An investigation of the feasibility of active boundary layer thickening for aircraft drag reduction

;JI~ C'e- 172/33

NASA-CR-177133 19860018592 (-"

A Reproduced Copy

(.

OF

@srl ~A!? - /77; /.33

Reproduced for NASA

by the

NASA Scientific and Technical Information Facility

.'

: fI,:; r, :. ~f r. f' ~y

f • I i-.

IT f; 1 8 1OF~7 111111111111111111111111111111111111111111111 NFOl710 U.h:.:I.t. ••.. 1"1 •• :..C:Ni"~R l.r8nl"~' . i lA:','\ I !1tIv.PTCN. VIRGINIA FFNo 672 Aug 65 "

I 3 1176 01348 3160

.

, ':" r ~ 1, ~t :: ...... 1 - 10 " ':" I ~ ;-:-r:' " ..... -r----- .. t,J../ ~~~~':'!lI'J-I1'lliII:\IE1""'OO;::':'!;=~..,-""" .....

-.........-\oi..~'"'\,;:-.. _to .'-A"~";;"'\:~"i -:;. : ~ .. ....:.<'r ......... +-:.D":1,,,.~~t" ,.'r"';~, '..;: ,~.~ !,..~ r!~~~~*!'f ... ~·~~~"'~:? ~&! ¥'Jl<i':aifi?~~~ ...... :!H4'$li2C( ... 9c .. .*Ji;e: C .~- 1\111 .~ _~ _ ..... _,,~ ... ~ ......... ..- .. If ,.~ ~'" .. ,,· .... ', ... .,"''lr • i

.----.-

, DEPARTMENT OF MECHANICAL ENGINEERING AND MECHANICS ~l SCHOOL OF EHGI~EERING OLD Do."'INION UfHVERSITY ,'7(1 ( ~ <: /

NORFOLK, VIRGItHA 23508 \L.I-' n I / I) L"

-··,1

I

. I AN INVESTIGATIOr. OF THE FEASIBILITY OF i 4CTIVE BOUNDARY LAYER THICKENING FOR I AIRCRI'.FT DRAG REDUCTIO:~ (NASA-Cn-177133) AN lNVESTIGA'I leN OF 'I HE N86- 280E: 4 IEaS1DILITY OF A~'IIVZ ECUN[AfiY LAY~R tHICKENING FOn AIECfiAFl CEA~ EELLCIION il.Dd1 Ref:ort, p~rl.od €DdJ.Dg 3 Har. 1986 :01J Unc1as Pcml.nl.o~ coIl.) :5 p HC 102/"1 AOl CSCL 01A G3/02 43481 Robert L. Ash, ~r1ncipal Investigator and C. Koodalattupur~n. Graduate Research Assi~tant Final Report For the perlod ended March 3, 1986 ?repared for the ) Nat;onal AeronautIcs and Spacp. Admlolstratlon , Lanqley Research Center i Hampton, VA 23565 Unaer Research Grant ~lAG-1-121 Hr. Hicha~l J. Wa'ish, Technlctl Monltor

."

HSAD-V1SCOUS Flow Branch Submitted by the Old Dominion Universlty Research Foundatlon

~7~

P.O. Box 6369

~6~g)

Norfolk, Vlrglnia 23508

'?9~©

Apnl 1986

/YY6-.;? PLJ0 ~ ~

.rr,:j~~\!l<?~-zt!l:",",*,iWI::;:":::;::i.:Li:":lC::#'i:Z:Z:'i:-:::;:'-c:!::!:':;::rrt<;:O::':l:.::::-::' :::::.~ .. ::---~-,.---. --- .,,- ..... - ... .........,...-,-,-.-.---- •. -::=:::-::::~~======::::::I"J

"v' - _ ... 0' •• _-._. ---.. -.. .•. . .• _ ... _._ ..• , .•.. .•.• - --

~ , . . ..

AN WVESTIGATIDr. OF THE FEASIBILITY OF ACTIVE BOUNDARY LAYER TtHCKErHNG FOR AIRCRAFT Ot~AG REDUCTION

~·"

:1

, ,,!

