Refurbishment of the Ballast-Water System of the Gravity-Based Structure Beryl Alpha
- Klaus Seume (Mobil North Sea Ltd.) | James M. Gilchrist (Mobil North Sea Ltd.)
- Document ID
- Society of Petroleum Engineers
- SPE Production Engineering
- Publication Date
- November 1989
- Document Type
- Journal Paper
- 473 - 478
- 1989. Society of Petroleum Engineers
- 1.14 Casing and Cementing, 4.2.4 Risers, 2.4.3 Sand/Solids Control, 1.7.5 Well Control, 4.2.3 Materials and Corrosion, 4.1.5 Processing Equipment, 4.1.6 Compressors, Engines and Turbines, 7.2.5 Emergency Preparedness and Training, 1.6 Drilling Operations, 5.1.1 Exploration, Development, Structural Geology, 4.2 Pipelines, Flowlines and Risers
- 1 in the last 30 days
- 115 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 5.00|
|SPE Non-Member Price:||USD 35.00|
In Sept. 1985, the Beryl Alpha platform in the North Sea experienced a leak in its ballast-water system in the utility shaft 328ft [100 m] below sea level. The importance of the ballast-water system for plat-form operations and the particular location of the leak led to the develop-ment and implementation of unique and extensive repair procedures. This experience may be of interest to design engineers and repair contractors. The total repair program took more than 3 years and has cost $6.5 million [francs 10 million*].
The Beryl Alpha platform stands in the British North Sea Block 9/13a, 180 miles [290 km] northeast of Aberdeen. The platform is a "Condeep"-type gravity-based structure with an oil storage capacity of 970,000 bbl [154x 10(3) m3]. The Beryl Alpha stores and exports through two articulated loading platforms and tankers all the oil it produces, as well as that produced on Beryl Bravo and Ness field. Oil movement in and out of storage is a displacement process in which ballast water displaces oil and vice versa. The ballast water enters and leaves the cells through a piping system, which is configured so that it controls the cells' internal pressure.
In Sept. 1985, during routine ballast-water piping inspection, a leak was discovered at a branch connection between a ballast-water subheader and an individual cell line 61 ft [18.5 m] above the seabed inside the utility shaft (Fig. 1). The rapid increase of the leak gave rise to the fear that pressure control might be lost, with potential consequential collapse of one or more storage cells.
Over a 3-year period. an immediate emergency response program and a long-term system refurbishment were implemented that rendered the system safe for the remainder of the expected platform life, currently projected as the year 2017.
Oil Storage and Export System
The Beryl Alpha oil storage system comprises 16 cells arranged in a honeycomb (Fig. 2). Each cell consists of a 138-ft [42-m] -high cylindrical part with a 61.7-ft [18.8-m] ID and dome-like closures at either end. The cell wall is 2 ft [0.6 m] thick.
By virtue of their construction, the cells have a limited capacity for internal over- and underpressure. The possible pressure regimes to which the cells can be exposed during their lifetime are governed by the sea state and the relative oil/water fill. The ballast-water system controls the cell pressure regime within permissible limits. It comprises a pipework system originating at each cell bottom, a header tank, and ballast-water pumps that control the ballast-water level at 321.5 ft [98 m], 82.3 ft [25.1 m] below the lowest astronomical tide of 403.9 ft [123.1 m].
Oil movement through the storage cells is a simple displacement process. Oil enters and leaves the cells through the cell tops. Ballast water enters and leaves the cells through the cell bottoms. A flow schematic is presented in Fig. 1. The cells must be filled with liquid at all times: otherwise, they would collapse under the hydrostatic head of the surrounding sea. Fig. 3 illustrates the cell pressure regime.
Detection. In Sept. 1985, the normally flooded utility shaft minicell was pumped dry for a routine pipe inspection. Within 4 hours after pumpdown, a leak developed at a welded branch connection between an 18-in. [457-mm] -diameter ballast-water subheader and a 10-in. [254-mm] -diameter cell ballast line. The initial leak rate was about 8 gal/min [1.8 m3/h], increasing to 79 gal/min [18 m3/h] within 24 hours.
At this stage, we decided to isolate the three affected cells by closing the individual ballast-water cell isolation valves and the respective subheader valve. This action, however, had no effect of the leak rate. Because no distinction was possible between the sealing capability of the cell isolation valves and the subheader isolation valve, determining whether the leak was from one or several cells or from the ballast-water riser was impossible.
Because of this uncertainty, and the possible danger of cell drainage and consequent damaging cell underpressure, the decision was made to flood the shaft to 223 ft [68 m]. This response guaranteed cell protection against excessive cell underpressure independent of the pipe-system integrity.
Temporary Repair. To get the leak under control as quickly as possible, a diving spread was mobilized to make a wet repair by installing a specially designed pipe clamp. With a leak at 75 ft [23 m], the water level was lowered to 207 ft [63 m] to permit air diving. Because of space constraints, saturation diving was not seen as a viable alternative.
To ensure the diver's safety, a personnel transfer basket was installed from the shaft entry at 486 to 236 ft [148 to 72 m] and a man riding basket winch from the spider deck at 223 ft [68 m] to the work site at 75 ft [23 m]. Diving control equipment was placed on the spider deck. A decompression chamber for emergency use, a breathing-air compressor, and a hot-water generator for the diving suits were installed next to the shaft entry. "Wet" decompression stops were used to prolong the diver's on-bottom time.
The actual diving operations were complicated by the fact that grating had to be removed from the top of the minicell at 167 ft [51 m] to gain access. Temporary scaffolding, ventilation, and lifting equipment also obstructed the dive path. Attempts to obtain live progress video recordings proved impossible because of poor visibility (1 to 2 ft [0.3 to 0.6 m]).
The total repair clamp installation required 6 days. Although a complete seal was not achieved, the leak rate was reduced to an acceptable 1.1 gal/min [0.25 m3/h]. In addition, the adopted pipeclamp design provided renewed structural strength to the defective joint.
Refurbishment of Minicell and Utility-Shaft Pipework
Engineering. The leak led to serious questions regarding the integrity of the entire ballast-water piping system. A task force was therefore assembled to examine the problem and to formulate a longterm ballast-water-system refurbishment plan. Five major recommendations were made.
1. Before any future dewatering of the minicell, install a crossover pipe between the ballast-water inlet line and the crude rundown line to protect the storage cells against unacceptable underpressure in case of pipe failures.
2. Once the minicell is dewatered, perform an extensive nondestructive testing of all seawater-handling pipework in the shaft to establish its condition.
3. Replace defective ballast-water subheaders and subheader isolation valves with materials suitable for the remaining platform life.
|File Size||519 KB||Number of Pages||6|