TRIAVIATHONTROPHY

CAFEFOUNDATION

PRESIDENTBrien Seeley

VICEPRESIDENT

Larry Ford

TREASURERCJ. Stephens

SECRETARYDaniel Waymaii

TEST PILOTC.J. Stephens

DIRECTORSCrandon Elmer

Otis HoltJack Norris

Gris HawkinsStephen Williams

Ed Vctter

CHALLENGETROPHY

AIRCRAFT PERFORMANCE REPORTSponsored and Funded by the Exper imenta l A i r c r a f t Association

and the Federal Aviation Administration

Aircraft ExhaustSystems IV

BY BRIEN A. SEELEY, ED VETTER AND THE CAFE BOARD

T he goal of this report isto improve the power,efficiency and reliability

of aircraft exhaust systems. Thereport summarizes the results ofa 16 month long study. Many ofthe systems tested here are sim-ilar to ones popularly used inlight aircraft. The tests include4 into 1 collector systems, 4into 2 "crossover" systems,"Tri-Y" systems and indepen-dent exhaust stacks. Additionalaspects of exhaust design in thisstudy are:

Intake wavesWave speedMegaphone effectsRPM effectsExhaust jet thrustCrossover/Tri-Y reflectionsFrequency analysis (FFT's)Header sizeCollector sizeCoanda nozzlesBends in the pipeEGT and CHT effectsBall joint effects

Over 350 separate ExhaustPressure Graph (EPG) record-ings were made using theLycoming IO-360 A1B6 enginein the CAFE test-bed MooneyM20E. All of these were madeat 125' MSL as static groundengine runs of approximately15 seconds duration.

CAFE, EAA and the FAAare grateful to the followingmajor contributors to this study:Aerospace Welders of Min-neapolis for the high qualityexhaust merges and ball joints,George Johnston of EAA Chap-ter 124 for the lathe-machinedmodel of the Coanda nozzle,Sam Davis at Tube Technolo-gies in Corona, California forthe stainless steel exhaust sys-tem derived from thes tests, andBill Cannam, a certified welderfrom EAA Chapter 124, for themajor effort to assemble thestainless steel exhaust system.Curt Leaverton, Jack Norris,Andy Bauer and Steve Williamseach contributed professionalscientific analysis of the EPG's.

ABBREVIATIONSEVO = exhaust valve openingEVC = exhaust valve closureIVO = intake valve openingFVC = intake valve closureTDC = top dead centerBDC = bottom dead centerW.O.T. = wide open throttlegph = gallons per hourdB = decibels, slow A scaleFFT = fast Fourier transformcyl = cylindercoll. = collectormsec. = millisecondsHz. = Hertz or cycles per sec"Hg. = inches of MercuryO.D. = outside diameter

34 JANUARY 1997

THE BASIC EPGA Review

Figure 1 shows a basic EPG. It wasrecorded on a well-tuned 4 into 1 col-lector exhaust system which wil l behereafter referred to as "File 411". Fig-ure 1 shows features which arcessential for understanding the othergraphs in th is report. The "X" axis,along the bottom of the graph, showsthe degrees of crankshaft rotation be-ginning at top dead center (TDC) ofthe firing stroke for cylinder #1. Thevertical "Y" axis shows the pressuremeasured in the pipe in inches of Hg.Since these runs were made at near sealevel, the zero pressure level representsambient pressure of about 29.92" Hg.

The typical EPG shows a steeplyrising "P" wave of exhaust pressure,shown in red, which starts upward atthe point of exhaust va lve opening(EVO). The tall P wave typically fallsto below zero (ambient) pressure laterin the exhaust cycle.

The intake pressure is shown inblue. There is a black vertical dottedl ine at BOC after the intake stroke,where the piston's descent ceases.

The amount of valve lift of the ex-haust and intake valves is shown at thebottom of the graph. At overlap TDC,both valves are open for a brief interval.

The EPG often shows addi t ionalwaves which come from reflections,turbulence and, in collector-equippedsystems, the firings of the other cylin-ders (cross-talk). These are labeled bytheir cylinder of origin as the R wavesin Figure 1.

The C waves arc those measured inthe collector, the common pipe intowhich arc merged the individual head-ers. Each cylinder produces a separateC wave. The time T between the rise ofthe P wave and the rise of the attendantC wave is very short and can be usedto calculate the velocity of the wave.

The "blowdown" cycle is definedas the period from EVO to firing BDC,and is labeled "B". It is during this in-terval that the steep rise of the P waveis seen, as the cylinder discharges or'blows down' through the exhaustvalve and the in-cyl inder pressurerapidly falls. Positive in-cylinder pres-sure during blowdown is still doingsome useful work by pushing down-ward on the piston.

Overlap TDC is a very importantinterval. When both the exhaust and

Files 411 Basic EPG: 4 into 1 as 1.75x34.5x2.25x19.5 equal length headers. 29.5" M.P.,2731 RPM 20.4 gph 86"F. 8-18-96. 125'MSL. Lycoming 10-360 A1B6 firing order: 1324See text for explanation of P, C, S and R waves shown below

411 Cyl#1 411 Cyl #Z —— 411 Intake --- 411 Collector Wave

Both cylinders #1 and #2 show low openingpressures at EVO and very low pressureduring overlap TDC. Some scavenging ofthe intake trace appears to occur. The Pwave (red) goes negative early in theexhaust cycle.

O

Figure 1Aft looking viewof collector entry

— Exhaust Valve

Intake Valve

30 60 90 120 150 180 210 240 270 300 330 360 390 420 450 480 510 540 570 600Crankshaft Degrees After Firing TDC

intake valves are open, the pressures inthe exhaust pipe, combustion chamberand intake tract can all influence oneanother. How much influence dependsupon the valve lift during overlap andhow long both valves remain open.

