VOL.7,NO.1,JAN.-FEB.1970 J.AIRCRAFT 3 32nd Wright Brothers Lecture Supersonic Air Transport-True Problems and Misconceptions P.SATRE Sud Aviation,Paris,France The transition from subsonie to supersonic transports will be the last opportunity of achiev- ing major time savings on long range fights for some time to come,as hypersonie transport will probably not be flying before a few decades.The state of the art has progressed far enough to enable designers to match supersonic airplane stage fuel and reserves requirements with acceptable operating weight empty and economic payload.A supersonic transport design program faces several major technical problems.Design features and solutions incorporated in the CONCORDE are discussed,in particular,the aerodynamic design compromise between high-speed and low-speed requirements.Kinetie heat problems and choice of materials are also reviewed.As far as reliability and safety are concerned,in trying to move forward,a more rational approach has been used as compared to subsonic airplanes, especially when it comes to certification regulations.Finally,in the area of operations,prob- lem matters are now well defined and the outlook is optimistic.Supersonic transports will cause new problems but they also have inherent advantages.In this area,certain problems have been somewhat exaggerated.On the whole,the supersonic transport looks promising. Introduction problems specific to the SST.First of all,however,it might be appropriate to consider these problems in relation to the HIS 14th of July is a double celebration as I have been invited by the AIAA to deliver the Wright Brothers general evolution of aeronauties or perhaps I should say of air transport. Memorial Lecture on the very day of our Bastille day.For One of the most practical yardsticks for measuring this me,this invitation is at once a great honor and a great plea- progress is the decrease in Direct Operating Costs (DOC). sure. I will refrain from going into the finer points of DOC defini- What tremendous changes have been wrought since 1903. tions.for we all know what is involved in broad terms.How, when Orville Wright made the first controlled flight!Aero- then,have the engineers managed to sbrink DOC's over the nautics,then a realm surrounded with an aura of mystery in years?They had two ways of doing this:to increase which moved only pioneers like the Wright Brothers,has capacity-and with it the gross weight-and to increase since spawned two of the world's most flourishing industries: flight speeds.In fact,we find that they have consistently the aviation industry and air transport.And for something applied both approaches concurrently. over ten years now,it has an equally prosperous junior: Figure I shows the evolution in transport aireraft speeds. astronautics. The gain in speed brings an attendant reduction in DOC's The intervening few decades have been marked with ad- as technological advances gradually allow this gain to be vances now taken for granted but which a little thought will show to be truly extraordinary.I do not propose to achieved at not too great a cost. This condition has indeed been fulfilled and we see from the relate these at length,but rather to talk to you about what I curve that flight speeds have steadily increased while,at the have been concerned with daily these last few years,namely same time,as shown in Fig.2,DOC's have tapered off- though how much of this is due to greater speed and how much to greater capacity is not readily apparent.(In B.2707 Fig.2,for comparison,DOC are re-estimated taking into account U.S.consumer prices variation). 1500 It would be a mistake to infer that the same trend in Concorde speed can be maintained for long.In fact,I am convinced Tu.144 that it cannot,and that the switch to supersonic transports 100 500 747 8 Dc 7c 8707120 60 0C8-30 940 195019601.97019801990 870x320 0c8:63877 Fig.I Cruise speed increase. Presented as Paper 69-759 at the AIAA Aircraft Design and 1950 1955 19601,9651970 Operations Meeting,July 14-16,1969,Los Angeles,Calif.;sub- mitted August 29,1969;revision received October 2,1969. Fig.2 Direct operating cost trend
VOL. 7, NO. 1, JAN.-FEB. 1970 J. AIRCRAFT 32nd Wright Brothers Lecture Supersonic Air Transport—True Problems and Misconceptions P. SATRE Sud Aviation, Paris, France The transition from subsonic to supersonic transports will be the last opportunity of achieving major time savings on long range flights for some time to come, as hypersonic transport will probably not be flying before a few decades. The state of the art has progressed far enough to enable designers to match supersonic airplane stage fuel and reserves requirements with acceptable operating weight empty and economic payload. A supersonic transport design program faces several major technical problems. Design features and solutions incorporated in the CONCORDE are discussed, in particular, the aerodynamic design compromise between high-speed and low-speed requirements. Kinetic heat problems and choice of materials are also reviewed. As far as reliability and safety are concerned, in trying to move forward, a more rational approach has been used as compared to subsonic airplanes, especially when it comes to certification regulations. Finally, in the area of operations, problem matters are now well defined and the outlook is optimistic. Supersonic transports will cause new problems but they also have inherent advantages. In this area, certain problems have been somewhat exaggerated. On the whole, the supersonic transport looks promising. Introduction T HIS 14th of July is a double celebration as I have been invited by the AIAA to deliver the Wright Brothers Memorial Lecture on the very day of our Bastille day. For me, this invitation is at once a great honor and a great pleasure. |fc What tremendous changes have been wrought since 1903, when Orville Wright made the first controlled flight! Aeronautics, then a realm surrounded with an aura of mystery in which moved only pioneers like the Wright Brothers, has since spawned two of the world's most flourishing industries: the aviation industry and air transport. And for something over ten years now, it has an equally prosperous junior: astronautics. The intervening few decades have been marked with advances now taken for granted but which a little thought will show to be truly extraordinary. I do not propose to relate these at length, but rather to talk to you about what I have been concerned with daily these last few years, namely Speed M.pH. 1,000 B-2707 ^mmm * Concorde Tu.144 1940 1950 1.960 1.970 1,980 Fig. 1 Cruise speed increase. 