Ram-jet Performance

S..

• • .

•• •S S •.. • •.. • 3 NACA RM E55G28 CO•F .DTI4L. : • • • . . . .

. . S • S • • • •• • S. • • ••.

• • .

. S S S. •.•

• • •• •••

• • S S

••

•• • •

• ••. ..

CHAPTER XIV RAN-JET PERFORMANCE X67

-87774

By A. J. Cervenka and R. Friedman INTRODUCTION The ram jet is basically one of the most dimple types of aircraft engine. It consists only of an inlet diffuser, a combustion system, and an exit nozzle. A typical ram-jet configuration is shown in figure 128. The engine operates on the Brayton cycle, and ideal cycle efficiency depends only on the ratio of engine to ambient pressure. The increased, engine pressures are obtained by ram action alone, and for this reason the ram jet has zero tbrut at zero speed. Therefore, ram-jet- powered aircraft must be boosted to flight speeds close to a Mach number of 1.0 before appreciable thrust is generated by the engine.

Since pressure increases are obtained by ram action alone, combustor-inlet pressures and temperatures are controlled by the flight speed, the ambient atmos- pheric condition, and by the efficiency of the inlet diffuser. These pressures and as temperatures, as functions of flight speed and altitude, are shown in figure 129 for the NACA standard atmosphere and for practical values of diffuser efficiency.

It can be seen that very wide ranges of combustor-inlet temperatures and pressures may be encountered over the ranges of flight velocity and altitude at which ram jets may be operated. Combustor-inlet temperatures from 500 0 to 15000 R and inlet pres- sures from 5 to 100 pounds per square inch absolute represent the approximate ranges of interest in current combustor development work.

Since the ram jet has no moving parts in the combustor outlet, higher exhaust- . gas temperatures than those used in current turbojets are permissible. Therefore, fuel-air ratios equivalent to maximum rates of air specific impulse or heat release can be used, and, for hydrocarbon fuels, this weight ratio is about 0.070. Lower fuel-air ratios down to about 0.015 may also be required to permit efficient cruise operation. This fuel-air-ratio range of 0.015 to 0.070 used in ram jets can be compared with the fuel-air ratios up to 0.025 encountered in current turbojets.

Ram-jet combustor-inlet velocities range from 150 to 400 feet per second.

These high linear velocities combined with the relatively low pressure ratios ob- tainable in ram jets require that the pressure drop through the combustor be kept low to avoid excessive losses in cycle efficiency. It has been estimated that, for a long-range ram-jet engine, an increase in pressure loss of one dynamic head would require a compensating 1-percent increase in combustion efficiency. Therefore, com- bustor pressure-loss coefficients (pressure drop/impact pressure) of the order of 1 to 4 are found in most current engines.

The operating conditions described impose major problems in the design of stable and efficient ram-jet combustion systems. This chapter presents a survey of ram-jet combustor research and, where possible, pointsout criteria that may be useful in the design of ram-jet combustion systems.

EXPERIMENTAL METHODS Data Sources Ram-jet combustor performance data have been obtained in connected-pipe, free- jet, tunnel, and flight tests. A connected-pipe facility (e.g., ref. 1) consists of .' '-'- 244 • • •.. • •• ' NACA RM E55G28 • S. • • • S S. S I .

• • I I • • I S S • • I • • • • • : : •• :: ..: .s diffuser, a combustion chamber, and an exhaust nozzle connected by suitable

a'ucn?c

ducting to air supply and exhaust pumps. A free-jet test installation (e.g., ref.

2) consists of a ram-jet engine complete with supersonic diffuser installed down-

stream of a supersonic nozzle. In subsonic or supersonic tunnel tests (e.g., ref.

3), the rain-jet engine has been installed either directly in the main air stream or

downstream of a connecting air supply duct. Ram-jet-combustor data have also been

obtained with engines attached to subsonic aircraft (ref. 4), with free-falling

engines dropped from high-flying aircraft (ref. 5), and with engines launched by

rocket power (ref. 6).

Data Reduction Methods

most

Combustion efficiency, one of the important performance parameters for

evaluating ram-jet combustors, is defined as the ratio of the actual enthalpy rise

across the combustor to the theoretical heating value of the fuel. Combustion-

efficiency data for a ram-jet engine are difficult to obtain directly from inlet and exhaust-gas temperature measurements because of the high exhaust-gas temperatures.

For Ahis reason, indirect methods have been evolved whereby combustián-chamber pres- sures, engine thrust, or heat-balance measurements are reduced to give combustion efficiency. For applications where thrust measurements are obtained, ram-jet perform- ance is usually expressed in terms of impulse efficiency, which is defined as the ratio of actual to theoretical specific impulse.

Pressure method. - In the pressure method, the increase in momentum of the

gases flowing thro.igh the combustion chamber is determined by means of total-

and

static-pressure measurements. These pressures can be related to. the temperature

rise across the combustor by the following compressible-flow equation (ref. 7): + 1)

P

A^g rst,5(rav

T4=

(1)

2 2R

(wa+wf)

where

ri

A area

g acceleration due to gravity

static pressure

p

R gas constant

T total temperature

w weight flow

specific-heat ratio

Tav average specific-heat ratio between static and total temperature at exhaust-

nozzle throat

Subscripts:

a air

f fuel

NACA RN E55G28 CO)A' ' .. , ••• • •••

.

S S • •' • • • • S.

S • • • • • • • . .

. • .

S S • •• • •• • ••• • .

S ..

• • • S..

S •• ••• • • S.. • • •• •• • S 5.5 •• static at 4 combustor outlet

5 exhaust-nozzle throat

Equation (1) is based on a choked exhaust nozzle where the Mach number M 5

is 1

and the temperatures at the combustor outlet and at the exhaust-nozzle throat

are

assumed to be equal. This equation is rendered more exact if the nozzle

throat area

is corrected by a discharge coefficient.

Exhaust total-pressure measurements are also used to determine combustion effi-

ciency directly without calculation of exhaust temperatures by defining combustion efficiency by the relation (ref. 8)

= (f/a)'

(2)

f/a

where f/a fuel-air ratio ideal fuel-air ratio that would produce same burner total pressure as (f/a)' actual fuel-air ratio

combustion efficiency

The calculation of (f/a)' is based on the fact that air flows for burning and. non-

burning conditions are the same for an engine with a diffuser operating supercriti-

cally. Thus, total pressure at the nozzle throat for a choked exhaust nozzle

is

calculated by compressible-flow equations similar to equation (1). By assuming no

change in total pressure and temperature between the combustor outlet and the nozzle

throat, the ratio of total pressures for burning and nonburning conditions is ob-

tained (refs. 8 and 9):

IE

y+1 1

R

y+l I

FT4-0 ,b

4 ) b

b)()ib

1 ^

Wa

4,nb

I.R\

[(L2--

I ')_Jnb

where

P total pressure

Subscripts:

b burning -

nb nonburning

free stream

(f/a)'

From measured is computed from

T4,b is calculated; values of 4,b/4,nb'

tables of ideal combustion temperature as a function of fuel-air ratio. Efficien-

cies are then calculated by means of equation (2).

•.

•S.

S S .

•S. • •• •• NACA EM E55G28 oNDfrIAL : : . • • • • • • • • • S S • • •• • • • •• • •SS • S S •• • • • ••S S • •5 S S ••S •S • .

•.

• • S. •S. • Combustor exhaust total temperatures are also Thrust-measurements method. - calculated from jet-thrust measurements in wind-tunnel installations where engines are mounted on thrust balances. The exhaust-nozzle total temperature, essentially equal to combustor-exhaust total temperature, is computed by the following energy equation derived in early NACA work: p5A(p5 - + F - A5 (p5 - p0)2 - p5A5F p0 ) T4 2gJ()5(w + Wf) T5 = gR(w + Wf) - gR(w + Wf) where constant-pressure specific heat at exhaust-nozzle throat jet thrust F mechanical equivalent of heat J Combustor efficiency is then calculated from the exhaust-gas temperature in the same way as described for the pressure method.

Heat-balance method. - Combustion efficiencies obtained by the pressure and thrust-measurement methods are only close approximations to true chemical combustion efficiencies if these efficiencies are defined as ratios of actual to ideal tempera- tare rise or as ratios of fuel-air ratio (eq. (2), e.g.). An exact combustion ef- ficiency is defined as a ratio of actual to ideal enthalpy rise. A method of obtaining this true combustion efficiency involves the use of a water quench spray at the nozzle exit. The temperature of the resulting steam-exhaust-gas mixture is measured at a station sufficiently past the spray to allow complete evaporation of the water. From enthalpy values corresponding to this measured temperature, combus- tion efficiency is determined by the following heat-balance equation (ref. 10): (AHw (5) = h(f/a) where h lower heating value of fuel All enthalpy rise Subscripts: e exhaust gases cooling-jacket water - j quench water for exhaust gases w Equation (5) is used for fuel-air mixtures leaner than stoichiometric. For mixtures richer than stoichiometric, the enthalpy rise of exhaust gases nsHe is determined from (e) - + p)m(e - Ti)] ) T + [(f/a)ac - ( f/ a)] [( L •.

•.S S S .

.• S. • ••• • • S • COTFI)E1IA . • • NACA RN E55G28 . • • S S • • S S • • • •• • •• • .

•..

S .• •.• •. • • • S.. • .

. • • . S.

•• • • . •.. ..

where mean constant-pressure heat capacity of fuel (cr ) In latent heat of vaporization of fuel L v Subscripts: ac actual i inlet mixture a stoichiometric FLAMEHOLDER AND C0USTI0N-CHAMBER GEOMETRY General Considerations The problem encountered in ran-jet combustors is that of initiating a stable flame in a fuel-air mixture traveling at velocities as high as 800 feet per second.

This stabilization can be accomplished by placing a bluff body such as a rod or disk in the gas stream. A flame initiated in the fuel-air mixture attaches itself to the eddy region behind the bluff body, and this stabilized flame serves to ignite the oncoming fuel-air mixtures. The subject of flame stabilization by bodies within the gas stream is covered in chapters IV to VI, and this chapter is concerned only with the direct ram-jet-combustor applications of flame stabilization.

The simplest type of flaineholder is a baffle placed at a single plane normal to the gas flow. Baffles may take the form of rods, disks, cones, or combinations of these. More advanced baffle designs consist of U- or V-shaped gutters with the open end facing downstream, arranged singly or in annular, radial, or grid-like planar combinations (fig. 130(a)).

Ran-jet flameholders are also designed in three-dimensional forms in which the axial dimension of the flaxneholder is appreciable. A gutter-type flaneholder may be constructed with axial sloping gutters to form a three-dimensional flameholder (fig. 130(b)). A refined type of three-dimensional flameholder of wide use is the conical can where the flaneholder consists of a conical surface perforated to allow the desired open flow area (fig. 130(c)).

Integral piloting systems are often used to assist flameholders in maintaining combustion under adverse conditions. The pilot creates the low-velocity region for stable combustion by channeling a small portion of the combustible mixture into a relatively large flow passage. Pilots are frequently combined with simple-baffle flaneholders or with three-dimensional flaneholders.

A number of other 'flaneholder designs have also been investigated in ran-jet combustors. Immersed-surface types in which plates or blades have been placed downstream of a gutter flaneholder directly in the flame zone have been employed successfully. Some work has been done with types 'that have the fuel-injection and flameholding systems combined in one unit.

The following section does not aim to select one of those general types of flaneholder as being superior to the others. In general, flameholder research has aimed at perfecting each of the various types of flameholder for its own specified purposes rather than in competing one type against another.

•.

• S• • • • •• • . .

• • • •, •.. • ••• • •.

. . S • • • . S CA RN E55G28 •• NA '00: .

• • . • S - • 5 ..• • • •• • S S •.

S • •5 SI. S Simple-Baffle Types Stability limits. - Some information concerning flame stability of simple baffles is treated in chapters III and VI. In addition, recent reviews of the sub- ject have been published (refs. 11 and 12). A theoretical-analysis of the effect of flaineholder dimensions and inlet-gas variables, upon stability has been developed by considering the fact that blow-out occurs when the heat supply rate from the eddy region behind the flameholder is infinitesimally less than the heat required to ignite the approaching fresh gases.

If viscosity is regarded as a function of the 0.7 power of temperature, then the following equation for fuel-air-ratio stability limits may be derived (ref. 13): Vbl (7) P5D56T) where diameter of disk-type flameholder normal 'to flow D V velocity functional notation Subscript: bl blow-out In additional studies of stability limits reported in references 14 to 21, the effects of such variables as fuel type, mixture temperature, stabilizer size and type, temperature, pressure, and turbulence were investigated experimentally.

In actual engines, the most common simple-baffle flameholder system consists of gutters arranged in grids or annular-radial combinations. Some work has been done on a flameholder system as simple as a sadden expansion in cross section from diffuser to combustor (ref. 22).

The blow-out limits obtained in free-flight investigations of ram-jet engines with V-gutter grid flameholders are described in references 4 and 23. Reference 23 reports that increasing the blocked area of the flameholder by increasing the number of gutters in the grid tended to widen the stability limits. The gutters were all 3/4 inch wide; the effect of gutter width was not determined. A comparison of U- and V-gutter grids of approximately the same blocked area in another investigation (ref. 24) showed no difference in stability limits between these types of gutters.

