The upstream oil and gas process

The oil and gas process is the process equipment that takes the product from the wellhead manifolds and delivers stabilized marketable products, in the form of crude oil, condensates or gas. Components of the process also exist to test products and clean waste products such as produced water.

An example process for the Statoil Njord floater is shown on the next page. This is a medium-size platform with one production train and a production of 40-45,000 bpd of actual production after the separation of water and gas. The associated gas and water are used for onboard power generation and gas reinjection. There is only one separation and gas compression train. The water is treated and released (it could also have been reinjected). This process is quite representative of hundreds of similar sized installations, and only one more complete gas treatment train for gas export is missing to form a complete gas production facility. Currently, Njord sends the oil via a short pipeline to a nearby storage floater. On gravity base platforms, floating production and storage operations (FPSO) and onshore plants, storage a part of the main installation if the oil is not piped out directly. Photo: Statoil ASA

A large number of connections to chemicals, flares, etc., are also shown. These systems will be described separately.

Nård main process illustration (next page): Statoil

Main Process - NJORD

4.1 Manifolds and gathering
4.1.1 Pipelines and risers
This facility uses subsea production wells. The typical high pressure (HP) wellhead at the bottom right, with its Christmas tree and choke, is located on the sea bed. A production riser (offshore) or gathering line (onshore) brings the well flow into the manifolds. As the reservoir is produced, wells may fall in pressure and become low pressure (LP) wells.

This line may include several check valves. The choke, master and wing valves are relatively slow. Therefore, in the case of production shutdown, the pressure on the first sectioning valve closed will rise to the maximum wellhead pressure before these valves can close. The pipelines and risers are designed with this in mind.

Short pipeline distances are not a problem, but longer distances may cause a multiphase well flow to separate and form severe slugs – plugs of liquid with gas in between – traveling in the pipeline. Severe slugging may upset the separation process and cause overpressure safety shutdowns. Slugging may also occur in the well as described earlier. Slugging can be controlled manually by adjusting the choke, or by automatic slug controls. Additionally, areas of heavy condensate may form in the pipelines. At high pressure, these plugs may freeze at normal sea temperature, e.g., if production is shut down or with long offsets. This can be prevented by injecting ethylene glycol. Glycol injection is not used at Njord.

The Njord floater has topside chokes for subsea wells. The diagram also shows that kill fluid, essentially high specific gravity mud, can be injected into the well before the choke.

4.1.2 Production, test and injection manifolds
 Check valves allow each well to be routed into one or more of several manifold lines. There will be at least one for each process train plus additional manifolds for test and balancing purposes. In this diagram, we show three: test, low pressure and high pressure manifolds. The test manifold allows one or more wells to be routed to the test separator. Since there is only one process train, the HP and LP manifolds allow groups of HP and LP wells to be taken to the first and second stage separators respectively. The chokes are set to reduce the wellhead flow and pressure to the desired HP and LP pressures respectively.

The desired setting for each well and which of the wells produce at HP and LP for various production levels are defined by reservoir specialists to ensure optimum production and recovery rate.

4.2 Separation
As described earlier, the well-stream may consist of crude oil, gas, condensates, water and various contaminants. The purpose of the separators is to split the flow into desirable fractions.

4.2.1 Test separators and well test
Test separators are used to separate the well flow from one or more wells for analysis and detailed flow measurement. In this way, the behavior of each well under different pressure flow conditions can be defined. This normally takes place when the well is taken into production and later at regular intervals (typically 1-2 months), and will measure the total and component flow rates under different production conditions. Undesirable consequences such as slugging or sand can also be determined. The separated components are analyzed in the laboratory to determine hydrocarbon composition of the gas oil and condensate.

Test separators can also be used to produce fuel gas for power generation when the main process is not running. Alternatively, a three phase flow meter can be used to save weight.

