C.P. No. 892

LIBRARI AL AIRCRAFT ESTABLlSHMRUl

BEDFORD.

MINISTRY OF AVIATION

AERONAUTICAL RESEARCH COUNCIL

CURRENT PAPERS

0

An Experimental Investigation of the Influence of

Base Bleed on the Base Drag of

Various Propelling Nozzle Configurations

BY

1. B. Roberts and G. J. Golesworthy

(wifh an Appendix by W. G. E. Lewis and M. V. Herbert)

LONDON HER MAJESTY’S STATIONERY OFFICE

1966

Price 12s 6d net

u.D.c. No. 621-225.1 I 533.6Y1.18.OY3 : 533.6.013.12

c.P. N0.892~

Pcbrunry, 1964

An experrmental lnvestrgatron of the rnfluence of base bleed on the base drag of various

propelling nozzle conflguratlons

- by -

J. B. Roberts and G. T. Golcsworthy

Various propelling nozzle configurations with all-Internal expan-

sion wers tested m external flow over the range of Mach number 0.7 to 2.2,

~.n order to determine the effect of bass bleed (i.e. the InJectron of low

energy secondary air) on base pressure. An ‘overall efflclency’ 1s

defined, which enables the effectiveness of base bleed, as a means of

reducrng the base drag9 to be assessed. The results lndrcate that with

supersonic external flow base bleed generally tends to ra‘a1se the level of

base pressure, but no Improvement m the overall effrclency IS obtalned.

At subsonic external speeds, the secondary air has a negligrble effect on

base pressure.

__ E . . . , / *m . . / * . . . -_ I * . . --*_l_ A - - - .ce--.. I * -1_1

>?

Replaces N.G.T.E. Report Ro. R.259 - A.R.c.~~ 228

-2-

1.0 Introduction

2.0 Doscrlption of test equpmcnt

2.1 The external flow rrg 2.2 Test models

3.0 Instnxnentatlon and au supplies

4.0 Node1 pcrformsnco

::: Base prossurc characterlstlcs Overall efflclencles

5.0 External boundary layer measurements

6.0 Conclusions

Acknowledgements

References

Detachable Abstract Cards

NO. -

I

NO. -

I

II

TABLZ

Tltlo

Nozzle operating condltlons

Xotatlon

APPi3NDICES

Tltlc

The overall thrust efflcuzncy of il propelllng nozzle vlth base bleed (by 8. G. E, Lowls and M. V. Herbert)

6

a

a 13

14

15

15

16

18

20

-3-

ILLUSTPATIOWS

Fig. No. Title

Model propelling nozzls test rrg

Sting carrier section in tranoonio line

Sting carrier section in supersonic line

General construction of test nozzle

Geometry of short and long coplanar models

Geometry of a&ended shroud and boat-tail wodcls

10

11

12-13

Q-15

16

17-18

19-20

21

22-23

24

Geometry of base blockage models

Locat?on of pitot rakes for boundary layer survey

A comparison of base pressure readings for short nozzle va;h coplanar cnt

A comparison of base pressure readings for long nozzle with coplanar exit

Dase flow patterns for a nozzle with a coplanar exit, operating in external flow

Performance of short nozzle with coplanar exit

Performance of long nozzle with coplanar exit

Base flow patterns for a nozzle with an extended shroud, operating in external flow

Performance of short nozzle with extended shroud

Construction of the isobaric jet boundary by axially-symmetric chnractoristrcs (with superimposed outor shroud looatron)

Performance of short nozzle with extended shroud

Performance of short nozzle with boat- talled exterior

Performanoo of short nozzle with base blockage

-4-

ILLUSTRATIONS

FlR. PO. Trtle

25-25 Comparison of results fmm short nozzles

27-28 Comparx3on of overall eff1clencles

29-30 External boundary layer velocity proflle

31 Eatlmatcd varzatlon of 6': with id,

32 Estimated vnrlatlon of 8 vuth Md,

-5- *

1.0 Introduction

For the engine installation of proposed supersonic transport air craft, the exit area of an internal expansion, convergent-divergent pro- pelling nozzle, such as would be appropriate to the exhaust pressure ratio at 0ru1*e, often proves to be considerably less than the cross-sectional drea of the engine nacelle. A relatively large base area is thus created, and the problem arises of minimising the dray associated with it. Proposed methods of reducing this drag are summarised as follows?

(1) Base drag may be completely eliminated by extending the nozzle until its exit area coincides with the nacelle cross-sectional area. The nozzle flow will then be overexpanded at cruise, and the internal pc3rformance will suffer. At off-design conditions the penalty may bo large.

(ii) The nacelle may be boat-tailed sufficiently to reduce the base area to zero, whilst retaining the correct nozzle exit area. Base drag is then replaced to some extent by boat-tail drag of the nacelle.

(iii) Base drag may bo reduced by allowing a small quantity of low energy, secondary air to escape into the base region (this technique is known as base bleed or base ventilation). The exit area of the nozzle appropriate to cruise conditions may then be retained, or the nozzle may be shortened. It has been suggested that base bleed is more effective with such a shortened nozzle at cruise, and some improvement in internal nozzle performance at off-design conditions could be anticipated. Various combinations of base bleed and boat-tailing are also possible.

With the intention of helping to decide which of these systems offers the best solution, and more especially to investigate the effectiveness or otherwise of base bleed as a technique, a series of convergent-divergent model nozzles has boon tested on cold dry air. These models lhere axisymmetric throughout, for convenience of manufacture and testing, although it should be remembered that for proposed supersonic transport aircraft the nacelle cross-section may be square or rectangular.

2.0 Description of test cquipmont

2.1 The external flow riK

A description of the external flow rig used for these tests will be found in Reference 1. The rig consisted of two alternative working sections, illustrated in Figure 1. Each line comprised a nozzle, viewing section and pressure recovery diffuser, the supply and exhaust arrange- ments being common to both lines. External flow from Uach number 0.7 to 1.5 could bo provided by the upper or ‘transonic’ line, whilst the lower ‘supersonic’ line oporated from Each number 1.3 to 2.4. Test models were carried on a long parallel, hollow sting, which passod through the throat of the test line nozzles, and could be interchanged between them.

Figures 2 and 3 show the arrangement of the sting carrier section, which fitted immediately ahead of either external flow nozzle. This consisted of a round duct with a streamlined bullet carried on its centre- line by a singlo hollolv arm. From the bullet, the sting, with the model on its downstream end, was supported so that the outlet plane of the model was located in the window of the viewing section. Air, controllad

-6-

externally, was delivered to the model through the tube. The lnstrumentatlon and control lines from the inner tubo and the outer cover of the sting.

Inner, load-carrying the model passed botweon

The transonic line was equipped with a slotted nozzle of circular cross-section, Il.3 inches in diameter (see Figure 2), tho outlet Slach numbor of which could be varied simply by adJustmg the applied pressure. A two-dimensional, flexible wall nozzle vas used for the supersonic llno, the nozzle outlet being 12 In. x 12 III. (see Figure 3).

2.2 Test models

All the various builds of model tested were founded on a common basic unit. A typical nozzle configuration built from this unit is shown m Flgurc 4.

