HOME

TheQuestForHorsepowerSeriesBringsYou:

TwoStroke TUNERSHANDBOOK
ByGordonJennings Illustrationsbytheauthor

Copyright1973by GordonJennings

Onlytenyearsagothetwostrokeenginewaswidelyandquiteunderstandablythoughttobea"reasonablealternativetothefourstrokeonlywhenminimum weightandmanufacturingcostwereallimportantconsiderations.Thetwostrokewasrecognizedashavingsubstantialtheoreticalpromise,asitdeliveredapowerstroke foreach360degreesofcrankshaftrotationbutthehardrealitywasthateachindividualpowerimpulsewastoofeebletoamounttomuchwhentotaledattheoutputend ofthecrankshaft.Averyfewengineshadbeguntoappearinwhichsomeofthetheoreticalpromisewasrealizedhowever,andthisencouragedengineersatMZ,Yamaha andSuzukitopersistintheireffortstowringcompetitivepoweroutputfromtheracingtwostrokeengine.Tosaythattheywereultimatelysuccessfulwouldbegross understatement. Thoseengineersweremotivatedbytheneedtodemonstratethatthetwostrokeengine,perse,wasworthwhileasthatwouldstimulatesalesoftheircompanies' ordinarytouringmodels.Myowninterestinthetwostroke,whichhadreachedthelevelofanobsessionby1963,wasgeneratedbycomparativepoverty.Iliketotinker withengines,andthecomplexitiesofthepoppetvalvefourstrokemakemodificationsveryexpensive.Onemaythinkthatachangeinvalvetimingwoulddowondersfor afourstroke'spower,butgettingacamshaftmadetoordercostshundredsofdollars.Incontrast,atwostrokeengine'svalvetimingmaybealteredsimplybyreshaping theholesinitscylinders,anditspoweroutputmarkedlychangedbyutilizinginertiaandresonanteffectsinitsintakeandexhausttracts.Noneofthesemodificationsare costly. Ontheotherhand,whilethetwostrokeenginedoesnotcommonlyrequirelargedollarinputstoraiseitspoweroutput,itdoesrequireanindepthunderstanding onthepartofthemandoingthemodifications.InanattempttoacquirethatunderstandingIbeganastudyofthehighspeed,highoutputtwostrokeenginethathasledto thecollectionofaminorlibraryoftextbooksandSAEpapers.Andtoanendlessseriesofexperiments,someofthemilluminatingandmanyothersraisingmorequestions thantheyhaveanswered.AtthisstageIhavearrivedatmoreorlesssatisfactoryexplanationsformostofthegrossphenomena,suchasthegeneralbehaviorofexpansion chambersandporttimeareavalues,andIflattermyselftothinkthatjustthatmuchisanacceptableexcuseforwritingthisbookfortheguidanceofthelayman experimenter.Ifitwillnotsupplyalloftheanswersitwillatleasttakecareofthefundamentalproblemsandpreventtheworstmistakes. MyspecialthankstoMr.JohnBrooks,ofMcCullochEngineering,whohasdonemuchtodilutemyoncepureignorance(butshouldnotbeheldaccountablefor theresiduefoundherein).AlsotothelateHenryKoepke,whomistakenlyassumedthatIknewsomethingabouttwostrokeenginesandsupportedmyearlyresearchto myoldfriendJoeParkhurst,whostartedmeworkingonthisbooknearlytenyearsagobutnevergotitandfinallytoTomHeininger,whowheedled,needled,pleaded, complainedandcajoleduntilIhammeredmyfileofnotesintopublishableform.

Foreword...iii Contents....v FUNDAMENTALS..1 PredictingPower3

Copyright1973by GordonJennings

Onlytenyearsagothetwostrokeenginewaswidelyandquiteunderstandablythoughttobea"reasonablealternativetothefourstrokeonlywhenminimum weightandmanufacturingcostwereallimportantconsiderations.Thetwostrokewasrecognizedashavingsubstantialtheoreticalpromise,asitdeliveredapowerstroke foreach360degreesofcrankshaftrotationbutthehardrealitywasthateachindividualpowerimpulsewastoofeebletoamounttomuchwhentotaledattheoutputend ofthecrankshaft.Averyfewengineshadbeguntoappearinwhichsomeofthetheoreticalpromisewasrealizedhowever,andthisencouragedengineersatMZ,Yamaha andSuzukitopersistintheireffortstowringcompetitivepoweroutputfromtheracingtwostrokeengine.Tosaythattheywereultimatelysuccessfulwouldbegross understatement. Thoseengineersweremotivatedbytheneedtodemonstratethatthetwostrokeengine,perse,wasworthwhileasthatwouldstimulatesalesoftheircompanies' ordinarytouringmodels.Myowninterestinthetwostroke,whichhadreachedthelevelofanobsessionby1963,wasgeneratedbycomparativepoverty.Iliketotinker withengines,andthecomplexitiesofthepoppetvalvefourstrokemakemodificationsveryexpensive.Onemaythinkthatachangeinvalvetimingwoulddowondersfor afourstroke'spower,butgettingacamshaftmadetoordercostshundredsofdollars.Incontrast,atwostrokeengine'svalvetimingmaybealteredsimplybyreshaping theholesinitscylinders,anditspoweroutputmarkedlychangedbyutilizinginertiaandresonanteffectsinitsintakeandexhausttracts.Noneofthesemodificationsare costly. Ontheotherhand,whilethetwostrokeenginedoesnotcommonlyrequirelargedollarinputstoraiseitspoweroutput,itdoesrequireanindepthunderstanding onthepartofthemandoingthemodifications.InanattempttoacquirethatunderstandingIbeganastudyofthehighspeed,highoutputtwostrokeenginethathasledto thecollectionofaminorlibraryoftextbooksandSAEpapers.Andtoanendlessseriesofexperiments,someofthemilluminatingandmanyothersraisingmorequestions thantheyhaveanswered.AtthisstageIhavearrivedatmoreorlesssatisfactoryexplanationsformostofthegrossphenomena,suchasthegeneralbehaviorofexpansion chambersandporttimeareavalues,andIflattermyselftothinkthatjustthatmuchisanacceptableexcuseforwritingthisbookfortheguidanceofthelayman experimenter.Ifitwillnotsupplyalloftheanswersitwillatleasttakecareofthefundamentalproblemsandpreventtheworstmistakes. MyspecialthankstoMr.JohnBrooks,ofMcCullochEngineering,whohasdonemuchtodilutemyoncepureignorance(butshouldnotbeheldaccountablefor theresiduefoundherein).AlsotothelateHenryKoepke,whomistakenlyassumedthatIknewsomethingabouttwostrokeenginesandsupportedmyearlyresearchto myoldfriendJoeParkhurst,whostartedmeworkingonthisbooknearlytenyearsagobutnevergotitandfinallytoTomHeininger,whowheedled,needled,pleaded, complainedandcajoleduntilIhammeredmyfileofnotesintopublishableform.

Foreword...iii Contents....v FUNDAMENTALS..1 PredictingPower3 PistonSpeed...5 PistonAcceleration7 CRANKTRAIN.17 ThePiston17 PistonRings.22 Wristpin/CrankpinBearings25 CrankAssembly.30 CYLINDERHEADS..33 TheCombustionProcess.34 SquishBands...37 PlugLocation..42 Head/CylinderSealing.46 EXPANSIONCHAMBERS51 TheBasicProcess52 TunedLength..55 DiffuserProportions.56 BaffleCones...................................................................................................61 OutletPipes....................................................................................................64 LeadinPipes.................................................................................................66 ExpansionChamberDesignFormulae..........................................................73 PORTTIMING......................................................................................................75 SpecificTimeArea......................................................................................76 AngleArea...................................................................................................80 TimeAreaCombinations.........................................................................81 EmphasisonArea.....................................................................................84 TimingLimits...........................................................................................85 RotaryValveTiming................................................................................88 CRANKCASEPUMPING...............................................................................91 ResonanceEffects.....................................................................................92 CarburetorLocation.............................................................................94 CrankcaseVolume....................................................................................95 ReedValves..............................................................................................98 TheRotaryValve.....................................................................................101 IntakePortShape......................................................................................102 CYLINDERSCAVENGING105 ExhaustPort.............................................................................................109 PortEdgeChamfers................................................................................112 FlowPatterns............................................................................................115 3lultipleTransferPorts.............................................................................119 Subtleties.............................................................................................121 CARBURETION,IGNITION......................................................................129 TheBasicCarburetor.................................................................................130 AdjustingMixture.................................................................................134 IgnitionFundamentals.............................................................................143

Copyright1973by GordonJennings

Onlytenyearsagothetwostrokeenginewaswidelyandquiteunderstandablythoughttobea"reasonablealternativetothefourstrokeonlywhenminimum weightandmanufacturingcostwereallimportantconsiderations.Thetwostrokewasrecognizedashavingsubstantialtheoreticalpromise,asitdeliveredapowerstroke foreach360degreesofcrankshaftrotationbutthehardrealitywasthateachindividualpowerimpulsewastoofeebletoamounttomuchwhentotaledattheoutputend ofthecrankshaft.Averyfewengineshadbeguntoappearinwhichsomeofthetheoreticalpromisewasrealizedhowever,andthisencouragedengineersatMZ,Yamaha andSuzukitopersistintheireffortstowringcompetitivepoweroutputfromtheracingtwostrokeengine.Tosaythattheywereultimatelysuccessfulwouldbegross understatement. Thoseengineersweremotivatedbytheneedtodemonstratethatthetwostrokeengine,perse,wasworthwhileasthatwouldstimulatesalesoftheircompanies' ordinarytouringmodels.Myowninterestinthetwostroke,whichhadreachedthelevelofanobsessionby1963,wasgeneratedbycomparativepoverty.Iliketotinker withengines,andthecomplexitiesofthepoppetvalvefourstrokemakemodificationsveryexpensive.Onemaythinkthatachangeinvalvetimingwoulddowondersfor afourstroke'spower,butgettingacamshaftmadetoordercostshundredsofdollars.Incontrast,atwostrokeengine'svalvetimingmaybealteredsimplybyreshaping theholesinitscylinders,anditspoweroutputmarkedlychangedbyutilizinginertiaandresonanteffectsinitsintakeandexhausttracts.Noneofthesemodificationsare costly. Ontheotherhand,whilethetwostrokeenginedoesnotcommonlyrequirelargedollarinputstoraiseitspoweroutput,itdoesrequireanindepthunderstanding onthepartofthemandoingthemodifications.InanattempttoacquirethatunderstandingIbeganastudyofthehighspeed,highoutputtwostrokeenginethathasledto thecollectionofaminorlibraryoftextbooksandSAEpapers.Andtoanendlessseriesofexperiments,someofthemilluminatingandmanyothersraisingmorequestions thantheyhaveanswered.AtthisstageIhavearrivedatmoreorlesssatisfactoryexplanationsformostofthegrossphenomena,suchasthegeneralbehaviorofexpansion chambersandporttimeareavalues,andIflattermyselftothinkthatjustthatmuchisanacceptableexcuseforwritingthisbookfortheguidanceofthelayman experimenter.Ifitwillnotsupplyalloftheanswersitwillatleasttakecareofthefundamentalproblemsandpreventtheworstmistakes. MyspecialthankstoMr.JohnBrooks,ofMcCullochEngineering,whohasdonemuchtodilutemyoncepureignorance(butshouldnotbeheldaccountablefor theresiduefoundherein).AlsotothelateHenryKoepke,whomistakenlyassumedthatIknewsomethingabouttwostrokeenginesandsupportedmyearlyresearchto myoldfriendJoeParkhurst,whostartedmeworkingonthisbooknearlytenyearsagobutnevergotitandfinallytoTomHeininger,whowheedled,needled,pleaded, complainedandcajoleduntilIhammeredmyfileofnotesintopublishableform.

Foreword...iii Contents....v FUNDAMENTALS..1 PredictingPower3 PistonSpeed...5 PistonAcceleration7 CRANKTRAIN.17 ThePiston17 PistonRings.22 Wristpin/CrankpinBearings25 CrankAssembly.30 CYLINDERHEADS..33 TheCombustionProcess.34 SquishBands...37 PlugLocation..42 Head/CylinderSealing.46 EXPANSIONCHAMBERS51 TheBasicProcess52 TunedLength..55 DiffuserProportions.56 BaffleCones...................................................................................................61 OutletPipes....................................................................................................64 LeadinPipes.................................................................................................66 ExpansionChamberDesignFormulae..........................................................73 PORTTIMING......................................................................................................75 SpecificTimeArea......................................................................................76 AngleArea...................................................................................................80 TimeAreaCombinations.........................................................................81 EmphasisonArea.....................................................................................84 TimingLimits...........................................................................................85 RotaryValveTiming................................................................................88 CRANKCASEPUMPING...............................................................................91 ResonanceEffects.....................................................................................92 CarburetorLocation.............................................................................94 CrankcaseVolume....................................................................................95 ReedValves..............................................................................................98 TheRotaryValve.....................................................................................101 IntakePortShape......................................................................................102 CYLINDERSCAVENGING105 ExhaustPort.............................................................................................109 PortEdgeChamfers................................................................................112 FlowPatterns............................................................................................115 3lultipleTransferPorts.............................................................................119 Subtleties.............................................................................................121 CARBURETION,IGNITION......................................................................129 TheBasicCarburetor.................................................................................130 AdjustingMixture.................................................................................134 IgnitionFundamentals.............................................................................143 SparkPlugs.............................................................................................152

MyspecialthankstoMr.JohnBrooks,ofMcCullochEngineering,whohasdonemuchtodilutemyoncepureignorance(butshouldnotbeheldaccountablefor theresiduefoundherein).AlsotothelateHenryKoepke,whomistakenlyassumedthatIknewsomethingabouttwostrokeenginesandsupportedmyearlyresearchto myoldfriendJoeParkhurst,whostartedmeworkingonthisbooknearlytenyearsagobutnevergotitandfinallytoTomHeininger,whowheedled,needled,pleaded, complainedandcajoleduntilIhammeredmyfileofnotesintopublishableform.

Foreword...iii Contents....v FUNDAMENTALS..1 PredictingPower3 PistonSpeed...5 PistonAcceleration7 CRANKTRAIN.17 ThePiston17 PistonRings.22 Wristpin/CrankpinBearings25 CrankAssembly.30 CYLINDERHEADS..33 TheCombustionProcess.34 SquishBands...37 PlugLocation..42 Head/CylinderSealing.46 EXPANSIONCHAMBERS51 TheBasicProcess52 TunedLength..55 DiffuserProportions.56 BaffleCones...................................................................................................61 OutletPipes....................................................................................................64 LeadinPipes.................................................................................................66 ExpansionChamberDesignFormulae..........................................................73 PORTTIMING......................................................................................................75 SpecificTimeArea......................................................................................76 AngleArea...................................................................................................80 TimeAreaCombinations.........................................................................81 EmphasisonArea.....................................................................................84 TimingLimits...........................................................................................85 RotaryValveTiming................................................................................88 CRANKCASEPUMPING...............................................................................91 ResonanceEffects.....................................................................................92 CarburetorLocation.............................................................................94 CrankcaseVolume....................................................................................95 ReedValves..............................................................................................98 TheRotaryValve.....................................................................................101 IntakePortShape......................................................................................102 CYLINDERSCAVENGING105 ExhaustPort.............................................................................................109 PortEdgeChamfers................................................................................112 FlowPatterns............................................................................................115 3lultipleTransferPorts.............................................................................119 Subtleties.............................................................................................121 CARBURETION,IGNITION......................................................................129 TheBasicCarburetor.................................................................................130 AdjustingMixture.................................................................................134 IgnitionFundamentals.............................................................................143 SparkPlugs.............................................................................................152

Throughoutthisbookitwillbeassumed,inconvenientthoughthatassumptionmayoccasionallybe,thatthereaderhasprogressedtoatleastasuperficial knowledgeofthemannerinwhichapistontypeinternalcombustionengine withparticularreferencetothoseoperatingonthetwostrokecycleprincipleconverts quantitiesoffuelandairintousefulpowerdeliveredattheendofitscrankshaft.Peoplewhoneedenlightenmentinthatregardwillfindawealthofexplanatoryliterature collectedontheshelvesofanypubliclibrarynorealpurposewouldbeservedbylingeringoverthematterhere.NeitherwillIattempttoinstructyouintheelementary mathematicsandphysicsrequiredtograspmuchofwhatfollows,asagainthepubliclibraryisanentirelyadequatesourceofinformation.Whatwillbeprovidedisakind ofstateoftheart reportabouthighspeed,highoutputtwostrokeenginesforlaymenwhoinmostcasesdonothaveaccesstotheliterature(SAEpapers,etc.) availabletoengineersandthusmustrelyuponhunches(oftenwrong)andfolklore(almostinvariablywrong)forguidance.Manyhavelearned,totheirsorrow,thatitis distinctlypossibletolavishenormousamountsoftimeandmoneyonthetwostrokeenginewithoutrealizingareturnappropriatetotheinvestment.Theinformationtobe providedherewillnotmakeyouaKaaden,orNaitoitwillhelpyoutoavoidsomeofthemoreseriousmistakes. Thefirstseriousmistakealaymanexperimentercanmakeistoassumethatthosewhodesignedandmanufacturedhisparticularenginedidn'tknowwhatthey weredoing.Inpointoffact,theprofessionalengineerknowsverywell,andiftheengineinquestionissomethingotherthanwhattheexperimenterhasinmind,thereare excellentreasons:allenginesarecompromised,fromwhatyoumightconsideranideal,intheinterestofmanufacturingeconomyandbroadusefulness.Forexample,ports mayderivetheirshapeasmuchfromwhatthedesignengineerintendedtobealowscrapratecastingasfromconsiderationofflowcharacteristics.Inotherwords,even somethinglikeportsdesignalwayswillbeinfluencedbythedemandsofmassproductionmanufacturing.Similarly,designingformassmarketsalesimpliesthatanengine mustbeagreeabletomanydifferentuseseventhoughthatinevitablymeansthatitwilldonosinglethingparticularlywell.Intheseareaswillwefindthelatitudefor improving anengine,andoneshouldalwaysbemindfulthattherealtaskissimplytotailoramassuseproducttoaveryspecificapplicationandthatinthetailoring processoneinevitablywillincurallthevariousexpensestheengine'sdesignerhasavoided.Hoursoflabormayberequiredtofinishroughcastportsdollarswillbespent correctingotherthingsthatarethecreaturesofmanufacturingeconomiespoweraddedatmaximumrevswillbepowersubtractedatlowercrankshaftspeeds,whilethe increasedspeedsrequiredtoobtainlargeimprovementsinpoweroutputwillbepaidforintermsofreliability. Anothermistakecommonlymade,sometimesevenbythosewhohaveenjoyedsomesuccessinmodifyingtwostrokeengines,istobelieveinakindof mechanisticmagic.Biggercarburetors,highercompressionratios,alteredporttimingsandexpansionchambersoftendobringanimprovementinpoweroutput,butmore

theresiduefoundherein).AlsotothelateHenryKoepke,whomistakenlyassumedthatIknewsomethingabouttwostrokeenginesandsupportedmyearlyresearchto myoldfriendJoeParkhurst,whostartedmeworkingonthisbooknearlytenyearsagobutnevergotitandfinallytoTomHeininger,whowheedled,needled,pleaded, complainedandcajoleduntilIhammeredmyfileofnotesintopublishableform.

Foreword...iii Contents....v FUNDAMENTALS..1 PredictingPower3 PistonSpeed...5 PistonAcceleration7 CRANKTRAIN.17 ThePiston17 PistonRings.22 Wristpin/CrankpinBearings25 CrankAssembly.30 CYLINDERHEADS..33 TheCombustionProcess.34 SquishBands...37 PlugLocation..42 Head/CylinderSealing.46 EXPANSIONCHAMBERS51 TheBasicProcess52 TunedLength..55 DiffuserProportions.56 BaffleCones...................................................................................................61 OutletPipes....................................................................................................64 LeadinPipes.................................................................................................66 ExpansionChamberDesignFormulae..........................................................73 PORTTIMING......................................................................................................75 SpecificTimeArea......................................................................................76 AngleArea...................................................................................................80 TimeAreaCombinations.........................................................................81 EmphasisonArea.....................................................................................84 TimingLimits...........................................................................................85 RotaryValveTiming................................................................................88 CRANKCASEPUMPING...............................................................................91 ResonanceEffects.....................................................................................92 CarburetorLocation.............................................................................94 CrankcaseVolume....................................................................................95 ReedValves..............................................................................................98 TheRotaryValve.....................................................................................101 IntakePortShape......................................................................................102 CYLINDERSCAVENGING105 ExhaustPort.............................................................................................109 PortEdgeChamfers................................................................................112 FlowPatterns............................................................................................115 3lultipleTransferPorts.............................................................................119 Subtleties.............................................................................................121 CARBURETION,IGNITION......................................................................129 TheBasicCarburetor.................................................................................130 AdjustingMixture.................................................................................134 IgnitionFundamentals.............................................................................143 SparkPlugs.............................................................................................152

Throughoutthisbookitwillbeassumed,inconvenientthoughthatassumptionmayoccasionallybe,thatthereaderhasprogressedtoatleastasuperficial knowledgeofthemannerinwhichapistontypeinternalcombustionengine withparticularreferencetothoseoperatingonthetwostrokecycleprincipleconverts quantitiesoffuelandairintousefulpowerdeliveredattheendofitscrankshaft.Peoplewhoneedenlightenmentinthatregardwillfindawealthofexplanatoryliterature collectedontheshelvesofanypubliclibrarynorealpurposewouldbeservedbylingeringoverthematterhere.NeitherwillIattempttoinstructyouintheelementary mathematicsandphysicsrequiredtograspmuchofwhatfollows,asagainthepubliclibraryisanentirelyadequatesourceofinformation.Whatwillbeprovidedisakind ofstateoftheart reportabouthighspeed,highoutputtwostrokeenginesforlaymenwhoinmostcasesdonothaveaccesstotheliterature(SAEpapers,etc.) availabletoengineersandthusmustrelyuponhunches(oftenwrong)andfolklore(almostinvariablywrong)forguidance.Manyhavelearned,totheirsorrow,thatitis distinctlypossibletolavishenormousamountsoftimeandmoneyonthetwostrokeenginewithoutrealizingareturnappropriatetotheinvestment.Theinformationtobe providedherewillnotmakeyouaKaaden,orNaitoitwillhelpyoutoavoidsomeofthemoreseriousmistakes. Thefirstseriousmistakealaymanexperimentercanmakeistoassumethatthosewhodesignedandmanufacturedhisparticularenginedidn'tknowwhatthey weredoing.Inpointoffact,theprofessionalengineerknowsverywell,andiftheengineinquestionissomethingotherthanwhattheexperimenterhasinmind,thereare excellentreasons:allenginesarecompromised,fromwhatyoumightconsideranideal,intheinterestofmanufacturingeconomyandbroadusefulness.Forexample,ports mayderivetheirshapeasmuchfromwhatthedesignengineerintendedtobealowscrapratecastingasfromconsiderationofflowcharacteristics.Inotherwords,even somethinglikeportsdesignalwayswillbeinfluencedbythedemandsofmassproductionmanufacturing.Similarly,designingformassmarketsalesimpliesthatanengine mustbeagreeabletomanydifferentuseseventhoughthatinevitablymeansthatitwilldonosinglethingparticularlywell.Intheseareaswillwefindthelatitudefor improving anengine,andoneshouldalwaysbemindfulthattherealtaskissimplytotailoramassuseproducttoaveryspecificapplicationandthatinthetailoring processoneinevitablywillincurallthevariousexpensestheengine'sdesignerhasavoided.Hoursoflabormayberequiredtofinishroughcastportsdollarswillbespent correctingotherthingsthatarethecreaturesofmanufacturingeconomiespoweraddedatmaximumrevswillbepowersubtractedatlowercrankshaftspeeds,whilethe increasedspeedsrequiredtoobtainlargeimprovementsinpoweroutputwillbepaidforintermsofreliability. Anothermistakecommonlymade,sometimesevenbythosewhohaveenjoyedsomesuccessinmodifyingtwostrokeengines,istobelieveinakindof mechanisticmagic.Biggercarburetors,highercompressionratios,alteredporttimingsandexpansionchambersoftendobringanimprovementinpoweroutput,butmore andbiggerisnotmagically,instantlybetter.Allmustworkinconcertwiththebasicengine,directedtowardtheparticularapplication,beforetheyconstituteagenuine

ReedValves..............................................................................................98 TheRotaryValve.....................................................................................101 IntakePortShape......................................................................................102 CYLINDERSCAVENGING105 ExhaustPort.............................................................................................109 PortEdgeChamfers................................................................................112 FlowPatterns............................................................................................115 3lultipleTransferPorts.............................................................................119 Subtleties.............................................................................................121 CARBURETION,IGNITION......................................................................129 TheBasicCarburetor.................................................................................130 AdjustingMixture.................................................................................134 IgnitionFundamentals.............................................................................143 SparkPlugs.............................................................................................152

Throughoutthisbookitwillbeassumed,inconvenientthoughthatassumptionmayoccasionallybe,thatthereaderhasprogressedtoatleastasuperficial knowledgeofthemannerinwhichapistontypeinternalcombustionengine withparticularreferencetothoseoperatingonthetwostrokecycleprincipleconverts quantitiesoffuelandairintousefulpowerdeliveredattheendofitscrankshaft.Peoplewhoneedenlightenmentinthatregardwillfindawealthofexplanatoryliterature collectedontheshelvesofanypubliclibrarynorealpurposewouldbeservedbylingeringoverthematterhere.NeitherwillIattempttoinstructyouintheelementary mathematicsandphysicsrequiredtograspmuchofwhatfollows,asagainthepubliclibraryisanentirelyadequatesourceofinformation.Whatwillbeprovidedisakind ofstateoftheart reportabouthighspeed,highoutputtwostrokeenginesforlaymenwhoinmostcasesdonothaveaccesstotheliterature(SAEpapers,etc.) availabletoengineersandthusmustrelyuponhunches(oftenwrong)andfolklore(almostinvariablywrong)forguidance.Manyhavelearned,totheirsorrow,thatitis distinctlypossibletolavishenormousamountsoftimeandmoneyonthetwostrokeenginewithoutrealizingareturnappropriatetotheinvestment.Theinformationtobe providedherewillnotmakeyouaKaaden,orNaitoitwillhelpyoutoavoidsomeofthemoreseriousmistakes. Thefirstseriousmistakealaymanexperimentercanmakeistoassumethatthosewhodesignedandmanufacturedhisparticularenginedidn'tknowwhatthey weredoing.Inpointoffact,theprofessionalengineerknowsverywell,andiftheengineinquestionissomethingotherthanwhattheexperimenterhasinmind,thereare excellentreasons:allenginesarecompromised,fromwhatyoumightconsideranideal,intheinterestofmanufacturingeconomyandbroadusefulness.Forexample,ports mayderivetheirshapeasmuchfromwhatthedesignengineerintendedtobealowscrapratecastingasfromconsiderationofflowcharacteristics.Inotherwords,even somethinglikeportsdesignalwayswillbeinfluencedbythedemandsofmassproductionmanufacturing.Similarly,designingformassmarketsalesimpliesthatanengine mustbeagreeabletomanydifferentuseseventhoughthatinevitablymeansthatitwilldonosinglethingparticularlywell.Intheseareaswillwefindthelatitudefor improving anengine,andoneshouldalwaysbemindfulthattherealtaskissimplytotailoramassuseproducttoaveryspecificapplicationandthatinthetailoring processoneinevitablywillincurallthevariousexpensestheengine'sdesignerhasavoided.Hoursoflabormayberequiredtofinishroughcastportsdollarswillbespent correctingotherthingsthatarethecreaturesofmanufacturingeconomiespoweraddedatmaximumrevswillbepowersubtractedatlowercrankshaftspeeds,whilethe increasedspeedsrequiredtoobtainlargeimprovementsinpoweroutputwillbepaidforintermsofreliability. Anothermistakecommonlymade,sometimesevenbythosewhohaveenjoyedsomesuccessinmodifyingtwostrokeengines,istobelieveinakindof mechanisticmagic.Biggercarburetors,highercompressionratios,alteredporttimingsandexpansionchambersoftendobringanimprovementinpoweroutput,butmore andbiggerisnotmagically,instantlybetter.Allmustworkinconcertwiththebasicengine,directedtowardtheparticularapplication,beforetheyconstituteagenuine improvement.Youcannottreatthemasavoodooincantation,hopingthatifyoumuttertherightphrasesandstirthechickenentrailsintheprescribedmanner,yourmild mannered,allpurposechufferwillbetransformedintoahyperhorsepowerfirebreather.Withalotofluck,youmightgetthatresultthechancesheavilyarethatyou won't. Withallthemysticismfilteredout,horsepoweratanygivendisplacementissimplyafunctionofaveragepressureinthecylinderduringthepowerstrokeandthe rateatwhichpowerstrokesoccur,minusworkabsorbedbyfrictionandscavenging.Raisepressureand/orthedeliveryrateofthepowerstrokes,orreducefrictionand pumpinglosses,andtheengine'snetoutputwillrise.Unfortunately,therearelimitationsonallsides:Pressuremustbelimitedbecauseofthermalconsiderations(andis furtherlimitedbyanengine'srestrictedabilitytorechargeitscylinderwithafreshair/fuelmixturebetweenpowerstrokes).Thelimitforpowerstrokesperunitoftimeis establishedbywhatistolerableintermsofcrankshaftrotationalspeeds,andtolerablehereiswhatthebearings,rodandpistonwillsurvive,ininertialoadings,forwhat youconsideranacceptableservicelifethedesignengineerhasalreadyexpressedhisopinioninthismatter.Pumpinglossescanbereducedrelativetothemassflow throughanenginewithaproperlydesignedexhaustsystem,butotherwiseareaninevitableandalmostinvariableconsequenceofpullingairfromtheatmosphere,moving itthroughtheengine,andouttheexhaustport.Someimprovementinoutputmaybeobtainedwithreductionsinfriction,butthescopeforsuchimprovementsisverysmall comparedtowhatmaybeaccomplishedwithcylinderpressureandenginespeed. Obviously,pressureinacylinderwillvarycontinuouslythroughoutanengine'sentirepowerstroke.Knowingwhatthosepressuresmaybeinagivenengineis useful,butmoreusefulstillisknowingwhattheyshouldandarelikelytobe,assuchknowledgecankeepyoufromthatfutileexercisecommonlyknownasfloggingadead horseandfrombelievingalotofliesabouthowmuchpowervariouspeoplearegettingfromtheirengines.Engineershaveanoverallefficiencyratingcalledbrakemean effectivepressure (bmep),whichtheycalculatebyworkingtheirwaybackthroughtorquereadingsobservedonthedynamometer,theleverageprovidedbycrankpin offset,andpistoncrownarea.Thus,bmepsayslittleaboutpeakcylinderpressures(thosemeasurementsbeingtakenwithapressuretransducerandoscilloscope)butitis anexcellentrelativeindicatorofperformanceandhighlyusefulinprojectingpoweroutputfromamodifiedengine.

PREDICTINGPOWER

Anaverage,welldevelopedstockengineintendedforuseinasports/touringmotorcyclewillhaveabmepofabout70psi.Itispossible,andImuststressthat word"possible",toraisethistoperhaps115psianimprovementofsome64percent,which(ifaccomplished)willyielda64percentincreaseinpoweroutputwithout raisingtheengine'soperatingspeed.Similarly,a64percentincreaseinoperatingspeedwithoutachangeinbmepwouldhavethesameeffectonoutput.Youwillseethis inthefollowingformulaforcalculatinghorsepower:

WhereBHPisbrakehorsepower Pisbrakemeaneffectivepressure,inpsi Lispistonstroke,infeet Aistheareaofonepiston,insquareinches Nisthenumberofpowerstrokesperminute Obviously,whenthevaluesofLandAareheldconstant,aswouldbethecasewithanenginehavingapistondisplacementatthelimitestablishedforaparticular racingclass,thenincreasesinpowermayonlybeobtainedbyincreasingthevaluesforPandNandyouwillfindthatinpracticeitisaloteasiertoincreasethelatterthan theformer. Asalreadystated,bmepfiguresforstock,touringtypeengineswithflowrestrictingaircleanersandmufflers,andwithporting/carburetioncompromisedinfavor

AdjustingMixture.................................................................................134 IgnitionFundamentals.............................................................................143 SparkPlugs.............................................................................................152

Throughoutthisbookitwillbeassumed,inconvenientthoughthatassumptionmayoccasionallybe,thatthereaderhasprogressedtoatleastasuperficial knowledgeofthemannerinwhichapistontypeinternalcombustionengine withparticularreferencetothoseoperatingonthetwostrokecycleprincipleconverts quantitiesoffuelandairintousefulpowerdeliveredattheendofitscrankshaft.Peoplewhoneedenlightenmentinthatregardwillfindawealthofexplanatoryliterature collectedontheshelvesofanypubliclibrarynorealpurposewouldbeservedbylingeringoverthematterhere.NeitherwillIattempttoinstructyouintheelementary mathematicsandphysicsrequiredtograspmuchofwhatfollows,asagainthepubliclibraryisanentirelyadequatesourceofinformation.Whatwillbeprovidedisakind ofstateoftheart reportabouthighspeed,highoutputtwostrokeenginesforlaymenwhoinmostcasesdonothaveaccesstotheliterature(SAEpapers,etc.) availabletoengineersandthusmustrelyuponhunches(oftenwrong)andfolklore(almostinvariablywrong)forguidance.Manyhavelearned,totheirsorrow,thatitis distinctlypossibletolavishenormousamountsoftimeandmoneyonthetwostrokeenginewithoutrealizingareturnappropriatetotheinvestment.Theinformationtobe providedherewillnotmakeyouaKaaden,orNaitoitwillhelpyoutoavoidsomeofthemoreseriousmistakes. Thefirstseriousmistakealaymanexperimentercanmakeistoassumethatthosewhodesignedandmanufacturedhisparticularenginedidn'tknowwhatthey weredoing.Inpointoffact,theprofessionalengineerknowsverywell,andiftheengineinquestionissomethingotherthanwhattheexperimenterhasinmind,thereare excellentreasons:allenginesarecompromised,fromwhatyoumightconsideranideal,intheinterestofmanufacturingeconomyandbroadusefulness.Forexample,ports mayderivetheirshapeasmuchfromwhatthedesignengineerintendedtobealowscrapratecastingasfromconsiderationofflowcharacteristics.Inotherwords,even somethinglikeportsdesignalwayswillbeinfluencedbythedemandsofmassproductionmanufacturing.Similarly,designingformassmarketsalesimpliesthatanengine mustbeagreeabletomanydifferentuseseventhoughthatinevitablymeansthatitwilldonosinglethingparticularlywell.Intheseareaswillwefindthelatitudefor improving anengine,andoneshouldalwaysbemindfulthattherealtaskissimplytotailoramassuseproducttoaveryspecificapplicationandthatinthetailoring processoneinevitablywillincurallthevariousexpensestheengine'sdesignerhasavoided.Hoursoflabormayberequiredtofinishroughcastportsdollarswillbespent correctingotherthingsthatarethecreaturesofmanufacturingeconomiespoweraddedatmaximumrevswillbepowersubtractedatlowercrankshaftspeeds,whilethe increasedspeedsrequiredtoobtainlargeimprovementsinpoweroutputwillbepaidforintermsofreliability. Anothermistakecommonlymade,sometimesevenbythosewhohaveenjoyedsomesuccessinmodifyingtwostrokeengines,istobelieveinakindof mechanisticmagic.Biggercarburetors,highercompressionratios,alteredporttimingsandexpansionchambersoftendobringanimprovementinpoweroutput,butmore andbiggerisnotmagically,instantlybetter.Allmustworkinconcertwiththebasicengine,directedtowardtheparticularapplication,beforetheyconstituteagenuine improvement.Youcannottreatthemasavoodooincantation,hopingthatifyoumuttertherightphrasesandstirthechickenentrailsintheprescribedmanner,yourmild mannered,allpurposechufferwillbetransformedintoahyperhorsepowerfirebreather.Withalotofluck,youmightgetthatresultthechancesheavilyarethatyou won't. Withallthemysticismfilteredout,horsepoweratanygivendisplacementissimplyafunctionofaveragepressureinthecylinderduringthepowerstrokeandthe rateatwhichpowerstrokesoccur,minusworkabsorbedbyfrictionandscavenging.Raisepressureand/orthedeliveryrateofthepowerstrokes,orreducefrictionand pumpinglosses,andtheengine'snetoutputwillrise.Unfortunately,therearelimitationsonallsides:Pressuremustbelimitedbecauseofthermalconsiderations(andis furtherlimitedbyanengine'srestrictedabilitytorechargeitscylinderwithafreshair/fuelmixturebetweenpowerstrokes).Thelimitforpowerstrokesperunitoftimeis establishedbywhatistolerableintermsofcrankshaftrotationalspeeds,andtolerablehereiswhatthebearings,rodandpistonwillsurvive,ininertialoadings,forwhat youconsideranacceptableservicelifethedesignengineerhasalreadyexpressedhisopinioninthismatter.Pumpinglossescanbereducedrelativetothemassflow throughanenginewithaproperlydesignedexhaustsystem,butotherwiseareaninevitableandalmostinvariableconsequenceofpullingairfromtheatmosphere,moving itthroughtheengine,andouttheexhaustport.Someimprovementinoutputmaybeobtainedwithreductionsinfriction,butthescopeforsuchimprovementsisverysmall comparedtowhatmaybeaccomplishedwithcylinderpressureandenginespeed. Obviously,pressureinacylinderwillvarycontinuouslythroughoutanengine'sentirepowerstroke.Knowingwhatthosepressuresmaybeinagivenengineis useful,butmoreusefulstillisknowingwhattheyshouldandarelikelytobe,assuchknowledgecankeepyoufromthatfutileexercisecommonlyknownasfloggingadead horseandfrombelievingalotofliesabouthowmuchpowervariouspeoplearegettingfromtheirengines.Engineershaveanoverallefficiencyratingcalledbrakemean effectivepressure (bmep),whichtheycalculatebyworkingtheirwaybackthroughtorquereadingsobservedonthedynamometer,theleverageprovidedbycrankpin offset,andpistoncrownarea.Thus,bmepsayslittleaboutpeakcylinderpressures(thosemeasurementsbeingtakenwithapressuretransducerandoscilloscope)butitis anexcellentrelativeindicatorofperformanceandhighlyusefulinprojectingpoweroutputfromamodifiedengine.

