sterling engine test

7/27/2019 sterling engine test

1/42

DOE/-78/lNASA TM-789f9

INITIAL TEST RESULTSWITH A SINGLE-CYLINDERRHOMBIC-DRIVE STIRLING ENGINE

J. E. Cairelli, L G. Thkme, and R. J. WalterNational Aeronautics and Space AdministrationLew15 Research Center

Prepared for tireU.S. DEPARTMENT OF ENERGYOffice of Conservation and Solar ApplicationsDivision of Transportation Energy C~nservation


2/42

T h i s r e p o r t was pr e pa r e d to d o c w n t work sponsor& byt h e Gniterf S t a t e s Government. N e i th e r t h e U n i te d S t a t e snor i t s agent , t t re Uni ted S t a t es Depar tment O f Energy.no r any Fe & r a l e m pl oye e s, no r a ny o f t h e i r c on t r a c t o r s .s u b c o n t r a ct o r s o r t h e i r e m pl oy ee s, makes any warranty.e xp r e s s o r i m p l ie d , or asscnaes a ny l e g a l l i a b i l i t y orr-esponsibi l it!, f o r t h e a c cu r a c y, c om ple t e ne s s , or useful-:,ess of an y i2format ior . . appara tus , produc t o r prec es sd i s c l ? s e d , or r e p r e s en t s t h a t i t s u s e vou!.d n o t i n f r i n g epr iva te ly mned r iq!x t s .


3/42

DOUNASA11040-7811NASA TM-78919

INITIAL TEST RESULTSWITH A SINGLE-CY LINDERRHOMBIC-DRIVE STIRLING ENGINE

J. E Cairelli, L G. Thieme, and R. J. WalterNational Aeronautics and Space Administ rationLewis Research CenterCleveland, Ohio 44135

Prepared forU. S. Department of EnergyOifice of Conservation and Solar ApplicationsDivision of Transportat on Energy ConservationWashington, D. C. 20545Under Interagency Agreement EC-77-A-31-1040


4/42

SUMMARYA s p a r t of a St i r l ing-engine technology s tu dy f o r the U.S. Dep ar t -

men t of Ene rgy , a 6 -kW (8-hp) , s ing le - cy l inde r , r hom bic -d r iveS t i r l i ng eng ine h a s b e e n r e s t o r e d t o o p e r a ti n g c o l d i ti o n , and p r e l i m i -rgm n a r y characterization t e s t s r un wi th hydrogen and he lium as t he work-(0m ing gases. The S t i r l i ng eng ine , pa r t of an eng ine -gene ra to r s e t des-Iw ignatcd as GPU 3 (ground power uni t ), w as bui l t by G e n e r a l M o t o rs

Re sea r ch Lab ora to r i e s (GRML) in 1955 fo r t he U. S. Arm y.I n i ti a l t e s t s at t h e L e w i s R e s e a r c h C e n t e r s ho w the eng ine brake

spec i f i c f ue l consumption (BSFC) with hydrogen working gas to bewithin the ran ge of BSFC o b s e r v e d b y t h e A r m y at Fort Belvo i r ,Virginla , ir, 1966 . The m ln lm um sy s t e m spec i f i c f ue l consum pt ion(SFC) obse rved dur ln g the Lewis t e s t s w i th hydrogen was 669 g / k W * h r(1.1 lb jhp - h r ) , com pared with 620 g /kW- h r (1.02 lb/hp- hr) f o r t h eArm y te s t s . Howeve r. t he eng ine ou tpu t power fo r a given m ea n com -p r e ss io n - sp a c e p r e s s u r e w a s l o w e r than f o r th e A r m y t e s ts . T h e o b -se rv ed outpu t power a t a w o r k i n g - a p a c e p r e s s u r e of 5 MPa (725 psig)wa s 3 .27 k W (4 .3 9 h p ) f o r t h e L e w i s t e s t s a n d 3 - 8 0 k W 5.09 hp) f o rthe Arm y t e s t s . A s expec ted , t he eng ine power wi th he l ium w as sub-s tant ia l ly lower than wi th hydrogen.

INTRODUCTION ,A St i r l i ng-eng ine-dr i ven ground-power uni t (GPU) wa s obta l ledf r o m t he U. S. A rm y M o b ~ l ~ t yq u ip m e nt R e s e a r c h a n d D e v el o p m en tC e n t e r an d r e s t o r e d t o o p er a t in g c on d it io n . T h e e n t i r e G PU w a s t e s t e dto ob tain S t l r l i ng eng ine p e r fo rm ance and ope ra t iona l in form a t ion . Ina dd it io n . t he S t l r li n g en g ln e d i m e n s ~ o n s n d p h y s i ca l c h a r a c t e r i s t i c sw er e m easu red and de fined.T h i s w o r k w a s d o ne in supp or t of the U.S. De pa rtm en t of Ene rgy'sS t i r l i ng Englne Highway Vehlc l e Sy s t em s Pro gra m . The Lewis Re-s e a r c h C e n t e r , t h ro u g h an in te ragency agr eem ent wi th DOE'S Off iceof Conservation IS respons:ble fo r pro jec t manag ement of th is e f for t .Th e Intent of the p ro gr am IS to deve lop the technology needed to pro -


5/42

vide the U.S. autom obile ind m tr y with the opt ion of moving into pro-duc t ion eng i nee r i ng f o r au t omot i ve Stirling engines in the mid-1980's.

T h e Stirling engine d r i v e n GPU used in this work was designeda d uilt by G e n e r a l M o t o r s R e s e a r c h L o b o r a t o r y (GMRL) in 1965 forthe U. S. Army. Th e GP U wa s t e s t ed by the Army in 1966 and thenretired. It was obta ined by Lewis , restored to operating condi t ion, andinstalled in a test facil i ty. The tests reported h e r e in w e r e m a d e withhel ium and with hydrogen as working f lu ids .

F ue l f low, a l t e rn a t o r power ou tpu t, mean compre s s i on - spacep r e s s u r e , h e a t e r - u b e gas t e m p e r a t u r e , m e a n c o m p r e s s i o n - s p a c e gast em pera t u re , coo li ng-wate r i n l e t t em pera t u re , and s e v e r a l o t h er pres-s u r e s and t em p e ra tu re s w e r e m e as ur ed , E w e u tput pow er andspec i f i c fue l consumpt ion w ere ca l cu l a ted and comp ared wi t h t he un-publ ished Arm y data. T h e e ng in e's p h ys lc a l c h a r a c t e r i s t i c ~ ndd i m e n s i o n s , i nc lu d in g in t e r n a l v o lu m e s , w e r e d e t e r m i n e d f o r use incomputer s imulat ion of the engine.

APPARATUS AND PROCEDUREGPU 3 St i r l ing Ground Po we r Unit

T h e G PU 3 ele ct r ic a l power g ene rat ing unit ( fig . 1) w as ob t a i nedf r o m the Army. A second . identical un it was bo r rowed f r om theSmithson ian Inst i tution; it i AS been used l a rge l y as a s o u r c e of s p a r ep a r t s f o r t3e f i r s t unit . Both GPU 3 un l t s a r e s e l f -con t ai ned , 3-kW,eng i ne -genera t o r sets bui l t by G e n e r a l M o to r s R e s e a r c h L a b o r a t o r i e sf o r the Army. The GP U deve lopmen t p rog ram is d i s c u s s e d i n r e f e r -e n c e 1, whi ch a l so i nc l udes sys t em per fo rm ance sp ec i f i ca t i ons andso m e t es t r e su l t s . The GP U componen t deve lopmen t and test r e s u l t sare p r e s e n t e d in r e f e r e n c e 2.

