Analog Computers

Reference / Paper · 1949

Determination of Rate, Area, and Distribution of Impingement of Waterdrops on Various Airfoils from Trajectories Obtained on the Differential Analyzer

Read the PDF (54 pp) ↗

NACA Research Memorandum (RM No. 9A05, February 1949) in which University of California researchers use a differential analyzer to solve the differential equations of waterdrop motion around Joukowski airfoils. Trajectories are computed for three airfoil/angle-of-attack cases, yielding the rate, area, and distribution of waterdrop impingement as functions of two dimensionless moduli; results are compared with impingement data for cylinders. The work demonstrates the differential analyzer as a practical tool for aeronautical icing calculations otherwise requiring laborious numerical integration.

Manufacturer
NACA
System
Differential Analyzer
Author
A. G. Guibert, E. Janssen, W. M. Robbins
Year
1949
Type
Reference / Paper
Language
English
Learning track
specific applications
Pages
54
Credit
National Advisory Committee for Aeronautics (NACA); University of California authors. Digitized via NASA Technical Reports Server.
  • Differential Analyzer
  • NACA
  • waterdrop impingement
  • airfoil icing
  • trajectory computation
  • differential analyzer applications

← Back to the Reference Library

Determination of Rate, Area, and Distribution of Impingement of Waterdrops on Various Airfoils from Trajectories Obtained on the Differential Analyzer

RM No. 9A05 -> b RESEARCH MEMORANDUM DETERMINATION O F RATE, AREA, AND DISTFUBUTION O F IMPINGEMENT O F TJ\rATERDROPS ON VARIOUS AIRFOILS FROM TRAJECTORIES OBTAINED ON THE DIFFERENTIAL ANALYZER BY A. G. Guibert, E. Janssen, and W. M. Robbins University of California N 66L80972 N 24: - i ,: LTHRU) ' $CODE1 ~ ( 3 ' t 4 \ : (NASA CR O R T M X O R AD N U M B E R ) . (CATEGORY) . - I I AiR~ j n ,-NATIONAL ADVISORY; COMMITTEE FOR AERONAUTICS WASHINGTON February 16,1949 I ^r- NATIONAL ADVISORY COMMTTTEX FOR AERONAUTICS RESEARCH r4EMORANIxTM DEFERMINATION QF RATE, AREA, AND DISTRIBUTION OF IMPINGEMENT OF WATWlwOPS ON VARIOUS AIRFOIIS FROM TRAJECTORIES OBTAINED ON W DIFFWENTIAL ANALYZER By A . G. Guibert, E. Janeeen, and W. M. Robbina SUMMARY The t r a j e c t o r i e s of waterdrops i n a i r flowing over a i r f o i l s a r e determined f o r three a i r f o i l - angle-of-attack combinations using the d i f f e r e n t i a l analyzer t o solve t h e d i f f e r e n t i a l equations of motion of the waterdrops. From these t r a j e c t o r i e s the r a t e of water impingement, t h e area of impingement, and the d i s t r i b u t i o n of impingement a r e determined as functions of two dimensionless moduli. Comparisons a r e made of the r a t e of water impingement on these a i r f o i l s and the r a t e of water impingement on cylinders. INTRODUCTION If a body of any shape and s i z e moves through a cloud, some of the waterdrops i n i t s path w i l l tend t o impinge on the surface of t h a t body over an a r e a which w i l l vary according t o the s i z e of the drops, the speed of t h e body, and s o f o r t h . Other drops o r i g i n a l l y i n i t s path w i l l be c a r r i e d around the body and w i l l n o t impinge- Studies have been made of the r a t e and d i s t r i b u t i o n of impingement of waterdrops on cylinders and two d i f f e r e n t a i r f o i l s by means of numerical i n t e g r a t i o n of the d i f f e r e n t i a l equations of the motion of the drops (references 1, 2, and 3 ) and on cylinders, spheres, and ribbons by s o l u t i o n of these equations on a d i f f e r e n t i a l analyzer (reference 4 ) . References 1 and 2 both made the assumption t h a t the velocity and s i z e of the drops were such t h a t Stokes' l a w of r e s i s t a n c e w a s followed. References 3 and 4 did not make t h i s assumption, which i s not applicable a t t h e v e l o c i t i e s of airplanes and f o r the drop s i z e s prevalent i n clouds. These references employed instead the experimentally determined drag c o e f f i c i e n t f o r spheres as a b e t t e r approximation t o the drag c o e f f i c i e n t of the drops. NACA RM No. 9AOg 2 I n the praoent study, the r a t e and d i s t r i b u t i o n of impingement of waterdrops on a symmetrical, 15-percent-thick, Joukowski a i r f o i l a t angles of a t t a c k of 00 (Case I ) and z0 (Case 11) and on a cambered (a = 1 mean l i n e ) , l?-percent-thick, Joukowski a i r f o i l a t 00 angle of a t t a c k (Case IV) are determined Using a d i f f e r e n t i a l analyze; f o r s o l u t i o n of the d i f f e r e n t i a l equations and employing the experimentally determined drag c o e f f i c i e n t of spheres t o approximate the drag c o e f f i c i e n t of the waterdrops. Figure 1 i s a drawing of t h e three a i r f o i l cases (Case I11 wa8 t o have been a study of the symmstrical, l?