During overlap TDC, the suction in atuned exhaust's header can help emptythe combustion chamber of its burnt gasresidues. This effect is called "scaveng-ing". The exhaust suction may evenenhance the combustion chamber's fill-ing from the intake valve, thus im-proving volumetr ic efficiency andhorsepower. With sufficiently longoverlap intervals, it is possible for thesuction to pull some cool intake chargeacross the hot exhaust valve, coolingthe valve face, stem, seat and guide.Such cooling comes at a price, which isthat raw fuel is being wasted out the ex-haust pipe. Higher compression pistonsshould scavenge better due to theirsmaller combustion chamber volume.

Note that in Figure 1, the intakepressure is greater than the exhaustpressure at overlap TDC. Such a pres-sure gradient w i l l encourage sca-venging. At part throttle, the intakepressure would be much lower, and

unfavorable reverse flow could occurat overlap. This is one argument for us-ing wide open th ro t t l e (W.O.T.)whenever possible in high altitudecruise flight.

PUMPING GAS

Normally, engine designers try toplace EVO about 40-75° prior to firingBDC so that the peak of the very highin-cylinder pressure can be dissipatedduring blowdown, before BDC. Atuned exhaust system, with a very lowopening pressure at EVO, can assist inevacuating the cylinder quickly, andcan thus allow EVO to be delayed untillater in the cycle. The later EVO al-lows the positive in-cylinder pressureto do more work pushing the pistondownward prior to EVO. Thus, a tunedexhaust system works best if the tim-ing of EVO is delayed to takeadvantage of the tuning.

After blowdown in the exhauststroke, the piston begins to rise fromBDC. A rising piston pushing against ahigh in-cylinder pressure causes a lossof power known as a "pumping loss".Instead, the rising piston should be

SPORT AVIATION 35

pulled upward by a negative pressurein the cylinder, thus producing a"pumping gain". Suction in a tuned ex-haust system can produce such apumping gain in mid to late exhauststroke. This is shown in Figure 1 wherethe exhaust pressure goes negative at260° of crank angle, which is 80° afterBDC. The earlier in the exhaust cyclethat the P wave subsides and goes neg-ative or below the ambient (zero)pressure, the more pumping gain canoccur, making for greater horsepower.

Thus, an ideal exhaust systemshould produce a highly negativepressure at the exhaust valve at bothEVO and again as soon as possibleafter dissipating the P wave. Thisnegative pressure should be made topersist throughout the overlapstroke so that favorable scavengingcan occur.

INTAKE PULSATIONSThe piston's descent during each

intake stroke exerts strong suction onthe intake pipe runner connecting thecarburetor to the cylinder. If all of theintake runners attach to a commonplenum, as in the Lycoming engines,the suction will affect all of those run-ners. The suction causes a flow to beinitiated in one direction which isabruptly stopped when the intakevalve closes. The flow stoppage cre-ates a reflecting wave which againaffects all of the runners. This leads tointake pulsations.

The intake pulsations on the Ly-coming IO-360 A1B6 engine aresizable and can be seen in Figure 1.These pulsations can show how muchscavenging effect might be expected,and the character of the cylinder fill-ing. The latter can serve as a guide tothe relative volumetric efficiency ofthe engine.

The W.O.T. intake pulse can reachas high as 6-7" Hg. above atmosphericpressure, as seen at "+S" in Figure 1.This effect thus gives an instantaneousmanifold pressure of about 37" Hg.,and, if timed correctly, can act some-what like supercharging. Ideally, thelowest point in the intake pulsationtrough should be timed to occur at "-S"or about 60° after overlap TDC. Thiswill tend to assure that the next posi-tive pulse will arrive just prior to IVC,enhancing flow through the intakevalve just as cylinder filling ends.36 JANUARY 1997

Files 411/413 The effect of a megaphone: 4 into 1 as 1.75x34.5x2.25x19.5 equal lengthheaders. File 413 has a 17Lx2.25x4" megaphone added to 411. Both at 29.5" M.P.(W.O.T.), 86°F. 8-18-96. Lycoming IO-360 A1B6 firing orderl324 Run at 125'MSL.

—— #411 2731 RPM 20.4 gph 106.9 dB. no meg

#411 intake pressure, no meg •

—— #413 2737 RPM 20.6 gph 107.8 dB, with meg

#413 intake pressure, with meg

Figure 2

The megaphone can increase power bylowering the opening pressure at EVO, andscavenging more at overlap. The noise levelis significantly higher, however.

cylinderfillingconcludes

-2030 60 90 120 150 180 210 240 270 300 330 360 390 420 450 480 510 540 570 600

Crankshaft Degrees After Firing TDC

Yagi et al17 have written an excel-lent paper on using induction systempulsations to force feed the engine'scylinder during the intake stroke.

We did not observe any pressurepulses in the intake waves due to thepropeller blade sweeping past the aircleaner intake. However, these testshad the air cleaner located 11-12" aftof the propeller disc. On those cowl-ings with a very far forward aircleaner intake, the EPG may be ableto detect whether the prop is produc-ing a pulse into the air cleaner at justthe right moment during the intakecycle.

WAVE SPEED

The EPG can show the averagespeed of a wave traveling through thepipe. The wave speeds observed actu-ally represent the sum of the averagesonic wave speed and the averagemass flow velocity.

A test using sensors 21.5" apart ona 1.625" primary header showed anaverage wave speed of 1751 fps.

In Figure 1, the time interval "T"

represents the time for the 2731 RPMP wave to reach the collector tap fromthe top of the header, a distance of47.0". This computes to about 1604fps average speed. This slower speedsuggests that some slowing occurs asthe header wave enters the collector.

File 412, at 2507 RPM, showed anaverage wave speed of 1508 fps, a7.5% reduction from an 8% reductionin RPM. The reduced speed is due toa lower EGT and the slower averagepiston speed which gives a slowermass flow.

The exhaust gas expands and coolsas it goes down the pipe, and the wavevelocity varies directly with thesquare root of the ratio of the absoluteexhaust gas temperatures.

MEGAPHONE EFFECTS

Figure 2 shows that a megaphoneadded to file 411 produced a loweringof the opening pressure at EVO andbetter scavenging at the expense ofmore noise. A megaphone was lateradded to a Tri-Y system and showedminimal influence on the EPG.

