Study of fuel systems for LH2-fueled subsonic transport aircraft, volume 1

TABLE OF CONTENTS Section Page Volume I Foreword ............................ iii List of Figures .........................

List of Tables ......................... xx & 3.3.1 Sensitivity of DOC to engine weight, SFC, and 3.3.2 Sensitivity of DOC to fuel pumping system pow e r 3.3.3 Sensitivity of DOC to volume and weight of fuel V

TABLE OF CONTENTS (Continued) Section Page 4.3.5 Operational characteristics ................... 87 , 4.3.6 Description of engine-mounted heat exchangers .......... 9t e 5.1.1 Concept descriptions ....................... I00 5.3.1 D e sign r e quirements ....................... 116 b 5.3.2 Candidate pump types ....................... 119 _ o vi TABLE OF CONTENTS (Continued) " Section Page 5.4.2 Candidate pump types and selection of preferred vii

TABLE OF CONTENTS (Continued) Section Page Volume II 7.1.2 Candidate insulation concepts .................. 19_ 7 3 . 1 W e ight con s iderati o ns ...................... 393 i 8 I LH 2 Aircraft D e s c ripti o n .................... 410 ' viii S T UDY O F FUE L SYS T EMS F O R LH 2 - F U ELED S U BS O NIC T RAN S POR T AIRCRAF T G. D. Brewer, R. E. Morr i s , G. W. Dav i s, E. F. Versaw G. R. Cunnlng to n, Jr., J. C. Ripl e , C.F. Ba e rs t , G. Garmong, Lockheed-Cal l forn l a Company Burbank, California S U M M ARY Conc e rn for t h e p o t e ntial s hort supply o f petroleum-base fuels has led to a series of studi e s sp on s o red by NASA whlch have explored the technol o sical asp e cts and e stablished the pot e nt i al of using l i quid hydrogen (LH2) for fuel in advanced cormmerclal transport a i rcraft. Pr e v i ous studi e s h ave i nv e st i gated most promising methods of producing hydr o g e n, proc e sses for liquefying the gas, aircraft configurations, and air t e rminal design and operations as they w o uld be affected by i n troductio n of L H 2-fueled a i rcraft.

T he present study was dir e cted at e x ploring the d e sign problems prese n ted by the fuel syst e m of a representative L H 2-fueled transport. T his encompasses e veryth i ng requ i r e d in th_ aircraft to contain, Control, or handl e the fuel.

Although hydrogen fuel systems have b ee n dev e loped for space m i ss i on appl i ca- ti o ns, the requirements f o r a i rcraft are s o d i ff e r e nt in regard t o mission dur- . a tio n , system l i f e, op e r a tin g cyc l es, a nd s afe ty a spects th a t en tirely di ffe r e nt d e s ig n pr ob lem s are presen t ed . The experience w it h LN 2 sy st ems i n t he U . S . S p ace P r og ram d id, h o wever , prov i de v a lu a b le reference da t a a nd s erve a s a p oi n t o f dep a r t ure in e st a b l is h i n g desi g ns for s o m e o f t he aircraf t c o mp o nen ts.

An a i rcraf t d e sig n fr o m a prev iou s stu dy p e rf or me d b y Lo ckh e e d f o r NASA ( R e f erence i ) was used a s b a s i s f o r devel op in g t he fuel s y st em des ig n . The a i r- c raf t i s sh o wn i n t he fr o n ti spiece . It carries 4 0 0 p as sen g er s i 0 190 km ( 550 0 n .mi$ a t a cru i se s p eed o f Ma c h 0. 85 . A de sign g uidel ine w a s tha t th e t ec h- n o l og y should re p re s ent i n iti al o perati o nal c a pab ilit y i n 1990-1995 .

I n o rder to pr o v i de m a x im u m c om pe te nce i n all a s pe ct s o f t he st udy L o ckhe e d-Cal l f o rn l a C o mp a ny w a s supp o r t ed by L o ckheed Mi ss iles and Space Co m pany , I n c. , t h e Cal i f o rn i a a n d Ariz o n a A i Re s earch Divi s i o ns o f t he G a r r e tt C o rp o ra tio n , and t he R oc ke t d y ne D i v i s io n of R o c kwell I n t ern a tio nal i n t he per - f or m a n ce o f t h e w o rk .

An i n iti al tas k i n t he s tudy w a s to define a n e ff ici en t eng i ne c ycle j o ne wh i ch w o uld take be s t advanta g e o f t he un i que pr o per ties o f hydr og en . F i ve ideas for exploiti ng t he advantages o ffe r ed b y th e large h ea t capac ity and th e l ow t e m pera t ure of hydrogen were explored. The s e i n c luded p reco mp ressor cool in g , co m pressor int ers t age cool i ng, cool i ng of t he t urb i ne cool in g a i r , regenera ti ve L ock h eed Missiles and Space Company AiR e s ea r ch D i vi sio n s o f t he G ar re tt C o rp o r a tio n R o cke t dyne D i v is i o n o f Ro ckwell In t erna tio nal heating of the hydrogen f u el by th e cor e e xhaust, and use o f a_ e xpansion c y c l e in connection with e xhaust heat regeneration to provide power for engine acces- sories. In additi o n, two levels of turbine rotor inlet temperatur e s, 1482 and 1760°C (2 7 00 and 3200°F), were evaluated, each in conJunctlon with an appropri- ate range of values of cycle pressure ratios and fan pressure ratios to p e rmit selection o f a pr e f e rred set of those parameters. All of this work was based on a definition of engine col a ponent performance and efflciencies agre e d upon as represen t ative of technology which can be developed for operational us e by 1990.

The selected e ngine cycle was based on the following characteristics at sea l e v e l static, s t andard day conditions.

Ro t or Inlet Temperatur e 1482°C (2700°F) Cycle Pressure Ratio 35 Fan Pressure Ratio (tip) 1.594 Bypass Ratio 10.25 Th e design uses hydrogen to cool th e turbin e cooling air, th e engin e oil a nd o ECS air, and also provides for heating the fuel to 677 K (1219°R) in an exhaus t r e g e n e rator b e for e inj ec tion into th e combu s t e r.

The e ngin e fuel supply system and the engin e d e livery and c ontrol syst e m re c eiv e d significant attention. The engine fuel supply syst e m takes the LH 2 out of th e tanks and deliv e rs it to th e inl e t of the engin e high pressure pump.

It consists, in main, of the boost pumps, valves, and fu e l d e liv e ry lines. For reasons of reliability, e ach of th e fo u r tank compartments in the airplane is pr o vided with a c luster of three boost pumps. The pumps are three-stage, v a rl a ble sp e ed, centrifugal designs which are driven by 270 Vdc motors. Th e y are d e sign e d to b e llne r e plscable units (LRU'_ for ready removal from th e air- pl a n e in cas e th e y ma]iun c tion.

Fuel delivery lines are stainless steel, 2.54 cm dis x 0.406 mm wall !

(i.0 in. x 0.016 in.). The lines are enclosed i_ 3.81 cm (1.5 in.) of closed I i cell foam insulation, which is i t self c o ntained in a 10.16 cm dia. x 0.406 mm wall (4 in. x 0.016 in.) aluminum tube which serves as a vapor barrier and provides mechanical protec t ion.

The engine delivery and control system consists of the engine-mounted high pressure pump, heat exchangers, and the fuel con t rol system, all mounted in the engine nacelle. The engine pump is a t wo-s t age centrifugal design, shaf t driven at a fixed speed ratio. It is designed to take saturated liquid hydrogen (3 NPSH ) at _45 kPa _50 psia) and provide a flow of 0.454 kg / sec tl.O ib / sec) at a pr ess ur e ri se o f 4813 k P a (698 psi). Its des i g n rotati o nal s p eed is 50 000 rpm.

• Th e e ngin e fuel cont r ol sys tem e m plo y• e le c t r onic c ont r ol ci r cuit r y and ha s a flowmete r and a flow-modulating and s hut-off valve lo c ated ju s t ahead of the engine combu s to r to cont r ol fuel flow to the en g ine. The s e unit s are located down s t r eam of the heat e x change rs to avoid lag in r espon se which would other- wi s e r e s ult f r om the capa c i t ies of t he hea t e x changers.

Fuel s ubsystem s w hi c h w e r e d e sign e d in c lud e th e followin g : Fu e ling / _ e fuel, V e nt and Pr e s•urizatiun, Fu e l Transfer, and Fu e l J e ttison. Th e d e sig - _ , , _ _e- ment s o f e a c h o f the s e sub s yst em s wa s e s tab l ish ed , then th e des igns _' - c_ c re a .' d s o wei ght s a nd c ost s c o uld b e e s timated an d op e rationql re q u irements a_sess_d.

A_l ezte_sive _nal l tiral stady was carri e d out to d etermin e th e b e st d e s ign for the fu e l conta in m ent sy s tem. This consi s t ed of investigation of vari o, ls tank structural conc e pts and 15 dif fe rent t ank insulation Jyst e m s . The s t r uc- tural i n v es tigation in c luded analy s i s o f both int eg r al and nonlnt e gral t a nk de sign s , p lu s sev eral pa r am e tric st ud ies invo l ving consi de r a tion of • dome s h a p e • pr ess ure s tabilization • pr ess ur e l e v e l • design lif e • t ank s uppo r t method • Th e tank in s ulation s t u dy consi s ted o f a c o n ce pt s creening pha s e in which 15 d e sign s w e re inv es tiga t ed, followed by • e l e ction and mor e detailed e xamin a tion of four pref e rre d candida t e s , two each for inte g ral and nonint eg ral t a nk s tru ct ural d e sign s . Th e s e f our ca ndida te s w e r e e a c h tr e ated as a ba s is f o r a s e parate airplan e d e sign so th e c omparison and final c hoi ce c ould b e mad e in t e rms of param e t e rs of primary interest to air c raft op e r a tors.

T ank in s ulation c on ce pt s whi c h ori g inal l y e nt ere d t he c o nce pt sc r ee ni ng phas e in c lu de d r e pr es entativ es of a l l con ce ivabl e type s in c l_dlng a c tive s y s - t e m s d e p e ndent on r e a s onably hard va c uum [0.01 3 3 Pa ( I x i0- Torr )] ; s ome which w e re s elf e va c uating by a pro cess of c ryopumping an in c lud e d g a s ; a n d tho se which w e r e c ompl e t e ly pa ss iv e .

T h e fue l co ntain me nt s y s t e m which r a n k ed high es t in t he ov e ral l ra tin g sche m e w a s a d e s i g n which u s ed an i n t egr a l t ank and an i n s ul a t ion s y st e m c o n- s isti n g of tiny, h ollow boro s i l i c ate s pher e s ( microsph e res ) c o ntain ed in an annulu s en c l osi n g t he t a nk wh i ch is p u mped to a s o f t vacuum . The des ig n pres- s ure i n t he annular space i s 13 . 33 Pa (0 . I T o rr) . A very c l os e s e co nd c h o i c e in t he f i nal e v alu a t i o n was a d esi g n wh ic h a ls o u sed an in t e g r a l t a nk bu t t he i n s ula tio n s y st e m wa s a wra p o f c l os ed c ell p las tic f o a m a r o un d t h e t ank , w it h a vapo r barr i e r th e n w rapped aro un d t ha t t o p reve nt ai r fr om p ene t ra ti ng t he f oa m . Th e se tw o ins u l a ti o n sys t e m s w ere so close i n t he ra ti ngs it is reco m - m e n de d bot h be fu rt her devel op e d. Th e nonint e g ra l ta nk d es ign s w ere e liminat e d b ecau s e of t h eir i nheren t t enden c y to be bot h heav i er a nd t h ic ke r.

!

!

Following th e d e sign o f all e lem e nts of th e airc r aft LH 2 fu e l s y s t e m it was r e quir e d that a comparison b e mad e betw ee n th e LH 2 and a corr e sponding Jet A-fu e led aircraft. To do this it w a s nec e ssary t o gen e r a te a J e t A e ngin e d e sign which ma t ch e d th e compon e n t p e rformanc e us e d in th e subj e c t LHp e ngin e .

This was accomplish e d and LH2-fuel e d ant J e t A-fu e l e d aircraft d e sign_ w e r e th e n e stablish e d so th e comparison of th e ir charact e ristics could b e mad e . A summary of som e of th e signifi c ant par a m e t e L= is pr e s e nt e d in th e following tabl e .

R a tio / Jet A_ LH 2 Jet A _ LH 2 ] G r os s weight kg 168 829 23 2 056 (ib ) (37 2 200) (511 6 0 0 ) 1.37 Op e ra t ing e mp t y wt k g 1 0 3 3 05 107 3 6 3 (lb) (22 7 750 ) ( 2 36 7 00) 1 . 0 4 Blo c k fu e l w e ight kg 2 1 6 2 1 7 2 3 65 (ib ) ( 47 670) ( 159 540) 3.35 Thrust per e n g in e N 135 000 184 900 ( I O) ( 3 0 3 50) ( 41 567 ) 1.37 Wi ng ar e a m2 2 96 .8 380 . 3 (f t 2) (3 195) (4 0 93 ) 1. 28 Span m 51.7 5 8 .5 ( ft ) ( 1 6 9.6 ) ( 191. 9) 1.1 3 B ody l eng th m 6 5 .7 6 0 . 0 (f t ) ( 21 5. 6 ) ( 1 9 7 .0) 0 .9 1 4 A i rcraf t price $1 0 6 4 3 . 39 44. 53 1 . 03 D O C * c / se a t k m 0 . 8 6 9 0 . 9 0 7 (C / se a t n.ml. ) ( 1 .609) (1 .67 9) 1 . 0 4 Energy u ti l i za tio n k J / s e a t k m 63 6 7 59 ( Btu / s e at n.mi ) ( 111 8 ) ( 1 3 34 ) 1.1 9 *Cal cula t ed fo r ba se lin e pr i ces o f e a ch f u e l; $5 .69 / GJ ($ 6 / 106Btu) fo r L H 2 an d $ 4. 7 4 / G J ($5 / 1 06 B tu ) f o r s yn t h et i c J e t A , ass um i n g bot h f u el s a ge m a d e fr om c o al .

Th e LH2- f ueled des i g n is s up er i or in n ea r ly e v ery pa r am e t e r. I n f ac t, th e a d vanta ges a re grea t e r t h a n t ho s e ca l c u l at ed o ri g inal l y a s p r ese nt ed i n R e f e r ence I. T h i s i s _u e prima r il y to a m o r e c ons e rvative app r ai s a l o f s ome en g in e co m p onent e ffi c i e n c ies, re fle c t ed in e n g in e s us in g bo t h fuel s ; ho weve r , t h r o u gh explo it a tio n ot t he pr ope rti e s of h y d r oge n, t he s p e cifi c fu el con- s u mptl on of t he LR2 des i g n w as ne arl y res t ored to it s o r i g ina l v alue. At th e b a s e l i n e pri c e f or s yn th e t i c Je t A, a p r i c e diff e r en tia l amo u nti ng to a n addi t i _l a l $1 . 59 per G J ($ 1. 6 7 / 106 Bt u) can be p ai d for LH2 f u e l a nd st il l provide equal D O(: .

A l i s t of 12 i t e m s is rec omm e n ded for develop m ent of t echnolo g y or t o p rov i de in for mati o n needed i n order for L H 2 - f uel ed ai rcra f t t o b eco m e a via b le p o ssibi l it y. T he i t e m s are arrange _ in order of p r i orit y accord i n g t c sc h edu lin g req ui re men ts . Dev e lo pm e nt of an a ircra f t t ank a nd i n sulat io n s yste m and LH 2 . ump s is cons i de r ed t o p p r i ority.

NOMENCLATURE I NOTE: Computations in this an a lysis were performed in U.S.

Customary units and then conv e rted to S.I. units.

AR - Aspect Ratio ATA = Air Transport Association BPR = Bypass Ratio CPR = Cycle Pressure .Ratio DOC = Direct Operating Cost E = YounE's Module' , of Elasticity (compression) C E t = Young's Modules of Elasticity (tension) ECS = Environmental Control System , FAR = Fedecal Air Regulation FCS = Fuel Containment System fg = Fiberglass , FN = Net Thrust FPR = Fan Pressure Ratio F t = Ultimate fiber stress, tension • U H = Head in feet HC - Honeycomb HP = High Pressure H.P.EXT = Horsepower Extraction 1 = Integral IGV = Inlet Guide Vanes IOC = Initial Operational Capability Jet A = Convention_l Hydrocarbon fuel KEAS - Knots Equivalent airspeed L / D = Lift-t o - Dr ag r atio L H 2 = Liquid Hydrog e n LHV = Fu e l Low e r H e ating Valu e LP = L o w Pr e ssur e M = Mach Numb e r M D = Des ig n Math Numb er NOMENCI , ATURE (Continued) N = Rotation a l speed, rpm NI = Nonintegral NPSH = Net Positive _Jction Head OPR _ Overall Pressure Ratio OEW = Operating Empty Weight P = Pressure PLA = Positive Low Angle of Attack PT2 = Average Fan Face Total Pressure PTo = Freestream Total Pr e ssure Q = H e ating rate or Volumetric flow rate in gpm , QEC = Quick Engine Ch a nge Nacelle RIT = Rotor Inl e t Temperature S = Wing Reference Area SLS = S e a Level Stn*Ic TOGW = Takeoff Gross Weight T / W - Thrust to Weight Ratio TCA = Turbine Cooling Air TIT = Turbine Inlet Temperature VJP = Primary exhaust jet velocity VJD = Fan Duct exhaust Jet velocity Vo = Flight velocib Vr = Takeoff rotate Velocity Vs - stall Velocity I w = Flow rate = Engine cor r e c ted airflow a & P 2 m Wpo d = Engine pod weight W / S = Wing Loading = Aircraft weight wing area ZFW - Zero Fu e l Weight - Angle of Attack - (PSIA / 14.7) 6p 2 - Delta r 2 PT2 NOMENCLATURE (Continued) ° 0T2 = Theta T 2 = TT2 (°K / 28g.2) q = efficiency p = density = heat exchanger effectiveness i. INTRODUCTION As a result of serious concern regarding the potential short supply of petroleum-base fuels, in 1973 the National Aeronautics and Space Administra- tion instigat e d a p ro gr a m to inv e stigate alternate fu e ls f o r co nwn e rcial transport aircraft.

Liquid hydrogen (LH2) , liquid m e thane (LCH4), and synthetic Jet A (synjet), all manufactured from coal and water, are leading alte c nate fuel candidates. To date, attention has been focused primarily on liquid hydrogen and on synjet, assuming the synthetic jet fuel would have the same properties as the present fuel for commercial airliners, Jet A or Jet A-I. Aircraft designs based on use of both of these alternate fuels have been created and compar e d (Ref e rence I), and the facilities, equipment, and operations needed at representative major air terminals to service liquid hydrogen-fueled air- craft have been studied (References 2 and 3).

Th e LH2-fueled aircraft d e sign from th e previous study (Referen c e I) was a conceptual design in which advanced technology features were in c orpo- rated representing an initial operational capability in the 1990 decade.

The aircraft was sized to carry 400 passeng e rs I0 190 km (5500 n.mi.) at a c ruise spe e d of Mach 0.85. N ec essarily, many assumptions were mad concern- ing the characteristics of the LH2-fueled e ngine, the fuel containment system, th e engine fuel supply system, and other elements of the complete air c raft fuel system.

In the present work, attention was focused on precisely those items so that a more realistic evaluation of the potential of a hydrogen fueled trans- port aircraft could be obtained. The objectives of this study wer e as follows: • D e fin e th e charact e ristics of a preferred design of fu e l system for the specified LH2-fueled transport air c raft.

• Establish the size, w e ight, c ost and performance of the LH2-fueled aircraft using the final fuel system d e sign.

• Compar e the LH2-fu e l e d aircraft with an equivalent t e chnology Jet A- fueled d e sign.

• Id e ntify r e lated r e search and technology dev e lopment r e quirements for the LH2 fuel syst e m.

An outlin e of th e a ppro ac h t a k e n in p e rforming this study is des c ribed in Section 2 .

2. TECHNICAL APPROACH 2.1 Team Organization The wide scope of technical expertise required to define adequately a practical fuel system for a liquid hydrogen-fueled aircraft led to formation of a team, the members of which were selected for their competenae in speci- fied technical areas. Lockheed-Callfornla Company reached agreement with . the following companies to participate in the study on a subcontract basis to provide special skills and innovative thinking in the areas indicated: • Lockheed Missiles and Space Company, Inc. - For analysls, design, and evaluation of cryogenic insulation systems, and for specialized tank structural analysis • Airesearch Divisions of the Garrett Corporation - For analysis and design of an advanced design LH2-fueled turbofan engine, a fuel con- trol system, pumps and other specific components • Rocketdyne Division of Rockwell International - For analysis and • design of the LH 2 engine fuel supply system, and for boost pump design This team provided an ideal combination of basic knowledge and familiarity with the reference aircraft design, plus experience with technology developed in the U.S. Space Program on cryogenic fuel systems in general, and use of LH 2 in particular.

2.2 Work Plan A schematic representation of the study work plan is shown in Figure I.

The work was performed in four phases. Phase I involved compilation of inpu t d a ta needed in the re m ainder of the study. These ite m s are described in Section 3.

Phase II, Syste m Studies, was the focus of the principal effort of the study. In this phase, the designs of the LH2-fueled turbofan engine, the engine fuel supply system, the fuel s u bsyste m s, a nd the fuel c o ntainmenu sys- tem were established. In essence, these tasks involved examining the require- ments, originating design concepts for evaluation, and choosing preferred , designs for each of these elemen t s of the fuel sys t em of an LH2-fueled air- craft. In addition, a comparable design of Jet A-fueled turbofan engine was also established to provide a basis for equivalent comparison of aircraft operated with the respective fuels.

• !

Phase III was an evaluation of the fuel system d_sign which resulted from the work of Phase II. Drawings of the engine f t ,-L supply system, the pressurization / vent system, and the fueling / defuel system, plus principal • components of each of these, were prepared so weight, reliability, mainte- nance, and operational requirements could be assessed.

Four designs of fuel containment systems (tank stlucture, insulation system, and support structure) were selected from 15 candidates originally con._ived and studied. These four selected designs were evaluated using the Lockheed Aircraft System Synthesis Evaluation Technique (ASSET) computer program to establish the potential of each in terms of aircraft size, weight, performance, and cost. Direct operating cost (DOC) was the principal mea- sure of merit used in selecting the final preferred desigh.

The characteristics of the aircraft with the preferred fuel containment system, which also used the LH 2 engine and fuel system designs from Phase II, were compared with those of an equivalent technology Jet A-fueled aircraft.

The Jet A design was subjected to the same optimization procedure using ASSET as the LH2-fueled _esign so that the comparison would be on an equitable | basis.

Phase IV consisted of summarizing the results of the work and preparing • the final report. A recommended research and technology development program for critical LH 2 fuel system elements was formulated as a part of this effort.

i i } 3. STUDY GUIDELINES ND INITIAL DATa Information required to perform analyses of th e LH2-fueled e ngine a:'J fuel system components was generated or assembled as an initial step in the program. This work included reaching agreement with NASA on specific guide- lin e s and requirements to be met; compiling basic data from Reference 1 on the baseline aircraft int o a convenient package; generating sensitivity factors so that the benefit or liability of changes in key aircraft param- eters could be assessed, and thus provide help in guiding design decisions; and finally, formulating a procedure for calculating direct operation cost (DOG) for the subject aircraft which would reflect a reasonable approxima- tion of curr e nt airlin e practices, and which would also account for differ- ences between LH2-fueled and conventionally fueled aircraft.

3.1 Guidelines and Requirements The guidelines and requirem e nts which were established for use during the study are listed in Table i. These items were either originally speci- fied by NASA as a basis for the study, or were perceived during the early stages of the program as being necessary for validity and consistency of results.

v 3.2 Basic Data Basic information on the reference LH2-fueled aircraft, and its design mission, which was needed to establish a starting point for the fuel system design requir e ments and analysis was derived, for the most part, from Refer- ence I. Some additional information was generated by making special runs of the ASSET computer program, and by separat e analyses. In all, the following items were assembled and transmitted to all study team members and to the NASA T e chnical Monitor as preliminary data: • Drawing CL 1317-I-I, General Arrangement - LH 2 M 0.85, 400 PAX, I0 190 km (5500 n.mi.)

• Drawing CL 1317-i-4, Engine Feed System - LH 2 Subsonic Transport (preliminary draft) • Drawing CL 1317-I-5, Fueling / Vent System - LH 2 Subsonic Transport (preliminary draft) • Design Mission Fuel Flow Schedule (shown in Appendix A).

TABLE I. - GUIDELINES A ND REQUIREMENTS • Baseline aircraft: Final design in Ref e rence I: 400 passengers, I0 190 km (5500 n.ml.) range, Mach 0.85 cruise speed. See Figure 2.

Initial Operational Capability: 1990 - 1995 Baseline fuel costs: LH 2 = $5.69 per GJ ($6 per 106 Btu = 31¢ / ib) Syn j et = $4.74 per GJ ($5 per 106 Btu = 9.2¢ / ±b = 62.2¢ / gai) D O C basis : !

1967 ATA equations updated to 1976 cos t experience and modified to _ more accurately reflec t airline practice, as well as differences resulting from use of alternate fuels. Assume production of 350 aircraft and 3600 engines.

Evaluation criteria: • DOC to be final measure of merit. All concepts must meet safety, reliability, maintainability, and operational requirements.

• Safety: Equal to or better than conventionally fueled commercial transport.

Design criteria: Meet all applicable or anticipated regulatory requirements including FAR 25.

• LH 2 Turbofan Engin e Thrust an d Fu e l Flow • Fu e l Flow Env e lop e - LH 2 Turbofan • D e s ign m i ss i o n flight profile • Lo ckheed L H 2 e n g ine char a c t er i s t ics , c o m po nen t efflc l enc i es , a n d i ns t all a t i o n fac t or s (i nle t rec o very , ins t a ll a t ion dra g, bleed air r e quir eme nt s , pow e r e xtraction r e quir e m en t s) .

1 3 f l ,, 3.3 S e nsitivity Fa c t o r s . Sensitivity factors were generated for the r eference LH2-fueled airplane to provide a basis for evaluation of the effect of changes from the baseline design. For example, in the study of the LH2 engine design, a given option may have offered a few percent reduction in specific fuel consumption (SFC), but at the expense of an increase in inert weight. Sensitivity factors were a m e ans of evaluating the net b e n e fit which might b e r e alized by incor p orating that option in the design of the engine. Note that the sensitivity factors were used merely as an evaluation procedure to assist in screening attractive candidates. Evaluation of final design concepts was made by incorporating appropriate data in the ASSET aircraft synthesis program.

The sensitivity factors, or exchange ratios as they are sometimes called, were established by using the ASSET computer program to define optimized v e hicles fo r e ach of a s e ri es of de viati o n s f r o m t he n omina l va lue o f it e ms l ik e s pecific fu e l consumption ( SFC ) , v e hi c l e in e rt w e ight (Wi ) , and thick- ness o f f uel tank insulation sy s tem. Th e re s ult s w e r e th e n plott e d v s variou s airp l ane characteri s tics s o the s lop e of th e curv e through th e d es ign point r e pr e s ent e d th e s e nsitivity of those c haract e ri s ti c s to s mal l chang es in the para me t e r b e ing studied.

• An ex ample of this proces s i s pr e s e nted in T abl e 2 and Figure 3. Th e e f fec t of I0 p e rc e nt and 20 per ce nt chang e s in SFC, both above and b e low the nominal valu e , on takeof f gro s s w e ight ( T OGW), fuel w e ight (W f ), op e rating • em pty w e ight (OEW ) , manufactur e r's e mpty w e ight (MEW), e ngin e size, a i rplane cost, and direct op e rating cost (DOC) were all e valuat e d. Th e aircra f t r e pr ese nt e d by each column in T a bl e 2 are real in th e s ense tlmt they hav e b ee n s iz e d u s in g th e ASSET pro g ram s o that they repre se nt a minimum g ross w e i g ht d es i g n to p e rform th e r e quir e d mi s sion, and that they m ee t all th e sp ec ifi e d d e sign c onstr a ints.

Th e results as plott e d in Figur e 3 illustrat e th e e ffect c hanges in SFC would hav e on TOGW, OEW, Wf, and DOC. Th e slop e of th e c urv e at th e de si g n p o int, shown f o r e ach ca se , i s the sensitivity f actor. It i s a ccur a t e in r e pr e s e ntin g th e e ff ec t on th e vari o us airplan e c hara c t e risti c s of small dev i a ti o n s in th e s u b j e ct pa ra m e t er . If l ar g e dev i a ti o n s are c o n te mplated, th e i r ef f e ct mu s t b e rea d fr o m the cu r ve s , or a sep a r at e e valu a tion m u s t b e perf o rme d .

Sim i lar d a t a are t abu l a t e d and pl ott e d in Table 3 a nd F ig ure 4 to i llus- t r a t e t he effec t c h a n g es i n i ner t w e ig h t of t he aircraf t w o uld have o n cer t a i n ch a r a c t er i s ti c s a ss u m i n g t he de sign h a s n ot be e n f ro zen . This assu mptio n a ll o w s de sIpn char a c t eris tic s o f t he a i rcr a f t such a s w i n g l o a d i n g a n d t h r us t - t o -w e l gh t ra tio t o be c hanged to accom mo d a t e th e i ner t w e i gh t var i a tion s i n t h e m o s t e ffi ci en t ma nner .

1 5 T ABLE 2. - EFFEC T OF C HANGE IN SPECIFIC FUEL CONSUMPTION (Reference SFCcruise = 0.203 (k g / hr) / daN (0.199 hl_br / ib)) SFC Basis 80% 90% Reference 110% 120% TGGW kg 16_ 440 169 467 177 67 2 186 717 196 510 (ib) (358 120) (373 610) (39 1 700 ) (411 640 ) (433 230 Fuel wt. kg 21 287 24 400 2 7 946 31 792 35 929 (ib) (46 930) (53 880) (61 610) (70 090 ) (79 2 101 OEW kg I01 242 1 05 iii 109 810 115 008 120 660 (ib) (223 200 ) (231 7 30) (242 090) ( 2 53 550) (266 010 MEW kg 92 106 95 894 I00 498 105 592 III 135 (ib) ( 2 03 060 ) (211 410) ( 2 21 560) (232 7 90) (245 010 Thru s t per N 116 677 121 748 127 619 134 114 141 142 Engine (ib) ( 2 6 23 0) (2 7 3 7 0) ( 28 69 0 ) ( 3 0 15 0 ) ( 3 1 7 30 1 Cos t / air c r a f t ( $ 1 06 ) 38 .96 4 0 . 25 4 1 . 81 4 3 . 53 45. 3 9 DOC ¢ / S km 0.862 0.922 0.990 1.063 1.143 (¢ / S n.ml.) (1. 597) ( 1.707) (1.833) (1.969) (2.116_ (% of ref.) 8 7 .1 93.1 I00 1.07 1.15 TABLE 3. - EFFE CT OF CHANGE I N INE RT WEIGHT VARIATION -9 0 7 2 -4 5 3 6 +4 53 6 +9 0 7 2 Inert Wt. kg(ib) (-20 000) (-I0 000) R e ference (+I0 0 0 0) (+20 000 TOGW kg 159 306 168 91 8 177 672 1 87 365 19 7 304 (i b ) (351 210) (37 2 400) (39 1 70 0) (413 070) (434 980 Fuel wt. k g 26" 1 50 2 7 238 2 7 946 29 039 30 218 (ib) (5 7 650) (60 050) (61 610) (64 020) (66 620] DEW k g 93 240 I01 763 109 810 118 410 12 7 169 (i b ) (205 560) (224 350) (242 090) (261 050) (280 360 MEW k g 8 3 983 92 474 I00 498 109 062 117 789 (i b ) (185 150) (203 8 70) (221 560) (240 440) (259 680 Th r u s t p er N 11 4 453 1 2 1 34 7 1 2 7 6 1 9 1 34 6 0 3 141 7 20 ! Eng ine (I b ) ( 2 5 7 30) (2 7 28 0) (2 8 690) (30 260) (31 8 601 Co s t / a ircr aft ( $10 6 ) 35 .77 38 . 88 41.81 44.90 4 8 . 0 2 D O C ¢ / S km 0.90 8 0. 9 5 2 0 .99 0 1.033 1.0 7 9 (¢ / S n.ml.) (1.6 81) ( 1.763) (1.833) (1.914) (1.998_ Q (% of r e f.) 91 .7 96.2 I00 104 109 " 45 0 -- . , !

TOGW

40 (REF. SFCcRuISE " 0.2031 _ - _ - r I da N = 816.5 kg J %SF C nr . A % SFC (18001b / % SFC) A TOGW [ (0"_ 99 I _ b / I _ _ 200 I / Z •r 300 = 362.9 k9 / % SFC 25 / A % SFC (800 I b/ % SFC) 125 k -

" / _ 25 0

" A % SFC ! _ 6 kg / %SFC 100 • (1075 Ib P ASFC) 20O

/

/

_Jf (1.833 ( I / S n.mi.) / 100 _ SS

(,, ..o oc : o,,,, ,/_. /

" / I

-- _ 0 . 00718 % I ;FC

- -/

80 J _s" J 80 gO lO q 110 120 . , _ SFC (PERCENT) , D " Figure 3, - Sens i t i vity o f CL 1317- 1 a i rcraft to changes in specific fuel consu m ption .

35 , 0 32. 5 JO, O - " :[ / , 10O ,._

" j i * '

"_ ( 5 i - . , z7.5 / L _ % ,( _ ..... _ ' e " 0.225 . _ ,.

: _ ...- F U E L W T j - _ = u _ . . . ,. " / ._i_ _ _ -12S < 2S , O ..,,_ , _ ' _ ,A .OEW. 1 .875 L _ W i 100 2Z5 1 1 20 11o / _ ' o -- I I

j .. , _ w, I O =_ _ - l_ ,b

x REF.DOC - 0.I_ 17 i / Skin (1. 1 33 J l S _ m L ) 20 . .

-30 -20 -1 0 REF , +10 _ ,20 +30 (I ooo Ib) t _ I I I 1 • I I . 16 .10 -6 REF +6 +10 _ , 16 +29 INERT WEIGHTCHANGE 11000k gj F ig ure 4 . - Sensi t i vity of CL 1 31 7 - I a i r c raf t t o i nert we ig ht chan g e.

1 8 S e n s itivity fa c tors fo r th es e two p a ram e t er s, S FC an_ Wi , in addition to t h e e ff ec t of v a riation in thi c kn e ss of fus e lage t a nk insulation sy s tems and the r e sulting influ e nc e t hi s would hav e an fu se lag e l e ngth, were the primary " tools nee d e d to e valuate design or c on ce pt t rad e -offs throughout Phase II.