-percent-thick, Joukowski a i r f o i l a t 4' a n g l e of a t t a c k , b u t i t w a s decided t o study Case I T i n preference t h e r e t o . ) This p r o j e c t w a s under the general d i r e c t i o n of L. M. K. Boelter. The authors wish t o acknowledge t h e advice of John W. Hazen i n the d i r a c t i o n and implementstion of the research program and the a s s i s t a n c e of R . Peck and M. P o t t e r i n making the necessary conputati'ons f o r presentation of the data and a l s o the a s s i s t a n c e of the operators of the d i f f e r a n t i a l analyzer under E. Janssen and G. N. B r i t t l e . This work w a s conducted under the sponsorahip and with the f i n a n c i a l a s s i s t a n c e of the National Advisory Committee f o r Aeronautics. SYMBOLS a acceleration of rtrop, f t / s e c * A projected area of waterdrop, f t * C chord length of a i r f o i l , f t CD drag c o e f f i c i e n t of drop, [1J* E EM percentage catch t o t a l percentage f drag force, # M r a t e of impingement of waterdrops on a body, l b / h r f t span P velocity of drop r e l a t i v e t o air, f t / s e c r radius of drop, f t * Dimensionloss The abbrsviation, l b , represents pound mass; t h e symbol, #, represents pound force. -K-x- based on maximum 113 thickness of a i r f o i l [ 13 c *P Reynolds Modulus f o r drop a t r e l a t i v e v e l o c i t y P, [l] Rv Reynolda Modulue f o r drop at free-stream velocity, [l] 8 p o s i t i o n of Impingement on eurface of a i r f o i l , measured fron chord line, divided by chord length, [l] S furthest p o s i t i o n of impiwement on surface of a i r f o i l (lee.-, trajectory t e n t ) , measured from chord l i n e , divided by chord length>l] t time, sec ua v e l o c i t y component of air parallel t o chord line, f t / a e c Ud v e l o c i t y component of drop parallel t o chord l i n e , f t / s e c U free-stream velocity, f t / s e c va v e l o c i t y component of air normal t o chord l i n e , f t / s e c Vd v e l o c i t y component of drop normal t o chord l i n e , f t / s e c W l i q u i d water content of cloud, l b / f t 3 X distance from the a x i s normal t o chord l i n e which i n t e r s e c t s leading edge a t chord l i n e i . e . , distance f r o n y-axis), divided by chord length, d? , U’ u - = #, [l] b d dT distance from the axis p a r a l l e l t o chord l i n e which i n t e r s e c t s leading edge a t chord l i n e ( i . e . , diatance from x-axis), divided by chord length, [l] YO distance of a t r a j e c t o r y from the x-axis a t x = -a, divided by chord length, [l] Ya fd Ya mass density of a i r , l b / f d . 4 7d NACA RM NO. 9A05 mass density of drop, l b / f t 3 P absolute v i s c o s i t y of a i r , Ib/sec f t $ Scale Modulus, 9 C A , [1] 7 U Time Scale, t- [l] C' e angle of r e l a t i v e velocity vector from x-axis [1] U angle of a t t a c k of a i r f o i l , deg Vd Subs c r ip t a : A airfoil C cylinaer L lower U upper 1 f i r s t impinging t r a j e c t o r y 2 second impinging t r a j e c t o r y ANALYSIS I n a cloud, the motion of a waterdrop which results when a body moves through t h a t cloud with f i n i t e v e l o c i t y i s caused by t h e drag created by f l o w of tha displaced a i r r e l a t i v e t o the waterdrop. This r e s u l t i n g motion i s the same as if t h e waterdrop had bsen suspended i n a i r flowing over tha s t a t i o n a r y body with the same v e l o c i t y . Making a force balance on the waterdrop (see f i g . 2) one obtains: ' I I F = O = m - f ZF, = 0 = m -- f dud at COS 8 at - f sin e E F Y = o = m dvd whera f , t h e drag force, is: NACA RM me. 9AQ5 i 5 c and COS e = - ('a - Ua) - U a - Ud P - P * Multiply both sides by L at b d C 2 p- 2 -r Y-a u 3 2 r IJ Y f l S u b s t i t u t i n g i n equations ( 5 ) and (6) the relatioriships: 6 give s Equations (7) and (8) are the desired equations for the twodimensional motion of a waterdrop in an air stream flowing over a b o a For solution of the differential analyzer, these equations must be m g e d as follows: r J 1 NACA i3M No. YAO3 n I n -J d(P/U)2 2(P/U) (Inverse integrator) Knowledge of the magnitude of the quantities CDEP ( t h e ratio of the 24 actual drag coefficient to the drag coefficient given by Stokes' law) and the velocity components of the air stream, ka and fa, as a function of the location of the waterdrop relative to the body, must be available for the solution to proceed on the differential analyzer. The variation of c Bwas taken from table I in reference 4. 