The motivation of aviation

1.3 Developments in transport technology, and in aviation in particular, are motivated not only by engineering incentives but also by its social implications and rewards. Moreover, apart from what society may want from aviation, the size of the job before us and what it will cost brings up problems. Many of the developments in aviation will have to be supported by society as a whole. So we must have good reasons why such deve­lopments are desirable and should be supported and paid for by society. We must admit that the technical case is not enough: the fact that something is technically feasible does not necessarily mean that it should be done. We need a wider and more rational basis for future developments in aeronautics, as a background to the problem of designing aircraft. However, we are only at the beginning in this search for guidelines, and we cannot yet offer a com­plete and rational answer to the question of motivation.

We are now going through a phase where many advocate that we should be guided by oomneroidl considerations: the creation of wealth. To put it crudely, a simple continuous cycle is envisaged: it begins at the "market place" where

we sell our wares and make a profit and where, at the same time, the blind or manipulated "market forces" dictate what the next sales line should be. The profits are then used to carry out the necessary research into the technical aspects of this new line and to finance the development and production of the new article. Then we go back to the market place, and so on (see e. g. F E Jones (1969)). This may be an arguable approach just now, but we cannot assume that it can serve as a guide in the long term. Less short-sighted is the view that "economic life is movement". This has always been true and is likely to hold also in the future.

More recently, the effects of technology on our environment and, in particular, what is called pollution have received attention (see e. g. P A Libby (1973)). These are serious matters which can provide some of the guidelines we are seeking and can lead to design requirements and criteria. Some of these, such as noise, will be taken up below.

If we really want a rational basis for future long-term developments, we must be guided by what we know about the nature of mm. If we knew enough about the behaviour of man and his natural make-up and about people and their insti­tutions in relation to their environment, we might conceivably evolve a system of organising our living conditions, which is at least sufficient in that it is so designed that our inborn instincts and our natural make-up and our legitimate interests are respected or, at least, not violated. Such a system is only thinkable on a global scale.

If we think of these social implications of transport technology, we are main­ly interested in the social and spatial mobility within society and how this affects its structure, that is to say, in those aspects which are influenced by the available means of transport and communications. Not very much is known about any of these. So only a few examples to think about can be given.

The first example is one of the first cases where some biological effects of transport technology can be demonstrated clearly and expressed in terms of numbers. It is concerned with the genetic structure of the human species and how it is influenced by the gene flow between contiguous populations. This, in turn, is influenced by the actual movement of people, which depends on the means of transport available. Two aspects of evolutionary importance are the amount of exogamy practiced by any particular population and the distribution of distances over which the individuals obtain their mates, i. e. the distri­bution of marriage distances. We may assume that marriages are partly the by­product of people moving about and meeting other people. Some of this travel may have been in the course of business and work, but much must have been what is now called "optional travel": not for business or holiday purposes but just to meet other people and places. This may be taken as an inborn instinct: to move about and to meet others. A demographic and genetic study of a group of Oxfordshire villages by C F KUchemann et al.(1967) gave some remarkable results. It was found that, between 1650 and 1850, one marriage partner came from another parish in about 1/3 of the marriages and that this ratio suddenly jumped to about 2/3 from 1850 onwards. The distance over which the outside partners were obtained was quite small and also almost incredibly constant over long periods, considering the many other factors which must come into this. The marriage distance was around 10 km up to about 1850 but then this, too, jumped and reached several times the former value, as is

Fig. 1.4 Marriage distances in Oxfordshire villages. After KUchemann et aZ/1967)

illustrated in Fig.1.4. This constancy of the average marriage distance and in social mobility and then the sudden changes must have something to do with the available means of transport and especially with the fact that a railway was built through the neighbourhood in 1850. This is a most remarkable result as it is concerned with one of the most fundamental aspects in human life, which provides staple material for poets, and yet we find that the dominating fac­tor is really the available means of transport. These can have dramatic con­sequences and provide a means for increasing human satisfaction.

We also note from the results in Fig. 1.4 that there is a preferred travel – ting time, at least in this important business of marrying. It would appear to be about two hours, and this is then independent of the available means of transport. This must be something fundamental in man’s natural make-up. It supports our contention that it is more significant to measure distances in hours than in kilometers.

