Current Events

Join us as we watch the crisis unfolding

May 3rd, 2010

Blowout

My apologies for the long gaps between Current Events postings. I'm working on the next-to-last version of the page proofs of When Oil Peaked. One more version, and the book should be out in October.

Meanwhile, back at the ranch, the Gulf Coast oil spill is getting a lot of attention. I realize that the oil company, the drilling contractor, and the government are focused on containing the oil spill. Right now, diagnosing the cause of the spill is of secondary interest. The information available through the news media is fragmentary, and sometimes contradictory. This is an overview about what to look for as the story unwinds.

The first line is not having blowouts. The second line is controlling a blowout after it happens.

I should explain that I have no official expertise about oil blowouts. However, my family history goes back to my father's experience with the 1928 Wild Mary blowout in Oklahoma City. Uncontrolled wells, blowouts, are an aspect of the oil business that you cannot ignore.

There are two lines of defense for avoiding blowouts. The first line is not having blowouts. The second line is controlling a blowout after it happens. Before 1900, all wells were drilled with cable tools; a chisel bit jerked up and down on the end of a cable. The hole was empty except for a few feet of water at the bottom. A major encounter with oil or gas was automatically a blowout. Those were the scenes that figured in old movies, with the oil gushing over the top of the derrick.

Controlling the flow of oil, gas, and water into the hole became possible with the 1901 introduction of the rotary rig. Drill pipe was used, with a cutting bit at the bottom. Mud, originally a clay-water mixture, was circulated down the drill pipe and back up in the annulus between the pipe and the rock. The pressure of the mud column was slightly greater than the pressure of the natural fluids in the rock. Oil, gas, or water would not flow into the hole.

The rule of thumb was that the natural pressure on the fluid would be equal to the pressure of a column of water extending all the way up to the surface. The rule of thumb broke down when abnormally high pressures were encountered beneath the Gulf Coast. An account of the discovery and an explanation is on pages 89 to 101 of Hubbert's Peak. (If you don't have a copy, the tag in the lower right corner of this website's opening page is a magic carpet that takes you to the book listing on amazon.com.) The higher pressures at depth were matched by lacing the mud with finely ground heavy material, like the mineral barite (BaSO4). Letting the mud density get too low could invite a blowout.

There are two possible scenarios for the Gulf Coast blowout. The drilling rig had just finished running casing in the hole and pumping down cement to fill the annulus between the casing and the surrounding rock. The blowout could have taken one of two forms:

  • The density of the newly-emplaced cement was too low and natural gas bubbled up, creating an open channel through the unset cement. Making the cement denser involves adding heavy material, like the barite used in the drilling mud.
  • At the bottom of the steel casing was a casing "shoe." There are several casing shoe designs, some of them with different names. All of them include a one-way valve that allows fluids, including unset cement, to flow out of the casing into the annulus, but nothing in the annulus can flow back into the casing. My initial guess as to the cause of the Gulf Coast blowout was failure of the casing shoe. On the Internet, there are previous examples of casing shoe failures.

At the moment, I have seen no news stories that even speculate on whether the Gulf Coast blowout was though the unset cement in the annulus or whether it was up the inside of the casing. It isn't likely to be both.

The second line of defense is the blowout preventer. It isn't a "preventer," the blowout has already happened. It's actually a "blowout container." With that warning, I'll still refer to it as a "blowout preventer."

The blowout preventer has to be anchored to something. The standard procedure is to drill an initial large-diameter hole from the surface down into hard rock and cement surface casing, typically about 20 inches in diameter, in the hole. The blowout preventer is attached, securely we hope, to the top of the surface casing. If you have to close the blowout preventer the cement bond between the surface casing and the surrounding rock better not break, and it better not leak. The 1969 blowout in the Santa Barbara Channel was a leak around the surface casing. Further, the surface casing has to extend deep enough to protect shallow groundwater horizons from being invaded by the upcoming natural gas and oil. My brief involvement with this problem is on page 101 of Hubbert's Peak.

Normally, on the drilling rig there are plenty of big red buttons that close the blowout preventer. The blowout preventer itself is a massive stack of hydraulic pistons that push sealing "rams" into the space immediately above the surface casing. One type of ram has semicircular indentations on each side to fit around the pipe in the hole; this would close a blowout from the annulus. The other type of ram is called a blind ram or a shear ram. It has the purpose of pinching the steel pipe shut or shearing off the pipe. We're talking heavy duty equipment.

The blowout preventer in the Gulf Coast well was manufactured by Cameron International, formerly the Cameron Iron Works, in Houston. Beginning in 1920, Cameron developed an outstanding reputation. That does not prove that the Gulf blowout preventer was not defective, but in my homemade estimate a mechanically-defective blowout preventer is one of the less-likely scenarios.

More likely, the blowout preventer was not hitched up properly. The blowout preventer needs two connections: one connection to a command to close the preventer, the other is a connection to a source of energy. Because none of the commands that were supposed to close the blowout preventer worked, my question is whether something was not connected properly. This can happen. I was the geological adviser to a series of oil and gas investment programs around 1980. One of our wells turned out to be the largest gas well in the history of New York State. There was a blowout preventer on the well, but the driller had not connected it up. Fortunately, the well did not catch fire and was brought under control the following day.

There are still options being explored for trapping the oil. Worst case is no containment until the relief well drills close to the bottom of the existing hole, about three months. The oil company, BP, that hired the drilling rig bears the primary fiscal responsibility for the blowout. However, lawyers are already suing the drilling company and Cameron as well as BP. BP was formerly British Petroleum, which was formerly the Anglo Persian Oil Company in Iran.

When I was running fall-break field trips to Mammoth Lakes, CA, Princeton University furnished us Exxon credit cards to buy gasoline. The year after the Exxon Valdez oil spill, we found that the only Exxon filling station in Mammoth had closed. The local people simply stopped buying at Exxon.

There will certainly be lots of new regulations about blowout preventers. It's called "locking the barn door after the horse was stolen" or "fighting the previous war." I'd suggest a broader approach to safety, starting with having fewer blowouts. There are plenty of other hazards on drilling rigs that should be reduced.

I should end by saying something nice about BP. Their well is producing 5000 barrels of oil per day. Put 5000 in your calculator, multiply by 365 days in a year, times $80 per barrel, and wheee . . . lots of money per year. Congratulations, BP. You drilled a good well. I'm sorry that 11 workers died.


 
 
© 2001—2011 - K. Deffeyes All Rights Reserved.
SD  site by Deffeyes Design