24 Plots of ia and ia, the velocity components of the air stream, as functions of position relative to the alrfoils under consideration were supplied by Ames Aeronautical Iaboratory, Moffett Field, California. Finally, having fixed .Jr, the Scale Modulus (presented in reference 5 ) , and Ru, the Reynolds Modulus of t h e drop based on free-stream velocity, solution of the equations c a n begin provided initial conditions for a trajectory are known. If it were possible to start the tradectory at infinite distance forward of the airfoil, there would be no question as to the initial conditions because the drop would have free-stream velocity at that distance. However, at a sufficiently large though finite distance ahead of the airfoil, the waterdrop still has essentially freestream velocity. It is then necessary merely to determine this distance and start the trajectory there. (See section ESTABLISHMENT OF INITIAL COiXDITIOKS ) - . As shown in figure 3 , waterdrops started at different points will have different trajectories. A waterdrop which has its trajectory tangent to the upper surface of the airfoil will start at some NACA RM No. 9AOg 8 position y = when a large distance ahead of the airfoil. Another drop at some position y = yoL when a large distance ahead of the airPoil will have a trajectory which is tangent to the lower aurface of the airfoil. A l l drops located between YW and 7% at this large distance ahead of the airfoil will have trajectories which intersect t h e airfoil surface, that is, the drops w i l l impinge on t h e aurface - specifically on that portion of the surface limited by the points of tangency of the tangent trajectories. Au. drops outside you 2 y . 2 ywill miss the airfoil. As mentioned previously, the area of impingement of waterdrops l i e s between the point of tangency on the upper surface and the point of tangency on the lower surface. Mstribution over this area can be found by determining additional trajectories starting from points intermediate between yolJ and yoL, such as yol and yo2 in figure 3. The differential analyzer also gives the drop velocities at the points of impingement. This information is incidental to t h e immediate purpose of this etudy, but is included with the more pertinent material in this report for possible future use. The more important asmptions which it has been necessary to make in arriving at the simplified probhm which admits of solution are: (1)A t a large distance ahead of t h e airfoil, the drops move with free-stream velocity (that is, at the same velocity aa the air) and with motion parallel to the free-etream path. (2) The flow of air around the airfoil is that of an ideal fluid without turbulence or compressibility. (The drag of the air on the drop is that of a fluid having viscosity.) (3) The drops are spherical. (4) No gravitational force acts on the drop. BSTABUS" OF INITIAL CONDITIONS In the study of waterdrop trajectories, the boundary conditions a r e that the waterdrops are traveling with free-stream velocity at x = -a (that is, at infinite distance ahead of the airfoil). At finite distances from the leading edge of the airfoil, the drops have velocity components and positions varying between those given by the free stream and the streamlines. For Airfoil Case I (shown at top in fig. l), the divergence of t h e streamlines is 0.15 percent at x = -3.05, 0.3 percent at x = -2.00, and 1.2 percent at x = -0.93. Since the divergence is so s m a l l at x zs -3.05 and even NACA KM No. 9A05 9 f at x = -2.00,postulating free-stream velocity and position for the d r o p at x = -2.00 should not 6ause great error in the trajectories. However, x = -2.00 is too great a distance for obtaining rapid results on t h e differential analyzer; x = -0.95 being about t h e ma~rimuqpermissible approach to $he airPail leading edge (for a scale of 20 in. per chord length on the output table). It waa determined on the analyzer that the assumption of free-stream values at x = -2.00, for a m a l l and intermediate 0 3 values of . (2-3, 2 , 2 gave values of y and yd at x = -0.95 which differed from the free-stream values by less than the sxpected precision of the analyzer, as seen in the following table: ), The deviation of from the free-stream value at x = -0.95 is nat inappreciable but it was determined in the course of the investigation that the results obtained on the analyzer were the same regardlem of whether id at x = -0.95 was chosen as the free-stream value or the streamline value. Further, if choice of free-stream values at x = -2.00 gave values of y and id at x = -0.95 which were still very close to free-stream values, then choice of free-stream values at any x further from t h e airfoil than x = -2.00 would give free-stream values of y and ya at x = -2.00 since the divergence in streamlines decremes as x becomes more negative and is a l r e a d y less than the expected precision of the analyzer at x = -3.05. For large values of $/F~J(&), choice of streamline values for y and yd at x = -2.00 resulted in obtaining values of y and id at x = -0.95 which differed from streamline values by less than the expected precision of the analyzer as shown in the following table. NACA FM N O * 9A03 10 6 */Ru = 2 I I x = -2.00 (Streamline values) Y 0.002 i -997 9 .00000 I = (-‘er) 0.002 9883 0 x = -o:g5 I I (Streamline Difference vdues) 0.002 0 .9882 .0001 0004 - 0004 Hence it would appear that for large values of $&, t h e initial conditions should be streamline position and velocity components. However, for large values of JI/Itv the positions of the drops whose trajectories are tangent to the upper and lower surfaces of the airfoil, respectively, are quite close together. At x = -0.93, the distance between the two