Leading Edge Retraction as a High Lift Device Citation Doss, Robert Franklin (1955) Leading Edge Retraction as a High Lift Device. Engineer's thesis, California Institute of Technology. doi:10.7907/PV47-6A63. https://resolver.caltech.edu/CaltechETD:etd-12102003-110111 Abstract A two-dimensional investigation was carried out in the Merril Wind Tunnel at the California Institute of Technology to determine the effect on low speed lift of retracting the leading edge of a thin, circular-arc airfoil. Several configurations were tried, some with a spanwise slot milled into the upper surface so that the leading edge recess formed by retraction could be utilized as an air intake to improve the flow. A comparison was made between the configurations with the leading edge retracted to various positions and the basic airfoil. The investigation showed that leading edge retraction caused a linear loss of maximum lift proportional to the percent reduction in chord up to a critical position where lift and the stalling angle of attack increased abruptly. Thereafter, maximum lift was reduced at a rate higher than the chord reduction. The effect of the slot was negligible. The critical position phenomenon warrants further study. Item Type: Thesis (Engineer's thesis) Subject Keywords: Aeronautical Engineering Degree Grantor: California Institute of Technology Division: Engineering and Applied Science Major Option: Aeronautics Thesis Availability: Public (worldwide access) Research Advisor(s): Stewart, Homer Joseph (advisor) Roshko, Anatol (advisor) Group: GALCIT Thesis Committee: Unknown, Unknown Defense Date: 1 January 1955 Record Number: CaltechETD:etd-12102003-110111 Persistent URL: https://resolver.caltech.edu/CaltechETD:etd-12102003-110111 DOI: 10.7907/PV47-6A63 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 4923 Collection: CaltechTHESIS Deposited By: Imported from ETD-db Deposited On: 12 Dec 2003 Last Modified: 28 Jun 2023 21:09 Thesis Files Preview PDF (Doss_rf_1955.pdf) - Final Version See Usage Policy. 3MB Repository Staff Only: item control page

LSADING EDGE RETRACTION AS A HIGH LIFT DEVICE Thesis by Robert F. Doss Lieutenant, United States Navy In Partial Fulfillment of the Requirements For the Degree of Aeronautical Engineer California Institute of Technology Pasadena, California 1955 The author wishes and Dr. Anatol Roshko worker in most of the excellent assistance. ACKNOWLEDGEMENTS to express his appreciation to Dr. H. J. Stewart for their interest and guidance, and to his co- investigation, Lt. Carl Birdwell, USN, for his il SUMMARY A two-dimensional investigation was carried out in the Merril Wind Tunnel at the California Institute of Technology to determine the effect on low speed lift of retracting the leading edge of a thin, circular~-are airfoil. Several configurations were tried, some with a spanwise slot milled into the upper surface so that the leading edge recess formed by retrac- tion could be utilized as an air intake to improve the flow. A comparison was made between the configurations with the leading edge retracted to various positions and the basic airfoil. The investigation showed that leading edge retraction caused a linear loss of maximum lift proportional to the percent reduction in chord up to 6 critical position where lift and the stalling angle of attack increased abruptly. Thereafter, maximum lift was reduced at a rate higher than the chord reduction. The effect of the slot was negligible. The critical position phenomerm warrents further study. Part II. Ill. IV. idi TABLE OF CONTENTS Title Acknowledgements Summary Table of Contents List of Mgures List of Symbols Introduction Apparatus Experimental Procedure Results and Discussion . Conclusions and Recommendations References Table I Figures Page ii iii iv vi 16 17 18 19 Figure nN 0 ONY NHN WA F&F WwW 10 11 12 13 14 15 16 17 18 20 21 22 iv LIST OF FIGURES Title Model Components and Leading Edge Positions Partial Sectional View of Model Leading Edge Slot Configurations Model and Balance System End View of Model in Test Section Three-quarter View of Model in Test Section Balance System Modification Pressure Survey Probe Set-up Static Pressure Probe and Set-up The Effect of Leading Edge Retraction on Maximum Lift for Configuration la2 The Effect of Leading Edge Retraction on Maximum Lift for Configuration 142 The Effect of Leading Edge Retraction on Maximum Lift for Configuration 143 The Effect of Leading Edge Retraction on Maximum Lift for Configuration 23 Cero vs.x for Configuration lal Ceo vs. for Configuration 1a2 Cho vs. ~ for Configuration 1b2 Cho vs. Cc for Configuration 142 CRO vs. x for Configuration 143 Chi vs. % for Configuration 23 Slot Effectiveness for Configurations 143-0 and 143-4 Slot Effectiveness for Configurations 1¢c2-0 and 1¢2-4 Probe Tuft Survey Sketch for Configuration 143-4 Page 19 20 21 22 23 23 24 24 25. 