Copyright 1989 by ASME. Garrett Multipurpose Small Power Unit (MPSPU) Program Status

Transcription

1 THE AMERICAN SOCIETY OF MECHANICAL ENGINEERS 345 E. 47 St., New York, N.Y GT-172 ]!L The Society Shall not be responsible for statements or opinions advanced in papers or in discussion at meetings of the Society or of its Divisions or Sections, or printed in its publications. Discussion is printed only if tie e paper is published in. an ASME Journal. Papers are available for fifteen months after the meeting Printed in USA. Copyright 1989 by ASME Garrett Multipurpose Small Power Unit (MPSPU) Program Status J. KIDWELL Advanced Technology Engineering Project Garrett Auxiliary Power Division Allied-Signal Aerospace Company ABSTRACT The Garrett Auxiliary Power Division's Multipurpose Small Power unit (MPSPU), Contract DAAJ02-86-C-0006, sponsored by the Aviation Applied Technology Directorate, Ft. Eustis, Virginia, has progressed through detail design and analysis to component and power unit development testing. The MPSPU Advanced Development program is structured to provide advanced technology for current and future United States Army and other Department of Defense auxiliary power unit/secondary power system applications for aircraft, combat vehicles, and mobile tactical shelters. The MPSPU has been designed for low specific fuel consumption, low weight and volume, low acquisition and life cycle costs and high reliability and durability. This paper discusses the design and current developmental status of the Garrett GTP50 MPSPU as reported by Kidwell (1988). INTRODUCTION The Garrett Auxiliary Power Division (GAPD) Multipurpose Small Power Unit (MPSPU), designated GTP50, is a U.S. Army sponsored Advanced Development program (Contract DAAJ02-86-C-0006) to provide the technology base for current and future military aircraft, combat vehicles, and mobile tactial shelter auxiliary power unit/secondary power systems. The GTP50, shown in Fig. 1, is designed to metric units, with low specific fuel consumption, low weight and volume, low design-to-unit production cost goals, low life cycle cost, and high reliability and durability. GAPD has designated the GTP50-1[A] for the baseline MPSPU and the GTP50-1[B] for the modified MPSPU. These power units utilize a demonstrated advanced technology base and are additionally supported by a large research and development investment providing emerging technology on an on-going schedule. 1-- G Fig. 1 GTP50 MPSPU is a self-contained power unit The development of the GTP50-1[A] and GTP50-1[B] will validate the small power unit family concept and will provide the technology from which a Full-Scale Engineering Development (FSED) program can evolve with minimum risk to schedule and cost. The goals of the program are: o Demonstrate significant performance improvements over existing power units in the 50 shp (37.3 kw) class. The GTP50-1[A] will provide a specific fuel consumption (SFC) of lb/hp-hr (0.441 kg/kw-hr) at the 50 shp (37.3 kw) design point. o Validate the concept of deriving a family of power units from a baseline MPSPU with design provisions to increase the continuous power rating with minimum changes in configuration and hardware. Presented at the Gas Turbine and Aeroengine Congress and Exposition June 4-8, 1989 Toronto, Ontario, Canada

