UFDC Home  Search all Groups  World Studies  Federal Depository Libraries of Florida & the Caribbean   Help 
Material Information
Subjects
Notes
Record Information

Full Text 
l~ci NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS WA RTI l M REPORT ORIGINALLY ISSUED February 1946 as Advance Confidential Report L5J29a EFFECTS OF SPECIFIC TYPES OF SURFACE ROUIDGESS ON BOUNDARYLAYER TRANSITION By Laurence K. Loftin, Jr. 1 Langley Memorial Aeronautical Langley Field, Va. Laboratory *""'i:.. . .. ... .. .. ." " v*eX S' .,e ,. ,. A. WASHINGTON NACA WARTIME REPORTS are reprints of papers originally issued to provide rapid distribution of advance research results to an authorized group requiring them for the war effort. They were pre viously held under a security status but are now unclassified. Some of these reports were not tech nically edited. All have been reproduced without change in order to expedite general distribution. L 48 K.i sL.. i..;; i... t .. DOCUMENTS DEPARTMENT /, I i 2 MACA ACR No. L5J29a I / :JATIO_;AL ADVISORY COi':TT'E FOR A OCIR.. jC A~DV ,, CO ::D IAL I.PORT EFFECTS OF SPECIFIC TYPiS OF SURFACE RO'.GHW'SS ON BOUNJkDAYLAYSR TrA;!3SITIO: ny Laurence K. Loftin, Jr. S I: ARY Tests .',ere conducted '::ith tvo typical lowdrag air foils of .Oinch chord to determine the effects of surface projections, grooves, and sandinc scratches on boundary layer transition. The Reynolds number at which a san'..ise reo of cylindrical projections w*.ould cause pre:.iaturo transition was determined for a range of Reynolds number fron approximately 5 x 10 to 10 x 10. Data :ere obtained for projections of various sizes and clord":ise locations on both. lov:drag airfoils. Th. results were analyzed on the assumption that the critical airfoil ReInolds number for a Liven projection .:as a fuilction only cf the localflow conditions around the projection. This Fssulption neglec:ted possible effects cf tun:;el tur bulence, pressure gradient, boundarylayer Reynolds number, and the crifin;al :.tent of the la:.i.nar flow. The data correlated on the basis of thick assur:tion a'ithin a range of critical airfoil Reynolds number of 0.5 x 10 and within a range of projection eight of C.00'2 inch. The tests of surface grooves and sanding scratches indi cated that, for the range cf Re:niolds ou.nber investigated, the laminar boundary layer :*as ;much less sensitive to surface rrooves and sanding scratches than to projections above the surface. INTRODTJCTIONi The development of the 5ACA lowdrag airfoils has aroused a greatt deal of interest in the problem of deter mining the amount of surface roughness necessary to lause pre.nature boundarylayer transition from laninar to tur bulent flow. A considerable amount of data has been published (references 1 ?rn 2) pcvtaining to th: effects NACA ACR i;o. L5J2Qa on airfoil characteristics of the application of carbo rundum grains to airfoil surfaces. Little data have been published, however, concerning the Reynolds number at which surface projections of a given size and chordwise location would cause premature transition. Fa"T has con ducted tests to determine the allowable size of three forms of surface ridge flat, arch, and wire located at various positions on lowdrag airfoils (references 5 and 4) and later extended the work to include the effect of smooth bulges and hollows (reference 5). Tani, Hama, and Mituisi (reference 6) have investigated the effect of spanwise wires on premature transition. The purpose of the present investigation was to determine the Reynolds nu, .er at which surface projections of a given type but of various sizes and chordwise loca tions would cause premature transition and, if possible, to establish a general relation between the projection size and critical Reynolds number. An attempt was also made to determine the effects of sanding scratches and imperfect sheetmetal butt joints. The tests of this investigation were conducted in the Langley twodimensional lowturbulence tunnel. Two typical lowdrag airfoils were tested and data were obtained for various combinations of projection size and chordwise location through a range of Reynolds number from approximately 3 x 106 to 10 x 106. Data were also obtained with the airfoil surfaces finished with various grades of sandpaper and carborundum paper. The imperfect sheetmetal butt joints were simulated by grooves cut into the surface. Tests were made with spanwise grooves of various sizes and chordwise locations. Although the projections tested simulated no definite type of roughness, the results of this investigation should prove useful as an indication of the order of mag nitude of the individual specks that may be tolerated on a lowdrag airfoil of given chord and pressure distri bution. T::e Reynolds numbers of these tests were low compared with usual flight values; however, application of the analysis to the prediction of allowable projection sizes at higher Reynolds numbers appears reasonable, par ticularly for projections on the forward part of the airfoil. CC I DEN. : IAL CO IDE'IAL 'A: AD3. 1:. Lj29a CC':'DE:.TIAL 3 SET7,CLS A:D COE'FICIZl.TS c. airfoil section drsa coefficient y distance normal to surface of lo,'.drag airfoil S bcun.'