AIRCRAFT PERFORMANCE REPORT

Sponsored and Funded by the Experimental Aircraft Association e Federath d an l Aviation Administration Aircraft Exhaust SystemV sI TRIAVIATHON TROPHY BY BRIEN A. SEELEY, ED VETTER AND THE CAFE BOARD

CAFE he goa thif o l s repors i t A FA e th d CAFEan A EA , FOUNDATION to improve the power, are grateful to the following T efficienc reliabilitd an y y major contributor thio st s study: PRESIDENT aircraff o t exhaust systemse Th . Aerospace Welder f Mino s - Brien Seeley report summarize resulte sth f so neapolis for the high quality a 16 month long study. Many of exhaust merge bald san l joints, VICE the systems tested her sime ear - George Johnston of EAA Chap- PRESIDENT ila o onet r s popularly usen di ter 124 for the lathe-machined Larry Ford light aircraft teste Th s. include model of the Coanda nozzle, int4 collectoo1 r systems4 , Sam Davis at Tube Technolo- TREASURER into 2 "crossover" systems, gie Coronan si , Californir fo a CJ. Stephens "Tri-Y" system independ san - the stainless steel exhaust sys- dent exhaust stacks. Additional tem derived from thes tests, and SECRETARY Bill Cannam certifiea , d welder Daniel Waymaii aspects of exhaust design in this study are: ChaptefroA mEA e r th 124 r fo , major effort to assemble the TEST PILOT stainless steel . C.J. Stephens Intake waves Curt Leaverton, Jack Norris, Wave speed Andy Baue Stevd ran e Williams DIRECTORS each contributed professional Crandon Elmer Megaphone effects scientific analysis of the EPG's. Otis Holt effectM RP s Jack Norris Exhaust jet thrust Gris Hawkins ABBREVIATIONS Stephen Williams Crossover/Tri-Y reflections Ed Vctter Frequency analysis (FFT's) EVO = exhaust valve opening Header size EVC = exhaust valve closure Collector size IVO = intake valve opening CHALLENGE TROPHY Coanda nozzles FVC = intake valve closure Bends in the pipe TDC = top dead center effectT CH sd an T EG BDC = bottom dead center Ball joint effects W.O.T. = wide open throttle gph = gallons per hour Ove separat0 r35 e Exhaust dB = decibels, slow A scale Pressure Graph (EPG) record fas= -tT FourieFF r transform ings were made using the cy cylindel= r Lycoming IO-360 A1B6 engine coll. = collector CAFe inth E test-bed Mooney msec. = milliseconds M20E thesf o l eAl . were made at 125' MSL as static ground Hz. = Hertz or cycles per sec engine runs of approximately "Hg inche= . Mercurf so y 15 seconds duration. O.D outsid= . e diameter 34 JANUARY 1997 BASIE G TH CEP Files 411 Basic EPG 1.75x34.5x2.25x19.s int4 a : o1 5 equal length headers. 29.5" M.P., A Review 2731 RPM 20.4 gph 86"F. 8-18-96. 125'MSL. Lycoming 10-360 A1B6 : 1324 See text for explanation of P, C, S and R waves shown below

Figure 1 shows a basic EPG. It was 411 Cyl#1 411 Cyl #Z —— 411 Intake --- 411 Collector Wave recorded on a well-tuned 4 into 1 col- Bot hsho2 # cylinderopenin w wd lo an g1 s# lector exhaust system which will be pressures at EVO and very low pressure hereafter referre "Fils a o det 411". Fig- during overlap TDC. Some of ure 1 shows features which arc e intakth e trace appearP occuro t se Th . wave (red) goes negativee th earl n i y Aft looking view Figure 1 essential for understanding the other exhaust cycle. graph thin si s report. Th" axise"X , of collector entry O along the bottom of the graph, shows the degree crankshaff so t rotatio- nbe ginnin deap to dt g a cente r (TDCf o ) the firing stroke for cylinder #1. The vertical "Y" axis shows the pressure measure pipe th . inche n ei n di Hg f so Since these runs were made at near sea level zere th , o pressure level represents ambient pressur abouf eo t 29.92. "Hg The typical EPG shows a steeply risin wav" gf exhaus"P eo t pressure, shown in red, which starts upward at the point of exhaust valve opening (EVO). The tall P wave typically falls beloo t w zero (ambient) pressure later exhause inth t cycle. The intake pressure is shown in — Exhaust Valve blue. There is a black vertical dotted Intake Valve lin t BOea C afte intake th r e stroke, wher piston'e eth s descent ceases. 30 60 90 120 150 180 210 240 270 300 330 360 390 420 450 480 510 540 570 600 amoune Th - f valvo ex t e eth liff o t Crankshaft Degrees After Firing TDC haus intakd an t e valve shows si tht na e bottom of the graph. At overlap TDC, intake valve opene sar pressure e th , n si unfavorable reverse flow could occur both valves are open for a brief interval. the exhaust pipe, combustion chamber at overlap argumen.e Thion - s si us r fo t The EPG often shows additional and intake tract can all influence one ing wide open throttle (W.O.T.) waves which come from reflections, another. How much influence depends whenever possibl hign ei h altitude turbulence and collector-equippen ,i d upon the valve lift during overlap and cruise flight. systems, the firings of the other cylin- how long both valves remain open. ders (cross-talk). These are labeled by During overlap TDC, the suction in a PUMPING GAS their cylinde waveR f origie ro th s s na 's header can help empty in Figur. 1 e combustioe th n chambe burns it s f ga rto Normally, engine designers try to waveC e thosc sTh ar e measuren di residues. This effec calles i t d "scaveng- place EVO about 40-75° prior to firing the collector, the common pipe into exhause ing"Th . t suctio evey nma n BDC so that the peak of the very high whic mergec har individuae dth l head- enhance the combustion chamber's fill- in-cylinder pressure can be dissipated ers. Each cylinder produces a separate ing from the intake valve, thus im- during blowdown, before BDC. A C wave. The time T between the rise of proving and tuned exhaust system, wit hvera w ylo attendane risth e th f wavth eP o d etan horsepower. With sufficiently long opening pressure at EVO, can assist in wavC usee vers eb i dn y ca shor d an t overlap intervals possibls i t i , r theefo evacuating the cylinder quickly, and to calculate the velocity of the wave. suction to pull some cool