positions (measured normal to the free-stream path) choosing streamline conditions differs by less than the expected precision of the analyzer from the distance obtained by choosing free-stream conditions.i On the basis of the above, free-stream values of drop position and velocity were taken as the initial conditions at x = -0.95 for all values of $/Rv considered. For Airfoil Case I1 (shown at center in fig. l), because of the effect of circulation, it was not possible to assume free-stream conditione at x = -0.95 for all cams, though the divergence of the streamlines was about 0.4 percent at x = -2.0 and about 1.4percent at x = -0.8. Preliminary trajectories were run from x = -0.8 to the airfoil surface for various values of $/Ru using free-stream co ditione as the initial conditions. For low values of $/Ru (2-3 and 2 - 9 the choice of freestream conditions as initial conditions seemed appropriate because the trajectories followed the path of the free stream for about 0.6 chord length before deviating appreciably and the y-component of velocity of t h e drop remained equal to the free-stream initial value for about the same distance. For higher values of $/Ru, the trajectories and y-component of velocity deviated from the free-stream values almost immediately, (about 0.1 chord length), indicating that free-stream conditions were not a suitable choice for initial conditions at x = -0.8. For these larger values of $&, the conditians at x = -2.0 were assumed to be free-stream conditions, and trajectories were run on the analyzer from x = -2.0 to x = -0.8 for various values of yo and I rJACA TIM No. 9A05 li d, f o r Jr/% = 2O, and 26. From these t r a j e c t o r i e s , the p o s i t i o n and v e l o c i t y components of a drop a t I = -0.8 were determined as functions of and of the p o s i t i o n of the drop a t x = -2.0. These data were then used aa the starting conditians at x = -0.8 for the determination of the trac)ectories from I = -0.8 t o t h e points of tangency o r impingement on the a i r f o i l Burface. +& Examinatim of the tFa3ectories and y-component of v e l o c i t y p l o t s = 2O, which were run from x = -2.0 t o x = -0.8 u s i n g freefor stream i n i t i a l conditions, revealed that t h e r e w a s l i t t l e deviation f o r about 0.3 chord length, an indication that choice of fme-stream conditions as i n i t i a l conditions a t x = -2.0 was valid. Choosing streamline conditions as initial conditions a t x = -2.0 = for p v e r e s u l t s which indicated that the waterdrops w e r e s t i l l following the streauiLine at I = -0.8 (xd = 0.964, f d = 0.0558; %a= 0 . 9 6 5 , i a = 0.056) and, consequently, that s t r e d i n e c o n d i t i o m w e r e probably more v a l i d than free-etmam conditlona as i n i t i a l condit i o n s a t x = -2.0. However f o r these l a r g e values of $/Rut the " i n i t i a l " positions of the drops whoae t r a j e c t o r i e s a r e tangent t o the upper and lower surfaces of the a i r f o i l , respectively, are quite close together. A t x = -2.0, the divergence of the streamlines i s about 0 - 3 5 percent, on t h e average, ( i n the region of the t r a j e c t o r i e s ) so postulating free-stream velocity and p o s i t i o n a8 t h e i n i t i a l conditions a t x = - 2 . 0 should not introduce too great an error even f o r the l a r g e valuea Of '#/Ru. 26 For A i r f o i l Case IV (shown a t the bottom i n fig. l),the "working" i n i t i a l conditiona, t h a t i s , those t o be used when starting t h e drop t r a J e c t o r i e s a t x = -0.8, w e r e detexmined by making preliminarg m a from x = -2.0 t o x = -0.8, (as was done f o r C a m 11) assuming t h e drops t o have free-stream v e l o c i t y and p o s i t i o n a t x = -2.0. From these runs, the p o s i t i o n and v e l c o i t y components of the drop a t r = -0.8 w e r e determined as functions of yo, t h e starting p o s i t i o n a t x = -2.0. The y-positions of the drops r e l a t i v e t o one another a t x = -2.0 are t h e same as a t x = -GO under the assumption that the drops have freeThe v a l i d i t y of t h i s stream v e l o c i t y and p o s i t i o n a t x = -2.0. JI/Etrr w a s substantiated, as f o r Case 11, assumption f o r most values of by examination of the t r a e c t o r i e s f o r t h e runs from x = -2.0 t o x = -0.8. Again, f o r 2-g < Jr/Ru < 26, the t r a j e c t o r i e s followed the free-stream path f o r about 0.3 chord length before beginning t o deviate and the y-components of the drop velocity did not change from the f r e e s t r e a m value (0) given them i n i t i a l l y over approximately the same distance. For higher values of \Ir/%, the choice of streamline conditions as i n i t i a l conditions seemed more v a l i d because t h e t r a j e c t o r i e s obtained followed the atreamlines even a t x = -0.8. However, f o r the same reasons given f o r Case 11, free-stream i n i t i a l conditions vere assumed even a t high values of $A. NACA RM NO. 9A05 12 RESULTS The d i f f e r e n t i a l - a n a l y z e r solutions of t h e equations of motion of the water”,-=ps were ir: t h e fern ef ~ l o t nnf t h e y-position of the waterdrop as a function of x and the x-component and y-component of v e l o c i t y of the waterdrop as a function of x, the distance ahead of the a i r f o i l leading edge. The y versus x p l o t s were drawn on an output t a b l e with a s c a l e drawing of the p a r t i c u l a r a i r f o i l mounted at one s i d e of the t a b l e t o e s t a b l i s h the x and y frame of reference. (See f i g . 