T he following ar e s om e of th e spe c ific trad e -off r e lation s hips which were de v e lop e d for application during this part of th e program. (No te th a t th es e s e nsitivity r e l a tionship s w e r e d e riv ed on th e b a sis of U.S. Customary unit s .

If S I units are employed, c onstant_ and c oefficients in the equations will require reevaluation.)

3 .3.1 Sensitivity of DOC to engine weight, 3FC, and maintenance cost. - The followin g pro ce dure p e rmitted trade-offs to be made of e n g ine cy c les or c on c epts in whi c h, for example, a more c omplex, heavier en g ine mi g hL deliver a re d u c ed SFC. To use this evaluation method, it w_s ne c essary that preliminary estimates 5e made of the installed S FC, wei g ht and main t enanc e requirements of th e propos e d en g ine relative to equivalent values for the baseline en g ine.

DOC = K . 89 + 0 .II M a lntbl ) w h e re : ¢ " DO C i s e v a luat e d in S n.mi.

. K = 1 .8 33 4 + 7 Maint = Estimat ed mai n te n a n ce ma nhou r s a nd m aterial r e lati ve t o th e Mai nt bl ba se lin e e n g in e b a sed on c o m plex i ty , oper atin g t e mp e ratu res , p ress u res , e t c .

W - W = Ch an ge i n w e i gh t o f t h e pr o p os e d e ng in e(s) or pr o pul- pro p p r oPb I s io " sys t em co mpa red t o t he f o llowing bas eline value s _ F s L s( in s talle d ) = 2 8 694 l b p er e n g in e W e i g ht (for 4 e n g in es): Engi ne s 22 141 I b Ex h aust (i n clu d ing t hr ,_t r ev.) 2006 Ib . Inlet s 255 8 ib Na c ell es 6559 lb S tart syste m 33 6 8 6 ib B thrust 4 x 28694 In sta l led • = 3.4 1 w eight 33 6 8 6 (Note that the pr o posed engine thrust level must b e the same as the baseline engine for a valid comparlson.)

= Estimated SFC of the proposed engine relative to values for _'IrCbl the baseline engine.

The scaling limits of this method are: +15% SFC -+4536 kg (+i0 000 ib) weight 3.3.2 Sensitivity of DOC to fuel pumping system power and weight. - This trade-off was for purpos e s of assessing the relative benefit (or liability) of weight vs pow e r requirements of c a ndidate pumping systems. ASSET vehicle synth e sis data w e re used, together with the baseline engine characteristics, to obtain an approximation of the horsepower-w e ight trade-off of tank-mounted ' aircraft fuel pumping systems. It was assumed that the pumps were driven by electric, hydraulic, or other suitable power source extracted from the engine accessory drive. The approach used was to compare systems on the basis of the incremental change in direct operating cost (DOC) as shown: DOC = 3.22 _hp cruise + 7.75 [Wsystem + 6 _ hPmax ] 105 _ 106 where: ¢ DOC is expressed in S n.mi.

hPcruis e = total input horsepow e r to all pumps running during cruise flight W = total weight of pumping system including pumps, drives, system installation, plumbing, etc.

6 _ hPmax = factor to account for the aircraft system installed weight penalty to provid e the necessary input power. This is based on th e total maximum horsepower of tank-mounted pumps. !

Q 2O r EXAMPLE : System No. l No. 2 i _ri_ _yp_ Electric Hydraulic

{

" 2 hp max 80 70 "2 hp cruise 50 40 Wt. pumps- ib 40 40 Wt. drives 58 90 Wt. installation 20 30 Wt. lines i00 120 W 218 280 system 7.75 " DOCsys No. 1 = 0.0000322 x 50 + 106 (218 + 6 x 80) t = 0.0016 + 0.00541 • = 0.00702 ¢ S n.mi.

7.75 DOCsyq No. 2 = 0.000322 x 40 + (280 + 6 x 70) 1 06 = 0.00129 + 0.00543 . - 0.006 7 2 ¢ S n . mt.

System No. 2 has the l o w e st incr e ment of DOC and would be favored over No. i. H owever, since this e valuation does not address the important aspects , of reliabil i ty and maint e nance, it c a n only be considered as a screenin g D means to eliminate the least likely candidates or to measure the relative impact of power consumption and weight trade-offs on the aircraft.

3.3.3 Sensitivity of DOC to volume and weight of fuel containment system. - The various candidate tank insulation systems and tank structural concepts offer trade-offs of thickness "_" (measured from inside surface of tank to exterior surface of aircraft) and weight. As thickness varies, the aircraft fuselage length must change to provide the required fuel volume within the fixed fuselage cross section. The following procedure, and associated values of influence coefficients, was derived from a matrix of ASSET cases which w e re run to simulate all reasonable combinations of fuel containment system thicknesses and weights. As noted earlier, all aircraft represented by the combinations of thickness and weight in the matrix are real in that they are sized to perform the design mission while meeting necessary design constraints.

I. Determine total fuel boiloff for the I0 190 km (5500 n.mi.) mission for the insulation concept and thickness being evaluated.

I 2. Calculate the fuel tank volume required using the _ollowing allowances: Integral i l Baseline New I Case Case i l • Ullage 2. 00% 2 .0 0 % i • Net tank contraction due 0.90 0.90 I to cooling*, plus expan- I sion due to i pressurization • Structure and equipment 0.64 0.64 • Trapped and unussble fuel 1.60 1.60 Subtotal 5.14 5.14 • Boiloff: Pr e ssurant gas 1.00 As Vented gas 2.56 Calculated Total Allowance (percent) 8.70 5.14 + % Boiloff *where insulation is on outside of tank " I 61 630 PH2 (Sat. liq. at 21 psia) Total Vol / tank - 2 x Total allowance (%) I+ For baselin e case: I (Ref) - 61 630 4.325 ft3 / Lank " 2 x 1.087 - 7746 3. From Figure 5, find the aircraft fuselage length (Lfus) knowing "t".

_. Calculate the installed weight of the total fuel containment system (_Wfcs). Assume weight of forward tank is same as aft tank.

NOTE: Installed weight includes tank, tank supports, shell struc- ture (nonintegral), insulation, adhesive, vapor barrier pro- tective cover, etc. For nonintegral tanks add th e weight of the fuselage shell structure in the tank region and the tank removal provisions. Assume forward tank shell specific weight is the same.

5. Calculate DOC using the following equation: DOC -- 1.8334 + 7.75 (_ Wfcs + _,WH2 _ 82 294 1 10 6 + - 224.3) 106 (Lfus whe r e: 1.8334 = Baseline aircraft DOC - ¢ / S n.mi.

(fuel cost - $6 / 106 Btu) 7.75 Wt. influenc e coeffic i ent 8 O ASSUME: !

VOL . FWD . TANK = VO L . AFT TA N K (t-in.) tocm 45 . 7 25 O (18) ++++ + 40. 4 75 - (16) BASELI N ES = N O N I N TEGRA L (14) 240 INTEGRAL _ E _ < +1 1 1 .2) z5 . , z + (11 Z _ ___ (8) ,,.,.I w 230 < -- (6) (4) uJ 4: ¢ n - (2) U.

I (o) 210 BASELIN E VOL. = 219 m 3 I (7746 ft 3) 200 [ 60 L 7 8 9 " - 1000 ft 3 t I I I 0. 2 00 0.220 0.240 0.260 VO LUM E / TAN K , _ 1000 m3 f Fi g u r e 5. - Fu se la ge l e n g th vs f u e l co ntainm e nt s y s t em thickn ess and tank volum e .

2 4 = Length influence coefficient 224.3 = Fuselage length (fe e t) for baseline integral tank (with t = 6.32 in.)

82 294 = _'Wfcs + _•WH2 for the baseline integral tank design _WH 2 = Weight of tanked liquid plus bot|_ pressurization and vented gas.

3.3.4 Sensitivity of DOC to aircraft ground losses. - The choice of fuel containment concept is influenced by the fuel boiloff losses during tank refueling and by the boiloff during the daily out-of-servic e periods. In the LH2 Airport Requirements Study (Reference 2) the economic desirability of collecting this vent gas for reliquefactlon was established. The follow- ing method of analyzing the worth of capturing and reliquefying this ground . boiloff hydrogen was derived to provide a basis for comparing competitive fuel containment system concepts. The assessment is in terms of DOC.

Assumptions: R I. The vapor is returned via a v a cuum jacketed c o llection header and insulated surge tank to the hydrogen liquefier at a point just upstream of heat exchanger X-8 (stream no. 56) as shown in Figure 7 of the report "Survey Study of the Efficiency and Economics of Hydrogen Liquefaction" (Reference 4). It enters at a pressure of 103 to ii0 kPa (15 to 16 psla) and a temperature of approximately 70°K (126°R). It then passes through the heat exchanger and, in turn, through the H2 flash and recycle compressors.

2. The cost of reliquefaction is assumed to consist of the cost of the electric power at 2C / kWh used in recompression, and a prorated share of the storage, distribution and plant costs using the discounted cash flow (DCF) method of accounting described on page 45 of R e ference 2.

On th_s basis the estimated cost of reliquefaction is: ($ / Ib) Electric power 0.05 Share of plant costs 0.027 V Total 0.077 !

The cost of the original GH 2 feedstock was, of course, included i_ the cost of the liquid hydrogen.

The effect of hydrogen boiloff losses on DOC of the baseline aircraft with a utilization of 4000 hours / year is: _WH 2 grnd.

_DOC grnd. = day x 3"75 x I0-6 ( ¢ )Sn_mi.

where: WH 2 grnd. = total of all daily ground boiloff losses 3.4 Calculation of Direct Operating Cost Dir ec t Op e rating Co s t (DOC) wa s used a s a primary se l ec tion c riteria in evaluation of design options for the LH 2 transport aircraft, and as a basis for comparing the economic performanc e of liquid hydrogen fueled air- craft of advanced design with that of conventionally fueled counterparts.

q Th e 1967 ATA DOC equations (Reference 35) were used as a starting point in the derivation of an improved method for calculating LH 2 and Jet A air- craft DOC. Th e 1967 ATA equations do not accurately reflect operation of either current or advanced technology aircraft and ther e fore required con- s iderable m odification. To provide a basis for reasonable evaluation of DOC for the subj e ct LH2 and Jet A aircraft, an e xt e nsiv e surv e y of airline op e rational practice s was made, actual CAB data analyzed, and Lockheed and e ngin e manufactur e r's sp e cialists consulted regarding probabl e maintenance r e quirements of the LH2 aircraft. Th e information derived as a result of thi s work wa_ used to formulate modifications to th e 1967 A T A formulas in t e rms of January 19 7 6 dollars for international trunk op e ration.

3.4.1 Background. - The DOC e l e ments, and variables which aff e ct th e ir eval- uation as reflected in the standard ATA formula, are listed in Tabl e 4.

The airline surveys involved a series of meetinzs with four major U.S.

air carriers to det e rmine the e lem e nts which are of significanc e to th e m and the param e ters and methods which they conventionally use in det e rmining P_'.

Th e following airlin e representativ e s cooperated in th e inv e stigation and c ontributed valuable data and advice: TABLE 4 . - COST ELEMENTS AN D VARIAB L ES IN 1967 ATA FORMULAS Evaluated as Functions of the Element Following Variables Crew cost GTOW Fuel and oil Lbs fuel, fuel cost, non-r e venue flying Maintenance Airframe Labor Airframe weight, speed, labor rate Material Airplane cost , Engine L a bor Thrust, labor r a te Material Engine cost Burden Ratio of m a intenance labor cost • Insurance % of airplan e cost Depr e c_atlon Cost of airplane, life, residual American Airlines - Mr. Jack Graef - Mr. Keith Grayson Pan American World Airways - Mr. William Hibbs | | Trans World Airlines - Mr. Walt Sherwood United Airlines - Mr. John Curry The participation of these airlines was solicited to provide representa- tion of a spectrum of route structures, operational procedures, and financial practices.* *It should be not e d that each of the airlines consulted has its own method- i ology for determining DOC according to its individual requir e ments. The method and procedure described herein should not be construed as represent- ing that of any o n e of the coop e rating airlin e s. Rather, th e method pre- s e nt e d in this r e port is the result of an attempt by Lockheed to represent _• nominal industry values, i !

27 i

|

Statistical a n alyses o f 19 73 , 1975 , a n d 1976 CA B a irline op e ratin g da ta were mad e in ord e r to provide realistic para m eters in addition to thos e pro- vided by the a irl_ n es. Various statistical te c hniques, including m uJtiple regression, w e r e used to id e ntify tr e nds in the data. The CA B data were also us e d to id e ntify a n d quantify variations betwe e n dom e stic and international trunk operation, variations betw e en airlines, and variations b etw e en types o f air f rames and e ngin e s. The variations between 1973, 19 7 5, and 19 7 6 d a ta were us e d to evaluat e and define trends and to provide e s ca lation factors.

Consult a tions with both airfr a me a nd engine spe c ialists w e re used to det e r- mine relative engine llfe, spares r e quir em ents, and maintenance valu e s f o r LH 2 op e ration.

Evaluation of DOC for the subject air c r a ft study was performed within the Lo c kheed AS S ET computer program which was used to dev e lop aircraft con- f igurations and mission p e rformance, as well as cost data . Airframe and e ngin e c os ts w e re d e rived fro m d e t a i l e d pa r a me tric for m u l a s within _SSET u s ing values for aircraft and engin e para m et e rs which w e re dev e lop e d f or the sub- J ect aircraft.

3.4.2 Para m eters re quired for DOC evaluation . - The following paragraphs present the basis for evaluation of the parameters involved in d e termination of DOC for th e sub j ect study.

3.4.2.1 Wei g ht: Weight is a pri m ary fa c t or in d e ve l o ping ai r cra f t co st a n d DOC. Formulas containing wei g ht as a parameter are bas e d upon w e i g ht-cost relationships resulting from current technology aircraft. When advanced materials such as composites are introduced, the historical w e lght-cost relationships ar e no lon g er valid and m ust be modified. These modifications w e re m ade by using weights equival e nt to current technology aircraft rather than c a lculated w e ights. Equivalent weights wer e us e d f or welght-r e lated parameters such as d e nsity.

3.4. 2 . 2 Aircraft c os t: Airfram e co s t and s chedul e w e r e bas e d u p on a fiv e a ircraft d e v e lopm e nt program and a 345- a ircraft producti o n for a t o tal of 3 50 aircraft. A maximu m production rat e of four aircraft per month wa s u se d.

Mo s t lab o r co s t s w e r e e s timat e d in term s of hour s with appli c able Lockh e ed- California C o mpany January 19 7 6 dir e ct and overh e ad rat es applied. Warr a nty c osts and a pro f it of 15 pe r cen t w e r e a d d e d, Engin e c osts w e r e bas e d upon us e o f th e e ngin e in two s e par a t e a ir c r a ft produ c tion progr a ms re quiring a tot a l of 3600 e ngin e s.

Air f r am e a nd e n gine sp ar e cos t s w ere es ti m at e d a s a pe r ce n t ag e of th e e n g in e or airfr a m e c o st . T h ese pe r ce nt age s w e r e d e r i v e d f rom cu rv e s prov ide d by T WA w h i c h r e l a t e p ercenta g e of spa r es to fleet s i ze . A f l ee t size of 20 w a s selec- t ed: 1 2 p e r cent was u sed for a i r f ra m e ; 29 p ercen t w as use d fo r th e con v ent i ona l !

,L t e ngin e , and 21 p ercent used for the LH 2 engine. The r e duction for the LH2 e ngines is considered appropriat e b e cause of an e x p e cted 30 percent increase In life for engines using that fuel. Pr e c e d e nt for this assumption lies in " experience usiL_g gas turbine engines fueled with natural gas to drive electri- cal g e n e rators and compressors in pipeline installations (Refer e nc e 44). In these applications with natural gas (85 percent m e thane) it has been observed that e ngine life and maintenance requirem e nts are both improved by about 20 to 25 percent compared to the same engines fuel e d with aviation kerosen e .

Th e oretical Justification for the additional improvement e xp e cted with hydrogen stems from considerations such as (I) gaseous hydrogen and air mix v e ry rapidly and thoroughly in the com b ustion chamber, which results in a very unifor m temperature profile, th e reby minimizing thermal stresses; (2) the very low e missivity of H2 / air com b ustion gas e s minimizes metal temperatures for a given t e mperature of the working fluid; (3) there ar e no carbon compounds to form coke or lacquer in the fuel lines, on the combustor walls, or in the turbine section; and (4) there is no sulfur or any other impurity in the fuel to cause either erosion or corrosion.

3.4.2.3 Mission characteristics: An average stage length (ASL) of 218 7 nautical miles was selected from prior route studies. This agrees _ery closely with Lockheed formulas for deriving ASL for an international route.

Block and fl_ght times were calculated in ASSET based upon the mission pro- file for the ASL. Block time equals flight time plus ground time. A utiliza- tion of 3993 block hours / year (10.9 block hours / day) is estimated from the ° Lockheed developed fomlula: U = 2 9 42 . 75 x (block tim e ) 0 "19 1 3.4.2.4 DOC elements: • Crew Cost. - An international crew cost of $450 / block hour was esti- mated from Lockheed-developed formula for 1973 domestic crew cost times a percentage for international bonus and adjusted for infla- tion from 1973 to January 1976. The formula and adjustment factors were derived from CAB data.

DCC (1976) = 38.38 x ASL 0'12 x OEW 0"202 ICe (1973) = DCC (1973) x I.i0 , ICC (J_N 1976) = ICC (1973) e scalated at 12.7Z / year Wh ere DCC = Do mes tic Cr e w Co s t I CC = In t er natio nal Cr e w C o s t OE W - O p e r a t in g E m p t y W e ig h t • Fu el Cost. - Fu e l c osts we r e giv e n by NA S A (s ee Tabl e I , G uid e lin e s a nd R e quir e m e nts), a ssuming bo t h fu e ls a r e produ ce d from c o a l and water in the 1990's .

LH 2 - $5.69 p e r GJ ($6 / 106 Btu = 3 1 ¢ / ib = 1 8 . 3¢ / gai) JET A = $4 . 74 p e r GJ ($5 / I06 Btu = 9. 2 c / Ib - 6 2.2¢ / gai ) Blo c k f u e l usag e is c al c ul a t e d by A SS ET.

A f a ctor o f 1.23 p e rcent for nonrev e nue f lying was applied, bas e d upon av e rage a irlin e op e ration s from CAB data.

• Maint enance Cost. - A maintenanc e labor rate of $9 .00 / hour was used as repres e ntative of the rates reported by the a irline s fro m the airlin e s urvey. A maint e nan c e burden fa c tor of 2 .27 was applied to m aintenan c e labor. Th e burden fa c tor was d e veloped f rom a s e l ec ted average for 197 5 es c alated at 3 perc e nt per year to January 1976.

Th e corr ec tion fa c tors for th e various el e ments of m a intenan c e ar e summarized in Table 5.

• In surance . - An av e rag e insuran ce rate ov e r the llfe of th e a ircraft of 0. 30 4 i s es ti m at e d f ro m an L CC -d e v e lop e d formul a .

Avg. Rat e = ( -1357.9 + 1 3 59 x Y e arsO'OOl ) / LIFE • D epreciation. - E s tima te d av er ag e air cr a f t c ost, in c luding spar e s l e ss r e sidual v a lu e , is divid e d by th e e stim a t e d li f e of the a ir c r a ft.

A 4-p e r ce nt r es idual valu e for wid e body w a s d e riv e d f rom t he a irlin e surv e y. T h e 16-y ea r llf e is normal for c urr e nt l a rg e a ir c r af t.

TABLE 5. - MAINTENANCE FACTORS FOR DOC CALCULATION Maintenance Correction Factors

® ® ® @ Total Factor

Maintenance Equiv. x ATA-to- LH2 Intntl Applied to X X = Element Weight Actual Cmplxty DMSTC ATA Formula Airframe Labor Jet A 1.408 0.52 1.0 1.07 0.783 LH2 1.388 0.52 1.02 1.01 0.788 Airframe (Uses (Incl.

Material cost) in cost) Jet A - 0.68 - 1.07 0.728 LH 2 - 0.68 - 1.07 0.717 Engine ( U ses Labor thrust) Jet A - 0.62 1.0 _.07 0.663 . LH 2 - 0.62 0.7532 1.07 0.50 Engin e (Uses Material cost) • Jet A - 1.31 1.0 1.07 1.402 LH2 - 1.31 0.7382 1.07 1.035 e quivalent @ - Airframe weight is us e d in airframe labor only. The weight factor adjusts the weight of advanced technology materials to an equivalent current technology weight.

- The ATA-to-actual ratio reflects a factor wh i ch must be applied to the ATA formulas to adjust to actual experi e nce on wide body air- fram e s and high bypass engines.

- The LH2 complexity factor accounts for var i ations between a Jet A- fueled aircraft and an LH2-fueled aircraft. A deta i led maintenance anal y sis of e a c h sub s yst e m indi c ated a n e t 2 pe r ce nt In c r ea s e in ai r - fram e lab o r for th e LH 2 -alrfram e . En g in e maint e n a n ce for th e LH 2 is reduced 30 percent from the J e t A-englne maint e nanc e for th e same reason s di s cus se d previously to account for lon g er llf e with LH2 e ngin e s.

@ - Th e Int e rnatlonal / domestlc adjustment is r e quired because ATA-to- actual f actor s were developed on dom e st i c t runk operation only and CAB data indicates a relativ e ly hi g her cost for international maintenance.

I

4. LH2 ENGINE DEFINITION Th e objectiv e of th e e ngin e d e finition task was to e_tablish a viabl e baseline concept for a liquid hydrogen-fu e l e d transport engin e considering th e requirem e nts of the aircraft, i. e ., mission profil e and p e rformanc e requir e - ments, and th e unique properties of the liquid hydrog e n fuel. Th e work was perform e d as follows: • The first element of this task addressed a feasibility investigation of various schemes to exploit the properties of hydrogen.

• The second ele m ent consisted of param e tric engine investigations oriented toward selecting cycle variables and the engine configura- tion which minimized direct operating cost. The factors considered in evaluating direct operating cost were specific fuel consumption and engine weight.

• The third element of the engine definition task was the detailed definition of the selected engine design. The definition included determining engine performance throughout the flight envelope; weight and geometry; scaling laws; engine estimated cost; noise and emission levels; and operatin" limits and capabilities.

• The final element consisted of for m ulating a list of technology development require m ents.

It is appropriate to point out that this task was not originally identi- fied as a major activity in the study. Although definition of an optimum de si gn o f a L H 2-fueled en gi n e is a topi c de s er vin g o f s e ri o us e ffort, it s e rved the purpo se s of th e pres e nt s tudy to limit th e work to a pr e liminary inv e stigation. A cc ordingly, th e re sults ar e pr e s e nt e d with th e r e s e rvation th a t m a ny o f th e d e sign c hoi ce s w e r e m a d e , n ece ss a rily, with l e ss th a n rigorous te c hni c al Justifi ca tion. A mu c h mor e c ompr e h e nsiv e d e sign study is r ec omm e nd e d to fully e xplor e th e pot e nti a l of LH 2 a s a fu e l for a dvan ce d turbof a n e ngin e s.

4.1 F e asibility St udi e s - Hydrog e n Exploi t ation Th e obj ec tiv e of this ta s k wa s to d e t e rmin e if th e uniqu e pr o pe rti es of hydro ge n c ould b e ca pit a li z e d on to p rovid e e ngin e p e rforman ce and / or w e i g ht b e n e fits. Th e c on cep ts whi c h w e r e s e l ec t e d for e v a luation In c lud e d: • C o mp r es s or a ir p r ec ooling • Com p r e s s or a i r in te r c ooling 3 2

I

a c • Cooling of turbine cooling air • Regenerative fuel heating • Expander cycle 4.1.1 Appr oach. - The approach used in th e feasibility studi e s was to s e lect a turbofan cycle compatible wltll the requirements of the liquid hydrogen- fu e l e d transport and to investigate the effects of the selected concepts on this baseline. Previous Lockheed work (Reference I) resulted in the defini- tion of a turbofan cycle for a liquid hydrog e n-fueled transport. Cheracter- istics of this cycle are shown in Table 6. The data shown in Table 6 were derived using AiResearch analysis and modeling t_chnlques and, therefore, differ sligbtly from Lockheed results as reported in Reference I. AIRese a _ch reviewed this _ycle and found it to be g e nerally consistent with technology projections for 1990. The bypass ratio and fan pressure ratio selected appeared to be high and low, r e sp ec tiv e ly, but the detailed param e teri c s r e qui._d to sele c t optimum valu e s wer e not completed until later in the study. Therefore, this cycle was used as a baseline for the hydrogen exploitation feasibility studi e s.

Th e high bypass ratio and low fan pressure ratio had little or no effect on th e results of the feasibility studies.

TABLE 6. - BASELINE ENGINE Maximum C ] _"h , i0 668 m Takeoff (35 000 feet) Paramet e r SLS, Std M = 0.85 Rot o r inlet t e mp e rature, °C 1416 13 7 9 (°F) (2580) (2514) Cycl e pressure ratio 35.2:1 41.13 Fan pr e ssur e ra t io 1 . 51:1 1. 634 C o r e pre ss u r e r at io 23.3 : 1 25 . 17 N o zz le -to -co r e - ve l o city r at io 1.022 1 . 17 Byp ass r a t io 12.95:1 13.0:!

N e t thrust, N 127 664 26 689 . ( i b) (28 700) (6 000) S p ec ifi c fu el c on s um p ti on (kg / h r ) / da N 0.096 0.2022 ((l b / hr ) / l b ) (0. 094 ) (0. 19 83) t S p ecif i c t h rust, N / ' k g / s ec ) 2 5 6 11 9 ( Ib / (ib / se c )) ( 26.1 0 ) ( 1 2. 14) # Two notable changes were made to the cycle shown _n Table 6 during the course of the feasibility studies. The first change adjusted the cycle for the low temperature of the hydrogen fuel as it entered the combustor. The _e¢ond change debited the cycle for the effects of turbine cooling air.

Cycle and performance characteristics associated with these changes are shown in Table 7.

The effects of turbine cooling air were incorporated only for the inves- tigation of the concept where compressor discharge air was cooled by the hydrogen fuel before it entered the turbine blades. Since the analysis method required evaluation of differential effects only, for all other concepts, zero turbine cooling air was assumed, The Lockheed-defined cycle assumed the use of sodium-potassium (NaK) fluid to cool the turbine. The NaK was cooled by the hydrogen fuel.

TA B LE 7. - BASELINE ENGINE C HARACTERISTICS (Ma x imum Cl imb, i0 668m ( 35 00G ft), M = 0. 8 5) Adjust e d Adjusted for C ycle and Performance for Fuel Turbine Ch a racteristics Temperature Cooling ° C Rotor inlet temperature, 1379 1379 (OF) (2514) (2514) ' Bypass ratio 13:1 13:1 Fan pressure ratio 1.634:1 1.634:1 Core pressure ratio 25.17:1 25.17:1 Nozzle-to-cor e velocity ratio 1.19 I.i0 Fuel temperature to combustor, OK 50 50 (OR) (9O) (90) Net thrust, N 26 689 26 689 (lb) (6000) (6000) Specific fuel consumption, (kg / hr) / daN 0.2082 0.2129 ((ib / hr) / Ib) (0.2042) (0.2088) Specific thrust, N / (kg / sec) 120 112 (lb / (ib / se c )) (1 2 . 2 0) (11.47) Horsepower extraction 12 5 125 Aircraft systems bleed extraction, % 4.1 4.1 Inlet total pr e ssure re c ov e ry 0.991 0.991 Noz z le thrust coefficients 0.995 0.995 , g All concepts were evaluated at the initial cruise flight conditions of i0 668m (35 000 feet), Mach 0.85. This flight condition determined engine sizing and was also typical of the cruise condition where the majority of . fuel is consumed.

The criteria used for evaluation of the concepts was direct operating cost. The sensitivity of direct operating cost to changes in specific fuel consumption and engine weight was based on a relationship presented in Section 3.3. The relationship used in the engine study was: [ SFC 7.75 (_englne weight) + I 332 _SFCbase - I) lO b _DOC (%) = x I00 DOCbase The change in specific fuel consumption was evaluated using a design point thermodynamic routine wh i ch allowed the various concepts to be modeled.

Engine weight for the various concepts was determined by adding the weight of the components associated with each concept to the baseline weight and adjt,_ting the baseline weight for changes in airflow, bypass ratio and tur- bine design considerations.

For all cycle investigations thrust, cycle pressure ratio, turbine inlet temperature, ano fan pressure ratio were held constant. Specific thrust , (FN / Wa) was held nearly constant by fixing the energy extraction of the low pressure turbine. This was accomplished by specifying a constant jet nozzle velocity ratio (Vcore / Vfan) in addition to the other constant parameters.

Holding specific thrust approximately constant allows the effects of the various concepts to be observed independently of propulsive efficiency changes.

It should be noted that holding the jet nozzle velocity ratio and fan pressure ratio constant does not hold sp e cific thrust exactly constant, but it results in only very minor changes in specific thrust and the analytical procedure is greatly simplified. The jet nozzle velocity ratio selected was 1.19 which was based on the original Lockheed cycle.

Installation effects that were included in the analysis were bl e ed and horsepower extraction for aircraft systems, inlet total pressure recovery, and exhaust system losses including fan scrubbing drag. Freestream cowl drag and inlet spillage drag were not included. To a first approximation, freestream cowl drag is a function of specific thrust and therefore, for thi s analysis, _s a constant. Spillage drag at the design point condition is insignificant.

" Other important assumptions inc]uded the temperature of the liquid hydro- g e n fuel at th e fuel pump ou t let, the specific heat of hydrogen, and the temperature of th e fuel into the combustor. The fuel pump outlet temperature of 50°K (90°R) was calculated based on an assumed temperature rise through fuel system lin e s and the temperature rise across the engine high pr e ssure fuel pump. Ov e r the range of temperatures encountered, the specific heat of LH2 is not constant but can be approximated by a constant 3.5 Btu / ib / °R.

The fuel temperature into th e combustor was assumed equal to the temperature out of th e last engine he a t exchanger for all concepts except the expander cycle. For that concept it was assumed equal to th e temperature out of the hydrogen expansion turbine, i 4.1.2 Compressor air precooling. - The concept incorporating compressor air precooling is shown in Figure 6. An annular heat e xchanger is requir e d in I the core stream in front of the compressor. Fuel would be routed to the } heat exchanger, entering at a temperature of approximately 50°K (dOOR) and, after I passing through the heat exchangel, to the burner at an increased tempera- 1_ tur e . Precooling th e c o mpr e ssor inlet air results in l e ss c o mpressi o n w o rk _ required and th e benefit is a reduced gas generator size. A second benefit i x is the fuel heating effect, As discussed, LH 2 would typically enter the tom- !

bustor at 50°K (90°R) and part of th e heat of combustion is required to heat the };, fuel to compressor discharge temperature. The elevated temp e rature of the i fuel at th e heat exchanger outlet minimizes this penalty. The benefit achieved !

is limited by the effectiveness and the air side pressure drop of the heat exchanger.

COMPRESSOR PRE C OOLER

• / / // I /// / // ' // 1 I

i l l / i I i ' / i il l I /A I i I FAN I I _ ._. __J I H 2 FUEL Figur e 6. - Sc hematic o f engin e c y c le with compressor precooling with H2 fuel.

3 6 t Th e r esults of this in v estigation ar e sh o wn in Figure 7 In terms of specific fuel consumpt i on versus precooler heat exchanger effectiveness and precool e r air side pressure drop, all for an exhaust nozzle veloclty ratio o f 1.19.

Pr e liminary heat ex c hanger design analysis indicated that an effective- ness (H2 side) of 0.8 and air side pressure drop of 6 percent was feasible.

This combination r e sults in a 1.86 percent improvem e nt in spe c ific fuel con- su m ption. Basic engine weight decrea&ed 29 kg (63 ib) but the heat exchanger added 34 kg (75 ib). The net e ff e ct was an improvement in DOC of ]..33 percent Detailed heat exchanger design condu c ted later in the study indicated a sev e re air side fre e zing p robl em . Recirculat_on of warm fuel was inv e sti- gated but did not solve the problem. Potential damage due to f_relgn object ingestion was also identified as a serious problem associated with this concept.

4.1.3 Compressor intercooling. - The potential performance improvement due to intercooling the compressor air at an intermediate point in its compres- sion process with the H2 fuel was evaluated. The benefit to the cycle was exp e cted to result from a reduction in core size due to compressor horse- " pow e r reduction per pound of core airflow a nd decrease in fuel flow due to heating of the fuel. An offsetting effect, as with compressor precooling, is the pressure drop on the air side of the intercooling heat exchanger.

' The cycle with co m pressor intercooling is illustrated in Figure 8. The point in the compression process selected for the heat exchanger was chosen as one giving approximately equal enthalpy rise in the compression process The results of this study are presented in Figure 9 in terms of specific fuel consumption _ , e_us intercooling effectiveness and p ressure drop , all for an e x haust nozzle velocity ratio of 1.19.

Pr e ilmlnary h e at exchanger design i ndi ca ted that an effectiven e ss (H2 s ide) of 0. 8 and an air side pr e ssure drop of 4 p e r c ent w as feasible. This combin a tion results in a 1.0 per c ent improvement in SFC relative to the base- lin e cycle. B a sic engine weight de c rea s ed 1 8 kg (40 ib) but the he a t e x changer a dd e d 45 kg (i00 ib). The net eff e c t was a n improvement in DOC of 0.57 perc e nt.

_ C ,,,I o i

- [

L C , I I \, _- 1 _ .

/ /i_t' / I I''l " J r- _ _ 4--#0 _" > 0 ,, o 4- ; i._ U m r- i_I 3 U 0 _ I-- 0 II _ _ I_l _ J l,I 0 N ¢ '_

- X /I

= d > / E • _ 14 ; 0 0 _ " ¢_,X_li_ i _ : l l _ 0 n II "- " _ _ . , .-i

° ° " - _ 1 ._ ' \

z - " _ - <:11" o o'_ 0 C _ " ( q l / J q / q l) 3 _S

L ....... I i .... I _ i

I, IEP / "_ ' O.-IS 3 8 • H P C I HPT LPT I I i , i FAN I I I -- I I I L , / ,I _ " - _ l • I H 2 FUEL , Figure 8. - Schematic of engine cycle wi t h compressor intercooling wi t h H 2 fuel.

4.1.4 Hydrogen cooling of turbine cooling air. - An evaluatio_ was made of the potential benefit which might be derived from cooling the HP turbine blade cooling air with the hydrogen fuel. Hydrogen cooling of the turbine cooling air would reduce the quantity of turbine cooling air extracted from the compressor and, simultaneously, would heat the hydrogen fuel.