intake charge can thu bso t e allo delayeO wEV d until The "blowdown" cycle is defined across the hot exhaust valve, cooling later in the cycle. The later EVO al- perioe th firins o t da froO g mBDCEV , the valve face, stem, seat and guide. lows the positive in-cylinder pressure and is labeled "B". It is during this in- Such cooling comes at a price, which is to do more work pushing the piston terval thasteee wavP th t pe eristh f eo that raw fuel is being wasted out the ex- downward prio EVOo rt . Thus tunea , d is seen, as the cylinder discharges or haust pipe. Higher compression pistons exhaust system work stime besth f -i t 'blows down' throug exhause hth t should scavenge theibetteo t e r du r s i delaye O o takt EV d inf eo g valve and the in-cylinder pressure smaller combustion chamber volume. advantag tuninge th f eo . rapidly falls. Positive in-cylinder pres- Note tha Figurn i t ee intak 1th , e After blowdown in the exhaust sure during blowdown is still doing pressur greates ei r tha exhause nth t stroke pistoe th , n begin riso st e from some useful work by pushing down- pressure at overlap TDC. Such a pres- risinBDCA . g piston pushing againsta ward on the piston. sure gradient will encourage sca- high in-cylinder pressure causes a loss vera Overla s i y importanC pTD t venging t parA . t throttle intake th , e of power know "pumpina s na g loss". interval. When both the exhaust and pressure woul muce db h lowerd an , Instead, the rising piston should be SPORT AVIATIO5 N3 pulled upwar negativa y db e pressure thn i e cylinder, thus producinga Files 411/41 effec e megaphonea Th f 3o t 1.75x34.5x2.25x19.s int4 a : o1 5 equal length headers. File 413 has a 17Lx2.25x4" megaphone added to 411. Both at 29.5" M.P. "pumping gain". Suction in a tuned ex- (W.O.T.), 86°F. 8-18-96. Lycoming IO-360 A1B6 firing orderl324 Run at 125'MSL. haust syste n producmca e suca h pumpin lato t egd exhausgaimi n ni t —— #411 2731 RPM 20.4 gph 106.9 dB. no meg stroke. This is shown in Figure 1 where #411 intake pressure, no meg • the exhaust pressure goes negative at —— #413 2737 RPM 20.6 gph 107.8 dB, with meg 260° of crank angle, which is 80° after #413 intake pressure, witg hme BDC earlie e exhause Th th . n ri t cycle e megaphonTh n increasca e e powey b r that the P wave subsides and goes neg- lowering the opening pressure at EVO, and ative or below the ambient (zero) scavenging mor t overlapea noise Th e. level pressure, the more pumping gain can is significantly higher, however. occur, makin greater gfo r horsepower. Thus, an ideal exhaust system Figure 2 should produce a highly negative pressur exhause th t ea t valv bott ea h EVO and again as soon as possible after dissipating the P wave. This cylinder filling negative pressure shoul made db o et concludes persist throughout the overlap strok thao es t favorable scavenging can occur. INTAKE PULSATIONS piston'e Th s descent during each intake stroke exerts strong suction on the intake pipe runner connecting the carbureto cylindere th e o th rt f o .l Ifal intake runners attac commoa o ht n plenum, as in the Lycoming engines, -20 30 60 90 120 150 180 210 240 270 300 330 360 390 420 450 480 510 540 570 600 suctioe th n will f thosaffeco l eal t run- Crankshaft Degrees After FirinC gTD ners suctioe Th . n cause e flosa b wo t initiate directioe on n di n whics hi Yag t ale i 17 have writte exceln na - represent 273e M timth e r 1sRP th efo abruptly stopped whe intake th n e lent paper on using induction system wavP reaco et collectoe hth frop rta m valve closes. The flow stoppage cre- pulsation forco t s e fee engine'e dth s headere th f o distanca ,thp eto f eo ates a reflecting wave which again cylinder during the intake stroke. 47.0". This compute abouo st t 1604 affects all of the runners. This leads to t observno Wpressurd y edi ean e fps average speed. This slower speed intake pulsations. pulseintake th e n si eth waveo t e sdu suggests that some slowing occurs sa intake Th e pulsation- Ly e th n so propeller blade sweeping past the air the header wave enters the collector. coming IO-360 A1B6 engine ar e cleaner intake. However, these tests File 412 t 250a , 7 RPM, showen da sizable and can be seen in Figure 1. had the air cleaner located 11-12" aft average wave spee f 150do 8 fpsa , These pulsations can show how much propellee ofth r disc thosn O . e cowl- 7.5% reduction from an 8% reduction scavenging effect migh expectede b t , ings with a very far forward air in RPM. The reduced speed is due to and the character of the cylinder fill- cleaner intake, the EPG may be abl slowee eth d ar an lowe averagT EG r e latteinge servn Th guid.a rca s ea o et deteco t t whethe proe rth produc s pi - piston speed which gives a slower the relative volumetric efficiency of ing a pulse into the air cleaner at just mass flow. enginee th . the right moment during the intake exhause Th expands ga t coold san s W.O.Te Th . intake puls reacn eca h cycle. as it goes down the pipe, and the wave as hig 6-7s h a . abov "Hg e atmospheric velocity varies directly wite hth pressure sees a ,t "+Sna Figurn "i . 1 e WAVE SPEED square root of the ratio of the absolute This effect thus gives an instantaneous exhaus temperaturess ga t . manifold pressure of about 37" Hg., The EPG can show the average and timef i , d correctly t someac n -ca , speed of a wave traveling through the MEGAPHONE EFFECTS what like supercharging. Ideallye th , pipe. The wave speeds observed actu- lowest point in the intake pulsation average allth yf o represen em su e th t Figur showe2 s tha megaphona t e trough shoul timee db occuo dt "-St ra " sonic wav eaverage speeth d dan e added to file 411 produced a lowering or about 60° after overlap TDC. This mass flow velocity. openine th f o d g an pressur O EV t ea will ten assuro dt e thanexe th t t posi- A test using sensors 21.5" apart on better scavenging at the expense of tive pulse will arrive just prio IVCo rt , 1.625a " primary header showen da more noise. A megaphone was later enhancing flow through the intake average wave speed of 1751 fps. added to a Tri-Y system and showed valve just as cylinder filling ends. In Figure 1, the time interval "T" minimal influenc EPGe th n .eo 36 JANUARY 1997 Files 502/422/510: Varied collector diameters. All use the same 4 nto 1, equal length THE EPG TEST METHOD headers of: 1.75x34.5 with -30" collector length. 73-82°F. 8-24-96. Lycoming IO-360 The EPG pressure sensor was con- A1B6 firing order:1324 Run at 125'MSL