.rd;:layer_ thicc'ness, defin.ri as t!lat distance normal to the surface at which =  1; h.iht of nrc action d diameter of projection c chord of ln':.dr.rz airfoil j stancee frnom airfcil leading e'l,e measxred alon.i chord line s distancee from atifcil leading edge mlieasur.ae along surl 1' .ce Ti, freestreaam velocity cI local velocity just *:.tsit,7 i tou'ndr 1 le:' u local veloc ltAy 'nide boundry a: 1er Uk local v.locit inside bou nirrl' l' er at to, of a projection qo freestrea:n dyrnam.ic npressiure p local static ?r assure V; local total pressure just outside boundary layer II0 "'rcestreamn total pressure h local total pressure inside boundary layer o e\ S pressure coefficient (o r' Scoeffiieo ', coefficient of kinematic viscosity. C ON F IE'ENT IAL 4 CC'TID; TIAL T..1CA AC(. "o. L5T2 a R airfoil Reynolds number; based on chord of lowdraj airfoil and freestream velocity  R' Reynolds number per unit length; bas,. on velocity just outside bounairry. layer () RF boundarylayer Reynolds number; based on boundary layer thickness and 1rcal velocity just outside boundary layer (R'6) Rk projection Reynolds number; based on hLiJht of pro jection ri'n velocity in boundary layer at top of projection () Rx Reynolds number bas3d on distance x and local velocity just outside boundary layer at position x (R'x) T boundarylayer transition parameter A constant for any1 chordwi*c, location of (cT ;2/3 projection CU C Subscripnt cr indicates conditions just :.fore transition from laminar to turbulent flo':. TEST TIET_ OLDS Models. The tests were conducted in the L.ranley twodimensional lowturbulence tunnel. The test section of this tunnel measures 3 by 7.5 feet and when mounted the models completely spanned the 5foot dimension. Tsts were ccreiducted with two typical laminarflow airfoils which hereinafter will be referred to as low drag airfoil 1 and lowdrag airfoil 2. On both airfoils C OT7 !i: TI AL IACA AC i No. L5J29a the position of rn.inuiou. pressure was at 0.7c; however, the pressure gradient was r.:ore favorable on lowdrag air foil 1 than on lowdrag airfoil 2. Lowdrag airfoil 1 was cambered for an ideal lift coefficient of 0.2 with a r.ean line of the typ a = 0.7; lowdrag airfoil 2 was a synmetrical section. Lxperirental pressure distributions are presented for the two airfoils at the given test con ditions in figures 1 and 2. Th3 models were constructed of wood and were painted and sanded ;o have aerodynam i cally smooth finishes. Zach r.odcl had a chord of 90 inches. Tests. The projections were cylindrical and con sisted of headless nails driven prp6enQicular into the surface until the desired height was attained. The pro jection heights were determined with an Ames dial gage. Tests 'were performed wlth one sparnis13 row of projections of constant size located at the desired chordwise station on the upper surface of the airfoil; the spanwise spacing was 3 inches in all tests. Projections of 0.055inch diameter and various heights w3re employed in the tests; check tests %.ere conducted with projections of 3.l15t.nch diaa.eter. he various combinations of projection size and chordise location tested with lowdrar airfols 1 and 2 are nipesented in table I. Drag data were obtained for airfoil Reynolds ni:umners varyir? from a:.proxi mately 3 x 100 to 10 x 10u for the airfoils with smooth surfaces and with each conrbiiat.Ion of projection size and location. The dra measurements 'w.ere made at a single spanuwse location by the .akazurvey retihod, a cort.plete description of vhi'ch appears in refere'.nce. 1. For each projections cctbination, the :.eynolds number at which the drag coefficient showed a definite increase over that o0 the smooth airfoil :a.s considered to be the critical Reynolds number. The dr.a data wo.re often Incon clusive, particularly when the projections were located at lare distances behind the leading. edge. In these instances the boundarylayer transition parameter (refer ence 7) was determined from measure,.ents of the velocity profile in the boundary layer. These neasuirements were made with a rac:: of totalpressure tubes (eference 7) located 2 inches behind the projectio.;s. The iReynolds number at which the boundarylayc.r transition parameter showed a definite increase was considered the critical value. The drag of the airfoil without projections was determined at frequent intervals to insure that all drag increments were caused by the projections and not some other surface imperfection. COCFIDEYTIAL CC' FIDLT IAL NACA ACd :Io. L5J29a Imperfect sheetmetal butt joints were simulated by grooves of several sizes cut into the surface of lowdrag airfoil 1. The various combinations of groove size and configuration that were tested are presented in table II. Grooves of Xplan form are illustrated in figure 3. The procedure followed in performing the tests of the airfoil with grooves was the same as that for the airfoils with projections. The various grades of sandpaper and carborundum pa'er used for determining the effects of sanding scratches on transition are indicated in table III. ilot only were various grades of abrasive used to determine the effects of sanding but the direction of sanding relative to the air stream .vas also varied. Enlarged photographs