This story is by David Hammer and Dan Shea

The Times-Picayune
May 6, 2010

The investigation into what went wrong when the Deepwater Horizon exploded April 20 and started spilling millions of gallons of oil into the Gulf of Mexico is sure to find several engineering failures, from cement seals that didn't hold back a powerful gas bubble to a 450-ton, 40-foot-tall blowout preventer, a stack of metal valves and pistons that each failed to close off the well.

There was, however, a simpler protection against the disaster: mud. An attorney representing a witness says oil giant BP and the owner of the drilling platform, Switzerland-based Transocean, Ltd, started to remove a mud barrier before a final cement plug was installed, a move industry experts say weakens control of the well in an emergency.

When the explosion occurred, BP was attempting to seal off an exploratory well. The company had succeeded in tapping into a reservoir of oil, and it was capping the well so it could leave and set up more permanent operations to extract its riches.

In order to properly cap a well, drillers rely on three lines of defense to protect themselves from an explosive blowout: a column of heavy mud in the well itself and in the drilling riser that runs up to the rig; at least two cement plugs that fit in the well with a column of mud between them; and a blowout preventer that is supposed to seal the well if the mud and plugs all fail.

In the case of the Deepwater Horizon, Scott Bickford, a lawyer for a rig worker who survived the explosions, said the mud was being extracted from the riser before the top cement cap was in place, and a statement by cementing contractor Haliburton confirmed the top cap was not installed.

Mud could have averted catastrophe

If all of the mud had still been present, it would have helped push back against the gas burping up toward the rig, though it might not have held it back indefinitely.

Click to see a full size PDF file about what happened on the Deepwater Horizon oil rig.When the gas got to the sea floor, the third defense - the blowout preventer -- also failed, and it has continued to fail for weeks as unmanned submarines have tried to get it to engage.

BP declined to answer questions about exactly how far along they were in the process of closing the well head 5,000 feet below the Deepwater Horizon rig when the explosion occurred.

But Halliburton said in a statement that it had completed pouring cement that lines the well 20 hours before the blowout. After that cement lining is done, the federal Minerals Management Service requires at least two prefabricated cement plugs to be placed at the bottom of the well and farther up, with mud packed in between. Halliburton's official statement shows there was still one more cement plug to be inserted.

"Well operations had not yet reached the point requiring the placement of the final cement plug which would enable the planned temporary abandonment of the well, consistent with normal oilfield practice," the Halliburton statement said.

Lawsuit disputes Halliburton statement

But Bickford's client, who was working immediately next to the drill floor at the time of the explosion, claims the rig operators had already started pumping mud out of the riser. Bickford said his client, whose identity he wants to protect for now, will allege human error in the decision to start removing the mud barrier before the well was totally capped.

Bickford said his client is the survivor of the rig explosions who called into the April 29 "Mark Levin Show," a nationally syndicated talk show out of WABC in New York, and gave perhaps the most detailed witness description available so far of what was taking place at the time. He used the assumed name "James" on the show.

BP photo

The bottom of the damaged and leaking marine riser sits atop the failed blowout preventer on the floor of the Gulf of Mexico. The main tube of the riser guides the drill pipe into the well, while the other pipes carry hydraulic fluid to operate the blowout preventer.

"We had set the bottom cement plug," the caller said. "At that point the BOP stack, the blowout preventer, was tested. I don't know the results of that test. However, it must have passed because at that point they elected to displace the marine riser from the vessel to the sea floor. They displaced all the mud out to the riser preparing to unlatch from the well two days later. So they displaced it with sea water."

Bickford's client went on to say that the crew opened a valve on the well head, allowing a huge kick of gas to push the seawater out the top of the marine riser and all the way to the top of the rig tower, 240 feet in the air. The resulting explosion probably instantly killed his colleagues who were in the path of the gas, "James" said.

Explosion doomed oil rig

Crew members were caught off-guard by a gas-bubble kick that spewed watered-down mud and an invisible plume of heavy gas onto the rig, igniting a fiery explosion that killed 11 crew members and doomed the rig.

Bickford said his client saw mud being pumped out of the riser and onto boats that normally collect the mud in tanks. Another lawyer, Stuart Smith, said he represents fishermen who witnessed the explosion and saw the mud being extracted beforehand.

BP spokesmen have declined to confirm or deny these descriptions of events, saying the details will come out as a result of the ongoing investigation.

Other lawsuits by rig workers paint a similar picture. Bill Johnson, a Transocean deck pusher with 35 years of experience on oil rigs, was injured in the explosion and has sued his employer, BP, Halliburton and others in Galveston County, Texas. Johnson's attorney, Kurt Arnold of Houston, said Johnson had a meeting with a BP supervisor about 10 hours before the explosion and was told "things were plugged in the well and good to go. He thinks in retrospect the company man was not following procedure."