2 The GTP50-1[B] will provide 110 shp (82 kw) within the same frame size by changing only seven power module components. Provide reliability and durability improvements over existing power units. The GTP50-1[A] and -1[B] are designed for mean time between removals (MTBR) of 3725 hours at maturity, a design life of over 6,000 hours and a low-cycle fatigue (LCF) life of over 15,000 cycles. Achieve a reduced design-to-unit production cost (DTUPC) compared with existing power units. The GTP50, has a DTUPC goal of $15,500. Current and future U.S. Army applications are projected to include auxiliary power unit/secondary power systems for the Light Helicopter Family, combat Armored Family of Vehicles, and Mobile Tactical Shelters. Gas Turbine Auxiliary Power Unit design draft specifications for these potential applications were utilized for design guidance to define major specification operational and performance drivers, in addition to the specific MPSPU requirements. POWER UNIT DESIGN The GTP50-1[A] and -1[B] MPSPU consist of a gas turbine power module, a removable inlet protection system and gearbox to provide a 12,000 rpm output pad, and to drive the controls and accessories necessary for self-sustaining operation. The GTP50-1[A] and -1[B] are a carefully engineered extension of the high technology GAPD commercial/military development and production base, the power module is a front drive, constant speed, single-shaft design featuring a single-stage centrifugal compressor, tangential can combustor, and a single-stage radial turbine. Gearbox-mounted accessories include the starter, with the capability for accepting either an electric or hydraulic starter being fitted to the same pad, the lubrication system pump, hydromechanical fuel control for fuel metering, and the inlet protection system scavenge and oil cooler fan. A full authority digital electronic control will be used to control power unit start and operation, provide overspeed and over-temperature protection, a life usage algorithm, and capability for line replaceable unit (LRU) fault detection and isolation. The baseline GTP50-1[A] and the modified -1[B] power units operate at the same speed and utilize a high degree of common hardware. The outline dimensions are the same for both units as shown in Fig. 2. GTP50-1[A] and -1[B] power demands and these are associated with the aerodynamic flow path. The same gearbox, controls and accessories, bearings and shafting are utilized on both power units. As air enters the radial inflow inertial inlet particle separator (IPS), Fig. 3, it is accelerated along a torturous highly inclined inlet ramp, deflecting incoming sand particles into the scavenge duct, which is above the hidden flow splitter. The scavenge vanes are located aft of the splitter lip to induce swirl into the scroll collector where a geardriven scavenge blower removes the particles and exhausts them. The cleaned inlet air proceeds through a short axial duct and is compressed to a pressure ratio of 5.5:1 in the single-stage titanium centrifugal compressor, operating at 114,915 rpm, with tandem vane diffuser. The flow is dumped at the diffuser exit with minimum swirl and exit Mach number, eliminating the need for a deswirl section. The flow then travels into an asymmetrical duct and is introduced to the tangential can combustor at approximately 500F (260C). The stacked ring, film cooled, non-coated HA -230 can combustion system will accept any of the four required fuels (JP-4, JP - 5, DF- 1 and DF-2) using a single piloted airblast fuel nozzle. The flow is raised to 1800F (982C) by combustion and delivered to the single-stage radial turbine through a NOMINALH NOMINA L ]NOMINAL df GBO NOMINAL ALL DIMENSIONS IN INCHES Fig. 2 MPSPU design is compact

3 G Fig. 3 MPSPU power module and inlet protection system are integrated low pressure loss combustor scroll assembly. Flow enters the cast MAR - M 247 coated stator/shroud and is directed onto the hot isostatic pressing (HIP) bonded dual alloy (cast MAR - M 247GX blading/power metal astroloy hub) radial turbine rotor where it is expanded, extracting work and then exhausted. Within the constraints of a single frame size, maximum hardware commonality is retained in the conversion of the GTP50-1[A] to the GTP50-1[B]. A total of seven components are traded to achieve better than twice the horsepower rating. Of these seven, two are variations of the same part number. The listing below describes the extent of hardware modification: ocompressor Shroud/Diffuser - Inlet shroud diameter, vane geometry, and height increases. ocompressor Rotor - New blading; adjacent component mating surfaces are identical. ocombustor Liner - Identical Stacked ring liner and cooling skirts with differing hole location and size, (minimum modifications). oturbine Stator/Shroud - New sidewall geometry for higher flow, vane shape identical. oturbine Rotor - New blading geometry. oturbine Exhaust Diffuser - Turbine exhaust diffuser modified. Inlet Particle Separator - Mating compressor inlet shroud contour is increased. Through judicial trade-off analyses, the method chosen for growth of the GTP50-1[A] to meet the higher horsepower requirements of the GTP50-1[B] is the flow scaling approach. Using this technique, GAPD has established a common shaft speed (114,915 rpm) and turbine rotor inlet temperature (TRIT) of 1800F (982C) which provides for maximum hardware commonality within the same frame size for the derivative family concept of small power units in the 50 to 110 shp (37.3 to 82 kw) class. Alternate methods of power increase were studied, including speed, temperature, variable geometry, and scaling; all of which compromise the commonality, cost, weight, or performance goals of the power units. High reliability and maintainability concepts have been designed into the MPSPU. In addition, GAPD recognizes Manpower and Personnel Integration (MANPRINT) requirements. GAPD has structured the overall program to identify and define improvements in these three major areas. GAPD has designed the MPSPU to be compatible with the U.S. Army two-level on-condition maintenance concept. The design provides the U.S. Army with a power unit that can be supported in the field with a standard U.S. Army tool kit, at the user level, with a mean-time-to-repair by one person of hours (14.7 minutes). The reliability goals for the GTP50 have also been established and the analysis planned for the program will be iterated to further improve the

