FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1958
1958 - 0685.PDF
4O° 3OO (OO - O TO 1 1 I 1 | \TL \ END OF ] N COPPER STRIP —-*i — — — - 1— 11 I NICKELTUBE — COPPERHEATING — — STRIP TS ^ FLUID BULK TEMPERATURE i TS ToTL 0 TW 00 O1 O2 O-3 O-4 0-5 O-6 O7 DISTANCE FROM CENTRE OF HOT WALL (in) 0 8 COMBUSTION AND PROPULSION . . . flame-tube wall temperatures are better understood and a con-sistent flame-tube life of 2,000 hours without inspection has been achieved; efficient skin-cooling at temperatures in excess of1,300 deg K has produced a flame-tube life greater than other components of the engine." Combustion theory can now beapplied by easy, reliable rules, but combustion pressure losses— mainly parasitic and due to primitive mixing methods—could beimproved. In the future, M = 2.5-3.0 turbine-inlet temperature and the high compressor intake temperature reduce the mechanicalpressure rise required, and the tendency is for the turbine to govern the frontal area. The combustion chamber would be madeas small and light as possible—even at the expense of some loss. Mr. Lombard then repeated his familiar arguments in favourof the by-pass turbojet for the optimum subsonic transport, and developed the theme of integration on the basis of the N.G.T.E.jet flap boundary-layer control on flaps, both the suck-and-blow and Attinello super-circulation, and laminarization. The opposite approach, the Rolls-Royce /Griffith jet-lift trans-port, discussed in Flight of April 18 and 25, was offered as an economic possibility at M=2.5, where the un-reheated turbojetattains an »/t of 38 per cent. The lecturer concluded upon nuclear power, and indicated thatpowerplant and shielding weight (say 100,000 lb) might be over half the aircraft weight and four times the density of normal pro-pulsion equipment, thus eliminating structural bending relief. After outlining the technical difficulties, including shielding andthe radiation crash hazard, he concluded: "What this adds up to is that a nuclear-powered aircraft is possible, and may be demon-strated within the foreseeable future, as a military scientific achieve- ment. As a practical military aircraft fulfilling a need incapable offulfilment in a better manner by other means it is some distance away. A radical development in nuclear reactors will be neces-sary before the picture can change—but this is not impossible. Although also presented in the first group, Mr. Robert S.Levine's paper Development Problems in Large Liquid Rocket Motors was as different in approach as it was in content. Intro-ducing his subject Mr. Levine mentioned non-scientific problems including reliability factors: if a hypothetical missile has 500 com-ponents of 99.5 per cent reliability, overall reliability will be (0.995)500, or 77.8 per cent; if a 95 per cent reliable propulsionsystem (also of 500 components) is required, each must be (0 95)i/50o; or 99.99 per cent reliable; "to prove that any com-ponent will not malfunction more than once in 10,000 tries takes much special testing, quality control and inspection. The neces-sity for lightness, normally incompatible with efficiency, entails careful design, raw-material selection, painstaking manufacture,extensive testing, and redesign—which is where the money goes. Only in a few special cases is any unusual technical knowledgerequired. The second source of expense is the vast infrastructure necessitated by safety precautions (which increases with the sizeof the project), the need for remote sites (because of noise, with consequent roads and pipelines) and, as engines get larger, themachinery to handle them. More than a whole eight-hour shiit is usually required for a single firing. To keep down costs, modelsare used wherever possible; often these are designed to simulate one parameter and frequently bear little resemblance to the original.An example of this last was a two-dimensional model used to study high-frequency combustion instability. This was a one-inchslice across the injector of a 2ft-diam. chamber, wherein alternate paired nozzle-holes inject ethyl-alcohol and Lox, impinging hke-on-like to form a flat fan-shaped spray. The model could simulate a tangential acoustical mode of the cylindrical chamber. Hign-speed photography (14,000 frames/sec) and Photocon Dynagage high-speed pressure pick-ups were used to establish that unstablecombustion could arise from the acoustical mode with pressure amplitudes of the order of 100 lb/sq in in a 200 lb/sq in chamber.From the tests it is possible to establish a "stability rating in terms of the magnitude of the disturbance which will render thecombustion unstable. With the use of such a criterion the effect of engine changes can be established in some five to ten tests—asagainst hundreds of runs to establish the statistical stability ot an engine that may already be near 100 per cent.The walls of large rocket engines are rather more easily cooled (Left) Temperature profile around a heated tube. The laboratory rig is dis- cussed in