THE EPG TEST METHODThe EPG pressure sensor was con-

nected to a 9" long copper tube of0.125" O.D. f lush-mounted to theheader pipe's inner wall. The mountingwas at a point 1.25" downstream of thecylinder head flange. The signals wereprocessed by the Vetter Sensor Acqui-sition Module and Digital AcquisitionDevice. Sensors were calibrated using awater manometer.

A new amplifier was used for th isstudy. I t s faster response t ime andhigher resolution provided a much bet-ter picture of the EPG relative to thosein previous reports. 1.2.3

The intake pressure recordings weremade 1.5" upstream of the intake valvethrough the fuel injector port in the Ly-coming cylinder head.

RPM, noise level, static thrust inpounds, fuel flow, wind incident to thepropeller and manifold pressure wererecorded manua l ly . Var ia t ions in theRPM. EGT, CHT and mixture were usedon several runs to study their effects.

In all of the EPG's shown here, thetiming of the waves with respect to thecrankshaft degrees has been shifted tothe lett (earlier) by 1.25 milliseconds tocompensate for a) the 12 inch distancewhich separates the pressure sensor andthe exhaust valve face (1.0 mi l l i sec-ond), b) the electronic rise time of thepressure sensor (0.15 mil l iseconds)and e) the amplifier delay (0.10 mil-liseconds). This places the wave timingat i ts correct phas ing w i t h the va lveopening cycles.

Fast Fourier t ransforms (FFT 's )were made on each of the runs to lookat the sonic frequencies which had thegreatest energy content . A n a l y z i n gthese transforms exceeds the scope ofth is report. See the b ib l iography forseveral references on wave theory.

Noise levels were taken from thearea between the front seats of the air-craft with the pilot's side vent windowopen using the A scale slow se t t ing .Noise was reduced when the ta i lp ipeexit was moved aftward relative to thenoise meter, as occurred w i t h thelongest tailpipes.

Peak RPM and fuel flow generallycorrelated wi th the thrust values andwere used as a rough guide to powero u t p u t . The anemometer showed achange in local wind speed and direc-t i o n as the propeller 's flow fieldreached full strength at maximum staticRPM. This flow was allowed to equili-brate before the RPM and fuel flowreadings were taken.

Files 502/422/510: Varied collector diameters. All use the same 4 nto 1, equal lengthheaders of: 1.75x34.5 with -30" collector length. 73-82°F. 8-24-96. Lycoming IO-360A1B6 firing order:1324 Run at 125'MSL

2x29" C 19.7 gph 2635 RPM #502

#502 Intake

2.25x30" C 20.3 gph 2687 RPM #422

#422 Intake

2.5x30" C 19.9 gph 2669 RPM #510

#510 Intake

The 2.25" collectorseems optimal.

1

4X2Aftward view ofcollector entrygeometry bycylinder*

Multiple small waves in#502 maybe due to multiple slipjoints used.

Exhaust Valve

— Intake Valve-20

60 90 120 150 180 210 240 270 300 330 360 390 420 450 480 510 540 570 600

Crankshaft Degrees After Firing TDCFiles 411, 415, 419, 422, and 424 to compare different collector lengths. All use the same 4into 1, equal length headers as: 1.75x34.5 with a 2.25" diameter collector. 78-86°F. 8-18-96. Lycoming 10-360 A1B6 firing order: 1324 Run at 125'MSL.40

30

20

EVO, blowdownand overlappressures allchange as collectorlength is altered.Fuel flows andRPM'S, suggest the19.5" or29.5"collector asoptimal. Intakesshow only minor |changes due tolimited valve

.j overlap.

————— 49.5" 20.2 gph 2666 RPM #415

• • • • • #415 Intake

_ . _ . _ . _ 29.5" 20.3 gph 2687 RPM #422

—.- - . . - #422Intake

————— 19.5" 20.4 gph 2731 RPM #411

- #411 Intake

10" 20.2 gph 2700 RPM #424

#424 Intake

O 30 60 90 120 150 180 210 240 270 300 330 360 390 420 450 480 510 540 570 600Crankshaft Degrees After Firing TDC

SPORT AVIATION 37

BENDS IN THE PIPEFile 41 I's header pipe bends were

as follows:Cyl#l: 25°+ 90°+170° = 285°Cyl #2: 85° + 90° + 90° = 265°Cyl #3: 35° + 170° +180° = 385°Cyl #4: 80°+ 70°+ 20° =170°

Individual testing of these separatecylinders did not show any significantchanges in their EPG waveforms. SeeFigure 1, cylinders #1 and #2.

Many aircraft use a downward bendin the tailpipe to keep exhaust soot offthe aircraft's belly. Keeping collectorlength constant, files 502 (a straight2x29" collector), 503 (2x29" with a90° bend at the exit), and 504 (1.5"nozzle on a straight 2x29" collector)were tested at W.O.T. The resultswere EVO opening pressures of-5.0, -4.0 and +3.0, respectively with overlappressures of-10.0, -10.0, and -4.0, re-spectively. The P wave widthremained the same.

File 503, with a 90° downwardbend of the collector at the exit, causedan insignificant increase in backpres-sure. The nozzle did impose a sig-nificant backpressure penalty.

COLLECTOR SIZESee Figure 3. These tests repeat-

edly showed that, for this particularengine, the 2.25" diameter collectorwas best for optimizing exhaust back-pressure at sea level. A 2.125"diameter collector would probablygive a good compromise betweenclimb power and high altitude jetthrust.

See Figure 4 and 6. Collectorlength appeared to optimize at 20-30".It must be long enough to developsome continuum of flow and fullycontain each pulse. r

COLLECTOR EFFECTS

See Figure 5. The addition of a col-lector to 4 separate independent pipesconsistently caused the entire EPG toshift to lower, more negative pres-sures. Some suspect that this effectmay be caused by the more continuousmass flow in the collector exerting aprolonged vacuum effect upon all ofthe headers. A suitable collector wasone with about 50-90% greater crosssectional area than each individual38 JANUARY 1997

Files 723/724/725 RPM effects: All of these headers are 1.625x28" with no collector except file 411which is 1.75x34.5x2.25x19.5. All at 97°F except 411 at 86°F. Lycoming IO-360 A1B6 firingorder:1324at125'MSL.

