Document
RESEARCH TURBINE FOR HIGH-TEMPERATURE
CORE ENGINE APPLICATION
I - Cold-Air Overall Performance
of Solid Scaled Turbine
by Edward M. Szanca, Hurold J. Schum, arzd Glen M. Hot2
Lewis Research Center
Clewland, Okio 44135
NATIONAL AERONAUTICS AND SPACE ADMINISTRATION WASHIHGTOW 0. C. FEBRUARY 1974 RESEARCH TURBINE FOR HIGH-TEMPERATURE CORE ENGINE APPLICATION
I - COLD-AIR OVER4LL PERFORMANCE OF SOLID SCALED TURBINE
by Edward M . Szanca, Harold J. Schum, and Glen M. Hotz
Lewis Research Center SUMMARY The cold-air performance of a half-scale model of a 50.8-centimeter (20-in. ) tur- bine suitable for high- temperature lvcorell engine a2plication is presented herein. The physical characteristic features of this turbine are i C W aspect ratio, thick leading an3 trailing edges, low solidity, and large relative rotor radial clearances.
The performance parameters presentc 3 herein are expressed in equivalent terms for an uncooled version (solid vanes and blades) of t h = aforementioned turbine.
Two rotor configurations were designed and tested. The first had untwisted rotor Mades with the same aerodynamic profile from ~ - A O to tip ta facilitate the fabrication of cooled blade:; utilizing complex cooling schemes. The second had twisted rotor blades designed for free-vortex flow and simple radial quilibrium. The second turbine w a s tested to determine the performance penalty incurred by using untwisted blades. Both turbines were tested over a range of speeds from 40 to 110 percent of design and inlet- to writ-total pressure r a t h frqm about 1.4 to 2 8. Over the entire range of speeds and pressure ratios tested, the efficiency of the tu?Xze with the twisted rotor blades was greater than the efficiency of the turbine with the untwisted rotor blades. At the design specific work output of 39.572 joules per gram (17.00 Btu/lbm) which occurred at a pres- sure n t i o of 1.817. the t c f r l efficiency of the tuiL,ne with untwisted rotor blades was87.1 percent. A t this pressure ratio the efficiency of the turbine with twisted rotor blades was 88.0 percent. These results compare favorably with the design efficiency of 87.0 percent.
INTRODUCTION Many future engines will ne of the turbofan variety having high bypass ratios and high pressure and temperature "care" engine turbines. These turbines are characterized by their relatively small-sizeci Slading (high hub- to tip-radiuc ratio) and low aspect ratio.
both of which a r e considcred detrimental to turbine cfficiency.
Research studies are presently being conducted at the Lewis Research Center to build a 50.8-centimeter- (20-in. - ) diameter. single-stage turbine to operate at stoichio- metric comhstion (2478 K; 4000' F) and a turbine inlet pressure of 413.7 N: cm abs (600 psia). At this temperature level, significant amounts of coolant flow a r e reyuired.
The turbine design w i l l incorporate full-film cooled vanes. blades. and end walls. The introc-tction of these large amounts of cooling air into the main o r primary f l o w stream can result in a significant deleterious effect on the turbine aerodynamic performance.
Prior to testing at the extremely high inlet-air conditions, candidate vane and blade designs will be tested in detail at much reduced inlet conditions in order to determine the aerodynamic effect of coolant air. To establish 5 basis of comparism, then, a solid (uncooled) version of the ?'core?? turbine w a s fabricated. The purpose of the subject report is to present the design ard test results of this turbine.
The turbine investigated was a half-scale model of the design turbine, and llad a rotor tip diameter of 2 5 . 4 centi- meters (10 in. ), with blade heights of 1.905 centimeters (0.75 in. 1. with no radial-axial flare. The rotor blades had a constant Rection profile from hub to tiF and no twist. This type of design facilitates fabrication of complex blades and inserts. U s e of untwisted rotor blading probab!y results in a turbine performance penalty, however. Accordingly, a sectjnd rotor was fabricated, d s i g n e d for free-vcrtex flow and simple radial equilib- rium, resulting in blade twist from hub to tip. This rotor was also tested with the same stator as used for the untwisted rotor, in order to determine the extent of the perform- ance penalty.
Both the untwisted and the twisted turbines were tested at an inlet pressure 9f 17.237 N/cm abs (25.0 psia) and an inlet temperature of approximately 306 K (550' R) and over a range of pressure ratio and speed. Tnis report presents performance ir, terms of equivalent mass flow, equivalent torque, equivalent specific work output, equiv- alent speed, outlet flow angie, and efficiency.
SY IMBOLS area, cm2; f t 2 A b vane or blade height, cm; in.
C chord length, cm; in.
