DERIVATIVE ENGINES FOR MARINE AND INDUSTRIAL USE 4.1 INTRODUCTION There are a number of industrial situations where the lightweight characteristics of an air- craft engine can be advantageously applied to generate mechanical power. The power, however, is mostly for driving other machines instead of developing aerodynamic thrust. Hence, the fan is no longer needed, and the developed mechanical power drives an electric generator, a compressor, or a ship’s propellers. Derivative engines based on aircraft engine technology have been developed expressly for such a situation for producing mechanical energy. Electrical power generation on an offshore oil platform is a good example. A tremen- dous amount of mechanical energy is required to operate drills, pumps, compressors, and rigs on exploratory and production oil platforms. The geometry of a typical rig calls for a platform weighing several hundred thousand pounds located 50 to 100 ft above the surface of the water, with the superstructure supported by four legs resting at the bottom of the sea. Strictly from stability considerations, it is of vital importance to limit the weight of the overall structure above the water line, as well as that of the individual components located on the platform. The weight of a typical industrial gas turbine is prohibitively large, and hence a design derived from aircraft engine technology would be ideal. Pipeline pumping and gas compression applications also require a large operating speed range, as opposed to a fixed speed requirement of a power-generation gas turbine. Add to that the capability of a combustion system to burn liquid or gas fuels that are abundantly available on an oil plat- form, and the suitability of a derivative engine becomes readily apparent. Shipboard prime movers and many other marine applications have similar weight restrictions. Low-speed reciprocating diesel engines have been traditionally employed for marine propulsion, but the engines tend to be physically large. For example, a supercharged and after-cooled 12-cylinder direct drive diesel engine of 30 in bore can produce 18,700 hp and weigh 74 lb per brake horsepower, for a total of 1,383,800 lb. The engine is 25 ft high, 8 ft wide, and 45 ft long, occupying a total volume of 9000 ft 3 . In comparison, an LM2500 derivative engine housed in its own 12-ft high, 11-ft wide, and 32-ft long module can develop 19,500 hp and have a gross weight of 50,000 lb, or 2.56 lb/bhp. Smaller bore reciprocating engines have a definite advantage over larger ones if weight is a prime consideration. In the above example, an eight-cylinder engine of 12 in bore may be employed, with gearing to the propeller shaft. The gearing is expected to weigh 20 percent as much as the engines and add 15 percent to the floor space of the engine to which it is CHAPTER 4 103 Copyright © 2005 by The McGraw-Hill Companies, Inc. Click here for terms of use. connected. To estimate the new dimensions and floor area required, assume that the smaller engines operate at the same stress level. For the same output at the same piston speed and mean effective pressure, the number of 12 in bore cylinders required will be 12 × (30/12) 2 = 75. Since there will be some losses in the gears, take 10 eight-cylinder engines, or total 80 cylinders of 12 in bore. Total engine weight calculates to be 18,700 × 74 × (12/30) × (80/75) = 590,400 lb, and this compares favorably with 1,383,800 lb for the single engine. To this must be added 20 percent for the gears, so the total weight of the engines and gears will be 590,400 × 1.20 = 708,480 lb. The height of the engines will be 25 × (12/30) = 10 ft, which will allow two more useful decks over the engine room. The floor area covered by the engines plus the gears will be 15 × (80/75) = 16 percent greater than that of the single engine because floor area is proportional to the piston area with engines of similar design. By eliminating the fan of an aircraft engine a few stages may be added to the low-pressure compressor. The axial compressor is split into low- and high-pressure modules, each pow- ered by individual turbine sections mounted on the same shaft. The power turbine is con- nected to the low-pressure compressor at the forward end and to a driven equipment at the other end. With concentric shafts the speed of both compressor sections offers more flexibility for optimizing. Three shaft derivative engines call for a separate power turbine that is directly connected to the driven machinery. All three shafts operate in their desig- nated speed range. Design innovations are incorporated to obtain the required long-life characteristics of most industrial applications (Fig. 4.1). Derivative engines offer a number of benefits. The size and weight of the complete engine lend them to assembly and packaging as a complete unit within the manufacturer’s 104 APPLICATIONS FIGURE 4.1 Solar Turbines Titan 130 engine for offshore oil production platform. facility. A generator or a compressor may