Metal Cases

Metal cases have several advantages compared to filament‐reinforced plastic cases: they are more rugged and will take the considerable rough handling required in many tactical missile applications, are usually ductile and yield before failure, and may be heated to relatively high temperatures (700 to 1000 °C or 1292 to 1832 °F, and higher with refractory metals) thus requiring less insulation. They will not deteriorate significantly with time or weather exposure and are easily adapted to take concentrated loads, if made thicker at a flange or boss. Since a metal case has much higher density and less insulation, it occupies less volume than a fiber‐reinforced plastic case; therefore, for the same external envelope, it may contain somewhat more propellant.

Figure 15–1 shows various sections of a typical large solid rocket motor case made of welded steel pieces. The shape of the case, particularly the length‐to‐diameter ratio in cylindrical cases, influences not only the stresses withstood by the case but the quantity of case material required to enclose a given amount of propellant. For very large and long motors both the propellant grain and the motor case have been made in sections; these case segments are then mechanically attached and sealed to each other at the launch site. The segmented solid rocket booster used in the Space Shuttle is shown in Fig. 15–2 and discussed in Ref. 15–1. Although this rocket motor is now out of production, it represents a good example of a segmented design; other similarly segmented rocket motors are still in use today. For critical seals between segments, a multiple‐O‐ring type of joint is often used as shown in Fig. 15–3 and discussed in Ref. 15–2. Segments are needed when an unsegmented rocket motor would be too large and too heavy to be transported over ordinary roads (i.e., cannot make road turns) or railways (will not go through some tunnels or under some bridges) and/or are too difficult to fabricate.

Small metal cases for tactical missile rocket motors can be extruded or forged (and subsequently machined), or made in three pieces as shown in Fig. 12–4. Such motor case is designed for loading a freestanding grain where the case, nozzle, and blast tube are sealed by O‐rings (see Chapter 6 of Ref. 15–4 and Chapter 7 of Ref. 15–5). Since mission velocities for most tactical missiles are relatively low (100 to 1500 m/sec), their propellant mass fractions are also relatively low (0.5 to 0.8) and the percentage of inert motor mass is high. Safety factors for tactical missile cases are often higher to allow for rough handling and cumulative damage. The emphasis in selecting motor cases (and other hardware components) for tactical missiles is therefore not on highest performance (lowest inert motor mass), but on reliability, long life, low cost, safety, ruggedness, and/or survivability.

High‐strength alloy steels have been the most common case metals, but others, like alloys of aluminum, titanium, and nickel, have also been used. Table 15–2 gives a comparison of rocket motor case material properties. Extensive knowledge exists today for designing and fabricating motor cases with low‐alloy steels with strength levels of over 240,000 psi.

Maraging steels have strengths up to about 300,000 psi in combination with high fracture toughness. The term maraging is derived from the fact that these alloys exist as relative soft low‐carbon martensites in their annealed condition and attain higher strengths from aging at relatively low temperatures.

Stress‐corrosion cracking in certain metals presents unique problems that may result in spontaneous failure without any visual evidence of an impending catastrophe. Emphasis given to lightweight thin metal cases aggravates stress corrosion and crack propagation that often starts from a flaw in the metal, with failures occurring at a stress levels below the documented yield strength of the original metal.

Wound‐Filament‐Reinforced Plastic Cases

Filament‐reinforced cases use continuous filaments of strong fibers wound in precise patterns and bonded together with a plastic. Their principal advantage is lower weight. Most plastics soften when they are heated above about 180 °C or 355 °F; they need inserts or reinforcements to allow fastening or assembly of other components and to accept concentrated loads. Thermal expansion of reinforced plastics is often higher than that of metal and the thermal conductivity is much lower, causing higher temperature gradients. References 15‐4 and 15‐5 explain the design and winding of these composite cases, and Ref. 15–6 discusses their damage tolerance.

Typical fiber materials are, ordered by increasing strength, glass, aramids (Kevlar) and carbon, as listed in Table 15–2. Typically, for the same load the inert mass of a case made of carbon fiber is about 50% of one made with glass fibers and around 67% that of a case mass made with Kevlar fibers.