T h e G PU 3 unlts are capab le of op era t ing on a var i e t y of f u e l s a n do v e r a broad rang e of amb ient condi t ions enc om pass ed by the fo llowingl imi t s :


6/42

Mininlum temperature fro111 sea level. . . . . . . . . . . . . .o2430m!8000ft), OC('F). 4(40)Maximunl temperature, OC PF),at -. . . . . . . . . . . . . . . . . . . . . . . . .ea level 46 (115). . . . . . . . . . . . . . . . . . . . .438 n~(8000 f t ) 32 (90)

Automatic controls, which were an integral part of the engine sys-tem, regulated f u e l flow to maintain a 677' C (1250~ )hydrogen gastemperature within the heater tubes, and maintained a 3000-rpm en-gine speed.

G PU 3 EnglneThe heart of the G P U 3 IS the S ti rl ~n g ngine, which is capable ofproducing about 6 k W (8 h$) at a mean con~pression-space ressure

of 6. 9 MPa (1000 psig: with hydrogen as the working gas. Figure 2 isa cro ss section of the G P U 3 engine showing the nlajor internal com-ponents. The displacer , rhombic -drive engine was designed to operatewith hydrogen as the working gas. The working gas occupies two spacesinside the engine: (1) the worklng space above the power pisto.1 wherethe thermodynamic cycle occurs and (2 ) the buffer space below the powerpiston, which is used to reduce the pressure forces on the mechanicallinkages and to minimize piston rlng seaiing requirements.

The gas in the working space IS distributed in three smaller, con-nected volumes; ( I ) A conlpression space at the cold end of the enginebounded by the cylinder wall. the top of the power piston, and the bottomof the displacer ylston; 12) an expansion space at the hot end of the en-gine bounded by the top of the displacer and the end of the cylinder; and(3) a dead space, which includes mainly the internal volume of the heat-er tubes, the regenerators, and coolers with the connecting passagesthat tie the conlpression space to the expansion space.

Strictly sptlaking, the clearance volumes w i t h the compressionand expansior, spaces a re also considered dead space. Both the com-


7/42

pression and expansion volumes a re continually varying as functionsof the engine crank rotation: As the crank rotates , the variation inthese volumes causes the hydrogen working gas to be alternately heatedand cooled by the engine's heat exchangers while passing back and forthfrom the expansion space and the compression space. The total volumeoccupied by the working gas is also varying as a function of crank rota-tion. The net work realized for one cycle is the difference between thework produced by expanding hot gas and the work required to compresscold gas.

The effect of dead volume is to diminish the pressure variationsduring the cycle. Although dead volume is necessary to meet the cycleheat-transfer requirements, it tends to reduce the work produced duringone cycle.

Each piston is connected to a shaft, which in turn is connected totwo rods, with each rod attached to one of the engine's two crank shafts.The displacer shaft passes through the center of the power piston andpiston shaft and attaches to the bottom pai r of connecting rods. Thepower piston shaft attaches to the upper connecting rods. Detailed de-scription and analysis of the rhombic drive can be found in Meijer'sthesis (ref. 3).

The following is a brief description of the major engine componentsand systems. Table I contains detai ls of the engine's internal dimen-sions.

Cylinder assembly. - The cylinder assembly (fig. 3) contains thecylinder in which both pistons ride, heater tubes, housings for eightcooler-regenerator cartridges, and water-cooling passages that supplywater to the coolers and cool the lower cylinder wall. The covershields and insulation have been removed in figure 3(a) to show thecooler-regenerator housing cylinders and the heater tube connections.Figure 3(b) is an inverted view of the part ial ly asserr-bled cylinderwitn some of the cooler-regenerator car tridges. The fuel flow controland governor-actuated hydrogen-control valves are also shown. Thermo-couples to measure the gas temperature inside the heater tubes and inthe passage between the coolers and the compression space are alsovisible on the left side of the assembly.


8/42

Heater tubes. - The eighty Multimet N-155 tubes used a re joinedat the top of the cylinder assembly to a common header ring; at theirlower ends they are alternately connected to either the hot end of thepiston cylinder or to the hot end of regenerator-cooler cylinders. Thepiston cylinder and regenerator-cooler cylinders are made of AISI 310stainless steel.

Seventy-six of the heater tubes are of the same diameter. The re -maining four, located at 90' intervals around the cylinder assembly,are of slightly larger diameter and are offset toward the center of theassembly. Each of these larger tubes contains a temperature sensorinstalled from the bottom of the cylinder zssembly. The sensors areused to measure the working-gas temperature inside the heater tubesand to provide a mechanical signal for the temperature controls. Twotypes of sensors ar e installed: Thermocouples a re used to in two ofthe larger tubes and special bimetallic sensors in the other two.

Figure 4 is a cross section of the bimetallic temperature sensorand fuel-control valve installation. The temperature sensor consistsof a stainless-s teel tube with a closed end and an inner tungsten rod.Variations in the heater-tube gas temperature causes a change in thelength of the outer tube. The length of the inner rod is relatively in-sensitive to temperature. The relative motion that results betweenthe bottom end of the tube and the bottom end of the rod actuates thefuel control valve.

Cooler-regenerator. - The cooler and regenerator (fig. 5) arefabricated and joined to form a single cartridge. The coolers areminiature shell and tube heat exchangers. Each cooler contains 39,1.52 -mm (0.060-in. ) o. d. by 1.02 -mm (0.040-in. ) i. d. tubes throughwhich the working gas flows. The cooling water makes a single cross -flow pass over the tubes. The cylinder assembly water passages a rearranged to form two parallel circuits with four coolers in series.

The regenerators consist of 308 layers of square-weave, 200-mesh,40.6- pm (0.0016-in. ) wire diameter, stainless-steel screens . Thelayers of screen a re carefully alined with alternate layers rotated about4 to provide the maximum heat trans fer and minimum flow loss. Thescreens are retained in a thin-walled, electro-deposited, nickel canister,


9/42

which is fastened to the cooler by roll forming the edge of the canisterinto a groove in the cooler. A split ring is placed in the groove to pre-vent the edge from unrolling when the cartridge is removed from thecylinder assembly. The split ring is not shown in this picture.

The cooler-regenerator cart ridges a re installed through eightopenings around the bottom of the cylinder assembly. Each cartridgeis backed by a re tainer assembly consisting of (1) a spacer, whichprovides a connecting passage from the cooler to the compressionspace, (2) a snap ring to hold the spacer in the cylinder, and (3) acover plate, which keeps the snap ring in proper position and preventsit from deforming when the engine is pressurized.

O-rings are used to seal the working gas from the cooling waterpasages. The cross section of the installation is shown in figure 2,and the arrangement of the components in the cylinder assembly areshown in figure 3(b).

A i r preheater and combustor assembly. - The air preheater andcombustion chamber (fig. 6(a)) are combined in one assembly whichmounts on top of the cylinder assembly.