The secondary or bleed flow was obtained by tapping off some of the primary nozzle flow, through vents upstream of the throat, and the amount tappod off was controlled by a remotely actuated ring valve, located in the annular secondary passageway between the innor and outer surfaces of the model. Secondary mass flow was estimated by passing the air through a constricted section downstream of tho ring valve, equipped with pitot tubes and static tappings, as shown in Figure 4.

Parts additional to those in Figure 4 enabled various builds of model to be constructed. These builds formed modifications of two ‘basic’ dosigns, which are illustrated in Figure 5. Both have parallol outer shrouds, orlth coplanar exists, and aroa ratios of 2.050 and 2.789 (the corresponding design pressure ratios arc 11.1 and 18.8 rospectivcly).

Two modifications appear in Figure 6. Figure 6a shows a model incorporating some dcgrco of cxtornal boat-tailing (1'30'). In othor respects, this build is identical to the ‘basic’ short nozzle, as in Figure 53. Figure 6b shows the construction of a non-coplanar (or extended shroud) nozzle. This was formed by fitting the outer shroud of the ‘basic’ long nozzle (seo Figure 5b) to the short primary nozzle of Figure 5s.

For some tests on the short nozzle, the base aperture was partly blocked. In one such arrangement, the bleed flou, w&s discharged adJaCent to the primary nozzle exit (see Figure Ta), whilst in another it passed on oithor side of the blocking piece (see Figure 7b).

3.0 Instrumentation and air suppiles

The models were fitted with the following prossurc instrumentation (the circled numbers in Figures 4 to 7 refer to the location of those tappings, and will be quoted where possible):-

(a) Two pltot and two static tubas situated in the primary nozzle inlet, upstream of the bleed flow take-off points (Numbers 1, 2, 3 and 4 in Figure 4).

(b) Two pitot tubes situated in the primary flow between the bleed take-off points and tho nozzle throat (Numbers 5 and 6 in Figure 4).

(cl

(d)

(e)

(f)

(d

(h)

(1)

(2)

(k)

(4)

(ml

-7-

Three pitot tubes and three wall static tappings in the bleed airflow measuring section, downstream of the control ring valve (Numbers 7, 8, 9, 10, 11 and 12, respectively in FWP~ 4).

A tapping In the primary nozzle base thickness (Number 15 in Figure 5a).

A tapping outside the primary nozzle at its outlet, indi- cating the static pressure in the base region (Number 16 m Figure 58).

Two tappings inside the primary nozzle at its outlet, indicating the static pressure in the primary flow (Xumbers 13 and 14 in Figure 5a).

Either one or two tappinge in the outer shroud base thickness (Number 17 in Figure 53 and Numbers 17 and 18 in figure 5b).

Either one or two tappings inside the shroud at its end, lndicatlng the static pressure in this region (Number 19 in Figure 5a and Numbers 19 and 20 m Figure 5b).

A tapping outside the shroud at its end (Humber 21 in ’ Figure 5a).

Five tappings along the outside of the boat-tailed shroud (Numbers 24 to 28 in Figure 6a).

In some builds, a reversed pi-tot tube in the base annulus (Number 22 in Figure 5a).

In some builds, a pitot tube placed in the base flow passageway (Number 37 in Figure 5a).

In the models with partial haze blockage, of the types shown in Figure 7, two sets of three base tappings each inserted in the blocking piece (Numbers 29 to 34 m Figure 7a). Also, two taspings located in the constricted region where the bleed flow escaped (llumbers 35 and 36 m Figure 7b).

Secondary mass flow was estimated from the readings of the single pitot and static tappings located in each of three of the six measuring section ports (Figure 4). The averages of these readings were treated as mean total and static pressures actmg over the constricted passage area, mass flow then being obtained according to one-drmensional isentropic relations. It is therefore likely that the values of secondary flow given here are over-estimated.

The height of the free-stream boundary layer at the nozzle exit plane was measured by attaching three pitot rakes (spaced circumferen- tially at 120°) to the end of the model. Their location is illustrated in Figure 8.

-8-

The temperature of the air supply to both model and test line nozzles was maintained within the range 25 to 35’C at all times. Air dry- ness was measured by an R.A.E. -Bedford pattern frost-point hygrometer, and held at better than -20°C frost-point throughout. An supply pressure *as at a level of 5 atmospheres and throttled independently as required for the modal and external flow.

4.0 Model performance

A summary of the operating conditions of the various model builds is given in Table I.

4.1 Base nrossure characteristics

The aim of all these tests was to examine, for each operating condition, the variation of base pressure with bleed mass flow. To obtain the results in a non-dimensional form, the conventional procedure is to

plot p against p (see Appendix I for notatron). In these tests u is m QS identical to the mass flow ratio, - 9

QP since the total temperature of the

bleed air is equal to that of the primary flow.

Some caution IS required when interpreting base pressure measure- merits. As mentioned in So&ion 3.0, a variety of pressure tappings was provided in the base region of these models. In many tests, particularly those on the short nozzle versions, these toppings indicated uniformity of base pressure for low quantities of bleed flow, the agreement steadily deteriorating as the bleed flow incrcssed. A typical set of base pressure readings is presented in Figure p for the short nozzle with no boat- tailing and a coplanar exit, and Figure 10 gives a similar comparison for ths long nozzle with coplanar exit. In the latter cast, the spread at high blood flow rates is much worse.

For tho purposes of comparing model performance one with another, it is necessary to assume a single bosc prassuro characteristic for each build. The base pressure results which follow have accordingly been obtained by averaging the available readings at each bleed flow value, despite the large differences occurring in some instances.

Before considering the expcrimontnl results in dotail, some des- cription will be given of the variations in flow pattern which are encountered.

The flow in the base region of a propelling nozzle with a coplanar exit is shown diagrammatically in Figure 11, for supersonic external flow

and various operating pressure ratios. In Frgurc 11 a we assume g < 1 p

so that the external stream expands around tho lip of the outer shyoud (at A). Also, in this example, the flom within the primary nozzle is under-expanded, and so continues to expand at the nozzle exit (at B) rn a similar manner to the cxtornal stream. Clearly, the ex t e rnal and internal boundary layers (of thicknessi; 6~ and 6I reopoctivoly, as sholm in Figure Ila) must detach at A and B, and free shcor layers will develop in the regions A-C and B-C, dividing the stagnant base region from the supersonic streams on either side. At G, the two shear layers evidently converge in a recompression region.

-9-

If sufflciont blood flow is now inJected to cause - to exceed ‘b PCC

unity, the expansion at A is replaced by a shock, and the flow pattern is modified to Figure Ilb.

Alternatively, if the pressure ratio across the nozzle is loworod, tho expansion fan at E may be replaced by a shock, and the flow pattom wll then be as Figure 11~. Further reduction in nozzle prossurc ratio may cause the internal jot to separate from the walls, thus effectively lncroasing the width of the base. The approprratc flow pattern for the scparatod condition is shown in Figure Ild.