of sur face areas sanded with circular and crosshatched strokes are shown in figure 4. The method for applying the rough ness is shown in figure 5. The roughness area was pro gressively increased from a strip from 0.7c to 0.5c to include the part of the airfoil between 0.7c and the leading edge. Drag data were taken through the range of Reynolds number after each area was sanded. RE S ULTS Projections. The results of the investigation of the effects of surface projections on transition are pre sented in figures 6 to 10. The variation of section drag coefficient with airfoil Reynolds number for the two low drsp airfoils with smooth surfaces is given in figure 6. The increase in d:'ag coefficient for lowdrag airfoil 2 at the higher airfoil Reynolds numbers is believed to have been caused by the increase in airstream turbulence with Reynolds nuInber. The drag of lowdrag airfoil 1 was not affected by the increasing airstream turbulence because of the more favorable pressure gradient of this airfoil. The results of the analysis given later in the discussion appear to indicate that the turbulence of the air stream had only a secondary effect on the Reynolds nrundoCr at which the projections caused premature transition. The increments of dra;: induced by projections of various sizes and chord'.ise locations are plotted against airfoil Reynolds number in figures 7 and 8. The boundary layer transition parameter (reference 7) is plotted as a function of airfoil Re nolds number in figures 9 and 10. CONFI DE r T AL COTF7TDE ITIAL !ACA AC1r. AC L' J.9a C .'iDECi;T.M L 7 Surfac; grooves ind. .rndil sc'ra.tches. The results of the in"' ti'.to&.t ..f 'che fi'' rt .t f s;E',h "c te.r..' . end sandin s!r.tches on 'zrcnsi: ion rccetn.r till tie test conditions at wh ilch the results :.ere obtaiind are presented in tables IM and III, resT.ectivly. DISC'.LSIO '' Pi.D A..'L T:S Projecttions. 's has already buen indicated, the airfoil Rei,.olds rurnbcr at ;::icl. either the ira injree:.ent or boundarylayer transition paioraeter sho'.s a definite incre.se is considered to be the critical .eynolds numtesr at 4hich preT.attre trr::nsition occurs. Tie ,Eccuacy with which an inic:rense in either of these paramenters est b lishes the critical i;.polds nu:.iber is indictxred in fig urs 7(c) to 7f' Several va'l.ues ol the critical ReynoJds nurttib:er. were ottaincd "Jith each si:.e of projection at 0.20c on lo'.dra airfoil 1. Tn.c values of the critical RcynolsQ niu,:ier obtained w'it, ac!h conflgurat ion genera;lly ar withn 1 x Althcuih Leter .,r w'i:: 1n ri.:'lt be considered s sirable, the results presented are thou.'ht to rive a jood indication of uhe order of rnmari tude of the Reynolds numberi at .;:_hichi pre.iatire transition may Ole expected with .roje.:tions of a g lv'n size: and location. The eneral eff.:. ts of :2.rojctjion siz. and 3.ccation on transition ae ini,'cated L;' the e::peri.iental curves. As .niht be exemcted, the arltical airfoil ,eynjolds nunber et a specified chord ise sc.:tion decreases vwith incvreasin projection height and Ciamet r. As project icns of t given sizo are. moved toward the leading efi6, the cri..cal Renolds nurber d.ecr,~ ses until th. ,rocjec icons al...~st reach the stagnation point ani then 'egine to increuae. rie lower critical Reynolds numbers for p'rojections of given height at C.65"c cs compareid wth . those for projec tions of the sar:e height at 0.50c (fi.s. 7 and ) may be due to the combined effects of a zero or slightly unfa vorable pressure gradient and larder values of the boundarylayer Heynolds number. The results obtained with projections near the stagnation pcint are explained by the low velocity over the surface and the steep velocity Gradient at the sta.nation oc.oint. The increase in critical Reynolds rnumnbor as nroj ntions are placed near the stagnation point should not, however, be taken CC. FIDE;TIAL Digitized by the Internet Archive in 2011 with funding from University of Florida, George A. Smathers Libraries with support from LYRASIS and the Sloan Foundation http://www.archive.org/details/effectsofspecifi001ang :[ACI ACRL I'o. L5J2P a span.vise ro.i across the ai'rfoil did n':t hav: exactly the same h,eiht and shape and therefore the same critical Reynolds number. In any case, the sharp rise ii drag seems characteristic of the projections tesctd. An attenpt was mT.ade to correlate, the critical airfoil Rcy.nols nutilbers at which the projections caused prema ture transition with localflow conditions around the projection (reference 10). In such an analysis of the results, certain var ables are neglected, r rch as tunnel turbulence, ..re sire "radi3nt, bo .ndrrylay.:l . ynolds number, and th :ri' inal extnt of 1.u aLe flcew, except insofar as 'h':s '.l. ibles affect loy lf>. conditions. Simi lar it: of 1 >".lflVo conIlt i .'. a':c'.tt rejectionsns in similar fieli' L'f 'lov is o:bcaz:.ed if tL.e projections are ,eo,.,etr;call L ".llar and if the :;Troids numfoer of the flow about the pro.cctions is the Za::.e. In the fol lowinr anal;,is, '.:. c pr:ojectiLc. are ta'en to be suffi clit l, srr.ma: tc ;a;:5 cu/,i in the a':ene of the pro jection es_tsetislly ostant fro! the .'v.ri'sc. to a height equal to the heil.t of the projection. Cylindrical pro jections are geometrically si.ilar if their iinness ratios d/i: are the sae.:e. For each vailu. o01 / the locslflo.v pattern is therefore co::pletely ,.'