1,990 Presented as Paper 69-759 at the AIAA Aircraft Design and Operations Meeting, July 14-16, 1969, Los Angeles, Calif.; submitted August 29,1969; revision received October 2,1969. problems specific to the SST. First of all, however, it might be appropriate to consider these problems in relation to the general evolution of aeronautics or perhaps I should say of air transport. One of the most practical yardsticks for measuring this progress is the decrease in Direct Operating Costs (DOC). I will refrain from going into the finer points of DOC definitions, for we all know what is involved in broad terms. How, then, have the engineers managed to shrink DOC's over the years? They had two ways of doing this: to increase capacity—and with it the gross weight—and to increase flight speeds. In fact, we find that they have consistently applied both approaches concurrently. Figure 1 shows the evolution in transport aircraft speeds. The gain in speed brings an attendant reduction in DOC's as technological advances gradually allow this gain to be achieved at not too great a cost. This condition has indeed been fulfilled and we see from the curve that flight speeds have steadily increased while, at the same time, as shown in Fig. 2, DOC's have tapered off— though how much of this is due to greater speed and how much to greater capacity is not readily apparent. (In Fig. 2, for comparison, DOC are re-estimated taking into account U.S. consumer prices variation). It would be a mistake to infer that the same trend in speed can be maintained for long. In fact, I am convinced that it cannot, and that the switch to supersonic transports 1 60^5——— ———— \ 8-63 B 747 1,950 1955 1.960 1.965 1,970 Fig. 2 Direct operating cost trend
P.SATRE J.AIRCRAFT Table 1 Subsonic vs supersonic weight breakdown comparison Subsonic Pistons Subsonie Supersonic Fuel,including reserves, 38-40 47-49 Operating weight empty, 47-48 44-45 : Payload, 12-15 6-9 Subsonic Jets and the figure claims to be no more than a mere forecast for the 21st century Supersonic Jets It is still fairly easy to forecast flight speeds,but gross weight prediction is not so simple.Figure 5 shows their Hypersonic 01， history.In point of fact,any extrapolation must allow for 0 traffic trends and,as you know,much can depend on whether 0 6 Mach number the growth rate is taken as 8%or 15%(which are for all Fig.3 Transatlantie stage;block time vs cruise speed practical purposes the limit figures quoted in the forecasts). Summing up,then,the way seems fairly clear:increase the flight speed in step with technological progress and will prove to be the last big step forward in this respect for increase the gross weight as and when such inerease is war- ranted or demanded by traffic growth.These improvements many years to come. While we are about it,let us dispose of one misconception will be automatically reflected in the DOC figure.Two As shown in Fig.3,aireraft in the Mach 2 and Mach 2.7 things remain paramount,however,in any such evolution, namely:1)to improve safety and reliability and 2)to categories belong to the same family;the saving in time due take operational requirements into account,which is not to the increase in speed is 334 hr with the currently envi- sioned SST's,and these aircraft are no more different from to say that they can remain static and I say so most em- each other than two subsonic aircraft Aying at Mach 0.75 phatically. Given this fairly clear pattern,how do SST's shape up on and Mach 0.90,respectively.Only,by achieving Mach 5 or the eve of their entry into service?What are their true 6,may we expect to pass yet another truly significant mile- problems?And what are the misconceptions being en- stone.And to establish that this would be the ultimate tertained about them? stage,we need only remember that the average passenger must not be subjected to accelerations in excess of g,which sets a lower limit of approximately 1 hr for the transatlantic Performance crossing. Let us bear in mind that it is no doubt perfectly reasonable We might as well admit right away that,like all their to make a Mach 2-2.2 aircraft if it is built now,using light forerunners,SSTs will not give their best performance as alloy,or a Mach 2.5-2.7 aircraft if it is built a little later, soon as they go into service.Like all other aireraft,they using titanium.Yet the missions devolving upon these will begin in a modest way and improve with time.As you aireraft do not differ fundamentally.The stage times are were able to see from Fig.2 just now,the 707 had a DOC close to that of the DC 7.It has since diminished by 30%. about the same,and above all the thermal balance is con- trolled in the same way:by using fuel-conventional The pattern of evolution in SSTs can be expected to be fuel-as the heat sink,with no further precautions.This comparable,as we shall see presently. no longer applies in the case of a hypersonic aircraft,which How will the first SSTs compare,performancewise,with the current long-haul subsonic jets?Table I gives you some would have to use a different fuel or resort to some other idea of the payload vs takeoff weight eapability. cooling device,quite apart from the acceleration problem I just mentioned.In short,from Mach 3 upwards,the The figures given are no more than