V!ith subsonic external flow, the expansion fan or shock ot A is repiaced by a zone of gradual expansion or compression, and the rccompres- sion region at C may bc expected to exhibit difforcnt characteristics,

Figure 12 presents tho performance of tho short nozzle with a coplanar exit, in supersonic external flow. The flow pattern hero is evidently of the typo dcpictod in Figure Ila. Small quantities of bleed flow (around 2 per cent of the prrmarJ mass flow) result rn a significant incrcaso in base prossurc (some 40 por cent), but little further improvc- merit is realised with higher quantities of bleed flow. For values of n m CXC~SS of 5 por cent, tho bass pro ssuro is sensibly constant. Figure 12 also indicates that a reduction in exhaust pressure ratio

i E.P.R. = 2 causes a fall in the lcvcl of base prossure ratio.

> The bnso pressure chsractcristics of the same nozzle in transonic

and subsonic external flow are given in Figure 13, for values of E.P.R. appropriate to a supersonic transport aircraft operating at thcso flight conditions. In the case of &,,= 1.1 (Figure 13a), reduction of E.P.R. has resulted in on increase of bosc pressure lcvcl. This trend is clearly opposite to that observed in Figure 12, whcrc tho external Mach number and E.P.R. were higher. An effect of this nature is consistent with tho results quoted in References 1 and 2, which indicate that, for a particular external Mach number, n critrcal value of E.P.3. exists at which tho base pressure ratio 16 a minimum. Ghcn E.P.H. is reduced below this critical value, the base pressure ratio incroasos rapidly; above it the bsso prcs- sure ratlo rums slowly with lncroase of E.P.R. Reid and Hastings2, work- mg only at L&,= 2.0, attribute this dlscontrnuity to a change of tho recompression shock pattorn in the oxtornnl flow at C (Figure Ila), but Reference 1 shows a very similar bchavlour of base prassuro with change of E.P.R. in subsonic external flow.

A consideration of Reynolds number suggests that the primary nozzle boundary layor is turbulont throughout,for the conditions of those tests. At E.P.R. 4 in Frguro lja, the nozzls applied prcssuro ratio

( A.P.R. = pt

> varies from 5.8 to 5.0, and the nozzle should thcrcforo be

pb running full. This will, of course, also be true at any highcr E.P.R. The flow pattern than rcscmblcs Figure Ilc.

In the case of Figure 13b, relating to an external Mach number of 0.69, base pressure ratio is soon to bc romarkabl indopcndont of both

$b bleed flow rate and E.P.R., the gcncral 1~01 of - being around 0.9.

PO3 The flatness of those curves is apparently associated with the prcsonco of a subsonic external stream, a rslatron borne out by other results to

- 10 -

be mentioned later. Just the same effect has bccn obtained at 0.N.E.R.A.4, using a baso blood modol very similar to that shown in Figure 5a, with an aroa ratio of about 2.6. Furthermore, roccnt theoretical work by Nash3 does in fact predict that base bleed is likely to bo ineffective in sub- sonic external flow at those values of Reynolds number.

Performance of thG long model with a coplanar exit is prcsonted in Figure 14, for external Mach Numbers of 2.2 and 1.8. Those base pressure characteristics have a difforcnt shape to those discussed carlicr, and the reason for this is not closr. Howcvor, the senso of variation of base prossuro ratio with E.P.R. is similar to that obscrvod in the short nozzle results. In Figure 15, the porformancc of this nozzle at subsonic speeds 18 glvon. Hcrc again, the trends are similar to those obscrvod for the short nozzle, and baso blesd is clearly ineffective.

Shown for comparison on Figures 14 and 15 arc some data extracted from Rofercnce 1. The nozzle tharo had an aroa ratio of 2.90 (dosign prossuro ratio 20) and a thin base, without bleed. It 1s intoresting to note that, for the case of zero bleed fiow, thsre is fairly close general agreement botwccn the values of base pressure ratro obtained from Referonce 1 and those from both ‘basic’ models in the prcsont tests, dcs- pito quite large differences of design pressure ratio and base height (comparo Figure 12a with 14a, 12b with l&b and 13b wltii 15b, for matching conditions of C.P.R.).

Now if the base region were able to be shielded from tho cxtcrnal flow, it is reasonable to suppose that the benefits accruing from base bleed might be increased. This should be ospccially so in the case of a shortened primary nozzle at cruise conditions, where the internal static pressure in the cxit plane is higher than ambient. To this end the extended shroud mod21 of Figure 6b was tested.

Figure 16 depicts the flow pattern in the base region of tho cxtendod shroud system, for supersonic extsrnal flow and various nozzle operating conditions. In Figure 160 the operating prossurc ratio is sufficient to cause the internal flow to expand at the primary nozzle oxlt (A). A frso shosr layer develops in the region A-B. Subsequently tho Jet reattaches on to tho outer shroud at 9, with accompanying recom- prossion. At tho same time, the free shear layer is rehabilitated on tho shroud to form a nGw boundary layer. Finally, further expansion or com- pression will occur at C, dopending upon overall conditions. Clearly, In this case, ths base region 1s completely ahioldod from tho external flow. So far as the internal flow is concerned, base pressure theory (e.g. Rcfcrsnoe 5 relating to backward-facing stops) indicates that a reduction in E.P.R. may be expected to result in a proportional lowering of the base pressurs ratio, the A.P.R. remaining constant for a given value of I”, provided always that the prim&rJ Jet reattaches on the shroud. This 1s sayrng no more than that the region A-B-C in Figure 163 beOOmGS cffoctivcly part of the internal expansion system, controlled only by tho pressure at C.

Corresponding to Figure 16b would be the casG when applied condi- tions are the samG as for Figure 16a, but the shroud length IS reduced so that rcattachmcnt cannot occur. The recompression zone will then lie outside the nozzle, and buhaviour ~11 be similar to that already described for the coplanar exit models.

- 11 -

At low values of A.P.R., the expansion at A is replaced by a shock which tends to deflect the internal flow away from the shroud, and ~111 eventually prevent reattachment regardless of shroud length (Figure 16~). St111 further reduction of A.P.R. gives Figure 16d, in which the internal flow is fully separated within the primary nozzle. In all these cases the external flow is able to influence ccndltions in the region A-C, so removing the immunity of base pressure to external flow enjoyed when reattachment takes place within the shroud.

At conditions when the external flow is subsonic, the internal situation will generally resemble that in Fq,ure 16d, with no reattach- ment. The system is therefore likely to behave in a manner similar to the coplanar exit models, which were then insensitive to bleed flow.

It can be realised from the foregoing discussion that, for an extended shroud model to operate efficiently at cruise, the shroud should be long enough to permit internal reattachment. The critical length will depend upon the value of base pressure created in the region A-B ~;;-,‘54 t which in turn is affected by the amount of bleed flow

.

The results of tests on the extended shroud model are presented in Figure 17, for supersonic external flow and various valuee of

E.P.R. pb %hen - Pp,t

is plotted against J.I, the same results fall fairly close

to a single curve (Figure 18), as would be expected if sufficient shroud length were available for reattachment to take place throughout the range of bleed flow.

‘b For sero secondary mass flow, p = If the internal flow pt.

