FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1940
1940 - 0597.PDF
FEBRUARY 29, 1940 THE AIRCRAFT ENGINEER SUPPLEMENT TO FLIGHT g German curves give generally ^AD so that mechanical, wheel friction and leakage losses are not included in the values given, which appear therefore very high. It is obvious that the use of the pressure head is independent of the unit system employed, if the same unit of length is used both for measuring H and the pressure ; all the enclosed German test curves can be used also in English units, by converting HAD from metres into feet and V1 from cubic metres into cubic feet, but the pressures calculated by y . HAD = p2 — Pi will be in lb. per sq. ft., not in 1b. per sq. in. (y being in lb. per cubic foot). The pressure head H is pro- portional to the square of the impeller tip speed ; the radial vane impellers used exclusively in air- craft engines give a theoretical H = u^/g, Ma being the tip speed, ofThis value o H can only be reached, even in theory, if the wheel has an infinite number of vanes ; a further big drop is caused by friction and eddy losses, so that the pressure head actually obtained is always much smaller than the above theoretical value. The German textbooks give elaborate formulae which should permit the actual pressure head HAn to be directly cal- culated from the theoretical values H. The theory on which these formulae are based is very involved, however, and practical tests are often in direct contradiction of it. These formulae can, perhaps, give reasonably accurate results for centrifugal pumps and industrial compressors ; for aircraft superchargers they have no practical value. The D.V.L. tests give simply the actual pressure head HAD as function of the theoretical value H by means of a coefficient of quality q^u ', therefore HAD — <?AD • H. Influence of the axial gap, A, between the impeller and the casing for star type impellers (right), and web type impellers (left). Coefficient of Quality A knowledge of the factors affecting the values of 17AD and the shape of the curve giving #AD in function of the volume flow Vj is most important, as it permits to forecast the performance and suitability of a supercharger design. It must be clearly noted, however, that qAo is not an efficiency, and is not affected by the same factors and in the same way as the efficiency. The expression " hydraulic efficiency " used in some textbooks for ^AD is therefore most misleading. A supercharger can very well have a low q\ D and a high efficiency 17AD or -q ; this will mean that the impeller can only impart a moderate quantity of energy to the air, but the energy is imparted without ex- cessive losses, and therefore without overheating the air. The term " quality factor " or " quality coefficient " express perfectly the meaning of <7AD (Giitezahl in German). After this foreword, long but necessary in order to interpret correctly the German test results, these can now be examined. Important tests were directed to find the most suitable type of impeller. Fig. 4 shows the different types of impeller tested at the D.V.L. ; in the open star type (Rateau) the effective part is reduced to radial spokes ; the web type, in which the radial vanes, often twisted at the inlet, are connected by a disc-shaped web on the side opposite to the air inlet ; this is the type generally used ; the closed box type developed by Junkers and still used by that firm, and the completely closed radial type developed by the D.V.L. The star type impeller, simple, suitable lor very high speeds, is very inefficient; it has the advantage of giving no axial thrust. The web type wheel is much more efficient, but gives a definite amount of axial thrust. The webs are often partly cut away between the spokes, as shown in Fig. 4 b, but D.V.L. tests have shown this to be useless or worse. Both star and web type are very sensitive to the axial gap between the rotating impeller vanes and the fixed casing side (the gap is marked a in Fig. 5). The curves on the right of Fig. 5 were taken on a star type impeller for two different gaps a : 0.25 mm. (o.oiin.) and 0.8 mm. (0.03m.). It will be seen that the small gap improves the efficiency in the region of low volume flow, but the curve is much steeper than with the larger gap, which gives much better results for larger flows. The maximum efficiency is low, and falls rapidly to very low values indeed. The curves on the left of Fig. 5 refer to a web type im- peller, with gaps of 0.9 mm. (o.o36in.), 1.5 mm. (o.o6in.)and 2.6 mm. (o.iin.). Here again, small gaps give better results, but only in a restricted field of low volume flow, while the larger gaps give much flatter curves. On the left of the line MN small gaps are better, on the right large ones. The efficiency is good, though not outstanding, and the type of blower tested corresponds closely to those to be found on the majority of engines now being used. The importance of this curve of Fig. 5 (left), cannot therefore be overestimated. It shows a way of varying and adapting the performance of existing superchargers simply by altering the axial gap. For instance, for very high altitude work, or when a two- speed gear is used, large gaps will be better, owing to the wide variation in the inlet volume ; while a smallish high • • WEB TYPE IMPELLERm.a-^O-Smm. 9 WEB TYPE IMPELLERH7. a~\mm. Comparative test of closed impeller and of web type impeller with different axial gaps.
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