4.2.2 Production separators

The main separators shown here are gravity types. On the right, you see the main components around the first stage separator. As mentioned before, the production choke reduces well pressure to the HP manifold and first stage separator to about 3-5 MPa (30-50 times atmospheric pressure). Inlet temperature is often in the range of 100-150 ºC. On the example platform, the well stream is colder due to subsea wells and risers.

The pressure is often reduced in several stages. In this instance, three stages are used to allow the controlled separation of volatile components. The idea is to achieve maximum liquid recovery and stabilized oil and gas, and to separate water. A large pressure reduction in a single separator will cause flash vaporization, leading to instability and safety hazards.

The retention period is typically 5 minutes, allowing gas to bubble out, water to settle at the bottom and oil to be taken out in the middle. In this platform the water cut (percentage water in the well flow) is almost 40%, which is quite high. In the first stage separator, the water content is typically reduced to less than 5%.

At the crude entrance, there is a baffle slug catcher that will reduce the effect of slugs (large gas bubbles or liquid plugs). However, some turbulence is desirable as this will release gas bubbles faster than a laminar flow.

At the end, there are barriers up to a certain level to keep back the separated oil and water. The main control loops are the oil level control loop (EV0101 20 above) controlling the oil flow out of the separator on the right, and the gas pressure loop at the top (FV0105 20, above). The loops are operated by the control system. Another important function is to prevent gas blow-by, which happens when a low oil level causes gas to exit via the oil output, causing high pressure downstream. There are generally many more instruments and control devices mounted on the separator. These will be discussed later.

The liquid outlets from the separator will be equipped with vortex breakers to reduce disturbance on the liquid table inside. This is basically a flange trap to break any vortex formation and ensure that only separated liquid is tapped off and not mixed with oil or water drawn in through these vortices. Similarly, the gas outlets are equipped with demisters, essential filters that remove liquid droplets in the gas.

Emergency valves (EVs) are sectioning valves that separate the process components and blow-down valves, allowing excess hydrocarbons to burn off in the flare. These valves are operated if critical operating conditions are detected or on manual command from a dedicated emergency shutdown system. This may involve partial shutdown and shutdown sequences, since the flare may not be able to handle a full blow-down of all process sections simultaneously.

A 45,000 bpd design production with gas and 40% water cut will give about 10 cubic meters from the wellheads per minute. There also needs to be enough capacity to handle normal slugging from wells and risers. This means the separator has to be about 100 cubic meters, e.g., a cylinder 3m in diameter and 14m in length at the rated operating pressure. This means a very heavy piece of equipment, typically around 50 tons for this size, which limits the practical number of stages. Other types of separators, such as vertical separators or cyclones (centrifugal separation), can be used to save weight, space or improve separation (to be discussed later).

There must also be a certain minimum pressure difference between each stage to allow satisfactory performance in the pressure and level control loops. Chemical additives will also be discussed later.

4.2.3 Second stage separator
The second stage separator is quite similar to the first stage HP separator. In addition to output from the first stage, it also receives production from wells connected to the low pressure manifold. The pressure is now around 1 MPa (10 atmospheres) and temperature below 100ºC. The water content will be reduced to below 2%.

An oil heater can be located between the first and second stage separator to reheat the oil/water/gas mixture. This makes it easier to separate out water when initial water cut is high and temperature is low. The heat exchanger is normally a tube/shell type where oil passes though tubes in a heating medium placed inside an outer shell.

4.2.4 Third stage separator
The final separator is a two-phase separator, also called a flash drum. The pressure is now reduced to atmospheric pressure of around 100 kPa, so that the last heavy gas components can boil out. In some processes where the initial temperature is low, it might be necessary to heat the liquid again (in a heat exchanger) before the flash drum to achieve good separation of the heavy components. There are level and pressure control loops.

As an alternative, when production is mainly gas, and remaining liquid droplets have to be separated out, the two-phase separator can be a knockout drum (K.O. drum).