6" stub Cyl #4 ©2670 RPM #727

- • - • • 28" Cyl #3® 2710 RPM #723

——— 45.5" Cyl #3 @ 2670 RPM #727

- - - 68" Cyl #3 @ 2700 RPM #728

——— 34.5" + collector @ 2731 RPM #411The 45.5" and 68" long headers eachhave too broad a P wave that delaystheir negative wave. All of theindependent headers (no collector)show characteristic "ringing" wavesfollowing the P wave. The 6" stub istoo short to contain a fully developedP wave. A collector equipped system(#411) shows much improved tuning.

Figure 5

-20O 30 90 120 150 180 210 240 270 300 330 360 390 420 450 480 510 540 570 600 630 660

;., ,-.-. Crankshaft Degrees After Firing TDC

Files 419/420/421/412/425 RPM and collector length effects: All headers are 1.75x34.5". Files 419,420, and 421 used a 2.25x40" collector. File 412 used a 2.25x19.5" and file 425 used a 2.25x10"collector. All at 78-86°F. Lycoming IO-360A1B6 firing order: 1324 at 125'MSL..40

2672 RPM, 20.0 gph, 105.4 dB, #419

2501 RPM, 15.2 gph, 103.2 dB, #420

2507 RPM, 106.3 dB, 19.5" coll., #412

2504 RPM, 14.4 gph, 105.8 dB, 10" coll., #425

— — — • 2289 RPM, 11.3 gph, 100.2 dB, #421

30

20

O)

OT<D

f 10

At 2500 RPM,shortening thecollector haslittle effect onboth the Pwave timingand the intakewaves. The10" collectorshows higherpressure atEVO andduringblowdown andoverlap. LowerRPM givessmaller Pwaves whichgo negativeearlier in theexhaust cycle.

-20100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400

Crankshaft Degrees After Firing TDC

Files 729/730/731/732 Crossover systems of various blind leg lengths. Cyl #1 has a1.625x28" primary header except in 732 where it is 45.5" long. The blind offtake is 2.5"downstream of #1 cyl head. 97°F Lycoming 10-360 A1B6 firing order: 1324 at 125' MSL.Note how blind end pressures are reflected back into the header to alter timing.

— 729 cyl #1 28" with 67" blind

730 cyl #1 28" with 51 .5" blind

——— 731 cyl #1 28" with 36.5" blind

- • - 732 cyl #1 45.5" with 36.5" blind

729 67" blind end

731 36.5" blind end

731 has the best combinationof low pressure during theexhaust cycle and goodscavenge at overlap. None ofthese systems shows the lowopening pressure at EVO orthe mid cycle pumping gainsfound with the 4 into 1 collectorsystems.

Figure 7

-20O 30 90 120 150 180 210 240 270 300 330 360 390 420 460 480 510 540 570 600

Crankshaft Degrees After Firing TDC

Files 511/516 to compare crossover 4 into 2 versus a Tri-Y system. All use the same equallength headers as: 1.75x34.5 with 511 having 2 each 1.875"x18" tailpipes and the Tri-Yhaving 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

#511 4 into 2, Cyl #1

#511 Cyl #2, 2660 RPM 20.0 gph

-.-.-•- #516 Tri-Y, Cyl#1

#516 Cyl #2, 2703 RPM 20.5 gph

Figure 8These two systems show that the mainbehavior of the system is related to thecrossover leg length, rather than theaddition of the Tri-Y's tailpipe-collector.However, the Tri-Y's collector does seem toshift the entire waveform toward lowerpressures and give higher RPM and fuelflow Both systems clearly show the P wavereflected to cyl #2 at X, and that reflection'sreturn into the cylinder #1 trace at R.

8 i.

-20

-360-240-120

OO 30 90 120 150 180 210 240 270 300 330 360 390 420 450 480 510 540 570 600

Crankshaft Degrees After Firing TDC

header and with a length of at least 18"or so. See "length formulae" below.

In Figure 5, the small waves whichoccur after the P wave in the indepen-dent pipes are called 'rings', as in adoorbell ringing. The negative por-tions of these rings are of such shortduration that they require the pipe de-signer to choose between positioningthem early in the cycle to obtainpumping gains or late in the cycle toscavenge at overlap. The collectorsystem has such a long duration nega-tive wave after the P wave, that itserves both purposes, i.e., gives pump-ing gain as well as scavenging atoverlap.

CROSSOVERS AND TRI-Y

A crossover exhaust joins the head-ers of cylinders whose firings occur180 crankshaft degrees apart. The Pwave of one cylinder will then travelupstream to the other cylinder where itwill bounce off of a closed exhaustvalve and return. Pipe lengths in thecrossover can be chosen so that the re-turning wave will produce a negativepressure for scavenging.

See Figure 7 and 8. Two differentcrossover systems were tested. Onewas a simulation model in which anindependent header on cylinder #1 hadan offshoot pipe welded on about 2.5"downstream from the cylinder flange.The offshoot pipe was a b l ind legwhose length could be adjusted and onthe end of which a pressure sensorwas attached, as shown in Figure 7.

The other crossover system (File511) was 1.75x34.5" headers pairingcylinder #1 with #2, and #3 with #4.Each of these pairs of cylinders fire180° apart. See Figure 8.

The reflected waves from the blindleg are powerful and their effect onthe P wave can be clearly seen here. InFigure 7, Xb marks the peak of the Pwave's arrival in the 67" blind leg offile 729 in the simulation model. XImarks the ill-timed return point of thatpeak into the cylinder #1 pressuretrace. This ruins the tuning. Nb marksthe trough in the 36.5" blind leg of file731. Nl shows this trough's return tothe cylinder to help it develop a nega-tive scavenging wave. This simulatorlacks the influence of cyl #2's firings.