" lonnecte9 a g o coppedt r tubf o e 2x29" C 19.7 gph 2635 RPM #502 0.125" O.D. flush-mountee th o t d header pipe's inner wall. The mounting #502 Intake The 2.25" collector 2.25x30" C 20.3 gph 2687 RPM #422 was at a point 1.25" downstream of the seems optimal. flange signale Th . s were #422 Intake processe Vettee th y drb Sensor Acqui- 2.5x30" C 19.9 gph 2669 RPM #510 sition Module and Digital Acquisition Device. Sensors were calibrated using a #510 Intake water manometer. Multiple small waves in 1 A new amplifier was used for this #50y 2ma study. Its faster response time and multiplo t e bedu e slip 4X2 joints used. Aftward view of higher resolution provide dmuca h bet- collector entry ter picture of the EPG relative to those geometry yb in previous reports. 1.2.3 cylinder* intake Th e pressure recordings were made 1.5" upstream of the intake valve throug fuee hth l injecto- Ly re porth n i t coming cylinder head. RPM, noise level, static thrust in pounds, fuel flow, wind incidene th o t t propeller and manifold pressure were recorded manually. Variatione th n i s mixturd RPMan . T eEGT werCH , e used severan o l run studo st y their effects. EPG'e th f o s l Ishownal n heree th , timinwavee th f go s with respece th o t t Exhaust Valve crankshaft degrees has been shifted to — Intake Valve the lett (earlier y 1.2)b 5 millisecondo st -20 compensat e 12th inc) a r h edistancfo e 60 90 120 150 180 210 240 270 300 330 360 390 420 450 480 510 540 570 600 which separate pressure sth e sensod ran Crankshaft Degrees After Firing TDC the exhaust valve face (1.0 millisec- Files 411, 415, 419comparo t 4 , 42 422 ed differenan , t collecto same rth lengthse 4 us l Al . ond), b) the electronic rise time of the into 1, equal length headers as: 1.75x34.5 with a 2.25" diameter collector. 78-86°F. 8-18- pressure sensor (0.15 milliseconds) 96. Lycoming 10-360 A1B6 firing t 125'MSLordera n :Ru 1324 . amplifiee ed )th an r delay (0.10 mil- 40 EVO, blowdown ————— 49.5" 20.2 gph 2666 RPM #415 liseconds). This place wave sth e timing and overlap correcs ait t t phasing wit e valvhth e pressurel sal ••••• #415 Intake opening cycles. chang collectos ea r length is altered. _._._._ 29.5" 20. #42 268h M 32gp 7RP Fast Fourier transforms (FFT's) Fuel flows and 30 —.--..- #422Intake were made on each of the runs to look RPM'S, suggest the 19.5" or ————— 19.5" 20.4 gph 2731 RPM #411 at the sonic frequencies which had the 29.5"collector as greatest energy content. Analyzing optimal. Intakes - #411 Intake these transforms exceed scope sth f eo show only mino| r changes due to 20." #42270h 10 M 2gp 4 0RP this bibliographe reportth e Se . r yfo 20 limited valve #424 Intake several reference n wavso e theory. .j overlap. Noise levels were taken froe mth area between the front seats of the air- craft wit pilot'e hth s side vent window open usinscalA e egth slow setting. Nois s reduceewa d whe tailpipe nth e exit was moved aftward relative to the noise meter, as occurred with the longest tailpipes. Peak RPM and fuel flow generally correlated wit thruse hth t valued san were used as a rough guide to power output. The anemometer showed a chang locan ei l wind spee direcd dan - tion as the propeller's flow field reached full strength at maximum static RPM. This flow was allowed to equili- brate before the RPM and fuel flow 0 60 0 57 0 54 0 51 0 48 0 45 0 42 0 39 0 36 0 33 0 30 0 27 0 24 0 21 0 18 0 15 0 12 0 9 0 6 0 3 O readings were taken. Crankshaft Degrees After FirinC gTD SPORT AVIATION 37 BENDS IN THE PIPE Files 723/724/72 effects f thesM o l 5RP eAl : header 1.625x28e sar " wit collectoo hn r except file 411 whic s 1.75x34.5x2.25x19.5hi t 97°a l FAl .excep t 41 t 86°F1a . Lycoming IO-360 A1B6 firing File 41 I's header pipe bends were order:1324at125'MSL. as follows: ©2674 #72# stu" l M 6 b7Cy 0RP Cyl#l: 25°+ 90°+170 285°= ° -•-•• 28" Cyl #3® 2710 RPM #723 265= ° °90 + ° 90 + ° 85 Cy: #2 l ——— 45.5" Cyl #3 @ 2670 RPM #727 Cyl #3: 35° + 170° +180° = 385° @3 # 270l Cy 0 " RP68 M - #72- - 8 Cy: 80°#4 l + 70° ° =170+20 ° ——— 34.5 collecto"+ r @ #41273M 1RP Th " lone68 g 45.5 d headeran " s each Individual testing of these separate hav broao waveto P da e that delays cylinder t sho significany no wd an sdi t their negative th f eo l waveAl . independent header o collector(n s ) changes in their EPG waveforms. See show characteristic "ringing" waves Figure 1. , cylinder#2 d an 1 s# followin" stu6 wave P s bi e e Th gth . too short to contain a fully developed Many aircraft use a downward bend waveP collectoA . r equipped system in the tailpipe to keep exhaust soot off (#411) shows much improved tuning. the aircraft's belly. Keeping collector Figure5 length constant, files 502 (a straight 2x29" collector), 503 (2x29" with a (1.54 exit) 50 e ben° "th d 90 t dan a , nozzl straigha n eo t 2x29" collector) were teste t W.O.Tda resulte Th . s were EVO opening pressures of-5.0, - 4.0 and +3.0, respectively with overlap pressures of-10.0, -10.0, and -4.0, re- spectively P wav e eTh . width remaine samee dth . File 503, with a 90° downward bend of the collector at the exit, caused an insignificant increase in backpres- -20 0 66 0 63 0 60 0 57 0 54 0 51 0 48 0 45 0 42 0 39 0 36 0 33 0 30 0 27 0 24 0 21 0 O18 0 315 0 0 12 0 9 sure. The nozzle did impose a sig- ;., ,-.-. Crankshaft Degrees After Firing TDC nificant backpressure penalty.