Results of a series of investigations with V-gutters arranged in annular-radial combinations are reported in references 7, 25, and 26. The flaineholders consisted of several rings of annular V-gutters of varying widths, staggered longitudinally, and interconnected by radial V-gutters or flat-plate struts similar to the configu- ration shown in figure 130(a). Changes in blocked area had little effect on stability limits, but the increased gutter widths improved these limits. A correla- tion representing a simplification of equation (7) applies to the data of these investigations (ref. 25) in the form (8) La (n45) bl S.

••• • • S • •.

• . . •• •• • ••• . ... •.

• • • • . S. .

S S • . • • • • • .

• •• • S • • •• • .

S..

S S • • • S .. ... .• NACA RM E55G28 C0NPIDTI1.

• • • . 0• •• • • •• •.. ..

where (f/a)bl fuel-air ratio at either lean or rich bl ow-out M3 combustor-inlet Mach number based on entire cross section n nominal gutter width The correlation is shown in figure 131, where the velocity - gutter-width param - eter is plotted against blow-out fuel-air ratio. A similar plot (ref. 17) shown in figure 132 correlates fuel-air-ratio limits with a parameter r/V 3 , where r is a nominal circular-baffle flameholder radius and V 3 the combustor inlet veloc- ity. In this case, an exponent of unity for the flameholder dimension correlates stability limits as well as the exponent of 0.45 in equation (8).

Some work has also been reported on the use of ceramic or ceramic-filled baffles rather than steel or nickel alloy types. A comparison between a steel flameholder consisting of four radial V-gutters and a similar flameholder of alundum and silicon carbide shows a much wider range of stability limits for the ceramic baffle, espe- cially at the rich limits (refs. 27 and 28). The wider stability range with ceramic flameholders is probably due to the reduction of heat losses by conduction from the flame zone. This is in conformity with the analysis previously presented which shows that blow-out occurs when heat losses exceed the heat supply rate to the flame zone.

Combustion efficiency. - An analysis of the combustion processes must necessarily consider fuel-air preparation and inlet parameters as well as flame stabilization and oxidation. Thus in order to compare flameholders, it is essential to control the fuel-air preparation and inlet variables. The role of the flameholder is discussed in this section, with particular reference to the simple-baffle type.

Preliminary combustion-efficiency investigations were performed on a variety of flameholder configurations (e.g., ref. 29). In early NACA work, screens, solid and perforated strips, flat plates, disks, cones, and V-gutters were used as flanieholders in a 20-inch-diameter ram jet. Another. investigation (ref. 30) showed that V-gutters in series were very successful for high combustion efficiency. In general, the most widely used simple-baffle type of flameholder is formed of V-gutters, although ram- jet engines have been designed with perforated-gutter flameholders (ref. 31), corru- gated gutters (refs. 32 and 33), and the simple sudden-expansion type of flaineholder (ref. 22) often with vortex blades at the entrance to the combustor (ref. 34).

The principles underlying the operation of efficient simple-baffle flameholders can be stated briefly. The stagnation region downstream of the baffle is a stable, high-temperature zone which acts as a torch for the adjacent high-velocity mixture.

For stability, a wide baffle is desired, but this in turn increases the velocity of the unburned mixture past the flameholder; therefore, a compromise between efficiency and stability must be made (ref. 12). A second principle observed is that a con- tinuous flame path connecting all of the flameholder baffles is desired. Thus, reignition can proceed if flame is locally snuffed out.

These principles areillustrated in the investigation reported in reference 35.

The four gutter-grid flameholders used, shown in figure 133, are typical ofhe simple-baffle type. Their combustion efficiencies were about the same, with a slight advantage for the standard-gutter flameholder. In general, the configurations that gave the highest peak efficiencies had the narrowest fuel-air-ratio range of operation. This compromise of efficiency with stability limits has been noted repeatedly in combustor investigations. As noted previously, this is the case •.

•II • • • • • • ••• ••• • • • I SlI • I I •• ••.

S 1I .

• • •• • • • • I : .

•cNh1ENTLAL NACA RM . • . • ••• S €: •• I • ••. •• : : •• • • S. ••. I primarily because large-width baffles for wide stability limits increase the veloc- ity of the unburned gases and reduce efficiency. A further explanation may lie in the fact that low-pressure-drop flameholders operated at high temperature ratios favorable for high efficiency tend to amplify pressure disturbances introduced in the diffuser (ref. 36). The intensified pressure fluctuations no doubt decrease stability. Also, reported in reference 35 is an investigation of an adjustable gutter-grid flameholder having a gutter angle that could be varied from 00 to 530 during operation to give a variation in the blocked frontal area from 14.2 to 59.5 percent of the combustion-chamber area. The combustion efficiency of the adjustable- gutter flameholder as a function of gutter angle is shown in figure 134, where the data show that variations in gutter angle from 25 0 to 500 had little effect on com- bustion efficiency. However, other results of the investigation showed that the stable limits of operation were improved slightly with increased gutter angle.

Gutter-grid flameholders were also used in the free-flight tests reported in reference 23. Three grids of 49, 55, and 60 percent blocked area were constructed of 3/4-inch V-gutters. The maximum combustion efficiencies obtained with the two

grids of smaller blocked area were greater than those obtained with the third flan

holder of 60 percent blocked area. Perhaps flame blow-out at some portions of the grid for the holder with 60 percent blocked area was responsible for these results.

The staggered annular-radial V-gutter combinations previously cited in the discussion of stability limits (refs. 7, 25, and 26) were also employed in a combus- tion-efficiency test program to determine the effect of flameholder geometry on per- formance. Figure 135 (ref. 26) indicates the effects of flameholder blocked area and gutter width upon combustion efficiency. Unlike stability, combustion efficiency is not greatly influenced by gutter width, and the gains afforded by increased area blockage are very slight.

Piloting Systems A pilot is a portion of the combustor in which a reduced air velocity is main- tained by expanding part of the air stream. This low-velocity region provides a stable burning zone from which flame may be propagated to the rest of the combustor.

Theories of flame propagation from a low-velocity region to a high-velocity region are discussed in chapters III, IV, and V. A typical pilot configuration is shown in figure 136. Since the pilot contains a low-velocity region, the design principles are different from those used for the main combustor. In the pilot a large pressure- loss coefficient may be tolerated and heat release per unit volume is small. Fuel- air ratios can be maintained at fixed optimum values and special fuels may be used.

Pilot heat release. - A program was conducted (ref. 37) to determine whether the heat release of a pilot or the production of active particles controls flame propagation from the pilot. Experiments were performed in a 2-inch-diameter burner with a pilot zone supplied with hydrogen and oxygen. The burner itself was run under fixed conditions of pentane flow, air velocity, temperature, and pressure. Specific impulse increased almost linearly as the heat release from the pilot was raised by increasing the flow of stoichiometric hydrogen-oxygen mixture. When more hydrogen was added, with the same oxygen flow, the heat release was kept constant while the production of hydrogen atoms dropped more than a hundredfold; nevertheless, the specific impulse remained nearly constant. Reference 37 takes this to be a negative type of evidence in favor of the importance of pilot heat release, as opposed to the production of actual particles in the pilot flame. However, it is by no means con- clusive evidence, because the effects of the temperature of the pilot exhaust gas were not considered.

•• ••. . S • • S • .. •5 • ••• • ••• ..

• •• S • • • • S • . .. .

. • S • • • S • • .. .

' NACA RM E55G28 CONFITh . • . .

S.

• • • • b .

•• • • S S..

•• Design and use of pilots. - The physical size of a piloting zone is an important factor in the design of combustors. It has been found experimentally that a circular pilot cross section is better than a rectangular one (ref. 38), and that a length equal to a diameter is adequate before pilot recirculation air is admitted (ref. 39).

Reference 40 states that the sum of the diameters of the first row of recirculation air holes should equal 40 percent of the pilot circumference.

A one-dimensional aerodynamic analysis of the required size of a pilot combustor is shown for one set of initial conditions in reference 41. A more extensive treat- ment of optimum pilot size is given in reference 42. In this report, an ideal pilot- ing system is considered in which all the combustion takes place in a low-velocity stoichiometric pilot zone. Secondary air is mixed with the exhaust products down- stream of the pilot combustor to give the desired over-all fuel-air ratio. The study shows that it is possible to maintain a large pilot area for efficient low-velocity combustion without incurring excessive total-pressure losses.

Percent pilot is defined as the percent of total fuel sent to the pilot zone.

This percentage may vary from 0 to 100, the latter value corresponding to the ideal- ized pilot of reference 42. An example of an experimental investigation of a ram-jet combustor operated at varying percent pilot is found in reference 43, where percent pilot ranged from 12 to 100 percent. In some cases, where low-drag flameholders are employed (refs. 9 and 44), piloting of 1 percent or less is sufficient for large gains in stability limits. Figure 137 illustrates the increase in efficiency with small percent pilot for a single V-gutter flazneholder described in reference 45.

I Pilot operation is not required where inlet conditions are very favorable for • combustion and over-all fuel-air ratios near stoichiometric are employed. In the investigation of reference 26, for lean over-all fuel-air ratios, where the fuel was concentrated locally, pilot operation was beneficial, but at rich fuel-air ratios where uniform fuel distribution was required, pilot operation was of little help.

Piloted flameholders. - Integral piloting systems have been combined with such well-known simple-baffle systems as V-gutter grids (ref. 46), radial gutters (ref.

47), annular-radial gutter combinations (ref. 48), and staggered annular-radial gutter combinations (refs. 8, 26, and 49). In general, the effects of flameholder geometry on combustion performance of the piloted flameholders were not different from those of nonpiloted flanieholders. Reference 47, for example, reports no appre- ciable effect of varied area blockage on combustion efficiency. Comparisons between otherwise similar piloted and nonpiloted configurations are given in references 26, 44, and 49. The piloted designs offered no improvement in peak combustion efficiency, but they did tend to widen the fuel-air-ratio range of operation. A comparison is made in reference 50 of three types of piloted configurations intended for use in a ram-jet combustor at low pressures and rich fuel-air ratios. The configurations consisted of a can that acted as a 100-percent pilot and two gutter combinations, one with five can-type pilots (fig. 138) and the other with a sloping-gutter pilot. The third type was the most satisfactory design. This configuration (also described in ref. 51) consisted of a perforated conical flow divider enclosing a sloping V-gutter basket (fig. 139). The sloping V-gutter and flow divider served first to confine the mainstream fuel to a portion of the air and then to promote good mixing of the pilot combustion products with the mainstream combustibles.

Another example of this type of design is given in reference 52.

Three-Dimensional Flameholders Three-dimensional baffle. - In contrast to the planar, simple-baffle flame- holder, the three-dimensional baffle has appreciable axial depth. The advantage of S.

•S• S •. • • • ••• S SS • •S S •• ••. • •*00 • S 55s• S • .

S • • • • .

•. S S •S• t2.

cNENTIAL NACA RN E55G28 S S S• S •S . S • • 5•5 . . •• •• S S S . S.

.• this design is that the volume of the primary zone can be large, and a means for con- trolling the introduction of dilution air is provided. The pilot configurations of references 50 and 51 (fig. 139) are an approach to the three-dimensional type.

Flaineholder designs have been evolved (refs. 43 and 53) in which the radial gutters slope at a comparatively small angle to the combustor axis and provide a conical flame-holding surface. The sloping-baffle configuration investigated in a 16-inch connected-pipe facility described in reference 53 consisted of two sets of U-shaped baffles separated by a conical section (fig. 130(b)). The 6 baffles in the primary zone and the 12 baffles in the secondary zone were inclined at 300 angles to the combustor axis. The fuel-mixing control sleeve, which extended from the fuel injectors to the flameholder, intercepted approximately 20 percent of the total engine air mass flow and ducted this air into the primary combustion zone.

Combustion originated in the wake of the upstream set of baffles and was substantially complete in the sheltered region downstream of these baffles. The use of a sloping baffle and conical shielded zone provided an expanding volume for the combustion region, thereby maintaining a low flow veloity which permitted combustion to be completed in a relatively short length.

The combustion performance of this configuration as a function of fuel-air ratio is compared in figure 140 with that of a baffle-pilot configuration investigated in the same facility and at the same test conditions. The advantage of providing fuel-air mixing control downstream of the point of initiation of combustion is an increase in combustion efficiency at low fuel-air ratios.

Another type of three-dimensional flameholder is the rake type, which consists of a pilot body with petal-like fins extending downstream and radially from the pilot (fig. 141). Usually several of these flameholder combinations are positioned at an axial station in the combustor. Multiple-rake flameholders were used in free- flight investigations reported in references 5 and 54 to 57, where these configura- tions were found to be superior to piloted gutter flameholders in stability limits and efficiency. Another comparison of a rake-type flaxneholder with a piloted ser- rated baffle flaineholder also showed improved results with the former type (ref. 58).

Further gains would have been possible with improved fuel injection systems. An investigation of more complex rake designs (ref. 59) showed that best performance was obtained with a rake-type flameholder which had alternate rakes connected to the pilot burner outlet by V-gutters. The three rakes that were connected in this manner appeared to be more effective as flanieholders than the other three rakes because of the continuous flame path from pilot to baffle.