4.2.5 Coalescer
After the third stage separator, the oil can go to a coalescer for final removal of water. In this unit, water content can be reduced to below 0.1%. The coalescer is completely filled with liquid: water at the bottom and oil on top. Internal electrodes form an electric field to break surface bonds between conductive water and isolating oil in an oil-water emulsion. The coalesce field plates are generally steel, sometimes covered with dielectric material to prevent short-circuits. The critical field strength in oil is in the range of 0.2 to 2 kV/cm. Field intensity and frequency as well as the coalescer grid layout are different for different manufacturers and oil types.

4.2.6 Electrostatic desalter

If the separated oil contains unacceptable amounts of salts, they can be removed in an electrostatic desalter (not used in the Njord example). The salts, which may be sodium, calcium or magnesium chlorides, come from the reservoir water and are also dissolved in the oil. The desalters will be placed after the first or second stage separator depending on GOR and water cut. Photo: Burgess Manning Europe PLC

4.2.7 Water treatment
 On an installation such as this, where the water cut is high, there will be a huge amount of water produced. In our example, a water cut of 40% gives water production of about 4,000 cubic meters per day (4 million liters) that must be cleaned before discharge to sea. Often, this water contains sand particles bound to the oil/water emulsion.

The environmental regulations in most countries are quite strict. For example, in the Northeast Atlantic, the OSPAR convention limits oil in water discharged to sea to 40 mg/liter (ppm).

Produced Water Treatment

It also places limits on other forms of contaminants. This still means that the equivalent of up to one barrel of oil per day in contaminants from the above production is discharged into the sea, but in this form, microscopic oil drops are broken down quickly by natural bacteria.

Various pieces of equipment are used. This illustration shows a typical water treatment system. Water from the separators and coalescers first goes to a sand cyclone, which removes most of the sand. The sand is further washed before it is discharged.

The water then goes to a hydrocyclone, a centrifugal separator that removes oil drops. The hydrocyclone creates a standing vortex where oil collects in the middle and water is forced to the side.

Finally the water is collected in the water de-gassing drum. Dispersed gas slowly rises and pulls remaining oil droplets to the surface by flotation. The surface oil film is drained, and the produced water can be discharged to sea. Recovered oil in the water treatment system is typically recycled to the third
stage separator.

4.3 Gas treatment and compression
The gas train consists of several stages, each taking gas from a suitable pressure level in the production separator's gas outlet, and from the previous stage.

A typical stage is shown on the right. Incoming gas (on the right) is first cooled in a heat exchanger. It then passes through the scrubber to remove liquids and goes into the compressor. The anti-surge loop (thin orange line) and the surge valve (UV0121 23) allow the gas to recirculate. The components are described below.

4.3.1 Heat exchangers

For the compressor to operate efficiently, gas temperature should be low. The lower the temperature, the less energy will be used to compress the gas for the given final pressure and temperature. However, both gas from separators and compressed gas are relatively hot. When gas is compressed, it must remain in thermodynamic balance, which means that the gas pressure times the volume over the temperature (PV/T) must remain constant. (PV = nkT). This ends up as a temperature increase.

Heat exchangers of various forms are used to cool the gas. Plate heat exchangers (upper picture) consist of a number of plates where the gas and cooling medium pass between alternating plates in opposing directions. Tube and shell exchangers (next picture) place tubes inside a shell filled with cooling fluid. The cooling fluid is often pure water with corrosion inhibitors.

When designing the process, it is important to plan the thermal energy balance. Heat should be conserved, e.g., by using the cooling fluid from the gas train to reheat oil in the oil train. Excess heat is dispersed, e.g., by seawater cooling. However, hot seawater is extremely corrosive, so materials with high resistance to corrosion, such as titanium must be used. Photo: SEC Shell and Tube Heat Exchanges

4.3.2 Scrubbers and reboilers
The separated gas may contain mist and other liquid droplets. Drops of water and hydrocarbons also form when the gas is cooled in the heat exchanger, and must be removed before it reaches the compressor. If liquid droplets enter the compressor, they will erode the fast rotating blades. A scrubber is designed to remove small fractions of liquid from the gas. There are various types of gas-drying equipment available, but the most common suction (compressor) scrubber is based on dehydration by absorption in triethylene glycol (TEG). The scrubber consists of many levels of glycol layers.