COLLECTOR SIZE Files 419/420/421/412/42 collectod an M r 5RP lengt h effects l headerAl : 1.75x34.5"e sar . Files 419, use1 42 d a 2.25x40420 d an , " collector. useFil2 e 41 2.25x19.5d a use 5 fild e 42 2.25x10d an a " " collector 78-86°Ft a l Al . . Lycoming IO-360A1B6 firing order: 132 t 125'MSL.4a . See Figure 3. These tests repeat- 40 At 2500 RPM, 2672 RPM, 20.0 gph, 105.4 dB, #419 edly showed that r thifo ,s particular shortenine gth engine 2.25e th , " diameter collector collectos ha r 2501 RPM, 15.2 gph, 103.2 dB, #420 bess optimizinr wa fo t g exhaust back- little effect on both the P pressur a levelse t 2.125A a .e " 2507 RPM, 106.3 dB, 19.5" coll., #412 30 wave timing diameter collector would probably and the intake waves. The 2504 RPM, 14.4 gph, 105.8 dB, 10" coll., #425 giv gooea d compromise between 10" collector climb powe higd an rh altitudt eje shows higher 228• 9— RPM— — , 11.3 gph, 100.2 dB, #421 pressure at thrust. EVO and 20 . Collecto 6 e Figur d Se an e4 r during length appeared to optimize at 20-30". blowdown and O) overlap. Lower It mus lone b t g enoug develoo ht p RPM gives OT some continuu f flomfulld o wan y

See Figure 5. The addition of a col- lector to 4 separate independent pipes consistently caused the entire EPG to shif lowero t t , more negative pres- sures. Some suspect that this effect ma causee more yb th ey d b continuou s mas scollectoe floth n wi r exertinga prolonged vacuum effect f upoo l nal the headers. A suitable collector was -20 one with about 50-90% greater cross 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 sectional area than each individual Crankshaft Degrees After FirinC gTD 38 JANUARY 1997 Files 729/730/731/732 Crossover systems of various blind leg lengths. Cyl #1 has a heade witd ran lengtha t leasa f hto 18" 1.625x28" primary header wherexcep2 45.573 s i n t ei i t " blinlonge Th d. offtak 2.5s ei " or so. See "length formulae" below. downstrea l headcy 1 .# f m97°o F Lycoming 10-360 A1B6 firing order: 132 t 1254a ' MSL. Note how blind end pressures are reflected back into the header to alter timing. In Figur smale th , el5 waves which occur after the P wave in the indepen- — 729 cyl #1 28" with 67" blind dent pipe callee sar d 'rings'a n i s a , 730 cyl #1 28" with 51 .5" blind doorbell ringing. The negative por- ——— 731 cyl #1 28" with 36.5" blind tions of these rings are of such short - • - 732 cyl #1 45.5" with 36.5" blind duration that they require the pipe de- blin72" d 967 d en signer to choose between positioning 731 36.5" blind den them early in the cycle to obtain e besth ts combinatioha 1 73 n pumping gains or late in the cycle to w pressurlo of e durine th g scaveng t overlapea collectoe .Th r exhaust cycle and good syste sucs m ha lonh a g duration nega- scaveng t overlapea . Nonf eo these systemsw showlo e th s tive wave after the P wave, that it opening pressure at EVO or serves both purposes, i.e., gives pump- the mid cycle pumping gains found with the 4 into 1 collector ing gain as well as scavenging at systems. overlap. CROSSOVER TRI-D SAN Y A crossover exhaust joins the head- f cylindero s er s whose firings occur 180 crankshaft degrees apartP e Th . Figure7 cylindee wavon f eo r will then travel upstream to the other cylinder where it will closea bounc f o f d eof exhaus t valve and return. Pipe lengths in the crossove chosee b - n re thao e rnca s th t turning wave will produc enegativa e -20 O 30 90 120 150 180 210 240 270 300 330 360 390 420 460 480 510 540 570 600 pressur scavengingr efo . Crankshaft Degrees After FirinC gTD differeno Tw . Figur8 e td Se an e7 crossover systems were tested. One Files 511/516 to compare crossover 4 into 2 versus a Tri-Y system. All use the same equal simulatioa s wa n mode whicn i l n ha length header : 1.75x34.sas 5 with 511 havin eacg2 h 1.875"x18" tailpipeTri-e th Yd san independent heade cylinden ro d ha 1 r# having a merging of those 2 tailpipes into a single 185x2" outlet. 77-78"F. 8-24-96. Lycoming IO-360 A1B6 firing order: 1324 Run at 125'MSL. BothatW.OT an offshoot pipe welded on about 2.5" downstream from the cylinder flange. #511 4 into 2, Cyl #1 e offshooTh tblina pips g dle ewa #511 Cyl #2, 2660 RPM 20.0 gph whose length coul adjustee db n o d dan -.-.-•- #516 Tri-Y, Cyl#1 f whico d pressurha en e th e sensor #516 Cyl #2, 2703 RPM 20.5 gph was attached, as shown in Figure 7. These two systems show that the main The other crossover system (File Figure8 behavio e systeth f o s mi relatee r th o t d crossove g lengthle r , rather thae th n 511) was 1.75x34.5" headers pairing e additioTri-Y'th f o sn tailpipe-collector. cylinder #1 with #2, and #3 with #4. However, the Tri-Y's collector does seem to shift the entire waveform toward lower Each of these pairs of cylinders fire pressures and give higher RPM and fuel 180° apart Figure Se . . e8 flow Both systems clearly wavP sho e ewth reflected to cyl #2 at X, and that reflection's The reflected waves fro bline mth d return into the cylinder #1 trace at R. leg are powerful and their effect on 8 i. the P wave can be clearly seen here. In Figure 7, Xb marks the peak of the P wave's arrival in the 67" blind leg of simulatioe th n i 9 fil72 en model. XI marks the ill-timed return point of that peak into the cylinder #1 pressure trace. This ruins the tuning. Nb marks troug e 36.5e th th n hi " f filblino eg dle 731 l showN . s this trough's returo nt cylindee th helo rt t developi pnegaa - tive scavenging wave. This simulator -360 lack influence sth #2'l cy sf efiringso . -240 -120 The crossover system showed bet- -20 O ter performance if the length from O 30 90 120 150 180 210 240 270 300 330 360 390 420 450 480 510 540 570 600 cylinder flang tailpipo et " e28 exis twa Crankshaft Degrees After FirinC gTD SPORT AVIATION 39 Files 516/517/601/603/411: Tri-Y system at differing RPM'S. Headers are 1.75x34.5" Into rather than 45". two separate 1.875"x18" intermediates which then merge into a single 2x18.5" collector for files 516/517 but into a 2x6" collector for files 601/603. Cylinders #1 and #2 are merged A commonly used, 'off-the-shelf . togetheFil cylinder#4 e e int d ar 4 #41 collectoos a an 1 a rs 13 i s # r system. 