26 27 28 29 30 31 32 33 34 35 36 37 38 Figure 23 24 25 26 LIST OF FIGURES (Continued) Title The Effect of Angle of Attack on Pressure Distribu- tion of Configuration 143-4 The Effect of Leading Edge Retraction on the Upper Surface Pressure Distribution of Configuration 1¢2 at = 10° Vortex Hypothesis The Effect of Slot Size and Configuration at the Fully Retracted Position Page 39 Lo 41 42 vi LIST OF SYMBOLS section lift coefficient section lift coefficient based on 10 inch chord section lift coefficient based on actual chord maximum section lift coefficient angle of attack angle of attack at stall pressure coefficient, upper surface I. INTRODUCTION Sharp leading-edged, thin airfoils have poor low speed characteristics, particularly maximum lift coefficient. Recently these airfoils have been introduced to advance the supersonic capabilities of fighter aircraft. In order to approach reasonable landing speeds the airfoils have had to be modified to incorporate high lift devices such as nose flaps, slats or boundary layer control, These devices have several disadvantages. Wose flaps and slats on thin airfoils must employ partially external actuators to provide and support the moment during low speed operation, but during high speed flight the actuators contribute considerably to drag. Boundary layer control is usually dependent on engine operation. Supersonic air- craft should have a “dead stick" landing capability to take advantage of the excellent glide ratio and the usual high altitude and high speed of operation. Boundary layer control further imposes a constraint, though of less importance, on engine performance in that maximum thrust cannot be developed at both the high lift and high speed configurations. The dis- advantages of presently employed devices prompted this investigation into the potentialities of leading edge retraction, a mechanically simple opera- tion. Thin airfoil stall is characterized by a separated region or "bubble" at the leading edge which grows chordwise with increasing angle of attack until it reaches the trailing edge and the wing stalls (Ref. 1). The separation at the leading edge is caused by the inability of the flow to negotiate the very small nose radius. The mechanism of reattachment of the flow is not fully understood (Ref. 1), but it seems logical that the flow is aided by a vortex pattern which appears within the bubble. It occurred to the author that by retracting a small portion of the -2— leading edge the flow might be turned from the stagnation point to the upper surface in stages. Interaction between the edges formed by the retraction might simulate a greater nose radius and cause a delay in stall. Another possibility of flow improvement at high angle of attack might exist in retracting the leading edge within the wing to form an air intake for re-energizing the confused flow within the bubble. The obvious disadvantage to leading edge retraction is the loss in wing area. There- fore, to make a significant contribution to low speed lift the retraction must increase the lift coefficient based on the original chord. The investigation was conducted in the Merrill Wind Tunnel of the Guggenheim Aeronautical Laboratory at the California Institute of Tech- nology during March and April, 1955. II. APPARATUS The tests were conducted in the Merrill Wind Tunnel, a closed-circuit type with a 32 by 45 inch test section and a 6:1 contraction ratio. Power was supplied by a 75 H.P. constant speed electric motor driving a three- bladed electric pitch controlled propeller giving a speed range of 0 to 180 mph, During the tests there were three 32-mesh wire screens mounted at the entrance to the contracting section to decrease turbulence. The model tested was a circular are 6 percent thick airfoil of 10 inch chord and 24 inch span. The model was fabricated from full span components in order to permit various configurations. The components were contoured from brass on