PREDICTINGPOWER

Anaverage,welldevelopedstockengineintendedforuseinasports/touringmotorcyclewillhaveabmepofabout70psi.Itispossible,andImuststressthat word"possible",toraisethistoperhaps115psianimprovementofsome64percent,which(ifaccomplished)willyielda64percentincreaseinpoweroutputwithout raisingtheengine'soperatingspeed.Similarly,a64percentincreaseinoperatingspeedwithoutachangeinbmepwouldhavethesameeffectonoutput.Youwillseethis inthefollowingformulaforcalculatinghorsepower:

WhereBHPisbrakehorsepower Pisbrakemeaneffectivepressure,inpsi Lispistonstroke,infeet Aistheareaofonepiston,insquareinches Nisthenumberofpowerstrokesperminute Obviously,whenthevaluesofLandAareheldconstant,aswouldbethecasewithanenginehavingapistondisplacementatthelimitestablishedforaparticular racingclass,thenincreasesinpowermayonlybeobtainedbyincreasingthevaluesforPandNandyouwillfindthatinpracticeitisaloteasiertoincreasethelatterthan theformer. Asalreadystated,bmepfiguresforstock,touringtypeengineswithflowrestrictingaircleanersandmufflers,andwithporting/carburetioncompromisedinfavor ofsmoothlowspeedrunning,willbearound70psi.Typicalfiguresforengineswithportingandotherplumbingarrangedsolely(andeffectively)formaximum horsepoweratpeakrevswouldbeabout115psiwithafewsmall,highlydevelopedtwostrokeenginesoperatingupat125psi.Theexactnumberwillvaryaccording tounitcylinderdisplacementandthewidthofanengine'susefulpowerband,butonemayreasonablyexpectthatenginessuitableformotocrosswillfallinthe8595psi rangewithbigcylinderstendingtowardthelowerfigureandsmallcylindersviceversa.Roadracingengines,tunedtoexertamaximumeffortoveraverynarrowspeed range,willusuallyshowabmepof100115psi,andofcoursethesameremarksregardingtheinfluenceofcylindersizeapply. Thesenumbershaveausefulnessbeyondthemeresatisfactionofvulgarcuriosity:theymaybeusedveryprofitablytodetermineanengine'ssuitabilityforsome particularapplication.Forexample,theyshedlightonthefutureprospectsofthosewhoaretryingtotransformKawasaki'sF5Bighornengine,a350ccsingle,intoa primemovercapableofendingtheYamahaTD2'sabsolutedominationinroadracing.Muchhasbeenmade,bytheKawasaki'ssupporters,oftheusefulnessofa broaderpowerrangeinherentwiththeF5'sdiscvalveinductionandthel00ccadvantageitgets,overtheTD2,byhavingonlyasinglecylinder(this,underthepresent AmericanMotorcycleAssociationrules).Nowwhileitistruethataracingmotorcyclehavingawidepowerbandiseasierforitsridertomanage,andmayofferan absoluteifveryslightadvantageonshort,extraordinarilytwistycircuits,onemustnotoverlookthefactthattheTD2hasbeenblessedwithanexcellentcloseratio transmissionandanumberofridersquitecapableofcopingwithanyproblemsintroducedbytheneedforfrequentgearchanges.Viewedrealistically,thesituationfacing

Thesenumbershaveausefulnessbeyondthemeresatisfactionofvulgarcuriosity:theymaybeusedveryprofitablytodetermineanengine'ssuitabilityforsome particularapplication.Forexample,theyshedlightonthefutureprospectsofthosewhoaretryingtotransformKawasaki'sF5Bighornengine,a350ccsingle,intoa primemovercapableofendingtheYamahaTD2'sabsolutedominationinroadracing.Muchhasbeenmade,bytheKawasaki'ssupporters,oftheusefulnessofa broaderpowerrangeinherentwiththeF5'sdiscvalveinductionandthel00ccadvantageitgets,overtheTD2,byhavingonlyasinglecylinder(this,underthepresent AmericanMotorcycleAssociationrules).Nowwhileitistruethataracingmotorcyclehavingawidepowerbandiseasierforitsridertomanage,andmayofferan absoluteifveryslightadvantageonshort,extraordinarilytwistycircuits,onemustnotoverlookthefactthattheTD2hasbeenblessedwithanexcellentcloseratio transmissionandanumberofridersquitecapableofcopingwithanyproblemsintroducedbytheneedforfrequentgearchanges.Viewedrealistically,thesituationfacing anyseriouschallengertoYamaha'ssupremacyisoneinwhichhorsepowermustbemetwithhorsepower.AndwhataretheKawasaki'sprospectsofdevelopingthatkind ofhorsepower?Let'shavealookatthenumbers: AssumingthatthemanwhomodifiestheKawasakiF5knowshisbusiness,butdoesn'thaveallthedevelopmenttimeintheworld,(probabilityfavorsthelatterfar morethantheformer)thenheverylikelywillarriveatacombinationofporting,etc.,goodforabmepofabout105psiwhichisaboutallthatcanbeexpectedwitha singlecylinderof350ccdisplacement.ToexpectmorewouldbetoignoretheconsiderabledifficultiesinscavengingefficientlytheF5'slargebore(3.17inch)cylinder. Furtherassuming(andasweshallseelater,thisassumptionisfarfromsafe)thattheF5enginewillremaininone,workingpieceforthedurationofalongishracewithits riderobservingaredlineof9000rpm,withapowerpeakat8500rpm,then, BHP= BHP=47.6 So,awelldevelopedF5woulddeliver47.6brakehorsepower.HowdoesthatcomparewiththeYamahaTD2?WithalltheyearsthathavegoneintotheTD 2'sdevelopment,andgivingduethoughttoYamaha'sprovenexpertiseinthesematters,itseemssafetoassumethatthisenginewouldbeoperatingwithabmepof115psi atitspowerpeakwhichseemstobeat11,000rpm.Thus,workingfromthosenumbersandthe250ccYamahatwin'sbore/strokedimensionsof56mmand50mm, respectively, BHP= BHP=48.0 Clearlythen,thosewhowouldtrytobeattheYamahawithaKawasakiF5havetakenuponthemselvesataskofconsiderablemagnitude.Theonlybrightspot inthepicture,forthem,isthatwhiletheyare0.4bhpdownontheYamaha(assumingnearoptimumworkontheirpart)theyprobablywillhavetheadvantageintermsof averagehorsepower,figuredfromthemomentagearisengagedwhenrevsfallsomewhatbelowthoseforpeakhorsepoweruntiltheredlineisreachedanditistime forachangetothenexthighergear.Therewillbenoadvantageinfrontalarea,foralthoughtheF5engineisnarrowerthanthatoftheTD2,thefairingmustbewide enoughtoshroudtherider,andtheminimumwidththatrequiresissufficienttoencompasseitherengine.Moreover,movingfromthetheoreticaltothepracticalfora moment,itishighlyunlikelythattheKawasakicouldbemadeasreliableat8500rpmasistheYamahaat11,000rpm,andnotbecausetheF5engineisbadlydesigned orshoddilyconstructed.Thesimpletruthisthatanysinglecylinder350ccenginewiththeF5'sbore/strokedimensionsandredlinedat9000rpmisgoingtobe stressedverynearitsabsolutelimitalimitimposedbythepropertiesofavailablematerials.

PISTONSPEED

Allthisasksthequestion,Howdoesonedeterminethelimit,withregardtoenginespeed?Unfortunately,establishingthislimitwithanyprecisionisnotonly extremelydifficultintermsofthemathematicsinvolved,butalsorequiresdataconcerningmetallurgy,etc.,seldomavailableoutsidetherecordroomsofthefactoriesfrom whichtheenginesoriginate.Still,thereareguidelineswhich,iflackinginabsoluteprecision,doatleasthavethevirtueofsimplicity,andwillprovideanindicatortokeep usawayfromcertaintrouble.Itisalmostimpossibletoestablishthepoint,inenginespeed,betweenzerotroubleandthepossibilityoftroublethereismuchlessdifficulty indeterminingaredlinebetweensometroubleandnothingbuttrouble. Aquickandeasymethodofestablishingalimitforcrankshaftspeedisbyworkingwithpistonspeed.Actually,with"mean"pistonspeed:pistonsdonottravelat uniformvelocitytheymovefromadeadstopateachendoftheirstroke,accelerateuptoamaximumspeedthatoftenisinexcessof120mph,andthenbraketoanother completestop.Forconvenience,weusejustthemeanpistonspeedandthesafelimitforthat,forengineshavingborestrokedimensionswithintherangeconsidered normalformotorcycles,isabout4000feetperminute.Andmeanpistonspeedmaybecalculatedveryeasilybyapplyingthefollowingformula: Cm=0.166xLxN Where:C mismeanpistonspeed,infeetperminute Lisstroke,ininches Niscrankshaftspeed,inrevolutionsperminute Thus,usingagaintheKawasakiF5engineasanexample,withLbeing2.68inchesandNgivenas9000,wefindthat Cm=0.166x2.68x9000 Cm=4000ft/min Herewehaveatheoreticallypredictedlimitthatseemstoagreequitecloselywithobservablerealityinthefield:ReportsfromthoseactuallyracingmodifiedF5 Kawasakisindicatethattheenginedoesinfactretainacceptable(withintheframeworkofthatword'smeaninginracing)reliabilitywhenredlinedat9000rpm,and ravelswithhorrifyingabruptnessifpressedfurther.Ofcourse,itmustbestressedherethatfewengines,theF5notexcepted,retainmorethanmarginalreliabilityatmean pistonspeedsof4000ft/min,andeventhispresupposesfrequentreplacementofthepistonandthecrank/rodbearings. Youwillbeonfarmoresolidgroundifyourengineisnotaskedtoenduremeanpistonspeedsabove3500ft/min.Anythingabovethattakesanengineintothe twilightzoneofreliability,andthegroundbetween3500ft/minandthenearabsolutelimitof4000ft/miniscoveredwithunpleasantpossibilities,buttheseoftenmaybe minimizedwiththeproperselectionofmaterialsandlubrication.Ishouldnoteherethatthereareexceptionstothisruleamongsomeoftheoldfashioned,longstroke engines,whichtendtohaveverylight(andstrong)reciprocatingpartsrelativetotheirabsolutestroke.AnexamplethatcomestomindistheBultaco125ccTSS,which hadastrokeofnoless2.36inches(decidedlylongfora125)butwhichwould,infactoryroadracingtrimrunupto11,500rpm,justliketheYamahaTD2(witha muchshorter,1.97inchstroke),andthatrepresentsameanpistonspeedof4500rpm.Obviously,Bultacoheldtheopinionthattheresultingthinishmarginofreliability wasacceptable,buttheirTSSneverwasaspredictablytroublefreeasYamaha'sTD2,whichatthesamecrankshaftspeed(11,500)hasameanpistonspeedofonly 3775ft/min. Whileonthesubjectofbore/strokedimensions,Iwouldliketosaythatthereismuchinfavoroflongstroketwostrokecycleenginesinmanyapplications.They arenotsuperior(asmanypeopleseemtothink)comparedtothepresentdayshortstrokedesignsintermsoflowspeedtorque,astorqueisentirelyafunctionof displacementandbmep,andwholelyunrelatedtobore/strokeratios.Withalongstroke,thereis(atanygivendisplacement)areductioninbore,andwithitalossof pistonareaagainstwhichgaspressurecanexertitsforce,thatexactlybalancesthelossofleverageinashortstrokeengine(whichis,inturn,compensatedbyagainin pistonarea).Theonlythingwrongwiththelongstrokeengineisthatitscrankshaftspeedislimitedbyinertialoadings,andthatinturnlimitsitsabsolutepowerpotentialas comparedwiththemodernshortstroker.Ontheotherhand,itiscompensatedbyhavingamuchmorecompactcombustionchamber,whichmakesformoreefficient burning,andbylowerthermalloadingsonthepistonasaresultofthesmallercrownareaintowhichheatfromthecombustionprocessmaysoak.Finally,thereisan advantageinportareaforthelongstrokedesignresultingfromitsrelativelylargecylinderwallarea.Thisareaincreasesinthelongstrokeenginebecausedisplacement

Aquickandeasymethodofestablishingalimitforcrankshaftspeedisbyworkingwithpistonspeed.Actually,with"mean"pistonspeed:pistonsdonottravelat uniformvelocitytheymovefromadeadstopateachendoftheirstroke,accelerateuptoamaximumspeedthatoftenisinexcessof120mph,andthenbraketoanother completestop.Forconvenience,weusejustthemeanpistonspeedandthesafelimitforthat,forengineshavingborestrokedimensionswithintherangeconsidered normalformotorcycles,isabout4000feetperminute.Andmeanpistonspeedmaybecalculatedveryeasilybyapplyingthefollowingformula: Cm=0.166xLxN Where:C mismeanpistonspeed,infeetperminute Lisstroke,ininches Niscrankshaftspeed,inrevolutionsperminute Thus,usingagaintheKawasakiF5engineasanexample,withLbeing2.68inchesandNgivenas9000,wefindthat Cm=0.166x2.68x9000 Cm=4000ft/min Herewehaveatheoreticallypredictedlimitthatseemstoagreequitecloselywithobservablerealityinthefield:ReportsfromthoseactuallyracingmodifiedF5 Kawasakisindicatethattheenginedoesinfactretainacceptable(withintheframeworkofthatword'smeaninginracing)reliabilitywhenredlinedat9000rpm,and ravelswithhorrifyingabruptnessifpressedfurther.Ofcourse,itmustbestressedherethatfewengines,theF5notexcepted,retainmorethanmarginalreliabilityatmean pistonspeedsof4000ft/min,andeventhispresupposesfrequentreplacementofthepistonandthecrank/rodbearings. Youwillbeonfarmoresolidgroundifyourengineisnotaskedtoenduremeanpistonspeedsabove3500ft/min.Anythingabovethattakesanengineintothe twilightzoneofreliability,andthegroundbetween3500ft/minandthenearabsolutelimitof4000ft/miniscoveredwithunpleasantpossibilities,buttheseoftenmaybe minimizedwiththeproperselectionofmaterialsandlubrication.Ishouldnoteherethatthereareexceptionstothisruleamongsomeoftheoldfashioned,longstroke engines,whichtendtohaveverylight(andstrong)reciprocatingpartsrelativetotheirabsolutestroke.AnexamplethatcomestomindistheBultaco125ccTSS,which hadastrokeofnoless2.36inches(decidedlylongfora125)butwhichwould,infactoryroadracingtrimrunupto11,500rpm,justliketheYamahaTD2(witha muchshorter,1.97inchstroke),andthatrepresentsameanpistonspeedof4500rpm.Obviously,Bultacoheldtheopinionthattheresultingthinishmarginofreliability wasacceptable,buttheirTSSneverwasaspredictablytroublefreeasYamaha'sTD2,whichatthesamecrankshaftspeed(11,500)hasameanpistonspeedofonly 3775ft/min. Whileonthesubjectofbore/strokedimensions,Iwouldliketosaythatthereismuchinfavoroflongstroketwostrokecycleenginesinmanyapplications.They arenotsuperior(asmanypeopleseemtothink)comparedtothepresentdayshortstrokedesignsintermsoflowspeedtorque,astorqueisentirelyafunctionof displacementandbmep,andwholelyunrelatedtobore/strokeratios.Withalongstroke,thereis(atanygivendisplacement)areductioninbore,andwithitalossof pistonareaagainstwhichgaspressurecanexertitsforce,thatexactlybalancesthelossofleverageinashortstrokeengine(whichis,inturn,compensatedbyagainin pistonarea).Theonlythingwrongwiththelongstrokeengineisthatitscrankshaftspeedislimitedbyinertialoadings,andthatinturnlimitsitsabsolutepowerpotentialas comparedwiththemodernshortstroker.Ontheotherhand,itiscompensatedbyhavingamuchmorecompactcombustionchamber,whichmakesformoreefficient burning,andbylowerthermalloadingsonthepistonasaresultofthesmallercrownareaintowhichheatfromthecombustionprocessmaysoak.Finally,thereisan advantageinportareaforthelongstrokedesignresultingfromitsrelativelylargecylinderwallarea.Thisareaincreasesinthelongstrokeenginebecausedisplacement risesonlyindirectproportiontostroke,butisincreasedbyafactorof3.14(theconstant,pi)withenlargementsinbore.Theseareveryrealadvantages,buttheyarenot enough,usually,toprevailagainsttheshortstrokeengine'ssheerabilitytorev.Crankshaftspeedistheonlythingsubjecttomuchjugglinginthehorsepowerequation andisafarmorepotentfactorindeterminingpoweroutputthantherelativelyslightimprovementsinbmepobtainablewiththemarginallybettercombustionchamberand portinginthelongstrokeengine.A10percentimprovementinourKawasakiF5engine'sbmep(alargeimprovementindeed)wouldraiseitsoutputto52.3bhpleave thebmepunchanged,butshortenthestrokeandspinit11,000rpmandyouwouldhave61.3bhp.Thereisindeednosubstituteforrevs.

PISTONACCELERATION
Sadly,whilethereisnosubstituteforrevs,thereareplentyofbarriers:pistonspeedisone,aswasalreadynoted.Butthatisaratherindirectlimit,asitignoresthe factthatitisnotspeedsomuchasallthestartingandstoppingofpistonsthatdoesthedamage,oratleasttheworstofanydamage.Theaccelerationforcesgeneratedby thestartingandstoppingarefelteveninanengine'smainbearings,buttheyareatapeakintheconnectingrodandpistonandhaveaparticularlydisastrouseffectonthe latter,asanyattempttomakeapistonstrongerisaptalsotomakeitheavierwhichaggravatestheverysituationthestrengtheningofthepistonshouldimprove.Evenso, anengine'strueAchillesheel,theproblemthatmaymoststronglyresistsolution,oftenisthedisastrouseffectspistonaccelerationmayhaveonthepiston'srings. Itoftenisthought,andquitewrongly,thatringsmaintainasealbetweenthepistonandthecylinder'swallssimplythroughtheirpropertiesassprings.Alittlethought shouldconvinceyouthatsuchcannotbethecase,formostrings,compressedintheprocessofinstallation,pressoutwardagainstthecylinderwithaforceamountingto about30psi.Gaspressureinthatcylindermayeasilyexceed750psi,anditshouldbeobvious

thata30psiforcewillnotholdbackonecirca 750psi.Still,equally obviously,pistonringsdoformaneffectiveseal.How?Becausetheygetalotofhelpfromthecylinderpressureitself:gaspressureabovetheringforcesitdownagainst thebottomofitsgrooveinthepiston,andalso(actingbehindthering,inthebackofthegroove)shovesitouthardagainstthecylinderwall.Thus,inthenormalcourseof

anengine'strueAchillesheel,theproblemthatmaymoststronglyresistsolution,oftenisthedisastrouseffectspistonaccelerationmayhaveonthepiston'srings. Itoftenisthought,andquitewrongly,thatringsmaintainasealbetweenthepistonandthecylinder'swallssimplythroughtheirpropertiesassprings.Alittlethought shouldconvinceyouthatsuchcannotbethecase,formostrings,compressedintheprocessofinstallation,pressoutwardagainstthecylinderwithaforceamountingto about30psi.Gaspressureinthatcylindermayeasilyexceed750psi,anditshouldbeobvious

thata30psiforcewillnotholdbackonecirca 750psi.Still,equally obviously,pistonringsdoformaneffectiveseal.How?Becausetheygetalotofhelpfromthecylinderpressureitself:gaspressureabovetheringforcesitdownagainst thebottomofitsgrooveinthepiston,andalso(actingbehindthering,inthebackofthegroove)shovesitouthardagainstthecylinderwall.Thus,inthenormalcourseof events,sealingpressureattheinterfacebetweencylinderwallandringalwaysiscomfortablyhigherthanthepressureitmustholdback. Thisverydesirablesituationwillbemaintainedunlesssomethinghappenstoupsetthings,andmostinsistentamongtheseveralsomethingsthatmayintrudeis excessivepistonacceleration.Whenpistonaccelerationexceedsthesumtotalofgaspressuresholdingtheringinplace,theringwillliftupward(asthepistonnearsthe topofitsstroke,andisbeingbrakedtoahalt).Instantly,astheringlifts,thegaspressurepreviouslyappliedaboveandbehindisalsoappliedunderneaththering,at whichpointitsinertiatakesovercompletelyandtheringslamsuphardagainstthetopofitsgroove.Thislastactionreleasesallpressurefrombehindthering,leavingit entirelytoitsownfeebledevicesinholdingbackthefireabove,andasits30psioutwardpressureisnomatchforthe750psipressureintheuppercylinder,itisblown violentlybackintoitsgroove.Thering'sradialcollapseopensadirectpathdownthecylinderwallforthehightemperatureandpressurecombustiongasesbutonlyfora microsecond,fortheactionjustdescribedinstantlyappliesgaspressureonceagainbehindtheringandthatsendsitssnappingbackintoplaceagainstthecylinderwall. Unhappily,itcannotremainthere,asgaspressureimmediatelybangsitbackintoitsgrooveagaintorepeattheprocessoverandoveruntilthepistonisvirtuallystopped andthering'sinertiaisnolongerenoughtocountergaspressure. Thenetresultofallthisactivityisthatoverthespanofseveraldegreesofcrankrotation,immediatelyprecedingthepiston'sreachingtopcenter,theringwillbe repeatedlycollapsedradiallyandatthesametimehammeredhardagainstthetopofitsgroove.Understandably,theringisdistressedbythis,asitnotonlyreceivesa fearfulbatteringbutalsoisbathedinfirewhilebeingdeprivedoftheclosecontactwithpistonandcylinderthatwouldotherwiseservetodrawoffheat.Equallydamaging isthatthepistonishavingmuchthesameproblem,withhightemperaturegasesblowingdownpastitsskirttocauseoverheating,toburnawaythefilmofoilbetweenitself andthecylinderwall,andwithitsring,orrings,allthewhiletryingtopoundtheirwayupthroughthepistoncrown.Amildcaseofwhatisquiteaccuratelytermed ring fluttereventuallyresultsinthedestructionoftheringandsometimesthedimensionalintegrityofitsgrooveamoreseriouscaseiscertaintoleadrapidlyintolubrication failure,overheating,andpistonseizure.Fortunately,thisdrasticproblemcanbeavoided,thankstotheworkoftheresearcherPauldeK.Dykes,whoseinvestigationof theringflutterphenomenonyieldedmostofwhatweknowaboutitandwhoinventedtheflutterresistantringthatbearshisname.Dykesshowedusthecauseofring flutter,andengineers'understandingofthecauseisreflectedintheirdesignsofthemodernpistonring,whichisverythin,axially,withaveryconsiderablewidth,radially. Thus,gaspressurebearsdownonalargesurface,providinganequallylargetotaldownforce,butisopposedbyarelativelysmallupwardloadasthering,beingthin,is lightandinconsequencehaslittleinertia.Still,evenwithverythinrings,flutterwilloccurifinertialoadingsarehighenough.Tosettlethequestion,withregardtoanygiven engine,applythefollowingformulafordeterminingmaximumpistonacceleration:

G max=
WhereGmax ismaximumpistonacceleration,infeetpersecondsquared Niscrankshaftspeed,inrevolutionsperminute Lisstroke,ininches Aistheratioofconnectingrodlength,betweencenters,tostroke

Toillustratehowhightheseforcesmaysometimesbe,let'suseasanexampletheYamahaTD2,using11,000rpmforN.Theformulatellsusthatatthatspeed, maximumpistonaccelerationwillbe(withtheanswerroundedoffbymyslideruleI'mtoolazytodoitallwithpaperandpencil)nolessthan135,000ft/sec2.Nowif youwillrecallforamomentthattheaccelerationofgravityisonly32ft/sec2,itwillbeclearthattheloadontheYamaha'spistonsandthusonitsringsisveryhigh indeed.ButistheloadinghighenoughtomaketheYamaha'sringsflutter?Obviously,itisnot,astheengineremainsnotonlyreliablebutcrispincomparativelylong races.Thelimit,fortheTD2engine,isslightlyhigherthan135,000ft/sec2butnotmuchhigher,asyouwillseeinthefollowingtablelistingringthicknessesandthe accelerationsatwhichtheybegintoflutter. Forringshavinga0.125inchthickness,40,000ft/sec 2 0.09453,000ft/sec 2 0.06380,000ft/sec 2 0.047106,000ft/sec 2 0.039138,000ft/sec 2 TheYamaha,withringshavingathicknessof1mm,or0.039inch,andamaximumpistonaccelerationof135,000ft/sec2at11,000rpm,wouldseemtobe operatingverynearthelimitasindeeditis.Butitprobablyisnotquiteasnearthelimitasthenumberssuggest,foraracingring(withitsexaggeratedthickness/width crosssectionalaspect)issomewhatlesssubjecttoflutterthanaringmadeforapplicationinatouringengine.Still,thenumbersgivenarefairlycloseforringswithnormal rangeproportions,andifyouhaveanenginewithringsforwhichButterispredictedat80,000ft/sec2andintendusingcrankshaftspeedsthatwouldraisemaximumpiston accelerationtosomethingmorelike100,000ft/sec 2,thenIstronglyurgeyoutofitnewpistonswiththinnerrings.Youmayinterpolatebetweenthefiguresgiventofindthe

0.039138,000ft/sec 2 TheYamaha,withringshavingathicknessof1mm,or0.039inch,andamaximumpistonaccelerationof135,000ft/sec2at11,000rpm,wouldseemtobe operatingverynearthelimitasindeeditis.Butitprobablyisnotquiteasnearthelimitasthenumberssuggest,foraracingring(withitsexaggeratedthickness/width crosssectionalaspect)issomewhatlesssubjecttoflutterthanaringmadeforapplicationinatouringengine.Still,thenumbersgivenarefairlycloseforringswithnormal rangeproportions,andifyouhaveanenginewithringsforwhichButterispredictedat80,000ft/sec2andintendusingcrankshaftspeedsthatwouldraisemaximumpiston accelerationtosomethingmorelike100,000ft/sec 2,thenIstronglyurgeyoutofitnewpistonswiththinnerrings.Youmayinterpolatebetweenthefiguresgiventofindthe safeaccelerationlevelsforringthicknessesnotlisted. Therearepistonringsthatresistverystronglypistonacceleration'seffortstowardmakingthemflutter.ThebestknownoftheseistheDykespatternring,which hasanLshapedcrosssectionandfitsintoasimilarlyshapedgrooveinthepiston.TheDykesringismadeflutterresistantbythefactthatitshorizontallegfitsquite closelyinitsgroove,ascomparedtoclearancesaroundtheverticalleg,andthereforeevenifaccelerationliftstheringitcannotlifthighenoughtocloseoffthepressure behindthering'sverticalleg.Inconsequence,thering'ssealingabilitiesaremaintainedataccelerationsthatwouldbetheundoingofringsintheconventionalrectangular sectionpattern.However,theDykesring'sabilitytomaintainasealdoesnotfreeitofalltheunpleasantnessattendingtoohighpistonacceleration:whileitmaysealunder

those conditions,itisstillbeingrattledaboutvigorouslyandiftherattling continueslongenough,theDykesring,andthegroovetryingtorestrainit,bothbecomebadlybattered.Atthatpoint,itsabilitytosealvanishesandmechanicalfailureof thering,piston,orboth,followsveryclosely.BultacohaslongusedDykespatternrings,ashavecertainothers,butmostmanufacturerspreferringsthatdonotrequire suchcarefulandintricatemachining.Thereareotherflutterresistantrings,andmanyexcellentreasonsforusingringsofconventionalconfiguration,butthesedetailsare discussedelsewhereinthisbookandingreaterdepththanwouldbeappropriatehere. Afterestablishingallthesemechanicallimits,withregardtopistonspeedandacceleration,andafterdecidinghowmuchpoweryouarelikelytogetfroma particularengine,youshouldsubjecttheenginetoacompletesurvey.Thiswouldincludethemeasuringofportheightsandwidths,combustionchamberandcrankcase volumes,andchartingpistontravelagainstcrankrotation.Thislasteffortmayatfirstseemratherpointless,butasyourworkprogressesyouwillfindthatthechart,which willshowalmostbutnotquiteasinecurve,providesaninstantreadoutbetweendegreesatthecrankshaftandthepositionofthepistonfromtopcenterthatismostuseful. Itwilltellyou,forexample,howmuchtoraisethetopedgeofanexhaustporttomakeagivenchangeintiming,andhowmuchtotrimfromthepistonskirt(inapiston portengine)togettheintakeperiodyouwantorthinkyouwant.Thechartalsowillprovideyouwithallthemeanportopenpoints,anditwillprovideanexceedingly usefulrelationshipbetweenignitiontimingexpressedindegreesandinpistontravelfromtopcenter.Youmaydeviseyourownmethodsforderivingallthisinformation accordingtoyourpreferenceandresourcesIhaveexplainedmyowntechniqueselsewhereinthistext,intheappropriatechapters. Anitemthatmustbeincludedinanydiscussionofthetwostrokecycleengine'sbasicsisgeneralgasdynamics.Youcangetinformationonthesubjectatyour locallibrary,buttheapplicableparticularsarelikelytobewidelyscatteredthere,soIwillcoverthesubjectinbriefhere.Themannerinwhichwhatfollowsappliesat specificpointsthroughouttheengineanditsrelatedplumbingwillbecoveredlater,butyoushouldknowafewofthefundamentalsnowandthussavemefrombecoming unnecessarilyrepetitiouslater. Onethingyoumustknow,forexample,isthattheairmovingthroughtheengine,amixtureofgases,hasmanyofthepropertiesofafluid.Itevenhastheabilityto wetasurface,andhasviscosity,whichmeansthatairwillclingtoallsurfaceswithinanengineinalayerthatmoveshardlyatallnomatterwhatthemidstreamvelocity maybe.Thisboundarylayer'sdepthisinfluencedbygastemperature,andbythetemperatureofthesurfaceonwhichitforms,aswellasbytheshapeofthesurface. Pleaseunderstandthatthelayerisnotsoliditisshearingwithgeneralflowthroughoutitsdepthwhichmaybeasmuchas0.100inchwithmovementincreasingas todistancefromthesurfaceonwhichitisformed.Andascloseas0.020inchfromthesurface,flowmaystillbeintheorderof80percentofthatinmidstream,which meansthattherestrictionformedbytheboundarylayerisnotverygreat.Nonetheless,itisthere,anditaccountsforsuchthingsasroundportshavinglessresistanceto flowthansquareports,areaforarea,andfortheabilityofasingleporttomatchtheflowofapairofportsofsomewhatlargerarea.Italsoaccountsforthefactthatflow resistanceincreasesindirectproportionwiththelengthofaport,andmuchoftheresistanceresultingfromtheshapeofaparticularportisduetothatshape'screatinga thickboundarylayer,whichbecomesliterallyapluginsidetheport. Generallyspeaking,boundarylayerswillbeheldtominimumdepthonsurfacesthatrise(relativetothedirectionofflow)andgaininthicknessonanysurface thatfallsaway.Thus,anintaketrumpet,forexample,shouldbetaperedinslightlyfromtheinletendtothecarburetorbyperhaps23degreesintheinterestofholding boundarylayerthicknesstoaminimum.Inthatconfiguration,itwillhaveappreciablylessresistancetoflowthanastraight,parallelwalltube.Similarly,transferports shoulddiminishincrosssectionalareafromtheirentranceinthecrankcasetowardtheiroutletinthecylinder. Thesegasesalsohaveinertia:oncesetinmotiontheytendtoremaininmotionwhenatresttheyresistalleffortstogetthemmoving.Inpractice,thismeansthat therealwaysisalagbetweentheintakeport'sopeningandthemovementofairintheintaketract.Fortunately,thislagcanbeamplycompensatedtowardtheendofthe intakeperiod,whenthepressureinsidethecrankcasehasrisentoalevelthatshouldpushpartofthechargebackouttheportbutcannotbecauseoftheeffectofinertia ontheincominggases.Inertiaalsohasitseffectontheflowofgasesthroughthetransferportsandouttheexhaustsystem,butIwilldealwiththatwhiletreatingthose subjectsseparately. Theseinertiaeffectsareuseful,butdifficulttomanageassomethingapartfromotherprocessesoccurringastheengineruns.Forexample,intaketractlength usuallyisestablishedmorewithaneyetowardresonancesthaninertia,anditsdiametersetbytheflowraterequiredbythecarburetortometerproperlybalancedagainst theresistancethatattendshighgasvelocities.Therefore,virtuallytheonlythingwecandoaboutinertiaeffectsistoattempttofindtheintaketimingthatwillmake maximumuseofthoseprovidedbyanintakesystemproportionedmostlytosuitotherrequirements. Resonancesareanothermatter.Soundwaveswilltravelthroughanyelasticmedium,suchasair,andintheirpassagetheypulltogetherorforceapartmolecules, justasthesimilarenergywavestravelingthroughtheoceanpullthewaterintopeaksandtroughsonitssurface.And,asintheocean,thewavesmovesteadilyonward

subjectsseparately. Theseinertiaeffectsareuseful,butdifficulttomanageassomethingapartfromotherprocessesoccurringastheengineruns.Forexample,intaketractlength usuallyisestablishedmorewithaneyetowardresonancesthaninertia,anditsdiametersetbytheflowraterequiredbythecarburetortometerproperlybalancedagainst theresistancethatattendshighgasvelocities.Therefore,virtuallytheonlythingwecandoaboutinertiaeffectsistoattempttofindtheintaketimingthatwillmake maximumuseofthoseprovidedbyanintakesystemproportionedmostlytosuitotherrequirements. Resonancesareanothermatter.Soundwaveswilltravelthroughanyelasticmedium,suchasair,andintheirpassagetheypulltogetherorforceapartmolecules, justasthesimilarenergywavestravelingthroughtheoceanpullthewaterintopeaksandtroughsonitssurface.And,asintheocean,thewavesmovesteadilyonward awayfromtheirsourcebutthetransmittingmediumdoesnot.Take,forexample,theactivitysurroundingasinglecondensation,orpositivepressurewave,asitmoves throughtheair.Initscenter,moleculeshavebeenpulledtogether,condensed,butasittravelsitreleasesthosemoleculesandcompressesothersasitreachesthem.In thesamemanner,ararefaction,ornegativepressurewave,pushesmoleculesapart.Bothwavesbehaveinacurious,butusefulwaywhenconfinedinatubeandthe effectsofinertiaaremixedwiththem.Foronething,theywillbereflectedbackwhenreachingtheendofthetubewhetherthatendisopenorclosed.Butatthetube's openend,thewavechangesinsign:acondensationisinvertedandbecomesararefaction,andviceversaattheclosedend,thewavewillbereflected,butretainsits sign. Howisallthatuseful?Forexample,intheintakesystemtheopeningoftheintakeportexposesthecrankcaseendofthetracttoapartialvacuum,andthatinturn sendsararefactionshootingofftowardtheopposite,atmospheric,endofthetract.Ittravelsouttotheintakebell,invertsinsigntobecomeacondensation,andinstantly movesbacktowardthecrankcasetoarrivethereasaclumpofcompressedmolecules,whichsurgeintothecrankcasetobetrapped,ifthepistonthenclosestheintake port,aspartofthescavengingcharge.Thateffect,overlayedwithinertiaintheinrushinggases,makesallthedifferenceingettingthejobofchargingdoneintwostroke engineswhichprovideonlyanabsurdlyshorttimeforsuchchores. Howshortatime?Thatisatthesametimeoneoftheleastcomplicatedandmostdepressingcalculationsyoucanperform.LetusconsidertheYamahaDT1, whichinfullydevelopedconfigurationhadanintakedurationof160degrees,atransferdurationof123degrees,andanexhaustdurationof172degrees.Yamahaclaims apowerpeakat7000rpm.Let'shavealookattheactualtime,infractionsofasecond,availableforthecompletionofthesefunctions.Toarriveatthesetimes,usethe followingformula: WhereTistime,inseconds Niscrankshaftspeed,inrevolutionsperminute isportopenduration,indegrees (Thisformulacanbeabbreviatedto ) Thus,tofindTforthe160degreeintakeduration, . Withapplicationofthesameformulatothetransferandexhaustperiods,wefindthattheformerisopen0.0029second,andthelatteropen0.0041second. Eventhelongestofthese,theexhaustopenduration,isonly41/10,000second,andthatisnotverymuchtimeinwhichtoemptyexhaustgasesoutofthecylinder. Actually,thatparticularprocessissubstantiallyfinishedinthe29degrees,or0.0007second,betweenexhaustandtransferopening.Inthatshortperiod,pressureinthe cylindermustfalltosomethingverynearatmospheric,ortheexhaustgaseswouldforcetheirwaydownintothecrankcasethroughthetransferports.Ofcourse,the exhaustgasesareprovidedquitealargeaperturebymeansofwhichtheymaymaketheirescape,andthattheydoso,successfully,islessremarkablethanthefactthatthe freshchargecompressedinatwostrokeengine'scrankcaseisabletomakeitswaythroughthefarmorerestrictedtransferports,propelledbyafarlowerpressure,to refillthecylinderintheextremelybriefmomentavailable.Itseemsnothingshortofastonishingthatthisrechargingoperationisaccomplishedinthe0.0027secprovided bytheYamahaDT1's114degreetransferperiodthatthesameprocesstakesplaceinaYamahaTD2engineinonly0.0017secappearsaminormiracle.Obviously, divineinterventionisnotreallyafactorinthefunctioningoftwostrokeengines,andcylinderrechargingispossiblesimplybecausetheprocessgetsalotofhelpfromthe activitiesoftheexhaustsystem,gasvelocitiesthroughthetransferportshaveameanvalueintheorderof300ft/sec,andthecrosssectionalareasoftheportsinvolved arerelativelylargeascomparedwiththevolumeofgasestobetransferred. Asithappens,itispossibletocalculatecorrectcombinationsofportopentimesandportareasforanymotorcycleengine,atanyenginespeed.Themaximum safespeedforanyengineisalsocalculable,asexplainedearlierinthischapter,alongwithexpansionchamberdimensions,carburetorsizeandmanyotherfactors influencingbothmaximumpoweroutputandoverallpowercharacteristics.Itshouldbenotedherethatnoneofthevaluesderivedpurelyfromcalculationsarenecessarily optima,andfineadjustmentsmustalwaysbemadeexperimentally,butitisfarbettertoemploythesimpleformulaepresentedinthechapterstofollowthantoattempta purelyexperimentalapproach.Themathematicsinvolvedarenotterriblycomplicated,thoughsometimesthearithmeticislaborious,andyoucanusepaperandpencilto arriveatabasicengine/pipecombinationthatwillbeveryneartheoptimum.Muchnearer,infact,thanwouldbeobtainedbyeventhemostexperiencedtuner's unsupportedguesswork,andnearenoughtoafullydevelopedconfigurationtominimizetheoutlayoftimeandmoneyentailedinthebuildingofaracingengine.Youstart bydetermining,mathematically,anupperlimitforenginespeed,thenusemoremathinestablishingamaximumforpistonringthickness,inestablishingalltheport dimensionstosuittheprojectedenginespeed,inselectingacarburetor,andindesigninganexpansionchamber.Suitablevaluesforcompressionratios,bothprimaryand secondary,areprovidedinthechaptersdealingwithcrankcasepumpingandcylinderheads,respectively,andwiththerestofthematerialincludedinthisbookitalladds uptobeingafairlycompleteengineredesignmanualforthetwostrokeenginefixatedtuner.Myownexperienceindicatesthatenginesbuiltalongthelinessuggested hereneverfailtodeliverhighspecifichorsepower(whichismorethanmaybesaidforanycutandtrysystem)evenwithoutthebenefitofexperimentindicated adjustments.Idislikeguesswork,havemadeaseriousefforttoeliminateitfrommyownprojects,andamhopefulthatthelessonslearnedandoutlinedinthistextwill reducethegenerallyhighlevelofguessworkamongmostexperimenters.IfIhaveforgottentocoveranything,theomissionisinadvertent,becausemydistasteforSpeed Secretsisevengreaterthanforguesswork.ThereisonlyoneSecretinthegame:toknowwhatyouaredoing,andtodoitthoroughly.