78"F. 8-24- Lycoming crossover syste s 1.75mha " 96. Lycoming IO-360A1B6 firing order; 1324 Run at 125'MSL 40 O.D. headers wherein cylinder1 s# 2703 RPM 29.7" M.P. 20.5 gph 2x18.5 coll. and #2 are joined about 11" down- — — — 2476 RPM 25.0" M.P. 15.2 gph 2x18.5 COll stream of cylinder #2 and 33" down- 29.8269M " 9 RP coll 6 M.P2x . . h 19.gp 9 strea f cylindemo r #1. This makesa 24.8M "251 RP coll6 M.P3 2x .. h 14.gp 8 44" "blind leg" or crossover length be- tween those two cylinders. Cylinders 2731 RPM 29.5" M.P. 20.4 gph File #411 #3 and #4 are similarly joined. These Figure 9 joined pipes then each exit througha 2.125"x 16" long tailpipe. The Tri-Y system Figurn s i wa e8 1.75x34.5" equal length headers which merged cylinders #1 with #2 and #3 with #4 into 1.875x18" intermediate pipes intermediatee Th . s merged into a 2x18.5" collector. In Figure 9, the Tri-Y showed a large amount (-15.0" Hg.) of suction during overlap t thibu ,s e camth t ea sacrific botf eo h opening pressurd ean pumping gain relativ greee th o net trace of file 411 's 4 into 1 collector. At lower RPM largea , pumping gain appears but the scavenge is lost. At Exhaust Valve graphe th n o larga , " e"F pressure Intake Valve trace arrives from cylinder #2'P s wave influence reflectioe th s i t I .f no -20 such large pressure waves that make 0 66 0 63 0 60 0 57 0 54 0 51 0 48 0 45 0 42 0 39 0 36 0 33 0 30 0 27 0 24 0 21 0 18 0 15 0 12 0 9 0 6 0 3 O negative th e pressure dramatio s n ci Crankshaft Degrees After Firing TDC the Tri-Y and crossover systems. See Figure 10. This expanded scale Files 516/517/51 comparo t 8 Tri-ea Y syste t differinma same th g e RPM'S equaus l Al .l length header : 1.75x34.sas 5 with 1.875"x18" intermediates merging Int singloa e 18.5x2" graph shows how the negative waves in outlet. 78-79°F. 8-24-96. Lycoming IO-360 A1B6 firing order:1324 Run at 125'MSL. the blind leg of cylinder #2 return and reduc pressure eth cylinden ei 's1 # r #516 2703 RPM Cy1 # l ——— #517 2476 RPM Cy2 # l header. See the paired arrows. The red #516 2703 RPM Cyl #2 #518 2255 RPM Cyl #1 double-ended arrow show remarke sth - ably early onset of negative pressure at #517 2476 RPM Cyl #1 #518 2255 RPM Cy2 # l thin i s M systemRP w .lo These primary Note the way that the cyl #2 trace shows the l arriva#1' cy wave P se pairf o l Th f arrowso . s header lengths (34.5") seeoptie b mo -t iFigur0 e1 below show that the negative wave in cyl #2 mal for about 2500 RPM, judging by helpd travelan s1 # lowes l bacpressurs cy it r o kt e good pumping gaiscavengd nan e th f eo trace. This gives some excellent scavenginl al t ga blu egraphe tracth n eo . \of these usable RPM'S. Shorter headers would probably better scavenge the higher RPM'S. The Tri-Y's wave timing is primar- ily controlled by the primary header length diametee Th . lengtd e th ran f ho common tailpipe see- en mshif o t e th t tire pressure trace up or down as a unit. In other studies, increasin lengte gth h of the intermediate pipes beyond 18" seeme raiso d t backpressure e eth . Crossover systems and Tri-Y systems are, in some ways, halfway between the independent pipe collece systeth d m-an r systemto . They still exhibit higher opening pressures int4 tha cole o1 n th - lector systems t thebu ,y enjoy larger, longer duration negative waves after thei wavrP e tha independeno nd t pipes. -20 Tri-Y tuning is more critical than the 4 100 120 140 160 1; O 200 220 240 260 280 300 320 340 360 380 400 420 into 1 system as to RPM. Crankshaft Degrees After FirinC gTD 40 JANUARY 1997 Files 432/428. Fast Fourier transform. (FFT) Interference seems to cause the higher identical but the opening pressure at frequencies to diverge even though the RPM's and firing frequencies were nearly identical. EVric e lowes th hO r wa mixtur rfo e 0.6 Cyl #1 Coanda nozzle + case. The FFT's for these runs do show __ 1.75x34.5x2x19.5, 4 into frequency changes EPG'e th t sbu , look Figure 11 22.22671 M z 05RP H remarkably similar. firing #432 __ Collector 89 Hz firing 0.5 #432 BALL JOINT EFFECTS 1.75x34.5x2x19.5, 4 into —— 1 2669 RPM 22.25 Hz collecto" 2 e th n I r test' s (506,507) firing #428 there was a very slight increase in This shows that the Coanda nozzle backpressure whe balna l joins wa t 0.4 raises the pressure energy in the chango usedn t fue,bu n e i l flowM RP , header's reflected wave greatld san y i increasee collectoth n i t i rs waves. or thrust was observed. oo e Coanda'Th s inner cone doubles alss - owa un G ThEP e waveP th f eo the surface area of boundary layer affecte additioe th y 2.25a b d f no " friction, slowing the gases and diameter ball join downstrea" 22 t f mo «0.3 reducing the effective cross-section w collectore oth f optimizen A . d design the collector merge when the total col- (D might need to enlarge the collector's lector lengt 43.75"s hdifferenwa o Tw . t cross-sectional area inside the ball joints were used. One (715) had a O) DC Coanda nozzle to provide for the smooth interna othee lth wald r an l 0.2 extra boundary layer caused by the inner cone. The 4 into 1 system more (717th ed commo)ha n internal shows that the firing frequency concave chamber collectoe Th . r wave, contains most of the energy, with less recorde poina t da downstrea" 4 t f mo in the higher frequencies. the ball joint, also showed essentially 0.1 chango n e from eithebale th l f jointsro . When a megaphone was added to the straight collector t showei , dmarkea d negative wave afte collectoe th r r wave peak. Ball joints can probably be used 0 12 0 10 0 8 0 6 0 4 0 2 140 160 180 for vibration isolation of the collector Frequency, Hz without detuning of the exhaust. FREQUENCY ANALYSIS intbot4 e exhauso1 h th t systed man COL DENGINT VS.HO E Tri-e th Y system Figure n I Se . . e12 The fast Fourier transform shown in both cases, no power gain was evident runo sTw (701,703) were made with depico t Figur y soune th twa ea d 1s 1i and the EPG did not show any striking identical pipes (1.625x28x2.25x21.75) frequencies most prevalent in a given change. The Coanda nozzle did seem except that one was at 67° F OAT and EPG. The 4 into 1 system shows the to give some noise reduction and mel- the other at 80° F OAT. The RPM's, firing frequency. Interestingly, some lowing of the exhaust sound. fuel flows and P wave shape were systeme th f o s pea t multipleka e th f so nearly identical. The 80° F run showed firing frequenc cylindere th f yo speA . - HEADER SIZE slightly lower pressure at EVO (-9" cial type of loudspeaker might Hg. vs. -6.5" Hg.) and overlap (-10" theoreticall usee yb d wit hnoisa e can- optimuFigure e Se Th . e13 m header Hg. vs. -7" Hg.). celling program (destructive inter- siz r thiefo s engin t 250ea 270o 0t 0 Two other runs (500,501) were ference nearlo t ) y eliminat- ex e eth RPM, at sea level appears to be 1.75" made with identical pipes (1.75 x34.5 haust noise by countering each of the diameter. The length of the headers in x2x20), one with 200° F CHT and the main frequencies showFFTe th n .no inta4 systeo1 