4 L following goals for mean-time-between-removal (MTBR) and mean - time - between - unscheduled - maintenance - action (MTBUMA): U.S. Army GTP50 Reliability Characteristic Requirements Goals MTBR 3000 hrs 3725 hrs Start Reliability with electric starter None with hydraulic starter None MTBUMA with electric starter 160 hrs 717 hrs with hydraulic starter 160 hrs 731 hrs Design-to-unit production cost (DTUPC) methodologies have been involved since inception of the GTP50 to achieve reduced cost compared with existing units in this power class. The GTP50 DTUPC is projected to provide a major improvement over currently marketed small power units. POWER UNIT PERFORMANCE Aerodynamic performance of the GTP50-1[A] was optimized based on iterative analyses concerning the thermodynamic components: varying shaft speed, cycle pressure ratio, and airflow for the compressor; and airflow, turbine inlet temperature (TIT), and turbine rotor tip speed for the turbine. Compressor design point efficiency was predicted over a range of pressure ratios and shaft speeds for an airflow rate of 0.5 lb/sec (0.23 kg/s). The turbine inlet temperature selection demands recognition of life characteristics for the rotor as well as adjacent structures. Fig. 4 shows the allowable turbine inlet temperature versus turbine rotor tip speed for selected turbine materials consistent with the life requirements. LL w 0 W N J 1 z z GB TIP SPEED, M/S TIP SPEED, FT/SEC U 1000 H 975 z 950 z Fig. 5 shows the result of designing a family of 50 shp (37.3 kw) power units at the indicated cycle pressure ratios and turbine inlet temperatures. Minimum SFC occurs between cycle pressure ratios of 5.5:1 and 5.7:1, depending on the turbine inlet temperature. I 0.7 I J vl 0.7 U GBB Fig. 5 MPSPU performance was optimized As stated, the turbine inlet temperature selection requires that the power unit life requirement (6,000 hour design life, 15,000 starts) be addressed. The selected dual alloy turbine approach affords the designer higher operating speeds and temperature to optimize engine performance. A turbine rotor inlet temperature of 1800F (982C) was chosen as the maximum inlet temperature which addressed rotor life, peformance, and adjacent structure life simultaneously. Based on the cycle analyses, the GTP50-1[A] incorporates a cycle design pressure ratio of 5.5:1, a turbine inlet temperature of 1800F (982C), and a turbine tip speed of 2270 ft/sec (692 m/s) T s LL The GTP50-1[B] is derived by maintaining a common envelope, shaft speed, and turbine inlet temperature, and by increasing the flow capacity of the power unit. The aerodynamic flow path components are redesigned for flow approximately twice that of the GTP50-1[A] design. This approach provides hardware commonality and maximum power spectrum while maintaining low fuel consumption for the GTP50 family. Fig. 6 depicts the GTP50 family of power units I m 0.72 J GB OUTPUT, KW ,915 RPM OUTPUT, HORSEPOWER I V Fig. 4 Optimum tip speed and life are maintained Fig. 6 GTP50 family spans a broad horsepower range 4