a horizontal planer and hand finished with a file and fine emery paper. The base of the model consisted of the aft 50 percent of chord and a tongue. The various configurations were achieved by bolting upper and lower contoured sections to the tongue, leaving a space for re- tracting the leading edge. The first configuration provided the largest recess. Thereafter, the configurations restricted the recess more each time and the leading edge piece was then milled in thickness to fit. This method prevented retesting a configuration after an alteration had been made but it permitted several configurations with only one leading edge piece, saving considerable time and money. Figures 1 and 2 show sectional views of the components. Significant chordwise dimensions are listed in Table I, A spanwise slot was incorporated in one upper-surface section. The slot was maintained by eight 0.0625 inch brass fingers which were secured in milled recesses in the fore piece by solder. The fingers were secured to simultaneously cut recesses in the main piece by glyptol glue. The slot —Ya was altered in shape once and in size several times during the tests (Fig. 3). The size change was achieved by shifting the upper leading edge out or in. This changed the uvper chord about 0.010 inches, but the effect on curvature was negligible. The model configurations were designated by a four-digit descriptive number as follows: (top piece) (slot configuration, if any) (bottom piece) dash (leading edge position). There were two top pieces, numbers 1 and 2, and three bottom pieces, numbers 1, 2, and 3. Top piece number 1 had four slot configurations: 2, b, c, and d. The leading edge posi-~ tions were described by numbers 0 through 7, by a position called "flush! which indicated when the leading edge was flush with the top piece edge, and by a position called "basic" which was the fully extended position making up the basic circular-arc airfoil. Intermediate positions were described by hundredths of the space between the numbers, i,e.-~~leading edge position half way between 4 and 5: 4.50. In all configurations the top piece extended ahead of the bottom piece and hence the top piece de- fined the chord when the leading edge was retracted beyond the flush posi- tion. Circular end-plates 15 inches in diameter were secured to the base piece. The outer edges were beveled 15 degrees to decrease the influence of the end plates on the model. ‘The inner faces shead of the base piece were milled to receive the ends of the retractable leading edge. A .125 inch wide slot was milled into each receiver so that two 125 inch locking bolts tapped into each end of the leading edge piece could ride during re-~ traction. The leading edge was fixed into the various positions by tight- ening the locking bolts. The positions were indicated by marking the end plates at approximately every 1/4 inch and using the base end of the lead- ing edge piece as a reference. -5- The model was mounted on a three point support in the wind tunnel. There were two forward struts which connected to trunion pins at the 50 percent chord point on the model outboard of the end plates. The tail support connected to a 1/2 inch diameter spanwise rod joined on each end to 12 inch tail stings mounted on the end plates. The three supports went through the test section floor to connect with the balance system. Figure 4 shows the model and balance system. Figures 5 and 6 show the model in the test section with leading edge retracted to position 4. The balance was modified to support the severe dynamic loads imposed by the model at stall (Ng. 7). A1/2 x 1x 8 inch bar was connected at the tail support pivot to the angle of attack setting moment arm. A similar bar was connected to a gusset on the inner lift frame. The free ends of the bars were connected by a wide-flange sheet metal clamp secured by two bolts and wing nuts. After the angle of attack was set, the clamp was tightened to permit the rigidity of the lift frame to damp the oscil- lation of the relatively weak balance moment arm. Static measurements showed the lift balance to be accurate to + .01 pounds. Accuracy of results is discussed in section IV. Besides force measurements upper surface pressure and tuft surveys were made. The pressure surveys were conducted with a static pressure probe made from .032 inch 0.D. stainless steel tubing. To minimize the ef- fect of orientation