D diameter, cm; in.
force-mass conversion ccnstant, 1.0 (32.174 ft/sec 1 g Ah specific work, J/g; Btu/lbm i rotor inlet relative incidence angle, deg mechanical equivalmt of heat. 1.0 (778.16 ft-lb/atu) J N turbine shaft speed. rad/sec: rpm pressure, N/cm abs: psia P R g i s -onstant. 287 J/(kg) (KI: 53.34 ft-lb/(lb) ( O R ) r radii: i, cm; in.
pitch cm; in.
S T absolute temperature, K; OR U blade velocity, m/sec; ft/sec
v absolute gas velocity, mjsec; it/sec
W relative gas velocity, m/sec; ft/sec W mass flow, kg/sec; lbm,/sec CY absolute gas f l o w angle measured from the aX;al direction, deg ratio of specific heats Y ratio of inlet total pressure to U.S. standard sea-level pressure, pb/p* c function of Y used in relating parameters b those using air inlet conditions at U.S.
standard sea-level conditions, Y total efficiency (based o r 1 inlet-total to exit-total pressure ratio) r7 squared ratio of critical velocity at turbine inlet io critical velocity at U.S. standard 'cr sea-level air (.cr/vTr)2
-
torque, N-m; ft-lb Subscripts : c r condition corresponding to Mach number of unity b blade h hub section m mean section t tip section u tangential compcnsnt V vane X axial component station at turbine inlet (fig. 8) 1 station a t stator exic (fig. 8) 2 station at turbine exit (fig. 8) Superscripts : ? absolute total state * U.S. E:andard sea-level conditions (temperature equal to 288.15 K (518.7' R).
pressdre equal to 10.13 N/cm abs ( 1 4 . 7 psia)) TURBINE DESCRIPTION Design parameters for an advanced high-temperature single-stage air-cooled axial- flow "core" engine turbine are summarized in table I. Also shown are the corresponding TABLE I. - TURBME DESIGN OPERATING VALUES Hot engine conditions Air-equivalent conditions Performance parameter (ASTM-A- 1,'Air = 0.0435) 50. P (20.0) 25.4 (10.0) Tip diameter. Dt. cm (in. ) 2 8 8 . 2 ~ m . 7) Inlet total temperature, ~ b , K (OR) 2200 (3960) Inlet total pressure, pb. N i c m abs (psia) 386.1 (560.0) 10.13 ( 1 4 . 7 ) 6 3 . 8 2 (140.72) 1.207 (2.660) Mass ilow , w, kg/sec (lbm/sec) 12 388 Turbine rotative speed. N , rpm 16 687 287.25 (123 4) 39. 572 (17.00) Specific work oGtput. Ah', J/g (Btu/lbm) 152.4 (500.0) Mean blade speed, Um, m / s e r (ftisec) 410.6 (1347) ---- _-------- Wet- to exit-total pressure ratio, pb/p* I . 818 ------------- 0.87 Total efficiency, ' 1 . percent air-equivalent design parameters for the half-scale uncooled model tested herein. The equivalent conditions were derived assuming no coolant-air effects.
The test-turbine stator was fabricated with untwisted vanes of constant mean-section profile, ignoring the relatively small amount of twist (about 3') from hub to tip if designed for free-vortex conditions. The rotor blades were also of constant profile from hub to tip, with no twist. These blaciing profiles were designed to facilitate fabrication of complex internal cooling- air-distribution inserts fQr future cooled turbine tests. The subject investigation, however, was conducted usi ig solid (uncooled) blades.
The velocity diagram evolved to meet the Gosign aerodynamic requirements is shown
b
f (a) Hub section; radius ratio rhht = 0.850.
1 (VIV,,)O = 0.231
(b) hkan section; radius ratio rmlrt = 0.925.
(VIVcr)0 = 0.231 .)1 = 9.329 ,iJ Tip section; radius ratio r$rt = 1.m.
figure 1. - Turbine design velocity diagram.
in figure 1. All quantities represent the f r e e stream uniform f l o w conditions. Pertinent test turbine geometry are presented in table X I .
A second tu: bine, designed for the same requirements, was built and testdd using the same vanes but employing twisted rotor blades as dictated from free-vortex and radial equilibrium considerations. All vanes and blades for the two turbine configurations are characterized by blunt leading and trailing edges, low aspect ratio, and high thickness to Because of the high rotor blade hub t o tip ratio, the radial clearance (blade chord ratios.