A large number of gas traps (enlarged detail) force the gas to bubble up through each glycol layer as it flows from the bottom to the top of each section.

Processed glycol is pumped in at the top from the holding tank. It flows from level to level against the gas flow as it spills over the edge of each trap.

During this process, it absorbs liquids from the gas and comes out as rich glycol at the bottom. The holding tank also functions as a heat exchanger for liquid, to and from the reboilers.

The glycol is recycled by removing the absorbed liquid. This is done in the reboiler, which is filled with rich glycol and heated to boil out the liquids at temperature of about 130-180 °C (260-350 °F) for a number of hours. Usually there is a distillation column on the gas vent to further improve separation of glycol and other hydrocarbons. For higher capacity, there are often two reboilers which alternate between heating rich glycol and draining recycled processed glycol. On a standalone unit, the heat is supplied from a burner that uses the recovered vaporized hydrocarbons. In other designs, heating will be a combination of hot cooling substances from other parts of the process and electric heaters, and recycling the hydrocarbon liquids to the third stage separator.

Glycol Regeneration

4.3.3 Compressors, anti-surge and performance
 Compressors are used in many parts of the oil and gas process, from upstream production to gas plants, pipelines, LNG and petrochemical plants. The overview given here will therefore be referenced from other sections.

Several types of compressors are used for gas compression, each with different characteristics such as operating power, speed, pressure and volume:

• Reciprocating compressors, which use a piston and cylinder design with 2-2 cylinders are built up to about 30 MW power, around 500-1,800 rpm (lower for higher power) with pressure up to 5MPa (500 bars). Used for lower capacity gas compression and high reservoir pressure gas injection. Photo: Ariel corp.

• Screw compressors are manufactured up to several MW, synchronous speed (3,000/3,600 rpm) and pressure up to about 2.5 MPa (25 bar). Two counter-rotating screws with matching profiles provide positive displacement and a wide operating range. Typical use is natural gas gathering. Photo: Mycom/Mayekawa mfg.

• Axial blade and fin type compressors with up to 15 wheels provide high volumes at a relatively low pressure differential (discharge pressure 3-5 times inlet pressure), speeds of 5,000-8,000 rpm, and inlet flows up to 200,000 m3/hour. Applications include air compressors and cooling compression in LNG plants. Axial rotor photo: Dresser Rand

• Larger oil and gas installations use centrifugal compressors with 3-10 radial wheels, 6,000–20,000 rpm (highest for small size), up to 80 MW load at discharge pressure of up to 50 bars and inlet volumes of up to 500,000 m3/hour. Pressure differential up to 10. Photo: Dresser Rand

Most compressors will not cover the full pressure range efficiently. The lowest pressure is atmospheric, for gas to pipeline, some 3 to 5 MPa (30-50 bar) pressure is used, while reservoir reinjection of gas will typically require 20 MPa (200 bar) and upwards, since there is no liquid in the tubing and the full reservoir pressure must be overcome. Therefore, compression is divided into several stages to improve maintenance and availability.

Also due to single unit power limitations, compression is often divided in several parallel trains. This is not the case in this example, since gas is not exported and reinjection can be interrupted during maintenance periods.
Compressors are driven by gas turbines or electrical motors (for lower power also reciprocating engines, steam turbines are sometimes used if thermal energy is available). Often, several stages in the same train are driven by the same motor or turbine.The main operating parameters for a compressor are the flow and pressure differentials. The product defines the total loading, so there is a ceiling set by the maximum design power. Furthermore, there is a maximum differential pressure (Max Pd) and choke flow (Max Q), the maximum flow that can be achieved. At lower flow, there is a minimum pressure differential and flow before the compressor will "surge" if there is not enough gas to operate.