be included in the package, together with the accessories purchased by the customer. Installation may also proceed at the job site by fac- tory personnel specially trained for debugging and performance matching. Because most customers strive to control operational costs, the engines may be readily adapted for remote control and automation. Offshore and remote pipeline pumping stations are normally designed for unattended operation. When auxiliary systems are uncomplicated, oil-to-air exchangers are used in place of water cooling. Starting devices requiring little energy are reliable. Hence, aviation-technology-derived engines lend themselves to automatic control from a distance. Aeroderivative engines can run continuously without inspection, until the monitoring equipment indicates a fault or sudden performance variation. Such an incident can best be handled by removing and replacing the gas generator section with a spare, so that the module can be inspected, evaluated, and repaired more efficiently at the factory. Under these circumstances, offsite maintenance plans offered by a number of manufactur- ers and leading service organizations play a useful role. Technical servicing is then mostly restricted to conducting minor running adjustments and related routine tasks. 4.2 SHIP PROPULSION PLANT A combination of diesel and gas propulsion arrangements has been selected by the Royal Netherlands Navy for its fleet of frigates (Broekhaus and Rand, 2002). A frigate is designed to act as an area air defense ship within a task group and as a command platform, and is capable of prolonged operation at sea. Each ship is equipped with two Rolls Royce Spey gas turbines and two Wartsila cruise diesel engines. The gas turbines are resilient- mounted in the forward section of the engine room, while the diesel engines are placed in the aft portion. The engines drive a controllable pitch propeller through a conventional gearbox with a clutch (see Fig. 4.2). Electrical load for the ship is generated by separate diesel generators. Spey gas turbines are developed from the TF41 military aviation engine. Proven by 500,000 h of marine operation, maximum power from the turbine is 19.5 MW. The turbine has two spools, operates on a simple cycle, uses modular construction, and enables Rolls Royce to compete with General Electric’s LM2500 engine in the marine market. Selection of DERIVATIVE ENGINES FOR MARINE AND INDUSTRIAL USE 105 FIGURE 4.2 Propulsion plant arrangement (Broekhaus and Rand, 2002). Rolls royce spey gas turbines Gear box Cruise diesels the engine is based on low initial purchase and operating costs, proven reliability, and acceptable technical risk. LM2500 engine was ruled out by the developers mostly because of the Navy’s preference for the Rolls Royce gas turbines used on their earlier frigates. Commonality of parts and the Navy’s familiarity with the earlier engines thus put General Electric’s engines at a disadvantage. Naval vessels have a number of unique requirements for onboard placement of the propulsion equipment. The ship’s outer surfaces must be sloped to reduce the radar cross section. The sides of the ship where the engine intakes are positioned are flared outward to facilitate maintenance chores such as filter replacement, cleaning, and installation of cov- ers. Another important consideration is impingement of exhaust gases upon the air intake manifolds; inadequate precautions may result in fouling of the filters. Air separator clean- ing may be needed if the problem persists. Gas turbines are required to meet contractual requirements for visible smoke during sea trials. Excessive emissions are noted mostly because of effusion holes in the canned com- bustor walls for increased cooling. A corresponding reduction in the number of air blow- holes maintains the pressure balance within the can. But this modification leads to an excessive reduction in primary zone combustion air, resulting in generation of smoke. The effect is observed when power output exceeds 14 MW. Additional design changes in the combustion cans are underway to increase primary air to reduce smoke emission. Power output of the gas turbines will then be increased to 18 MW. This brings into question the manufacturer’s power-setting guarantees. The difference in top speed of the ship between 18 and 19.5 MW power output is 1 / 2 knot, which may be acceptable. In the absence of other proven propulsive technologies, gas turbines compare favorably with their rival, diesel engines. Gas turbines offer greater power density than a diesel engine, but have higher specific fuel consumption and initial purchase cost. Another impor- tant consideration is the use of integrated propulsion and electric power generation as opposed to the more traditional separate approach. In the final reckoning, operating costs for maintenance (material and labor) throughout the engine’s life against the cost of fuel burn represent the crux of the financial argument in either engine’s favor. Based