and yaw, three evenly spaced No. 80 holes were drilled circumferentially .75 inches from the nose of the probe. Fine piano wire (.012" dia.) was soldered into the front end of the probe and extended for ward in the tunnel through a small grommet attached to an upper hanger made of the same wire and out the bottom on the tunnel just ahead of the test section. The probe was 13 inches long. A rubber hose and enother fine wire were attached to the rear of the probe at an expanding section. The -6- hose and wire passed through a nylon housing which was secured to the aft model support. The nylon housing carried the hose outside the test sec- tion. The pressure was measured on an alcohol micromanometer referenced to stream static pressure. The probe was adjusted by the fore and aft wires. The upper hanger was used to support the fore wire tangent to the leading edge, requiring adjustment with change in angle of attack. The probe positions were identified by 1/2" marks numbered 0 through 20 on the wing starting at the trailing edge with 0. A stripe of bluing behind the static holes on the probe aided their alignment with the marks of the wing. The survey was accomplished nose to tail without interruption, the only de- lay being caused by the time lag in pressure stabilization due to the damp- ing of the small tube, Figures 8 and 9 show the probe and set-up. The tuft surveys were carried out with a probe made from 1/4 inch drill rod, tapered and fitted at one end with .032 inch stainless steel tubing. A single strand of cotton thread 3/8 inches long was inserted and glued into the tube. ITI. EXPERIMENTAL PROCEDURE Much of the testing of the model was carried out jointly with another investigator (Ref. 2). Lift measurements were made for all of the model configurations; drag measurements were made for some configurations. The balance modification precluded any moment measurements. Since this report is concerned with the effect of leading edge retraction on lift only, no further reference will be made to the drag measurements. Lift data was measured directly by the balance system and corrected by an aerodynamic tare taken with the model removed from the endplates and supports. No wind tunnel boundary corrections were made. The tare was run at a dynamic pressure, a, of 20 lbs. /ft.”. All other runs were made at a = 25 lbs. /ft.2. The Reynolds number based on a 10 inch chord was 710,000. The lift measurements were converted to coefficient form and plotted during each run so that questionable points or observed phenomena. could be double-checked. The basic 10 inch chord was assumed for all con- figurations and leading edge positions in converting to lift coefficient so that realistic comparison could be made between the configurations and the unaltered airfoil. A record of the repeatability of the data was kept in order to determine the experimental error. A basic airfoil configuration was run after each modification to the model, On the initial basic configuration test, the leading edge piece junctures with the upper and lower conponents were covered with scotch tape to minimize any effect of some sharp edge irregularities. Repeating the run with the scotch tape removed showed the tape to have no noticeable effect. Scotch tape was also utilized to cover the various spanwise slots. Comparing these runs with the open-slot runs enabled the slot effectiveness -3- to be determined. Runs during which tape was employed are designated by (ts) at the end of the configuration number. | The leading edge was retracted from the basic position step by step to the fully retracted position. At each step force measurements were made, usually over a range of angle of attack from -2 to 14 degrees. Several times the range of angle of attack was limited to the upper por- tion of the lift curve. Pressure and tuft surveys were made over the upper surface of some configurations in an effort to describe variations in lift measurements between leading edge positions. The model was set at the desired angle of attack and a survey accomplished. The angle of attack was then alter ed, and in the case of the pressure survey the probe was re-aligned to travel tangent to the model surface. The run was repeated. The tuft sur- vey was conducted by positioning the tuft with the probe, observing the action, and sketching it. The tunnel speed for the surveys was the same as for the force