Alexander L. Kielland (oil rig)
The Alexander L. Kielland was a semi-submersible drilling rig named after the Norwegian writer Alexander Lange Kielland and used as living quarters , which capsized after an accident on March 27, 1980 in the North Sea . 123 of the 212 crew members died. The cause of the accident was signs of fatigue on the rig's structure . The loss of the Alexander L. Kielland led to a number of radical changes in the design , testing and safety equipment of drilling rigs.
development
In 1963 the Institut Français du Pétrole entered into a cooperation with the Schlumberger Group- owned exploration company Neptune to design an oil rig resting on five buoyancy bodies. The first platform, P81 , was delivered in 1969. In 1970 the design was revised in collaboration with several other companies and the P82 was built in Brownsville, Texas . This was the basis for nine other platforms, three of which were built in Finland and six by the French company Compagnie Française d'Entreprises Métalliques (CFEM) in Dunkirk . Alexander L. Kielland was the seventh platform of the modified basic type, was built in France and called the Pentagone 89 .
Description of the Alexander L. Kielland
The oil rig was a semi-submersible of the Pentagon type. The platform with the superstructures rested on five pillars, some of which were submerged in the sea and which were supported by buoyancy bodies located under water. The floats had a diameter of 22 m and a height of 7.5 m. The columns had a diameter of 8.5 m and were, including the float, 35.6 m high. They were connected to each other and to the platform by a series of struts. The horizontal struts had a diameter of 2.6 m and wall thicknesses of 25 mm, the diagonal struts had a diameter of 2.2 m. They were made of C-Mn structural steel (corresponding to Lloyds' ship's steel "Grade EH") with a yield strength of at least 355 N / mm² (comparable to structural steel material number 1.0570 S355J2 + N (according to EN 10025-2: 2004-10)).
Tanks for ballast and drinking water, fuel and other operating materials were housed in the pillars and floats . Three of the pillars also contained machine rooms that could be reached via elevators mounted in the middle of the pillars. Propellers on these three pillars could move the oil rig. A total of five diesel generators , including one for emergencies, ensured the power supply for Alexander L. Kielland . They could be operated up to an incline of 20 °. There were also pump rooms in the pillars for draining in the event of water ingress.
The actual platform had dimensions of 103 m × 99 m and a mass of 10,105 t. In normal operation, it was about 15 m above the water surface. On top of it was a 40 m high derrick . Since the Alexander L. Kielland was to be used as a drilling platform in the foreseeable future, all the equipment required for drilling was also on board. The Alexander L. Kielland had been used as a living platform since it was commissioned. For this purpose, residential containers were set up on the platform on several floors. These containers were each set up for four people. The capacity was increased from 80 people to 348 people.
A system of ten anchors held the Alexander L. Kielland in position. The anchor ropes ran in pairs over mooring winches arranged on the pillars . These winches were controlled by three hydrophones , which were located on the horizontal struts of the Alexander L. Kielland . The hydrophones received their signals from a sound wave transmitter installed on the sea floor. Alexander L. Kielland always kept this system in the same position.
Use and certification
On July 5, 1976, the drilling platform was delivered to the Stavanger Drilling company in Norway. She was then chartered by the US Phillips Petroleum Company . Their area of operation was the Ekofisk oil field in the geographic center of the North Sea . In 1980, the majority of the around 80 drilling rigs and production platforms in the North Sea at that time were located in this area.
When it was first used in July 1976, the oil rig was used as accommodation (so-called flotel, a combination of words from “floating hotel”) for the H-7 compressor platform . From July 15, 1976 to August 1, 1979, the Alexander L. Kielland was used, among other things, as residential accommodation for the Henrik Ibsen and the Dyvi Alpha . On August 1, 1979, the platform was Albus Kjell 2.4 F for Edda 7.2 C spent. Here it served as a living platform for the workers of Edda 2/7 C and was connected to this platform by a movable bridge for most of the time.
The security certificate was issued by Det Norske Veritas , the Norwegian equivalent of Lloyd's Register of Shipping . Only a few small errors were found during acceptance. The last annual inspection took place at sea in September 1979. The large four-year inspection had been postponed from April 1980 to June 1981 at the request of the owners.
Disaster on March 27, 1980
Starting position
The Alexander L. Kielland was next to the Edda 2/7 C platform , which was at 56 ° 27 '53.4 " N , 3 ° 6' 16.1" E, was. The location corresponded roughly to the height of Edinburgh and was 385 km off the Norwegian coast. The distance to the Dutch coast was about the same. The Alexander L. Kielland had been assigned to the Edda 2/7 C production platform as a living platform for nine months and was connected to it by a movable bridge about 25 meters long.