3 . ) In obtaining t h e tangent t r a j e c t o r y , the analyzer was operated such that a trial t r a j e c t o r y , s t a r t e d a t some i n i t i a l y-position, yo, was drawn up t o the v i c i n i t y of t h e a i r f o i l surface. If the t r a j e c t o r y missed the a i r f o i l surface o r impinged a t some point s h o r t of t h e point of tangency, a new estimate of the i n i t i a l y-position of the tangent t r a j e c t o r y w a s made and a second t r a j e c t o r y run on the analyzer. This t r a j e c t o r y w a s usually close enough t o the tangent one t o permit i n t e r p o l a t i o n (or e x t r a p o l a t i o n ) , though occasionally ( i n the f i r s t runs f o r any a i r f o i l ) one o r two more t r i a l s might be necessary t o determine t h e tangent t r a j e c t o r y s a t i s f a c t o r i l y . Supplementary t r a j e c t o r i e s , with i n i t i a l yo values intermediate between the values f o r the t r a j e c t o r i e s tangent t o t h e upper and lower surfaces of t h e a i r f o i l , were run t o t h e i r points of impingement on the airf‘oil t o permit determination of the d i s t r i b u t i o n of the impingement. The r a t e of impingement of water on t h a t portion of the surface of a body bounded by the point of tangency (SL) on the lower surface and the p o i n t of tangency ( ~ u ) on the upper surface, i a (per u n i t span): = (You (= - Y0L)Uw - where Ayo yoTJ yo,-) is the distance between the i n i t i a l p o s i t i o n s of t h e upper and lower tangent t r a j e c t o r i e s , U is t h e free-atream velocity., and w i s the l i q u i d water content of the cloud. Equation ( 9 ) may be rewritten i n terms of an e f f i c i e n c y of water catch, EM, and t h e maximum catch based on the maximum thickness of t h e a i r f o i l ( t h a t is, the catch of t h e a i r f o i l when the waterdrop t r a j e c t o r i e s a r e along the free-stream path), then MA = UWEM (Maximum thiclmess) and (10) For the intermediate trajectories (see fig. 3 ) , the rate of impingement of water on that portion of the surface of the body bounded by the point of tangency (sL) on the lower surface and the point of impingement of the intermediate trajectory (1)is where yol being the initial position of the intermediate trajectory. From equations (10) and (12) MA1 MA -=- 7 E EM These equations are the defining equations for the quantities (%, E/%) which are plotted as functions of RU and 4f and which, with the plots of q~ and s~ versus Ru, $, permit the computation of the rate and distribution of impingement of waterdrops on a particular airfoil, given the necessary data to calculate Ru and 4f. Tables I, 11, and I11 are summaries of the data obtained using the differential analyzer for Airfoil Case I, Airfoil Case 11, and Airfoil Case IV, respectively. The values of xd and yd are the drop velocities at the points of impingement or tangency. This information is incidental to the immediate purpose of this study, but is included with the more pertinent material because of the possible need for it at some future time. These values of 5, and jTd are reliable except at high values of $/qT when the velocity components of the drop change rapidly near the nose of the airfoil. Figures 4, 9, and 18 are plots of EM, the total percentage catch versus the Scale Modulus, \c; with the Reynolds Modulus, Ru, as parameter, for Airfoil Cases I, 11, and IV, respectively. At low values of $, the curves of constant Ru approach a value of EM which is the m a x i m is attainable for the particular airfoil case. This maximum value of equal to the ratio of the projected frontsl thicbess of the airfoil to the maximum thickness of the airfoil (17 percent chord in each Airfoil Case). The values of yo upon which the values are based are estimated to be good to 0.0001, as f a r a8 the precision of the differential analyzer is concerned. Since is essentially the difference between two values of yo, a t worst the e r r o r i s about 0.0002. For values of EM 2 100 percent, the percentage e r r o r i s about 0 . 