Online Water Wash Tests of Ge J85-13
Total Page:16
File Type:pdf, Size:1020Kb
Proceedings of GT2005 ASME Turbo Expo 2005: Power for Land, Sea and Air June 6-9, 2005, Reno-Tahoe, Nevada, USA GT2005-68702 ONLINE WATER WASH TESTS OF GE J85-13 Elisabet Syverud and Lars E. Bakken NTNU, Norwegian University of Science and Technology, Department of Energy and Process Engineering, N-7491 Trondheim, Norway ABSTRACT publications give a historical review of online washing systems This paper reports the results of a series of online water and a classification of available systems [1,2]. wash tests of a GE J85-13 jet engine at the test facilities of the Today’s online washing equipment for aeroderivative Royal Norwegian Air Force. The engine performance was engines can be categorized in two main pressure ranges: Fluid deteriorated by injecting atomized saltwater at the engine inlet. pressures up to 10 bar are considered low pressure systems , Then the engine was online washed with water injected at three while high pressure systems have fluid pressures above 50 bar. different droplet sizes (25, 75 and 200 µm) and at water-to-air The atomized droplets produced by high pressure systems have ratios ranging from 0.4% to 3% by mass. Engine performance a droplet diameter typically less than 150 µm and resembles the was measured using standard on-engine instrumentation. Extra water occurring naturally in clouds and fog. Low pressure temperature and pressure sensors in the compressor section systems will have larger, drizzle-like droplets with diameter provided additional information of the propagation of deposits ranging from 100 to 500 µm and larger. Low pressure systems in the aft stages. The measurements were supported by visual with air assisted nozzles will generate smaller droplet sizes, observations. resembling the droplets of the high pressure systems. Large The overall engine performance improved rapidly with droplets may cause blade erosion in the compressor. [2-5] online wash. The build-up of deposits in the aft stages was The fluid injection rate (water-to-air ratio) has an impact on influenced both by the droplet size and the water-to-air ratio. the internal surface wetting of the compressor. Due to possible The water-to-air ratio was the most important parameter to control instabilities, flame-out, or blade erosion the water-to-air achieve effective online washing. ratio has generally been kept low. A typical online washing system for aeroderivative engines has fluid injection in the Keywords: Compressor cleaning, axial compressor, stage range from 0.2% to 0.8% (mass based) [2-5]. characteristics, GE J85-13 The flow field within an axial compressor subject to water injection is complex to predict due to the two-phase nature of INTRODUCTION the flow. The motion of water droplets inside axial compressors Online water washing has become increasingly important was studied by Marchik [6] and Tsuchiya and Murthy[7]. While with operators of industrial gas turbines due to the potential for Marchik studied droplets smaller than 70 µm, Tsuchiya and reduced degradation rate and increased operating intervals. Murthy tested a six stage axial compressor with water injection Successful online washing requires close attention to the gas rates up to 15% and with droplets of 90 and 600 µm. Several turbine flow path geometry, the operating profile and the nature papers offer theoretical approaches to wet compression[8-10]. of the airborne fouling at the compressor inlet (after filtration). These studies are related to tiny droplets, less than 15µm, with Several manufacturers offer online washing equipment and no velocity slip between the droplet and air. A study of the there exists many patents on gas turbine water wash. There is effects on gas turbines of naturally occurring water in the currently no consensus on a recommended method for effective atmosphere is given by AGARD [11] and gives additional online water washing. System properties like droplet size, insight into the operating limitations of gas turbine engines. droplet velocity and fluid injection rate vary from one system to To understand and reveal the mechanisms related to online another. This makes it difficult for operators to select the best water washing a systematic test campaign was performed on a online water wash system for their application. Two recent GE J85-13 jet engine. The engine performance was deteriorated 1 Copyright © 2005 by ASME by injecting atomized saltwater at the engine inlet and the mix Mixture (humid air) engine performance was restored using online water washing. s Static condition The water was injected at three different droplet sizes (25, 75 t Total condition and 200 µm) at water-to-air ratios from 0.4% to 3% by mass. The tip Tip cleaning effectiveness was measured in terms of improved VMD Volume median diameter engine performance using standard on-engine instrumentation. w Tangential velocity component In addition, extra temperature and pressure sensors installed in 2.1, 2.2,... 