A schematic of this concept is illustrated in Figure I0. Air for cool- ing the turbine hub is extracted from the compressor and routed to the turbine by conventional means. Hub cooling air is not cooled by the hydrogen as the flow requirements are set by the pumping characteristics of the turbine disks and n o t by heat transfer r e q u ir e ments. If th e r e was n o flow o f c o o l air through the cavities in front of or behind a disk, the air in the cavity would quickly reach t he tempera t ure of the main stream. Th_s is due to a significant recirculaticn between the cavity air that is pumped in a toroid a l flow pattern by the ro t ating disk and the high velocity main stream air. To maintain the cavi t y a ir at an acceptable level, cool air mus t be introduced into the cavity to avoid recirculation, or at least limit it. The quantity required is set by the rotating flow process m ore t han by the t e m perature , of th e a ir. If s om e r e clrculation w e re allowed, cooling the cavity purge / c ,q I-" • J f t .. (._) .,.n t'-., ,4 ,,, w d ,._ u J ,_ 0 3 0 o o _ -- I'- " Z _ "_ . - O0 ,i_ rj * II 3 C : _ i._ II m _ cO 0 *- .- r- _ j × C 0 N e,,I 0 0 "- " ", - 4 Z _ 0 E * II II u J _ c :_ ( IJ 0 0 _ - -- Z _ 14. I -- I_0 • r ,_ _ 0 L._ q L / ( _q / q L ) -3.-IS

[ t........ J...... I, _ ..... ._. J

_ 0 c_ _ c_ - o o _ 40 N_P / _-_ q - 3 .-I S H 2 FUEL

I

: I

' -- EX C HANGER I BLAD E

_ 4 HEAT 1 .... TURB I N E I I

I l I COOLING AIR I FAN NOZZLE

" I

FAN I.. _ --s TURBINE HUB COOLING AIR Figur e lO. - S c hemati c of c ooled turbin e c ooling air c y c le.

air could reduce the flow. However, design standards whJ c h account for worn seals, varying engin e power lev e ls, and allowabl e disk temperatur e environ- ment would need to b e established. Furth e rmore, diff e rent approa c hes to routing of the air to the disks would have to be established.

Turbine blade cooling air is extracted from the c ompressor and dire c ted to a heat exchanger and from there to th e HP turbine at a temperature lower th a n c ompr e ssor dis c harge temperature. The air used to cool the HP turbine vanes does not bypass any work produ c ing stag e s and, therefore, is consider e d n on c hargea b l e p ro vided th a t th e tu rb ine inl e t temperature is quoted at the rotor inlet rather than the combustor outlet. (For an e x planation of the terms chargeable and non c hargea b le, see Paragraph 4.2.3.1.4.) Cooling of the vane c ooling air would redu c e t h e amount of air required, but it would not impact the cy c le. Another reason for c ooling th e coolin g air would b e to reduce the van e flow in order to diminish the e ff e ct _; c ooling flow on turbine efficiency. As the vane cooling flow exits, it disrupts the vane flow field. At high flows, this effect can be significant. However, for the time period specified and at the turbine inlet temperature level assumed, the vane cooling flow is low enough to preclude significant efficiency penalties.

The analytical method used to evaluate the effects of cooling was to penalize the turbine efficiency for pumping losses, etc., and to assume that the bla4e and hub cooling air do no work in the turbine being cooled and re- enter the cycle behind the HP turbine. This assumption is valid only in the ca se of a sing l e_stage turbine. For a mu l t is t a ge turb i ne, coo l ing air byp as ses on l y th e stage being cooled and reenters the cycle behind that stage. Turbin e vanes in mult is tag e turbin es a l s o require d i f f eren_ handling.

The cooling air to a second stage vane bypasses the first turbine stage and there f ore, must be considered as chargeable cooling air.

Turbine efficiency was penalized for the amount of cooling air required.

In other words, turbine efficiency increased as required blade cooling decreased. The efficiency penalty schedule us e d was 0.2 points of efficiency for each percent cooling air.

The results of this study are summarized in Figures II, 12, and 13.

Pr e sented in Figures ii and 12, are curves of sp e cific fuel consumption, bypass ratio, and specific net thrust as functions of j et nozzle velocity ratio and heat exchanger effectiveness. Fi g ure 13 prese n ts curves of HP turbine e fficiency, and HP tur b ine cooling airflow ver s us heat exchanger e f fectiveness.

At a jet nozzle velocity ratio of 1.19 and a heat exchanger effective- ness of 0.8, the maximum benefit to the cycle was approximately 0.53 percent improvem e nt in specific fuel consumption. This improvement was relative to the base l ine cy c le with c ooling a ir (see e arli e r di s cussion of b aselines).

B a sic e n g ine wei g ht decr e a se d 12.2 k g ( 27 ib) per engin e but the heat ex c h a ng- ers a dded 4.5 kg (i0 ib) per en g in e . The net improvement in DOC w a s 0.41 per- c ent. The benefit to the c ycle was s m a ll b ec ause the pro j ected bl a de coolin g air requirement for the 19 8 5-1990 tim e frame is s mall.

Although the benefit to the cycle wa s s m all, hydrogen cooling of the blade cooling air s eemed promising from oth e r aspect s . It suggested that th e combined benefit of h i gher turbine inlet temperature plus inexpen s ive c oolin g might b e attractive. A cc ordin g ly, a mo r e d e tail ed study was c ondu c t e d at 1760°C (3 2 00°F) turbin e inl e t temperature. The results of this stud) will be dis c ussed in a subsequent section.

4.1.5 Fu el heating with exhaust gas. - The concept of regenerative fuel heating was suggested because of the effect of the low temperature fuel at the combustor inl e t. A s n o ted e arlier , the introduction of low temperature

!

• 14 o13 < PRECOOLER EFFECTIVENESS = 0 I1 0.22 I I i I N OTE S : .a I . HYDROGEN FUEL I N LET T EM PER A T U RE • . 4._ T O HE A T S X CH ANG ER " 5 0° K ( 90°R ) i.

_ : 2 . M AX . CRUISE OPE R ATING C O ND ITIO N 0.220 - ._.

.o A T lO 668 m ( 35 OOO FEET ), MA C H 0 . 8 5, , _ S T ' D A T MO SPHE R E , 3. H P C O M P RE SSO R DI S CHARGE TOT AL z c TE M PE RA TU R E - 7 86°K ( 1415 ° R ) P C O . 21 5 -- :) " %4_ NET THRU S T = 26 689 N I( 6000 L B ) J_ /

_, _ 0.2 " Z

"-,_, _._- _ . ,<_ c,oo, _ R_%, w,_ss i .. _

z 0. 2 1 0 -- o

t

0. 2 0 - _ U J u _ 0 . 20 . 0.6 0.8 .O I.2 1.4 1.6 C O R E N OZZLE - TO - FAN N OZZLE V ELO C IT Y R ATIO (Vc oRE / VFA N ) Fig u r e 11. - E ff ect on SF C o f cooling tu rbine coo l ing air w ith fuel at max. c ruis e.

I

NO T ES : 1. HY DROG E N FU E L IN LET TE M PERAT UR E TO 1 5. 0 -- H EATE X CHA NG E R = 5 0°K ( 90°R) 1 5 2. MA X . CRUISE OPERATING CONDITIONAT 1 0 668 m (3 5 000 f t) , M A CH 0 . 85 , STiD ATMOSPHERE 3. . ,oC O M P RESS O R D I S CHA R G E TOT A L TE M PE R AT e'R E = 7 86°K (14 1 5 °R )

14o --

E .D U _ m13.0 __ .o

o /

z _ 13 .....

F- Z __ ,2o - _ , I11 ' • _ p - , _ Z U U

_ II.O --,' :"

-

10.0 -- I 0 0.6 0. 8 I . 0 I. 2 I.4 .6

9 . 5

CORE NOZZLE-TO- F AN NOZZLE VELOCITY RAT I O, VCORE / V F A N S-2 1 0 55 Fi gu re 12. - E ff ect on specific ne t thr u st o f coollng tu rbln e coo ll ng 41r wit h fue l a t m ax. cr ui se.

0.91 NOTES: I . HYD RO G E NFU EL I N L E T T EM PE R AT URE T O u H EAT EXC HA N GE R = 5 0° K (9 0°R ) Z o w 2 . MAX. TURB I NE I NLE T TE M PE R AT UR E A T TAK E O F F = 17 5 6 °K ( 3 1 6 0° R ) 3. M A X I M U M COM P RESSOR DIS C H ARG E 0.905 - T EM P ERA TU RE = 881°K (15 8 5°R) ....

_ .

0 . 90 ............

0.04, • -.

O- J Ju I, >- _° " ) . - . O " 03 l . , / 2. I% _._ .T O T A L _ . _C OOLI NG A I R O- J

"°

B L A D E COO LI NG A I R ONL Y ,,,, 0 .0 1 I-- " I- • iii _ _ J i T i i ii ii J i _ i , w _ . . ....

" 0 o.2 0 .4 0. 6 o. 8 1.o

HP T URB I N E COOLING A I R H EAT EXCH ANG E R EFFECTI V ENESS F ig ure 13 . - HP tu rbi ne e ff i c ien cy a n d c ool i n g a ir s c hedul e s used i n study of fu el co oli n g of tu rb ine c o ol i n g ai r, !

fu e l into th e c ombustor r e sults in a p e rforman ce p e n a lty wh e n c o m par e d with th e fu e l t e mp e ratur e s typi ca l of c onv e ntion a l k e ros e n e typ e fu e ls. The s c heme is e ss e nti a lly a r e g e n e r a tiv e g a s turbin e c y c l e e x ce pt th a t th e e n e rgy is a dd e d to th e fu e l r a th e r th a n th e a ir. It "s illustr a t e d in Figur e 14. P a r e nth e ti ca lly it is int ere stin g to n ot e that, with th e hi g h pr e ssur e r a tio c y c l e s b e i n g e v a lu a t e d, re g e n era tion o f th e c ompr e ssor dis c h a rg e a i r would not b e f ea sibl e b eca us e its t e mp era tu re is high er th a n th a t o f th e e xh a ust g a s.

The r e sults of this study are pr e s e nt e d in Figur e s 15, 16, a nd 1 7 . P r e- s e nt e d a r e c urv e s o f sp ec i f i c n e t th r ust, a nd sp e c i f i c f u e l c onsumption a s f un c tions o f e xh a ust h ea t e x c h a ng e r e f fec tiv e n e ss (H 2 sid e ) a nd e xh a ust noz z l e v e lo c ity r a tio. Figur e 15 is f or th e ca s e w h e r e th e e xh a ust g a s sid e pr e ssur e drop is z e ro, Figu re 16 shows 4 p erce nt, a nd Figur e 1 7 shows 8 p e r ce nt.

Th e r e sults indi ca t e th a t this s c h e m e o ffe rs th e m a ximum b e n ef it o f a ny of th e c on ce pts e v a lu a t e d. Th e improv e m e nt in sp ec i f i c f u e l c onsumption a t a c om- bin a tion of 0.80 e ff ec tiv e n e ss a nd 4 p e r c ent p re ssur e drop at a 1.1 9 J e t nozzl e velo c ity ra tio is 4.31 p e r ce nt, r e l a tiv e to th e r e fer e n ce v a lu e a s sho wn on Figur e 16. Engin e w e ight in c r ea s e d 1 2 kg ( 2 7 Ib) a nd th e h ea t e x c h a ng e r a dd e d a n additional 11 2 pounds. Th e n e t improv e m e nt in DOC wa s 2 . 9 p e r ce nt r e lativ e to th e bas e lin e c y c l e wh e r e th e f u e l w a s introdu ce d into th e c ombust e r a t 5 0 K (90R ).

T h e d ecrea s e in e ngin e e ff ec tiv e n e ss whi c h would normally re sult f rom r e - moving h ea t from t he e xh a ust is mor e th a n mad e u p by the incr eas e in wo r k output f r om both th e fan a nd core en gin e s ec tion as a r e su l t of th e in c r ea s e in e nth a lpy of th e fu e l / a ir c ombustion produ c ts a ppli e d to both th e high- and lo w -pr e ssur e turbin e st a g e s, a nd a s a r e sult of th e multiplying f a ctor st e mming from th e i0:I byp a ss r a tio.

FAN H 2 F U EL HEAT EX CHA N GER 7 HPC f HPT LPT I I "---- H 2 F UEL @ F ig ure 1 4. - S c h e m a tic o f en gi n e c y cle w it h ex hau st gas hea ti n g o f H2 f u e l.

4 7 ¢ 4.1.6 _HI expander cycle. - A study was completed to evaluate the performance improvement which might be obtained by providing a_rcraft accessory horse- power using a hydrogen expansion turbine rather than extracting it from the engine through a gearbox. Originally, it was hoped that the expander cycle could provide some of the fan or compressor horsepower requirements. However, preliminary calculations showed that the fuel would have to be pumped to very high pres s ures to pr o vid e significant amounts of pow e r r e lative to t he r e quire- m e nts of the fan and compr ess or. For example 13 7 90 kPa, (2000 psi) is r e quir e d to obtain 580 hp, which i s only 5 p e rc e nt of the compre ss or hor se power. Furth e r increases in power would require higher hydrogen pressures since th e turbine flow rate which is the engine fuel flow rate, and the heat addition from the exhaust stream is constant. It is believed that th e increase in engine com- plexity requir e d to use the hydrogen expansion turbine to provide only a small amount of the compression horsepower requirem e nts is unwarranted.

The emphasis shifted to the investigation of providing aircraft acces- sory horsepow e r requirements. The expander cycle is illustrated in Figure 18.

This scheme consists of pumping the hydrog e n fuel to a pressure level above that required for delivery to the engine, heating it in an exhaust gas heat exchanger, and expanding it through a turbine. The benefit to the cycle was

/ co 4 .ozz I

" / A EXCHANGER_ / I BO OST PUM P FAN I L__ H 2 FUEL

I

I I. -- A C CESS O R Y SHAFT

' ( + HORS E POW E R

EX P AN SIO N TUR B I NE t Figure 18. - S che mati c of ex p a n de r c y c l e .

5 0 expected to r esult f r om the elimination of the accessory power load and the decrease in fuel flow due to fuel heating. There would also be a decrease • in engine weight associat u d with a smaiier engine gearbox, but it was assumed that was offset by the increased weight of the fuel pump and the additional weight of the hydrogen turbine and the associated gearing.

The heat exchanger effectiveness and pressure drop (_P / P) selected were 0.8 and 0.04, respectively. The pressure to which the fuel must be pumped to yield a net output of 125 hp for the aircraft accessories was calculated.

Only the power required to pump the hydrogen to pressures above 2482 kPa (360 psia) was charged to the expander cycle.

The relationship of net horsepower available to the aircraft accessories and the hydrogen tu_blne inlet pressure is shown in Figure 19. The fuel pump efficiency was selected as 65 percent and hydrogen turbine efficiency including mechanical losses was assumed to be 80 percent. The fuel tempera- ture out of the fuel pump was calculated as a function of fuel pressure rise.

The hydrogen temperature into the H ) turbine was calculated using all the engine exhaust flow and the effectiveness of 0.8. The hydrogen was expanded across the turbine to the required combustor inlet pressure of 360 psia.

The I I _ temperature at the outlet of the turbine was calculated based on a 90 percent adiabatic efficiency. Engine performance was based on the fuel temperature at the turbine outlet and includes the effects of exhaust gas pressure drop and cooling in the heat exchanger. Specific fuel consumption and core jet-to-fan duct velocity ratio are shown versus bypass ratio in Figure 20. The specific fuel consumption at a core-to-fan nozzle velocity ratio of 1.19 is 0.1993 (kg / hr) / daN (0.1954 (ib / hr) / ib which is a 4.31 percent improvement over the baseline SFC of 0.2042 (Ib / hr) / \b listed in Table 7.

The regenerator weight is 51 kg (112 ]b) engine and the decrease in engine weight Js 12 . 2 kg (27 ib) / engine. The improvement in DOC is 2.9 per c ent .

a.l.7 Selection of preferred concepts. - A summary of the feasibillty studies is included as Table 8. The concepts which yield the largest reduction in DOC are fuel heating and the expander cycle. The fuel heating concept was recommended for use in the study as it is less complex and provides an equal DOC benefit. As noted ear l ier, however, hydrogen cooling of the turbine cooling air is attractive in many resp,_cts and offers advantages if higher turbine inlet temperatures were selected. This concept is further dis- cussed in section 4.2.3.1.

4.2 Cycle Definition and Configuration Definition The cycle definition and configuration work was accomplished in three parts. The first part was a review of prior studies of advanced turbofan engines. The second part was selection of the initial cycle for a LH2-fueled • e n g ine. It r e lied heavily on thes e pri o r studies and cycle variables were 800 .............

5000 -- 7 00 m

- f

_ z 4000 v W Z _ _ _ 600 // ' 3 0 00 - 40 80 120 160 200 NET OUTPUT SHAFT HORSEPOWER l I , l I 20 60 I00 140 NET OUTPUT POWER, kW Figure 19. - N e t o u tput shaft h o rs ep ower v s turbine inlet pressure.

I -- CO Z .0 CM _ _ O c:_ eY U

-

l-J O e,_ , -t L- I L l N _' 4 N X r , 4 !

n o u _ L _ N

_ _ _ _ o

_ " 0 N ", 4 o _ _ r_ 0 c_l 0 _ I - , * r0 _ ' _ _ . _ II II I.¢.I _ .M 0 0 O.

Z _ Li. I " -" 0 0 C) O _ TABLE 8. - HYDROGEN EXPLOITATION SUMMARY _P _SFC** Engine HX _DOC* fH2 Pair % _wt _wt % Precooling 0.8 0.06 -1.86 -63 + 75 -1.33 Inter cooling 0.8 0.04 -0.93 -40 +I00 -0.57 Cool_d turbine cooling air 0.8 N / A -0.53 -27 + I0 -0.41 Fuel heating 0.8 0.04 -4.31 +27 +112 -2.90 H2 Expander cycle 0.8 0.04 -4.31 +27 +412 -2.90 **Relative to the baseline SFC = 0.2042 (.l--b) / Ib hr 7.75 (_wt) + 1.332 SFC 1 106 SFC_I *ADOC(%) = x I00 DOCbase selected consistent with them. In the third part, a more detailed study was made of a high temperature, high pressure ratio cycle. Work completed in the second part was updated and compared to the results of the high tempera- ture study.

4.2.1 Review of previous studies. - In a previous LH2 transport study (Ref. I), Lockheed established a baseline engine cycle which is defined in Table 6. More recently, the General Electric Company and Pratt and Whitney Aircraft have studied turbofan engines designed for low fuel consumption under contract to NASA Lewis. The work is reported in References 36, 37, and 38. These studies were reviewed in detail as they represent the latest engine cycle and config- uration studies for conventionally fueled subsonic transports which might become operational in the late 1980's, and because they were of a much greater depth and scope than the engine studies accomplished in the subject program.

The aircraft used in these studies are very similar to the LH2-fueled trans- port in mission and payload. It was considered that cycle and configuration characteristics for LH2-fueled transport engines should be very similar to the cycle characteristics selected by G.E. and P&W Jn the E3 (Ener_v Efficient Engine) study. A brief summary of the results of the General Electric and Pratt and Whitney Studies is given in the following sections.

4.2.1.1 Turbine temperature and cycle pressure ratio: General Electric found that low-pressure turbine (LPT) cooling requirements became the over- riding factor at rotor inlet temperatures greater than 1538°C (2800°F). Min- imum SFC occurred at 1427°C (2600°F) for cycle pressure ratios from 32:1 to > 45:1. Minimum engine weight occurred at approximately 1538oc (2800OF) for cycle pressure ratios from 32:1 to 45:1. Minimum DOC occurred between 1482 and 1510°C (2700 and 2750°F) for cycle pressure ratios from 32:1 to 45:1.

r GE selected a 1427°C (2600°F) takeoff rotor inlet temperature and a cycle pressure ratio of 38:1.

• Following similar logic, Pratt and Whitney selected a 1427°C (2600°F) maximum combustor outlet temperature and a cycle pressure ratio of 45:1.

4.2.1.2 Fan pressure ratio and bypass ratio: General Electric's results showed that higher fan pressure ratios (up to 1.8:1) yielded improved direct operating cost (DOC). GE selected a fan pressure ratio of 1.7:1 for initial rating with possible growth to 1.8:1. For a separate flow exhaust system, an exhaust nozzle velocity ratio of approximately 1.5 yielded minimum DOC.

Pratt and Whitney found that the lower SFC possible at low fan pressure ratio and higher bypass ratios was offset completely by increased propulsion !

system weight. Nacelle drag drives high bypass ratio engines to more compact, high fan pressure ratio levels. A fan pressure ratio of 1.7:1 and a bypass ratio of 8.0:1 was selected.

4.2.1.3 Engine configuration: The General Electric Energy Efficient Engine is comprised of a single-stage fan driven by a 4-I / 2 stage low-pressure tur- i bine; a three-stage low-pressure ratio compressor providing a pressure ratio of 1 7 driven by the LPT; a nine-stage high-pressure compressor providing a pressure ratio of approximately 14:1 driven by a single-stage cooled axial turbine; a double-annular combustor; and a mixed flow exhaust system. I !

The selected P&W engine consists of a high-speed, single-stage 1.7 pres- sure ratio fan, a three-stage low-pressure compressor with a pressure ratio of 1.53, and a two-stage, 18.2:1 pressure ratio high-pressure compressor.

A low emission, two-stage vortex c_mbustor with aerating pilot nozzles is included to provide a 1427°C (2600UF) maximum average combustor exit tempera- ture. The compression system is powered by a two-stage, cooled high-pressure turbine and a five-stage low-pressure turbine. The exhaust system consists of a fan nozzle and a core nozzle.

4. 2 . 2 Initial L H 2 e ngi ne c _ c l e s e l e ction. - The initial LH 2 e n g in e c y c l e s e lection procee d ed o n t h e b asis t h at a rotor inlet tempe r aE u re o f 1427°C ( 2 600°F) t o 153 8 °C ( 28 00°F) was optimum. Th e assumption was based on findings th a t show t e mp e r a tur e s a bov e this l e v e l r e quir e c ooling for th e low-pr e ssur e turbin e v a n e s a nd bl a des. Cooling th e low-pr e ssur e turbin e r e sults in sig- nifi ca nt p e rform a n ce p e n a lti e s and is e xp e nsiv e . This a ssumption wa s t e st e d l a t e r in th e study through th e investig a tion of a 1760°C (3 2 00°F) e ngin e th a t used hydrog e n to cool the turbine cooling air and thereby minimize the per- , forman c e p e nalty. The results of this inv e stigation ar e c overed in a subse- qu en t sec tion.

4. 2 . 2 .1 Ba selin e eng in e de s cr i p tion: The b a s e lin e e n g in e ch osen for th e initi a l c y c l e se l ec tion s t udy iR a two- s pool, separa tel y exh a u st e d tu r b o fan c onsisting of th e following c ompo ne nts: % Lp _.- ....... _.... , ........................ .._ . _,..,._._._ ..__ ..... _, , ,_ , -r_ ,n_-_._r_mL _w_-_ ................... - Y • Single-stage fan • Two-stage low-pressure compressor (booster stages) • Ten-stage high-pressure compressor • Annular combustor • Axial cooled HP turbine (single stage) • Axial uncooled fan turbine (4-6 stages) • Exhaust regenerator (for fuel heating) • Separate fan and core convergent exhaust nozzles The cycle characteristics of the baseline engine were selected to approxl- ma t e the cycle used in the feasibility studies discussed in 4.1; however, additional intercomponent pressure drops, cooling flows and leakage were added.

A definition of the b a seline cycle for the cycle selection studies is pro- vided in Table 9. Also shown in Table 9 are three other cycles which pro- vide a summary of how the cycle was changed from the initial Lockheed cycle to the baseline cycle derived for the cycle selection studies. The cycle labeled 4 is the baseline cycle used in the hydrogen exploitation feasibility, studies and is quite close to the original Lockheed cycle (see 4.1.1). Cycle number 3 resulted when the exhaust regenerator was added for fuel heating.

A 4.3 percent improvement in specific fuel consumption resulted when the exhaust regenerator was added.

There was some optimism in cycle 3, however, and cycle nomber 2 incorpo- rated the following changes: • HP turbine efficiency was reduced I point to 0.90 to allow for losses due to cooling.

• LP turbine efficiency was reduced 3 poi._s to 0.88 as a result of turbine preliminary design.

• 21 horsepower allowed for bearing losses, etc., and to drive engine accessories.

• Fan duct and intercompressor pressure drops were modified.

J • 3.5 p e rc e nt tu r bine c ooling ai r was add e d.

Th e n e t e ffect of these changes was to increase sp e cific fuel consump- tion 7 .8 p e rcent with re s p e ct to the cy c le nu m b e r 3.

The b a selin e cycle for c ycle s e l e ction studi es r e sult e d from Slow path and compon e nt analysis. Th e core compre ss or ratio of 25.17 postulat e d for cycl e s 2, 3, 4 wa s considered too high. Ext e nsiv e varia b le geom e try (all stag e s) would be requir e d and th e tur b in e work lev e ls for a singl e stage TABLE 9. - BASELINE ENGINE CYCLE, INITIAL CYCLE SELECTION • I 2 3 4 E x h a ust Bas e line Adjusted I H e at i ng Feasib i li t y Bas e li ne Cy c l e St u d y S tu d i e s F_ Inl et Correcte d F low, k g / s e c 6 8 4 659 61 8 619 (Ib / se¢) (1507) (145 3 ) (1362) (1364) Press u r e R a tio 1. 6 1.6 1. 6 34 1.634 A d i a ba t i c E ffi c l en e _ ' 0 .8 92 0. 8 92 0 . 8 89 0 .8 89 Bypass Ra t i o 12.0:1 12 . 0:i 12.5:1 1 3 .0:1 L P Com pressor (Bo oster ) P ress u re R at i o 1. 3 - A diabat i c Effi c i e nc y 0. 8 65 - - HP C o mp ressor P ressure Rat i o 19.5 2 5 .17 25. 1 7 2 5 . 17 Ad ia bat ic Ef f ici e nc y 0.8 6 2 0.8 62 0 .86 2 0 . 8 62 Com bus t or !

E f fi c ie ncy 1 . 0 1 . 0 1 . 0 1 . 0 P r essur e Drop, _P / P 0.0 4 5 0 . 045 0.045 0.045 I Hp T u rbine R o to r lnlet T o t al T e mpe ra tu re , °C 13 7 9 13 7 9 1379 1379 (OF) (2514) (2514) (2514) (2514) A d i a ba tic E ff i c iency 0 .90 0. 90 0 . 91 0 .91 I Horsepo we r Extracti o n 125 125 12 5 125 , LP Tu r bin e Inl et Total Te m per a ture , °C 959 948 975 9 7 8 (OF) (1 7 5 8 ) (173 8 ) ( 1 7 8 7) (1793) Ad i a b a t i c E ff i c i e n cy 0 .88 0. 8 8 0 . 91 0.91 Horsep ow e r E xt ra ct ion 21 21 0 0 E x hau st R egener a tor E ffect i veness 0. 8 0.8 0. 8 - Gas S i de P ress u re D r o p ( _P / P) 0 .04 0.04 0.04 - Co re N o zz t e Thr u s t C oeff i c i e nt 0.9 88 0. 9 8 8 0 . 995 0 .995 F a n N o z z l e Thr u s t Co ef fi c i en t 0. 9 8 0 . 98 0 .99 5 0 .995 Fa n Du c t Pr ess ur e Lo ss , AP / P 0. 015 0 . 0 15 0. 03 0 . 03 LPC-HPC Pr e s s u re Lo ss A P / P 0.0 1 5 0 . 015 0 0 L PT -N o zzl e Press u re L os s, AP / P 0 .005 0 .0 05 0.0 0 5 0 . 0 0 5 Ai rcr a ft Bl ee d Ex t r a c tion, Perce n t 4.1 4.1 4 . 1 4 . 1 T ur bin e C ooli ng A ir , P erce nt 3. 5 3. 5 0 0 L e akag e , P ercent 1.0 0 0 0 Net T h ru st, N 2 9 68 7 2 9 5 76 29 68 7 2 9 b 8 7 • (I b ) (66 7 4) (6649) ( 6 674) ( 6674) S pe c i f i c Fu e l Co ns u mp t ion, (k g / hr ) / d aN 0 . 2 1 8 3 0.2 14 8 0. 199 3 0.208 2 (Ib / h r ) / Ib ) (0. 2 141 ) (0. 21 061 (0.1954 ) (0.204 2) Net T hru s t, N (I nc l u d es N ace lle D rag ) 2 6 68 9 26 68 9 2 6 6 8 9 26 689 (Ib ) (6000 ) (600O ) ( 6 000 ) (6000 ) " S F C. ( kg / hr ) / daN (In c lu des Na ce ll e D r a g ) 0 .24 28 0. 2379 0 . 22 1 7 0 . 23 1 7 (I b / hr ) / Ib ) (0. 23 8 1 ) (0 . 2 3331 (0.2 174) (0 . 2 2 7 2) 5 7 !

high-p re ssure turbine w e r e consid e red exce ssiv e . To maintain c y c le p re s s u re ratio at 40: 1 with a lower core pressure ratio, booster stages were _44_!

to the fan spool. Provis i ons for seal leakage ( I percent) were also incorpo- rated. This resulted in a 1.7 percent increa se in specific fuel c " _umption with respect to cycle number 2.

With respect to the cycle defined by Lockheed in th e previous study, specific fuel consumption was higher by 7.7 percent, including th e e ffe c t of the exhaust reg e nerator.

The flight condit i on chosen for the cycle sel e ction studies was for in i - tial crui s e at i0 788m ( 35 000 feet), M = 0.85, max i mum cl i mb power setting.

The studies were made on the basis of installed performance and included the effects of cowl drag but not inlet spillage drag.

The selection crit e ria for all investigations was minimum DOC. DOC was evaluat e d using the equation given in 4. 1 .1.

4.2.2.2 Selection of rotor inlet temperature and cycle pressur e ratio: As stated e arlier, turbine inl e t t e mpe r ature was limited to a restricted rang e 1427-1538°C (2600-2800°F). The G.E. Energy Effici e nt Engine study showed minimum DOC occuring b e tw e en 1482-1510°C (2700-2750°F). A maximum rotor inl e t temperatur e of 1482°C (2700°F) was s e lect e d on th e basis of the prior studi e s.

Cycle pressure ratio was selected primarily on th e basis of utilizing a single stage high pressure turbine. Although extensive tradeoff studies cou]d be made addressing single and two stage HP turbines, AiResearch experi- ence has shown that a single stage turbin e minimizes turbin e cooling required and that minimizing the number of cooled stages results in lower e ngine cost.

Th e a c tual maximmn turbin e work l e vel is based on tip spe ed , flow path, cooling and stresses, but for cycle sel e ction a value of 488.1 kJ / kg (210 Btu / Ib) at the temp e rature s e lected is a reasonabl e maximum. Likewis e , the maximum core compressor ratio is rightfully the subj e ct of a detail e d study but for this program, th e selection of a 20:1 maximum core pressure ratio is r e asonable and avoids consideration of mismatch and stability problems.

Shown in Figure 21 is the relationship of high pr e ssur e turbine work, fa n pr e ssure ratio and high pr e ssur e compressor pressure ratio. A maximum tur- bine inl e t temperature of 1482°C (2 , J0°F) and a booster pressure ratio of 1.30:1 was assum e d. Within the constraints of 488.1 kJ / kg ( 210 Btu / ib) turbin e work and core c o mpressor ratio, an ov e rall pressure ratio of 45:1 is r e ach e d only with high fan pressure ratios. At an overall pressur e ratio of 40:1, fan pr e s- !

sur e ratio can b e approximately 1.55 to 2.0:1. Furthermor e , a cycl e pr e ssur e ratio of 40:1 achi e v e s most of the benefit of i_igh cycl e pressur e ratio. Based on th e s e consid e rations, cycl e pressur e ratio was s e l e cted at 40:1.

' 5.8 - 250 .....

FPR R I T ,,1379°C (2514° F ) 2.2

R B - , . 3 : , I

5 . 6 -- 240 -- lO 668 m (35 000 FT )_ • , - 5.4 -

._s . 2 -

u_ -- 220 / _5. 0 - 6 _ - _ o

_ 21o • --

o.

" _ 200 4 , 6 -

/

4.2 - 18o / I

4.0 -- 170 ....... L

,6 , / IB _9 20 2, 22

CORE PRESSURE RATIO F i g u r e 21. - S electio n o f c y cle ov e ral l pr e ss u r e ra ti o .

: 59 4.2.2.3 Fan pressure ratio and bypass ratio selection: Fan pressure ratios from 1.4 to 2.2 were studied. The turbine inlet t e mperature of 1482°C (2700°F) maximum (1379°C (2514°F) cruise) and the maximum cruise cycle pressure ratio of 40:1 were held constant in this po[tion of this study. Primary stream energy extraction was varied by considering a range of bypass ratios.

Fan pressure ratio was select e d on the basis of minimum DOC which includes the effects of specific fuel consumption and engine weight. Maintenance was not included.

The baseline e ngine is as described in 4.2.2.1. Figure 22 shows specific fuel consumption versus exhaust nozzle velocity ratio for the range of fan pressure ratios considered. Figure 23 shows bypass ratio versus fan pressure ratio and exhaust nozzle velocity ratio. Figure 24 shows estimated engine relative weight versus fan pressure ratio and bypass ratio.

The trends in DOC are shown in Figure 25. Each fan pressure ratio has a minimum DOC. The curve of minimum DOC and fan pressure ratio is shown in Figure 26. Based cn this curve, the fan pressure ratio for minimum DOC is 1.7:1. The related bypass ratio is 9.3:1 as shown in Figure 27.

4.2.3 Hi gh temperatlre investigation. - As th e results of the hydrog e n exploitation studies and the initial cycle s e lection work became availabl e , it was apparent that some ben e fit might accrue to a high temp e rature cycle which used hydrogen to cool the cooling air and th e reby m inimize th e coolin penalty for both the high and low pressure turbines. In order to tak e full advantage of th e higher turbine inlet t e mperatur e s, high e r cycle pre ss ur e ratios are requir e d and a two-stage, high pressur e turbine becom e s necessary.

At a fan pr e ssure ratio of 1.6, for example, high pres s ure compr e ssor pr es sur e ratio would be between 21.5:1 and 26:1 at cycle pr e s s ure ratios b e tween 50:I and 60:1 respectively. Attaining the s e pres s ure ratios in a r e asonable number of stages and avoiding mismat c h and stability problems i s quite a formidable task without even considerin g weight, complexity and c ost p e nalties.

Th e introduction of a two-sta g e high pres s ure turbine and a cooled fan turbin e r e quired a m ore detailed turbine coolin g flow analysis.