Some further conjectures about the characteristics of human behaviour may be inferred from the same study. One is to move only during day time; another is to have a home base. If this turned out to be generally true, it would have drastic consequences on the design of aircraft. It would make it even more important to keep the travelling time short, irrespective of distance.

More generally, we may conclude with P L Roe (1972) that, in all history of travel, we may observe two constants which, because they concern human nature, may confidently be extrapolated into the future: the significance of personal contact between people, and the reluctance of most people to undertake fre­quently journeys which last more than a few hours. Regardless of how any of us personally regards the prospect of a global village in which all men are members of a truly international society, it does seem very probable that this is the eventual destiny that a peaceful earth must tend toward. But this can­not come about until all major cities and centres of population are brought within a few hours of each other: the means of travel must grow to embrace the globe and allow everyone to communicate readily and cheaply with everyone else. To do this in a way which suits human nature is the contribution that aviation can make, and this should be our ultimate aim.

Following P L Roe (1972), we may think in terms of regions which will have to be brought within reach of convenient travel. We may suppose that the number of journeys people will wish to make from one region to another depends in some way on the number of "attractions" to be found in the other region, such as trading centres, political capitals, mineral wealth, holiday resorts, or just "people" and "places" they would like to meet and see. If we suppose that the attractiveness of a region is simply proportional to its area, then the requirement for journeys over a distance R is

J(R) = sin (it R/Rg) (1.14)

for a spherical earth. Very roughly, the actual distribution of population in large cities, shown in Fig. 1.5, looks like that, with a maximum for the potentially most heavily used transport routes at about one quarter of the way around the globe, and with a secondary peak at short ranges in the already developed regions. This is a striking enough conclusion, and we may expect that the actual transport requirement will, in time, approach something like that given by (1.14). It would be quite unrealistic and also irresponsible to assume that future developments in aviation will still be restricted to serve mainly those relatively few people in Europe and North America. If we con­sider what appears to be technically possible, we can begin to think of a

Population in city poirs Fig. 1.5 Population distributions ("potential traffic"). After Naysmith (1969)

global network of routes, from very short ranges to global ranges, where no two places are further than about two hours apart.

We may speculate on how the availability of such means of transport may even affect the genetic structure of the human species. A model for the possible resulting changes has been constructed by R W Iliorns et al. (1969) on the assumptions that there are, at present, an infinite number of populations in the world, that the exchange rates are symmetrical, and that the exogamy component for each population is distributed evenly with respect to all other populations and is 20%, which is a relatively low value. It then follows that about 30 generations – or 600 to 800 years – would be required for the resulting population to become homogeneous over the whole world, but that the population variability would be much greater in this global village than it is now. We do not want to pronounce on the desirability, or otherwise, of such a development, but we should be aware of the possible consequences.

Our next example is again concerned with the population and how it is distri­buted in cities and regions. This is the population problem as formulated by Lord Florey (1965). To quote, "there is now overwhelming evidence that rapid population growth is bringing with it dire consequences. Evidence is slowly accumulating that the question is not simply whether food can be supplied for an ever-increasing population, but whether overcrowding per se does not lead to obscure and so far ill-defined difficulties of mental and social adjustment to a crowded and rapidly changing environment. Perhaps we should be paying more attention to the generally unpleasing form that life is assuming in great cities. It may be that to relate population to environment optimally is the greatest technological task of the end of this century". In this task, trans­port technology and aviation, in particular, must play an essential part. It

can be used to design cities, towns, villages, which would serve Florey’s purpose better. They make it possible for designers and planners to think out new layouts and arrangements. The roughly circular shape of present cities probably had many good reasons in the past. One of these is to make it poss­ible for everybody to communicate readily with everybody else. This made good sense in antiquity and in the middle ages, with the means of transport avail­able then. But it does not make sense now. We should not base our thinking on the assumption that everything will remain as it is now, only scaled up in size; that we can extrapolate from the present to the future. This implies that we should not assume that we shall be lumbered for ever with great lumps of roughly circular cities, ever growing and eating into the countryside. Instead, we should think of finding other ways of living together; perhaps, by planning our cities on a linear rather than a radial principle. We can indeed begin to think of planning urban developments on a large scale in terms of linear cities stretching over continents, leaving plenty of "countryside" in between and allowing for excellent communications between all parts. It is clear that such a plan will have to rely heavily on air transport. Indeed, this may turn out to be the socially most useful application of aviation.

