Control Laws

Each manufacturer has different control strategies, and each engine type has detailed differences in its control laws. However, aerospace applications place certain common requirements on control:

This is necessarily only a small subset of the engine control requirements and consequences on the system (see Figures 9–1 to 9–6).

Engine Electronic Controller

At the center of the FADEC system is the engine electronic controller (EEC). The stringent requirements for safety and availability of an aero engine cannot be met with simplex electronics. For this reason, FADEC designs generally include two channels of electronics, sensors, wiring harnesses, and duplicated electrical parts of actuators. So that the system is fully operational following a single electrical or electronic failure. The two channels within the EEC include features that enable them to exchange data, which are used to detect failures in the system and to allow continued operation. However, the channels must be designed so that a fault in one channel cannot propagate to the other.

In some cases, the two electronic channels are housed in separate enclosures, but more usually, they are contained in a single unit. The EEC may be installed in the airframe, particularly in military aircraft or in civil fuselage-mounted engine installations. For wing-mounted engine applications, typical of large civil turbofans, the EEC is mounted on the engine. This installation places particularly harsh environmental requirements on the electronics, while further emphasizing the need for low weight and volume—a need reflected in the components used, the construction techniques, and mounting arrangements. One environmental threat, particular to electronic systems, is electro-magnetic radiation from, for example, lightning (both on the ground and in the air) and airport radar. The substantial connector housings used are in part designed to help alleviate these threats.

The EEC reads data from the sensors, other information from the aircraft avionic systems, and the pilot's inputs to calculate the new required position of the actuators, and uses its drive circuits to move them, often by means of secondary servos in the actuators. It also transmits data relating to the engine condition back to the aircraft, along industry standard serial data busses. The aircraft manufacturer is responsible for deciding which data are displayed to the pilot, subject to certification rules and the engine manufacturer's installation manual.

The EEC gathers information on any failures it has diagnosed within the electronics, the remainder of the FA DEC system, or in some cases in the gas turbine itself. This information is transmitted to the aircraft systems, but if the system considers itself to be in a safe configuration and no action is required in flight, the information is often not displayed in the cockpit—it is available to the pilot if required, but is intended for use by maintenance personnel on the ground.

This fault information may also be stored within the EEC itself for retrieval by the ground crew and may include more detail than is transmitted to the aircraft. Should the EEC be removed as a result of a suspected failure, these data are also used to assist in the diagnosis of the fault at the repair base.

Actuation

Actuators can use various power sources, in addition to high-pressure fuel:

Fuel Pumps

The pumping system has to be able to deliver sufficient fuel flow to the engine under all conditions and at pressures high enough to overcome the gas pressure in the fuel spray nozzles generated by the engine compression system. Flow is also required to power the servo systems. The pumps are driven from the engine accessory gearbox.

The input pressure to the pumping system depends on the aircraft fuel system; in some cases, the engine pumps are required to continue to feed the engine with fuel if the aircraft fuel pumps fail—the engine pumps would then have to deliver fuel from the aircraft tanks. Generally, it is not possible to provide the suction and fuel pressure required with a single pump, and so two pumps are often used in series: a low-pressure centrifugal pump provides suction and delivers fuel to a high-pressure pump (most commonly a positive displacement gear pump), which in turn delivers fuel to the FMU.

In all cases, the pumping system must deliver enough fuel for the engine. The nature of the positive displacement pump sets a unique relationship between speed of rotation and flow delivered, resulting in more flow from the HP pump than is required by the engine. To accommodate this, the FMU returns unwanted fuel to the inlet of the HP pump, which then recirculates it. A consequence of this can be excessive heating of the fuel as it is repeatedly compressed and decompressed. A cooler is often required to dissipate this heat.

Bleed Valves

Many engines include bleed valves in the compression system, which allow intercompressor flows to be matched at low speeds. The position (open or closed) of these valves is signaled by the EEC, and fuel, electrical, or pneumatic servo systems are used to actuate them.

Electrical Power Supply

The reliability and availability of the electrical and electronic parts of the system are dependent on the quality and reliability of the electrical power supply. Each channel of the FADEC system requires independent power and, for aircraft, this is typically provided by a dedicated, gearbox-mounted generator. Aircraft power is also supplied for starting, and to provide a backup should the dedicated generator fail. In some military aircraft, only aircraft power is used, mainly because of the dependability of the aircraft supplies.

Software

The software embedded in the EEC defines the system behavior. The performance of this software is therefore vital to the operation of the engine. The software is generated from the requirements using disciplined processes and extensive testing. These processes are defined in industry standard documents and guidelines. Software developed to these standards is expensive to generate and can take a considerable time, particularly due to the effort required in testing and qualification. Software tools and techniques are becoming available to reduce this effort but these are far from mature.

Indication Systems

In modern systems, data from the control system and other sources is displayed on one or more display units mounted in the instrument panel; multi-function screens, which display basic engine data such as rotor speeds, turbine temperature, and power, have replaced the multitude of dials and individual instruments found in older aircraft. The multi-function screens are programmed to reconfigure themselves to display other data in response to abnormal circumstances, or as required by the operator. The information is displayed on the screen in the form of virtual dials with digital readouts and warnings; cautions and advisory messages are shown as text. A mimic diagram representing the physical layout of the equipment may be provided to assist in locating a problem. The displays are color-coded and, when necessary, linked to audible warning systems so that the operator is aware of the severity of any problem.