The counterflow preheater is made of vertical rows of horizontaltubes separated by vertical baffles. The tubes and baffles are arrangedin spiral fashion radiating from the inside diameter of the preheater toit s outside diameter. The entire assembly is made of stainless steel.Figure 6(b), a cross section through the preheater-combustor as-sembly, indicates the airflow path through the preheater and combustor.A i r enters the bottom of the preheater on the outside diameter. A plenumchamber distributes the air around the base of the preheater. The ai r isthen distributed vertically through 16 passages which serve a s headers.Then it flows between the baffles picking up heat from the tubes as itflows inward along a spiral path. Another set of 16 passages a t the in-side diameter collect the heated air and channel it to a plenum at the topof the preheater . This plenum feeds the heated ai r into the combustionchamber. The hot exhaust enters the tubes at the inner diameter of thepreheater and flows outward on a spiral path to an annular space at theouter diameter of the preheater and then upward through the end of the


10/42

annulus to the atmosphere. This preheater configuration i s one ofse ve ra l Investigated by GMRL. These a r e discussed in some detai lby Percival (ref, 2).

The combustion chamber is mounted inside the preheater . Fuelis atomizeci by the fuel ??ozzle hrough use of a sepa rat e atomizing ai rcircui t fed by two smal l, vane type ai r pumps. The atomized fuel ismixed with the preheated conlbustion air and is burned in the combus-tion chamber, A specla1 spa rk plug i s provided to initiate combustion.Once conlbustlon 1s st ar ted, the spark plug is deenergized. Burner de-sign and development are discussed In reference 2 . GPU 3 exhaustenlissions with and without exhaus t-gas recxrculation were reportedby researchers at Wayne State University (refs. 4 and 5) .

Pistons, shafts. and seals. - Figure 7 shows the displacer and- --- -A - - I .power pistions assembled w~ t hheir respective shaft s, which a r e con-nected to the rhombic -drive nlcchanism through the power-piston anddisplacer-pis ton yokes. Both shaft se al assemblies ar e also shown.The displacer shaft passes through the hollow power-piston shaft.The displacer shaft i s tinplated at i ts lower end and i s connected toits yoke with an interference fit to insure that the shaft is perpendicularto the center line of the dlsplacer yoke pins. Thi s is necessary to in-su re shaft se al alinement. The power piston is attached to its yokewith a hollow pin. Thi s allows the driving force to be divided equallybetween the t-vo crank s h a f t s , whlle allowing the shaft to be guided bya bushing j u s t below the power-piston seal,

Shaft seals, - The GPU 3 engine uses sliding seals on both the------ -isplacer and power-plston shafts., Figure 7 shows the cross sectionthrough the seals, and figure 8 is a photograph of the disp lacer shaftand its sea l parts.

The s h a f t se al s for the GPU 3 were made by GMRL using theirown Buna-N (nitride) formulation. The sea l has a T-shaped cr os ssection and uses phenolic an t~ ro ll in g hields (backup rings) on eitherside. The design of the seal is very similar to the Palmetto GT sealmanufactlrrcd by Greene Tweed & Company.

A Disogrin rod wiper is installed b e l w each sea l to help preventcrankcase oil f r o m passlng through the rod seals. To insure rod


11/42

a l inement wi th each sea l , a G lac ie r "DW' guide bushing is locatedbe tween the se a l and the wiper . In the lower se a l assem bly , a spr in gloaded Teflon cap seal is mounted above the T-s ea l to reduce the pre s-su re d i f fe re n ti a l a c ro s s the pow e r -p is ton se a l . The c a p se a l is n~ oun te dso that i t a c t s as a check v alve, a l lowing flow out of the s pa ce betweense a l s t he re by e xpos ing the T -se a l t o a r e l a t ive ly c ons ta n t p r e s su re a p -proximate ly equa l to the minimunl buf ie r -space pr ess ur e .

Later versions of the GPU 3 engine made use of rol lsock sealss im i la r to those used by N. V. Phil ips. The seal work done a t G M R Lis descr ibed in re fe rence 2.

Pi s ton se a l r i ngs. - The pi ston s e a l r i ng s a r e show n in c r os s s e c -t ion in f igu re 7. F i g u r e 9 is a photo gra ph of the GPU 3 pis tons and ase t of pis ton rings. The se al s for the dis pla cer piston and power pistonto the cylinder is accom plished using squ are- cut , Rulon LD piston rings.T h e r i n g s a r e m a d e up 9f an inner and an oute r r ing . The inne r r ing ,w hich ha s s qua re e nds hav ing a me ta l -c o i l- c ompre s s ion sp r ing s e t i n tothe end gap, expands to for ce the outer r ing again st the cyl inde r wall.The oute r r ing has a s tepped end gap. The two r ings are kept in a l ine-ment by a sm al l meta l p in s o tha t the end gaps a r e 180' apar t . Twose ts of p i s ton r ing s a r e used on each p iston. The r ing s are radial lygrooved on one face. Each s e t of r in gs is instxlled with the groovesfac ing the o ther se t . The grooves caus e the r in gs to ac t as check valvesthereby tending to bui ld up pressure in the space between the r ings.Th i s p r e s su re e x e r t s a r a d i a l fo rc e on the r fngs and improve s t he i rsea l ing capac i ty . On the d isp lacer p i s ton the upper r in gs a r e smooth(not grooved) to preven t hot ga s flow p ast them.

Engine o i l sys tem . - The engine o i l sys tem suppl ies o i l for lubr i -ca t ion a s wel l a s the ene rgy fo r opera t ion of the engine cont ro ls . Aposi t ive displacem ent pump geare d dire ct ly to one of the output shaft ssuppl ies o i l a t about 0.4 MPa (60 psig). A s t r a igh t SAE 10W oil, with-out addi t ives, i s used to minin lize de te r iora t io n of the e las tom ericse a l s . Th i s sys t e m is somewhat complex and since i t is not pert inentto our d i scuss ion of the basic S t i r l ing engine, i t wil l not be given anyfu r th e r d i sc uss ion .


12/42

h x n e-------ontrols, - Control of the GPU 3 is accomplished throughunique hydron~echanical evices. Two independent control systemsare used: one to maintain nominal heater temperature a t 950' C(1250' F) and one to maintain engine speed at 3000 rpm. Both controlshave manual adjustments to change the set point within a narrow range.Safety devices ar e a lso provided to overr ide the normal controls in theevent of heater overtemperature o r engine overspeed..

Heater-tube temperature control. - Figure 10 is a greatly simpli---ied schematic diagram of the heater - tube temperature control.Bimetallic temperature sensors are mounted within two of the special,larger-diameter heater tubes. (See section----eater tubes. ) The posi-tion of the rod protruding from the bottom of each sensor is proportionalto the heater-tube gas temperature. One of the bimetallic sens or s mech-anically positions the gas temperature-controlled fuel valve, therebycontrolling the fuel flow. As the temperature ri ses toward the desiredheater-tube gas temperature (950' C , 1250' F), the fuel flow rate drops;a s the temperature drops fro111 he desired temperature, the fuel flowra te ris es - The second bimetallic sensor actuates the fuel cutoff valve:If the heater-tube gas temperature exceeds 990' C (1320' F) this valvecloses the control oil supply line causing the shutoff valve to close andthe engine to make a normal stop.

Speed contnoi. - Figure 11 i s a simplified diagram of the speed-control system. Hydrogen gas, sufficient to supply the needs of theGPU for extended operation, is stored in a high-pressure tank. Thetank is initially charged to about 13.8 MPa (2000 psigj. A supply pres-sure regulator within the hydrogen-cont ro l mainfold limits maximumhydrogen supply pressure to the engine to 6.9 MPa (1000 psig).