Aswasnotedinthechapterofthisbookdealingwithbasics,poweroutputfromanengineofanygivendisplacementisafunctionofgaspressureinthecylinder duringthepowerstroke,andthenumberofpowerstrokesperunittime.Implicitthereinisthesuggestionthatthehorsepowerultimatelytobehadfromanenginehaslittle todowithportshapesandporttimings,exhaustsystems,carburetionorindeedanyofthethingsonwhichourattentionusuallyisfixed.Why?Foronething,increasesin gaspressurebringcorrespondingincreasesinheatflowintothepistonandnohighoutputtwostrokeenginecanoperatebeyonditsthermallimit.Similarly,youcannot increasetherateatwhichpowerstrokesoccurwithoutincreasingcrankshaftspeeds,withincreasesinthisdirectionsoonerorlatertakingyoubeyondtheengine's mechanicallimit.Thehorsepoweryouultimatelywillextractfromanygivenenginedependsthereforeverydirectlyuponyourabilitytoexpandthosethermaland mechanicallimitstothegreatestextentpossible,andonlythentomakethemostoftheterritorythusgained.

THEPISTON

reducethegenerallyhighlevelofguessworkamongmostexperimenters.IfIhaveforgottentocoveranything,theomissionisinadvertent,becausemydistasteforSpeed Secretsisevengreaterthanforguesswork.ThereisonlyoneSecretinthegame:toknowwhatyouaredoing,andtodoitthoroughly.

Aswasnotedinthechapterofthisbookdealingwithbasics,poweroutputfromanengineofanygivendisplacementisafunctionofgaspressureinthecylinder duringthepowerstroke,andthenumberofpowerstrokesperunittime.Implicitthereinisthesuggestionthatthehorsepowerultimatelytobehadfromanenginehaslittle todowithportshapesandporttimings,exhaustsystems,carburetionorindeedanyofthethingsonwhichourattentionusuallyisfixed.Why?Foronething,increasesin gaspressurebringcorrespondingincreasesinheatflowintothepistonandnohighoutputtwostrokeenginecanoperatebeyonditsthermallimit.Similarly,youcannot increasetherateatwhichpowerstrokesoccurwithoutincreasingcrankshaftspeeds,withincreasesinthisdirectionsoonerorlatertakingyoubeyondtheengine's mechanicallimit.Thehorsepoweryouultimatelywillextractfromanygivenenginedependsthereforeverydirectlyuponyourabilitytoexpandthosethermaland mechanicallimitstothegreatestextentpossible,andonlythentomakethemostoftheterritorythusgained.

THEPISTON
ForaverylongtimesubsequenttoDugaldClerk'screationofthetwostrokeengine,thethermallimitwastheonlylimit,butitwasenoughtoholdpoweroutput fromsuchenginestoextremelymodestlevels.Then,asnow,itwasprimarilyalimitimposedbyavailablepistonmaterials.Castironhasitsadvantagesintermsofwear resistance,hotstrengthandlowthermalexpansionrates,anditwasusedquitefrequentlyinthelowspeedenginesofyearspast.Unfortunately,ironisheavy,andheavyis thelastthingyouwantinapistonwhichinmodernenginesissubjectedtoaccelerationswellinexcessof100,000ft/sec 2.Aluminum,usedastheprimaryconstituentin virtuallyallpistonalloystoday,isconvenientlylight,butdisagreeablyinsistsonmeltingatmuchlowertemperaturesthanthatofthefiretowhichitisdirectlyexposed. Moreover,itlosesstrengthveryrapidlywithincreasesintemperatureaboveambient,sothatpistonfailuresdooccuratcrowntemperatureswellbelowthematerial's meltingpoint.Finally,aluminumisahighexpansionratemetal,whichmakesapistonmadeofitavariableclearancefitinanycylinder.Butaluminumisaverylightmetal, andthatalonewasenoughtorecommenditforuseinpistons,eventhoughthedrawbackslistedwereenoughtoseverelylimitthespecificpoweroutputsattainablewith twostrokeenginesforalongtime. Aluminumbasedpistonalloysimprovedslowlyovertheyears,withtheadditionofsmallpercentagesof,say,copper,toimprovetheirhotstrength,butitwasnot untilmeanswerefoundtoaddconsiderableamountsofsiliconthatlargeimprovementsweremade.Today,thebestpistonalloyscontainbetween15and25percent silicon,andthisadditionhasallbuttransformedthealuminumpiston.Admixturesofsiliconinexcessof15percentnotonlydrasticallyreducealuminum'sexpansion rate,theyalsoaffectaproportionateincreaseinhotstrengthandimprovethepiston'swearresistantproperties.Inalloftheserespectstheimprovementislargeenough toalmostexactlyequalthepercentagegainsinhorsepowerduringtheyearsinwhichaluminumsiliconalloyshavebeeninuse.Iaminclinedtothinkthatmostofwhatwe considertobemodernimprovementsintwostrokeenginedesignwithparticularreferencetoexpansionchambertypeexhaustsystemsmighthavebeenappliedas muchasfiftyyearsagohadgoodpistonsbeenavailable.Therewaslittlepointinsuchdevelopmentworkwithoutthealuminumsiliconpistonaluminumoraluminum copperpistonswouldmeltatspecificpoweroutputswellbelowwhatwenowconsideronlyaverage. Withallthat,highsiliconcontentpistonalloysstillarenotuniversallyemployed.Asithappens,suchalloysdohavetheirdisadvantage,whichisthattheyare difficulttomanufacture.Justcastingpistonsofaluminumsiliconalloyisataskforspecialistsusingspecializedequipmentmachiningtherawcastingsintofinishedpistonsis anevenmoreformidabletask.Youmayencounterthislastdifficultyifyouhaveoccasiontomodifyacylindercastfromthematerialinquestionandyouwillfindthatit bluntscuttingtoolsofanykindwithremarkablerapidity.Foryou,thatwillbeaninconvenienceforthemassproducerofpistonsitisadisaster,astheneedforfrequent resharpeningoftoolbitsentailslosingoutputfromhismachinerywhilesuchrepairsaremade,anditmeanstheexpenseofthemanhoursrequiredfortherepairs.Thus, themanufacturerhaseveryreasontorestrictthesiliconcontentofthepistonalloysheusestotheminimumrequiredbytheusetowhichhisengineswillbeput,whichisthe reasonwhyYamaha,forexample,usesdifferentalloysfortouringandracingpistons. Inpointoffact,theJapaneseseemtomanagehighsiliconcontentpistonsbetterthananyoneelse,whichmaywellaccountfortheirnotablesuperiorityincoaxing powerfromtwostrokemotorcycleengines.AllofthemajorJapanesemanufacturersemploypistonalloysintheirtouringengineshavingpercentagesofsiliconhigh enoughtobeconsideredracingonlyinmuchoftherestoftheworld.And,sadtosay,manyoftheracingpistonsbeingofferedbyspeedequipmentmanufacturersare inferiorinthisregardtotheordinaryofftheshelfpartsyou'llfindatyourlocaldealerinJapanesemotorcycles.Forthatreason,IaminclinedtouseeitherstockorGYT kitpistonswhenIamworkingwithenginescarryingamadeinJapanlabel,ratherthanwastemymoneyonaspecialtyreplacement.Thereare,ofcourse,exceptionsto thisrule,whichevolveprincipallyaroundringwidths,andIwilldealwiththatinduecourse. Unlessyouhappentobeapistonmanufacturer,thereisn'tmuchyoucandoaboutpistonalloys,beyondseekingoutpistonshavingahighsiliconcontent.Neither i s t h e r e a n y t h i n g y o u c a n d o a b o u t p i s t o n s h a p e whichismostunfortunate,becauseapistonis

not,asitfirstappears,simplycylindrical.Evenwiththeuseofaluminum siliconalloys,pistonsdoexpandastheyareheated,andtheydonotexpandatallevenly.Thegreatestincreaseindiameterwilloccurupatthecrown,becausethatis boththeareaofmaximummassandhighesttemperature.Sotheremustbemoreclearance,measuredcold,upatthepiston'scrownthanisrequireddownaroundthe lowerskirt.Infact,clearancesvarycontinuouslyfromthepiston'scrowntothebottomofitsskirtandfromsidetoside,asthepistonisellipticalratherthanround. Someday,someonemaybeable,withthehelpofacomputer,toactuallycalculatealltheclearancesandellipseratiosinvolvedforthepresenttheyaredecidedina processoftrialanderrorbyeventhemostexperiencedofmanufacturers.

Secretsisevengreaterthanforguesswork.ThereisonlyoneSecretinthegame:toknowwhatyouaredoing,andtodoitthoroughly.

Aswasnotedinthechapterofthisbookdealingwithbasics,poweroutputfromanengineofanygivendisplacementisafunctionofgaspressureinthecylinder duringthepowerstroke,andthenumberofpowerstrokesperunittime.Implicitthereinisthesuggestionthatthehorsepowerultimatelytobehadfromanenginehaslittle todowithportshapesandporttimings,exhaustsystems,carburetionorindeedanyofthethingsonwhichourattentionusuallyisfixed.Why?Foronething,increasesin gaspressurebringcorrespondingincreasesinheatflowintothepistonandnohighoutputtwostrokeenginecanoperatebeyonditsthermallimit.Similarly,youcannot increasetherateatwhichpowerstrokesoccurwithoutincreasingcrankshaftspeeds,withincreasesinthisdirectionsoonerorlatertakingyoubeyondtheengine's mechanicallimit.Thehorsepoweryouultimatelywillextractfromanygivenenginedependsthereforeverydirectlyuponyourabilitytoexpandthosethermaland mechanicallimitstothegreatestextentpossible,andonlythentomakethemostoftheterritorythusgained.

THEPISTON
ForaverylongtimesubsequenttoDugaldClerk'screationofthetwostrokeengine,thethermallimitwastheonlylimit,butitwasenoughtoholdpoweroutput fromsuchenginestoextremelymodestlevels.Then,asnow,itwasprimarilyalimitimposedbyavailablepistonmaterials.Castironhasitsadvantagesintermsofwear resistance,hotstrengthandlowthermalexpansionrates,anditwasusedquitefrequentlyinthelowspeedenginesofyearspast.Unfortunately,ironisheavy,andheavyis thelastthingyouwantinapistonwhichinmodernenginesissubjectedtoaccelerationswellinexcessof100,000ft/sec 2.Aluminum,usedastheprimaryconstituentin virtuallyallpistonalloystoday,isconvenientlylight,butdisagreeablyinsistsonmeltingatmuchlowertemperaturesthanthatofthefiretowhichitisdirectlyexposed. Moreover,itlosesstrengthveryrapidlywithincreasesintemperatureaboveambient,sothatpistonfailuresdooccuratcrowntemperatureswellbelowthematerial's meltingpoint.Finally,aluminumisahighexpansionratemetal,whichmakesapistonmadeofitavariableclearancefitinanycylinder.Butaluminumisaverylightmetal, andthatalonewasenoughtorecommenditforuseinpistons,eventhoughthedrawbackslistedwereenoughtoseverelylimitthespecificpoweroutputsattainablewith twostrokeenginesforalongtime. Aluminumbasedpistonalloysimprovedslowlyovertheyears,withtheadditionofsmallpercentagesof,say,copper,toimprovetheirhotstrength,butitwasnot untilmeanswerefoundtoaddconsiderableamountsofsiliconthatlargeimprovementsweremade.Today,thebestpistonalloyscontainbetween15and25percent silicon,andthisadditionhasallbuttransformedthealuminumpiston.Admixturesofsiliconinexcessof15percentnotonlydrasticallyreducealuminum'sexpansion rate,theyalsoaffectaproportionateincreaseinhotstrengthandimprovethepiston'swearresistantproperties.Inalloftheserespectstheimprovementislargeenough toalmostexactlyequalthepercentagegainsinhorsepowerduringtheyearsinwhichaluminumsiliconalloyshavebeeninuse.Iaminclinedtothinkthatmostofwhatwe considertobemodernimprovementsintwostrokeenginedesignwithparticularreferencetoexpansionchambertypeexhaustsystemsmighthavebeenappliedas muchasfiftyyearsagohadgoodpistonsbeenavailable.Therewaslittlepointinsuchdevelopmentworkwithoutthealuminumsiliconpistonaluminumoraluminum copperpistonswouldmeltatspecificpoweroutputswellbelowwhatwenowconsideronlyaverage. Withallthat,highsiliconcontentpistonalloysstillarenotuniversallyemployed.Asithappens,suchalloysdohavetheirdisadvantage,whichisthattheyare difficulttomanufacture.Justcastingpistonsofaluminumsiliconalloyisataskforspecialistsusingspecializedequipmentmachiningtherawcastingsintofinishedpistonsis anevenmoreformidabletask.Youmayencounterthislastdifficultyifyouhaveoccasiontomodifyacylindercastfromthematerialinquestionandyouwillfindthatit bluntscuttingtoolsofanykindwithremarkablerapidity.Foryou,thatwillbeaninconvenienceforthemassproducerofpistonsitisadisaster,astheneedforfrequent resharpeningoftoolbitsentailslosingoutputfromhismachinerywhilesuchrepairsaremade,anditmeanstheexpenseofthemanhoursrequiredfortherepairs.Thus, themanufacturerhaseveryreasontorestrictthesiliconcontentofthepistonalloysheusestotheminimumrequiredbytheusetowhichhisengineswillbeput,whichisthe reasonwhyYamaha,forexample,usesdifferentalloysfortouringandracingpistons. Inpointoffact,theJapaneseseemtomanagehighsiliconcontentpistonsbetterthananyoneelse,whichmaywellaccountfortheirnotablesuperiorityincoaxing powerfromtwostrokemotorcycleengines.AllofthemajorJapanesemanufacturersemploypistonalloysintheirtouringengineshavingpercentagesofsiliconhigh enoughtobeconsideredracingonlyinmuchoftherestoftheworld.And,sadtosay,manyoftheracingpistonsbeingofferedbyspeedequipmentmanufacturersare inferiorinthisregardtotheordinaryofftheshelfpartsyou'llfindatyourlocaldealerinJapanesemotorcycles.Forthatreason,IaminclinedtouseeitherstockorGYT kitpistonswhenIamworkingwithenginescarryingamadeinJapanlabel,ratherthanwastemymoneyonaspecialtyreplacement.Thereare,ofcourse,exceptionsto thisrule,whichevolveprincipallyaroundringwidths,andIwilldealwiththatinduecourse. Unlessyouhappentobeapistonmanufacturer,thereisn'tmuchyoucandoaboutpistonalloys,beyondseekingoutpistonshavingahighsiliconcontent.Neither i s t h e r e a n y t h i n g y o u c a n d o a b o u t p i s t o n s h a p e whichismostunfortunate,becauseapistonis

not,asitfirstappears,simplycylindrical.Evenwiththeuseofaluminum siliconalloys,pistonsdoexpandastheyareheated,andtheydonotexpandatallevenly.Thegreatestincreaseindiameterwilloccurupatthecrown,becausethatis boththeareaofmaximummassandhighesttemperature.Sotheremustbemoreclearance,measuredcold,upatthepiston'scrownthanisrequireddownaroundthe lowerskirt.Infact,clearancesvarycontinuouslyfromthepiston'scrowntothebottomofitsskirtandfromsidetoside,asthepistonisellipticalratherthanround. Someday,someonemaybeable,withthehelpofacomputer,toactuallycalculatealltheclearancesandellipseratiosinvolvedforthepresenttheyaredecidedina processoftrialanderrorbyeventhemostexperiencedofmanufacturers. Presumably,youwillnothavethefacilitiestoalterwhatevershapeyourengine'spiston(s)mayhave,butyoucanvaryrunningclearancesbychangingcylinderbore

not,asitfirstappears,simplycylindrical.Evenwiththeuseofaluminum siliconalloys,pistonsdoexpandastheyareheated,andtheydonotexpandatallevenly.Thegreatestincreaseindiameterwilloccurupatthecrown,becausethatis boththeareaofmaximummassandhighesttemperature.Sotheremustbemoreclearance,measuredcold,upatthepiston'scrownthanisrequireddownaroundthe lowerskirt.Infact,clearancesvarycontinuouslyfromthepiston'scrowntothebottomofitsskirtandfromsidetoside,asthepistonisellipticalratherthanround. Someday,someonemaybeable,withthehelpofacomputer,toactuallycalculatealltheclearancesandellipseratiosinvolvedforthepresenttheyaredecidedina processoftrialanderrorbyeventhemostexperiencedofmanufacturers. Presumably,youwillnothavethefacilitiestoalterwhatevershapeyourengine'spiston(s)mayhave,butyoucanvaryrunningclearancesbychangingcylinderbore diameter.TheproblemhereisoneofHowmuch?andIregrettosaythatitisaproblemforwhichthereisnoconvenientsolution.Clearances,measuredatthepiston's maximumdiameter,acrossitsthrustfaces,mayvaryfromabout0.002toasmuchas0.007inch,dependingon:theshapeandcompositionofthepistonitselftheabsolute cylinderborediameterthematerialfromwhichthecylinderismade,aswellasitsconfigurationandthethermalloadingstowhichthepistonwillbesubjectedwhichwill themselvesvaryaccordingtogaspressure,fuelmixture,cylinderconfigurationandthevehicle'srateofmotion.Manypeoplehaveexpressedgreatfaithinrulesrelating clearancetocylinderborediameterIhavenotfoundthechoicetobethatsimple.Ifthereisarule,itwouldbethatyoucanaddperhaps0.0005to0.001inchtothe clearancerecommendedbyyourengine'smaker,buteventhisisagrossoversimplificationandImentionitonlybecauseitissomewhatbettertohavetoomuchclearance thantoolittle.Intheformer,theexcessiveclearanceadverselyinfluencesheattransferfromthepistontotherelativelycoolercylinderwallsandmayleadtoanyofthe severalunpleasantriesassociatedwithoverheatingthepiston,whichrangefromatendencyforoiltobecomecarbonizedintheringgrooves,totheappearanceofalarge holeinthepistoncrown.Toolittleclearancewillrevealitselfintheformofscuffing,oroutrightseizureunlessthepistonisonlymarginallytootight,inwhichcasetheonly symptomofdistresswillbeapowerlossintheorderof2to3percent. Often,inmodifiedengines,youwillfindthatthestraightforwardincreaseinoverallpistonclearancebyslightlyenlargingthecylinderboreisnotacompleteanswer. Ifthemanufacturerhasdonehisworkproperly,hispistonswill,astheyexpandwithtemperature,assumearoundshapewhentheengineishot.Yourproblemwillbe thatwiththemodificationsyouhavemade,moreheatwillbeforcedintothepiston'scrown,raisingitstemperatureabovethelevelanticipatedbythemanufacturer,which resultsinacompletelydifferentsetoftemperaturegradientsdownthelengthofthepiston.Specifically,whilethewholepistonwillassumeadiameterslightlylargerthan thatplannedforbyitsmaker,theareaaroundthecrownwillgrowmorethantherest.Itwillthusbeimpossibletocorrectforthealteredconditionssimplybyhoningthe cylinderborelarger,forifyouenlargetheboreenoughtoproviderunningclearanceforthetopofthepiston,itsskirtwillbegiventoomuchclearance(leadingtorocking, andtroublewiththerings).Insuchcases,whicharenottheexception,buttherule,thesolutionistomachinewhatiscalledaclearancebandaroundthetopofthe piston.Usually,thisbandwillextenddownfromthecrowntoapointabout0.125inchbelowtheringgroove,orgrooves,andthepiston'sdiameterreducedbyperhaps 0.002inchovertheentireband'swidth.Althoughtheclearancebandisnotaparticularlycleansolutiontothepistonexpansionproblem,itisonethatcanbeappliedby anyonewithaccesstoalathe,andithasoneadvantageoverthegenerallymoredesirablepurecontouringofthepiston:ifapistonwithaclearancebandseizespartially, aluminumwillnotbesmearedaboveandbelowtheringgroove aneventwhichwilllocktheringinitsgrooveandupsetitsabilitytosealagainstgaspressure,In

practicalterms,thismeansthatthe clearancebandedpistonwillabsorba lotofpunishmentbeforeitisdamagedsufficientlytocauseretirementfromarace. Excessivedeepclearancebandsmustbeavoided,fortheyexposethesealingringtotoomuchheat,andheathasadevastatingeffectontheservicelifeofapiston ring.Butfortheseeffects,therewouldbeeveryreasontolocatetheringasclosetothepistoncrownasismechanicallypossible,becausewewouldthenobtainthe cleanestopeningandclosingoftheportswiththeringinitsusualposition,about0.200inchbelowthepistoncrown,thereisatendencyforgasestoleakdowntheside ofthepiston,andtheportopeningprocessthusbecomesmoregradualthanisdesirable.Theeffectisslight,butitisthere,andforthatreasonringlocationalwaysisa matterofjugglingtheconflictingrequirementsofkeepingtheringcool,andobtainingsharp,cleanportopeningcharacteristics.Andinmostinstances,thebalanceofthis compromisewillbeinfavoroftheformer,foranoverheatedringquicklyfails.Thecauseofthisfailureistwofold:first,excessivelyhightemperatureseffectivelyannealthe ring,anditlosesitsradialtensionsecond,anoverheatedringwarpslikeapotatochip,andnolongermaintainsclosecontactwiththebottomofitsgroove.Inbothof thesecases,thering'sabilitytosealisreduced,whichallowsfiretostartleakingdownpastthering,andthatfurtherraisesitstemperaturestartingacyclethatsoonresults inoutrightringfailure. ThesingleexceptiontotheunpleasantnessjustdescribedistheLshapedDykesring,whichalsoisexceptedfromtheimmediateeffectsofringflutter(described elsewhere).AnumberofengineshavebeenfittedverysuccessfullywithDykesringslocatedrightatthetopsoftheirpistons,andthedireeffectsofexcessiveheatingare avoidedbecausetheDykesring'sverticalleghasenoughareaincontactwiththecoolercylinderwalltodrawawayheatfasterthanitcanbeaddedbythering'scontact withhightemperaturegases.Atleast,that'sthewaythesituationcanbe,ifeverythingisright.Ontheotherhand,itisworthrememberingthatmanyusersofDykes patternringshavebeenforcedtofabricatethemfromstainlessalloystoovercometemperaturerelatedtroubles,andeventhenhaveexperiencedproblemswithoil carbonizingintheringgrooves.ProbablythebestthingtobesaidforDykespatternringsfromtheexperimenter'sviewpointisthattheycanbeusedtoovercomethe problemofusingstockpistonsatverymuchhigherthanstockcrankshaftspeeds.If,forexample,youwouldliketousethestockpiston,butcannotbecauseithasbeen groovedforrings2.0mmthickandyoumustuse1.5mmringstoavoidringflutter,youcansimplycutanewgrooveatthetopofthepistonforaDykesringandthe problemissolvedunlessyouencountersomeoftheotherdifficultiesjustdiscussed.

PISTONRINGS
Ofalltheproblemsthatcanbeexperiencedwithamodifiedengine,thoseconnectedwiththepistons'ringsarethemostinsidious.Borderlinesealingfailurescan sendfireshootingdownalongthepistons'sidestocauseseizuresand/orholingofthepistoncrownthatappeartobetheresultofleanmixture,excessiveignitionadvance ortoohighcompression,butarenot.Thesefailuresare,Isuspect,muchmorefrequentthaniscommonlysupposed,forthe2.0mmringsthathavebecomealmost

problemissolvedunlessyouencountersomeoftheotherdifficultiesjustdiscussed.

PISTONRINGS
Ofalltheproblemsthatcanbeexperiencedwithamodifiedengine,thoseconnectedwiththepistons'ringsarethemostinsidious.Borderlinesealingfailurescan sendfireshootingdownalongthepistons'sidestocauseseizuresand/orholingofthepistoncrownthatappeartobetheresultofleanmixture,excessiveignitionadvance ortoohighcompression,butarenot.Thesefailuresare,Isuspect,muchmorefrequentthaniscommonlysupposed,forthe2.0mmringsthathavebecomealmost standardwillbegintoflutterwhenpistonaccelerationrisesaboveabout60,000ft/sec 2anditisentirelytooeasytoexceedthatlimitwithamodifiedtouringengine. Therefore,Iwouldagainurgeyoutodoyourhomeworkbeforestartingadevelopmentprogramwithanyengine.Aformulaforpredictingtheonsetofringflutteris providedinthechapterheaded,Fundamentals,andyoumaysaveyourselfalotofgriefbydeterminingyourengine'sredlinewithpaperandpencilinsteadofthrough experimentation.Atthesametime,Imustcautionyouagainstsimplyassumingthatverynarrowringsareanadvantageinallengines.Infact,thereisnodetectablepower differencebetweenthestandard2.0mmringandtheracing1.0mmringbelow7000rpm,andthewiderringhastheadvantageofbetterdurabilityrightuptothepoint wherepistonaccelerationstartsitfluttering.Neitheristhereanyadvantage,below7000rpm,intheuseofsingleringpistons.Abovethatlevelthelowerfrictionofthe singleringpistonbeginstomakeadifference,butinthelowerspeedrangesyoumayaswellusetworingpistonsandtakeadvantageoftheirsecondlineofdefense capability. Selectionofringtypewillusuallyhavebeenmadeforyoubythepistonmanufacturer,andmyadviceisthatyoudonottrytoimproveuponhisjudgment,which willbealmostimpossibleinanycase.Youcannot,obviously,remachineapistonmadefor2.0mmringstotake1.0mmringsunlessyoucutanewringgrooveabovethe existinggrooves,andthatwouldpositionyourringperilouslyclosetothepistoncrownandalmostcertainlyleadtoimmediateringfailure.Theonlywayaroundthisistofit aDykespatternring,rightupatthepistoncrownaswasnotedpreviously.Suchmodificationscanbeverysuccessful,ifyouhavetherightringfortheapplicationand cutthegroovecorrectlyforthering,butIcannotrecommendtheproceduresimplybecausethereissomuchroomforerror.Ingeneral,Ithinkitisfarbettertoreplace thestockpistonwithonefittedwiththinnerringsevenifthereplacementpistoniscastofsomewhatinferiormaterial,asisoftenthecase.Afterall,thebestofpistonswill failifitsringsarenotsuitedtothejobitisbeingaskedtoperform.Ontheotherhand,ringsoflessthandesirablematerialwillperformverywellinracingapplicationsif replacedfrequently,andiftheyhavenotbeencrudelyfinished.Muchofthering'sabilitytofunctionisrelatedtothislatteraspect.Theordinarycastironringisfragile, andwillshatterveryquicklyifallowedtoflutter,butitwillperformentirelysatisfactorilyifitslowersurfaceissmoothandtrue,andsealsagainstthebottomofthering groove.Ringsmadeofnodularcastironhavethesamewearresistantproperties,andarevastlystronger,forwhichreasonsthismaterialisalmostuniversallyused. Surfacecoatings,rangingfromchromiumtoTeflon,areoftenappliedtothepiston'sring'sface,toimproveservicelifeand/orpreventscuffingduringbreakin. Ringstickingisaproblemtobefacedwithallhighoutputtwostrokeengines.Carburizedoilmaylocktheringinitsgrooveafteraremarkablyshortperiodof runningiftheringlackssufficientverticalclearance(usually,from0.0015to0.0040inch)oriftheringislocatedtoonearthepistoncrown.Morefrequently,theproblem stemsfromtheoilbeingusedforlubrication,anditismostunfortunatethattheveryoilsprovidingthebestlubricationaretheonesmostlikelytocauseringsticking. Castorbasedoils,particularly,willbuildupthicklayersofvarnishinsidetheringgroove,unlesstheoilcontainsaconsiderablepercentageofdetergentchemicals. ApartfromtheLsectionDykesring,mostpistonringshaveabasicallyrectangularcrosssection,butyouwillfindmanyminorvariationsonthisarrangement. Currentlyverypopularisthekeystonering,whichhasataperedsection,witheithertheupperorlowersurface,orboth,slopingawayfromthering'souterface.The reasonforthisprimarilyistokeeptheringanditsgroovescrubbedfreeofcarbonandvarnish.Infourstrokeenginestheringsarefreetorotate,anddo,andtheirrotation performsthisscrubbing.Twostrokeenginesnearlyalwayshavetheirringspinned,topreventthemfromrotatingandthering'sendsfromspringingoutandbecoming trappedinaport.Hence,theneedforsomeotherformofscrubbingaction.Seldomisthetaperinakeystonetyperingmorethan7degrees,anditisalltooeasyto attemptinstallingoneofthemupsidedown,soyoushouldgiveparticularattentiontothering'smarkings.Suchmarkingsvaryinkind,butwithoutexceptiontheywillbe onthering'suppersurface. Anotherpointoftroublecanbethering'slocatingpin,andifyouencounterdifficultieswithlocatingpinsworkingloose,thesourceofthetroublenearlyalwayswill beintheexhaustport.Theracingengine'sverywideexhaustport(widthrepresenting,inextremeinstances,upto70percentofcylinderborediameter)leavesalotofthe ring'sdiameterunsupportedwhenthepistonisdowninthelowerhalfofthecylinder,whichallowstheringtobulgeoutintotheport.Makingtheportopeningovaland chamferingitsedgeswillpreventtheringfromsnagging,asthesethingseasetheringbackintoitsgrooveasthepistonsweepsbackupward.However,whiletheringmay

notsnagonthe port,itdoesgetstuffedbackintoitsgroovefairlyrudely, andthatmayhaveaverybadeffectonthelocatingpin:Onmosttworingpistons,thelocatingpinsarepositionedadjacenttotheareasofblindcylinderwallbetweenthe intakeandtransferportsplacedabout90degreesaparttoprovidealongpathforgasleakage.Thus,whentheringbulgesoutintotheexhaustportandthenisstuffed back,theendoftheringispushedintohardcontactwiththepin,andafterasufficientnumberofhardblows(andtheseaccumulaterapidlyat,say,10,000rpm)thepin beginstoloosenanditwillgraduallyenlargetheholeinwhichitisinsertedenoughtoworkcompletelyloose.Thentheringisfreetorotate,anditquicklyworksitsway aroundtocatchtheendinaport.Atriskofseemingimmodest,IwilladmittohavingisolatedthisproblemforYamahaseveralyearsagoandtodaythatfirm'sracing engineshavepistonswithlocatingpinspositioned180degreesfromtheexhaustport.Touringengines,whichhavemuchnarrowerexhaustportwindowsandthustreat theirringsmoregently,usuallybenefitfromhavingtheirtworings'endgapsplacedmorenearlyonoppositesidesofthepiston,asdescribedbefore. Insomeracingapplications,thestandardringsareadequatetotheenginespeedsanticipated,butoverallperformancemaydictateamuchwiderthanstock exhaustport.Then,theoffsetringlocatingpinmayprovepronetopreciselythesortoflooseningandsubsequentfailuredescribedintheprecedingparagraph,which willleadyouintoapistonmodificationthatcanbeverytricky:installinganewlocatingpininthebackoftheringgroove.Thisgetstrickybecauseinmanycasesthepin willbehalfin,halfabove,theringgrooveanditisimpossibletodrilltheholeforanewlocationafterthegrooveismachined.Impossible,unlessyoucutasmallpieceof aluminumtoexactlyfittheringgroove,fillingitflush,inwhichcaseyoudrillyourholehalfinthepistonandhalfinthefillerpiece.Thenyouremovethefillerandyourhole isreadyforthepinwhichintroducesyetanotherproblem:whattouseforapin?Steelwireisagoodchoiceongroundsofstrength,butislikelytoworkloosesimply

theirringsmoregently,usuallybenefitfromhavingtheirtworings'endgapsplacedmorenearlyonoppositesidesofthepiston,asdescribedbefore. Insomeracingapplications,thestandardringsareadequatetotheenginespeedsanticipated,butoverallperformancemaydictateamuchwiderthanstock exhaustport.Then,theoffsetringlocatingpinmayprovepronetopreciselythesortoflooseningandsubsequentfailuredescribedintheprecedingparagraph,which willleadyouintoapistonmodificationthatcanbeverytricky:installinganewlocatingpininthebackoftheringgroove.Thisgetstrickybecauseinmanycasesthepin willbehalfin,halfabove,theringgrooveanditisimpossibletodrilltheholeforanewlocationafterthegrooveismachined.Impossible,unlessyoucutasmallpieceof aluminumtoexactlyfittheringgroove,fillingitflush,inwhichcaseyoudrillyourholehalfinthepistonandhalfinthefillerpiece.Thenyouremovethefillerandyourhole isreadyforthepinwhichintroducesyetanotherproblem:whattouseforapin?Steelwireisagoodchoiceongroundsofstrength,butislikelytoworkloosesimply becausethealuminuminwhichitispressedgrowsandcontractssomuchwithchangesintemperature.Asmalldiametersplitpin(whichislikearolledtube)isabetter choice,butifsuitablesizesarenotavailable,thennpinmadeofhardbrassisatleastasgood.