0.036.

is assumed to be inviscid, then once t e baoh number at the primary nozzle exit is specified, the position of the jet boundary streamline may be calculated by the method of c aracteriotics (in this case for axisymmetric flow). A computer programme %

% has boen written for this purpose, and the

result obtained when - PP t

= 0.036 1s shown m Figure 19. In this compu-

tation, the outer shroud is assumed to be absent, and the boundary of the freely expanding jet 10 dotormined. The external shroud location 1s then superimposed on the drawing, to give an approximate indication of the minimum length for reattachment. Viscous effects will introduce some modification to the picture. At eero bleed flow, the shroud is of ample length. Pb The corresponding flow pattern for - = 0.085 (corresponding to n = 0.08) appears in Figure 20. PP, t In this case, It ~011 be observed that the outer shroud, no matter what its length, cannot contain an inviscid free-jet. However, m the presence of secondary flow, the reattachment streamline can no longer be assumed to coincide with the inviscid jot boundary. This could mean that tho picture given by Figure 20 is to some extent pessimistic. The scatter of the experimental points in Figure 18 at higher bleed flows may be due to partial failure of the primary Jet to reattach.

Values of base prossure ratio obtained with tho oxtendod shroud model in subsonic external flow are given in Figure 21. In their ‘flat’ character these curves resemble those for coplanar models, as previous consideration of the flow patterns suggested.

- 12 -

Test results for the boat-tailed model of Frgure 6a are shown m Figures 22 and 23. Figure 22 is for Ma 2.2 and 1.8. These curves exhibit a maximum value of base pressure ratio at bleed flow rates varying between 4 and 6 per cent. The pronounced fall in the level of base pres- sure ratio at high bleed flows is thought to be due to the increased velo- oity at which the secondary air is discharged through the constricted bass passage. Figure 23 relates to transonic and subsonlc external flow, and comparison with Frgure 13 for the 'basic' short nozzle shows little difference as a result of boat-tailing'at these conditions.

Before any test data became avaIlable from other builds, some opinion held that the performance of the short nossle with P coplanar exit mrght be improved by introducing partial blockage in the base region, and causing the bleed air to be discharged at high velocity. The models illustrated in Figure 7 were accordingly tested at & 2.2. Figure 24 indicates that the high rnJection velocity of the bleed air has In fact produced the opposite effect to that required. As can be seen, tho base pressure ratio remaans either almost indopendcnt of bloed flow, or actually falls with increasing flow.

In Figure 25 is presented a comparison at M',2.2 between the base pressure characteristics of all the mAdo builds with a short primary nozzle. It 1s evident that the extended shroud model makes much the most offcctivo uso of bleed air. At low bleed flows, the boat-tailed build is superior to the 'basic' short nozzle, although inferior to the oxtended shroud. This advantage disappears at larger bleed flow, when the velocity of blood discharge becomes too high. Still greater discharge velocltios occurring in those models with partial base blockage are clearly dotrrmental.

It should be borne in mind that some further drag is associated with the boat-tail. This component is included in the overall thrust efficiency, to be discussed in the next Scctlon, and was derived from pres- sures measured on the boat-tail surface.

A srmilar comparison is made in Figure 26 for three builds with a short primary nozzle at b&,0.7. No si@rficant differences can be detected.

To sum up the effects of gcomotry and bleed flow on base pressure ratio alone, without regard to the overall picture, it IS apparent that:-

(1) Elced flow can increase base pressure ratio in supersonic external flow, rapldly at first and subsequently more gradually, so long as discharge velocity is low.

(ii) Boat-tailing as a general principle offers some improvement ln SuperSOnlC eXtCrna1 flow, SUbJeCt to (1) above.

(iii) An cxtcndod shroud, of length sufficient to allow reattach- ment of the primary Jet, successfully shields the bleed discharge from a supersonic external stream, and permits much higher values of base pressure ratlo to be attained for a given bleed flow. The effect IS particularly marked at high bleed flow.

- 13 -

(iv) With subsonic external flow and appropriate primary nozzle operating conditions, the value of baso pressuro ratio is virtually constant regardless of bleed flow or geometry.

4.2 Overall efficioncios

To enable the performance of the various nozzle configurations to be compared directly, a form of ‘overall efficiency’ (n) has been employed. The expression for n is given in Appendix II.

pb This enables base pressure

characteristics in terms of i;- versus n to be converted into efficiency characteristics (i.e. q versuzu) , which provide a true measure of the effectiveness of base bleed as a method of reducing base drag. It should be noted that in the derivation of the efficiency expression, it is assumed that the secondary air is captured at free-stream conditions, and its inlet momentum has been computed on this basis. This momentum tenn can be very important at high bleed flows.

It 1s unnecessary to convert all the base pressure data into efficiency characteristics, since it has already been seen that the base blockage modifications to the short nozzle proved unsatisfactory. Furthermore, it has been observed that the performance of the short nozzle is considerably improved when the outer shroud is extended, and so it only remains to compare the following configurations on an efficisncy basis:-

(a) short nozzle with extended shroud,

(b) ‘basic’ long nozzle with coplanar exit,

(cl short nozzle with boat-tailed exterior.

In figure 27, the overall efficiency of these three configurations is plotted against the bleed flow ratio p for l&,2.2 and E.P.R. 20. The corresponding characteristic for the ‘basic’ short nozzle with coplanar exit is included for comparison. These results reveal the following points of interest I-

(i) For both the ‘basic’ short nozzle with coplanar exit and the boat:tailed model, the inlet momentum of the secondary air becomes a dominant factor in the expression for n

pb as bleed flow increases, which the rise of - with n is POD

insufficient to offset. This causes n to fall as u increases.

(ii) As the base pressure characteristics would lead one to expect, the efficiency of the extended shroud model is markedly better than that of the ‘basic’ short nozzle.

pb In this case, tho rapid increase of F with p at low bleed flows is sufficient to OvercomeOOthe effect of the secondary inlet momentum and the efficisncy mcreases slightly up to a value of u = 0.02, whereafter it falls. The maximum efficiency is, however, only marginally better than that at zero bleed.

- 14 -

(rrl) The best effrcrency 1s obtarned with the ‘basic’ long nozzle and no blood.

On Frgure 27 1s also shown a value of n obtarned from Reference 1 for a nozzle of design pressure ratlo 20, wvlth a thin base and no bleed, operating at the same condltlons. This nozzle had a value of overall area ratlo (maxrmum cross-sectlonal area/throat area) of j.06, as compared with 3.39 for the present models. It should be noted that the efflclency values quoted In Reference 1 ordlnarrly exclude the drag force on the annular base. For comparison with the overall efflclancy used In tho present tests (Append= II), It 1s necessary to deduct the base drag term as given m Reference 1.

Slmllar comparrsons to that 1n Figure 27 have been made for various values of %.P.R. and external hfach number. At supersonIc external speeds, the trends arc consistently as lndrcated In Figure 27.

It ~111 bo observed that the nozzle tested In Reference 1 has a hrgher efflclency at cruise condltlons than any of the base bleed models considered In this report.

A typical comparison with subsonrc external flow (I&,= 0.70, E.P.R. = j) 1s shown In Figure 28. It ~11 be observed that under these condltlons, nono of the models bcneflted from the lntroductlon of base bleed, n consistently fdlllng as n I.S Increased. A corresponding value of q obtalncd from Reference 1 1s agarn mcluded.

