Subsea/Surface Systems
Testing begins on dual gradient drilling with seawater-driven pumps
Diamond Offshore's second generation semisubmersible Ocean New Era has been outfitted with the first seafloor drilling fluid lift system. The system, being used to drill the world's first dual gradient well, was developed over the last five years during a three-phase joint industry project (JIP) headed by Conoco and Hydril.
The SubSea MudLift Drilling (SMD) JIP partners include Conoco, Chevron, BP, Texaco, Schlumberger, Hydril, Diamond Offshore, and Global Marine. The rig will drill a well for Texaco in Green Canyon Block 136 of the Gulf of Mexico. This is well number eight on the field, and if all goes well, the system will be used in Texaco's production plan for the field.
Kenneth Smith, Project Manager for the SMD JIP, said the goal was to find a prospect where the geology was simple and well understood. "This allows us to focus more intently on the test of the SMD system," he said. The rig will drill in 910-ft water depths, which is deep enough to prove the system concept, but nowhere near its hyperbaric-chamber tested depth of 9,000 ft.
The concept of mudlift drilling has been embraced by several companies as a possible answer to the narrow drilling pressure margins present in deepwater and ultra-deepwater fields. Often, it is necessary in such wells to run contingency strings of casing to isolate troubled zones where lost circulation may occur. Operators address this problem today by trying to squeeze in more casing strings through the existing riser and blowout preventer (BOP) systems.
This is an expensive contingency, and it leaves the risk of not reaching total depth (TD) with a large enough wellbore through which to run completion strings and produce the well. Dual gradient drilling reduces the hydrostatic head of the mud column above the mudline, so that the well's casing program can be designed as if the well were onshore.
Rather than taking mud returns up the annulus of the riser, the returns are diverted into a pump at the wellhead on the seafloor, and routed to surface on a return line. This means there are two fluid densities imposing a load on the well: the drilling fluid below the mudline, and the seawater above the mudline.
While simple as a concept, the many components involved proved challenging to design. Smith said the final version of the system, the one being used on the New Era, was designed by Hydril. During Phase 2 of the JIP, all of the system components cleared the detailed design engineering, fabrication, and extensive flow-loop testing at Hydril.
Dual gradient drilling and well control procedures also had to be developed to teach drillers how to operate this revolutionary system. The development of these procedures alone accounted for 20-25 man-years of work, Smith said. The hardware and the procedures had to be developed in parallel because they were inter-related, a change in hardware capabilities could affect the procedures, and vice-versa. Phase 3, the integration phase, began in mid-2000, and involved:
- Training: The project team evolved the drilling and well control procedures into a training program. Training experienced drillers to operate a dual-gradient system was challenging since it is so different from what they are accustomed, but Smith said more than 100 people have been through the program.
- Test system: The integration of components began and a test system was built. This required that a vessel be selected. The New Era was converted to use this system. Smith said this provided the JIP with an opportunity to demonstrate how scalable the system was.
While the system on the New Era is designed to pump a maximum of 900 gal/min, the full-blown system would pump twice as fast and could handle mud weights up to 18.5 lb/gal. These might be the requirements if the system were to be used in 10,000-ft water depths. Smith said that if the first system proves successful, the full-blown system could offer operators ultra-deepwater wells with 12.25-in. open hole at total depth. That would allow an almost limitless variety of completion options, including multilaterals and extended reach.
The system uses three diaphragm pumps, landed on top of the lower marine riser package below the flex joint. The system is run with the BOP after the 20-in. casing is set. The diaphragm pumps are installed on three of the four corners of the unit. The fourth corner is dedicated to the solids processing unit.
The pumps are actually modified pulsation dampeners. Smith points out that the subsea pumps transfer energy from the seawater to the mud. Conventional triplex mud pumps on the surface actually do the pumping, which eliminates high-power subsea pumps. The design allows for a much smaller subsea system to do the job.
Drilling fluid travels down the drillpipe as it typically would, but passes through a drillstring valve prior to reaching the bit. The valve holds the mud in position when drilling is not underway. Otherwise, the mud in the pipe would fall out and the MudLift pump would continue pumping it out. Smith said much of what was done on this project focused on this fluid balance. "In every operation, the goal is managing the U-tube. So, when we stop the surface pumps, everything stops moving," Smith said.
Upon returning from downhole, the mud is diverted by a subsea-rotating device. This device provides a mechanical barrier between the mud in the well and the seawater in the riser. A stripper element runs on the drillpipe and lands inside the device, with a rotating rubber seal. The drillstring actually moves through this device when drilling ahead. The design is not intended as a well control device, but has handled up to 2,000 psi during testing, Smith said, which offers added safety before the blowout preventer closes.
Seawater is collected, filtered, and pumped down through a special dedicated line on the riser, and into the pumps. The diverted mud enters the diaphragm pump from one side and is displaced by the seawater. The seawater is ejected into the sea and the mud, other fluids, and cuttings are transported up a second line attached to the drilling riser to the surface for treatment.
Because the returns are taken to the surface, it was necessary that the subsea system handle all types of cuttings. The project team did not want to risk degrading smaller cuttings, affecting the quality of the mud, but needed to ensure that larger cuttings did not damage the system or block the return line. Thus, the return line in the test system is a 5-in. line, run outside of the riser. Cuttings larger than 1.5-in. in diameter travel into a solids processing unit where they are broken down into manageable pieces before entering the pump. This is critical for a system operating in the Gulf of Mexico where gumbo soil can deliver chunks of clay the size of basketballs.