In military aircraft, these data may be displayed using a “head up display” (HUD). The HUD system projects information and instrument images onto the screen in front of the pilots.

Using this technology means that pilots do not have to divert their attention from the view around them. On some applications, information can also be shown on the visor as part of the headgear worn by the pilot.

Engine Health Monitoring

It is in the interests of all customers to minimize the cost of operation of the gas turbine and its associated equipment.

The costs of operation include fuel, scheduled maintenance, and unforeseen events that result in the engine not being available when required. Monitoring systems can help to reduce all of these costs. Scheduling of major engine maintenance (for example, to restore performance after many hours of operation) is a complex economic decision for which monitoring systems can provide important supporting data.

Although it is not strictly part of the control system, the EHM electronics are often housed within the control system enclosure, and the two systems are to some extent integrated.

It is important to note, however, that the safety requirements of the two systems are different and the design of each, and their integration, must reflect this. A function in the monitoring system cannot be adopted for use in the control system without considering the reliability of its implementation.

Data from the EHM systems are not generally available to the flight crew. Large amounts of data are stored although data reduction and analysis algorithms are used to make storage requirements more reasonable. Aircraft systems are used to transmit the data to a ground station, which in turn will forward the data to a center where further analysis can be carried out in order to inform maintenance logistics (Figure 9–7).

Sensors

Whatever the particular application, a series of parameters needs to be measured at various locations around the engine system in order to control the engine and provide useful indication of performance to the operator. Typically, these are temperature, pressure, and speed measurements. The transducers used to take these measurements are chosen on accuracy, response time, and durability requirements.

Temperature Sensors

Thermocouples

Thermocouples are used to measure high temperatures, typically at HP compressor exit, and in and around the turbines. The temperature of the gas in the turbine is measured at several radial and circumferential positions in order to even out any local temperature variations due to turbine entry temperature traverse effects. Thermocouples have the advantage of being very reliable, small, and cheap; they also have a relatively quick response time over a large temperature range, and generate their own output—and so do not require an external power supply. However, thermocouples are easily damaged, and can lose accuracy through oxidation (Figure 9–8).

Pressure Sensors

Pressure sensors broadly divide into those required to provide high accuracy, and those that focus on transient response. Accurate measurement is required when pressure ratio is used to measure engine thrust. Transducers based on a variety of technologies are used for this purpose, but they generally need electronics to support their operation or to provide calibration information, and therefore the assembly is usually housed within the EEC, which can involve quite long pipe runs. If high bandwidth is required, simpler transducers, often based on strain gauge technology, are used and may be mounted close to the engine to avoid pipe delays.

Rotor Speed Sensors

Types of speed sensor include tachogenerators and magnetic variable reluctance (VR) probes. A tachogenerator is a shaft-driven, electrical generator with a variable frequency output, which is related to speed. These devices are very rugged, but produce a relatively low output signal. If a VR speed probe is used, it is positioned on the compressor casing in line with a small disc, which has accurately machined notches on its circumference and is mounted concentrically on the shaft. Rotation of the shaft results in a current being induced in the probe with a frequency content proportional to engine speed (Figures 9–9 and 9–10).

Position Measurement

Position measurement is used to confirm that actuators are operating correctly and to assist in closed loop control. There are three main types of device used: the LVDT (linear variable differential transformer), RVDT (rotational variable differential transformer), and the resolver.

An LVDT consists of three adjacent coils of wire wound around a hollow form through which a core of permeable material (such as steel) can slide freely. The middle winding is known as the primary coil, and is excited by a relatively high frequency AC voltage. This sets up a magnetic flux, which is then coupled through the core to the other two, secondary windings, inducing a voltage in them. When the moving core is centered between the secondary coils, the voltage induced in them is equal and opposite. If the core is displaced, then an imbalance is set up, creating a voltage that can be read and calibrated to give a position.

RVDTs and resolvers are based on similar principles, but are used to measure rotational angles.

Vibration

Many engines are fitted with sensors that continuously monitor the vibration level of the engine. Indication of excessive vibration is shown on the control display unit using signals from engine-mounted transducers. There are three main types of vibration sensor:

Safety and Availability

Safety is the most important design consideration in any gas turbine or installation; another high priority is availability—the loss of power from an engine, although not necessarily a safety hazard, can cause severe operational disruption. The duplication of the electrical elements of the system is evidence of this concern. Rigorous analyses and testing are necessary to ensure that faults in the system are correctly accommodated to allow for continued engine operation.

Just as it is safe to complete a flight during which a failure has occurred in the duplicated part of the system, it can also be shown by analysis that the aircraft can continue to operate for subsequent flights for a defined period before a fault is repaired. A “time limited dispatch” analysis is carried out to establish which faults can be treated in this way and for how long. This is of considerable benefit to the aircraft operator, who can continue to operate the aircraft normally and repair the fault at a convenient time, for example, when the aircraft next returns to the operator's main base.

Other safety features may also be required, implemented either in the software or in dedicated hardware to address the effects of adverse operating conditions, or of particular engine or control system failures, which could represent a threat to the aircraft if not accommodated.

