1、RM L57BO7 RESEARCH MEMORANDUM- A TRANSONIC INVESTIGATION OF THE MASS-FLOW AND PRESSURE RECOVERY CHARACTERISTICS OF SEVERAL TYPES OF AUXILIAR Y AIR INLETS By John S. Dennard Langley Aeronautical Laboratory Langley Field, Va. -.- L=- -l . ; 8 Provided by IHSNot for ResaleNo reproduction or networking
2、permitted without license from IHS-,-,-.B NACA RM L5iBOT T . NATIONAL ADVISORY COMMITTER,FQKARRONAUTICS A TRANSONIC lwvEsTIGA!IION OF THE MASS-FLOW AND PRESSURE REZOVEEIY CHARACTERISTICS OF SEKERAL TYPES OFAUXILIARYAIR INLETS ByJohnS.Dennard SUMMARY Several basic inlet shapes have been tested throug
3、h a Mach number range from 0.33 to 1.3 to determine their mass-flow and pressure recov- ery characteristics when operated as auxiliary inlets. Results indicate that, for flush inlets, the inclination of the inlet axis is the major geometric parameter influencing recovery and the smaller angles with
4、respect to the surface offer decidedly superior performance. For simple flush openings, a circular or low width-depth ratio recw opening offers slight advantages over a higher width-depth ratio rectangular opening. Boundary-layer fences for low width-depth ratio inlets are advantageous in increasing
5、 the maximum mss-flow rate and pressure recov- ery at high mass-flow rates. For low mass-flow rates, the NfUCA submerged inlet with diverging wall ramp offers improved -pressure recoveries. A circular scoop inlet with its inner wall tangent to the surface surpassed all flush inlets tested in regard
6、to both total-pressure recovery and the yaw angle far several of the inlets was varied from Oo to 600. Rectangular inlets with various inclination angles, width-depth ratios, and ramp approaches were also tested. The Mach number range was from 0.55 to 1.3 and mass- flow rates were varied from 0 to c
7、hoke. Inlet mass-flow ratios are pre- sented as a function of the required pressure drop and of the total- pressure recovery. a D H L 2 M n P 4 9 u U X Y 6 SYMBOLS i , velocity of sound, ft/sec hydraulic diameter of inlet, in. total pressure, lb/sq ft length of inlet duct (constant area), in. ratio
8、of mass flow through inlet to mass flow in free-stream tube of area equivalent to cross-sectional area of inlet Mach number, U/a boundary-layer profile exponent, 0 Yk g= 6 static pressure, lb/sq ft static pressure differential, p, - pi dynamic pressure, $oU2, lb/sq ft local velocity-in boundary laye
9、r, ft/sec free-stream velocity, ft/sec distance measured downstream from inlet lip, in. distance measured from surface to point of local velocity u in boundary layer, in. boundary-layer thictiess measured to point where u/u = 1.0, in. . . 3 . . + . Provided by IHSNot for ResaleNo reproduction or net
10、working permitted without license from IHS-,-,-NACA RM L57BO7 . . 8 IncUnation angle of inlet axis, deg J# yaw angle of inlet axis, deg q rwe angle, deg Subscripts: 1 inlet conditions measured at downstream end of constsnt-area throat W free-stream conditions measured outside of boundary layer APPAR
11、ATUSANDHETHODS The general arrangement of the test setup is shown in the Une drawing of figure 1. Air is supplied through the entrance bell at the Apparatus left at a maximum pressure of 2 atmospheres absolute from a centrifugal blower without benefit of aftercooling or drying. Air from the entrance
12、 bell flows through the slotted test section which consists of parallel walls 17 inches long and k$ inches high by a valve in the auxiliary pump line determines the rate at which air is removed and thus controls the speed for M, 1. The Mach number distributions for this test section are discussed in
13、 reference 1. The inlet models were mounted on the solid tunnel wall opposite the slotted wall and centered in one face of a cylindrical settling chamber which has a diameter of $- inches, 4 anda length of g$ inches and is located 6 inches downstream of the beginning of the slots. A screen spanned t
14、his chamber at a station 3 inches above the inlet. Downstream of the screen, the air flowed through a calibrated venturi. A valve in the exhaust line controlled the mass flow withdrawn from the inlet and the system was aspirated by an auxiliary vacuum pump. The free-stream total pressure and tempera
15、ture were measured in the upstream duct and the static pressure was measured in the settling chsm- , .- ber. Other test-section pressure instrumentation included static-pressure orifices in the venturi and a calibrated static-pressure orifice in the plenum. Provided by IHSNot for ResaleNo reproducti
16、on or networking permitted without license from IHS-,-,-4 NACA RM L57Bo7 A sketch showing typical examples of the models tested and the range of variables .investigated is shown as figure 2. The inlets tested had circular and rectangular shapes. The circular inlets included inlets with ducts of vary