tip to casing) of 0.0432 centimeter (0.017 in. ) represe about 2.3 percent of the blade height. All of these geometric factors are considered detrimental to high turbine effi- TABLE I I . - TEST TURBINE GEOMETRY
I StatL r
Mean diameter, D , , , , cm (in.) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23.495 (9.25) Vane height, bv, cm (in.) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.905 (0.75) Axial chord, cxv. cm (in.) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.905 (0.75) Axial solidity, (cxdsv)m . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.929 Aspect ratio, b/cx . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1.000 Number of vaneti. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Leading edge rMius, cm (in.). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.254 (0.100) Trailing edge radius, cm (in. ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.0445 (0.0175) Rotor ~~ ~ ~~ ~ Mean diameter, Dmb,cm (in.) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23.495 (9.25) Blade height, l+,, cm (in.). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.905 (0.75) Axial chord, cxb, cm (in. 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.715 (0.675) Axial solidity, (cxH's,,)m . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.652 Aspect ratio, b,,/cxb. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.111 Nnmbs- o f blades . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Tip clearance, c m (in. ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.043 (0.017) Leading edge radius, cm (in.). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.1494 (0.0588) Trailing edge radius, cm (in.) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.0445 (0.0175) ciency. The overall design efficiency of the turbine based on inlet- to exit-total pressure ratio was determined to be 87 percent. This efficiency was based on the value cvf mean speed-work parameter U,/gJ Ah of 0.587 and corrections found in the literature to ac- count for the low aspect ratio, size effect due to scaling, and radial clearance. No cor- rection was made for the incidence angle caused by the untwiste.3 design since reference 1 indicates no significant loss w i t h ircidence from approximately - 10' to approximately +loo (depending on the amount of reaction). For the subject turbine with untwisted rotor blades, the incidence angles were calculated to be +7.4' at the hub and -10.4' at the tip which a r e within the tolerance limits of reference 1.
Vane and rotor blade design surface pressure distributions at the mean section a r e shown in figures 2 and 3, respectively, calculated by the method of reference 2. Fig- u r e 2 shows a decreasing pressure drop on the suction surface of the vane from the lead- ing edge to about 40 percent of the axial chord. From this point on the pressure ratio re- mains nearly constant up to about 85 percent of the chord length, at which point the pres- sure increases to the stator exit. In t e r m s of velocity along the suction surface, this can be interpreted a8 an accelerating flow up to a value near the stator leaving velocity with some diffusion over the final portion of the suction surface. The low diffusion of the stator ia considered favorable for high stator efficiency.
The pressure dlrstribution along the pressure surface o f the vane (fig. 2) i R typical, with gradually decreasing pressure from leading edge to trailbig edge.
Fraction ol axial chord length Figure 2. - Design surface static pressure distribution at stator mean section.
Figure 3. - Design surface static pressure distribution at rotor mean section.
Figure 3 shows a low pressure spike on the suction surface of the rotor blade near the leading edge. The sudden drop in pressure is caused by the accelerating flow around the leading edge of the blade. A s the fluid passes around this circular leading edge, an abrupt change in curvature results in a deceleration of the flow and the resultant high pressure. After the leading edge the pressure along the suction surface drops to a pres- sure approximately equal to the rotor exit static pressure.
Along the pressure surface a pressure spA e OCCUTS for the same reasons as dis- cussed for thz suction surface. The spike, however, occurs further from the leading edge as measured along the axial chord than that for the suction surface. I h e result was an tthour-gl w?t pressure distribution near the forward portion of the blade.
AlthoJgh it was not suspected that the low pressure spikes contributed to any appre- ciable loss, a more gradual transition in curvature from leading edge to blade surface TABLE 111 - STATOR \'ASK COORDINATFS Mean Section Orientation angle, ,.-, deg 4 4 % X
- - -
cm in. cm 0 I. 2540 1. I00 ----- .064 .025 ----- -127 ,050 - - - - _ .I91 .075 ,2210 ----- ,254 .loo .2390 - _ _ _ _ .318 .I25 .381 .l50 .0318 ,445 .I75 .0597 .2740 .OB26 .508 .200 .2810 ,572 .225 .lo29 .2860 .635 ,250 .I207 ,2890 .699 .275 ,1334 ,2900 ,0575 ,7379 .762 . IO0 .1461 .2905 .06W ,7315 .E89 .350 ,1651 ,2880 .0710 .I163 I. 016 .2820 .40G .le03 ,0740 .2720 ,6909 I. 143 .450 .I880 I . 270 500 .1918 .2605 I. 397 ,1905 .2475 .550 ,2340 I. 524 ,600 .1867 ,2185 I. 651 .650 ,1765 I . 778 ,700 .16;1 ,2030 I . 905 ,750 ,1486 ,1860 1.032 . aoo .1308 ,1675 1 . 159 .650 ,1105 ,1470 2.286 .goo .0902 .1255 1.413 .950 ,0660 ,1020 1.540 1.000 ,0406 ,0770 2.667 1.050 .0127 .0510 1 . 7 7 6 1.093 ,0445 ,0175
- -
stat ng axls cwrdinates cm I in. I cm I in.
