Part of a crane on the south side of the Draugen platform broke during trials, following an extension to the boom in order to lengthen its reach. Part of the extension with a 15-tonne test bag and rope went into the sea.
The only personal injury was suffered by a crane operator who complained afterwards of neck pains. No damage was caused to the platform because the break occurred 25 metres out over the sea.
Things could have gone worse in somewhat different circumstances, according to an investigation report composed by Shell itself. When the crane rope broke, it whipped back into a topside area which had not been cordoned off. Serious injuries could have been incurred if anyone had been passing by.
After the incident, all personnel on Draugen were accounted for and a helicopter was requisitioned to look for possible oil leaks in the sea. Underwater cameras were used to find the section of boom, while also checking to see whether any damage had been done to subsea installations.
The boom section was retrieved four days later. An inspection revealed that it had been badly designed, with its bending moment transferred to the wrong position.
Regular production was unaffected by the incident.
Historically, a number of serious crane and lifting incidents have occurred on the NCS and such accidents offer a big potential for damage. Even though far too many incidents involving dropped objects had been recorded on Norwegian offshore facilities, the Draugen event was not investigated by the PSA.
Its investigations were confined to incidents regarded as the most serious, with the operators left to conduct an inquiry and report their findings to the regulator.
An undesirable incident occurred on Draugen during a wireline operation on 4 December 2010 to replace a gas lift valve with a new type.
The problem was that this device was installed some distance down the well. To retrieve it, a subsurface safety valve which sat higher up the borehole had to be pulled out first. This valve, which is found in all wells to provide a downhole barrier against a possible blowout, got struck as it was being manoeuvred through the wellhead on the platform.
This left only one barrier in the well – the upper swab valve, which could be operated manually. The remaining barriers in the Xmas tree were blocked. Shell, the PSA, and Seawall – responsible for the wireline job – joined forces to come up with a solution to this challenge, which was now serious.
The approach adopted was to return the safety valve to its position in the well. Two mechanical bridge plugs were then inserted in the borehole above the valve. After four days of intensive work, the well was secured.
Shell had notified the PSA of the incident immediately after it occurred. The regulator considered the event to be so serious that it launched an investigation.
This concluded that, since the well at one point only had a single remaining barrier to a hydrocarbon outflow, the incident involved a major accident potential.
The PSA also identified a big threat of dropped objects during rigging up, and that a possible wireline failure could have caused an uncontrolled fall of the toolstring down the well.
According to the report, Shell had committed a number of regulatory breaches – including inadequate management, risk assessment and well control. The failure to halt production from the rest of the wells was identified by the PSA as a gross breach of the regulations by Shell and Seawell.
A possible wireline failure could have eliminated the final well barrier, leaving nothing to contain the wellstream and leading to a blowout. Shell received a notice of an order from the PSA. This is part of the PSA’s administrative process where the recipient is asked to assess the factual basis, and represents only a first step before an administrative decision is made.
An order, on the other hand, is an administrative decision made pursuant to the regulations and legally binding on the recipient. The one issued to Shell in May 2011 was in three parts, which had to be complied with by 1 July of the same year.
First, the company had to assess and test its own criteria for continuing well intervention while maintaining production from other wells in an emergency when barriers are lost.
Second, its own criteria for conducting internal investigations or investigating serious incidents with loss of barriers in the well area on Draugen had to be assessed.
Finally, Shell had to assess its own criteria for prudent well control and the need for risk-reducing measures when implementing well intervention on the Draugen facility.
Wind speeds up to 40 knots combined with temperatures down to -7°C proved too much for the Draugen platform, and production had to be shut down on 22 February 2010.
That same evening, the alarm sounded and everyone on board except those with emergency response duties mustered to the lifeboats.
The alert had been given because equipment in the fire-fighting cabinet on the helideck had frozen. However, none of the 79 people on board were sent ashore.
This was the second time Draugen was shut down because of subzero temperatures. The previous occasion was the unusually cold January of 2001.