Another one of Arnold's clients, roustabout Nick Watson, said mud came back up the h*** so suddenly before the explosion that he was trying to wipe it away from his eyes on the deck when the power went out and the first explosion came, Arnold said.

'Mud weight is the first round of defense'

If the final cement plug wasn't in place yet, removing the mud would be at odds with "good oil-field practice" outlined in 2003 by the federal Minerals Management Service. The MMS report, prepared by WEST Engineering Services, warns against single-point failures -- counting on one mode of protection -- by saying that "mud weight is the first round of defense against a kick, followed up by" the blowout preventer. Removing the mud left the blowout preventer as the only failsafe.

"To displace mud above the position of the upper plug with water before setting the upper plug means that you are relying on one barrier for the duration; this is not good," said a deepwater drilling expert who did not want to be identified because he does business with BP. The expert is not involved in the Deepwater Horizon project.

While the mud could have given workers more time to react to the blowout, the accident itself and the oil spill originated in the failure of the cement and the failure of the blowout protector.

Blowouts often related to cement problems

Blowouts are not unprecedented, and often they are caused by cementing failures. An MMS study found that half of 39 blowouts on offshore rigs from 1992 to 2006 were related to cement problems.

Cement has two roles in oil exploration: It seals the pipe lining the well from the bedrock around it, and it is used to seal wells on the inside before abandoning them. It's not known which of the two cementing jobs was the culprit in the BP accident.

Even with the problems with cement seals and the weakening of the mud barrier, the blowout preventer, or BOP, a contraption built by Cameron International, still could have blocked the oil gusher. Unfortunately, those devices, too, have had doc**ented troubles. Transocean Chief Executive Officer Steve Newman reported "a handful of BOP problems" during a call with stock analysts last year, although he said "they were anomalies."

According to internal BP doc**ents obtained by The Times-Picayune, the preventer on the Deepwater Horizon's well head had a series of six valves and "pipe rams" that are activated by hydraulic pistons and constrict around the drill pipe to close off the well. BP said those valves failed to close the well before the rig was abandoned. In addition, there's a last-ditch mechanism, called a shear ram, that is supposed to use high pressure to slice clear through the drill pipe and shut off the wh*** opening.

But shear rams have a weakness. They are not engineered to cut through tool joints, the knuckles where sections of the drill pipe are connected every 30 feet. That means that about 10 percent of a pipe is made up of tool joints that a shear ram isn't strong enough to penetrate, said Per Holand, a drilling expert from Norway who has advised the MMS.

"If they do not know the exact location of the tool joint, they would normally close a pipe ram and lower the drillpipe until it stops against the pipe ram to ensure that the shear blind ram does not hit a tool joint," Holand said. "This may of course be difficult if you have a crisis on the rig."

The removal of the mud could have limited the amount of time the crew had to work through the process Holand described.

The shear ram is activated by a button on a control panel on the drill ship. An MMS safety alert in 2000 urged drill operators on the Outer Continental Shelf to have a backup method for activating the blowout preventer.

Blowout preventer backup not required

But the United States does not require the acoustic backup system that must be used in Norway, Canada and Brazil. Holand said such an acoustic system could have helped avert such a massive spill from the Deepwater Horizon well if the section of pipe inside the blowout preventer was normal-sized. But if there were tool joints inside the preventer, an acoustic trigger system "may not have worked" anyway, Holand said.

Robots on the seafloor have been unable to activate the shear ram using a manual switch.

Even if a tool joint wasn't in its way, the shear ram may not have been strong enough to cut through the pipe under the intense conditions at the bottom of the sea, where fluid inside the well bore may be as hot as 400 degrees and the water on the seabed outside can be just above freezing. The shear rams are rarely, if ever, put to the test in real-life emergencies.

Because the shear rams are the prevention method of last resort and would destroy any drill pipe if used -- costing oil companies a tremendous sum of money -- they are tested on location just to see if they move, without any pipe getting cut. The standards for manufacturing them with enough force to actually cut a drill in two at the bottom of the sea are all based on formulas.

In a September 2004 study for the MMS, researchers from WEST Engineering found that BOP manufacturers were not using the best models for calculating the necessary force and were not adjusting the force according to different types of pipes.

With all of these potential Achilles' heels, it's amazing that oil companies and regulators haven't prepared for the possibility that all of the redundant protections could fail at once, said Mark Davis, director of the Institute on Water Resources Law and Policy at Tulane University Law School.

"It doesn't matter how many levels of sophistication there are in the blowout prevention device; if you have nothing to fall back on that's when a spill becomes catastrophic," he said. "We in New Orleans know, this is almost like building levees, you can build them with the expectation that they will hold in every event, but we know there's risk of something unknown and unprepared for. The risk of harm is so great, that's why we need a backup system on the two-to-three-day horizon, not 60-to-90 days."

David Hammer can be reached at dhammer@timespicayune.com or 504.826.3322. Dan Shea can be reached at dshea@timespicayune.com or 504.826.3391.