An associated problem is how to prevent people from congregating in the big cities, how to keep them sensibly distributed within countries and over con­tinents. Part of the solution to this problem must again require the provision of suitable means of transport and communication to keep people mobile and in contact with others while they live otherwise isolated all over the country­side. So here is another case where aviation can prove itself to be vital to our society in providing mobility to people at all economic levels. In fact, this may turn out to be one of the first and main cases where civil aviation is not motivated by commercial profitability or considerations of prestige but by the contributions it can make to many social and economic goals by affect­ing regional developments, population distribution, and land use. We are only at the beginning of such developments and much more work is needed to determine what society really needs and wants. We may suppose that developments of this kind are already going on in some way in some countries (see e. g. the Joint DOT-NASA Study, Anon (1971)). In these cases, we are concerned with short – haul and medium-haul aircraft. When we said earlier that the demand for long – range transport aircraft was likely to increase quite out of proportion, we can now add that the same is likely to be true also for aircraft to fly over short and medium ranges. Thus a whole spectrum of aircraft will be wanted, from the shortest to the longest ranges.

Another example is concerned with another strong streak in our natural make­up: the need for comfort. What may loosely be called comfort for our purpose

is the influence of the environment on the senses of man and the resulting reactions which man classes as agreeable, bearable, or annoying. This leads to a "comfort scale" which is primarily a function of time: what is quite

bearable for a short while may become intensely annoying if it lasts too long. As far as travelling is concerned, H Busch (1970) has given a useful survey and has included under comfort in a wider sense also punctuality and waiting at the beginning and the end of the journey as well as during intermediate stops or changes. He explains how the degree of comfort must be increased with journey time to make the travelling bearable in a civilised society.

People may stand up to half an hour and, up to about two hours travelling time, the provision of a seat will do. Beyond that, means of ever-increasing com­plexity must be provided, such as meals, entertainments, lounges and bars and showers, sleeping accommodation, at greater and greater expense of space, weight, and real cost. In the end, we have a rolling, floating, or flying home

Prolegomena

or hotel. These devices have nothing to do with transport as such and their provision seems utterly absurd. This is proof in the nature of reduotio ad absurdum, if ever there was one. Technology can now be used to eliminate such irrelevant absurdities and to provide pure and sensible means for transporting people and goods from any one place to any other place. Strictly, the purpose of travelling is not fun or entertainment. All this implies that the means of transport should be designed to keep the travelling time short: about two

hours comes up again and would seem to be a reasonable limit and target.

Lastly, to put the prospects of air transport into perspective, we may look at some results of 6 Gabrielli & Th von Karman (1950) who discussed all types of transport vehicle: living, terrestrial, marine, and aerial. Some of their

results, together with some recent ones, are shown in Fig.1.6. To find some

measure for the price to be paid for speed, they considered the work needed to be done to achieve a given transport performance and concluded that all vehicles give results which lie above a certain "limiting line": for every

class of vehicle, there is a certain limiting speed beyond which the vehicle becomes uneconomical. It appears that there is a price to be paid for speed: to go faster requires a greater tractive force per unit weight. Ships, trains on rails, and classical and swept aircraft touch the same limiting line at various points according to their increasing speeds. For the latter, a family of aircraft can, in fact,, be defined, which follows this limiting line

precisely, as we shall see later in Section 4.2. This puts each class of vehicle into its proper place, if we accept that journey time in the signifi­cant parameter for covering a certain distance. Within the meaning of Fig,1.6, it is also more efficient to go by rail rather than by road, over distances up to about 200 or 300 km. Beyond that, aircraft may take over as the favoured means of transport. What is surprising and important is that, as will be shown below, the recent results for new types of aircraft depart from this limiting line and do not appear to require the large specific tractive powers to reach their speeds, as predicted only 25 years ago when "the commercial airplane" was thought to be a propeller-driven monoplane with unswept wings, flying at 320 mph! One is tempted to conclude that, as a "universal law", one should never work harder to reach one’s destination two hours away than the man who walks on his feet.