Turbin e inl e t t e mperatur e was h e ld constant at 1760°C ( 32 00°F). This temp e ratur e was arbitrarily s e l e ct e d but is rep re s e ntativ e of the maximum tur- bin e inl e t temperatur e feasible in th e study time p e riod. Cycle pr e ssur e ratios o f 40, 50, a nd 60 wer e inv e stigat ed . Fan pressur e ratio was also varie d althou g h it was b e li e v e d that chang e s f r o m th e fan pr e ssur e rati_ s e lected in th e e arli e r investigation would b e s ec ond order.

!

6 1 LH2 ENGINE I NITIA L CYC L E SE L ECTI ON I 0 668 m (3 5 000 ft), 0. 8 5 M, M A X I MUM CRU I SE SE TT ING)

I

TIT = 1379 ° C ( 2 514°F) I CPR = 40:1

I

25 LPPR = 1 .3 : 1 ( bo o s ter stages) EXHAUST REGENERATOR I FN - Dcowl = 26 689 N (6000 Ib) I • 2 0 -- ] -- ,, - , o ii i .0 .2 .4 1.6 . 8 2 .0 2 , 2 CORE TO - FAN NOZZLE VELOCITY RATIO Fi g ure 23, - E _f e ct o5 fan p r essure ratio an d nozzle vel o city ra t io o n bypass ratio.

LH2 ENG I NE INITIAL CYCLE SELECT I ON 1.8 T I T = 1 482°C ( MAX) 700°F) / 1 6 CPR = 40:I EXHAUST REGENEIATOR • LPPR = 1.3:1 ( o oster stages)I = 1 . 4 / _ 1.2 Z Z _ 1.0 - - FAN PRESSIRE RATIO 0.8 / . . " 2 , 0 I 2.2 _ L i O, t _ ..............................

0 5 0 5 20 25 BYPASS RATIO F ig u r e 24. - Effect o f f a n pres sur e r ati o and b y p a ss r a ti o o n e n g ine wei g h t .

, s J • ., . ,, ,., mJ . _, _ - LH2 ENGINE INITIAL CYCLE SELECTION 10 668 " m (35 000 ft), 0.85 M, MAXIMUM CRUISE SETTING) I0.... I 2.01.8 1.6

\

FAN PRESSURE RATIO

6 \

1.4 O ,',4 W _ q "r

CPR - 4 0' 1 I

L PP R - 1 .3 :1 ( boos r st a g e s) EX HAUST R EGENE RA TO R F N - D co w l = 2 6 6 8 9 ( 6 0 0 0 I b ) -4 .......... I I 2 4 6 8 0 12 14 BYPASS RATIO Figu r e 2 5. - E f fe c t o f f a n pressu r e ra tio a nd byp a ss ra tio on direct operating cost.

LH 2 ENGINE INI TI A L CYC L E SELEC TI ON I 0 6689 M (35 000 ft), 0. 8 5 M, MAXIMUM CRUISE SET T ING) 4 I TIT = 1379 ° C (2514°F) C PR = 40: 1 L P PR = 1.3 : 1 ( b o o s ter s tages) EXHAUST REGENER A TOR F N - Oc o w l = 2 6 6 8 9 N (6000 Ib) 2 - " O _ 0 Z -2 _D FAN PRESSURE RATIO -4 i .4 1.6 1 . 8 2.0 2.2 FAN PRESSURE RATIO P Figure 26. - Effect of fan pressure r a tio on ch a ng e In direct ope r ating c o st, LH. ENGINE INITIAL CYCLE SELECI ! ON 10 668 m (35 Z o00 ft), 0.8 5 M , M AXI M U M CRUISE SETTING) I 4 I I : TIT = 137 9°C (2514 0F) CP R = 4 0:1 LPPR = 1 . 3 : 1 (booster stages) EXHAUST REGENERATOR FN - DCO W L= 26 689 N (6000 l b) 12 ,-.

F - m I0 Q.

"_ RA TIO I N SELEC T E D C YC L E _ ., 1 .4 1 .6 1 . 8 2 . 0 2 . 2 FAN PRESSURE RATO Fi gure 27 . - Bypass r ati o vs fan pr e ssure r atio.

66 c _ Ill I I I I II I The flight condition, installation factors, and other related assump- tions were identical to those assumed earlier. The baseline engine is identical to that described in 4.2.2.1.

Cooling flow requirements were defined at maximum power, hot day, sea level. An engine having a fan pressure ratio of 1.7:1 and a bypass ratio of 9.3:1 was assumed in computing the cooling flow requirements. A schemati c representation of the hydroge n cee!ing of the turbine cooling air was shown previously in Figure I0.

4.2.3.1 Turbine cooling requirements: 4.2.3.1.1 Engine operating conditions: Turbine cooling requirements were establish e d for hot day, sea level, maximum power conditions. Cycle tempera- tures are related to temperatures at the cruise conditions using ratios established in prior studies. The temperatures that are important are com- pressor discharge temperature, combustor outlet temperature, high pressure rotor inl e t temperature, and low pressure turbine inlet temperature. These locations are shown schematically in Figure 28. The temperatures used in this study are shown in Table 10.

Temperatures at turbine stations other than those listed in Table 10 were calculated based on equal t e mperature drop across each turbine stage.

For example, if the _T across the low pressure turbine is 1000 ° and there are four stages, tne temperature drop across each stage is assum e d to be 250 ° . The temperatures between each blade or vane row are based c the mass averaged temperatures of the gas stream and the cooling flows. Also, combustor outlet temperature is that required to provide the rated rotor inlet temperature after mixing of the first high pressure turbine vane cool- ing air. i !

TABLE i0. - INTERNAL CYCLE TEMPERATURES (SEA LEVEL, HOT DAY, TAKEOFF THRUST) Cycle Pressure Ratio Station 40:1 50:1 60:1 Compressor Discharge i' 3 859 (1546) 916 (1648) 968 (1742) Combustor Outlet T3. 9 2087 (3756) 2087 (3756) 2087 (3756) ° HP Rotor Inlet T 4 203 3 ( 3 660) 2 0 33 ( 3 660) 2 0 33 ( 3 660) LF Turbine Inlet T4. 2 1617 (2910) 1 577 (2838) 153 7 (2766) ...... i " T e mperatures in OK (OR) 6 8

I

4.2.3.1.2 Turbine cooling air heat exchanger: The turbine cooling air heat exchanger is similar to those used in prior work which has been dis- cussed earlier. At first, it was assumed that the heat exchanger had an effectiveness of 0.8. Further study indicated that the resulting cooling air outlet temperature was approximately 211°K (380°R). Heat e x c hang e r |, freezing would b e a problem at this temperature and cooling air outlet temperature was limited to 311°K (560°R) to avoid this problem, A bypass _ arrangement or a lower effectiveness heat exchanger is required. (Refer to heat exchanger design, Section 4.3.6.) i | J 4.2.3.1.3 Turbine design criteria: Minimum design crit_ , rla were established _ to allow the determination of turbine cooling requirements. The criteria | established included the follo_Ing: _ I R • Allowable metal temperatures !

• Combustor pattern factor • Blade and vane heat transfer effectiveness • Blade relative gas temperature • Turbine work limits • Turbine cooling efficiency penalties The allowable metal temperatures assumed for this study are shown in Table ii.

They are based on the results and prolectlons of turbine material technology pro- grams and are applicable t o l o ng-life transport engines. !, TABLE l l. - ALLOWABLE METAL TE}_E_TURE LIMITS !

i Temperature I,tmit t ltigh Pressure Turbine Vanes 1204°C (2200°F) High Pressure Turbine Blades 1093°C (2000°F) Low Pressur e Turbine Vanes i149°C (2100°F) Low Pressure Turbine Blades i093°C (2000°F) J L l I The combustor pattern factor (T3.9max - T3.9avg) / T3. gavm - T 3) deter- mines the peak temper a ture that the turbine vanes feel. Bla_es are not influenced by the pattern factor as their rotation tends to average the temperatures to which they are exposed. The combustor pattern factor per- sists throughout the turbine, although it is attenuated and tends to shift both radially and circumferentially. Hydrogen-fueled engines will have l o wer pattern factors than J e t A fueled e ngin e s . H , w e v e r , th e in he r e nt l o wer pattern factor can be traded to some extent for smaller combustor volumes.

The patt e rn factors assum e d for the study are show_ in T able 12. They ar e significantly better than can be achieved with conventi o nal fuels and are consistent with the combustor size sel e cted. Furth e r improvem e nt would b e possible if a larger combustor wer e select e d; however, the combustor was sized bas e d on flowpath, weight and cost consid e rations which g e nerally favor small size.

Blade and vane cooling requirements were calculated o n th e basis of simple effectiveness correlations, a simplified approach. To establish cooling flows precisely requires consideration of a number of factors not included in the simple effectiv e ness correlations and is beyond the scop e I of this study. T_ ; o levels of effectiveness versus cooling flow were us e d.

One_ us e d for the high e r pressure turbine, reflects a sophisticat e d, high effectiveness, high cost approach. The s e cond is a lower eff e ctiveness, low e r cost approach which was used for the low pressur e turbin e .

The temperature environment of the rotating blade is a function of the stage work, mean blade speed and _he gas temperature. For this study, it was assumed that the t e mperature felt by th e blade was 90 percent of the gas temperature for high pressure turbine blades and 93 p e rcent of the gas temperature for the low pressure turbine.

An analysis of the turbine work required to driv e the high pressure compressor in the 40 : 1, 50:1 and 60:1 cycle pres s ure ratio engin e s indicated that the 40:1 engine r e quired a single-stage turbine and the 50:1 and 60:1 cycl e s required two-stage high pressure turbines.

TABLE 12. - LH 2 COMBUSTOR PATTERN FACTORS Turbine Vane Pattern Factor i !

First High Pressure Vane 0.15 I Second High Pressure Vane 0.15

[

Fi r st Low P re ssu r e V ane 0. 1 25 Second Low Pressure Vane 0. I0 Turbine cooling flow leaving the turbine blades or vanes disrupts the flow field and causes losses. These losses are small when trailing edge discharge is feasible. However, to achieve the high effectiveness cooling schemes required, f_Im cooling is required and efficiency penaltles are incurred. Baseline turbine efficiencies were, therefore, adjusted to reflect the type of cooling air discharge and the quantity of cooling air.

4.2.3.1.4 Turbine cooling flow quantities: Turbine cooling flow require- m e nts are shown in Table 13 for the three pre=gure ratios being investigated.

As just stated, a slngle-stage high pressure turbine is satisfactory for th e 40:1 pressure ratio cycle, but a two-stage high pressure turbine is required for the 50:1 and 60:1 pressure ratio cycles.

The flow requirements are separated into cooled chargeable, uncooled chargeable, and, in the case of the high pressure turbine, nonchargeable cooling air. Cooled chargeable air is cooling flow that is cooled by hydrogen and which bypass e s one or more work producing stages of the turbines. It therefore reduces horsepower produced by the turbine. Uncooled chargeable air is cooling flow that is not cooled by the hydrogen and which bypasses one or more work producing stages. Nonchargeable air is the air used to cool the first high-pressure vane. First hlgh-pre_sure vane cooling air does not bypass any work producing stage and, ther e fore, does not diminish horsepower produced by the turbine. First vane cooling air does have an TABLE 13. - TURBINE COOLING AIR FLOW REQUIREMENTS 1760°C (3200°F) ROTOR INLET TEMPERATURE HP Turbine* L P Turbine* Cycle Pressure Cool e d Uncooled Nonchargeable Cooled Uncooled Ratio Chargeable Chargeable (Ist Vane Chargeable Chargeable 40:1 (I STG HPT) 3.0 1.8 3.0 5.0 2.4 50:1 (2 STG HPT) 6.2 2.3 3.0 4.1 2 .4 60:1 (2 STG HPT) 5.0** 2.3 3.0 2.7 2.4 * C ooling flow ex pr e s se d in p e rcent of c ompr e ssor flow. Includ e s coolant for blad e s, vane s , s hroud, and disks, plus l e a k ag e .

" **Note that cooling flow requirements are lower for the 60:1 pr e ssur e ratio design due to the low e r inl e t temper a ture at th e se c ond stage of the high pr ess ure turbin e and at the inl e t of the low pr e ssure turbine. Th e s e lower t e mp e ratures result from the greater work extra c tion at higher pr es sure ' r ati o s , !

effect on the temperature of the gas entering the first turbine rotor.

However, all cycle calculations are based on the mixed temperature at the first rotor inlet. Therefore, the gas temperature at the first vane inlet (combustor outlet) is higher than the temperature at the first rotor inlet.

The am o unt of c oo ling flow t o the vane is only _mportant in determining the temperature environment of the vane and the efficiency of the turbine.

High vane cooling can effect the ef f iciency of the turbine as discussed earlier. To minimize the amount of vane cooling and the efficiency penalty, hydrogen cooling of the vane cooling air is utilized. Cooling air for the second high pressure turbine vane and the low pressure turbine vane is chargeable as it bypasses one or more work producing stages.

As described in Paragraph 4.1,4, disk cooling air is not cooled by the hydrogen.

4.2.3.1.5 Thermodynamic accountability of turbine cooling air: Turbine cooling air results in two penalties to the cycle. The first is a reduc- tion in turbine efficiency due to disturbing t he blade and vane flow fields.

This has been covered in earlier discussion. The penalty assessed was 0.2 points in efficiency for every percent cooling _ir used. The second penalty results from bypassing one or more work producing stages of the turbine. In the thermodynamic model of the engine, flow to any of the blade or vane rows, other than the first vane, is assumed to completely bypass the high pressure turbine. This is also true for all air going to the low pr es sure turbine. This approach simplifies the model considerably but results in a more severe penalty than is actually incurred. For e xample, vane cooling to the second vane of the HPT bypasse s only one stage of the H PT. To account for this, the actual cooling air to any cascade row was r e duced by the ratio of the number of work producing stag es it bypa s se s and the t otal number of stages in the turbine. This is a simplification which is considered satisfactory for this investigation.

4.2.3. 2 Cy c le sele c tion: Th e cooling flows list e d in Tabl e 13 were u se d in the cycle s el e ction studi es for the high-temp e r a tur e inv es tigation, Ro t or inlet t e mp e ratur e was held constant at 1760°C (3 2 00°F) and c y c le pres s ur e ra t io, fan pre s sure ratio and byp as s ratio were varied. Boo s ter pr e ssure ratio was h e ld c onstant at 1.45:1. All en gin e c ycl e s w e r e e valuat e d at i0 66 8 M (35 000 ft), Math 0. 8 5. Engin e thru s t minu s c owl drag w a s h eld c onstant at 26 6 8 9 N (6000 Ib). Th e e xh au s t r e gen e rator was includ e d in th e c ycle. Figures 29 through 34 show t he r e sult s of the inve s tigatio n for t he range of c y c le pr es sur e ra t i o s and fan pre ss ure ratio s inves t igat e d. The w e ight and performanc e d a t a shown in Figure s 2 9 thr o ugh 34 w e r e u se d t o d e termin e ADOC r e l a tiv e to th e b aseline, Tr en d s o f DOC v e csus f an pr ess ur e ra tio a n d by pass ra tio are s h o wn in F i g u res 3 5, 3 6, 3 7 f o r th e t h r ee s e l ec t ed v a lu e s of c y c l e pr e ssur e r at io. Th e minimum DOC a t eac h f a n pr e ssur e r a tio and c y c le p ress u re r at i o is s h o wn in Figur e 3 8 . Figu re 3 8 a l s o sh o ws th e byp a ss ra tios a s a fun c tion of f a n p re s s ur e ra tio a nd c y c l e p re cqurc ra tio.

7 2 !

J

LH 2 EN G I NE H IG H TEMPER A TURE CY CL E SF.LECTI ON !'

I0 670 m (35 -,_0it), 0. 8 5 M, MAXIMUM CRUISE SE TT ING) I.

FAN PRESSURE 16 _ RATIO = !

o I 1 6 _: 12 ...... _ ..........

a_ >" I 10 ----NOT_. --- t ............... 1.8 r 6 _ r stages) I _2.0 EXHAUST REGENE&ATOR 0.28 H2 COOLING OF TCA z 0 . 28 m _o = FN - DCOWL=26 689N (6000 Ib) I _ I

" - " _:I,,- .I .8 1.5

= _ 0.24 _ I __0.24 U Z u. _ 0 . 22 0.22 .......................................

u w- o..

0. 2 C 0.20 , i , , , • 0.6 0.8 1.0 1.2. 1.4 1.6 1 . 8 2.0 2 2 CORE-TO - i-,:\N f.', , (HALIS TN OZZLE ISENTROPIC VELOCITY R ' _TIO !

II Fi g ur e 29. - E ffec t of co r e e n er g y e x t rac tio n a n d f an p r ess ure r a ti o on SF C and b yp ass r a t io ( C P R = 40 ) .

7 3

LH2 EIJG I NE HIGH TEMPERATURE CYCLE SELECTION

.... I •

TIT = 1760% ( 32 00 ° F) (_X.)

C P R - 4 0 : I 1.4 L PPR - 1.45:1 (boost e r st a g e s) EXHAUST REGENERATOR H2 CO O L ING OF TC A /

/

1.2 ........ / | . I z FAN PRESSURE RATIO

/

z 1.0 " L

" 6 /

0 . 8 ...........

O. 4 .......

6 8 0 2 4 6 8

_)FA_3 RAT I O Ffgure 3 0 . - Effect of fan p ressure ratio and by pas s r at io on en g ine w eig ht ( CPR = 40).

7 4

W w : i LH2 ENGINE STUDY HIGH TEMPERATURE CYCLE SELECT I ON I0 670 m (35 000 ft), 0.85 M, _XIMUM CRUISESETTING) 16 • , _ o 14 _ .....

_ 12 .......

' 6 N OTES: ,,,,,2.0 TIT - 1 6 50° C (3OO 0 °r) [1760°C(3200°F) MAX] 0 . 28 - CPR = 5 0 : 1 EXHAUST REGENERATOR _ _ 0 . 2 8 LP PR = !.45 : i ( boo st e r s ta g e s) _ H2 C OOLI N G OF T C A

I _ o. 26 =

0. 2 6 FN " D OWL = 26 68 9 N (60 0 0Ib ) o /2.b

= = 0 .24 - _l_ - "' .6

_ 0 . 22 F AN P R ESS U R E RA TIO = 0 , 20 - 0 . 20 _- - _- 0,_ O.b . 0 1 , 2 1.4 1. 6 1.8 2.0 2 . 2 COR F - ]O-FAN ^,t EXH , ,_S , NOZ L LE ISENTROPIC VELOCITY RAT I O • Figur e 31 . - . Effec t of cor e energy ex t rac t i o n an d fan p ress u r e r at i o o n SFC and B PR (CPR - 50 ) .

/ ! .

LH 2 ENG l,,E HIGH TEMPERATURE CYCIE SELECTION ,6 '-' 1.4 LPPR = 1.45:1 (booster stages " EXHAUST REGEkERATOR TIT cPR = = 50 : 11760°C (3200°F) (M A X.) / H 2 COOLING Of' TCA 1.5 / 1.2_=_ -

/

FAN PRE3SURE RATIO 20 - 0.6 ....

0.4 . ,.i ,,- 6 8 IO 2 4 6 18 BYPASS RATIO i Figure 32. - Effect of fan pressule ratio and bypass ratio on engine weight (CPR = 50).

, ' L H 2 ENG I NE HIGH TEMPERATURE CYCLE SELECTION (10670 m (35 000 ft), 0.85 M, MAXIMUM CRUISE SETTING) 14 ....... FAN PRESSU _ 6 NOTES: 2.0 T IT = 1650 ° C (3000 ° F) [1760°C (3200°F) MAX] CPR = 60:1 4 LPPR = 1.45:1 (BOOSTER STAGES) I I EXHAUST REGENERATOR H ^ C OOLI NG OF TCA ; N'" OCOWL = 26 689N (6000 Ib) Z O 0.26 _ FAN PRESSURE RA T IO 2.0

_ 0 .26[

Z = _ _ I _ _ 1 .6 _ _ I .5

o . I ..... -_-_-- i

=_'= ° .221_'- 0.22 ............. t ......

:._

0. 2 0 L 0.20 I __.

m O. 0.8 1.0 1.2 1.4 1.6 .8 2.0 2.2 CORE-TO-FAN EXHAUST NOZZLE ISENTROPICVELOCITY RATIO !

' Figure 33. - Effe c t of c o r e energy e> , t r a c tlon and fan pre s sure ratio on SFC and BPR (CPR = 60).

' Z ?

LH2 ENGINE HIGH TEMPERATURE CYCIE SELECTION T I T = 1760°C (3200° F ) MAX.

CPR = 60:1 1 4---- LP P R = 1.45 : 1 (bo o s ter s t a ge s ) ..................

EXHAUS T REGENERA T OR H2 COOLING OF TCA

l 5

1 2 -- I-- '- i- (. 9 L I J :_ FAN PRESSURE RATIO hl hi

7 - 1 6

,. J

06 -L

0 . 4 ............ i__ ....--I... . . _, 6 8 I0 12 14 16 18 I BY P ASS RATIO Figure 34. - Effect of fan pressure ratio and bypass ratio on engine weight (CPR = 60).

LH 2 E N GI N E H IGH T EM P ERA T UR E C YC L E SELECT I ON ( 1 0 668 m ( 35 000 ft), 0 . 85 M , M AXI M UM CRUISE SETTING ) G

I

TI T-- 1649°C (3 00 0 °F )[I t ° C (3200 °F ) MAX . ] 4 C P R = 40 : 1 1 LPPR =1.45:1 (boos.er staces) EXH Ab3 T REGE N ERATOR H2 COOLING OF TCA ,-, ° _I. 2 __ 2 .o F N " DC O W L = 2 6 6 8 9 N ( 600 0 Ib) .... / --_,,, _ RE RATIO = Z -,r

< \

_0 .......

_. "' I. 6..._ -2 - 4 i ......

6 8 I 0 12 4 16 18 5YP . asS RATIO g Figure 35. - Effect of bypa ss rat i o and FPR on DOC for C PR- 40.

7 9 ! !

L H2 ENG I NE HIGH TEMPERATU R ECYCLE S ELECTI O N ( i0 66 8 m ( 35 0 0 0 ft) , 0 . 8 5 M, MAXIMUM CRUI S E SETTING )

•6 I I I

TIT = 1 6 4 9 °C ( 3 000°r) [1760°C 3200°F) MAX] C P R = 50 :1 I LP P R = 1.4 . 5 : 1 (BOOST E R S T AG E S) I E XH AU S T REGENER AT OR FAN P R E SS U RE R A T I O = : -- H 2 COOL I NG DF TCA 1.5- O " +2 - -

° I

F N - DC O W L = 26 6 8 9N ( .6000 I b) / Z "1" Z 0 , U , Y ?

,,, w O .

-2

_ 4_[ ............ l

6 8 I0 12 14 6 1 8 BYPASS :',ATI 0 Figure 36. - Effect of bypass ratio and FPR on DOC for CPR = 50.

8 0 LH2 ENGINE STUDY HIGH TEMPERATURE CYCLE SELECTION (10 668 m (35 000 ft),0.85 M, MAXIMUM CRUISE SETTING)

+° L I I

I T I T : 1649°C ( 3 000 °F ) [1760°C (3200°F) MAX] FAN CPR = 60:1 PRESSURE LPPR ,, 1 .4 5 : l (BOOSTERSTAGES) RAT I O +4 E X H AUS T R EG ENERAT O R = I . 5 ---- F N - D COWL = 26 689 N (6000 Ib) H2 COOLING O F TCA /

/

o +2 g .

6 8 I0 12 14 16 18 BYPASS RATIO • Figure 37. - Effect of bypass ratio and FPR on DOC for CPR = 60.

LH 2 ENGI N E H I GH TEMPERATURE CYCLE SEL EC TION (IO 668 m (35 O00 ft), O.85 M, MAXIMUM CRUISE SETTING) 2O TIT ::1649°C(3COO°F)[1760°C(3200°F) MAX.]

LPPR : 1.45:i (BOOSTER STAGES) EXHAUST REGENERATOR H2 CO O L I NG O F TCA FN - DCOWL = 26 689 N (6000 Ib) F- o 15 ....

ro >- I0 _ CYCLE PRESSURE -_,,._ _ RATIO 5 ) CYCLE_ - (REV I SED F O R H.

C OOLI NG OF T CA_ O '-' CYCLE PRESSURE z 0 _ RAT I 0 ,., 40 z 50 ", e" U __ 60 Z _ 2 L LJ

J

- 4L_ .... _ i . •

1.2 1.4 1. 6 1.8 2.0 2.2 FAN PRESSURE RATIO Figure 38. - Effect of fan pressure ratio and cycle pressure ratio on DOC and bypa s s ratio.

Minimum DOC occurs at a fan pressure ratio of approximately 1.75:1 for all three cycle pressure ratios and a 60:1 cycle pressure ratio yields the most improvement. There is approximately 0.8 percent difference in DOC between a cycle pressure ratio of 40:1 and 60:1. This small difference in DOC was not believed to be high enough to justify the significant complexity and cost penalties associated with the very high pressure ratio engine.

The 1482°C (2700°F) cycle initially selected offered a_DOC of 2 percent.

However, cooling flows were calculated on a different basis and hydrogen cool- ing was not utilized. Accordingly, cooling flows were calculated on the basis of the revised methodology (4.2.3.1) and weight and SFC recalculated.

The _DOC for this cycle is shown in Figure 38.

The 1482°C (2700°F), 40:1 cycle pressure ratio engine incorporating hy- drogen cooling of the turbine cooling air and the exhaust regenerator was selected as the cycle to represent technology and performance appropriate for the subject LH 2 fuel system study. This selection results in a DOC approxi- mately i percent higher than the 60:1, 1760°C (3200°F) cycle. The high- temperature, hlgh-pressure ratio engine would be significantly higher in cost than the selected engine. If the cost were more than 6 percent higher, which is very likely, the DOC advantage would be negated.

4.3 Selected Engine Concept The final cycle selected as a resu l t of the hydrogen exploitation studies and cycle selection investigations has the following significant features at the engine design point (maximum cruise power, i0 668 m (35 000 ft) M 0.85): • Fan pressure ratio of 1.7:1 and a bypass ratio of 10:1 • A booster pressure ratio of 1.45:1 • A compressor pressure ratio of 16.5:1 • A rotor inlet temperature of 1379°C (2514°F) [1482°C (2700°F) maximum] rotor inl e t temperature • , cycle pressure ratio of 40:1 4.3.1 Description and performance. - The selected engine is a twin spool, direct drive, separately exhausted turbofan. A single stage fan and two booste_ stages are driven by a multistage, uncooled, axi_l turbine. The • gas generator consists of a lO-stage axial compressor, a through-flow cir- cular combustor and a slngle-stage cooled axial turbine. The spool shafts are concentric and the low pressure spool shaft _asses through the high pressure shaft.

g

83 i

!

!

Four heat exchangers are included as part o f the engine to provide (a) hydrogen cooling of the turbine cooling air, (b) engine oll c ooling, (c) hydrogen cooling of the aircraft environmental control system air and (d) fuel heating. They are described in Section 4.3.6.

Basic cycle and performanc e data are listed in Table 14 at the engine design point and at sea level static takeoff conditions. The performance includes the effects of inlet pressure recovery, horsepower extraction, aircraft bleed extraction and fan stream scrubbing drag. Freestream cowl drag and inlet spillage drag is not included. The cycle and performance characteristics shown in Table 14 are the final results of cycle o_timlza- tion. They r e flect final estim a tes of c omponent performance, pressure losses, coolin g flows, et c . The primary r e finements in c luded increases in low pressur e turbin e efficiency and nozzle thrust co e f f icients, c ompared to thos e used in the early part of th e study. Typi c al e n g ine p e rforman ce c urv e s ar e pres e nted in App e ndix G.

4.3. 2 Weight_ geometry, and scaling relationships. - An envelope drawin g of the s e lected en g in e is included as Figur e 39. Dim e nsions, mount loc a tions, a cce ssory g e arbox and thrust reverser details are shown.

The estimated dry weight of the bare baselin e -size engine is 1 7 15 kg (3780 ib). This w e ight in c ludes e ngine ac ce ssories, i.e., fuel control, fu e l pump, lubrication pumps, h e at e xchan ge rs a nd acc e ssory gearbo x . Aircraft acc essories, inl e t, nozzles, fan thrust revers e r and noise suppression are n o t includ e d. Th e estimat e d a eight of the inner and outer fan ducts, fan and c o r e nozzles, and fan thrust reverser is 367 kg (809 ib). The tot a l dry w e ight of th e e ngine exclusive of inl e t, aircraft a cc ess o ries and noise sup- pr e ssion is 2082 kg (4589 ib).

The engine may be scaled within _25 percent of its base size according to the following relationships: Scaled Weight = Wbl Bas e Thrust Scaled Thrust) 1.0 Scaled Length _ Lbl Base Thrust Scal ed Thrust ) 0.25 S c aled Diameter = Dbl " Bas e Thrust S caled Thrust) 0.5 ' TABLE 14. - CYCLE AND INSTALLED PERFORMANCE CHARACTERISTICS - SELECTED L H2-FUELED BASELINE ENGINE M 0.85 i0 668 m SLS, Std Day (35 000 ft) Power setting Takeoff Max. cruise Net thrust N, (ib) !36 587 (30 706) 29 i00 (6542) SFC, (kg / hr) / daN ((ib / hr) / ib) 0.1045 (0.1025) 0.2054 (0.2014) Bypass _atio 10.25 I0,0 Fan airflow, kg / sec (ib / sec) 483.7 (1066.4) 217._ (478.8) Fan pressure ratio (tip) 1.594 1.7 Fan pressure ratio (hub)* 2.26 2,466 Compressor pressure ratio 15.5 16.5 Rotor inlet temperature, °C, (OF) 1482°C (2700) 1379°C (2514) *Hub pressure ratio includes booster stages 4.3.3 Engine cost. - Engine cost was established using techniques developed for estimating the cost of Jet A-fueled engines, with suitable allowances made for the differences between S et A-fueled engine technology and H2-fueled engine technology. These differences include the previously discussed pro- visions for cooling the turbine cooling air, the engine oil, and the cabin air; heating the hydrogen in a core exhaust heat exchanger; and the fuel control ar 4 delivery system. Cost_ were developed for the base pngine, and also for th_ installation of nozzles and thrust reverser. The cost data were in 1976 dollars, and were provided as input to the ASSET computer program.

4.3.4 Nois e and e missions. - 4.3.4.1 Noise: It is estimated that the engine selected for the LH 2 trans- port will allow the requirements of FAR36 minus i0 EPNdB to be met. The penalty to specific fuel consumption to meet these requirements is estimated to be negligible. The penalty to engine weight and cost is estimated to be " l ess than two p e rcent.

8 6

Noise reduction in the following areas will be necessary: • Fan source noise through improved airfoil design and proper blade- - to-stator spacing • Combustor noise • Turbine source noise through optimization of blade and vane numbers and spacing It is anticipated that the technology for achieving the reduction in the areas listed above will be available by the 1990 time period. The acoustical treatment of the inlet, fan duct and turbine exhaust will con- tinue to be a requirement. With respect to noise, there is no difference between equivalent Jet A- and LH2-fueled engines.

4.3.4.2 Emissions: The use of liquid hydrogen as a fuel simplifies the • emissions problem as products of combustion do not include hydrocarbons, carbon monoxide, or impurities such as sulfur or carbon. The exhaust from a LH2-fueled engine is basic a lly water vapor. The only pollutant that will be produced are some oxides of nitrogen (NOx) as a result of nitrogen and oxygen from the air combining at the high combustion temperatures encountered in aircraft Jet engines. The NOx output is an exponential function of tem- perature and residence time in the combustor.

Hydrogen, injected in gaseous form into the combustor, has the char- acteristic of diffusing rapidly into the air so that mixing occurs thoroughly and very quickly. Combustion of H2 / air also occurs at a high rate so the result is smooth, complete burning with a much more uniform temperature profile than is characteristic with Jet A fuel. Elimination of the high temperature p e aks, which occur with Jet A, and reduction of the residence time can si g nificantly reduce the production of NO x from a I . H 2-fuel_d _ngin e, even though the average temperat,,re ..:th= combustion chambers of comparable engines _ th e s a m e .

4.3.5 Op erational c ha r a c t er i s ti c s. - 4.3.5 . 1 Rat e d p er fo r man ce : P er forman ce ra tings f o r th e b a s e lin e -siz e e ngin e a r e shown in T a bl e s 15, 16, a nd 17 , Th e p e rform a n ce shown in c lud e s th e eff e c ts of 1 2 5 ho r s e pow er ex t r a c tion an d 3 . k p erce nt bl ee d a ir ex tra c tion. I t a lso in c lud e s th e e ff ec ts of int e rn a l nozzle p e rform a n ce , inl e t rec ov e ry, and fan str e am s c rubbin_ dra_.

@ TA BLE 1 5. • INST A LLE D PERFORMAN C E R A T ING S A T U.S. S T ANDARD A TMOSPHERE SE A -LEVEL S TA TIC CONDI T ION S I I T _ r _ t Co_p r,, m T _ r bln e I _ l e Low P re ss u re _,g_ P r e ss ur e [ : d a n r ko / hr / JjN T e _ p e r a t u re 1 Spoo l 5p e rd Sp o o l Sp ee d i ° C q °F) rp_ r_ m P u w e r S e t ti n g ( Ib) l _]_ / n r / Ib) I i ( J O 7 0 6 j | (0 10 2Sl I I i , )9 8 _ 17 0 9 _ !

| I ( 29 ] 27 ) | ,0 10] 2 ) [ [ l TA BLE 1 6 . - IN S TALLED P E R F O RMANCE RATINGS AT 34.2° C ( 93.6°F) S EA-LEVEL STATIC CONDITION S Th ru st Co nsum ptio n [ T u r b Jne In le L i L o w P r e ss u r e H l_ h Press u r e d a N _ g / h r l d aN T e m p e r t u r e Spo o l _p eed S pool S p ee d I i (Ib ) (I b l hr l l b) r p * ' I rp_ TABLE 17. - INSTALLED PERFO R MANCE RATINGS AT U.S. STANDARD ATMOSPHERE i0 688 m (35 000), 0.85 MACN I N et S p e¢ _ t , c Fu _l i Thrust Co, ,_ur , p_ o n Turb in e I nl e ' ' L Ow Pres s ur e M_qh P re ssu r e d _ N k g / hr / d aN T em p e rat u r e _ S p o or _p eed C , p o ol _p e e d Po_ er _ e tt , ng (I b l (I b / hr / I b) o c ( O F ) rp_ rpr M ax _r, C l_,_b I J2J6 , 0. 2096 I_5_ ( _ 6 50 ) I I *) C c 1 7 ]_I , ( 727 5 1 ( 0, 2055_ ! i !