5.0 Extrrn,al boundary layer measurements

Theoretical work3 lndlcatcs that the level of base pro-sure achieved In tbs type of oxperrment IS Influenced, to a great extent, by the thick- nesses of the rntornol and sxtcrnal boundary layers. To provide a datum for tile test results, It 3.6 therefore necessary to measure or cstlmato the approprratc boundary laycr parameters.

No attempt was mado here to take measuromonts of the lntcrnal boundary layor. A method of estrmating boundary layer thlckcess In a pro- pcllrng nozzle may be found In Appendrx IV of Reference 7.

The external boundary layer In the model cxrt plane was explored by means of the pltot rake dcscrlbod In Section 2.2. Cxperlmcntal velocity profiles were found to be lndepcndont of free-stream Mach numbor, and the results were therefore averaged. Vcloclty profiles obtalnod In the 12 m.‘x 12 In. supersonIc line and the 11.3 m. transomc llno aro dls- played ,n Frgures 29 and 30 respoctlvely, where they arc conparod nlth one- ninth and one-eleventh power law profllcs. It wrll be observed that the expcrlmontal results do not conform to clthcr of these power laws. The approprlatc boundary layer thlckncss for the supersonrc line was 0.80 m., whrlst for the transonlc llnc It was 0.62 m.

These values amount to 23 per cent and 18 @r cent respcctrvely of the overall model diameter. For comparison, the englnc nnccllc of a supersonIc alrcraft might be expected to grow a boundary layer around 12 per cent of Its diameter by the nozzle outlct plane. For a g-lven aftcrbody gcomstry, this ~mpllos that tho ratlo of boundary layer thick- ness to base height 1s unropresentntlvely high In the present work. On the cvldcnce of Reference 8, there IS In consequence a goneral tendency to over-estimate base pressures.

- 15 -

By assuming that the total temperature throughout the boundary layer 18 constan<, It 1s possible to calculate, for a given external Mach number, the displacement thickness (o*) and momentum thickness (f3) from the experi- mental velocity profile. Variation of these quantities with LI, is pre- sented in Figures 31 and 32.

6.0 Conclusions

In most of the nozzle configurations which were tested, base bleed was observed to raise the level of the base pressuro ratio, so long as the external flow was supersonic and the nozzle was running full. When the external flow was subsonic and appropriately low pressure ratios mere applied to the nozzle, base bleed was found to be consistently ineffective.

On comparing the results in terms of an overall efficiency, and debiting ths fres-stream inlet momentum of tho bleed air, it is found that no significant increace in efficiency can be obtained by em@oymg bass bleed in any of the configurations under any conditions. It would appesr that base bleed does not afford a gonor method of improving the perform- ance of a propelling nozzle insta113tion.

There is dcfinitoly no case for cutting short tho internal expansion surfaces of a nozzle in order to provide an aren for introduction of baso bleed. In the case of a long-range transport aircraft, better all-round nozzle performance can be obialncd by continuing tho primary expansion at least as far as is required to reach ambient pressure at cruise.

The authors wish to acknowledge the assistance given in this work by Mr. C. Overy, Nilss EL Falers and Kiss V. Soarle.

This experimental investigation was undertaken at the request of the British Aircraft Corporation, viho designed and built the test models. Tho experimental work was carried out at the National Gas Turbine Cstablishment, and the results presented m this report have been independently asscssod. Figures l+ to 8 have been taken from the ralevant dosign drawings, with ths kind permission of the British Aircraft Corporation.

- 16 -

1 G. T. Golesworthy M. V. Herbert

2 J. Reid R. C. Hastings

3 J. F. Nash

4 P. CarAre M. Sirieix

5 J. F. Nash

6 J. B. Roberts

7 11. V. Herbert D. L. Mertlew

8 G. T. Golesnorthy J. B. Roberts c. Ovary

Title. etc.

The performance of a conical oonvergent- divergent nozzle with en area ratio of 2.9, in externel flow. A.R.C. c.p.891 November, $963

The effect of a central jet on the base pressure of a oylindricsl sfterbody in a supersonic strenm. A.R.C. R. & M.3224 December, 1959

Private communication 1963

Unpublished work at O.N.E.R.A. Ch&illon-sous Bagneux, France, 1763

An znnlysis of two-dimcnsiond turbulent base flow, including the effect of the approaching boundary l.ver. A.R.C. R. & ht.3344 July, 1962

Unpublished work at N.G.T.E. 1963

The design point porformctncc of model internal expansion propelling nozzles, with area ratios up to 4. R.R.C. R. & Ia. December, 1963

The perforwnce of conical convcrgent- divergent nozzles of area ratios 2.44 and 2.14 in external. flow. S.R.C. C.P.893 February, 1964

- 17 -

TABLE I

Wozzlo operztmg condltlons

,1.

J3od.01 build

; Short nozzle wth coplanar exit

j/

;’ Long nozzle wth coplanar exrt

Short nozzle with :/ oxtondcd shroud

/, Short nozzle wrth

I boat-tollmg I

ji

Short nozzle mth partml j base blockago - typo (a)

S‘ _ 1 j’ Short nozzle wth partml

bsso blockage - type (b) ::

Mach No. Exhaust pressure ratios

2.

2.20 20.44, 15.52 1.80 14.92, 12.19 1.10 7.01s 4.05 0.69 5.66, 3.01

2.20 20.52, 18.45, 16.45 1.80 11.75 0.89 yji, 3.01 0.70 5:07: 2.96

2.20 20.01, 11.94, 15.94, 14.02 2.01 19.96, 17.98, 15.91, 24.03 0.88 6.06, 3.02 0.70 5.06, 3.02‘

2.20 1.80 1.09 0.70

19.50, 17.84, 16.00, 13.95 14.95, 12.00 1.12, 3.97 5.00, 3.00

2.20 19.95, 17.95, 16.25, 14.40

2.20 20.05, 17.95, 15.50, 14.02

_- - -.

- 18 -

Ayp~mIx I

Nototzon

A cross-sectional area

D drag force

F

K

P pressure (st&tlc unless otherwise stated)

Q mass flow

R gas constant for xl.r (= gel ft lb/lb%)

T temperature (static unless othermse stated)

v , velocity

a primary nozzle divergence half-angle

P afterbocly boat-tail engle

6 . boundary lay& thickness

60 boundary layer dwplacement thickness

e boundary layer momentum thi&ness

rl overall thn& efficiency of nozzle and bleed system

(see Appendix II)

% primary nozzle internal gross tllot efflclency

"s bleed a~ momentum efficiency

IJ bleed mass flow ratlo ( ’ Qs% = \ ?PRp,t

A$.R.

E.P.R. Exhaust pressure ratio = Primrf nozzle entry total pressure Ambient pressure

- 19 -

AITm1x I (cont'a)

* isentropic conill+aons 111 prunery nozzle throat

b base

e

m

n

P

s

t

E.T

co

pr2.ma-y n0wJ.e exit

maximurr, cross-section of nacelle

cross-section of nacelle at ex3.t plx~e

primary nozzle flow

secomla-y or bleed flow

tot&l head condltiona

boat-t&l

amblent condxtions

- 20 - 0

AW?NJIX 11 _- _ --

The overall thrust ef'ficlency of a propellmg nozzle _------- with base bleed

- by -

Y. G. E:. Lewis and M. V. Herbert

\ Case A. Conical prunary nczzle running full.

