FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1935
1935 -2- 0785.PDF
NOVEMBER 28, 1935 85 THE AIRCRAFT ENGINEER SUPPLEMENT TO' FLIGHT lube. If, however, it is being slowed while passing over a surface, as soon as the boundary layer exceeds a certain thickness breakaway occurs, with a considerable loss of head. If the breakaway depends on the kinematic viscosity (v) velocity gradient and boundary layer thickness, the non-dimensional expression must bz (8'/f) [dXjds), where a is the boundary layer thickness and s is the distance back from the nose,, and this must not exceed the critical numerical value. It is obvious that the air entry must be kept well forward where 8 is small, and if for some reason the entry must be back, the air speed should be very gradually reduced, or some means taken to get rid of the thick boundary layer. Tests in the 24ft. Wind Tunnel The tunnel was formally opened in April, and apart from preliminary calibration tests it has been working con- tinuously on cooling problems. The first problem was to determine the real magnitude of the engine and cooling drag of existing engines, a point on which there was con- siderable uncertainty, and the second was to see how far these items could be reduced by the application of the principles just considered. Two installations were chosen for our preliminary investigations :—- (1) The Gauntlet, fitted with a radial air-cooled Mercury YI.S. engine having a simple ring cowl, and (2) The Bulldog, fitted with a Rapier air-cooled in-line engine. The Gauntlet is typical of the best modern practice of a fixed cowl unbaffled installation, while the Rapier Bulldog, on the other hand, is a completely baffled installation which lends itself to experiments on controlled cooling flow. In both series of tests, measurements were made to determine ;— (a) The quantity of air passing through the cowl. (b) The cooling drag. This was deduced by difference from measurements of the drag of the aircraft with engine, and the drag of the aircraft with the engine removed and replaced by a well- shaped nose fairing. The drag difference found by these tests was taken to be the cooling drag of the installation. The measurements, of course, could only be made without slipstream. (c) The drag loss inside the cowl. ' ., Pi^Nt of ENTRY prrerrs. — PLANES or prrora AKOUNO CYUNOB? . Fig. 6. Diagram of the installation of the Mercury VI.S. and engine cowl on Gloster Gauntlet. This was obtained from measurements of the total head and velocity of the air leaving the cowl. The method is analogous to the determination of the drag of a wing from the pressure and momentum in the wake. The measure- ments were made with and without slipstream. The difference between (b) and (/:) gives the spoiling; effect on the engine and cowl installation. I shall describe first the Gauntlet tests. The cowl i.s shown in Fig. 6. In these tests the total head and velocity over the inlet and outlet areas of the cowl were deter- mined from readings of combined pitot static tubes. Three such tubes were used to explore the cowl entry and four tubes were fixed in the exit. These tubes were spaced at various angular positions around the cowl and were moved in turn to four positions radially across both the entry and exit of the cowl. The tests were made with and without airscrew running, and in the tests without airscrew the airscrew was removed and a small gear case fairing was fitted. The tests with airscrew were made at a series of speeds and at engine .revolutions corresponding to level flight and climb advances respactively. In addition to the flow measurements the drag of the Gauntlet was measured and the cooling drag was deter- mined by repeating the drag experiments with the engine and cowl replaced by a nose fairing to the fuselage. The main results are given in the table below. Tlic cooling losses are given in li.p. and as percentages ol tlie engine power. Kngine b.h.p. Flow through cowl (lbs./sec.) Flow through cowl (lbs. /sec. /h. p.) H.P. absorbed in cowl deduced from flow measurement Cooling drag from .. balance readings (lbs. at 100 f.p.s.) H.P Ivevel flight at 15,000ft. at 230 IT With airscrew 675 33* 0.049 34 (5%) — — l.p.h. No . airscrew. i_ ' *-. • 38* * • — 19.5 (3%) •• - '' - 1 ' 12.4 54 («%) Climb at at 130 With airscrew. 640 TO.4 ; 0.03a 6-5 (1%) - 13,OOOft. m.p.h. No airscrew. — — — — — — Inspection of the table shows that under level flight conditions :— T. With no airscrew the cooling drag is 8 per cent, of the b.h.p. ; 3 per cent, of the b.h.p. is absorbed in the cowl, so the spoiling drag is about 5 per cent. 2. The airscrew increases the power absorbed in "the cowl from 3 to 5 per cent, of the engine power and seriously reduces the flow. This is due to the breakaway of the flow behind the boss. Under flight conditions the total cooling drag must thus be at least 10 per cent., i.e., 2 per cent, above that deduced from the tests without airscrew. 3. If the engine is adequately cooled on flic <-limb whort the power absorbed is only 1 per cent. 01 th< ii.h.p , th<- flow at top speed must be greatly in excess ot what is required. An excess of 50 p:.*r cent is computed. In any air-cooled engine installation (IK- c-iwl mint provide sufficient cooling air to satisfy the renditions experienced on the climb. These conditions are always more severe than those met with at top and cruising speeds, and consequently the cowl that satisfies the < limb conditions invariably"permits more cooling air <'• pr-.s through the cowl than is required to cool the engine xl< quately at the higher speeds. We have just seen in the case of the Gauntlet cowl, that a mass flow approximately 50 per cent, in excess of the amount required for cooling the engine was passing through the cowl under level flight conditions. T*v -nii;il>i" throttling of the emergent stream from the cowl, 11K nf' 1 •, it should be possible to decrease the power losses iniide the CDWI and also to increase the efficiency of the induction of the flow by a relative increase of the velocity of its wake relative to the aeroplane. To investigate the possible reduction of drag resulting from the application of this principle, some controlled
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