M, * _ _ Cru, se [ 0 205_. l J )'9 ( 2 51_,) WI 2 7 I 6 88 6 2910, I (6 _4z) (0 2 0 1 ,* ) I _ i 4.3.5.2 Capabilities and lim i ts: 4.3. 5 .2. 1 E n g in e fl ig ht e n v el o pe : T he e n gin e flight o perat i n g envelope i s shown i n F i gure 40.

4.3.5.2.2 Fl l ght- r a an euv e r l oa ds: T he fli g ht ma ne uver operat i n g load d i a g rams a r e sh o wn i n Fi g ur e 4 1 .

4.3 .5 .2.3 St ar ting: The e n gi n e shall be c a pa b l e o f gro u nd st ar ts at altitu d es f rom s ea l eve l t o 457 2 m ( 1 5 00 0 ft) . T h e al r s t a rt envel o pe i n t e rm s of a lt i tud e a n d M ac h numb er i s s h own on F i gure 37 . T h e a mb ien t t e mper a tur e r a n ge fc r gr o u nd st a rt i n g is f rom -40° C (-40°F ) t o 51°C ( 12 5°F ) . .

50 000 • 15 000- E W 9000 - 30 000 -- 6000 20 000 .......

M IN IMUH START I EN' ] ELOPE 3000 I0 000 --- 0 0.2 0.4 0.6 _ _ 1.0 MACH NO t Figure 40. - LH2 engine flight operating envelope.

(a) L OAD FACTORS AND ANGULAR VELO C I T IES A N D A N GULAR ACCELERAT;ONS SHOULD BE TAKEN AT OR ABOUT THE CENTER O F G R AVITY OF THE FLI G HT ENG I N E .

(ID L E _; _AKEOFF T HF ; T UP (b) SIDE L OAD FAC TO RS ( S. L .)

ACT TO EITHER SIDE.

e " : _ o r ad / se c 2 "- - | (c) () A ND ()' A R E PIT CH I N G (d) "_" I S YAWING ACCELERAT I ON.

; S .L. _ ° " = ' ra d /s ec 2 = + I // _i_ VE L OC I TY AND ACCELERATION.

A PP | ICAB L E TO - , _ - -_ 2 . - _ (e ) _ o " ,S ROLL ACCE L ERA T ,ON.

CONP'ETE RECTANGLE __'_ (f) DOWN LOADS OCCUR DUR I NG • FRuM 5 UP T O 7 D O WN A FT _ . ._ _--" F ORE PU LL O U T OR UP - GUST.

p .... • "_I " ( ,_) F ORE L O ADS OCCUR DURING = +2 tad / see L AND I NG.

LI M IT -- (h) S.L. and e A RE N O T A C T ING S.L . = +2 S I M ULT AN EOUUSL Y " APPLICAB L E TO ( i ) AT M AXI M U M RA T E D EN GIN E S P EED_ T H E E N G IN E AND IT S C O M PLETE C RO S S - / SUPPORTS S HALL W ITH S T A ND HAT C HED A R EA _ A G_ROSC O P I C P O M ENT !

See Not e (h) DOWN I MPOSED BY A STEADY ANGULAR VELOC I TY OF -- 2 .5 rad l s ec F OR A TOTAL E N GINE LIF E PERIOD UP OF 15 S ECO N DS .

TAKEOFF AND LANDING FO R E D RA FT ) SHA LL NO T F RACT U RE WHEN ( O T O MAXIMUM THRUST, F- - -_i'-_ _ (j) THE ENG , NE AND I TS SU P P O R T S S .L. = - 2 . O ] _L_ + I ! MA T E L OADS O F '. 5 TI MES 'B" = + 12 tad / see2 AFT _ _ _-FORE THE ABOVE L I M I T LOADS.

2 I _1 (k) U LT IMA T E- L OAD D I TCH I N G '_ ' = ± 0 t ad / s ec T " - . 1_ CAPAB I L ITY W ITH TH E EN GINE _ AT I D L E THRUS T: ! 4 I. 12 g ' s FOR W A RD W ITH 15 ULTI M ATE - 6 g' s DOWN.

____!_ _-- A - 3 0o CON E W ITH THE 2 . 12 g' s FOR W A R D W ITHIN DOWN C O NE AP EX AT THE ENG I NE CG AND TH E CON E AXI S • PARALL F L TO THE ENG I NE LONQI T UDI flAL AXIS.

( 5 ) 99 ' s FOR W AR D CO M BI N ED W ITH A 1 . 5 - g SI O E LOAD , A N D EITHER 4._ g ' S OOWN O R 2 g ' s UP , Figure 41. - LH 2 engine operating load limits.

4.3.5.2.4 Ambient temperature limits: The engine ambient temperature flight envelope is as sho_n on Figure 42.

4.3.5.2.5 Engine speed llmits: Maximum low pressure spool (fan) speed is 4430 rpm and maximum high pressure spool (compressor) speed is 17 860 rpm.

4.3.6 Description of englne-mou°.ted heat exchangers. - Based upon the engine analysis, problem statements for the four-engine mounted heat exchangers were prepared, and heat exchanger preliminary designs were established to meet these requirements. Engine-mounted heat exchangers are required to perform the following functions: • An air to hydrogen heat exchanger to cool compressor bleed _ir for use in cooling the HP turbine vanes and rotor blades.

• An oil to hydrogen heat exc|mnger to cool the engine oil.

• An air to hydrogen heat exchanger to cool compressor bleed air for i use in the aircraft environmental control system.

• An exhaust gas to hydrogen heat exchanger, located in the engine flow path downstream of the low-pressure turbine and upstream of the exhaust nozzle, to transfer heat from the engine exhaust gas to the hydrogen fuel.

Design point data for the four heat exchangers are presented in Table 18.

Because of the high hydrogen inlet pressure 2758 kPa (400 psia), only tubular heat exchangers were considered for these applications. The heat exchangers were designed to eliminate freezing problems which can occur when moisture is condensed out of the air as it is cooled and is then exposed to tube wall temperatures below 0°C (32°F).

To provide compact heat exchanger designs and eliminate freezing problems, the turbine cooling air heat exchanger and the aircraft ECS heat exchanger both utilize finned tubes and hydrogen reclrculation. The fins provide high thermal conductance to the air and the recirculation preheats the hydrogen to raise the tube wall temperature above the freezing point.

The engine lubrication oil heat exchanger is also in the recirculation loop and was designed as a shell and tube heat exchanger. An ejector (Jet pump) is use_ to produce the hydrogen reclrculatlon flow with only a small • additional pressure drop in the hydrogen. The turbine cooling air heat ex- changer, the engine lubrication oil cooler, and the aircraft ECS air heat exchanger heat the hydrogen in series from 50°K (90°R) to a temperature of 264.1°K (475.4°R) to eliminate freezing problems in the engine exhaust gas ' heat exchanger. The turbine cooling air heat exchanger was arranged in a pattern as shown in Figure 43. The engine lubrication oil cooler was o

I "/7//7. ] o_o°.o _.o.. , o._s_._,._ _.v_o_

_\\\\" I _,. , G,T S TART , NG ENVELOPE "

14 0 I i , , I 60 I MAXIMUM GR O U N D "=_" 1 STARTI N G TEMPERATURE 12 0 % I I I

s o 7__ /_ iN

, o

-20 _

- 4 o _\\,\\\ , _\\ , _\\_\\ ; I / . 4 o

_ ,MINIMUM GROUND STARTING TEMPERATURE \\\\\\ \\\\\\\\\1 -60 \ \ \ \\ \\ \\ \\\\\\ _!

- 80 ........... -_'_ 6 0

., o o ;\

................. -_, - 80 -120 -1_0 0 10 000 20 000 30 000 40 000 60 000 ft

l 1 I I I I

O 3000 600 0 9000 12 000 15 000 PRESSURE ALTITUDE, m F i gure 42. - LH 2 e ng l n e ambie n t f li g ht and starting temperatur e env e lope .

!

- 4

H2 IN

__ H 2 OUT

AIR FLOW

,p AIRFLOW LENGTH 48.59 cm (19.13 in.)

WIDTH 12.7 cm (5.0 in . ) HEIGHT 9.42 cm (3.71 in.)

FINNED TUBES ' TUBE O . D. 0.64 cm (0.25 in . ) TUBE WALL 0.04 cm (0 , 016 in . } TUBE MATERIAL 304 C RES FIN O.D. 1.27 cm (0.50 in.)

FIN S PA C ING 0.06 cm (0.025 in.)

FIN THI C KNESS 0.010 cm (0 . 004 in.)

FIN MATERIAL OFHC COPPER FIN AND TUBE C OATING NiCr TOTAL NUMBER TUBES 308 NUMBER HYDROGEN PA S SES 1 TUBE WEIGHT 7.4 kg (16.3 Ib) TOTAL H E AT EX C HANGER WT 18 kg (40 Ib) Fi g ur e 4 3 . - Tu r bin e c oolin g a ir to h y d r o ge n h e at exchange r .

designed as sho_1 in Figure 44. The aircraft ECS air cooler was arranged as shown i:iFigure 45.

The engine exhaust gas heat exchanger heated the hydrogen for entry to the combus=or. This heat exchanger is exposed to a large flow rate of air (exhaust gas) and had to be designed with low air pressure drop in a location with limited air flow area. To satisfy these requirements, an inline tubular heat exchanger such as used by Pratt and Whitney in the 304 engine was selected.

The proposed design has involute curved tubes running from a 56.4 cm (22.2 in.)

inner diameter to a 126.5 cm (49.8 in.) outer diameter air passage. The in- volute tubes are arranged in a pattern as shown in Figure 46.

4.4 Technology Development Required The technology postulated for the LH2-fuel_d engine is representative of that which would be incorporated in an engine entering service in the 1990 time period. Much of the technology is not, however, unique to use of LH 2 fuel. Much of the aerodynamics, materials, mechanical design and manufacturing processes, while advanced, are equally applicable to future kerosene-fueled advanced transport engines.

Elements which are unique to the LH 2 fueled engine are: • Combustor • H 2 cooling of the turbine cooling air • Heat exchangers • Fuel control The fuel control system is discussed in section 5,5 of this report.

4.4.1 Combustor. - Technology development is required to take advantage of the properties of hydrogen and to execute a combustor design which is smaller, provides an improved pattern factor, and is low in oxides of nitrogen emissions.

The design of hydrogen combustion systems is particularly amenable to analysis relative to conventional kerosene combustion systems. The kinetic schemes and reaction rates are well established except for turbulent flow.

Therefore, a technology program to develop a hydrogen combustion system • could consist of analytical design augmented by an experimental program to provide turbulent flow kinetics and to verify the analytical design.

' 4.4.2 H2 coolin_ cf turbine cooling air. - There are two problems introduced when hydrogen cooling of turbine cooling air is incorporated in an engine.

OIL OIL IN OUT H2 OUT _ -.. 4 -- H2 IN (3 .7 5 in,) I 9 .5 3 tc m 7 -, J " a - " (3 . 5 o In .

TUBES TUBE O.D. 0.32 cm (0.125 in.) i TUBE WALL 0.030 c m (0.012 in,)

I

TUBE MATERIAL 304 CRES I m

I

TOTAL NUMBER OF TUBES 600 NUMBER OF HYDROGEN PASSES 1 NUMBER OF OIL PASSES 2 TUBE WEIGHT 1.2 kg (2.6 Ib) TOTAL HEAT EXCHAN GERWEIGHT 2.6 k9 (5.75 Ib)

I'

Fi g ur e 44. - Eng l n e o li to h ydrog e n h ea t excha ng e r , 9 6 .

t

]

!

• i

!

i

H2 I N b . _ , |

._ _ - H 2 OUT

, , .

AIR FLOWLENGTH 7 .9 cm (3.1 in.)

WIDTH 7.6 cm (3.0 in.)

HEIGHT 60.7 cm (23.9 in.)

FINNED TUBE TUBEO.D. 0 . 64 cm (0 . 25 in.)

TUBEWALL 0.041 in. (0.016 in.)

TUBE MATERIAL 304 CRE S FIN O.D. 1.2 7 cm (0 . 50 in.)

FIN SPA C ING 0.064 cm (0.025 in.)

FIN THICKNESS 0.010 cm (0.004 in . ) FIN MATERIAL OFHC COPPER FIN AND TUBE COATING NiCr TOTAL NUMBERTUBE S 31 5 NUMBERHDYROGEN PASSES 1 " TUBEWEIGHT 4 . 5 kg (10.0 Ib) ,. TOTAL HEAT EXCHANGERWEIGHT 11.3 kg (25 . 0 Ib) F ig ur e 45. - E CS b lee d alr to h y dr o g e n h ea t exc h a n ge r .

9 7

H2 IN

OUTSIDE DIAMETER 126.5 cm (49.8 in.)

IN SIDE DIAMETER 56.4 cm (22 . 2 in.)

AIR FLOW LENGTH 19 . 1 cm (7.5 in,) INVOLUTE TUBE LENGTH 56.9 cm (22.4 in . ) CIRCUMFERENTIALTUBE SPACING 6 DIAMETERS AXIAL TUBE SPACING 1 . 25 DIAMETERS TUBEO.D . 0.478 cm (0 . 188 in . ) TUBEWALL 0.030cm (0.012 in . ) TUBEMATERIAL 304 CRE S TOTAL NUMBEROF TUBES 1984 NUMBEROF HYDROGEN PA SS ES 8 TUBEWEIGHT 38 . 3 kg (84.5 Ib) TOTAL HEAT EXCHANGERWEIGHT 77.1 kg (170 Ib} F ig ur e 46. - Eng i n e e xhaust f u e l h e at er.

9 8 L The first is a design problem. Normally turbine cooling air is routed internally through the engine from the compressor to the cooled turbine.

The routing is different when the turbine cooling air is hydcogen cooled.

Complex design problems would have to be addressed but the task could be best undertaken concurrently with engine design.

The second problem is caused by the lower temperatures of the turbine cooling air. Thermal gradients in the blades would be more severe than presently experienced for a similar blade heat transfer system. These high thermal gradients can result in low cycle fatigue damage. In order to realize the advantages of H2 cooling of the turbine cooling air, it is recommended that parallel technology programs be undertaken to i. Develop heat transfer systems which produce more uniform temperatures 2. Extend development of single crystal turbine blades which have higher cyclic fatigue strength.

5. ENGINE FUEL SUPPLY SYSTEM 5.1 Candidate System Concepts 5.1.1 Concept descriptions. - The basic fuel system performance requirement is that a pump or combination of pumps must supply fuel according to a specified flow-pressure schedule. The schedule shown in Table 19 was used for initial design considerations. Additionally, a tank-mounted pump must be included in the system, to provide a pressure higher than the vapor pressure to the fuel lines and the engines. The wide range of pumping systems that could achieve these requirements is constrained by cost considerations, as expressed through theADOC equation applied in Section 5.1.2. The objective of the concept selection was to determine the general arrangement of the fuel supply system.

which could most efficiently meet the basic requirements of the system.

, Two concepts for the arrangement of pumps in the engine fuel supply sub- system we e considered initially. In concept I, a low-pressure-rlse (nominally 50 psi rise) boost pump would be in the tank and a high-pressure rise main pump would be on the engine. The boost pump would provide a positive pressure !

to move LH2 from the tank to the engine and would meet main pump inlet pressure requirements. In concept 2, the main pumD on the engine would be eliminated, with tbe total flow-pressure requirements of the engine met by a single tank- mounted pump. The analytical concept selection procedure described below (Section 5.1.2) also included evaluation of a concept 1-1 / 2 pump arrangement.

Here, an engine-mounted main pump and a tank-mounted boost pump would be per- formance matched so that the boost pump would supply some intermediate pressure rise (determined as a parameter of the analysis) and the main pump would supply the balance of the pressure rise. In such an arrangement, the performance requirements of the main pump (and hence its weight and power requirements) could be reduced and conceivably provide benefits to the overall system.

5.1.2 Results of evaluation. - The concept selection process utilized a para- metric trade-off study approach based upon the _DOC sensitivity equation appropriate to the fuel subsystem. With this approach, the entire range of system configurations could be evaluated. In performing this analysis, both the pump concept and the optimum fuel line diameter were selected. Pressure dropline diameter and calculated line dlameter-llne weight data (see Sec. 5.2,1) were introduced into the analysis at the appropriate points. For example, as the llne diameter was decreased, the line weight decreased but the pressure drop necessarily overcome by the tank-mounted pump increased, thus increasing the required size and weight of the boost pump.

i00 TABLE 19. - LH 2 TURBOFAN ENGINE NET THRUST, FUEL FLOW AND FUEL PRESSURE SCHEDULE FOR INITIAL DESIGN CONSIDERATIONS A i _iLude 1 d / Engln e N e t Thru st / E n g. Co mp. D i sc h ar g e Condit i on m ( i000 f t ) Ha c h Ikg / sec ( Ib / sec) N ( ib ) P r ess u r e-kPa (ps la )

T

M ax S L S 0 0 i 0.340 (0. 7 49) i2 7 66 4 ( 28 7 0 0 ) 4 8 75 ( 707 ) Ta k eoff 0 0 1 0 . 3 4 0 ( 0 . 75 0) 1 2 7 6 6 4 ( 28 7 00) 4 8 75 ( 70 7 ) Climb O 0.38 0 . 351 (0. 7 74 ) 83 538 (18 7 80 ) 5068 ( 7 35 ) 610 (2) 0.39 0 .342 (0, 7 55) 8 0 246 (18 040) 1 219 (4) 0.41 0 . 332 (0,732) 76 243 (17 140) 1 8 2 9 (6) 0 . 4 2 0.313 (0,690) 7 1 1 7 2 (16 000) 2 4 38 (8 ) 0 .44 0.3 0 1 ( 0 .664) 67 1 2 4 (15 0 90) 3 048 ( i0) 0.46 0 .2 88 ( 0 . 635) 63 83 2 (14 350) 4165 (604) 3 048 (I0) 0.64 0.303 (0.667) 56 715 (1 2 750 ) 43 92 (637) 4 57 2 (15) 0. 7 1 0. 2 78 (0.613) 50 354 (ii 320) 6 096 (20) 0 .7 8 0. 2 51 (0.553) 44 7 05 (i0 050) 3 7 30 (541) 7 620 (25) 0.85 0. 2 26 (0.498) 38 655 (8 690) 9 144 (30) 0.85 0.195 (0.429) 32 828 (7 380) 3061 (444) i0 668 (35) 0. 8 5 0 .1 5 8 (0.348) 27 357 (6 150) 266 8 ( 3 8 7 ) C r uise i0 668 (35 ) 0.85 0. 1 59 (0.351) 2 6 689 (6 000) 2 641 (383) ii 58 2 (38) 0,85 0, 1 32 ( 0. 2 9 2 ) 2 3 576 (5 300) 2 448 (355) Flight Idle i0 668 (35) 0,85 0,029 (0 , 06 3 ) -91 2 (-205) 1551 ( 22 5 ) II 58 2 (38 ) 0 . 85 0.023 (0.051_ - 7 30 (-164 ) 0 0.4 0.078 (0.1 7 1) 3 6 48 (820) Ground Idle 0 0 0.039 (0.0 8 5) 8 140 (i 830) 1868 ( 27 1) I 5 2 4 (5) 0 0. 0 32 (0.071 ) 6 7 61 (1 520) G ro und S t a rt 0 0 0,011 (,0 2 4) - Ib 2 (2 2) I Con c ept selection w a s translated into a problem of optimiz a tion of the tank-mounted boost pump pressure rise, If the optimum rise were small,

i

• c oncept I would be chosen; if large, concept 2; and if som e intermediat e pressure rise were found optimum, c>n c ept I-I / 2 would be c hosen.

An a nalysis b a sed on ADOC was c onducted to determJne the optimum design • pressure rise ( a t maximum flow) for the boost pump. If the englne-mounteu main pump effi c iency (qm) is assumed to be a constant 45 percent regardless of the boos = pump pressure rise, and if the main pump weight is assumed to be I01 a constant 25 pounds, the minimum direct operating cost occurs at the lowest possible boost-pump pressure rise. The reason for this is shown by -n investigation of the _DOC sensitivity equation.

DOC × 105 = 3.22_hp c + 0.775 (Ws + 6EhPmax) where ¢ DOC is expressed in seat n_i.

_,_hPc = horsepower at cruise for all pum_s (tank an_ engine mounted) W = total w_ight s 6 = coefficient approximating the aircraft weight penalty to provide and transmit the required horsepower _hPmax = maximum horsepower into pump drives that are tank mounted (not engine mounted) Parameters on the right_hand side were related to pressure rlse_P through the relations HP _ P and W _ p0.6, the latter equation being based on exten- ump slve Rocketdyne experle_ce with cryogenic pumps. Results of the analysis are shown in FiEure 47. For this case of constant main pump weight, increasing the boost pump pressure rise decreases HP c slightly because the boost pump is more efficient than the main pump. However, this factor is far outweighed by the increased boost pump weight (which increases Ws) and the 6 _hPmax term, which is directly proportional to boost-pump pressure rise. The net effect is that, for constant main pump weight, the minimum feasible _DOC occurs at the minimum boost pump pressure rise that will provide sufficient pressure to the main pump inlet.

If, as is usually the case, the main pump design speed is assumed to be NPSH limited (resulting in a weight that decreases with increasing boost-pump pressure rise), the optimum_DOC is shown to occur at a boost pump pressure rise of 317 kPa (46 psi). An investigation of theADOC relation shows the reason f o r this. At boost pump pressure rises (_PB) below 276 kPa (40 psla), the main pump NPSH is so low that the main pump weight becomes large and consequently dominates theADOC equation. This causesADOC to increase with decreaslng_ B.

If, on the other hand, _ P B is greater than 345 kPa (50 psi), the main pump NPSH is so large t hat the main pump weight reduction has little effect: However, the boost pump power is directly proportie-al to APB and, therefore, the tank- mounted pump power term (6 _hpmax) becomes the dominant facLor. As a result.

_DOC increases with increasing _PB if _?B is greater than 345 kPa (50 psi).

t These results show that it is not economical to increase the boost pump design pressure rise much above 46 psi. Therefore, the concept 1-1 / 2 approach does not appear to be practical. This, in conlbination w_th the even higher _DOC for concept 2 (main pump in the tank) resulted in the selection of concept i, with a minimum boost pump design pressure rise, for the hydrogen- fueled aircraft. These pazametric results were verified by preliminary calcu- lations using several system configurations. Again, concept 1 with a pressure rise of 46 psi was found to be favored.

5.1.3 Characteristics of selected systems. - Figure 48 depicts the selected engine feed system concept, This figure, together with the flow-pressure schedule found in Table 19 defined the requirements for the selected boost pump / drive system described in Section 5.3 and the engine-mounted fuel pump described in Section 5.4.

5.2 Engine Fuel Supply Lines l Selection of the configuration of the engine fuel supply lines that carry the fuel from the tank to the engines involves determination of two basic factors. First, the diameter of the lines which contain the fuel must be selected. Second, the appropriate insulation system for the cryogenic lines must be found. Diameter affects system performance in establishing both fuel-line pressure drop that must be overcome by the tank-mounted boost pump, and also line weight. The choice of insulation enters the system calculation in weight and heat leakage. Qualitative insulation effects such as safety and fabricability must also be considered. This section presents the results of the feed-lines portion of this investigation and the methods used.

5.2,1 Size Selection. - The line diameter for the fuel-feed lines was opti- mized in the concebt selection analysis described in Section 5.1.2 above.

As a precursor to that calculation, it was first necessary to determine the line pressure loss as a function of line diameter. Feed Line No. 4 (aft tank to right-hand outboard engine), was selected as the most severe configuration in terms of total line length and number of bends. Total line length, inclu- ding a growth factor of 1.2, was calculated to be 54 m (176 ft). Line losses for eight right-angle bends were calculated assuming utilization of optimum line bend radius ratio (r / dtube) of about three to five. This results in loss coefficients (K L) of about 0.2 for a 1.57 rad (90 deg) bend. The results of a conventional analysis for the flow of incompressible liquid hydrogen in pipes are summarized in Figure 49 for the maximum engine fuel flow rate condition of • 351 kg / sec (0.774 ib / sec), Table 19. It has been assumed that: (i) vapor / liquid ratio of the fluid delivered to the engine must be zero at maximum flow, and (2) low loss valves have been utilized in the system, so that line loss du e to valves is approximately equal to their equivalent line length. Ball valves when used in liquid hydrogen systems sa t isfy this assumption. A nominal one- q inch diameter llne results in a pressure loss of 9 7 kPa (14 psi) at max. flow, i J t" 0 0 0 40 80 120 160 p | l I I I I I I 0 2 0 0 40 0 600 800 1000 B O OST PUMP PRESS , RISE A PB, kPa Fig u re 47. - Boost pump pressure rise effects.

LINES I ENGI N E FUEL SUPPLY N OMINAL TA N KCO N DITIO N S ] / SHUTOFF VALVES HIGH-PRESSURE ON SATURATED LH 2 LO W.PRESSURE _ )_ / T = -253°C ( - 423°F)"-- "' -- PUMP | P = 317 kPa (46 psig) _" PUMPENGINE P = 145 kPa (21 psia) FLOW REQUIREMENT IN TABLE 19 VAPOR / LIQUIr , = 0 Fig u r e 4 8 , - Co n ce pt I s che m a ti c ( tank-mo u nted l o w- p r e ss u r e p u m p / en g ln e - mount e d hlgh-pr e ssure pump), , i FEED LINE # 4 L (X1.2) = 53 . 6 m 11 7 6 ft ) 15 0 1 I TUR N L OSSCOEFF. = 0.02 RICTION OEFF. _ 0.013 100 ....

,°4

HYDROGE N 5 0 _ --- - ' _ - P = 72.1kg / m 3 14 . 5Ib / ft3 ) 30 0 _ _ = 0 . 351 kg / sec (0. 7 74 Ib / sec), I _ (MAX. FLOW) O L o _ ,, 0 .5 1.0 1 . 5 2 9 (in.)

I I I I I 1. 0 2. 0 3 . 0 4. 0 5. 0 L INE I N NER DIAMETER , m Figure 49. - LH 2 fuel feed llne pressure los_.

Figur e 50 sh o ws th e in c rem e ntal ADOC determin e d fr o m th e s e nsitivity equation as a function of line diameter for the selected pump concept, de t er- m ined in Section 5.1.2. A llne diameter of 2.54 cm (I.0 inches) was deter- m ined to be optimal based upon this trade approach.

5. 2 . 2 Insulation system comparison. - Fr o m th e system o ptimizati o n pr oce s s des c ribed in S e ctions 5.1.2 and 5.2.1, a fuel-llne inner dlamt e r of 1.0 in c h was c hos e n. B ec aus e th e fuel to be pumped, LH 2, is cry o geni c , t he fuel lin e s must be insulated to prevent excessive heat input to, and consequent vaporiza- tion of, the fuel as it flows from the tanks to the engines, a distanc e of up t o 4 8 .8 m (160 fe e t). Two prin c ipal types of insulation systems wer e c on- sider e d: va c uum and foam.

Wh e r e pra c ti c al, vacuum insulation syst e ms are usually utiliz e d for ground-bas e d c ryogeni c transf e r systems, as the overall heat transf e r c an be : mini m ized b y m e ans o f an eva c uated spa c e fill e d with radiati o n shi e lding t o co ntr ol th is m ode o f h e a t l os s. Th e exp e rien c e wi t h v a cuum -i n su l a t e d syst e m s for fligh t -weight systems is limited, but a s i gn i f ic ant drawb a c k i s found in manufa c t uring, installation, maintenan c e, and safety of the thln-walle d tubing U I necessary for a flight-weight system. On the other hand, to achieve the Insu- lation properties of a vacuum system, foam-insulated lines must be relatively large and bulky. Foam systems can also degrade in performance over long periods of time. Thus, there is no clear-cut choice for the insulation approach. This section presents the results of the determination of the best present choice for insulation system. The approach incorporates consideration of both insulation properties and practical considerations such as weight, manufacture, maintenance, safety, etc.

5.2.2.1 Insulation requirements: Thermal insulation considerations are one factor in estimating the total feed-system weight. A primary requirement that was arbitrarily established for the proposed aircraft engine design was that the hydrogen vapor volumetric fraction should not exceed 0.5 at the engine inlet under any flow condition. This was a consideration affecting the design of the engine pump which reflects a limitation that allows up to one-half the cross-sectional area of the feed line to be gaseous hydrogen at any given instant. A second consideration relates to ground-hold conditions (i.e., zero H 2 flow) after initial line chilldown. The ground-hold condition will result in line venting and some fuel loss (boil-off) at various time intervals dependent on the insulation effectiveness. Excessive pressure in the line is prevented by thermal relief devices incorporated in the shutoff valves and a small hole in the pump check valve to allow venting into the fuel tank.

Various techniques have been developed for insulating cryogenic components.

Some, such as those utilizing helium or nitrogen purges, do not appear suitable for aircraft feed line application. A vacuum jacket and / or closed cell type foam insulation appears suitable in terms of basic simplicity. A comparison of a typical foam insulation and a simple vacuum Jacket in terms of heat leak rate is shown in Figure 51. It is apparent that the vacuum-jacket approach is supe- rior in terms of minimizing heat leak rate. (Joints are not included).

Foam: The effect of line diameter (i.e., pressure drop) on required foam insu- lation thickness and total insulation weight is presented in Figure 52 for the conditlon _ = 0.023 kg_ec (0.051 Ib / sec). For the nominal 2.54 cm (i.0 in) line diameter, the total insulation weight (all four feed lines) is about 23 kg (50 Ibs). Both the line and insulation weights are related to fe , _d llne dia- meter as shown previously. Since the line diameter determines hydrogen pressure drop, it is possible to relate the line and insulation weight to either the pump i discharge pressure or pressure rise as shown in Figure 53. This approach in combination with a pump weight versus pressure rise curve permits direct deter- mination of the minimum weight system, see Section 5.1.2.

D : The vacuum-Jacketed line approach is superior t o the foam insulation in t erms of minimum heat leakage a s noted previously in Figure 5 1. If the insul a - t ion system is a vacuu m a nnulus only, the insulation weight is zero. If an a luminized mylar radiation shield is wr a pped on the inner line, as is al m os t certainly necessary, the insulation weight is 0.104 kg / m (0.07 ib / ft), or 18.3 kg (40.3 ]b) for 176 m ( 5 76 ft) total llne length.

0. 0 0265 0 . 002 6 0 0 . 0048 E • 0 . 0049 _ 0 .O 0 2 55 - Z 0.0 0 47 _ Z = ¢ 0 . 00 2 50 _ 0.0046 _ 0 . 00 2 45 - 0 . 0045 , 0.0 0240 e • 0 . 0044 0 . 7 0 . 8 0 . 9 1 . 0 1.1 1.2 in.

I I I I I J 1.75 2.0 2 . 25 2.5 2.75 3 . 0 LH2 LI N E DIAMETER , cm Figur e 50. - Slz e o ptimization f o r engine fu e l supply line.

50- L INE DIAMETER = 2.54 cm 11.0 in.)

4 0 4 0 CLOSEDCE LL FOAM

5 0 _ I i

I | I N SU L ATIO N _ I _ CO N DUCTIVITY ' _ 0.0173W i re OK Z 30 30l _ _ (0 . 01 Btu / hf-ft-F) 20 20 ' I-< I I _"_ , _ , ,, = ,. . 295°K( 5 30R) (GROUND.

= "' lO l O _ - I _ M - O l

250°K (46 0 °R) (ALT. = : | 11 580m (38 0 00 ltl (M = 0.85) | | VACUUM ,JAC K ET ' ol----t - - - t" --- 'i' - -"- .....

0 2 4 6 8 10 in.

I I i I l I 0 5 10 15 20 INSULATION THICKNESS , crn Figur e 51. - Eff ec t of insulation on h e at l e ak rate.

10 7 !'

60 m - 4 . 2 120 I / 1.6 IN S ULATION 50 I _ = 0.0173 W / m - OK (0.01 Btu / hr-ft-F) W - 3.8 P = 32 . 04 kg / m3 12 Ib / ft 3 t I Q / L = 20.4 W / m (21 . 2 Btu / hr / ft) j / __ " _ 1.4 8 0 ' // _" //, 1.2 -- 3.0 30 Z v z --2.6 o O _ 60 1.0 "d _ : "_ --2.2 -J < 20 o _ _,= /// / ,.£ .

I . - z 0 ,.. 0.8 -

" 4 0 / _ _

/ -- 1.8

Y

20 0.6 -- 1 .4 BASED ON MI N IMUM L H 2 FLOW = 0.023 kg / sec (0. 0 51 Ib / sec} SATURATED LIQUID AT ENGINE INLET 0 0.4 _ 1 .0 ) . 60 0.8 1.0 1.2 1 .4 1.6 (in.)

_ , I I I I I I I ; 1.6 2.0 2.4 2.8 3.2 3 . 6 4.0 INNER LINE DIAMETER, cm W Figure 5 2, - Foam insulation r e quir e ments.

10 8

!' i

,. I J : 1 40 " 60- 5O 100 _ __ LINES (4) PLUSINSULATION (FOAM) (LINE tWALL = 0.071 cm ' 40 / (0.028 in.)

z 30 - - ul _ 60 • _ _ _ . _ , _ ' ;tWAL L = 0.071 cm (0.028 in.)

20- __

I / - 40 LINESONLY

(ALUM. 6061) 10- PUMP AP=50-21 + APLINE 0 - 0 I I L I I Z J I I I I 30 50 70 90 110 130 (psi) I I 1 I I I I I j 200 300 400 500 600 BOOSTPUMPPRESSURERISE , kPa Figur e 53, - Fu e l f e ed llne w ei ght r e quire m ents.

1 0 9 w 5.2.2.2 Line configurations: Vacuum Jacket insulation: The inner llne is wrapped with approximately 20 al- ternate layers of fiberglass cloth and aluminized mylar. The mylar acts as a radiation shield and the fiberglass prevents contact between the mylar sheets.

Commercial manufacturers of cryogenic piping have found that vacuum insulated lines without radiation shields lose significant amounts of heat through radi- ative mechanisms. The space between the inner and outer lines is vacuum pumped to approximately i micron.

Line wall thickness calculated from flow pressure or minimum handling gage (Sec. 5.2.3) determines the tubing weights. However, tilevalues chosen have been called into question during discussions with one manufacturer of cryogenic piping with whom the problem was discussed, CVl Corporation. CVI recommends somewhat thicker walls for three reasons: easier fabricability, easier repair, and greater structural strength. CVI suggested inner and outer wall thicknesses of 0.089 and 0.122 cm (0.035 and 0.048 in.) respectively, compared to Rocketdyne's estimates of 0.030 and 0.064 cm (0.012 and 0.02 5 in.). However, CVl's experience lles primarily in the area of nonfllght weight systems produced without advanced welding techniques. CVI's gener a l concern must be considered, however. There- fore, a reco m mended technology effort for subsequent work includes fabrica t ion, testing, and repair of fllghtweight cryogenic lines to establish minimum wall thicknesses that may be utilized for an operating system.