To supplement this overall picture of the rightful places of various means of transport, we must also consider costs. At present, air transport is probably already the most suitable and cheapest for journeys over longer ranges, from transatlantic onwards. But it is still dearer in terms of the price to be paid to travel over a certain distance than many other means of transport, such as motorcars, buses, and trains, for the shorter distances up to one or two thou­sand kilometres. However»the cost of air transport, in real terms, has come down steadily over many years, and the expected advances which we discussed earlier can all be used to lower the costs further. This should bring the price of air transport more in line with other modes also over the shorter ranges *

As to the time taken by various means of transport to cover given distances, we reproduce some estimates due to R Smelt (1971) and E G Stout & L A Vaughn

(1971) shown in Fig.1.7. An envelope is drawn, which touches what has been estimated for the pedestrian, the motorcar, and aircraft. Rail transport appears to take rather a long time so that, by comparison with the placing in

Fig.1.7 Some present "transportation gaps". After Stout & Vaughn (1971)

LIVE GRAPH

Click here to view

Fig.1.6, rail and motorcar interchange positions. Various "transportation gaps" are identified as the difference between the envelope and the actual curves, and further evolution of aircraft is considered to have to fill both the short-haul and the long-haul gaps. On the basis of the future prospects we have discussed earlier, we do not have to accept that travelling time will have to continue to increase with range. Thus another boundary has been drawn in Fig.1.7, which limits the time to roughly two hours. The gap to be filled is then much wider, but there is no reason to suppose that it cannot be eliminated (see also G J Schott & L L Leisher (1975)).

To sum up this brief overall review: we have seen that aviation is only just

growing up and that it has reached a stage in its evolution where an overall pattern is beginning to emerge. A whole spectrum of types of aircraft will be required, and we shall see later that this requirement can probably be met and that we can now define mayor types of aircraft to provide such a global network of transport operations ‘.

1.4 The design problem. Design work is the ultimate purpose of aerodynamics and all other activities should lead up to it. The design of an actual air­craft provides the final and most severe test of hypotheses, concepts, and methods. In view of this, it is important that workers in aerodynamics should give some thought to the question of whether or not their own individual pieces of work are well-aimed towards application in design; they should also have some notion of the design strategies at their disposal. The design pro­blem may be approached in several different ways. The approach adopted here follows E C Maskell (1961), J A Bagley (1961), and D Ktlchemann (1968) and will be set out in more detail below. Before we do that, we should explain briefly why we do not want to make use of other possible design strategies.

In some sense, Nature is faced with the same kind of problem and the method by which she solves it may be described as a process of natural evolution: a Darwinian empirical approach. Progress in Nature proceeds by adaptation and evolution. Adaptation is the adjustment of populations to their environment by the operation of natural selection. Evolution is the observable result of adaptation at different points in time and space. Changes are brought about by the processes of mutation and recombination. Mutation is an incoherent if not random process which provides the novel changes in genes and chromosomes. Thus truly new genetic variations arise only by mutations. Recombination, by far more frequent in occurrence, provides new individual variations within populations, but these are variations limited to a range set by a pre-existing genetic theme. Nature also seems to know what the design criteria and aims are: it is simply the survival of the species. It selects the "fittest" and operates in this way without mercy.

It might be argued that this process could be imitated in engineering design. In principle, even random changes of an existing design might be admitted and their usefulness investigated. It might also be argued that, if only we in­vestigated possible changes systematically and thus covered the field fully,

*) It is sometimes argued that telecommunications will, in future, satisfy most of these needs. While it cannot be accepted that they will ever eliminate the need for personal contact, we must hope that they will be developed to the full because it may well turn out that, otherwise, the actual means of transport could not possibly cope with the demand, once the world population becomes mobile.

we should come upon some possible variations which constitute some advance on the existing design. In this way, we could proceed in small steps by small improvements on "pre-existing genetic themes" in a Darwinian empirical manner, governed by natural selection. Occasionally, an "inventor" would provide us with a random "mutation" and a new theme. Such a strategy, which should lead to the optimisation of technical systems according to the principles of bio­logical evolution, has been described and advocated by I Rechenberg (1973).