2 percent b u t f o r Z 10percent and lower, the percentage e r r o r i s 2 percent values of and higher. Hence, a t very high values of q/Ru (Jr/Ru = 26) when there may be some question of the v a l i d i t y of free-stream conditions as i n i t i a l conditions a t x = -2.0, the precision of the d i f f e r e n t i a l analyzer i s such t h a t even i f the correct i n i t i a l conditions had been used, the percentage e r r o r would s t i l l have been a t least 2 percent o r higher. The precision of the t r a j e c t o r i e s could be increased by enlarging the scale, but then consideration must be made of the runnhg t i m f o r each t r a j e c t o r y on t h e d i f f e r e n t i a l analyzer. The question of the s c a l e necessary to give the desired pracision while n o t causing t h e running time per t r a j e c t o r y t o be excessive i s one which i s posed whether the i n t e g r a t i o n be performed numerically o r on any kind of computer. Figure 5 is a p l o t of Su, the distance along the upper a i r f o i l surface t o t h e point of tangency of the tangent t r a j e c t o r y (that i s , the f u r t h e s t point of impingement on the upper surface of the a i r f o i l ) , as a function of with R u as parameter f o r Case I . SL, the distance along t h e lower a i r f o i l surface t o the point of tangency of the tsngent t r a j e c t o r y ( t h a t is, the f u r t h e s t point of impingement on the lower surface of the a i r f o i l ) , i s equal in magnitude t o % because the a i r f o i l i s symmetrical and a t a = Oo, f o r Case I. A l l curves of constant Ru approach the value %(= SL) = 0.283, the point on the surfaca a t which the a i r f o i l has i t s m a x i m thickness, as J, decreases (waterdrops increase i n diameter). For Case I, there can be no impingement bsyond t h i s point on e i t h e r surface. + Figures 1 0 and 11 a r e p l o t s of S, and SL versus and R u f o r Case 11. The maximum value of Su i s now 0-241 and t h a t f o r SL i s 0.321. These values correspond t o t h e case when +/Rv = 0 ( t h a t i s , when there i s no d e f l e c t i o n of the drop by the streamlines). Figures 19 and 20 a r e similar p l o t s f o r Case IV. The maximum value of S, i s 0.325 and t h a t f o r SL i s 0.220. A s before, these values correspond t o the case when there i s no d e f l e c t i o n of the drop by the streamlines (*/Ru = 0 ) . The data p l o t t e d i n f i g u r e s 5 , 10, ll, 19, and 20 were obtained by s c a l i n g off the distances t o the points of tangency on the output p l o t s of the d i f f e r e n t i a l analyzer. The l o c a t i o n of the exact point of tangency wa3 not accurately d e t e d n a b l e bscause of the thickness of the i n k - l i n e representing the t r a j e c t o r y and because of the l a r g e radius of curvature of both the t r a j e c t o r y and a l s o of the a i r f o i l surface when the t r a j e c t o r y i s tangent i n t h e region aft of the nose of t h e a i r f o i l . The precision of location of the points of tangency i s estimated t o be such as t o give a "maximum e r r o r " of about G . 0 0 2 ( i n terms of chord) a t the lower ends of t h e curves, W.005 ( i n terms of chord) a t the center, and kO.015 ( i n terma of chord) a t the upper ends of t h e curves. This maximum error i s n o t a measure of any inherent error i n the a n a l y z e r t r a j e c t o r y , b u t i s only a measure of t h e indeterminacy of the l o c a t i o n of the point tangency. The lowest and highest estimations of the l o c a t i o n of t h i a point were used i n determining the magnitude of the maximum error and it is probable t h a t the a c t u a l e r r o r was much less than t h e maximum. Figures 6 and 7, 1 2 t o 14, and 21 t o 23 are p l o t s of E/% versus s, the distance along the a i r f o i l surface ( i n terms of chord length) f o r various values of */% with % as parameter, f o r Case I, Case 11, and Case I V , respectively. The qjmntity E/% is t h e r a t i o of the percentage catch between the point of tangency an the lower surface and any point of impingement on t h e a i r f ' o i l to the t o t a l percentage catch between the p o i n t of tangency on t h e lower surface and point of tangency on t h e upper surface. . Figures 8, 15 t o 17, and 24 t o 26 are r e p l o t e of the data of the preceding paragraph, $/Ru now being the variable parameter and R u being t h e f i x e d parameter. I n the former f i g u r e s , v a r i a t i o n of Ru with $/Ru constant did not a f f e c t the d i s t r i b u t i o n greatly (except a t high values of '#/Ru) but v a r i a t i o n of $/Ru w i t h RU constant changes the d i s t r i bution g r e a t l y f o r all values of Ru, ae shown i n these l a t t e r figures. The dashed curve f o r $/% = 2-=(lf/Ru = 0) drawn i n figures 15 t o 17 and 24 t o 26 i s based on computed values and i e a l i m i t i n g d i s t r i b u t i o n which i s obtained when the drops are n o t deflected by the streamlines ( t h a t is, when the drops a r e very large). This dashed c m e i s n o t drawn i n f i g u r e 8 because i t almost coincides with the curve dram f o r $/Ru = 2- 6 . DISCUSSION OF R E m n T S A s can be seen from reference 5 , the t o t a l percentage catch, EM, - t h e mea of Impingement per f o o t of span, Su SL, and the d i s t r i b u t i o n of Fmpingement, E/&, a r e functions of $, the Scale Modulus, and Ru, the Reynolds Modulus. The range of 9 and RU used in these s t u d i e