17、ing length-diameter ratio (L/D), thin-plate inlets, circular flush inclined inlets, and a scoop inlet. The rectangular inlets were flush and had vying inclination angles and width-depth ratios (where depth is measured from the inletlip perpendicular to the duct axis). Four of the rectangular inlets
18、were instrumented with static-pressure tubes along the center line of the ramp and through the length of the inlet duct. Figure 3 is a photograph showing several of the inclined recte inlets, some with modifications to the ramp, and the circu- lar scoop. The pressure at the downstream end ofthe inle
19、t constant-area duct was assumed to be equal to the settling-chamber static pressure. This assumption, which has long been recognized as valid for subsonic flows and which is the basis for most nozzle flow and orifice flow measure- ments, was verified during the test program by a compsrison between
20、the downstream throat static pressure (where available) and the settling chamber static pressure for unchoked inlet operation. Tllfs pressure would correspond to the static pressure at the entrance to a diffuser if such were used. The total pressure Hi is calculated from the measured mass-flow rate
21、and the static pressure with the assumption of uniform velocity distribution at the exit of the constant-area duct. The test- section Mach number is determined by means of the- calibrated plenum static-pressure orifice and the upstream total pressure. Pressures were read to a reading accuracy of 1mi
22、lUmeter of mercury and this accuracy gives errors in the computed data as follows: Quantity 22 . . . . . . . . . . . % Hw-p . , . . . . . . . Klo y. 5 . . . . . . . . . . . M 2 * . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MaximumError . . . . . . . . . . . 0.002 . .
23、. . . . . . . . . 0.002 . . . . . . . . . . . 0.006 . . . . . . . . . . . 0.000 . . . . . . . . . . . 0.004 . . . . . . . . . . . l 0 .oog Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-5 Tests and Methods When the tests were conducted, the inlet ai
24、r flow was varied from zero to its maximum value for each of several constant values of Mach number ranging from 0.55 to 1.3. The settling-chamber static pressure, free-stream total pressure and total temperature, plenum-chamber static pressure, and mass-flow data were recorded simultaneously. Ramp
25、static pressures were measured on models where pressure-orifice instrumentation was available. The Reynolds number of these data range from 3.45 x lo6 per foot at a Mach number of 0.55 to 7.0 x 106 per foot at a Mach number of 1.3. RESULTS AND DISCUSSION Auxiliary inlets are usually required to oper
26、ate in a surface on which the boundary-layer thickness is comparable to the dimensions of the inlet. For these tests the boundary-layer velocity profile and thickness varied with Mach number as shown in figure 4. The profiles indicate a turbulent boundary layer at sll Mach numbers. The exponent n of
27、ten used to describe the shape of the boundary-layer increases from 5.5 to 8 as the Mach number varies from 0.55 to 1.3. The boundary-layer thickness, defined as the point where ub = 1.0, varies from 0.15 inch to 0.115 inch; this variation is linear from M = 0.7 to M = 1.3. Inlet Performance The inl
28、et performance data in this paper are presented in three forms: the inlet pressure differential these inlets would be expected to have a relatively high total-pressure ratio. Met static-pressure differential.- In the second presentation (fig. 6), the inlet static-pressure differential p, - pi or 4 i
29、s expressed in terms of b and is plotted as a function of the mass-flow ratio. These curves are directly comparable with the inlet chsracter- istics in reference 2 and to the perforated-plate characteristics of reference 3. These results are directly applicable to selection of porosity requirements
30、for perforated-tunnel calculations when a given Mach number distribution has been specified. For sny given flush inlet configuration, the inlet static-pressure differential at zero m the curves at various Mach numbers are separated further at low mass-flow ratios and cross over near The separation n
31、oted here is due to the com- pressibility factor F,. This factor is equal to w and, for this scoop inlet, which is essentially a totsl-pressure tube when operated at a mass-flow ratio of zero, the inlet static-pressure differential is equivalent to -F,. As the mass-flow ratio increases, this correla
32、tion becomes less accurate; nevertheless, the general position of the curves relative to one another persists until the inlet mass-flow ratio approaches its choke.value, Inlet total-pressure ratio.- The third presentation (fig. 7) which will form the basis for the bulk of the discussion to follow is
33、 the ratio of the inlet total pressure to the free-stream total pressure. The curves l Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NACA RM L57BO7 7 . presented here have been faired to match those of figure 5 which, in turn, were faired as famili