m seems optimizet da other with 400° F CHT. These showed about 28-36". Longer length probably no significant difference in the EPG. COANDA NOZZLES raises backpressur delayd ean s onsef o t scavenging while shorter length- sre MORE HORSEPOWER The Coanda nozzle is shown in the duc e abiliteth contaio yt nfulla y covee photth n roo page t consistI . f so developed, powerful wave. Bruce Arrigoni extensivs ha o wh , e a megaphone inside whic places hi da experience in dyno race tuning of the solid cone whose taper ratio produces LEAN vs RICH EPG's Subaru engines with Formula Power in no net change in cross-sectional area Concord, California, states thasine th t- throughout the megaphone. The outlet A richer mixture produce lowesa r gle most effective way to increase the Coande th f o a nozzl sharpla s eha - yta EGT and thus a slower wave speed Subaru horsepower output was to pered trailing cone intended to produce than doe leasa n mixture EPG'o Tw . s smoot sharp-edgee hth d transitiof no a low pressure vortex. Theory has it wermanifolf usino n " eru g25 d pres- the exhaust valve's seat bevet cu l that this vortex will reduce backpres- sure and 2500 RPM wherein one used where it blends into the valve's tulip sure and give more horsepower. 15 gph and the other 11 gph. The P portion. This apparently greatl- yim Coande Th a nozzl addes ewa o dt waves and scavenging were nearly prove floe sth w pas valve th t e bott ha

SPORT AVIATION 41 initial opening and during the small Files 428/432: 428 = 1.75x34.5x2x19.5. File 432 has a Coanda megaphone of 2x11x4 as valve openings at overlap. drawn below added ont 19.5e oth " collecto f filro e 428 botn .i h Win runsmp d3 . 8-18-96. Lycoming IO-360 A1B6 firing order: 1324 Run at 125'MSL LENGTH FORMULAE 77°42 B 106.266d 8h M F 620. gp 9RP 1 428 Intake Most simple mathematical formu- 432 Coand 105.267ah M 20.gp B 00RP d calculatinr fo e la ideae gth l lengtr hfo 75°F exhaust pipes fail to account for to 432 Coanda Intake recognize that there is a Doppler phe- The apparent quieting from the Coanda nozzl partls ewa y nomenon occurrin exhausn a n gi t pipe artifact becaus t ei move e dth becaus sonie eth c exhaust wav rids ei - Q outlet further aft of the cabin. ing on the "wind" of the streaming e powe*^Th f o theso r tw e systems seems nearly mass flow of fuel and air. The sonic identical e despitth e wave move t 1500-180sa whils 0fp e Coanda's slightly better scavengin t overlas ga it d pan the mass flow moves at 200-400 fps. lower pressure t EVOea Th . sonie Th c wave thus travels fasteo t r Coanda might sho wpowea r werO gaiEV e f i nmove o dt the tailpipe than doe returnine sth - gre cyclee lateth n ri . flected sonic wave which must "swim upstream" to reach the exhaust valve. Figure 12 Computer programs can address these complexities using wha calles i t d the "method of characteristics". One such program is Curt Leaverton's "Dynomation", available from V.P. Engineering, 5261 NW 114th St., Suite J, Grimes, IA. 50111. Ph. 515- 276-0701

SIZIN PIPEE GTH S -20 0 60 0 57 0 54 0 51 0 48 0 45 0 42 0 39 0 36 0 33 0 30 0 27 0 24 0 21 0 18 0 15 0 12 0 9 0 36 0 Crankshaft Degrees After Firing TDC It mus rememberee tb 0 d 20 thae tth HP engine becomes a 130 HP engine Files 737/738/739/411 Comparing straight stacks of 35.25" using various header diameters at cruise altitudes of 8000-12,000 ft. to the collector equipped system #411. 96°F Lycoming IO-360 A1B6 firing order: 1324 at Optimizatio f exhausno t tunint ga 125' MSL. these altitudes, wit attendane hth - re t 40 The 1.625" header 1.6257 73 " 270M 0RP duce densityr dai e , wilus e l th calr fo l reache highesa r smallef o r diameter header cold -san pressure (clipped) 738 1.75" 2690 RPM showd an sa 7391.875" 2700 RPM ,,; , ; lectors compromisA . e mus foune b t d broader P wave to not rob the engine of its sea level 30 _ than the 1.75" and 411 1.75"x34.5x2.25x19.M 5 273RP 1 climb power stainlesA . s steel multi- 1.875" headers. However t doei , s segmente t nozzle/megaphondje e scavenge well at Figure 13 whose outlet area could be adjusted overlay pb delaying the arrival for altitude coul worthwhile db r efo firse th tf o _ringin g 20 optimizing both low and high altitude wave at R. performance. RECOMMENDATIONS

The 4 into 1 exhaust system (File 411) used on the CAFE testbed Mooney can be reproduced by Sam Davis at Tube Technologies in Coro- mandres a naA C , l bent pipes requir- weldinG TI g theio gin t r exhaust flanges. Alternative designs can be File 411, with Its made in mild steel from "U" bends collector, again and then sent to Sam for duplicating in e b appear o t s 321 stainless steel. Aerospace Welders superio n i reducingr ! in Minneapolis providn ca N eM , very backpressure. -240

can be used to compute the jet thrust available #439 10.4 gph 2195 RPM 0.94 psi 1.5" noz 19.5 gph 2645 RPM 5.25 psi from the exhaust. Several formulae can be used for this, however, they require several assump- Figur4 e1 tions, which are listed in bold below: 30

p P/gRTe= abs =M (C/1728 (RPM/60)x p)x sT|ix v 1/2 V= M/peA = (2q/pe) where 20 ave = velocit s V ga . exitt ya , ft/sec M = mass flow rate in slugs/sec A = 7r(tailpipe diameter-2(w))2/4 w = wall thickness 10

q = 1/2 x pe x Vp2 = pitot pressure at CD tailpipe th e exit. £I pea= p kV velocity derived fro. mq OL

V = Vpx.817** Thrust in pounds = (WxV)/g = MxV Pounds of Jet Thrust = (WxV)/g Assuming a 0.14" boundary layer and 0.049" wall thickness outlee th , t area, A rough check can be made using the wher == e32.17g 4 ft/sec2 s " collectoit 2 d e oan th f 2.0n s i i r q 6s W = slugs/sec ((gph)x6x12)/3600.= . fuer ai l d ratioan h : gp thrust force calculate 10-1o st 2 lb.t a , Vnozzl= velocits ega y -10 20.d an 2 268M gph e 1.5RP 5 Th . " tex e computationr fo tSe f so lb/se= W (gphx6xl2)/360c= 0 nozzle gives 17-21 lb of thrust, but thrust. probably has reduced horsepower **See reference 19. The .817 is derived from -— -du higheeo t r backpressure. complea x analysi thesf so e pipes' Reynold's numbers and boundary layer thickness. Psi = 0.0023769 p = exhaust gas density Crankshaft Degrees After FirinC gTD e 'Fuel = 6 lb/gal and air/fuel ratio is 11, giving the 6 and 12 above. q = dynamic pressure, psf T| 0.9v= 5volumetri= c efficiency ball joint r straisfo n relief placed both into 1 exhaust system usually pro- pip= eA exit excludin, areaft q s , g b.l. moute collectoe ath tth f ho r entrs ya duces an increase in the negative b.l. = boundary layer well as about halfway down the pressure achieved at the exhaust R54.