on this experience from the Royal Netherlands Navy, the following rules may be shaped for future prime mover selection: • Diesel engines offer a better fuel burn argument over simple-cycle gas turbines through- out the operating regime. But when the turbines are loaded up to their top capacity, or when advanced cycle gas turbines are employed, the diesel engines’ superior fuel econ- omy is challenged. • Diesel engines provide a cheaper initial propulsion plant and lower fuel burn cost, but experience higher through life cost because of higher maintenance requirements. Exceptions to this trend do come up occasionally. • Diesel engines are a must for the export market from a commercial angle, where the higher technology risk may play against its adoption. • During operating profiles calling for sprinting and loitering, diesel engines achieve lower cost compared to gas turbines, but this must be weighed against increased damage and maintenance costs incurred in operating partially loaded diesels. Thus, it would be prefer- able to install two 4-MW diesel engines for part load running than a single 8-MW unit. Note, however, that the weight and space requirements of the two engines do not differ from each other substantially. Also, gas turbines have greater specific power output, and do not suffer penalties when running at part load. • Diesel engines require a large maintenance envelope on all sides. Gas turbines, on the other hand, need accessibility from only one or two sides for the removal of modules. 106 APPLICATIONS • Diesel engines do not have a high air consumption rate, so large intake and exhaust man- ifolds and stacks are not needed. Some treatment of exhaust gases is needed to meet emis- sion regulations, but space requirements are not large. • Infrared emissions are lower in diesel engines than in gas turbines. Gas turbine compressors used to be cleaned by crank soak washing or by injecting solid compounds such as nutshells or rice husks at full speed with the unit on line. With the advent of coated blades for compressors, this method of online cleaning by soft erosion was no longer preferred because it caused pitting. Additionally, unburned solid cleaning com- pounds and ashes cause blockage of the carefully designed turbine blade cooling systems if ingested into the stream. Wet cleaning with detergents was introduced in the 1980s, and time intervals between online washing and a combination with offline washing required establishment. An airflow reduced by 5 percent due to fouled compressor blades will reduce the out- put by 13 percent and increase the heat rate by 5.5 percent (Hoeft, 1993). Marine engines are particularly susceptible to the phenomenon of fouled compressor blades. The intake of sea air near shorelines and further away in the ocean increases a gas turbine’s specific fuel consumption because of soot, salt, and dirt adhering to the surface of compressor blades and vanes. Aerodynamic performance of the airfoils is reduced by the restricted airflow, while also increasing frictional losses from the associated surface roughness. Cleaning the blades by spraying a detergent solution while motoring the gas turbine using the starter mechanism has been reported to restore compressor performance to a certain level. Even with the wash- ing after specific periods, compressor performance continues to degrade. Fouling rates vary considerably, and are specific to each application (Stalder, 1998). Surrounding environment, climatic conditions, and plant layout play a role in the frequency required for washing. Site weather parameters probably have the most impact on the foul- ing rate and consequent performance degradation. Most fouling deposits are mixtures of water-wettable, water-soluble, and water-insoluble materials. The deposits progressively become more difficult to remove if left untreated, as the aging process bonds them more firmly to the airfoil surface, thus reducing the cleaning efficiency. Water-soluble com- pounds tend to promote corrosion when chlorides are present. Water-insoluble compounds may be from hydrocarbon residues or from silica. Demineralized water is preferable for online cleaning, and the detergent must fulfill the fuel manufacturer’s specifications. Hot wash water (140 to 170°F) will soften the deposits better than cold water, and will prevent thermal shocks; however, equipment must be available for heating the water. One method of arresting the rate of deterioration calls for the application of a fouling resistant coating on the compressor airfoils (Caguiat, 2002). In a series of tests conducted by the U.S. Navy, two different coatings were selected to determine their effectiveness in eliminating surface roughness. Made by Sermatech International, one coating was used for the first two stages and the other for the remaining stages, with both possessing an inert top layer and an anticorrosive aluminum-ceramic base coat. A chemical similar to Teflon was added for improved fouling resistance. The