By l R. L. Ash and C. Koodalattupuram2 ABSTRACT The feasibility of using a forward mounted windmilling prapeller to extract momentum from the flow around an axisymmetric body to reduce total drag has been studied. Numerical calculations indicate that a net drag reduction is possible when the energy extracted is returned to an aft mounted pusher propeller. /lowever. net drag reduction requires very high device efflciencies.

Results of an experimental progran to study the coupling between a propeller wake and a turbulent boundary layer are also reported. The

i

.'

1" experiments showed that a canplex coupllng exists and sllnple modes for the 1 flow field are not sufflciently accurate to predict total drag.

",f j

I

'l

·1

.;'1

t

h

.r.

lChalrman/Eminent Professor, Department of Mechanical Englneerlng, Old Oomlnion Universlty, Norfolk, Vlrglnla 23508.

2Graduate Research Assistant, Department of Mechanical Engineerlng, Old Oomlnlon Unlversity. Norfolk, Virglnla 23508.

Ii ~:gZ;;7!::: :;%$ '%jU;;: : =. se:.:... .. ~: __ 4_' - '~-:~~",::_-!..!....:. -=.~~~~__ _ -- .... '-- ... --- -- ~ .¥~ I rnRODUCT I ON This project was concerned with investigating the feasibility of using propeller systems to reduce total drag on an axisymmetric body. Net drag

-

reductions of up to 10 percent have been predicted by Lobert (Refs. 1, 2) for bodie~ of revolution wit~ rtimensions similar to the fuselage of a 1arge transport aircraft travelling at subsonic speeds. Those predictions were based on simple models for the propeller wakes and for the turbulent bound- ary layer beneath the wakes. Since this drag reduction schem~ could have applications in general aviation. remotely piloted vehicles, torpedo and

-

~ubmarine systems, the reliability of the early estimates of Lobert needed to be demonstrated. In order to assess its potential for drag reduction, an accurate model of the flow field was required along with s~~e level of ex- perimental verification.

Combined numerical and experlmental lnvestigation were undertaken in the present study. The computational effort used a potential flow code (Ref. 3) and a boundary layer code (Ref. 4) to estimate skin frictlon and pressure drag over a body wit,l dlmenslons slmllar to a transport alrcraft.

Experiments were co~ducted using a cylindrical body in a low-speed wlnd tunnel to investigate the coupllng between a windmilling propeller \~ake and a fully turbulent boundary layer.

This project was supported under NASA grant NAG-1-121, and monltored by Mr. Michael J. Walsh, HSAD-V1SCOUS rlow Branch.

NLMERICAI. S JUDY ·1 A typical body of r~volctl0n is shown in Fig. 1 (at zero angle of

~

attack). Several locations and propeller sizes were exanined for the for· t .. to ;; ; L ., j" DC,. .". H· .,.' --~'-.T"~ .. -- ""-'''''-''~--.:' ....... -~~''~~ -,'- - Il'~~

V :=:J

---'r==== 7~'-

"w I NDf1 ILL" PUSHER

PROPELLER ~ ~ MOTION ~ y

~

~ PROPELLER ~ I y , \lAKE _J

-

-.,..

J -"-0./ ~ u u mmODIFIED HODIFIED BOOIDARY LAYER (REOOCED FRI CfI 00) BJ.MY\RY UWER N Figure 1. Strategy for active boundary layer thickening • • _____________ ~. ____ w _. ---- "_""'101' , f,i .. p ,.....---. .... ~ .. ---- .... :. __ w .,-_~- ---:-:=~-:..==--!--===~.::..-=-:::"--- --v- ....... --~--a ward-mounted wlndmi 11 propell e)'. The locations and blade lengths considered are tabulated In Table 1.

Table 1. Typical Numerical Results.

SKIN PRESSURE TOTAL PROPULSIVE TURBINE CASE FRICTIOr. DRAG DRAG PO'.lER REQO POWEll No Turbine 5887 lb ~181 11;) 8068 lb 6.42 M ft-lh sec Wind Turbine 3914 6061* 9976 7.94 M ft-lb 2 .1S ~, f..~:.!.~_ sec sec *Pressure drag includes drag of turbine fan.