A flyball governor is used to sense engine speed. The governorregulates the control- oil pressur e, which actuates the governor-actuated hydrogen valve. If the engine speed is below 3000 rpm, thegovernor causes the valve to increase the pressure in both the com-pression and buffer spaces by allowing flow from the high-pressurestorage tank to the engine. If the speed is too high, the governorcauses the valve to vent. both engine spaces to the hydrogen compressor,


13/42

which pumps the hydrogen back into the h igh-pressu re s to rag e tank .S ince the hydrogen co m pre sso r capac i ty is not adequate to reduce en -g ine power q ~ ic k ly , he governor -ac tua ted hydrogen va lve provides abypass be tween the work ing space and buf fe r spac e to a s s i s t govern ingdur ing sudden load dec reas es .

An overspeed device is buil t into the speed governor . I t prev entsopera t ion a t sp eeds above 3450 rpm. If th i s speed is exceeded, thecont ro l -o il p res su re to the hydrogen overspeed bypass va lve is cut offTh is opens the va lve , r educ ing the engine power , and prev ents fu r th e rinc reas e in speed . In addit ion , con t ro l-o i l p re ss ur e is dumped, thuscausing the fu el f low to be cut off and the engine to com e to a n o r m a ls top .

Auxi l ia ry sys tems . - Severa l aux i l ia ry components are groupedand mounted in one as se m bl y (fig. 12(a)). These include the combus-tion air blower , two vane type fuel-a tomizer air pumps, the magneto,and fue l pump. They sh a r e a d r ive sys tem which is powered by e i therof ttvo methods: (1)a pulley on the fro nt of the eng ine through an o v e r -running c lu tch o r (2) the hand s ta r t in g c rank . To insure adequa te air-f low, bo th a to miz e r pumps a r e dr iven dur ing s ta r tup . But on ly oneop era tes when dri ve n by the engine. The cooling-water pu mp (fig. 12(b))and rad ia to r fan ( f ig . 12(a)) are dr iven d i rec t ly f rom a pu l ley on thefro nt oi the engine . Th ese rota te only when the engine output shaf t isrota t ing. A tachometer to read engine rpm and an engine tota l- run-t ime m ete r w ere a l s o inc luded wi th the or ig ina l un it . F igure 12(b) showsthe gages and m ete r s ins ta l led on the G P U 3 as it was rece ived a t Lewis .

E ng ine s t a r tup sy s t e m. - Although ea r l ie r vers ion s of the G P Uuse d a n e l e c t r i c sl . r t e r , t he GPU 3 is sta r te d by hand. Th is changewas made to r educe i t s we ight. A hand c rank loca ted a t the a l te rn a torend of the unit (fig. 12(a)) is used ini t ia l ly to rota te the engine to placethe power pis ton a t top dead cen ter (TDC) thus posi t ioning the enginea t the beginning of the expan sion str ok e. Once the power piston isprop er ly posi t ioned, the crank is used to rota te the combust ion airblower , fu e l a tomiz ing air pumps , fue l pump, and spa rk magneto .This p rov ides the means to warm up the hea te r tubes . Selection of the


14/42

function of the hand cran k is made by m anual engagement af e i t h e r oft ~ oand clutches (not visible in the figures). The power-piston posi-tioning mechanism disengages actomatically when the power pistan isa t TDC, The gearbox clutch is disengaged automatically when the en-gine tabs up the torque load f o r the auxiliaries drive.

When he gas tempera ture in the hea ter tubes is at the normalopera t ing tem perature of 950 K (1250' F), the stutlg pull cable (fig.12(b)) is pulled. The plll cord, whick is attac.\ed to a recoil mech-a n is m s im i i a r to those used on lawn mowers , ro ta tes the c rank sha f t sand moves the pistons. A s the engine speed builds, the combustionsys tem dr iv e is taken o v e r by the engine through the overrunning clutch,the gearbox clutch disengages, and the hand cranking is discontinued.

InstrumentationTh e GPU 3 Grourd Pow er Unit cam e to Lewis fitted with B ourdon

tube pres su re gages, e lec t r ica l me te rs , a d hermocouples. Table I1l i s t s the pa ram ete rs measured , the type of ins t rument used and therange of reading, The pr ess ure s we re read on se pa ra te gages. Thethermocouples we re read on two m e te r s , one for heater-tube gas tem-pera tu re and one for the remaining temperatures . The thermocouplet o be read on the second m eter w as determined by positioning a se lec torswitch.

Figure 13 shows the GPU 3 as i t was te s ted at Lewis. Table IIIl i s t s the instrumentation added to the GPU 3, its type and range.S e v er a l p r e s s u r e g a e s w e r e r eloc a te d to the gage panel on the opera torside of the unit. (All inst ~ttm enta tion sed durin g these te st s read inU. S. Customarv units.)

Te st SetupThe tes t se tup for the ', testing of the GPU 3 ls shown in fig-

u re 14 . Although our inter ;rim arily with the engine, the completeGPU 3 unit, with only minor modification, was tested to minimize the


15/42

facility support systems. Some modification was required to circum-vent operational problems and facilitate testing. The following changeswere made to the GPU 3 for the initial tests.

(1)Two externally mounted fuel tanks with nitrogen pressurizationwere substituted for the original self -contained tank and fuel pump.There tanks were cgnnected to the engine through a selector valve whichallowed use of the smal ler tank for startup and between data points, anduse af the larger during the gathering of data. This was done to providea means for determining the fuel consumption. The fuel tank used duringthe data runs is shown in figure 14 on the lower level of the test stand onthe right hand side.

In addition, chronic malfunction of the temperature control made i tnecessary to install a manually adjusted needle valve in the nozzle f u e lline. The overtemperature fuel cutoff system appeared to operate andwas left intact. The fuel-control needle valve is shown in figure 13 onthe upper gage panel.

(2) The high-pressure-hydrogen tank (see fig. 13)was not used.An external, gas -supply panel was made up to control the supply ofeither hydrogen or helium directly to the hydrogen-supply regulatoron the hydrogen-control manifold. The gas supply panel is shown onthe right of the test stand in figure 14. The hydrogen compressor wasdisconnected, and the hydrogen vent, which would normally be connectedto the compressor, was tied directly to an atmospheric vent.

(3) A leak developed in the original oil cooler, which was locatedinside the cooling-water radiator. To avoid complicated repair, anexternal oil cooler was installed in f roct cf the water radiator. Theconnections to the original oil cooler were capped to prevent loss ofcoolant.

(4) The alternator output power was abscrbed by the separate re-sistance load bank shown at the left of figure 14. Engine load waschanged by varying the resistance of the lc . ink. A voltage regulatormounted on the GPU maintained the voltr 30 volts. The regulatorwas not altered for these tests.


16/42

!S j A hood was installed over the generating unit to prevent hydrogenaccumulation as well as to remove the combustion products from the testarea.

Figure 15 is a simplified schematic diagram of the GPU 3 test setup.It shows the major engine components and indicates the instrumentationused during the tests.

Test ProcedureThe GPU 3 was installed i n the test stand (fig. 14), and the various

systems and instrumentation were connected (fig. 15). The hydrogeno r helium supply was set. The engine was then purged of a ir by al te r-nate pressure-vent cycles of the working and buffer spaces, fi rs t withhelitun and then with hydrogen, if hydrogen was to be the working fluid.The fuel tanks were filled with No. 1diesel fuel. The run tank wasweighed, and then both tanks were pressurized to about 1.86 MI%(27 psig) with nitrogen gas, The startup tank was valved to the engine.The electr ical load on the alternator was s et a t zero.