The crossover system showed bet-ter performance if the length fromcylinder flange to tailpipe exit was 28"

SPORT AVIATION 39

Files 516/517/601/603/411: Tri-Y system at differing RPM'S. Headers are 1.75x34.5" Intotwo separate 1.875"x18" intermediates which then merge into a single 2x18.5" collector forfiles 516/517 but into a 2x6" collector for files 601/603. Cylinders #1 and #2 are mergedtogether as are cylinders #3 and #4. File #411 is a 4 into 1 collector system. 78"F. 8-24-96. Lycoming IO-360A1B6 firing order; 1324 Run at 125'MSL

40

Figure 9

2703 RPM 29.7" M.P. 20.5 gph 2x18.5 coll.

— — — 2476 RPM 25.0" M.P. 15.2 gph 2x18.5 COll

2699 RPM 29.8" M.P. 19.9 gph 2x6 coll.

2513 RPM 24.8" M.P. 14.8 gph 2x6 coll.

2731 RPM 29.5" M.P. 20.4 gph File #411

Exhaust Valve

Intake Valve

-20O 30 60 90 120 150 180 210 240 270 300 330 360 390 420 450 480 510 540 570 600 630 660

Crankshaft Degrees After Firing TDC

Files 516/517/518 to compare a Tri-Y system at differing RPM'S. All use the same equallength headers as: 1.75x34.5 with 1.875"x18" intermediates merging Into a single 18.5x2"outlet. 78-79°F. 8-24-96. Lycoming IO-360 A1B6 firing order:1324 Run at 125'MSL.

iFigure 10

——— #517 2476 RPM Cyl #2#516 2703 RPM Cyl #1

#516 2703 RPM Cyl #2 #518 2255 RPM Cyl #1

#517 2476 RPM Cyl #1 #518 2255 RPM Cyl #2

Note the way that the cyl #2 trace shows thearrival of cyl #1's P wave. The pairs of arrowsbelow show that the negative wave in cyl #2travels back to cyl #1 and helps lower its pressuretrace. This gives some excellent scavenging at all

\of these usable RPM'S. Shorter headers wouldprobably better scavenge the higher RPM'S.

-20100 120 140 160 1; O 200 220 240 260 280 300 320 340 360 380 400 420

40 JANUARY 1997Crankshaft Degrees After Firing TDC

rather than 45".A commonly used, 'off-the-shelf

Lycoming crossover system has 1.75"O.D. headers wherein cylinders #1and #2 are joined about 11" down-stream of cylinder #2 and 33" down-stream of cylinder #1. This makes a44" "blind leg" or crossover length be-tween those two cylinders. Cylinders#3 and #4 are similarly joined. Thesejoined pipes then each exit through a2.125"x 16" long tailpipe.

The Tri-Y system in Figure 8 was1.75x34.5" equal length headers whichmerged cylinders #1 with #2 and #3with #4 into 1.875x18" intermediatepipes. The intermediates merged intoa 2x18.5" collector.

In Figure 9, the Tri-Y showed alarge amount (-15.0" Hg.) of suctionduring overlap, but this came at thesacrifice of both opening pressure andpumping gain relative to the greentrace of file 411 's 4 into 1 collector.At lower RPM, a large pumping gainappears but the scavenge is lost. At"F" on the graph, a large pressuretrace arrives from cylinder #2's Pwave influence. It is the reflection ofsuch large pressure waves that makethe negative pressures so dramatic inthe Tri-Y and crossover systems.

See Figure 10. This expanded scalegraph shows how the negative waves inthe blind leg of cylinder #2 return andreduce the pressure in cylinder #1 'sheader. See the paired arrows. The reddouble-ended arrow shows the remark-ably early onset of negative pressure atlow RPM in this system. These primaryheader lengths (34.5") seem to be opti-mal for about 2500 RPM, judging bygood pumping gain and scavenge of theblue trace on the graph.

The Tri-Y's wave timing is primar-ily controlled by the primary headerlength. The diameter and length of thecommon tailpipe seem to shift the en-tire pressure trace up or down as a unit.In other studies, increasing the lengthof the intermediate pipes beyond 18"seemed to raise the backpressure.

Crossover systems and Tri-Y systemsare, in some ways, halfway between theindependent pipe system and the collec-tor system. They still exhibit higheropening pressures than the 4 into 1 col-lector systems, but they enjoy larger,longer duration negative waves aftertheir P wave than do independent pipes.Tri-Y tuning is more critical than the 4into 1 system as to RPM.

Files 432/428. Fast Fourier transform. (FFT) Interference seems to cause the higherfrequencies to diverge even though the RPM's and firing frequencies were nearly identical.0.6

Figure 11

0.5

0.4

ioo

«0.3w(D

O)DC

0.2

0.1

Cyl #1 Coanda nozzle +__ 1.75x34.5x2x19.5, 4 into

1 2670 RPM 22.25 Hzfiring #432

__ Collector 89 Hz firing#432

1.75x34.5x2x19.5, 4 into—— 1 2669 RPM 22.25 Hz

firing #428This shows that the Coanda nozzleraises the pressure energy in theheader's reflected waves and greatlyincreases it in the collector waves.The Coanda's inner cone doublesthe surface area of boundary layerfriction, slowing the gases andreducing the effective cross-sectionof the collector. An optimized designmight need to enlarge the collector'scross-sectional area inside theCoanda nozzle to provide for theextra boundary layer caused by theinner cone. The 4 into 1 systemshows that the firing frequencycontains most of the energy, with lessin the higher frequencies.

20 40 60 80 100 120Frequency, Hz

140 160 180

FREQUENCY ANALYSIS

The fast Fourier transform shown inFigure 11 is a way to depict the soundfrequencies most prevalent in a givenEPG. The 4 into 1 system shows thefiring frequency. Interestingly, someof the systems peak at multiples of thefiring frequency of the cylinder. A spe-cial type of loudspeaker mighttheoretically be used with a noise can-celling program (destructive inter-ference) to nearly eliminate the ex-haust noise by countering each of themain frequencies shown on the FFT.

