1、7_aM“NASA TECHNICAL NOTE“ GPO PRICE $“1“I CFSTI PRICE(S) $ZI- Hard copy (HC) ,_Z_/ C),=C, Microfiche (MF)Zff 653 July 65NASA TN D-4445 N68-17564(ACCESSION NUMBER) (THRU), 21-,(NASA CR OR TMX OR AD NUMBER) (C._ )NONCAVITATING PERFORMANCE OF TWOLOW-AREA-RATIO WATER JET PUMPSHAVING THROAT LENGTHS OF 7.
2、25_,by N:lson L. Sanger _ _:.iLewts Research Center _ _“T_:_“ _i_/Cleveland, Ohio _NATIONAL AERONAUTICS AND SPACE ADMINISTRATION WASHINGTON, D. C. MARCHi968Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-,J!_ LProvided by IHSNot for ResaleNo reproduc
3、tion or networking permitted without license from IHS-,-,-aiiiiItwfNASA TN D-4445NONCAVITATING PERFORMANCE OF TWO LOW-AREA-RATIOWATER JET PUMPS HAVING THROATLENGTHS OF 7.25 DIAMETERSBy Nelson L. SangerLewis Research CenterCleveland, OhioNATIONAL AERONAUTICS AND SPACE ADMINISTRATION/For sale by the C
4、learinghouse for Federal Scientific and Technical InformationSpringfield, Virginia 22151 - CFSTI price $3.00FProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-lwProvided by IHSNot for ResaleNo reproduction or networking permitted without license from I
5、HS-,-,-_ : _ ,-_; _ . iCONTENTSSUMMARY .PageiINTRODUC TION 23: PERFORMANCE PREDICTION !| 3, PRINCIPLE OF OPERATION .= ANALYSIS 5Performance Parameters 55Conventional and Modified Analyses .Component Losses . 78Axial Static Pressure Distributions .APPARATUS 9TEST FACILITY . 9Research Pump Loop . 9Aux
6、iliary Systems . 11TEST PUMP DESCRIPTION . IIINSTRUMENTATION . 14EXPERIMENTAL PROCEDURE 15TESTS OF SMALLER AREA RATIO JET PUMP, R = 0.066 15TESTS OF LARGER AREA RATIO JET PUMP, R = 0.197 16| RESULTS AND DISCUSSION 16- EXPERIMENTAL RESULTS 16I-_ Overall Performance . 16Comparison of Experiment and Th
7、eory 18_ Mixing Characteristics . 21_- Axial static pressure distributions 21Total pressure surveys . 23COMPONENT LOSSES 31|immmCONCLUDING REMARKS . 36THROAT LENGTH 37NOZZLE SPACING 37,111Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-DIFFUSER GEOME
8、TRY . _.OVERALL PERFORMANCE SUMMARY OF RESULTS .APPENDIXESA - SYMBOLS .B - DEVELOPMENT OF JET PUMP EQUATIONS CONVENTIONAL ANALYSIS .Performance Parameters .Dimensionless Loss Expressions .Dimensionless StaticPressure Rise inthe Throat MODIFIED ANALYSIS - PERFORMANCE PARAMETERS .DETERMINATION OF FRIC
9、TION LOSS COEFFICIENTS REFERENCES .373839414343434851525559ivProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-|-_Pr.anZzztlImEENONCAVITATING PERFORMANCEOF TWO LOW-AREA-RATIOWATER JET PUMPS HAVING THROATLENGTHS OF 7.25 DIAMETERSby Nelson L. SangerLewis
10、 Research CenterSUMMARYThe performance of two jet pumps having low nozzle- to throat-area ratios wasevaluated in a water facility and compared to theoretically predicted performance. Thepurposes of the investigation were to gain a better insight into the flow mechanisms in-volved in low-area-ratio j
11、et pumps, and to compare the abilities of two existing one-dimensional analyses to predict jet pump performance over a wide range of operatingconditions.Two nozzles were evaluated experimentally and the test pump consisted of one ofthe two nozzles and one test section, the latter having a throat dia