It is expected that the greatest stress loadings will be experienced by the outer llne. A technique for wrapping the outer llne with a nonmetallic composite material reinforcement may allow simultaneous reductions in weight at a fixed strength level and also a backup insula t ion system. In a NASA- funded study, engineers at Martin-Marietta Corporation showed that serviceable cryogenic lines could be constructed by wrapping thin metallic tubing with glass-flber reinforcement. The metallic tube carried the cryogenic fluid, while the wrapped reinforcement provided both strength and t hermal insulation.

If the outer vacuum Jacket were wrapped with glass reinforcemen t , the single wrap could act both as a reinforcement and as a backup insulation.

Another option for the ou t er Jacket is the use (for all or par t of the o uter Jacket) of seml-flex llne. This approach would elimina t e differential thermal expansion problems.

Provision of a one micron vacuum in the vacuum annulus may be accomplished in either of two ways; first, each fuel llne could consist of a single annulus extending the entire length of the line. This annulus would be pumped by an : onboard vacuum pump or by periodic pumping by ground-based equipment. Alter- natively, the annulus could be pumped down and sealed during assembly and re- pumped only if measurements indicated a vacuum leak. Second, the vacuum llne could be built from independent stand-alone units. Again, the individual seg- ments, or spools, could be pumped by onboard or ground-based pumps, or they could be pumped out and sealed when constructed.

ii0 L

f

The spool approach is preferred due to its greater reliability. In the case of a single vacuum annulus, a single large leak imposed by some accident . during flight could c onceivably disable that fuel ] ine . Further, v a cuum pumping of a thin annulus is very slow, requires a large pump, or may require an extensive network of pumping lines throughout the aircraft. For tile spool approach, as discussed in Section 5.6, loss of vacuum on a single, 3.05 m (i0 ft) spool will not cause vaporization of the fuel that passes through the spool. Thus, a single spool failure does not endanger the aircraft at any point of the flight profile.

Foam insulation: The foam insulation approach offers several distinct advan- tages compared to the vacuum approach principally in the areas of safety and reduction of technical complexity. From the standpoint of safety, foam insulation is not lost by catastrophic failure of spool sections, and the presence of the foam protects the inner fuel line from damage during handling and normal operations. Foam-insulated lines also present significant advan- tages in manufacturing and maintainability. Vacuum spool sections must be f_bricated as complex double-concentric units and pumped at the fabrication or installation points. Construction for foam lines is much simpler: indivi- dual single tubes can be welded together and then covered by a foamed-in- place insulation. Repair is accomplished simply by cutting and removing the foam, repairing the inner line, and refoaming an insulation layer. Foam layers must be protected against the phenomenon of cryopumping, in which condensation of gases within the foam cells eventually degrades the insulation , properties. Lightweight, metallic coverings can successfully protect against cryopumping problems. An additional advantage of foam is the elimination of one type of thermal expansion problem. For a concentric-tube vacuum insula- tion system, the differential contraction between the cold inner line and the warm outer line can be sufficient to damage the lines in the absence of a bellows in one of the lines to absorb the change in length. In current practice, no provision for differential thermal contraction is provided for foam-insulated lines, since the foam cells are sufficiently resilient to expand and compress to absorb the length change. Thus, the only thermal contraction which need be considered is the net length change of the metallic inner line. This effect is discussed in Section 5.2.3 below.

The disadvantages of foam are weight, fire resistance, and long-life embrittling. Foam is expected to be 25 percent heavier than a vacuum llne on per foot basis, but the elimination of complex spool connections should essentially offset _eigh t penalities. Care must be taken that the foams selected for use are resistant to burning in short, relatively intense hydrogen fires. Finally, many existing foams tend to become embrittled during long exposure to cryogenic temperatures. At the present time, no " known foams are completely unaffected by such conditions. Long-life foam development programs are presently underway, and it is expected that by 1985 fully stable foams will be available.

iii It is concluded that a foam-insulated cryogenic piping system is the best choice for use in hydrogen-fueled aircraft. A further comparison of foam ver- sus vacuum lines is presented in Sect. 5.6.2.

5.2.3 Design description. - This section includes discussion of materials for use in the fuel lines, thermal contraction, and weight.

5.2.3.1 Materials selection: Liquid hydrogen feed system materials which have been utilized successfully at Rocketdyne are summarized in Table 20.

The cast aluminum alloy (Tens-50) and the highly alloyed stainless steel (A-286) materials are utilized mainly for fittings, valve bodies, or other complex shapes. The candidate materials for feed lines are the wrought aluminum (6061) alloy and 321 stainless steel. Use of a working-stress level equal to the lower value of either one-half of yield or one-fourth of ultimate strength results in the aluminum alloy having the highest strength-to-weight ratio. Because of its much lower thermal conductivity and thermal expansion characteristics, 321 stainless steel was chosen for the inner line. 6061 aluminum was used for the outer line.

5.2.3.2 Thermal contraction provisions: Two types of thermal contraction must be considered for lines which experience temperature changes from ambient to cryogenic temperatures. First, the overall length of the line may change, thereby affecting the system geometry and linerattachment provisions.

This length change is on the order of i0 x 10 -6 in / in / °C, or 4.57 cm (1.8 in.)

for a 15.24 m (50 ft) run of line cooled from ambient to cryogenic temperatures.

Practically, this length change may be rendered harmless through the provision of sufficient bends in the llne, a line-space envelope which allows the normal i portion of the bend to absorb the length change elastically, and compliant mounting provisions (such as a cable-tray type of approach to supporting the fuel lines). Bellows might be provided where necessary, but normal practice has shown that the provisions suggested above are sufficient under ordinary operating conditions.

The second type of thermal contraction problem is the differential thermal expansion between the inner and outer lines. When the inner is cooled from ambient to cryogenic temperature and the outer line remains at essentially ambient temperature, differential thermal strains may be developed• For the case of foam-insulated lines, current practice with long, large diameter lines ; in the Space Shuttle has shown that thermal strains are accommodated without any need for special provisions such as bellows. Vacuum-lnsulated lines, on the other hand, generally require some type of mechanical strain-absorbing element such as a bellows. In addition, special configurations of foam- insulated lines may require large-deflection capability, so bellows arrange- ments were investigated for this application. Rocketdyne has extensive experience in cryogenlc-llne applications for rocket engines, such as the P TABLE 20. - CANDIDATE MATERIALS FOR LIQUID HYDROGEN APPLICATION, 217.2°C (-423VF) Y i e ld Ultim a t e * W or king D en s ity Str e ss S tr ess S tr ess Ma t e rial k g / e m 3 (ib / f n 3) kPa ( p si ) k Pa ( p si) kP a (p bl ) T e ns- 5 0 Aluminum - Cast 0.00 2 7 (0 . 0 9 6) 37 2 3 1 7 ( 5 4 000 ) 4 55 0 5 4 (66 00 0) 11 3 76 3 (16 5 00 ) 6 0 0 1 Aluminum - Wr oug h t 0 .00 27 ( 0 .0 9 6 ) 3 1 7 159 (46 00 0) 413 685 (60 000) 10 3 4 2 1 (1 9 000) 32 1 Sta in less S tee l 0.00 7 9 (0. 2 86) 2 96 4 7 5 (4 3 000 ) 1 31 C 004 (190 000 ) 14 8 2 3 7 ( 2 1 500 ) A-286 S t a inl ess Ste e l M ac hin e d 0.00 7 9 (0 . 2 87) 896 31 8 (130 000) I 365 16 2 (198 000) 341 29 0 (49 5 0 0) *W ork ing Str ess = Lo w e r va lu e o f ei th e r 1 / 2 yie ld o r 1 / 4 u l timat e s tr e ss Space Shuttle Main Engine. The general approach for insulation in the engine system is to apply foam over exposed lines, joints, valves, etc., wherever possible. Because the engine operates intermittently and then at very high fluld-flow rates, more efficient insulation approaches were rejected due to weight or complexity. A prime purpose for covering the exposed surface is preclusion of formation of LOX that could lead to an engine fire, and foams are effective in this role.

At some locations, however, foams cannot be used. The rocket engine is gimballed, and the cryogenic transfer lines must incorporate sufficient flexi- bility to permit several degrees of rotation. The bellows units are double walled with insulation provided by a vacuum in the annulus. This vacuum is produced by pumping the annulus to i micron, backfilling with pure argon gas, and sealing. When cryogenic fluid flows in the lines, the argon llqulfles and a vacuum is produced. To achieve 1 micron vacuum in a bellows unit that has many slowly pumping regions requires several days of laboratory pumping, and the approach chosen allows attainment of a good vacuum without requiring heavy on-board pumping equipment. All lines and bellows are welded wherever possible.

5.2.3.3 Fuel feed llne weight: Minimum fuel line wall thickness can be estimated from the hoop stress produced by the contained fluid, using the relatlon: Pd t -- -- 2_ W where t = wall thickness P = internal pressure _w = working stress and prior experience for llne fabrlcability. In the case of the selected con- cept which uses a tank-mounted boost pump, the hydrogen pressure levels are so low (approximately 317 kPa (46 psla)) as to result in unrealistically thin walls (approximately 0.003 - 0.008 cm (0.001 - 0.003 inches)) if only hoop stresses are considered. Rocketdyne manufacturing experience and Lockheed CL-400 expe- rience indicate minimum wall thickness of 0.041 cm (0.016 inch) are required for practical considerations.

For the inner fuel-containment llne, 321 stainless steel was chosen because of its thermal properties and because of ease of fabrication and welding, as well as proven structural integrity. 6061 aluminum was chosen for the outer line because of weight saving and compatibility with liquid hydrogen. Table 21 summarizes line-only and llne-plus insulation weights. It is noted that the total weight of vacuum insulated line is only 4.5 kg (i0 pounds), or 3.4 percent less than foam-insulated llne.

T A BLE 2 1 . - LINE WEIGHT SUMMARY Weight per m Total F e ed Syst e m (ft), kg (ib ) Weight per 1 82 m ( 5 96 ft ) , k g (ib) L Inner Line 0.27 (0.18) 49 (i0 7 ) I (0 . 016 St a inless Ste e l,

l

2. 5 4 cm (i.0 in.) o.d.)

Out e r Line 0.34 0.23 62 ( 13 7 ) ( 0.316 Alu m inu m , 10._6 c m (4 in.) o.d.)

Total 0.61 0.41 iii (2 44) Ins ula t ion W e ight Foam 23 (50) Vacuu m 18 (40) Total Wei g hts Foam 134 (294) Vacuum 129 (284) • 5.2.3.4 Fuel llne summary: The selected f_el line configuration is summ a rized in Figure 54. Although vacuum insulation saves a small amount of weight in the fuel-line, the foam insulated line was se ] e c t,_d due to safety, manufacturing, and repair considerations. In addition, sn _,alysis was performed to deter- mine the difference in cost which might be e_pected between a typical vacuum Jacketed fuel line and a foam insulated _ , _Bn. It was found that the foam insulated line would cost only 62 perce_ as much as the vacuum design; $39 i00 versus $62 700; thus adding another reaso_ _or selecting the foam insulation system.

6061 ALUMINUM JACK _ .T 0.041 cm (0.016 , n.) THICK 321 STAINLESS ST EE L INNER LINE 0.041 cm ( 0.016 i n ) THICK CLOSED CELL FOAM ,. , ,10.16 cm (4 . 0 m.} Figure 54. - S e lected fuel-line configuration.

5.3 Boost Pump In th e fuel- s y s t e m optimization effort (Sec. 5.1.2), it was d e term i ned th a t th e boost pump should provide a minimum 317 kP a (46 psi) boost ov e r the e ntir e rang e of fu el -flow, as sp e cified in th e s ch e dul e of T a bl e 19. Th e o bj ec tiv e of th e boost pump d e si gn e ffort w a s to s e l ec t th e m o st a ttr ac tiv e pump within this c onstr a int a nd oth e r d e sign re qui re m e nts d e tail e d in S ec . 5.3.1. This pump s hould not only attain minimum in c r e m e ntal _DO C , but it should also be amenable to long-life and easy maintenance in airline operations. It has been determined that two- or three-stage centrifugal pumps operating at speeds up to 36 000 rpm are the leading candidates for this apnlication. However, there is no data for the opera t ion of this or any other _ype of flight-weight pump in liquid hydrogen for t imes approaching the 8000 hours deslred for airline operations. Development of long-life, reliable cryogenic pumps of this type appears to rest on development of satisfactory bearings.

5 .3.1 B__ e si_n r equirements. - T h e d e sign philosophy for th e bcost pu mp syst e m is based uoon th e pr em is e that a single pu m p failure shall not compromise air- craft s_f_ty. In addition, aircraft op e rators ar e relu c tant to ground an air- craft if on e boost pump in any of its fuel tanks is incapabl e of b e ing operat e d.

In accordance with this philosophy, each tank in the hydrogen-fueled subsonic transport will incorporate a minimum o f three b o ost pumps. Th e Justification for this conclusion is discuss e d in the following paragra p hs.

Although hydrocarb o n-fu e l e d aircraft can takeoff and c limb tc "rulse al- t itudes with b o ost pumps inoperative most o f th e tim e , hydro g en-fu e led air- craft engines would flameout if the boost pumps fail e d, due to vaporization in th e llne with l o ss of pressure. Henc e , the boost pump system must e ntail a r e dundancy which precludes loss of thrust from any e ngine in th e event of pump failure imm e diately after th e aircraft b ec omes airborne. During tak e off and initial climb, this phil o s o phy di c tates that on e tank suppli e s one e ngine and that two pumps in each tank must b e op e rat e d simultan e ously. Thus, with two pumps operating, a singl e boost pump failure Jus t after tak e off could not caus e a loss of engin e thrust. Th e r e dundan c y r e quir e ment furth e r di c tat e s that n o two pumps within a given fuel tank can be supplied electrical power from the same source.

Th e abov e r e quirements indicat e that e ach pum p must be capabl e of supply- ing fu e l at the pr e ssure and flowrate required by one en g ine at the maximum flow conditi¢ , , which occurs during se a level climb operation. To permit en- gine performance growth without the necessity of redes l gning the engine pumps, a margin of 10 percent excess capaci t y has been specified in the pump func- tion a l requirements. A draft of a general functional requirement specifica- ti o n f o r a pump s y s t em is shown in Table 22. [ l

i

!

I

TABLE 22. - FUNCTIONAL REQUIREMENTS Title: Pump, Fuel Boost, tank-mounted, Plug-in Motor Driven i. Scope This document defines the functional requirements for a sub- merged, motor driven, liquid hydrogen fuel boost pump. The pump shall have provisions for quick removal and replacement from the tank without having to remove fuel, plumbing_ or electrical wiring from the aircraft.

2. Applicable Documents (to be added) I 3. Requirements 3.1 Ports 3.1.1 Discharge - The discharge port shall be a four bolt flange type , sized for one inch tubing.

3.1.2 Pressure Sensing - Pressure sensing bosses shall be provided at each pump for sensing discharge pressure.

3.2 Lubrication - The pump and its driving motor shall be lubricated with a system compatible with hydrogen.

3.3 Pump Housing - A pump housing shall be provided which permits removal of the pumping element and driving motor without re- quiring that fuel be removed from the tank during the operation.

3.4 Check Valves - Check valves shall be provided in the inlet and discharge passages of the pump housing such that no fuel leakage can occur when the pump elements are removed.

3.5 Thermal Relief - The pump discharge check valve shall have a small hole vented to the tank to provide thermal relief.

3.6 Electrical - 3.6.1 Power - The pump motors shall be "Y" connected and shall be i rated for continuous duty at 115 / 200 volts, 3 phase, 400 Hertz, or as an alternate, 270 Vdc power.

3.6.2 Power Consumption - The power consumption shall be optimized for . the cruise operation.

TABLE 22 - Concluded.

3.6.3 Electrical Connection - The electrical connection between the removable element and the pump housing shall be automatical]y d_sconnected concurrently with removal of the pumping element and the motor subassembly, 3.7 Performance - 3.7.1 Fluid - The pump shall be compatible with liquid hydrogen fuel.

3.7.2 Operating Pressures - The pump inlet and discharge pressures shall be in accordance with the requirement of Figure __ 3.7.3 Flowrate - The pump flow requirements shall be as dictated by Figure_.

3.7.4 Environment - The pumping element, housing and driving motor shall be capable of operating in an environment established by the presence of liquid hydrogen stored at a pressure of 145 kPa (21 psia) absolute.

3.7.5 Priming - The pumping element shall be capable of priming itself if initially filled with gaseous hydrogen at start up, 3.7.6 Maximum Pressure - The maximum pressure output of the pump under any condition shall be compatible with the limitations of the engine systems.

3.8 Reliability - 3.8.1 MTBF - The mean time between failur e s per element shall not be less than 2500 hours using the definition: _BF = (Cumulative Flight Hours) (No. of Units /Aircraft) Cumulative Number of Chargeable Failures 3.8.2 TBO - The scheduled time between overhauls shall not be less than 8000 flight hour's.

3,8.3 Shelf Life - The unit shall have a shelf life of not less than 5 years with a capability of immediate service.

3.8.4 Safety - Safety concepts and design features shall be incorpo- rated in the pump and drive design. The pump shall be capable of operating dry in a hydrogen gas environment without hazard.

3.9 Pump Mounting Attitude - The pump assembly shall be mounted vertical]y with the pump inducer located at the low point in the storage tank.

During cruise, when fuel flowrates and the hazard resulting from engine flameout are considerably reduced, each pump must be capable of supplying two engines by means of crossfeed for added redundancy and to reduce electrical power requirements if the operator chooses.

Boost pump performance requirements were determined from the flow-pressure requirements of Table 19 and as a result of the concept selection trade de- scribed in Section 5.1.2. The flow-pressure schedule is an unusually wide range from the standpoint of thro t tling of the pump ou t put, and this factor has influenced pump selection considerably. The performance of the boost pu m p must be matched t o t hat of the englne-mounted main pump since no englne-to-tank return line is provided in the hydrogen aircraf t . The boost pu m p must provide a minimum NPSP of 3.4 kPa (0.5 psi) to the engine-mounted pump with minimum weight and power consumption.

Other design requirements are determined from the intended mission of the pump system within the aircraft. The pump m ust be designed for long llfe, 8000 hours being the baseline goal, The bearings m us t operate in LH 2 or, = alternatively, an acceptable thermal isolation system must he found. The pump drive must operate on available aircraft power systems. Several candidates were considered, but the choices soon narrowed to electrically driven pumps.

F o r the evaluation performed here, two aircraft electrical systems were con- sidered. Present conventional syste m s utilize 400 cycle power. It is pzo- Jected that by 1995 c om mercial aircraft may utilize 270 volt dc power systems, which have considerable advantages for aircraft applications. Both of these electrical systems were considered, and the details are given in Section 5.3,3.

The tank- m ounted boost pump must be safe in operation and easily maintainable.

The manufacturing costs should be as low as possible consistent with meeting other operating needs. Finally, the boost pu m p must meet all general require- ments of FAR 25.

5.3.2 Candidate pump types. - Four basic candidate pump types were considered: inducer, vane, piston, and centrifugal. In preliminary calculations, tandem row inducer pump designs were shown to have the lowest values of ADOC. How- ever, at the minimum flow condition (flight idle at ii 582 m (38 000 ft) and M = 0.85), they did not deliver enough pressure rise to meet t he specified main pump NPSP (pressur@ above vapor pressure) requirement of 3.4 kPa (0.5 psi).

This pressure rise might have been met by using the wide range, tandem row in- ducer design along with 50 percent flow reclrculation around the motor-boost pump unit. However, this would result in a pump inlet vapor volume fraction that might be too high for the pump to operate because an inducer pump cannot pump two-phase flow if the inlet flow coefficient is too far off design. This, in combination wlth the fact that such a design would have to approach an un- stable operating condition (which occurs in an axial pump that is operated at too low a flow) in order to meet the NPSP require m ent, resulted in a decision to use a pump design with a wider operating range capability.

" Positive displace m ent pumps would have design rotational speeds less t han i0 percent of those of centrifugal or inducer pumps and, therefore, would be I too hea_ w for this particular combination of head rise and volume flowrate.

Included in this positive displacement category are piston pumps and vane pumps. Additionally, positive displacement pumps typically have large sur- face areas that require some lubrication. This requirement would likely re- duce operating life significantly, since liquid hydrogen is not a good lub- ricant. For the_e reasons, vane and positive displacement pumps were not considered further. The only rer,aining candidate is the centrifugal pump, with its wide operating range, forgiving stall characteristic, and relatively small lubrication requirement. Constant-speed and variable-speed centrifu- gals were evaluated as to ability to meet flow requirements and ADOC minimi- zation. This evaluation is described in detail in Section 5.3.4.

For centrifug_l ,_umps that are designed for maximum efficiency, the performance characterisLics are shown parametrically in Figure 55 for design point operation (sea level climb at M = 0.38) and Figure 56 for minimum flow operation (flight idle at il 582 m (38 000 ft) and M = 0.85). It is apparent that stage numbers and d_sign speeds can be varied over wide ranges to g_ve whatever combination of characteristics is desired.

From Figure 55, it is appacent that, in this rotational speed range (less than 40 000 rpm to obtain an inlet diameter greater than i inch so as to pass the flow), multistaging is necessary in order to operate down to shutoff (Pm > 345 kPa (50 psi)). With these high efficiency types of designs, operation down to shutoff is possible if the stage specific speed is greater than about 3000. However, this may be done at a stage specific speed of only 1150 by designing specifically to obtain a wide operating range. This is achieved at the expense of approximately a 14 point penalty in efficiency.

It may be concluded that simplicity can be achieved at the expense of per- formance. Since both objectives a_e of interest here, wide range designs as well as high-efficiency designs were investigated.

5.3.3 Candidate pump drive systems. - Hydraulic, engine bleed air, and electrical pump drives were initially considered. Preliminary calculations Ehowed th a t the fluid line and system weights necessary for the first two choices for use with remotely-located tank-mounted boost pumps were prohib- itively high so that the choice was narrowed to an electrical drive. The aircraft electrical system may be either the standard 400 cycle ac system or a 270 volt dc system th a t has shown promis e for future aircraf t : applica- tions. Special controls are required if variable speed is to be used, whereas they are not if constant speed (which requires pump operation nearly down to shutoff) is to be used.

Brushless motors were assumed for the 270 volt dc case. The weights of these motors are shown in Figure 57 along with the weights of the correspond- ing electronic equipment required to operate a brushless motor over infinite ranges of torque and speed. In Figure 58 these motor and electronic equip- ment weights are summed to give the overall brushless motor assembly weights.

For this brushless motor data, four additional assumptions were made: (i) the stator is hydrogen cooied to reduce resistance and, consequently_ size ( 120 0 0 ,, 80 I'" _ . _ ' 7O i11 p- 60 ' > . - r. 3 Z " ' 50 u .

u.I I- . 4 0 z = - r 3 30 .......

DESIGNPOINT: HYDROGE N FLOW ' 20 RATE = 0.351 kg / sec (0.774 Ib / sec) . ; ALT . = SEA LEVEL M = 0 .38 24 28 32 36 40 • ROTATIO N AL SPEED , N , rpm X 10.3 Figure 5 5. - D e sign point p er f o r m ance f or c e nt r ifugal pu m ps designe d for maximum efficiency.

3 50 < " _ 40

_ 2s o

¢<3

_ :o _2 00 _ o

_ 1 5 o =

= - 20 _ . _ z 100 5 0 10 j __ --

_ 3 0

' - J U.

ul < 10 - " 24 28 32 36 40 DESIGN SPEED , N , rpm X 10.3 i Fig u r e 56. - Of f d e s i gn pe rf o rm a n ce f o r ce ntrif u gal pumps d e signed for max i mum e f fi c i e n c y.

1 22 tl 1.2 3 MO' r OR J 0 . 4 1 • I - - " 0 0 1.2 ELECTRO N ICS ' 0.8 J 0.4 1 0 0 2 3 4 5 f HORSEPOWER , hp z _ I J, I , _l ,_l 1.4 1.8 2,2 2 . 6 3.0 3,4 3,8 ' " POWE R, kW Figu re 57. - 27 0 V brushl e ss d c m o tor w e ight br ea kd o wn .

1 2 3 _k . I ,_1 ° l-I • • _ _ ' _ 0 ¢ } u t l ,,l "_ o

i "

Q U c 4 ° r ,t I ¢ q l '°PM ' ASS V OlN O H J.D=II:I + I:I010 N 40 1HDF:IM _" I 1 2 4 Q and weight, (2) there is no fluid in the air gap between the rotor and the stator, (3) hydrogen cooling does not result in any hydrogen vaporization, and (4) electromagnetic interference is suppressed to existing military - standards. Regarding these assumptions, it must be noted that keeping the hydrogen out of the air gap may be difficult to achieve and, in fact, may not even be desirable. Further investigation and, possibly, some technology efforts are required to establish the proper approach.

For 400 cycle ac motors, the weights for 24 000 rpm, 2 pole, constant speed motors are shown in Figure 59. If variable speed is desired, a rather large inverter weight must be added. This is also shown in Figure 59.

Finally, the efficiencies of these motors are shown in Figure 60. Due to the electronic equipment losses, a brushless dc motor is slightly less efficient than a constant speed ac motor. However, if an inverter is'used to make the AC motor variable speed, the large losses in the inverter drop the overall efficiency more than i0 percentage points. This is also shown in Figure 60.

5.3.4 Boost pump and drive candidate evaluation. - In order to evaluate candidate pumps, alternates must first be sized to meet specified flow conditions.

5.3.4.1 Pu m p sizing: The pump inlet diameter require m ent in the tank was determined from isentropic equilibrium expansion (from a saturated liquid) curves for hydrogen (Figure 61) and the maximum hydrogen flowrate per engine of 0.351 kg / sec (0.774 ib / sec). The resulting line size requirements are shown in Figure 62 as a function of tank vapor pressure and pump inlet vap o r volume fraction (assuming the flow has reached equilibrium). As shown, a one- inch diameter hole will pass the flow at a low vapor fraction for a tank vapor pressure of 145 kPa (21 psla). Since this is about as small a pump inlet as is practical from a manufacturing standpoint, this value was used throughout the pump selection procedure.

5.3.4.2 Pump selec t ion: Seven pump-drive combinations were investigated. As shown in Table 23 three were analyzed with ac motors and four were analyzed with dc motors. Within each m otor category, two of the combinations useo constant speed motors because there are two methods for obtaining throt t ling down to shutoff; (i) using a single stage, wide range pump (which has a lower e ffi c i e ncy), a nd (2) u sing a multistage, high ef fi c ien c y pum p . T h e o th e r c ombination within e a c h motor category us e d a variabl e sp ee d mot o r and, sin c e varlabl_ speed redu c es the requir e d number of stages, a high efficien c y t ype pump.

o i 1 25 i c _ u'J

o

o o o (' 4 I ., ' i w _

_\ o

o o 7 IL co : _ d _ N ql '.LH DI 3M ,-: t t i I, I , I [ J s... r ., B >I '.LHOF_ M 1 2 6 % • __ l l_ , u_ q_ , _ ,,- • j %, • 90 BRUSHLESSdc_ 8 5 g.- z U.

Ul 0: 80 -.

o I- O ' J " 24 000 rpm ¢j 90 ., J a o . J ,. I . J D 85 8O 2 3 4 5 HORSEPOWER, hp

, I I _ I I J. I

1.4 1 . 8 2 . 2 2.6 3.0 3.4 3 . 8 POWER,kW o Figu re 60. - El ec t r i cal m oto r eff l c l enc i es a t full l oad.

2400 - 82.7 450 -- 75 . 8 2200 -- 11 45 310.3 09 . 0 40 2 7 5.8 2000 - 400 -- 55 . 2 Po " PI' kPa 8 35 241.3 ¢" 4 m E 1800 350 _ 27 . 6 ._ . _ " _ " o psi a.

300 _ 25 172.4 < 1400 w 13.8 2 20 138.0 ,, n - < _ 250 -- I - 1200 -- b 1 5 - 103.4 x 1000 -- 200 -' _ . !

CHOKED W P. (MAX . _ v / A) _ " 150 -- _, F L OW PROCESS = EQUI L IBRIUM ISENTROPIC 600 -- TANK CONDITIONS = SATURATED LIQUID u_ w 100 -- SOLUTION METHOD NOT CRITICAL Z OFFICIAL HYDROGEN PROPERTIES "1 400 -- SUBSCRIPT 0 = TANK CONDITIONS 1 = INLET LINE COND I TIONS 50 -- Po P1 REFERSTO INLET TOTAL MINUS 200 -- I N LET STATIC PRESSURE

o- o J I " 1 I I I I

0 10 20 30 40 50 60 LINE VAPOR TO VOLUME FRACTION, PERCENT Fi g u re 6 1. - Z ero NP SH pu mp in g cap ab lZl ty re qui re m en t s £o r hyd roge n .

, 4 .., Il l I I -- I I I II I I I 2.0 " 5 . 0 -- 4.0 1,5 .....

• E

; '- k

\ VA P OR VOLUME PERCE N T " 1. - _ 10% w 3. 0 -- •

_o TM

I ,. . . c 1 . 0 • z , _ o .5 ....

1.0 -- 0 -- 0 0 10 20 30 40 50 psi

I [ I I I I I I , I J

_ 0 40 80 120 160 200 240 280 320 360 TA N K PRESSURE( S ATURATED) , PT kPa Fi g ur e 6 2. B o os t p um p inl e t dia me ter r e quir e m en ts .

12 9 1 3 1 As shown in the last data block (Minimum Flow) in Table 23, all the constant speed designs delivered a high pressure rise of 62 psi during mini- mum flow operation. This high pressure rise provides more than enough net positive suction pressure (NPSP, pressure above the vapor pressure) to meet the main pump requirements. However, with a variable speed boost pump, the boost pump speed can be adjusted so that the pump puts in only enough pres- sure (approximately) to meet the main pump NPSP requirements. As shown, this was assumed for the three variable speed cases (configurations 3, 6 and 7 in Table 23). This is also true during cruise. As a result, the boost pump power requirements are lower for the variable speed cases during off-design operation. However, in the analysis that was made, the main pump was assumed to be independent of the boost pump discharge pressure during cruise opera- tion, as is also shown in Table 23. As a result, the main pump discharge pressure probably exceeds the cruise requirements for the cases in which a constant speed boost pump is used. In these constant speed boost pump cases, variable speed main pumps would have to be used in order to match with the boost pump during all modes of operation. In summary, exact matching between the boost and the main pumps can be achieved only if one of the two pumps has a speed that can be set independently.

As shown in Table 23, configuration 2 (a 5-stage pump driven by a con- stant speed motor) has the lowest direct operating cost _DOC) of the ac driven candidate_ and configuration 7 (a 3-stage pump driven by a variable speed motor) is best for the dc candidates. Configuration 3, which has a variable speed ac drive, cannot compete with configuration 2 because the frequency converter required to obtain variable speed with ac is very heavy.

This is not true with dc bec a use the converter is much lighter for dc and because it is required for bo t h constan t and variable speeds. As a resul t , t he vari a ble speed drive (configuration 7) was the best wi t h dc. Also shown is th at the use o f wide range, single s ta ge pumps to ob t ain simpler configu- rations results in a decrease in pump efficiency and, consequently, an in- crease in oper a ting cost. This is particularly true with a c where the lower design speed results in a lower single stage pump specific speed which, in turm, results in a greater pump efficiency penalty.

Because the minimum allowable boost pump pressure rise during minimum flow opera t ion was originally unknown, several variable speed desigBs, each with a different'pressure rise at the mini m um flow condition, were analyzed to determine their pump and motor efficiencies during minimum flow operation.

The results are su mm arized in Table 24. These datap in turn, were used in a he a t transfer an a lysis to determine the res u lting NPSP's delivered to the main pump. Of the candidates lis t ed in Table 24, the one with the lowest _P minimum 34 kP a ( 5 psi) delivered 6.2 kPa (0.9 psi) NPSP to t he main pump, which exceeds the requiremen t . All of the other designs had even higher NPSP's. As a resul t , a ll the variable speed configurations in Table 23 (conflguratio_s 3, 6 and 7) will meet the main pump NPSP requizements.

13 2 _ 0 c _ c ; _ c _ 6 , 4 zo CY Z k 1 .- I _z fT .] _ 0 o 0 o 0 0 U l _ 0 - 5 _1 _=_ " 0 , _ _w . _ _ ' _ _ I

°[t

u_ u _ o . _ 1 3 3 I 5,3.5 Selected b oost pump a nd drive sys tem. - The selection o f the final boost pump configuration is de p endent on the e n gine pump and fuel control characteristics and is described in Section 5.6.

tk 5.3.6 Boost pump mountin g and changing tool. - An important part of this program has been to find a method by which the f u el pumps may be replaced qu ickly and safely, without req u iring that the liquid-hydrogen fuel tank be draxned. In this section, a configuration devised with this requirement in mind is described. Incl u ded are drawings and descrip t ion of the physical configuration of the pump mounting, the method of cha n ging the pump witho u t draining the fuel tank, and a tool designed to accom p lish the changing of the p ump. Particular c o nsideration has been given to ensuring the safety and reliability of the configuration and the method for changing the pumps.

Figure 63 shows the physical configuration of the pump mounting. Three pumps are pl a ced on a single mounting unit. Since FAA reg u lations re q uire that two pump_ be operable for all takeoffs and landings, this choice allows continuation of missions where one p u mp has failed at some interme d iate time.

! This capability is desirabl e , since all intermediate stops of a flight may not be equipped or convenient for changing a cryogenically cooled pump.

The pump and housing are shown in cross section in Figure 63. Each pump 4 has its o_i inlet while the discharge is common for the three pumps. The pump cavity housing is roughly cylindrical with the inlet located at the bottom of the fuel tank, so that the fuel may be used entirely. Fuel enters at the pump inlet and passes through the inducer and three impellers and ' exits through a check valve into the fuel line leading to the engine. The check valve is provided to ensure that fuel will not be pumped in reverse direction through a pump(s) th a t is n ot operating.