34、es by cross plotting. Restriction of test points to those for which the indicated Mach number at the downstream end of the duct was less than or equal to 1.0, that is, unchoked inlet flow, has resulted in some curves being presented for which there is only one test point. The total-pressure ratio al
35、ways decreases with increasing Mach number but the variation of Hi/H, with rni/mC is dependent upon the inlet configuration and orientation with respect to the stream. The incllned inlets exhibit a rising total-pressure ratio with increasing mass-flow ratio up to a point nesr choke; thereafter, as m
36、i/mO increases, the pressure recovery falls. This increasing recovery is probably due to the ingestion of air from further out in the stream where the total pressure is higher than in the boundary layer. As the inlet approaches a choked condition, local accelerations and fric- tion become increasing
37、ly important and the total-pressure ratio decreases. The flush inlets which have their axes more nearly perpendicular to the surface require greater turning angles and, therefore, incur higher losses; for these inlets the pressure recovery decreases throughout the flow range at all except the lowest
38、 speeds. Intermediate configcrations may be found wherein the total pressure remains essentially constant throughout the unchoked mass-flow range for a given Mach number. The curves of figure 7 have been cross plotted at constant mass-flow ratios in figure 8 which shows the variation of total-pressu
39、re ratio with Mach nuLmber for several geometric variables. The symbols do not repre- sent actual test-data points but were obtained from cross fairing. Solid symbols indicate the Kzh number and total-pressure ratio corresponding to choking at the particular mass-flow ratios involved. Absence of a s
40、olid symbol indicates that choking did not occur in the Mach number range of these tests for that psrticular mass-flow ratio. Lines representing the ratio of static pressure to total pressure in the free stream are drawn on all figures. These lines represent the pressure which would be measured by a
41、 wall static-pressure orifice and are observed to be in fair agreement with the measured pressures at s = Mach nuniber range. mg 0 throughout most of the Flush Circular Inlets Effect of L/D.- The pressure recovery of a series of circulsr inlets followed by a straightduct set perpendicular to the sur
42、face is presented as a function of stream Mach number in figure 8(a). These inlets had a diameter of 0.375 inch which is 2.88 times the boundary-layer thicbess at a Mach number of 1.0. At all mass-flow ratios for which data are pre- sented, Hi/ thus, the total-pressure ratio and the maximum mass-flo
43、w rate decrease. All curves for these flush circular inlets show a rapidly decreasing total pressure at mass-flow ratios near choke. This condi- tion is probably the result of the increased friction losses as the inlet velocity increases, the increased momentum losses as air flows over-the sharp lip
44、s as was reported in reference 5, and.nonuniformity of the velocity distribution at-the discharge of the inlets. Effect of yaw angle.- The flush circular inlet with 8 = 15 and L/D = 5 was tested at yaw angles of 30 and 60 in addition to the unyawed position. These results are presented in figure 8(c
45、). There is a striking simils.rity between these curves and those for the inlets of figure 8(b) where the axis became nearly normal to the surface. As the yaw angle goes to 30 and to 60, and, at J, = 60, the total-press-ye ratios fall rapidly approachclosely to the free-stream static-pressure curve.
46、 It is noted that this inlet at $ = 60 offers a very wide Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-JUCA RM L57BO7 9 opening to the approaching boundary layer; thus, the inlet takes in a proportionately larger quantity of low-energy boundary-la
47、yer air. In addition, there is the simple effect of increased turning required because of yaw. These results suggest that for highest total-pressure recovery the yaw sngle should be maintained as low as possible. This condition is increasingly important as the mass-flow ratio or Mach num- ber is inc
48、reased. Effect of size of circular thin-plate inlets.- Three circular thin- plate inlets, t = l/16 inch, having diameters of 0.25, 0.375, and 0.50 inch were tested to determine the effect of the ratio of inlet diameter to boundary-layer thickness. (See fig. 8(d).) The variation of boundary-layer thi
49、ckness with Mach nuuiber prodded an overall range of D/6 from 1.67 to 4.35. The total-pressure recovery for the zero mass-flow (vent) condition agrees well with the stream static-pressure curve up to M = 1.0. Above M = 1.0, however, expansions into the opening together with a shock against the downstream face of the inlet cause a slight pressure rise. At the two higher mass-flow ratios for which data are
