased stress concentrations. The cyclical stresses experienced by the rig in the North Sea served to further exacerbate the situation.
Eschede Train Disaster, 1998
On June 3, 1998 a high speed train traveling from Munich to Hamburg was derailed when a single train wheel failed, causing a chain reaction which lead to a bridge collapse and over a dozen derailed train cars.
A steel tire on car #1 started the chain reaction when it failed, released from the train, and became embedded in the floor of the first car. As the train passed through a switch, the embedded tire slammed against the guide rail of the switch, causing the guide rail to also embed into the train, which lifted the trains axles off of the track. When the train approached the second switch, one of the derailed wheels struck the switch, which changed its setting. This caused the rear axle of car #3 to be pulled onto a parallel track, violently derailing the car, which struck and destroyed the main supports of an overpass bridge. Several more cars, traveling at 120 mph, struck the bridge until it collapsed completely, blocking the entirety of the track. The remaining cars hit full speed into the rubble, causing a large pileup.
Overall, 101 fatalities were reported along with nearly 100 injuries. Among other factors,investigators determined that the design of the wheel was flawed and lacked sufficient validation testing prior to implementation. Engineers had placed a rubber damping ring between the tire and wheel body in an effort to reduce vibrations during cruising. This led to increased fatigue susceptibility in several ways:
- The tires were flattened into an ellipse as the wheel turned through each revolution (approximately 500,000 times during a typical day in service), with corresponding fatigue effects.
- In contrast to the pure monobloc wheel design, cracks could also form on the inside of the tire.
- As the tire became thinner due to wear, the dynamic forces were exaggerated, resulting in crack growth.
- Flat spots and ridges or swells in the tire dramatically increased the dynamic forces on the assembly and greatly accelerated wear.
Other contributing factors included improper maintenance (records indicated that this particular wheel has failed to pass inspections on several occasions leading up to the crash, although it was never replaced), overbridge design (wasn’t designed with spans), and the use of welds in the carriage bodies (led to “unzipping” during the crash) As a result of the disaster, all wheels with a similar design were replaced with monoblock wheels.