Since the velocity distribution produced by the propeller wa~e was three-dimensional and dependent upon the number of propeller ~lades, rota- tional speed, forward speed, and propeller blade geometry. it was impossible to develop either an opttmized mean wake velocity profile or an optimized propeller location and geometry. After nunerous tri al and error i1ttenpts at produclng a realistic propeller wake velocity profile, the boundary layer velocity profile was assur,ed to take a one seventh power 1 aw f(,rm and the wake profile was assumed to be in the fOrn1: u = Co + C y + C (y- a)2 • That profile was sufficient to allow coupllng between the wake and the tur- bulent boundary layer using the 8eckwith-eushnell code (Ref. 4). Further- !,!

more, It was possible to calculate the power extrdcted by the ~ropeller and the pressure drag due to the power extraction, ihe performance of the aft-mounted pusher propeller has not been con- sidered in any detail. The difficulty associated wlth that elenent is the , f,t __ foil 1 • 'i€ j, t ;: 1 ' L t~! "" ,',.::t ... , J,..! s;.",l.., ~ ..... _,'" .... ~,l.l .. ; ",.".,.,tJtf}t W .. :~ '4 qtPJltl<.;oqt¢fh .. "'_ ;:; ll):;:;'ZW;X:::: ¥J~.t-.., "'i'""' ..... ;~4'C'l-i.o> t# ... "'~St4.,.t.lt,.or r.tP"t *-ft' }1'l.~~~1 .. :1't;"'~,,,,,: ..,:..t\t)~$+J; ... W*=(i!i¥}4~."~.,£~- ,.. bt~·..: p '1 ... rJt'"i-t::i ........ - "Iij..;i.

---------- WITHOUT PROP.

1.51 WITH PROP.

~ ~

1.0 I I !

, I

~ "

u .5 ~ \ f

,.

O~ \ J

- _-= I

----- .... , /1"'\ l

'--' ~

-.51 C-------- ~

_ =:> f'

Ul Figure 2. Predicted coefficient distribution over an axisymmetric body w;thout propellers and with forward and aft-mounted propellers.

t. - i : c =.,;;.;;:' ~,;: -.....,." • i "t > > .. ,»~L. '" I; '. "', >' "%; ." it • "r.--'. ;, >i>f\ "."*'1 "1 .... \:1 ~h~~ .. c : :;;::;.~Z: ;, ." ., lrJ<§ .. ~~ ..... ......,._«.~ .... i:'*"'wJt. .. ~-......I- ...... ~~~~JIi_ ... ~~ ... ..,..\.;.:" ".#J .... t .... i··~fft.~,J.,...i'~"!,;..."y..., .. ~~~r.!.,.~"'''kotl..,..ot!-<-..:~ ..... 'fIi .... ~1o-,-o ...... t ~~ .... t<U' ... ,~ ... !4·.,,--t"<.~J~.~i;.ibh.,.,.,\ .... ..,.;

REPRESENTATIVE BODY USED IN CQ~PUTER STUDIES

r

FLOH~ ~

" 11

PROPELLER

II ../ , I, ~ r

~

C II

.

ti " , I " 220 FT r )lao i- " .

~

~

• I'j

r'1AXIMU~l BODY D IAI1ETER: 24 FT

PROP. BLADE LEr;~TH: 1 FT

Figure 3. Representative mociel dimensions for computational study. (Aft I'JOunted propeller 0\ not shO\~n).

'''' ol ,~ '-1 flow physics to the extent that net drag redtlction can be translated into precise device efficiency requirements and this led to the experimental

, :1

phase of the investigation.

EXPERIMENTAL NEASUREMEfHS The low speed wind tunnel at Old D~ninion University was usea to study the coupling between turbulent boundary layer velocity profiles and the flow field behind a wind~illing propeller. The wlnd tunnel has a 3 (0.914 m) by 4 (1.219 m) feet test section and can be operated at speeds up to 175

·1

~

,1

ft/sec (53.3 m/sec).

An axisJrnmetric model 2.37 inches (6.02 cm) in diameter and 55.75 inches (1.42m) long with an elliptic nose and a blunt base was mounted in the wind tunnel test section. The model was designed to acc~nodate a pro- peller, 27 inches (68.6 cm) behind the nos~ as shown in Fig. 4. The Inodel was held in place by a vertical airfoil strut located 7 inches (17.28 on) fr~n the nose and an adjustable sting attached to the aft end of the model.