Engine Thrust

The thrust of an engine is shown on a thrust-meter, which will be one of two basic types; the first measures turbine discharge or jet pipe pressure, and the second, known as an engine pressure ratio (E.P.R.) gauge, measures the ratio of two or three parameters. When E.P.R. is measured, the ratio is usually that of jet pipe pressure to compressor inlet pressure. However, on a fan engine the ratio may be that of integrated turbine discharge and fan outlet pressures to compressor inlet pressure.

In each instance, an indication of thrust output is given, although when only the turbine discharge pressure is measured, correction is necessary for variation of inlet pressure; however, both types may require correction for variation of ambient air temperature. To compensate for ambient atmospheric conditions, it is possible to set a correction figure to a sub-scale on the gauge; thus the minimum thrust output can be checked under all operating conditions.

Suitably positioned pitot tubes sense the pressure or pressures appropriate to the type of indication being taken from the engine. The pitot tubes are either directly connected to the indicator or to a pressure transmitter that is electrically connected to the indicator.

An indicator that shows only the turbine discharge pressure is basically a gauge, the dial of which may be marked in pounds per square inch (p.s.i.), inches of mercury (in. Hg.), or a percentage of the maximum thrust.

E.P.R. can be indicated by either electromechanical or electronic transmitters. In both cases the inputs to the transmitter are engine inlet pressure (P₁) and an integrated pressure (P_INT) comprised of fan outlet and turbine exhaust pressures. In some cases either fan outlet pressure or turbine exhaust pressure is used alone in place of P_INT.

The electro-mechanical system indicates a change in pressure by using transducer capsules (Figure 9–13) to deflect the center shaft of the pressure transducer causing the yoke to pivot about the axis A.A. This movement is sensed by the linear variable differential transformer (L.V.D.T.) and converted to an A.C. electrical signal, which is amplified and applied to the control winding of the servo motor.

The servo motor, through the gears, alters the potentiometer output voltage signal to the E.P.R. indicator and simultaneously drives the gimbal in the same direction as the initial yoke movement until the L.V.D.T. signal to the motor is canceled and the system stabilizes at the new setting.

The electronic E.P.R. system utilizes two vibrating cylinder pressure transducers that sense the engine air pressures and vibrate at frequencies relative to these pressures. From these vibration frequencies electrical signals of E.P.R. are computed and are supplied to the E.P.R. gauge and electronic engine control system.

Engine Torque

Engine torque is used to indicate the power that is developed by a turbo-propeller engine, and the indicator is known as a torquemeter. The engine torque or turning moment is proportional to the horsepower and is transmitted through the propeller reduction gear.

A torquemeter system is shown in Figure 9–14. In this system, the axial thrust produced by the helical gears is opposed by oil pressure acting on a number of pistons; the pressure required to resist the axial thrust is transmitted to the indicator.

In addition to providing an indication of engine power, the torquemeter system may also be used to automatically operate the propeller feathering system if the torquemeter oil pressure falls due to a power failure. It is also used, on some installations, to assist in the automatic operation of the water injection system to restore or boost the takeoff power at high ambient temperatures or at high altitude airports.

Engine Speed

All engines have their rotational speed (rpm) indicated. On a twin or triple-spool engine, the high-pressure assembly speed is always indicated; in most instances, additional indicators show the speed of the low pressure and intermediate pressure assemblies.

Engine speed indication is electrically transmitted from a small generator, driven by the engine, to an indicator that shows the actual revolutions per minute (rpm), or a percentage of the maximum engine speed (Figure 9–15). The engine speed is often used to assess engine thrust, but it does not give an absolute indication of the thrust being produced because inlet temperature and pressure conditions affect the thrust at a given engine speed.

The engine speed generator supplies a three-phase alternating current, the frequency of which is dependent upon engine speed. The generator output frequency controls the speed of a synchronous motor in the indicator, and rotation of a magnet assembly housed in a drum or drag cup induces movement of the drum and consequent movement of the indicator pointer.

Where there is no provision for driving a generator, a variable-reluctance speed probe, in conjunction with a phonic wheel, may be used to induce an electric current that is amplified and then transmitted to an indicator (Figure 9–16).

This method can be used to provide an indication of rpm without the need for a separately driven generator, with its associated drives, thus reducing the number of components and moving parts in the engine.

The speed probe is positioned on the compressor casing in line with the phonic wheel, which is a machined part of the compressor shaft. The teeth on the periphery of the wheel pass the probe once each revolution and induce an electric current by varying the magnetic flux across a coil in the probe. The magnitude of the current is governed by the rate of change of the magnetic flux and is thus directly related to engine speed.

In addition to providing an indication of rotor speed, the current induced at the speed probe can be used to illuminate a warning lamp on the instrument panel to indicate to the pilot that a rotor assembly is turning. This is particularly important at engine start, because it informs the pilot when to open the fuel cock to allow fuel to the engine. The lamp is connected into the starting circuit and is only illuminated during the starting cycle.

Turbine Gas Temperature

The temperature of the exhaust gases is always indicated to ensure that the temperature of the turbine assembly can be checked at any specific operating condition. In addition, an automatic gas temperature control system is usually provided, to ensure that the maximum gas temperature is not exceeded.

