Subsea Control System 

All subsea sensors must be powered from, and relay their information to the Subsea Control Module.

The SCM provides a source of power for the sensors, typically 12 or 24 volts for a 4-20mA interface and reads the signal from the sensor, digitises it via an analogue-to-digital converter, then relays it to the surface via the telemetry system, usually in response to a request from the MCS for data from that sensor.

Many SCM designs have the ability to switch off any particular sensor circuit i.e. putting the sensor “out of service” This safeguards the rest of the system in case of a faulty sensor or damaged interface cable, which might otherwise short-circuit the supply voltage. The drive electronics therefore also has to be protected against short circuits, and individual supplies are preferable, so that the other sensors are not affected if one fails.

It is not usual to use this facility to switch off a sensor between readings. Whilst most modern sensors have little or no warm-up time, they may still require 1-2 seconds to stabilise and if the SCM had to wait for this period when reading each sensor, the overall system scan time would be slowed down.

Furthermore, repeatedly switching an electronic component on and off reduces its reliability. A formal reliability analysis does not consider this. An SCM typically reads all subsea sensors as fast as the telemetry system allows, nominally every 2-3 minutes, and so there is little to be gained by de-powering the sensors.

Similarly, the circuitry also should provide for clearing any digital readback value in case of failure of the sensor. If this is not provided locally within the SCM itself, then the MCS will usually change the displayed colour of a sensor reading to indicate fault, as well as displaying any detected under or over range alarm.

Most subsea sensors now use this interface, particularly after the acceptance of subsea conductive connectors, which allow dc connections to be made, and allow 2 or 4-wire operation.

The connection of the cable harness to the sensor itself is a potential leak path and must be tested during manufacture; they are sometimes not mated to each other until after delivery of the sensor itself from the manufacturer, and should then be hyperbarically tested by the Subsea System Supplier.

An alternative is an integral connector in the sensor itself, which can then be tested during manufacture by the Sensor Supplier. In any event, the 'final' joint of the various sub-assemblies must be tested under hyperbaric conditions.

Where there are a large number of sensors to be connected to an SCM, a Supplier will sometimes provide a subsea junction box, which enables a single high-density connector to be 'broken-out' to many individual sensors.

Care should be taken to verify if a proposed sensor is, or is not, isolated from the SCM circuitry; if not, earth loops could arise once the sensor is installed onto a Manifold/Template/Tree and could affect the overall SCM operation as well as the corrosion protection system. Sometimes the manufacturing process of the sensor itself leaves the sensing element in contact with the metal body, so isolation is not always achieved. Other methods encapsulate the sensing element within a sheath and avoid the problem.

Some sensors can be prone to Radio-Frequency Interference in that the sensing element and its wires can pick up induced voltages from nearby radio sources such as walkie-talkies. Obviously this is not a problem once installed subsea but it is preferable to take precautions in the SCM design to avoid unsuspected problems during onshore (or offshore on-deck) commissioning,

Sensors must be calibrated before being installed subsea. Once subsea there is no easy way of detecting any drift in calibration and the manufacturer's specifications are the only practicable guide to this parameter. However, it is usual to test (again) all sensors once they have been installed onto a Tree/Manifold/Template, as there is a high risk that they could have received severe physical shock/vibration or mishandling during installation and possible over-pressurisation during Manifold proof-pressure tests.

Subsea pressure gauges typically operate using the force-balance technique, in which the current required by a coil to resist the movement of the detecting diaphragm gives a measure of the applied pressure. The diaphragm therefore does not actually deform and therefore devices can be built to withstand high pressures. Such devices can achieve an accuracy of +/- 0.15%, although the output current circuitry may reduce this to an overall error-band of around +/- 0.5% of full-scale.

Other devices operate using a similar technique in which the change in capacitance of a sensor element with pressure is detected and typically, accuracy of +/- 0.5% can be obtained.

However, it should be noted that further 'errors' are introduced into the overall system via the digitisation process between the sensor and its eventual display on the MCS. It is unreasonable to expect (and to specify) better than about +/- 2% for a pressure reading.

As pressure transducers are flange-mounted, they cannot be removed subsea if they fail. It is possible to specify a transducer with dual gauges fitted in one housing providing a dual redundant sensor.

Note that retrievable PT/TT sensor (intrusive PT/TT sensor from PRESENS) has been recently developed by TECHNIPFMC has been implemented on EGINA water injection Xmas Tree.

Figure 20.4 - Retrievable PT/TT sensor installed subsea (TECHNIPFMC Water Injection Xmas Tree of EGINA Development)

Higher accuracy can be obtained using quartz-crystal gauges and these are often used for high-accuracy Downhole Pressure Transducers.