Individual fibers are strong only in tension (2400 to 6800 MPa or 350,000 to 1,000,000 psi). Fibers are held in place by plastic binders of relatively low density; this prevents fibers from slipping and thus weakening in shear or bending. In a filament‐wound composite case (with existing tension, hoop, and bending stresses) filaments are not always oriented along the direction of maximum stress and the materials may include low‐strength plastics; therefore, the strength of composites may be a factor of 3 to 5 less than the strength of the filament itself. The traditional plastic binder is a thermosetting epoxy material, which limits maximum temperatures to between 100 and 180 °C or about 212 and 355 °F. Although resins with higher temperature limits are available (295 °C or 563 °F), their fiber adhesion has not been sufficiently strong. Typical safety factors used (in deterministic structural analysis) are for failure to occur at 1.4 to 1.6 times the maximum operating stresses, and proof testing is done up to 1.15 to 1.25 times operating pressures.

A typical case design is shown schematically in Fig. 15–4. The forward and aft ends and the cylindrical portion are wound on a preform or mold which already contains the forward and aft rings. The direction in which the bands are laid onto the mold and the tension that is applied to the bands is critical in obtaining proper case designs. Curing is done in an oven and may be done under pressure to assure high density and minimum voids in the composite material. The preform is then removed. One way is to use sand with a water‐soluble binder for the preform; after curing the case, the preform is washed out with water. Since filament‐wound case walls can be porous, they must be sealed. The liner between the case and the grain can be the only seal that prevents hot gases from seeping through the case walls. Scratches, dents, and moisture absorption degrade case strengths.

In some designs insulator material is placed on the preform before winding and the case is cured simultaneously with the insulator, as described in Ref. 15–7. In another design the propellant grain with its forward and aft closures is used as the preform. A liner is applied to this grain, then an insulator, and the high‐strength fibers of the case are wound in layers directly over the insulated live propellant. Here, curing has to be done at relatively low temperatures so that the propellant will not be adversely affected. This process works well with extruded cylindrical grains. There are also cylindrical cases made of steel with an overwrap layer of filament‐wound composite material, as described in Ref. 15–8.

Allowable stresses are usually determined from tensile tests with roving or band clusters and rupture tests on subscale composite cases made by an essentially identical filament winding process. Some companies further reduce the allowable strength value to account for degradation due to moisture, manufacturing imperfections, or nonuniform densities.

In a wound motor case the filaments must be oriented in the direction of the principal stress and must be proportioned in number to the magnitude of that stress. Compromise occurs around parts needed for nozzles, igniters, and so on, where orientations are kept as close to the ideal as practicable. Filaments are customarily clustered in yarns, rovings, or bands, as indicated in Fig. 15–5. By using two or more winding angles (i.e., helicals and circumferentials) and by calculating the proportion of filaments in each direction, a balanced stress structure may be achieved. The ideal balance is for each fiber in each direction to carry an equal tension load. Realistically, filaments supported by an epoxy resin must also absorb stress compressions, bending loads, cross‐laminar shears, and interlaminar shears. Even though the latter stresses are small compared to the tensile ones, each must be analyzed since they can lead to case failure before any filament fails in tension. In a proper design, failure should occur only when filaments reach their ultimate tensile strength, rather than from stresses in other directions. Figure 18‐5 shows the cross section, made of ablative materials, of a Kevlar filament motor case and detail of its flexible nozzle.

Classification

Nozzles for solid propellant rocket motors can be classified into five categories as listed below and as shown in Fig. 15–6.