COANDA NOZZLESThe Coanda nozzle is shown in the

photo on the cover page. It consists ofa megaphone inside which is placed asolid cone whose taper ratio producesno net change in cross-sectional areathroughout the megaphone. The outletof the Coanda nozzle has a sharply ta-pered trailing cone intended to producea low pressure vortex. Theory has itthat this vortex will reduce backpres-sure and give more horsepower.

The Coanda nozzle was added to

both the 4 into 1 exhaust system andthe Tri-Y system. See Figure 12. Inboth cases, no power gain was evidentand the EPG did not show any strikingchange. The Coanda nozzle did seemto give some noise reduction and mel-lowing of the exhaust sound.

HEADER SIZE

See Figure 13. The optimum headersize for this engine at 2500 to 2700RPM, at sea level appears to be 1.75"diameter. The length of the headers ina 4 into 1 system seems optimized atabout 28-36". Longer length probablyraises backpressure and delays onset ofscavenging while shorter lengths re-duce the abi l i ty to contain a fu l lydeveloped, powerful wave.

LEAN vs RICH EPG'sA richer mixture produces a lower

EGT and thus a slower wave speedthan does a lean mixture. Two EPG'swere run using 25" of manifold pres-sure and 2500 RPM wherein one used15 gph and the other 11 gph. The Pwaves and scavenging were nearly

identical but the opening pressure atEVO was lower for the rich mixturecase. The FFT's for these runs do showfrequency changes, but the EPG's lookremarkably similar.

BALL JOINT EFFECTS

' In the 2" collector tests (506,507)there was a very slight increase inbackpressure when a ball joint wasused, but no change in fuel flow, RPMor thrust was observed.

The P wave of the EPG was also un-affected by the addition of a 2.25"diameter ball joint 22" downstream ofthe collector merge when the total col-lector length was 43.75". Two differentball joints were used. One (715) had asmooth internal wall and the other(717) had the more common internalconcave chamber. The collector wave,recorded at a point 4" downstream ofthe ball joint, also showed essentiallyno change from either of the ball joints.When a megaphone was added to thestraight collector, it showed a markednegative wave after the collector wavepeak. Ball joints can probably be usedfor vibration isolation of the collectorwithout detuning of the exhaust.

COLD VS. HOT ENGINETwo runs (701,703) were made with

identical pipes (1.625x28x2.25x21.75)except that one was at 67° F OAT andthe other at 80° F OAT. The RPM's,fuel flows and P wave shape werenearly identical. The 80° F run showedslightly lower pressure at EVO (-9"Hg. vs. -6.5" Hg.) and overlap (-10"Hg. vs. -7" Hg.).

Two other runs (500,501) weremade with identical pipes (1.75 x34.5x2x20), one with 200° F CHT and theother with 400° F CHT. These showedno significant difference in the EPG.

MORE HORSEPOWER

Bruce Arrigoni, who has extensiveexperience in dyno race tuning of theSubaru engines with Formula Power inConcord, California, states that the sin-gle most effective way to increase theSubaru horsepower output was tosmooth the sharp-edged transition ofthe exhaust valve's seat bevel cutwhere it blends into the valve's tulipportion. This apparently greatly im-proves the flow past the valve both at

SPORT AVIATION 41

initial opening and during the smallvalve openings at overlap.

LENGTH FORMULAE

Most simple mathematical formu-lae for calculating the ideal length forexhaust pipes fail to account for torecognize that there is a Doppler phe-nomenon occurring in an exhaust pipebecause the sonic exhaust wave is rid-ing on the "wind" of the streamingmass flow of fuel and air. The sonicwave moves at 1500-1800 fps whilethe mass flow moves at 200-400 fps.The sonic wave thus travels faster tothe tailpipe than does the returning re-flected sonic wave which must "swimupstream" to reach the exhaust valve.

Computer programs can addressthese complexities using what is calledthe "method of characteristics". Onesuch program is Curt Leaverton's"Dynomation", available from V.P.Engineering, 5261 NW 114th St.,Suite J, Grimes, IA. 50111. Ph. 515-276-0701

SIZING THE PIPESIt must be remembered that the 200

HP engine becomes a 130 HP engineat cruise altitudes of 8000-12,000 ft.Optimization of exhaust tuning atthese altitudes, with the attendant re-duced air density, will call for the useof smaller diameter headers and col-lectors. A compromise must be foundto not rob the engine of its sea levelclimb power. A stainless steel multi-segmented jet nozzle/megaphonewhose outlet area could be adjustedfor altitude could be worthwhile foroptimizing both low and high altitudeperformance.

RECOMMENDATIONSThe 4 into 1 exhaust system (File

411) used on the CAFE testbedMooney can be reproduced by SamDavis at Tube Technologies in Coro-na, CA as mandrel bent pipes requir-ing TIG welding to their exhaustflanges. Alternative designs can bemade in mild steel from "U" bendsand then sent to Sam for duplicating in321 stainless steel. Aerospace Weldersin Minneapolis, MN can provide veryhigh quality stainless steel collectorsand merges for any desired system.All systems must include slip joints or42 JANUARY 1997

Files 428/432: 428 = 1.75x34.5x2x19.5. File 432 has a Coanda megaphone of 2x11x4 asdrawn below added onto the 19.5" collector of file 428. Wind 3 mph in both runs. 8-18-96.Lycoming IO-360 A1B6 firing order: 1324 Run at 125'MSL

428 20.1 gph 2669 RPM 106.6 dB 77°F

428 Intake

432 Coanda 20.0 gph 2670 RPM 105.0 dB75°F

432 Coanda IntakeThe apparent quieting fromthe Coanda nozzle was partlyartifact because it moved the

Q outlet further aft of the cabin.*^ The power of these two

systems seems nearlyidentical despite theCoanda's slightly betterscavenging at overlap and itslower pressure at EVO. TheCoanda might show a powergain if EVO were moved tolater in the cycle.

Figure 12

-2030 60 90 120 150 180 210 240 270 300 330 360 390 420 450 480 510 540 570 600

Crankshaft Degrees After Firing TDC

Files 737/738/739/411 Comparing straight stacks of 35.25" using various header diametersto the collector equipped system #411. 96°F Lycoming IO-360 A1B6 firing order: 1324 at125' MSL.40

30

20

The 1.625" headerreaches a higherpressure (clipped)and shows abroader P wave

_ than the 1.75" and1.875" headers.However, it doesscavenge well atoverlap bydelaying the arrival

_ of the first ringingwave at R.