tests were designed to focus primarily on the effects of high levels of concentration of salt in the air. The salt was injected by means of air atomized spray nozzles mounted at the turbine inlet. The ingestion rate of salt is based on 0.01 ppm of air. The average mass flow rate for the test engine is 38 lb/s. Using 0.05 lb of salt per pound of water, the flow rate from the nozzles was set at 3.0 gal/h over a period of 0.25 h. It may be assumed that in the accelerated test environment, salt would have a propen- sity to deposit in the same locations on the blades as it would in a nonaccelerated shipboard environment. The assumption has validity since salt tends to adhere to the stagnation points on the compressor blades as the air stream moves through the compressor. DERIVATIVE ENGINES FOR MARINE AND INDUSTRIAL USE 107 The test engine speed and load from the mechanically coupled generator are held nearly constant during the evaluation. As fouling progresses, the compressor discharge pressure will decrease from the clean engine level, while the fuel consumption rate will need to increase in order to maintain the load. Increased surface friction losses will increase tem- perature at the compressor discharge and at the turbine inlet. Figures 4.3 and 4.4 provide results from the test. The fuel consumption and compressor discharge temperatures indi- cate a mostly linear upward trend as the salt ingestion increases, while the compressor dis- charge pressure shows a nearly linear downward trend. The combination equates to a downward trend in adiabatic efficiency. A loss of 7 percent in the compressor discharge pressure and an increase of 3 percent in fuel consumption required merely 0.065 lb of salt. Comparisons were also made between coated and noncoated blades in a similar manner. 108 APPLICATIONS FIGURE 4.3 Temperature variation due to compressor blade fouling (Caguiat, 2002). FIGURE 4.4 Compressor performance degradation due to blade fouling (Caguiat, 2002). A clear reduction in degradation of each of the parameters is observed with the coated blades. The results were verified at a number of load levels. 4.3 GAS COMPRESSION SYSTEMS FOR PIPELINE PUMPING Gas turbine driven compression systems may be used for many different situations. Some examples are: gas transport in pipelines, pressure boosting, reinjection of natural gas into oil wells, gas lift to support oil production, storage and withdrawal of gas from storage facilities such as caverns, and gathering from diverse areas of gas fields. Often the fuel gas for the turbine is withdrawn from the line. Although fuel efficiency is an important goal in transport applications, the reinjection gas is virtually cost free since it would have to be flared if it were not reinjected. Even in pipeline applications the gas appreciates in value the further along the line it travels. And for pressure boosting the prime consideration might be the capability to attain the desired level. Consequently, most applications tend to focus on the capability to provide acceptable performance over a range of operating scenarios rather than at a well-defined operating point (Kurz and Fozi, 2002). Since each project has operating conditions that are critical to the success of the project, effort is expended in meeting that goal rather than achieving often-contradictory objectives. The criteria may not be performance related—rotor dynamic stability, reliability, and avail- ability are some examples. A typical plant may opt for extreme guarantee points such as near surge or deep in choke conditions of the compressor. Little emphasis is then placed on performance at operating points that are seldom used and have minimal impact on the oper- ating cost. Those parameters that most affect profitability must then be analyzed. Some typ- ical situations will be reviewed. In gas transport applications, operating costs are linked to the amount of fuel used and the maximum amount of gas that can be compressed for a given operating condition. Note that efficiency and maximum power output of the gas turbine and efficiency and head developing capacity of the compressor will affect the outcome of the required criteria. None of them alone will determine the outcome. Gas gathering is another example. Costs in this form of operation hinge around the capa- bility to produce heavier hydrocarbons that are part of the associated gas. The preference for gas turbine fuel is at the lighter end of the associated gas. Hence, fuel efficiency of the compression system has little impact on the operating costs. In fact, a premium is placed on the reliability and availability of the unit. Another important parameter is the maximum flow that can be achieved by the package, but this may be redundant if the flow rate from the source is not high. The operating characteristics of a gas compressor for pipeline oper- ation and for a gathering and storage situation are shown in Fig. 4.5. The key issue in gas reinjection application is the ability