On March 27, 1980, the weather had deteriorated during the day. There were wind speeds of 16 to 20 m / s with gusts of wind in force 10. The wave height was 6 to 8 m. The air temperature was between 4 and 6 degrees and the water temperature was a little over 6 degrees. At end of the shift on the Edda 2.7 C , the workers returned after 18 pm at the Alexander L. Kielland back. The connection between Edda 2/7 C and Alexander L. Kielland had been removed due to the bad weather. Around 50 to 80 people were in the two cinema rooms, around 50 in the mess hall and others in their living quarters. A total of 212 people were on the Alexander L. Kielland .
There was an emergency plan for the Ekofisk oil field , which provided for the presence of three standby ships so that each platform could be reached in a maximum of 25 minutes. The motor ship Silver Pit , a converted trawler , which, in addition to a fast lifeboat, also carried a lifeboat with water jet propulsion for up to three rescuers and twelve people to be rescued, was intended for the platforms Edda 2/7 C , Alexander L. Kielland as well as Eldfisk Alpha and Eldfisk Bravo to back up. For this purpose, the master had been instructed to stay in "Area 3" in the middle between Eldfisk Alpha and Edda 2/7 C. However, the ship only stayed near Eldfisk Bravo for several months . In March a new captain took over the ship. He received insufficient information about the tasks of the Silver Pit . In his testimony to the Commission of Inquiry, he said that he assumed he was only responsible for Eldfisk Bravo . At the time of the accident, the Silver Pit was one nautical mile southeast of Eldfisk Bravo and thus about 6 nautical miles from the Alexander L. Kielland . Thus, the Silver Pit only reached the site of the accident after 7.15 p.m. and could not save anyone.
Course of the accident
Shortly before 6:30 p.m., a strong shock was perceived on the Alexander L. Kielland , which was followed by vibrations . Most people are said to have taken it for a wave and ignored it. After a second bump, the rig "shook" and leaned to starboard until it reached an incline of 30 to 35 °. Worker Tony Sylvester described the situation: “Everyone believed it was over now. [...] There was a terrible crash, and shortly afterwards again, and then the whole thing tilted 45 degrees on its side. "
As was determined later, it was at this point that the D-6 horizontal strut broke. As a result, the other struts in the lower area of pillar D were overloaded and also broke. The float lifted the column and turned it so that the remaining struts also broke. Pillar D separated completely from the rig and drifted away. As in the column D no longer a boost was given heeled the rig to this page. The actual platform was partially washed over by water and the columns C and E sank so deep that they were almost completely immersed in the water. Instead, the port side rose sharply.
Objects slipped to starboard across the rig. In the makeshift cinema on the drilling deck, parts of the drilling equipment broke through the wall and injured several men. In the living quarters, loose cabinets fell over and locked doors. A little later the lights and alarm signals went out because the diesel generators no longer worked when the heel was too steep.
For a short time the rig was stable. During this time, water ran into rooms and tanks in the superstructure and into pillars C and E. The water penetrated through doors, hatches and ventilation openings that were otherwise well above the water surface. About 20 minutes the skew steadily increased until the oil rig to 19:00 capsized and upturned swam.
Rescue operation
At the rescue stations of the Alexander L. Kielland there were eight motorized lifeboats for 50 people each, four deployable and six disposable life rafts for a total of 400 people as well as eight containers with a total of 125 life jackets. There were a total of 541 life jackets on the Alexander L. Kielland . The regular crew of the Alexander L. Kielland also had rescue suits . Of the other people on board, only a few had emergency suits; these were not generally prescribed. Most of these suits were also on Edda 2/7 C, as it was very bulky equipment that the workers did not always carry with them.
The cinema-goers tried to reach the highest point of the platform, the B column , through a hatch on the port side of the cinema room. Since many paths in the living area were blocked by loose furniture, some people jumped out of the windows. According to reports, many people did not reach the rescue equipment.
A relatively large number of people had gathered at pillar B. Boats 5 and 7 were there. Boat 5 could only be boarded by 14 people. The fully closed boat tore itself loose and swam up keel up in the sea until it could be raised again by the occupants and by swimmers in the water. 19 people were then taken on board from the water. Since smoke escaped when attempting to start the engine, the boat was allowed to drift away without being propelled. An unspecified number of people were able to save themselves in boat 7.