s encompasses most combinations of t h e following range of variables: Variable Maximum value Minimum value 2r U 100 400 20 microns 100 mph (20,000 f t ) 1.267 x 10-3 l b sec2/ft4 3.4 x 10-7 l b s e c / f t * 0.25 f t 1.94 l b sec2/ft4 L 7a CI C yd (sea x 10' 2.378 3.75 x 10-7 30.0 1.94 16 NACA RM No. 9A05 Figuras 4, 9, and 18, which a r e p l o t s of versus with Ru as parameter f o r Case I, Case 11, and Case m, respectively, show t h a t EM, based ori the maximum thickness of the a i r f o i l , becomes g r e a t e r than 100 percent when the projected f r o n t a l thickness of the a i r f o i l becomes g r e a t e r than the maximum thiclmess of t h e a i r f o i l , as it does f o r Case I1 and Case IT. The shape of the curve9 of constant Ru i s t h e same, i n general, but the slopes tend t o d i f f e r a t t h e upper and lower ends. The f i v e values of the parameter % f o r Case I bracket t h e f o u r values f o r Case I1 and Case I V . The reduction i n number of values was d e s i r a b l e because the number of runs w a s correspondingly reduced while the range of variables was s t i l l encompassed, f o r the most p a r t . I n order t o b s able t o compare f i g u r e 4 d i r e c t l y with figures 9 and 18, the dashed l i n e s f o r the intermediate values of Ru were obtained by i n t e r p o l a t i o n . Cornsarison of f i g u r e 5 with f i g u r e s 1 0 and ll shows t h a t a t an a n g l e of a t t a c k of 20, t h e symmetrical 15-percent-thick Joukowski a i r f o i l e x h i b i t s points of tangency of t h e tangent t r a j e c t o r i e s which a r e c l o s e r t o the loading edge on the upper surface and f u r t h e r from the leading edge on the lower surface than f o r the same a i r f o i l a t angle of a t t a c k of 00, as was t o be expected. Also, comparison of t h e d i s t r i b u t i o n curves i n figures 6 t o 8 with those shown i n f i g u r e s 1 2 t o 18 shows t h a t the curves i n the l a t t e r a r e n o t symmetrical about the point s = 0 and, consequently, t h a t some 60 t o 80 percent of the t o t a l catch impinges on the lower aurface of the a i r f o i l inertead of the catch being d i s t r i b u t e d evenly between upper and lower surfaces. Inspection of the d i s t r i b u t i o n curves f o r Case I V (ftgs. 21 t o 23 o r f i g s . 24 t o 26) shows t h a t from 50 t o 60 percent of the t o t a l catch impinges on the upper surf'ace of the cambered a i r f o i l except when the very high. I n t h i s instance, about drops a r e small and the velocity 60 percent of the catch I s on the lower mrf'ace of the cambered a i r f o i l . (m) I n general, the d i 8 t r i b u t i o n curves f o r all three a i r f o i l cases show t h a t f o r a given value of $/Ru the e f f e c t of varying % is n o t too g r e a t b u t t h a t f o r a given value of % the e f f e c t of varying $/Ru i s q u i t e great, t h a t i s , drop s i z e i s r e l a t i v e l y more important than v e l o c i t y i n determining the d i s t r i b u t i o n of catch. Figures 27 t o 29 show comparisons of t h e rate of water impingement f o r the respective a i r f o i l s t o the rate of water impingement on two cylinders; one with a diameter equal t o twice t h e radius of the leading edge of the a i r f o i l , the other with a diameter equal t o the maximum thiclmess of the a i r f o i l (15 percent chord). The comparisons &re made f o r a low and a high value of Ru. A t high values of $ the former comgarison i s somewhat b e t t e r , whereas a t low values of $, the l a t t e r comparison i s much b e t t e r . This was t o be expected because l a r g e waterdrops ($ low) are not deflected g r e a t l y by the a i r flow and t h e catch p e r f o o t span is dependent only on t h e projected f r o n t a l thickness which i s the same i n the l a t t e r instance previously mentioned. CONCLUSIONS 1. The rates of water impingement on the t h r e e airPoil cases studied cannot be determined satisfactorily by assuming these rates to be equal to the rates of water impingement on cylinders except for limited ranges of $, the Scale Modulus. However, they can be determined within f25 percent for values of d( between 1 and about 100 or 10,000 (depending upon by aasuming the rates to be equal to the rates of water the value of impingement on a cylinder whose radius is equal to the maximum thickness of the airfoil. m), 2. With respect to distribution, the effect of drop size is greater than the effect of velocity. 3 . Increase of angle of attack of a symmetrical 15-percent-thick, Joukowski airfoil from a = Oo to a = 2 O , or change from a symmetrical l?-percent-thick, Joukowski, to EL caoibered, a = 1 mean line, 15-percentthick, Joukowski airfoil, does not change the rate of water impingement greatly (especially at low Jr and high Ru) but does change the area of impingement and the distribution of impingement to a greater extent. ~ . Department of Jbgineering '-University of California Lo8 Angeles 24, Calif., October 5, 1948 18 NACA RM No. 9AOg 1. Glauert, Muriel: A Method of Constructing t h e Paths of Raindrops of Different Diameters Moving in the Neighbourhood of (1)A Circular Cylinder, (2.) an Aerofoil, Placed in a Uniform Stream of Air; and a Determination of the Rate of Deposit of the Drops on the Surface and t h e Percentage of Drops Caught. R. & M. No. 2025, British A. R. C., 1940. 