Can flameholder. - The principle of an expanding combustion zone is inherent in the design of a conical can-type flameholder. Can combustors fall into two general types of designs: simple cans and annular cans. A typical simple can is shown in figure 142(a). The flameholder consists of a continuous or segmented conical surface, expanding in a downstream direction with a pilot assembly usually situated at the upstream end of the cone. An annular can (ref. 60), shown in figure 142(b), consists of two conical cans, the vertex of inner can facing downstream, and the vertex of the, outer can facing upstream. Can flameholders may differ in cone angle, distribution of holes and hole sizes, arrangement of holes, and shape of holes. Cans are usually specified in terms of open area and pitch alinement. Open area is the ratio of the area of the perforations to the cross-sectional area of the combustion chamber; the pitch alinement is the number of rows of perforations that spiral around the conical surface counted along the intersection of an axial plane on the surface.

The effect of cone angle on combustion performance has not been systematically 150.

How- investigated in the literature, the standard half angles being from 5 0 to ever, the effect of pitch alinement on combustion efficiency and stability limits •. •S. • • • •5 •• • . • S S S S ••• • ••• •• • • • • S S • S S. • S • • • • 5 • • • •• S.. .

.

. • . .

NACARME55G28 CO1t1'IDl1pJp. ••• ON: •• •, , , has been investigated (ref. 61). A slight improvement in efficiency was found through the use of a two-pitch-alinement can rather than a zero-pitch-alinement can, but stability limits were unaffected by the alinement.

In an investigation of can combustors intended mainly for piloting applications (ref. 38), stability limits were unaffected by hole size, total open area, or pitch alinement. Reference 41, on the other hand, reports that for a single-rowed can flameholder, increasing hole size increases the rich limits but does not affect the lean limits.

CA The effect of open area and hole type on combustor performance is described in reference for a 10-inch-diameter quarter-segment combustor in a free-jet instal- lation. Figure 143 compares the combustion efficiency of four can configurations, three having open areas of 100, 145, and 177 percent with round holes the same size in each case, and one having an open area of 100 percent with transverse rectangular slotted holes. The flameholder with 100 percent open area yielded a maximum effi- ciency of 91 percent, a value slightly higher than the maximum combustion efficiency obtained with the cans of greater open area. The transverse-slotted can exhibited a maximum efficiency close to that of the corresponding round-hole can, but efficien- cies at low fuel-air ratios were much less for the slotted can, probably because the increased frontal area of the slots admitted too much air to the primary portion of the can.

Immersed-Surface Plameholders The use of surfaces in the flame zone of a combustor stems from the observed catalytic effects of certain materials in the ignition of quiescent combustibles.

The employment of similar materials in a high-velocity combustor was a logical extension of this principle. In addition to the thermal effect, a beneficial in- crease in mixing rate was envisioned. Some of the effects of immersed surfaces on combustion efficiency and stability are described in the following paragraphs.

A systematic program of investigations of immersed surface combustors was con- ducted and is reported in references 10, 44, 63, and 64.

In reference 63, the flameholder employed was a wedge-shaped block of graphite that had been spray-coated with aluminum. Two such wedges placed parallel across the cross section of the combustor represented a conventional type flanieholder. Additional rows of similar wedges were introduced downstream of the original row to evaluate the effect of sur- faces immersed in the flame zone. The flameholder configurations used in this program are shown schematically in figure 144.

The effect of the immersed sur- faces on pressure loss and combustion efficiency was slight, but the additional sur- faces did widen the combustion stability limits.

In reference 64, Inconel surfaces were employed because carbon blocks have inadequate shock resistance. The results indicated that, within the requirements of providing a low-velocity path for flame propagation, the number of surfaces and their geometry were unimportant. A conventional V-gutter configuration is compared in reference 10 with another in which 12 Inconel blades were positioned in the flame zone. Variation of the immersed-surface temperatures confirmed that the performance gains were due to aerodynamic rather than thermal influences. Both stability limits and combustion efficiency were improved by the immersed surfaces. The combustion- efficiency effect is illustrated in figure 145, where efficiency is plotted against equivalence ratio. The difference between efficiency of cooled and uncoo led blades is small in comparison with the difference between efficiencies of the config- urations with and without immersed surfaces.

•.

••.

• I S. .

•. • •II

• • ••' • • • .. S.. II . S I • ••

• •

S • S S I . I . .

.

.

I I .. C0DTTIAL NACA EM E55G28 :0 . •• " s 4 . ' " • • : •• I.

• • • .. •1.

A systematic investigation of simplified immersed-surface configurations in reference 44 concluded that the greatest gains in combustion efficiency could be effected by the use of a single blade in the flame zone close to the flameholder.

Optimum efficiency results were obtained with a blade across the combustor perpendic- ular to and 4 inches downstream of the single V-gutter flanieholder (figs. 146 and 147). Apparently the flame-immersed blade is best positioned close to the flame- holder, but not close enough to disturb the normal wake region behind the flan holder (less than 4 in. in this case). At a slight compromise in efficiency and pressure drop, a considerable improvement in stability limits was obtained by combining this single perpendicular blade with three parallel blades farther downstream. A further widening of stability limits was posible by removal of all of the flame-immersed surfaces.

In studies described in reference 65, the effect on combustion of a grill of heated, coated molybdenum strips submerged in the combustion zone of a 6-inch- diameter combustor was determined. The anticipated acceleration of combustion was not realized because the strips did not act as surface combustion aids, but instead conducted heat from the flame zone to the combustor walls.

The advisability of employing immersed surfaces in the combustion zone appears controversial, even though the aerodynamic influence of these surfaces was shown to be important (ref. 10). Mixing can possibly be controlled in a simpler manner, for example, by proper spacing and sizing of holes in a can-type flameholder. However, these tests very dramatically emphasized the fact that the rate of mixing of unburned and burned gases can be the controlling step in the combustion process.

IGNITION The basic principles of ignition discussed in chapter III apply to ram-jet com- bustors. The techniques generally used in starting ram-jet combustors incorporate either pyrotechnic flares or spark plugs located within the pilot zone. The spark system is more desirable for ground tests since repeated starts can be obtained.

However, under certain operating conditions, chronic failures of the spark system have been experienced, and ignition with a hypergolic fuel such as aluminum boro- hydride is advantageous. Flight-test data reported in reference 4 show that up to pressure altitudes of 14,000 feet spark systems are satisfactory, but above that altitude flare systems are necessary. However, in recent tests with a 48-inch ram jet (ref. 66), reliable spark ignition was obtained at 1/8 atmospheric pressure. A further aid to spark ignition at low pressure is the addition of small amounts of hydrogen in the region of the spark (ref. 12). A magnesium-flare ignition system that has proved to be very satisfactory is described in reference 6. Additional ease in starting flight models is afforded by the use of magnesium flares in con- junction with rake-type flameholders (refs. 6 and 67) FUELS AND FUEL SYSTEMS Combustion proceeds most favorably in a near stoichiometric fuel-air mixture, where flame speed and temperature are at a maximum. Thus, the principles underlying the design of a fuel system are the achievement of a near stoichiometric mixture and complete vaporization of fuel in the precombustion zone.

Fuel Injection Fuel-injection types. - In ram-jet combustors, the energy of the high-velocity airstream is used to atomize the fuel, thus allowing simple fuel-injector designs S.

••• • • S • • . • •• • ••• .

. S • • • • • •• S.

S • S .5 • • • • S S • . ' • NACA RM E55G28 C0DEAZ,•: S • • • • • • .• , .. .. S S •• •• • • • • • and low fuel pressures to be used. Air-blast atomization is discussed in chapter I, where it is shown that drop size can be predicted from certain fuel properties and • the relative air and fuel velocities and volumes by the following equation: \O.45 / 1.5 585.f 597 (\l000 -) d32 Ur + (9) () where Sauter mean diameter, microns fuel-air volume ratio relative air-fuel velocity, meters/sec Ur viscosity, poises 0 surface tension, dynes/cm density P Subscripts: a air fuel f For a typical ram-jet combustor-inlet velocity of 300 feet per second and with JP-4 fuel, the drop-size calculation can be approximated by d 32 = , since Ur ', Pf is small. The value of d32 at this condition is about 35 microns. This drop-size diameter is roughly one-third that predicted for a 17-.5-gallon-per-hour Monarch spray nozzle in still air by the following equation from chapter I: d32 = 251LP 07 microns (10) where LP pressure drop across nozzle For a pressure differential of 100 pounds per square inch, d32 is 115 microns.

These drop-size estimates show that the air flow past the fuel injector provides remarkably good atomization and that further mechanical improvements are unnecessary.

Thus it is not surprising that injector types of a wide variety have proved satis- fact6ry. At low fuel-air ratios 'it is reported (ref. 41) that simple fixed-orifice nozzles give the highest efficiencies because less dilution of the fuel-air mixture occurs. However, fewer injection points are required with the Monarch nozzle because the radial component of the fuel-droplet velocity distributes the fuel over a greater area than the simple-orifice type. At higher fuel-air ratios, air-atomizing types of nozzles provide a more homogeneous mixture and are more satisfactory. Variable- area spring-loaded nozzles appear promising for ram-jet applications, but difficulty has been encountered in providing equal flow through each nozzle in multinozzle installations.

.•

••.

• • S S.

S. S•. • ••• • •. . . .

••

S'..

:96NFAEN-2411,

NACA EN E55G28

• •• ..• • • •5 S. 555 •

A program was conducted (ref. 68) to determine the effect of nozzle size in a

six fixed-area nozzles evenly spaced at an axial station. The

configuration of

results showed that there is an optimum size of nozzle for maximum efficiency. With

overly large nozzles, distribution is poor; similarly, with undersized noz1es, fuel

pressure is so high that fuel particles strike and flow along the 'chamber/ wall and

do not mix with air properly, decreasing efficiency.

Other methods of fuel injection include impinging jets and spray bars. Excel-

lent efficiencies are reported (refs. 69 and 70) for fuel-injection systems consist-

ing of multiple impinging jets of compressed air and fuel. Spray bars, usually per-

forated tubes, are also widely used since they have the advantage of low area

blockage.

to

Location and direction of fuel sprays. - The location and direction of fuel

sprays are determined by the combustor-inlet conditions and the type of fuel injector.

Since it is desired that a vaporized, locally stoichiometric mixture be produced at

the flameholder, the direction of fuel spray is of some importance, particularly for

fuel injectors just upstream of the flameholder. Injection of fuel in a contrastream

direction allows a greater path of travel of the fuel droplets and produces a more

homogeneous mixture than injection in the costream direction. Thus reference 51

reports excellent combustion efficiencies at rich fuel-air ratios by contrastream

injection, but recommends costream or cross-stream injection, which produce a more

concentrated fuel mixture, at lean fuel-air ratios.

The significance of fuel-injector location is best illustrated by some examples

from reference 71. Combinations of upstream fuel injection, flameholder injection,

and split injection between the two positions were employed, as shown in figure

148. The combustion efficiencies of the four injector configurations are shown

in figure 149 as a function of fuel-air ratio. The broadest range of operation

was obtained with flanieholder injection, where operation was possible from fuel-air

ratios of 0.012 to 0.047. The upstream-injection case gave only a narrow range of

operation near stoichiometric fuel-air ratio and peak efficiencies of less than 80

percent. This range was broadened by the use of split injection, and the peak eff i-

ciency increased slightly by the use of flameholder split injection. Thus the results

indicate that upstream or split injection, which gives a nearly homogeneous mixture,

produces the best performance at rich fuel-air ratios, and localized laineho1der

injection produces the best performance at lean fuel-air ratios.

These findings have been widely confirmed. Results from reference 72 plotted

in figure 150(a) show that for a can-type combustor peak efficiencies at low

fuel-air ratios are obtained with internal (flan1eholder) injection; peak efficiencies

at high fuel-air ratios are obtained with upstream injection. For operation over a

wide range of fuel-air ratios, a combination of internal and upstream operation was

most satisfactory.

The principle of fuel-mixture stratification to obtain efficient combustion at

lean fuel-air ratios also applies to the radial positioning of the fuel injector.

Figure 150(b) (ref. 72) shows that for upstream fuel injection, efficiency is

better at low fuel-air ratios for injector positions near the centerbbdy but the

reverse is true at high fuel-air ratios. Figure 151(a) shows the effect of

injector-ring diameter on combustion efficiency (ref. 73). The fuel-air ratio for

peak efficiency increases with increasing fuel-ring diameter as seen in the cross

plot (fig. 151(b)). The solid line in figure 151(a) illustrates the fact that

injector rings of two different diameters may be combined to broaden S the range of

operation.

.. . ..

. .

..

.. S •S• S ••( • • • S • S. • • • S S S • • S S • • • S S• • • •• • .. S NACA RN E55G28 C(NIITIL . ..• . • • • • • ,27 SS ••• •• ••. S 555 •• . • SS •• Staged fuel injection. - For operation over a wide range of fuel-air ratio, the principle of fuel-injection staging is used. A set of injectors, denoted as primary fuel injectors, supply fuel for low-fuel-air-ratio operation. For operation at richer fuel-air ratios, additional fuel is injected through secondary sets of in- jectors, which are positioned for the most efficient combustion.

