UNDERGROUND BLOWOUT KILLED WITH QUICK SNUBBING OPERATION

Robert Grace, Grace Shursen
Moore & Associates Inc.
Amarillo, Tex.

Gilbert Stanislaus
Trinmar Ltd. Point
Fortin, Trinidad & Tobago

Bob Cudd
Cudd Pressure Control
Woodward, Okla.

A shallow underground blowout off the island of Trinidad required quick action and the importing of snubbing equipment to kill the well and avert cratering the sea floor beneath the platform. The blowout was controlled in 16 days.

The blowout at Trintomar's Pelican platform on the east coast of Trinidad posed a most challenging well control problem. Most of the service companies with equipment to control the well were not available in this relatively remote area. The well control plan required all the pumping services on the island, and the snubbing equipment had to be flown in from the U.S.

Because of the high gas and condensate flow rates and high pressure, the blowout at the Pelican platform had the potential to destroy the entire platform, endanger the lives of many crew members, result in the loss of natural resources, and interrupt the supply of natural gas to the island of Trinidad.

PELICAN PLATFORM

Trintomar was formed in 1988 to develop the gas reserves off Trinidad's eastern shelf to provide a supply of natural gas to the National Gas Co. (Fig. 1). The geology of the eastern shelf is faulted and complex, making the drilling operations there difficult.

The geological targets required deep, high-angle, extended reach wells. In two of the first four wells on the Pelican platform, the total depths were deeper than 19,000 ft with inclination angles up to 60 and with displacements of approximately 9,000 ft.

Pelican A-4X was completed in January 1991. The well had a total depth of 14,235 ft measured depth (MD), or 13,354 ft true vertical depth MMscfd of gas and 2,200 b/d of condensate at 2,800 psi flowing tubing pressure, and the bottom hole pressure was 5,960 psi. Fig. 2 shows the well bore schematic for the A-4X.

By April 1991, the total daily production from the platform was 130 MMscfd of gas and 7,200 b/d of condensate, and the cumulative production was 23.8 bcf of gas and 1.29 million bbl of condensate.

The Pelican A-7, the seventh well on the platform, was spudded on Apr. 23, 1991. The 18 5/8-in. surface casing was run to 1,013 ft and cemented to the surface. Drilling operations continued with a 12 1/4-in. open hole (Fig. 3).

UNDERGROUND BLOWOUT

The directional data indicated that the A-7 and the A-4X were approximately 10 ft apart at a depth of 4,000 ft. Early on May 1, the A-7 inadvertently intercepted the A-4X at 4,583 ft. The bit penetrated the 13 3/8-in. casing, the 9 5/8-in. casing, and the 4 1/2-in. production tubing. The A-4X wellhead on the production deck lost pressure, and the A-7 began to flow.

The A-7 was diverted and bridged almost immediately, but the A-4X continued to blowout underground. With only 1,013 ft of surface pipe set across unconsolidated sands in the A-7, the entire platform was in danger of being lost if the blowout fractured to the sea floor under the platform.

Underground blowouts can quickly become the most difficult and dangerous of well control operations. The conditions are typically hidden, causing difficulty in analyzing the problem.

Because underground blowouts seldom occur, they are not often properly respected or diagnosed. Underground blowouts have a hidden danger because many people only associate danger with visible events, thus possibly leading them into a false sense of security.

If an underground blowout is near the sea floor, the potential for cratering is high, and the crater could form almost anywhere around the platform. A crater is most destructive when it forms directly under the rig or platform; entire rigs and production platforms have been lost in craters formed from underground blowouts.

Quick, decisive action was necessary to avoid a major disaster at the Pelican platform and to control the underground blowout in a minimum time frame. Deep, high-pressure gas operations were new to Trinidad; therefore, local services were limited.

The blowout was modeled to determine the magnitude of the flow, the zone of loss and the potential dangers of continued operations from the platform. The model was then used to design the kill plan.

INTERCEPTION

After the A-7 intercepted the A-4X, the shut-in surface pressure on the A-4X stabilized at 2,200 psi. An analysis of the surface shut-in pressure data indicated that the flow was being lost to a zone at approximately 3,600 ft. A deliverability evaluation, confirmed by Trintomar's reservoir engineers, determined that the A-4X was flowing underground at a rate of approximately 38 MMscfd of gas and 6,000 b/d of condensate.