The engine was then rotated using the hand crank to position thepower piston at top dead center, The control-oil pressure accumulator,which is contained in the control oil mainfold, was then pumped up usingthe small hand-actuated pump, also located on the control-oil mainfold.The engine was then pressurized with helium or hydrogen gas to about3.45 MPa (500 psig).

Next, the combustion system was started. The manual clutch inthe gearbox was energized, and the hand crank rota'& to s tar t the com-bustion-airflow and spark. Fuel flow was controlled by adjusting thefuel needle valve. When the heater-tube temperature reached approxi-mately 704' C (1300~ ), the pull cable was pulled, and the enginebegan rotating. Once the engine attained sufficient speed to overtakethe overruming clutch, the cranking was stopped. The engine pressurecontrol was then allowed to operate normally with the speed governor.The engine speed increased to the normal 3000 rpm. Warmup to normaloperating temperatures required about 15 minutes. The resistance of


17/42

the load bank was then adjusted to load the engine as desired. Thespeed-governor system automatically adjusted engine pre ssures toi n a i n + Aconstant 3000 rpm. The heater-tube gas temperature wascontrolled zxnually by adiusting the fuel needle valve. When necessary,ihe speed governor was adjusted to the co rrec t speed. The engine wasallowed to stabilize for about 15 minutes. The run fuel tank was thenvalved to the engine and the time noted. A data reading was taken ofalternator voltage and current and of all other instruments after 5 min-utes had passed. A seconu reading was taken after 10 minutes. After15 minutes the fuel supply was switched back to the startup tank. Thefuel tank was then weighed on the balance scale. The amount of fuelused was determined by comparing the initial and final weights of thetank. This procedure was repeated for each data point. A ser ies ofload points were run at 3000 rpm with helium as the working fluid andthen with hydrogen. The heater-tube temperature was maintained at677' C (1250' F) for all testing.

The indicated heater-tube-gas temperature varied slightly becauseof the method of control but never was fa fther than 6 . 6 CO (a10 Ffrom 677' C (1250' F) for all data points.

The cwling-wate- temperature was not controlled. Fo r the heliumte st s the cooling-water-inlet temperature ranged from 27' to 43 C0(80 to 109' F). The water temperature tended to increase as thealt ernator load increased. The thermocouple used to monitor thewater-inlet temperature failed to operate during the hydrogen testing;therefore. i t was not recorded during that portion of testing.

RESULTS AND DISCUSSIONThe resul ts of the init ial GPU 3 tests with helium or hydrogen as

the working gases a re presented in two ways: (1) those related to theoverall engine-generator syst em and (2) those related to just the en-gine. Our data ar e compared with heretofore unpublished Army datataken in 1966. Selected data from the Lewis te st s are listed in tabiesIV and V; Army data are given in the appendix (p. 19). Both the Lewis


18/42

and the Army data contained in these tables were o r i g i d l y read inU. S. Customary units; they have been converted to the SI System forreport ing purposes. All of the Lewis data were obtained with thealternator loading the engine. Since most of the Army testing wasdone to verify system endurance, only a smallportion of their data,representing the extr emes in GPU 3 performance, are presented.These data include two se ts of data with alternator and one set withthe engine loaded with a dynamometer. The maximum power absorb-ing capability of the alternator is abolrt 3 kW (5 hp). The Army dyna-mometer test data showed the engine to be capable of producing5.97 kW (8 hp). The overal l GPU 3 system results are presented infigures 16 and 17.

Figure 16 shows the alternator output as a function of mean com-pression-space p ressure. Curves of Lewis data are plotted for bothhelium and hydrogen. Two se ts of Army hydrogen test data are alsoplotted for comparison, The maximum alte rnator output of 2.59 k W(3.48 hp) with hydrogen occur red at 5.00 MPa (725 pig). The maxi-mum a lte rnator output with helium at about the same pres sure was1.33 kW (1.79 hp), o r about half rhe hydrogen output. The pr es su re sfor the no-load condition (1.76 MPa (255 psig)' for hydrogen and2.69 MPa (390 psig) for helium) indicate that a considerable portionof the power is needed to dr ive the engine auxil iaries and to overcomewindage and friction.

The Army data with hydrogen indicated higher alternator outputfor a given mean compression-space pr essure than the Lewis data.This was the case over the ent ire operating range. At 4.90 MPa(710 psig) the output during the army test s was about 3 k W (4 hp) orabout 0.45 W (0.6 hp) above that measured in the Lewis tests withhydrogen. The lower output power for the sa me pr es su re is probablydue to excessive leakage past the power piston rings. A new pistonhad been made after the discovery of a crack in the original piston.Through an oversight, the new piston was made according to the latestrevised drawing rather than as originally fabricated. As a result, thering grooves were made too wide. An attempt to salvage the piston by


19/42

repairing the ring grooves resulted in a poor surf ace finish on thefaces on which the ring was to seal. Another fac tor that may havecontributed to the lower power was the dete rioration and part ial block-age of the regenerators.

Figure 17 shows the overall system SFC with alternator output forhydrogen and helium. The minimum SFC with hydrogen was 669 g/kW- hr(1.1 lb/hp-hr) and occurred at slightly less than 2.4 kW (3.2 hp). TheArmy tes ts yielded slightly higher SFC over most of the power range.At about 2.6 kW (3.5 hp) the SFC's for both tes ts were about 670 g/kW- hr(1.1 lb/hp-hr ). Minimum SFC observed by the Army was 623 g/kW*hr(1.02 lb,/hp= r) at 3k W (4 hp) . The Lewis tes ts were run with heater-tube gas temperature between 671' and 682' C (1240' and 1260' F) andcooling-water temperatures from 27' to 43' C (80' to 110' F). TheArmy tests of June 1966 were run with the heater-tube gas tem peratu resbetween 649' and 666' C (1200' and 1230' I?) and with the cooling-watertemper atures ranging from 48 t~ 56' C (118' to 132' F). The July1966 arm y te st s were conducted with the heat er temperature lowered tobetween 593' and 607' C (1 100' and 1125' F) with the cooling-watertemperatures ranging from 46' to 60' C (115' to 140' F).

The Carnot efficiencies determined by these t emperatures a r econsistent with the ranking of the system SFC observed in the Lewisand Army tests . However, the re la t~ve alues of SFC appeared to bemore sensitive to the operat~ngemperatures than would be predictedby the relative Carnot efficiencies.

The data required to make a detailed heat balance were not taken;the refore , it i s not possible to determine how much of the differencein SFC could be attributed to operating temperature and how much tothe difference in parasltic load caused by removal of the fuel pumpand hydrogen compressor in the Lewis tes ts. Although neither ofthese parasitic loads is a large power consumer, their removal mayhave produced the same or de r of magnitude SFC changes a s the oper -ating temperatures.

The minimum SFC with hellurn was 1274 g/kW-hr (2.1 lb/hp*hr)at about 1 .3 kW (1.8 hp) It appears that the SFC could be even lower


20/42

at a higher al ternator output, but the limitation of the operating pr es -su re precluded further testing.

