1、REPORT No. 680THE EFFECT OF NACELLE-PROPELLER DIAMETER RATIO ON BODY INTERFERENCEAND ON PROPELLER AND COOLING CHARACTERISTICSBy Jums G. MCHUGH and ELDRIDGE H. DERRINGsuMMARYAn investigation was conducted in the N. A. C. A.$O$oot tunnel to determine the slipstream drag, the bodyinterference, and the
2、cooling characteristics of nacelJe-propeller combinations m“th different ratios oj nucellediameter to propeUer diameter. Four combinations ofgeometrically similar propellers and nacelks, mountedon standard wing support8, were te8tedwith ?xduesof theratio of nuce.?lediameter to propeUerdiameter oj 0.
3、fi6, 0.%?,and O. (fi thenet ejkiency of a naeelkopeller combination decreusesrapidly m“th increasing dues oj the ratio oj nucellediameter to propeller diameter; (3) the presence of aspinner ozer the propeller hub increases the propulsiveem”ency by an amount vaying from 1%to 4 percent; and.4)the maxi
4、mum prewure drop available wiih adjustablecowlingjlaps is about gOpercent greaterthun the maoirnumpressure drop available w“th an adjustable-kngth cowlingskirt. INTRODUCTIONConsiderable information has reoentIy been madeavailable concerning the propukh-e and the codingcharacteristics of a fu-scale a
5、ir-cuded radial-enginenaceUe-propelIer combination having a ratio of thenacelIe diameter ta the propelkr diameter of approxi-mately 0.41. Very little information is available con-cerning the eikts of variation of that ratio on the a.lip-stream drag, the body interference, and the efficiencimof a pro
6、pelkr-nacelle combination or on the cooling-air-flow characteristics of a nacelle-propeller combination.LMostpre5ent-day estimates of the variation in pro-pulsive efficiency with the ratio of nacelle diameter toprcpekw diameter are based on the results reported inreferences 1 and 2. Those investigat
7、ions were con-ducted with an uncowled radid engine and low-pitnhpropders, and the results are not applicable to presentpractice. Only a few isolated tests are avaiIabIe fordekrminkg the efteot of variation in the ratio ofnacelle diameter to propeller diwneter on the cooling-air-flow characteristics.
8、In order to suppy adclitiona information on thissubject, the N. A. C. A, has instituted an investigationof wing-naceIIe-propeHerinterference and cooling char-acteristics. The investigation incIudes: (a) determina-tions of the drag and of the propeller and cooling char-acteristics of four combination
9、s of geometrically simikrmodel propellers and nacelles having values of the ratioof nacelle diameter to propder diameter of 0.25, 0.33,and 0.44; and (b) determinations of the lift, the drag,and the propeller and cooling characteristics of the samecombinations of propellers and nacelk operating incon
10、junction with a 5- by 15-foct N. L C. A. 23018airfoil. This report presents the remdti of part (a);part (b) is reported in reference 3.The present report gives the results obtained fromtests of geometrically eimilax 3-bIade propellera ofdiameter D of 36 and 48 inches (3 and 4 feet) operatingin conju
11、nction with geometrically similar nacelles ofdiameter d of 12 and 16 inches, making possibk the d/Dratios of 0.25, 0.33, and 0.44. Results obtained fromother tests in which free-propelIer conditions were ap-proached are ako prwented. The effeots of a variationin the ratio of nacelIe diameter to prop
12、eIIer diameter onthe propeIIer characteristics and on the a1.ipstreamdragM weII as the effects of nacelle interference on propekrpower and thrust are shown. AIso included are theresults of determinations of the cooling characteristicsof all the combinations tested in addition to comparisons,on one n
13、acelIe, of adjustable cowhg flaps with an ad-justable-kmgth cowling skirt as a means of controlledcooling.APPARATUSAND METHODSThe N. A. C. A. 20-foot wind tunnel in which thesetesti were conducted is described in reference 4. Theteds were conducted at air speeds from 20 to 80 milesper hour.Two geome
14、trically similar sheet+ahuninum naceks,12 and 16 inches in diameter (fig. 1), with nose 7 ofreference 5 were used in the investigation.727Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-728 “- “”-REPOkT 0: 680-NATIONAl-AIJVIfJORY COJTTEE FOR AERONAUT
15、ICSThe engine was simulated by fine-mesh wire screens,the resistance of which had been adjustad to give thedssired conductivity. The conductivity was deter-mined from measurements of the quantity of air flowthrough the cowling and of the pressure drop across theCowling flop Stic pressureStole, in. W
16、ire!-mesh screen16-in. nacelle assly.0Y8 ._.Stoic pressure“ Electric motor ,_Wire-mesh sckeen W1h $-“12-in. nacelle ossemby nocelieFIGUBEL-SPlnneI andmcellemcddswd in prorwllw-namlleIrwWon.screensthat simulated the engine. From these measure-ments, the conductivity K (reference 5) was found tobe 0.0
17、85 for the 16-inch nacde and 0.072 for the 12-inchnacelle.For certain of the teds, t cowling-exit area of the16Ach nacelle wae varied bo.fk by adjusting the cowlingflaps (fig. 1) and .by reducing the length of the cowlingskirt.Two 3-blade propellers, 36 and 48 inches in diameter,having Clark Y secti
18、ons and geometrically similar topropeller 6101 (reference 5) except for vwiabe-.insteadof controllable-pitch hubs, were “ Thepower output of the motor was obtained from a cali-bration involving motor torque, revolution speed, andactive current.The test set-ups were moted in the airstream on thestand
19、ard airfoil supports (reference 6) aridall thiust anddrag forces were measured by automatic recordingbalances on the tast-charnber floor.For. that portion of the teet program in which it wasdesired to obtain free-propeller characteristics, the pro-peller wae driven through a 3-foot-extension shaft.