Turbine gas temperature (T.G.T.) sometimes referred to as exhaust gas temperature (E.G.T.) or jet pipe temperature (J.P.T.), is a critical variable of engine operation and it is essential to provide an indication of this temperature. Ideally, turbine entry temperature (T.E.T.) should be measured; however, because of the high temperatures involved this is not practical, but, as the temperature drop across the turbine varies in a known manner, the temperature at the outlet from the turbine is usually measured by suitably positioned thermocouples. The temperature may alternatively be measured at an intermediate stage of the turbine assembly, as shown in Figure 9–17.

The thermocouple probes used to transmit the temperature signal to the indicator consist of two wires of dissimilar metals that are joined together inside a metal guard tube. Transfer holes in the tube allow the exhaust gas to flow across the junction. The materials from which the thermocouple's wires are made are usually nickel-chromium and nickel-aluminum alloys.

The probes are positioned in the gas stream so as to obtain a good average temperature reading and are normally connected to form a parallel circuit. An indicator, which is basically a millivoltmeter calibrated to read in degrees centigrade, is connected into the circuit (Figure 9–18).

The junction of the two wires at the thermocouple probe is known as the “hot” or “measuring” junction and that at the indicator as the “cold” or “reference” junction. If the cold junction is at a constant temperature and the hot junction is sensing the exhaust gas temperature, an electromotive force (E.M.F.), proportional to the temperature difference of the two junctions, is created in the circuit and this causes the indicator pointer to move. To prevent variations of cold junction temperature affecting the indicated temperature, an automatic temperature compensating device is incorporated in the indicator or in the circuit

The thermocouple probes may be of single, double, or triple element construction. Where multiple probes are used they are of differing lengths in order to obtain a temperature reading from different points in the gas stream to provide a better average reading than can be obtained from a single probe (Figure 9–17).

The output to the temperature control system can also be used to provide a signal, in the form of short pulses, which, when coupled to an indicator, will digitally record the life of the engine. During engine operation in the higher temperature ranges, the pulse frequency increases progressively causing the cyclic-type indicator to record at a higher rate, thus relating engine or unit life directly to operating temperatures.

Thermocouples may also be positioned to transmit a signal of air intake temperature into the exhaust gas temperature indicating and control systems, thus giving a reading of gas temperature that is compensated for variations of intake temperature. A typical double-element thermocouple system with air intake probes is shown in Figure 9–18.

Oil Temperature and Pressure

It is essential for correct and safe operation of the engine that accurate indication is obtained of both the temperature and pressure of the oil. Temperature and pressure transmitters and indicators are illustrated in Figure 9–19.

Oil temperature is sensed by a temperature-sensitive element fitted in the oil system. A change in temperature causes a change in the resistance value and, consequently, a corresponding change in the current flow at the indicator. The indicator pointer is deflected by an amount equivalent to the temperature change and this is recorded on the gauge in degrees centigrade.

Oil pressure is electrically transmitted to an indicator on the instrument panel. Some installations use a flag-type indicator, which indicates if the pressure is high, normal or low; others use a dial-type gauge calibrated in pounds per square inch (psi).

Electrical operation of each type is similar; oil pressure, acting on the transmitter, causes a change in the electric current supplied to the indicator. The amount of change is proportional to the pressure applied at the transmitter.

The transmitter may be of either the direct or the differential pressure type. The latter senses the pressure difference between engine feed and return oil pressures, the return oil being pressurized by cooling and sealing air from the bearings.

In addition to a pressure gauge operated by a transmitter, an oil low-pressure warning switch may be provided to indicate that a minimum pressure is available for continued safe running of the engine. The switch is connected to a warning lamp in the flight compartment and the lamp illuminates if the pressure falls below an acceptable minimum.

Fuel Temperature and Pressure

The temperature and pressure of the low pressure fuel supply are electrically transmitted to their respective indicators and these show if the low pressure system is providing an adequate supply of fuel without cavitation and at a temperature to suit the operating conditions. The fuel temperature and pressure indicators are similar to those for oil temperature and pressure indication.

On some engines, a fuel differential pressure switch, fitted to the low-pressure fuel filter, senses the pressure difference across the filter element. The switch is connected to a warning lamp that provides indication of partial filter blockage, with the possibility of fuel starvation.

Fuel Flow

Although the amount of fuel consumed during a given flight may vary slightly between engines of the same type, fuel flow does provide a useful indication of the satisfactory operation of the engine and of the amount of fuel being consumed during the flight. A typical system consists of a fuel flow transmitter, which is fitted into the low-pressure fuel system, and an indicator, which shows the rate of fuel flow and the total fuel used in gallons, pounds or kilograms per hour (Figure 9–20). The transmitter measures the fuel flow electrically and an associated electronic unit gives a signal to the indicator proportional to the fuel flow.

Vibration

A turbo-jet engine has an extremely low vibration level and a change of vibration, due to an impending or partial failure, may pass without being noticed. Many engines are therefore fitted with vibration indicators that continually monitor the vibration level of the engine. The indicator is usually a milliammeter that receives signals through an amplifier from engine-mounted transmitters (Figure 9–21).