1. Fixed nozzle. Simple and frequently used in tactical weapon propulsion systems for short‐range air‐, ground‐, and sea‐launched missiles, also as strap‐on propulsion for space launch vehicles such as Atlas and Delta, and in spacecraft rocket motors for orbital transfer. Typical throat diameters are between 0.25 and 5 in. for most tactical missile nozzles, but some are larger.
2. Movable nozzle. Provides thrust vector control to a flight vehicle. As explained in Chapter 18, one movable nozzle can provide pitch and yaw control and two are needed for roll control. Movable nozzles are typically submerged and use a flexible seal or a ball‐and‐socket joint with two actuators 90° apart to achieve omniaxial motion. Movable nozzles are primarily used in long‐range strategic propulsion ground‐ and sea‐launched systems (typical throat diameters are 7 to 15 in. for the first stage and 4 to 5 in. for the third stage) and in large space launch boosters. Titan's boost rocket motor and Ariane's V solid rocket booster have throat diameters in the 30‐ to about 90‐in. range.
3. Submerged nozzles. A significant portion of the nozzle structure is submerged within the combustion chamber or case, as shown in Figs. 15–2 and 15–6 b, c. Submerging the nozzle reduces the overall motor length somewhat, which in turn reduces the vehicle length and its inert mass. It is important for length‐limited applications such as silo‐ and submarine‐launched strategic missiles as well as their upper stages, and for space propulsion systems. Within a fixed motor length, it also allows an increase of propellant mass inside the case. The splitline between fixed and movable sections is more favorably located than with external subsonic splitline movable nozzles (see Ref. 15–9). Reference 15–10 describes the sloshing of trapped molten aluminum oxide that can accumulate in the groove around a submerged nozzle; this accumulation is undesirable, but can be minimized by proper design.
4. Extendible nozzle. Commonly referred to as an extendible exit cone (EEC), although it is not always exactly conical. It is used on strategic missile propulsion upper‐stage systems and upper stages on space launch vehicles to maximize motor‐delivered specific impulse. As shown in Fig. 12‐3, this nozzle has a fixed low‐area‐ratio section, which is enlarged to a higher area ratio by adding one or more nozzle cone extension pieces. The extended nozzle improves specific impulse by doubling or tripling the initial expansion ratio, thereby significantly increasing the nozzle thrust coefficient. This system thus allows very high expansion ratio nozzles to be packaged in a relatively short length, thereby reducing vehicle inert mass. The nozzle cone extensions reside in their retracted position during the boost phase of the flight and are moved into place before their rocket motor is started but after separation from the lower stage. Typically, electromechanical ball screw actuators deploy exit cone extensions.
5. Blast‐tube‐mounted nozzle. These are used with tactical air‐ and ground‐launched missiles. The blast tube allows the rocket motor's center of gravity (CG) to be close to or ahead of the vehicle CG. This limits the CG travel during motor burn and makes flight stabilization much easier. See Fig. 12‐4.

As stated in Chapter 12 nozzle throat areas often enlarge slightly during hot firing. An enlargement of more than 10% is usually considered unacceptable. With slight enlargements of the throat area, both the chamber pressure and the specific impulse will slightly decrease. When single value for specific impulse applicable during their entire operation is used in rocket motor specifications it is named the effective specific impulse.

Design and Construction

Almost all solid rocket motor nozzles are cooled by heat absorption in ablative materials. In general, construction of a rocket nozzle features steel or aluminum shells (or housings) designed to carry structural loads (where motor operating pressures and nozzle TVC actuator loads are usually the biggest), and composite ablative insulators (with or without liners) which are bonded to the cases or to the housings. These ablative liners are designed to insulate metal housings, provide the internal aerodynamic contour necessary for efficient combustion gas expansion to generate thrust, and ablate and char in a controlled and predictable manner in order to prevent heat buildups that could substantially weaken the structural housings or the bonding materials. Solid rocket motor nozzles are designed to ensure that the thickness of ablative liners is sufficient to maintain the liner‐to‐housing adhesive bond line below temperatures that would degrade the adhesive structural properties during rocket motor operation. Nozzle designs are detailed in Figs. 1–5, 12–1 to 12–4, and 15–7.

The construction of nozzles ranges from simple, single‐piece nonmovable graphite nozzles to complex multipiece nozzles capable of controlling the direction of the thrust vector. The simpler, smaller nozzles are typically of applications with low chamber pressure, short durations (perhaps less than 10 sec), low area ratios, and/or low thrust. Two typical small, simple built‐up nozzles are shown in Fig. 15–7. Complex nozzles are necessary to meet more difficult design requirements such as providing thrust vector control, operating at higher chamber pressures (and thus at higher heat transfer rates) and/or higher altitudes (large nozzle expansion ratios), producing very high thrust levels, and surviving longer motor burn durations (above 30 sec).

Figures 15–8 and 15–9 illustrate design features of a large and relatively complex solid rocket motor with thrust vector control. It was used as the Reusable Solid Motor rocket (RSRM) booster of the Space Shuttle Launch Vehicle, which was retired in 2011. It is typical of other large flexible nozzles. The two RSRMs provided 71.4% of the lift‐off thrust of the Space Shuttle launch vehicle shown in Figs. 1–14 and 15–2. The nozzle was designed to provide structural and thermal safety margins during the Shuttle booster's 2‐min burn time and consisted of nine carbon‐cloth phenolic ablative liners bonded to six steel and aluminum housings. The housings were bolted together to form the structure of the nozzle. A flexible bearing (described further in Chapter 18), made of thin rubber sheets vulcanized to spherically shaped steel shims, enabled the nozzle to vector omni‐axially up to 8 degrees from centerline to provide thrust vector control. Since these metal housings were recovered and reused after each flight, an exit cone severance system (a circumferential linear shaped charge) was used to cut off a major section of the aft exit cone just below the aft exit cone aluminum housing to minimize splashdown loading on the remaining components. NASA's new first‐stage boosters (segmented solid propellant rocket motors) are being patterned after the Space Shuttle's SRM design.