.terr.lined by thel Re:,rolds number oi the fio' 0.Lcub t' :h p.rjection. The Rcrnolls numob'r 3f thl. flow about the projection lR ulI k was taI:tn to be . Tlhe critical nrojEct'on Reynolds number RF, was calculated iromt the experimental data corresponding to the critical airfcil Reynolds nuraber as indicated by the curves of figures 7 to 10. The Blasius relation for 5u/6y expressed in terms of the boundary layer thickness vns e..2ployed for calculating uk. The variat.c.n. of the Lcundarylayer thickness with chordwise position at "eynolds nznbars cf 3 10 and 6 x 10 is Clven in firure 11 for lowdrag airfoils 1 and 2. T'h boundarylayer thicknesses were calculated b'y means of equation (31) of appendix 3. The final equation for the critical projection Rey.nolds number is as follows: 2 Ri = 0.74/ Rt (1) "cr k6cr/ cr C ONF IL, 'IT L!.L CO:rIDEIITIAL COITF1DLrTIAL '/.CA ACR !To. L5J29a The derivation of equation (1) is given in appendix A and the method of reducing the experimental data to obtain Rker is outlined in appendix B. Values of r are plotted against the corre V kcr spending values of the projection fineness ratio d/k in figre 12. In this figure N R_k was used as a variable rather than Rkr because v/Tkcr is directly propor cr vcr tional to the critical projection height. The considerable scatter of the points shown in figure 12 apars to be unsystematic. The scatter may have Leen caused primarily either by the neglect of some of the variables previously mentioned or by exerimental inaccuracies in determining the critical airfoil Reynolds number for a given projection. Check tests of the critical Reynolds number have been shown to differ by approximately 1 x 10 In order to indicate the practical significance of the scatter of the data shown in figure 12, curves of projection height against critical airfoil Reynolds number have been calculated from the faired curve of figure 12 by means of the relation presented in appendix C. The comparison of the experimental points with the calculated curves (figs. 13 and !4) shows that virtually all the experimental points of the critical Reynolds number can be made to agree with the calculated curves by shifting the points not more than 0.5 x 10 on the Reynolds number scale and not more than 0.002 inch on the height scale. The results therefore indicate that, with the exception of the points obtained for projections close to the stagnation point, the effects of small pro jections on transition can be correlated with localflow conditions within the limits of experimental error in this investigation. The data from which \R was correlated with d/k were taken at Reynolds numbers from approximately 3 x 106 to 10 x 106. It is reasonable to believe, however, that the correlation would be valid at higher Reynolds nu.1:ters. A consideration of the parameters describing the boundary laysr indicates that conditions near the leading edge at high Reynolds numbers are equivalent to conditions farther back at low Reuiolds nur.mb~rs. The analysis pre~sernted is then particularly applicable when small vlueLSs of x/c COC, IDE:I AL EIACA ACR fo. L5J29a are considered with relation to high eynolds nu.bers. Inasmuch as all tests were made with one soanwise row of cylindrical projections, the critical Reynolds number may be somewhat optimistic for projections likely to occur in practice because of variations in the shape of the reo jections from the type investigated and possible combined effects of a nunter of projections at various chordwise locations. It should also be noted that the analysis was based on data in which the height of the projection was small compared to the boundarylayer thickness and can be expected to apoly only when this condition is fulfilled. Page has presented the results of experiments con ducted for the purpose of deterininin a criterion for the critical height of either a single arch or a flat ridge located in a spanwise direction at various chordwise positions on a lowdrag airfoil (reference 3) and a flat plate (reference 4). The criterion as determined from the airfoil tests was presented in the form of a corre lation of Rk with k/c, here c is the airfoil cr chord and k the ridge height. The values of . cr determined from the flatplate tests were correlated wjth : /L, where L is the original length of lamrinar flow. Although the drag data presented in reference 5 show that the critical Reynolds number is somewhat dependent upon the design length of laminar flow, the values of Rkc determined from these two investigations were plotted in figure 15 as a function of d/k, where d, in this case, was taken to be the ridge width. '1he parameter d/k is similar to the projection fineness ratio in that it describes the forrm or geometry of the boundarylayer disturbance. The vales of VR obtained from tests made with projections and ridges are not strictly comparable, since ridges represent a two dimensional disturbance and projections are th'ec dimensional. Values of Rkcr obtained from th' inves tigation of threedimensional