A vibration transmitter is mounted on the engine casing and electrically connected to the amplifier and indicator. The vibration-sensing element is usually an electro-magnetic transducer that converts the rate of vibration into electrical signals and these cause the indicator pointer to move proportional to the vibration level. A warning lamp on the instrument panel is incorporated in the system to warn the pilot if an unacceptable level of vibration is approached, enabling the engine to be shut down and so reduce the risk of damage.

The vibration level recorded on the gauge is the sum total of vibration felt at the pick-up. A more accurate method differentiates between the frequency ranges of each rotating assembly and so enables the source of vibration to be isolated. This is particularly important on multi-spool engines.

A crystal-type vibration transmitter, giving a more reliable indication of vibration, has been developed for use on multi-spool engines. A system of filters in the electrical circuit to the gauge makes it possible to compare the vibration obtained against a known frequency range and so locate the vibration source. A multiple-selector switch enables the pilot to select a specific area to obtain a reading of the level of vibration.

Warning Systems

Warning systems are provided to give an indication of a possible failure or the existence of a dangerous condition, so that action can be taken to safeguard the engine or aircraft. Although the various systems of an aircraft engine are designed wherever possible to “fail safe,” additional safety devices are sometimes fitted. Automatic propeller feathering should a power loss occur, and automatic closing of the high-pressure fuel shut-off cock should a turbine shaft failure occur, are but two examples. On some engine types, the fuel system is fitted with a control to enable the engine to be operated by manual throttling should a main fuel system failure occur.

In addition to a fire warning system, a number of other audible or visual warning systems can be fitted to a gas turbine engine. These may be for low oil or fuel pressure, excessive vibration or overheating. Indication of these may be by warning light, bell, or horn. A flashing light is used to attract the pilot’s attention to a central warning panel (C.W.P.) where the actual fault is indicated.

Other instruments and lights warn the pilot of the selected position of the thrust reverser, the fan reverser, or the afterburner variable nozzle, when applicable. Gauges also inform the pilot of such things as hydraulic pressure and flow and generator output, which are vital to the correct operation of the aircraft systems.

Aircraft Integrated Data System

The aircraft integrated data system (A.I.D.S.) is an extension of the “black box” aircraft accident data recorder. By monitoring and recording various engine parameters, either manually or automatically, it is possible to detect an incipient failure and thus prevent a hazardous situation arising.

Selected performance parameters may be recorded for trend analysis or fault detection. Existing instruments are used, wherever possible, to provide the signals to a magnetic tape. Further instrumentation, recording air pressure from points throughout the engine, oil contamination, tank contents and scavenge oil temperature, may be provided as required for flight recording.

After each flight the magnetic tape is processed by computer and the results are analyzed. Any deviation from the normal condition will enable a fault to be identified and the necessary remedial action to be taken.

Electronic Indicating Systems

Electronic indicating systems consolidate engine indications, systems monitoring, and crew alerting functions onto one or more cathode ray tubes (C.R.T.s) mounted in the instrument panel. The information is displayed on the screen in the form of dials with digital readout and warnings, cautions and advisory messages shown as text.

Only those parameters required by the crew to set and monitor engine thrust are permanently displayed on the screen. The system monitors the remaining parameters and displays them only if one or more exceed safe limitations. The pilot can, however, override the system and elect to have all main parameters in view at any time (Figure 9–22).

Warnings, cautions, and advisory messages are displayed only when necessary and are color coded to communicate the urgency of the fault to the flight crew. Provision is made to record any event or out of tolerance parameter in a nonvolatile memory for later evaluation by ground maintenance crews.

Electronic indicating systems offer improved flight operations by reducing the pilot workload through automatic monitoring of engine operation and a centralized caution and warning system. Reduced flight deck clutter is another feature as the multitude of instruments traditionally present are replaced by the C.R.T.s.

Nonintrusive Wear Monitoring

A few users have tried nonintrusive wear monitors (in applications where bearing or other component deterioration might be an issue). This technology uses neutron bombardment and has not grown in popularity over the past decade although it aroused considerable interest initially.

Oil Debris Monitoring

Oil debris monitoring is better suited to reciprocating machines, because of the lead time involved in analyzing the readings. The technique may be used online but is a long way from being done in real time. Reciprocating machines turn slower, so a sample may be taken and analyzed in time to indicate a trend. This is rarely the case with rotating machinery. Once metal particulates are detectable in an oil sample on a rotating machine, progression to failure is generally swifter than the analysis cycle (see Figure 9–23).

Protective Systems

Generally, the GT I&C has the following protective measures:

Permissives (Interlocks)

Permissives are conditions that must be met before a startup or operation can continue. Typically, they are as follows:

Liquid fuel permissives:

Startup Sequence, GT

This proceeds as follows:

Then the generator CB closes automatically or manually.

This system compares CDP with EGT and signals the fuel control valve to make changes as required. As CDP rises, the controller changes the set point of the EGT; it must drop. The controller signals the fuel control valve to close to the extent required. Power output stays at 100% while the change (less fuel flow) is made. This indicates the turbine runs more efficiently when ambient air temperature drops. However, this does not imply that the turbine is controlled by ambient air values. This is because CDP is also affected by:

Note: Other control mechanisms include:

Transmitters for pressure and temperature have no control function. They provide a 4–20 mA signal over the range specified. Their function is to indicate parameter values only.