measurements. IV. RESULTS AND DISCUSSION The lift measurements as determined by repeatability were accurate to + .04 pounds, equivalent in lift coefficient to + .001, up to the stal- ling angle of attack. In the immediate vicinity of the stalling angle the lift measurements were accurate to £ .12 pounds, and at higher angles the accuracy was +t .20 pounds. The most important observation in the investigation was the occurrence of a favorable but critical leading edge position (Figs. 10-13). Note in particular Figure 12. Retracting the leading edge caused a decrease in the maximum lift coefficient based on the original chord up to a certain position defined here as the critical position. At this point, Ch max made an abrupt increase to equal or exceed the basic maximum lift coefficient. Further retraction caused an abrupt decrease in Cp max with the least Ce ax occurring in most cases where the upper edge and leading edge were in the same vertical plane. Retraction within the wing had only a slight effect over the already reduced Co ax’ This behavior suggests three regions for discussion: (1) the positions before critical, (2) the critical position, and (3) the positions after critical where the leading edge is fully within the wing. (A). Before critical. The reduction of lift due to retraction of the leading edge from the basic position to a position approximately one percent of chord from the flush position was linear. Correcting the lift coefficient for reduction in chord produced a constant coefficient for this region showing that the reduction was due entirely to loss in wing area (Figs. 10, 12, and 13). The upper portions of the lift curves for all configurations are shown in Figures 14 through 19. Com ~10- paring the different configurations at positions 5 and 6 showed no significant effect due to configuration. Throughout this region the nose piece was not retracted sufficiently to open the slot. (B). The critical position. As the leading edge was retracted to the vicinity of the upper edge, lift abruptly increased. Retraction in this region over a range of less than one percent chord caused a change from the low, chord re- duced lift coefficient to a maximum lift coefficient, followed by a decrease to a minimum lift coefficient. The stalling angle of attack was 10 degrees for every position except the critical where it was 12 degrees. The slope of the lift curve for the critical position was less than the slope of the lift curve for the basic airfoil. The critical position was not achieved for every configuration because of the extreme sensitivity of the position. The first con- figuration tested was configuration lal (Fig. 14). At position 4 the lift increased substantially and w increased to 12 degrees. The second configuration when tested at position 4 showed only a slight increase in lift (Fig. 15). The third configuration showed no increase. The other tests were similarly inconsistent at position 4. The results were further confused when scotch tape was placed over the slots and the tests were repeated showing in some cases the lift to be higher and in some cases the lift to be lower. When the maximum lift coef- ficients for all configurations at position 4 were plotted, the curves showed a variation of nearly 30 percent of the basic lift coefficient. This inconsistency pointed to the high sensitivity of position. The model was re-assembled in configuration 1d2 and carefully tested in the critical region (Fig. 17). Variations of .005 inch in -ll- retraction produced considerable change in maximum lift coefficient (Fig. 11). This .005 inch variation was much less than the position- ing error and explained the inconsistency first noted. In fact, the model leading edge accuracy when compared to a spanwise straight—edge showed a variation of 4 .020 inches. The critical position was locat- ed by trial and error: and, when maximum lift was obtained, the posi- tion was called 4.00. Configuration 143 was then carefully tested and the critical position achieved (Figs. 12 and 18). Prior to assembling and testing configuration 23, the leading edges were worked over and the straight-edge variation was decreased to + .010 inches. The slot effect in the critical region was negligible. Figure 20 shows the comparison of 143-4 and 1d3-4(ts) where the critical posi- tion was achieved, and Figure 21 shows the comparison of 1lc2=4 and 1c2-4(ts) where the