Nobody was hurt when four people ended up in the sea after a man-overboard (MOB) boat overturned near the Draugen platform.
All were quickly recovered from the water.
Those aboard this small, fast rescue craft hailed from the Ocean Sky standby ship stationed on Draugen, and the weather was good with a virtually calm sea when the accident happened. The four were picked up by another MOB boat from the Tordis Knudtsen tanker. All were wearing survival suits and in good condition, without physical injuries from the incident.
A break occurred in the loading hose while discharging oil from Draugen to the Navion Scandia shuttle tanker, allowing three to 10 cubic metres of oil to escape to the sea.
This was regarded as a small spill, but the PSA nevertheless took a serious view of the incident and launched a formal investigation. The leak followed hard on the heels of a broken loading hose on Statfjord just before Christmas 2007, when about 4 000 cubic metres of oil were spilt. According to the PSA, these two incidents had to be seen together. Shell had also experienced a number of crude oil discharges.
It now transpired that the Draugen spill was caused by the fact that a safety catch in the loading system had behaved exactly as it was supposed to. An overpressure had occurred in the hose, releasing the safety catch. The discharge comprised oil contained at that moment in the hose, which was voided to reduce the pressure.
Why the overpressure had arisen in the first place, however, was something Shell needed to investigate more closely.
The PSA nevertheless ordered Norske Shell to identify and make necessary improvements to its management of loading operations, and to assess whether other aspects ought to be improved.
Teekay Shipping Norway AS, which was responsible for Navion Scandia, also received orders following the oil discharges on both Draugen and Statfjord.
Plans were already afoot to replace the loading buoy. Investment in new equipment was needed because Draugen’s producing life had been extended.
Read more in the article on the loading buoy.
The freefall lifeboats on Draugen created problems when it transpired that their superstructure was unable to cope with the stresses experienced when being launched into the sea.
When the Petroleum Safety Authority Norway (PSA) became aware of the problems, it ordered Norske Shell to cease Draugen production and told Statoil to shut down Snorre A. Both companies had applied to remain on stream with a minimum workforce. The PSA also received exemption requests for six other facilities on the NCS with the same type of lifeboat. These were Statoil’s Troll A and Heidrun platforms and the Oseberg C, Brage, Troll B and Njord A installations operated by Norsk Hydro.
The applications for Draugen and Snorre A were rejected, but the operators opted to stay on stream – to the surprise of the PSA, which believed output should have ceased immediately. Exemptions were given to the other six facilities because their lifeboats could also be lowered manually. Extra requirements were also imposed.
In addition to a dedicated standby vessel, these included increased helicopter availability, expanded radar coverage and plans to shut down in bad weather.
The exemption applications for Draugen and Snorre A had not included lowering the lifeboats as an alternative. In the PSA’s view, compensatory measures for evacuation from these facilities were insufficient to achieve acceptable safety.
Shell had no option but to shut down Draugen and send its personnel ashore. Thirty-eight people remained on board while the lifeboats were reinforced to cope with a freefall launch.
The operator was less than pleased with the PSA order. While the regulator had initially approved the lifeboats, the initiative to upgrade them came from the industry itself.
Draugen could come back on stream after less than a day when the PSA gave an exemption. Shell had sent it a new application which outlined compensatory measures.
These included a new release mechanism for manual lowering of the lifeboats and intensive training in such launching. The exemption ran until 1 December, when all the lifeboats had to be upgraded.
The shutdown on Snorre A continued.
It does not happen often. But a helicopter en route from Åsgard B to Kristiansund with 18 people on board was hit by lightning on 15 March 2005 and made an emergency landing on Draugen. Nobody was hurt.
The Super Puma machine was flying through snowclouds when everyone suddenly heard a loud bang, which the pilots took to be a lightning strike. They decided to land immediately, for safety’s sake. Draugen was the nearest safe haven, and the pilots opted to land there even though no fault reports or alarms were showing.
An inspection of the machine revealed soot marks on one of the rotors. The pilots were probably right in assuming that the bang heard by the passengers was a lightning strike.