orders of magnitude difficulties add up very fast while the returns diminish.In but they speak for themselves:fuel consumed plus reserves fact,it appears that hypersonic aircraft will not be worthwhile are up by some 9%and this must be made good by a reduction until the day they can be justified by a sufficient traffic in empty weight and/or payload.But the payload cannot growth over stage lengths of 5000-10,000 miles.This would be allowed to drop below the indicated figures,for notwith- standing faster turnrounds and a greater potential (CON- produce the breakdown shown in Fig.4 (on which the log- CORDE,for example,is designed for some 14,000 trans- arithmic scale enables the respeetive traffic magnitudes to be more clearly portrayed).However,we are not there yet, atlantie flights as against 7000 for a subsonie aireraft),prof- itability would otherwise be impossible. We had therefore to gain 3%on the empty weight.You who are familiar with the problems involved in reducing AIR TRAFFIC aireraft empty weights,can appreciate the magnitude of this achievement. It must be realized that designing a supersonic airplane isn't quite like building a powered buggy. You all remember what the first automobiles looked like- -a motor mounted SUBSONIC JETS SUPERSONIC JETS Table 2 DOC breakdown Fuel Approx.30% Man-hours Crew HYPERSONIC Maintenance Approx.20% Depreciation Items affected by Insurance 100200 30010002000 50o05i688 aireraft and Financing Approx.50 Fig.4 Air traffic share breakdown including hypersonic spare prices Spares Tooling aireraft
P. SATRE J. AIRCRAFT Block Time . Hours I Table 1 Subsonic vs supersonic weight breakdown comparison Subsonic Pistons Subsonic Jets Supersonic Jets Hypersonic ^-LL'UL/' 0 12345 6 Mach number Fig. 3 Transatlantic stage; block time vs cruise speed. will prove to be the last big step forward in this respect for many years to come. While we are about it, let us dispose of one misconception. As shown in Fig. 3, aircraft in the Mach 2 and Mach 2.7 categories belong to the same family; the saving in time due to the increase in speed is 3^-4 hr with the currently envisioned SST's, and these aircraft are no more different from each other than two subsonic aircraft flying at Mach 0.75 and Mach 0.90, respectively. Only, by achieving Mach 5 or 6, may we expect to pass yet another truly significant milestone. And to establish that this would be the ultimate stage, we need only remember that the average passenger must not be subjected to accelerations in excess of ^g, which sets a lower limit of approximately 1 hr for the transatlantic crossing. Let us bear in mind that it is no doubt perfectly reasonable to make a Mach 2-2.2 aircraft if it is built now, using light alloy, or a Mach 2.5-2.7 aircraft if it is built a little later, using titanium. Yet the missions devolving upon these aircraft do not differ fundamentally. The stage times are about the same, and above all the thermal balance is controlled in the same way: by using fuel—conventional fuel—as the heat sink, with no further precautions. This no longer applies in the case of a hypersonic aircraft, which would have to use a different fuel or resort to some other cooling device, quite apart from the acceleration problem I just mentioned. In short, from Mach 3 upwards, the difficulties add up very fast while the returns diminish. In fact, it appears that hypersonic aircraft will not be worthwhile until the day they can be justified by a sufficient traffic growth over stage lengths of 5000-10,000 miles. This would produce the breakdown shown in Fig. 4 (on which the logarithmic scale enables the respective traffic magnitudes to be more clearly portrayed). However, we are not there yet, 200 Stage Length N.M 500O 100OO Subsonic Supersonic Fuel, including reserves, % Operating weight empty, % Payload, % 38-40 47-48 12-15 47-49 44-45 6-9 and the figure claims to be no more than a mere forecast for the 21st century. It is still fairly easy to forecast flight speeds, but gross weight prediction is not so simple. Figure 5 shows their history. In point of fact, any extrapolation must allow for traffic trends and, as you know, much can depend on whether the growth rate is taken as 8% or 15% (which are for all practical purposes the limit figures quoted in the forecasts). Summing up, then, the way seems fairly clear: increase the flight speed in step with technological progress and increase the gross weight as and when such increase is warranted or demanded by traffic growth. These improvements will be automatically reflected in the DOC figure. Two things remain paramount, however, in any such evolution, namely: 1) to improve safety and reliability and 2) to take operational requirements into account, which is not to say that they can remain static and I say so most emphatically. Given this fairly clear pattern, how do SST's shape up on the eve of their entry into service? What are their true problems? And what are the misconceptions being entertained about them? Performance We might as w r ell admit right away that, like all their forerunners, SSTs will not give their best performance as soon as they go into service. Like all other aircraft, they will begin in a modest way and improve with time. As you were able to see from Fig. 2 just now, the 707 had a DOC close to that of the DC 7. It has since diminished by 30%. The pattern of evolution in SSTs can be expected to be comparable, as we shall see presently. How will the first SSTs compare, performancewise, with the current long-haul subsonic jets? Table 1 gives you some idea of the payload vs takeoff weight capability. The figures given are no more than orders of magnitude but they speak for themselves: fuel consumed plus reserves are up by some 9% and this must be made good by a reduction in empty weight and/or payload. But the payload cannot be allowed to drop below the indicated figures, for notwithstanding faster turnrounds and a greater potential (CONCORDE, for example, is designed for some 14,000 transatlantic flights as against 7000 for a subsonic aircraft), profitability would otherwise be impossible. We had therefore to gain 3% on the empty weight. You who are familiar with the problems involved in reducing aircraft empty weights, can appreciate the magnitude of this achievement. It must be realized that designing a supersonic airplane isn't quite like building a powered buggy. You all remember what the first automobiles looked like—a motor mounted Table 2 DOC breakdown Fig. 4 Air traffic share breakdown including hypersonic aircraft. Fuel Man-hours Items affected by aircraft and spare prices Crew Maintenance Depreciation Insurance Financing Spares Tooling Approx. 30% Approx. 20% Approx. 50%