If variations in flow are expected or differences between common shaft compressors occur, the situation will be handled with recirculation. A high flow, high pressure differential surge control valve will open to let gas from the discharge side back into the suction side. Since this gas is heated, it will also pass through the heat exchanger and scrubber so as not to become overheated by circulation.

Compressor State Diagram

The operating characteristics are defined by the manufacturer. In the diagram above, the blue lines mark constant speed lines. The maximum operating limits are set by the orange line as described above. The surge domain is the area to the left of the red surge curve.

The objective of compressor performance control is to keep the operating point close to the optimal set point without violating the constraints by means of control outputs, such as the speed setting. However, gas turbine speed control response is relatively slow and even electric motors are not fast enough, since surge response must be in the 100 ms range. Anti-surge control will protect the compressor from going into surge by operating the surge control valve. The basic strategy is to use distance between operating point and surge line to control the valve with a slower response time, starting at the surge control line. Crossing the surge trip line will cause a fast response opening of the surge valve to protect the compressor.

Operation with recirculation wastes energy (which could result in unnecessary emissions) and produces wear and tear, particularly on the surge valve. Each vendor supplies several variants of compressor control and anti-surge control to optimize performance, based on various corrective and predictive algorithms. Some strategies include:

• Set point adjustment: If rapid variations in load cause surge valve action, the set point will be moved to increase the surge margin.

• Equal margin: The set point is adjusted to allow equal margin to surge between several compressors. 

• Model based control: Outside the compressor itself, the main parameter for the surge margin is the total volume from the surge valve to the compressor suction inlet, and the response time for the surge valve flow. A model predictive controller could predict surge conditions and react faster to real situations while preventing unnecessary recirculation.

Since compressors require maintenance and are potentially expensive to replace, several other systems are normally included:

Load management: To balance loading among several compressors in a train and between trains, the compressor control system often includes algorithms for load sharing, load shedding and loading. Compressors are normally purged with inert gas, such as nitrogen during longer shutdowns, e.g., for maintenance. Therefore, startup and shutdown sequences will normally include procedures to introduce and remove the purge gas.

Vibration: Vibration is a good indicator of problems in compressors, and accelerometers are mounted on various parts of the equipment to be logged and analyzed by a vibration monitoring system.

Speed governor: If the compressor is turbine driven, a dedicated speed governor handles the fuel valves and other controls on the turbine to maintain efficiency and control rotational speed. For electrical motors this function is handled by a variable speed drive.

The final function around the compressor itself is lube and seal oil handling. Most compressors have wet seals, which are traps around shafts where oil at high pressure prevents gas from leaking out to atmosphere or other parts of the equipment. Oil is used for lubrication of the high speed bearings. This oil gradually absorbs gas under pressure and may become contaminated. It needs to be filtered and degassed. This happens in smaller reboilers, in much the same way as for the glycol reboilers described earlier.


4.4 Oil and gas storage, metering and export
The final stage before the oil and gas leaves the platform consists of storage, pumps and pipeline terminal equipment.


4.4.1 Fiscal metering
Partners, authorities and customers all calculate invoices, taxes and payments based on the actual product shipped out. Often, custody transfer also takes place at this point, which means transfer of responsibility or title from the producer to a customer, shuttle tanker operator or pipeline operator. Although some small installations are still operated with a dipstick and manual records, larger installations have analysis and metering equipment.

To make sure readings are accurate, a fixed or movable prover loop for calibration is also installed. The illustration shows a full liquid hydrocarbon (oil and condensate) metering system. The analyzer instruments on the left provide product data such as density, viscosity and water content. Pressure and temperature compensation is also included.

Figure 9. Metering system

For liquids, turbine meters with dual pulse outputs are most common. Alternatives are positive displacement meters (pass a fixed volume per rotation or stroke) and coriolis mass flow meters. These instruments cannot cover the full range with sufficient accuracy. Therefore, the metering is split into several runs, and the number of runs depends on the flow. Each run employs one meter and several instruments to provide temperature and pressure correction. Open/close valves allow runs to be selected and control valves can balance the flow between runs. The instruments and actuators are monitored and controlled by a flow computer. If the interface is not digital, dual pulse trains are used to allow direction sensing and fault finding.