projections rre, however, also included in figure 1i and show the similarity between the results obtained with the two distinctive types of disturbance. Although the values of Rkcr obtained with the two types of disturbance do not form a con tinuous curve, they are of the same order of magnitude. CONFIDENTIAL CONFIDETI AL TTAJA ACR No. L5J29a In order to check Fage's results, a strip of "Scotch" cellulose tape simulating a spanwise ridge was applied to lowdrag airfoil 1. Two thicknesses were employed and the results, which are plotted in figure 15, are in fair agreement with rage's results. "age also made tests with a spanwise wire located at various chordwise positions (reference 75). Wires of three diameters were tested; the value of d/1 was, of course, 1 in all cases. The values of Vcr obtained were 13.1, 15.5, and 8.6. Tani, T .::a, and : ttuisi (reference 6) conducted similar tests with wires located on an airfoil and a flat plate. The values of F1cr were 13 for a flat plate and 15 for an airfoil. Surface grooves and sanding scratches. The results of the investigation of the effects of surface grooves and sanding scratches indicate that within the range of Reynolds number from 3 x 10 to 10 x 10 at which these tests were conducted, the boundary layer is relatively insensitive to surface scratches. Only deep Xplanform grooves located near the ledi ,g edge caused premature transition (table II). No definite indications of pre mature transition were noticed with any of the types of sanded surface. The drag was soi. 'hat hiht. I.en the 1 surface was finished with TTo. 1 nd.pae.r, but there 2 was no definite break in the drag curve. It is thought that at higher Reynolds numbers than those at which the tests were made the t;)e of sanded surface would show a more definite effect upon transition. A comparison of the results obtained with various types of surface imp.er fections indicates clearly that, within a given rfnge of Reynolds number, the laminar boundary layer is much more sensitive to surface projections than to inS entations in the surface. C ONCLTS I OIS From tests cc!~ducted with two typical lowdrag air foils of 90inch chord to determiino the effects of surface projections, grooves, and sanding scratches on bou.rndtry layer transition, the following conclusions were reached: 1. The Reynolds number at which one row of spanwise projections causes prezmrature transition is primarily a C CO'rI DL:r IAL CC:iIDEIITIAL ITAC.A ACT' No. L5J29a CC'TIDEN TAL 1x function of the projection geonetry and the R?;e ynolds r1.nmbr based on the height of :h'~ projection and tiie velocity at the top of the projection, provided the height of the projection is small co.;pared *.ith: the bou.nd&.rlaYver thickness. 2. The laminar boundary layer is more sensitive to surface projections than co surface roovec or car.dini scratches. Langley Tem:iorial ..eron:.'utical La':ori or.'y national Advisory Co..imittee for aeronautics s Lan;ley Field, Ia. C OUI DNT AL 14 CO0 TDENTIAL IACA ACR :Io. L5J29a APPLD IX A ,/ k2 DRIVATIOI: OF Rk = 0.76 T) Rgc er 0r/ or The parameter Rk may be thought of as a Reynolds number based on the projection height and the boundary layer velocity at the top of the projection; that is, uk Rk = For small values of y, the velocity u in the laminar bojindary layer may be expressed as a linear function of y by du u =y then Sdu Uk = k so that k2 du Rk dy In order that Rk may be more easily calculated, the Dlasius relation (reference 11) for the slope of the laminar boundarylayer velocity profile is introduced u = 0.3352 (A2) dy x A The substitution of equation (A2) into the expression for the projection Reynolds number gives Rk = k20.2 (A5) V ~x C C:F i Di':7'AL NACA ACEP. o. L5J29a The Blasius expression layer thickness, which to the wing. surface at (reference is defined u whichh = U 11) for tie boundary as the distance normal 0.707, is 2 5x  Since in equation (AS) x . eIuation (AS) may be .written as k2 k = 0.7 6 ' but U therefore rp. = o.76L,,' 6 If the numerator and denominator are multiolied by 5 2 Rk = 0.76h ) ( If 6 is taken as the boundarylayer thick:ncss just before transition from laminar to tarliblent flour, then i:6 is tLhs critical boundarylayer Rer:olds number and equation (AL) may be written as follows: S.72 p Cr 0.76L.. ) Rg (A "ker 'cr'. Oc 1) ,) CONFIDENTIAL COFIDENT.IAL 'ACA ACR No. L5J2Qa APPLEDIX B DETB1C 'I NATIONN OF Rk, FROM EXPERI'.E!;IAL DATA From equation (A5), it is seen that the values of R6 and 5 whric' correspond to the airfoil Reynolds number at transition must be calculated. A suitable equation for 6 is obtained by assuming a Blasius velocity distribution and integrating the von Kir:nan momentum relation. The following equation results (reference 12): U \ J \Uo7) c A more convenient relation is obtained if equation (El) is multiplied by R' .2 p/Uo8.17 is/c 8.17 s R r)2 = 5.0 0 d (B2) i ,so \Uo The numerical value of equation (D2) is a constant for any chordwise position and need be calculated only once for each position at which tests are being conducted. The critical values of 6 and Rg may be calculated from equation (B2) when the critical Reynolds number Rcr has been experimentally determined. By definition, U R  but Uoc "cr therefore, U R' = Rcr (B3) Uc The boundarylay,_r thLickness is then obtained by dividing t:le square root of R', as determind from equation (35), into the constant 6/,'. In order to obtain R3, it is only necessary to multiply 6 by R'. All the variables in equation (A5) are now known, and Rkcr may be calculated. COTIFIDE T IAL CO ILr IDE NT IAL I;ACA ACR iTo. L5J29a DERIVATIOi1 OF C CI IDE iIAL AFPPEJlDIX C /' .555 Rcr Si'ce 2 or r r U r then R2 !,2C .7 l4 p,r  11 If both sides of equation (E11 arc mn.ltiplled by 6or 'rc?