TABLE lV. - ROTOR BLADP COORI INATES X ;Iub section Mean section Tip section L Orientatiw angle, cp, deg - 3 0 ' 1 3 '
I 2 4 '
I
I YL
-
cm , .in. i cm [ in.
c m in. cm
-
- -
0 0.1451 I0.05G8 1.1494 0.0588 0.1494 0 1.0588 0.1494 0.0588 0.1434 0.0588 0.1494 1.0588 ------ - - _ - _ - 025 _ _ _ _ _ _ I - - - _ _ _ - - - - .064 .1275 ,3023 .1190 .3239 .3102 .1300 _ - - - - - - - - - - - - - - - _ - -_---- .127 ,3874 .1525 .050 .4318 .4229 .1665 ------ _ - - - - - ----. - _ _ _ _ _ - .191 .zoo0 .4483 .1765 ,075 .5080 .4877 .1920 .254 .2215 -0419 .0165 ,4978 ,1960 .loo .5626 .0165 .5r4E ,0419 ,2145 .381 .1321 .0495 .5664 ,2330 .150 .6553 .6223 . "250 .2580 .a520 ,1257 .508 ,6007 ,2365 .2805 . Z O O 7 .0710 .200 .7125 .0790 ,6655 . ,620 .1803 .635 .2920 .2515 ,0860 .6071 .2390 .250 .7417 .0990 .6795 ,2675 .2184 .762 .2930 .2807 ,0953 ,5931 ,2335 .300 .7442 ,1105 ,6693 .2635 ,2426 .E89 .2835 .2921 .0970 .5664 2230 ,350 .7201 .1150 ,6414 2525 2464 l.ul6 .2670 .2C96 .0940 .5283 .2080 ,400 .6782 .1140 .6007 .2365 .2388 1.143 .2455 .2718 .OB70 .4801 .1890 .450 .lo70 .5486 .2210 .6236 .2160 1.270 .2388 ,0765 .4305 .1695 . 500 . C740 .1943 .2205 .5601 .4877 .1920 .553 1.397 .1925 .1956 .0620 .3747 ,1475 .4890 .0770 .4191 ,1650 .1575 1.524 .1585 .1461 ,0460 .3099 .1273 .600 .4026 .0575 .3454 1360 .1168 .650 1.651 .1210 ,0927 .02W .2375 ,0935 ,3073 .0365 ,2629 .I035 .0737 .0800 1 . 7 7 8 .0356 .0110 .1575 ,0620 .IO0 .2032 .0140 ,1715 . E 7 5 .0279 .----- _ _ _ - _ _ _----- ------ _ _ _ - - _ 1.935 .0445 .750 .0175 .0445 .0175 .----- - - - _ - - - - - _ - - ------ . _ _ _ _ _ 1.908 .0175 ,0445 ,0175 .751 .0445 _ _ _ - - _ _ _ - - _ _ - - - - - - _ _ _ _ _ _ - - - - - - - - - - - _ - - - - - - 1.975 0175 .758 .044E
-
Stacking axifi coordinates x X X Y
Y I Y I
I
cm in. cm in. cm in. cm in. cm in. cm in.
-
0.810 0.319 0.386 0,152 0 . 7 7 2 0,304 0.361 0.142 0.742 0.292 0.328 0.129 would improve the blade loading profile.
Vane profile coordinates are presented in table In; rotor blade coordinates a t the hub, mean, and tip sections for the twisted configuration a r e given in table Tv. For the untwisted rotor blade configuration, the mean section profile was mzintained constant from hub to tip. A l l dimensions are f o r the half-scale test model. A photograph of the stator is presented in figure 4. Similarly, photographs of the twicted and the untwis d rotor assemblies a r e shown in figures 5(a) and (b), respectively.
APPARATUS The apparatus consisted of the turbine as described ir. the trtxcaing section; P speed- decreasing gear box, a cradled dynamometer, .J an inlet md erhwdt piping system with flow cwtrols. A 223.7-kilowatt (300-hp) elrckiz dynamometer was used to absorb the pover autput of the turbine, cc-itri/l it6 speed, and measure torque output. Both the gear box and the d y n a m o m c k were supported on hydrostatic bearings to minimize beiring friction losses. The arrangement of the experimental equipment is shovn schematically in figure 8.
Pressure Thin plate regulator orifice High pressure air T m
1 - _
w71
To burst - disk
T Turbine inlet pressure
control valve r I I Dynamomettr TurMne gtlt
- Gear I
c- pressure
- h -
control MIW--, \
I i I I
Fiaure 6. - Test Installation schematic.
Figure E . - Schematic d turbin test section.
of the stator-blade trsiling edges, and in the center of the projected free stream channel.
The turbine outlet meaaclrirg station (station 2) was located about m e blade chord downstream of the rotor. The instrumentation include15 sta+l.c pressure, total pressure.
total temperature, and flow angle. The static pressure w a s measured with eight wall taps located as described for station 0. The outlet flow angle was measured with an angle sen- sirive probe and a self-alining probe actuator. Ah.0 mounted on ;he angle rneasurixig probe were provisions for determining total pressure. The angle and total pressure sensing elements were located at the mevl blade height.