20、Thamotor. with its extension shaft was supported lmtwcenthe standard airfoil supporte as shown in figure 3, Thecompl r, sfdfon radios+b, nect!onchord;k wtlon thtcknowz%WxnMo PkcluI%blademgk.The extension shaft was not used for the tests of thenacelle-moeller unite. The motor was built into the.Anace
21、lle and was supported between the airfoil supportsas shown in figure 3, At the beginning of this part ofthe teat program, the supporting strut (which will sub-sequently be called shut 1) was lmwe and made a bad-.intmsecfion with the n-aceIIe,thereby causing separationProvided by IHSNot for ResaleNo
22、reproduction or networking permitted without license from IHS-,-,-EFFECT OF NACELLE-PROPELLER DLAMETER RATIO ON BODY I.NTERFEREITCE 729The I!2-fnehmdle. The propelleralone.The 12-fuchnncellewith.orer. The propellerfn the presenceof the Wnrh nacelhThe l the air speed was then held constant and thepro
23、peller revolution speed was varied ta cover the restof the propeller operating range. Simultaneous read-ings of power, thrust, revolution speed, air speed, andpnmure drop through the engine were taken at fre-quent intervals.Provided by IHSNot for ResaleNo reproduction or networking permitted without
24、 license from IHS-,-,-730 REPORT NO. 680-NATIONAL ADVISORY COMMITTEE FOR AERONAUTICSThe tare drag of the supports was determined byattaching to the nacelle a single dummy strut, geo-metrically similar to the ones that supported the na-celle. Drag tests “were rde. Wth. and witho thedummy strut in pla
25、ce. The tare drag was then de-termined on the mssumptiog_J.at theige?se in dragdue to the presence of the. dummy strut was-one-h-ilfof the total strut and interference drag. The taredrag with the propeller operating was not determined.The nacelle arrangements tested are shown in ure 3.The various co
26、mbinations on which measurements ofpropeller characteristics were.obtained are listed in theoll - - - -, - - - .- - - -.8 - - -,- .- -b.6 Strut 3 (wiihod sepordion-” I (with *) T.4 -.2 -01 I I I I I I I I I I I I I I 1.4 .6 .8 t.o 1.2 /.4 i, 6V.ITL.D!. 8.strut 3. The cause of this dfierence in propu
27、lsiveefficiency was determined from a study of the air flowover the naceHe by attaching streamers to iti surfacetmd studying their actions in the ah stream with andwithout the propeller operating. When the nacellewas supported by strut 1 with the propeller removed,the air flow separated over all tha
28、t portion of the nacelleback of the strut intersection but, with the propeUeroperating, the tied of the slipstream was h shift theseparation point downstream by several inches. Theslipstream thus caused an effeotive reduction in nacelledrag and a high value of propulsive efficiency wastherefore obta
29、ined. A similar study of the flow whenthe nacelle waa supported by strut 3 revealed that theseparation point was near the tail and was apparentlyuninfluenced by the propeller slipstream.As a result of this study, it is desired b stress thecZ/DOrithe eflicienoiesof the naceUe-propeUercombina-tions, t
30、he envelope curves of apparent, propukive, andnet efiiency for the various arrangement tested megiven as a function of V/nD in gure7 and as a functionof C, in figure 8. Attention is called ta the fact thatthe readts given for values of d/D of 0.10 and 0.13 infiguras 7 and 8 were obtained km teats wi
31、th no nacellebehind the propelIer. In those cams, the value of d/Dis the ratio of the diameter of the extension-shaftfairing to the propeller diameter.The hmreaaein both propulsive and net efficiency thatoan be obtained through the use of a spinner is alsoahown in figures 7 and 8. At tha value of d/
32、D of 0.25,at which tests were made with a spinner over the pro-peller hub, the gain obtained varied from about 1 per-cent in the take-off re to about 4 pereent in thehigh-speed range.Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-EFFIWT OF NACISUJJ3
33、-PROPELLER DL4METER R4TI0 ON BODY FERENCE 733Figure 9 Sununarl“zes the results given in figures 7 Irapid deorease in net efficiermy with increasing valuesand 8 and illustrates quite cIearIy the variation at both of d/D. The divergence of the two sets of curves fromconstant V/nD and constrmtC. of the
34、 various efficiencies each other is explained on the premise that, aIthoughLo -. -. -.8 -= _- _ -+ -.6Lo.8?.61.0.8- - A.6V* Propeller Nacellediamefcr,in. diame fer,in. d/D.4 36 12 0.33- - 48 16 .33- 36 16 .44- No nacelle .10 :1 No nacelle .13 -48wifh spinner )2 .25-48 12 .25.20 .6 .8 f.o /.2 f.4 1.6