JAN.-FEB.1970 SUPERSONIC AIR TRANSPORT Table 3 Safety reserves according to TSS-OPS 5.7 M668w 1 Standard mission plus: 001 8747 Missed approach Climb,diversion and 30 min alternate hold at 15,000 ft 600 Enroute reserves (to be determined for each route;might be 3%of block fuel on North Atlantic) 500 2 One-engine failure: L1011 Destination or alternate must be reached with fuel avail- 400 DC 10 able for 30 min hold at 15,000 ft ·CONCORDE 3 Two-engine failure or pressurisation failure: 300 8707320 Any airport appropriate for landing must be reached 8707.1200 No holding 200 100 8314pC4 DC6B on a buggy chassis.Our job is altogether different.Building a supersonic airplane means not only a change in engines and 1940 1950 1960 1970 1980 geometry but also a shift in the state of the art. Indeed, the compromise must be a little more advanced in all respects Fig.5 Maximum takeoff weight trend. than on subsonic aircraft:refinements in design to save from 7.5-10.35%of the takeoff weight.A simple calculation weight,and more automation to alleviate the crew's work- shows that the DOC then drops from 1.3 to 0.94. load since the same functions must be performed in a shorter Assuming the same gains,the subsonic transport's time.Yet thesc advances must be made without com- DOC drops from I down to only 0.83.Now subsonic aireraft promising safety.On the contrary,parallel efforts are are not going to make any weight savings without a corre- made to increase it.Many of the solutions adopted stom sponding increase in the buying price,and,as for fuel con- from these three joint requirements. sumption,there has been far more time to experiment,so that But it so happens that we nourish great hopes from the there is certainly less latitude.Ultimately,as SST technology very hurdles which we find we must clear from the perfor- becomes more commonplace,so the supersonie transport's mance standpoint.For the fact that the weight of fuel is DOC will tend toward that of its subsonic counterpart. 5-8 times the payload also means that 1 saved on fuel DOC will tend toward that of its subsonie counterpart.And consumption means a 5-8%gain in payload. as I was just saying,it is those present narrow margins of ours In other words,the improvement in propulsion efficieney precisely,which give us broad seope for the future. will be far more effective than on subsonic aircraft.On I would like to revert for a moment to the question of the whole,we think that there is every reason to believe that reserves.The over-all design does not depend on them a SSTs will offer a greater development potential than subsonic great deal,but the way the plane is to be operated does to a aireraft. great extent.The quantities of fuel quoted previously This can be established cursorily by reference to the fig- inciude reserves amounting to about 9%of the takeoff ures.Our current estimates are that,given substantially weight.This is a maximum and should be compared with the same weight,the SST will have a DOC of 1.3,if we take the FAA's projected figure of about 8.5%and with 7%or so that of the subsonie transport as 1. of the Anglo-French regulations (TSS-OPS 5.7). To see whether it can be dropped to 1 or even below,let Table 3 provides a very concise summary of the TSS-OPS, us take a look at Table 2,which gives a D0C breakdown per from which you ean see that the requirement provides ample types of cost.The purchase price of the SST may come safety,which,of course,is as it should be. down:the techniques involved,which seem very sophis- In fact even the figure of 7%is arguable,for it hardly ticated at present,will become more conventional,hence seems right to apply worldwide a rule that was formulated less costly.A 10%saving on the buying price would mean primarily with New York and a few other major airports in a 5%saving on the DOC. mind.In fact,TSS-OPS 5.7 does provide for a number of But it is the weight breakdown which will provide the special cases. most signifieant gain.Let us now turn to Fig.6 which Let us nevertheless assume a figure of 7%.This means reproduees the weight breakdown in Table 1. that,under the TSS-OPS 5.7 regulation,airlines are left Assuming coustant prices,highly likely gains of 1%on the with 2%of the takeoff weight with which to optimize the OWE and 5%on fuel consumption would raise the payload flight regularity-payload compromise as they see fit.We be- Pierre Satrc Pierre Satre,born May 4,1909 at Grenoble,France,was educated in Marseille.He is a graduate of Ecole Polytechniqne (Class of 1929)and Ecole Nationale Superieure de l'Acronautique (Class of 1934).He started his career as an Aeronautics Engineer in varioux ministerial offices. In March 1941,he was appointed Chief Engineer at SNCASE-Toulouse (later to be known as Sud Aviation).In this capacity,he was in charge of a number of military and commercial aircraft,among them the Armagnae,Grognard,and Durandal,a Mach-2 fighter plane,and finally the well known Caravelle with rear engines. Appointed Technical Director of Sud Aviation in 1959,he is,specifically,Technical Director of the Concorde projeet. Mr.Satre is an officer of the Legion d'Honneur and Commander of the National Order of Merit and a member of AIAA:he has been awarded the British Silver Medal,as well as numerous other French and foreign medals. He is married and has five children