From a performance perspective, any nozzle should efficiently expand the gases originating from the motor's combustion chamber to produce thrust. Simple nozzles with noncontoured conical exit cones can be designed using the basic thermodynamic relationships presented in Chapter 3 to establish the throat area, nozzle half angle, and expansion ratio. More complex contoured (bell‐shaped) nozzles are used to reduce divergence losses, improve the specific impulse, and reduce nozzle length and mass. Section 3.4 gives data on designing bell‐shaped nozzles with optimum wall contours (that avoid shock waves and minimize impact of particulates in the exhaust gas).

Two‐dimensional, two‐phase, reacting‐gas method‐of‐characteristics flow codes are used to analyze gas‐particle flows in nozzles for the determination of optimal nozzle contours and of shortened nozzle lengths that maximize specific impulse while yielding acceptable erosion characteristics. Such codes provide results for all identified specific impulse loss mechanisms (in units of secs), which produce a less than ideal performance. An example is given in Table 15–3.

Figure 15–10 illustrates amounts of carbon cloth phenolic liner removed by chemical reactions and by particle impingement erosion, the liner char depth, and gas temperature and pressure at selected locations in the RSRM nozzle. Erosion on the nose cap (1.73 in.) is high primarily as a result of impingement by particles traveling down the motor bore. The mechanical impact of these particles removes charred liner material. In contrast, the radial throat erosion of 1.07 in. results primarily from the carbon liner material reacting chemically with oxidizing species in the combustion gas flow at the region of greatest heat transfer. At the throat location, impingement erosion is low because any particles present are traveling parallel to the nozzle surface.

The often used acronym ITE means integral throat/entrance and refers to a single‐piece nozzle throat insert that also includes a part of the converging entry section. ITE nozzle inserts can be seen in Figs. 1–5, 12–1, 12–3, and 12–4. Most ITEs are made from carbon‐carbon (either 3 or 4 directionally reinforced). The introduction of ITEs has improved reliability and simplified designs (see Ref. 15–22).

As already stated, nozzle throat erosion causes throat diameters to enlarge during operation and is one problem encountered in nozzle design. Usually, a throat area increase larger than about 10% is considered unacceptable for most applications, since it causes a significant reduction in thrust and chamber pressure. Erosion occurs not only at the throat region (typically, at 0.01 to 0.25 mm/sec or 0.004 to 0.010 in./sec), but also at the sections immediately upstream and downstream of it, as shown in Fig. 15–10. Nozzle assemblies typically lose 3 to 12% of their initial inert mass during operation. Erosion is caused by complex interactions between the high‐temperature, high‐velocity gas flow, the chemically aggressive species in the gas, and the mechanical abrasion by small particles. Carbon in nozzle materials reacts with hot species like , , , or becoming oxidized; the cumulative concentration of these species is an indication of the likely erosion. Tables 5–6 and 5–7 give chemical concentrations for the exhaust species from aluminized propellants. Fuel‐rich propellants (which contain little free or ) and propellants where some of the gaseous oxygen is removed by aluminum oxidation show fewer erosive tendencies. Uneven erosion in a nozzle causes thrust misalignment. Some tactical designs use a particular grain design that burns very progressively but counter this with a high erosion‐rate material to control the maximum system pressure.

For bell‐shaped nozzles finding optimum nozzle wall contours requires analysis (computer codes using the method of characteristics are mentioned in Section 3.4) to determine wall contours, which effectively turn the gas to near axial flow without introducing strong shock waves. Results given in Section 3.4 are for gaseous exhaust jets (without entrained solid particles), as may occur with liquids or double‐base solid propellants. Figure 3‐13 illustrates key parameters, which describe such bell shaped nozzles. The initial angle (angle through which supersonic flow is turned immediately downstream of the nozzle throat) is typically 30 to 50° and the exit angle is 10 to 15°. Such bell‐shaped contour allows a 1 to 2% increase in specific impulse, when compared with an equivalent conical nozzle exit having a 15° half angle. For some applications, designers have used a parabolic curve or a circular arc without performance penalties.