the text (Leyine). (Right) Conditions under which engine inlet area becomes greater than com- pressor frontal area. 25 23 2-1 1u ^HUB/TIP RATIO 1-5 701FLIGHT, 23 May1958 TURBOJET' COMPRESSOR DESIGNS •3 -4 -5 -6 -7 8 COMPRESSOR INLET AXIAL MACH 9 I-O NUMBER than are small ones because the propellant flow-rate (by weight)to the surface requiring cooling is greater. Boiling of the liquid normally suffices to overcome the heat-transfer to the inner walland maintains the contact-surface within 50-60 deg C of the boil- ing point at the jacket pressure. However, increasing combustionpressures have naturally been accompanied by higher jacket pres- sures until the critical pressure of the fluid has been exceeded, andevaporation is no longer available. The heat-transfer phenomena are complicated by the fact that one side only of the jacket isheated. The laboratory rig consisted of a rectangular nickel tube plated to a copper heater bar, the whole being the length of therocket jacket, and contained within a pressure tube. Ten welding sets could generate 10,000 A, giving a local heat-transfer rate upto 15 B.Th.U./sq in/sec, while a flow of 47 Imp. gal/min could be maintained at 2,500 lb/sq in. Tests up to 6 B.Th.U./sq inwith a temperature difference up to 480 deg C compared closely with the Sieder-Tate heat flux equation—an empirical formulaevolved with hydrocarbon oils in round tubes with temperature differentials below 150 deg C.Large rockets have some peculiar ignition problems, particularly local explosions, due to inadequate ignition allowing unlit pocketsof fuel to form, which can trigger unstable combustion. Full-scale "slice" two-dimensional models of the chamber, with heavy Lucitesides having a thin internal layer of Pyrex glass to prevent burning, are used to study flame behaviour. The answer to the ignitiontrouble is to use sufficient energy, distributed across the injector face, plus precise control of the propellant entry. The time-scaleof such tests is indicated by the 7,000 frames/sec photography used—with detonation occurring at the fifth and sixth frames inan example. In other respects large engines have a higher perform- ance than small, probably because their scale permits multiplepropellant injectors, thus avoiding over-rich concentrations, while the greater physical length of the nozzle (for a given area ratioand expansion angle) allows more time for chemical equilibrium changes in the exhaust. Large rockets present special problems in fluid dynamics. "Waterhammer" can be disastrous in a 20ft pipeline of thin-gauge metal where, perhaps, a 2,000 h.p. turbopump comes to full speed in afraction of a second. Another trouble can occur in a large diameter Lox pipe, perhaps 20ft high, in which the fluid warms as it standsbefore firing. The hydrostatic head means that there is a higher boiling point at the bottom of the pipe and a bubble will formand force up the column of liquid, releasing the pressure on the superheated Lox, so that a large bubble will "percolate" to thetank, causing severe water-hammer shock as the pipe refills. This cycle must be suppressed. [Typical geometry can be seen in thedrawing on pages 706-7—Ed.] Low-frequency instability arises from a sudden increase inchamber pressure retarding the propellant flow, which in turn reduces the chamber pressure, which is followed by a surge ofpropellant—the cycle repeating at 10 to 100 c/s. The rule-of- thumb for stability is a pressure drop of one-half the chamber pres-sure to achieve damping. In large rockets elasticity of the vehicle structure affects pipeline resonance, normally a useful stabilizingadjunct, while the elasticity of the piping and the casing of the turbopump tend to decouple line resonance. In consequence, theminimum injector/chamber pressure drop for stability is 0.15. Theoretically, a high-speed flow control system could be usedinstead, but it seldom is because of its weight and complexity. Mr. Levine's final words crystallized his introduction andpointed his moral: ". - . the attainment of good reliability depends on achieving very high reliability in components and assemblies—this is done by straightforward engineering methods." Mr. J. Howard Childs. of the N.A.C.A. Lewis Flight Propulsion Laboratory, opened Group II with Effects on Turbojet Combustors and Afterburners of other Engine Components. Although the "combustor" of a new turbojet is developed on a rig which simu- lates first-order effects of airflow delivery, it is usual to meet trouble from second-order effects, such as time-varying flow, warped velocity profiles, or acoustic resonance between the rotor and the intake duct. . Transonic compressors today have potential air mass flows otthe order of 40 lb/sec/sq ft with high inlet Mach numbers and low hub-tip ratios, which necessitate a turbine with two or morestages if the frontal area is to be kept within compressor dimen- sions Above about M = 1.7 the air-intake area will be greater than
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