This represents a 39 percent increase in SFC over the SFC fo rhydrogen at 1.3 kW (1.8 b?). The Lewis computer model (ref. 6),which does not include combustion or mechanical losses, indicatesthat the SFC with helium should be about 12 percent higher than theSFC with hydrogen.

Recently acquired, previously unpublished GMRL repo rts indicatethat the GPU 3 performance with helium was f ar below that predictedby their engine analys is program. They found that the piston-ringfriction was affected by the working gas. When using the same se t ofpiston rings in the test rig and changing only the gas, the power requiredto drive the r ig with hydrogen and nitrogen was the same, but morepower was required to drive the r ig with helium.

The engine brake horsepower was calculated from the al ternatoroutput by making use of alternator efficiency data also obtained from theArmy. Figure 18 shows alternator efficiency a t 3000 rpm with the out-put voltage regulated to 30 volts.

Figure 19 shows the engine brake output power a s a function ofmean working pr es sure for both helium and hydrogen. The range ofArmy data f or hydrogen is also plotted for comparison. The upperboundary of the Army data is dynamometer data taken in February 1966;the lower boundary is alternator data obtained in July 1966. The enginepower in the Lewis test s fe ll slightly below the range for the Army tes ts.The maximum output with helium was 1.68 kW (2.25 hp) and with hydro-gen 3.27 kW (4.39 hp). Both maximums were obtained at a meancompression-space pr es su re of about 5 MPa (725 psig). The differencein brake horsepower between operating with helium and operating withhydrogen was 1.60 kW (2.14 hp). The Lewis computer model predicteda 1.42-kW 1.9 -hp) decrease in power when operating with helium.The maximum power from the Army test data, using a d y ~ m o m e t e rload, was 6 kW (8.04 hp) at a mean compression-space pre ssu re of6.9 MPa (1000 psig) and heater-tube gas temperatures from 680' to706' C (1256' to 1302' F).


21/42

Th e engine BSFC's with hydroge n and with hel ium are shown inf i g u r e 20 as a func t ion of engine bra ke output. Co mp aris on is m a d ewi th the ran ge of BSFC o b s e r v e d d u r i n g t h e A r m y t es t s . T h e BSFCobse rve d dur ing the Le w is t e s t s w i th hydroge n t ende d to fa l l w ith in t herange of the Arm y da ta , a s would be e x p e c te d f r o m the o p e r a t i n g t e m -pe ra tu r e s . The Le w is he lium da t a t e nde d tow a rd the uppe r l imi t s ofthe rang e of Arm y data . The h igher fue l consum pt ion wi th he l iumgives s om e indicat ion of the i n c r e a s e i n i n t e r n a l w in d ag e l o s s e s .

C O X LUDING REMARKSThi s l imi t e d i nve st i gat i on p rov ide s s om e of t he d ime ns iona l and

e xpe r im e n ta l i n fo rma t ion ne e de d to e va lua t e a compute r mode l of aSt i r l i ng e ngine . Th i s w ork is expec ted to cont inue so as to provide abody of S t i r l ing-engine te s t da t a fo r va l ida t ion of com pute r mo de ls ,s u c h m o d e l s to p r ov id e t h e b a s i s f o r f o r m u l ~ i i o n f s p e c i f ic d e s ig nc r i t e r i a f o r a d va n ce d S t i r l i n g e n g in e s.

A s p re s e n t ly p l a nne d , t he G PU e ngine w i l l be re w orke d in to a re-se a r c h eng ine c onfigura t ion . A ux i l i ary c ompone n t s suc h a s t he c om-b u st io n a i r b l o w e r , f u e l p um p , a ~ l d o f o r i h w i ll be re move d a nd re-p lac e d w i th fa c il i ty i t e ms . Eng ine c ompone n t s , suc h a s t he c oo le r -r e g e n e r a t o r s , t h a t are w orn w i l l be re p l a c ed . The e ng ine w i l l beextens ive ly ins t rumented.

Fu tur e work m ay a l s o include tes t ing of advanced component con-cepts . The GPU engine tes t ing wi l l cont inue unt i l a m o r e s u i t a b le re-s e a r c h t o ol is avai lable . Such a t oo l, t he S t i r l i ng ge n e ra l pu rposetest engine, is now being designed .


22/42

APPENDIX - ARMY TEST DATATABLE A-I. - DYNAMOMETER TEBTS OF FEBWARY 1866

Mean Coolfly- Heater-tuba Wodthg-p. EIlplw ?bl E-wo mt4r Inlet gm c o l d - w e output flow G8FCp r e s a l r e tampantun, t empemture tampera-

Sld t dMP. "C OC OC kW g h g h W . hr6.69 61 706 130 6.00 2272 3195.52 55 702 116 4.61 1914 3984.14 4.9 700 102 3.50 1558 4392.93 43 694 90 2.10 1211 5772.07 39 683 62 1.0 3 962 9341 .23 36 680 74 0 739 ---U. S. customary unitsPa@ OF OF OF hp I b h lhhp . hr1000 142 1302 266 6.04 5.01 0.623

800 131 1296 240 6.45 4.22 .654600 119 1292 215 4.69 3.39 ,723425 109 1282 194 2.62 2.67 .946300 102 1262 179 1.38 2.12 1.536178 96 1256 166 0 1.63 -----

TABLE A-n. - SYSTEM TESTS OF JUNE 1966[Working luld. hyd-n; opsnting m p 4 . 3000 rpmr fuel. k.1 uel lower b ea ti ng d u e . 43 070 J / g

(8 530 Btuhb).]Mean Coollng- Heater-tube Alternator output Ntemtor Englne h e 1 System Engine

working wrter M et gas effictency output flow SFC BSFCpressure t emperature temper8ture

SI unltaMP. OC OC V A kW percent kW g/hr g h W . hr g h W . hr4.90 54 643 30 100 3.00 79.0 3.797 1901 634 5014.36 52 654 30 90 2.70 79.5 3.396 1774 657 5223.56 49 663 30.1 71 2.14 80.0 2.671 1560 729 5842 .83 4a 666 30.3 49.5 1.50 80.0 1.875 1379 919 7 352 .10 48 .796 78.0 1.021 1166 1492 11641.46 53 666 ---- ----- 1012 ---- ----4.81 54 79 .0 3.747 1869 631 4994.83 56 649 30.3 99 3.00 79.0 3.797 1669 623 492

U.S . customary unitepa41 OF OF v A hp percent hp Ib/hr lb hp . hr Ibhp . hr7 10 130 1190 30 100 4.023 79.0 5.092 4.19 1.042 0.82 3

212 128 12 30 ---- -----698 130 .820


23/42

TABLE A-m. SYSTEM TEWS OF JULY 10dd[ W o r k h g fluld, hydrogen, c&%anUngpeed. 3000 rpmc fuel, CITE, lower bating value, 43 070 J / g g8 530 B t u b ) . ]

