FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1941
1941 - 0626.PDF
MARCH 13TH, 1941- AIR-COOLED v. LIQUID-COOLED AIRCRAFT' Their Drag, Weight, Cooling and Fuel Consumption : Three Groups 0/ Aircraft Designed and Compared : Air-cooled Found Superior in All Three By JOHN G. LEE (United Aircraft Corporation) THERE are many rea-sons for choosing anair-cooled or a liquid- cooled power plant. They may be technical, economic, or practical. The present paper is confined to the technical reasons. The presentation of a technical paper on this subject is made difficult at this time because of the restrictions which military necessity imposes upon the publication of new data. Those engineers who are engaged in military work and who have access to confidential material will agree, it is believed, that such material does not controvert the conclusions presented here. Aerodynamics The basic drag of a streamline object is first considered. In Fig. 1 the drag coefficient is shown plotted against fine- ness ratio for a variety of both new and old streamline shapes, ranging in form from nacelles to airship models. This curve serves as a reminder that, practically speaking, drag is not caused by frontal area alone, but is largely caused by skin friction. In a perfect fluid, a streamline object has no drag at all. Turning to Fig. 2, the skin friction drags of the streamline forms shown in Fig. 1 have been plotted. These skin-friction drags are presented as coefficients based on the frontal area of the streamline form, and arc thus directly comparable with the total drag coefficients shown in Fig. 1. It is clear that when we increase the diameter of a fuselage to take a larger engine the increased drag is due, not to the increase of frontal area, but to the increase of surface area, other things being equal. The data shown -in Figs. 1 and 2 have been used to determine fuselage drags in subsequent calculations— increased, of course, for the effects of roughness, scoops and other improprieties. It is interesting to pursue the subject of frontal area and fineness ratio a little further. From Figs. 1 or 2 it can be demonstrated that the total drag of a fuselage 4.5ft. in diameter and 27ft. long (Fineness Ratio 6) is actually less than that of a fuselage 4.0ft. in diameter and 32ft. long (Fineness Ratio 8), even though the latter has 20 per (dit. less frontal area. The short, fat fuselage is not as bad as one would like to think. Of course, if the length of the fuselage or nacelle is * Paper presented before the Airplane Design Section of the Institute of theAeronautical Sciences, New York, January 31, 1041. THE author of this article is assistant director of research of the United Aircraft Corporation, which includes the Pratt and Whitney concern. He may thus be accused of being pre- judiced in favour of the air-cooled engine. Nevertheless, the arguments put forward do not appear unfair, and the conclu- sions are distinctly interesting, to say the least. In England, too, research on cooling and cowling of air-cooled engines has not been standing still. zo O ,14u n O ,-.12 $* gj.,0 c< ir Z 5 n >HAC1 - Tft 2ti •JOUHNAC. I/JI-MIT. flE^T 3«J,44* • * COWR CTCD TO HI i i IS already determined by other factors, increasing its dia- meter does increase its drag. This may be seen for two practical cases in Fig. 3. Here the drag, expressed as the familiar "i" factor, is plotted against cross-sec- tional area for a nacelle 15ft. long and a fuselage 30ft. long. The two ver- tical lines represent two extremes of area ; that at the left is the minimum fuselage cross-section which will house a "jockey" pilot (ellipse 30m. x 48m.), and that at the right is the size which will accommodate a 2,000 h.p. P. &W. 2,800 air-cooled pursuit engine (circle 55m. diam.). The cross-section required to accommodate a practical liquid-cooled engine, in spite of our best efforts, lies much nearer the air-cooled engine than the "jockey" pilot. Even if we imagine a liquid-cooled engine so compact as to fit, with all its accessories, inside the cross-section required by a "jockey " pilot, the drag of the bare fuselage housing such an engine would still be at least 70 per cent, of the drag required for the present P. & W. 2,800 air-cooled engine. Engine Nacelles In the case of the drag of an engine nacelle the situation is complicated by the presence of the wing behind it. In Fig. 4 some unpublished nacelle tests are presented which were run in the 7|ft. tunnel at M.I.T. on a wing having a 36m. chord and 6oin. span. The tests were confined to low-lift conditions, hence the low aspect ratio is not impor- tant. In Fig. 4 the minimum drag and minimum drag co- efficient of each nacelle is tabulated, together with a sketch of the nacelle. Unfortunately, the drags of the extension shaft models may be somewhat in error, since they represent the small difference between the measure- ments of two relatively large quantities. The drags of the other models should be reasonably accurate. In Fig. 5 the minimum drag coefficients of the nacelles in Fig. 4 have been plotted against the ratio of nacelle diameter to wing thickness. A reasonable order prevails. The smaller the nacelle relative to the wing, the lower the drag co- efficient. In subsequent calculations the more conserva- tive values given by the dotted line have been used. A word about the "submerged" engines may be in order at this point. During the past few years wings have been getting thinner for planes of a given gross weight, 234 5 6 7 0 FINENESS JATIO ;o 11 J 13 .10 |.o. * 07 2 06 *! .04 JS -03 O 02 ) 1 I \ \ \ • V. FINE rnow TO CSS * FI6 1 rAL D r 1 ATlO 1 • RAG corr-. * I • s FINENESS RATIO Fig. 1. Variation of minimum drag coefficient with fuselagefineness ratio. Fig. 2. Relationship between skin friction drag and total drag.
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
