FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1942
1942 - 0196.PDF
74 FLIGHT JANUARY 22ND, 1942 SUPERCHARGERS entry did not correspond for the impeller and the volute; the influence of this factor being further enhanced by the fact that with varying head the change in volume between the impeller entry and the commencement of the volute varies according to the number of blades. It is therefore difficult to arrange the experiment in a manner to define the separate influence of the number of blades ; however, the tests do show that a smaller number of blades than usual is practicable, and in certain cases more favourable than increasing the number of blades. It remains to underline the importance of the number of blades in connection with the first cost and weight of the blower, and par ticularly the bearing loads produced by centri fugal forces. Obviously, it is cheaper and simpler to machine a smaller number of im peller blades. The question of the most suitable design for Q 2,000 4,000 3,000 Among the important and insufficiently clarified prob lems of impeller design is the reciprocal relationship between blade-loading and intake velocity; in the case of centrifugal pumps this is known to influence the value of the suction head. To determine this relation ship, two sets of experiments were made with the blower shown in Fig, 35; the result will now be examined. In the first series of experi ments the diameter of the original entry impeller was reduced by stages, without any other modi fy* Entry Diameter of original entry impeller •a I the intake to the blower (casing inlet) is par- 55 1,000 ticularly debatable. This apparently sub sidiary component of the blower can have a disproportionate effect on the installation of the power-unit in the airframe. Casing inlets can be: straight axial, single or multiple intake pockets, single or multiple intake manifolds. Figs. 34 and 35 show three curve-charts ex plaining the influence of the above-mentioned inlet types on the characteristics of the blower. In cases A and B, Fig. 34, the curves for similar tip speed differ only slightly, and in case B the efficiency is even 2 per cent, less than in case A. In the straight axial intake of case C, Fig. 35, on the other hand, the curves of equal tip speeds are considerably flatter, i.e., a larger throughput is possible, and the efficiency is greater by some 4 per cent. In all three cases, pressures in advance of the casing inlet were measured at approximately the same distance from the impeller entry along the central line of the flow. The improved result obtained with a straight axial intake is probably due prin cipally to the decreased throttling effect of a straight entry. Furthermore, the flow entering through an intake pocket or ma-nifold is probably not entirely free from twist, and the entering flow is consequently not evenly distributed over the cross-section. The difference is, how ever, not great. More recent experiments indicate that intake pockets, if amply dimen sioned and properly shaped, even if the axial length is restricted, are rela tively little less favourable than a straight radial intake. The correct • mathematical evaluation of such casing inlets is not yet defined ;* rela tively slight modifications in form can have a great effect; particular care should be taken to secure freedom from twist and an evenly distributed flow on the impeller. The absolute aif velocity will probably also exert a certain influence at the impeller entry—earlier tests of intake pockets also showed the efficiency to be some 3-3 per cent, less than for a straight axial intake. Hovvever, for reasons connected with the installation on to the aircraft, straight axial intakes are not usual. In a number of cases the casing inlet has been designed as a normal volute, the entry rotor being omitted and the impeller having straight radial blades. h- 1 turned-down - •d I Q < LU I m 2,000 I.OOO 242 md^ 223m/sT\ 205 rW 186 W %-idr^% \ \J \ 4/5 f\\a Fe=Q4F0 Pi=90° 40 N% 35 ijj, 7? O 0,2 0,4 m?/s 0.6 INDUCTION VOLUME V| (cu.m/sec) O 0.2 in. 3/s Q4 INDUCTION VOLUME ^ (aun/sec) Characteristic curves of the same experimental blower at different entry diameters of the entry impeller and varying blade design. The charac teristic curves of the experimental blower in the original form (entry impeller as shown at a, Fig. 36) had the characteristics of Fig. 35. In Fig- 37 (left) the full lines apply to an entry impeller turned down to the dimensions of case 6, Fig. 36. The dotted lines are for the dimensions of case c of that Figure. In Fig. 38 (right) the curves apply to an entry impeller as c, Fig. 36, but with straight radial instead of curved blades. fication of the blower (Fig. 36). The characteristic curves for 70 per cent, and 40 per cent, of the original intake area of the rotor, corresponding o " b" and " c" of FJg- 36» are shown in Fig. 37. The measurements showed the suction volumes V, for rjmttx to be in the ratio of Fig. 36. Experi mental rotor with different d i a - meters of entry impeller. An experimental French blower. Fig. 39 (left) shows it to be of semi-shrouded type, with subdivided entry impeller. Fig. 40 (right) shows the exit diffuser of the same blower, with subdivided vanes. 1:0.73:0.375, instead of the theoretical ratio 1:0.7:0.4. The difference is inconsiderable in view of the flat gradient of the curves of >; with respect to the suction volume at n = const., particularly since the volute spiral was nor J> modified, which somewhat reduced the value of these ex periments. For an intake area of Fc = o.7 F0, the value
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