JAN.-FEB. 1970 SUPERSONIC AIR TRANSPORT Table 3 Safety reserves according to TSS-OPS 5.7 . Standard mission plus: Missed approach Climb, diversion and 30 min alternate hold at 15,000 ft Enroute reserves (to be determined for each route; might be 3% of block fuel on North Atlantic) I One-engine failure: Destination or alternate must be reached with fuel available for 30 min hold at 15,000 ft > Two-engine failure or pressurization failure: Any airport appropriate for landing must be reached No holding on a buggy chassis. Our job is altogether different. Building a supersonic airplane means not only a change in engines and geometry but also a shift in the state of the art. Indeed, the compromise must be a little more advanced in all respects than on subsonic aircraft: refinements in design to save weight, and more automation to alleviate the crew's workload since the same functions must be performed in a shorter time. Yet these advances must be made without compromising safety. On the contrary, parallel efforts are made to increase it. Many of the solutions adopted stem from these three joint requirements. But it so happens that we nourish great hopes from the very hurdles which we find we must clear from the performance standpoint. For the fact that the weight of fuel is 5-8 times the payload also means that 1% saved on fuel consumption means a 5-8% gain in payload. In other words, the improvement in propulsion efficiency will be far more effective than on subsonic aircraft. On the whole, we think that there is every reason to believe that SSTs will offer a greater development potential than subsonic aircraft. This can be established cursorily by reference to the figures. Our current estimates are that, given substantially the same weight, the SST will have a DOC of 1.3, if we take that of the subsonic transport as 1. To see whether it can be dropped to 1 or even below, let us take a look at Table 2, which gives a DOC breakdown per types of cost. The purchase price of the SST may come down: the techniques involved, which seem very sophisticated at present, will become more conventional, hence less costly. A 10% saving on the buying price would mean a 5% saving on the DOC. But it is the weight breakdown which will provide the most significant gain. Let us now turn to Fig. 6 which reproduces the weight breakdown in Table 1. Assuming constant prices, highly likely gains of 1% on the OWE and 5% on fuel consumption would raise the payload L 1011 m DC 1 * CONCORDE Fig. 5 Maximum takeolf weight trend. from 7.5-10.35% of the takeoff weight. A simple calculation shows that the DOC then drops from 1.3 to 0.94. Assuming the same gains, the subsonic transport's DOC drops from 1 down to only 0.83. Now subsonic aircraft are not going to make any weight savings without a corresponding increase in the buying price, and, as for fuel consumption, there has been far more time to experiment, so that there is certainly less latitude. Ultimately, as SST technology becomes more commonplace, so the supersonic transport's DOC will tend toward that of its subsonic counterpart. DOC will tend toward that of its subsonic counterpart. And as I was just saying, it is those present narrow margins of ours, precisely, which give us broad scope for the future. I would like to revert for a moment to the question of reserves. The over-all design does not depend on them a great deal, but the way the plane is to be operated does to a great extent. The quantities of fuel quoted previously include reserves amounting to about 9% of the takeoff weight. This is a maximum and should be compared with the FAA's projected figure of about 8.5% and with 7% or so of the Anglo-French regulations (TSS-OPS 5.7). Table 3 provides a very concise summary of the TSS-OPS, from which you can see that the requirement provides ample safety, which, of course, is as it should be. In fact even the figure of 7% is arguable, for it hardly seems right to apply worldwide a rule that was formulated primarily with New York arid a few other major airports in mind. In fact, TSS-OPS 5.7 does provide for a number of special cases. Let us nevertheless assume a figure of 7%. This means that, under the TSS-OPS 5.7 regulation, airlines are left with 2% of the takeoff weight with which to optimize the flight regularity-pay load compromise as they see fit. We bePierre Satre Pierre Satre, born May 4, 1909 at Grenoble, France, was educated in Marseille. He is a graduate of Ecole Poly technique (Class of 1929) and Ecole Nationale Superieure de I'Aeronautique (Class of 1934). He started his career as an Aeronautics Engineer in various ministerial offices. In March 1941, he was appointed Chief Engineer at SNCASE-Toulouse (later to be known as Sud Aviation). In this capacity, he was in charge of a number of military and commercial aircraft, among them the Armagnac, Grognard, and Durandal, a Mach-2 fighter plane, and finally the well known Caravelle with rear engines. Appointed Technical Director of Sud Aviation in 1959, he is, specifically, Technical Director of the Concorde project. Mr. Satre is an officer of the Legion d'Honneur and Commander of the National Order of Merit and a member of AIAA; he has been awarded the British Silver Medal, as well as numerous other French and foreign medals. He is married and has five children