/12 ) 3/2 bi "2 r v c/2 R V cr kcr D.7. 64 1 poto o t b:ut o' r is pc.3,_ticn so thiat  /j o r r cr 5/2c /2 C. = 3/2 o'r constant for any riven chordwisJ 2/3 cr \ /.., ^=\J. vjh coe , o .'. TJ 0.7o../ COITFIDE lTIAL (Cl) (3u u / r 33\  \ 1 : 18 COIIDE;TIAL I!ACA ACI No. L5J29a RLE'ZhENCES 1. Abbott, Ira H., von Doenhoff, Albert E., and Stivers, Louis ., Jr.: Summary of Airfoil Data. NACA ACR No. L5005, 1945. 2. Abbott, .rank T., Jr., and Turner, Harold R.,' Jr.: The Effects of Roughness at High Reynolds Numbers on the Lift and Drag Ch.:asct eristics of Three Thick Airfoils. I..CA ACER .,. LL21, 19$4. 5. Fage, A.: The Effect of Harrow Spanwise Surface Ridges on the Drag of a La!inaprFlow Aerofoil. 5950, Ae. 2019, British A.P.C., July 13, 191,2. 4. Fage, A.* The Effect of i!arrow Spanwise Surface Riges on the Drag of a LaminarFlow Aerofoil. 6126, Ae. 2019a, British A.R.C., Sept. 22, 1942. 5. Fage, A.: The Smallest Size of a Surface Bulge, a Ridge or Hollow, Which Affects the Drag of a LaminarFlow Aerofoil. 6i45, Ae. 2148, British A.R.C., Jan. 22, 1943. 6. Tan., Itiro, Hama, Ryosuke, and Mituisi, Satosi: On the Permissible Rouighness in the Laminar Boi.nd'ry Layer. Rep. 'o. 199 (vol. XV, 13), Aero. Res. Inst., To:..yo Imperial Tniv., Oct. 1940. 7. von Dosnhoff, Albert E.: Investigation of the Boundary Layer about a Sr. metrical Airfoil in a Wind Tannel of Low Turbulence. L".C. ACR, Aug. 1940. 8. SciLubauer, G. B., and Skramstad, H. K.: Laminar BoundaryLayer Oscillations and Transition on a Flat Plate. NACA ACR, April 195. 9. Tollmien, *;. The Production of '_':'u ul ncr. AV.CA T: *,!. 60o 1931. 10. Schiller, L.: Stromung in Rohren. Handbuch der x.olimentai:liv;ik Bd. IV, 4. eil, Hydro und Aero';dna,]il.; Lidvrig Schiller, 3Irsg.; Akad. Verl'.,,gesellschaft m. b. 1. (LeirZig), 1952, p. 191. C C '7T DE' T TAL NACA ACR No. L5J29a 11. Frndtl, L.: Thae ;echanics of Viscous Tluids. The lit Plate. Vol. III cf A.i otAn,l..ic fheory, .iv. sec. I.L, V. c. ur..nrd, co., Julius rringer (TBerlin), 1Q55, pp. j4QO. 12. Jacobs, E. N., and von DoerLinhof, E.: Fcrmr.1l3 for Use in floundLaryLaC.e Calculations on Low rM,., :.'ings. .i. AC ., A'i '19 1. COn FTENr T IAL :,; I TIAL I:ACA ACR No. L5J29a TAELZ I CO!I:ATIONS OF SIZE AND CHORD.N'ISL LOCATION OF PROJECTIONS TESTED WITH LOVDRAG AIRFOILS 1 AND 2 Lowdrag airfoil 1 Lowdrag airfoil 2 Chordvise Diam. Height Chordwise Diam. i Height location (in.) (in.) location (in.) (in.) 0.o0) .o .08 0.005 .010 .015 0.20c 0.035 0.009 .20c .035 .05 .20c .055 .020 .20c .055 .025 .20c .0j .025 0.010 .015 .020 .025 0.010 .015 .025 .050 .QLtO 0.05c .05c .05c .050 0.20c .20c .20c .20c .20c .20c .20c .20c .20c .20c 0.20c .20c .20c .20c .20c .20c .20c .20c .20c .o3515 0.055 .035 .035 .055 0.055 .035 .035 .035 .055 .055 .055 .035 055 .05 I  t+ t 1 0.015 .015 .015 .015 .015 .015 .015 .015 .015 0.010 .020 .025 .050 0.010 .011 .012 .015 .020 .025 .035 .040 050 0.015 .016 .018 .019 .021 .025 .25 o.650 0o.o5 o.01o o.50c 0.055 o.oo2 .650 .055 .015 .50c .055 .c25 .65c .055 .020 .50c .c35. .030 .65c .055 .025 .50c .055 1 .'5 .650 .055 .000 .50c .035 . .65c .055 .05 .50c .055 .050 .65' .055 .oo i .65c .035 .c!5 5 'ATICl'.L ADVISORY CC :::I"TT FOR ,.LONAUTiICS CO0iiFIDEiiTIAL 0.0007c .0007c .0007c 0.058c .058c .058c 0.055 .055 .055 0.055 .055 .035 .o 355 o.55c .35c .35c .35e 0.50c .50c .50c .50c .55c .50c 0.035 .055 .055 .055 0.055 .055 .055 .055 .055 I CO!TFIDE'T IAL PIACA AC io. LSJ2' a COrFFIDEi'.TIL 21 TABLE I I F2?QCTS OF 'P SURFAC. GRO' '.S 0; DR.'G' CrEAACTLF ISTTCS OF LOiDR.iO AIPFOIL 1 ircove description ,FLemarks I  IC___  Sp"an ise :rve ro ve 0. 0 J.i. io .")asu.r:,ble increase in deep n id 0.00C5 i,. wide Idrag o'er that of s:nooth at .20,c wi 'n;g for ranie of ;eyniolds n .u.be from 5 10 to 10.7 >' 10 3nan,.'jise r'oc've 0 L,,'3 in. deep and 0.010 .n. v.ie Do. at .'C:c j ,Spanw.i.se grooves O.l'.. in. deep and C0.0 10 in. ',ide Do. E..t .C0Cc and C.ic.: . ~ . 2n. Span"nie groves 0.0 in. deen and 0.0'10 in. wide Do. at .20c, i.. %f c, and 0.0c 1 Span.wise roove e. 00I in. deep and O.C'l in. "'ide Do. at ".05 :,c Spant'ise .rccve ,J.CO. in. deep an'd c.021 in. .'ide i Do. at 0.C053  I _  Grooves 0 .C ini. deep Premratiure transition indi and U.C') i. 