2. Kantrowitz, Arthur: Aerodysnamic Heating and the Deflection of Drops by an Obstacle in a n Air Stream in Relation to Aircraft Icing. NACA TN No. 779, 1940. 3. Bergrun, Norman R.: A Method for Numerically Calculating the Area and Distribution of Water Impingement on the Leading Edge of an Airfoil in a Cloud. NACA 'I" No. 1397, 1947. 4. Langmuir, Irving, and Blodgett, Katherine B.: A Mathematical Investigation of Water Droplet Trajectories. General Electric Co. Rep., 1943. (Also available as Army Air Forces Tech. Rep. No. 5418 and as Dept. of Commerce Pub. €9 No. 27565.) 5. Boelter, L. M. K., Young, George, and Tribus, Myron: The Limitations and Mathematical Basis for Predicting Aircraft Icing Characteristics from Scale Model Studies. Section V of Army Air Forces Tech. Rep. NO. 5529, NOT. 6, 1946, pp. 63-78. TABLE I 4/27/48 1-1-3-1 4/27/48 1-1-3-1 0 15sThlok -trloal a- 0 1 1 I I i i i i i I 1 I I i i i i i 2l 27 0.074 upper” . 0.265 1.0 0 0.148 1.0 -0.074 Lover+ -0.265 1.0 0 0 0 0.273 0 . 9 ~ 0.004 4/22/48 1-1-4-2 Z3 Z9 0.074 Upper’ 4 f22 /48 1-1-4-2 23 29 -0.074 Lower+ - 0 . 8 3 0.99 0.148 1.0 -0.W4 0 0 1.o I 4/22/48 4/22/48 1-1-5-3 25 211 0.072 Upper’ 0.262 0.m 0.013 O.IW+ 1-1-5-3 25 211 -0.072 Lover+ -0.62 0.997 -0.013 0 1-2-2-1 22 25 0.073 Uppei* 0.273 1.0 0.012 0.146 1-2-2-1 22 e5 -0.073 Love+ -0.273 1.0 -0.012 0 0 . 1.o 4 0 1-2-3-1 24 27 0.070 Upper’ 0.244 1.005 0.023 0.140 1.o 1-2-3-12 24 27 0.045 Upper 0.068 0.9 0.013 O.ll5 0.821 1-2-3-11 24 27 0.020 Upper 0.021 0.962 0.009 0.Op 0.643 1-2-3-11 24 27 -0.020 Lover -0.021 0.982 -0.009 0.050 0.357 1 0.933 I I 4/23/48 1-2-5-21 I 2’ I 9I 0.020 I Upper I 0.023 I 0.931 1 0.029 I 0.078 I . XTangent T r a j e c t o r y . 0.672 I ~~ ~ NACA RM No. 9AO5 20 TABLE I - Concluded YATWWOP TRAJECTORY VALUE O B T m Pam D Smtrlcal m I A L AApLyzw FOR JOUKOWSKI AIREUIL, 15%Thick a Date Run No. 9 Ro yo 4/30/48 RR 1-4-1-3 26 Z3 0.0255 u p p + 0.321 0.0510 0.078 0.870 - 4/21/48 1-4-1-22 $ 23 0.018 Upper 0.031 0.693 4/21/48 1-4-1-21 $ Z3 0.008 Upper 0.010 0.69e a .061 Lower -0 . O ~ O 0.698 -0.061 Lower -0.19 -0.031 0.693 - Love+ -0.078 0.870 -0.321 5/4/48 FR 1-4-4-1 212 Z9 I I 0' a .iw 0 .Ob35 0 J3335 o .657 - 0.0175 0.343 0.147 0 0 I Upperi 0.073 0.628 0.042 Love+ -0.073 0.628 0 Uppe+ 0.050 0.681 a A76 0.0290 1.o Uppar 0.020 0.572 0.198 0.0245 0.a5 1-0 0 Upper I 0.009 10.563 I 0.109 I0.0195 I 0.672 Lover -0 .Om 0.572 o. ~ 4 5 0 J55 Lover* -0.050 0.681 -0.476 Upperi 0.052 0.741 Lovelx -0.052 0.741 -0.451 ~ . I ~ I -0.198 0.280 0 J93 I 0 0 1.o 0.451 0.030 0 0.200 0 ~ I I I 1 0.022 10.329 I 0.469 IO.0080 I Love+ I -0 .022 I O .329 I -0.469 I 0 I I I I I Upper, I I 1.o 0.053 0 I I NACA RM NO. 9 ~ 0 5 21 TABLE I1 NACA RM No. 9AO5 22 TABLE I1 - Concluded Spmetrical I a = 2' 15% Thick 6/30/48 2-3-4-2U 21° 21° -0.0743 Upperf 0.063 0.914 0.424 0.0617 1.Oo0 6/30/48 2-3-4-2L 21° 21° -0.1360 Lower* -0.118 0.902 -0.195 O.oo00 O.Oo0 I 2-4-2-2U I I 26 I-0.1005 I Upper* I 0.038 10.767 I 0.528 1 0.0377 I 7/1/48 I 2-4-2-lD 129 1 Z6 )-0.1080I Upper I 0.009 10.553 1 0.303 I 0.0297 I 7/1/48 g9 1.ooO 0.603 7/7/48 2-5-2-9 212 26 -0.1243 Upper* 0.003 O.Og0 0.434 0.0035 1.ooO 7/7/48 2-5-2-5 212 26 -0.1278 Lower* -0.015 0.246 -0.235 b.OKXl O.Oo0 7/7/48 2-5-3-4U 214 28 -0.1254 Upper* .O.oOe 10.103 0.506 0.0021 1.000 7/7/48 2-5-3-4L 28 -0.1275 Lower* -0.014 10.175 -0.295 O.oo00 O.Oo0 214 ---I *T'Tangent Trajectory. 0.251 0.802 I 2-4-2-ZD 129 I 26 1-0.1155 I Lower I -0.004 10.542 I 0.132 1 17.0232 I 7/1/48 0.411 0.023 0.014 NACA RM No. 9A05 23 TABU m b . 7/13/48 4-2-3-2U 25 '2 0.0775 Upper' 0.275 L O P 0.0% 0.1375 lxxxl 7/13/48 4-2-3-u) 0.0503 Uppr 0.092 O.g@ 0.042 0.m3 0.8M 0.0225 Upper 0.034 0.g6 0.09 O.Ote5 0.600 25 28 - 3 . ~ 4 5 - Upper 0.001 0.92 0.0017 0.0555 0.404 25 '2' 7/13/48 4 - 2 - 3 4 25 7/13/48 4-2-3-9 Z8 0.1005 0.8J7 0.0745 0.598 0 . 0 ~ 0.402 0.02% 0901 o.oo00 0.000 0 -917 I . 7/15/40 4-3-3-20 1 Upper Z8 2' -0.0165 7/15/48 4-3-3-9 '2 2' -0.03b Lasr 7/15/48 2' '2 -0.0510 Lasr 4-3-3-4D 7/15/48 4-3-34, 28 28 -0.068~ b r + 0.1147 1.000 o.oo00 oiwo 0.765 NACA RM No. 9 A 0 5 24 T W I11 - Concluded WA'ENBOP TRAJECTORY VALUES CBTAINEU FROM DIFFWmIAL ANALYZER FOR JOUKOWSKI AIKFOIL, Cambered 7 16 48 44-2-31 29 Tangent Trajectory. a = 1 Mean Line 15% Thick a = Oo NACA RM No. 9A05 25 . 0 C II II 3 3, ) 3 II d 3 h x0 4 4 0 g 8 h xc) Yl 1 0 $ I a, a, h k a, a, k jI!i l a, a, Fr k 26 Drop velocity ‘d Figure 2.- +X Diagram of velocity components of air stream and waterdrop. 7d 3 G Q) k Q) 2 a Q) 9 w 0 rn 4 0 4 a 4 3 a 4 3 0 E 0 k w G 9 4 cd cd 4 a d cd 0 EE-l I c6 . 2 .I+ k NACA RM No. 28 0 4 0 9A05 us .r( Frc . 8 0 us c? 0 ? 0 us 0 In 1 1 0. 0 0 0 0 30 NACA RM No. 9A05 32 NACA RM No. 9A05 33 , NACA RM N o . 34 In c‘? 0 *. In 1 0 1 0 9A05 .