P.SATRE J.AIRCRAFT 00C VORTICES (d] 10 CL PREDICTED +38 IN FLIGHT.BETWEEN M O AND M.0.8 Fig.6 Anticipated improvements. lieve this would be by far preferable to imposing statutory reserves not strictly necessary for safety,for,like the empty weight,the fuel weight must be determined even more pre- cisely than on subsonic aireraft beeause of its economic (d implications. 10 15 OF ATAC In short,the performance problem with SSTs is a genuine one but let us not add to it through excessive conservation Fig.8 Concorde lift curve and lift increase due to vortices. on reserves based on a misconception.Presently,when I go on to examine operating problems,I shall show why ean be reduced.3)It must possess excellent flying qual- ities throughout the flight envelope. it is no exaggeration to speak of excessive conservation, especially when reserves equal to or greater than those on Now although the third requirement can be met by making subsonic aireraft are being contemplated even through the the necessary refinements,the first two are in direct conflict. contingencies likely to be met on the journey are far fewer. The subsonic regime is of considerable importance in the And while we are on the subjeet of performance,let me turn over-all economies of the SST,as portrayed diagramatically right away to another misconception. in Table 5. The takeoff and landing speeds associated with delta You will observe that the requirements imposed for holdings wing aircraft are higher than with other aircraft.Some and diversions have a determining effect.This being so,the have inferred from this that landings in particular would attractions of variable geometry as a means of resolving present difficulties.At no time was this view shared by this contradiction are manifest,and I am not surprised that the chief pilots of our customer Airlines,who from the outset Boeing attempted a breakthrough along these lines.Their has posed the problem in stark and simple fashion:speed design study,the only one to have been taken far enough. is important,but less important than flving qualities.The was necessary to establish that the variable geometry solution CONCORDE's flight tests have shown how right they were: is not yet ripe.This is not to say that we shall not see back thanks to excellent flying qualities,landings at 160 kt can with us some day.But so far the 3 SST design projects be controlled without any trouble,and you will find confir- feature near-fixed geometry,I say "near"because the droop mation of this in Table 4. nose and the air intakes do feature variable geometry,which This leads me straight on to the origin of these relatively makes my task simpler.I shall simplify it still further by high takeoff and landing speeds,namely the aerodynamic confining myself to what I know well:the solutions adopted for the CONCORDE. compromise between high and low speeds Starting with a pure delta wing with 63.5 of sweepback, we set about looking for improvemnents,with special emphasis High-Speed/Low-Speed Aerodynamic on low-speed qualitics. Compromise A sharply swept apex (about 76),as shown in Fig.7, produced a triple advantage:1)a reduction in the thick- There are at least three requirements to be met in de- ness ratio at the wing root,plus an arrow planform,both signing an SST:1)The airplane must be configured for favorable factors for supersonic flight,2)a forward supersonic cruise fight.2)It must adapt readily to shift in aerodynamic center location,which in turn shifted subsonie fight,notably for takeoff,for landing,and also, for holdings prior to lauding until such time these holdings Table 4 Concorde 001--first fandings Touch down WING TIP:55 Approach Air Vertical Bank No.of speed, speed, speed, angle BASIC DELTA:63,5 fight kt kt m/sec 167 T60 1 0.9 APEX:76 2 171 165 0. 0.6 3 175 165 0.6 0.7 4 171 166 0.4 5 179 171 0.2 6 175 16 0.5 0 7 176 0.s 0 174 16S 0.8 0.4 165 10 0.2 Mean values 172 165 0.7 0.4 Fig.7 Wing plan form. Cinetheodolite failure
P. SATRE J. AIRCRAFT 100 CL Fig. 6 Anticipated improvements. lieve this would be by far preferable to imposing statutory reserves not strictly necessary for safety, for, like the empty weight, the fuel weight must be determined even more precisely than on subsonic aircraft because of its economic implications. In short, the performance problem with SSTs is a genuine one but let us not add to it through excessive conservation on reserves based on a misconception. Presently, when I go on to examine operating problems, I shall show why it is no exaggeration to speak of excessive conservation, especially when reserves equal to or greater than those on subsonic aircraft are being contemplated even through the contingencies likely to be met on the journey are far fewer. And while we are on the subject of performance, let me turn right away to another misconception. The takeoff and landing speeds associated with delta wing aircraft are higher than with other aircraft. Some have inferred from this that landings in particular would present difficulties. At no time was this view shared by the chief pilots of our customer Airlines, who from the outset has posed the problem in stark and simple fashion: speed is important, but less important than flying qualities. The CONCORDE'S flight tests have shown how right they were: thanks to excellent flying qualities, landings at 160 kt can be controlled without any trouble; and you will find confirmation of this in Table 4. This leads me straight on to the origin of these relatively high takeoff and landing speeds, namely the aerodynamic compromise between high and low speeds. High- Speed/Low- Speed Aerodynamic Compromise There are at least three requirements to be met in designing an SST: 1) The airplane must be configured for supersonic cruise flight. 2) It must adapt readily to subsonic flight, notably for takeoff, for landing, and also, for holdings prior to landing until such time these holdings WING TIP:55° ACL DUE TO VORTICES (X (d*} i PREDICTED IN FLIGHT,BETWEEN M=OANDM = 0.8 <X (d") Fig. 8 Concorde lift curve and lift increase due to vortices. can be reduced. 