The performance of a combustor operated with careful proportioning of primary and secondary fuel flows is shown in figure 152 (ref. 72). High efficiencies were achieved by operating with primary fuel up to fuel-air ratios of 0.034 and then by maintaining the primary fuel-air ratio constant at 0.014 and increasing the sec- ondary fuel flow. A mixture control sleeve of the type described in the following section separated the primary and secondary fuel-air mixtures.

Such combinations of primary and secondary fuel injection are very common in all types of ram-jet combustors. A three-stage fuel-injection system for a can flame- holder is described in reference 62, one stage located at the can entrance and the other two downstream of the first stage. Best results were obtained when the first stage was used for very lean fuel-air ratios, the first and second stages for ratios upto 0.045, and all three stages for ratios above 0.045. It is important to reduce the primary fuel flow to a low value at rich fuel-air ratios, as is illustrated in figure 152 and further shown in a plot from reference 73 (fig. 153), where the rich limits of combustion narrow as primary fuel flow is increased from 18 to 34 percent of the total fuel flow.

Mixture control sleeves. - Good combustion efficiency at lean fuel-air ratios can be obtained with upstream injection as well as with flameholder injection if excessive mixture dilution upstream of the flameholder is prevented by a control sleeve. The combustor configurations shown in figures 130(b) and (c) and 152 illustrate the placement of such a control sleeve. Although primary and secondary fuel injection are from the same axial station upstream of the flameholder, the primary fuel-air mixture within the control sleeve is channeled directly to the upstream portion of flameholder where it maintains a rich concentration. The control sleeve prdduces the same effect as does the near-stoichiometric pilot zone of ref- erence 42, described in the section on piloting.

An early type of mixture control sleeve is described in reference 74, but the device had not been fully exploited until recently (refs. 1, 9, 34, 43, 53, 75, and 76). An increase in combustion efficiency from 30 to 75 percent at a fuel-air ratio of 0.025 with no increase in combustor pressure loss is reported in reference 75.

The advantages of a sleeve system are reviewed in reference 34. Besides the improve- ment of combustion efficiency at low fuel-air ratios provided by rich local mixtures,.

the control sleeve produces more consistent fuel distribution and sharp-edged fuel- air profiles and allows the use of simple fuel orifices rather than atomizing nozzles.

Effect of Fuel Variables Fuel type. - The effect of fuel type on combustion performance in a rani jet would depend on the controlling step in the combustion process. The combustion process may be considered as successive steps of fuel vaporization, mixing, and chemical oxidation. If the vaporization step were controlling, then the physical properties of the fuel would be of great importance; if oxidation were controlling, the chemical properties would determine the performance of the fuel. Small-scale tests of efficiencies of several fuels bear out these hypotheses. (Fundamental com- bustion properties are discussed in chapters I to VI.) Investigations of gaseous fuel-air mixtures (ref. 45) where fuel volatility is eliminated as a factor show that fuels of low ignition energy, or short ignition lags, such as hydrogen, acetylene, carbon disulfide, or propylene oxide, produce the best efficiencies.

SS• •• S S S. • .. . ..• . ..

..

S • 0S• . S S • • • • • .00NFIDTd NACA EM E55G28 2 8.

. . P .

• • • • S • ••I •.

S S ••• •• ••S IS •• • S S • •• S. •SS S Although large-scale engine tests cannot be as easily analyzed, the same con- clusion may be reached regarding the effect of fuel properties. In reference 77, the use of gasoline rather than JP-3 fuels increased combustion efficiency 10 per- cent at an inlet-air temperature of 920 0 R. These results indicate that a 53-inch length for a 20-inch-diameter combustor was insufficient for complete vaporization of the less-volatile fuels. Similarly, in other investigations where low inlet tem- peratures and, short combustor lengths caused the vaporization of liquid fuels to govern the over-all combustipn efficiency, gasoline was superior to the less-volatile kerosene (refs. 32 and 45). Other studies have found n-heptane to be superior to the less-volatile Diesel oil (refs. 26 and 78) and propylene oxide superior to ker- osene (refs- . 3, 32, 58, 59, 79, and 80), but this effect with the latter pair of fuels may be attributed to flame speed as well as to increased volatility.

Where nearly complete vaporization is ensured, fuels that differ from one another mainly in volatility give nearly the same performance in ram-jet combustors.

Reference 81 reports little difference in combustion efficiencies between 80-octane gasoline, JP-3 fuel, and special low-vapor-pressure fuels. Similarly, almost iden- tical results with gasoline and JP-4 fuel were found in the work of reference 82, 10600 R con- and samples taken in this investigation at an inlet-air temperature of firmed that vaporization was substantially complete in 17 inches of mixing length.

In ram-jet combustors where vaporization does not control the over-all effi- ciency, appreciable gains in efficiency result from the use of high-flame-speed fuels. For premixed and prevaporized fuel mixtures, reference 83 states that com- bustion efficiency increases with the 1.1 power of flame speed.

Octane rating of gasoline-type fuels is of little consequence in ram-jet com- bustors. Reference 84 shows that 62-octane gasoline performed as well as 100-octane gasolin•e.

Fuel preheating. - For combustor operating conditions where fuel vaporization is an important factor, increased preheating of fuel aids combustion efficiency.

Reference 85 reports a 10-percent increase in combustion efficiency with preheating 400 to 2000 F. In combustors where vaporization of the of 62-octane gasoline from fuel is complete for cold injection, such as the combustor discussed in reference 82, preheating the fuel is of no consequence.

OPERATING VARIABL Pressure Effect on combustion efficiency. - Because pressure affects all of the funda- mental processes in the combustor, finding occasional contradictory results from pressure investigations is not too surprising. Evaporation of liquid fuels is retarded by increased combustor pressures (ch. I), whereas this pressure increase aids the oxidation reaction (ch. III). Thus, in general, at low and moderate pres- sures where fuel vaporization is rapid, efficiency increases with pressure. Con- versely, where vaporization is the controlling step, as at high pressures or with low-volatility fuels, efficiency may remain independent of pressure or even decrease slightly with pressure.

Results fora 2-inch-diameter gaseous-propane - air burner (fig. 154) show a continuous increase in combustion efficiency with pressure increases from 5 to 85 inches of mercury absolute (ref. 45). Similar plots of combustion efficiency as a function of inlet pressure have been established from larger-scale tests of com- bustors with simple-baffle flameholders (ref. 86), rake-type flameholders (ref. 55) .. S • • .. S. • S.. • IS. • • • S .. S •• S I • • • • . . S • S • I • • • • . .. . .. I cN'INTjAL 59 NACA EM E55G28 S •SI I • S • I • • as great as from 8 to 85 inches ranges and can flameholders (ref. 87) over pressure of mercury absolute. The more usual findings ) however, are that efficiency increases with pressure up to a certain value and then remains independent of pressure. Figure 155 illustrates this trend from results given in reference 77, where at a fuel- air ratio of 0.06, combustion efficiency increases with pressures from 7 to 55 inches of mercury absolute and remains constant at higher pressures.

In combustors where fuel vaporization is more critical, the pressure effects shown in figures 154 and 155 may not apply. In reference 60, for example, data are presented for can combustors where efficiency increases with pressure only up to a pressure of 20 inches of mercury absolute, above which efficiency decreases slightly. References 3 and 71 state that efficiency was virtually independent of pressure at pressures as low as 17 to 28 inches of mercury absolute. That these results are due to an increased importance of fuel vaporization is confirmed by the work of reference 88, where over a pressure range of 30 to 65 inches of mercury abso- lute, combustion efficiency decreased with pressure when radial simple-orifice fuel injectors were used and increased with pressure when hollow-cone spray nozzles were used. The spray nozzles had better vaporization characteristics than the fixed- orifice injectors.

Effect on stability limits. - Increaed inlet pressure usually tends to widen both the lean and rich limits of combustion, especially at very low pressures (refs.

55, 56, 73, 77, and 86). As in the case of combustion efficiency, where fuel vapor- ization is a critical factor, pressure may have an adverse effect upon limits. In figure 156, the effect of pressure on stability limits is shown for three inlet temperatures (ref. 73). At the two higher temperatures, the expected widening of 0 R, where vaporization may the limits with increased pressures is seen; but at 810 control, the limits are independent of pressure above 18 inches of mercury absolute.

A fundamental treatment of the effect of pressure on flame stabilization is given in chapter VI.

Temperature Effect on combustion efficiency. - The two important combustion steps of vapor- ization and oxidation are both accelerated by increased temperatures (chs. I and III); hence, combustion efficiency would be improved appreciably by increased inlet tem- peratures. This effect of temperature has been confirmed by many combustion studies, although the results of some of these reports are of doubtful significance because of lack of control of other variables, principally pressure.. Investigations where inlet temperature was the only variable have shown that combustion efficiency in- creases almost linearly with increasing inlet temperature (e.g., refs. 34, 59, 62, 77, 78, and 88). Since the temperature contribution to combustion efficiency is not the result of two competing effects, efficiencies increase asymptotically with temperature to limiting values of 100 percent. Studies of ran-jet combustors are pesented in which increased inlet temperature improved combustion efficiency greatly wider conditions of low inlet pressure (ref. 60) and off-stoichiometric fuel-air ratios (ref. 76). However, at atmospheric pressure and near_stoichionletric fuel-air ratios, where efficiencies were normally near 100 percent, increasing inlet tempera- tures could not improve combustion efficiency.

Effect on stability limits. - The stability curves for three inlet temperatures shown in figure 156 indicate that increased inlet temperatures widen stability temperature on of limits, especially the rich limits. These findings of the effect stability limits are in agreement with those of other investigations (refs. 27, 78, 81, 89, and 90). A discussion of the effect of temperature on stability is also given in chapter VI.

•• •S • • S S. • .. ... . .•• . ..

• S S • S S S • .

• • .

• • _. S S. S S • S COFINIfL 26i.. . .. . NACA EM : . ...

• . .

• • S • •• S S 900 •S •SS S S •S •• .. S.. S Velocity Effect on combustion efficiency. - Increased inlet velocity improves fuel atom- ization (çh. I); but, for a given chamber length, increased velocities allow shorter reaction times for the evaporation and oxidation steps in the combustor. Thus, as with inlet pressure, competing contributions to combustion efficiency are influenced by inlet velocity. Experimental information has been presented wherein combustion efficiencies may increase up to maximum and then decrease with increasing velocity (ref. 61), decrease with increasing velocity (refs. 78 and 88), or be unaffected by velocity changes (ref. 3). These contradictory findings are apparently due to dif- ferent relative importances of the atomization, vaporization, and oxidation steps in the combustor. This has been borne out by the work of reference 71, where combustion efficiency reached a maximum with velocity for upstream or split fuel injection runs and decreased with increasing velocity for flameholder fuel injection. The most usual findings, however, are that combustion efficiency decreases with increasing air velocity.

Effect on stability limits. - Increased velocity has been found to both widen (ref. 74) and narrow (refs. 7-and 78) lean limits. With a homogeneous fuel-air mix- ture, increased velocity would normally result in poorer lean limits of combustion.

However, under certain conditions of stratified fuel-air-mixture distribution, the increased mixing rates associated with increased velocity would tend to improve the local combustible concentration. The only available literature on rich limits (ref.

73) shows that the limits are decreased by increased inlet velocities.

Angle of Attack Angle-of-attack operation distorts the velocity profile at the combustor, and therefore would be expected to be detrimental to combustion efficiency. Investiga- tions of 16-inch can-type combustors at several angles of attack (refs. 91 and 92) showed that efficiency decreased about 10 percent with an increase in angle of attack 100, but stability limits were unaffected.

from 00 to Correlations of Operating Variables It is obvious from the experimental literature that the same change in an inlet variable may affect combustor performance in opposite directions, depending on which step in the over-all process predominates. Semitheoretical correlations must of necessity assume a controlling step in the combustor. Most combüstor investigations are conducted under conditions to which an oxidation-controlled mechanism would be most closely analogous. For this particular case, the usual correlation of combus- tion efficiency takes the form (pTl = (II) Tib V where combustion length as proposed in reference 14. A function of this type was successful (ref. 93) in correlating the performance of turbojet combustors under conditions of constant combustor length and limited fuel-air-ratio ranges. With these limitations, the simplified parameter PT/V will suffice for correlation.

.. •.. . .

S •• •• • • S. • S..

•• NACA RM E55G28 FIEWPAL : : .

: : . : :. : :. : : 261 • . S • • ••• S S • S • • I I SI •I• SI •S• S • •• •S • • . ..• SS A correlation similar to the preceding was derived for application to rain-jet combustors (ref. 94). The theoretical basis for the correlation was the assumption • that the mixing of burned and unburned gases and their movement through the flame front controlled the over-all combustion rate. Thus, combustion efficiency was defined as the ratio of the mass flow of gases through the flame front to the over- all mass flow of unburned gases through the combustor Tib - puAffu (12) where fundamental flame speed u Subscripts: C combustor cross section ff flame front u unburned gases Flame-front area and flame speed were evaluated in terms of inlet variables to give a function of inlet pressure, temperature, and velocity. Empirical data from runs with a 5-inch ram-jet combustor with V-gutter flameholders could be correlated by the expression = (o.3) (13) This expression could be applied to data of many different fuels by multiplying the * correlation parameter by the term ( uf/ uref) 1 , where uf is the flame speed of the fuel employed and u.ef is the flame speed of a reference fuel (ref. 83). The ref- erence fuel used in references .83 and 94 was gasoline with a flame speed of 1.4 feet per second. The correlation is shown in figure 157, where combustion efficiency is plotted against the correlating parameter. Reasonable agreement between the results with different fuels is shown with the exception of carbon disulfide, a compound with a very low ignition energy. A linear relation between combustion effi- ciency and the correlation parameter is exhibited up to efficiencies of 80 percent, above which the effect of the parameter on efficiency is slight. At high values of the inlet variable parameter, therefore, combustion efficiency would be high and would be unaffected by moderate changes in inlet variables. This conclusion has also been shown in tests in a 16-inch combustor reported in reference 76.