The pump is contained within the housing by means of l o cking lugs. To change a pump t he insula t ed p a nel covering all three pumps is rem o ved a nd the pump changing t o o l sh o wn in Figure 64 is secured to the p u mp housing ex t ernal locking l u gs. An "O" ring is provided to seal the tool to t he pu m p io wer housing surface. The helium lines are then at tached to t he pump ch a nging tool, the GHe inlet valve and a ir escape valves opened a nd t he changing tool (and new pump) purged of a ir. When this is c o mpleted, the GHe line is c o nnected t o t he pump purge port located in t he p u mp lower housing plate. The t oo l is then rotated 0.785 rad (4 5 deg) to disengage t he lugs and the pump pulled down s o th a t the pump inle t closure sleeve blocks the pump inle t thereby preventing the escape of the LH 2 in the tank. At this point the pump GHe purge valve is opened admitting GHe to the top of the pump, forcing the LH 2 trapped in the pump back into the tank vi a the s m all check valve and dis- ch a rge passage in the pump housing. When this is co m ple t e, the operating handle is turned another 0. 7 85 rad (4 5 deg) to disengage the closure sleeve locking lugs a _ the pump is completely withdrawn and placed on the carrier plate in the ch a nging tool. The new pump is t hen moved in t o position by means of the carrier plate operating rod and the procedure reversed t o replace the pump (no further purging is required h o wever). The purging lines are then re m oved, the changing tool removed and the cover plate replaced. The fluid discharge and / i 1 35 Figure 64. - B o ost pump r e pla ce m e nt co n c ept.

electrical connections are made simultaneously during the final travel of the pump into its locked position. Since the new pump is warm and quickly cools to liquid-hydrogen temperature, the bellows-mounted seal is provided to absorb " the thermal expansion of the pump during its temperature change. Such seals have proven effective on previous Rocketdyne pumps producing considerably higher pressure than the approximately 414 kPa (60 psi) maximum required here.

This approach to an easily replaceabl e pump unit has several advantages.

Because both the pump cavity and the pump-changing tool are purged, there is no chance foreign matter such as particulates or vapor that might condense or freeze at liquid-hydrogen temperatures can find its way into the pump cavity, even if the pump changing operation is delayed with the pump cavity exposed.

Second, the proposed approach ensures the safety of the mechanic who performs the operation. Should there be any failure during the pump-changing opera- tion, the tool prevents the escape of hydrogen that might injure the mechanic or result in a hazardous condition. Third, the mechanism of the pump-changing tool, together with appropriate guides built into the pump cavity ensures that the replacement pump will be inserted into precisely the correct position to seal with the bellows-mounted seal. Finally because the replacement pump is protected inside the pump-changing tool, there is little chance of its being damaged prior to insertion. All handling of the pump itself may be accom- plished within a controlled workshop environment rather than on the field.

It is estimated that a pump may be changed in 10-15 minutes by means of the approach described here, by one or two mechanics. Because the tool may be constructed of aluminum alloy, it should weigh on the order of 9.1 kg (20 ibs). Thus, the replacement pump and tool unit should weigh approximately 13.6 kg (30 ibs) and may be carried by a single person. However, a second per- son may be required to attach the tool to the lower side of the fuel tank, par- ticularly if the access is in an awkward position. The tool will become cold as the pump-replacement operation is conducted through its contact with the cold pump and the tank. Thus, the mechanic must take the precaution of wearing gloves during the operation, but no other special protection is required.

5 .4 Engine Fuel Pump The engine fuel pump requirem e nts are to provide a high pressure rise and to comply with the severe demands of air transport service while oper a ting in the liquid hydrogen environment represented by a low net positive suc t ion head (NPSH), cryogenic tempera t ure, and low viscosity. This secti o n discusses the implications of these requirements for the engine pump, presents the more significant design trade-offs, provides the results of a selected design , approach, and recommends certain items for advanced tech_Lology development.

1 3 7 B 5.4.1 Design requirements. - The pump requirements start with the engine fuel flow and delivery pressure requirements which were developed in the engine study reported in Section 4.0. Table 25 presents a summary of the more sig- nificant operating conditions for the baseline engine, and tabulates the re- _ qulred fuel flow and fuel delivery pressure for each of these conditions. The engine LH2 fuel pump is required to meet this set of requirements for engine fuel flow and delivery, with stable nonpulsatlng flow.

In order to avoid severely penalizing the engine fuel supply system which delivers LH 2 to the engine fuel pump, it is necessary that the engine pump not require an excessively high fuel inlet pressure. Based upon pre- liminary studies of the engine fuel supply sys t em, it was agreed to consider the condition of a saturated liquid at 345 kPa (50 psla) as a defi n ition of the state of the LH 2 at the engine high pressure pump inlet for design pur- poses. All subsequent pump investigation was based upon this assumed pump inlet condition for steady state operation. Other pump design requirements were the following: • During starts, it was assumed that the engine pump may encounter significant vapor associated with heat soak _nto the aircraft fuel line, and that either this vapor would have to be vented, or some scheme would have to be established for passing it through the pump.

• Pump rotational speed was not constrained, except as it may be by the selected drive system.

• T h e pump should be designed to minimize aircraft DOC, associated with pump weight and required input power.

• The minimum time between overhaul (TBO) upon entry into air transport service was established at I000 hours.

• Design for flight reliability and flight safety was an overriding requirement.

• Requirements for vapor venting, and pump thermal preconditioning, or other unusual operational constraints associated with the use of LH 2 fuel were to be eliminated or minimized.

5.4.2 C andidate pump types and selection of preferred concept . - P um p types which were considered to be potentially feasible for the proposed application were • Centrifugal pu m ps, single or multistage • Positive displacement piston pumps • Positive displacement vane pumps.

.......... o ° o _v Z o . _ _ _ . , _ H o _ ,9 ° I ' ,-_ <r C,,I • _ ^ o _

_ o _ _ ° _ o _

5 o v° _° 8 o " _5 u : = u 1 3 9 References 5 and 6 describe the preliminary development of a five cylinder piston pump for LH 2 service. This work demonstrated the feasibility of piston type pumps, but also highlighted the problems of leakage and l_fe, and the necessity to operate at very low speed which resulted in a physically large pump for a given flow rate. Reference 7 reports similar work with a LH 2 p_ton type pump which was tested by the General Electric C o . In addition, Refer- ence 7 reports that tests of a cryogenic vane pump were unsuccessful.

Because of these reported limitations of positive displ a cement pumps in the existing state of the art, and because centrifugal type pumps for LH 2 service have been relatively successful, the decision was made to concentrate the re m ainder of this limited investigation exclusively on the use of centrif- ugal type pumps. Further serious consideration of positive displacement LH 2 pumps for air transport service must be preceded by successful detailed in- vestigation aimed at resolving the presently known design deficiencies of this type of equipment.

Both single and multistage centrifugal pumps were considered. The single stage pump represented the simplest design, whereas the use of multistages had the advantage of greater efficiency since it permitted the pump to operate at a more favorable specific speed.* Illaddition, the multistage pumps were smaller in diameter thus reducing the impeller thrust loads, and facilitating packaging. The following table shows the comparison of size and efficiency for various numb e rs of pump stages.

Number Impeller Estimated of Stage Die Efficiency Stages Ns cm (in.) % 1 257 13.4 (5.28) 50.5 2 432 9.5 (3.73) 60.2 3 585 7.7 (3.05) 66.0 B a s e d upon t his comparison, the 2 stage centrifugal design was selected as a reasonable compromise betw e en design simpli c ity, pump efficiency, t hrust load, and packaging feasibility. Subsequent work was bas e d upon use of a two-stage design.

I 1 9 * Spe c ific speed Ns = _ where N = rpm Q = gallons per minu t e H = head in feet

140 i

5. 4.3 Candidat e 2_m_ . d_ive systems. - F our candidate pump driv e syst em s were considered: • • Bleed air driven turbopump • 2 7 0 Vdc electric motor driven pump • Fixed r a tio _haft driven pump • Variable rauio shaft driven pu m p.

Figure 65a shows the bleed air driven turbopump syste m which was considered.

The power source was high pre. sure bleed air ex t racted from the engine co m - pressor. This was then ducted t o a high speed air t urbine and t he flow t o the turbine was modulated to an : nlet valve. The air turbine was used t o direct drive a centrifugal type LHp pump to provide LH 9 to the engine. One main advantage of this type of dri g e was the ease wlth'which variations in pump speed could be obtained.

Another type of drive which was considered was based upon the recen t development of high efficiency, 2 7 0 Vdc electric generators and m otors having at once a capabili t y for both high speed and variable speed. Control flexi- bility was a main advan t age of t his type of drive, if it should prove possible to design a system having a competitive weight. Figure 35b shows a , chematic of a 270 Vdc pump drive system.

" The simplest type drive which was considered was a fixed speed ratio shaft drive, using power extraction from the engine gearbox. This type drive is shown schematically in Figure 65c. The inabili t y to vary the speed of the pu m p was seen as a possible major problem wi t h this concept.

A variable speed mechanical drive can be used t o obtain pump speed variation with a sh a ft driven syste m , and this scheme is shown in Figure 65d.

Added m echanical co m plexity was seen as a possible m ajor drawback for thl8 approach.

The co m parative evaluation of these drives is discussed in the next sec t ion.

5 . 4.4 Engin e pump and drive candidate evaluation. - A trad e -off an a lysis wa s conduct e d to s e l e ct a pr e ferr e d pump and drive sy s tem. In p er forming th e analysis, e ngin e operating conditions w e r e first r e vi e w e d with p a rti c ular r efe r e nc e to f u e l d e liv e r y r e quir e m e nt s as summ a rized on Tabl e 25. Two con- , ditions were selected as having particular significance in the pump drive t rade-off analysis: • Ta keo ff (Condition 4) was s e l ec t e d as on e point for c omparison, sin c e it r e pr e sent e d th e maximum pump power requirem e nt and h e nc e would d e t e rmine th e drive w e ight.

• Start of cruis e (Condition 6) was s e l e ct e d as anoth e r point for comparison, sinc e it w a s typical of long t e rm cruis e op e ration, and would r e flect the influence of pump drive overall effici e ncy on engine SFC.

Th e pump driv e s to b e comp a r e d r e pr e s e nt two basic ty pe s, variabl e sp ee d ratio and fixed sp ee d ratio. The first thr ee candidate driv e typ e s, discuss e d Jn Section 5.4.3, had a variabl e sp ee d ratio. The variable sp ee d typ e has th e advantage of p e rmitting adjustm e nt of the pump sp ee d and r e - suiting fuel d e liv e ry to mor e n e arly mat c h the engin e requir e m e nts. This results in a r e duction cf required pump input power at off-d e sign conditions, such as cruis e . A fixed spe e d ratio pump driv e do e s not p e rmit such a sp ee d adjustm e nt and h e nc e r e quir e s r e l a tiv e Iy great e r power input during th e off- d e sign operation, when compar e d to th e vari a ble sp ee d s ystems. Tabl e s 26 and 2 7 show summaries of pump oper a ting conditi c ns for both the variabl e speed ratio driv e s and a fixed sp e ed ratio driv e .

Th e s e data w e r e th e n us e d in d e v e loping weight estimat e s and pow e r r e - quir e m e nts for the various drive systems. Th e a tta c h e d s,_=ary c hart, Tabl e 28, shows a c omparison of th e significant cnaract e risti c s of th e al- t e rnativ e pump drives. Drive syst e m overall e ffici e ncy is pr e s e nt e d for both takeoff and start of cruise conditions, and also th e r e sulting r e - quir e d e ngin e pow e r e xtr a ction. Syst e m w e ights ar e shown. Th e chang e in DOC was calculated for th e differ e nt syst e ms, and is shown tabulat e d as a relativ e r a nking. In addition, th e various driv e s wer e rank e d according to design simplicity, inh e rent r e Ziability potential, and system cost.

In e v a luating th e r e sults of trad e -off analy se s pr e s e nt e d in this su_ a ary, it w a s d e cid e d that th e delta DOC numb e rs d e velop e d for th e start of cruis e condition should be giv e n r e lativ e ly littl e w e ight sinr" the absolut e v a lu e s w e r e quit e small. Primary signific a nce was assign e d to syst e m d e sign sim- plicity and inh e r e nt rel_ability pot e ntial, with syst e m weight and cost con- sid e r e d n e xt. On this b_si s , th e fix e d ratio s haft driv e was s e l e ct e d as th e pr e f e rr e d pump d r iv e . T_e bl ee d air driv e n turbopump is consid e r e d to be th e s e cond most a ttra c tiv e alt e rnativ e . Th e 270 Vdc syst e m has th e d e cided ad- vantag e of not r e quiring a shaft dynamic s e al in th e pump, but h a s th e dis- advantag e of r e lativ e ly high syst e m weight. Th e variabl e sp ee d r a tio shaft driv e has th e disadvantage of m e chanical c ompl e xity.

5 .4.5 Pum p bearin_ con sid erations . - Th e e ngin e EH 2 f ue l pump b e arin g s y s t e m r ep / e sents on e of the k e y t e chnical probl e ms in th e d es i g n of th e ov er all syst e m. Th e b e arin g syst e m must b e ca pabl e of op e rating at hi g h rotational sp ee d s (50 000 rpm); must b e c a p abl e of carryin g hi g h load s, particula r ly hi g h • _hru s t loads under ce rtain c onditions; a nd must be c omp a tibl e with th e pumping of LH 2 .

Rocket engine turbopumps have typically used rolling element bearings operating in LH2. The LH 2 provides cooling but essentially no lubrication because of the very low viscosity; lubrication is provided by the use of teflon type separators, wherein the teflon transfers as a solid film lubri- cant to the rolling elements and races. These systems have demonstrated the required speed and load capability but have a demonstrated maximum life of about I0 hours, although extremely lightly loaded bearings have been run over i000 hours. Therefore, if rolling element bearings operating in LH 2 are to be used in the engine LH 2 pump, it will be necessary to assure extremely light loading under all operating conditions.

Hydrostatic LH 2 bearings, and hybrid bearings consisting of a hyd[ostatic bearing used in combination with a rolling element bearing have been proposed and tested experimentally, but have not yet demonstrated the capability of meeting the engine fuel pump requirements.

Foil bearing systems have been tested by AiResearch with air and cryo- genic helium, and appear to offer an attractive design alternative for LH 2 systems, although the loading would have to be controlled to a low value.

Industrial cryogenic turbomachlnes utilize oil lubricated bearings, both rolling element and plain journal bearing types, but require strict thermal " control within the machine to prevent freezing the oil. However, the oil lubricated bearings have both high rotational speed capability, and substantial load carrying capacity.

t Based upon these considerations, a preliminary selection of a bearing system was made for the proposed shaft driven pump described later. This system used an oil lubricated rolling element bearing at the shaft drive end of the pump. The bearing would receive oil from the engine gearbox and would operate at gearbox temperatures. Careful thermal design of the pump would be required to successfully use this design approach. The high load capacity of the oil lubricated bearing would be used to carry the pump thrust loads, which can be substantial under certain conditions of impeller seal wear and leakage.

In addition, the oil lubricated bearing would carry the local radial loads at the shaft drive end of the unit. A foil type bearing (or rolling element alternative) was located between the two pump impellers and operated in the cryogenic hydrogen. With careful attention to dynamic and hydrodynamic balance, the loads on this bearipg can be maintained at a suitable low value.

It was considered that this hybrid approach offered the potential for meeting the engine LH2 fuel pump bearing system design requirements, and it was recommended for advanced technology development.

p 5.4.6 Selected pump and drive system. - 5.4.6.1 General description: The selected engine high pressure pump is a two stage centrifugal design, and is shaft driven from the engine at a fixed speed ratio. The pump is designed to provide a flow of 387 _ / min (102 gpm) at a pressure rise of 4723 kPa (685 psid) with a design rotational speed of 50 000 rpm. At the design point, the required shaft power input is 50.5 kW (67.7 hp) and the required condition for the hydrogen at the pump inlet is 345 kPa (50 psia) minimum at 25°K (45°R) maximum. These limits correspond to saturated liquid (0 NPSH) at 345 kPa (50 psla). Table 29 provides a more detailed summary of the pump operating conditions and design chailcteristlcs.

Referring to the pump cross-sectional drawing shown in Figure 66, it may be seen that the pump rotating group consists of the two impellers and the shaft, fastened together with curvic couplings and an axial tie bolt.

This construction is typical for modern, small, high speed turbomachines.

The curvic couplings provide the required accuracy in alignment of the pump rotating parts and, in addition, can be made with convex-convex generated surfaces, thus reducing the surface contact area and increasing the resistance to conductive heat flow along the shaft. Minimizing the heat flow from the warm engine gearb o x to the cryogenic end of the liauid hydr o gen pump was an important design requirement.

A splined torsion drive shaft is used to connect the pump rotati_ig group to the engine gearbox. The torsion shaft provides the torsional compliance necessary to isolate the gear tooth generated excitation from the inertia of the pump rotating group and, in addition, it provides additional resistance to conductive heat flow. As shown later in Figure 72, a fluid coupling is in- corporated in the drive train in the engine gearbox. This is to permit the pump to be disengaged from a windmilling engine in the event of an in-flight shutdown.

The pump rotating Broup Ls carried in an oil ]ubrlcated ball bearin_ at the gearbox end, and in a foil type journal bearing running in hydrogen at the pump end. This bearing arrangement has a particular advantage in that the high load capacity oil lubricated bearing not only carries the radial load at the gearbox end of the pump, but also carries all of the axial thrust load, so that the foil bearing running in the cryog,nic hydrogen at the im- peller end of the pump is only required to carry the local radial load. With accurate dynamic balancing of the impellers, the magnitude of this local radial load can be kept relatively small. This provision for carrying the thrust load was an important consideration, since the possibility of encountering high thrust loads always exists for high pressure centrifugal pumps, for instance if a labyrinth type thrust balance seal were to develop an abnormally high leakage rate.

¢ The ball bearing is a preloaded duplex pair carried in a ring type flexible mount. The flexible mount provides the necessary radial compliance to accommodate the greater than normal temperature range expected for this application and, in addition, provides some angular compliance for the shaft, TABLE 29. - ENGINE MOUNTED, HIGH PRESSURE LH 2 PUMP CHARACTERISTICS Conditions: Inlet Pressure 345 kPa (50 psia) (Total Net) Inlet Temperature 2 5.2OK (45.4°R) (Saturated) Discharge Pressure 5068 kPa (735 psia) Flow 386 _ (102 gpm) (Liquid) Design Data: No. of Stages 2 Total Head 7 509.7 m (24 638.13 ft) Speed 50 000 rpm First Stage TSH (a) 70.6 m (231.75 ft) First Stage SV (b) I0 000 Stage Specific Speed 431.5 Overall Efficiency 60. 2 % Overall Power 50.5 kW (67.7 hp) Impeller Diameter 9.47 cm (3.730 in._ Impeller Reynolds No. 2.29 x 101 Impeller Eye Dia. First St. 2.50 cm (0.985 in.)

Impeller Eye Dia. Second St. 2. 2 2 cm (0.875 in.)

Impeller Tip Width 0.14 cm (0.055 in.)

(a) TSH = Thermal Suction Head NQ I / 2 (b) SV = Suction Specific Speed = 3 / 4 (N_H + TSH) necessary to accommodate the radial compliance of the foil bearing at the pump end. The ball bearing pair is lubricated by a slight oll mist from the engine gearbox, and a drain-back port is provided to return any collected mist to the gearbox. Direct spray lubrication was not planned for this bearing and may be objectionable because of the possibility for very low operating temperatures. A dynamic shaft seal was provided to retain the oil mist in the gearbox, and an overboard drain was provided to accommodate any oil leakage from the seal.

Because the thrust bearing is some distance from the impellers in the " selected bearing arrangement, there is some possibility for axial misalign- ment of the impeller and diffuser center lines. To accommodate this without a harmful degradation in pump performance, the impeller tip width was made somewhat larger than the diffuser entrance width.

The foil bearing located between the two pump impellers is an AiResearch proprietary design similar to the bearings used in a variety of other high speed turbomachines. The advantage of the foil bearing is that it offers the potential for a long service life in the cryogenic end of the pump where oil lubrication was not feasible. An alternative design approach was to use a rolling element bearing operating in the cryogenic hydrogen, in lieu of the foil bearing.

Labyrinth seals are used for controlling leakage and for obtaining thrust balance across the pump impellers. Referring to the cross-sectional drawing of Figure 66, it can be seen that the first stage impeller inlet labyrinth seal is vented to a location about six inches upstream of the pump inlet. This was done to minimize vapor flashing in the pump inlet, which would be detrimental to pump performance. The rear labyrinth seal for the second stage impeller is also vented, for the purpose of reducing t he re- quired design pressure of the hydrogen dynamic shaft seal. A hydrogen vent is provided in t he housing to accommodate a ny leakage from the hydrogen dynamic shaft seal.

The pump inlet housing assembly is mounted to t he bearing housing asse m bly by three radial pins. This arrangement facilitates radial contrac- tion of the pump housing at cryogenic temperatures, and also reduces the conductive heat transfer. A thin gage convoluted seal is provided t o pre- clude leakage. The pump assembly is m ounted to the engine gearbox by a standard AND 20002 15.24 cm (6.00 in.) flange.

5 .4.6.2 Materials: Materials selected for use in significant parts of the engine LH 2 pump were: • Impellers - Inconel 718 • Diffuser housing - aluminum alloy • Main housing - 300 series corrosion resistant steel • Rolling element bearings - 400 C corrosion resistant steel • I m peller shaft - Inconel 718 • Splined drive shaft - Nitralloy • Oil seal assembly - carbon, 400 C corrosion resistant steel, and 300 series corrosion • resis t an t s t eel.

o 5.4.6.3 W e i ght: The estimat e d t otal weight f o r t h e tw o st ag e sh a ft dr i v e n c entrifugal pump was 5.9 kg ( 13 . 1 ib ) . This was obtain e d by detail e d e stim a t e of the weight of th e individual compon e nts shown on the layout of Figure 66.

5 . 4.6.4 P e rformance: Performan c e maps showing pump h e ad and efficiency as a function of pump flow and rotational speed are shown in Figur e 67.

5.5 Fuel Control System 5.5.1 Design r equirem ents. - Th e initial und e rtaking in th e design of a fuel deliv e ry and control system was to revi e w the r e quired functions, establish a list of inputs needed to perform those functions, and itemize th e required output.

Functions of th e engin e fu e l d e livery and c o ntrol syst e m include: , • Provide the interface between the engine fuel supply system and th e engine. The control syst e m receives fu e l from the e ngin e fuel supply system within a limited range of thermodynamic stat e s, and d e livers this fuel to the engine in a condition which provid e s for effici e nt combustions and at the proper flow rate for all engine op e rating conditions.

• Provide scheduling of compr e ssor bleed valves used during starting, and compressor variable vane positions.

• Pr o vide sch e duling of other valves and / or ignition r e quir e d during engin e pr e start conditioning, and starting, and shut-down.

I Inputs to the engine fuel delivery and control system include: i Physi cal Inputs • Fuel from the e ngine fu e l supply system, provid e d in ac c ordan ce with a flow s c hedule establish e d by th e engin e fuel flow requir e m e nts, and in a c cordan c e with a minimum pr e ssur e s c h e dule e stablished by the e ngine high pressur e pump suction p e rforman c e limits.

• Electrical pow e r from the aircraft system for use in the engine fuel deliv e ry and control system e le c tronic control, for use in operation of valves and actuators, and for possible use as a pump drive.

1 5 2 1 5 3 P • Engine compressor bleed air for use in the engine compressor variable vane actuators, and for possible use as a fuel turbopump drive.

• Engine shaft power for possible use as a fuel pump drive.

• Heat for use in vaporizing and heating the hydrogen fuel, to be obtained from the aircraft environmental control heat load, from the engine turbine cooling air, ard from the engine exhaust.

Informational Inputs • Command signals, including Electrical system master switch Engine fuel system purge operation Engine start signal Engine power level setting Engine stop signal.

• Informational inputs from the engine, including Fan rotational speed • Compressor rotational speed Compressor inlet total temperature Compressor inlet total pressure Compressor disch a rge total pressure Compressor variable guide vane and stator vane positions Low pressure turbine inlet total tempera t ure, or exhaust gas total temperature.

• Informational inputs generated within the fuel delivery and control system, including Pump inlet housing temperature Pump discharge pressuce Pump discharge temperature Pump rotational speed Fuel flowmeter rotational speed Outputs from the engine fuel delivery and control system include: Physical Outputs • Fuel deliv e red to the engine combustor, in a c ondi t ion whi c h provides for e ffici e nt c ombus t ion, and at a proper flow rat e for all engine starting, transient, and steady s t ate operating c onditions.

1 54 r • Vent gas during cool down or purging operations.

• Pump shaft dynamic seal vent gas.

Informational Outputs • Engine speed or thrust indication* • Scheduling of compressor bleed valve • Scheduling of compressor variable vane positions • Signals for monitoring of fuel delivery and control system significant parameters, such as fuel pump rotational speed, control valve positions, etc.

5.5.2 Candidate concepts. - Candidate concepts for the fuel control system which were studied and evaluated were based on use of the following pump drive systems: • Bleed air driven turbopump system • 270 Vdc motor driven pump system . • Engine shaft driven pump system, fixed speed ratio • Engine shaft driven pump system, variable speed ratio.

System schematics for these concepts are shown in Figures 68 through 71.

In each of these systems, electronic control circuitry was used in conjunc- tion with the fluid pumping and metering elements. This use of electronic circuitry is consistent with modern engine design technique and will prob- ably be used e xclusively on this class engine in the 1985 time period.

All of the schemes use a flow modulating and shut-off valve downstream of the heat exchangers to reduce the effect of heat exchanger capacitance in fuel system transient performance. A turbine type flowmeter is included for fuel flow measurement upstream of the flow modulating valve.

5.5 . 3 Sel e c t e d system.

" 5. 5 . 3 . 1 Descr ip t i on: Sel ec t i on of the de si gn o f the fuel co ntrol sy s t e m wa s dependent on the c ho ic e of drive for the engine fuel pu m p . Wi t h the sel ec t i on of the fixe d sp e ed rat i o eng i ne shaf t dr i ven pu m p sys t em as • * O t h e r e n g in e moni t oring param e t e r s a re not c onside r ed par t of th e e ngin e fuel delive r y and c on t rol sys t e m .

L 0 m (_. .d --t_ - . 1- u 4 l : _ _-r _ r i I_I Qo GO l : _ ' _ a: u _ 0 Z CC l - _-< o _ .1 l ' ;-_J II ,[, z 5 _ I I_ II ILl = _ _<_T_.LI I_ _ ' .

Z [ .................... J CC E 0 " . • k_ j u. 3E r --> t "' _

t - I tl ,,,._

_ _ _ _ t I i I I :.' 1 _ -- _

<_"_, i i l I _ l_ _ >_

_ -_ ......... _ : i_ , .=__ ..q - __._._ , _ , Zll_ -- , ,,, m , < t--

_ X

I ,- t- _ Z < _ u.

! k _ °°o

1 5 7 discussed in Section 5.4.4, the contr o l syste= represented in Figure 70 beca m e the preferred concept. Figure 72 shows the actual design arr a nge- ment of the components in the overall aircraft system. A description of the operation of this system is discussed in the following section.

5.5.3.2 Control response considerations: With the design arrange m ent shown in Figure 70, the potential problems associated with the effects of the large heat exchanger volume capacitance and the H 2 fluid compressibility are minimized. It is expected that control response characteristics com- parable to that of a conventionally fueled engine can be readily achieved.

Note that particular attention will have to be paid to the design and development of the fuel flow modulating valve and the turbine type flowmeter.

5.6 Engine Fuel Supply System Final Design and Performance The critical problem in the delivery of fuel from the tank to the engine is to ensure that the engine mounted pump is delivered a supply of l_quld hydrogen at a pressure such that no significant amount of vapor is present (two-phase flow), although it is estimated that at low speeds the main pump could han d le a vapor-to-volu m e fraction of approximately one half. This means that the he a t added by the tank boost p u mp, mo t or, supply llne, valves, etc., cannot exceed the fuel saturation enthalpy associated with the pressure at the engine pump inlet (zero net positive suction head). Since the heat added by the lines and system is proportional to the area of the llne and inversely prop o rtional to the flow rate, the lowest flow rate is t he most critic a l. The heat added by the boost pump is proporticn a l to the pressure rise buu inversely propor t ional t o the pump efficiency. A high p_essure £1se across the boost pump is desirable to suppress vaporization in the delivery system but the pump efficiency corresponding to the pressure rise must also be considered.

Another important consideration which influenced the system configur a - t ion and concept was the deslrabili=y of being able to use the hydregen vapor in the delivery system during engine start. This would preclude the necessity for either having a long vapor return llne back to the tank fro m the engine or providing a m ethod to safely vent or du m p the vapor. In a ddition, the residual vapor would then be use a ble in the engine. This could be accom- plished by designing the engine fuel control qystem to handle vapor during the starting transient condition. Th e pressure r e quired to insure fu e l delivery to the engine and into the combustor could be supplied by the boost pump which is immersed in llquid. The increasing flow rate @uring accelera- tion from s t ar t ing to ground idl e would chill down the system so that liquid hydrogen would be available at the engine pump prior to reaching Idle.

The abov e concept was pursu e d during the course of the study and was selected as the final approach described in the following sections.

!

1 5 9 ,t J I 5.6.1 Heat added to hydr ogen. - During system operation, the liquid hydrogen temperature at the engine pump inlet will be higher than the tanked value due to heat inputs from the pump and (submerged) electric motor as well as heat "I input along the length of the fuel feed line. The primary concern is to en- cure that the hydrogen vapor volumetric fraction ne v er be greater than 0.5 at the engine pump inlet under all operating conditions. In actuality, it is desirable to ensure that some excess in engine d e liver e d pressure compared to inlet saturation pressure is maintained. This excess p r essure is usually referred to as net positive suction pressure (NPSP) and a minimum value of zero was specified fo r normal operations.

The hydroger : saturation pressure at the engine pump inlet cqn be ob- tained from the cslculated fluid enthalpy at the inlet in conjunction with hydrogen property tables. The engine inlet pressure is simply the ta,k pressure plus the boost pump pressure rise minus the feed line pressure loss. Therefore, we have NPSP = PENG - PSAT = PTANK +Z_PP u MP " 'ZXPL I NE - PSAT (i ) The engine pump inlet enthalpy is equal to the hydrogen entha!py at tanked conditions plus the e nthalpy rise attributed to line, pump and motor heating, respectively: q heng = hTANK + &h L + _hp + Z_ M (2) : The enthalpy rise due to line heating _h L is determined simply from: Q / L • L _h L = . (3) ' WH 2 The enthalpy rise due to p.',,p heating is given by: &h = _P " - + 0.017 (4) P J which for liquid hydrogen with a density of 4.3 Ib / ft 3 and using the proper conversion factors to obtain consistent units reduces to: _%h = _P ' 0.0412 i _ + 0.017 (5) p J _ / _ , m where _P is pump pressure rise (psi) and qp is pump efficiency. The numeric a l factor 0.0175 accounts for compressibility effects on the hydrogen internal energy.

i The fluid enthalpy rise due to th e submerged electric motor inefficiency is _ given by: _hM 778 WH2 where hp is motor horsepower and qM is motor efficiency.

i 5.6.2 Final system selection. - The a bove relations, together wi t h the line, joinu, a nd valve heat leaks were used to determine conditions during the critical ground start of the engine. The boost pump selected to give max- imum efficiency at low flows is a three staEe, 2 7 0 volt dc driven cen t rif- ugal, variable-speed pump as shown in outline in Figure 73. This pump is the same as configuration 7 in Table 23.

The assumptions used in the analysis were: • Starting flow rate is 0.011 kg / sec (0.024 Ib / sec) • The longest line run was used (tank #4 to engine #4) • Compartment temper a ture = 54.4°C (130°F) at sea level • One inch diameter stainless steel line • Line is chilled down at engine start.

The o bjective was to c o mpare foam versus vacuum insulati o n sys t ems for the engine fuel supply llne.

Since the in-service reli a bility of light weight v a cuum jacke t ed line is unknown, but based on experience with static ground equipment is not expected to be very high, the foa m jacketed concept was included in t his analysis. The foam line consists of concentric tubes filled wi t h 1.5 inches of closed cell foam with suitable bellows a nd connectors. Being a passive . system, t he consequences of a leak in t o the closed cell foam space will not, in the short term, incre a se the heat leak rate and it is expected to be m ore reliable and rugged overall.

The analysis was based on the boost pump and motor characteristics " shown in Figure 74. Transient conditions during start are shown in Figure 75.

Heat rejection to the fluid from the pump and motor as well as line, Joint, and valve heat leaks, are included in the fuel temperature rise.

T L 11.43 cm _ " (4 . 50 in.) dia \.

28.24 cm (11.12 in.) j _ _ (6.75 in.) dia 17.15 cm __ POWER 27 0 Vdc P UM P WEIGHT 0.5 0 8 kg (1.12 Ib) 3-STAGE , VARIAB L E SPEED CE N TRIFUGAL rpm 36 000 max.

MAX. kW (hp) 2.36 (3.16) SYSTEM WEIGHT 2 8 .8 kg (63.5 Ib) Figure 73. - Sele c ted LH 2 bo o st pump.

!

I J !

500 - THREE STAGE - 270 Vdc - VARIABLE SPEED CE N TRIFUGAL PUMP * I i

_ o !

o 4 0 0 -- 60 .......... "

i

5C ° _ 300 _ .

// __ , : _ 4C 200 3C 2(

/ i

100= -- 1

4 0 i

!

i 30 / E 20 !

T 40 I 0.039 kg h ec i 30 100 (0,085 Ib / sec) . _ - 'r/ PUMP Jr o : _ , 20 80 _ " _ . _ " r/ MOTOR : _ ; 0 40 ; 0 0.05 0 . 10 0.15 0.20 " L, I ! I I I 0 0. 0 2 0.04 0.06 0.08 0.10 Ib / sec. i FUEL F LOW - kg / sec , l F i gur e 75. - Pump c ha r a c t er isti c s (t r ans ie nt ) .

Figure 76 shows the pump NPSP, V / L volume ratio, and percent of vapor at the pump inlet versus fuel flow. Also shown is the effect of GH? leakage into either i or 2 of the average 3 m (i0 ft) long vacuum-jacketed line see- : tions. Since it would not be reasonable to ground the airplane with a single vacuum leak, the assumption must be that a second leak could develop.

either inflight or at a location where repairs were not possible. With this assumption, the performance of the foamed li_e is seen to be nearly identi- cal to the vacuum line with a GH2 leak in two 3m (i0 ft) sections. Since the weight of the foamed line is only about 10-15 percent greater, it would appear that the foam concept is more attractive in safety, manufacturing, reliability, original cost and maintenance.

In the case of an unchilled vapor filled line, the engine acceleration to idle will take a longer time than with a chilled line since the engine will not accelerate until the engine pump receives liquid H 2 (V / L = 0). This time could be reduced by using the intertank transfer system to increase the boost pump flow rate.

5.6.3 Final system configuration. - The final system layout and details are shown in Figure 77. The lines are foam insulated and protected by an outer aluminum cover which would contain any H2 leakage in the inner line. All i components are purged and ventilated. An outer shroud (unpressurized) is provided where the fuel line runs through pressurized compartments The • i motors and actuators of all shutoff and crossfeed valves can be removed i without disturbing the line itself• Pump replacement can be done with LH2 i fuel in the tank. (See Section 5.3.6). 1

i

5.6.4 Engine operational procedures• - The procedures and requirements for I operation of the engine, and its fuel delivery and control system, are as follows: !