The lifeboat 1 in the stern of the Alexander L. Kielland could be boarded by 26 people. By heeling it took only a maximum of two meters eased to be. Since the releasing hooks could not be released under load, an ax had to be used to get the boat free. In the time required for this, the boat was thrown against the oil rig and damaged. However, it was able to move away from the oil rig with engine power.
Lifeboats 2, 3 and 4 were not brought into the water and were hit against pillars by the waves and destroyed. Boat 6 broke off together with pillar D. Thus only half of the lifeboats were used.
The life rafts were probably not launched, but tore themselves loose when they capsized and inflated themselves. They and the life rafts launched from Edda 2/7 C enabled a further 16 people to be rescued.
The radio operator of the Alexander L. Kielland had immediately after the first heeling over FM - voice a " Mayday deposed" ring and then go to the lifeboat 5, from where he Edda 7.2 C to the radio of the boat on further action informed.
The first Mayday call was received on a handheld radio from Baste Fanebust, the ship coordinator for the Ekofisk complex. "Charly Transport", as the radio call name , then dispatched most of the ships in the Ekofisk field to the site of the accident.
The rescue coordination center in southern Norway was also notified. Ships in the North Sea and rescue stations in Norway, Scotland, Denmark, the Netherlands and Germany were alerted until 18:42. The first Norwegian rescue helicopter took off around 7:30 p.m. At the same time, two helicopters and a reconnaissance plane took off in Great Britain . More planes followed; Fog made it difficult for them to work.
A transport helicopter in the oil field took off, but had no rescue equipment on board and could not land on the inclined landing area.
Lifeboat No. 5 put its emergency radio buoy into operation and was located at 7:30 p.m. by the Normand Skipper , a supply ship not equipped for rescue operations. Twelve occupants were able to climb onto the Normand Skipper via a net , then the operation was canceled as too dangerous. The remaining 21 people were picked up by two helicopters between 2:30 a.m. and 4:00 a.m.
Lifeboat No. 1 had radio contact with the Silver Pit and the supply ship Normand Skipper . However, the Silver Pit could not find the boat. The Normand Skipper reached the lifeboat together with the Normand Vibran at around 1:20 a.m. on March 29, but could not take over any people due to the waves. The 26 men were recovered by two Norwegian helicopters until 3:00 a.m.
The Edda 2/7 C platform saved seven swimmers from the North Sea with one of its two cranes. Other people were rescued by other ships and other helicopters.
Of the 89 people rescued, only 59 wore a life jacket. Only eight people had put on a rescue suit, seven of whom had not properly closed the suit. Four people in rescue suits were found dead. Christian Naess, the captain of Normand Skipper , reported that a person in a rescue suit could not be brought on board because the suit was wet and slippery. At that time the suits did not have loops to hold on to.
71 civilian ships, nine military ships , 19 rescue helicopters and seven airplanes took part in the rescue operations until the suspension of operations on March 29 at 7:00 p.m. The injured were taken to Rogaland Hospital. Although the hospital was not prepared for such an emergency, it also set up an emergency medical station at Sola Airport and dispatched a team to the Ekofisk field.
A survey of forces involved in the rescue operation nine months later showed that 67% saw themselves exposed to considerable dangers during the rescue operation; many suffered from post-traumatic stress syndrome .
Recovery of the oil rig
The wreck of the oil rig was towed into the Gandsfjord off Stavanger . About three and a half years after the accident, she was raised there again. This action served both further investigations and, above all, the recovery of corpses so that they could be buried by their families on land. The erection was prepared for several months, among other things special software was written for the calculations. In addition, visual inspections and drilling into the pillars determined how much water had penetrated into the pillars. There were floats welded and secured residential container with additional steel cables.
The rig was later Nedstrandsfjorden (between the local Tysvær and Finnøy ) m sunk at a depth of about the 700th
Causes of accidents
The accident was triggered by the breakage of the D-6 strut. A drainage opening was cut into this strut, which was reinforced by a flange . In addition, a downward-facing piece of pipe was welded in as a support for one of the three hydrophones needed to position the drilling rig.
The pipe section for the hydrophone had a diameter of 325 mm, was 228 mm long and had a wall thickness of 26 mm. The metal was of poor quality because the platform manufacturer did not consider it a statically effective component. To use it, a hole had been cut in the strut with a cutting torch . Then it was connected to the strut by a fillet weld . The material of the strut had been heated twice and was under tension . The weld seam belonged to the lowest of the three weld classes used in the construction of the drilling rig and was made very thin. This point was the starting point of the strut break.