737 1.625" 2700 RPM

738 1.75" 2690 RPM

7391.875" 2700 RPM ,,; , ;

411 1.75"x34.5x2.25x19.5 2731 RPM

Figure 13

File 411, with Itscollector, againappears to besuperior in reducing!backpressure.

-20

-240 <D-120 "S

60 90 120 150 180 210 240 270 300 330 360 390 420 450 480 510 540 570Crankshaft Degrees After Firing TDC

-o600

EXHAUST JET THRUSTFigure 14 shows the results of measuring the

pressure at the tailpipe exit with a pitot tube (to-tal pressure). The pressure measured this waycan be used to compute the jet thrust availablefrom the exhaust. Several formulae can be usedfor this, however, they require several assump-tions, which are listed in bold below:pe = P/gRTabsM = (C/1728) x (RPM/60) x psi x T|vV= M/peA = (2q/pe)1/2 where

V = ave. gas velocity at exit, ft/secM = mass flow rate in slugs/secA = 7r(tailpipe diameter-2(w))2/4w = wall thickness

q = 1/2 x pe x Vp2 = pitot pressure atthe tailpipe exit.

Vp = peak velocity derived from q.V = Vpx.817**Thrust in pounds = (WxV)/g = MxVA rough check can be made using thegph and air fuel ratio:W = lb/sec = (gphx6xl2)/3600**See reference 19. The .817 is derived froma complex analysis of these pipes' Reynold'snumbers and boundary layer thickness.Psi = 0.0023769pe = exhaust gas densityq = dynamic pressure, psfT|v = 0.95 = volumetric efficiencyA = pipe exit area, sq ft, excluding b.l.b.l. = boundary layerR = 54.0 = exhaust gas constantg = accel of gravity = 32.174 ft/sec2Tabs = 1760° R = exit temperatureP = 2116 psf = sea level pressure6 = pounds per gallon of avgas12 = 1 + air fuel ratio of 11 to 1 (rich)3600 = seconds per hour1728 = cubic inches per cubic footC = 180 cubic inches = effective full

time engine displacementUsing these formulae and 20.2 gph at 2685

RPM with 2.25 psi outlet pressure (Figure14), a peak exhaust jet thrust of about 10-12pounds is found at wide open throttle with a2" tailpipe. The 1.5" tailpipe nozzle gives 17-21 pounds of thrust. Simultaneous solution ofthese equations can be used to find the un-known values.

The ambient pressure will determine theexhaust gas density upon exit and thus the exitvelocity. The low ambient pressure at 10,000-14,000' would give an increase in exhaustthrust, especially with turbocharging, whichmaintains a higher exhaust mass flow atthose altitudes 6

Files 438/439/440/505 Average exhaust jet thrust variation with RPM and fuel flow. All arethe same 4 into 1, equal length headers: 1.75x34.5 with 2x19.5" collector, except 505which uses a 1.5" nozzle outlet on a 2x29" collector. 72-74°F. 8-18-96. Lycoming IO-360A1B6 firing order: 1324 Run at 125'MSL. A collector exit pitot probe was used here.40

30

20

10

CD

I£

OL

-10

—— #438 20.2 gph 2685 RPM 2.25 psi --• #440 4.6 gph 1571 RPM 0.28 psi

#439 10.4 gph 2195 RPM 0.94 psi 1.5" noz 19.5 gph 2645 RPM 5.25 psi

Figure 14

Pounds of Jet Thrust = (WxV)/gwhere g= = 32.174 ft/sec2

W = slugs/sec. = ((gph)x6x12)/3600.V = nozzle gas velocitySee text for computations ofthrust.

Assuming a 0.14" boundary layer and0.049" wall thickness, the outlet area,of the 2" collector is 2.06 sq in and itsthrust force calculates to 10-12 lb., at2685 RPM and 20.2 gph. The 1.5"nozzle gives 17-21 lb of thrust, butprobably has reduced horsepower

-— -due to higher backpressure.

Crankshaft Degrees After Firing TDC'Fuel = 6 lb/gal and air/fuel ratio is 11, giving the 6 and 12 above.

ball joints for strain relief placed bothat the mouth of the collector entry aswell as about h a l f w a y down theheaders The joints must always besecured with redundant spanningbolts, compression springs and cotterpinned castle nuts.

CONCLUSIONS1. Substantial negative pressure

waves can be generated in tunedaircraft exhaust systems and thetiming of their suction can bearranged so as to improve enginepower. Such improvement shouldproduce more power, better effi-ciency and a cleaner combustionchamber.

2. The 4 into 1 collector exhaustsystems appear to offer the bestcombination of low opening pres-sure, some pumping gain and goodscavenging, though the crossoverand Tri-Y systems can also obtaingood scavenging during the over-lap stroke.

3. The addition of a suitablemegaphone to the collector of a 4

into 1 exhaust system usually pro-duces an increase in the negativepressure achieved at the exhaustvalve, but at a substantial penaltyin noise. 12. 14.

4. The use of swiveling ball jointson the collector of a 4 into 1 ex-haust system has a neglible effecton the EPG and provides an im-portant vibration-isolation benefitto the system.

5. The optimization of pipegeometry for the crossover, Tri-Yand 4 into 1 exhaust systems can befound by study of the EPG.

6. Fast Fourier transforms, de-rived from the EPG, could facilitatethe development of an electronic,active noise-cancelling muffler.Aircraft exhaust systems, by theirlimited RPM range, are particu-larly well-suited to such a muffler.