critical position was not quite achieved. The tuft surveys for these configurations showed the flow to be up-stream on the surface of the wing in: the vicinity of the slot (Mg. 22). In one case the tuft was sucked into the slot from the upper surface but the suction was very weak. The tuft could usually be made to lie across the slot or in it at will, pointed in the up-wind direction. The slot and taped slot configurations of 143-4 at the stalling angle of attack were compared by pressure surveys. As is shown in Figure 23 the slot effect was to decrease the absolute magnitude of pressure over the first 15 percent of the upper surface and to slightly increase it over the remainder of the wing. The net effect was a zero change in lift. The small change in pressure distribution would probably have a very slight effect on drag and pitching moment. Fig- ure 24 shows a similar slot effect. The small pressure peaks occur -12- ring 82.5 percent from trailing edge are caused by the influence of the slot on the probe. Flow here was not always parallel to the static orifices causing a dynamic component of pressure to be vicked UD. The mechanism for the achievement of higher lift and later stall is not known. It is suggested, however, that at the critical position a vortex forms between the leading edge and upper edge and permits the flow to attack the upper edge at an angle more favorable for negotiation than if a stagnation point existed. Sketches in Figure 25 illustrate the hypothesis. The supposition of an aiding vortex is advanced by a test on configuration 142-4 where scotch tape was used to cover the three edges to form a triangular shaped nose (Fig. 17). The tape prevented the action of the supposed aiding vortex and the resulting lift was considerably less than the non-taped critical posi- tion lift. The stalling angle of attack of the taped-edge configura- tion was 10 degrees whereas the stalling angle of attack of the non- taped configuration was 12 degrees. The taped-edge configuration lift curve resembled the curves of the configurations on either side of the critical position. An attempt was made to more clearly define the critical position. The angle formed with the chord line by a line connecting the leading edge and the upper edge was measured. Because of the spanwise ir- regularities this angle varied considerably. The angle was determined from the average of several measurements made at different spanwise stations to be 59 degrees. Probably the critical position was not achieved everywhere at the same leading edge setting because of this spanwise irregularity. ~13- An opportunity to evaluate the scale effect on the critical posi- tion occurred between configurations 143 and 23. The increment of Ce ax between the basic position and the critical position when cor- rected for decrease in chord was .077 for configuration 143 and .o44 for configuration 23. The critical angle was approximately the same in each configuration, but the vertical distance between the edges was less for configuration 23 than for configuration 143. The ratio of vertical distances was determined from the data in Table I to be 0.58. The ratio of A Cy ex was 0.57. These ratios indicated a linear increase in Co ax with an increase in scale. A further pursuit of similarity could not be made because the governing parameters were not all known. (C). After critical. Retraction of the leading edge into the wing was to provide an intake for air to re-energize the flow in the bubble and thereby aid flow re-attachment and delay stall. | The method was ineffective as is shown in Mgures 10 and 12. In no configuration of this region did the maximum lift coefficient in- crease above the chord-reduced coefficient. In configuration la2 the fully retracted position produced a slightly higher Cy ax than the flush or intermediate positions (Mig. 10). In configuration 143 the flush position was higher (Fig. 12). The effect of the slot for the fully retracted position is il- lustrated in Figures 20 and 21. For configuration 143-0 the slot caused an earlier stall than for the taped~slot. The taped-slot lift curve slope was not linear, whereas the slope of the slot lift curve was linear. Figure 24 shows the pressure distribution for configura- tions 1c2-0 and 1c2-0(ts). The effect of the slot was opposite to that -14- of position 4 previously described. The slot caused a negative pres- sure