Turbine torque was transmitted to a commercial strain-gage load cell through is torque a r m attached to the dynamometer stator. The rotational speed W ~ S detected by a magnetic pickup and shaft- mounted gear.
All pressures were memured w i t h calibrated electrical transducers. A 200- rnnel data acquisitinn system w a s used to rnehsure and record the electrical signals from the transducers.
PROCEDVRE Turbine performance with eac3 rotoi w a s obtained at nominal turbine inlet conditions of 17.237 N/cm abs (25.0 psia) pressure and ambient temperature (approximatsly 3% K (550' R ) ) . Data wcre obtained over a range of inlet- to exit-total pressure ratios from 1 - 4 to 2 . 8 and m e r sn equivalent speed range from 40 to 110 percent of design.
Torque calibrations were obtained before aid after eath performance xun. The cal- ibrations were cbtained with the dynamometer in the motoring mcde, rotating the turbine at a speed of approximately 4500 rpm. Thus, moFt cf the effects of bearing and seal fi-ic- tion were accounted in the calibratior!. Turbine windage !osses were minimized cixricg the calibrations by evacuating the air f m m the turbine housing.
The turbine efficiency w a s rated on the basis of inlet- to exit-total pressare ratio.
Inlet and outlet total pressure were calculated from mass flow. static pressure. total temperature, and f l o w angie from the following equation: At the S e t , the flow angle Q w a s assumed b be axial (cos Q~ = 1). A t the exit, the total temperature was complted from measured values of torque, mass floa and inlet total temperature.
The results of this investigation are presented in three parts. The first section covers the turbine p e r L o r w i c e obt;air,ed using a rolur with untwisted blades. The second section covers the performance obtained with twisted rotor blades. The third section is a comparison between the performance of the two turbine configurr tims to determine the effect of twist irr the rotor blades. A l l data are shown in t e r m s of epivalent air values.
The experimental results include overall performance in t e r m s of equivalent mass flow, quivalect specific work, and efficiency A l s o included a r e the variations in turbine exit fiow angle and static pressure within the turbine.
Turbine Performance W i t h Untwisted Pator Blades
Overall performa-,cc. - The overall performaace map for the turbine with untwisted
rotor blades is show? in figtire 9 in t e r m s of specific work outmt Ah/ecr and a mass flow-speed parameter E wN/b for lines of constant total pressure ratio and equivalent rotor speed. Contours of constant values of efficiency based on the tctrii pressure ratio across the turbine are also included.
The turbine design specific work output of 39.572 joules oer gram (17.00 Btu/lbm) was obtained at a total pressure ratio of I . 817, and cortesponae to an efficiency of 8 7 . 1 percent. This efficiency compares well to the 87 percent predicted efficimcy. The map Mas, flow-speed parameter, ~ w N i 6 , (Iq)(rad)/secL I I I I I I I I I I I I j . 2 1.4 1.6 1.8 2 0 22 24 2.6 2.8 3.0 3.2 3.4 3.6 ,.&!Id Mass flow-speed parameter, mNfb, (IbmHrpin,/sec 9. - Overall turbice performance map for turbine with untwisted rdor blades.
Figure also shows that the design specific work is well below limiting loading indicating a con- servative design.
Over the range of pressure ratios and speeds investigated, the turbine efficiency varied from abuut 65 percent near the 40 percent speed line, to 88 percent near the 110 percent speed line.
Stator mass flow calibration. - A mass flow calibration of the stator was made with
the turbine rotor removed. This calibration was run over a range of pressure ratio (tur- bine inlet total ta turbine exit tip static pressure pdpZt) from 1 . 1 5 to 2.9. The vari- ation in mass flow with pressure ratio is shown in figure 10, The choking mass flow for the stator was 1.233 kilograms per second (2.718 Ibm/sec) and occurred at a pres- sure ratio pupZt of 1 . 7 5 . This flow was about 5 percent less than design, indicating a smaller throat area than designed.
1 5 1.24 -I-___ , I - I I 2 . 7 r- i I
7 I
I 1.m .%
2 . 1 =.;I
-
. ' T I 1
8 2.2 I Turbine inlet total to turbine exit tip static pressure, #pa figure 10. - Stator mass floK calibration.
orque and mass flow characteristics. - The variation of equivalent torque E ~ , , ' 6 ail #f equivalent mass flow c w K r / 6 with pressure ratio for the speeds investigated is sho dn in figures 11 and 12, respectively. Data from the faired curves of these two fig- vrr::: were used to calculate the performance map shown in figure 9.
The torque curves are typical and show that torque continually increased with in- creasing pressure ratio for all speeds over 50 percent design. At the design equivalent s p e ~ II and a pressure ratio of 1.817, which corresponds to design specific work output, 70 1 Y i ! i Corresponds to design equibalent 1.2 1.6 1. a 2.0 2.2 2.4 2.6 2. a inlet- to exit-total pressure ratio, @ p i 11. - Variation 31 torque with pressure ratio and speed for untwisted r d o r blade configQration.