WRISTPIN/CRANKPINBEARINGS
Backinthedayswhenpistonswereuniformlypoorandtwostrokeengineswouldn'tberunveryfast,wristpinbearingswerealmostalwaysasimplebrassbushing.Such bushingsworkverywellinfourstrokeengines,butlubricationismuchlesslavishinthecrankcasescavengedtwostrokeandaddeddifficultiesarecreatedbythe essentiallyunidirectionalloadsplaceduponit,whichpreventthepistonpinfromliftingawayfromthelowerpartofthebearingandadmittingoiltotheloadcarrying surfaces.Forthosereasons,theplainbushinghasnowalmostuniversallybeenreplacedbyneedlerollerbearings,whicharemoreeasilypenetratedbysuchoilasis availableandinanycaseneedmuchlessoil.Thislastisofveryparticularimportanceinhighoutputengines,astheheatflowingdownfromthepistoniscertaintothinany oilpresenttoaviscosityapproachingthatofwater.Butallthesedifficultiesnotwithstanding,theneedlerollerbearingiswonderfullytroublefree,andifyouencounter problemsatthehingebetweentheconnectingrodandpistonpin,thoseproblemswillalmostinvariablybewithbreakageofthebearingcage.Giventheextremelylow rotationalspeedofthebearinginquestion,nocageisreallyneededexcepttomakeengineassemblyeasier:thecageholdsalltheneedlerollersinplacewhilethepistonis beingfittedtotheconnectingrod.Thearrangementcertainlymakesworkingontheenginelesscomplicated,butasithappens,thecagebecomesthebearing'sweakest link.Pistonaccelerationathighspeedsisalsoappliedtothebearingcage,anditmayshatterunderthestrainwhichsendsashowerofparticlesfromthebrokencage andlooseneedlesdownintothecrankcase.Thedebristhusliberatedinvariablygetspumpedupthroughthetransferports,intothecylinder,andmoreoftenthannota rollerwillgettrappedhanginghalfoutofaportbythepistonwithdireconsequencestoboth. Yamaha'sTD1wasparticularlypronetosmallendbearingcagefailures,andIlearnedthehardwaytoreplacethesebearingsifIsawover11,000rpmonthe tachometerevenforamoment,fortheircagesrequiredonlyamoment'sbatteringbeforecrackswouldstarttospreadandoutrightdisintegrationsoonfollowedevenifI indulgedinnomoreexcursionspasttheredline.Thisdifficultyhasbeenovercomewithcagesmadeoftoughermaterialitispossibletoaccomplishthesamethingby usingcrowdedneedlesandnocageatall,whichdoesrequirethatawasherbefittedoneachsideoftheconnectingrod,totakeupclearancesothattherollerscannot escape.Gettingthethingassembled(withtherollergluedinplacewithgrease)isenoughtomakestrongmenweepwithfrustration,butitabsolutelyinsuresreliabilityat thispointintheengineandisameasureworthrememberingifproblemswithbrokenwristpinbearingcagesdooccur. McCulloch,thechainsawpeople,haveusedanarrangementsimilartotheonejustdescribedforyears,buttheyhavereasonsotherthansimplyworkingaround bearingcagefailuresatthewristpinendoftherod.ItwasdiscoveredatMcCullochthatfailuresatthecrankpinbearingweretraceabletothethrustwashersmost manufacturersoftwostrokeenginesusetocentertherodonthecrankpin.Thesewashersusuallyaremadeofbrass,orsteelwithacoppercoating,andtheydonotfind highrubbingspeedsandscantylubricationatallagreeable.Atveryhighcrankshaftspeeds,theyregistertheirprotestbyoverheating,andthiscausesariseintemperature

allaroundtheconnectingrod'sbigend, whichthinstheoilpresentenough tocreateyetmorefriction,moreoverheating,untilatlastthethrustwashers,rollerbearingandcagearehotenoughtoflashtheoil.Atthatpoint,lubricationisniland frictionquicklymeltsthebearingcageandwearsflatsontherollers.McCulloch'sengineersreasonedthatthepointoffailurecouldbepushedupwardmateriallysimplyby removingthethrustwashers,whichiswhattheydid.Ofcourse,theconnectingrodstillhadtobecenteredoverthecrank,butthistaskwasgiventoapairofthrust washersupinsidethepiston.TheimprovementintermsofelevatingtheMcCullochkartengine'smaximumcrankspeedwasintheorderof1500rpm,anditisworth notingthatYamahaborrowedthisideaforuseinthe17,000rpmGPenginesthecompanyracedin1968.Itisinterestingthatinthoseengines,thepistonringswereonly 0.6mminthickness. Crankpinbearingfailuresalsostemfromtheuseofexcessivelyheavybearingcages.Sheerrotationalspeedisnotenoughtoburstacageofsuchsmalldiameter andmass,butthefactthatthecagemustaccelerateanddecelerate,relativetothecrankpinastheconnectingrodswings,willcausedifficultiesunlessthebearingcageis verylight.Ineffect,therollersmustpushthecageuptospeedandthenslowit,andifthecagehasenoughinertiaitwillresistthispushingandpullingenoughtoskidthe

0.6mminthickness. Crankpinbearingfailuresalsostemfromtheuseofexcessivelyheavybearingcages.Sheerrotationalspeedisnotenoughtoburstacageofsuchsmalldiameter andmass,butthefactthatthecagemustaccelerateanddecelerate,relativetothecrankpinastheconnectingrodswings,willcausedifficultiesunlessthebearingcageis verylight.Ineffect,therollersmustpushthecageuptospeedandthenslowit,andifthecagehasenoughinertiaitwillresistthispushingandpullingenoughtoskidthe

rollersatwhich pointtheymomentarilybecomeaplainbearingajobfor whichtheyarepoorlyconstituted.Theskiddingrollersgeneratealotofheat,throughfriction,andtheheatleadsthebearingintothesamedeterioratingcycletooutright failureaswasoutlinedforthethrustwashers.Mostmodernengineshavesteelcrankpinbearingcages,copperortinplatedtoprovidealowfrictionsurfacetobear againsttherollers,crankpinandconnectingrodeye.Thesereplacethephosphorbronzecagesoftherecentpastwhichreplacedtheinelegantaluminumandbrasscages ofayetearlierera.Butthebestcurrentbigendbearingcagesaremadeoftitaniumandsilverplated.Experimenterswithnearunlimitedfundsmayliketotrytitanium bearingcages,butwhenhavingthemmadetheyshouldknowthatthebearingretainingslotsmustbemachinedwithedgesparalleltowithin1/200witheachotherandwith thecrankpin(assumingaparallelconditionbetweencageandcrankpinaxis).Itisnotajobforsomeonewithabenchviseandafile.Ontheotherhand,ifemploying silverplatedtitaniumcagesandmovingthethrustwashersfromthecrankpintothepistonwillelevateyourengine'sredlineby2000rpm,thentheyclearlywillpay dividendsinhorsepowerifporttiming,etc.,isadjustedcorrespondingly. Connectingrodsshouldnotbelightened,orevenpolished,unlessyouintendgoingallthewayinthisdirectionandwillfinishthejobbyhavingthepartshot peened.Forgingsacquireatoughskinintheprocessofbeingpoundedintoshape,andIknowofinstanceswhereconnectingrodsthatwereentirelysatisfactoryin standardconditionpromptlybrokeafterhavingbeenpolished.Idothink,ontheotherhand,thatthereisamarginofsafetytobegainedbysmoothingofftheroughedges wheretheflashhasbeenshearedawayfromtheforgings.Notchesare,intheengineer'slanguage,stressraisersandyoucandotheconnectingrodnoharminremoving them.Lighteningtheconnectingrodis,however,apoorchoiceofwaystouseone'stime,becausearodintendedfortheloadsat,say,8000rpmisgoingtobe overstressedat10,000rpmandifanything,materialshouldbeaddedtotherod,notremoved.Ontheotherhand,onesometimescanimprovebearingreliabilityby openingslightlytheoilchannelsattheendsoftheconnectingrod.Idonotrecommendthatyouactuallycutintothebearingsurface,butoildeliverytothebearingwillbe improvedbytaperingtheentry.Donotextendthetaperallthewaytothebearingsurface,asthesharpedgesthusformedwillflakeawayastheenginerunsandcausea bearingfailure. Crankshaftmainbearingsseldomaretroublesome,exceptinenginesthathave

beeninstorageforalongtimeandhavehadcorrosionatworkinthese bearingsorunlessthebearingshavebeenmishandled.Bearingsteelsareverytough,butyoudefinitelycanpoundsmallpitsintheracesbyinjudicioususeofahammer, andpitsalsocanbeformedbyrusting.Bearingsdamagedineitherfashionshouldbereplaced,asthepitswillsoonspreadandbecomeminortrenches,asaresultofan activitycalledBrinelling,whichactuallyisaformofworkhardening.Thebearing'srollersandraceshavecasehardenedsurfaces,butthemetalunderthisthincaseis relativelysoft,anditiscompressedandreleased(atanygivenpoint)asthebearingturnsunderaload.Iftheloadishighenough,orthebearinginservicelongenough,the repeatedcompressionswillliterallyfatiguethemetal,andtinyparticlesofthesurfacewillstartflakingawaywhichbecomesvisibleasthetrackingseenintheracesofa wornoutbearing.Anybearingwillstartflakingatsomepointinitslifebearingswithracesdamagesbyrust,etc.willbeginsuchflakingalmostimmediately.Incidentally, inveryhighlyloadedbearingstheflakingmaybestartedbythesharpedgesaroundanyinterruptioninthebearing'ssurface,iftherollerspassoverthoseedges.Oiling slotsintherod'sbigendarepronetodevelopthiskindoffailure,andthesamesortofflakingissometimesobservedaroundtheoilfeedholesinthecrankpinsofengines

andpitsalsocanbeformedbyrusting.Bearingsdamagedineitherfashionshouldbereplaced,asthepitswillsoonspreadandbecomeminortrenches,asaresultofan activitycalledBrinelling,whichactuallyisaformofworkhardening.Thebearing'srollersandraceshavecasehardenedsurfaces,butthemetalunderthisthincaseis relativelysoft,anditiscompressedandreleased(atanygivenpoint)asthebearingturnsunderaload.Iftheloadishighenough,orthebearinginservicelongenough,the repeatedcompressionswillliterallyfatiguethemetal,andtinyparticlesofthesurfacewillstartflakingawaywhichbecomesvisibleasthetrackingseenintheracesofa wornoutbearing.Anybearingwillstartflakingatsomepointinitslifebearingswithracesdamagesbyrust,etc.willbeginsuchflakingalmostimmediately.Incidentally, inveryhighlyloadedbearingstheflakingmaybestartedbythesharpedgesaroundanyinterruptioninthebearing'ssurface,iftherollerspassoverthoseedges.Oiling slotsintherod'sbigendarepronetodevelopthiskindoffailure,andthesamesortofflakingissometimesobservedaroundtheoilfeedholesinthecrankpinsofengines equippedwithdirectinjection oilingsystems,liketheSuzukisandKawasakis.Removethesharpedges,andyouremovetheproblemifany.Thereissufficient marginofstrengthinstockproductionenginessothattheproblemdoesnotoccuryoumayfinditinthecourseofreachingforcrankspeedssubstantiallyabovethestock specification. Somebodyisalwaystellingmeabouthavinganenginebalanced,andIalwayssmilenastilywhentheengineinquestionhasfewerthanfourcylinders.Inpointof fact,thesinglecylindermotorcycleenginecannotbebroughtintodynamicbalance,forifyoucounterweightthecrankshafttocompensateforthefullweightofthepiston androd,youwillsimplyhavemovedtheshakingforcefrombeinginplanewiththecylinderaxis90degrees."Balancing"oneoftheseenginesconsistsoffindinga balancefactor,inpercentageofreciprocatingmass,whichiskindtotheengine'smainbearingsanddoesnotexciteresonanceinthemotorcycle'sframe.Inlinetwinand threecylinderenginesalwayshavearockingcouple.Byandlarge,thestockcrankshaftcounterweightingwillbecorrectformostapplications,andunlessyouwantto getintoareallylengthyexperimentalprogramthereisnothingtobegainedinmakingchanges.

CRANKASSEMBLY
Therearegainsinpowerandreliabilitytobehadfromcarefullyaligningyourcrankshaftandmainbearingbores,andingettingthecylinderaxisprecisely perpendiculartothecrankshaft.Asithappens,thereismorevariationinproductiontoleranceswhenthevariouspartsofacrankshaftaremadethancancomfortablybe toleratedinaracingengine.Crankpinholesinflywheelsarenotallpreciselythesamedistancefromthemainshaftaxisfactories"selectfit"theseparts,andyoucanbe fairlycertainthatanewcrankshaftistrue,butifyoumanagetoruinanyofitsflywheels,donotassumethatareplacementflywheel,selectedatrandomfromthenearest partsbin,willbeasatisfactoryreplacement.Crankpinholes,infacingflywheels,shouldbematchedtowithin0.0002inchwithregardtotheiroffsetfromthemainshaft. Ifyourlocalsourcecannotsupplyasinglereplacementwheelwithinthattolerancelimit,Istronglyurgethatyoupurchaseacomplete,newcrankshaftwithflywheels matchedatthefactory.Andwhenrebuildingacrankshaft,withnewcrankpinsandbearings,becertainthatitisalignedtoatleastthetolerancessuggestedbythe manufacturer'sworkshopmanual.Also,checkyourcrankcasesformainbearingborealignmentand,moreimportantyet,thatthecylinderisexactlyperpendicularwith thecrankaxis,foranytiltingwillbereflectedinaddedfrictioninthehearings(especiallyatthethrustwashers)andinthepistonitself. Donotattempttosecondguessthemanufacturerwithregardtocrankshaftandcrankpinbearingsunlessyouhaveveryspecializedknowledgeinthisfieldorcan obtaintheadviceofsomeonewhoisanexpert.Mainbearings,particularly,shouldnotbereplacedwithjustanythingthatwillfit,asnveryspecialkindofbearingis employedintheseapplications,withclearancestoaccommodatetheexpansionandcontractionofaluminumbearinghousings.Andthesamecautionarynotemustbe addedwithregardtocrankshaftseals,whichinthehighspeed,twostrokeenginemustsurviveextremesintemperaturesandrotationalspeedswithveryscanty lubrication.Notsoverylongago,sealfailureswerecommon,butnowthatmeanshavebeenfoundtoTefloncoatseal'slips,troubleisusuallyencounteredonlywhenthe sealshavebeendamagedinthecourseofinstallation.Sohandlethesealscarefully,andprecoatthemwithagoodhightemperaturegreasebeforeassemblingyour engine.Youcanalsoimprovetheirreliabilitysomewhatbypolishingtheareaonthemainshaftsagainstwhichtheyheartoaglassyfinish.Thesealsthemselveswillpolish theshafteventually,butatconsiderableexpensetotheirworkinglife. Byandlarge,problemswithpiston,connectingrodbearings,crankshaftandsealscanbeavoidedsimplybyfollowingtherecommendationsmadeinthe manufacturer'sshopmanual.Thesingleexceptiontothisisinthefitbetweenpistonandwristpin,fortheveryhightemperaturesinamodifiedenginetendtocausea breakdowninthelubricationbetweenpinandpiston.Troublecanbeavoidedintheracingengineifthewristpinisalight,slidingfitthroughthepistonitshouldslip throughofitsownweight,withoutforcing,forifitistightenoughsothatyouhavetotapitthroughwithamallet,youeventuallymayhavetoremoveitwithahydraulic press.Toolightfitsmaybecorrectedbyusinganoldwristpinasalap,andadashofsomefine,nonimbeddinglappingcompoundtopolishoutthepiston'spinboreto size.

FortheOttocycleengine,ofwhichthetwostrokeisanexample,thereisatheoreticallevelofefficiency,intermsofconvertingheatintowork,referredtoin basicengineeringtextsasairstandardefficiency.Inthis,itisassumedthatthecylinderisfilledonlywithdryair,andheatthenadded,whichignoresthefactthatin practicetheaircontainssomemoistureandaconsiderablepercentageofhydrocarbonfuel.Evenso,thistheoreticallevelofefficiency,calculatedagainstcompression ratio,providesausefulyardstickagainstwhichactualefficiencycanbemeasured andittellsusalotabouttheeffects,onpoweroutput,ofcompressionratio.For example,atacompressionratioof5:1,airstandardefficiencyis47.5percent,whileat10:1,itis60.2percent.Thatis,ofcourse,averygreatgain,andthe consequences measuredatanengine'soutputshaftarethereasonformanyexperimenters'fixationonraisingthecompression.Certainly,increasesincompression ratio,whichmaybeaccomplishedsimplybytrimmingafewthousandthsofaninchfromthecylinderhead'slowersurface,canworkminormiracleswithanengine's performance. Buthighercompressionratioscanalsobringaboutamechanicaldisaster:improvementsinpowergainedinthismannerarepurchasedatadisproportionatecostin peakcylinderpressure,leadingtoreducedbearinglifeandsometimestoanoutrightfailureofaconnectingrodorcrankpin.Moreover,becausethehigherpressuresare reflectedinaproportionatelygreatersidethrustatthepiston,frictionallossesaresuchthatnetpowergainsarealwayslessthantheimprovementonewouldexpectfrom thecalculatedairstandardefficiency.Finally,heatflowfromthecombustiongasesintothesurroundingvessel(pistoncrown,cylinderhead,andcylinderwalls)rises increasinglysharplywithcompressionratio,sothatanumberofthermalrelatedproblemsintrudeintothealreadycomplicatedrelationshipbetweencompressionratioand power. Theworstoftheseproblemsistheoverheatingofthepistoncrown.Atoohighcompressionratiowillraisepistoncrowntemperaturestothepointwhereheating ofthemixturebelowthepiston,inthecrankcase,reducestheweightofthechargeultimatelytrappedinthecylinderduringthecompressionstroketosuchextentthatnet powersuffersnomatterwhatMr.Otto'sairstandardefficiencyformulamaysay.Andifthecompressionratioishighenough,heatinputintothepistonmayraisethe crowntemperaturetothepointwheredetonationandthenpreignitionoccur.Thesephenomenawill,inturn,veryquicklyfurtherraisepistoncrowntemperaturetosuch extentthatthepistonmateriallosesenoughofitsstrengthtoyieldtothegaspressureabove thepistoncrownthenbecomingeitherconcave(whichdropsthe compressionratiotoatolerablelevel)ordevelopsalargehole(andthatreducesthecompressionratiotozero:zero). Manypeoplehaveencounteredthislasteffect,andthetuner'sonetimefavoriteployofmillingthehead hasfallenintodisrepute.Butitalsoispossibleto encountertroublewithoutrecognizingit:Thereisadelicatebalancebetweengainsfromincreasedcompressionratiosandlossesduetoincreasedtemperatureswhich appearnotonlyatthepiston'sinterior,butalsothroughoutthecrankcase,crankshaft,rodandalltherestoftheengine'sinteriorcontactedbytheair/fuelmixture.When thesepartsarehotter,themixture'stemperatureisalsoraised,alongwithitsfreevolume.Thus,themixture'stemperatureinducedeffortstoexpandinevitablyforcepart ofitouttheexhaustport,andaspowerisrelatedverycloselytotheweightofthechargecapturedinthecylinder,thisheatingshowsupasapowerloss.Thetrickisto balancecrankcaseheatingandcompressionratio.Thereisanoptimumcombinationforeverysetofconditions,butfindingthatoptimumwithoutheatsensingequipment andadynamometerisexceedinglydifficult. THECOMBUSTIONPROCESS Nottoosurprisingly,theequilibriumdescribedisinfluencedbycombustionchamberdesignasisthepointatwhichsmoothburninggiveswaytotheoutright

size.

FortheOttocycleengine,ofwhichthetwostrokeisanexample,thereisatheoreticallevelofefficiency,intermsofconvertingheatintowork,referredtoin basicengineeringtextsasairstandardefficiency.Inthis,itisassumedthatthecylinderisfilledonlywithdryair,andheatthenadded,whichignoresthefactthatin practicetheaircontainssomemoistureandaconsiderablepercentageofhydrocarbonfuel.Evenso,thistheoreticallevelofefficiency,calculatedagainstcompression ratio,providesausefulyardstickagainstwhichactualefficiencycanbemeasured andittellsusalotabouttheeffects,onpoweroutput,ofcompressionratio.For example,atacompressionratioof5:1,airstandardefficiencyis47.5percent,whileat10:1,itis60.2percent.Thatis,ofcourse,averygreatgain,andthe consequences measuredatanengine'soutputshaftarethereasonformanyexperimenters'fixationonraisingthecompression.Certainly,increasesincompression ratio,whichmaybeaccomplishedsimplybytrimmingafewthousandthsofaninchfromthecylinderhead'slowersurface,canworkminormiracleswithanengine's performance. Buthighercompressionratioscanalsobringaboutamechanicaldisaster:improvementsinpowergainedinthismannerarepurchasedatadisproportionatecostin peakcylinderpressure,leadingtoreducedbearinglifeandsometimestoanoutrightfailureofaconnectingrodorcrankpin.Moreover,becausethehigherpressuresare reflectedinaproportionatelygreatersidethrustatthepiston,frictionallossesaresuchthatnetpowergainsarealwayslessthantheimprovementonewouldexpectfrom thecalculatedairstandardefficiency.Finally,heatflowfromthecombustiongasesintothesurroundingvessel(pistoncrown,cylinderhead,andcylinderwalls)rises increasinglysharplywithcompressionratio,sothatanumberofthermalrelatedproblemsintrudeintothealreadycomplicatedrelationshipbetweencompressionratioand power. Theworstoftheseproblemsistheoverheatingofthepistoncrown.Atoohighcompressionratiowillraisepistoncrowntemperaturestothepointwhereheating ofthemixturebelowthepiston,inthecrankcase,reducestheweightofthechargeultimatelytrappedinthecylinderduringthecompressionstroketosuchextentthatnet powersuffersnomatterwhatMr.Otto'sairstandardefficiencyformulamaysay.Andifthecompressionratioishighenough,heatinputintothepistonmayraisethe crowntemperaturetothepointwheredetonationandthenpreignitionoccur.Thesephenomenawill,inturn,veryquicklyfurtherraisepistoncrowntemperaturetosuch extentthatthepistonmateriallosesenoughofitsstrengthtoyieldtothegaspressureabove thepistoncrownthenbecomingeitherconcave(whichdropsthe compressionratiotoatolerablelevel)ordevelopsalargehole(andthatreducesthecompressionratiotozero:zero). Manypeoplehaveencounteredthislasteffect,andthetuner'sonetimefavoriteployofmillingthehead hasfallenintodisrepute.Butitalsoispossibleto encountertroublewithoutrecognizingit:Thereisadelicatebalancebetweengainsfromincreasedcompressionratiosandlossesduetoincreasedtemperatureswhich appearnotonlyatthepiston'sinterior,butalsothroughoutthecrankcase,crankshaft,rodandalltherestoftheengine'sinteriorcontactedbytheair/fuelmixture.When thesepartsarehotter,themixture'stemperatureisalsoraised,alongwithitsfreevolume.Thus,themixture'stemperatureinducedeffortstoexpandinevitablyforcepart ofitouttheexhaustport,andaspowerisrelatedverycloselytotheweightofthechargecapturedinthecylinder,thisheatingshowsupasapowerloss.Thetrickisto balancecrankcaseheatingandcompressionratio.Thereisanoptimumcombinationforeverysetofconditions,butfindingthatoptimumwithoutheatsensingequipment andadynamometerisexceedinglydifficult. THECOMBUSTIONPROCESS Nottoosurprisingly,theequilibriumdescribedisinfluencedbycombustionchamberdesignasisthepointatwhichsmoothburninggiveswaytotheoutright explosionswecalldetonation.Thisaspect,too,iswidelyappreciated,butnotwidelyunderstood.Intruth,mostpeoplehaveverylittleunderstandingoftheeventsthat followignitioneventsthatarehighlycomplexifstudiedwithregardtotheirchemistrybutreallyquitestraightforwardtakeninlessnarrowterms.Muchofthe misunderstandingthatexistshasbeencreatedbythepopularpress,whichinsistsuponsayingthatapistonisdrivendownwardonitspowerstrokebyaburningmixture. Inreality,theburningoffuelinthecylinderissimplyameansofraisingthetemperatureoftheworkinggas(airactuallyamixtureofgases)andtherebyraisingits pressure.ThisrelationshipwasformulatedlongagobyBoyleas: Where,ofcourse,PispressureandTistemperature.Thewholebusinessgetscomplicatedintheinternalcombustionenginebythechangesinthecylinder'scontentsdue tothecombinationofelementsintheworkinggaswithfuel,butitstillbasicallyisacaseofraisingtheworkinggases'temperatureandthusraisingtheirpressure,anditis thatwhichpushesthepistondownandmakesthehorsepower.Infact,burningwillhavebeenallbutcompletedbythetimethepistonstartsdownwardonitspower stroke. Here,foranyonewhocares,iswhathappensfromthemomentofignition:Severalthousandthsofaninchoftravelbeforethepistonreachesthetopofits compressionstroke,representingsomewherebetween20 and45degreesofcrankrotation,thetrappedair/fuelchargeisignitedbythesparkplugandburning commences.Atfirst,theprocessproceedsquiteslowly(relativetosubsequentcrankrotationbeforeTDC).Asmallbubbleoffireexpandsgentlyawayfromthepointof ignitionbetweenthesparkplug'selectrodeandgroundwire,andifallcombustionweretocontinueatthispaceitwouldhardlybecompletedintimeforthefollowing compressionstroke.However,thissmallflamequicklyheatstheremainingmixtureenoughtoenormouslyincreasetherateatwhichburningoccurs,andaftertheinitial delay,theflamefrontacceleratesoutwardfromitspointoforiginwitheverincreasingrapiditysweepingthroughoutthecombustionchamber.Andiftheenginehasbeen giventheproperamountofsparkadvance,thepistonwillhavejustmoveduptothetopofitsstrokeastherapidphaseofcombustionbegins,sothatthebulkofthe burningisdonewhilethepistonisvirtuallystoppedandthemixturecompressedtominimumvolume.Bythetimethecrankshafthasrotatedafewmoredegrees,andthe pistonisonceagainmovingdownward,thecombustionprocesswillhavebeenalmostentirelycompleted. Theprecedingiswhathappensinthenormalcourseofeventscombustiondoesnotalwaysoccurthatneatly.Themostcommon,regrettablecombustion irregularityisdetonation,theharshknockingyouhearjustbeforeanengineseizes,ormeltsapiston andthenoiseyouwouldhear,whenrunninganengineona dynamometer,astheneedleonthescalebeginsanominousretreat.Unhappily,theveryprocessbywhichthemixtureinthecombustionchamberisreheatedbeforeits actualcontactwiththeflamefrontadvancingfromthesparkplug,andrapidcombustionthusmadepossible,istheprocessthatmayalsoleadtothesuddenexplosionof thecombustionchamber'scontentsthatwecalldetonation.Here'showithappens:Ithasalreadybeennotedthatastheflamefrontadvances,thecombustionchamber's remainingunburnedmixtureisheated,andthisheatingiscausednotonlybydirectcontactwiththeflame,butalsobyradiationandtheoverallpressurerisewithinthe chamber.Ifthetemperatureofthisremainingmixtureisraisedtoitsignitionpoint,allofitisconsumedatthesameinstantinasingleexplosion.Thisexplosioncreatesa shock,duetoafantasticallyrapidpressurerise,thatstrikesoutagainstallitssurroundingshardenoughtomakedetonationscharacteristicknockanditisashockwitha forceoftensufficienttobreakthesparkpluginsulator'stipanddamageboththepistonandbearings.Evenso,itsworsteffectistoforcealotofheatoutintothepiston, cylinderheadandthecylinderwalls.Thesearethusbroughttoabnormallyhightemperature,whichtendstooverheatthenextair/fuelchargeandmakeitdetonateeven morequicklyandseverely. Shouldthisdetonationcontinue,itwilloverheattheengine'supperendtothepointwhereignitionoccursbeforethereisaspark:compressionheatsthemixturein anycase,andwhenalotmoreheatisaddedfromthepistoncrown,etc.,themixturewillbe

cylinderheadandthecylinderwalls.Thesearethusbroughttoabnormallyhightemperature,whichtendstooverheatthenextair/fuelchargeandmakeitdetonateeven morequicklyandseverely. Shouldthisdetonationcontinue,itwilloverheattheengine'supperendtothepointwhereignitionoccursbeforethereisaspark:compressionheatsthemixturein anycase,andwhenalotmoreheatisaddedfromthepistoncrown,etc.,themixturewillbe

broughttopreignite.Detonationhasaverybadeffectonpoweroutput preignition(thoughtbysometobethesamephenomena)isevenworseinthatregard,butwillnotlongcontinueunnoticedasitwillveryrapidlyoverloadthepistonin boththethermalandmechanicalsense beyondthepointoffailure.Knowingthat,youwillappreciatethatdetonationistobeavoidedifatallpossible.Onewayto avoiddetonationwouldbetosimplyholdthecompressionratiotosomeverylownumber,astheywouldreducetheprecombustiontemperaturesandtherebymake detonationunlikelyifnotimpossible.Butthatmethodismostly(theexceptionIwilldealwithshortly)tooexpensiveintermsofpoweroutputefficiency.Abettermethod isoneemployedinmostenginestoday:useofa"squishtypecombustionchamber,inwhichthemixtureistrappedinasmallpocketunderthesparkplug,andtherestof thecylinderheadsurfaceovertheboreismadetofitcloselyagainstthepistoncrownwhenthepistonisattopcenter. WehaveEngland'sHarryRicardotothankforthistypecombustionchamber,whichhecreatedtocopewithconditionsthatceasedtoexistlongbeforemostofus wereborn.DuringtheconflictthatwrackedEuropejustaftertheturnofthiscentury,therewerenotonlyshortagesofinternalcombustionenginefuels,butthefuels a v a i l a b l e w e r e o f v e r y p o o r q u a l i t y a n d w o u l d d e t o n a t e s e v e r e l y i n t h e s i d e v a l v e e n g i n e s o f t h a t

periodunlesstheengineswereoperatedwithamuchretardedspark,or theircompressionratiosloweredtoabout4:1,orboth.Thesemeasureshadaterribleeffectonfueleconomy,naturally,andtheproblemledRicardotodoserious researchintothenatureofdetonation.Wenowknowthatthesidevalveengineisparticularlypronetodetonation,asitofnecessityhasaverylongcombustionchamber. Igniteafireatoneend,anditwillbealongwhilereachingthefarcornersofthechamber.Intheintervalbetweenignitionandthecompletionofburningthereisample opportunityfortheunburnedpartofthechargetooverheatandignite.

SQUISHBANDS
Ricardosolvedtheproblem,oncehehaddetermineditsnature,byloweringtheundersideofthecylinderheadinthatpartofthechamberoverthepiston.Thus, mostofthemixturewasconcentratedrightattheignitionsource,andwouldbemorelikelytoburnwithoutdetonating.Thesmallpartofthemixturecaughtbetweenthe cylinderhead'ssquishbandandthepistonwasstillsubjecttocompressionheating,butwasfairlyeffectivelyshieldedfromradiationandwas,moreover,spreadinsucha thinlayerthatitwouldresistignitionfromanycauseasitwouldloseheatintotherelativelycoolpistonandcylinderheadtoofasttoignite. Thatstillisthesecretofthesquishtypecylinderhead:Itconcentratesthemainchargeinatightpocketunderthesparkplug,andspreadsthemixtureatthe cylinderbore'sedgestoothinlytobeheatedtothepointofignition.Theseendgasesdonotburnwiththemaincharge,andareonlypartlyconsumedasthepiston movesawayfromtopcenterandreleasesthemfromtheircoolingcontactwiththesurroundingmetal.Andrightthereisthedisadvantagethatcomeswiththesquish band cylinderhead,formixturethatdoesnotburnismixturethatcontributesnothingtopoweroutput.Oflesserimportance,thoughonlyinthiscontext,isthatthoseendgases contributeheavilytothereleaseofunburnedhydrocarbonsouttheexhaustpipeandintotheatmosphere,andforthatreasonautomobilemanufacturersarenowrelying

cylinderhead'ssquishbandandthepistonwasstillsubjecttocompressionheating,butwasfairlyeffectivelyshieldedfromradiationandwas,moreover,spreadinsucha thinlayerthatitwouldresistignitionfromanycauseasitwouldloseheatintotherelativelycoolpistonandcylinderheadtoofasttoignite. Thatstillisthesecretofthesquishtypecylinderhead:Itconcentratesthemainchargeinatightpocketunderthesparkplug,andspreadsthemixtureatthe cylinderbore'sedgestoothinlytobeheatedtothepointofignition.Theseendgasesdonotburnwiththemaincharge,andareonlypartlyconsumedasthepiston movesawayfromtopcenterandreleasesthemfromtheircoolingcontactwiththesurroundingmetal.Andrightthereisthedisadvantagethatcomeswiththesquish band cylinderhead,formixturethatdoesnotburnismixturethatcontributesnothingtopoweroutput.Oflesserimportance,thoughonlyinthiscontext,isthatthoseendgases contributeheavilytothereleaseofunburnedhydrocarbonsouttheexhaustpipeandintotheatmosphere,andforthatreasonautomobilemanufacturersarenowrelying muchlessheavilyonsquishbandchambersforcombustioncontrol.Youmaybeinterestedtoknow,too,thatinmanycasesanonsquishcombustionchamber,withits completeutilizationofthemixturetooffsetthepowerlimitingeffectsofanecessarilylowercompressionratio,hasproventobebestinabsolutetermsofpowerand economy.McCulloch,forexample,makeengineswithbothsquishandnonsquishcylinderheadconfigurationshavingfoundthatbothhavetheirapplications. Ourapplicationhere,ofcourse,isstronglybiasedtowardmaximumhorsepower,andthatpointstowardasquishbandheadwhichiswhatyouwillhaveinmost motorcyclesinanycase.Iwillwarnyou,now,thatitmaybeunwisetofollowtheoldtimetuner'shabitofincreasinganengine'scompressionratioasanopeninggambit inthequestforbetterperformance.Indeed,beforeyourworkisdoneyoumayfinditnecessarytoreduceyourengine'scompressionratiobelowthestockspecification. Yousee,inthefinalanalysisitisnotsomuchcompressionratioascombustionchamberpressurethatdeterminesthelimitandthesearenotatallthesamethings.Your stockengine,withacarburetorsizeandportingchosentolenditasmoothidleandeasystarting,doesamuchlesseffectivejobofcylinderfillingthanwillbethecaseafter ithasbeenmodified.Moreimportant,itwillprobablyhaveanexhaustsystemthathasmoretorecommenditasasilencerthanasaboosterofhorsepower.These factors,incombination,makeaverygreatdifferencebetweenthecylinderpressuresatthetimeofignitioninthestockandmodifiedengine.Evengivenacertain willingnessonyourparttouseafairlycoldsparkplugchangingitfrequentlyandafurtherwillingnesstoreplacepistonsandbearingsmoreofteninpaymentforadded power,itmaystillbenecessarytostaywiththestockspecificationforcompressionratio.Or,asIhavesaid,tolowertheengine'scompressionratiofromthestock condition.Thislastwillbeparticularlytrueifyousucceedincreatingamuchbetterthanstockexhaustsystem. Byandlarge,youwouldbewelladvisedtoignorethewholebusinessofcompressionratiosinfavorofcrankingpressures.Thereis,afterall,abigdifference betweenthekindsofnumbersyougetbyperformingthetraditionalcalculationstofindcompressionratio,andwhatishappeningastheengineturns.Myexperiencehas beenthatyoucanusecrankingpressuresof120psiwithoutworryingmuchaboutoverheatinganything.Maximumpowerwillbeobtainedatcrankingpressures somewherebetween135and165psi.Goinghigherwithcompression,inaconventionalmotorcycleengine,cangiveaneatboostinlowspeedtorque,butthethermal repercussionsofhighercrankingpressureswillsurelylimitmaximumoutput.Ontheotherhand,fancooledkartenginesperformverywellatcrankingpressuresupat200 psi,andwatercooledenginesbehavemuchthesame. Oneofthemostundesirablesideeffectsthatcomeswithtoohighcompressionratiosisanenormousdifficultyingettinganengineto"carburet"cleanly.Whenthe compressionratioistoohigh,you'llfindthatanengine'smixturestrengthrequirementhasasharphumprightatitstorquepeakthatnomotorcyclecarburetorcan accommodate.You'llrealize,afterworkingwithhighoutputtwostrokeengines,thatallofthemaretosomedegreeliquidcooledandthatthecoolingliquidisgasoline.