-fe shall define the following fcrce terms:-

FP = momentum + pressure thrust of pruner au at primary nozzle outlot

..=.(I)

Fs = momentum of bleed au at outlet + base pressure thrust

QSVS = 11, *-g--+ :A,- &) @I, -pm) ..“(2)

Da = Inlet momentum of bleed au, assumed to bc taken from free stream

Qs% = g

DB,T = boat-tall drag of nacelle afterbody

. ...(3)

= (il, - A,) (pm - PB.T) . ...(4)

- 22 -

,

the signs being taken such that

Total force = Fp + Fe - De - 4-T

We further define

F. jJ&

p,ldeal = g

where Vis is the fklly-expended rsentropx velocity correspondir& to

PO t -p" ( = E.P.R.) M

The overall efficiency is given by

I' + Fs - Ds - DB T rl==

. Fp,kieal

It IS oonvenlent to express ell force quantities non-dlmenelonelly in the f 0x-m

Fp , ideal

AQP,t = !pL

J.PI',t (5) . . . .

$6 where % = AVp,t

2 ,J,.- ’ ( ... E.E'.R. 1

. . . . (6)

where f~e.is the isentropu veloolty corresponding to prunary A,

nozzle area rat10 - ( 2

A* 1 + cosib ) p t

D.P.R. is the primary nozzle deblgn press UT0 rat10 = -p 0

,mx is the design-pomt efflcxncy of the prun~!f nozzle

- 22 -

Nwi 8,RT, vs = A&

and putting p = Q&i- QPPG

Q 1' we get .2.2=

I-I~K~'R ~ (-@Pp,t)a , T,

g g *34, Ta,t

FS . .-. .D A*Pp,t

D S

Gp %a

*"Pp,t = -.-

"G

.‘.‘(7)

. ...(a)

. ...(Y)

Thenq 1s evaluated as (6) + (7) - (8) - (9) --

(5)

Notas

1. PB.T may be determuwd apprrximately for a given boat-tail angle 9 by using the appropriate two-dimenslonal Prandtl-Meyer relation, or from experimental pressure measurements if available.

2. In the present tests Y for all three streams can be taken as 1.4, SO that Kp = 0.3966.

3. In (7) and (8) It will usually be pussible t? write Ts,t = Tm,t. It2 these particular tests It 1s true tc put Tp,t = T,,t = T",t*

4. In many cases It ~11 be suffxlently accurate to take Ts/T3 t c 1 in (7), the first term of which amounts t,) a very small'part of the wtile fzce. For the same reasonns 1s also taken to be unity in the absence of other wf'ormat~on.

5. If the velocity of bleed air discharge 1s sufficiently great for the approximation in Note 4 to be urqustifled, the value of Ts/T3,t can be fgund as follows:-

- 23 -

T Y -1 while + = 1 + -+ Ias"

s

u Ts,t , Hence T

( 1 Ts =

s Lg. c!gz ($(A.P.R.)

s s

6. The value of prmsry n~zzlc design-paint effu?iency'~,mBx depends

on D.P.R. and cone angle c,, It may either *be calculated approxi- mately or derived from static thrust calibrations. For the primary nozzles of these tests, vnth a = IO0 and D.P.R. values 11.12 and 18.76, a fqure of 0.988 was estunated from Reference' as a~prop- riatc to these test corxiltlons.

7. II' the primary nozzle should be other than conical, e.&. two- dimensional, the same relations are applicable except as regards

the te,rms %),max d 2

1 + cosa *

Case B --. * Conical prunary nozzle wzth internal separation,

Relatwns (2), (3), (h), (5), (7), (S), am3 (9) remain as l~1 Case A, ad Notes 1 to 5 apply. Alternative treatment must, holvever, be used for the qwntlty F Ve ml1 now wr E

when the prxmary nozzle is no longer running "1. te

I”P = [ i, 2P.I 5 g + @Ll - %J or &+== +%(A;, -E;R > L 1 . . . . (IO)

p,t i~R * .;* . * *

where 1s the isentroplc velocity term corresponding

APR

to f$$ (= A.P.R.)

-9 1s the internal efficiency measured for the primary nozzle opcratm@; under stntxc condltlons at thus value of A.P.R:

Then q 1s evaluated a~~") + (7) r (8) - (') 5

- 24 - Report No. R.259

Notes

8. To take an example, appropriate to curve 2 of Figure 28, for which

h$$ = 0.7; E.P.R. = 3; D.P.R. = 11.12

we see from Figure 13 that Pb/p, a 0.9, giving A.P.R. = 3.33. At thx condition a suitable value of boundary layer separation.

% would be 0.92 with turbulent

9. It 1s worth observing that any primary nozzle of sdar geometry but different D.P.R., provided that turbulent separation still occurs at the same A.P.H., wiLl have a value of tion quite close to that given 111 Note 8 above. 'TZL3%L"Z$~e understood by reference to the sketch below, d.ra%vn for a partzcular A.P.R.

a b o

P P- P,t

A F

Due to the nature of the pressure riso occurring in a nozzle with turbulent boundary-layer separation, most of which is concentrated at the separation point as depicted, it will make very little difference to the internal performance, as represented by the shaded area, whether the nozzle ends at station a, b or C. For instance, e,t the condition of A.P.R. 3.33 with a turbulent boundary layer mentioned in Note 8, a IO0 nozzle of D.P.R. 20 has ?jp = 0090 (Refs. 1 and 7), as compared with the value 0.92 taken for D.P.R. ?1.12. It is therefore generally unnecesscxy for work of the present neturz to carry out static thrust calibrations for every nozzle of different D.P.R. encountered.

IO. In the special cx,e of no bled flow (cl = 0) an< no boat--tail (P = 0), the expressIon for overall.effudency reduces to

- 25 -

It my be noted that, rmth the, further condition of 41 = A, (zero base thickness), this reduces to the famllxr symmetrical ralatlon for angle-stream nozzle internal performance

A, + F - 1 E.P.R. =

m whxh the notation is now as used, for mstmce, in Reference 1.

D 76932/l/125875 I& IO/'66 R & XT,

FIG. I

MODEL PROPELLING NOZZLE TEST RIG.

FIG. 2

NOZZLE

SLOTTED

----PORTION

OF NOZZLE

INLET OF

’ SLOTTED

NOZZLE

STING

STING CARRIER SECTION IN TRANSONIC LINE. ,’

I FIG. 3

I 9 I

T

STING CARRIER SECTION- 1N SUPERSONIC

LINE.