Shutdown Sequence, GT

Starting Systems

The two energy sources for starting a gas turbine are

Engine starting systems can also be used to

If the latter use is not employed, note that the rotor must be turned at slow speed (about 6 rpm for 4 hours) after emergency trips to prevent shaft sag.

Turbine Blade Optical Pyrometry Mapping (Rotamap 1™)⁵

Turbine Blade Optical Pyrometry Mapping (Rotamap 2™)⁵

Use of Optical Pyrometry with the Siemens Vx4.3 Gas Turbines⁶

This section will discuss two pyrometer systems that have been built that meet the design requirements for use in heavy-duty gas turbines. They are of the “test bed” standard probe design that consists of a 2000 mm long cylindrical probe with an outer diameter of 27 mm that is reduced to 22 mm outer diameter for the part that goes into the hot gas path (Figure 9–25). These versatile probes are designed for use in turbine stages one, two, and three. Thus, the maximum insertion depth ranges from 80 mm in the first stage of the V64.3A gas turbine (70 MW) to 400 mm in the third stage of the V84.3A (180 MW).

In the first stages the thermal stress is dominant with peak values of several hundred MPa, whereas in the second and third stage the bending stress becomes more and more important (up to 650 MPa). It is evident that only superior chrome nickel alloys and state-of-the-art manufacturing techniques such as electron beam welding have to be applied in order to achieve probes that are suitable for this wide field of application.

With regard to the high temperatures, the probe is water cooled by a stainless steel 24 stage high-pressure impeller pump. Furthermore, the optical parts are purged and cooled by high-pressure nitrogen, a portion of which is discharged through tiny slots in the mirror, in order to cool the very tip of the probe. A closed loop control allows a predetermined nitrogen overpressure to be kept constant at all operating conditions, e.g., Δp = 1 to 5 bar. Even though the actual pyrometer is miniaturized a 22 mm outer diameter of the hot gas part is required, due to space for the sturdy sight tube, the cooling water and the nitrogen channels.

Application to Heavy-Duty Gas Turbines

The main purpose of temperature measurements on turbine blading is the experimental evaluation of numerical calculations that have been applied during the design process. Figure 9–26 shows a schematic of traverse units for linear and angular movements of the probes mounted on a gas turbine. Usually at least two pyrometer systems are applied at the same time, such that the leading edge (pressure side) and the trailing edge (suction side) of one blade stage can be investigated simultaneously. In addition, new vanes and blades can be investigated, using the test bed gas turbine as component carrier. As a side effect, quality assurance of all rotating blades may be performed by optical pyrometry. Even though thermocouples and thermal paint are largely applied for temperature measurements, optical pyrometry is widely used by several gas turbine manufacturers for turbine blade and vane testing. The systems applied are based on commercially available sensors, e.g., by LAND.

Since the required technique for the test bed was not readily available at the time, a high-resolution pyrometer has been developed in a joint effort between Siemens Power Generation and the Technical University of Berlin, where the basic research work was carried out at the Institute of Internal Combustion Engines.

A visual impression of the high-resolution prototype pyrometer probe and the data acquisition unit is given in Figure 9–27. The optical access at the tip of the probe and the flexible light guide, which may be as long as 15 m, are clearly visible. During operation, the sturdy data acquisition unit, containing both the electronic receiver and a robust industrial PC, is mounted adjacent to the gas turbine.

Hence, measurement data is transferred via a local area network to the host PC in the remotely located control room. The pyrometer system is characterized by a temporal resolution of 1 μs, a minimum field of view of 1 mm, and a measurement range from 600 to about 1500°C. The cylindrical probe is of the traverse type with a modular design. Thus, the minimum field of view can be adjusted over a wide range and all components can be exchanged easily, e.g., in order to use up- or down-looking mirrors or to use optical components that are suitable for long or short infrared wavelength measurements.

Figure 9–28 displays the probes and traverse units mounted on a model V84.3A 180 MW gas turbine ready for testing. At the rear, the connections and flexible hoses for cooling water and nitrogen can be seen. Different traverse units with extension distances ranging from 500 to 3000 mm for probe diameters from 6 to 30 mm with maximum speeds of 1200 m/s are available for a large variety of investigations in the test bed.

Since the pyrometers are traversed into the hot gas path, it has to be demonstrated that the probes have no influence on the temperature distribution to be measured. This has been thoroughly investigated, making use of different traverse velocities and especially of different probe positions as shown in Figure 9–29(a).

The first stage blades can be measured directly from pyrometer 1 positioned at the first stage vane outlet or from the combustor outlet (pyrometer 2). The result of the measurement is independent of the probe position for identical test conditions. Slight deviations are due only to decreased spatial resolution when the pyrometer is in position 2. Thus, unlike for flow probes, the 22 mm outer diameter size of the pyrometers has no influence on the results of measurement.

Figure 9–29(b) shows a schematic of the cylindrical pyrometer probe partly inserted into the hot gas path at the row 2 vane outlet in order to investigate the leading edge and pressure side of the row 2 blades. The blades are rotating past and the pyrometer receives temperature data from a certain radius. Additionally, a key phasor gives a once per revolution signal in order to determine the beginning of a new measurement.