Ij TIle sting could be adjusted to eliminate angle of attack problens and both supports were adjusted to minilnile any bowlng of the model. A 2 watt D.C.

I

j

motor/generator was attached to the propeller shaft to vary the propeller t1

']

load and thereby control propeller speed. The Inotor was too small to produce any significant thrust and thus precludes measurements on the

.1

coupling between a turbulent boundary layer and a propulsive propeller.

A fully developed turbulent boundary layer was produced by emploYlng a 0.08 inch (2w~) di~neter wir~ trip located 3 lnches (7.62 cm) from the nose at the shoulder. Pitot probe boundary layer surveys were made along the ,1 model to establish the qualit.y of the turbulent boundary layer. lhe propel-

-. '1

ler slot was sealed to prevent any dlsturbances and typlcal bOtlndary layer velocity surveys are ~hown in Fig. 5. The boundary layer thickness varied

·1

'1

• .. t. '" " • ,f: .... _" :-',' .... Ii ;c,~./, e;:::: .!"~ (""T ~-::(- ... ':'J" '; ... =' ..

E;~1: f::: :~! 'c:>.; L:, :: ~'

.. '" ,~ .. .;; • ....... / .... ~,.;v~'·~.lq,. .. - ............ ,....,' ......... ~, ~ r"".; _ '"~ _... ' _"" Met 9 tr t "..~; ........... - / ''''9 pb.,..c. t." ';, ~.: • .';~; .~' f'e ,-,:~~,:{!"","~,:~-,' • ~~'''''~'hn~''1t''~..tr..~--::..t''1J~~r'''~i-''' ' Propeller ring I ..

~<.""'~ .

..

,

, /

----- .. _--

I'

-

/

2.t711

'~J

t

k-*:-- __ _ JII 24 ~~ ~ -28.5 ~ I: j .25

p'

Figure 4. . i Schematic diagr~m of the wind tunnel model showing dimensional data (in inches).

CD

I

,... c (C • t, ; '. , 0:'., ; S l' tI i i. :; A!£ ;a ., $ , ;;: A J ZC,,(' Sf .". 3 • ... pc R, , 1 . ..4Ji}j_' '! ... ,., Z :;~ 'V j t , p ~~J"'.!)'of'-";'" ~ ... ~ .... "',:" ... 'fll~.1,'::~·-t""~t''''l1~''' .•• : oJ' ,.. ..... \.. t' ..... ...:~ ... 1- ~ ".",.""- "-, .. I~ .. i,.~, J ,-J.- ~·"'·l{-h ...... <~l"".l~'joW ~ ... ~.,,~. ~I"'''' t-"~"''''\'~''''''-N1\~''t.~\~ ¥ ......... ~ ........ ..,."""'., .......... J~~ •• ,l'hf~ .. ..,..;..:.. ..... ~'iJ"u.NI,t~.I~~~'6(J~·,..~~~,&~Nj tr~.k •• v'Mo¥-""'" fI ~.,t;li" .. J.:t'.~""~ ~, "'~~l~ Distance from model nose tip = x lIN U N = inJex in the power law = (~) Uco .

x = 16.998 in: I

r 25.597

1 28.223

i

: I

I : .~

!

- .

' : I N = 7 : J " ': B = 6.5 t' r f.

j/

/1

."

f .'

I. ... J: •• I I I .. , .

I I • '" I, ,I. .. . ............

35.8 M/SEC

..

"I

. ;, I

: "1

·

x = 16.993

: .

· 25.595 •

1 32.507

~ :

·

,.

• : : ..

.

t; 6.5 l N = 6.5 ..

i ' .

.

i

/ .

~ .

.,,~' /

.. '

• I j.

I

I • • • i t i • , " 1/ ,< f=' , » • i i

I

I

x ~I

1-

26.8 M/SEC \0 Figure 5. Reference ~urbulent boundary development.

r-;r';'-"'2&V ~)' -- -. - -- ___ _ between .61 inches (1.55 cm) and .703 inches (1.79 cm) at a speed of 88 ftlsec (26.~ m/sec) and between .555 inches (1.41 cm) und .609 inches .