Itis truethatanoverrichmixturetendstodampenthecombustion process,andreducepower,buthereagainwefindourselvesfacedwiththenecessityforfindingabalancebetweenevils:Wehaveoverheatingtorobpowerononeside, andwecancooltheenginewithgasoline,buttoomuchfuelalsorobspower.Thesolutionisabeggar'schoice,inwhichwetrytofindthecrossoverpointbetween overheatingandoverrichmixtures. Inanengineintendedpurelyforroadracing,withatorquepeakvirtuallycoincidentalwithitspowerpeakanddrivingthroughaverycloseratiotransmission (enablingtheridertoholdenginespeedwithinnarrowlimits),makingthisbeggar'schoiceisafairlystraightforwardproposition:youplaywithjettinguntilthemotorcycle runsfast.However,roadracingconditionsallowyoutostayrightonthemixturerequirementhumpyoudon'thavetoworryaboutwhathappenstwothousandrevs belowthepowerpeak,becausethat'sbelowwhatyou'lluseinarace.Motocrossracingisanothermatterentirely,andanenginewithamixturecurvehumpwilldriveyou absolutelymad.Jetamotocrossenginesothatitdoesn'tmeltapistoneverytimeitpullshardatitstorquepeak,and(ifitsmixturecurveishumped)itwillbehuffingsoot andlosingpoweraboveandbelowthatspeed. Theanswertothisproblemistoironoutthatmixturerequirementhump,becausenomatterhowmuchworkyoudowiththecarburetor,itneverwillbeableto copewiththeengine'sneeds.Allthecarburetorknows,really,ishowmuchairismovingthroughitsthroat,anditaddsfueltotheairinproportiontotherateofairflow don'texpectittoknowwhenthepistonisgettinghotandrespondbyheavinginsomemorefuel.Howdoyougetridofthehump?Youdoitmostlybysubstitutinga somewhatlesseffectiveexpansionchamber:onethatgivesmorenearlythesameboostallthewaythroughthespeedrangeyouareobligedtousebyracingconditions, withoutanybigsurges.Thatwillresultinadropinpeakpower,obviously,butyoucancompensateforittoaconsiderableextentwiththehighercompressionratioyou previouslywereforcedtoforegointheinterestofkeepingthepistoncrownintactwhentheexpansionchamberdiditsbigboostroutine.Again,itisallamatteroffinding thebalance. Nomatterwhatthecompressionratioyouultimatelyuse,itwillhavebeeninfluencedmuchmorethanyouprobablysuspectbythecombustionchamber configuration,andbycertaingrosscharacteristicsoftheheaditself.Overtheyears,Ihaveseenthefashionincombustionchamberformsswingbackandforth,hitherand yon,withfirsthatsectionchambersinfavorandthentrenchtypechambers,andtorustypechambersandsoonandsoforthadinfinitum.Iwasnot,andamnot, impressed.Combustionchamberformshouldbeestablishedwithaneyetowardonlyaveryfewspecialconsiderations,andthesecannotaccountforevenhalfthe chambershapesIhaveseen.Listed,thoughnotreallyinorderofimportance,theseare:surface/volumeratiosparkpluglocationthermalloadingsandcombustion control.Wewillconsidereachoftheseinturn. Surfacetovolumeratioisimportantbecauseeveninthepartofthecombustionchamberfullyexposedtotheadvancingflamefront,therewillbeamixturelayer adheringtothemetalsurfacesthatdoesnotburn.Theselayers,likethattrappedwithinthesquishband,arecooledbytheirproximitywiththecylinderhead,orpiston,

yon,withfirsthatsectionchambersinfavorandthentrenchtypechambers,andtorustypechambersandsoonandsoforthadinfinitum.Iwasnot,andamnot, impressed.Combustionchamberformshouldbeestablishedwithaneyetowardonlyaveryfewspecialconsiderations,andthesecannotaccountforevenhalfthe chambershapesIhaveseen.Listed,thoughnotreallyinorderofimportance,theseare:surface/volumeratiosparkpluglocationthermalloadingsandcombustion control.Wewillconsidereachoftheseinturn. Surfacetovolumeratioisimportantbecauseeveninthepartofthecombustionchamberfullyexposedtotheadvancingflamefront,therewillbeamixturelayer adheringtothemetalsurfacesthatdoesnotburn.Theselayers,likethattrappedwithinthesquishband,arecooledbytheirproximitywiththecylinderhead,orpiston, andsimplyneverwillreachignitiontemperature.And,liketheendgasesfromthesquishband,theyeventuallyfindtheirwayouttheexhaustport,havingtakennopartin theconversionoffuelandairintohorsepower.Thus,thebestcombustionchambershapetakenstrictlyfromthestandpointofsurface/volumeratiowouldbeasimple sphericalsegmentsweepinginacontinuousarcfromonesideofthecylinderboretotheoppositeside.Notrickychangesinsection,nosquishbands,nonothing.And thatis,inpointoffact,preciselytheshapeemployedinnearlyallnonsquishcylinderheads. Butifyouwanttouseatrue(measuredfromexhaustclosing)compressionratiomuchover6.5:1,onahighoutputengine,combustioncontrolbeyondthat a f f o r d e d b y a n o n s q u i s h c y l i n d e r h e a d w i l l b e n e c e s s a r y . C o n s i d e r a b l e v a r i a t i o n i s p o s s i b l e , b u t a

goodruletofollowistomakethecylinderhead'ssquishbandabout50 percentofthecylinderborearea.Forexample,ina3inchborewhichhasatotalareaof7.07inches2thesquishbandwouldbewideenoughtorepresentanareaof justabout3.5in 2.Assumingthatyouhavecenteredthecombustionchamberproperontheboreaxis,thenyoursquishbandwouldbearinghavingthesameouter diameterasthebore,andaninnerdiameterofjustover2inches.Thecombustionchamberitself,tomeetthepreviouslystatedminimumsurface/volumerequirement, wouldagainbeasphericalsegmentwitharadiusthatprovidesthetotalvolume,addedwiththatfromtheclearancespacebetweenpistonandsquishband,togivethe desiredcompressionratio. Theclearancespacebetweenpistonandcylinderheadmustbeenoughtoavoidcontactathighenginespeeds,yetcloseenoughtokeepthemixtureheldthere cooledduringthecombustionprocess.Thisverticalclearancebetweensquishbandandpistonshouldnotbegreaterthan0.060inch,anditismyopinionthatthe minimumshouldbeonlybarelyenoughtopreventcontactusuallyabout0.015inchinsmallengines(withtightbearingsandcylinder/rodcombinationsthatdonotgrow, withheat,disproportionately)anduptoabout0.045inchinbigengines. Somedisagreementexistsastothevalidityofclaimsthatthesquishbandaidscombustionbycausingturbulenceinthecombustionchamberasaresultofthe piston"squishing"partofthechargebetweenitselfandthehead.Idon'tknowaboutthat,butIdoknowthatholdingsquishbandclearancetoaminimummeansthatthere willbethesmallestvolumeofendgasesescapingthecombustionprocess,andthatcanbemoreimportantthanyoumightthink.Forexample,a250cccylinderwitha fullstrokecompressionratioof10:1willpackitsentireair/fuelchargeintoavolumeofonly28ccbythetimeitspistonreachestopcenter.Assumingthatithasa3inch bore,anda50percentsquishbandwithapiston/headclearanceof.045inch,thenthevolumeofthechargehidinginthesquishareawillbeintheorderof2.6cc,or almost10percentofthetotal.Thatcanbereducedto5percentmerelybyclosingthesquishband'sclearanceto0.020inchandyou'llneverfindaneasier5percent horsepowerdifference.True,thedifferencemeasuredatthecrankshaftmightprovetobemorelike21/2percent,buttheadditionofthosesmallpercentagescanmake averylargefinaldifference.

PLUGLOCATION
Testshaveshownthatthebestlocationforthesparkplugis,byandlarge,squarelyinthecenterofthecombustionchamber,andwithitsgapasclosetothecenterofthe volumeoftrappedmixtureaspossiblewhichislogical,asthatpositionprovidestheshortestflametravelinalldirections.However,anumberofotherconsiderationsdo intrude.First,thepluggapwillnecessarilybeattheperipheryofanypartlysphericalchamber,andnotatitscenter,andtryingtoformaknobinthechamberroofto movetheplugdeeperintothemixturevolumewillupsetthechamber'ssurface/volumeratio.Secondly,movingtheplugtooclosetothepistonseemstocausealocal overheatingofthepistoncrown,whichcanimposeanunnecessarilylowceilingoncompressionratio. Thislastconsiderationhas,inmanyinstances,leddevelopmentengineerstousecombustionchamberswithformsthatallowtheplugtobepositionedwellaway fromthepiston:modifiedspheroidsconicalsections,etc.Also,chamberswithhigherroofs(likethoseshapedascones)withtheirsparkplugsupatthetopandthe broaderbasedownatthepiston,provideaslightlyslowerpressureriseascombustionprogresses,andareinconsequenceabitmorekindtobearings.Otherswitchesin pluglocationmaybemadeintheinterestofeasingthejobofplugreplacement:itisdifficulttochangeaplugcenteredinthecylinderheadwhenthebottomofafueltank, orframetube,isdirectlyoverhead. Fortunately,mostenginesusuallyarerelativelyinsensitivetopluglocationaslongasthegapisn'tmovedtooclosetothepiston.Whichraisesaninterestingpoint: Thecommonpracticeofshavingmaterialfromthecylinderhead'slowersurfacenotonlyraisesthecompressionratio,andthusthethermalloadonthepiston,butitbrings t h e p l u g g a p c l o s e t o t h e p i s t o n c r o w n c o m p o u n d i n g t h e p r o b l e m . A b e t t e r a p p r o a c h t o o b t a i n i n g

orframetube,isdirectlyoverhead. Fortunately,mostenginesusuallyarerelativelyinsensitivetopluglocationaslongasthegapisn'tmovedtooclosetothepiston.Whichraisesaninterestingpoint: Thecommonpracticeofshavingmaterialfromthecylinderhead'slowersurfacenotonlyraisesthecompressionratio,andthusthethermalloadonthepiston,butitbrings t h e p l u g g a p c l o s e t o t h e p i s t o n c r o w n c o m p o u n d i n g t h e p r o b l e m . A b e t t e r a p p r o a c h t o o b t a i n i n g

increasesincompressionratioistopurchaseacylinderheaddevelopedto dothejobproperly.Yamaha'sGYTkitheads,forexample,providetherightcompressionboost,havetheirsparkplugsproperlylocated,etc.Othermadeforthejob cylinderheadsofferthesamefundamentaladvantage,whichisthatyougettobuyalotofotherpeople'sengineeringataverylowcost. Notallcylinderheadshavetheirsparkplugsandcombustionchamberpocketscenteredoverthecylinderbore,andtherearegoodreasonsformostofthe variationsinformoneseesintheproductsofthemajormanufacturers:Forinstance,pistoncrowntemperaturesseldomareeven,andwhiletheoveralltemperature distributionpatternisunderstandablyinclinedtowardmaximumsinthecenterofthecrown,circumstancecanalsolendabiastowardtheexhaustport.Thatbiascomes notfromanyheatinputpattern,butratherfromthemannerinwhichthepistoncrowniscooledbyheattransferenceintotheair/fuelmixturebelow,andintothepiston skirt,fromwhenceitistransferredoutintothecylinderwalls.Coolingprovidedbytheturbulentcrankcasechargeismoreorlesseventhesamecannotbesaidofheat lossesintothecylinder,forthetemperaturegradientsaroundthecylinder'swallsaremostuneven.Theareaaroundtheexhaustportishotterthanthatbackattheintake port,eventhoughtheexhaustsideofthecylinderisinmostinstancestherecipientofthedirectcoolingairblast.Moreover,theexhaustportsideofthepistonskirtis bathedinfireeverytimetheportopensattheendofapowerstroke.Theoverallresultistomovethemaximumtemperaturepointonthepistoncrowntowardthe exhaustport. Now,whenthatmaximumtemperaturebiasbeginstoseriouslyoverheatthesideofthepiston,youarelikelytoseesomeseverepistonringproblemsdevelop: Toohightemperatureswilleventuallybeadisasterfortheringitself,butmoreoftenitwillnothaveachancetoshowitsdispleasurebecauseanotherdisastroussituation willalreadyhavedeveloped,withthelubricatingoil.Sometimes,ifarelativelyhighashcontentorinadequatelydegummedoilisused,theringwillbegluedsolidlyinits groovebyvarnishandcarbonizedoils.Moreoften,thetemperaturesprevailinginthatsectionofthepistonskirtadjacenttotheexhaustportwillcauseabreakdownof theoilfilminthatareaandthepistonwillseize.Andthiscanhappeneventhoughagenerousmarginofsafetystillexistsallaroundtherestofthepistonskirt.Acommon, andhighlysensiblesolutiontothisproblemistomovethecombustionchamberpocketawayfromtheboreaxis,towardtheback(inlet)sideofthecylinder.Thismeasure shroudsmoreofthepistoncrown'sexhaustsideunderthesquishbandwhichbecomescrescentshaped,insteadofbeingasymmetricalringandreducesheatinput therefromcombustion(theskirtwillstillbegettingplentyofheatwhentheexhaustportopens)enoughtoprovideamoreevendistributionofheataroundthepistonskirt. Then,withpistonskirttemperaturesevenedout,aslightlyhighercompressionratiomaybeusedwithoutincurringseizure,orlocalizedoverheatingofthepistonring. Thereisanothersolutiontotheproblemthathasnothingwhatevertodowiththecylinderhead:yousimplyaddmetaltothepistoncrown,andthat,too,willtend toequalizeskirttemperaturesbutitalsomakesthepistonheavier.Evenso,itisasolutionmuchlovedbymanufacturers,asaddingthicknessinthepistoncostsvirtually nothing,whileanydeparturefromsymmetryincombustionchamberconfigurationentailsmultiplemachiningoperations(itbeingextremelydifficulttocast,withsufficient accuracy,thecombustionchamber'ssmallvolume)andmachiningtimeisexpensive. Theremaybeanotherreasonforemployinganasymmetricalcombustionchamber,and/orrelocatingthesparkplugfromitsnormalpositionovertheboreaxis.In loopscavengedtwostrokeengines,thefreshchargeisdirectedupward,andat,therearcylinderwall,asitemergesfromthetransferports.Ideally,themixturestreams convergeandsweepupandoveratthetopofthecylindertoclearawayexhaustproductsandpushthemouttheexhaustport,followingtherearcylinderwallupward,

Thereisanothersolutiontotheproblemthathasnothingwhatevertodowiththecylinderhead:yousimplyaddmetaltothepistoncrown,andthat,too,willtend toequalizeskirttemperaturesbutitalsomakesthepistonheavier.Evenso,itisasolutionmuchlovedbymanufacturers,asaddingthicknessinthepistoncostsvirtually nothing,whileanydeparturefromsymmetryincombustionchamberconfigurationentailsmultiplemachiningoperations(itbeingextremelydifficulttocast,withsufficient accuracy,thecombustionchamber'ssmallvolume)andmachiningtimeisexpensive. Theremaybeanotherreasonforemployinganasymmetricalcombustionchamber,and/orrelocatingthesparkplugfromitsnormalpositionovertheboreaxis.In loopscavengedtwostrokeengines,thefreshchargeisdirectedupward,andat,therearcylinderwall,asitemergesfromthetransferports.Ideally,themixturestreams convergeandsweepupandoveratthetopofthecylindertoclearawayexhaustproductsandpushthemouttheexhaustport,followingtherearcylinderwallupward, andthencurlingbacksmoothlyunderthecylinderhead.Inpractice,thescavengingstreamtendstobemuchlessorderedinitshabits,andthegeneralturbulencecan makeitleapanddodgeallovertheplace,impingingstronglyatonepointandonlyeddyingatothers.Thisleads,insomeengines,toareshapingandrepositioningofthe combustionpocketthepurposeofsuchchangesbeingtoaidscavengingbyusingthecombustionchamber'sformtogivethescavengingstreamdirection. Insuchcases,thesparkplugmayalsobemovedtoapositionwhereitwillbewashedbythemixturestream,whichtendstocooltheplugbetweenfirings,and thusmaketheenginesomewhatlesssensitivetoplugheatrange.Also,asnotedbefore,theplugmaybemovedawayfromthecombustionchambercentertocreatea slightlylongerpathforflametravel,whichlowerstherateatwhichpressureinthecylinderrisesduringthecombustionprocessand,insomeinstances,makesforsmoother running.Toalesserextent,thesametreatmentmaybeusedtocombatatendencytowarddetonation,asthelowerpressureriserategivesallthepocketsofendgases timetolosetheirheatintothesurroundingmetal.Thislasteffectis,ofcourse,betterobtainedwithaconicalcombustionchamber,ratherthanbyoffsettingtheplug. Incidentally,movingthesparkplugovertooclosetoanyedgeoftheboreisusuallypoorpractice:Attimes,particularlywhenstartingfromcold,thepistonringwill scrapeoiloffthecylinderwallsandpitchitupatthecylinderhead,andifyouplacethesparkpluginthelineoffire,itdefinitelywillshowaweaknessforoilfouling. HEAD/CYLINDERSEALING

Amajorproblemwithcylinderheadsonhighoutputenginesthatbeganlifeaslowoutputenginesispersistentleakagearoundthehead/cylinderjoint.Thecombined increasesintemperatureandpressureseemalwaystobetoomuchforthejoint,andyouwillfindevidenceoffirepuffingpastonthesurfacesafterdisassemblyeven thoughyoumaynothaveobservedanythingoutoftheordinarywhentheenginewasrunning.Thisleakingwilloccurevenifyouhaveretainedtheengine'sstock compressionratio,anditmaybecomeveryseriousiftheheadhasbeenthinnedtogetacompressionratioincrease.Manymanufacturers,perhapsmost,feelsomeawful compulsiontoskimponsectionthicknesseswhentheymakeacylinderhead,ahabitthatoftenstandsrevealedasaquestionableeconomywhenyoutesttheirhandiwork onadynamometer:first,thethinsectionsoftendonothavethecrosssectionalarearequiredtotransferheatawayfromthehead'slowersurfacequicklyenoughtokeep thesparkplugtemperaturesstabilizedsecond,mostofthesecylinderheadsaresecuredtotheircylindersbyonlyfourwidelyspacedbolts,whichpresumesheavilyon theirbeamstrengthtomaintainatightsealatthejoint. Thislastsituationbecomesespeciallymarginalwhenmetalhasbeenmachinedawaytoraisetheengine'scompressionratio,andthestockheadgasket(usuallycut fromlightgauge,softaluminum)willinmanycasesnotbestrongenoughtoholdeventhepressureincreasesinvolvedinasimpleswitchingofexhaustsystems.Shavethe head(whichbothweakensthehead'sbeamstrengthandincreasestheforcesactinguponit)andyou'llverylikelyfindthatitbecomesimpossibletoholdthehead/cylinder sealthegasketwillfailafteronlyminutesofrunning.Also,attemptingtousethestockcylinderhead,ineitherstandardormodifiedform,oftenwillincreasetheheatinput aroundthesparkplugtosuchextentthattheenginebecomesimpossiblyfussyaboutplugheatrange.Useaplugcoldenoughtoavoidtroubleatmaximumoutput,andit willfoulatanythinglessthanfullthrottleoperation.Thereisnothinglikemassesofmetaltoequalizethetemperaturegradientsthroughthecylinderhead,andsadtosay thosemassesarenotprovidedinmanystockcylinderheads. Cylinderheaddesignalsocanstronglyaffectoverallcylindercooling.Whenthecylinderhead'slowersurfaceiscoolerthanthecylinderitself,heatwillbedrawn awayfromthelatterconversely,acylinderheadcanalsoputheatintothecylinderifthesituationisreversed.Allthingsconsidered,theengine'sbestinterestsprobably areservedbyisolating,tosuchextentasispossible,thecylinderandheadwhichmeansrestrictingthecontactareaatthecylinder/headjointtoanarrowsealingband whichbulgestoencompasstheholddownbolts,orstuds.Inthatwayanycoolingproblemswillbeisolated,andcanbedealtwithseparately.That,ofcourse,assumes thatitwillbepossibletoimprovecylindercoolingshouldsuchimprovementbecomenecessary.Actually,makinganewcylinderheadisfairlyeasy(itcanbeeithercastor simplymachinedfromablockofaluminum)whilethecylinderitselfpresentsafarmoredifficultprobleminfabrication.Soyoumayverywellwanttouseanoversized, deeplyfinnedcylinderheadtohelpcoolaparticularengine'sstock,castironcylinder.Andifthatshouldbethecase,rememberthatyou'llneedamaximumcontactarea betweenheadandbarrel,andsurfacesthatwillsealwithoutanykindofgasket.Thereisaverysharptemperaturegradientacrossanyjoint,andevenasolidcopper gasketpresentsonemorepairofsurfacesacrosswhichheatmustflow. Youmayfindthatprovidingasealbetweentheheadandbarrelisoneofthemoredifficultfacetsoftheoveralljob.AsIhavesaid,stockaluminumgasketsare almostcertaintofail,beingabitweakatambienttemperaturesanywayandimpossiblyfrailatthetemperaturestowhichtheywillbesubjected.Copperisabetter material,forwhileitisnearlyassoftasaluminumatambient,itshotstrengthpropertiesarebetter.Copperissoftenoughtomakeagoodgasketintheannealedstate,but hardensinuse,andmustbereannealedfrequentlytokeepitsoftandthusretainitspropertiesasagasket.Brassshouldneverbeusedasagasketmaterial,butsteelmay b e u s e d i f i t i s v e r y t h i n a n d h a s o n e o r m o r e c o r r u g a t i o n s r o l l e d , i n r i n g s , a r o u n d t h e b o r e i n t h e m a n n e r o f t h e h e a d

gasketsusedinsomeautomobileengines.Youcanalsogetagoodseal bymachininganarrowgrooveinthecylinder'supperfaceandinsertinginitasoftcopperring(madefromwire)tobearagainstthehead'slowersurface.Other,even bettersealsmaybehadwithgasfilledmetalOrings,pistonrings(they'llworkhere,too)andoneofthebestsealingringsI'veseenhasaVshapedsection,laidonits side,withtheV'spointaimedawayfromthebore.GaspressuretriestoforcetheVopen,bringingonearmtosealdownagainstthecylinderwhiletheotherispressed againstthecylinderhead.AnothersealingringthatworksinroughlythesamefashionisahollowmetalOringwithventholesdrilledthroughfromitsinnerdiameterto

gasketsusedinsomeautomobileengines.Youcanalsogetagoodseal bymachininganarrowgrooveinthecylinder'supperfaceandinsertinginitasoftcopperring(madefromwire)tobearagainstthehead'slowersurface.Other,even bettersealsmaybehadwithgasfilledmetalOrings,pistonrings(they'llworkhere,too)andoneofthebestsealingringsI'veseenhasaVshapedsection,laidonits side,withtheV'spointaimedawayfromthebore.GaspressuretriestoforcetheVopen,bringingonearmtosealdownagainstthecylinderwhiletheotherispressed againstthecylinderhead.AnothersealingringthatworksinroughlythesamefashionisahollowmetalOringwithventholesdrilledthroughfromitsinnerdiameterto admitgaspressurefromthecylinderwhichexpandsitoutwardandthuscreatesasealevenbetweensomewhatunevensurfaces. Nominalcompressionratios,asIhavesaidbefore,havelittlemeaninginhighoutputtwostrokeengines.However,youcanworkwithtrappedcompression ratiosalmostaseffectivelyasbymeasuringcrankingpressures.Anengine'strappedcompressionratioistheratiobetweenthecylindervolumeatthemomentofthe exhaustport'sclosingandthevolumewiththepistonatthetopofitsstroke.Tofindthis,youmustfirstmeasurethecombustionchambervolume,withthepistonin positionattopcenter.Thejobcanbedonewiththeengineassembled,usingngraduatedcylinderandpouringinoiluntilthelevelcomesuptothesparkplughole.Or youcancalculatethevolume.Whenthecombustionchamberhasasimpleshape(partspherical,conicalorcylindrical)Iprefertodothejobbycalculation,butmore complexshapessendmescurryingforacanofoilandagraduatedcylinder.Infact,theprocessofactualmeasurementmayappealtoyouasaregularthing,becauseyou willneedagraduatedcylinderformorethanthissingletask,andasliderulemaynotbeapartofyourbasicequipment.Inanycase,rememberwhenfiguringthe compressionratio,thatitisnottheratiobetweenpistondisplacementandcombustionchambervolume,butbetweencylindervolumesfromthepointofexhaustport closingtotopcenter,asinthefollowingformula: WhereCRiscompressionratio V1iscylindervolumeatexhaustclosing V2iscombustionchambervolume

Traditionally,compressionratioshavebeenmeasured"fullstroke".Thatistosay,V,wouldrepresentthecombustionchambervolumepluspistondisplacement frombottomcentertotogcenter.Thus,acombustionchambervolumeof28ccandapistondisplacementof250cc,calculatedfullstroke,wouldbe CR=9.93:1

ButafarmorerealisticfigureisobtainedwhenV 1representsthecylindervolumeabovetheupperedgeoftheexhaustport,andifweassumethatourhypothetical enginehasanexhaustportheightequalto45percentofstroke,thenV 1becomes55percentofpistondisplacementplusV 2,andcalculationgoeslikethis: CR=5.91:1 Coincidentally,thatcompressionratio(5.91:l)isverynearlyallanonsquishcombustionchamberwillpermitinanotherwisefullydevelopedtwostrokeengine. Withsmallboreenginesyoumaypushthecompressionratiouptoperhaps6.5:lwithoutseriousconsequences,usinganonsquishcylinderhead,butthatisverynearthe limit.Goodsquishbandcylinderheads,ontheotherhand,permitcompressionratiosuptoasmuchns9.5:linmotocrossengineswithexhaustsystemsthatprovidea wideboostwithoutanysubstantialpeaks,butforroadracingenginesIcannotrecommendanythingabove8.5:levenwhenunitcylindersizeisonly125cc.Youwillfind thathighercompressionratiosthanthosesuggestedcanproducemarvelouslyimpressiveflashreadingsonadynamometerassoonastheenginehasachancetogetupto fulltemperature,theoutputwilldropwellbelowthatsustainedbyanotherwiseidenticalenginewithalowercompressionratio.Sustained,andnotflashhorsepower,is whatwinsraces. WalterKaadenwaschiefengineerofMZ'sracingdepartmentthroughthatfirm'sgloryyearsontheGrandPrixcircuit,andinthatcapacityKaadenadvancedthe stateoftheartwithregardtoexpansionchamberdesignveryconsiderably.Andonedaywhilediscussingthesubjectheremarked,onlyinjest,You'llknowwhenyou havethedesignright,becausethechamberwillthenbeimpossibletofitonthemotorcyclewithouthavingitdragtheground,burntherider'sleg,orforcetherelocationof oneormoremajorcomponents.Ofcourse,allpresenthadafinelaugh,butthejokecontainedalargeandbitterkerneloftruth.Inpointoffact,thatodd,bulkybitof exhaustplumbingwecallanexpansionchamber(apoortermforthedevice,butwidelyused)isexceedinglydifficulttoaccommodateneatlyonamotorcycle.Routed underneath,itisanacuteembarrassmentintermsofgroundclearanceevenonaroadracingmachineandfightsalosingbattlewithrocksonanoffroadbike.Curled backalongthemotorcycle'sside,itcanforcechangesinthepositionoffueltanksandframetubesandalwaysroaststherider'slegand/orforceshimtoridebow legged.Justasbad,itfiendishlyassaultstheearsofeveryoneforseveralhundredyardsineverydirection.andhasdonemoretomakethemotorcycleandtheman astrideoneunpopularthanalltheWildOnesmovies,andtabloidheadlinesofOnePercenter'smisdeeds,puttogether. Attendedasitisbythesemanifoldinconveniences,onealmost(butnotquite)wonderswhywebotherwiththeexpansionchamber.Unfortunately,damnable nuisancethatitunquestionablyis,thereisnothingelseintheengineer'sbagoftricksthatcomesanywhereclosetomatchingtheboostatwostrokeenginegetsfroma properlydesignedexpansionchamberexhaustsystem.Forthatreason,ithasbecometheubiquitoushelpmateofthehighoutputtwostrokeengine,andforthatreasonit willbewithusuntilweallchangeovertoelectricmotorsorgasturbines.Anduntilthattime,experimenterswillbetossingawaystockmufflersandtryingdifferent expansionchambersasamajorpartoftheirendlessquestforeverhigherperformance. Actualpercentageimprovementsbetweenenginesfittedwiththeirstandardmufflersandthesameengineswithexpansionchamberswillvarygreatly.Alot dependsonhowgood(orbad)theirmufflerhappenedtobe,andoncarburetorsize,porting,etc.anyofwhichcanimposelimitsthatcannotbetotallycompensatedby eventhebestofexpansionchambers.Inmostcases,however,theimprovementwillbeintheorderof10to25percent,andwhentheexpansionchamberisgivenabit ofhelpfromchangesintiming,andtherest,itthenbecomespossibletogetimprovementsrangingfrom50to(insomeinstances)over100percent.Thisdifferenceis widelyappreciated,evenbythosewhoknowabsolutelynothingabouttheexpansionchamberitselfandhavenodirectexperiencewiththedevice,andthataccountsfor thebrisksaleofaccessorychambersasreplacementsforstockexhaustsystems.Italsohasledmanyanenthusiasttoconstructanexpansionchamberofhisowndesign. Sadly,therealresultofmostpeople'sshadetreeexperimentalworkissimplytodiscoverthatitremainspossibletobringdownonone'sheadalloftheexpansion chamber'sconsiderabledisadvantageswithoutbeingcompensatedbyanincreaseinperformance.Or,asIheardoneexperimentercomment,lookingbemusedatthe chamberhehadcobbledtogetherforhismotorcycle,Itdoesn'tmakemuchpowerbutitsureisnoisy.Hewasbeingfunny,butIdidn'tlaugh,becausetheonlything thatdistinguishedhimfromhisfellowswasthathewashonestabouttheresultsmostoftheothersdonobetterbutaren'twillingtoadmitthattheyhavemadeabig mistake. Wheredoeseveryonegowrong?Usually,itistheresultofsimple,uncomplicatedignoranceregardingtheinnerworkingsoftheexpansionchamber,whichallthe folkloresurroundingthedevicenotwithstandingareabsurdlyuncomplicated.Usingamixtureofsonicwavebehaviorandcontrolledbackpressure,theexpansionchamber helpspullexhaustgasesoutofthecylinderduringtheinitialpartsoftheexhaust/transferprocessandhaulsthefreshchargeintoplaceandthenreversesitselftoprevent thechargefromescapingouttheexhaustport.Toillustratethepoint,let'swatch(inslowmotion)theactivitythroughasingleoperatingcycle,fromthetimetheexhaust portopensandthroughthetransferphaseuntiltheexhaustisonceagainclosed.Frombeginningtoend,theprocesstakesonlyabout3to4thousandthsofasecond.

fulltemperature,theoutputwilldropwellbelowthatsustainedbyanotherwiseidenticalenginewithalowercompressionratio.Sustained,andnotflashhorsepower,is whatwinsraces. WalterKaadenwaschiefengineerofMZ'sracingdepartmentthroughthatfirm'sgloryyearsontheGrandPrixcircuit,andinthatcapacityKaadenadvancedthe stateoftheartwithregardtoexpansionchamberdesignveryconsiderably.Andonedaywhilediscussingthesubjectheremarked,onlyinjest,You'llknowwhenyou havethedesignright,becausethechamberwillthenbeimpossibletofitonthemotorcyclewithouthavingitdragtheground,burntherider'sleg,orforcetherelocationof oneormoremajorcomponents.Ofcourse,allpresenthadafinelaugh,butthejokecontainedalargeandbitterkerneloftruth.Inpointoffact,thatodd,bulkybitof exhaustplumbingwecallanexpansionchamber(apoortermforthedevice,butwidelyused)isexceedinglydifficulttoaccommodateneatlyonamotorcycle.Routed underneath,itisanacuteembarrassmentintermsofgroundclearanceevenonaroadracingmachineandfightsalosingbattlewithrocksonanoffroadbike.Curled backalongthemotorcycle'sside,itcanforcechangesinthepositionoffueltanksandframetubesandalwaysroaststherider'slegand/orforceshimtoridebow legged.Justasbad,itfiendishlyassaultstheearsofeveryoneforseveralhundredyardsineverydirection.andhasdonemoretomakethemotorcycleandtheman astrideoneunpopularthanalltheWildOnesmovies,andtabloidheadlinesofOnePercenter'smisdeeds,puttogether. Attendedasitisbythesemanifoldinconveniences,onealmost(butnotquite)wonderswhywebotherwiththeexpansionchamber.Unfortunately,damnable nuisancethatitunquestionablyis,thereisnothingelseintheengineer'sbagoftricksthatcomesanywhereclosetomatchingtheboostatwostrokeenginegetsfroma properlydesignedexpansionchamberexhaustsystem.Forthatreason,ithasbecometheubiquitoushelpmateofthehighoutputtwostrokeengine,andforthatreasonit willbewithusuntilweallchangeovertoelectricmotorsorgasturbines.Anduntilthattime,experimenterswillbetossingawaystockmufflersandtryingdifferent expansionchambersasamajorpartoftheirendlessquestforeverhigherperformance. Actualpercentageimprovementsbetweenenginesfittedwiththeirstandardmufflersandthesameengineswithexpansionchamberswillvarygreatly.Alot dependsonhowgood(orbad)theirmufflerhappenedtobe,andoncarburetorsize,porting,etc.anyofwhichcanimposelimitsthatcannotbetotallycompensatedby eventhebestofexpansionchambers.Inmostcases,however,theimprovementwillbeintheorderof10to25percent,andwhentheexpansionchamberisgivenabit ofhelpfromchangesintiming,andtherest,itthenbecomespossibletogetimprovementsrangingfrom50to(insomeinstances)over100percent.Thisdifferenceis widelyappreciated,evenbythosewhoknowabsolutelynothingabouttheexpansionchamberitselfandhavenodirectexperiencewiththedevice,andthataccountsfor thebrisksaleofaccessorychambersasreplacementsforstockexhaustsystems.Italsohasledmanyanenthusiasttoconstructanexpansionchamberofhisowndesign. Sadly,therealresultofmostpeople'sshadetreeexperimentalworkissimplytodiscoverthatitremainspossibletobringdownonone'sheadalloftheexpansion chamber'sconsiderabledisadvantageswithoutbeingcompensatedbyanincreaseinperformance.Or,asIheardoneexperimentercomment,lookingbemusedatthe chamberhehadcobbledtogetherforhismotorcycle,Itdoesn'tmakemuchpowerbutitsureisnoisy.Hewasbeingfunny,butIdidn'tlaugh,becausetheonlything thatdistinguishedhimfromhisfellowswasthathewashonestabouttheresultsmostoftheothersdonobetterbutaren'twillingtoadmitthattheyhavemadeabig mistake. Wheredoeseveryonegowrong?Usually,itistheresultofsimple,uncomplicatedignoranceregardingtheinnerworkingsoftheexpansionchamber,whichallthe folkloresurroundingthedevicenotwithstandingareabsurdlyuncomplicated.Usingamixtureofsonicwavebehaviorandcontrolledbackpressure,theexpansionchamber helpspullexhaustgasesoutofthecylinderduringtheinitialpartsoftheexhaust/transferprocessandhaulsthefreshchargeintoplaceandthenreversesitselftoprevent thechargefromescapingouttheexhaustport.Toillustratethepoint,let'swatch(inslowmotion)theactivitythroughasingleoperatingcycle,fromthetimetheexhaust portopensandthroughthetransferphaseuntiltheexhaustisonceagainclosed.Frombeginningtoend,theprocesstakesonlyabout3to4thousandthsofasecond. THEBASICPROCESS Whentheexhaustportcracksopen,gasesstillunderaconsiderablepressureburstoutintotheexhausttract,formingawavefrontthatmovesawayathighspeed downtheportandheadedforlessconfinedquarters.Aftertravelingacomparativelyshortdistance,thiswavereachesthefirstpartoftheexpansionchamberproper whichisadiffuser(commonlycalledamegaphone).Thediffuser'swallsdivergeoutward,andthewavereactsalmostasthoughithadreachedtheendofthesystemand is,inthemannerofwavesexplainedinthefirstchapterofthistext,reflectedbackupthepipetowardthecylinderwithitssigninverted.Inotherwords,whathadbeena positivepressurewaveinverts,tobecomeanegativepressurewave.Thebigdifferencebetweentheactionofthediffuserandtheopenendofatubeisthattheformer returnsamuchstrongerandmoreprolongedwaveitisamuchmoreefficientconverter(orinverter)ofwaveenergy.