peak over the area ahead of it which was about 70 percent higher than the pressure of the taped-slot configuration. If a substantial pressure recovery could be gotten by the intake, this pressure peak could probably be increased sufficiently to cause a significant in- crease in lift. The required pressure recovery was not possible with the configurations tried because the stagnation point was below the lower leading edge at high angles of attack. A comparison of the effects of the various slots and configura-~ tions with the leading edge fully retracted is shown in Figure 26. The poorest configuration was 1b2-0 where the slot gap was the greatest. The best configurations were 1¢2-0, 1c2-0(ms), and 1a2-0. Slots a and c were the same size, but smaller than slot b. The configuration 1c2-(ms) is shown in Mgure 21. It was an intake modification made in an attempt to improve the flow, but very little difference in lift was noted. To summarize, leading edge retraction from the basic airfoil configura- tion to the fully retracted position produced three distinct regions of interest. The first was the region of initial retraction where the lift decreased in proportion to the chord. The second was the region where the leading edge and upper edge interacted. Favorable interaction occurred at the critical vosition, causing a 2 degree increase in the stalling angle of attack and an increase in lift coefficient. The slope of the lift curve for the critical position was less than the slope of the lift curve for the besic configuration. The third region was the range of position where the leading edge was retracted past the flush position. The air intake and slot were ineffective in producing a significant increase in -15- lift because the pressure recovery was very small. The effect of the lower edge was insignificant in the configurations tested. However, in view of the critical position phenomena, there may be a benefit realized by having the lower edge in a forward stagger posi- tion. To investigate this both the lower edge and leading edge must te movable. The possibility of using leading edge retraction as a high lift de- vice is indicated by the results. The lift increase achieved here was not in itself sufficient, but the indication of an increase in lift with an increase in scale leads to the possibility that a substantial lift in- crease may be gained with a very small chord retraction. -16- V. CONCLUSIONS AND RECOMMENDATIONS The following conclusions can be drawn from the investigation on the effect of leading edge retraction on low speed lift of a 6 percent thick cirecular-are airfoils 1. Retraction of the leading edge caused an advantageous flow phenomenon at a critical position which warrants further study. 2. Retraction of the leading edge past the critical position was dis- advantageous. The use of the space as an intake at high angles of attack was not satisfactory. 3. The slot did not significantly improve lift for the fully retracted leading edge. It was ineffective at the critical position. It is recommended that the investigation of the critical position phenomenon be continued. Larger scale tests with accurate positioning devices for the lower and leading edges would be advisable. -17- REFERENCES McCullough, G.B. and Gault, D.#.: "Examples of Three Representative Tynes of Airfoil-Section Stall at Low Speed", NACA TN 2502, 1951. Birdwell, Lt. Carl, USN; "Investigation at Low Speed of a Six-percent Thick Symmetrical Circular-Are Airfoil with Leading Edge Retraction in Combination with Nose Slot", Ae= Thesis, Calif. Institute of Tech., 1955. SIGNIFICANT DIMENSIONS FOR VARIOUS CONFIGURATIONS -18- TABLE I Leading Hdge Configurations Chord Leading Edge Percent Position (defined by (inches) Position Retraction (number) top piece) (inches from (of basic trailing edge) chord) basic la,lb,lc,ld 10.000 10.000 9) 7 la,lb,1le,ld 9.867 9.867 1.33 basic 2 9. 