System ideal for long flowlines
R. G. HarrisThe Britannia field is a large gas condensate reservoir located in the Central North Sea. The field is operated by Britannia Operator Ltd, which is 50/50 joint venture by Conoco and Chevron. The platform has living quarters, production and drilling facilities, and two interlinked subsea centers, located 7.5 km and 15 km, respectively, from the platform.
Genesis Oil & Gas Consultants
J. Clapham
Britannia Operators Ltd
Two 7.5-km-long bundles containing flowlines link the subsea centers to the platform. A new approach to hydrate prevention has been successfully applied for the first time in the world - offshore - by circulating a heating medium through the annulus of each bundle to keep the flowlines warm.
On the platform, a primary heating medium (glycol/water) system recovers waste heat from three Solar Mars power generators. The primary heating medium is supplied at 140°C to plate heat exchangers which are used to supply up to 15 MW of heat to a secondary heating medium (water treated with corrosion inhibitor, biocide and pH adjusting chemical).
This is circulated at 550 cu meters/hour and at a supply temperature of around 50°C through the annulus of each bundle, and returned via a pipeline in each bundle back to the platform. The flow direction can be reversed for a faster warm-up at start-up. An expansion drum is provided to allow for initial "hydraulic" expansion/contraction of the bundle heating system at startup and shutdown, and for some thermal expansion. A large atmospheric tank is also provided for thermal expansion.
The bundle carrier pipes are each 37.25 in. in diameter, and are insulated with 13 mm of polypropylene foam system to give an overall U value of 11.4 W/sq meters K. The bundles contain the 14-in. production and 8-in. test flowlines operating at up to 160 bar arrival pressure, a 3-in. methanol flowline, and a 12-in. heating medium pipeline. Although these bundles are large and have flow through the annulus, they are conventional in design, construction, and installation. The direction of flow for the secondary heating medium is normally down the annulus, with return via the pipeline. This ensures that the gas arrives on the platform at around the same temperature as the heating medium entering the bundle (50°C). Therefore, production remains outside the hydrate region (above 17°C) downstream of the topsides pressure control valve, which lets down the pressure to 55 bar in the HP and/or test separator.
Methanol is only required for short durations at startup to avoid hydrate formation at the subsea wellhead chokes and jumper spools. It is also injected during long term shutdowns into the unheated spools at the subsea manifold and riser base.
Choosing heated bundles
During front-end studies and development planning the use of heated bundles was compared with other more proven techniques to avoid hydrates such as methanol or MEG (monoethylene glycol) injection or insulation. Heated pipe-in-pipe (coaxial) flowlines were also considered. The heated bundles and heated coaxial flowlines were carried into conceptual design for the following reasons:- Formation water production rates of up to 4,000 b/d were estimated for the subsea flowlines.
- With insulation applied to the flowlines to achieve a U value of 1 W/sq-m K, continuous inhibitor injection would still be required around 10 years into the field life due to a decline in FWHTs and production rates. Inhibitor injection would also be required during periods of low production flowrates during early years of operation.
- The heated system also accounts for the potential for wax formation at the low platform arrival temperatures expected at low production rates. The use of the heated system eliminated the need for pigging to remove wax.
- The heated system gives greater operating flexibility.
- The heated system gives greater reservoir production flexibility.
- The heated system is a neat means of minimizing environmental impact, as waste heat is recovered from turbine exhausts and chemical consumption is minimized.
Heated bundles and the coaxial pipe-in-pipe system were compared in conceptual design. The coaxial system warms up quicker than the bundle, and expansion volume is considerably less. However the bundle was finally selected for detailed design as there was a lower risk of cost escalation and extended schedule compared to the coaxial system which required a laybarge for installation. It was also easier to ensure the annulus was kept clean, thus reducing the risk of biological growth and corrosion, which was a major concern.