Astheinitialwavemoveddownthediffuser,theprocessofinversioncontinuesapace,andanegativepressurewaveofsubstantialamplitudeanddurationis returned.Also,overlaidonthisistheeffectofinertiaonthefastmovingexhaustgases,andthetotaleffectistocreateavacuumbackattheexhaustport.Thisvacuumis verymuchstrongerthanonemightsuppose,reachingavalueofsomethinglikeminus7psiatitspeak.Addthattotheplus7psi(approximately)pressureinthe crankcaseworkingtoforcethefreshchargeupthroughthetransferportsandyouwillbetterunderstandhowthetransferoperationisaccomplishedinsuchaveryshort time.Obviously,too,thiscombinedpressuredifferentialofalmostoneatmosphereisveryhelpfulinsweepingfromthecylindertheexhaustresiduefromtheprevious powerstroke.It'sallalotlikehavingasuperchargerboltedonoverattheengine'sintakesidebutwithoutthemechanicalcomplication. Yearsago,theexhaustsystemendedrightbehindthediffuser.ThatwasthearrangementontheoldsuperchargedDKWs,andwesawstubmegaphonesusedon theGreevesscramblersofthefairlyrecentpast.Thosedevicesdidajobinclearingexhaustgasesfromthecylinder,andhelpedthefreshchargeupfromtheircrankcase, buttheirvacuumingeffectwasverymuchamixedblessing:theirproblemwasthattheydidn'tknowwhentostopvacuuming,andwouldpullasizableportionofthefresh chargerightoutofthecylinder.Horsepowerbeingmoreorlessadirectfunctionoftheair/fuelmasstrappedinthecylinderattheonsetofthecompressionstroke,this aspectofthepuremegaphone'sbehaviorwashighlyundesirable,andthetwostrokeenginewasnottocomeintoitsowninracing(wherepowerisvitallyimportant)until

time.Obviously,too,thiscombinedpressuredifferentialofalmostoneatmosphereisveryhelpfulinsweepingfromthecylindertheexhaustresiduefromtheprevious powerstroke.It'sallalotlikehavingasuperchargerboltedonoverattheengine'sintakesidebutwithoutthemechanicalcomplication. Yearsago,theexhaustsystemendedrightbehindthediffuser.ThatwasthearrangementontheoldsuperchargedDKWs,andwesawstubmegaphonesusedon theGreevesscramblersofthefairlyrecentpast.Thosedevicesdidajobinclearingexhaustgasesfromthecylinder,andhelpedthefreshchargeupfromtheircrankcase, buttheirvacuumingeffectwasverymuchamixedblessing:theirproblemwasthattheydidn'tknowwhentostopvacuuming,andwouldpullasizableportionofthefresh chargerightoutofthecylinder.Horsepowerbeingmoreorlessadirectfunctionoftheair/fuelmasstrappedinthecylinderattheonsetofthecompressionstroke,this aspectofthepuremegaphone'sbehaviorwashighlyundesirable,andthetwostrokeenginewasnottocomeintoitsowninracing(wherepowerisvitallyimportant)until afteracurewasfoundfortheproblem. Here,ouroriginalwavereachesthatcure.Followingthediffuser,andafterperhapsacoupleofinchesofstraightwalledchamber,thewaveencountersa convergingconethateffectivelyconstitutesaclosedendtotheexpansionchamber.Apartofthewaveenergywillalreadyhavebeeninvertedbythediffuserandsent backtothecylinder,butthereisenoughofitsoriginalstrengthlefttoreboundquitestronglyfromthatclosedend,anditreflectswithitsoriginal,positive,sign.Indue courseoftime,thiswavearrivesbackattheexhaustportitself,stallingtheoutflowofthefreshcharge.Indeed,itwillmomentarilyreversetheflowthere,stuffingwhat mightotherwisehavebeenlostbackintothecylinder.Thenetresultofallthisactivityonthepartoftheexpansionchamberfirstpullingandthenpushingatthefresh chargetoholditinthecylinder i s a b i g b o o s t i n p o w e r . I n f a c t , i t i s t h e o n l y t h i n g y o u c a n d o t o a t w o strokeenginethatwill

clearlybefeltintheseatofyourpantsyoudon'tneedadynamometerto findthedifference. Aswasmentionedbefore,theexpansionchamberisnotpurelyasonicwavedevice:Backattheclosedendofthechamberthereisanoutletpipe,anditistoo smalltokeepthepressuresinsidethechamberequalizedwithatmosphericpressure.Consequently,thereisanabruptpressureriseinsidethechamber,towardtheendof itsoperatingcycle,whichisfeltattheengine'sexhaustportandplaysaverylargepartinpreventingchargeloss. Thisentireprocesscanworkwonderfullywellanditalsocanfailmiserablyifthevariouselementsoftheexpansionchamberarenotproperlydimensioned.All ofthevariouswavesandpressuresuckingandsurgingabouttheexhaustportmustoperateinagreementwiththeengine'srequirements.Whentheydisagree,theresultis worsethancanbeobtainedatamuchlowerpricepaidintimeandmoneywiththestockmuffler.Asithappens,themotionsofthosewavesarestubbornlytiedtoexhaust gastemperature,andsupremelyindifferenttowhattheenginewouldpreferintermsoftheirarrivals.Thetimeintervalsbetweentheinitialwavedeparture,andthereturn ofitsreflectedcomponentsisafunctionofwavespeed,andthesystem'slengths.Thus,aswavespeedissubjectonlytothelawsofphysicsandexistsassomethingone mustsimplyusewithoutaltering,thetaskofdesigninganexpansionchamberforsomeparticularapplicationistoestablishlengths,diametersandtapersthatwillusethe pulsationswithintheexhaustsystemtotheengine'sbenefit. TUNEDLENGTH Wemaystartbydeterminingtheproperlengththroughtheentiresystembacktotheexpansionchamber'sclosedend.Thattaskrequiresthatweknowthespeed atwhichsonicwavestravelwithinthechamber,andthereinliesagreatdifficulty.Asnotedpreviously,thesewaves'velocitiesaredeterminedlargelybythetemperatureof thegasesthroughwhichtheyarepropagatedandthatfactor,temperature,variouscontinuouslyinthecourseofasingleoperatingcycle.Exhaustgasesemergefromthe cylinderatabout1200Fandhaveverynearly(about800 o F)thesametemperaturebackintheoutletpipe.Butexpansionwithinthechamberitselfcoolsthem(priorto recompressionandreheatingbackinthebafflecone)toperhaps500F.,orless,inthemidsection,andawavedocsnotmoveasrapidlythroughthosecoolergases.Itis possibletocalculatefairlyexactlythetemperaturesatallpointsthroughoutthesystem,butthatisaverycomplexthermodynamicproblemandcertainlybeyondthe capabilitiesofthelayman.Indeed,honestycompelsmetoadmitthatitisnotaproblemIwouldliketofacewithoutacomputerandtheassistanceofsomeone experiencedinthatkindofwork. Happily,inthisinstanceitispossibletoarriveatasatisfactorysolutiontotheproblembydeterminingwavespeedbystartingwiththeanswerandworkingback. Inshort,youcanmeasurealotofexistingexpansionchambersknowntobeeffective,andbycomparingtheirlengths,exhaustporttimingsandthespeedsatwhichthe enginesdeveloptheirpower,eventuallycomeupwithafigureforwavespeedrepresentingaworkableaverageforawholerangeofhighoutputengines.Myown research,conductedalongthepragmaticlinesjustdescribed,wasbeguninabout1960andIarrivedataconclusionin1964thathasrequiredonlyslightmodificationover thesucceedingeightyears.Thatconclusionwas,andis,thatonemayuseawavespeedfigureof1700ft/secincombinationwiththeanticipatedenginespeedatmaximum powertoarriveatasystemlength(measuredbetweentheexhaustportwindowandthepointofmeanreflectionintheconethatconstitutestheclosedendofvirtuallyall expansionchambers).Thatfigureprovidesanexcellentstartingpointforthesystem,asitrepresentsahighaverageandanyerrorwillmerelyresultinalowerthan projectedpowerpeak.Actually,theadditionofmoreexamplestomychartsinrecentyearsmakemeinclinedtothinkthatsomethinglike1670ft,/secismoreaccurate, butIstillusethe1700ft/secfigureasastartingpoint,andsubsequentlyshortenthesystemslightly,perhapsaninch,iftestsindicatethatthepowerpeakobtainedwiththe chamberistoolow. Usingthathighaveragefigureforwavespeed(orindeedanyfigureyourfancydictates,ifyourfindingscontradictmyown)youmayestablishtheexhaustsystem's tunedlengthbymeansofthefollowingformula: WhereL tisthetunedlength,ininches

chamberistoolow. Usingthathighaveragefigureforwavespeed(orindeedanyfigureyourfancydictates,ifyourfindingscontradictmyown)youmayestablishtheexhaustsystem's tunedlengthbymeansofthefollowingformula: WhereL tisthetunedlength,ininches Eo istheexhaustopenperiod,indegrees Vs iswavespeed,infeetpersecond Niscrankshaftspeed,inrevolutionsperminute Forexample,inanenginewithanexhaustopenperiodof180degrees,andapowerpeakat7000rpm,andusingthe1700ft/secfigureforwavespeed,then, inches Thatlengthis,Imustagainstress,measuredfromtheexhaustportwindowbacktoapointslightlymorethanhalfwaydownthebaffleconeattheendofthe system.Theexactpoint,andhowtofindit,willbedealtwithshortly,alongwithanexplanationofwhyweuseaconetoclosethesysteminsteadofaflatplateandhow thetaperofthatconeinfluencesanengine'spowercurve.First,we'llconsiderthesizeandtaperofdiffusers. DIFFUSERPROPORTIONS Arrivingatthepropersizeforadiffuseris,thankstotheworkofseveralresearchers,almostentirelyacutanddriedproposition.Diffusersshouldhavean8 degreeincludedtaperformaximumenergyrecovery,andanoutletarea6.25timesthatoftheirinlet.Thus,adiffusertobeattachedtoaninlethavinga1.5inchdiameter shouldhaveanoutletof3.75inchdiameter.Outletdiametersfordiffusersofallinletdiametersmaybedeterminedinthefollowingmanner:

WhereD 2isthediffuseroutletdiameter Dlisthediffuserinletdiameter 6.25istheoutlet/inletratioconstant

Althoughthesediffuserdiameters,tiedtothe6.25constant,remainthesame,diffuserlengthmaybevaried,astherearereasonsforusingdiffusertapersotherthan 8degrees.Thattaperdoesthebestoveralljobofenergyrecovery,butitispossibletogetastrongerinvertedwavewithdiffusertapersgreaterthan&degrees,atthe expenseofwaveduration.Conversely,onealsomayextendthewavedurationbyacceptingsomediminishingofitsamplitudewithshallowertapers.Alongwave durationspreadsanengine'spowerbandashortdurationwavewithhighamplitudeisbestformaximumpoweratpeakrevs.Diffusershavingtapersofmorethan10 degreesreturnawaveofsuchbriefdurationastobealmostuselessevenforaroadracingenginecoupledtotherearwheelviaamultispeed,ultracloseratio transmission,andalsoareratherinefficientintermsofenergyrecovery.Forthatreason,Idonotrecommendthatyouuseadiffusertapergreaterthan9degreeseven whenplanninganexpansionchamberforaroadracingmachine,asyoumayotherwisefinditimpossibletokeeptheengineoperatingwithinitspowerband.Atthe oppositeextreme,donottrytouseanythingbelowa5degreetaperdiffuserinanexpansionchamberforanoffroadmotorcycle.Youwillfindthatevena5degree taperresultsinadiffuserthatisalmostimpossibletoaccommodatewithinthesystem'stunedlength,andthatitreturnsaninvertedwavetotheexhaustporttoofeeblein amplitudetobeveryeffectiveinscavengingatwostrokeengine.Asaruleofthumb,Iwouldsuggestemployingdiffusershavingtapersof8to9degreesforroadracing, smalldisplacementmotocrossbikes(upto250cc)shouldhavetapersof7to8degrees,andforbigmotocrossbikes,tapersof6to7degrees.Theselastusuallyhave morehorsepowerthantheycancomfortablyapplytothegroundanyway,andonedoesnotthereforegiveawayanythingbyextendingtheirpowerrangeattheexpenseof peakhorsepower. Allofthetapersgivenareforsingletaperdiffusers.Inthefutureweshallbeseeingmoreofmultitaperdiffusers,whichIfirstsawontheYamahaGPracing machinesandwhichhavesincebeguntoappearonmostofJapan'sexpansionchamberequippedproductionmotorcycles.Thesemultitaperdiffuserswouldseen)tobe intendedtoaccommodatethemselvestothefactthatawave,passingclownadiffuseroranycontainingvesselwithdivergingwallstendtoseparatefromthosewallsat thediffuserentryunlessthetaperisslight,butiswillingtoacceptsubstantiallygreateranglesofdivergenceonceithasmovedintothediffuserfarenoughtohaveexpanded andslowed.Fromthisbehavior,ifIunderstanditcorrectly,onemustconcludethatthebestdiffuserwouldbeonedivergingalongexponentiallineslikeatrumpet.Of course,suchashapewouldbeexcruciatinglydifficulttofabricate,andIthinkitisforthatreasonYamahaandSuzukichosetoemployinsteaddiffuserswithtwoorthree tapers.Myowninvestigationintothismatterisstillinitsveryearlystages,andIcannotofferanyopinion(muchless

Allofthetapersgivenareforsingletaperdiffusers.Inthefutureweshallbeseeingmoreofmultitaperdiffusers,whichIfirstsawontheYamahaGPracing machinesandwhichhavesincebeguntoappearonmostofJapan'sexpansionchamberequippedproductionmotorcycles.Thesemultitaperdiffuserswouldseen)tobe intendedtoaccommodatethemselvestothefactthatawave,passingclownadiffuseroranycontainingvesselwithdivergingwallstendtoseparatefromthosewallsat thediffuserentryunlessthetaperisslight,butiswillingtoacceptsubstantiallygreateranglesofdivergenceonceithasmovedintothediffuserfarenoughtohaveexpanded andslowed.Fromthisbehavior,ifIunderstanditcorrectly,onemustconcludethatthebestdiffuserwouldbeonedivergingalongexponentiallineslikeatrumpet.Of course,suchashapewouldbeexcruciatinglydifficulttofabricate,andIthinkitisforthatreasonYamahaandSuzukichosetoemployinsteaddiffuserswithtwoorthree tapers.Myowninvestigationintothismatterisstillinitsveryearlystages,andIcannotofferanyopinion(muchless

facts)excepttonotethatYamahahasbeenworkingwithdiffuserstapered 7degreesfor70percentoftheirlength,andtheremaining30percenttapered12degrees.BothYamahaandSuzukihaverecentlybeguntoconnectthesetwostage diffuserstotheexhaustportwithaleadinpipethatisalsotapered2 to3degrees.Thisslighttaperprobablyisprovidedtominimizeflowresistanceandthereby enhancetheleadinpipe'sabilitytoinertiaevacuatethecylinder,ratherthantofunctionaspartofthediffuser. Incidentally,theuseoftwostagediffusersalsofacilitatescopingwiththeoftenconflictingrequirementsofexpansionchambervolumeandleadinpipelength.A diffuserproportioned7 o /12 o ,70%/30%,issomewhatshorterthanadiffuserhavingastraight8degreetaperwhenbothareconstructedtothesame6.25:loutlet/inlet arearule.Thisreductionindiffuserlengthmaybeaddedattheleadinpipe,oratthechamber'sparallelwallsection,wherethediffuserandbaffleconearejoined.In eithercase,withalongerleadinpipeoraddedchambervolume,theoveralleffectistoincreasepoweroutputbelowthepowerpeakwithvolumehavingitsmost pronouncedeffecthighontheenginespeedscale,nearthepowerpeaklengthaddedtotheleadinpipebringsaboutasomewhatmoreseriousdropinmaximumpower, butalsolendsamarkedincreaseinlowspeedtorque.Theseeffects,itshouldbenoted,tendtomodifythechoiceofdiffusertapers,asashort,steeptaperdiffuser providesroomforalongerleadinpipeoraddedchambervolumebothofthesetendingtooffsetthepowerrangenarrowinginfluenceofsuchdiffusers. Somedeparturesfromthe6.25:larearatiorulemayalsobedesirableonoccasion:whenaleadinpipediameterisexaggerated,relativetoexhaustportareaand cylindersize,intheinterestoflowrangeperformance,youmayfindthatfollowingtheratioruleresultsinanexpansionchamberofsuchmonstrousdiameterthatitsimply cannotbefittedonthemotorcycleforwhichitisintended,andinthatcaseanappropriatedownwardadjustmentindiameterisindicated. Shouldthisconsiderationofinconvenientbulkleadyoutodepartfromthedimensionsrequiredbymyformulae,thinktwicebeforeyousuccumbtothetemptation toflattentheexpansionchamber.Inthefirstplace,you'llupsetalltheareaprogressionsthroughthediffuseran8degreediffuser,flattenedeversoslightly,isnolongeran 8degreediffuser.Moreover,evenifyoucalculatetheareassothatyouhavearoundedwedgewiththecorrectinlet/outletareaproportions,waveenergyrecoverywill stillsuffer.Thosewavessimplydonotlikebeingpuffedthroughanythingbutaconeevenlessdotheylikeaconethathasbeendentedornotchedtoclearaframetube ortoprovidegroundclearance.Theycan"feel"everychangeincrosssectioninthecontainingvessel.Theyare,however,willingtofolloweventhemostabruptjogin thesystem:youcanresectionthediffuserconeandkinkitallovertheplacetomakethe

expansionchamberfitthemotorcycle,andthewavewillneverknowthe difference.Sonicwavesmaybeabletofeeleventhemostminutechangesinsectiontheywillmakeanyturnyoucanbuildintothesystemwithoutslowingorlosinganyof theirenergy. Theonlypartofthesystemwhereyoumustbecarefultoprovidesmoothturnsisupattheleadinpipeandattheentrytothediffuser.Throughthatsection,gas velocityisveryhigh,andwhilethewavewon'tcareaboutsharpjogs,suchjogswillhaveabadeffectongasflowwhichisadifferentmatterentirely(gasflowinvolves themovementofmatterawaveisjustenergy,andbeingwithoutmass,isalsowithoutinertiaandthereforecaresnothingaboutsharpcorners.Atleast,thatisvery substantiallythecasehere,wherethevelocities,etc.involvedarelowenoughtofallwithinthescopeofNewtonianphysicsEinstein'sunifiedfieldworkishardlyapplicable atwavespeedsofsuchlimitedmagnitude). Ifefficiencyweretheonlyconsiderationincurrentexpansionchamberdesign,itwouldbepossibletodepartfairlysubstantiallyfromtheroundcrosssectionback

theirenergy. Theonlypartofthesystemwhereyoumustbecarefultoprovidesmoothturnsisupattheleadinpipeandattheentrytothediffuser.Throughthatsection,gas velocityisveryhigh,andwhilethewavewon'tcareaboutsharpjogs,suchjogswillhaveabadeffectongasflowwhichisadifferentmatterentirely(gasflowinvolves themovementofmatterawaveisjustenergy,andbeingwithoutmass,isalsowithoutinertiaandthereforecaresnothingaboutsharpcorners.Atleast,thatisvery substantiallythecasehere,wherethevelocities,etc.involvedarelowenoughtofallwithinthescopeofNewtonianphysicsEinstein'sunifiedfieldworkishardlyapplicable atwavespeedsofsuchlimitedmagnitude). Ifefficiencyweretheonlyconsiderationincurrentexpansionchamberdesign,itwouldbepossibletodepartfairlysubstantiallyfromtheroundcrosssectionback atthechamber'smiddleportion.Actuallyflatteningthechamberisstillnotagoodidea,butitcanbesquaredoffsomewhatwithoutgreatlycompromisingitsbasicpower enhancingproperties.Twoyearsago,Iwouldhaverecommendedpreciselythatsortofmodificationininstanceswhereinstallationwasaproblem.Now,withthedrive againstnoisewellandtrulyunderway,anydeparturefromroundistobeconsideredpoordesignpractice.Why?BecausethosewavesIhavebeentalkingaboutarevery strong,andwillmakeevenaroundsectionchamber'swallsringlikeabell(tobespecific,acowbell)justliketheengineisshootingmarblesoutitsexhaustport.These pulses,whicharestrongenoughtosetuparingingevenintherelativelystiffwallsofaroundsectionchamber,willmakeanyflatareasintheexpansionchamber'swalls pantinandoutlikeadrumhead.Thisvibrationisofcoursetransmittedintothesurroundingatmosphereasahellishlyloudnoise,andnomatterhoweffectiveamuffler youmayaddbackatthechamber'soutletpipe,themotorcycle'soverallnoiseoutputwillnonethelessbeveryhigh.Thenoisesourcejustdescribedcanonly.be minimizedbyeithermakingthechamberoutofveryheavysteel,orbygivingitashapethatresistspulsingtheroundsectionchambermayringsomewhat,butitcannot actuallypulseinandoutevenwhenmadeofverylightgaugematerial.Thispulsingofthechamber'swallshasanotherhighlyundesirablesideeffect:itmakesthe permanentattachmentofabracketorheatshieldverydifficult.Mostfastenerswillfairlyquicklyfracturefromtheseverevibration,leavingtheheatshieldtodropaway whichisbad,butnotasbadaswhenthesamevibrationfracturesamajormountingbracketandtheentireexpansionchamberconiesadrift.Forallthesereasons,the roundsectionexpansionchamber,althoughinconvenientlybulkyattimes,reallyseemstobethebestchoice. BAFFLECONES Wehavealreadynotedthatthebaffledendofmostexpansionchambersisconical.Thisconelendsthechamberrathermorepleasinglinesthanitwouldhavewith aflatend,butthatisnotitsreasonforbeing.Thereasonisthatifweendthechamberveryabruptly,withaflatplate,thewavereflectionsawayfromitwillalsobevery abrupt:strong,butofadurationtoobrieftoprovidethedesiredportpluggingeffectexceptwithinextremelynarrowlimitsinenginespeed.Aconicalbaffle,ontheother hand,extendsthewavereflectiontime(asreflectionoccursdownitsentirelength)and,becauseitseffectsarethusfeltoverawiderenginespeedrange,theengine'suseful powerbandisbroadened.Obviously,here,along,gentlytaperedbaffleconewillextendanengine'spowerrangemorethanashorter,moresharplytaperedconeinthe customarytradeoffbetweenrangeandpeakpower.Thesetapersshouldbe,inmostcases,twicethatofthediffuserusedintheexpansionchamber.Thus,inachamber havingan8degreediffuser,thebaffleconeshouldbetapered16degrees.Thatistheruleingeneral.However,widevariationsarepossibleandmaybeemployedto copewithaspecificsituation.Thelargesttaperangleyoushoulduseis20degreesthesmallest,14degrees.Andyoumay,toobtainaparticulareffect,mismatch

diffusersandbaffleconesinanycombination. Thethingtorememberis thatthereisapeculiarsidetothepowerrangebroadeningeffectofthebafflecone:mostofitisonthepartoftheengine'spowercurvepastthehorsepowerpeak. Thus,foranenginethathasproventoberatherfragilewhenpressedbeyonditsrpmredline,youmayterminatetheexpansionchamberwitha20degreebafflecone,and restassuredthatifthesystem'stunedlengthisestablishedtoplacethehorsepowerpeak,say,500rpmbelowthedangermark,theenginewillresistverystronglyany efforttogetitspinningfaster. Myexperienceindynamometertestingvariousexpansionchamberconfigurationshasshownthata20degreebaffleconegivesagood,strongpowerpeakand thensimplycutstheenginedead,intermsofoutput,ifyoutrytoforcetherevsanyhigher.A15degreebafflecone,incontrast,givesasomewhatlowermaximumoutput buthelpstheenginemaintainitsoutputwellafterthepeakingspeedhasbeenexceeded.Theimplicationsofthisinfluenceonanengine'spowercurveshouldbeobvious: motocrossbikescangaininengineflexibilityfromalong,gentlytaperedbafflecone,butifyouaretuningforsomeyoungmanwhooftenforgetstoprotecttheengine fromoverexertingitselfbychanginggears,thenyoucanusea20degreebafflecone.Itwillremindhimaboutthegearleverbychoppingthepowerdrasticallyeverytime hetriestousetoomanyrevs. Thereisanotherthingyouneedtoknowaboutthosebaffleconesbeforeyoucandesignyourownexpansionchambers:theydo,asstatedearlier,reflectover

motocrossbikescangaininengineflexibilityfromalong,gentlytaperedbafflecone,butifyouaretuningforsomeyoungmanwhooftenforgetstoprotecttheengine fromoverexertingitselfbychanginggears,thenyoucanusea20degreebafflecone.Itwillremindhimaboutthegearleverbychoppingthepowerdrasticallyeverytime hetriestousetoomanyrevs. Thereisanotherthingyouneedtoknowaboutthosebaffleconesbeforeyoucandesignyourownexpansionchambers:theydo,asstatedearlier,reflectover

theirfull lengthanywaveenteringthem,butthereisnotanevenreflection, forreasonstoonumerousandinvolvedforseriousdiscussionhere.Whatyoudofind,however,isthatthereisameanpointofreflectionwhichis,asIsaidearlier, slightlymorethanhalfwaydownthebafflecone'slength.Thetunedlengthwediscussedbefore,thelengthsocriticaltotheexpansionchamber'sperformance,is actuallythedistancefromtheexhaustportwindow,atthepistonface,measuredalongtheexhaustsystem'scenterlineouttothispointofmeanreflection.Thismidway pointseemstobeatthehalfwaypointofthecompletecone,whichwouldbehalfthelengthoftheconeiftheconewerecomplete,rightouttoasharptip,insteadofbeing truncatedatitssmallendtomakeroomfortheoutletpipe.Youcanfindthispointofmeanreflectioneitherbydrawingthecompletecone,measuring,anddividingits lengthinhalf,orbyusingthissimpleformula:

WhereL risthedistancefromthebafflecone'sinlettothemeanpointofreflection D,isthebafflecone'sinletdiameter A,ishalfthebafflecone'sangleofconvergence(i.e.fora16degreeconeuse8degrees,etc.) Whileweareonthisparticularsubject,Iwillalsogiveyoutheformulaforfindingthelengthofacone,givenitstaper,majorandminordiameters.Theformulaisas follows: WhereLislength D2isthecone'smajordiameter D1,isthecone'sminordiameter Aishalftheangleofdivergence,orconvergence. OUTLETPIPES Havinggottenpastthebusinessofdiffusersandbafflecones,wecanproceedonwardtothelengthsanddiametersofleadinandoutletpipes.Thelatterwill,if therestoftheexpansionchamberisproportionedfairlycloselyinaccordwiththeadviceIhavegiven,haveadiameterbetween.58and.62timesthatoftheleadin pipe,andalengthequalto12ofitsowndiameters.Itissimplyapressurebleedresister,whichpreventsthefreeescapeofexhaustgasesfrominsidetheexpansion chamberandtherebycreatesabackpressuretoenhancetheportpluggingeffortsofthewavereflectedbythebafflecone. Thisoutletpipeismuchmoresensitivetodiameterthanlength.TheruleIhaveprovidedfordiameterwillgetyouveryclosetotheoptimum,iftherestofthe chamberisproportionedcorrectly(oratleastasIhaveindicated,whichIhopewillamounttothesamething)andthelength,of12diameters,willalsobeveryclose. Evenso,theywillonlybecloseandIhavefoundfewinstances,inmyownexperimenting,whenasessionofcutandtryonthedynamometerdidnotleadmetoa slightlydifferentdiameterandlengthforbestresults.Youprobablydonothaveadynamometer,andthatbitofnewsisinalllikelihoodnotsomethingyouwantedtohear. Nonetheless,itistrue.YouwillnotgetoptimumresultsheresimplybyfollowingtheinstructionsIprovide.Infact,atadofadjustingisrequired,tomeetaparticularset ofrequirements,tomakeanystrictlyformulaexpansionchambergivemaximumperformance.Butthereisaveryparticularproblemassociatedwiththeoutletpipethat youshouldknowabout:Simplystated,theproblemisthatifyoumaketheoutletpipetoosmallindiameter,orgiveittoomuchlength,(bothtendingtooverrestrictthe chamber'soutletpassageintermsofflowcapacity)thentherewillbeapricetobepaidinoverheating.IfranklydonotknowthemechanisminvolvedinthisIonlyknow thatiftheoutletistoorestricted,enginetemperatureisverygreatlyelevated andYamaha'sNaitohassaidthatthesameunfortunateresultwillbeobservedifthe expansionchamber'sbaffleconeistaperedtooabruptly.FromwhatIhaveseen,thisincreaseintemperatureisespeciallysharpatthepistoncrown,whichtendstobe thepowerlimitingpartinatwostrokeengineinanycase.So,youshouldkeepasharpeyeonyourengine'spiston.Therewillbetemperatureshighenoughtodarken theundersideofthepistoncrown,duetobakedonoilformingthere,whentheengineishealthyjustdon'tignorethewarningsignyouseewhenthatoilbeginstochar. Thenextthingyousee,afteryouseeash,willbeahole. Actually,theworstsideofthiswholematterofselectingtherightoutletpipeisthatwhilepowerrisesquitesharplyasyouworkyourwaydownfromatoolarge outletpipediameter,therewillbeonlyaslightdecreaseinpoweroutputwhenyouhavegonetoofarinrestrictingtheoutlet.Therefore,unlessyouhaveadynamometer andinstrumentationformeasuringenginetemperatures,makeallerrorsonthelargesideofthediameteryougetbymultiplyingleadinpipediameterbythe0.580.62 factorIhaveprovided.Inotherwords,ifthefactorderiveddiameter(takenfromaleadinpipediameterofsay,1.75inches)isfrom1.015to1.085inches,thenyou shouldnottrytouseapieceofoneinchtubingforthatoutletpipe,asitsnominaldiameterisanoutsidemeasurement.Subtractforawallthicknessof,say,0.049inch, andyou'llhaveapassageonly0.902inchindiameter.Thatisn'tlargeenoughtobesafenorwouldatubehavinganominaldiameterofl 1/8inches,withaninside diameterof1.027inchesbeentirelysafe.Whenskirtingthatclose,gouptothenextlargestavailablediameterinthiscaseitwouldbel 1/4inchesandruntheengine longenoughandhardenoughtopermitavalidreadingofthepistoncrown'sunderside.Ifitshowsnosignofexcessiveheat,youcantryaslightlysmalleroutletpipe,

outletpipediameter,therewillbeonlyaslightdecreaseinpoweroutputwhenyouhavegonetoofarinrestrictingtheoutlet.Therefore,unlessyouhaveadynamometer andinstrumentationformeasuringenginetemperatures,makeallerrorsonthelargesideofthediameteryougetbymultiplyingleadinpipediameterbythe0.580.62 factorIhaveprovided.Inotherwords,ifthefactorderiveddiameter(takenfromaleadinpipediameterofsay,1.75inches)isfrom1.015to1.085inches,thenyou shouldnottrytouseapieceofoneinchtubingforthatoutletpipe,asitsnominaldiameterisanoutsidemeasurement.Subtractforawallthicknessof,say,0.049inch, andyou'llhaveapassageonly0.902inchindiameter.Thatisn'tlargeenoughtobesafenorwouldatubehavinganominaldiameterofl 1/8inches,withaninside diameterof1.027inchesbeentirelysafe.Whenskirtingthatclose,gouptothenextlargestavailablediameterinthiscaseitwouldbel 1/4inchesandruntheengine longenoughandhardenoughtopermitavalidreadingofthepistoncrown'sunderside.Ifitshowsnosignofexcessiveheat,youcantryaslightlysmalleroutletpipe, andthencheckthepistonagain. LEADINPIPES Thereprobablyisabetter,closer,rulefordeterminingoutletpipediameterifonlybecausetheruleIhaveofferedistiedtothediameteroftheexpansion chamber'sleadinpipe,whichisvery,verydifficulttodetermineonastrictlytheoreticalbasis.Forengineshavingexhaustporttimingsandportwidthstypicalinracing terms,theexpansionchamber'sleadinpipeshouldhaveadiameterprovidinganarea10to15percentgreaterthanthatoftheportwindow.Butthatonlyapplies,I havefound,whentheparametersareasstated,andonlythenwhenapowercurveverydistinctlybiasedtowardmaximumoutputatmaximumrevsisdesired.For motocrossapplications,thediameterchosenwillbeonethatcanbeusedincombinationwithaconsiderablepipelengthtobroadenthepowerrange.Insomeinstances, leadinpipesareusedwithcrosssectionalareasrepresenting150percentoftheexhaustwindowarea.Asageneralrule,youmayassumethatthemanufacturerofyour particularengineknowsmoreaboutleadinpipeareasthanyou,orI,andyoucan'tgetintotroublefollowinghislead.Youmay,byignoringthepossibilitiesinother directions,missanoptimumbysomesmallishpercentagebutyouwon'tgetintotrouble. Thethoughtmayhaveoccurredthatmyinstructionsare,inthisregard,somethingfarshortofprecise.Andsotheyare,forgoodreason:thechoiceofleadinpipe diametermustbeshapednotonlybyunitcylinderdisplacement,porttiming/area,andaccordingtotheapplicationyouhaveinmindfortheenginebutalsowithaneye towardtheleadinlength,andtheconfigurationofthediffusertowhichthepipeattaches.Allofthesethingshavetheireffect,butIhavenotasyetbeenabletoisolate eachitemwellenoughtoarriveataquantitativepattern.So,forthemoment,Imakedowithaqualitativeunderstandingandacoupleofhandyrulesofthumb:thefirst youalreadyknow(regardingtheroughrelationshipbetweentheportandpipecrosssectionarea)thesecondrule(moreasuggestion,really)isthatformaximum horsepoweronly,youshouldgivetheleadinpipealengthequaltofrom6to8ofitsdiameters,whileforabroaderpowercurve(andatsomeexpensetothemaximum) youmaywanttouseupto11timespipediameter.Alltheselengthyare,ofcourse,notjustthatofthepipesthemselves,butalsoincludethedistancefromthepipe mountingflangethroughtheporttothepistonface.Afinalnoteonleadinpipelength:if,forreasonsofeasierinstallation,orabroadeningofthepowerrange,orboth, youdecideuponaleadinpipelengthgreaterthan8timesitsdiameter,plantoincreaseitsdiameterslightly.Resistancetoflowisincreasedwithlength,andthiscanand shouldbeoffsetbygivingthegasesalargerpassage.Ideally,thisreductioninresistancewouldbeaccomplishedbyusingaleadinpipehavingaslightlydivergingtaper(2 or3degrees)butthatmayrepresentamoredifficultfabricationjobthanyoucaretoundertake. Afteryouhavefinishedthejobofdesigninganexpansionchamberexhaustsystem,thethingwillhavetobetranslatedfromasetofdimensionsintometal,oryou'll neverknowwhetherornotthedesignisanygood.Youcan,ofcourse,buildthewholethingyourself,butIwouldnotrecommendthatapproach.Especially,Iwould cautionagainstattemptingtorollyourowncones,asthisisthekindofjobthattakesendlesshourswhenyoudon'thavethepropertoolsandislikelytoturnoutbadly anyway.Gotoasheetmetalshopandpaythemtomakeyourconesthey'llgetitdoneinmuchlesstimeandifthey'resharpyou'llgetconeshavingtherightlengths, tapersanddiametersforverylittlemoremoneythanyouwouldspendjusttobuythematerials.Benttubing,tobeusedinmakingtheleadinpipe,canbeobtainedina varietyofdiametersandbendradiiatmostmufflershops.Welding?Unlessyouhavetheequipmentandarecompetentatweldinglightgaugemetal,thattooshouldbe hiredfromaprofessional,forthefinisheditemcangettolookveryscabbygiventhebestofworkmanship.Asfortolerances,theonlythingyoureallymustwatchisthat thefractionsofaninchdonotadduptoupsetthedistancefromtheexhaustportwindowbacktothebafflecone.Aninchthere,willmovethepowerpeakabout500 rpm.Thesameincrementinlength,upattheleadinpipe,producesaslightreshapingoftheentirepowercurve.Throughmostofthesystem,trytoworktowithin 1/16 inchtolerances. Whentestingthefinisheditem(torecapsomethingsalreadynoted),rememberthatchangingthetaperofthebaffleconechangesthatDartofthepowercurve pastthepowerpeak,whilechangesintheleadinpipelengthmostlyinfluencethelowrevspartofthecurve.Increasesinvolumeareeffectivemostlyinaddingareaunder thepowercurverightatthepowerpeakreducingoutletpipediameterwill,iftheexistingpipeistoolarge,boostpowerovertheentirerange,butrequiresthatyoukeepa closewatchoverpistoncrowntemperature.Thereareother,moresubtlethingsthatmayneedcorrecting:forexample,thoseohsohelpfulsonicwavesarenotalways contentjusttoreverberateupanddowntheentirelengthoftheexpansionchamber.Asithappens,eachpartofthesystemtendstodevelopitsownindividual resonances,andtheleadinpipe,diffuser,baffleconeandoutletpipewilleachhavetheirownlittlewavesystemrattlingbackandforth,withotherresonancesoflower frequenciesoccurringinpairedpartsofthechamber.Inmostcases,theseincidentalwavepatternsgounnoticed,butsometimestheywillfallintomutuallyreinforcing activityatcertainenginespeedsandcombinetoforcehumpsandhollowsintowhatwouldotherwisebeanice,evenpowercurve.Whenthesedooccur,oftenaslight shiftingonlengthsisenoughtokeepthemfrommarchinglockedstepandinterferingwiththejobyouaretryingtodowiththeprimarywavemotions. Farmorecommon,whenyoufitanexpansionchamberexhaustsystemonsomeengineintendedforanunassumingmuffler,isthatthebalancebetweenthemixture deliveredbythecarburetorandtherequirementsimposedbyconditionsinthecombustionchamberisdestroyed.Regrettably,thehighoutputtwostrokeengineistoa verylargeextentgasolinecooled,requiringrathermorefuelmixedwithitsairthanisneededtosatisfythesimplechemicalcombinationofoxygenandhydrocarbonsinthe burningprocess.Inlightofthis,youwillappreciatethatanexpansionchamberarrangedtoprovideanenormousboostatsomepointalongtheenginespeedrangemakes itabsolutelyimpossibletojetthecarburetorforcleanrunningatallenginespeeds.Thewayitworksisthatanengine'storquepeaksetsthemixturestrengthrequirement withregardtocoolingthepistoncrownbelowitsmeltingpoint,squelchingdetonation,etc.andifthatmixtureistoorichforlower,orhigher,enginespeeds,thenthe powerrangesuffers.Sometimes,jettingforthattorquepeakwillmakethemixturesorichthattheenginewillhardlyrunatallinitslowandmiddleranges.Therewillbe moreonthissubjectinthechaptersoncarburetionandcylinderheaddesignitshouldbeenoughtosay,here,thatfortrulybroadrangeperformance,itmaybenecessary tocompromiseintermsofexpansionchambereffectivenesstostaywithinthetolerancespermittedbythemixtureanordinarycarburetorwilldeliver.Youcandothat, andthentakeupsomeoftheslackinengineperformancewithahighercompressionratiothanyouwoulddareusewithanexpansionchamberthatgaveabignarrow rangeboost. Twoyearsago,youwouldn'thavebeenbotheredwithanythingbeyondtheexpansionchamber'soutletpipetoday,you'llhavetofitamuffler,anditisimportant thatyouknowwhatsuchanendfittingmeansinoverallperformance.Properlymanaged,theadditionofamufflertoyourexpansionchamberwillhaveabsolutelyno effectonpoweroutput,andwillincreaseonlyslightlytheexhaustsystem'ssizeandweight.Tryingtoaccomplishthesamethingwitharacingfourstrokeenginewouldbe difficulttothepointofimpossibility,butthetwostroke'sexpansionchambermusthavearestrictedoutletinanycase,anditcareslittlewhethertherestrictionisprovided byasimplebitofpipe,orbyamuffler.Ihavedynotestedanumberofexpansionchambers,madeasreplacementsforstockmufflers,thatactuallyshowedagainin powerwithanaddonmuffler.Ofcourse,thesechambersclearlyhadtoolargeoutletpipes,andthesameresultcouldhavebeenobtainedwithpiecesofsmaller diametertubing.Still,itwouldbethesameresult,whichmeansthatthereisnoperformancepenaltytobepaidwhenanexpansionchamber'soutletpipeiscappedwitha muffler. Theonlyrealproblemsyouwillencounterinmufflinganexpansionchamberexhaustsystemare:mufflingeffectivelyandmufflingwithoutaddingtoomuchflow resistance.Thelatterproblemisofminorimportanceifyouhaveaccesstoadynamometer,butassumescriticalproportionswhenyoumustrelystrictlyonseatofthe pantstesting.Addingtoomuchrestriction,withthemuffler,willproducepreciselythesameeffectaswhentheoutletpipeismadetoosmall:drasticoverheatingofthe engine.Predictingflowresistancethroughamufflerisverymuchmoredifficultthanforastraightpipe,sothetrickistouseamufflerthatoffersastraightthroughpassage fortheexhaustgases,andapassageofsuchdiameterthatitwilladdlittleornoresistancetothatalreadyprovidedbythechamber'soutletpipe.Withamufflermeeting thatspecification,theexpansionchambersimplywillnotknowthatitisn'treleasingitspulsesstraightintotheatmosphere.Therewillbenospecialprobleminfindingthat kindofmuffler,asseveralmanufacturerstheirintereststimulatedbytherecentAMArulerequiringthatallmotorcyclesinsportsmanclassracingbemuffledarenow producingstraightthroughsilencerstospigotmountonoutletpipesofvariousdiameters. Thisbringsustotheproblemofeffective silencing,andItakenopleasureintellingyouthatnoneofthemufflersbeingofferedforsaleasthisiswrittenare