35、 f.8v3FZIYJEE7.-The vfufatfon of apparent, propuklv and ne$eiliclency enwlopa with V/nI).with reIative body size. The chosen vahms of V/nDand C, roughly correspond to the take+ff end cruisingconditions of a representative transport airplane. Thereis a rapid increase in apparent efficiency md a914the
36、 increase in body size causes an increase in apparentthrust owing to the greater reaction createdbetweenbody and propeller, the net thrust of the combination hasbeen reduced owing to the increase in body drag.-Provided by IHSNot for ResaleNo reproduction or networking permitted without license from
37、IHS-,-,-REPORT NO. 680-NAONAL ADVISORY COMMITTEE FOR AERONAUTICS .- . - - - - - - - - - - . -. . *i-w- - -/- - w .- -= - - - - - .- - i Propeller, Nate I/ediameter,in. diumeter,in. d jD.36 )2 0.33- _. .16 .33- .$? !6 .44-48 Nonocelfe ,0 36 Nonacelle .3 -48(with spinner .25-48 .% .25_ - - . - - - - -
38、 _Y .-.-%/.4 1.6 1.8 2.0 2? 2.4 2.6C.qProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-EFFECT OF NACELLE-PROPEWR DIAMETER RATIO ON BODY INTEEI?ERIKNCE 735Figure 10 is presented to show the effect of the d/Dratio on the parameters that iufhmnce the sek
39、otion ofa propeller. The ideal efficiency of a propeller is di-rectly dependent on power disk loading. The tiectof change in value of d/D is, in turn,to alter the powerdisk Ioding. The use of the power disk-Ioadimg co-efficient P. as the independent .vm-iable in figure 10 istherefore very convenient
40、 in that it aLlowscomparisonsof the other importsmt design caeflicients at a constantvalue of power disk loading, The other coefficientsshown in figure 10 were obtained from design chartsof the type shown in ure 11 and from similar chartsin which 1 was the independent variable. Figure10 is of extrem
41、e interest because it shows in conciseform, for all the combinations tested with strut 3, thefoIlowing important information: C, at V/nD for .z;V/nD for .a,; m d the vaIues of qm=zand B that areusually obtained from the conventional C* designcharts. At the same time it permits their comparisonat a c
42、onstant value of power disk Ioading.The values of nm.zobtained at d/D=O.33 are nearlythe same as those obtained at d/D= O.25. The differ-ence is of the order of one-half of 1 percent and maybewithin the experimental error of the remdts. At the10WWratio, however, there may be a loss in propulsiveeffi
43、ciency owing to the fact that a relatively small bodydoes not tend to make inoperative the inefhient huband root sections of the propeller and therefore a largerportion of the power is wasted than if the propellerwere operating in front of a linger body.The propulsive e%iciency obtained at d/D=O.44
44、is ofthe order of 4 or 5 percent Iower than that obtained ateither of the other ratios. Inspection of the resultsgiven in figure 10 and of the enveIope curves of pro-pulsive efficiency given in figures 7 and 8 indicates that,for practicaI installations, the eflect of d/D on propul-sive efficiency is
45、 relatively unimportant at values ofd/D 1sssthau 0.33. At higher vA= of d/D, q decreasesrapidly with increasing values of d/D. This result is inagreement with the remdt anticipated from extrapola-tion of previous test results.The difierw.mes in proplve efficiency that havebeen discussed thus far me
46、attributed to the fact thatthe presence of a body behind an operating propellerhas two opposing E the representation of thevariation of ATC with Tcf by a single straight linefor all blade angles was therefore considered to bejuetiiied.The method used to evaluate the slipstream drag wasbasimlIy simil
47、ar to that used to determine the body-interference effects. The slipstream drag is the difler-tion, there can be but one value of the power coefhcientat a given value of 17/nD.The variation of slipstream-drag coefficient TCdetermined by the method prev-iousIy outlined withthe apparent propeller thru
48、st-loading coeflibient isshown for the various propeller-nacelle combinationsin figure 14. There is considerable dispemion of theplotted points but, by the same reasoning used in the.8.706.5n,8II111/1/ l/l YIA/t111111 lllzll=”ll.4.3/.5.2Lo.o “%.50 .5 1.0 .5 2.0 25 0cFmrim 11.Representativedesign ch M-inch propelk end 16-fnchnrwdle.ence between the propeh thrust T and the propulsivethrust (TAD) and, in coefficie