To obtain the required accuracy, the meters are calibrated. The most common method is a prover loop. A prover ball moves though the loop, and a calibrated volume is provided between the two detectors (Z). When a meter is to be calibrated, the four-way valve opens to allow oil to flow behind the ball. The number of pulses from it passes one detector Z to the other and is counted. After one loop, the four-way valve turns to reverse flow direction and the ball moves back, providing the same volume in reverse, again counting the pulses. From the known reference volume, number of pulses, pressure and temperature the flow computer can calculate the meter factor and provide accurate flow measurements using formulas from industry standard organizations such as API MPMS and ISO 5024. The accuracy is typically ± 0.3% of standard volume.

Gas metering is similar, but instead, analyzers will measure hydrocarbon content and energy value (MJ/scm or BTU, Kcal/scf) as well as pressure and temperature. The meters are normally orifice meters or ultrasonic meters. Orifice plates with a diameter less than the pipe are mounted in cassettes. The pressure differential over the orifice plate as well as pressure and temperature, is used in standard formulas (such as AGA 3 and ISO 5024/5167) to calculate normalized flow. Different ranges are accommodated with different size restrictions.

Orifice plates are sensitive to a buildup of residue and affect the edges of the hole. Larger new installations therefore prefer ultrasonic gas meters that work by sending multiple ultrasonic beams across the path and measure the Doppler effect.

Gas metering is less accurate than liquid, typically ±1.0% of mass. There is usually no prover loop, the instruments and orifice plates are calibrated in separate equipment instead.

LNG is often metered with mass flow meters that can operate at the required low temperature. A three run LNG metering skid is shown above.

At various points in the movement of oil and gas, similar measurements are taken, usually in a more simplified way. Examples of different gas types are flare gas, fuel gas and injected gas, where required accuracy is 2-5% percent.


4.4.2 Storage

On most production sites, oil and gas are piped directly to a refinery or tanker terminal. Gas is difficult to store locally, but occasionally underground mines, caverns or salt deposits can be used to store gas.


On platforms without a pipeline, oil is stored in onboard storage tanks to be transported by shuttle tanker. The oil is stored in storage cells around the shafts on concrete platforms, and in tanks on floating units. On some floaters, a separate storage tanker is used. Ballast handling is very important in both cases to balance the buoyancy when oil volume varies. For onshore, fixed roof tanks are used for crude, floating roof for condensate. Rock caves are also used for storage

Special tank gauging systems such as level radars, pressure or float are used to measure the level in storage tanks, cells and caves. The level measurement is converted to volume via tank strapping tables (depending on tank geometry) and compensated for temperature to provide standard volume. Float gauges can also calculate density, and so mass can be established.

A tank farm consists of 10-100 tanks of varying volume for a typical total capacity in the area of 1-50 million barrels. Storage or shuttle tankers normally store up to two weeks of production, one week for normal cycle and one extra week for delays, e.g., bad weather. This can amount to several million barrels.

Accurate records of volumes and history are kept to document what is received and dispatched. For installations that serve multiple production sites, different qualities and product blending must also be handled. Another planning task is forecasting for future received and delivered products. This is for stock control and warehousing requirements. A tank farm management system keeps track of all stock movements and logs all transport operations that take place.


4.4.3 Marine loading

Loading systems consist of one or more loading arms/jetties, pumps, valves and a metering system.

Tanker loading systems are complex, both because of the volume involved, and because several loading arms will normally interact with the tanker's ballast system to control the loading operation. The tanks must be filled in a certain sequence; otherwise the tanker's structure might be damaged due to uneven stresses. It is the responsibility of the tanker's ballast system to signal data to the loading system and to operate the different valves and monitor the tanks on board the ship. Photo: Statoil