SUMMARY OF RAM-JET COMBUSTION PRINCIPLES Ram-jet combustion work has been conducted mainly along experimental lines.

Because the field is comparatively new, considerable effort was expended in explora- tory programs, and only a few of the many variables affecting ram-jet combustor per- formance have been investigated systematically. Some of the ram-jet design principles which have been evolved are briefly reviewed in the following discussion.

•• • S •S• •• S .. ... . ..• S •S S S • • S S S S S S S S • . • S • NACA RM E55G28 ••• CNFI1NIL : • • • • • • • •SS •• •• _.. S • •SS •S • • S. •S. • Flameholder Geometry The single-plane, simple-baffle flameholder can be applied satisfactorily at favorable inlet-air conditions and at rich fuel-air ratios. Its simplicity, low weight and drag, and structural reliability are very desirable. However, at less- favorable inlet conditions (particularly at inlet pressures below one-half atmosphere and at lean fuel-air ratios), the three-dimensional flameholder, as exemplified by the can type, has more desirable performance characteristics. The most satisfactory simple-baffle shape was a V-gutter arranged to form an annular-, radial-, or grid- type flameholder. Combustion performance is insensitive, within certain limits, to I) variations in baffle width, spacing, angle, and area blockage. Baffle widths of 1 -i 300 to 600 have been most generally used.

t) to 2 inches and angles from The three-dimensional flanieholder type most widely investigated is the conical can. It is believed that the superior performance of this type at lean fuel-air ratios and at low pressures is due to fuel-air mixing control downstream of the point of initiation of combustion, which is absent in the simple-baffle flameholder.

Within the range investigated for the can flameholder, geometric variables of hole number, size, shape, pitch, and open area do not influence combustion efficiency appreciably, and most of the geometric investigations were directed toward reduction of flameholder pressure loss. The flmneholders investigated had open areas ranging from 75 to 175 percent.

The sloping-baffle flameholder, a combination of the simple-baffle and the can- type flameholders, appears to have the desirable performance characteristics of the can type. However, the many variables involved in this design have not as yet been investigated.

In reference 95 a comparison is made of five configurations tested in the sane 20-inch ram-jet engine. From. these tests of can, sloping-baffle, annular-piloted, and piloted V-gutter flanieholder designs, it was found that all types gave combus- tion efficiencies of 80 to 90 percent. A comparison on the basis of specific ml- pulse, where the effect of friction loss is included, showed the same high level of performance for all designs.

Piloting The use of a large-size pilot in conjunction with either the simple-baffle or the can flameholder is a desirable design principle for improved combustion performance, particularly at lean-fuel-air-ratio conditions. It has been shown analytically and experimentally that a shielded, low-velocity pilot zone can occupy a relatively large percentage of the combustor area without causing or producing excessive pressure losses (ref. 42).

Both high combustion efficiency and wide stability limits are desired in pilots; but it has been found experimentally that a pilot which has wide stability limits usually has poor combustion efficiency. The importance of pilot burning efficiency on over-all combustion performance has not been definitely established, except to the extent 'that, where the pilot heat release represents a large portion of the total heat release, the pilot combustion efficiency should be high. A circular pilot cross section gave better' efficiency than a rectangular cross section, and a pilot length equal to its diameter was adequate. Pilot performance was insensitive to hole size, number, and spacing within the limits investigated. For large-size combustors, an annular-shaped pilot is believed to have an advantage over the centerbody pilot, since for a given pilot cross-sectional area, the radial distance from pilot to outer wall can be reduced.

.. •.e . • ' .. S. S •• • • S• •• • . . S S S S S • : NACA RM E55G28 DETA • • .

. I •S I •• S S • • .

• S S.. S • I S • S S I •• S..

•• ••• • . I.

•• • I S •II IS Fuel Injection Location of injectors. - The location of the fuel injector with respect to the flameholder is one of the most significant variables influencing ram-jet combustor performance. In general, for efficient combustion at lean fuel-air ratios it is desirable to stratify the fuel-air mixture, thereby producing a locally stoichio- metric region. For near-stoichiometric operation, a homogeneous mixture gives best results. Information on mixture distribution downstream of fuel injectors is given in chapter I. A stratified mixture can be obtained with some degree of effectiveness by radial and axial location of the fuel injector so that the fuel mixes with only a portion of the combustion air. A more effective technique for obtaining stratifica- H tion is mechanical control of fuel-air mixing by means of a mixing control sleeve (refs. 1 and 75).

For efficient combustion at lean as well as rich operation, the use of dual fuel-injection systems is necessary. For best combustion efficiency, the primary injector supplies the fuel at lean fuel-air ratios, and both primary and secondary injectors are used for rich mixtures. At increasingly rich operation, a smaller fraction of the fuel is supplied by the primary and the remainder by the secondary injector.

Locating the fuel injector upstream of the flameholder is a generally accepted practice. However, care must be exercised to prevent flame from seating upstream of the flameholder, especially with the can-type flameholder. It is possible to inject the fuel within the flameholder, and combustion efficiency, though not impaired at lean fuel-air ratios by this method of injection, is poor at rich conditions.

Types of injectors. - Simple-orifice, solid, and hollow-cone sprays have been used with equal success. The spray bar with multiple orifices provides a large num- ber of injection points with little area blockage; this is desirable for rich opera- tibn. However, spray nozzles do not plug as easily, and changes in nozzle tip size can be readily made. Variable-area fuel nozzles also seem suited for ram-jet application.

Fuel Type The majority of the investigators have concluded that for inlet-air tempera- tures corresponding to flight Mach numbers of 2.0 or greater, combustion is satis- factory with either gasoline or JP fuels. High-volatility and high-flame-speed fuels such as propylene oxide show some gains at severe operating conditions such as at low inlet-air temperatures or with short combustor lengths.

Inlet-Air Conditions The combustion efficiency of ram-jet combustors follows well known trends with inlet-air conditions. Efficiency under usual conditions increases with pressure and temperature and decreases with velocity. These effects have been correlated for an idealized combustor in which the vaporization and mixing steps have been eliminated as possible rate-controlling mechanisms (ref. 94). For these reasons, and because the effects of other variables such as combustor length and operating fuel-air ratio have not been definitely established, these correlations can be applied specifically only in certain cases.

* •S• 00: •• S S 0 0 S.• ••• • •.. • S.

S S S • • S S S S : 26 4 . .. • •• • • • • • coFIDNTL.: NACA EM E55G28 S III S I I S I I I •IS SI SI S • 151 SI • I •• •5 ••. S 1. Cervenka, A. J., and Dangle, E. E.: Effect of Fuel-Air Distribution on Perform- and of a 16-Inch Ram-Jet Engine. NACA EM E52D08 1 1952.

a 2. Wentworth, Carl B., Hurrell, Herbert G., and Nakanishi, Shigeo: Evaluation of Operating Characteristics of a Supersonic Free-Jet Facility for Full-Scale Rain- Jet Investigations. NACA EM E52108, 1952.

3. Sterbentz, W. H., and Nussd.orfer, T. J.: Investigation of Performance of Bumblebee 18-Inch Ram Jet with a Can-Type Flame Holder. NACA EM E8E21, 1948.

4. Disher, John H.: Flight Investigation of a 20-Inch-Diameter Steady-Flow Ram Jet.

NACA EM E7105a, 1948.

5. Carlton, William W., and Messing, Wesley E.: Free-Flight Performance of 16-Inch- Diameter Supersonic Rain-Jet Units. I - Four Units Designed. for Combustion- Chamber-Inlet Mach Number of 0.12 at Free-Stream Mach Number of 1.6 (Units A-2, A-3, A-4, and A-5). NACA EM E9F22, 1949.

Disher, John H., Kohl, Robert C., and Jones, Merle L.: Free-Flight Performance 6.

of a Rocket-Boosted, Air-Launched 16-Inch-Diameter Ram-Jet Engine at Mach Numbers up to 2.20. NACA EM E521,02, 1953.

Jones, W. L., Shillito, T. B., and Henzel, J. G., Jr.: Altitude-Test-Chamber 7.

Investigation of Performance of a 28-Inch Ram-Jet Engine. I - Combustion and Operational Performance of Four Combustion-Chamber Configurations. NACA RN E50F16, 1950.

S. Smolak, George R., and Wentworth, Carl B.: Altitude Performance of a 20-Inch- Diameter Ram-Jet Engine' Investigated in a Free-Jet Facility at Mach Number 3.0.

NACA EM E52K24, 1953.

9. Sniolak, George R., and Wentworth, Carl B.: Altitude Investigation of Can-Type Flame Holder in 20-Inch-Diameter Ram-Jet Combustor. NACA EM E54D08, 1954.

10. Male, Donald W.: Use of Flame-Immersed Blades to Improve Combustion Limits and Efficiency of a 5-Inch Diameter Connected-Pipe, Ram-Jet Combustor. NACA EM E53B16, 1953.

11. Longwell, J. P.: Flame Stabilization by Bluff Bodies and Turbulent Flames in Ducts. Fourth Symposium (International) on Combustion, The Williams & Wilkins Co. (Baltimore), 1953, pp. 90-97.

12. Longwell, J. P., and Petrein, R. J.: Ramjet Technology. Ch. 10 - Design of Baffle-Type Combustors. APL/JHU TG 154-10, Esso Labs., Standard Oil Dev. Co.

(Publ. by The Johns Hopkins Univ.)

13. DeZubay, E. A.: Characteristics of Disk-Controlled Flame. Aero. Digest, vol.

61, no. 1, July 1950, pp. 54-56; 102-104.

14. Friedman, J., Bennet, W. J., and Zwick, E. B.: The Engineering Application of Combustion Research to Ram-Jet Engines. Fourth Symposium (International) on Combustion, The Williams & Wilkins Co. (Baltimore), 1953, pp. 756-764.

15. Williams, G. C.: Basic Studies on Flame Stabilization. Jour. Aero. Sci., vol.

16, no. 12, Dec. 1949, pp. 714-722.

S. •SS • S S •• .5 • ••. • •5• •• • . S S • S • S • • • .

S • S • • . .. 5 .

S S S S • • s S •• I NACA RM E55G28 . OIFF1DENIAL. ... . • • . . .

. 25 .. S..

•s S.. S S •S •• • • . ... ..

16. Williams, G. C., Hottel, H. C., and Scurlock, A. C.: Flame Stabilization and • Propagation in High Velocity Gas Streams. Third Symposium on Combustion and Flame and Explosion Phenomena, The Williams & Wilkins Co. (Baltimore), 1949, pp. 21-40.

I 17. Longwell, J. P., Chenevey, J. E., Clark, W. W., and Frost, E. E.: Flame Stabili - zation by Baffles in a High Velocity Gas Stream. Third Symposium on Combustion and Flame Explosion Phenomena, The Williams & Wilkins Co. (Baltimore), 1949, pp. 40-44.

18. Caldwell, Frank R., Ruegg, Filimer W.; Olsen, Lief 0., and Broida, Herbert P.: Sixty-fifth Report on Progress of the Combustion Chamber Research Program Jan.-Mar. 1950. U.S. Dept. Commerce, Nat. Bar. Standards, Apr. 15, 1950.

19. Caldwell, Frank B., Ruegg, Fillmer W., Olsen, Lief 0., and Broid.a, Herbert P.: Sixty-sixth Report on Progress of the Combustion Chamber Research Program Apr.-June 1950. U.S. Dept. Commerce, Nat. Bar. Standards, July 15, 1950.

20. Caldwell, Frank B., Ruegg, Filimer W., Olsen, Lief 0., and Broida, Herbert P.: Sixty-seventh Report on Progress of the Combustion Chamber Research Program July-Sept. 1950. U.S. Dept. Commerce, Nat. Bar. Standards, Oct. 15, 1950.

21. Williams, Glenn C., and Shipman, C. W.: Some Properties of Rod-Stabilized Flames of Homogeneous Gas Mixtures. Fourth Symposium (International) on Combustion, The Williams & Wilkins Co. (Baltimore), 1953, pp. 733-742.

22. Annunziata, F. J., and Edwards, A. E.: Low Altitude Performance of the UAC 5-Inch Diameter Ramjet Engine at JfllJ/APL-Jan. 1949. Meteor Rep. UAC-40, Res.

Dept., United Aircraft Corp., Sept. 1949. (U.S. Navy, Bar. Ord. Contract NOrd 9845 with M.I.T.)

23. Black, Dugald 0., and Messing, Wesley E.: Effect of Three Flame-Holder Config- urations on Subsonic Flight Performance of a Rectangular Ram Jet over Range of Altitudes. NACA RM E8I01, 1948.