With solid particles in the exhaust gas bell‐shaped nozzles (designed for a purely gaseous exhaust) quickly erode, particularly at the nozzle exit region where a diverging flow is turned back to a near axial direction. With the insulation eroded, the nozzle exit metal housing quickly burns out. All solid composite propellants have some amount of small solids (mostly ) in their exhaust. A solution for not losing excessive nozzle insulation materials is to go to a smaller value of the initial angle (20 to 32°) on a gentler bell‐shaped contour. Compared to a 15° conical nozzle, the specific impulse gain of such contour is no more than 0.7%. In general, nozzle throat inlet contours are not critical and can be based on any hyperbolic curve that uniformly accelerates the combustion gas flow to sonic velocity at the throat. In the supersonic region the task is more complicated, see Section 3.4. for details.

Heat Absorption and Nozzle Materials

Since solid rocket motors do not reach thermal equilibrium, the temperature of all components exposed to heat flow increases continuously during operation. In proper thermal designs, critical locations only reach their maximum allowable temperature a short time after the rocket motor stops operation. Nozzle components rely on their heat‐absorbing capacity (high specific heat and high energy required for material decomposition) and slow heat transfer (good insulation with low thermal conductivity) to withstand the stresses and strains imposed by thermal gradients and loads. The maximum allowable temperature for motor materials is taken as just below the temperature at which excessive degradation occurs (e.g., the material loses strength, melts, becomes too soft, cracks, pyrolyses, unglues, or oxidizes too rapidly). Operating durations are thus affected by nozzle design and by the amount of heat‐absorbing and insulating material present. Stated in a different way, one principal design objective here is to arrive at a nozzle with just sufficient heat‐absorbing‐material and insulation so that its structures and joints will do the job for the duration of the application under all likely operating conditions, while complying with mandated margins or factors of safety.

Selection and application of proper materials is the key to successful designs of solid rocket nozzles. Table 15–4 groups various typical nozzle materials according to their usage. High‐temperature exhausts from solid rockets present an unusually severe environment for nozzle materials, especially when metalized propellants are employed.

About 80 years ago nozzles were manufactured from a single piece of molded polycrystalline graphite and some were supported by metal housing structures. They eroded easily, but were low in cost. We still use them today for short duration, low chamber pressure, low altitude flight applications with low thrust, such as in certain tactical missiles. For more severe conditions a throat insert or ITE is placed into the graphite piece; this insert is a denser, better grade of graphite; later pyrolytic graphite washers and fiber‐reinforced carbon materials came into use. Figure 15–7 shows a set of non‐isotropic pyrolytic graphite washer in the throat insert of small nozzles (they are not used now). For a period of time tungsten inserts were used; they had very good erosion resistance, but were heavy and their melting point was eventually exceeded as higher motor pressures and hotter propellants came into use. The introduction of high‐strength carbon fibers in a carbon matrix has been a major advance in high‐temperature materials. For small and medium‐sized nozzles, ITE pieces have been made of carbon–carbon, the present abbreviation for carbon fibers in a carbon matrix (see Ref. 15–12). The orientation of the fibers can be two‐directional (2D) or three‐directional (3D), as described below. Some properties of these materials are listed in Tables 15–4 and 15–5. For large nozzles the then existing technology did not allow the fabrication of large 3D carbon–carbon ITE pieces, so layups of carbon fiber (or silicon fiber) cloth in a phenolic matrix were used. An example of a self‐supporting ablative exit cone is the Naxeco‐phenolic structure used in the first stage of the Vega launch vehicle (Ref. 15–13).

Regions immediately downstream of the throat have less heat transfer, less erosion, and lower temperatures than the throat, and less expensive materials are usually satisfactory here. This includes various grades of graphite, or ablative materials, strong high‐temperature fibers (carbon or silica) in a matrix of phenolic or epoxy resins, which are described later in this section. Figure 18‐5 shows a movable nozzle with multilayer insulators behind the graphite nozzle pieces directly exposed to heat. These insulators (between the very hot throat piece and housing) limit the heat transfer and prevent excessive housing temperatures.

In the diverging exit section the heat transfer and temperatures are even lower and similar but less capable ordinary materials can be used there. This segment can be built integral with the nozzle throat (as it is in most small nozzles), or it can be a separate one‐ or two‐piece subassembly, which is then fastened to the smaller diameter throat segment. Ablative materials without oriented fibers as in cloth or ribbons, but with short fibers or insulating ceramic particles, can be utilized. For large area ratios (upper stages and space transfer vehicles), the nozzle will often protrude beyond the vehicle's boattail surface. This allows for efficient radiation cooling since any exposed exit cone can radiate directly to space. High‐temperature metals (niobium, titanium, stainless steel, or a thin carbon–carbon shell) have been used for radiation cooling in some upper‐stage or spacecraft exit cone applications. Since such cooled nozzle sections can quickly reach thermal equilibrium, their duration can be unlimited.