'ide in c,ted .;: sudden increase X:plan; form at ir. dra at a Reynolds approx. ,.,0, :j ua. er of 6.5 x 1C6 (see fi f) :JIIOL .AL ADVISORY COa'.:ITTTi FOR A Ci:A TICS CO FIDEITIAL NACA ACR No. L5J29a TABLE III S'r::0.F.Y OP DRAG RES'TjTS FROM TrLSTS OF LOSJDRAG AIRFOIL 1 FITIST D .'ITI! VARIOUS GRADES OF S~FDPAPZER A iD CARBORUITDTM. PAPLR All tests were tradee of lowdrag airfoil 1 at Feynolds numbers from approximately 5 x 106 to 10.57 x 10o6 Abrasive No. 320 carborundum paper No. 520 carborundum paper Ch ord: ise extent of roughness 0.7c to 0.Oc Sanding strokes Parallel to sanrea in windc steps as direction indicated ini figure 5 do perpendicular Sto wind direction Effect on drag No measurable increase in drag over that of smooth wing Do. F, _  No. 520 do 145 to wind Do. carborundum direction ?, per No. 280 do Parallel to Do. carboru.,di.u i wind p,alper direction No. 280 carborundum paper No. 260 carborundum paper do Perpendicular to wind direction Complete surface lErratic Do. Do. NATIONAL ADVISORY COC' ;ITTL. FOR A..OIAUTICS CCO i, .mTIAL ~i ~ C i ___t__ ~  t ~ I~L CONFIDENTIAL ! I I IACA AC;?, 'o. L5729a TABLE III Concluded SUMrM."RY OF DRAG RESULTS FRO"' TESTS Concluded Abrasive 4 To0. Cl i Com rlete carborundum surface ise Sanding of strokes eSS __I irratic I paper . 120 ddo Crosshatched carborundum i (see fig. 4) paper  . i* ___ No. 120 do Circular carborundm I (see fig. 4) paper 1 To. 12 sandpaper 0.7c to 0.5c IPerpndicular to wind direction  1 I No. 1 .o Erratic sandpaper . NATIONAL ADVISORY COMMITTEE FOP; AERONAUTICS CONFIDEIT IAL r . l CO DNFID ;;TIAL Effect on drag Hi :o r.ieasura.ble increase in drag over that of smooth wing Do. Do. Drag slightly high at Reynolds number of 10.57 x 10 Do. ___J NACA ACR No. L5J29a Fig.l 2.0 , I CONFIDENTIAL o tUpper surface i + wratr surface %4 1.0 .2 COMMITTEE F06AMA S UT S1.. .2 .3 .5 .6 .8 .9 2.0 Chordwise station, z/o Figure 1 . Preasure distribution of lowdrag airfoil 1 at a lift coerricient of 0.347. .8   NATIONAL ADVISORY CONFIDENTIAL 0 .1 .2 .5 .J .5 .6 .7 .8 .9 1.0 Chordwise station, z/o Figure 1 . Pressure distribution or lowdrag airfoil 1 at a lift coerficient of 0.h7. NACA ACR No. L5J29a o Upper surface + Lower surface CONFIDENTIAL I 4 NATIONAL ADVISORY COMMITTEE FOR AIONAUTICS CONFIDENTIAL L I I I ) .1 .2 .3 .4 .5 .6 .7 .8 .9 z Chordwiae station, x/o Figure 2 . Pressure distribution of lowdrag airfoil 2 at a lift coefficient of 0. 2W. 1.8 1.6 1.4 1.2 S1.0 0 0 5 .8 .6 .4 .2 0 Fig. 2 NACA ACR No. L5J29a Fig. 3 o S0 _j o0 O 0 w 0 0 0 o I a (i S o s o 0 it S00 4. Id 1rl \ I < " \~ 1 it \ s l^ \ ^ I! :! ^^ .^r> B *f ^,^^ ^ ^ >C ^ Is X s il . NACA ACR No. L5J29a Fig. 4a,b us 0 0 i a to : C, o0 43 4 zI z 2 O 2 4O * oO 0 a 544 0 4 3 10 0 C) N NACA ACR No. L5J29a Fig. 5 a a OUt 44 o 0o O $ I oo 0 on z PO 0 E4 O4 0 Fig. 6 NACA ACR No. L5J29a x 0 0a 0 z zO OS * r. * za z 0 z 0 * tZ id I  a P A el     _ 0 ' o' e+ r *   o NO 0 \ 00 ^" g \^ i  2 1 Z '4 PO 'aeoTOjjQOo Swap Unoroeg NACA ACR No. L5J29a .006 C o o.oB 005 (in.)J .06 .oo4 06 .o4 .001 0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 Reynolds number, R 1o (a) Projectionl of 0.035inch diameter at 0.0007c. aO Fig. 7ac AL .. .006  k (in.) .005o 0.015 .010 .004 .005 .00? 5.0 4.0 5.0 6.0 7.0 8.0 9 10.0o 1l. x 106 Reynolds number, R (b) Projections of 0.035inch diameter at 0.0580. .006 Teat .005 + 2 7 NATIONAL ADVISORY .004 C ITTEE FI *, AMITICS n fl  nn       i  /  .001 4.0 5.0 6.0 7.0 8.0 9.0 10.0 Reynolds number, R (c) Projections of 0.055inch diameter and 0.009inch height at 0.200. Figure 7 . Increment of section drag coefficient as a function of Reynolds number for projections of various mimes and cbordwise locations on lowdrag airfoil 1. 1] .0 x 106 CONFIDENTIAL a o Fig. 7df .00 o I I I Teat CONFIDENTIAL o 1 ____ .ook   .005   + 2 4003 L+ .001 .00 __ ___ C' ^ ^   S 3.0 NACA ACR No. L5J29a 5.0 6.0 7.0 8.0 9.0 10.0 l1.ox 106 Reynolds number, R (d) Projections of 0.035inch diameter and 0.015inch height at O.20e. . U U b I 1  .005 Teat e 1 + 2 .00    .002  .001 3.o 4.0 5.0 6.0 7.0 8.o 9.0 10.0 11. Reynolds number, R (e) Projections of 0.035inch diameter and 0.020inch height at 0.20c. 5.0 6.0 7.0 8.0 Reynolds number, R 0 x 106 (f) Projections of 0.055inch diameter at 0.20c. Fijre 7 . Continued. k Test (in.) o 0.025 1 + .025 2 x .02 NATIONAL ADVISORY COMMITTEE FOA AIONAUTICS CONFIDENTIAL I I I 1 .I......... 4 4.. J  I I. h I t 9.0 10.0 11.0 106 NACA ACR No. L5J29a .oo6 .005 .005 .003 .002 .001 0 3 Fig. 7gi k CONFIDENTIAL (in.) o 0.010 .015 .0 L.0 5.0 6.0 7.0 8.0 9.0 10.0 11. Reynolds number, R Projections of 0.035inoh diameter at 0.55o. 5.0 6.0 7.0 8.0 9.0 Reynolas number, R Projections of 0.055inch diameter at 0.35c. 5.0 6.o 7.0 8.0 9.0 heynoiis rumLer, R Projections or 0.0O5jnch diameter at 0.50o. Concluded. 10.0 11.0 x 10 ,0 x 106  k I (in.) o 0.020  * "02 __ __ __ < __ / __ __ __ __ __ __ __ __ q .006 .005 .002 .001 0 5 0 .oo6 .006 .005 .003 .002 .001 0 5 lin.)  e 0.010 + .015 S .02 NATIONAL ADVISORY CONFIDENTIAL COMMITTEE FOR AERONAUTICS ~ ~~~ ~ ~~~~~~~~~~~ f^~ ~ ^^^^^^^^ 1 ^ ** o^ ~^^ .0 (1) Figure 7. K f~ IfLI I I i i [[1 i:I I  . 10.0 11.0 x 10b J Fig. 8ac 5.0 6.0 7.0 8.0 9.0 Reynolds number, R Projections of 0.035inch diameter at 0.05e. NACA ACR No. L5J29a 11.0 x 10 11.0 i 06  Reynolds number, R (b) Projections of 0.035inch diameter at 0.20c. k I (in.) CONFIDENTIAL o 0.025 * .025 * .021 ___ .021 t .019 0 .018 A .016 & .016__ V 015  NATIONAL ADVISORY S .01g COMMITTEE FM AELMAUTICS .. .01 SI. IJ247d~7 . 