pantstesting.Addingtoomuchrestriction,withthemuffler,willproducepreciselythesameeffectaswhentheoutletpipeismadetoosmall:drasticoverheatingofthe engine.Predictingflowresistancethroughamufflerisverymuchmoredifficultthanforastraightpipe,sothetrickistouseamufflerthatoffersastraightthroughpassage fortheexhaustgases,andapassageofsuchdiameterthatitwilladdlittleornoresistancetothatalreadyprovidedbythechamber'soutletpipe.Withamufflermeeting thatspecification,theexpansionchambersimplywillnotknowthatitisn'treleasingitspulsesstraightintotheatmosphere.Therewillbenospecialprobleminfindingthat kindofmuffler,asseveralmanufacturerstheirintereststimulatedbytherecentAMArulerequiringthatallmotorcyclesinsportsmanclassracingbemuffledarenow producingstraightthroughsilencerstospigotmountonoutletpipesofvariousdiameters. Thisbringsustotheproblemofeffective silencing,andItakenopleasureintellingyouthatnoneofthemufflersbeingofferedforsaleasthisiswrittenare particularlyeffective.Theyare,ofcourse,betterthannothing,buttheywillnotreducethenoiseoutputfromyourmotorcycledowntoeventhe92dbAmaximum specifiedbytheAMA'srule.Worse,mostofthemrelyonfiberglasspackingfortheirsounddampingpropertiesandthisfiberglassfairlyquicklydisintegratesandis blownoutwiththeexhaustgases,orloadswithcarbonandoil(whichkillsitsabilitytoattenuatesoundwaves),orboth.Onlyfrequentreplacementofthefiberglass packingwillkeepsuchmufflersinworkingorder,andwemayveryshortlyfacefederalregulationsthatspecificallyforbidtheuseoffiberglassinmufflers.Andthesame regulationsprobablywillspecifyamaximumnoiselevelforevenoffroadvehiclessomewhatbelowtheAMA's92dbAlimit alimitthatisnotnowbeingmetby fiberglasspackedmufflersinperfectworkingorder. Theseseveraldifficultiesshouldconvinceanyonethatadifferentapproachtotheproblemofeffectivelysilencingtheexpansionchamberisrequired.Lackinga moreeffectivesolutiontotheproblem,wemayeventuallybeforcedtoreverttoastraightforwardmufflerinplaceoftheexpansionchamberandlivewiththelossof powerandperformancethatentails.Idonotbelievethatwillbenecessary,asIstumbleduponaphenomenonafewyearsagothatmeantverylittleatthetimebutnow assumesmajorimportance:Thethenexistinggeneralracingregulationsrequiredthatamotorcycle'sexhaustsystemterminateatsomepointforwardoftherearmostedge ofthebacktire,andIwasplanningtoraceabikewithitscylindersreversedtoproviderearfacingexhaustports(forreasonsthatwereimportant,butnotpertinenthere). Theonlymajorflawinthisschemewasthatevenwiththemotorcyclebuiltonalongishwheelbaseandwithitsenginelocatedwellforward,therewasnotquiteenough roomfortheexhaustpipeswithinthelengthallowedbytherules.Theexpansionchambersthemselveswouldfit,buttherewassome12inchesofoutletpipelefthanging backbehindthereartire,andnotenoughroomtocurltheseoutletpipesbackwithinthelimit.Whilegropingforasolution,Ihitupontheideaofsimplyslidingthem forward,insidethebafflecones.There,theywouldstillfunctionaspressurebleedresisters,andfurthercontemplationledmetotheconclusionthattheexpansion chambersmightevenworkbetterwiththeiroutletpipespositionedinside.Withtheforwardendoftheoutletpipelocatedatthechamber'smaximumdiameter,aheadof thebafflecone,thereshouldbeasomewhatstrongerreflectionfromthebaffle,andthatmightverywellgivetheengineasomewhatbetterboost.OrsoIthought. Anyway,Igavetheschemeatry,andwhilecertainothermodificationspreventeddrawinganyabsoluteconclusionsfromtheexperiment,thebikedidprovetobe veryfast,anditseemedcertainthatwhilemyinsidestingersmightnotofferanyrealpoweradvantage,theyprobablywereatleastaseffectiveasthoseattachedinthe moreconventionalmanner.ButthatisnottosaythatIdidnotnoticeadifferenceandthedifferencewasinsound.Withthoseinsideoutletpipes,thetypicalexpansion chambercracklewasverynoticeablysubdued.Thatmadesense,asthechambers'outlettoatmospherewastakenfromapointwherethepulsewasatitslowest amplituderatherthanfromthehighpressureareaatthetipofthebafflecone. Morerecently,Ihavebeenabletoperformaseriesoftestsusingadynamometer,toseeifmyinsidestingeridea(which,Ihadlearned,wasanideaalso advancedbyanobscureGermanresearchersomeseveralyearsearlier)hadanyrealmerit.Anexpansionchamberwasconstructedwithitsbaffleconeterminatingina clamp,toallowrapidchangesinoutletpipes.andwetriedpipesofdifferentdiametersandlengths,andmovedthembackandforthintheclamptofindthechangein outputastheoutlet'sforwardendwasadvancedupthebafflecone.Theresultsweremostinteresting:Therewasabsolutelynodifferenceinpoweroutputwiththeoutlet pipeinfullforwardandfullbackpositions,butwedidfindaquitenoticeabledropinpowerwiththepipe'sforwardendpusheduptotheapproximatelyhalfwaypointin thebafflecone.Wealsofoundthatthesameoutletpipediameterproducedoptimumresultsnomatterwhatthelocation,butthatthesystemwasratherlesssensitiveto outletpipelengthwhenthepipe'sforwardendwaslocatedaninchorsoaheadofthebafflecone'sforwardend.Finally,wefoundthatthenoiseoutputwiththeforward locatedoutletpipewasverymuchreduced:toaboutthesamelevelasaconventionalexpansionchamberfittedwithabolton,fiberglasspackedmuffler.Iwas,of course,abitdisappointedthatmyinsidestingerpipedidnotshowabigadvantageinpowerovertheconventionalvariety(theremaybeaslightbroadeningofthepower curve,butthedifferencesobservedweretoosmalltoofferconclusiveproof).Still,bythattime,thesounddampingpropertiesoftheinsidestingerarrangementhad becomeextremelyimportant,astheycouldbeusedinconjunctionwithalowresistancemuffler,locatedfartherdownstream,tomeettheAMA'snewnoiselimitrule withoutanylossofperformance.Forthatreason,thisidea thecreature,indirectly,oftheAMAcompetitionrulesbookdidprovetobeasuccess,evenifnotin preciselytheformIhadanticipated. Giventhedisadvantagesofthefiberglasspackedmuffler,betterdesignsareneededandalreadyarebeginningtoappear.Yamaha,forexample,havequitean effectivemufflerfortheirexpansionchamberequippedmotocrossmotorcycles.Thisoneconsistsofaperforatedtubepassingthroughacanister,withthecenterofthe tubepluggedtoforcetheexhaustpulsesoutthroughtheperforationsinthefirsthalfofthetubesand,intothecanister,wheretheyescapebackthroughtheholesinthe tube'ssecondhalfandthenoffintotheatmosphere.Passagethroughtheholes,whichhavediametersofabout5mm,breaksupthepulse,anditisfurtherattenuatedby expansioninsidethecanister.MyonlyconcernhereisthatYamaha'snewmufflermayoverrestricttheexpansionchamberoutlet,butgiventhatcompany'sthorough approachtoengineeringandtesting,thatseemsaremotepossibility.Howeverandthisisnotmyconcern,buttheAMA'sIdoubtthatYamaha'smuffledmotocrosseris reallyquietenoughtomeettheAMA's92dbAlimit.Perhapssoperhapsnot.Inanycase,theexpansionchamberisheretostay,andsoisthemovementtorestrict noise.Theproblemsofeffectivemufflingwillbesolved,andIthinkmyinsidestingermayhelpwiththesolution.

approachtoengineeringandtesting,thatseemsaremotepossibility.Howeverandthisisnotmyconcern,buttheAMA'sIdoubtthatYamaha'smuffledmotocrosseris reallyquietenoughtomeettheAMA's92dbAlimit.Perhapssoperhapsnot.Inanycase,theexpansionchamberisheretostay,andsoisthemovementtorestrict noise.Theproblemsofeffectivemufflingwillbesolved,andIthinkmyinsidestingermayhelpwiththesolution.

EXPANSIONCHAMBERDESIGNFORMULAE
Where:L t =tunedlength,ininches Eo=exhaustopenperiod,indegrees
Vs =sonicwavespeed(use1700feetpersecond)

N =crankshaftspeed,inrevolutions per minute

Dl= (seetext)

D3=D 1x(0.57to0.62,seetext) A = (halfthediffuser'sangleofdivergence)

reallyquietenoughtomeettheAMA's92dbAlimit.Perhapssoperhapsnot.Inanycase,theexpansionchamberisheretostay,andsoisthemovementtorestrict noise.Theproblemsofeffectivemufflingwillbesolved,andIthinkmyinsidestingermayhelpwiththesolution.

EXPANSIONCHAMBERDESIGNFORMULAE
Where:L t =tunedlength,ininches Eo=exhaustopenperiod,indegrees
Vs =sonicwavespeed(use1700feetpersecond)

N =crankshaftspeed,inrevolutions per minute

Dl= (seetext)

D3=D 1x(0.57to0.62,seetext) A1= (halfthediffuser'sangleofdivergence)


Dl= (seetext)

D3=D 1x(0.57to0.62,seetext) A1= (halfthediffuser'sangleofdivergence) A2= (halfthebafflecone'sangleofconvergence) USEFULCOTANGENTS: cot3.0 =19.0811 o cot3.5 =16.3499 o cot4.0 =14.3007 o cot4.5 =12.7062 o cot5.0 =11.4301 o cot5.5 =10.3854 o cot6.0 =9.51436 o cot6.5 =8.77689 o cot7.0 =8.14435 o cot7.5 =7.59575 o cot8.0 =7.11537 o cot8.5 =6.69116 o cot9.0 =6.31375 o cot9.5 =5.97576 o cot10 =5.67128 o cot11 =5.14455 o cot12 =4.70463 o cot13 =4.33148 o cot14 =4.01078 o cot15 =3.73205
o

CENTERLINELENGTHOFCURVEDPIPE L =Rx0.01745xangleofbend Mostoftheliteraturedealingwiththetwostrokeengine'sporttimingsmakesaseriousandfundamentalerrorinconcentratingontimingtotheexclusionof everythingelse.Inmanyinstancesveryimpressivelookinggraphswillbepresented,toillustratehowexhaustporttiming,forexample,willadvancewithincreasesin enginespeed,etc.,etc.,butallseemtoslipquietlypastacoupleoffactsthatshouldarresteveryone'sattention:First,itshouldbecleartoanyonewhohasmadeevena cursorystudyofthevarioustwostrokeengines'porttimingsthatnoclosecorrelationbetweenthesetimingsandenginespeedexists.Second,itshouldbeequallyclear thatthebasicfunctionoftheseportswhichistopermitavolumeofgasesunderpressuretoescapeconfinementisnotsolelydependentupontime.Timeisrequired, obviously,buttheprocessisonethatalsoissensitivetoarea.Inotherwords,atanygivenpressuredifferentialthetimerequiredtomoveavolumeofgasesthroughaport inthesideofacylinderwallwillbemostlyafunctionoftheport'scrosssectionalarea.Thus,itisnotpossibletotreatwithtimingalonewhenplanningtheportingina twostrokeengineyoumustfindthecombinationoftimeandareathatwillpermitcompletepassageofthegasvolumeathand. ThistimearearequirementhasbeenknowntomealotlongerthanIhavebeenabletoeffectivelyquantifytheconcept.Itisonethingtoknowthattimearea numbersapplicabletothewholerangeoftwostrokemotorcycleenginesprobablyexist,andquiteanothertofindthem.WhatIneededfirstwasamathematical expressionthatwouldcombinethetimefactorwithportarea,andincludeanadjustmentfortheratiobetweenportareaandunitcylindervolume.Butwhilethismight appeartobefairlysimple,thereisaseriousdifficultyinarrivingatavalueforportarea.Justmeasuringthetotalportareawouldnotsuffice,fortheportisfullyopenonly forthatinfinitesimallybriefmomentwhenthepistonrestsattheendofitsstrokeatallothertime,duringtheperiodwhenaportisnominallyopen,somepartoftheport windowismaskedbythepiston.Worse,thedegreeofmaskingvariestosomeextentaccordingtodifferencesinratiobetweentheoncenterslengthofconnectingrod andstroke,andalsowiththeabsoluteportopenduration.Inconsequence,anystudyofporttimeareawouldhavetobebasedonnumbersnotreadilyavailableunless onehasendlesshourstodevotetocollectinginformationeitherlockedawayinsidetheenginesthemselvesorinsidemanufacturer'sarchives.Itispossibletoworkout timeareanumbersonapurelytheoreticalbasis,butthisapproachisevenmoretimeconsuming,involvingasitdoessomereallynastyworkwithintegralcalculus equations.Also,thisapproachvirtuallydemandsthatonemakesomefairlyshakyassumptionsinanumberofareasandthatacomputerbeathandifallthecalculations aretobecompletedinthiscentury. AllthischangedwhenIacquiredanSAEpaperpresentedbyYamaha'sNaitohandNomura.Inthispaper,timeareavaluessuitablefortwostrokemotorcycle enginesaregiven,andwhiletherewasnoexplanationofhowthesewerederivedparticularlywithreferencetothedeterminationofportareatheydidprovidethekey thatunlockedwhathadbeenagreatmystery.WhilepuzzlingovertheNaitohNomuratimeareavalues,Irecalledthatoneofthepioneersinthefield,Schweitzer,had basedhiscalculationsonameanportareathatwassimplytheaperturepresentedwhenthepistonwashalfwaytowardthefullyopenpositionattheendofitsstroke. Thatistosay,halfwayintermsofdegreesofcrankangle,andnotwherehalftheportisuncovered.Forexample,withanexhaustportthatopens90degreesbefore bottomcenter,themeanportareawouldbetakenwiththepiston45degreesofcrankanglefrombottomcenterwhichinmostengineswilluncoverabout70percentof thetotalportwindowarea.AssumingthatNaitohandNomurawereworkingalongthoselines,IworkedouttimeareavaluesforacoupleofYamaharacingenginesfor whichIhadalltheapplicabledata,andtheyfellneatlyintoplace.Actually,theNaitohNomurapaperwasratherobscureonthispointandIwasforcedtodoalotof assuming,butitdidprovideapatternintimearearelationshipsthat,onceunraveledandcheckedagainstconcreteexamplesprovidedbyYamaha'sengines,enabledmeto presentwhatIconsidertobevalidnumbers. SPECIFICTIMEAREA Thenumbersgivenhereexpresstimeandarea,andtheratiobetweenportwindowareaandcylindervolume.Theyrepresentnarrowlydefinedguidelinesfor intake,transferandexhaustporttimeareasasfollows: Forpistoncontrolledintakeports,0.00014to0.00016seccm2/cm3 Fortransferports..,0.00008to0.00010seccm2/cm3
2 3

CENTERLINELENGTHOFCURVEDPIPE L =Rx0.01745xangleofbend Mostoftheliteraturedealingwiththetwostrokeengine'sporttimingsmakesaseriousandfundamentalerrorinconcentratingontimingtotheexclusionof everythingelse.Inmanyinstancesveryimpressivelookinggraphswillbepresented,toillustratehowexhaustporttiming,forexample,willadvancewithincreasesin enginespeed,etc.,etc.,butallseemtoslipquietlypastacoupleoffactsthatshouldarresteveryone'sattention:First,itshouldbecleartoanyonewhohasmadeevena cursorystudyofthevarioustwostrokeengines'porttimingsthatnoclosecorrelationbetweenthesetimingsandenginespeedexists.Second,itshouldbeequallyclear thatthebasicfunctionoftheseportswhichistopermitavolumeofgasesunderpressuretoescapeconfinementisnotsolelydependentupontime.Timeisrequired, obviously,buttheprocessisonethatalsoissensitivetoarea.Inotherwords,atanygivenpressuredifferentialthetimerequiredtomoveavolumeofgasesthroughaport inthesideofacylinderwallwillbemostlyafunctionoftheport'scrosssectionalarea.Thus,itisnotpossibletotreatwithtimingalonewhenplanningtheportingina twostrokeengineyoumustfindthecombinationoftimeandareathatwillpermitcompletepassageofthegasvolumeathand. ThistimearearequirementhasbeenknowntomealotlongerthanIhavebeenabletoeffectivelyquantifytheconcept.Itisonethingtoknowthattimearea numbersapplicabletothewholerangeoftwostrokemotorcycleenginesprobablyexist,andquiteanothertofindthem.WhatIneededfirstwasamathematical expressionthatwouldcombinethetimefactorwithportarea,andincludeanadjustmentfortheratiobetweenportareaandunitcylindervolume.Butwhilethismight appeartobefairlysimple,thereisaseriousdifficultyinarrivingatavalueforportarea.Justmeasuringthetotalportareawouldnotsuffice,fortheportisfullyopenonly forthatinfinitesimallybriefmomentwhenthepistonrestsattheendofitsstrokeatallothertime,duringtheperiodwhenaportisnominallyopen,somepartoftheport windowismaskedbythepiston.Worse,thedegreeofmaskingvariestosomeextentaccordingtodifferencesinratiobetweentheoncenterslengthofconnectingrod andstroke,andalsowiththeabsoluteportopenduration.Inconsequence,anystudyofporttimeareawouldhavetobebasedonnumbersnotreadilyavailableunless onehasendlesshourstodevotetocollectinginformationeitherlockedawayinsidetheenginesthemselvesorinsidemanufacturer'sarchives.Itispossibletoworkout timeareanumbersonapurelytheoreticalbasis,butthisapproachisevenmoretimeconsuming,involvingasitdoessomereallynastyworkwithintegralcalculus equations.Also,thisapproachvirtuallydemandsthatonemakesomefairlyshakyassumptionsinanumberofareasandthatacomputerbeathandifallthecalculations aretobecompletedinthiscentury. AllthischangedwhenIacquiredanSAEpaperpresentedbyYamaha'sNaitohandNomura.Inthispaper,timeareavaluessuitablefortwostrokemotorcycle enginesaregiven,andwhiletherewasnoexplanationofhowthesewerederivedparticularlywithreferencetothedeterminationofportareatheydidprovidethekey thatunlockedwhathadbeenagreatmystery.WhilepuzzlingovertheNaitohNomuratimeareavalues,Irecalledthatoneofthepioneersinthefield,Schweitzer,had basedhiscalculationsonameanportareathatwassimplytheaperturepresentedwhenthepistonwashalfwaytowardthefullyopenpositionattheendofitsstroke. Thatistosay,halfwayintermsofdegreesofcrankangle,andnotwherehalftheportisuncovered.Forexample,withanexhaustportthatopens90degreesbefore bottomcenter,themeanportareawouldbetakenwiththepiston45degreesofcrankanglefrombottomcenterwhichinmostengineswilluncoverabout70percentof thetotalportwindowarea.AssumingthatNaitohandNomurawereworkingalongthoselines,IworkedouttimeareavaluesforacoupleofYamaharacingenginesfor whichIhadalltheapplicabledata,andtheyfellneatlyintoplace.Actually,theNaitohNomurapaperwasratherobscureonthispointandIwasforcedtodoalotof assuming,butitdidprovideapatternintimearearelationshipsthat,onceunraveledandcheckedagainstconcreteexamplesprovidedbyYamaha'sengines,enabledmeto presentwhatIconsidertobevalidnumbers. SPECIFICTIMEAREA Thenumbersgivenhereexpresstimeandarea,andtheratiobetweenportwindowareaandcylindervolume.Theyrepresentnarrowlydefinedguidelinesfor intake,transferandexhaustporttimeareasasfollows: Forpistoncontrolledintakeports,0.00014to0.00016seccm2/cm3 Fortransferports..,0.00008to0.00010seccm2/cm3 Forexhaustports..,0.00014to0.00015seccm2/cm3 Forrotarydiscintakevalves,,,,,,,,,0.00018to0.00019seccm2/cm3 Someofyoumaynotquiteunderstandthatexpression,seccm2/cm3.Itmaylookatrifleintimidating,butitmerelyindicatesthatthenumberprecedingitwas derivedbydividingcylindervolume,incubiccentimeters(cm3),intothemeanareainsquarecentimeters(cm 2)oftheportinquestion,andthenmultiplyingtheresulting numberbythetotaltime,inseconds,duringwhichtheportisopen.Itsays,inshort,timeareaperunitdisplacement,anditremovesallthemysterythathassolong surroundedporttiming. Toworkanytimeareaproblem,youmustfirstbeabletoconvertanengine'stiming,indegrees,intoactualtimeatsomegivenenginespeed.Letusbeginwitha hypotheticalexhaustporttimingof170degreesduration,andanenginespeedof7000rpmthespeedatwhichweintendthattheengineshoulddevelopmaximum power.Toconverttheseknownfactorsintotime,weusethefollowingformula: Where,Tistime,inseconds Nisenginespeed,inrpm istheportopenperiod,indegrees Thus, T=0.004seconds DeterminingT,ortime,isquitesimpletheprocessoffindingaport'smeanareaissomewhatmorecomplicatedandinfinitelymoretimeconsuming.Thereare twobasicapproachestofindingmeanportarea:Themostdirectistosimplyboltadegreewheeltotheendofthecrankshaft,alignitforTDC,thenfindthepointatwhich theportbeginstoopenand,finally,cranktheengineovertoapointhalfwaybetweenthepointofportopeningandbottomcenter(ortopcenter,iftheintakeportisbeing checked).Withthepistonatthathalfwaymark,measuretheareaoftheportwindowthatisexposedandyou'llhavethemeanportareawhichrepresentstheaverage sizeoftheportwindowaperturethroughthewholeperiodfromopeningtoclosing. Thesecondmethodforfindingmeanportarea,andtheoneIprefer,istotransferallofthepertinentdimensionstoasheetoffinegridgraphpaperandworkfrom that.Beginbydrawingaverticallinetorepresentthecylinderaxisandthen,towardthebottomofthatline,addacirclerepresentingthepathfollowedbythecrankpin. Withthatdone,measureupfromthetopofthecircleadistanceequaltotheoncenterslengthoftheconnectingrodandaddahorizontallineatthatpoint.Repeatthis, measuringfromthebottomofthecircle,andyouhavecreatedthespacesweptbythepiston.Theportsarethendrawnin,flankingthelinethatrepresentstheboreaxis, withtheexhaustandtransferportsdownagainstthebottomofthespaceandtheintakeportsituatedupatthetop.Finally,markthemeanportopenpointsonthecircle, usingaprotractor,andconnectthesepointsandtheboreaxislinewithlinesofpreciselythelengthoftheconnectingrod,centertocenter.Youwillfindthatthe connectingpoints,ontheboreaxisline,fallabout70percentdownontheexhaustport,about75percentdownthetransferports(ofwhichonlyonesideneedbe shown)andabout65percentupontheintakeport.Drawhorizontallinesthroughtheportwindowsatthesepoints,andyou'llthenbeabletomeasurethemeanopen areasassumingthatyouhaveaccuratelyreproducedallthedimensions.

usingaprotractor,andconnectthesepointsandtheboreaxislinewithlinesofpreciselythelengthoftheconnectingrod,centertocenter.Youwillfindthatthe connectingpoints,ontheboreaxisline,fallabout70percentdownontheexhaustport,about75percentdownthetransferports(ofwhichonlyonesideneedbe shown)andabout65percentupontheintakeport.Drawhorizontallinesthroughtheportwindowsatthesepoints,andyou'llthenbeabletomeasurethemeanopen areasassumingthatyouhaveaccuratelyreproducedallthedimensions.

Timeareastudiesmaybemadedirectlyfromanengine,butabettermethodistotransfertheappropriatecrank,rod,andportdimensionstoasheetofgraphpaper,thentake meanportareasfromthedrawing.

Youwillfindthatthemostdifficultjobistoaccuratelyreproducetheportwindowshapes.Exhaustportstendtobemorenearlyroundthansquareinhighoutput engineswithwide,unbridgedportsintakeportshaveroundedsidesevenwhentheyarebasicallyrectangulartransferportsoftenhave\angledupperedgesandin additiontothesedifficulties,youmustdetermineaccuratelytheradiiexistingatthecornersofallports.So,findingtheareaofagivenportapertureisnotsimplyamatter ofmultiplyingwidthtimesheight.Myownmethodforresolvingthisdifficultyistomakeaclosefittingpapersleeve,slideitinsidethecylinderbeingchecked,tapeitin placeandthen,byrubbingaroundtheedgesoftheportwindowswiththesideofapenciltip,transferthewindowshapestothepaper.Pleaseunderstandthatthese shapesmaynotthenbetransferreddirectlytothegraphpaperonwhichthebore,connectingrod,andcrankshaftlineshavebeendrawn.Why?Becauseinlayingthe sleeveflat,thewidthoftheportsisexaggerated,andwhatyouneedisthetruesizeoftheaperture.Myapproachistomakethispapercylinderoutoffairlystiffishstock, andaftergettingtheportshapesmarkedonitsinnersurfacetoreverseitssides,creatingacylinderwiththeportwindowdrawnoutsidewheretheycanconvenientlybe measuredwithprecisecalipersandthetruedimensionsthentransferredtomyworkingdrawing.Aseriesofhorizontalguidelinesdrawnacrosstheportshapesbeforere rollingthecylinderisanaidinmeasuringtheport'sshape,butinthefinalanalysisaccuracydependsonone'swillingnesstobemeticulouswithallmeasurementsandtobe imaginativeinconcoctingwaysofrecreatingaportshapeonpaper.Ihavefoundthatadraftsman'sellipsetemplateisveryusefulinthiskindofwork,butitisonly

usingaprotractor,andconnectthesepointsandtheboreaxislinewithlinesofpreciselythelengthoftheconnectingrod,centertocenter.Youwillfindthatthe connectingpoints,ontheboreaxisline,fallabout70percentdownontheexhaustport,about75percentdownthetransferports(ofwhichonlyonesideneedbe shown)andabout65percentupontheintakeport.Drawhorizontallinesthroughtheportwindowsatthesepoints,andyou'llthenbeabletomeasurethemeanopen areasassumingthatyouhaveaccuratelyreproducedallthedimensions.

Timeareastudiesmaybemadedirectlyfromanengine,butabettermethodistotransfertheappropriatecrank,rod,andportdimensionstoasheetofgraphpaper,thentake meanportareasfromthedrawing.

Youwillfindthatthemostdifficultjobistoaccuratelyreproducetheportwindowshapes.Exhaustportstendtobemorenearlyroundthansquareinhighoutput engineswithwide,unbridgedportsintakeportshaveroundedsidesevenwhentheyarebasicallyrectangulartransferportsoftenhave\angledupperedgesandin additiontothesedifficulties,youmustdetermineaccuratelytheradiiexistingatthecornersofallports.So,findingtheareaofagivenportapertureisnotsimplyamatter ofmultiplyingwidthtimesheight.Myownmethodforresolvingthisdifficultyistomakeaclosefittingpapersleeve,slideitinsidethecylinderbeingchecked,tapeitin placeandthen,byrubbingaroundtheedgesoftheportwindowswiththesideofapenciltip,transferthewindowshapestothepaper.Pleaseunderstandthatthese shapesmaynotthenbetransferreddirectlytothegraphpaperonwhichthebore,connectingrod,andcrankshaftlineshavebeendrawn.Why?Becauseinlayingthe sleeveflat,thewidthoftheportsisexaggerated,andwhatyouneedisthetruesizeoftheaperture.Myapproachistomakethispapercylinderoutoffairlystiffishstock, andaftergettingtheportshapesmarkedonitsinnersurfacetoreverseitssides,creatingacylinderwiththeportwindowdrawnoutsidewheretheycanconvenientlybe measuredwithprecisecalipersandthetruedimensionsthentransferredtomyworkingdrawing.Aseriesofhorizontalguidelinesdrawnacrosstheportshapesbeforere rollingthecylinderisanaidinmeasuringtheport'sshape,butinthefinalanalysisaccuracydependsonone'swillingnesstobemeticulouswithallmeasurementsandtobe imaginativeinconcoctingwaysofrecreatingaportshapeonpaper.Ihavefoundthatadraftsman'sellipsetemplateisveryusefulinthiskindofwork,butitisonly essentialthatyouhaveagoodprotractor,astraightedge,acompass,andasupplyoffinegridgraphpaper.Byassigningavalueofonemillimeterperline,anddoingall

Timeareastudiesmaybemadedirectlyfromanengine,butabettermethodistotransfertheappropriatecrank,rod,andportdimensionstoasheetofgraphpaper,thentake meanportareasfromthedrawing.

Youwillfindthatthemostdifficultjobistoaccuratelyreproducetheportwindowshapes.Exhaustportstendtobemorenearlyroundthansquareinhighoutput engineswithwide,unbridgedportsintakeportshaveroundedsidesevenwhentheyarebasicallyrectangulartransferportsoftenhave\angledupperedgesandin additiontothesedifficulties,youmustdetermineaccuratelytheradiiexistingatthecornersofallports.So,findingtheareaofagivenportapertureisnotsimplyamatter ofmultiplyingwidthtimesheight.Myownmethodforresolvingthisdifficultyistomakeaclosefittingpapersleeve,slideitinsidethecylinderbeingchecked,tapeitin placeandthen,byrubbingaroundtheedgesoftheportwindowswiththesideofapenciltip,transferthewindowshapestothepaper.Pleaseunderstandthatthese shapesmaynotthenbetransferreddirectlytothegraphpaperonwhichthebore,connectingrod,andcrankshaftlineshavebeendrawn.Why?Becauseinlayingthe sleeveflat,thewidthoftheportsisexaggerated,andwhatyouneedisthetruesizeoftheaperture.Myapproachistomakethispapercylinderoutoffairlystiffishstock, andaftergettingtheportshapesmarkedonitsinnersurfacetoreverseitssides,creatingacylinderwiththeportwindowdrawnoutsidewheretheycanconvenientlybe measuredwithprecisecalipersandthetruedimensionsthentransferredtomyworkingdrawing.Aseriesofhorizontalguidelinesdrawnacrosstheportshapesbeforere rollingthecylinderisanaidinmeasuringtheport'sshape,butinthefinalanalysisaccuracydependsonone'swillingnesstobemeticulouswithallmeasurementsandtobe imaginativeinconcoctingwaysofrecreatingaportshapeonpaper.Ihavefoundthatadraftsman'sellipsetemplateisveryusefulinthiskindofwork,butitisonly essentialthatyouhaveagoodprotractor,astraightedge,acompass,andasupplyoffinegridgraphpaper.Byassigningavalueofonemillimeterperline,anddoingall drawingtoscale,youcannotonlyworkwithconsiderableaccuracy,butthedeterminingofwindowareasbecomemucheasieryoujustcountthenumberofsquaresand fractionsofsquareswithinthemeanaperturetogetarea,andifyouhavedrawneverythingcarefully,themarginoferrorwillbetoosmalltobeofanyparticular significance. Havingfoundtheports'meanareas,youarereadytocompleteyourcalculationsinfindingeachport'svalueinseccm2/cm3.Foranexample,we'llreturntothe hypotheticalexhaustportforwhichwehaveaportopenduration(derivedfrom170degreesand7000rpm)of0.004sec.Wewillfurtherassumethatthisporthasa meanareaof8.0cm 2,andisinacylinderhavingadisplacementof250cm 3(or,asitismorecommonlyexpressed,250cc).Toarriveatthefigurewewant,insec cm2/cm3,itisfirstnecessarytodividecylindervolume,250cm3,intoportarea,8.0cm 2,andthenmultiplytheresultsbytime,inthefollowingmanner: 0.032x0.004=0.00013seccm2/cm3 And,referringbacktothespecifictimeareafiguresalreadyprovided,whichestablisharangefrom0.00014to0.00015,welearnthatthisparticularhypotheticalengineis timeareadeficient,intheorderof10percent.Increasingtheportwidth,tobringthemeanopenareaupto9.0cm 2,willraisetheexhaustport'sspecifictimeareato 0.000145,whichisrightinthemiddleofthespecifiedrange.Increasingthisengine'sexhaustopendurationwillincreasebothtimeandmeanarea,andthat'swherethe drawingonthegraphpaperbecomesveryhandy,asitenablesyoutocharthowmuchareaisgainedbyincreasesintiming,andtoestablishwhatcombinationofport width,heightandtimingwillyieldthespecifictimeareaneedfortheparticularenginewithwhichyouareworking.Inmostinstancesyouwillnotbeabletoobtainthe correctvaluesbymerelywideningportsacombinationofincreasesinbothtimeandareaarerequiredifyourstockenginehasapowerpeakat,say,6000rpmandyou wanttopushitathousandorsorpmhigher. ANGLEAREA Withallthejugglingofnumbersentailedinfindingcorrectcombinations,itisadvisabletofindshortcuts,andoneoftheseistoworkwithspecificanglearea numbersinsteadoftimearea.Onecannotsimplysubstituteangleareafortimeareaandignoretheenginespeedfactor,obviously,soIhaveprovidedchartsfromwhich youmaywork.Take,forinstance,theexhaustporttimearea/angleareachart:you'llfindthattherearetwolinemarkingtheexhausttimearealimitsof0.00014and 0.00015seccm2/cm3,withverticallinesmarkingoffenginespeedandhorizontallinesforanglearea.Thespecificangleareaofourhypotheticalexample,takingthe exhaustopendurationas170degreesandthecm 2/cm3figureas0.032(8.0/250),is170x0.032=5.44degcm2/cm3.Referringtothechart,wenotethatat7000rpm thespecificanglearearequirementisfromabout5.88to6.3degcm2/cm3,meaningthatmoretimeand/orareawillbeneedediftheengineistooperateeffectivelyat 7000rpmandthatanangleareavalueintheorderof6.1degcm2/cm3isagoodtargetfigure.Thisprocessis,ofcourse,intendedtoberepeatedforallofanygiven engine'sports. Thesetimeareaandangleareanumbersareenormouslyusefulinplanningtheportingofanytwostrokemotorcycleengine,andIwouldbehappytotellyouthat theyrepresentthefinalwordonthesubjectbuttheyunfortunatelydonot.Manyotherfactorsintrudeuponthisseeminglyuncomplicatedpicture:Referringonceagain tothetimearea/angleareacharts,youwillobservethatthereis,ineveryinstance,arangeforwhatconstitutescorrectvalues,whichreflectsthefactthattheoptimumis influencedsomewhatbyfactorsbeyondthosealreadydiscussed.Sadly,itisinaccountingforthesethatwemustdepartfromthecomfortablecertitudesofmathematics, fortheotherinfluencesareinthefieldofexhaustsystemdesignandcharacteristics,flowcoefficients,nominalandeffectiveportareas,crankcasecompressionratios, resonanceintheintaketract,andverylikelyahostofminorinfluencesthatwillforeverescapeourattention.Alltheseadduptoasituationinwhichacertainamountof adjustment,doneaccordingtoresultsobservedintesting,willalwaysbenecessary.Calculationbringsusclosetotheoptimum,afterwhichweworkwithinthetime area/anglearearangeaccordingtoaneducatedguess,andfinallywetest.Professionalengineersdonotdothejobanydifferently. TIMEAREACOMBINATIONS EventhoughIcannotprovideyouwithformulae,orevenanythingotherthangeneralizedcomment,Icangiveyouatleastsomeindicationastowhatcombinations

resonanceintheintaketract,andverylikelyahostofminorinfluencesthatwillforeverescapeourattention.Alltheseadduptoasituationinwhichacertainamountof adjustment,doneaccordingtoresultsobservedintesting,willalwaysbenecessary.Calculationbringsusclosetotheoptimum,afterwhichweworkwithinthetime area/anglearearangeaccordingtoaneducatedguess,andfinallywetest.Professionalengineersdonotdothejobanydifferently. TIMEAREACOMBINATIONS EventhoughIcannotprovideyouwithformulae,orevenanythingotherthangeneralizedcomment,Icangiveyouatleastsomeindicationastowhatcombinations oftimeareas/angleareasworkbestinspecificinstances:Exhaustsystemcharacteristicsare,forexample,verycloselyrelatedtobothexhaustportandtransferporttime areas.Inamotocrossengine,wherepowerrangeismoreimportantthanmaximumpower,youwouldwantanexhaustsystemthatprovidesarelativelyweakbut extendeddurationresonantpulse(whichmeansthatitwouldbeeffectiveoveraverybroadenginespeedrange)and,relativelyspeaking,alowexhaustporttimearea valueincombinationwithahightransferporttimearea.Infact,iftheengineinquestionisasinglecylinder400,orsomethinginthatgeneraldisplacementrange,thenyou probablywouldselectatransfertimeareaattheverytopoftherange,andanexhausttimeareaestablishedatthebottom.Thatistosay,atransferporttimeareaof 0.00010seccm2/cm3andanexhaustporttimeareaof0.00014seccm2/cm3.Thiscombinationoftimeareasprovideanexhausttimingthatborrowsminimallyfromthe workingstroke,whichmaximizesbroadrangetorque,anditgivesthenegativewavereturningfromtheexhaustsystemabetteropportunityoffindingthetransferportstill openmeaningthatthisnegative,orscavengingpulsewillbeabletohelppullthefreshchargeupfromthecrankcase.Relativelylargetransferporttimeareasalsogive thefreshchargeconditionsallowingittomakeitswayupintothecylinderevenifthereislittleornoassistfromtheexhaustsystem.Thesoledisadvantageofthecondition describedisthatitdoesappeartopermitasomewhatgreaterdegreeofmixingbetweenresidualexhaustgasesandtheincomingcharge,aswellasanenhancedtendency towardshortcircuiting. Roadracingenginesrepresentanentirelydifferentsituation,astheyrespondbest,beingtunedformaximumpowerandwithpowerrangeasecondary consideration,toamaximumvalueforexhaustporttimeareaandaminimumtransferporttimearea.Atleast,thatwillbethecasewhentheengineinquestionisfitted withatypicallyroadracingexpansionchamberexhaustsystem,whichreturnsverystrongbutshortlivedscavengingandpluggingpulsestothecylinder.Also,forreasons thatwillbediscussedlater,roadracingengines'scavengingisaidedbyhighcrankcasepressuresresultingfromstrong,albeitnarrowbandsuperchargingeffectsofsonic waveactivityintheirintaketracts.Withtheextractoreffectsoftheexhaustsystemandtherammingeffectsontheintakesidecombiningtoaidcylindercharging,maximum powerisobtainedbyupperlimitexhaustporttimeareas(tomakebestuseoftheexhaustsystem)andlowerlimittimeareasonthetransfersidetominimizecharge dilutionandshortcircuiting. Carriedtoitsextremes,thedescribedcombinationofsharp,powerfulpulsesfromtheexhaustsystemandlowlimittransferporttimeareacanelevate,and narrow,anengine'spowerbandtoaremarkabledegree.Thetwocylinder,350ccYamahaTR3,forinstance,hasanexhaustporttimeareavalueof0.000148sec cm2/cm3andatransferporttimeareaof0.000081seccm2/cm3,andthesenumbersrepresentverynearlythemaximumandminimumtimeareavalueswithintheir respectiveranges.Poweroutputfromthisenginemustbeintheorderof6365bhp,butthepowerisdevelopedoversuchanarrowrangethata6speedtransmission withultracloseratiosisrequiredtoholditwithinlimits.IaminformedbyaverygoodsourcethatthelatestYamaha250ccTD3hasanevennarrowerpowerband,and thatthemostskilledofridershasconsiderabledifficultykeepingitonthepowercurveunderactualracingconditions.Mysuggestionto