“The fillet weld is almost completely traversed by a crack on the nozzle side, which again probably started from the root layer, because there it is closest to the fusion line. As in case E, the melt line runs almost parallel to the rolling plane; the seam hardly engages in the socket. The high inclusion density was noticeable. It was up to 0.3 percent with a sulfur content of 0.02 percent; In other words, there must have been a lot of oxide inclusions. "
Paint residues on cracks indicated that these cracks must have already appeared during the construction of the oil rig. Further cracks were caused by poor quality of the welds and high stresses on the strut. From highly stressed areas, fatigue cracks extended to the perimeter of the strut. After the cracks had expanded to two thirds of the circumference, the strut broke in the storm. The other struts on pillar D were now overloaded and also broke.
The rig would not have overturned so quickly if openings on pillars C and E and on the platform had been properly closed and weather-appropriate. This allowed the columns to fill up faster.
Consequences
Det Norske Veritas
On Easter Sunday after the accident, Henrik Ibsen , who comes from the same production series as Alexander L. Kielland and serves as a flotel with 625 beds, got a list of 20 ° during a lowering exercise, which was only not bigger because one leg of the island was in shallow water touched the seabed. Det Norske Veritas then prohibited the use of Henrik Ibsen . Subsequently, it was planned to subject all 40 or so floating oil rigs and flotels to a four-week inspection on the coast. It was assumed that production losses were around 7.5 million crowns (1.5 million euros) plus inspection costs. A detailed examination of other semi-submersibles revealed that several semi-submersible had cracks similar to those of the Alexander L. Kielland . These would not have been noticed during a visual inspection at sea, only during a large 4-year inspection. Platforms with cracks had been certified by both Det Norske Veritas and Lloyds .
State commission of inquiry
A commission of inquiry was set up the day after the accident and presented its report in 1981. Up until then, the steel structure and the rescue facilities in particular had been examined in detail.
The Commission made a large number of recommendations. The state should continue to bear overall responsibility for drilling platforms. However, the monitoring during planning, construction and operation should be carried out by classification societies, since more different specialists would be available here. Manuals should also be created for this.
In the future, platforms would have to be built in such a way that errors in construction and operation would be minimized, inspections would be easy to carry out and relatively minor damage would not lead to complete failure of the steel structure.
Around 1970 experts were generally more concerned with the extent to which weld seams can affect the fatigue of steel structures. In 1976, the year the Alexander L. Kielland was shipped , new design guidelines from the British Welding Institute appeared. At that time, no classification society had set guidelines for fatigue. No one thought about the effect of inserting such a small part as the hydrophone. Only after the accident were regulations put in place.
It was also not common to use redundant systems. The Alexander L. Kielland had no reserve buoyancy on this side when support D was broken off. The commission of inquiry then demanded that future platforms must be constructed in such a way that the failure of a support does not trigger a critical situation and that reserve buoyancy is available.
It was found that only a few people on board the Alexander L. Kielland had rescue training. Of the approximately 4,000 people working on mobile platforms, only around 1,000 had such training in 1980. Exceptions had been applied for for 75% of the staff and approved by the State Maritime Directorate. One reason was the lack of funds for training. When asked about this, Ivar Sandvig, the head of the directorate, called this approach a “sense of reality”.
International Maritime Organization
At the 46th Maritime Safety Convention of the International Maritime Organization , the Norwegian delegation proposed a review of the stability requirements of the MODU Code (Code for the Construction and Equipment of Mobile Offshore Drilling Units). The MSC decided to carry out a periodic review in order to incorporate the ongoing development and the experiences from the disasters of the Alexander L. Kielland and the Ocean Ranger . At the 28th meeting of the Subcommittee on Ship Design and Equipment it was decided to change the general part of the MODU Code and to set up ad hoc working groups to regulate the installation of machinery and electrical systems. It was also agreed that there was an urgent need to revise the rescue equipment guidelines.
For example, lifeboat spaces were required for 200% of the people on board, as it had repeatedly been found that some of the boats could not be used in the event of accidents due to fire, listing or damage.
It was also required that every person on board should have a personal rescue suit in the cabin. On the possible evacuation routes and at the rescue stations, rescue suits should also be available for 200% of the people on board. In the following years, rescue suits were also heavily revised. For example, they were given loops on the back, with the help of which a person floating in the water can be better grabbed and pulled on board.
A rescue ship should still be assigned to each platform less than a nautical mile away.