A somewhat related approach is to procede by investigating "systematic series" of geometric shapes. We know of many such investigations, – of aerofoil sec­tions, of wing-fuselage combinations – resulting in catalogues out of which the designer was expected to pick out what he needed. Nowadays, one could pro­ceed experimentally by statistically-designed tests in windtunnels and theor­etically by the application of mathematical tools in the form of multivariate analysis, using computers. It is often said that this process, too, leads to "aircraft design optimisation".

It is suggested here that these processes are unrealistic, unsuitable, and wasteful (see e. g. D Kllchemann (1974)). In investigations of "systematic series", one would always have doubts on whether one had hit upon the right series in the first place and whether one had really found and covered all the relevant parameters systematically, assuming there existed such parameters in the first place. Besides, one would not know how to find the aerodynamic para­meters without having a model of the flow and of the aircraft in mind. Further, what guarantee does one have that, when the work is done, one does not come up with a whole series of duds and not a single useful answer? Is it good enough merely to show that this particular set does not get us anywhere? What deters one most is the intrinsic wastefulness of this process. Nature can proceed in this way because her resources and time appear to be unlimited. But evidently we are not in this happy position. When we consider the resources which we need for our work and also the responsibility we take on when we ask society to provide them, we cannot really take such enormous chances on whether any­thing will come out of our work or not.

It is now becoming clear that it is also mistaken to assume that computers could produce optimum designs in an empirical manner: it cannot be carried out in practice. What can be done is the application of numerical methods for locating the constrained minimum of a function of several variables to the problem of choosing values of the parameters in a mathematical model of a hypo­thetical aircraft so as to give the best design according to a given criterion (see e. g. BAM Figgott & В E Taylor (1971)). This implies that we must know in advance what is a reasonable model of the aircraft we want to consider. It also implies that we can make in advance a reasonable choice of all the con­straints which are physically realistic. Again, it implies that the para­meters to be chosen have some physical significance. Altogether then, such work can only be done in a meaningful and realistic manner if a conceptual framework for the type of aircraft to be examined already exists.

We must, therefore, look for an alternative design procedure which leads to conceptual frameworks, where we can state our aims beforehand and then pursue them in a rational manner and at least with a reasonable hope of success. In aerodynamic design, we want to suggest and explain in detail later, that a reasonably safe way to good and practical designs is to start on the basis of fluid mechanics and to select types of floti which appear to be suitable for engineering purposes and might be used with some confidence. This then leads to corresponding types of aircraft and frameworks of design concepts and methods for each of them. We cannot and do not claim that this will neces­sarily lead to an "optimum" and that no risks are involved. But the sound basis of fluid mechanics and engineering should lead to good and practical designs. This has been proven many times throughout the history of aviation.

Possibly the strongest argument against the use of any other design strategy is that it is very hard to imagine how the actual types of aircraft, which we have, with their controls and means for generating lift and propulsion, could have come out of a computer in an evolutionary manner. The shapes which result from considerations of desirable flows are really very odd indeed in the sense that the chances of arriving at them from purely geometric considerations must be regarded as very remote. The oddest of them all would seem to be that of the classical aircraft, but here it has helped that this class of shape had a counterpart in Nature, in the shapes of birds and insects. These have always intrigued observant men, but real progress was made only when Cayley introduced radical abstractions of what he saw and adapted these to human engineering, and when scientists like Lilienthal and the Wright Brothers recognised the nature of the type of flow involved and then proceeded to design their aircraft to exploit this type of flow. The concepts underlying classical aerofoils are really more complex and more difficult to understand than those of the other basic types of flow and types of aircraft, and this is probably one of the reasons why it took man so long to learn how to fly (see e. g. E von Holst &

D KUchemann (1941)).

Our approach has some important repercussions on the research needed to pro­vide the foundations before actual design work can begin. Much work needs to be done in the field of fluid mechanics on finding out about types of flow and their potential suitability for engineering applications. It is the kind of work that Lichtenberg wanted, and it may be called aimed research. The scientists who do the work must then be perceptive and imaginative, and they must have a clear idea in mind which way they are going. New findings are then not a random process but are guided by aims and conjectures. The terms "pure" research and "applied" research are then not appropriate and become rather meaningless as far as aerodynamic research is concerned.