The housing or structural support of the nozzle uses the same material as the metal case, such as steel or aluminum. Housings are never allowed to become very hot. Some of the simpler, smaller nozzles (with one, two, or three pieces of mostly graphite) do not have a separate housing structure, but use the ITE for the structure.

Estimates of nozzle internal temperatures and temperature distributions with time can be made using two‐dimensional finite element software for transient heat transfer analyses. These are similar in principle to the transient heat transfer method described in Section 8.5 where some results are shown in Fig. 8‐20. After firing, nozzles cool by conducting heat from hotter inner parts outer sections. Sometimes these outer pieces will end up exceeding their temperature limit and suffer damage. Structural analyses (of stresses and strains) of key nozzle components are highly interdependent with heat transfer analyses, which determine the component temperatures. Design must also allow for thermal growth and for differential expansion of adjacent parts.

Typical materials used for ITEs or nozzle throat inserts are listed in Table 15–5. These materials are exposed to the most severe heat transfer conditions and undergo high thermal stresses and temperatures. Their physical properties are often anisotropic, that is, vary with the orientation or direction of the crystal structure or of reinforcing fibers. Polycrystalline graphites are extruded or molded units. Different grades with different densities and capabilities are available. As already mentioned, they are used extensively for simple nozzles and for ITE parts. Pyrolytic graphite is strongly anisotropic and has excellent conductivity in one preferred direction. A nozzle using pyrolytic graphite is shown in Fig. 15–7. It is fabricated by depositing graphite crystals on a substrate in a furnace containing decomposing methane gas. While pyrolytic graphite use is declining, it is still installed in current rocket motors of older design.

Carbon–carbon materials are made from carefully oriented sets of carbon fibers (woven, knitted, threaded, needled, or laid up in patterns) in a carbon matrix. Two‐directional (2D) materials have fibers in two directions, 3D have fibers oriented in three directions (usually at right angle to each other), and 4D has fibers oriented in one of two ways, often called “tetrahedral” and “hexagonal” (Ref. 15‐14). In the tetrahedral construction, the carbon bundles are the same diameter in all four directions represented by the four diagonals of a cube. In the hexagonal construction, three of the four directions of reinforcement are in a plane perpendicular to the axial direction. The three directions of yarn bundles in the perpendicular plane are at 60° to each other and are usually smaller in diameter. Highly densified materials are superior in high heat transfer regions, such as the throat. Multidirectional fiber reinforcements allow them to better withstand the high thermal stresses introduced by the steep temperature gradients within the component. There are two methods in use for matrix deposition: chemical vapor deposition (CVD) and liquid impregnation followed by further pyrolyzation. Carbon‐carbon densification is discussed in Ref. 15–9.

The French (Herakles) Naxeco Sepcarb carbon‐carbon and the Naxeco 3‐D reinforced phenolic materials are presently in production and have flown in Vega launches. Carbon‐silicon carbide and refractory metal nozzle components have reached production in the last decade.

Ablative Materials

These commonly used materials in rocket motors nozzles are also used in some insulators. They are usually made from a composite material with high‐temperature organic or inorganic high‐strength fibers, namely, high silica glass, aramids (Kevlar), or carbon fibers, impregnated with organic plastic materials such as phenolic or epoxy resins. The fibers may be individual strands or bands (applied in a geometric pattern on a winding machine), or come as a woven cloth or ribbon, all impregnated with resin.

Ablation may result from a combination of surface melting, sublimation, charring, decomposition in depth, and from film cooling. As shown in Fig. 15–11, progressive layers of ablative material undergo an endothermic degradation, that is, physical and chemical changes occur that absorb heat. While some ablative material evaporates (and some types also have a viscous liquid phase), enough charred and porous solid material should remain on and below the surface to preserve the basic geometry and surface integrity. Upon rocket start, ablative materials act as a thermal heat sink, but their poor conductivity causes their surface temperature to rise rapidly. At temperatures of 650 to 800 K some of resins start to decompose endothermically into a porous carbonaceous char and into pyrolysed gases. As the char depth increases, these gases undergo an endothermic cracking process as they percolate through the char in a counterflow direction to the heat flux. These gases then form an artificial fuel‐rich, protective, relatively cool but flimsy, boundary layer over the char.