Figirre I I I 1, Corresponds to design equinlent I I
11 s p e d and work output
1.001 1.2 1.4 1.6 1. a 2 0 22 2.4 2.6 2.8 Inlet- to exit-total pressure ratio, @pi Figure 12 - Variation d mass flow with pressure ratio and speed for untwisted rotor blade configuration.
the equivalent torque was 36.133 newton-meters (26.65 ft-lb).
The equivalent mass flow of the turbine with the untwisted rotor blades is shown in figure 12 as a function of pressure ratio and speed. The separation of the speed lines in the figure indicates the flow is being controlled by the rotor. At 40 percent of equivalent design speed and the rotor choked at a total pressure ratio of 1.82 and flow of 1.225 kilo- grams per second (2.700 lbm/sec). A s the speed was increased, the flow at any given pressure ratio decreased. A t 110 percent of design speed, the rotor choked at a pres- sure ratio of 2.15 and a flow of 1.202 kilograms per second (2.650 lbm/sec).
At the design equivalent speed and at a pressure ratio of 1.817, which corresponds to the design specific work output of 39.572 jotiles per gram (17.00 Btuilbm), the equivalent mass flow was 1.188 kilograms per second (2.620 lbm/sec) as compared to the design equivalent flow of 1.207 kilograms per second (2.660 lbm/sec).
Turbine-outlet flow angle and static pressure distribution. - The average flow angle
at the turbine outlet is shown i n figure 13 as a function of the speed and the total pressure ratio across the turbine. Negative angle data correspond to a positive contribution to work output. The average <angle indicated at the conditions of equivalent design specific work output at equivalent design speed is -11.5' as compared to -17.8' at design. The smaller angle leaving the rotor indicates a lower s w i r l component of velocity leaving the rotor. Also, Eince AVu must remain constant for a given design specific work and speed, the tangential component of velocity leaving the stator must be large? than de- was cawed by a pressure drop across the stator greater than design and signed. This resulted in increasing the positive incidence angle into the rotor at design work and speed. The effect nf incidence will be discussed in a later section where the performance a Inlet- to exit-total pressure ratio, @pi Figure 13. - Variation d turbiqe exit absolute flan angle with pressure ratio and speed for untwisted rotor blade configuration.
between the two rotor configurations is discuss&.
The cause for the stator overexpansion can be partly attributed to higher rotor losses caused by the larger rotor radial clearance (percentage) of the scaled version turbine as compared to the 50.8-centimeter (20-in. 1 design turbine. Because of the mechanical limitations of the test faci_lity, the ratio of radial clearance to annulus area of the scaled version was almost twice that of the 50.8-centimeter (20-in. design turbine.
The variation in static pressure through the turbine is shown i7 figure 14 as a function of t o t a l pressure ratio for design speed. All pressures are ratioed to the inlet total pres- sure. The static measurement ; at the hub are shown in figure 14(a), tip measurements in figure 14(b).
Choking in a blade row i s indicated when the static pressure at the inlet to a blade row r e m a h s comtant as the total pressure ratio across the turbine is increased.
Referral to figures 14(a) and (b) shows that the hub and tip sections of the rotor choked at approximately the same overall total pressure ratio of 2.25. At choking conditions the pressure ratio to the rotor inlet at the huh was 0.55. The corresponding pressure ratio at the rotor tip w a s 0.61.
t (a) Hub.
5 ) : 1.0 a.
U .- c m _” .8 - 8 r .- c d . 6 . 4 . 2 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 inlet- to exit-total prpssure ratio, pijpb (b) Tip.
figure 14. -Variation of static pressure through turbine with inlet- to exit-total pressure rdio d equivalent design speed for untwisted rotor configuration.
Turbine Performance With Twisted Rotor Blades
Overall performance. - The overdl performance map for the turbine with twisted
rotor blades is shown in figure 15 and is similar td that shown in figure 9 for the turbine with the untwisted blades. Over the range of speeds and pressure ratios investigated.
the total efficiency of the turbine with untwisted rotor blades was less than the efficiency of the turbine with twisted rotor blades At design speed 2nd at a tota: pressure ratio of 1 . S I 7. the specific work output was 3 9 . 9 9 joules per kilogram ( 1 7 . 1 8 Btu/lbm). This ccrrzsponds to an efficiency of 8&. 0 per c ent .
T o q u e and mass f l o w characteristics. - The variation of equivalent torque ET/^ and of equivalent flow EW K d 6 with pressure ratio for i-he speeds investigated is shown in 1 5 U - - U L U - U - . 6 . 7 .8 . 9 1.0 1.1 1 . 2 1 . 3 1.4 1 . 5 1.6 1.7 1 . E ~ l d Mass flw-speed parameter, EWNIB, IkglIradllsec2
-
1 . 2 1 . 4 1 . 6 l!8 210 2!2 h 4 2.'6 2 ! 8 3f0 312 3 ! 4 316 3.8!18 Mass flow-speed parameter, EWW~I, IIbm)(rpm)Isec Figure 15 - Overall turbine performance map for turbine with twisted rotor blades.