12、meter of 1.35 inches(3.43 cm), a throat length of 7.25 diameters, and a diffuser included angle of 8o6 (0. 141 rad). The nozzles had exit diameters corresponding to nozzle- to throat-arearatios of 0. 066 and 0. 197. Each nozzle was operated at several spacings of the nozzleexit upstream from the thr
13、oat entrance over a range of from 0 to 3 throat diameters.For an area ratio of 0.066, a m_ximum measured efficiency of 29.5 percent wasachieved at a fully inserted nozzle position (nozzle exit plane coinciding with throat en-trance plane); for an area ratio of 0.197, a maximum measured efficiency of
14、 35.7 percentwas achieved, also at a fully inserted nozzle position. Performance at maximum effi-ciency levels was maintained for both area-ratio jet pumps over a range of nozzle spac-ings from 0 to 1 throat diameter.At small nozzle spacings (up to 1 throat diameter) a simple one-dimensional analy-s
15、is predicted performance quite closely. The theory also demonstrated that low effi-ciencies exhibited at low ratios of secondary to primary (high pressure) flows are due toinefficient mixing, whereas low efficiencies at high flow ratios are due largely to frictionlosses. A modified theory, which att
16、empted to account for the effect of the mixing pro-file in the throat, required more computational effort and did not improve correlationwith experimental performance.Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-INTRODUCTION ,-1FOne method being c
17、onsidered for generating relatively large quantities of power forapplications in space vehicles is the use of a Rankine cycle system utilizing liquid metalas the working fluid (ref. 1). The jet pump has been utilized in the development of thesespace-oriented systems (refs. 2 to 4) because of its sim
18、plicity and inherent reliability;furthermore, it requires only a modest net positive suction head for cavitation-free oper-ation and it can be installed in locations remote from mechanical power sources. A jetpump may have several applications in these systems. It may be used as an inducerstage for
19、the boiler feed pump, a recirculating boiler pump, or as an auxiliary lubri-cation or cooling pump. The probable source of the high-pressure driving (primary)fluid for such jet pump applications will be from mechanical or electromagnetic pumps.In order to keep size, weight, and power requirements of
20、 the main pumps low, it will benecessary to keep the amount of fluid recirculated (primary fluid) to the jet pump low.As a consequence, jet pumps which operate at high ratios of secondary to primary flow-rates are required, and they are specified geometrically by low ( 1.05 28 (1.77x10-3)-A 1.54 28
21、(1.77x103) 1.54 35 (2. 21x10-3)-tl 2.08 28 (1.77x10-3)3 2.58 28 (1.77x10-3)-3.04 28 (1.77x10-3) 3. 04 35 (2.21xi0-3)-261l. 1 ,l _10 I i.0 1 2 3 4 5L_ _ !,16 i=.,. L2-_ . _Nozzle spacing,.o8 s/_ “_: _ “l 3.040 1 3 4 5363224I20160(a-l) Efficiency._ I.4/.8 3.2 1,Flow rate, M - Q2/QI(a) Area ratio, R -
22、0.066.I I t l INozzlespacing,sldtO 0o . g6 |_ A 1.36 /r, 2.26_ C3 2.68_ -. .2.0 2,4 2.8(a-2) Headratio.38r_._0- 26 - 22 - 18 -.14 -_ -I 10 -.06 _L_0-_- I t t t INozzle spaclng,_ s_ _-i c_ l_8I1 i 1-I, .8 _2 16 2.o 2.,12.8Flow ratio, M = Q2IQI(b-l) Efficiency. Ib-2) Headratio(b)Area ratio, R = 0.197.