3) It must possess excellent flying qualities throughout the flight envelope. Now although the third requirement can be met by making the necessary refinements, the first two are in direct conflict. The subsonic regime is of considerable importance in the over-all economics of the SST, as portrayed diagramaticallv in Table 5. You will observe that the requirements imposed for holdings and diversions have a determining effect. This being so, the attractions of variable geometry as a means of resolving this contradiction are manifest, and I am not surprised that Boeing attempted a breakthrough along these lines. Their design study, the only one to have been taken far enough, was necessary to establish that the variable geometry solution is not yet ripe. This is not to say that we shall not see back with us some day. But so far the 3 SST design projects feature near-fixed geometry, I say "near" because the droop nose and the air intakes do feature variable geometry, which makes my task simpler. I shall simplify it still further by confining myself to what I know well: the solutions adopted for the CONCORDE. Starting with a pure delta wing with 63.5° of sweepback, we set about looking for improvements, with special emphasis on lowr -speed qualities. A sharply swept apex (about 76°), as shown in Fig. 7, produced a triple advantage: 1) a reduction in the thickness ratio at the wing root, plus an arrow planform, both favorable factors for supersonic flight, 2) a forward shift in aerodynamic center location, which in turn shifted Table 4 Concorde 001—first landings Touch down Fig. 7 Wing plan form. No. of flight 1 2 3 4 5 6 7 8 9 Mean values Approach speed, kt 167 171 175 171 179 175 176 174 165 172 Aiispeed, kt 160 168 165 166 171 169 172 168 150 165 Vertical speed, m/sec 1.1 0.5 0.6 . . . a a 0.5 0.8 0.8 . . . a 0.7 Bank angle d° 0.9 0.6 0.7 0.4 0.2 0 0 0.4 0.2 0.4 1 Cinetheodolite failure
JAN.-FEB.1970 SUPERSONIC AIR TRANSPORT 7 Table 5 Importance of subsonic regime SECONDARY AIR YALVE FIRE DOOR Route Paris-J.F.K.-3200 nm LAVER Block time Fnel consumption Subsonic AUXILIARY DOOR BAY COOLING Super- Sub- Super- Sub- DOOR sonic sonic sonie sonic Standard mission—lo RAMP ASSEMBLY PRIMARY NOZZLE holding (ideal case) 79% 21% 840 16% 15-min holding (average case) 74% 267% S1% 19% Supersonic 200-min diversion 30- SHOCK PATTERN FUSER NOZZLE min holding (eritical case under which Fig.9 Concorde power plant. requirements are set) 62% 38% 75元 25% the CONCORDE,the fact that the nacelles were grouped in pairs imposed a two-dimensional air intake.This is the heaviest loads into the most rigid structural area and schematized in Fig.9.I will spare you the deseription of facilitated accommodation of the landing gear,3)a these intakes and merely repeat that it is all very difficult more intense attached leading edge vortex resulting in greater to match up and optimize.I believe one of the hardest lift at low speeds. problems in designing a supersonic transport to be the con- Much work was also done on the shape of the wingtip, figuring of its propulsion system.This is indeed a thorny and the process of improving it aerodynamically has gone technical problem,for just now we saw the enormous economie on until just recently.It turned out that reducing the importance of fuel consumption. sweep to 559,by truncating the delta slightly,improved the On CONCORDE,so far,the system has functioned sat- lift drag ratio in subsonic flight against a very small tradeoff isfactorily in subsonic flight,and while it is difficult to dis- in the supersonic regime. sociate drag from thrust,we can say that the theoretical This ultimately led us to the ogee delta planfori shown figures have been borne out to within the measurement in the figure.It is more difficult to depict the wingtip error. As you know,I am not yet in a position to tell you camber and twist,though optimizing them is importaut. how the system will behave in supersonie flight. Finally,the dihedral must be chosen with due allowance for wing deformation in flight (which is significant despite the Thermal Problem and Choice of Materials small span:16 in.at the wingtip in cruise relative to static on the ground)and for the ground clearance. I am obviously still less in a position to talk to you about Optimally configured in this way,the wing has already fatigue in supersonic flight.In fact only the operational demonstrated its qualities in subsonie flight (Fig.8).This aircraft themselves will provide us with in flight fatigue confirmation fills us with confidence that our expectations data since our objective is a service life of 45-50,000 hr. for supersonic flight will be borne out and that the changes over half of which will be under high-temperature conditions. decided upon as between the prototype and the production The structural behavior will be investigated by fatigue aireraft will have the desired effect. tests on a full-scale structure which will be subjeeted to Studying the propulsion system on the ground is more mechanical and thermal loads,with suitable amplification difficult.First,the interactions between the intake,the coefficients to insure safety. engine and the nozzle are highly complex;second,none of Figure 10 represents the equivalent of a flight eyele:the our test facilities provided complete simulation of flight mechanical loads are applied twice,the thermal loads once, conditions.The margin of uncertainty is therefore greater. amplified by 15-20%.This procedure gives the speeimen Thus,making the final choice between major options is time to cool down. doubtless less simple.Although all 3 SST projects ulti- Our confidence in these tests is founded,in the main, mately settled for a fixed delta wing by the end of their firstly on the component tests and secondly on experiments design studies,the powerplants are of the direct flow type with the CARAVELLE,the results of which are summarized in two cases and of the bypass type in the third.The ad- in Table 6. vautages,from the drag and weight standpoint,offered by During these CARAVELLE fatigue tests,99 damages the direct-flow type made us decide in its favor,and we use were noted.Seven modifications were decided upon and a development of the British Olympus turbojet,the Olympus introduced,since which only 9 damages have been recorded 593. in service,even though many of the aireraft have now logged The problem of configuring the propulsion system for more than 20,000 hr (and some as many as 27,000-28,000 hr) supersonie/subsonic operation is not so much an engine problem as an air intake and nozzle problem.In the case of Thermal loads 1+151o202 Table 6 Caravelle structural fatigue test results (without landing gear) Number of cycles applied to specimen: 100.000 Number of damages during testing: 99 Number of modifications decided: Number of flight cycles completed: upt030,000 (0.5 cracks Same damages as during testing 0.3 fastener failures 9 (0.1 fretting corrosion New types of damage 0.2 cracks including one dne to stress corrosion Fig.10 Fatigue testing