., .~ 75 ~~ A
52 T 70- ___~. ~~ _ ~ _ .~ [--. --- _ _ _ _ ~ _ - - + - . _ ~ ~ _ _ _
48 c
I
o m I d turbine with kntwisted rotor blades 1.2 1.4 1.6 1.8 7. D 2. 2 2.4 2.6 2.8 Inlet- to nit-total pressure ratio, p b l ~ ! ~ Figure 16. - Variation of torque wiin pressure ratio and speed for twisted rotor blade configuration.
Percent of 2.6 equivalent design Speed p 2.4 E c W 1.04 Y 1. m 2 . 2 1 . 2 1.4 1.6 1.8 2.0 2. 2 2 . 4 2.6 2.8 Inlet- to exit-total pressure ratio, w p ) Figure 17. - Va, ,ation of mass flow with pressure ratio and speed for twisted rotor blade configuration.
n work output of turbine 1 . 2 1.4 1.6 1.8 2.0 2.2 24 2.6 2.8 Inlet- to exit-total pressure ratio, @ p i Figure 18. - Variation oi turbine exit absolute flow angle with pressure rztio and speec ,r twisted rotor blade configuration.
figures 16 and 17, respecti7ely. These curves a r e similar i n trend to those obtained with iintwisLLJ rotor blades.
At the design equivalent speed and at a pressure ratio of 1.817, the equivalent mass flow was 1.189 kiiograms per second (2.621 lbm/sec). Tne corresponding equivalent torque was 36.607 newton-meters (27.00 ft-lb).
The turbine outlet flow angle data a r e presented in figure 18; the static pressure dis- tribution through the turbine for the design speed is shown in figure 19. These data have the same trends as found for the turbine with the twisted rotor blades (figs. 13 and 14) and are included for completeness.
(a1 Hub.
Pressure ratio corresponding to design work output 31 tlr b i n e with untwisted rotor blades
O L - ! i- 1.4 1-1- 1.6
2.0 2. 2 2.4 2.6 7 8 1.8 1 . 2 Inlet- to exit-total pressure ratio, p$pp (bl Tip.
Figure 19. - Variatio. of static pressure through turbine with inlet- to exit-total pressure ratio at equivalent design speed for twisted rotor configuration.
Turbine P e r f x m a n c e Comparison A comparison of total efficiency between the turbine with the untwlsted rotor blades and the turbine with twisted rotor blades is shown in figure 20. At a pressure ratio of 1 . 8 1 7 , the efficiency of the turbine with the untwisted rotor blades was 87.1 as compared to 88.0 percent for the turbine with the twisted rotor blades. The difference in efiiciency is attributed to the difference in rotor inlet incidence angle between tile twr - '-JI config- urations.
Rotor blades Untwisted
--r Twisted
:r- I
c I c a l
\ I
85 .-
\ e )I i c 0 ) ._ u I t80 - m c I- ~. ~
75 .. 4 1.6 1.8 2.0 2 2 2 4
I n l e t - to exit-total pressure ratio. pblpi F i g u r e 20. - V a r i a t i o n of total efficiency with i n l e t - t n exit-tetal p r e s s u r e rario at equivalent design speed.
The incidence angles at the hub, mean, and tip sections a8 a function of overall total pressure ratio for the design speed are shown in figure 2 1 for the b o rotor coafigurations tested. The incidence angles were determined by using the measured exit flow angle, cal- culating AVu from the speed and specific work output and solvipg for the stator exit ve- locity by assuming a stator exit angle, and iterating until continuity at the rotor ir,lel was satisfied. A stator efficiency of 9G percent was assurr.d. Free-vortex flow a t the stator exit w a s also assumed. This assumption was reasonable because Jf the high hub to tip ratio ( 0 . 8 5 ) and the small amount of twist necessary for a free-vortex design. At a pres- sure ratio of 1.817, the incidence angles f o r the untwisted rotor a r e approximately +13.0° at the hub, +5.0° at the mean, and - 5 . 5 ' at the tip.
Reference 1 indicates that the inci- dence at the mean and tip sections should not significantly affect turbine performance; 2 5 I Pressure ratio ' - .
corresponding to , d w g n work output
i - - -
of turbine with un- 1 I twisted ro blades I I 1.4 1.6 1.8 2.0 2.7-1.4 1.6 1.8 2.0 2.2 Inlet- to exit-total pressure ratio. pblpp (a) Untwisted rotor.
(b) Twisted rotor.