During evaluation the 3D geometry software determines the intersection of the pyrometer beam and the rotating blade. Hence, each individual measurement point can be correlated to its origin on any blade.

The temperature vs. time plots of both systems for 10 blades are shown in Figure 9–30. The temperature values have been normalized by their mean value respectively. Both first stage blade measurements were accomplished during liquid fuel operation and are qualitatively in close agreement with one another.

However, maximum and minimum values are, as expected, considerably higher resp. lower for the high-resolution pyrometer. Thus, the state-of-the-art pyrometer provides a more realistic thermal image than the conventional system.

In order to evaluate the measurement results for the first stage, additional measurements are necessary, in particular for the stationary vanes. The radial hot gas temperature profile is determined by thermocouple probes that are traversed slowly into the flow channel, alternately to the pyrometer probes. Therefore, one can account for inhomogeneous temperature distributions at the outlet of the ring combustor. Furthermore, the brightness temperature of the combustor outlet is measured by the pyrometer itself for natural gas operation at base load in premix mode. The results of these measurements are normalized by the maximum hot gas temperature and plotted against the normalized radius (Figure 9–31).

The hot gas temperature reaches its maximum relatively close to the hub, whereas the brightness temperature of the combustor shows maximum values close to the tip. The latter can be used for analytical corrections of the influence of combustor radiation. Please note that the brightness temperature of the combustor for liquid fuel operation is about 200 K higher at the same load conditions due to the bright yellow oil flames.

Thus, the error of the vane measurements can easily be in the magnitude of 100 K. Hence, the electronic receiver has been equipped with optical high pass filters that limit the effective wavelength from 0.75–1.10 μm, i.e., the visible portion is excluded. However, for vane measurements under natural gas premix operation (light blue flames, no soot) the combustor influence can be estimated as follows: a “true” vane temperature of 1000°C, a measured vane emissivity of 0.9 (±0.05) and a measured combustor brightness temperature of 1200°C yields a maximum error of about +40 K at the vane’s leading edge; further downstream it is considerably lower.

In order to get temperature information from most of the blading surface, the pyrometer probes are operated in a scanning mode. For vane measurements, a certain viewing angle is chosen and the pyrometer traversed quickly (1200 mm/s) from the tip to the hub and subsequently withdrawn. Only the data acquired during the inward motion is evaluated, in order to process only undisturbed temperature values. After a waiting period of four thermal time constants of the vane a steady thermodynamic state was established again.

Pyrometer Calibration

The pyrometer was calibrated using a black body furnace before and after each engine testing. The calibration data were fitted to the following equation:

$R=\frac{\epsilon \cdot K\cdot C_1\cdot {\lambda }^{-5}}{\exp \left(\frac{C_2}{\lambda \cdot T}\right)-1}$ (2)

(2)

T is the target temperature and R is the radiance measured and normalized by the pyrometer. ε is the emissivity of the target and is the value of 1 for a black body furnace. λ and K are the effective wavelength and the efficiency of the pyrometer respectively and these are determined using a minimum least squares error method.

Figure 9–33 shows a calibration curve obtained before and after a certain engine testing. Slight degradation of sensitivity was found after engine testing. During the engine testing, high-pressure air supplied from a commercial compressor was used as purge air. Although a micro mist filter was installed between the compressor and the pyrometer, the optical system in the pyrometer was probably contaminated by the pollutant in the compressed air. Before next engine testing, the pyrometer was disassembled and the optical system was cleaned.

Repeatability of the pyrometer was also examined. Calibration was repeated six times and a calibration curve was obtained for each case. Figure 9–34 shows the temperature calculated with each calibration curve at a fixed and normalized radiance of 800. The deviation is within +/− 2°C, indicating that the pyrometer has sufficient repeatability.

In the calibration, the effective wavelength of the pyrometer is determined to be approximately 1.6 μm. Although gaseous medium like water vapor and CO₂ in the main gas stream also absorbs and emits radiation, these absorptions and emissions do not occur in the range between 1.5–1.75 μm.⁸

So in the engine testing, radiance measured by the pyrometer is not influenced by gas absorption and emission.

The short wavelength pyrometer which utilizes wavelength near 1 μm is not suitable for blades coated with TBC.⁹

Although TBC is applied to the production type of GGT-1B, several blades without TBC were assembled together.

Engine Installation

The pyrometer setup is shown in Figures 9–35 to 9–37. The pyrometer can be attached to several locations on the engine casing depending on the measurement target. At each position, it can be fixed at different circumferential angles. Thus, the detector can collect radiance emitted from a broad area on the blade surface.

As the probe is exposed to high-pressure and high-temperature main stream gas, it needs to be air-cooled during the engine testing. High-pressure air from a commercial compressor continued to be supplied to the probe at a maximum flow rate of 0.1 kg/s, which is far less than the main gas flow rate of 86.5 kg/s. In the engine testing of M7A, we confirmed that the cooling air of the pyrometer had little influence on the blade temperature. For L30A, the influence is even smaller since the ratio of cooling airflow rate to main gas flow rate is much smaller than that of M7A.