= 0exhaus= constans ga t t header jointe Th s s must alwaye sb valve, but at a substantial penalty g = accel of gravity = 32.174 ft/sec2 secured with redundant spanning in noise. 12. 14. Tabs 1 ° temperature bolts, compression spring cotted san r = 760 R = exit 4. The use of swiveling ball joints P = 2116 psf = sea level pressure pinned castle nuts. on the collector of a 4 into 1 ex- 6 = pounds per gallon of avgas haust syste neglibla s mha e effect 12 = 1 + air fuel ratio of 11 to 1 (rich) CONCLUSION Sprovided - an im G n sa EP e onth 360 second0= hour spe r 1. Substantial negative pressure portant vibration-isolation benefit 1728 = cubic inches per cubic foot to the system. C = 180 cubic inches = effective full generatee waveb n sca tunen di d time engine displacement aircraft exhaust systems and the 5. The optimization of pipe timin f theio g r suctioe b n nca geometry for the crossover, Tri-Y Using these formulae and 20.2 gph at 2685 arrange improvo t s a o ds e engine witM hRP 2.2 i outle5ps t pressure (Figure and 4 into 1 exhaust systems can be power. Such improvement should found by study of the EPG. 14)peaa , k exhaus t thrustje abouf o t t 10-12 produce more power, better effi- pounds is found at wide open throttle with a ciency and a cleaner combustion 6. Fast Fourier transforms, de- 2" tailpipe. The 1.5" tailpipe nozzle gives 17- chamber. rived from the EPG, could facilitate 21 pounds of thrust. Simultaneous solution of the development of an electronic, these equations can be used to find the un- 2. The 4 into 1 collector exhaust active noise-cancelling muffler. known values. systems appea offeo rt bese rth t Aircraft exhaust systems theiy b , r ambiene Th t pressure will determine eth combinatio openinw lo f no g pres- limited RPM range, are particu- exhaus densits tga y upon exithud exie tan s th t sure, some pumping gain and good larly well-suited to such a muffler. velocity. The low ambient pressure at 10,000- scavenging, though the crossover 14,000' would give an increase in exhaust and Tri-Y system alsn sca o obtain Coande Th . 7 at nozzlno d edi thrust, especially with turbocharging, which good scavenging durin overe gth - produce a noticeable increase in maintain highesa r exhaust mass flot wa lap stroke. power. Fabricatio durabilitd nan y 6 those altitudes e additioTh . suitabla 3 f no e problems make this nozzle of lim- megaphone to the collector of a 4 ited attractiveness.8 SPORT AVIATION 43 8. Exhaust jet thrust was measured haust-Stack Nozzle Are Shapd aInr an -eFo and calculated for several exit sizes, dividual Cylinder Exhaust Gas Jet IMPORTANT NOTICE RPM's and fuel flows. It can produce Propulsion System. NACA Report No. Every effort has been made to significant thrust at high power set- 765,1943. 6 obtai mose nth t accurate informa- ; tings, especiall cruisint ya g altitudes. 7. Smith, Philip H., and Morrison, John tion possible dat e presentee a.Th ar d 9. The stock camshaft used in Scientifie anTh , C. c Desig Exhausf no d an t measures a subjece - ar er d o dt an t aircraft engine is typically optimized Intake Systems, Third Edition, Robert rors fro mvarieta sourcesyof i Any . for reliability and tractability and is Bentley, Inc, June 1978. reproduction, sale, republicationr o , other use of the whole or any part of not optimized for the tuned exhaust 8. Blair, Gordon P, Design and Simula- this report without the consent of •; systems tested here. To fully realize tio Two-Strokf no e Engines, Societf yo Experimentae th l Aircraft Associa' - potentiae th l benefit tunef so d exhaust Automotive Engineers, Inc, 1996. systems for the aircraft engine, the CAFe th tiod En an Foundatios ni camshaft timing must be suitably al- 9. Harralson, Joseph, Design of Racing strictly prohibited. Reprints of this tere makiny db g exhaust valve closure and High Performance Engines PT-53, So- obtainee reporb y writinma y t db g occur later and the overlap period of ciety of Automotive Engineers, Inc, 1995. to: Sport Aviation, EAA Aviation ; 9 longer duratio highed nan r lift. Many 10. Seeley, Brien Technologe Th , f yo Center, P. O. Box 3086, Oshkosh, of the scavenging systems here do not CAFE Flight Testing. Sport Aviation Vol. WI, 54903-3086. J exhibit as much effect upon the intake 43, No. 5, pg. 51, May, 1994. manifold pressure durin overlae gth p ACKNOWLEDGEMENTS as might occur if the camshaft had 11. Goldstein, Norton Coande ,Th - aEf greater valve overlap. Magazined Ro fectt Ho , , December 1962. This work was supported in part by 10. The effect of cylinder head tem- 12. Creagh, Joh . R.nW , Hartmann, FAA Research Grant Number 95-G- perature, mixture (EGT), and outside Melvin J, and Arthur Jr, W. Lewis, An In- CAFe 037Th . E Foundation gratefully air temperature on the EPG were vestigation of Valve-Overlap Scavenging acknowledge assistance sth f Anneo e studie found dan produco dt e mini- Ove Widra e Rang Inlef Exhauseo d an t t Seeley, Mary Vetter, Lyle Powell, Jim mal changes. Pressures, NACA Technical Note No. Griswold ChapteA EA , r 124 Sante th , a 1475, November, 1947. Rosa Airport Tower, the Rengstorf 11. Further study should include family, Hartzell Propellers, and several the correlation of climb and cruise 13. Tabaczynski, Rodne , yEffecJ f o t helpful peopl engineerine th n ei - gde airspeeds with EPG's take flightn i s Inlet and Exhaust System Design on En- partmen Avco-Lycomingt ta . controlle power dfo r settin aird gan - gine Performance PapeE SA ,r 821577, craft weight. These shoul pere db - 1982. SPONSORS formed using exhaust jet nozzles, 14. Jameson, Renee T, and Hodgins, Experimental Aircraft Association megaphones, altered ignition timing Patrick A, Improvement of the Torque Federal Aviation Administration and other variations• . Characteristics of a Small, High-Speed Aircraft Spruce & Specialty Co. Engine Through the Design of Helmholtz- Aerospace Welding BIBLIOGRAPHY Tuned Manifolding, SAE Paper 900680, Minneapolis, Inc. 1. Seeley, Brien, and Vetter, Ed, The March 1990. , , , ,=,,„• Fluke Corporation Aircrafd an G t EP Exhaust Systems. Sport B & C Specialty Company TuninBikesg m 15Bi g Ra .r Bi ,gfo Aviation, Vol. 45, No. 1, pg. 39, January, Engineered Software: Bike magazine, pg. 52-57, November, 1996. "WildTools/PowerCadd" 1970. Bourns & Son Signs 2. Seeley, Brien, and Vetter, Ed, EPG. Johnny Franklin's Muffler Shop 16. Ewing, William H, and Nemoto, Spor, t Aviation48 . pg , 3 . , VolNo , .45 Sam Davis at Tube Technologies Hiroshi, A Computer Simulation Approach March, 1996. Factory Pipes Exhauso t t System Noise Attenuation, AeroLogic's Personal Skunk Works SAE Paper 900392, March 2,1990. 