More on stopping leaks

Henry Fountain has an article in The New York Times about blowout preventers:

At 53 feet tall and 325 tons, the blowout preventer that occupies center stage in the worst American oil spill in two decades is a huge stack of equipment. But it is also a jumble of contradictions. It’s simple: a brute monster capable of strangling a pipe or, in more desperate circumstances, beheading it, to block the flow of oil and gas at enormous pressures from a formation deep underground. But it is also complex, controlled by an elegant brain of elaborate circuits custom built from pipes, not wires, and using hydraulic fluid instead of electrons.
Most of the time the device, commonly called a BOP, does little but sit on the seabed as drilling pipe and liquids travel through its main bore, which is usually about a foot and a half wide. But, in the occasional frenetic moments when the BOP is called into action, it must work. Ordinarily no one ever hears about blowout preventers. Now the world has.
Blowout preventers are essential for drilling on land or underwater, and the rig accident has prompted talk of improvements, like the use of modular parts or better materials. And the broader application of different drilling techniques may help operators become less reliant on the devices.
Catastrophic failures are rare, but the devices are not without problems. Hydraulic circuits can leak, seals can erode, and other problems can crop up when the devices are tested, as they are supposed to be regularly. And as the 20 April blowout in the Gulf of Mexico showed, a device’s redundant systems and backups may not help. Investigators still do not know exactly why this BOP, which was tested ten days before the accident, did not do its job.
But there is currently no alternative to the use of blowout preventers on many wells. “As the second of two barriers to containing formation pressures, the BOP is integral to doing our job,” said John Rogers Smith, an associate professor of petroleum engineering at Louisiana State University.
The key to safely drilling for oil or gas is controlling the pressure in the well-hole. The primary method involves circulating special fluid, generically called “mud”, down through the drill pipe and back up the space between the pipe and a larger pipe called a casing. The mud recipe can be altered to make it lighter or heavier as needed. As long as the hydrostatic pressure of the column of mud exceeds the pressure in the formation being drilled, the well remains under control.
But, if the drill bit hits an area of higher pressure, there can be a surge of oil or gas into the mud— a “kick” in oil-speak. That is when operators on the drilling rig will activate the blowout preventer to block the upward flow of higher-pressure mud which, if not controlled, can quickly be followed by oil and gas.
In the blowout preventer, one or more massive rams mounted perpendicular to the flow can be activated, sealing the space between the drill pipe and the bore of the preventer, covering the opening if there is no drill pipe, or even shearing the pipe if necessary. Another device on the stack, a doughnutlike rubber ring called an annular preventer, can seal the space between the drill pipe and the bore but still allow the pipe to function.
However the flow is blocked, the mud can be diverted into a separate line with a valve called a choke. By closing this valve, the open loop of circulating mud becomes a closed one, and back pressure builds until it exceeds the pressure of the kick. Then heavier mud can be circulated and drilling can be resumed.
“The whole point of using the BOP to react to a kick, and control it properly, is to prevent it from becoming a blowout,” Dr. Smith said.
The principle of using brute-force rams to control a well was developed nearly a century ago. “The basic function hasn’t changed,” said Bob Sherrill, who built and repaired blowout preventers for 20 years and now runs Blackwater Subsea, a Houston company that supplies personnel for deepwater work. “What has changed are the materials— they’ve gotten a lot more sophisticated, a lot stronger.” They have also been made more corrosion resistant, to counteract problems caused largely by hydrogen sulfide gas found in oil deposits. Still, Mr. Sherrill said, the harsh conditions mean that preventers must be rebuilt every seven years or so. “As you cycle these things back and forth, you end up with small scratches,” he said. “Cavities wear out.”
Blowout preventers also changed when offshore drilling began in earnest in the 1950s and ’60s. Preventers used on land are far easier to repair, and the rams can be locked in place manually or closed with wrenches if hydraulics fail. In water, below about 1,000 feet, they can be serviced only by robotic submersibles, and locking the rams in place requires a second hydraulic system. There is also no way to close them by hand if the hydraulics fail. So the control systems on subsea BOPs are far more elaborate and redundant, with two identical pods on each stack.
Those pods are huge, twenty feet tall in some cases, and filled with a hundred or more hydraulic valves, electrically operated solenoids, and other devices. The works are enclosed to protect them from pressure and moisture but, when exposed, the gleaming array of pipes and switches, fabricated from high-strength steel, looks like a techno version of an old telephone operator’s console.
Graeme Reynolds, manager of BOP controls at Oceaneering International, a company that is best known for its robotic submersibles used in deepwater work, said pods had to be custom-built for each blowout preventer. “We can’t go into the industrial hydraulics market and buy stuff that will satisfy us,” he said. “It won’t meet our thermal criteria, it won’t meet our pressures. So we have to make all that stuff ourselves.” As a result, they can be extremely expensive, as much as $18 million or more for the controls on a typical deepwater BOP.
In normal use, the controls are activated by an electrical line that accompanies a hydraulic line running from the drill rig. If a decision is made to close a ram, a signal activates solenoids that open valves, allowing water-based hydraulic fluid to flow into the proper cylinders on the stack. Special pressure tanks on the drill ship called accumulators, which contain hydraulic fluid and a charge of nitrogen, provide a burst of power to close the rams, usually in about thirty seconds.
But the control pods have backup systems, including accumulators on the stack itself that can provide enough hydraulic power to close rams if power is lost from the surface. A deadman device fires some of the switches if both electric and hydraulic power are lost. (A 2003 report for the Minerals Management Service, the federal agency that oversees offshore drilling, found that deadman devices often were not armed because of fear that they would activate prematurely, necessitating costly fixes. BP said the deadman switch did not activate in the 20 April blowout.)
In Norway and Brazil, another backup is required: a switch that works on acoustic signals received from the drill ship. The industry in the United States has successfully fought proposals to require the switches, arguing that they are unreliable. And industry experts pointed out that in the current spill, if the regular and deadman switches could not activate the rams, an acoustic switch would not have worked either.
As a final backup, BOPs must be able to be activated by robotic submersibles. So the control units have special valves that can use hydraulic fluid provided by the submersible using a probe called a hot stab. BP officials said that since the accident they had been able to activate some of the rams to some degree using this method.
If the blowout preventer is damaged or contains an unsealable section of pipe, the best hope for stopping the leak, other than drilling a relief well, is to route heavy mud around the preventer stack and into the well. This would involve first reconfiguring the preventer, something that is difficult but not impossible, experts say. BP officials say they are exploring this option. Other than that, though, a damaged blowout preventer is really not repairable until it is brought to the surface.
Mr. Reynolds of Oceaneering has devised a modular system of swappable parts so that a submersible can “go down, grab a hold of something, pull it out, and plug another one in.” The approach is similar to what NASA has employed to repair the Hubble. While it would not be useful in accidents like this, it would help with maintenance. But for the approach to really work, BOPs would have to be redesigned.
For the industry, a different approach to drilling might mean less reliance on blowout preventers, although they would still have to be used. In so-called managed pressure drilling, the wellhead is sealed by a rotating rubber cone, creating a closed-loop mud circulation system that allows for more precise control over pressures.
The technique raises other problems, and so far has not been used much with floating oil rigs or subsea blowout preventers, said Dr. Smith of LSU. “But it would allow us to respond faster to abnormal conditions,” he said.

Rico says having an acoustic switch might have saved BP a lot of money...