to operate safely and reliably at a significantly high discharge pressure of up to 700 bars, or 10,000 lb/in 2 . Performance and efficiency issues may then be almost irrelevant when compared to the rotor dynamic stability and structural integrity. When multiple operating points are defined, the compressor design may not be opti- mized for best operation at the usual point, and compromises may then be needed to cover the array of points within the operating envelope. Efficiency-related points should not be defined at the edges of the operability of the compressor, but rather at the most likely oper- ating points. The number of identical units operating at a station also plays a role. Two or more compressor sets may operate at a location, with another unit as a spare for increased operating and maintaining flexibility (Fig. 4.6). DERIVATIVE ENGINES FOR MARINE AND INDUSTRIAL USE 109 110 APPLICATIONS FIGURE 4.5 Gas compression characteristics: pipeline operation (upper); gathering and storage (lower) (Kurz and Fozi, 2002). FIGURE 4.6 Multicompression train. 4.4 OPERATIONAL EXPERIENCE OF LM2500 ENGINE An integrated electric propulsion and power service system provides for greater flexibility, efficiency, and survivability of naval ships. Examples of this concept include the type-45 destroyer program for the Royal Navy and DD (X) program for the U.S. Navy (Harvey, Kingsley, and Stauffer, 2002). The U.S. Navy system comprises a General Electric LM2500 gas turbine engine directly coupled with a Brush Electric Company synchronous generator. The unit is capa- ble of producing 21.6 MW at 0.8 power factor. To determine system response during start- up, a ramp load is conducted during the system test to characterize power system interface, stability, and control performance. Step loads are in the form of propulsion motor acceler- ation at an average of 4 percent power per second up to steady-state condition. System oper- ating conditions are 25, 50, 75, and 100 percent of shaft output power of the propulsion motor. Figure 4.7 provides details of turbine performance during the 100-percent ramp. Turbine performance and voltage regulation were exceptionally well controlled. The gen- erator power turbine speed followed the U.S. Navy’s 3.33 percent droop curve standard with very little deviation during each ramp-up operation. Since motor load is proportional to the third power of speed, motor acceleration was shaped to provide a linear power ramp and a cubic speed profile. Ramp unloads were also carried out during the system test to a similar set of charac- teristics as in the ramp-up mode, but with the engine decelerating. Since the motor con- verter is nonregenerative, the motor’s deceleration is restricted, and it ramps down at a rate of 2 percent power per second. Figure 4.8 illustrates this feature, with the deceleration time DERIVATIVE ENGINES FOR MARINE AND INDUSTRIAL USE 111 FIGURE 4.7 Turbine generator and propulsion motor characteristics—ramp acceleration. period more extended than the period during acceleration. The motor slows down in a lin- ear speed profile as opposed to a cubic acceleration schedule. Step unloading of the power generating system represents a considerably more severe situation than a gradually unloaded ramp operation. The engine is driving at a constant mechanical power and the generator exciter is providing a constant excitation when the electrical load is suddenly lost. The frequency and voltage of the machine overshoot momentarily, hence mechanical power input to the generator must be reduced rapidly to prevent overspeed and return the power turbine to normal speed. This is accomplished by adjusting the fuel-metering valve. In a similar manner the automatic voltage regulator must rapidly reduce the excitation current to the generator to return the voltage to a normal level. On the loss of motor load at 100-percent power, the frequency of the LM2500 genera- tor rises from its initial steady-state droop setting of 58.0 Hz, or 3480 rpm, as is shown in Fig. 4.9. The upper steady-state tolerance limit is 63 Hz (3780 rpm) or 5 percent, which is met. The final frequency is 60 Hz (3600 rpm), which represents the no-load frequency. A high-power propulsion motor trip may result in a flameout condition in the turbine engine. Step response tests of the fuel-metering valve from the maximum to the minimum positions indicated some undershoot in the metered flow rate, but it is not enough to cause a flameout in the engine. The problem lies more in the control algorithms. The imple- mented core software may not have sufficient deceleration limitation to prevent a flameout on rejection of the full load. By changing the logic to allow fuel flow to be maintained at or above the minimum schedule while keeping the gas generator’s deceleration rate within acceptable limits, a flameout can be prevented. The main protection against reignition is to shut down when a flameout occurs. In this connection ultraviolet-based detection sensors have been proved to be fast enough to indicate a flameout. The detectors will then initiate a shutdown of the turbine, if a loss of flame is detected following a step unload. Instances 112 APPLICATIONS FIGURE 4.8 Turbine generator and propulsion motor characteristics—ramp deceleration. [...]