Figure 21. - Variation of rotor inlet relative incidence angle inlet- to exit-total pressurt ratio.
however, the +13.0° a:: &~e r e c f i m could result in some loss due to local flow separa- ticn. The reference shows the losses a r e dependent upon the amount of reaction across
- the greatest losses occur with impulse turbines. The tsst results indicate
the rotor that the +13.0° incidence did contribute to some l o s s since the efficiency did improve as Ihe pressure ratio wils reduced and a more favorable incidence angle was produced at the hub section (see fig. 20). A s the pressure ratio was further reduced, the incidencd angle at the tip w a s beccn..ng highly negative, thus quickly offsetting any gain at the hub. The highest efficiency obtained for the untxisted rotor configuration was 87.5 percent at a pressure ratio of 1.65.
A t a pressure ratio of 1.817. the incidence angles for the twisted rotor are 4 . 5 ' at the tip section, 6 . 0 ' at the mean section, and +3.0° at the hu*) section. As in the case for the untwisted rotor blade version, the large positive incidence may have contributed to some loss in turbine efficiency. This is evidenced by an increase in turbine efficiency a s the pressure ratio w a s reduced below 1.817 (fig. 20), and the incidence anglep ap- proac!icd zero. At pressure r;ttios below 1.6 the efficiency cnntinued to increase indica- ling 8ome negative incidence to be desirable. The turbine w a s not tl;sted below a pres- s u r e ratio of about 1.4.
At pressure ratios above 1.817, the efficiency of Suth turbine6 decreased. This was due to an increase in positive incidence into the rotor. A t a pressure ratio cf awroxi- mately 2. I , both rotors choked and ths incidence angle into the rotor for both configura- tions remained constant. The difference in efficiency between the two turbines beyond a pressure ratio of 2.1 is a constant 2 points and is attributed to the difference in incidence between the two turbines. The drop in efficiency as the pressure ratio increases is due lnstrwwnt statim of t i r b n c t 6 t sation Figure 22. - ComDarison d stat?
cressure varia'ion through ttrbine betwetn turbinewith untwisted roto: blades and :;vbine with tv.:jted rotor blades.
to normal off-design effects.
The btatic pressure variations through the turbines a r e Shawn in figure 22. The static pressure distribution for both the cntwistd and tv.isted rotors are approximately the samc. The high radius ratio of the twbines (hub diameter to tip diameter) resulted in the small radial pressure gradients observed at the stator exit locations (station 1, fig. 22).
S u m Y OF RESULTS
AP experimental investibgtioa of a turbine suitable for high-temperature "core' '
engine application was made tc d e t c m i n e the performance level of the basic blading over a range of speed and pressure ratio, and the effect of using untwidted constad profile rotor blades rather t!!an blada design& fcr tree-vortex flow. AX experimental data a r e for a solid, uncooled, half-scale version of the actual turbine. The pertinent results are as follow8: 1. At the design equivalent speed o f 12 388 rpm, the d e s i g specific work output of 39.57 joules per gram (17.00 Btu/lbm) wa8 obkined at an overall total pressure ratio of 1.817 for the turbhe w i t h untwisted rotor blades, The corres,xmding efficiency was 87.1 percent. This agrees well with the design effin,iwxy of 87 percent.
2. A t the aforzmentioned pressure ray the specific work outplt of the turbine with the twisted blades was 29.99 joules per gram (17.18 Rtu/Pm\ wM-h restdts in an effi- ciency of 88.3 percent. Thus, an approxims:te one u J 9, incurred by using untwisted rotor blades. The difference in r-i. -;."- . p b o rotor configurations is attributed to the higher incidence losses of I .ar blade version.
3. The twist in the rokr blade had iittle IX;,.L. ,.L; ~ n t m w capacity of the turbine. A t 2: a pressure ratio ol 1.817 ?he mass fhws :or the untwisted iotor blades and twisted rotor blades were 1.18s kilograms per second (2.620 1bm;’sec) and 1.18’? kilog-am? per seconc!
i2.621 1.bn sec), respectively. The aesign flow w a s 1.207 ki1OgI’aliAS per second (2.660 1bm:’sec).
4. The turbine efficiencv obtained agreed well with the predicted design efficiency which factored in corrections for LG-N aspect ratio, Reynolds number effects, and large radial clearances.
Lewis Research Center.
National Aeronautics and Space Administration, 24, 1973, Cleveland, Ohio, September 501-24.
REFERENCES 1. Ainley, D. G. ; and Mathieson, G. C. R.: An Examination of the Flow and P r e s s u r e Rows of Axial-Flow Turbines. Rep. R&M 2891, Aeronautical Losses in Blade R-esearch Council, 9. Britain, 1955.
2. Katsanis, ThedoTe: FORTFUN Program f o r Calculating Transonic Velocities on a Blade-to-Blade Stream Surface of a Turbomachine. NASA TN D-542’;. 1969.
hASA-Lanyiey. 19i4 E-7592