8y-mS KEnglwoutput

E w eBLIE

Meanw o w

praalumSI units

RwIQowWorklog-guaold-.pee

temperstursCoding-

water inletternperare

Heabr-brbbg u

tampenlure

g/hr129312791497186020591139

percent73.57 3 . 579.280.079.0----

MPI1.931.972.724.005.241.52

U . 8. cuctormry ~ n i b

A l t e r ~ t o rutput

kW0.108.408

1.5152.8133.791----

gAW . hr431042631248827666----

OC494752526046

Altdr~DOrefflcleney

g W hr31693135988661512----

OC888510210413291

Ibbp . r5.2105.1551.6251.088.a92-----

v

301'k599593I593604

hp0.547.547

2.0313.1705.090----

A101040751000

bp0.402.4021.6093.016

1004.021----

A101040750

kW0.30.301.202.253.00----

lb/hr2.852.823.304.104 .2-51

percent73.573.579.280.079.0----

L/hp hr7.0897.0152.0511.3591.129----

OFF190185

' 215220no195

Pa@280285

1 3955110760220

VF120117125125140115

11101 CO

11001120


24/42

R E F E R E N C E S1. Heffne r , F. Ea r l : Highl igh ts f ro m 6500 H ou rs of S t i r l ing EngineOperat ion. SAE Paper 650254, Jan. 1965.2 . P e r c iva l , W. H. : Hi s to r ic a l Review of S t i r l ing Engine Deve lopm ent

in the Uni ted S ta te s f ro m 1960 to 1970. (Rept. 4 -E8-00595, Ge ne ra lM o t o r s R e s e a r c h L a b o r. ; E nv i r onm e n ta l P r o t e c t i v e A ge nc y Con-t r ac t 4 -~ 8-0 05 95 . ) NASA CR-12 1097 , 1974.

3. M eije r , R. J. : T h e P h i l i p s S t ir l i n g T h e r m a l E n g in e. PhD. T h e s i s ,Techn ica l Univ e rs i ty De l f t , 1960. (Also publ i shed as P h i l i p s Re -s e a r c h R e p o r t s , S u p p le m e n ts No. 1, 1961, pp. 1 -126 .)

4. D avis , S tephen R.; Henein, Naeim A. ; a n d L u n d s t r o m , R i c h a r d R. :Combus t ion a nd E m is s io n F o r ma t io n in t he S t i r l i n g E ng ine w i thExh aus t Gas Rec i rcu la t ion . SAE Pa pe r 710824, Oc t . 1971 .

5. D avis , S tephen R. ; and Hen ein, Naeim A. : C o m p a r a t i v e A n a l y s isof S t i r l ing and Ot he r Com bust ion Engines . SAE Paper 730620,July 1973.

6. Tew, Roy C . , Jr. ; Je f f e r i e s , K e n t S . ; a nd Mia o , D av id : S t i r l i ngE ng ine Com pu te r Mode l f o r P e r f o r m a nc e Ca lc u l a ti ons . NASATM-78884, 1978.


25/42

ISIS

TABLE I . . PU 3 ENGINE DIMENSIONS

- iFPF-

AND PARAMETERSI Number of cyllndere. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Type of englne . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . apincerIhlve. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RbomblcWorking fluid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . HydrogenDeeignspeed.rpm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3000Deal@ mean compreeelon-space pressure. MP8 $dg). . . . . . . . .19 (1000)Design brake output power. kW @p) . . . . . . . . . . . . . . . . . . . . . 6 @ )Deeign heater-tube gas emperature. OC eF) . . . . . . . . . . . . 677 (1250)Design cooling-water-inlet temperat ure. OC . . . . . . . . . . . . 37 (100)Cvlinder bore. m (in. . . . . . . . . . . . . . . . . . . . . . . . . . .99 (2.7 5)Stroke. cm (In.) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.15 (1.24)nisplacement (maximum change in total working

space volume). cm3 (ln. , . . . . . . . . . . . . . . . . . . . . 119.6 0.30)I)is..lacer rod diameter. cm (in.) . . . . . . . . . . . . . . . . . 0.953 (0.375)Piston rod diameter. cm (In.) . . . . . . . . . . . . . . . . . . . 2.223 (0.875)Cooler:Tube ength. m (In.). . . . . . . . . . . . . . . . . . . . . . . . 4.470 (1.76)Heat-transfer length. m (In.) . . . . . . . . . . . . . . . . . . . 3.480 (1. 7)Tube lnelde diameter. cm (In.) . . . . . . . . . . . . . . . . . . .102 (0.040)Tube outaide diameter. cm (in.). . . . . . . . . . . . . . . . . . .152 (0.060)Number of t u b e per cylinder . . . . . . . . . . . . . . . . . . . . . . . 312Number of tubes per cooler . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Heater:Mean tube length. cm (in.):

Regenerator el& . . . . . . . . . . . . . . . . . . . . . . . . . 12.90 (5.08)Cylinder side . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.63 (4.58)

Heat-transfer length. c m (In.) . . . . . . . . . . . . . . . . . . . 7.772 (3.06)Tube nside diameter. cm (In.) . . . . . . . . . . . . . . . . . 0.302 (0.119)Tube outaide diameter. cm (ln . . . . . . . . . . . . . . . . . . 0.483 (0. 90)Number of tubes per cylinder . . . . . . . . . . . . . . . . . . . . . . . . 80

Regeneratore:Length (inside). cm (in.). . . . . . . . . . . . . . . . . . . . . . 2.260 (0.89)Diameter (Ineide). cm (In.) . . . . . . . . . . . . . . . . . . . . 2.260 (0.89). . . . . . . . . . . . . . . . . . . . . . . . . . . . . .umber per cylinder 8Matrix:

Wire cloth material . . . . . . . . . . . . . . . . . . . . 304 etainlees steelCloth meeh. pe r 2.5 cm (1 in.) . . . . . . . . . . . . . . . . . . 200 by 200Wire diameter. pm (In.). . . . . . . . . . . . . . . . . . . . 40.64 (0.0016)Number of layere . . . . . . . . . . . . . . . . . . . . . ...308F'lller factor. percent . . . . . . . . . . . . . . . . . . . . . . . . . . 28.6

h i v e :Connecting rod length. cm (In. . . . . . . . . . . . . . . . . . . 4.597 (1.81)Crank radfua. cm en.). . . . . . . . . . . . . . . . . . . . . . . .379 (0.543)Eccentricity. cm @I.) . . . . . . . . . . . . . . . . . . . . . . . 2.083 p. 2)


26/42

TABLE 11. - INSTRUMENTATION ORIGINALLY SUPPLIED WITH GPU 3Item123456789

10111213141516

ParameterHydrogen bottle pr es su reHydrogen supply pressureOil pump pressureCompression-space mean pre ss ureBuffer-space mean pres sureNozzle fuel pressureEngine oil pre ssu reEbel pump pressureHeater-tube F?S temperature (2)Compression-space temperatureOil sump temperatureWater-inlet temperatureAlterna tor output voltageAlternator output currentBuffer-space gas temperatureEngine speed (not operable)

Item17181920212223

InstrumentThermocoupleThermocoupleGageCageGageGagepulse type with

frequency meter

ParameterWaterloutlet temperaturePreheater air-inlet temperatureCombustion-blower outle t pr es su reFuel nozzle atomizing ai r pre ssur eGovernor oil control signal pressureCoollng-water supply pressureEngine speed

InstrumentGageGageGageGageGageGageGageGageThermocoupleThermocoupleThermocoupleThermocoupleVoltmeterAmmeterTherlnocoupledc generator

Range0 - 500' F0 - 500' F0 - 35 in water0 - 15 psig0 - 60 psig0 - 15 psig--------------

Rarae0 - :MOO psig0 - 1500 psig0 - 100 pslg0 - 1000 psig0 - 1500 psig0 - 15 psig0 - 100 psig0 - 30 pslg0 - 1 5 0 0 ~0 - 500' F0 - 500' F I0 - 500' F0 - 3 0 V d c0 - 1 5 0 A0 - 500' F2600 - 3200 spm


