The problems arose during a coiled tubing operation, a method completely new to Statfjord which was being conducted by service company BJ. This technique involves running a long, thin hollow tube down a well in order to carry out various maintenance jobs or repairs. The well remains under high pressure – almost 400 bar in the Statfjord one, which was used for gas injection.
The operation involved using hydraulic force to drive the tube down the well through an assembly of safety valves known as the Xmas tree. A number of additional safety valves were installed above the tree to help prevent the well contents from escaping. But dense gas was spotted seeping over the drill floor as the tube was pushed down the well. Efforts were made to close all available valves.
Toolpusher Hans Bergsland finally managed to shut the valves, and the actual coiled tubing was cut off with a shear ram. The blowout was halted, but the pressure in the well continued to rise and the outlook caused great concern.
Harald Luckman, at that time secretary-general of the Organisation of Petroleum Exporting Countries (Opec) and later Nigeria’s oil minister, happened to be visiting Statfjord C that day with the usual kind of press coverage which accompanies such trips. When the helicopter carrying Luckman and his retinue arrived over Statfjord, two standby ships were lying close to the A platform and spraying water over it to cool the structure down. This was done to reduce the consequences of a possible explosion and fire if the well should start leaking again. Nobody at that time knew what was happening or might occur.
The first reaction was to pump drilling mud down the borehole in an attempt to drive the gas into the formation and kill the well. This effort continued for four days, interspersed with bleeding off the pressure. But it was impossible to pump down sufficient mud, and the pressure did not decline. All production from the shaft containing the problem well was shut down, but continued from the other shaft. Before further action could be taken, it was necessary to establish what had actually happened in the well during the coiled tubing operation. The cause of the problems needed to be identified.
In an attempt to form a picture of conditions inside all the valves, X-ray film was packed around and above the wellhead so that pictures could be taken. These were a surprise – they showed not just one but five tubes stacked next to each other. That complicated a possible operation to fish out the tubing. What had happened was that the tubing, as it was being driven down the well, encountered a plug of hydrocarbon ice or hydrate. This forms when gas combined with a little water freezes under special conditions and at relatively high temperatures.
The hydraulic motor which had been pushing the tubing down the borehole was so powerful that it did not stop when the tubing encountered resistance. Instead, the tubing broke. Nobody noticed that anything unexpected had happened, and tubing continued to be driven into the bore hole. Each time a new length of tubing hit the hydrate plug, it broke off again. That continue right up to the moment gas was discovered on the drill floor.
Once it was found that the tubes were stacked side by side, the need for expert assistance to restore full control over the well became clear. Help was sought from a well-killing company in Louisiana, which had personnel in place on 20 June – 15 days after the leak was discovered. To kill the well, it had to be sealed in some way and all equipment installed above it removed. Ice plugs had previously been used very successfully to seal piping which was to be cut off, but never before on a well under such high pressure.
This was nevertheless proposed as a solution and, after long discussion, installing an ice plug immediately beneath the wellhead was identified as the best way to restore control. A detailed plan for the operation was developed. The proposal was submitted to the Norwegian Petroleum Directorate (NPD), which had been involved from the start. It was worried that the steel used for the wellhead might crack when exposed to a low temperature. For its part, operator Statoil presented calculations which showed that the plan could be implemented with an acceptable level of safety, and the NPD gave its approval.
A provisional ice container was constructed around the conductor pipe, beneath the Xmas tree. Pure drilling mud was simultaneously circulated in the same area. The actual freezing operation was primitive, using dry ice (frozen carbon dioxide) as the coolant. Batch after batch of dry ice was dispatched from land and loaded onto Statfjord A, where it was crushed, packed into the container and compressed. Then it was necessary to wait until the ice froze. The temperature was constantly monitored, and the next phase could start once it had been confirmed that an ice plug had formed on the inside of the conductor pipe.
Production on the platform was completely shut down and everyone not involved in the next phase – some 310 people – was sent across to the Polycrown flotel. Only 103 personnel were left.
The flare was extinguished at 04.30 on 23 June. At 05.36, platform manager Geir Pettersen gave the go-ahead for the operation. Residual pressure in the well above the ice plug was bled off and the Xmas tree opened. Bit by bit, the valves and the riser were disconnected and an ordinary hacksaw used to cut the coiled tubing. After just over three hours, it was possible to install a new set of valves and end the danger of losing control. But the problem had not been overcome, in that both coiled tubing and mud remained in the well. It took several months to clean it out. A snubbing rig, used to enter a well under pressure, was installed to fish out the rest of the coiled tubing. Several thousand metres had to be extracted.
Limited production from the rest of the platform was resumed on 26 June. Statfjord A was out of operation for almost two weeks, but this did not have dramatic consequences for total production from the field. Both the B and C platforms flowed at full throttle. The request from the Organisation of Petroleum Exporting Countries (Opec) to reduce output by 7.5 per cent was also still in force.
This cut followed the oil price slump in 1986, when Opec wanted to reduce the supply of crude in order to stabilise and preferably increase prices. In support of this goal, the Norwegian government had withdrawn 80 000 barrels of oil per day from the market during the last two months of 1986. It then resolved to reduce production by 7.5 per cent from the planned level for 1987. This measure continued until 30 June 1988. The restriction did not apply to day-to-day output, but to the figure for the half-year as a whole.
Interview with Geir Pettersen.
Norwegian Wire Service (NTB), 18 June 1987.
Norwegian Wire Service (NTB). 9 July 1987.
Austvik, Ole Gunnar. “Den petroleumspolitiske utfordring” . Norwegian Foreign Policy Yearbook 1988. Norwegian Institute of International Affairs (Nupi).