The A-4X was then put on production at a rate of ID-MMscfd of gas and 1,000 b/d of condensate to minimize the flow underground, to reduce the extent of shallow formation charging, to conserve the natural resources, and to minimize the potential for cratering. Based on the reservoir modeling, the flow underground was reduced to 23 MMscfd of gas and 5,000 b/d of condensate.

TEMPERATURE SURVEY

A temperature survey was run in the A-7 (Fig. 4). Fig. 4 also shows the static temperature measurements from the A-3; these measurements were used to establish the geothermal gradient. The interpretation of the temperature data was complicated because the hydrocarbons lost from the A-4X flowed into another zone inside the A-7 well bore.

The temperatures down to 3,600 ft were as high as expected and consistent with other data analyses, indicating the zone being charged was at approximately 3,600 ft.

The temperature anomaly above 1,000 ft was interpreted as a charging of the sands between the sea floor and 1,000 ft. Thus, cratering and subsequent loss of the platform were possible. Further analysis was therefore warranted.

The temperatures were analyzed taking into consideration the relative distance between the two wells at each depth. The well bores are interpreted to be 2 ft apart at a depth of 1,000 ft and 5 ft apart at the sea floor. Therefore, the temperature anomaly above 1,000 ft was thought to be caused by the proximity of the two well bores and not' by gas and condensate flowing into a shallow zone. No gas or condensate was observed in the sea around the platform at any time during the kill operation, confirming this theory.

KILL PLAN

Fig. 5 is a rough schematic showing the conditions during the underground blowout. The gas and condensate flowed from the A-4X at 4,583 ft through holes in the tubulars to the A-7 well bore. As long as these conditions remained constant, operations from the platform could continue safely; kill alternatives were evaluated based on these flowing conditions.

Wire line operations in the A-4X determined that the internal diameter of the 4 1/2-in. production tubing had been reduced from 3.826 in. to 1.75-2.25 in.

Based on the kill model and the fluid dynamics required to control the well, the best and most expeditious operation involved opening the 4 1/2-in. production tubing to full internal diameter, followed by snubbing 2 7/8-in. tubing in the well down to 5,600 ft and killing the well with 18 ppg mud pumped at 10 bbl/min (Fig. 6).

The kill plan required a snubbing unit with experienced operators, but snubbing units were not readily available because high-pressure gas operations were new to Trinidad. The snubbing unit and other necessary support equipment were available in the U.S. and could be transported on a chartered 747 aircraft.

The charter company was reluctant to ship the snubbing equipment because it believed the equipment required for safe unloading from the aircraft was not available in Port of Spain, Trinidad. The charter company transferred the equipment to Trinidad only after Trintomar accepted complete responsibility for any possible lost time because of repairs or damage to the aircraft. The equipment was shipped and unloaded on May 6 without incident only 5 days after the blowout began.

The designed kill procedure required approximately 4,750 hydraulic hp. This horsepower requirement represented all the available pumping horsepower from all the service companies in Trinidad. The service companies did an excellent job of coordinating all the pumping equipment, which completely filled all the available space on the Pelican platform.

SNUBBING

The snubbing operations began on May 13. A 2/8-in. work string with a full-gauge broach was snubbed into the 4/2-in. production tubing in the A-4X. The broach encountered a tight spot at 4,623 ft. The 6-ft-long tight spot was opened to the full 3.826-in. internal diameter.

The 27/8-in. work string could then pass through the 41/2-in. tubing inside the 91/sin. casing, but not through the 41/2-in. tubing inside the 7-in. liner. To ensure that the 4/2-in. tubing was not parted, the 21/8-in. work string was snubbed to a total depth of 6,700 ft, which was 48 ft below the top of the 7-in. liner.

The work string was then snubbed out of the hole, and on May 16, the 2,-/s-in. kill string was snubbed into the well down to 5,660 ft.

The pumping equipment was tested throughout the night, and the kill operation began at 7:43 a.m. the next morning. The gas and condensate stopped flowing by 7:50 a.m., and by 8:30 a.m., the well was dead. The design and actual pumping schedules are shown in Fig. 7.