(1.55 em) at a speed of 117 ft/sec (35.8 m/sec).

-

A 10 inc~ (25.4 cm) diameter two bladed, wooden model airplanp propel- .t ler with a 6:1 pitch has been used in the preliminary phase nf the exper- iments. The D.C. motor generator was used to control the windmilling speed of the propeller when diff~rent loadings were consldered. In all cases, the propeller speed wa~ measured using a strobe light.

The free-wheeling or unloaded propeller velocity profiles are shown in Fig. 6. The apparent hysteresls effect in one of the 117 ft/sec (35.8 inl ~' f: sec) velocity profiles was due to a piotting error and should ~e ignored.

The influence of loading on the velocity profiles is shown in Fig. 7 for ~ free-stream velocit; of 88 ft/sec (26.8 m/sec).

'lISCUSSION The exper1ments have shown thus far that the velocity proflles beh1nd a windmllling propeller are net necessarily modelled by a simple coup11ng ;,,~ between a wake and a turbulent boundary layer of the type assumed 1n thp numerical rlnalysis (as sketched in Fig. 1). Obviously, the vortical compo- nents of the propeller wake can account for the inflectional properties of the outer velocity profile but the flow fleld between the helical vor~lces " and the boundary layer sugge~ts other complex flow phenomena are present.

, ~~ Questions concerning whether a model airplane propell~r operating as a f wlndmill produces a generic f1o~ field must be addressed, along with more

'{

detailed measurements. Those experiments are continuing at thlS time.

!

~'l'\.

~

f . }I

I • ~ t { ~ ~ ;i < ,4 , ,.

)" , ' r. ~~ ,",~~l;.".j~j. ~~ 'C'<i~ ... --_~<a; :;"'. :1 t ... ~,(toJJ"""-=-''''':C::O::::~ "~=:::S--::O~~;:;:;I;V-*"':;';:;;S:;: ~~~iJ3tiQ'-J , ... ~.....t ..\t>otts::e:JtZU:t~.!-~".; ~.~¥O>;;::::.o#OJ:4!4'""#1J"'1l~", , /~ :.

\ I MEAN VE~J:ITY PROFILES Fon WINDMILLING PROPELLER

I'

.l

.' ..

• t

, . . . .

.

.

~ I ~ o '> > 0 . .

I' 'r .

. i i y' J ..

1:.-

t

"1' ~ ."t \

• l_ .. _ 1:"~ ';.~ '", ___ , __ ..... !..!-:~ ___ -' .. )

. , or..f

ill n. i!~ --~~- ---, • Ie ,. .J.-rIf (ft" ,. II 'i') U ,I.. ,~ • .M ~I • ',' I

FREE STREAM VELOCITY: 117 FT/S~C (35.8 MIs)

f:(' J, I • If~l I I f ~ It

·

·

·

·

,.r I ·

.. ~

rl' · .

·

· : . !

~ f ·

. .

h

· • 0

. ~. . .

·

I

.~ J

I ,. j .: .: l . . .

I ~ .J

: l ........ ) ........ \ __ ..... J _

~q.

• II •• ,. •• 1 U ... , I U ..... ..

· J --.:;-~--::'--.--.,

~~

L l L 88 FT/SEC

~

. '

,3 ;- c .- r.: .

t

-, -<. ..

4:- PROP. TIP

ELEVATION

c ___ ,_

__---11.&-1'--0.1.1$ tw.! R ..... \' E.Y I OCAT I ON S

~

....

F1Jure E. \'elocity profilec: .

....

,'4 ,....,-- ..... .....,,- .... ~-~... . ...

~: .. '" • ,i ..... j_.: A~;:',J ( ~ ~'';;1~ q;;:.: .. ~\"r:{-~~,,'...,--";.,P;---.,.;:-".t··-;-~---t-;!~ -. -·~-;-~~-~p';.-.-'·~-T-.·C'F:r::,;--i· ,·-.-----,.,.-".,r:.---';- ... ;<..- -1~·"--'------·--T-"~-'"T.:-r""~·~~'~T';---·~1-~~~~~-iII ~~...." A ,(.J< .. "- ("~ ';"\4- '.'It~~ ;-J~i..c ... .r!J/.,. • .i...-:tiV' ..... -\"" ... ,I.~, ..... ~~,,! Vi .... ~ ... ~ .. :;;,,~~Jo-a,. .... l.~Jr~ .......... ".'<- I~'" ot~~:;l-' ~~-I.l-~"" ....... - ......... , .... ~.J;~,. rl 'JfI:~ .. "-.'... _ -l [,>: ~:)'~ i,:,~:';,::~>:;,~"~·~·,;;~?,J'>,,