840 9.840 fe) 6 la,lb,le,1ld 9.617 9.617 3.83 2 9.617 9.617 2.25 5 la,lb,le,ld 9.367 9.367 6.33 5 2 9.367 9.367 4,72 flush 2 9.285 9.285 5.65 by 2 9.285 9.100 4 la,1b,le,1ld 9.100 9.100 9.00 flush 1b 9.060 9.060 9.40 flush la,le 9.040 9.040 9.60 flush 1d 9.030 9.030 9.70 3 8.850 2 8.600 1 8.350 0 8.100 PAN pe j {OLD & bite [ME Nh 27) JPAAAL i OL/ a Cue iff ion ia er ; Ud Oyy aa Ay > ~. ee | sc weyag oe EEE ~ | pou W020 | f Oe weIjaq t uae 12 aftpa Biupna 1 7 Lua bs | — a rs Br wo ! a a a eee 7 _ 2 ~~ tees ST soy fe os J a L ee t a“ 7 a tee 7 fo ed o to yay oo : aby buipvay fAaPOjN JO MalA /OLIOIQIAG fOl]AQd 2 blf ~ OF ee 7 ED jos - 0606 MN L926 Ss 7 mG a -21- Fig 3 Slot Configurations 2, > SS = 2 /a i S a —.. | A hae . Fig. 4 Model and Balance System. -23- Fig. 5 End View of Model in Test Section, End Plate Removed. Leading Edge at Position 4, Fig. 6 Three-quarter View of Model in Test Section. Fig. 8 Pressure Survey Probe Set-up. -25- Fig. $. STATIC PRESSURE PROEE AND SET-UP Ssoder — A \ \ = pe | ~ 032” SS tube / | “™ “O06 or OW jmItInG Wire ne. EC. drill ~~ > Section AA Pebe wire hanger C/ a a > | Test section ee 2 pore ~ \ L grommer . i Clamp .. i Lm ££ > — mee “ \ ricdel ; \ . — vylon ‘ ho: using aa | \ = contro! urive . we tube to % © micromanoneter contro! wire CA ice) cy ! = CO epee eee | | G01) pious quested uy abe buspos} 40 4OH!s OW 6 8 2 9 $F + ¢ 2 /} .%:6 G@ 4 9 wet ie npn ee i : i i i ‘ i aie eed i t | ! H ! : i i a in | et ONDIEVENDIANOD Hod LITT WAWIXV HO HOTLOWELEY GOCE ONIGVET £0 LORMAN HHS | i nae end i } : ! | -27= wey eee pe ee gee ee ny ee a cia : ! : i 3 i i i H $ { t wepee enced i ‘ 6iypo3/ 72 5 4/012, 1S 4 3 , 4 on’ ; $ | NOLLO : oi i { ! H i : Pe a ee Pua ed ‘ i i a ma the pee = a am 2hee Peary) pias puede Ur’ 5 ; “wojzsad ~- Soopemecpenss beserpe pene op ue 07 6 @ £ 9 & » © &@ | 6 6 @ 4° 9: & EF 2 Ft noe es ' \ ! : t - : { she whee e ferent od esition. of Jeadlog edge. t Beep t Ss F_LEADING ED YT beshe percent chird. (10 “0. ws 99 FOR CONFIGURATION 23 | i tm ar t an i + { IGURATION Jal)... i ' or -31- shee a 3,50 3 ae "4.26 375 (Stush) bee ere eared mip me ee sengen tone i eee it eet eed oe _ + —_i— 7 - Basle: dy 2 ett badhiig edge estore ; | FOR webeeee * 110 i i | _— 4 wad -32— Se ca epee pif} yy Popgom Ss so oo a : 4 pe ee eee ee fading edge { H ' : i if j- : | CONFIGURATION |Ip2 fio. $ sneer pe tree emcees wet ~4 mg area. ee nip q TET AE oe ey pod ope Tq a | i | @ i ~ a. owed . ——C _ i ~ : soe as aes Sane penne vee eee & g be + poi oe ‘- 4 . i : es oe Ys o> tf i | of 4k i i 7 ~ : ! SS | a Mo — ! ar \ pee 7 : en ee Nae w | : : : i. 5, a a ce oes . Fie - s : i we dee eee pot * - 7 ~ | i S : os : 7 : SRN : : he Oe . : : ae i. heb po ote. i : {- Nw ~ Syne Sw i _ S. — a i fee oa ~ -_ | men : Day aw i NS SY eet aon ee ns noe a re Ooo be hock db ib Ped oe. : : = : . wet do dich bibs deeeeb edict. : H : : i oe ee es Dee - : i ; i H eb ads iL - t + i i : i oes : : i i _ aL. ‘ a | : : Hi ete a he ere ee pee fee 2 : \ i i | pooh eet | 4 ! ne aS Sn a nD SDSS ct eS eee nn Seer vt if moet H > — _ i i ONFICURATION..14 1 ee . t 1 pepe PING : i \ Soe bee NE ERS boo. ors rrr j { fe . : : ‘ : _ | | I : i : jl Si : ; i jens Cone . : ee oe a pot i poe : 7 : i . : i: fen ee bomen wok i es Sens Senne nneT a Pe, a PT NID et a BIE + -4 49D 4 ' : . ' : i Dba bee. ese ne ee fee nen anes eee aeee a FIG, 19.6 5 we. cx... FOR CONF] GURAPION..23 a wos ofeced opto pf Goth . sete 7 bees ae : J eetenavtdie vewteninbe eet des ! k i : -36- genet : . Seppe - : ~ 1x _— = ~ ey . : sesee : _ ; — ~ ~ ead AQ ne ie —. eee an on of ? . . ‘ : i pel : i : ee ee eee eee poy i. a : a - : - - SY mane ogee er. ae Fr i ! i cotersepccres Pes peers et a. 4 i i a Sn See as a4 : i ~ H i + wt 2: ae sere 4 . seen 2 $y QQ tp _ a aka | & Ps ote che peed ne. webs . eran ees BERR bg poe Bo enn : a hk. Longe eke i SL pon meters i i } ! 4 Scape te tne nape te 50 Al..| ” O81 24 . > (78) fat ' i + i Le2=0.AND 1e2—4 s~- woo 1 4 Qo Be _ ley. 8 Bs : a re Ry et? a eee? eee ete Nob BO mE a No dB ; i . ' bees K' . a + Pod 4 * N a a Se re Nie g ; : os Oy - BN ar nee: pone +- ann ta @ . ‘eg Lease ae a} cafes ~ : — DOOR 8 Pn ee ee ee ee b- aaa cain sop FIG. 22 PROBE TUFT SURVEY OF CONFIGURATION 1d3-4 «= TURBULENT FLOW | : | ~— SMOOTH FLOW | | i “CONFUSED FLOW a ae % CHORD .. THE EFFECT OF. ANGLE. OF... ATTACK OW PRESSURE DIs—. 24 wm ee 2 Bo. : ; : Ode beeeee i. i eee 1 sence ne oe ene dee arene FIG. 24....THE. EFFECT OF LEADING EDGE |... RETRACTION ON THE UPPER. SURE. DISTRIBU-. _ TION! OF CONFIGURATION 12. } ___. SURFACE. PRES a res t } AT oF 210° i re ee rt ar a 40 ad, perce + ~ + -Ai..-. . ; i { i i i 3 j on - 4{[- FIG. 25 VORTEX HYPOTHESIS a = 10° LEADING EDGE BEFORE CRITICAL | LOWER EDGE Cann — > Aiding Vortex UPPER EDGE mn LEADING EDGE CRITICAL LOWER EDGE aan ; , AFTER CRITICAL Lower EDGE ezine ; ay an a — “+ ~ iets | sae | sng lags AB. AND CONFIGURATION 2.0... TH POSETION, _...