5 Table 1 identifies the design point cycle parameters for the GTP50-1[A] and -1[B]. Table 1 GTP50-1[A] and GTP50-1[B] design points have been defined GTP50-1[A] Baseline Power Unit Parameter Sea Level Standard 59F Day GTP50-1[B] Modified Power Unit Core total pressure loss, was 0.68-percent compared to a design value of 0.58 percent. Scavenge flow pressure loss was 3.7 percent compared to a design value of 2.9 percent. Initial tests identified a flow separation zone adjacent to the splitter lip on the hub flowpath. Modifications were made to eliminate this separation zone as shown in Fig. 8. Dramatic improvements were noted for the AC coarse and C- Spec separation efficiencies. Table 2 shows the results of several test configurations. Additional testing is planned to raise C - Spec efficiently consistent with the program goals. 114,915 Speed, RPM 114, :1 Pressure Ratio 5.5: Compressor Corr Airflow, lb/sec 1800(982) Turbine Inlet Temp, F(C) 1800(982) 50 Output Horsepower, HP SFC, lb/hp-hr COMPONENT DEVELOPMENT Component development activities in support of the MPSPU are divided into three areas of technology: inlet particle separation, aerothermodynamics (compressor, combustor, turbine) and mechanical integrity. Inlet particle separator testing has been initiated with sand separation, flow profile and blower tests. All aerothermodynamic components have completed initial performance tests for both baseline and modified development. Mechanical integrity of the duplex bearing, roller bearing, shaft seals, gearbox and rotor dynamics has been established. The following paragraphs describe those activities. Inlet Particle Separator Inlet particle separator (Fig. 7) initial tests concentrated on core flow pressure loss, scavenge flow pressure loss and sand separation. The test rig replicated the MPSPU inlet with power unit hardware being used for the splitter lip, scavenge vanes, and scroll. The hub and shroud flowpath is identical in aerodynamic contour but has been segmented at selected points to allow for easy test configuration changes to enhance separation and/or reduce inlet losses Fig. 8 Inlet particle separator flowpath modifications were made to all surfaces Table 2 Inlet particle test summary i GB Fig. 7 Inlet particle separator includes shroud splitter and scroll hardware Configuration Separation Efficiency AC Splitter Hub Shroud Coarse, % C-Spec, % Baseline Baseline Baseline Baseline 3 Baseline Baseline Baseline Baseline Baseline Baseline Baseline Coated Baseline Baseline

6 Compressor The small single-stage backward-swept centrifugal impeller for the GTP50-1[A] has 16 full blades and 16 splitter blades followed by a tandem vane diffuser. Testing has been completed for the as-designed compressor stage components. Impeller exit survey testing was also conducted to define the exit flow field parameters including pressure, temperature, and angle from shroud-to-hub across the passage. Fig. 9 shows the impeller and diffuser during partial build of the test rig DESIGN GOAL P GBB GCB A We Fig. 9 GTP50-1[A] impeller and diffuser hardware is ready for test Fig. 10 GTP50-1[A] compressor rig test are encouraging Initial test results for the GTP50-1[A] compressor indicate the following: Q DESIGN GOAL o Impeller flow was as-designed o Impeller work was four-percent higher than design values o Diffuser loss coefficient is significantly higher than anticipated due to increased impeller work. o Stage pressure ratio work efficiency are below design intent Test data review indicated a higher than anticipated blockage at the diffuser leading edge resulting in flow acceleration at the diffuser throat. Existing diffuser rework (leading edge cutback) was accomplished to improve the high blockage. Testing was resumed and the results are shown in Fig. 10. P Based on the results of Test 1 and Test 2 the diffuser is currently being redesigned to gain a more favorable match and improve stage efficiency and pressure ratio while maintaining design flow. Testing also has been accomplished for the GTP50-1[B] compressor hardware. Fig. 11 presents the test results for the GTP50-1[B]. Again an impeller diffuser mismatch exists for the GTP50-1[B]. As evidenced from the data; the design pressures ratio was achieved, design flow is 1.5-percent low and efficiency is below design intent. Design recommendations are underway for the GTP50-1[B] to rematch the impeller and diffuser. GC A Fig. 11 GTP50-1[B] compressor rig test data indicates goals are achievable we 6