Some temperature sensors operate by measuring the output of a thermocouple, which is a simple device whose output is proportional to the difference in temperature between a hot and a cold junction. The hot junction is the one measuring the process and the cold junction is at the head itself.

To compensate for change in ambient temperature (cold junction) most such transmitters incorporate a compensating circuit, which needs to be co-located with the cold junction, but there is still a potential time-lag between the junction and compensation circuit which affects overall accuracy.

A resistance sensor has an element, which depends only upon the absolute temperature and therefore has no cold junction compensation. The disadvantages of a resistance element are its strength, but modern encapsulation techniques can improve this. They are also larger than the thermocouple probe, which can give it a slower response time. Modern sensors are typically made by film-deposition onto a ceramic substrate.

Temperature sensors present somewhat of a dilemma as regards their installation. In theory, the sensing element should be as close to the process fluid as possible. However, if the sensor were simply fitted into the production bore, there would be no other physical barrier between the process and the environment, other than the sensor itself. The usual approach therefore is to install a thermowell, or use a pocket drilled into the Tree Block.

Care must be taken in matching the size of the thermowell and the sensor probe - ideally, the probe should be just in contact with the wall of the process pipe (but not damaged by differences in tolerance or expansion etc.). Thermally conductive grease can improve this contact. Some installations do not consider this and therefore the temperature reading will be subject to the thermal inertia of the system.

The API standard for thermowell lengths is that they should protrude sufficiently into the process flow path to be within the mainstream turbulent flow. However in pipeline applications where through pigging is required, the thermowell must not protrude into the pipeline bore.

A design of sensor is available in which a Pressure and Temperature element are combined into one package. In this design, the temperature sensor is located in a probe, which is designed to be flush-mounted into the process pipe work. This also helps reduce errors due to hydrate formation. For sensors upstream of the Production Master Valve Regulations may however require that there is one or two block valves before the sensor, so, whilst the pressure measurement is unaffected, the temperature measurement may be subject to thermal inertia depending on its location.

The two devices are electrically independent, so the connector has to be a four-pin device. Care should be taken, however, to test that both devices are truly independent of each other as regards variations in power supplies on each device and temperature effects.

Figure 20.5 - Combined Pressure / Temperature Transducer

Subsea chokes usually operate on the principle of stepping actuators adjusting the trim via ratchet and pawl mechanisms. Other types operate via a hydraulic motor. All however, require the position of the trim to be accurately monitored from the surface.

Positional transducers operate by a variety of mechanisms:

Conventionally an LVDT is used which measures the change in inductance of a coil when an armature is moved into it. This has no contacting parts and is very reliable. Proper electrical design can cancel out the effect of variations in supplied voltage. Again, the SCM supplier must supply the necessary energising voltage and detect the output variations and a special circuit is required.

The SCM or MCS must then calculate the choke position from the LVDT signal. A magnetic fluxgate device is a recent addition to the available techniques. In this, a fixed magnet connected to the rotating part of the choke actuator via a simple shaft causes electronic variations in a sensing circuit.

All the electronics are contained within one small can which can be bolted to the outside of the choke and retrieved if necessary. The output is available in digital or standard 4-20mA form, making interfacing with a SCM straightforward; no further computation within the SCM is required. With all of these sensors, care must be taken to match the available accuracy, hysteresis and resolution to the overall system requirements.

If the MCS is simply displaying choke position as a readout, then these are less important. If 'flow rate' is being calculated, whose value will depend on the C_v of the choke (i.e. position) as well as differential pressure across it, then they are of more importance.

If the measured position is being used as part of a choke-position control algorithm, in which the choke is moved until the measured position matches the desired position, then all of these are significant (for example, if hysteresis or resolution is too great, the choke might never actually reach the desired position, even when its direction of motion is reversed to attempt to approach it from the other direction, so it could continually 'hunt' for the correct position.

Systems that do suffer from this are obliged to put in an 'error-band', of +/- 1%, within which the choke can approach the desired band before its movement is stopped. The system then usually has to provide a means to 'single-step' the choke to allow the Operator to fine-tune its position.

A Differential Pressure Sensor is similar to a Pressure Sensor except that pressure is applied on each side of the measuring element; thus, its output is proportional to the difference between the two pressures. The devices are physically similar, as are their electrical connections (4 - 20mA).

Care must be taken to match the rating of the device to the expected range to be measured - sometimes this is difficult to achieve whilst still providing a device that can withstand the full process pressure across it.