Since char is almost all carbon and can withstand 3500 K or 6000 R, porous char layers allow the original surface to be maintained (but with a rough surface texture) and provides geometric integrity. However, the char's carbon may be oxidized by certain combustion gas species in which case there is a slow regression of the charred surface. Char is structurally weak and can be damaged or abraded by direct impingement of solid particles from the flowing gas. Ablative material construction is used for some or all parts of the chambers and/or nozzles shown in Figs. 1–5, 6–9a, 12–1 to 12–4, and 15–10.

Ablative parts are formed either by high‐pressure molding (˜55 to 69 MPa or 8000 to 10,000 psi at 149 °C or 300 °F) or by tapewrapping on a shaped mandrel followed by either hydroclaving, typically at 1000 psi and about 300 °F, or by autoclaving at about 250 psi and 330 °F. Tapewrapping has been a common method of forming in very large nozzles. The wrapping procedure normally includes heating a shaped mandrel (∼54 °C or 130 °F), heating the tape and resin (66 to 121°C or 150 to 250 °F), pressurizing the tape of fiber material and the injected resin in place while rolling (∼35,000 N/m or 200 lbf/in. per inch of width), and maintaining the proper rolling speed, tape tension, wrap orientation, and resin flow rate. Experience has shown that the as‐wrapped density is an important indicator of procedural acceptability, with the desired criterion being near 90% of the autoclaved density. Resin content usually ranges between 25 and 35%, depending on the fabric‐reinforcing material and the particular resin and its filler material. Normally, the mechanical properties of the cured ablative material, and also the durability of the material during rocket operation, correlate closely with the cured material density. Within an optimal density range, a low density usually means poor bonding of the reinforcing layers, high porosity, low strength, and high erosion rates.

We note here that in liquid propellant rocket engines ablatives have been used effectively in very small thrust chambers (where there is insufficient regenerative cooling capacity), in pulsing, restartable spacecraft control rocket engines, in the exit region of large nozzle and in variable‐thrust (throttled) rocket engines. Figure 6‐9a shows an ablative nozzle extension for a large liquid propellant rocket engine.

Heat transfer properties of many available ablative and other fiber‐based materials are highly dependent on their design, composition, and construction. Figure 15–12 shows sketches of several common fiber orientation approaches. The orientation of fibrous reinforcements, whether in the form of tape, cloth, filaments, or random short fibers, has a marked impact on the erosion resistance of composite nozzles (Fig. 15–10 shows erosion data). When perpendicular to the gas flow, erosion resistance is the greatest but heat transfer to outboard surrounding materials is also the highest. Good results have been obtained at throat and entrance locations when the fibers are at 40 to 60° relative to the gas flow, as the taped‐wrapped ply at these angles can still provide high erosion resistance. However, at the exit cone, shallower ply angles (0 to 20°) are used as they reduce the heat transfer while simultaneously providing erosion protection and reducing nozzle weight (Ref. 15–14.)

Pyrotechnic Igniters

In industrial practice, pyrotechnic igniters are defined as igniters (other than pyrogen‐type igniters) that use solid explosives or energetic propellant‐like chemical formulations (usually small pellets of propellant that provide large burning surfaces and short burning times) as the heat‐producing materials. This definition fits a wide variety of designs, known in the trade as bag and carbon igniters, powder can, plastic case, pellet basket, perforated tube, combustible case, jellyroll, string, or sheet igniters. A common pellet‐basket design is shown in Fig. 15–14, being typical of the pyrotechnic igniters. Ignition of the main charge (in this case pellets consisting of 24% boron–71% potassium perchlorate–5% binder) is accomplished by stages; first, on receipt of an electrical signal the initiator releases the energy of a small amount of sensitive powdered pyrotechnic housed within the initiator, commonly called the squib or the primer charge; next, the booster charge is ignited by heat released from the squib; and finally, the main ignition charge propellants are ignited.

A special form of pyrotechnic igniter is the surface‐bonded or grain‐mounted igniter. It has its initiator included within a sandwich of flat sheets; the layer touching the grain has the main pyrotechnic charge. This form of igniter is used with multipulse rocket motors with two or more end‐burning grains. The ignition from the second and successive pulses of these motors presents unusual requirements for available space, compatibility with grain materials, life, and for the resulting pressures and temperatures from booster grain operation. Advantages of the sheet igniter include light weight, low volume, and high heat flux at the grain surface. Any inert material employed (such as wires and electric ceramic insulators) is usually blown out of the motor during ignition and resulting impacts have at times either damaged or plugged the nozzle, particularly if the blown material has not been intentionally broken up into small pieces.