3. Seeley, Brien, and Vetter, Ed, EPG Software III. Spor, 80 t. Aviationpg , 5 . No , , Vol45 . May, 1996 17. Yagi, Shizuo, Ishizuya, Akira, and Fujii, Isao, Researc Developmend han f to COMP AKA TIVE AIRCRAFT . Lord4 , Alber , HeinickeM. t , Orville High-Speed, High Performance, Small FLIGHT EFFICIENCY, INC. H., and Stricker, Edward G.: Effect of Ex- Displacement Honda Engines, SAE Paper The CAFE Foundation: haust Pressure on Knock-Limited 700122, January 16,1970. .., ,• : Profitn Volunteerl No Al ,A , Performanc Air-Coolen a f eo d Aircraft-En- Tax-exempt Educational Foundation 18. Hosomi, Mikiya, Ogawao, Sumio, gine Cylinder. NACA Technical Note No. 4370 Raymonde Way, Santa Rosa, Imagawa, Toshiyuki, and Hokazono, 1617, June 1948. CA 95404. Yuichi, Developmen Exhausf o t t Manifold FAX 544-2734. . He5 y wood, John B.: Internal Com- Muffler PapeE SA , r 930625, Marc, h5 Aircraft Performance Evaluation Center: bustion Engine Fundamentals, 1988, 1993. 707-/45-CAFE (hangar, message) McGraw-Hill, Inc. 19. Schlicting, Herman, Boundary America Online: 6. Pinkel, Benjamin, Turner L. Richard, Layer Theory 402-403g p , . Pergamon [email protected] Voss, Fred, and Humble, Leroy V.: Ex- Press• ,- 1955. Internet: [email protected] 44 JANUARY 1997 READERS7 RESPONSE

AIRCRAFT EXHAUST to 40 minutes, about as far north as Portland, would use a single tailpipe with mega- SYSTEMS IV Oregon or Bangor, Maine, so folks further phone, containing a coanda nozzle south may wish to adjust the value for g. Mi- adjustable fore and aft. The aft position would give a large tailpipe opening for Dear Editor: ami, for example, would be about 32.12, and minimum back pressure, and the forward "Aircraft Exhaust Systems IV" (January Memphis would be 32.1435. With the pressure, density and tempera- position would give a smaller tailpipe 1997) by Brien Seeley, et al, is one of the opening for increased jet thrust at altitude. best articles I've seen in your magazine. ture ratios and g calculated, the suggested equation modifications reduce the spread This is really good work, and they should be * Jim Cunningham in calculated thrust ranges mentioned in Cunningham Engineering Associates complimented for it. I'd like to make a few the article and would allow the approxi- 185 Center St., Ste. 210 supplemental remarks regarding calcula- mate thrust to be computed for any power Collierville, TN 38017 tions for the sidebar on exhaust jet thrust. setting, altitude, temperature and latitude. 901/853-8377 The equation given for M is actually only For an O-320 or O-360 at cruise, jet the induction air mass. Before it can be used ** For a PA-28-180 at 2400 Ibs. gross in the Thrust equation (T=M*V), the Fuel thrust can be on the order of 5-10% of the total thrust**, so it potentially is an effec- weight in level flight at: TAS 135 mph, mass must be added. If the Induction Air tive contributor to both speed and range. 3000' PAlt, OAT 65 deg F, MP 24"Hg, 2400 mass is called Ai rather than M, and the Fuel One simple but inefficient operational pos- rpm, prop dia. 6'2", & 0.81 prop eff. Then mass is computed as F=gph*6/(g*3600), sibility would be a single muffler with twin the plane is flying at a power setting of then M=Ai+F, and the Mixture Ratio is Ai/F tailpipes pointing aft, one with a flap gate, 76.67% (138 Hp) and Thrust = Drag = 310 rather than the estimated 11 to 1 (rich) given one without. Flap gate open, both pipes pounds. If the tailpipe is turned back and in the examples. And also, in the W equa- functioning***, you get minimal back sized for 15 pounds thrust, the thrust increase tion, (1+Ai/F) would replace the number 12. pressure and maximum horsepower near is 4.8% less pumping losses. This is worth 2- Of course, this makes W/g exactly equal to sea level, but minimal jet thrust. Flap gate 3 mph or the equivalent of removing twice M (which it should be), and it means the vol- sealed, the pipe remaining open would give the drag of the horizontal stabilizer. umetric efficiency is the only assumption in this part of the calculations. significant thrust at altitude with back pres- *** Exhaust tuned and sized as per This modification will make a signifi- sure still minimal. A second possibility CAFE recommendation. ^ cant difference in part throttle operations like Test 440 where an assumption of 0.95 for volumetric efficiency would show an KIS SPORT AIRCRAFT inconsistent and unlikely mixture ratio of about 26:1 rather than the 11:1 specified. It • FASTER BUILDING • BETTER VALUE is apparent that the severe throttle plate re- striction on Run 440 means the volumetric efficiency was actually about 0.4 and the mixture was indeed on the order of I 1:1. This makes a big difference in the thrust output calculations. The same is true of Test 439, where the volumetric efficiency was on the order of 0.6 or 0.65. Also, the q and velocity equations given in the sidebar are for incompressible flow (low Mach Number) even though on at least one of the tests (#505) the Mach number is KIS Cruiser TR-4 4-Place Cruise Speed-75% 185 mph up near 0.91, and the Mach Number is high • VERSATILE Stall Speed 55 mph enough to matter on several of the other runs. Useful Load 1100 pds Using the compressible versions of the equa- Powerplants 160-210 hp tions changes the computed thrusts by several Build Time 1200-1500 hrs percent. In addition, the computed thrust is KIS TR-1 2-Placc very sensitive to the assumed thickness of the Cruise Speed-75% 185 mph boundary layer. For example, in Test 505, Stall Speed 55 mph changing the boundary layer assumption Useful Load 610 pds from 0.140" to 0.125" changes the thrust by Powerplants 80-125 hp several pounds. Since the boundary layer Build Time 800-1000 hrs thickness is not constant from run to run and its calculation is straightforward, I recom- KIS two and four place all composite kitbuilt aircraft are designed to Keep It mend a computed approximation for each Simple. Pre-molded high temperature composite and prewelded metal parts run. The same statement holds for the make KIS assembly easy. Both KIS aircraft will perform well with many weighted radial velocity distribution (it isn't different engine options. always 0.817). Although it's a minor factor, INFO PACK $15 • VIDEO $15 • OVERSEAS ADD 50%- the acceleration of gravity ranges from TRI-R Technologies, Inc., 1114 E. 5th St. Oxnard, CA 93030 USA 32.0878 fps 2 at the equator to 32.2577 at the Phone (805) 385-3680 • Fax (805) 483-8366 • VISA/M.C. Accepted poles. The value of 32.174 used in the article implies a test site Latitude of 45 degrees 20 For information, use SPORT AVIATION'S Reader Service Cord SPORT AVIATION 103