Further consequences
One of the problems with the evacuation was that a lifeboat did not come loose because one of the release devices was always under tension in the swaying boat. At first, no decision was made about a change, as there had been an accident with a lifeboat some time before, in which the release mechanism opened too early and the boat hit the water from a great height. Three people died in the process. Ultimately, Norway decided on an extreme solution in the form of free fall lifeboats . However, the people on board must be trained more intensively for this. Other countries have resorted to systems in which the lifeboat can be released by means of an internally operated, mostly hydraulic, locking of the releasing hook .
Web links
- Norwegian Petroleum Museum (Norwegian, only parts in German)
- Website of the Norwegian broadcaster NRK with photos and videos (Norwegian)
- Animation of the accident on YouTube (2:39 min) on YouTube (alternatively also on Exponent.com )
- Review of "The true story of the sinking of the Alexander Kielland" by David Schraven and Vincent Burmeister
supporting documents
- Stig S. Kvendseth: Giant Discovery - A history of Ekofisk through the first 20 years , Phillips Petroleum Company Norway, Tanager (Norway), ISBN 82-991771-1-1
- Victor Bignell, Joyce Fortune: Understanding systems failures , Manchester University Press ND, 1984, ISBN 0-7190-0973-1
Only excerpts, quotations and drawings were available from these texts:
- The "Alexander L. Kielland" -accident : From a Commission Appointed by Royal Decree of 28th March, 1980: Report Presented to Ministry of Justice and Police, March 1981 (NOU 1981: 11)
- A. Hobbacher: Damage investigations into the misfortune of the half-diver "Alexander L. Kielland" , machine damage , Munich 56 (1983) 2
- J. Hoefeld: Alexander L. Kielland - causes and consequences of an oil rig accident , HANSA, Hamburg 1982
Individual evidence
- ↑ John Frederick Lancaster: Engineering catastrophes: causes and effects of major accidents . Woodhead Publishing, 2000, p. 102, ISBN 1-85573-505-9 . (English)
- ^ Conceptual Designs of Platforms . (PDF; 3.3 MB) Norwegian University of Technology and Sciences , pp. II-2-11a (English, accessed June 2, 2014)
- ^ "" Alexander L Kielland Accommodation Platform "" The Welding Institute, (English, accessed June 2, 2014).
- ^ John Frederick Lancaster: Engineering catastrophes: causes and effects of major accidents , Woodhead Publishing, 2000, p. 122, ISBN 1-85573-505-9 .
- ↑ Broken hip . In: Der Spiegel . No. 14 , 1980, pp. 238 ( online ).
- ^ Günter Bossow, Hellmut Hintermeyer: Mayday, Mayday ... Ship disasters from the Spanish Armada to the present day. Pietsch, Stuttgart, p. 180, book no. 19935 6.
- ↑ Stig S. Kvendseth: Giant Discovery - A history of Ekofisk through the first 20 years. Phillips Petroleum Company Norway, Tanager (Norway), ISBN 82-991771-1-1 , p. 117
- ↑ S. Ersland, L. Weisaeth, A. Sund: The stress upon rescuers involved in an oil rig disaster. "Alexander L. Kielland" 1980 . In: Acta Psychiatr Scand Suppl . 355, 1989, pp. 38-49. PMID 2624133 .
- ^ The erection of the death island Alexander Kielland . In: The time. No. 40/1983, p. 69.
- ↑ Marc von Lüpke: drilling platform disaster 1980: "I swam and swam". In: Spiegel Online . March 28, 2015, accessed June 9, 2018 .
- ↑ Martin Möser: Break through hydrogen - cold cracks under seam cracks . (PDF; 1.1 MB) In: Schweißtechnik , Berlin 1985, pp. 45–47 (accessed November 23, 2009)
- ↑ a b Heavy list . In: Der Spiegel . No. 17 , 1980, pp. 174 ( online ).
- ^ A b John Frederick Lancaster: Engineering catastrophes: causes and effects of major accidents . Woodhead Publishing, 2000, ISBN 1-85573-505-9 , pp. 111-112.
- ↑ Samir Mankabady: The International Maritime Organization, Volume I: International Shipping Rules . Croom Helm, 1984, ISBN 0-7099-3591-9
- ^ Günter Bossow, Hellmut Hintermeyer: Mayday, Mayday ... Ship disasters from the Spanish Armada to the present day. Pietsch, Stuttgart, p. 184, book no. 19935 6.
- ↑ John Frederick Lancaster: Engineering catastrophes: causes and effects of major accidents . Woodhead Publishing, 2000, p. 112, ISBN 1-85573-505-9 .