Digital Telemetry

Digital telemetry offers many “space age” possibilities. It is what enables a diagnosis team on the ground, a test pilot, or the diagnostic interface of an engine in flight to read and transmit data that, when interpreted by human or automated means, can save millions of revenue dollars. For instance, the data sent by telemetry (air to ground) could mean, for instance, that a module on a specific engine needs to be removed when the plane lands and a spare rotor used to replace it, before the engine can take off again.

Similarly, telemetry has great uses in offshore platform or remote plant and field operation. It offers the same potential to marine end users, too; however, ship and ferry engineers tend to be a rather self-sufficient crowd by necessity and culture. Their engine data can certainly be transmitted to on-shore offices; however, the engineer on board often has taken action before getting a call from a “land” office.

The two hardware units described next reflect the current state of digital telemetry instrumentation. The reader may contact the source for further operating data, as well as internet search for manufacturers of similar products.

Rotatel™ Telemetry Unit

Minitel™ Telemetry Unit

Specialized Diagnostic Systems

Today, different diagnostic techniques are used individually or in complementary application, as the following case studies illustrate.

Selection of Relevant Operating Parameters

Taking the requirements of the previous sections into account (time, skill of personnel) and the fact that most operating parameters vary with varying ambient conditions, it becomes obvious that monitoring and displaying only directly measured parameters is not sufficient.

In order to allow an easy interpretation and financial evaluation in a timely manner, the following parameters must be provided by the system:

And last but not least, the combustor pulsation must be monitored. This parameter requires a little more attention than the previously mentioned parameters. The evaluation of the failures that were caused by elevated pulsation shows the importance of the frequency spectrum of the pulsation. Figure 9–41 shows a typical pulsation spectrum obtained from an ABB GT11N-EV gas turbine. Obviously a well-defined peak at a given frequency is more damaging to the equipment than is a “white noise” type of pulsation, even though the overall RMS level of the two signals may be the same.

Thus the system must provide the following values:

In addition, the system must be capable of monitoring user-defined operating parameters.

Data Archiving

The system must provide a data archive that allows the following tasks:

The system is designed to provide the following:

Modular Program Structure

The monitoring system is a PC-based tool that reads process data from up to three different sources simultaneously (see Figure 9–42).

The modular structure was chosen, among other reasons, so that:

Instrumentation

For the measurement of the combustor pulsations a heat resistant piezoelectric pressure transducer is used. The transducers have been used in several gas turbines for several years without failing. The transducer must be installed as close to the burner as possible. By placing the transducer close to the source for the pulsations, the likelihood of measuring in the vicinity of the pressure node is relatively small.

Pulsations

The combustor/burner specific spectrum is best seen when using a one-minute peak hold spectrum. A typical example for this turbine is shown in Figure 9–42.

This type of burner arrangement exhibits two characteristic frequencies.

The amplitude of the one in the 124 Hz range can be influenced by the overall fuel/air ratio and by the flow pattern downstream of the burners. Increased air leakage between the combustor inner liner and the burners will impact the amplitude of this frequency.

Large amplitudes in the range of 10–12 Hz indicate that some burners are starved of fuel. If the fuel and air distribution is correct, this frequency should not be dominant.

A crack in the general vicinity of the transmitter or a shift in the position of the casing, in which the transmitter is supposed, can falsify the signal significantly. An example is shown in Figure 9–43.

This is a particular good example of why the frequency spectrum is important. Would the above signal only be measured and displayed as an RMS value, one would only observe a huge increase in the signal level. If this were the only information available, the above signal would require the unit to be shut down and inspected. The spectrum, however, permits us to make the following observations:

If this phenomenon is observed and the spectrum is available, the unit can be kept operational until an inspection can be scheduled. Naturally the unit needs to be monitored closely to watch for other indications of component deterioration.

Summarizing the above, the following statements can be made:

Typical magnitudes of the characteristic frequencies but otherwise increased amplitudes across the spectrum indicate possible shifting casing positions and/or leakage airflow around the transmitter.

NO_X Emissions

On a GT11N1-EV the NO_x emissions are mainly a function of the fuel/air ratio. The NO_x emissions are very sensitive to the local fuel/air ratio. Thus if changing emissions are observed for a given operating condition, they can be caused by any of the following conditions:

Item 2 through 4 can usually be ruled out fairly easy by checking the condition of the gas filter, gas valve positions and the gas quality.

Thus if a change in the NO_x emissions is detected and items 2–4 can be eliminated as a root cause, one has a strong indication that one or more hot gas path components are deteriorating.

Turbine Exhaust Temperature Profile

Unlike the NO_x emissions that for a given operating point are pretty constant the temperature profile after the turbine is affected by ambient conditions and the period of time that the gas turbine has been operating at the given load. The latter is only significant during the first 4 hours after a start. On a unit that starts daily and only operates for 10–14 hours per day, this effect can make the detection of changes in the profile difficult as the ambient conditions typically change during operating period as well.

An inspection of the unit revealed that the combustor inner liner indeed had a crack that most likely was created during operation on fuel oil with high pulsations on January 8, 1999. Thus, the TAT spread is a good indicator for cracks in the hot section of the turbine, but on daily cycling units the change in the profile may be difficult to detect.