• Initial Condition Engine is stopped Electrical system is de-energized Engine fuel flow control valve is closed Tank shut-off valve is closed Boost pump is not running Entire fuel system down stream of the liquid hydrogen tank has reached a soak temperature of 311°K (560°R). The system is full of " hydrogen gas at a temperature of 311°K (560°R), and the pressure has relieved to the 145 kPa (21 psia) tank pressure through a reverse flow check valve in the tank shut-off v a lve.

14 _ ' SEA LEVEL I Tamb = 54°C (130°F) °

1 2 I

10 _/ _ --3 . 8 cm (1.5 in , ) ; FOAM INSULATION VACUUM LINE:.

\

-I %10m 120ft) , GH2 LEAK VACUUM LINE: +300 +40 - 6 . 1 0 m // _ 3 . 05 m (10 ft), GH2 LEAK / . (20 ft), GH2 LEAK +20 +100 +200 . __ - _ -1 0 0 I I ' _ -20 - _ b -2 0 0 - 300 -40 - - 500 / , - STARTING F L OW

/ • -600 -80 -

p -700 S - •-100 -- ' i , , l J ..... _ . I 0 .05 .10 . 15 .20 Ib / sec , I I, I I I I • . 0.02 0.04 0.06 0.08 0.10 F ig ur e 7 6 . - En g in e p u m p inl e t c ond i t i ons .

l ' E -I VIEW C FROM TANK 4 EN G 4 _ . _ ._ J FROM ENG 3 _ . . _ . _ ENG I _JJ FR FROM TANK 2 TANK 1 // / f -- CROSSFEEDVALVES 13) DET. A - CROSSFEEDVALVE SCHEMATI C -- _ - _ __ .:--------- INSU LATION VALVE ACTUATOR _1= " i - _ , :_..,, - RE M OVABLE (3) -[ iJ I ' i VIEW B r ' _,,, I CROSSFEEDVALVEAS S Y ' = .1 _ ; i ' _ . _ ' -INSTALLATION VALVE ACTUATORS _ _. _ .

, " ! , , FWD

rr "4 VIEW C C R OSSFEEDVALVE ASSY.

Fig ur e 77. - Con ti n u ed.

1 6 8 I N SULATION CLOSEOUT _ . VALVE j S EE DET. F FOR - / VALVE • TANK RUBBER SEAL _ _ WALL _ CLAMPS(2) - / FILLER BLOCK (SPLIT) _ AM VAPOR SEAL DET. E - TANK FTG AND VALVE INSTL.

PRESS.SIDE Ii A M BIENT SIDE _ j _ STRUCTURE SEAL _ 1 0 TRAVEL 0 . 8 " 2 . 0 3 cm t 0 . 8 in . )

_ - __ _ . . _ ' _ _ _ .._ _ ...... - _ --, - _ _ 3 o_

I I I _ i / BUSHING " BELLOWS (2 PLY) DET. D - BULKHEAD FTG. - FLEXIBLE ' Figur e 77 . - C ontinu e d.

I I_ , j_ ACTUATOR (270 V dc) GEARBOX EVACUA rED BELLOWS _ _ _ MO OR ASSEMBLY _ " _ I : _ J¢': (TEFLON) "_ ! I J CLOSEDCELL GUIDE BU S HING _ _ ' • / FOAM " _ [ _ / u I )SEAL

7 - - : --

i i F VE SELF-ALIG N I N G POPPET DET. F - TA N K ISOLATIO N VALVE Figure 7 i . - C o ntl n u e d.

. . 17 0 • Start Procedure Energize engine e lectrical syst e m. This automati c ally opens th e high pressure pump drive fluid coupling fill valve. The valve lat c _, e _ open and will remain open until electrically energized to close.

Open tank shut-off valve.

Start boost pump. Since the boost pump inlet is immersed in liquid hydrogen, the boost pump will pressurize the hydrog e n gas in the line up to the engine fuel flow control valve to approximately 2 90 kPa (42 psia).

Start cranking engine. Engine oil pressure develops during cranking, and fills the high pressure pump fluid drive coupling thus driving the pump.

At i0 per c ent speed, turn on ignition, move throttl e to Idle position and the engine fuel flow c ontrol valve starts the admission of fue± t o the eng i ne fuel inje c tor.

Engine cranking continues until the engine is self-sustaining, and then the cranking is terminated.

Engine continues acceleration to Idle speed.

Continue operation at idle speed until chill-down of the engine high pressure pump is completed. The time required for chill-down is approximately one minute. After chill-down of the engine high pressure pump is completed, normal engine operations may be c ommen c e d .

• Ground and Flight Oper ation Sys t em respon d s to command inputs in a m anner similar to that of a c onventional Jet A-fueled t urbofan engine.

• Shut Down Redu c e eng=ne speed to idle.

Move throttle to idle cut-off.

De-energlz e engine electrical system. This automatically energized the hlgh-pressure pump drive fluid coupling fill valve to close. The valve latches close, and will remain closed aft e r the electrical system is de- energized. With the valve closed, the fluid drive coupling drains, dis- connecting the pump drive.

Turn off boost pump.* Close tank shut-off valve.

*Beca us e the en g ine pump int e rsta ge bearing op e ra t es in the hydrogen working fluid and i s dependent upon it for c ooli n g , and in the c a s e of the fo l l bear- ing is dependent upon it for load carrying ability, it is desired to maintain b oos t Dump pr ess ure until af t er th e e ngine pump fluid coupling has di s connec t ed.

This time delay in t urning off the boos t pump can be short, perhaps 10-15 s econds.

• P_ ocedure After Aborted Ground Start Turn off ignition.

Crank en g ine for a n r_ A _=_-_ time _o ven L ilate englne a n d re m ove any hydrogen vapor.

Determine reas o n for aborted start and take c o rrective action if required.

S t art engine using n o rmal start procedure.

• Procedure for In-Flight Shut Down Reduce engine speed to Idle.

Move throttle to idle cut-off.

De-energlze engine electrical system. This automa t ically energizes the hlgh-pressure pump drive fluid coupling fill valve to close.

With the v a lve closed, the fluid coupling drains permitting the fuel pump to stop, although the engine may be windmilllng at m oderately high speed.

Turn off boost pu m p.

Close t ank shut-off valve.

• Procedure for In-Fllght Start Op e n tank shut-off valve.

Start boost p ump.

En e rgize engine el e ctric a l syst e m. This autom a tic a lly e n e rgizes th e high-pressure pump drive fluid coupling fill valv e to op e n.

Sin c e th e e ngin e is windmilling, oil pressure is a vailable and the coupling fills thus driving the pump at windmilling sp e ed.

Ini t iate ignition.

Mov e throttl e to idle position.

When e ngin e is s t art e d a nd op e rating normally at fligh t idl e , move throttle to th e desired power s e tting.

5.7 Technology Developmen t Required • The st u dy of the en g in e fu e l sup p ly syst e m id e nti f i e d and brou g ht into f oc us various ar e as u f risk in the t e chnnlogy where advances in the state of th e ar t a re either nece s sary or high l y d e sir a ble to f aci lit a te t he t i m e l y a nd ec onomi c d e v el op ment of a full s c ale sys t em. Thls se c tion lis t s the " mor e s igni f i c an t o f th e se t ec hnical risk item s , a nd pr e sen t s r ec ommen da- " tions r eg a r din g ap p ro pr iat e advanced de ve lop m en t , r 1 7 3

t

!

5.7.1 Engine fuel pump. - The e ngine high-pressure pump bearing system is a major technical risk item requiring a dvanced development. The c u rrent state of the art in advanced high pressure LH 2 pumps has evolved mainly from the development work which _as been done on rocket engine turbopumps.

As a result of this work, the problems of designing for pump performance (bead, flow range, and suction performance), and also the problems of mechanical design and materials selection for cryogenic service, have been adequately resolved and may be considered state of the art.

However, all rocket engine components inherently have a very short mission duty cycle. This has resulted in very short specified life require- ments, even for reusable equipment such as Space Shuttle, where the main engine design life requirement is I0 houra and i00 missions. On the other hand, air transport equipment such as that being considered in this study, is at the other end of the life requirements spectrum, with airliner utilization running to i0 hours a day operation, equipment overhaul periods of 5000 hours minimum being conm_on, and equipment service life of 40 000 hours being typical.

This vast dl ; ference in life requirements poses a specific problem in that the rocket engine turbopump bearing technology is not transferable since the bearing systems developed for rocket engine turbopumps can only meet the very short life requirements and do not have the inherent potential for development of the long life capability required for air transport service. This limitation of life development potential is well demonstrated by the vast amount of work which has been necessary to achieve even the limited life required for rocket turbopump applications.

For these tea ns, it is considered that the most critical problem in the d e v=ivpme_., of a high-pressure LH 2 pump suitable for airline service is th !,um - _ring system, and it is recommended that the newer approaches described _,luhis report be investigated.

It should be noted that these comments apply only to the bearings of the high-pressure LH 2 pump, which are relatively highly loaded and which operate at high rotational speed. The very lightly loaded bearings of a LH2 fuel boost pump, which run at lower rotational speed, can probably be developed adequately for airline service as a further evolution of the existing design approach, using rolling element bearings and separators having a dry lubricant capability.

It is recommended that engine high-pressure pump bearing system ad- vanced development be undertaken, and that such advanced development start with the preliminary design of an engine high pressure pump in sufficient depth to establish the bearlng requirements. This would then be followed by design, fabrication, and feasibility testing of a bearing system having !

the objective of meeting t_ese requireme,ts. Initial bearing tests would be in a bearing test rig, followed by tests in an actual pump.

5.7.2 Engine fuel control system. - Operation of the cryogenic hydrogen fuel control system presents several new problems such as starting with the supply line full of vapor, the necessity for extremely rapid chill I down of the engine high pressure pump, the probable necessity to control the flow of fuel in both the vapor and liquid states, and the presence of significant volume capacitance in the fuel system combined with the use of the relatively compressible H 2 fuel. These new problems suggest the de- sirability of analysis and computer simulation of the selected engine fuel delivery and control system, to verify performance capability including flow, pressure, and thermal transients. Following analysis and computer simulation, fabrication and test of a breadboard system would be highly desirable.

5.7.3 Overall system. - It is desirable to make a preliminary investigation of sys t ems interactions involved in utilizing H 2 as a heat sink for cabin air conditioning, engine oil cooling, engine stator vane and r o tor blade cooling, in combina t ior with the engine exhaust fuel heating concept. This may be done by co m pu t er simulation, and particular a t tention should be paid to identifying critical off-deslgn conditions.

J

I

L I 6. F UEL S U BS Y S TEMS The aircraft fuel subsystems, illustrated schematically in Figure 78, consist of all fuel-oriented systems up to the interface at the engine fuel control system. These systems cover the functions of storing fuel, fueling / defueling, supplying fuel to the engine and auxiliary power unit (APU), transferring fuel, pressurizing / venting, Jettisoning, and purging and / or in- erting system_. These systems are discussed in the following paragraphs.

6.1 Fuel Tank Arrangement Fuel is stored in thermally insulated tanks located within the fuselage.

There are four separate tank compartments, corresponding to the number of engines, in accordance with the convention that each engine be fed from an independent source during takeoff and landing. They are numbered sequentially beginning at the forward end of the airplane. Tanks i and 2 are located between the flight station and the forward end of the passenger compartment.

Tanks 3 and 4 are located aft of the passenger compartment. Each tank has a nominal usable fuel capacity of 6985 kg (15 400 Ib) of LH 2. Tanks i and 2 in the forward storage area are separated by a bulkhead which isolates the liquid fuel in each of the tanks. However, a vent system which is common to both tanks maintains an equal ullage pressure on both sides of the bulkhead.

Tanks 3 and 4 are separated by a similar bulkhead in the aft fuselage area.

6.2 Fueling and Defueling A fueling system shown in Figure 79 is provided which interfaces with the airport ground supply through two adapters located at the aft end of the fuselage below the vertical tail. Liquid hydrogen is supplied to the fuel- ing adapter [Appendix B, Figure B-2) and displaced hydrogen gas from the aircraft fuel tanks is returned to the airport liquefaction facility for re- cycling by means of the vapor recovery adapter (Figu e 80). A 5-1nch vacuum- insulated fueling manifold conveys fuel to Tanks 2, 3, and 4, reducing to a 3-inch manifold between Tanks i and 2. Fuel is discharged into each tank by means of a perforated fueling manifold located near the bottom of the tank below the normal reserve fuel level. The perforations are sized to maintain a low discharge velocity to minimize turbulence in the bulk liquid.

The fuel level control system consists of a shutoff valve (Figure 79) actuated by a signal from a level sensor which terminates flow to each tank whet, it is full. When a given flight requires less than full tanks, the shutoff v a lves are actuated by a signal ini t i a ted by bugs on the tank fuel quan t ity indicat o rs, loca t ed in the aircr a ft flight s t ation, which have pre- vlously been set a t the desired fuel quantity. The fuel quantity can also be selected at the refueling panel in the t ail (Figure 7 9).

e ¢ |

!

Q) i I i i ,,' _ , iJ ! _ - 0 _i i _ r .... •II h , " . , ¢ ? '_- , , . - . J ! , / | t 78 f , _ U'GI _ T A T. ; PAGE T$ 0_,POOR QUALITy' DET. A - FUELING VALVE INSTL.

I Figure / 9, - Continued.

t j" • / " _ ' _ ,. ACTUATOR •MOTOR "",,t, (270V DC) EVACUATED / / B E LLOWS AS S Y. ._ \ \\ /

/

GUIDE BUSHING (TEFLON) PUMP-OUTPORT VACUUM _ ACKET , . , " I COUP L I N G (1 1 2) 7 -62 cm (3 in.) REF I : L PRESS.RELIEF VALVE SELF ALIG N I N G PO P PET DET. B - FUELING SHUTOFF VALVE q F ig ur e 79. - Continu e d.

1 8 0 ,L DRIGINA L PA G _ r_ D R IX _ OR Qu_zry |,

' t

SET POINT TEST - ON WITH : , | .

POWER - OFF WHEN POI N TER I ! AN D BUG MATCH

!

/ FUEL QUANTITY INDICATOR

\

POIN TER A N D DIGITAL DEFUEL VALVE INDICATOR \ _ / _ ' - - LIGHT ON WHEN VALVE _ / READOUT i

/

NOT IN SELECTED POSITION _ DEFUELIN G VALVE SWITCH (4) / F + __ EFUEL _ . _ ' -- +_ __ (: _ _ - DEFUE _ ' / _J -- _ V A LVE CLOSED , N D,C A TOR TAN KN O 2 UNLOCK I LOCK l UNLOCK _ LOCK | I F TAN K N O 3 " -

t FUEL

.t TANKNO 4 / # J DEFUEL "_ LIG H T S , , REFUEL VALVE SWIT C H (4) I , P SET KNOB - POSITIONS BUG / ADAPTORS - DU S T COVER AND SECONDARY SEAL - REFUEL DOOR CANNOT BE CLOSED UP'LESS COVER S ARE IN PLA ;E SET BUG - INDICATES QUANTITY AS SET BY KNOB . FUEL FLOW STOPS WHE N POINTER MATCHES BUG VIEW C- C - REFUEL PANEL Fig u re 79. - C o eclud e d.

1 8 1 I Vapor released during the fueling operation flows through the absolute tank pressure regulators (Figure 79) into the common vent line where it is diverted to the vapor recovery adapter by means of a vent bypass valve built into the adapter and operated by the actuating linkage of the adapter (See Figure 79). The absolute pressure regulators prevent flashing of the fuel by maintaining the tank pressure above the saturation pressure of the delivered fuel.

The fuel tanks are protected from o v erpressurization in the event that a shut o ff valve fails to cl o se when the tank is full by limiting the ground sys t em delivery press u re (Reference 2) a nd by sizing the vent lines to a llow liquid hydrogen overflow through t he vent system to the vapor recovery adapter.

Defueling is acc o mplished wi t h b ot h fueling a nd vap o r recovery adap t ers connected to t he airp o rt defueling f a cility (Reference 2). The fuel transfer and refueling llne tank isolation valves are then opened and t he f u el level control v a lves are cl o sed. Oper a ti o n o f the tank boost p u mps will star t the def u aling op e ra t ion. To maintain t a nk pressure above outside ambient, some heat may have to be a dded to the st o red hydrogen by means of the fuselage- mou n ted tank pressurization heat exchanger which u tilizes a c a lrod heating element to convert liquid hydrogen to gas. The tanks may be defueled indl- vldually or simultaneously.

6.3 Engine F u el Supply The engine fuel supply sys t e m is shown in Figure 77. E a ch engine is l norm a lly supplied fuel fro m i t s iden t ically numbered fuel ta nk. In t he event of engine failure, fuel fro m the tank which normally supplies t he i, failed engine can be m a d e available to the opera t ing engines by a crossfeed sys t em. However, the c rossfeed syste m is not required for a ircraft cen t er of gravi t y c ontr o l a s will be discussed in 7 .2, Opera t ion a l Re q uir e ments of the Liquid Hydrogen F u e l System. A signifi c a nt featur e is t he lo c at ion and arr a ngemen t o f the crossfeed valves. T | ._v are c ontained in one a sse m bly for c onvenien c e in servi c ing and a ls o t o preclude l o ng sec t i o ns of t ransfer lines which w o uld c o n ta in vapor and c ould res u lt in engine flameo ut when switching from direct to crossfeed.

Lines leading t o t he e_glnes are l o cated in the wing b o x for pr o tec t ion and isola t ion. The lines are foam insulated within a protective met a l outer tube. Evacu a ted double bellows lines with an outer braided cover are used where required for flexibility.

i A s u rge box l oc a t e d at th e l o w point in each fuel t ank houses three boost pumps which supply fuel to e a ch engine. The surge box t raps fuel in the vicinity of the pu m ps to minimize unusable f u el, and to ensure its availabili t y during unusual t ransient maneuvers. The present design util- izes a presurized ac c umulator downstream of the p um p check valves t o p re-

i

clude engine starvation if the fuel migrates t o the top of t he surge box during neg a tive or zero g flight.

I ¢ The reasons for selection of a three-pump system :a ther than the two-pump arrangement used in conventional hydrocarbon-fueled aircraft are discussed in 5.3.1.

6.4 Auxiliary Power Fuel Supply The APU is supplied liquid fuel, normally from Tank No. 2, but available from any tank by crossfeed. During the initial APU startup, before electrical power is available to the tank-mounted boost pumps, it is expected that the normal tank pressure 145 kPa (21 psia) will preclude the need for a separate APU tank-mounted boost pump. It is also possible that an external combustion engine may be a feasible method of driving an APU. This represents a change from a statement in Reference 1 which had indicated that the APU might operate on boiloff hydrogen. More detailed studies showed that this was impractical because of the wide variation in boiloff rates, and also because of the high compressor power required to raise the gas to conventional APU combustor pressure.

6.5 Fuel Transfer I Fuel transfer between tanks csn be accomplished by opening the appropri- ate fuel transfer valves and fuel level control valves while operating the tank boost pumps. A fuel transfer system is incorporated to preclude trap- ping of fuel in any tank should the feed line tank isolation valve fail in the closed position. The effect of this type of failure on center of gravity travel is discussed in 8.3.3, Fuel Management.

6.6 Fuel Jettison Inasmuch as this airplane meets the climb requirements of FAR 25.1001 (b) and (c) with full fuel load, a fuel Jettison system is not legally re- quired. Howev e r, some situations can be postulated in which a Jettison sys- tem might be desirable. For example, if a wheels up landing is anticipated fuel could be Jettisoned to reduce the landing speed. Assuming a full fuel load and climb to 3048 m (I0 000 ft) the time to Jettison fuel down to the reserve level would be approximately 1.4 hours. The Jettison arrangement could be as shown on sheet 1 of Figure 79.

6,7 Tank Vent and Pressurization System The forward pair of tanks and the aft pair of tanks have separate pres- surization and vent systems but share a common overboard vent system down- stream of t,e pressure regulators (see Figure 80). The tanks are maintained at an absoidt, pressure of 145 kPa (21 psia) by a primary absolute pressure regulator located Just downstream of the point where the vent line emerges from each pair of tanks. A secondary pressure regulator set at an absolute pressure of 159 kPa (23 psia) is mounted in parallel w i th the primary regu- lator to protect the tan_ from exca=sive pressure in the event of failur e of 18 3 D t " ORIGINA L PAG_ IS VAPOR SEAL

' / , v., . z ,; qT_ .... ff;_

/ VACUUM JACKETED LINE AS S Y __ / . , _ / "_ <J _\ - " _,- / - _ , i / L . / _ X X . CO U PL I NG - / _ - '- SLEEVE A N D " _ // / \ _ SEE DET .

/ / . / / _ _ __ _ -'- " - " - _ FOAM FILLER

/ , - w =.ox / / \ \

\ . , , ,' , " _ N SU LATI O N J / CLOSE OUT TA N K WAL L -- " 1 DET. B - TA N k PRESSURE CO N TROL VALVE INSTL.

ABSOLUTE REF. BELLOWS F PRESS. REGULATOR A S SY . (145 kPa (21 psia)) A N D SERVO VALVE ASSY. / (145 kPa (21 p s ia)) / j - -"I UAL DIAPHRAGM (CRES.)

- REM O VAB L E " --- _ / • ,_ . J J _ r = ' IDE BUSHING . _ ; _ T . _ . _ . r_ ,[; _ " _'_ / -- _ O V ERRID E SOLENOID (N.O.)

34 . 3cm 13.5 i n.) _,.?_ ' _l / _d!_ - '_ . , ORIF:ICE (PRECHECK) r r , , - ; "

'-"41 - ""

FROM i- 41- /" TRAVEL ' _ ' 7 " SYM _ E I y;. l , , , _ - / _ " _ - - -. _ A C' rU A TOR: P O PP ET VALVE j I . ' ' r " _ D.C. MOTOR I / . t: l - -- - _ . .._ -- _ / _ _ EXIT POR T _ . _ AND RED. GEAR INSULATION NOT i / 23 . 6 c 2 I 8 . 1 cm (3.2 in. ) P URGE VALVE ASSY.

i SHOWN F O R CLARITY-- _j (9.3 in.) -- / ;6 , 9 cm (- '"_ 2.4 in.) -- PRES S . RE L IEF ASSY. ( 1 59 kPa (23 psia)) VIEW C.C - P RESS. CO N TRO L VALVE t Figur e 8 0. - Co ntinu e d.

ALTERNATE VENT BACK PRESSURE VALVE 24.1 kPa (3.5 psig) _ .

FROM TANKS -" i i I r ' - _ J / / .

SOLENOID OVERRIDE - PREVENTS VALVE VENTING DURING REFUEL TO TAIL VENT " _ . _ _

/

/ / VENT ADAPTOR SEE DET. E PANEL / / . / !

/

D ET , D - ALTERNATE V ENT I NSTL .

t F igure 8 0. - Con t t _ 0 1e d .

1 8b I VACUUM JACKET ASSY.

CLAMPS141 12.7 cm (5 in.) OIA x 0.041 cm OUTER TUBE RIGID TUBE CO k ,.LING / -- / (o.o16 in.)WALL / END CLOSURE / / / RUBBERSEALS(2)

- ,J- l -_ /

/

Eitlh .....

I LA YER MYLAR / FIBERGLASS EVACTUATED INSU LA TION FELT INSULATING SLEEVES(2) INNER TUBE - 10.16 cm (4 in.) DIA x 0.041 cm DET. F - TYPICAL VACUUM SYSTEM (0.016 in . ) WALL RIGID JOINT NACA FLUSH VENT SCOOP b _ _ I AIRFLOW

j - -_

PRIMARY VENT SE _ / BACK PRESSURE RVO VALVE ('--- _ VALVE (10.3 kPa (1.5 p$i9)) . , . , ....

(TEFLON COATED) , . , _ -_ - _. - --- _ ..... " " OVL:RRIDE SOLENOID (N.O.) I DET. G -- PRIMARY VENT INSTL. FROM TANKS (PRECHECK) Figur e 8 0 . - Con c lud e d . , , , - / 188 / - / the primary regulator. A purge gas discharge valve for use during the initial tank fill and during tank purging for repair or inspection completes the valve assembly at this location. If the tank absolute pressure drops below 124 kPa (18 psia) when the boost pumps are operating, as might happen, for example, if a takeoff is attempted immediately after the tanks are filled with subcooled hydrogen, a backup absolute pressure regulator allows liquid flow normally sup- plied from Tank No. 4 to be evaporated at the fuselage-mounted tank pressuriza- tion heat exchanger. Tank pressures are, thus, always maintained above the minimum level.

Vent boxes located within Tanks 2 and 3 act as liquid traps to preclude liquid from passing overboard through the common vent line. A drain valve at the bottom of each trap allows the liquid fuel to drop down into the tank below when the fuel level is below the float in the drain valve. Each vent box communicates with the tank it serves through a single vent line with its inlet in the ullage bubble above the point of intersection of the fuel sur- faces for maximum pitch attitude extremes with full fuel tanks. This repre- sents the simplest and most reliable vent design. If detailed aircraft attitude studies reveal that no single inlet location will always be void of liquid fuel, an alternative design is available which incorporates two inlets in each vent line. The inlet which is remote to the vent box would be open at all times and the inlet near the vent box would be closed by means of a float-oper- ated vent valve when under fuel but open when not covered by fuel. The added complexity of this system is to be avoided if possible since it places moving components within the fuel tanks which would ultimately require maintenance.

The common vent line downstream of the absolute pressure regulators serves a dual purpose. In flight, gas relieved through the pressure regula- tors is conveyed through the vent line to a llghtning-protected overboard vent mounted in the vertical stabilizer. The overboard vent assembly in- cludes a servo operated back-pressure valve set at 10.3 kPa differential (].5 psig) to prevent air from being drawn into the vent where it could con- stitute a hazard. During fueling operations, the common vent serves as a means to recover large quantities of boiloff gases by routing them back to the vapor recovery adapter so that they can be recycled by the airport hy- drogen liquefaction and distribution system. In the event of the failure of the primary vent an alternate servo operated vent set at 24.5 kPa (3.5 psig) is located in the tail cone area (Figure 80). This valve is closed by an override solenoid to prevent opening during fueling.

6.8 Nitrogen Inerting System An investigation was made t o determine the characteristics of a GN 2 inerting system which might be required to inert the space surrounding the n o nintegral fuel tank (candidate A), and the e ngine supply system down to and including the engine pump and fuel control. The ground rul e s esta b lished f o r this study were: | i. The purge system is a flight dispatch it e m and must have cual redun- dancy in all functi o nal aspects.

2. The quantity of N2 carried must be sufficient to meet the most severe of the following: (a) one flight of I0 190 km (5500 n.ml.) + alternate destination + ground hold, or (b) two flights of 4050 km (2187 n.mi.) wlth ground hold and diversion to an alternate destination.

3. The minimum purge space pressure must be at least 1 psig above ambient to preclude ingestion of air due to local flight pressures above ambient free stream.

4. Leakage rates through fuselage structure based on improved L-1011 production aircraft experience.

5. N2 purge system must prevent air ingestion during maximum emergency rate of descent.

6. N2 to be stored as a liquid.

The most difficult aspect of the analysis was to predict what leakage might occur in a service aircraft. As a starting point the functional test procedure (FTP) required of all production L-lOll's was reviewed. This requires that after blocking all valves, vents, and drains, the alr leakage s ho ul d n ot exceed 3 9 k g / m l n (8 7 ib / min ) w i th a c a bi n diffe r e nt ial p r essu r e o f t 4 8 to 55 k Pa ( 7 to 8 p s i). C o rrec t in g t h i s r a te f o r le a ka g e a rea , o ne ps i d l f- fe r en ti al and N 2 prop e rt ies an d t em p e r a t u r e s , it was calcula t ed t ha t a pp r o x i - m a t ely 4 990 k g ( ii 0 00 I b) o f N2 w o uld b e re q uired f o r t he i0 1 9 0 k m ( 5500 n . m l.)

f lig ht p ro fi l e . Sin c e th is is not rea s o n a ble, di scuss io ns w ere h e l d w ith the L - 1011 t est pers o nnel a s to h o w t his l e akage mig h t be reduced . The c o nclusi o n was t h at s i nce m uch o f t he L-lO ll l ea k age was due to th e m an y d oor an d win do w seals , fee d - thr us , an d h i dd e n h ole s i n s t ruc t u re, t ha t in t h e LH 2 ai r craf t, w lt h minimal acc e ss doors and ca r ef u l a tte n t i o n to s ealing o f hol e s, t h i s r a t e m i g ht b e reduc e d to I 0 o r 15 perc e n t . Fo r pur pos e s of t his analys is, t he i0 p e rc e n t value was a ssum ed.

A s chema t ic o f t h e sys t e m i s sh ow n i n Fi g u r e 8 1 . Ca b in di sc harg e a lr is u s e d to hea t t he cryo g enic LN2 .

The significan t surface areas a nd v o lumes are : S ur face Ar ea m 2 ( ft 2) V o l . m 3 ( f t 3 ) F orwa rd L H 2 tank c ompartm e nt 216.7 ( 2333 ) 80 . 8 ( 285 4) Aft LH 2 ta n k co mp a rtment 2 20 .6 ( 23 7 5 ) 7 1 . 8 ( 253 6) Engine supply purge Jacke t ( including engine p ump and f uel c o n t r o l ) 4 1 .1 ( 44 2 ) 2 . 8 ( 99) a To t a l 4 78. 4 (5 1 50) 1 55. 4 (5 4 891 Compari s on o f th e mission pro f il e s sh owed that th e two 4 0 50 km ( 2 187 n.mi.)

trip s co ns t itut ed the mo r e s ev e r e r e quir e m e nt wi t h r e g ar d to qua n tity o f N 2 b e cause of the more fr e qu e nt c limb s and d esce nt s . N 2 venting i s requir e d on c limb with s ubsequent repleni s hm e nt being nec ess ary on le sce nt. On t h i s basi s th e f ollowi n g weights wer e estimat e d for th e s y s tem: L__N 2 Required: C om p artm e nt c ha r gin g 838. 7 kg (1849 i b ) L e akage 467.2 (10301 Residual N2 3 2 . 2 ( 7 1) Total 1338.1 k g (29 50 Ib ) S yst e m W eights: LN 2 Dewars (60 psig) 11 7 .0 kg ( 2 58 ib) Equipment 45.4 ( 1001 Plumbing and shrouds 179.6 (396) 342.0 kg ( 754 l b) ) + 10 p e r ce nt co nt inge n c y 3 4 . 5 76) Total S yst e m 3 7 6.5 kg ( 8 3 0 Ib ) + LN 2 133 8 .1 (2 950 ) T otal i n s t a ll ed s y s t e m + gas 1 7 14 .6 kg (3 7 80 l b) The e ff ec t o n DOC (assuming t he LN ? cos t s n ot hin g since i t is required f o r H p liquefac t ion) i s e q u i valen t to an increase o f 1 . 53 p ercen t in t he base ll ne va l ue o f 0.99¢ / s. k m ( 1 .833 4 C / se at n . m l.). This is c l e a r ly u nde- slrable D n ot o nly f o r t he direc t ec o n o mic penal t y , bu t also fr o m t he p o in t o f view o f l og i st ics and servlclng .

An a l t ern at e co ncep t w a s a ls o ex am ine d in which t h e co m p aztm e nt s a r e held a t a c o n st an t a b s o lu t e p re s sure . Thi s saves t he q uan tit y o ¢ N 2 required f o r char g ing and rechar g in g b u t increa s es t he leak a ge so t ha t t h = fi rs t qua n ti ty o f LN 2 r e q u ired is 1252 k g (2 76 0 ib ), al most as much a s b ef or e.

Further st ruct u r a l p en a l ti e s w o uld re s ult fr o m d e si gn i ng the t ank c ompa r tm en t s t o w it hstand t h is p re ss ure .

If t he i n t e r nal p re s sure d i fferen ti al c o uld b e re d uce d t o 3 .4 5 kP a ( 1 / 2 psig ) t h e total system w ei g ht w o u ld be 1 55 6 k g (3 4 3 0 lb ) , a 9 perce n t re d uc tio n . The 3 .4 5 kP a (1 / 2 psi ) mig h t b e m ar gi n a l, h o wever , i n p reven ti n g al r i n g e stio n under cer t a i n fl ig h t co n ditio n s.

1 92 The approach selected for the final candidate fuel containment systems A, B, C, and D consists of air purging and compartment ventilation where applicable, together with the leak detection at purge exits. This approach is t further described in 8.6.1. N2 inerting is used, however, in the flexible foam outer insulation layers of candidates no. 3 and no. 4 as described in 8.6.2.

The need for active inerting and choice of a final concept is dependent on service experience with an actual LH 2 fuel system in a developmental aircraft program.

• 6.9 Technologv Developments Required To establish the most promising fuel system design will require component and system design evaluations followed by detailed laboratory developmental test- ing. Areas in which this effort should be concentrated include the following: 6.9.1 Negative "g" operation. - The availability of fuel to the tank boost pumps must be assured at al-lqtimes to prevent engine starvation. The pres- ent design proposes a pressurized accumulator downstream of the boost pumps.

Other methods, such as double-ended boost pumps, should be investigated in- asmuch as they may be lighter in weight and more reliable than the proposed accumulators.

J 6.9.2 Engine starting without boost pumps operating. - If the airplane is to be self-supporting, the engines and / or the APU must be capable of being started without the aircraft fuel tank boost pumps operating. Hence, the minimum inlet fuel pressure for starting the engines and / or the APU should be determined and compared to available pressure at the engine / APU inlet as a result of fuel tank pressurization.

6.9.3 Float-operated valve development. - Because of the low density of liquid hydrogen and permeability of most materials when subjected to hydro- gen, the design of a float presents problems in sizing and material selec- tion. The feasibility of floats to operate shutoff valves or switches should be investigated at an early date since they offer the simplest and most reliable method of sensing liquid levels.

6.9.4 Fuel quantity gauging. - A neutron radiation fuel gauge should be in- vestigated for fuel quantity gauging. Neutrons will pass through the walls f of the fuel tank quite easily and yet are attenuated proportionately to the density of the hydrogen they pass through. This would allow the gauging components to be placed external to the tank.

As an alternate, capacitance gauging is feasible and has been used in LH_ but would require the insertion and support of long probes at multiple lo_atlons in each tank.

19 3 6 . 9.5 t _U concepts - It is al s o d e sirable to mak e a pr e limina r y inv e stigation of of APU c on c epts for the H 2 f u e l e d air c raft, i nc ludi n g inv e stigation of the utilizstion of H 2 boiloff as th e fuel. T hi s inv e sti g ation would c ente r a r ound study of £he f e a s ibility of u s in g th e ext er nal c ombustion conc e pt to facilitat e th e utili z ation of H 2 boiloff.

l !

: i

I