24. Douglass, Wm. M.: Tests of a Direct Connect Segment of a Rectangular Wing Ramjet. USCAL Rep. 3-5, Aero. Lab., Univ. Southern Calif., Dec. 1, 1947.

(Navy Contract NOa(s) 8164.)

25. Shillito, T. B., Jones, W. L., and Kahn, R. W.: Altitude-Test-Chamber Investi- gation of Performance of a 28-Inch Ram-Jet Engine. II - Effects of Gutter Width and Blocked Area on Operating Range and Combustion Efficiency. NACA RM E50H21, 1950.

26. Shillito, T. B., and Nakanishi, Shigeo: Effect of Design Changes and Operating Conditions on Combustion and Operational Performance of a 28-Inch Diameter Ram-Jet Engine. NACA RM E51J24, 1952.

27. Anon.: Survey of Bumblebee Activities. Rep. No. 54, Appi. Phys. Lab., The Johns Hopkins Univ., Feb. 1947. (Contract NOrd 7386 with Bar. Ord., U.S.

Navy.)

28. Anon.: Survey of Bumblebee Activities. Rep. No. 64, Appl. Phys. Lab., The Johns Hopkins Univ., July 1947. (Contract NOrd 7386 with Bar. Ord., U.S.

Navy.)

••S SS S S S •• •S I •. •.. • ••• . • S C01ThflE1WIAL. NACA EM E55G28 • • •• • ,• • • • . .

S • I S S • ••• S • S • S • S ••• IS • •S• •S S S SI •S • as 050 I 29. Weedinan, J. A.: An Initial Dig€st of the Literature on Combustion in a Ram Jet.

CAE Rep. No. 324, Continental Aviation and Eng. Corp., Detroit (Mich.) . , Feb.

3, 1947.

30. Longwell, J. P., Eames, J. P., Yahnke, R. L., and Newhall, R. M.: Study of Com- bustors for Supersonic Rain-Jet for period Feb. 1-Mar. 31, 1946. Rep. PDN-4174, Esso Labs., Process Div., Standard Oil Dev. Co., Apr. 30, 1946. (Contract NOrd-9233.)

31. Lustenader, E.: Hermes B-i Ramjet Combustion Tests in a Twelve-Inch Duct. Rep.

No. R50A0524, Apparatus Dept., General Electric Co., Dec. 1950. (Proj.

Hermes (TU1_2000 A).)

32. Nussdorfer, T. J., Sederstrom, D. C., and Perchonok, E.: Investigation of Com- bustion in 16-Inch Rain Jet under Simulated Conditions of High Altitude and High Mach Number. NACA EM E5OD24, 1951.

33. Perchonok, Eugene, and Farley, John M.: Internal Flow and Burning Characteris- tics of 16-Inch Rain Jet Operating in a Free Jet at Mach Numbers of 1.35 and 1.73. NACA EM E51C16, 1951.

34. Chamberlain, John: Development of an 11-Inch Unit for a 30-Inch Diameter Multi- Unit Ramjet. Rep. R-50484-33, Res. Dept., United Aircraft Corp., Sept. 1952.

(Contract NOa(s) 9661, U.S. Navy, Bar. Aero.)

35. Wilcox, Fred A., Perchonok, Eugene, and Wishriek, George: Some Effects of Gutter Flame-Holder Dimensions on Combustion-Chamber Performance of 20-Inch Ram Jet.

NACA EM E8C22, 1948.

36. Dangle, E. E., Cervenka, A. J., and Perchonok, Eugene: Effect of Mechanically Induced Sinusoidal Air-Flow Oscillations on Operation of a Ram-Jet Engine.

NACA EM E54D01, 1954.

37. Garmon, R. C., Moomaw, C. E., and Fenn, J. B.: Pilot Heat and Altitude Burner Performance. TM-150, Experiment, Inc., July 26, 1949.

38. Frost, E. E., Morris, K. G., Petrein, R. J., and Weiss, M. A.: Quarterly Progress Report on Study of Combustors for Supersonic Ram-Jet for period Jan. - 1-Mar. 31, 1951. Rep. No. PDN 5621, Esso Labs., Standard Oil Dev. Co., June 15, 1951. (Contract NOrd-9233.)

39. Frost ) E. E., Morris, K. G., Petrein, R. J., and Weiss, M. A.: Quarterly Progress Report on Study of Combustors for Supersonic Ram-Jet for period Oct. 1-Dec.

31, 1950. Rep. No. PDN 5591, Esso Labs., Standard Oil Dev. Co., Feb. 5, 1951.

(Contract NOrd-9233.)

40. Longwell, J. P., Weiss, M. A., and Van Swerigen, R. A., Jr.: Report on Design Variables in Small-Scale Ram-Jet Altitude Pilots. Rep. No. PEN 5582, Esso Labs., Process Div., Standard Oil Dev. Co., Jan. 9, 1951. (Contract NOrd-9233, SOD/CM 656.)

41. Farley, John M., Smith, Robert E., and Povolny, John H.: Preliminary Experiments with Pilot Burners for Ram-Jet Combustors. NACA EM E52J23, 1953.

42. Dangle, E. E., Friedman, Robert, and, Cervenka, A. J.: Analytical and Experimen- tal Studies of a Divided-Flow Ram-Jet Combustor. NACA EM E53KO4, 1954.

... . S • •S •S • • • • •S .S• • • . • • :6 - . I S • • • S • • . .

S I • • • S • I NACA RM E55G28 : cNDETIA ••• I • S S .. S.. 01 •I• I • •I •• • I I ••• II 43. Henzel, James G., Jr., and Wentworth, Carl B.: Free-Jet Investigation of 20- Inch Rain-Jet Combustor Utilizing High-Heat-Release Pilot Burner. NACA EM E53H14, 1953.

• 44. Reynolds, Thaine W., and Male, Donald W.: Effect of Immersed Surfaces in Combustion Zone on Efficiency and Stability of 5-Inch-Diameter Ram-Jet Com- bustor. NACA EM E54C25, 1954.

45. Mullen, James W., II, and Fenn, John B.: Burners for Supersonic Ram Jets. Some Factors Influencing Performance at High Altitudes - A Resuxn. Tech. Memo. No.

TM-188, Experiment, Inc., Richmond (Va.), Jan. 11, 1950.

Cn 46. De Vault, R. T.: Summary of Subsonic Ramjet Development. USCAL Rep. 2-11, Aero.

Lab., Univ. Southern Calif., Apr. 15, 1947. (Contract NOa(s) 7598, Navy Res.

Proj.).

47. Beckelman, B. F.: Ram-Jet Burner Development. Doc. No. D-9487, Boeing Jet Lab., Boeing Airplane Co., Seattle (Wash.), Dec. 21, 1948. (Contract W-33-038 ac- 13875.)

48. Justice, D. A.: Progress Report for November and December, 1952 - Development of XRJ-30-Mk-8 Subsonic Rats Jet Model MP19G and Model FM8 Fuel Control. Reps.

PR-36-20 and PR-36-21, Marquardt Aircraft Co., Dec. 31, 1952. (Contract NOa(s) 51-762-c.)

49. Shillito, Thomas B., Younger, George G., and Henzel, James G., Jr.: Altitude- Test-Chamber Investigation of Performance of 28-Inch Ram-Jet Engine. III - Combustion and Operational Performance of Three Flame Holders with a Center Pilot Burner. NACA EM E50J20, 1951.

50. Meyer, Carl L., and Welna, Henry J.: Investigation of Three Low-Temperature- Ratio Combustion Configurations in a 48-Inch-Diameter Ram-Jet Engine. NACA • EM E53K20, 1954.

51. Rayle, Warren D., and Koch, Richard G.: Design of Combustor for Long-Range Rain- Jet Engine and Performance of Rectangular Analog. .NACA EM E53K13, 1954.

52. Henzel, James G., Jr., and Trout, Arthur M.: Altitude Investigation of 20-Inch- Diameter Ram-Jet Engine with Annular-Piloted Combustor. NACA RN E54G12, 1954.

53. Cervenka, A. J., Bahr, D. W., and Dangle, E. E.: Effect of Fuel Air Ratio Con- centration in Combustion Zone on Combustion Performance of a 16-Inch Ram-Jet Engine. NACA EM E53B19, 1953.

54. Messing, Wesley E., and Sinipkinson, Scott H.: Free-Flight Performance of 16- Inch-Diameter Supersonic Ram-Jet Units. II - Five Units Designed for Combustion-Chamber-Inlet Mach Number of 0.16 at Free-Stream Mach Number of 1.60 (Units B-1, B-2, B-3, B-4, and B-5). NACA EM E50B14, 1950.

55. Disher, John H., and Rabinowitz, Leonard: Free-Flight Performance of 16-Inch- Diameter Supersonic Ram-Jet Units. III - Four Units Designed for Combustion- Chamber-Inlet Mach Number of 0.245 at Free-Stream Mach Number of 1.8 (Units D-1, D-2, D-3, and D-4). NACA EM E50D07, 1950.

56. Rabb, Leonard, and North, Warren J.: Free-Flight Performance of 16-Inch- Diameter Supersonic Ram-Jet Units. IV - Performance of Ran-Jet Units Designed for Combustion-Chamber-Inlet Mach Number of 0.21 at Free-Stream Mach Number of 1.6 over a Range of Flight Conditions. NACA EM E501,18, 1951.

.

••••' . . ..

268.9 . NACA EM E55G28 . S S S . .

. .

• .

• •S • S • S• S S S S • SS S • S .

S •S• S • S S S ••• •• •• S S •S• •• • S •S .. •5• S 57. North, Warren J. Summary of Free-Flight Performance of a Series of Ram-Jet Engines at Mach Numbers from 0.80 to 2.20. NACA EM E53K17, 1954.

58. Wilcox, Fred, Baker, Sol, and Perchonok, Eugene: Free-Jet Investigation of a EM E50G19, 1950.

16-Inch Ram Jet at Mach Numbers of 1.35, 1.50, and 1.73. NACA 59. Howard, Ephraim M., Wilcox, Fred A., and Dupree, David T.: Combustion-Chamber Performance with Four Fuels in Bumblebee 18-Inch Ran Jet Incorporating Various Rake- or Gutter-Type Flame Holders. NACA EM E8IOla, 1948.

LO 60. Anon.: Bumblebee Semi-Annual Survey. Convair Rep. 5001-9, July-Dec. 1950, Consolidated Vultee Aircraft Corp. (Contract NOrd 9028.)

61. Dupree, D. T., Nuasdorfer, T. J., and Sterbentz, W. H.: Altitude-Wind-Tunnel Investigation of Various Can-Type Burners in Bumblebee 18-Inch Ram Jet. NACA RME8L2O, 1949.

62. McFarland, H. W.: Development of Supersonic Ramjet Burners. USCAL Rep. 8-2, Aero. Lab., Univ. Southern Calif., Feb. 19, 1951. (U.S. Navy Bur. Aero.

Contract NOa(s) 9961, Item 2.)

63. Breitwieser, Roland: Performance of a Ram-Jet-Type Combustor with Flame Holders Immersed in the Combustion Zone. NACA EM E8F21, 1948.

64. Male, Donald W., and Cervenka, Adolph J.: Design Factors for 4- by 8-Inch Ram- Jet Combustor. NACA EM E9F09, 1949.

65. Cunningham, W., ed.: Summary Report on Jet Engine Development. Continental Aviation and Eng. Corp., Detroit (Mich.), Aug. 1, 1949. (Air Forces Contract W-33-038-ac-13371.)

66. Rayle, Warren D., Smith, Ivan D., and Wentworth, Carl B.: Preliminary Results from Free-Jet Tests of a 48-Inch-Diameter Ram-Jet Combustor with an Annular- Piloted Baffle-Type Flameholder. NACA EM E54K15 1 1955.

67. Anon.: Survey of Bumblebee Activities. Rep. No. 82, Appl. Phys. Lab., The Johns Hopkins Univ., June 1948. (Contract NOrd 7386 with Bur. Ord., U.S. Navy.)

68. Messing, Wesley E., and Black, Dugald 0.: Effect of Variation in Fuel Pressure on Combustion Performance of Rectangular Ram Jet. NACA EM E8128, 1948.

69. Reid, J.: Development of Six Inch Diameter Ramjet Combustion C}iamber. Rep. No.

AERO 2282, British R.A.E., Sept. 1948.

70. Reid, J.: The Further Development of a Six Inch Diameter Ramjet Combustion Chamber. Rep. No. G.W. 12, British R.A.E., Mar. 1952..

71. Sterbentz, W. H., Perchonok, B., and Wilcox, F. A.: Investigation of Effects of Several Fuel-Injection Locations on Operational Performance of a 20-Inch Rani Jet. NACA EM E7LO2, 1948.

72. Cervenka, A: J., Perchonok, Eugene, and Dangle, E. E.: Effect of Fuel Injector Location and Mixture Control on Performance of a 16-Inch Ram-Jet Combustor.

NPCA EM E53F15, 1953.

73. Bennet, W. J., and Maloney, J.: Altitude Tests of the Convair 20 Inch Diameter Can Type Combustor. Rep. No. ZM-9136-010, Consolidated Vultee Aircraft Corp., San Diego (Calif.), May 10, 1950.

.. ... . .

.

.. .. . ••. . ...

S.

NACA RN E55G28 • OOIDE*T4A . . . . . .