... AND INDUSTRIAL USE 22 20 5200 3700 18 16 5000 3650 Voltage 10 46 00 8 Power, MW 12 48 00 3600 3550 Power turbine speed, rpm 14 6 44 00 4 3500 2 42 00 0 3 6 9 12 15 18 21 24 27 0 30 345 0 Time, s FIGURE 4. 9 Turbine generator and propulsion motor characteristics—100-percent step unload of overshoot in the fuel valve have also been observed during a partial step unload A loop simulation for the valve may then... 78.6 mm 27.0 mm −52° 344 K 0.678 kg/s 59,828 rpm 2.91 0.616 135 DIESEL AND AUTOMOTIVE ENGINE TURBOCHARGERS φ = 130° 45 ° mirror Photomultiplier z Window 6 R Window 7 C0 Window 4 φ = 0° (tongue) FIGURE 5. 14 Laser Doppler velocimetry setup (Karamanis, MartinezBotas, and Su, 2000) within set limits Statistical uncertainty is determined to be less than 1.8 percent for mean velocity and 4. 4 percent for root... can reduce the specific fuel consumption from about 3 to 14 percent in the engine’s speed range The reduction in fuel consumption becomes more marked as the engine’s load is reduced, as can be seen in Fig 5.1 from the family of constant load curves ranging between 1 /4, 1/2, 3 /4, and full engine load However, at full load below 140 0 rpm and 3 /4 load below 1000 rpm the specific fuel consumption is inferior... 1 14 APPLICATIONS Besides a good volume-to-shaft horsepower (shp) ratio, the turbine must contain turbomachinery components to address a number of constraints and needs: • • • • Means for power augmentation to cover short-term needs Components with wide-operating mass-flow range Parts designed for high thermal cycle count and shock loads Durability and ease of maintenance of engine systems Figure 4. 10... exhaust discharges will be 144 °, 120°, and 90° respectively To overcome exhaust gas interference in the manifold, manifolds are subdivided so that in the case of an in-line six-cylinder engine, cylinders 1, 2, and 3 are grouped together and, similarly, cylinders 4, 5, and 6 are grouped together, and there is now an extensive exhaust interval between subdivided manifolds of 240 ° The exhaust discharge... Gabrielson, R., “Progress on the AGATA project—A European ceramic gas turbine for hybrid vehicles,” ASME Paper # 95-GT -44 6, New York, 1995 Szwedowicz, J., “Harmonic forced vibration analyses of blade assemblies modeled by cyclic systems, part I—theory and vibration,” ABB Technical Reports HZX-ST 5 849 , Baden, Switzerland, 1996 Walsh, P P., and Fletcher, P., Gas Turbine Performance, ASME Press, New York, 1998... from 10:1 for a naturally aspirated engine to 9:1 for a low boost pressure or 8:1 for a medium to high boost pressure turbocharged engine In a direct injection diesel engine having a normal 16:1 compression ratio, the turbocharged engine compression ratio may be lowered to 15:1 or 14: 1 The fundamentals of supercharging are based on changes in pressure exerted on the gas being delivered to the cylinders,... 1.0:1 or less 1.0–1.5:1 1.5–2.0:1 2.0:1 and higher DIESEL AND AUTOMOTIVE ENGINE TURBOCHARGERS FIGURE 5.7 125 Naturally aspirated and supercharged engine PV diagrams specific fuel consumption against engine speed are shown in Fig 5.8 for three different stages of engine tune: (1) naturally aspirated, (2) turbocharged, and (3) turbocharged and intercooled The specific fuel consumption curves indicate that... either side of the 140 0 to 1800 rpm speed band, but the difference is more significant toward maximum speed Turbocharged petrol engines generally have reduced compression ratios to accommodate the high cylinder pressures, and under load the ignition timing is automatically retarded to prevent detonation from taking place, while at full load a rich mixture is necessary Consequently, the turbocharged petrol... 2.2:1, the b.m.e.p also rises from 7 .4 bar to 10.6 bar respectively The intercooler has very little effect on the b.m.e.p below a pressure ratio of around 1 .4: 1 However, if the air charge is intercooled by an air-to-liquid intercooler, so that the temperature is maintained at about 80°C, then there is a marked increase in b.m.e.p for a boost pressure ratio of 1 .4: 1 to 2.2:1, which raises the b.m.e.p . 113 FIGURE 4. 9 Turbine generator and propulsion motor characteristics—100-percent step unload. 5200 5000 48 00 46 00 44 00 42 00 0 3 6 9 12151821 242 730 Time, s 22 20 18 16 14 12 10 8 6 4 2 0 Power,. family of constant load curves ranging between 1 / 4 , 1 / 2 , 3 / 4 , and full engine load. However, at full load below 140 0 rpm and 3 / 4 load below 1000 rpm the specific fuel con- sumption. noncoated blades in a similar manner. 108 APPLICATIONS FIGURE 4. 3 Temperature variation due to compressor blade fouling (Caguiat, 2002). FIGURE 4. 4 Compressor performance degradation due to blade fouling (Caguiat,