7. The Coanda nozzle did notproduce a noticeable increase inpower. Fabrication and durabilityproblems make this nozzle of lim-ited attractiveness.8

SPORT AVIATION 43

8. Exhaust jet thrust was measuredand calculated for several exit sizes,RPM's and fuel flows. It can producesignificant thrust at high power set-tings, especially at cruising altitudes.6

9. The stock camshaft used in anaircraft engine is typically optimizedfor reliability and tractability and isnot optimized for the tuned exhaustsystems tested here. To fully realizethe potential benefits of tuned exhaustsystems for the aircraft engine, thecamshaft timing must be suitably al-tered by making exhaust valve closureoccur later and the overlap period oflonger duration and higher lift.9 Manyof the scavenging systems here do notexhibit as much effect upon the intakemanifold pressure during the overlapas might occur if the camshaft hadgreater valve overlap.

10. The effect of cylinder head tem-perature, mixture (EGT), and outsideair temperature on the EPG werestudied and found to produce mini-mal changes.

11. Further study should includethe correlation of climb and cruiseairspeeds with EPG's taken in flightscontrolled for power setting and air-craft weight. These should be per-formed using exhaust jet nozzles,megaphones, altered ignition timingand other variations. •

BIBLIOGRAPHY1. Seeley, Brien, and Vetter, Ed, The

EPG and Aircraft Exhaust Systems. SportAviation, Vol. 45, No. 1, pg. 39, January,1996.

2. Seeley, Brien, and Vetter, Ed, EPG.Sport Aviation, Vol. 45, No. 3, pg. 48,March, 1996.

3. Seeley, Brien, and Vetter, Ed, EPGIII. Sport Aviation, Vol. 45, No. 5, pg. 80,May, 1996

4. Lord, Albert M., Heinicke, OrvilleH., and Stricker, Edward G.: Effect of Ex-haust Pressure on Knock-LimitedPerformance of an Air-Cooled Aircraft-En-gine Cylinder. NACA Technical Note No.1617, June 1948.

5. Hey wood, John B.: Internal Com-bustion Engine Fundamentals, 1988,McGraw-Hill, Inc.

6. Pinkel, Benjamin, Turner L. Richard,Voss, Fred, and Humble, Leroy V.: Ex-44 JANUARY 1997

haust-Stack Nozzle Area and Shape For In-dividual Cylinder Exhaust Gas JetPropulsion System. NACA Report No.765,1943.

7. Smith, Philip H., and Morrison, JohnC., The Scientific Design of Exhaust andIntake Systems, Third Edition, RobertBentley, Inc, June 1978.

8. Blair, Gordon P, Design and Simula-tion of Two-Stroke Engines, Society ofAutomotive Engineers, Inc, 1996.

9. Harralson, Joseph, Design of Racingand High Performance Engines PT-53, So-ciety of Automotive Engineers, Inc, 1995.

10. Seeley, Brien, The Technology ofCAFE Flight Testing. Sport Aviation Vol.43, No. 5, pg. 51, May, 1994.

11. Goldstein, Norton, The Coanda Ef-fect, Hot Rod Magazine, December 1962.

12. Creagh, John W. R., Hartmann,Melvin J, and Arthur Jr, W. Lewis, An In-vestigation of Valve-Overlap ScavengingOver a Wide Range of Inlet and ExhaustPressures, NACA Technical Note No.1475, November, 1947.

13. Tabaczynski, Rodney J, Effect ofInlet and Exhaust System Design on En-gine Performance, SAE Paper 821577,1982.

14. Jameson, Renee T, and Hodgins,Patrick A, Improvement of the TorqueCharacteristics of a Small, High-SpeedEngine Through the Design of Helmholtz-Tuned Manifolding, SAE Paper 900680,March 1990. , , , ,=, ,„•

15. Ram Tuning for Big Bikes, BigBike magazine, pg. 52-57, November,1970.

16. Ewing, William H, and Nemoto,Hiroshi, A Computer Simulation Approachto Exhaust System Noise Attenuation,SAE Paper 900392, March 2,1990.

17. Yagi, Shizuo, Ishizuya, Akira, andFujii, Isao, Research and Development ofHigh-Speed, High Performance, SmallDisplacement Honda Engines, SAE Paper700122, January 16,1970. .., ,• :

18. Hosomi, Mikiya, Ogawao, Sumio,Imagawa, Toshiyuki, and Hokazono,Yuichi, Development of Exhaust ManifoldMuffler, SAE Paper 930625, March 5,1993.

19. Schlicting, Herman, BoundaryLayer Theory, pg 402-403. PergamonPress, 1955. - •

IMPORTANT NOTICEEvery effort has been made to

obtain the most accurate informa- ;tion possible. The data are presentedas measured and are subject to er-rors from a variety of sources. Any ireproduction, sale, republication, orother use of the whole or any part ofthis report without the consent of •;the Experimental Aircraft Associa- 'tion and the CAFE Foundation isstrictly prohibited. Reprints of thisreport may be obtained by writingto: Sport Aviation, EAA Aviation ;Center, P. O. Box 3086, Oshkosh,WI, 54903-3086. J

ACKNOWLEDGEMENTS

This work was supported in part byFAA Research Grant Number 95-G-037. The CAFE Foundation gratefullyacknowledges the assistance of AnneSeeley, Mary Vetter, Lyle Powell, JimGriswold, EAA Chapter 124, the SantaRosa Airport Tower, the Rengstorffamily, Hartzell Propellers, and severalhelpful people in the engineering de-partment at Avco-Lycoming.

SPONSORSExperimental Aircraft AssociationFederal Aviation AdministrationAircraft Spruce & Specialty Co.Aerospace Welding

Minneapolis, Inc.Fluke CorporationB & C Specialty CompanyEngineered Software:

"WildTools/PowerCadd"Bourns & Son SignsJohnny Franklin's Muffler ShopSam Davis at Tube TechnologiesFactory PipesAeroLogic's Personal Skunk Works

Software

COMP AKA TIVE AIRCRAFTFLIGHT EFFICIENCY, INC.

The CAFE Foundation:A Non Profit, All Volunteer,

Tax-exempt Educational Foundation4370 Raymonde Way, Santa Rosa,

CA 95404.FAX 544-2734.

Aircraft Performance Evaluation Center:707-/45-CAFE (hangar, message)

America Online:CAFE400@aol.com

Internet: CAFE400@sonic.net