. . . • 269 • S • SI S • SS I • III I •• S S • I I • • •.• S • • S • • • • •5 •SS •• 5...5•S •• .

S • ••5 •• 74. Huber, Paul W.: Preliminary Tests of a Burner for Ram-Jet Applications. NACA RM L6KO8b, 1947.

75. Trout, Arthur M., and Wentworth, Carl B.: Free-Jet Altitude Investigation of a 20-Inch Ran-Jet Combustor with a Rich Inner Zone of Combustion for Improved • Low-Temperature-Ratio Operation. NACA RN E521,26, 1953.

76. Cervenka, A. J., Dangle, B. E., and Friedman, Robert: Effect of Inlet-Air Tem- perature on Performance of a 16-Inch Ram-Jet Combustor. NACA RM E53103, 1953.

77. Bean, Thomas T.: Results of Direct-Connected Altitude Proof Tests of the XRJ-43- MA-i Model C20-2.5A6 Test Ramjet at the W-P AFB May 12-June 22, 1950. Memo..

CA Rep. No. M-1207, Ramjet Design Section, Marquardt Aircraft Co., Dec. 5, 1950.

78. Kahn, Robert W., Nabanishi, Shigeo, and Harp, James L., Jr.: Altitude-Test- Chamber Investigation of Performance of 28-Inch Ran-Jet Engine. IV - Effect of Inlet-Air Temperature, Combustion-Chamber-Inlet Mach Number, and Fuel Volatility on Combustion Performance. NACA RN E51D11, 1951.

79. Wilcox, Fred A.: Free-Jet Performance of 16-Inch Ram-Jet Engine with Several Fuels. NACA RN E50106, 1950.

80. Wilcox, Fred A., and Howard, Ephraim M.: Comparison of Two Fuels in Bumblebee 18-Inch Ran Jet Incorporating Rake-Type Flame Holder. NACA RN E8F11, 1948.

81. Lundquist, W. G.: Third Quarterly Progress Report on Barn Jet Development.

W.A.C. Ser. Rep. No. 1485, Wright Aero. Corp. (N.J.), Oct. 16, 1950. (Contract AF 33(038) -9000.)

82. Dangle, E. E., Cervenka, A. J., and Bahr, D. W.: Effects of Fuel Temperature and Fuel Distribution on the Combustion Efficiency of a 16-Inch Ram-Jet Engine at a Simulated Flight Mach Number of 2.9. NACA RN E52J14, 1953.

• 83. Reynolds, Thaine W.: Effect of Fuels on Combustion Efficiency of a 5-Inch Ram- Jet-Type Combustor. MACA RN E53C20, 1953.

84. Perchonok, Eugene, Sterbentz, William H., and Wilcox, Fred A.: Perl'ormance of a 20-Inch Steady-Flow Ram Jet at High Altitudes and Ram-Pressure Ratios.

MACA RN E6L06, 1947.

85. Perchonok, Eugene, Wilcox, Fred A., and Sterbentz, William H.: Invetigation of the Performance of a 20-Inch Ran Jet Using Preheated Fuel. NACA RN E6123, 1946.

86. Curry, Richard: Development of a Four-Inch Ramjet Burner Unit for Multi—Unit Test Burner Application. Rep. R-50133-17, Res. Dept., United Aircraft Corp., June 1950. (U.S. Navy, Bur. Aero. Contract NOa(s) 9661, Lot II.)

87. Dailey, C. L., and McFarland, H. W.: Development of Ramjet Components. USCAL Rep. 13-5, Prog. Rep. June through Sept. 1951, Aero. Lab., Univ. Southern Calif., Oct. 8, 1951. (U.S. Navy, Bar. Aero. Contract NOas 51-116-c.)

88. Cervenka,-A. J., and Miller, R. C.: Effect of Inlet-Air Parameters on Combustion Limit and flame Length in 8-Inch-Diameter Ram-Jet Combustion Chamber. NACA RN E8C09, 1948.

89. Anon.: Survey of Bumblebee Activities. Rep. No. 46, Appl. Phys. Lab., The Johns Hopkins Univ., Nov. 1946. (Contract NOrd 7386 with Bur. Ord., U.S.

- Navy.)

..

• • ••.

S ..

S. •SS • ••' • •• . . .

S _''" S S

Cck'I ti1NTi#L . • NACA RM E55G28 . • • •

.. .

• •. •

S • • • •S. • '0' S S • • • • S • ..• ..

55•S••S•• •S • S. •S. S 90. Anon.: Survey of Bumblebee Activities. Rep. No. 49, Appi. Phys. Lab., The Johns Hopkins Univ., Dec. 1946. (Contract NOrd 7386 with Bur. Ord., U.S.

Navy.)

91. Perchonok, Eugene, Wilcox, Fred, and Pennington, Donald: Effect of Angle of Attack and Exit Nozzle Design on the Performance of a 16-Inch Ram Jet at Mach Numbers from 1.5 to 2.0. NACA EM E51G26, 1951.

92. Hearth, Donald P., and Perchonok, Eugene: Performance of a 16-Inch Rain-Jet Engine with a Can-Type Combustor at Mach Numbers of 1.50 to 2.16. NACA RM E54G13, 1954.

H 93. Childs, J. Howard: Preliminary Correlation of Efficiency of Aircraft Gas- Turbine Combustors for Different Operating Conditions. NACA EM E50F15, 1950.

94. Reynolds, Thaine W., and Ingebo, Robert D.: Combustion Efficiency of Homogeneous Fuel-Air Mixtures in a 5-Inch Ram-Jet Combustor. NACA EM E52123, 1952.

95. Wentworth, Carl B.: Performance of Five Low-Temperature-Ratio Rain-Jet Com- bustors over Range of Simulated Altitudes. NACA RN E54H13, 1954.

S.

••• S • •S S. • ••S • ••• • .

:- 271 :0: .

NACA PM E55G28 S S •• S IS I I • S.

I S •SS S S S S S I . •I •S III 15....... .. S S 555555 co

r

II

I

I N 0:::::::::::: I '1 •::::::::::::::::

_I

In

r1

4-)

a3

.........:::•::::•:::::•:•::•::::::::::: rd .: H ::...::...:: .......:::: .::::

C)

ci) .................:::: ............... ::::..:::..:..

.::

4-)

ci)

H

to Cd

::. .........

: ........

C) ::::::::::

EA

H Ii) to LO H rd 4-,

::::. .:.:: 0)

:•::•:•

H U H S r.

co

1T

S. ••S • S• . .

NACA EN E55G28 • 4) C) C a) C') CA I—' (0 4)

01 PRIM 111111, Illill

I

C) V) H a)

_

'Ii-1Iii1

4C0• C') )

EMEMMEMMMMOMMS

.0 (0 a).

I d 0 cC)

M E

MMEMEMEM M E EkI

C') 4) HH Cd Id

EMENOMMUMMEMEN

o a5 (04) o a)

EMEMSEEMENESION

• a)tf) C') C)iO' •0 4) •C) C)

EMEMEEMEMMON's-LE

c 4) CC S CC C) ObO

EMEMEMMEMEMMION

;I tto r C')H

MEMEMEMENUMEom _

(0 (0

MIN

•0 (U H 0 H 0 +'C) a) H

;. •

C) OW

EMEMEM

ki k

rd

MERMENCEL

000LlULI1L1 0

rcf CQ

• MOMMEMELIMM

co

_____

I11U1111

S 0 0 0 0 C') 0 CC) (C) H H J 'PTW :. S.. S S S •• 5. S •S5 S 555 •S . S S S • S S • • if ' 273 CIAWMIIAOL.5 . NAAR1E55G28 . . ..

.5• S S I S S •S• • . • .

S • S.

••• •• ••• S S SS •• S I • ••S •S S Upstream fuel manifold Flaimeholder Water-cooled combustion r Downstream fuel (a) V-gutter flameholder.

a t',,1 —Conica1 shield (b) Sloping-baffle flameholder (ref. 53).

Figure 130. - Typical ram-jet flameholder configurations.

NACA RN E55G26 274 •! '! •: :'' - . . S • •S S • • S I I S I •• • S •S I • S I S S •S• I I I I S • I • S ••. •• ••• •• a •. •. • • • • •• ,•• p Ea -1 41) H 4) rd r4 CaN LO 0 L0

N4 91

0) (a

I

.r4 4-) a, ,-1 rcl a, 0 0 0 o- .0 C) Z 0 .0 H r > H H I 0 41' -H 0 4-' .0 a, a, 41) P1 1 w 00 bO I a, 0coo 9 ' ca 0 Pi a5 +'H P1 0 P 0 594 -4 H0 4., 0 00 U) 00 C H V '-4 4) H r C) a -4-' OU) '-4 O: ,0+' 4-'P '- Q) H C) a, 4-4 4., 0 g C) + Ca 04 00 'd 'd N OH H H +' P4 H O o bD 91 P4 4-i -i HH a) Q) • -1 a, 41) H !

p., a) a) .4-' '-I 0 p OH '-4 I 41) -4 t 4-1 ..

• • S • •I m.

• •, • •• •• • . • . . . • S • S •5 • S S NACA EM E55G28 OFThEN']!'IAL. • . . • • .. • •.

, S S S ••• • S S S S S S S S. •SS S. S..

• • •• 05 • S • ••• •s 0.

LI) .lE c; 13 V r.

H U, I .14 Lean a blow-out Flameholder _ - 1 a) Rich 4) blow-out A a) PA V __ > 6 4-, a) H a) .1-c .04 .05 .06 .07 .08 .09 Blow-out fuel-air ratio Figure 131. - Correlation of blow-out data with gutter size. Combustion- chamber-outlet total pressure, 2000 pounds per square foot absolute.

I __ El I P: 4-, a) Baffle diam, in.

0 0.75 •H l El 0 CO 1.90 2.87 - __ 4 12 20 28 44X105 Baffle radius r Mixture velocity' V3 p Figure 132. - Effect of mixture velocity and baffle size on stability range. Pressure, 1 atmosphere; temperature, 300° F (ref. 17).

55. •S S • S 55 •• 27 '• •• • ••• . C IITG2b . . a*. IDETfT IL• • • S • • •S S S S • S • S S •• • •• • • S • S S S 555 • • S • S S • S S •S• SI S •SS SI •• • • S 50 151 S S.

* (a) Standard flameholder mounted in (b) Three-quarter-scale flameholder flameholder section.

S C-39465 (c) Double-scale flameholder. (d) 1.4 Spaced flatneholder.

S Figure 133. - Gutter-grid flameholders used in 20-inch ram jet (ref. 35).

S •S •S • ••S S •SS •S .. S.. S . S S S S S • • S • S • • .0. • 5 5 5 . S S • • NACA EM E55G28 CC1FDTLL • ••• S • S• • •• •S • t • SS • •5• •• .. ... •5••I • •S • S U) C) Ambient-air 4) a) - pressure, p0, C.)

lb/sq ft abs CH a)Q-) - 0 1190 C) 60 - ft-4 Pq 4-) ME 10 20 30 40 50 60 Gutter angle, deg Figure 134. - Effect on combustion efficiency of gutter angle of adjusta- ble three-gutter ulaniehold.er. 20-Inch ram jet with 5-foot combustion chamber; fuel-air ratio, 0.072 to 0.078; combustion-chamber-inlet static pressure, 1500 pounds per square foot absolute (ref. 35).

1 .1 1

Plameholder gutter width, in.

S 1.0 - 0 1.38 to 1.5 )0 iC) 2.0 a) 4-).'

C) o .,- o C) - r jaLaeLlwu.er u.Luuzs.eu.

Figure 135. - Effect of flameholder blocked area on combustion efficiency.

Fuel-air ratio, 0.050; combustion-chamber-exit pressure, 1800 pounds per square foot absolute (ref. 26).

•S .. ... S •SS S •S S S S SSS • S I • S S S • S • S S S S S S S S • • II S I a 05 • S S S . CFfiIZL F1 55G28 2'8. • • • . . • .,• S S 55 55 5 I SSS IS S. ISS I 555 55 S H c-fl 4) Cd ) -I .3 .3 ) -4 j -1 ) -4 p • •• S• S 555 • ••S 55 .. ... S • S • • • • S S.. S S I • • I • S • • S • S SI I IS • NACA PM E55G28 C c1p DTIJL • ... a . . . . .

• •S• •• .. ... .. •5• • S •S Se • • a

I I I

-1tO C\C'JC'ir- ___ ___ ___ ___ ___ ___ ___ ' \ +1+1+1+1- -1-)U r—cOlI)N ' % • ' !

0)11) \ If LI) \ N H H a) \ V) r1 0000 \ V 0000- H -P (011)00 .. \ -,

:L

ii: _ __ _ _ _ _ c Q+) ) a) -I C) -4 Q) -p () -I O'd LI) C) a) N C) 04) H P1 •r-I CH -P0 LI) Q4-' cl—I cf1 • NO I)) H __ _______ C') ID ;g3 V 0 0 0 0 0 0) N (0 In uo.zad T 'iouoçjj osTncnhI a •SS S 55 55 S. ••S • S S S 555 • I S • • S S • S S S • • • • S S • • S IS S S S •S S S • I

.•. CETIL NACA F4 E55G28

.

2 Q • • • • • • S S 55

S •SS • SI S • 555S• 555 SI D Cn (0 c

I

Iq

If) H '-4 ci) p UD ci) -H • .• S • S• • ••• •• .. S.. • • .