0 x 106 Reynolds number, R (c) Projections of 0.015inch diameter at 0.20c. Figure 8 . Increment of section drag coefficient as a function of Reyn.:,is number for projections of various sizes and chordwise locations on lowdrag alrfoll 2. Sk) CONFIDENTIAL (in.) 15  o 0.015 + .010 0   I5   I= = B    e   5.0 0 44 4'.oo6 <I . .006  o 0 0 .004 0 .002  a * S.001  OL g .001 0 a 14 .006 .oo6 .005 .004 .005 .002 .001 '"" "' r .  1 NACA ACR No. L5J29a .5 1. .3 .2 .2 0 I I .24 m .1 0 Fig. 9a,b 2.0 5.0 4.0 5.0 6.0 Rern:.lds rn moer, R (a) Projections of 0.055inch diameter at 0.50c. i In.) 0 .040 .035  .030 o .025 A .020 015  P .010 CONFIDENTIAL f I NATIONAL ADVISORY coNNiTTEE FO ALMONAUTICS 2.0 5.0 4.0 5.0 6.0 7.0 Reynolds number, R (b) Projectlons of 0.03lnch diameter at 0.65c. Figure 9 . Boundarylayer tranaltion parameter as a function of Reynolds number for lowdrag airfoil 1 with projections of various sizes and chordriae locations. .r t 8.o 9.o0x 0 S (n.) CONFIDENTIAL  0 0.0h0  + .030 o .025 IJ i ==  ..O 8.0 X00 "''''"'"' 1.0 Fig. lOac NACA ACR No. L5J29a 1.0 2.0 4.0 5.0 6.0 7.0 8.0 Reynolds number, R (a) Projections of 0.035inch diameter at 0.05c. k (in. o O.0o + .035 Z .030 S .021 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 .Ox 106 0.ox 206 Reynolds number, R (b) Projections of 0.035inch diameter at 0.20c. k (in.) 0 0.050 + 0'  = .035 El .030 .020 NATIONAL ADVISORY CONNITTEE FOI AERONAUTICS CONFIDENTIAL 0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0x 1 Reynolds number, R (c) Projections of 0.035inch diameter at 0.50c. Figure 10 . Boundarylayer transition parameter as a function of Reyricids number for lowdrag airfoil 2 with projections of various sizes and chordwise locations. CONFIDENTIAL (in. S0 0.030 + .025 x .020 5. S.6 5 9 .64 S.2 [.1 0 i 06 NACA ACR No. L5J29a 0N o o0 0 *u g '8oSemg(oq .zjeTLzmpunog Fig. 11 Lir b 0 U 4 #4 a al. 0 0 O .aI 4 3 0 o to 04 C a 4 *P 0) i4 > a a eo o C S >** X;(>k ' en *^ *VS I (0 a * 0) r o c (mm Ix. Fig. 12 NACA ACR No. L5J29a 0  ,4 oc O .. 0 I #1 .1 5 Q0 O Q O 0 0 C 4 a a 2 1 (< 0 1 d& Z SI 00ri S0 0  S0 0 04+E 4[ .4 Al +_ 0 Ow t 2 W 0 0 SUa    f    *(4 a T *4 *o k 0 '4 O + Or0   I 0 0 0 0 0 0 0 0 P\ N H Nu 'to~ovj jequinu sptou~eU jsLBTLjvUDunoS  i  f . NACA ACR No. L5J29a 0 1.0 2.0 5.0 4.0 5.0 6.0 7.0 Critical Reynolds number, Rcr (a) Projections at 0.0007c. CONFIDENTIAL 0  Calculated 0 Experimental \srzrz 8.0 9.0 10.ox 10 .U( .06 6 C calculated + Experimental .05  S .04   I04 0 3 .: .05       c \ NATIONAL ADVISORY So COMMITTEE FOR AERONAUTICS .02 0 CONFIDENTIAL 0 x 1 0 6 0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.3 10.0 x1o6 Critical Reynolds number, Rcr (b) Projections at 0.058c. Figure 13 . Calculated and experimental values of maximum allowable projection height as a function of Reynolds number for projections of 0.055inch diameter at various chordwise locations on a 90inch chord model of lowdrag airfoil 1. " Fig. 13a,b NACA ACR No. L5J29a 1.0 2.0 3.0 4.0 5.0 Critical Reynolds 6.0 7.0 8.0 number, Rcr 9.0 l0.OX 10 ) Projections at 0.20c. .uo   Calculated B Experimental .05 .04 03  SNATIONAL ADVISORY L CONNITTEE FOR AERONAUTICS .02 a_ CONFIDENTIAL 0 I 1 I 1 6 0 1.0 2.0 5.0 4.0 5.0 6.0 7.0 8.0 9.0 l0.ox 10 Critical Reynolds number, Rcr (d) Projections at 0,35c. I I I CONFIDENTIAL SCalculated K Experimental I Fig. 13c,d 6 NACA ACR No. L5J29a Fig. 13e,f CONFIDENT AL Calculated 0 Experimental 0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 l0.x 106 Critical Reynolds number, Rcr (e) Projections at 0.50c. 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Critical Reynolds number, Rcr L 9.0 10.0 x 10 (f) Projections at 0.65c. Figure 13 . Concluded. .04 4.3 * .02 0 a .02 .01 0 Calculated A Experimental _a II]IL I NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS CONFIDENTIAL . l l NACA ACR No. L5J29a .02 .01 n .06 .05 .04 .05 .02 .01 n o i.o 2.0 5.0 4.0 5.0 6.0 7.0 8.0 9.0 1 Critical Reynolds number, Rcr (a) Projections 0.055 inch in diameter at 0.05c. 0.0 x 106 0 1.0 2.0 5.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 x106 Critical Reynolds number, Rcr (b) Projections 0.055 inch in diameter at 0.20c. FIijre 14 . Calculated and experimental values of maximum allowable projection height as a function of Reynolds number for projections at various chordwise locations on a 90inchchord model of lowdrae airfoil 2. I I I CONFIDENTIAL Calculated o Experimental 0 0 Calculated \ + Experimental NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS CONFIDENTIAL w Fig. 14a,b  V NACA ACR No. L5J29a 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Critical Reynolds number, Rcr 9.0 10.0x 10 (c) Projections 0.015 inch in diameter at 0.20c. 0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 11 Critical Reynolds number, Rcr (d) Projections 0.035 inch in diameter at 0.50c. Figure 14 . Concluded. CONFIDENTIAL Calculated u Experimental \, \ s Calculated 0 Experimental NATIONAL ADVISORY COMMITTEE FOI AEOIIAUTICS CONFIDENTIAL S I I 0.0 x 106 f Fig. 14c,d Fig. 15 NACA ACR No. L5J29a o 0 *nO _ 0 0. ._ E  0U 0 UN *40. + * N 4 0 oo < I II "I Q t 0 *o H 0 S z 4z L o( o, ( wr ca 0 00 aI IL 1  0 x0 "4 4 0 O a)4 d Cd 0B * ss 00 a 'S. ni 4 NH S t.00.4? c s. , 0 0 0 0 'I z oto.j eqgmnu epfouboe .sGsLAjspmuno9 UNIVERSITY OF FLORIDA I11 11 l 111 lllll1n 3 1262 08104 993 3 SINVERSITr OF FLORIDA . DOCUMENTS DEPARTMENT 1'.0 ,MARSTON SCIENCE LIBRARY 7 O. BOX 117011 S.NESVILLE, FL 326117011 USA 