J,/ ."'" JIt """, ....~ l Ji'~--- ...... -~/.:::;,.,. ..... ,.-~ ¥.~ ){ "

~~i-L~!.-~~ __ ~ ___ . ~ .... 7 __ -_ .... _ __ p .. "- ... _-,,_ ..... - .- ......... \ ...... ,..,...-,';.---;---.._ ... 1""""-... - .... - '; ~J,"~,..E. t1~

INFWEIICE OF IIIfIDMILL LOrJlJIIG ON VELOCITY PHufILES AT 88 FT/sEc (26.8 HIs) ~~ /,---

! .""c'" .. !~ .! .

I~; :< & Po!\J: 1 S. ,f-l ~ /'1. , .

.

l,2~I" .

.

'l" , "

J

l

" <;.- t: 1~~ ; l .

.

· 'l

..

~, .

.

• i : ,/~~,:

'I ~

, .) ',), .l \

f

. .J

• ' .« ., J • r-t: a~ • 'U'L:(1~&lr- 4 ~t..c:;...

nr--'-ii ---.:----.:~ • • I. •• .. .. .u · l' •• 1. I' 1 ;,

r' l

PROPELLER SPEED: 1100 RPM

t J,: & I I !l : • i ~} .

.

.'

·

· "

!

·

I

:~ J

.,1 .

• ~ ! , ~ ::

· •

t , :

· ~

·

.

·

.

·

YlO

·

.,- f

~'i'1 ~

I

>

J

-.. .. --\

d .. ,. ,........... • .... ~ __ .J I ) ."C:- .:1"

.. ' ., 'It ~ ... .. tu .. j

~

." 0-" .~

."

t t' , 0 ... · " :a r" I ...

.t

1600 RPM ~

0-' c: . • ~ ': .

I

.- t': =i ._ I ~ fi: I I

·

I

· . j

I

t , I t ~ ..

~ : •.

• I t !

i ;) Figure:. Influence of wind~il1 10301n9 on velocity profiles at 68 ft/sec (26.8 mjs). ~.

,,~~"'

!

, ~1 i ~ I "~.~"'7J"F~p-.:- ~;...:= .:::re:"]-_"1:"7~';·f" ff .. --=~ ..... ~-~ l"--'--"""'''''''' - --_ . ...,.,...-- ....... ~ - ~,~ .. ~ - •• '>:.. ....

~4J;.,:;,;,"~':;'::.:i~ :;:"-"~",&'-:'- -;:,-"/'·~~\,,;..;~niw~~~-~':"·:";.~-f;"'''':'T''-;:' , .

REFERENCES ~ ...

" 1. Lobert, G., "Drag Reduction by Means of Active Boundary Layer Thicken- ing," 23rd Israel Annual Conference on Aviation and Astronautics, Tel Aviv and Haifa, February 11-12, 1981.

J ., ., 2. I.obert, G., "New Drag Reduction Methods for Transport Aircraft," leAS- 8~-2.4.3. International Council of Aeronautical Sciences, 14tn Congress, Proc. Vol. 1, pp. 410-420, 1984.

3. Keller, J.u., and South, J.e., Jr. "RAXBOD: A FORTRAr~ Progran for Inviscid Transonic Flow OYer Axisymmetric Sodles," NASA TM X-72831, February 1976.

4. Beckwith, I.E., and Bushnell, O.M., "Calculation by Finite Difference Method of Supersonic Turbulent Boundary Layers with Tangential Slot Injection," NASA TN 0-6221, Apri 1 1971.

5. Hixon, B.A., Beckwith, I.E., and Bushnell, O.M., "Computer Program for Compressible Laminar or Turbulent Nonsimnar Boundary Layers," NASA TM X-2140, Apr11 1971.

~ End of Document