2.7 Compressor stage 1, 2,... to stage 7 the compressor section provided added information on the 3 Compressor discharge propagation of the deposits into the engine. The measurements 5 Turbine discharge were supported by visual observations through the borescope and by laboratory analysis of the stator vane deposits. The TEST FACILITIES AND ENGINE DESCRIPTION results of the axial compressor deterioration are reported by Engine testing was carried out at the RNoAF’s test facilities Syverud, Brekke and Bakken [12], and that paper should be seen at Kjeller, Norway. in context with the present work. The test facilities, the GE J85- The General Electric J85-13 engine is a compact, light 13 engine and its initial condition, the engine instrumentation weight, single-spool turbojet engine. It has an eight-stage axial- and the deterioration method is described in the present paper flow compressor with bleed-off valves, adjustable inlet guide only when required for completeness. vanes and a variable exhaust nozzle. The compressor pressure ratio is 6.5:1. The variable geometry is controlled by the throttle NOMENCLATURE angle, but the timing is ambient temperature biased. At ISO A Flow area, [m2] conditions, IGV will be at maximum deflection and bleed-off- cp Specific heat at const. pressure, [kJ/kgK] valves will be fully open below ~81% corrected speed and fully C Absolute air velocity, [m/s] closed at >96% corrected speed. At ambient temperatures above CIT Compressor inlet temperature, [K] ISO conditions, the bleed-off-valves will close at lower speed CIP Compressor inlet pressure, [kPa] settings. The nozzle is controlled by the throttle as long as the GE General Electric engine is running below the exhaust temperature limit. When the IGV Inlet guide vanes maximum exhaust temperature is reached, the exhaust nozzle will ISO Int. Organization for Standardization (*) increase the flow. This reversal in the exhaust nozzle schedule m Mass flow rate, [kg/s] occurs close to the maximum throttle angle. N Shaft speed, [rpm] GE J85-13 compressor geometry is published by Tesch and NASA National Aeronautics and Space Steenken [13] and compressor stage characteristics are Administration published by Milner and Wenzel [14]. P Pressure, [kPa] r Blade radius, [m] ENGINE INSTRUMENTATION R Gas constant, [kJ/kgK] Additional sensors were installed to provide more R.H. Relative humidity, [%] information than available from standard test-cell RNoAF Royal Norwegian Air Force instrumentation. The engine instrumentation is shown in Fig. 1. RTD Resistance temperature detector The temperatures at stages 1, 3, 4, 6 and 8 were measured T Temperature, [K] using a single resistance temperature detector (RTD) at each U Blade velocity, [m/s] stator row. The entire 15 mm sensor length was immersed into V Relative air velocity, [m/s] the airflow, giving a representative measure of the bulk average w/a Water-to-air ratio, [kgH2O/kg air] temperature. Because the sensors are unshielded, the velocity α Absolute air angle, [deg] error will be significant in the temperature reading. The velocity β Relative air angle, [deg] error was calculated as recommended in AGARD AR-245[15] ρ Air density, [kg/m3] with a recovery factor of 0.65. η Polytropic efficiency The static pressures at stage 5 and at compressor discharge γ Ratio of specific heats were measured at a single point on the circumference. The stage T 2 2 Ψ =cp∆Tt/Utip Stage work coefficient, [(kJ/kg)/(m/s) ] 5 static pressure was measured in the bleed channel. 2 Φ =Ca/Utip Flow coefficient Subscripts a Axial velocity component amb Ambient condition * ISO reference conditions assume an ambient temperature of 288.15 K, a barometric pressure of 101.325 kPa and a relative humidity of 60%. 2 Copyright © 2005 by ASME nozzles used. The droplet size could not be measured in the largest flow rates due to multiple scattering of the laser beam. The measured droplet size is reported in Table 3 together with the droplet size data provided by the manufacturer. In the following, the water nozzles are referred to using the droplet sizes provided by the manufacturer Figure 1 GE J85-13 engine instrumentation Engine throttle and nozzle position were recorded manually at each setting. Relative humidity and ambient temperature were recorded manually throughout the testing and were measured at the same location outside of the test-cell intake. Due to test-cell recirculation, the recorded compressor inlet total temperature (CIT) was slightly higher than the measured ambient temperature. CIT was measured using four sensors located at the engine inlet screen. Figure 2 Water wash manifold with 12 nozzles giving All instruments were calibrated prior to the test program 25 µm droplets at 4.4 l/min and the measurement uncertainties were calculated based on VMD methods given in the ASME Performance Test Codes [16,17]. ENGINE ONLINE WATER WASH EQUIPMENT Accelerated engine deterioration was done through the ingestion of atomized saltwater[18].