7 a Combustor Development testing of the GTP50-1[A] and GTP50-1[B] combustion systems has progressed. The small high heat release design represents a significant challenge, in that the axial length is extremely small (Fig. 12). Test results to date have been very encouraging on both the GTP50-1[A] and GTP50 - [B]. Table 3 presents the combustion system development current status. As shown, for the GTP50-1[A] wall temperature, pressure drop, lean blowout fuel/air ratio and multifuel demonstration goals have been achieved. Pattern factor has been significantly reduced from early values of 0.24 to 0.15 with a goal of Development testing continues on the GTP50-1[A]. Test data for the GTP50-1[B] shows an impressive pattern factor below design goal. The aggressive 0.14 goal was demonstrated during the development program, however, a higher than desired wall temperature remained. In efforts to reduce the wall temperature, an asymmetric primary and dilution orifices design was tested with positive results. As shown in Table 3 additional work is required and is ongoing to reduce the wall temperatures below 150OF for life considerations. Turbine Initial cold air turbine rig testing has been completed for both the GTP50-1[A] and GTP50-1[B]. Test hardware includes power unit scroll, turbine plenum, turbine rotor (Fig. 13) and exhaust diffuser. The rig nozzle fully replicates the power unit flowpath hardware. Test results, shown in Fig. 14, indicate the designs to be at or above design intent. Testing was conducted over a range of speeds and pressure ratios. The GTP50-1[A] turbine stage efficiency is within 0.5- percent of design intent at operating conditions. The scroll losses are as predicted ensuring uniform flow at the stator inlet. The GTP50-1[B] turbine efficiency exceeds design goals by approximately 1-percent. Again scroll losses were as predicted. Exhaust diffuser recovery is close to prediction. Fig. 12 GTP50-1[A] combustor can and scroll is a small high heat release design Table 3 Combustor test results POWER UNIT DEVELOPMENT GTP50 power unit testing was initiated on schedule in December 1987, eighteen months after contract award. Initial test runs were conducted using an air turbine starter to motor the power unit to full operating speed with nonvitiated air. This approach allowed observation of critical running parameters such as compressor axial clearances, turbine radial and axial clearances and shaft dynamics. Following successful mechanical checkout tests the unit was configured for hot runs as shown in Fig. 15. Initial lightoff and accleleration runs were accomplished during Build 1 with GTP50-1[A] GTP50-[B] Design Goal Test Design Goal Test Design Point Conditions Inlet Pressure, psia " Inlet Temp, F Airflow, lb/sec Fuel Flow, lb/hr Outlet Temp, F Program Objectives Combustor Efficiency, % 99.7 TBD 99.5 TBD Comb Pressure Loss, % Pattern Factor Comb Skin Temp, Max F Smoke Below Visibility TBD Below Visibility TBD Threshold Threshold Multifuel Capability Diesel TBD Diesel Yes Lean Blow-Out F/A < <0.003

8 GB Fig. 13 GTP50-1[A] turbine rotor operates at optimum tip speed Fig. 15 Power unit testing is underway Table 4 Power unit operational time >- I..) 2 w U W Electric Hydraulic Total Time Power Unit Starts* Starts* hrs S/N A S/N B S/N C Totals *Starts to 100 percent speed, idle conditions GBS PRESSURE RATIO J -1[B] DESIGN POINT 0-1[A] DESIGN POINT Fig. 14 Cold turbine test results are very encouraging self-sustaining operation achieved in January Testing was conducted using the full authority digital electronic control and accessories. All components functioned as required. Development testing has progressed through mechanical integrity testing on two GTP50-1[A] power units and one GTP50-1[B] power unit. Preparations are now underway to establish initial performance parameters consistent with the component development. Table 4 summarizes the power unit operational time and logged starts. In addition to mechanical development testing, static bench leakage tests, heat rejection tests, and thermal survey tests have been conducted. Results of these tests indicate the design philosophy to be sound. SUMMARY Significant development effort has been completed for the Garrett MPSPU program. Component testing supports the design intent with additional work planned. Power unit testing has progressed through initial mechanical integrity and development testing. Performance tests are currently underway with initial as designed components prior to final component evaluation. Both electric and hydraulic starting systems have been demonstrated with the fully functional electronic control unit. Development activities will proceed through final component test and power unit demonstration of the program goals. REFERENCE Kidwell J.R., 1988, "Garrett Multipurpose Small Power Unit (MPSPU) Program Status", Aerospace Technology Conference and Exposition, Society of Automotive Engineers. 8