FIG. 4

c I r c .

i 7-

MEASURNC

/PRIMARY NOZZLE

/OUTER SHROUD

~~~NN~JLAR hmst.m~~c

SECTION MVIDED

INTO SIX PORTS

i PRI MARY PITOTS

F ORTS

RING VALVE

.- INLET PITOT STATIC

I RAKE

UTER STING TUBE

INNER’ STlNC TUBE

GENERAL CONSTRlJCTlON OF TEST NOZZLE.

0 b

THROAT DIA. I.90

PRIMARY EXll DIA. 2.720

SHROUD OUTYDE DlA 3510

THROAT DlA I.90

PRIMARY EXIT DIA 3 172

SHROUD OUTSIDE DIA 3.510

GEOMETRY OF SHORT 8 LONG COPLANAR MODELS

FIG.6 (a a b>

THRQI\T DIA I .90’

PRIMARY EXIT UA. 2 720”

SHROUD OUTSIDE DIA. 3.275”

ON’PARALLEL

PORTION OF 0 a

SHROUD

0 b

THROAT DIA. 1.90”

PRIMARY EXIT DIA. 2.720”

SHROUD OUTSIDE DIA. 3410”

SHROUD OVERLAP I, 268”

GEOMETRY OF EXTENDED SHROUD 8 BOAT-TAIL

MODELS

FIG.7(0 L b)

THROAT DIA I. 90”

PRIMARY EXIT DIA 2.720”

SHROUD OUTSIDE DIA 3.510

GEOMETRY OF BASE BLOCKAGE MODELS.

FIG. 0

LOCATION OF PITOT RAKES

FOR BOUNDARY LAYER SURVEY

FIG. 9

pb

x0

0

0

0,

0

0

0

0

0

0

--

____

0

4 -

r

I I

I I

?.A,=2 20 E P R=IS.SO

I I LOCATION OF STATIC TAPPINGS

AXIS OF NOZZLE -- --

I

I

2 O-04 0.06 0 08

0

I

-

A COMPARISON OF BASE PRESSURE READINGS FOR SHORT NOZZLE WITH COPLANAR EXIT.

2

‘b - PCL

I

I

I

1.;

I.(

OE

06

04

02

A

o-

e-

6-

4-

1 LOCATION OF STATIC TAPPINGS

_ AXIS OF- NOZZLE _

I

I

Me = 2.20

EPR I 18.45

--

0.02 0

4

/

D

-x-

0

FIG. IO

A COMPARISION OF BASE PRESSURE READINGS

FOR LONG NOZZLE WITH COPLANAR EXIT

08

06

04

02

A I-

l-

+

#

L 0

A-

,-

1

!

0

IC

Pb

c

08

da

04

0*2

a

2 0

M&2.20

k4 0 6 0

0 E P R.=20.44 - x EPR =I5 52

3 0 0 P

X EP.R =I4 92 =I4 92 -@EPP r12 19 r12 19

0 12 0 14 0 6 0 08 0.

PERFORMANCE OF SHORT NOZZLE

0 P WITH

COPLANAR EXIT.

pb PC0

0.8

06

0.4

0.:

C

I’0

pb PC

0.0

0.6

0.4

02

0 0

0 a

,

@?I L

C IC

D.0

FIG. 13(ae b)

I

2 0.04 0 06 c

hi#O 69

)2 0.04 0.06 0

m ERR-7 01 a E PR14.05

I BE 0.10

P

- Q E.PR -5.66 X E P.f?=3 01

-I 1s 0.10 P

PERFORMANCE OF SHORT NOZZLE WITH COPLANAR EXIT.

FIG. l4(a a ti

04

0.:

I.0

pb

z

x E P FL= 20.52 0 E.PR.- I e.45 BEPA- 16.45

FROM AEFI t&=2 40 k-2 20 0 EP.R - 20.89

0 &RR = 18-16 A EPA= IS-05

@

4 0’06 O-08 0.10 P

0.02 0.04 0 06 0

- 14.72 X ERR.= I I 75

FROM REF I M-91.75 m ERR =I5 25 0 ERR=11 93

PERFORMANCE OF LONG NO2 ZLE WI TH COPLANAR EXIT.

04

pb Pm

0.1

0.

0 .,

(

I.(

pb PO2

0.1

0

0. J

0:

(

B-

w

b-

4-

2-

3, 0

W 0 2 0

FIG. lS(ar b)

0 E P R.= 6.02 x E P.R - 3.01

I 0

t&,=o~e9

--I-

0 02 0.04 0 06 ,

7 0

-I

@EPA=507 XE.P R.= 2.96

l REF. l REF

E.P.R=5 0 E.P.R=3.0

IO 7

WITH PERFORMANCE OF LONG NOZZLE COPLANAR EXIT.

I *b

IO

08

0.6

04

0.2

0

: 17.96

=-IS 96

PR=l4 03

0.06 C

0.10 F

PERFORMANCE OF SHORT NOZZLE WITH EXTENDED SHROUD.

0.10

pb

PRt

0 09

0 00

0 07

0.06

0 05

0 04

0.03

0.02

0.01

0

,

0 0 02 0. I

4 0.06 0

7 0

x k-220 6 &=a-oI

-

FIG. I8

0 C ‘0

PERFORMANCE OF SHORT NOZZLE WITH EXTENDED SHROUD.

FIG. 2l.(aeb)

Of

02

3- d

/

i -

I-

)- 0

pb

P,

0

0

0

0

--I-- (4

I 0 02 O(

T / d,=o~Be

c

0 06 0.

0

x

+

X

” *

r t&=0 70

I

0 02 0 04 0 06

PERFORMANCE OF SHO ?T NOZZLE WITH EXTENDED SH 4OUD.

x EPR=6 06 0 E PR.=3 02

x ~1?1?=5.06 0 EPR.53 02

FIG. 22 (ad)

0t

04

02

C

‘b - PU3

I-

,-

)- 0 A

I-

t

08

06

04

0*2

C

A

It&‘2 20 --I-- EPR= I AEPR=I XEPR=rl QEPR=l

(b>

T-i 9 50 7.84 6.00

3 95

I I * 2 004 0.06 0 08 0 IO P

0

X E.P.R - I4 95 0EP.R =I2 00

16 008 0 IO

c

P

PERFORMANCE OF ShORT NOZZLE WITH BOAT-TAILED EXTERIOR,

FIG. 23 (tub)

0.6

04

02

0

IO

pb

P,

08

06

04

02

0

f 0

t&-I 09

XEPRt7 I2 - OEPR=3 97

I

0 0 02 0’04 0 06 008 0 IO P

III (b) i

I OEPR-5.00 XE.PRr3 00

*=o 70 (

0 0 02 0 04 0 06 0 08 0 IO P

PERFORMA ‘JCE OF SHORT NOZZLE WITH- BOAT-TAILED EXTERIOR.

0t

‘b - PCC.

01

0.

0;

FIG. 24 (a u b)

x EP.R =I9 95 0 EPR.El7 95 + EPR-16 25 A E.P R -14.40

06

0.4

02

C

7 0

OEPR-2001 + E.PR-I7 95 0 E.PR.sl5.50

X EPR ~14.02

t&*2 20

--r

(&OMETRY AS IN FIG 7b)

(b)

I I I I I I B

0.02 004 006 008 0 IO A

PERFORMANCE OF SHORT NOZZLE WITH BASE BLOCKAGE.

FIG. 25

I 6

1.4

I.2

O-6

0.4

0 CObLANAR EXIT’(BASIC) ’ + BOAT-TAILED EXTERIOR

X EXTENDED SHROUD A BASE BLOCKAGE (AS FIG 7a)

Q BASE BLOCKAGE (AS FIG 7b)

I 7

+ #C 0 0 0

T 2’

\

r-

MO. r 2.20

E P.R.eZO

L 0.02 0.04 01

I .

ks

36

-I- --cc El ‘--

, 0 08 < 0 P COMPARISON OF RESULTS FROM SHORT