This approach has also repercussions on the toots needed to do the work and on how to use them. On the theoretical side, we need mathematical models of the flow, in which all the essential features of the flow are recognised and represented. It is not very useful to have answers of great numerical accur­acy from a computer, say, for a flow model which is not adequate. It is not good enough to use a mathematical model which indiscriminately represents the shape of a body, for example, and the flow in an egalitarian and undifferen­tiated fashion. Any useful flow model must have built into it all the indivi­dual characteristic features which distinguish the particular flow from others. Thus the essentials of the behaviour of the flow in significant regions such as leading edges, near separation lines, near planform kinks, and near junc­tions, must be thought out carefully beforehand and fed into the flow model and into the computer program, if one is to be used. It is then that com­puters and multivariate analysis can be really useful. Examples of this kind of work may be found in D H Perry (1970), D H Peckham (1971) , D L I Kirkpatrick (1974) and J Collingbourne (Dll Kirkpatrick (1974). Most valuable for practical design purposes are those methods which bring out clearly the physi­cal concepts and provide conceptual frameworks which can guide the designer towards the realisation of those characteristics which he wants his aircraft to have. Above all, conceptual frameworks which are firmly based on physics allow the designer to practice the art of the soluble (see P В Medawar (1967)) and prevent him from being deluded into chasing phantoms which cannot be realised in practice.

On the experimental side, aerodynamic research and design is characterised by the extensive use of model testing, probably more than in any other branch of science and technology (see e. g. D KUchemann (1964), J Zierep (1971). Simi­larity laws and nondimensional parameters and scaling functions are exploited to the full, and windtunnels for model testing are the main tools. Again, it is of overriding importance to represent in such model tests all the signifi­cant individual features of the type of flow to be investigated. Windtunnels and testing techniques must be designed to suit this purpose if they are to give meaningful and useful results. To recognise the significant features and to find out what these are is partly a matter for experiments in the mind, and this is where conceptual frameworks can again help in the design, carrying out, and analysis of meaningful and crucial experiments. Thus theory and experiment must go together in aerodynamics, and there is little room for the pure and isolated mathematician or for the pure and one-sided experimentalist. To put on blinkers one way or the other will not do. But, as we shall see, theoretical aerodynamics is also exceedingly difficult and complex, and this is why aerodynamics is still largely an experimental science.

In view of this, it is important to have a good understanding of experimental techniques. A description of these goes beyond the scope of this book. They have been discussed in some detail in AGARD in recent years, and much informa­tion may be found in AGARD publications (Conference Proceedings CP-83 and 174, Advisory Reports AR-60, 68, 70, 83, Reports R-600, 601, 602, and R C Pankhurst (1974)).

When all these tools are available and properly used, the main task that remains is to establish enough confidence to believe that, for the type of aircraft and mission under consideration, there exist regions of no conflict between the various essential characteristics, within which a set of design requirements can be met naturally. What we are really seeking is probably that "harmony" between elements, which some see in the motions of the planets in the heavens since the ancient Greeks, and which some see in the Darwinian model of biological evolution (see e. g. D G King-Hele (1971)). So we are not out for a "compromise" in the sense that we can achieve some desirable charac­teristic only by degrading another and where a "deal" is made at somebody else’s expense. We shall endeavour to explain what is meant by this by giving examples of good design concepts. On the other hand such a "good design" is not likely to be one where the overall result is an "optimum" with regard to any single parameter at just one design point. Instead, all the significant parameters are in harmony and not in conflict for a set of design points and off-design conditions, and the final solution is sound and healthy. We are not interested in pathological flows and aircraft. It was Prandtl who intro­duced the concept of healthy flows, and we are well-advised to follow him and to search for sound and healthy engineering solutions when designing aircraft and to avoid "sick" and "lousy" flows which cannot be relied upon.

We hope to show later that such sound and healthy engineering types of flow and types of aircraft do indeed exist. In fact, the most important develop­ment during the past two decades or so has probably been the realisation that there is more than one such major type of flow and aircraft; and also the knowledge that matters work out well if all the design elements "click" and fit together and if a design stays firmly within the bounds of sound physical design concepts.