23、 Primary flow rate, 63 gallons per minute (3. 9/x10-3 m3/sec).Figure 7. - Noncavitating performanceof jet pumps17Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-tion and at a flow ratio of M = 1.40 (fig. 703-1). The head ratio N corresponding tothis
24、flow ratio and nozzle position was N = 0. 255 (fig. 703-2).Figure 8 (a cross plot of figures 7(a-l) and 703-1) illustrates the effect of nozzlespacing on jet pump maximum efficiency for each area ratio. The maximum efficiency,after remaining relatively constant for nozzle spacings up to about 1 thro
25、at diameter foreach configuration dropped off gradually as spacing was increased. This effect will bediscussed later.Comparison of Experiment and TheoryPlotted in figure 9 are theoretical curves for N and _ as calculated by equa-tions (13), (14), (34), and (35) (appendix B) for the fully inserted no
26、zzle position. Theloss factor K_ used in equations (34) and (35) was determined after studying the totalpressure surveys presented later in figures (14) to (16). Although this information wouldnot be available to the designer, it was used in this instance to provide the most accuratemeasure of the p
27、otential usefulness of the modified theory. It is immediately evident thatthere is very little difference between the original conventional theory and the modifiedversion. Furthermore, the experimental points, in general, compare closely to the con-ventional theory. At the best efficiency flow ratio
28、 the conventional theory predicted per-formance within about 3 percent.One of the fundamental assumptions of the conventional theory is that nozzle spacingIs zero. However, since jet pump performance changes little between nozzle spacingsof 0 and 1 (figs. 7 and 8), the theory was also in general agr
29、eement with experiment in36 I32_g 28._uE_ 24-x200Area ratio,R - O.197 - -0 .056t I I.4 .8 1.2“-o,b,1.6 2.0 2.4 2.8 3.2Nozzle spacing, sld tFigure 8. - Noncavitating jet pump performance. Effect of nozzle spacing on maximumefficiency.18Provided by IHSNot for ResaleNo reproduction or networking permit
30、ted without license from IHS-,-,-i,-_|gZiiR,-,4.3.2.i0a_w,5iii-ric,.lIivaZ._o-_,wm _N Ex_rim(_ntal l)rima_flow rate,gallmin Q1(m31sec)0 28 (I.l/x10-3) 35 (2.21xi0-3)-TheoryConventional -Modified_ _-1 2 3 4 5Flow ratio, M =Q21QI(a Area ratio, R, 0.066.I I I I IhprimaryExperimental- flow rate, -gallmi
31、n Q1im31sec)O 3 63 (3.97xi0 -3) _ 83 (5.24xi0-3)- - Theory -Conventional - - “_ I_., Modified_ -(_=t %“%.4 .8 1.2 1.6 2.0 2.4Flowratio, M - Q2/Q1(b) Area ratio, R, O.197.Figure 9. - Comparison of theory with experimental points for fullyinserted nozzle position (s/dt = 0).19iProvided by IHSNot for R
32、esaleNo reproduction or networking permitted without license from IHS-,-,- 12O8.040-,04CI. Z4o.20i,oEz 1208 m.04-.040“ Nozzle spacing, Primary ilow rate,sldt gallmin Q_m_sec)0 28 (1.77xi0-3)0 35 (2.Zlxl0“3)1.05 28 (i./7x10-3)3.04 28 (1.77xi0-3)-L - IOAZ0 01 2 3 4Flow ratio, M =Q2/QI(a) Area ratio, R
33、, 0.066.Conventional theory5 6 Nozzle Primary flow rate,. spacing, QI,-._ 0 Sialt gallmin ,mlsec,0 63 (3.g8x10-_ 0 83 (5.24x10A i.36 63 (3.98xi0-_ 0 3.02 63 (3.98xi0-II theoryConventiona J_X-0-0.4i.8 1.2 1.6 2.0 2.4Flow ratio, M : Q21Q1(b) Area ratio, R, 0. 197.Figure 10. - Nondimensional pressure r
34、ise in throat.2.8“k-20Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-that rahge of s/d t. As the nozzle is retracted, mixing occurs in the converging second-ary inlet region as well as in the constant diameter throat. Therefore, it was not unex-pect
35、ed that at large nozzle spacings experimental performance did not agree with thetheory.An additional demonstration of the applicability of the conventional theory is shownin figure 10. With the same friction loss coefficients as were used in the calculations forfigure 9, equation (28) was used to pr