27/42

TABLE N. W OPU STEBT M $A -EN W-0 IZUID

Uun ikatw-tubo A l t u r u O r o u ~ t AlUnutar Eaglm f in1 Q-mworldal lu .mbm9 o@ul now SIC BSIC

p m a ~. m p n t u r r5 un\'#

LB. OC V A kW pram$ LW g/br # / LW .h r s/kW.hr1.74 674 20.0 0 0 -- --- 872 - -.86 671 38.0 6 .114 18.4 0.240 798 4601 33a91.03 674 29.6 11 73.7 1 873 2003 10112.05 674 99.5 16 .w 76.0 .64n 1024 alc, 1578a. s1 bn 99.6 2s .7rr 77.2 .w 1 o a IS= 10742.71 874 20.0 56 1.044 78.8 I.-?! 1064 1019 8013.00 077 20.3 40 1.348 79.0 1.706 1236 917 7263.36 077 29 .3 66 1.612 86.0 9.016 1904 809 0473. 90 074 39.5 69 3.036 8C.3 2 . 6 s lr76 726 6824.46 677 29.1 61 2.367 80.0 2.846 1576 069 6365.00 677 28.6 91 2.594 79.3 3.271 1768 682 541

C.8. - w r y unit.p.Cr OF V A bp percent bp I b h Ib/hp. hr I b h hr

262.6 1246 99.0 0 0 -- -- 1.922 -- --270 l240 29.0 0 .233 72.4 0..?%2 1.762 7.662 6.472180 1246 2 9 . 5 1 1 . 4 3 5 73.7 .690 1,924 4.4M 3.268297.6 1 2 20.6 16 .652 76.0 .a89 2.267 3.402 2 . W336 1260 2 9 . 5 2 8 1 . 0 2 8 77.2 1 , ' s 2.364 2.290 1.761598.5 1246 29.0 36 1.388 78. 6 1.780 2.346 1.677 1.318435 1260 2 9 . 3 4 0 1 . 8 0 7 7 9 . 0 2 . 2 8 7 2 . 7 3 5 1.608 1.192187.5 1260 2 9 . 3 5 6 2 . 1 0 0 8 0 . 0 2 . 7 0 0 2 . 8 7 5 1.331 1.066

29.5 69 2.729 80.3 3.399 3.264 1.192 .95729.1 81 3.160 80.0 3.960 3.474 1.099 ,87928.5 91 3.477 79 .5 4.386 3.888 1.121 .889

TABLE V. - LEWIBG P U 3 TEST DATA WITH HELN M WORKING FLUID1-m .psed. 9 6 ~ )pni b 1 . No. 1 *mlc fuel lower vrlw, 43 197 J/g (18 664 muAb).j

M a n Coollag- ReUer-tube A l t u r ~ t o r utput Almrmtor Englue -1 dyrurn Er gh eworldng vlserinlet rn etllelency mt Ilow SFC 88?Tp m - t e m p sn u m tenrprrhre

SI unit.L6p. OC OC v A k'g percent kW g/hr gkw hr g/lrW hr2.69 27 671 28.5 0 0 ---- -- 11% --- A2 . ~ 1 ae 677 2 9. 0 7 . 2 ~ " 7 2. 8 0.279 iiao 6714 11583.09 32 679 29 .0 12 .W 73.9 .471 1197 3440 "5413.33 29 677 2 9.0 17 ,49 3 76 .3 .666 1286 2629 8103.87 32 682 29.0 n .783 77.4 1.012 14% 1 . 9 ~ 141s4.22 37 679 29.0 30 1.OU 78.9 1.323 1608 1444 11405.02 43 677 29.0 46 1.334 70.5 1.678 1700 1274 1013 -

U.8.cu.tom.ry unit.P* O F OF V A hp percent hp IbAr l b b h r I b b . r390 80 1240 28.5 0 0 --- --- 2.504 -- ---407.6 82 1260 29.0 7 .272 72.8 0.374 2.667 9.401 6.837447.6 90 1255 29. 0 12 ,466 73.9 .631 2.640 6.773 4.184

1 2 1 20 .0 1 1 71 . 0 01 1 76-3 8 7 2 . S 4.322 3.264 I,533.6 1260 29.0 27 1.060 77. 4 1.366 3.100 3.015 2.336612.6 lZ56 2 9 . 0 3 0 1 . 3 8 9 78.9 1.773 3.326 2.377 1.174727.5 100 la60 2 3 . 0 4 8 1 . 7 8 8 79.6 2.249 3.746 2.0# 1.667


28/42

ORTGINAI: PAGE EQOF POOR QUALITY

n p r ~ - CPU 3 Stirlina around p e r unit.


29/42

Figure 2 - Cross section of GPU 3 engine.


30/42

ORIGINAL PAGE ISOE POOR QUALITY


31/42


32/42

Figure 5. - Cooler-rrqmerator cartridqe tssmbly.


33/42

ORIGWAL' PAGE ISOF POOR QUALITY


34/42

Ru lm guide su-4 C k i u "DU" bushingDisqpin rodwiper .

n e e it 1ith tin platingon shaft-,Figure 7. - CW 3 pisbns. shafts. and seals assenbly.


35/42

C-77-7146Rqureg - Displacer shaft and seal VnS.

ntj:lx*r[a0 L I I ~ V~ntr l n n ~ rinqC-75-4)9;Fiourr 9. - CPU 3 Pistons and piston rinqs.


36/42

'B imetdl ic ;'JIlL ,j sensors in ,,: h d e r t u b s ' F ue l c uto ffI1

l"6770 C( l a 4 )

A .L F u e l

o m ' AGE IS- ue 1-- ontrol oi l OF POOR QUATJTY

Figure 10. - Heater-tube temperature control, simplified sche matic

high-pressur

WarkingspacemGovernor-actuated Hydrogen 'hydrogen overspeed JBuffe r bypass valveA-J-&$ydrogenontrol - ydrogen--- ontrol oil

Figure 11. - Engine speed control, simplified schematic.


37/42


38/42

Fuelnw

-controlIle va lve-- - -

FIOUTP 13. - GPU as tested.

F iqur~ d. - GPU 3 test sdup.


39/42


40/42

June 1966 Armdata hyd

~L sw is ydrogenLb

I I .-Lewis helium data

- a., 1 : I d u n e 1966 Army\ h f l r w m dab

Mean compression-space pressure. MPaI I 1 I I I0 200 4 0 0 d 0 0 8 0 0 1 O O OMean compression-space pressure. psig

Figure 16. - GPU 3alternator output versus mean working pressureat Ma) pm.

-. .m/- hydrogen data-'

Alternator output. kW

Alterndor output. hpFigure 17. - GPU 3 system specific fuel consumptionversus llternator output at 3000 rpm.


41/42

1 2Load, kWF l y r e 18. -Alternator eff ic iency versus loadat M volts ano MOOrym.

e"~ ~ e w i sydrogendata

Mean working pressure. MPa1 I 1

0 200 4M) WO 80 0 1000Mean working pressure, psig

Figure 19. - GPU 3 engine output versus mean working pressure.


42/42

I I I I 1 I I I0 1 2 3 4 5 b 1 f8Brtt cnjw wlplt hp

Ciwrea. WLi 3 sgaific ice! anrumptionwrus atginc oulplt.

Documents

sterling engine test