Pyrogen Igniters

A pyrogen igniter is basically a small unit, containing all the elements of a rocket motor, used to ignite a larger rocket motor, see Ref. 15–11, and is not designed to produce thrust. They all consist of one or more nozzle orifices (both sonic and supersonic types) and most use conventional rocket motor grain formulations (sometimes the same as the main propellant grain) and design technology. Heat transfer from the pyrogen gas to the motor grain is largely convective, with hot gases contacting the inner grain surface, in contrast to pyrotechnic igniters that transfer heat by high‐energy radiation. Figures 12–1, 12–2, and 12–20 illustrate rocket motors with a typical pyrogen igniter; in Fig. 18–5 the igniter has three nozzles and a cylindrical grain with high‐burn‐rate propellant. For pyrogen igniters, initiator and booster charge designs are often very similar to those used in pyrotechnic igniters. Reaction products from the main charge impinge on the inside surface of the rocket motor grain, producing motor ignition. Common practice on very large motors is to mount them externally, with the pyrogen igniter pointing its jet up through the large motor nozzle, in which case the igniter becomes a piece of ground‐support equipment.

Two approaches are commonly used to safeguard against rocket motor misfires, or inadvertent ignition; one is to use of the classical safe and arm device and the second is to design safeguards into the initiator. Energy sources causing unintentional ignitions (usually disastrous when they happen) can be (1) static electricity, (2) induced electrical currents from electromagnetic radiation—such as radar, (3) induced electrical currents from ground test equipment, communication apparatus, or nearby electrical circuits in the flight vehicle, and (4) heat, vibration, or shock from motor handling and operations. Functionally, the safe and arm device serves as an electrical switch to keep the igniter circuit grounded and interrupted when not operating; in some designs it also mechanically misaligns or blocks the igniter's gas flow passage so that unwanted ignition is precluded even when the initiator charge may fire. As the device is moved into the arm position, the electric ignition circuit is no longer blocked and an ignition flame can reliably propagate from the igniter's booster and main charges to the propellant surface.

Electric initiators in motor igniters are also called squibs, glow plugs, primers, and headers; they always constitute the initial element in the ignition train and, if properly designed, can act as a safeguard against unintended ignition. Three typical designs of initiators are shown in Fig. 15–15. Both (a) and (b) types form structurally a part of the rocket motor case and generically act as headers. In the integral diaphragm type (a), the initial ignition energy is passed in the form of a shock wave through the diaphragm activating the acceptor charge, with the diaphragm remaining integral. The same principle is also used to transmit shock waves through metal case walls or metal inserts in filament‐wound cases; here, the combustion case would not need to be penetrated and can remain sealed. The header in type (b) resembles a simple glow plug with two high‐resistance bridgewires buried in the initiator propellant charge. The design in type (c) employs a small bridgewire (0.02 to 0.10 mm) of low‐resistance material, usually platinum or gold, that explodes by the application of a high‐voltage discharge.

Initiator safeguard aspects appear as a basic design feature in the form of (1) a minimum threshold electrical energy required for activation, (2) voltage‐blockage provisions (usually, air gaps or semiconductors in the electrical circuit), and/or (3) responsiveness only to a specific energy pulse or frequency band. Invariably, such safeguards may compromise to some degree the safety provided by the classical safe and arm device.

A more recent method of initiating igniter action of is to use a laser as an energy source to start the combustion in an initiator charge. Properly designed, there are no problems with induced currents or other inadvertent electrical initiations. The energy from a small neodymium/YAG laser, external to the motor, travels in fiber‐optical glass cables to the pyrotechnic initiator charge (Ref. 15–16). Sometimes an optical window in the case or closure wall allows the initiator charge to be inside the case.

Igniter Analysis and Design

The basics of initiating ignition are common to all designs and applications of pyrotechnic and pyrogen igniters. In general, current analytical models of physical and chemical processes relevant to igniter design (including heat transfer, propellant decomposition, deflagration, flame spreading, and chamber filling) are far from complete and accurate. See Chapter 5 by Hermance and Chapter 6 by Kumar and Kuo in Ref. 14–1, and Ref. 14–15. Analyses and design of igniters, regardless of the type, depend heavily on experimental results. When data on past successes and failures with full‐scale motors is included, the effects of some important parameters have become quite predictable. For example, Fig. 15–16 can be useful in estimating the mass of the igniter main charge for rocket motors of various sizes (i.e., motor free volumes). From these data,

15–3

where is the igniter charge mass in grams and is the motor free volume in cubic inches (the void in the case not occupied by propellant). A larger igniter mass flow means a shorter ignition delay. Ignition time events are shown in Fig.14‐3.