NOZZLES.

FIG. 26

b >-

3 5-

5-

6-

Z-

>- 0 a 2 0, 4 0

7

X COPLANAR EXIT (BASIC) Ei EXTENDED SHROUD B BOAT- TAILED EXTERIOR

6 O-08 C C

IO k

COMPARISON OF RESULTS FROM SHORT NOZZLES.

FIG. 27

0 93

0.92

0.91

0.90

FROM REF.1 I 0 P.R 20 THIN BASE

b&=2*20

E.P.R.e?O

Q) SHORT NOZZLl? WITH EXTENDED SHROUD

@ SHORT NOZZLE (BASIC)

@ SHORT NOZZLE WITH BOAT-TAILED SXTtRloA

0 LONG NOZZLE (SASIC)

/ I I I 1 I w

0 0.02 0 04 0.06 0.08 0.10 P

COMPARISON OF OVERALL EFFICIENCIES.

FIG. 28

I I.00 -

0.90 -

0.70 -

0 60 r

ul$.eo 7 E.PRa3 0

m FROM REF 1 DPR 20

THIN BASE

0 SHORT NOZZLE WITH EXTENDED SHROUD

@ SHORT NOZZLE (BASIC)

0 SHORT NOZZLE WITH BOAT-TAILED EXTERIOR

@ LONG NOZZLE (BASIC.

0.501 I I I I I 0 0.02 0.04 0 06 008 0 IO

/u

COMPARISON OF OVERALL EFFICIENCIES.

I I I j I I

4 ’ I I I I I (Y ,” 0 m 0 0 04

m3 - A 6 t, 6

z

EXfERVNAL BOUNDARY LAYER VELOCITY PROFILE.

I.0

0.f

0. t

0.4

02

0 EXPERIMENTAL POINTS II 3”RIG (AVERAGED)

_ = ;’ 5-0.62 u ---- U 0

--------I, U 0

2” .g=l3.L52

0 0

6) 0

II= LOCAL VELOCITY U FREE STEAM VELOCITY

FIG. 31

ESTIMATED VARIATION OF S* WITH Moo.

FIG. 32

?ESTIMATED VARIATION OF 8 WITH Moo.

a-- l I A.R.C. C.P. No.892

Febru~. l%i bbei-~~, J. B. and lIole~~orUW, (1. T.

62l-225.1r533.69l .18&93: 533.6.013.12

I AN EWERIIIEWAL INVEBTIOATION OF ‘RIE INFWENC!S OF

Bk3E BLEED ON THE Bh% DR.&i OF VARIDUS

, PFUJPEUIM: KGZLE CONFICUFKTITICNS

Var,ous prppell,ng nazle conflguntfons nlth all-Internal exw3.9lo” : were ccsced In external flosi over the llnBe of Nach Number 0.7 Co 2’2. 1” 1 nrder CO determine the effect or base bleed i1.e. the lnje~tl”” Or low

energy secondary air) on base preSsWe. A” ~ovenlll erflclency’ 1s jellned, i tilch e”“b.bles the effectiveness of base bleed, BS 8 me-“s of reducl”B the / base dma, Co be asseSSed. The results 1”dlcat.e Chat with supenonlc 5 exCen,al llm ba.398 bleed BenWalL” tends Co mlse the level Or base presS”=e, ’ b”t no lmpmeoent I” the oVeZ’“ll ertlclency 1s ObtaIned. AC SubSOnIC

external speeds, the saandary air h”s a “eBllBlble errect 0” bxse p2SSUre.

t-

A.R.C. C.P. No.&? Febnmy, 1 %lt Roberts. J. B. .aM GolesworC~, 0. T.

621-225.1:533.6vl .1&w: 533.6.913.12

AN EXPERIK?iTkL INVFSTIG~TICN OF TIE INFWZSCE OF B.SE BLED ON THE BXE DRM OF VARIOW

PROPELLING NOZZLE O3HFIGUBATICMB

vap10us pmpe111nB nozzle c0nrlBwcl0~ alth nil-l”ca?le1 erpas1on nere cesce, 1” excemn1 rim over the rrnge or Uach Nwlber 0.7 to 2.2. I” otier Co detemlne the errect or b8.w bleed (I..?. the lnj?ctlon or law energy sHmdary air) on b&Se pressure. :a *t~eraii erriciencp~ ~8 mhd, Iihlch enables the errectweneSs or base bleed. OS a meals or lY&cl~ Che base drag, Co be assessed. The results lndlcote Chat alth superso”lc external rlw, base bleed generally tend3 Co raise the level oi b”se pressu-e WC no lmpmvemenc 1” the o~epall erriclency is obceined. 4 ~U~SO”IC external 8pee18, the seamday air has ” “PBllBlble efrecf On ~a.56 pl-esure.

-~..-- __---_ ------___-_ .-- - _- .___ __

A.R.C. C.P. No&C? 621-225.1:533.G9.18.aj3: FebNWY. 1964 Robelts. J. 8. and GalesnorChy. 0. T.

533.6.013.12

AN EXPERIHEtiT& INKSTIG,‘TION OF THE INFLUENCF OF BMSE BIBD CN TIE LLSE DRAG OF VARIOUS

PROPELLING NOZZLE COtiFICURATITpJs

V,ai-loup pmpelllng nozzle confleuraClans wlth all-lntemnl expansion wre ce~ted in external rlow over the rylge or t&h Nlrmber 0.7 CO 2’2, In order Co ?etemlne the errect or bpse bleed (Le. the lnjectlo” or lorr

energy secondary all-1 on base pressure. h ‘0~ei-a efrichcyr IS der~ned which enables the eriectlveness of base bleed, zs 8 mea”3 Of red”M”B Chha base drag, Co be ~a%WSsed. lhe results 1”dlcaCe Chzt with su~ersonlc external llm baSe bleed generally tend.3 Co IXlS.2 the level Or bzsse pZX%We but no lmpmv’aent 1” the memll elrlclency 1s obtnlned. ht sUbSO”lC external SpWd3, the secondary nlr has a “egllglble MreCC 0” base pressure.

-

C.P. No. 892

0 Crown copyrght 1966

Prmted and pubhshed by

HER MAJESTY’S STATIONERY OFFICE

To be purchased from 49 High Holborn, London w c I 423 Oxford Street, London w I. 13~ Castle Street, Edmburgh 2

I09 St Mary Street, Cardiff Brazennose Street, Manchester 2

50 FaIrfax Street, Bristol I 35 Smallbrook, Rmgway, Brmmgham 5

80 ChIchester Street, Belfast 1 or through any bookseller

Printed m England

C.P. No. 892 S.O. Code No 23-9016-92