36、edict the pressure rise in the constant diameterthroat. The theoretical nondimensional pressure rise in the throat ACp, t is plottedagainst flow ratio in figure 10. Experimental values of the parameter at selected nozzlespacings for both area ratio pumps are also plotted in the figure. As in figure
37、9, thecomparison between theory and experiment is good at zero and relatively low nozzlespacings, but they diverge significantly at large spacings where the premises of thetheory are violated. A major effect of large nozzle spacings is a failure to achieve the-oretically attainable pressure rise in
38、the throat.Mixing CharacteristicsA better understanding of the internal fluid mechanics can be gained by an examin-ation of the axial distributions of static pressure, and the radial surveys of total pres-sure.Axial static pressure distributions. - Two distinct effects can be differentiated byan ins
39、pection of the static pressure distributions in the axial direction.(I) Effect of nozzle position - A further experimental representation of the effect ofnozzle spacing is shown in figure II, a plot of the axial distribution of static pressurecoefficient for both area ratios. Figure ll(a) presents d
40、ata for an area ratio of R = 0.066and for three different nozzle positions at the best efficiency flow ratio of M = 3.5. Animportant effect of nozzle spacing, evident at all flow ratios, is a general rise in theoverall pressure level in the secondary inlet and throat entrance region as the nozzle is
41、retracted from the throat entrance. This is due directly to an increase in annular areaof the secondary inlet as the nozzle spacing is increased. Similar effects are exhibitedby the R = 0. 197 pump at three comparable nozzle spacings (fig. If(b). The major ef-fect of a low-pressure level at very sma
42、ll nozzle spacings is a high susceptibility tocavitation. This is an important consideration because the best efficiency was obtained ata nozzle spacing of zero. If cavitation were to limit operation at this nozzle position,then some design compromises would have to be made. For this reason the zero
43、 nozzlespacing position should not be inflexibly categorized as the “optimum“ nozzle positionfor this secondary inlet-throat-diffuser configuration.21Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-.20 N-zzlespacing,o Nozzlespacing, a3 .-LI-I.16 O 0
44、- - O 0 - _, I. 05 .96A 2.58 A 2.6812 I.08(j_ _ /.O _13 -E l - _“G .04 . ,-_ i-End of throat -End of throatI0 14 18 22 -2 2 0 10 14 18 22Axial location from throat entrance, xldt-. 12-.16-2 2(a) Area ratio, R, 0.066; flow ratio, M, 3.5. (b)Area ratio, R, 0.197; flow ratio, M, 1.4.Figure 11. - Effect
45、 of nozzle position on axial static pressure distribution.Mixing length can be equated with throat length only at zero nozzle spacing. Whenthe nozzle is in a retracted position it is not practical to define an equivalent mixinglength because of the unknown effects on the turbulent mixing of the pres
46、sure and velocityfield in the secondary inlet region. The length of the throat should be sufficient to permitthe static pressure to increase continually with length, but not so long as to cause staticpressure to become constant or decrease due to friction losses. The length of throatnecessary to per
47、mit static pressure to reach a maximum is plotted in figure 12 as a func-tion of nozzle spacing for both area ratios. The curves represent the limit of staticpressure increase in the throat for all flow ratios and are general in nature. The figureshould not be interpreted as a plot of optimum or rec
48、ommended throat lengths. What issignificant about the figure is that the two area ratios demonstrate different static pres-sure trends in the throat. The static pressure reached a peak earlier in the throat forthe smaller area ratio pump (R = 0.066). Another way of stating this is that a longermixing length was required at all nozzle positions for the larger area ratio pu