JAN.-FEB. 1970 SUPERSONIC AIR TRANSPORT Table 5 Importance of subsonic regime SECONDARY AIR VALVE Route Paris-J.F.K.-3200 nm Fuel consumption SubBlock time Super- Sub- Supersonic sonic sonic sonic Standard mission—no holding (ideal case) 79% 15-min holding (average case) 74% 200-min diversion -f 30- min holding (critical case under which requirements are set) 62% 21% 84% 26% 81% 38% 75% 16% 19% the heaviest loads into the most rigid structural area and facilitated accommodation of the landing gear, 3) a more intense attached leading edge vortex resulting in greater lift at low speeds. Much work was also done on the shape of the wingtip, and the process of improving it aerodynamically has gone on until just recently. It turned out that reducing the sweep to 55°, by truncating the delta slightly, improved the lift drag ratio in subsonic flight against a very small tradeoff in the supersonic regime. This ultimately led us to the ogee delta planform shown in the figure. It is more difficult to depict the wingtip camber and twist, though optimizing them is important. Finally, the dihedral must be chosen with due allowance for wing deformation in flight (which is significant despite the small span: 16 in. at the wingtip in cruise relative to static on the ground) and for the ground clearance. Optimally configured in this way, the wing has already demonstrated its qualities in subsonic flight (Fig. 8). This confirmation fills us with confidence that our expectations for supersonic flight will be borne out and that the changes decided upon as between the prototype and the production aircraft will have the desired effect. Studying the propulsion system on the ground is more difficult. First, the interactions between the intake, the engine and the nozzle are highly complex; second, none of our test facilities provided complete simulation of flight conditions. The margin of uncertainty is therefore greater. Thus, making the final choice between major options is doubtless less simple. Although all 3 SST projects ultimately settled for a fixed delta wing by the end of their design studies, the powerplants are of the direct flow type in two cases and of the bypass type in the third. The advantages, from the drag and weight standpoint, offered by the direct-flow type made us decide in its favor, and we use a development of the British Olympus turbojet, the Olympus 593. The problem of configuring the propulsion system for supersonic/subsonic operation is not so much an engine problem as an air intake and nozzle problem. In the case of Table 6 Caravelle structural fatigue test results (without landing gear) Number of cycles applied to specimen: Number of damages during testing: Number of modifications decided: Number of flight cycles completed: Same damages as during testing New types of damage 100,000 99 7 up to 30,000 0.5 cracks "} 0.3 fastener failures /• 9 0.1 fretting corrosion/ 0.2 cracks including one due to stress corrosion 2 BOUNDARY^ LAYER BLEED Subsonic AUXILIARY DOOR BAY COOLING DOOR TERTIARY AIR DOORS RAMP ASSEMBLY PRIMARY NOZZLE A_____\ Supersonic SHOCK PATTERN DIFFUSER SECONDARY NOZZLE Fig. 9 Concorde power plant. the CONCORDE, the fact that the nacelles were grouped in pairs imposed a two-dimensional air intake. This is schematized in Fig. 9. I will spare you the description of these intakes and merely repeat that it is all very difficult to match up and optimize. I believe one of the hardest problems in designing a supersonic transport to be the configuring of its propulsion system. This is indeed a thorny technical problem, for just now we saw the enormous economic importance of fuel consumption. On CONCORDE, so far, the system has functioned satisfactorily in subsonic flight, and while it is difficult to dissociate drag from thrust, we can say that the theoretical figures have been borne out to within the measurement error. As you know, I am not yet in a position to tell you how the system will behave in supersonic flight. Thermal Problem and Choice of Materials I am obviously still less in a position to talk to you about fatigue in supersonic flight. In fact only the operational aircraft themselves will provide us with in flight fatigue data since our objective is a service life of 45-50,000 hr, over half of which will be under high-temperature conditions. The structural behavior will be investigated by fatigue tests on a full-scale structure which will be subjected to mechanical and thermal loads, with suitable amplification coefficients to insure safety. Figure 10 represents the equivalent of a flight cycle: the mechanical loads are applied twice, the thermal loads once, amplified by 15-20%. This procedure gives the specimen time to cool down. Our confidence in these tests is founded, in the main, firstly on the component tests and secondly on experiments with the CARAVELLE, the results of which are summarized in Table 6. During these CARAVELLE fatigue tests, 99 damages were noted. Seven modifications were decided upon and introduced, since which only 9 damages have been recorded in service, even though many of the aircraft have now logged more than 20,000 hr (and some as many as 27,000-28,000 hr) Fig. 10 Fatigue testing