S S S S • S • S S 00 S SS • S • .. ' S • S S S S NACARNE55G CON}LTN1AL' • S •S• 'S .. •SS S S •S S. e S S S •. is, • Air flo S CD-3033 - Figure 139. - Annular-piloted combustor configuration (refs. 50 and 51).

S.. • •• ... ... . •S S 0 0 see 00

. S • S •

• • • S S • .. S :cJrAJ. NACA PM E55G28 : 2;2.. .

a • . . S S I S • ••• S • S • S .. IS• S • S •I •S S • •.• S• ••I •• S lOC - 4-) (i) C) -a — ci) - / .' 9.

C) / ci) C-) -'-I Cl—I a) / Baffle Three-dimensional -

j

(ref. 53)

I - - - -

0 Single-plane .02 .01 .03 .04 .05 a Fuel-air ratio Figure 140. - Comparison of combustion performance of planar and three-dimensional flameholders. Inlet-air temperature, 6000 F; velocity, 230 to 260 feet per second; pressure, 31 to 35 inches of mercury absolute.

•• S. ••• • S S 55 •• S •SS S •S• SS S • S S S• S • • S S S S • • SI •S • S 55 S S S • S • • •5 . 5. • I • S . S I C ON eI]IET7]IAL. NACA NM E55G28 •• . •.. id •I • •• • • •• • • C- 21694 Figure 141. - Rake-type flameholder with flare pilots.

S. •SS I •IS S II II S I I 555 II • • • S S I I I S S I €ONIIDfrIAL NACA EY E55G28 • I I I I S I S ISS S I I I S I S SI IS I S 515 II S. •II I III IS - 4-, H a) H a) H Ii. i..

H a) 4-, I, • S..

S.. S SI I..

•g I, 04.

I. SI a) H

gi

N ; •

N

ooco

H a) -I H a) 4-, 0 'a

ooI

o H :: C H ,-1 a p-I a) El H 0 I

I

/ a) I /' H a) C :: a) ON L !

cri H La -. a) +, 0 0 H H 4-,

h

U, a) H H H U) a

\Ltg

I H U) a • S •• •• • •S• S 555 •• S. •S• S • S• S • • . S . 0 . • • • S • NkCA RN E55G28 C4DT13L ' . 5 •2 ••• S S . S S ' S S S S 555S• •S 555 •• • 55 •S • • CD o - / "--I G LI) r-1 to ° Ily (0 •r4+I I0 Ha) 0 a If) • •.- OW -p D w 0 -4 -4 0 w+) r4 o0\ rH OH ow \ \

Ii i __

\

\

a)

- I 0

I

- 8

• I I I\

I

a)

I '

Wa) P4 a) O cil 0 C) a) • •1-1 H 'd'd'dwaw - 000,.4___ P4 P4 P4 ()0 a -- O e

coa) OLONO 'E

0'i4N0 • 0.

Fi HHHH Wa) ') c' P4 P4 -o o ci '1) -4 Oa)

I

0 0

0 0

. 0 0) CO

N (0

H

.uoo.xd 'SouaToTija uosnquio

ri

•. •S ... S ••• • • S • S S •SS 55 S • S S S S S • S S • S • S S S S IS S IS S S S S S • S IS S S , S S S S S 2i6. • • NACA RM E55G28 • C®TIIELAL S. SS• S S S as 05 S S •5I 55 •SS 55 Air flow go- L.

CA H (TI One row

L

S Two rows S L..

Three rows

1l21

L..

ICD-J]

Four rows Figure 144. — Flaineholder configuration employing immersed surfaces (ref. 63).

•• •• • ••• . S•• •• .. S.. • • S • • • • • •. • ••

E55G28 CCE1TIk ' ' . S ' ' S .287

NPLCA RN S • • S • • S S • S.. S • •SS • S•• •• ••• S • •• •• S • .. •5• •S -t U).

'-4 V) ncooled blades 1!0 blades^\ a) 70 '-I C) CH CH - - / /V-gutter alone .8 .7 .9 1.0 1.1 1.2 1.3 Equivalence ratio Figure 145. - Combustion efficiencies of immersed-surface configurations with both cooled and uncooled blades. Inlet-air pressure, 1 atmosphere; inlet-air temperature, 6600 B; velocity, 200 feet per second; fuel, gaso- line with high-pressure injection.

SI •SI S •S• • 55 50 S S • S•• NACA RN E55G28 :VE1I •:. •:.

.: .

I

CA I-I C Instrumentation and mounting strips- (a) Single-perpendicular-blade configuration.

-- S (b) Four-blade configuration.

Figure 146. - Typical arrangements of blades (immersed surfaces).

•• •. S.. S • •• S. S ••S S •S• • 00. S • • • • S • • . S • S • aS I •S• S • • • • S NACA RN E55G28 ••, . . .• .. • .

. ...

1.

H H S H H -1 S Ao t—.

Y) (a) Comparison of stability limits.

I I I o V-Gutter alone o One blade at 1 inch, perpendicular, and three blades at 2 4, and —.A_.

inches, parallel 8( A One blade at 4 inches, perpendicular — a) C, U 7( C) a) '-I C) -4 C.-.

C) --- --—--o-- 6( 5( .7 .8 .9 1.0 1.4 1.1 1.2 1.3 Equivalence ratio

I

(b) Comparison of combustion efficiency.

Figure 147. - Effect of location of immersed surfaces on combuatton efficiency and stability limits or 5-inch-diameter combustor. Inlet-air temperature, 6600 R; inlet-air velocity, 220 feet per second.

I

NACA RN E55G28 Iq holder Upstream fuel injector,, Airflow Fuel-spray nozzles (a) Upstream-injection burner.

(b) Split-injection burner.

(c) Flamaholder split-Injection burner.

L°

(d) Flameholder injection burner.

Figure 148. - Fuel-injection arrangements used in oper- ational performance Investigation of 20-inch ram jet (ref. 71).

. .. .• . ... . ... ..

.. ... . .

••S S S • S S • S S • S • • S • • •21 coFLA NACA RM E55G28 .

. S.. • • • . . .

S 55 •• . S.. ••

•5 •S• •• •S• • • •

OV Combustion- chamber- inlet static pressure, / lb/sq ft abs 2000 - 1800 - - -.

S 1600 C) S - I I i .05 .o6 7 .03 .04 .05 .06 (a) Upstream-injection (b) Split-injection burner.

burner.

'-I C) 9-i 9-4 a, I80--/--- \13 OL

!hl

.03 .04 .05 .01 .02 .03 .04 .05

Fuel-air ratio, f/a

(c) Fla2nehalder split- (d) Flwneholder injection burner.

injection burner.

Figure 149. - Effect of fuel-injector location on combustion efficiency at various conditions of combustion-chamber-inlet static pressure. 20-Inch ram jet with 8-foot combustion chamber and 17-inch-diameter exhaust nozzle; ambient-air pressure, 20 inches of mercury absolute; equivalent pressure altitude, 10,100 feet (ref. 71).

•. ... . ,.. S .5 •S • S I ••• •• • . . . . . . . • • ONF D'IAL NACA RN E55G28 •s S SI • • S I I I • I I a S S I • S S • • III • I •SI •1••I• S I II II9 •.i I. •II II Injection

Internal I

- - ---Combination upstream and internal ---Upstream (3f.-in. radial position)

r1=

4., o 40 a) (a) Linear location.

lOG Distance of 7 7 fuel injector / - from outer wall, - - in.

N Me

d o4

.01 .02 .03 .04 .05 .06 .07 Fuel-air ratio (b) Radial location.

I

Figure 150. - Effect of fuel-injector location on performance of can- type combustor. Inlet-air pressure, 32 to 36 inches of mercury abso- lute; inlet-air temperature, 1050 0 to 1070 0 R; velocity, 230 to 260 feet per second; fuel, MIL-F-5624A grade JP-4 (ref. 72).

NACA RM E55G28 C ONFIWM%T.P6L 293 . .

.

.. ..

S •SS • ••s .. • . .

.

S S • • • • . S • 5 • • • . •5 • • 5S S • S S. • •• S • • S • S • ••• S • • 5 5• S S .. SI. S. ••5 S • S.

•• • • . ... I.

a lOC to a) C) a) a) hi

N

8( a, Diameter, in.

'-I 12.5 and 15 - •1• (combined)

I 12.5 -

\ / - 13.5

I

14.5 15.0 - - - Primary only .01 .0 .03 .04 .05 .06 .07 Fuel-air ratio (a) Combustion efficiencies.

S C) 0) O•rl C) .p rl tl) ,-1 a, H a) 1a4 15 14 Secondary-fuel-injector ring diameter, in.

(b) Fuel-air ratio at maxiimim efficiency.

S Figure 151. - Effect of Secondary-fuel-injector ring diameter on com- bustion. Inlet-air pressure, 15 inches of mercury absolute; inlet-air U temperature, 580 0 R; fuel, ANF-48; 20-inch-diameter combustor; 20-per- cent primary flow (ref. 73).

S 294 NACA FM E55G28 • , ca'ITIA1 .

•• • SS • •• •. •.. . S S • • S S • I I • S S • • . •S S • • • S S S S • SS • SI I S • I • S .

•SS • S S I S S S •• • S •S• •SS SI •• . S •• • • S. •S• S U Control sleeve M1111 SIEMENS!

L1

Li

I I I Primary fuel-air ratio o 0.014 .02 O .034 o Primary fuel only —0--- O O4- . .-1 p • ,—i ci) - I O4- C) U) .01 .02 .03 .04 .05 .06 Fuel-air ratio Figure 152. - Effect of fuel staging on combustor performance.

Inlet-air pressure, 32 to 36 inches of mercury absolute; inlet- air temperature, 1050 to 1070 0 R; velocity, 230 to 260 feet per second; fuel, M]1-.F-5624A (ref. 72).

I S

I

S NkCA RN E55G28 C OIF]T3iiU S • S • S S • • • . .

. . • . •.

.

. • . .

. • .

5 . S. • • • . .

. . .•.

.

. . .

S.. ..

•• ••• • S •• •• • • . ..• ..

.07 .06 t 4-, a a, .05 S a, t .04 I I I .03' I I I I I I I 18 20 22 24 26 28 30 32 34 Ratio of primary to total fuel flow, percent Figure 153. - Effect of primary fuel flow on combustor operating limits of 20-inch-diameter can combustor.

Inlet-air pressure, 15 inches of mercury absolute; temperature, 910 0 H; velocity, 280±40 feet per second; fuel, ANF-48 (ref. 73).

Hurner inlet Mach number 0.21 ID C, a, U) a, ,-1 i-a U) o 70 •4 C) 27/ .27 50 L 0 10 20 30 40 50 60 70 80 90 Inlet-air pressure, in. Hg abs Figure 154. - Effect of inlet-air pressure on burner efficiency. Burner length, 14 inches; fuel, pentane; atoichiometric conditions; pilot heat, 10 percent (ref. 45).

•S S . ..• S •S NACA RN E55G28 • • S •S S • • • S S • •• • • . S. S S S S S I S •I• • I I S • SI • I •l •SI • ••S •S SI S • S S. 151 • air 4-, 80 - C, a P, C) a C) '-4 a 0 - '-I 4-, a 2r - 6 8 10 20 40 60 80 100 Inlet-air pressure, in. Hg abs Figure 155. .- Combustion efficiency of 20-inch-diameter rain-jet com- bustor as function of fuel-air ratio and burner inlet pressure. Inlet- 9200 8; fuel, ANF-58; exit nozzle, 55 percent (ref. 77).

air temperature, I.

I

Inlet conditions - - - .06 - Temperature, Velocity, OR ft/sec 1060 290±30 910 270±30 - 810 245j20 Rich limit - .07 - — —. — — / .06 -14 / 4-, a -.-- '-4 a -4 a .05 Lean limit r 8 10 12 14 16 18 20 22 Inlet-air pressure, in. Hg abs Figure 156. - Effect of inlet-air pressure on combustion limits for three inlet temperatures in 20-inch-diameter rem jet with can-type flameholder. Fuel, ANF-48 (ref. 73).

NPCA RM E55G28 CON IML • S • S • • • • • S S S S • S S • S •S • • • S • • • •• . •. • S • S S • • ••• S S • • • • • . S.

••• •• ••• .

S •S •• • • S ••5 55 U -1-' r.

a) C) o'n-Pentane Pq - -c - - 6C - 0 n-Heptane - C) Isooctane a) CO V Cyclohexane K) C) - - Pentene-2 - csi o Benzene CH o - a, 0 P t- Propionaldehyde —<--- — — - )1 —.4 Diethylether '-I V Propylene oxide M Acetone - I Methylcyclohexane - t- Carbon disulfide D Isopentane - 0 Neohexane - • Blend A • Blend Gasoline I I 6 8 10 12 /P'T\(uf \l.l Correlation parameter, Q8 / U.ef) Figure 157. - Correlation of combustion efficiency for V-gutter flarrieholder in 5-inch ram-jet combustor for 14 pure fuels, a gasoline, and 2 fuel blends.

(Blend A contains 2/3 propylene oxide plus 1/3 isopentane by weight; blend BY 1/3 propylene oxide plus 2/3 isopentane by weight.)

S