Design considerations
Steady state calculations to prove the concept were initially carried out by heat and mass balance. Thermal modeling of steady state and transient operation was completed using a modified version of OLGA in detailed design. The temperature profiles of the gas and heating medium was generated for time intervals along the length of the bundles to enable optimization of insulation thickness on the carrier pipes and heating medium pipeline. This modeling was a complex task but was essential to:- Ensure the test and production flowline temperature profiles were outside the hydrate region throughout their length
- Predict the time to warm up the system
- Predict temperature profiles for stress analysis of the bundles.
Pressure drop in the bundle heating system was predicted by estimating the pressure drop in each element of the system. Due to the large cross-sectional area, the pressure drop in the carrier pipe was predicted to be relatively low even though there are a large number of spacers holding the flowlines in position. Most of the pressure drop occurs in the heating medium flowline, and across a restriction orifice on the topsides which restricts flow in the event of flowline failure (covered later).
The circulation pumps are each sized for 550 cu meters/hour and 43 bar differential pressure. A surge analysis was also completed to predict pressure transients at startup and shutdown. Due to the low velocities through the bundle, the pressure rises were found to be relatively modest in either flow direction.
Prediction of thermal expansion of the secondary heating medium was completed by conventional methods. Due to the large inventory in the bundles, it was necessary to allow for most of the expansion in a 250 cu meter atmospheric tank elevated above the drum. Initial hydraulic expansion of the bundle upon pressurization at startup required the expansion drum to be sized with sufficient hold-up to allow for this relatively rapid event, and conversely at shutdown, the drum needed to accept the hydraulic contraction of the bundle.
Transfer from the drum to the tank is by nitrogen pressure of 5 bar in the drum, and return from the tank to the drum is by pump. It is essential that a supply of heating medium be maintained to the bundle in a shutdown as it cools and contracts to prevent carrier pipe collapse.
Preventing problems
A problem that required considerable development during design was handling gas ingress into the secondary heating medium in the event of flowline failure. The final solution was to install a restriction orifice at the inlet to the expansion drum, that will limit the flow of gas and liquids to within the capacity of the drum with a pressure of up to 140 bar upstream of the restriction orifice.Gas is relieved to the HP flare via bursting discs, and heating medium overflows from the drum into a caisson. If the pressure rises to greater than 140 bar upstream of the restriction orifice due to a large rupture with high pressure in the flowline, a high-integrity overpressure protection system closes valves at the top of the secondary heating medium supply and return risers. This will protect the topsides, but will result in overpressurization of the bundle carrier pipes which are designed for 60 bar.
Another issue that required considerably more effort than originally anticipated was to determine the measures necessary to control corrosion in the heated annulus of the bundle. Originally it was thought that conventional water treatment chemicals would be sufficient to prevent biological growth and corrosion in the bundles. However, a major difference exists with this closed loop system as compared with onshore cooling water systems where water treatment chemicals are commonly applied.
The offshore system is to survive for 25 years without blowdown or draining of the system. Considerable effort was spent by Conoco, Nalco Exxon, and other specialists on the development of a biocide and corrosion inhibitor for this application. Monitoring of the system in operation will be essential to ensure the design life of 25 years is achieved.
Performance
The system has now been commissioned and is in operation. The thermal and hydraulic characteristics of the system are remarkably similar to those anticipated in design studies and no significant problems have been found so far. Great care was taken over system cleanliness at the onshore fabrication sites for the bundles and topsides, as well as during the initial fill offshore.Temporary filters were installed offshore for the initial few weeks of circulation when found that there were considerable quantities of an anti-corrosion being swept from the bundle. Apart from some initial problems with waste heat recovery unit reliability and leaks from the plate heat exchangers the system is performing well.
Heating a bundle using a circulating fluid is now a proven technology offshore. It could be considered for any subsea field with long flowlines to a platform, low flowing wellhead temperatures or high water production rates. It could be considered to either prevent hydrates or wax, or for a heavy oil, it may overcome restartability problems. The flowline could be installed either in a bundle or in a pipe-in-pipe configuration.
Author's Note:
For more information, contact Rob Harris at Genesis Oil & Gas Consultants Ltd, London Tel: 44 0171 430 0040, Fax: 44 1071 831 3934, E mail: [email protected])Copyright 1999 Oil & Gas Journal. All Rights Reserved.