FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1949
1949 - 1748.PDF
512 FLIGHT, 20 October 1949 CAS TURBINE FUEL SYSTEMS A Survey of the Primary Factors in Control System Design By W. A. Andrews, B.Sc, A.C.G.I. IN the course of the last few years the gas turbine hasbeen utilized in many different types of power unit.In the simplest of these air is compressed, heated by combustion of fuel, and expanded through a propelling nozzle, the compressor being driven by a turbine placed immediately downstream of the combustion chamber. In discussing fuel control for such an engine, the first essential fact is that for any given set of nacelle conditions and aircraft speeds there is a unique relationship between the engine r.p.m., gas temperatures, gas pressures, and so forth, and the fuel supplied. From this it follows that only one of these quantities may be chosen as the basic quantity to be controlled, the remainder being covered by individual limitations im- posed either by the pilot or by auto- matic means. The two most important factors in determining the life of an engine are flame temperature and r.p.m., since the former affects the turbine strength and the latter its loading. Since the two are related, either may be selected as a basis for control although, due to the influence of atmospheric tempera- ture, the ideal is probably a combina- — —~ ~~~~ tion of the two. "•'"'"'. Since pilots have, in the past, been accustomed to flying by engine r.p.m., this has become generally accepted as the proper indication of running conditions, and the duty of the control system has been to vary the r.p.m. in accord- ance with pilot's lever movement, independently of atmospheric or aircraft conditions. Such a requirement can be met in two ways: by providing a speed-selector device in which any deviation from the r.p.m. selected would give a signal changing the fuel supply so as to restore the original condition; or by calculating the fuel required and devising a fuel system to meter the appropriate fuel to the engine and so ensure the r.p.m. dictated by the pilot's lever. The former is referred to as a monitoring, or sensing system; the latter as a scheduling system. In general, scheduling has so far been adopted through- out this country, whilst sensing systems have been used in America. The Rolls-Royce Derwent, and the De Havilland Ghost and Goblin fuel systems are examples of the former. The reason underlying the British use of scheduling is probably that hydraulic servo-systems had been widely used prior to the introduction of jet aircraft, and in conse- quence it was logical to use a system dependent upon known factors for stability, as against a system with an operating loop including the engine. It should be appreciated at this juncture that, although the steady running characteristics of a turbojet are susceptible to comparatively simple analysis, transients present an entirely complex problem of which much has yet to be learned. Before proceeding to discuss scheduling r.p.m. control (this being the system in general British use), it is necessary to outline the fuel requirements. It is usual when analysing results of centrifugal compressors to express the charac- teristics in terms of non-dimensional parameters, which are interrelated according to known curves, or functions. The same method is applicable to complete jet engines. The relationships of interest for present purposes are as follows: — F / N THE author of this article, which introducesthe basic elements of fuel control system design, has been actively engaged in this department of gat turbine technology for several years. He is thus able to bring to his treatment of the subject a specialized know- ledge of high order, and this, in conjunction with the clarity of the exposition makes the presentation of the subject matter doubly valuable. Tj = Nacelle temperature. r1= = Blower compression ratio. The first part of this equation is true only for constant component efficiencies, and for the static case when Tj = atmospheric temperature, but it is also close enough for most purposes over the cruise / take-off range, where efficiencies are high. It should also be noted that F should be in heat units per unit time, and is modified by a constant (dependent on the calorific values of the fuel) to convert it to flow rate. Thus if F is written as, say, g.p.h., the equa-.' tion applies to one fuel only. The second part is of interest in that it gives the means for calculating blower delivery pres- sure, which is that obtaining at the fuel burner discharge orifice. The importance of this pressure depends on the type of fuel control used, but usually its effect is of secondary order. On the Derwent V engine in service it is (very approximately) one-twentieth of the pump pressure and, conse- quently, produces about two to three per cent effect upon the flow, the system being simply a pair of orifices in series with a constant upstream pressure. The Rolls-Royce Dart turboprop is, by contrast, fitted with a flow control system in which fuel flow is determined by throttle pressure drop, with the result that the effect of back pressure is negligible. Reverting to the first part of equation (i), it will be seen that for fixed nacelle conditions Tlf Plt the engine r.p.m. is a function of fuel consumption only. Further, for fixed r.p.m., which is the requirement for any given throttle position, the .fuel consumption is: — F = Pl/t(Ti) •- •• ^. ,: -.> <«) One of the idiosyncrasies of hydraulic systems is that, whilst lending themselves readily to the addition of forces, they are less adaptable as multipliers. Fortunately a natural relationship exists in the atmosphere between tem- perature and pressure. This is such that on an I.C.A.N. day at o m.p.h., a fairly close approximation to equation (ii) is given by : — F = Cj V - C3 Po .. • Where Po = atmospheric pressure and Clf C.,, C3 are constants. (iii) where F = Fuel consumption. N = Engine r.p.m. P, = Nacelle pressure. l 3Equation (iii) is the type of characteristic obtained if the pressure drop across a fixed orifice is made proportional to P,,. It is the approximate characteristic of the Barometric Pressure Control on ~iOO] ' ' ' r~ "• the Derwent V en- gine or the Flow Control on the Dart and Mamba engines. The absence of tem- perature in the equa- tion at once halves the functions re- quired of the con- trol, i The disadvantage of the system is that the atmospheric tem- peratures vary from I.C.A.N. at all alti- tudes and conse- quently it is impos- sible accurately to schedule for every combination by 6 2 1OO Fig. I. Curve of consumption against r.p.m. for a typical turbojet at 0 m.p.h. at sea level (I.C.A.N.).
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