The helicopter blocked the helideck on Draugen, which meant it was impossible to land another machine to take the passengers on to Åsgard.
So how did this story end? A report was drawn up by the Sintef research institute for Norske Shell, a number of other oil companies and the Norwegian Civil Aviation Authority. It concluded that three of the 13 helicopter accidents in the North Sea – 11 in the UK sector and two on the NCS – between 1999 and 2009 were related to lightning strikes. None of these occurred on the Norwegian side, and none led to fatalities. The emergency landing on Draugen in 2005 was not included in this study, since it covered the North Sea alone.
In Sintef’s view, helicopter flights would always be vulnerable to this kind of risk. No satisfactory method existed for detecting lightning ahead of an actual strike.
It had nevertheless become significantly safer to fly in this period since new technology meant that the damage caused by lightning would be lower in 2009 than a decade earlier.
No accidents related to lightning occurred on the NCS in 1999-2009, but an annual average of two-three incidents involving such strikes were reported during this period.
The field was shut down again after a minor leak was found in a condensate pipeline on the platform. Output was suspended for safety’s sake, but none of the 70 people on board were evacuated.
Draugen had ceased production a number of times in recent years as a result of leaks.
The leaking pipe was found to be corroded. Since the platform was not producing anyway, all condensate piping in the relevant area was checked and turned out to have more rust than expected.
Draugen remained shut down for more than a week to correct this extensive corrosion damage.
Storms with wind speeds well above hurricane force struck the Halten Bank in January 2005. In itself, this weather posed no threat to either platform or production on Draugen.
Before it began, however, damage had been caused to the hose used to transfer oil from Draugen’s loading buoy to shuttle tankers for shipment to land. Although a new loading hose was to hand and personnel were in place, the replacement could not be carried out until wind and waves had calmed down.
This meant in turn that oil could not be transferred to tankers. The platform’s storage cells filled up, and production had to be suspended because no more space was available.
The wind was gusting up to 85 knots – which is pretty high when a hurricane starts at 65. Draugen was shut down for more than a week. January 2005 was characterised by a lot of bad weather, starting with the Gudrun cyclone over southern Norway on 8-9 January.
Then the Hårek storm hit the northern coast from Nord-Trøndelag to the Lofoten islands on 10 January, and Inga struck further south from Egersund to Kristiansund on 11-12 January.
The extreme wind speeds created significant wave heights – the mean measurement (from trough to crest) of the highest third of the waves – of up to 10 metres. Individual waves can be up to twice as high.
A helicopter was flying from Kvernberget airport to Draugen when a door to the passenger cabin fell off. Nineteen passengers and two pilots were on board, but nobody was injured.
The warning lights for the right-hand door had come on immediately after departure, while the machine hung in the air about 10 metres above ground before setting off westwards.
Air controllers in the tower had seen the door fall off onto the tarmac. The helicopter set down and taxied over to the apron before the passengers were given a debriefing. They were then transferred to a new helicopter in order to get to work on Draugen. The incident was an unpleasant experience both for them and for the pilots.
The SHT investigated the incident and reached the following conclusion:
The SHT considers that this was a potentially serious incident. Had the door fallen off at cruising speed, it could have hit vital components on the helicopter with the damage and possible accident that this would have involved. In addition, third parties could have been injured.
This flight had been the second for the helicopter that day, and the co-pilot reported problems with locking the door for the first journey. A flight technician had corrected the fault before take-off.
The SHT’s report also noted that operator CHC Helicopter Service had stated that a concentration on keeping to the timetable could have contributed to the failure to identify and correct the fault before the first flight of the day.
The report found that the incident could primarily be traced back to work done a little earlier on the emergency opening system. Poor assembly meant that vibration caused to door to loosen.
According to the commission, the company took the affair seriously and had introduced a number of changes to its routines and inspections.
A similar incident occurred in January 2010 when the cockpit door fell off a Super Puma helicopter en route to the Sleipner field in the North Sea. This was also a CHC machine.