Shownhereingraphformistherelationshipbetweentimeareaandangleareaoverarangeofenginespeeds.

thosewhohavethismachine,andthedifficulties,istoincreasetheengine'stransferporttimeareaslightly.Anincreaseinthetransferopendurationofonly4to6 degreeswouldprobablybroadentheTD3'seffectivepowerbandenoughtomakethemachineverymucheasiertoride,reducingmaximumpowerbyperhapstwobhp andaddingaboutthreeorfourbhpatthelowerlimitofthepresentrange.Thesameappliestoalltwostrokeengines:increasesintransferporttimeareatendtodepress thepowerpeak,butaddtothepowercurveatlowerenginespeeds.

cm2/cm3andatransferporttimeareaof0.000081seccm2/cm3,andthesenumbersrepresentverynearlythemaximumandminimumtimeareavalueswithintheir respectiveranges.Poweroutputfromthisenginemustbeintheorderof6365bhp,butthepowerisdevelopedoversuchanarrowrangethata6speedtransmission withultracloseratiosisrequiredtoholditwithinlimits.IaminformedbyaverygoodsourcethatthelatestYamaha250ccTD3hasanevennarrowerpowerband,and thatthemostskilledofridershasconsiderabledifficultykeepingitonthepowercurveunderactualracingconditions.Mysuggestionto

Shownhereingraphformistherelationshipbetweentimeareaandangleareaoverarangeofenginespeeds.

thosewhohavethismachine,andthedifficulties,istoincreasetheengine'stransferporttimeareaslightly.Anincreaseinthetransferopendurationofonly4to6 degreeswouldprobablybroadentheTD3'seffectivepowerbandenoughtomakethemachineverymucheasiertoride,reducingmaximumpowerbyperhapstwobhp andaddingaboutthreeorfourbhpatthelowerlimitofthepresentrange.Thesameappliestoalltwostrokeengines:increasesintransferporttimeareatendtodepress thepowerpeak,butaddtothepowercurveatlowerenginespeeds. Itshouldbeunderstood,however,thatexcessivetransferporttimearea,incombinationwiththewrongexhaustsystem,canleadtoseriousinstabilityinrunning yieldingamajordropinpeakpowerwithoutadequatecompensationinpowerrange,andapowercurvemarkedbyhumpsandhollows.Thus,whileenginesexistin whichexhaust/transfertimeareaimbalances(relativetothevaluespresentedhere)havenotpreventedquitegoodpoweroutputs,suchimbalancesmayberegardedas extremeexampleanomaliesmoreinterestingfortheirvalueascuriositiesthenaspatternsfromwhichtowork.Inmostenginesthecorrectapproachwillbetoestablish timeareavaluesthatfallwithintherangessuggestedhere,andtomakeadjustmentswithinthoserangesaccordingtotheconditionsforwhichtheengineisintended.Road racingenginesforwhich6speedand/orcloseratiogearsetsarenotavailableshouldbebiasedtowardthemotocrossendofthetimeareaspectrumsmalldisplacement motocrossengineswhichcommonlyarecoupledtofairlycloseratio,5speedtransmissionsshouldbebiasedtowardtheroadracingspecificationsimplybecausethe horsepowerthusgainedhasbecomeacompetitionnecessityeventhougharatherpeakymotocrossengineisnojoytoarider. EMPHASISONAREA Takingeachportindividually,thereiseveryreasontomakeanyportaswideaspossible,acquiringthenecessarytimeareavalueinthismannerinsteadof extendingtheportopenduration.Reasonsformoderatingthisapproachdoexist,however,intheinteractionbetweenportsandintheeffectsexaggeratedexhaustport widthshaveuponpistonsandrings.Thesereasonsarediscussedindetailinthechaptertitled,Scavenging,butIwillIincludeabriefreminderhere.Atoowideexhaust portwillcauseringstosnagandbreak,orwearveryrapidly,andifthewideningbringstheexhaustportwindow'ssidestoonearthetransferports,therewillbean increasedtendencytowardshortcircuitingoftheincomingcharge.Obviously,excessivewideningofthetransferportscanalsoresultinringtrappingand/orchargeshort circuiting.Youshouldalsounderstandthatwideninganengine'sexhaustport,increasingitstimeareavaluewithoutactuallyincreasingitsopenduration,hasmuchthe sameeffectasobtainingthesameincreasebyraisingitsheightandthusincreasingbothtimeandarea:thatistosay,wideningtheexhaustportincreasestheenginespeed atwhichmaximumpowerisrealized,whilereducinglowspeedpower.Andthesamepatternistobeobservedinincreasestotransferporttimearea,thoughinthe oppositedirection.Theseeffectsshouldbecomefamiliartoyou,particularlyasregardstheexhaustport,foranyincreasesinexhausttimeareashouldbeginwithwidening theporttothemaximumtolerabletothepistonandrings,movingontothebusinessofraisingthetopoftheexhaustportonlyafterthelimitforwidthhasbeenreached. Thereisgoodreasonfortakingthisapproach,forwhileincreasesinexhaustporttimearea,gainedbywhateverchangeintheportwindow'sshape,certainlywillhavethe samegeneraleffect,increasingwidthtogetmoretimeareahasamuchlessnarrowingeffectonthepowerbandthanincreasesinheight. TIMINGLIMITS Widthisevenmoreimportantontheintakesideofanyenginewithapistoncontrolledintakeport,astherearesharplimitstotimeareaincreasesgainedby


Shownhereingraphformistherelationshipbetweentimeareaandangleareaoverarangeofenginespeeds.

thosewhohavethismachine,andthedifficulties,istoincreasetheengine'stransferporttimeareaslightly.Anincreaseinthetransferopendurationofonly4to6 degreeswouldprobablybroadentheTD3'seffectivepowerbandenoughtomakethemachineverymucheasiertoride,reducingmaximumpowerbyperhapstwobhp andaddingaboutthreeorfourbhpatthelowerlimitofthepresentrange.Thesameappliestoalltwostrokeengines:increasesintransferporttimeareatendtodepress thepowerpeak,butaddtothepowercurveatlowerenginespeeds. Itshouldbeunderstood,however,thatexcessivetransferporttimearea,incombinationwiththewrongexhaustsystem,canleadtoseriousinstabilityinrunning yieldingamajordropinpeakpowerwithoutadequatecompensationinpowerrange,andapowercurvemarkedbyhumpsandhollows.Thus,whileenginesexistin whichexhaust/transfertimeareaimbalances(relativetothevaluespresentedhere)havenotpreventedquitegoodpoweroutputs,suchimbalancesmayberegardedas extremeexampleanomaliesmoreinterestingfortheirvalueascuriositiesthenaspatternsfromwhichtowork.Inmostenginesthecorrectapproachwillbetoestablish timeareavaluesthatfallwithintherangessuggestedhere,andtomakeadjustmentswithinthoserangesaccordingtotheconditionsforwhichtheengineisintended.Road racingenginesforwhich6speedand/orcloseratiogearsetsarenotavailableshouldbebiasedtowardthemotocrossendofthetimeareaspectrumsmalldisplacement motocrossengineswhichcommonlyarecoupledtofairlycloseratio,5speedtransmissionsshouldbebiasedtowardtheroadracingspecificationsimplybecausethe horsepowerthusgainedhasbecomeacompetitionnecessityeventhougharatherpeakymotocrossengineisnojoytoarider. EMPHASISONAREA Takingeachportindividually,thereiseveryreasontomakeanyportaswideaspossible,acquiringthenecessarytimeareavalueinthismannerinsteadof extendingtheportopenduration.Reasonsformoderatingthisapproachdoexist,however,intheinteractionbetweenportsandintheeffectsexaggeratedexhaustport widthshaveuponpistonsandrings.Thesereasonsarediscussedindetailinthechaptertitled,Scavenging,butIwillIincludeabriefreminderhere.Atoowideexhaust portwillcauseringstosnagandbreak,orwearveryrapidly,andifthewideningbringstheexhaustportwindow'ssidestoonearthetransferports,therewillbean increasedtendencytowardshortcircuitingoftheincomingcharge.Obviously,excessivewideningofthetransferportscanalsoresultinringtrappingand/orchargeshort circuiting.Youshouldalsounderstandthatwideninganengine'sexhaustport,increasingitstimeareavaluewithoutactuallyincreasingitsopenduration,hasmuchthe sameeffectasobtainingthesameincreasebyraisingitsheightandthusincreasingbothtimeandarea:thatistosay,wideningtheexhaustportincreasestheenginespeed atwhichmaximumpowerisrealized,whilereducinglowspeedpower.Andthesamepatternistobeobservedinincreasestotransferporttimearea,thoughinthe oppositedirection.Theseeffectsshouldbecomefamiliartoyou,particularlyasregardstheexhaustport,foranyincreasesinexhausttimeareashouldbeginwithwidening theporttothemaximumtolerabletothepistonandrings,movingontothebusinessofraisingthetopoftheexhaustportonlyafterthelimitforwidthhasbeenreached. Thereisgoodreasonfortakingthisapproach,forwhileincreasesinexhaustporttimearea,gainedbywhateverchangeintheportwindow'sshape,certainlywillhavethe samegeneraleffect,increasingwidthtogetmoretimeareahasamuchlessnarrowingeffectonthepowerbandthanincreasesinheight. TIMINGLIMITS Widthisevenmoreimportantontheintakesideofanyenginewithapistoncontrolledintakeport,astherearesharplimitstotimeareaincreasesgainedby lengtheningtheportopenduration.Pistonportengineshavetheadvantageofsimplicity,butaresomewhat(sometimesseriously)handicappedbythefactthattheirintake timingissymmetricallydisposedbeforeandaftertopcenter.Thereis,therefore,astrongtendencyforthemixtureaspiratedintothecrankcaseduringtheperiodbetween intakeopeningandtopcentertobepushedbackoutduringtheequalportopenperiodbetweentopcenterandintakeclosing.Thistendencyaccountsforthe extraordinaryinfluenceofintaketractresonanceandgasinertiaonthepistonportengine'spowercharacteristics.Thecombinedactivitiesofsonicwavesandtheinertia ofthehighvelocitymixturestreamcansimplyoverpowertherisingpressureinthecrankcasecreatedbythedescendingpiston. Ideally,intakeclosingshouldoccurattheprecisemomentwhenrammingpressureisatitspeakandwhenthatpressureisequaltothepressureinsidethe crankcase,asthisconditionwilltrapthegreatestvolumeofair/fuelmixtureinsidethecrankcase.Unhappily,thisidealcanonlyberealizedwithinverynarrowengine speedranges,asinertiaeffectsdiminishrapidlyatlowerthanplannedspeedsandthenaturalfrequencyoftheintaketractisdeterminedalmostsolelybyits(andthe crankcase's)dimensions,whichmeansthatitpulsesatafixedrate,andonlyatoneparticularenginespeedwillittrulybeworkinginphasewiththemotionsofthepiston. Worse,atverylowenginespeedsneithersonicwaveactivitynortherammingeffectsofgasIinertiawillbestrongenoughtopreventthepistonfromdisplacingpartofthe chargeaspiratedintothecrankcaserightbackoutthroughthecarburetor.Allofwhichmeansthatatcrankingspeeds,whenyouaretryingtostartanengine,thetotal

volumeofthechargebeingdeliveredintothe cylinderwillbedetermined bythatwhichthepistondisplacesbetweenthepointatwhichthetransferportscloseandthepointofintakeopening(whichalsoisintakeclosing).Forexample,ina pistonportroadracingenginewithtransferclosingat115degreesbeforetopcenterandanintakeportopeningandclosing100degreesbeforeandaftertopcenter,the volumeofgasesactuallybeingpumpedthroughthecrankcase,perrevolution,wouldbeonlythatdisplacedbythepistoninamere15degreesofcrankangle. Itisentirelypossiblethatintheexamplegiven,startingwouldprovetobeimpossibleunlesstheenginewasactuallycrankedfastenoughtobringituptothepoint whereintaketractresonanceandinertiabegantohavesomeeffect.Quiteobviously,thisexampledoesrepresentanextreme,butnotoneIthatseriouslydistortsthe conditionbeingillustrated.Yamaha'sTR3racingIengineactuallyhastransferportsthatclose120degreesbeforetopcenterandIanintakeportthatopensandcloses 94degreesbeforeandaftertopcenter,leavingonly26degreesofcrankangleforpumpingenoughmixturetostarttheengine.Iwouldthinkthatthisissomethingvery neartheabsoluteminimumevenforanenginetobestartedbyvigorouspushingofthemotorcycle,anditwouldseverelylimitanyeffortstoimprovethisengine'spower rangethroughincreasesintransferportopenduration. Consideringtheintaketiminglimitationsimposedbythejustdiscussed,itshouldbeclearthatthetaskofobtainingadequatecrankcasefillinginhighspeedengines

Itisentirelypossiblethatintheexamplegiven,startingwouldprovetobeimpossibleunlesstheenginewasactuallycrankedfastenoughtobringituptothepoint whereintaketractresonanceandinertiabegantohavesomeeffect.Quiteobviously,thisexampledoesrepresentanextreme,butnotoneIthatseriouslydistortsthe conditionbeingillustrated.Yamaha'sTR3racingIengineactuallyhastransferportsthatclose120degreesbeforetopcenterandIanintakeportthatopensandcloses 94degreesbeforeandaftertopcenter,leavingonly26degreesofcrankangleforpumpingenoughmixturetostarttheengine.Iwouldthinkthatthisissomethingvery neartheabsoluteminimumevenforanenginetobestartedbyvigorouspushingofthemotorcycle,anditwouldseverelylimitanyeffortstoimprovethisengine'spower rangethroughincreasesintransferportopenduration. Consideringtheintaketiminglimitationsimposedbythejustdiscussed,itshouldbeclearthatthetaskofobtainingadequatecrankcasefillinginhighspeedengines isnotconfinedtoestablishingasuitabletimeareavalue.Enginesformotocrossarerestricted,intermsofportopenduration,bytheneedforaverybroadrangeoutput characteristic,toanintakeperiodofnotmuchmorethan160degrees.Morethanthatvirtuallyguaranteesthattheywillbetoopeakytoberideable,oratleasttobe effectiveintermsofcompetitivelapstimesonmostcircuits,nomatterwhattheirtimeareanumbermaybe.Roadracingengineshaveanultimatelimitimposedbythe startingproblemalreadyoutlined.Allofwhichmeansthatyoumayregardtheupperlimitofintakeopendurationfortheformerasbeingabout160degrees,andabout 200degreesforthelatterabout80and100degreesbeforeandaftertopcenter.Exceedthoselimits,andtheroadracingenginewillnotstartthemotocrossengine's powerbandwillnarrowbeyondthepointofbeingusefulnomatterhowimpressivethemaximumpowerfiguremaysound. Intaketracttuningwillbevitallyimportantnomatterwhatkindoftimeareaisprovidedattheportwindow,anditisalltooeasytogetthepulsationsoutof phasewiththepistonbyalteringtheintaketiming.Allalterationsinintaketimingshouldbefollowedwithacarefulchecktodetermineifmatchingalterationoftheintake tractlengthisnotalsorequired.Althoughthiskindofworkshouldbevalidatedbyactuallyrunningtheenginewithastubexhaustattached,asoutlinedelsewhereinthis book,apreliminarycheckmayberunmathematically,usingtheformulaforfindingtheresonantfrequencyoftheneckedflaskformedbythecrankcaseandintaketract providedinthechapteroncrankcasepumping. Forallofthereasonsoutlinedinthatchapter,whichdealsatlengthwithintaketracttuning,Ihavelittleconfidencethattheuseofthisformulawillprovidemore thanaroughguideasregardsanengine'spropertunedintakelength,butforsomethisroughguidemaybealltheguidetheywillhave.Itisbetterthannothing,ifyou comparethefrequencythusderivedforyourengine'sstockconditionwiththerealityandmakeappropriateadjustmentsinthetheoreticallyobtainednumbersforyour modifications.Itmayalsohelptoknowthatoneofthebetterresearchersinthefield,FujioNagao,ofKyotoUniversity,hasverifiedthatmaximumairdeliveryoccurs whentheintakepipe'snaturalfrequencyprovidesawaveperiod75percentoftheintakeportopenperiod.Withallthat,Istillaminclinedtobelievethatthereisno substituteforactualtesting,usingastubexhaustsystemtoisolateintakeeffects. ROTARYVALVETIMING Blowbackduringtheperiodbetweentopcenterandintakeclosingisthelimitingfactorinestablishingtimeareavaluesforalltwostrokeenginesexceptthose fittedwithautomaticintakevalves(i.e.reedvalves).Buttheproblemismuchlesssevereandmoreeasilymanagedwhenanenginehasrotarydiscintakevalving. Becauseoftheconsiderablemechanicalcomplicationattendingthismethodofcrankcasefilling,rotaryvalveshavenotbeenusedmuchinmassproducedengines,and theinconvenientwidththeyaddnowseemsunacceptableasquitegoodresultscanbeobtainedwitheitherreedorpistonportinduction.Bethatasitmay,therotary valvestillisbestintermsofsheerengineperformance,whetherarrangedformaximumpowerorforanultrabroadpowerrange.Therotaryvalveisfreeofthereally seriousblowbackproblemafflictingpistoncontrolledvalving,anditoffersmuchlessresistancetoflowthanreeds.Peoplewhoarecurrentlysoinfatuatedwiththereed valveconceptshouldconsiderthatintheworldofkarting,wherethereismuchmoreexperiencewithbothreedsandrotaryvalvesthanmotorcyclistshaveaccumulated, thetwotypesofengineshavebeenseparatedintodifferentclasses.Why?Becausewhilethereedvalveenginesareinexpensive,theycannotmatchtheperformanceof thosewithrotaryvalves.So,eventhoughtherotarydiscvalveisforthemomentoutoffashion,thepressureofcompetitionmayeventuallyreturnittotheforedespiteits handicapincostandbulk. Principallybecauseitissofreeofblowbackproblems,thepropertimeareaofarotaryintakevalveismuchhigherthanforthepistoncontrolledport.Thisdoes notmeanthatonemayindiscriminatelywhackawayatthevalvediscwithoutgettingintotrouble,especiallyontheportclosingsideofthedisccutaway.Thevalve's openingpointhasaninfluenceonpoweroutput,butingeneralitisrelativelyinsignificantascomparedwithportclosing.Curiously,thebestportclosingtimingforavery widevarietyofdiscvalveenginesisabout65degreesaftertopcenter,butbeforedecidingtoapplythattimingyoushouldconsiderthatanyappreciabledelayinclosing theport,afterthepistonhasstarteddownfromtopcenter,willcausesomelowspeedblowback.Athigherenginespeedsinertiaeffectsintheintaketractwillovercome theslightblowbackcausedbythedelayedintakeclosing,buttherewillbealossoflowspeedpower.Allthingstakenintoaccount,thebestapproachhereistoincrease thevalveclosingdelayinverysmallstages,notmorethan2degreesatatime,untilthedesiredresultisobtained.Rememberthatretardingtheintakeclosingpointmoves theengine'spowerpeakhigher,whilereducingpoweratthelowerendoftherange.Remember,too,thatchangesinthelengthordiameteroftheoverallintaketract,such aswouldoccurinsubstitutingacarburetorofsomedifferentsize,willalterthepointatwhichportclosingdelayreachesitsoptimum. Itisfortunatethatrotaryvalveenginesarerelativelyinsensitivetothepointofintakeopening,becausethereisnoclearpatterninexistingexamplestolendour effortsdirection.Ifthereisarule,itisthatthepointofopeningforrotaryvalvesisbestestablishedrightatthepointoftransferclosingifyouwantaverybroadrangeof power.Ontheotherhand,maximumpowerisobtainedbyopeningtheintakeportsomewhatearlier:from130to145degreesbeforetopcenter,whichmeansthatthe intakeandtransferportshaveoverlappingopenperiods.Thispresupposesthattheenginewillhavebeenfittedwithaproperexpansionchamber.Thediffusersectionin suchchambersiscapableofreturninganegativepressurewavetothecylinderhavingabelowatmosphericvalueofsomethinglikeminus7.0psi,andasthispartial vacuumiscommunicatedtothecrankcaseviathetransferports,thepressureinsidethecrankcaseitselfwilldropwellbelowatmospheric,Atlowerthanpeakengine speeds,particularly,thereisatendencyforthisbelowatmosphericcrankcasepressuretoequalizeitselfbydrawingpartofthechargebackdownthroughthetransfer ports,whichneatlycancelsanequalandimportantpartoftheworkdonebytheexhaustsystem.Thereislittlethatmaybedonetocounterthisinapistonportengine, beyondworkingwithanexhaustsystemproportionedtogivealongdurationscavengingpulsethatwillmaintainthenegativepressureinthecylinderuntilaftertransfer closing.Butwhentheengineinquestionhasarotaryvalve,itispossibletoopentheintakesidejustastransferpassagebackflowingisabouttooccur,andbalancethe pressureswithmixturedrawninthroughthecarburetorinsteadofrobbingfromthecylinder.Asyoumightguess,thisneattrickdoesn'tworkproperlyunlesstheintake openperiodisdelayeduntilafterpressureinthecrankcasehasbeenpulleddowntoatmosphericorbelowwhichmeansthatintakeopentimingisverycloselytiedtothe exhaustsystem'spulsingandtheoverallflowcharacteristicsofthetransferports.Afewhoursworkwithanoscilloscopeandpressuretransducerswouldgetyourighton target,becauseyouwouldthenknowwithahighdegreeofcertaintytheprecisepointatwhichcrankcasepressuredidinfactfalltoatmospheric,butfewpeoplehavethat veryexpensiveequipmentandmostwillhavetoachievethesameresultthroughalaboriousprocessofcutandtry.Thesamekindofcutandtryinfact,asisneededto locatetheoptimumpointforintakeclosing. Allofthemajorinfluencesontimearearequirementshavebeencovered.Thereareothersthatcouldbemajor,butforpracticalreasonsarenot.Crankcase compressionratioisone.Bothintakeandtransfertimearearequirementsdovaryininverseproportionwithcrankcasecompressionratios,simplybecauseascrankcase pressuresarereducedsoarethepressuredifferentialsthatcausegasestomove.Youaresparedhavingtoworryovermuchaboutthisfactorbecausevirtuallyallmodern engineshaveprimary(crankcase)compressionratiosverynear1.5:1thishavingproventobebestforalmosteveryapplication,andismechanicallyeasy(itiswhatyou getwithflywheelsandcrankcaseofnormalproportions).Exaggeratedanglesofentryintothecylindercouldalsocreateportswithwindowareasmisleadinglylargerthan thepassagesthemselves,butagainthiscondition whilealwayspresentinsomedegree isseldomseriousenoughtowarrantone'sworrywhileworkingtimearea calculations.Finally,extraordinarilypoorcastingtechniquesand/orbadlymisshapenportcavitiescouldlendportssuchlowcoefficientsofflowastoupsetone's calculations,butonealmostneverseesoutstandinglybadworkfromanyofthemajormanufacturers.Alloftheircastingworkisquitegood,andwillnotbemuch improvedbyeventhemostpainstakingcarvingandpolishingonyourpartwhichmaybediscouraging,butdoestendtopreservethevalidityofthetimeareavaluesI havepresentedhere. Backbeforetheunpleasantnessoftheearly1940s,anumberoftwostrokemotorcycleengineswerebuiltwithexternalscavengingairpumpsmuchinthe

improvedbyeventhemostpainstakingcarvingandpolishingonyourpartwhichmaybediscouraging,butdoestendtopreservethevalidityofthetimeareavaluesI havepresentedhere. Backbeforetheunpleasantnessoftheearly1940s,anumberoftwostrokemotorcycleengineswerebuiltwithexternalscavengingairpumpsmuchinthe fashionoftoday'sGMCtruckenginesdespitetheresultingpenaltyinbulkandmanufacturingcost.Itwasthenthoughtthatnoenginerelyingonordinarycrankcase pumpingforscavengingairdeliverycouldhopetocompetewiththebetterfourstrokesintermsofspecificpoweroutput,andtherewaseveryreasonforthatpessimism: Foronething,nooneinhisrightmindwoulddesignapistontypeairpumpwithasmuchclearancevolumeasinevitablyexistsinanengine'scrankcase.Neithercouldit beconsidereddesirabletopreheatthescavengingairbeforeitisdeliveredtothecylinder,whichispreciselywhathappensinthecrankcasescavengedtwostroke engine.Finally,usingthetwostrokeengine'scrankcaseasascavengingpumpcondemnstheconnectingrodbearingstoadietoftoomuchrawfuelandtoolittleoil.Only thefactthatresortingtoaseparate,externalscavengingpumproughlydoublesthecostofasingleortwincylinderengine,makingitmoreexpensivethanafourstroke engineofequivalentpoweroutput,haskeptusatworkonthecrankcasescavengedtwostroke.Andovertheyears,outofnecessity,wehavelearnedtomakethe crankcasefunctionasapumpwithaneffectivenessthatwouldastonishtheengineersoffiftyyearsago. Muchoftheimprovementintheairdeliverycapabilitiesofcrankcasepumpingcanbetracedbackahalfcenturytoatwostrokestationaryengine(usedtodrive anelectricalgenerator)thatneitheremployedanexternalscavengingpumpnorrelieduponitscrankcasetodothejob.Instead,theworkofmovingairthroughthe cylinderwasperformedbytheeffectsofsonicwavesandinertiaintheengine'sintakeandexhaustpipes.Theformerwasconnecteddirectlytothecylinder'sscavenging portsthelatterwasaconventional,iflengthy,straightpipe.Thisenginehadtobemotoreduptoitsoperatingspeedandablastofcompressedairdirectedintoitsintake pipetostartitfiring.Butthenacombinationofwaveandinertiaactivityinitsintakeandexhaustpipeswouldtakeover,toscavengeandrechargethecylinder,andthe enginewouldthudthudmerrilyalongatthatspeeduntilitranoutoffuelorbroke.Beingutterlydependentontheresonantfrequenciesoftheattachedplumbing,itwould ofcourserunatonlyonespeedbutthatisavirtue,ratherthanadisadvantage,inanengineusedtoturnagenerator.Andwhiletheenginedescribedwasneithervery powerfulforitssizenorparticularlyefficient,itwassimpleandtroublefreeandpointedthewayforthehyperpowerfulracingenginesofanother,latertime. Obviously,enginescapableofrunningatonlyonespeed,andwhichmustbecrankeduptothatspeedbeforetheywillstart,areoflimitedusefulnessinthecontext ofmotorcycling(althoughacoupleofroadracingenginesapproachingthatconditioncometomind).Evenso,fewofthecurrentcropofhighperformanceengineswould performatallwellwithouthelpfromresonantpipes.Rotaryvalveenginesfunctionverynicelywithoutmuchassistancefromintaketractresonance,butthoseusing pistoncontrolledintakeportswhichnecessarilyhavetheirintakeopenperiodspreadsymmetricallybeforeandafterthepistonreachestopcentertendtoregurgitatea lotofthemixturedrawnintothecrankcasebytheascendingpistonunlessthisbackflowisinsomemannerprevented.Combinedinertiaandwaveactivity,whichmay attainpressurevaluesinexcessof1.5atmospheres,areenoughtobalanceanypressurescreatedbythedescendingpistonevenifportclosingisdelayeduntilasmuchas 100degreesaftertopcenter,asisthecaseinhighlydevelopedroadracingengines.Ofcourse,theseeffectsareextremelytransitory,andwillpreventbackflowonlyif theirarrivalisproperlytimed. RESONANCEEFFECTS Timingwavearrivalsonthetwostrokeengine'sexhaustsideisarelativelysimpletask,astheexhaustsystem'sresonatefrequencyisalmostexclusivelyafunction ofitslength.Afarmorecomplexsituationexistsontheengine'sintakeside,forthereyouhavenotstraightforwardorganpiperesonance,butaresonatingflask consistingofthecrankcaseandintaketract.Asithappens,thereisafairlysimpleformulafordeterminingtheresonantfrequencyofflasks,whichis

WhereV s issonicspeed(usuallyabout1100ft/sec) Aisthecrosssectionalareaoftheinlet Listheinletpipelength Vcistheflask(crankcase)volume Fromtheformula,youwillseethatresonanceinaflask,orinanengine'sintakesystem,(whichincludesthecrankcase),frequencyisdirectlyproportionaltothe squarerootofintakepipecrosssectionalareainverselyproportionaltothesquarerootofcrankcasevolumeandalsoinverselyproportionaltothesquarerootofintake pipelength.ResearcherFujioNagao,ofKyotoUniversity,whohasdonemuchofthebasicinvestigationinthisfield,hasestablishedthatintaketractlengthformaximum airdeliveryshouldbethatwhichprovides75pressurefluctuationscoincidentwiththeperiodofinletportopening.Thatissosay,ifanengine'sintakeperiodindegrees is120degreesanditstorquepeakisat6000rpm,thenitsintakeperiodintimewillbe0.003secondandtheintakesystem'sfrequencycoincidentwith75percentof that,or125cyclespersecond. Unfortunately,thisstraightforwardpicturebeginstogetverycomplicatedasonetriestoapplyanyoftheprecedingtotheconcreteexample.Wecan,forinstance, onlyguessatsonicspeed,asitisdependentontemperatureandonecanonlyguessatthetemperatureofastreamofgasessimultaneouslybeingcooledbyevaporating gasolineandheatedbysurroundingmetal.Butthatisnotthegreatestdifficultyonefacesincalculatingtheproperlength:Theflaskinquestionherehasavolumethat variescontinuouslywiththepiston'smotions,andtheintaketractisinmostenginesfarfrombeingasimpletube(notonlyareabutshapewillvaryfromthecarburetor intakebelltotheportwindow).Moreover,theinertiaofthefastmovingcolumnofgasesintheintaketractmustalsohaveaneffect.Inshort,calculatingintaketract lengthisahighlycomplexproblem,andwhileatrainedengineerwithaheavybackgroundinhighermathematicscouldprobablydothejobgivensufficienttime,itreallyis ataskforthecomputerandeventhenitwillbenecessarytouseabbreviatedformulaetoavoidaccumulatingtoomanyexpensivecomputerhoursontheproblem. Inallprobability,thosewhowillreadthisbookwillhaveneitheracomputernorthesparehoursforatheoreticaldeterminationofintaketractlength.Thus,it becomesnecessarytoarriveatthislengthexperimentally,whichfortunatelyisamuchshorter(andmoreaccurate)meansofaccomplishingthesamething.Thefirststepin thatdirectionistoisolatetheinfluenceofintaketractlengthontheengine,whichmeansremovinganyeffecttheexhaustsystemmayhaveontheresultsobtained.Todo this,youwillhavetofityourenginewithastubpipetooshorttobeeffectivewithintheengine'sprojectedoperatingrange,yetlongenoughtopreventthebackbreathing ofairinthroughtheexhaustporttoupsetmixturestrengthandthusmaskintakelengthinfluence.Forsmalldisplacementcylinders,atotalexhaustlengthof3 1/2inches (measuredfromportwindowtotheendoftheexhauststub)shouldbesatisfactory.Middledisplacementenginesshouldhavea4inchstubexhaust,andthosewitha 350400cccylinderdisplacementalengthof4 1/2inches.Quiteobviously,thesheernoiselevelproducedbystubexhaustswillmakesomeformofmufflinganecessityin manyareas,andwheresuchisthecaseyoucanaddmufflingwithoutupsettingtheexperimentbyintroducingalargevolumechamberintowhichtheexhauststubempties. Thischambershouldbeaslargeascanbefittedonthemotorcyclewithoutmakingitimpossibletoride,andtheattachedmufflershouldhavesufficientinternaldiameterto preventanybackpressurefromdevelopinginthesystem.Thehardwarethatmakesupthisnoeffectexhaustsystemneednotbebeautiful,nortuckedinonthe motorcycleneatlyenoughtopermitanythingmorethanstraightlineriding,butitshouldbeeasytoinstallbecauseyouwillhavetoretunetheintakelengthwithevery changeinporttimingand/orcarburetordiameter.Therewill,ofcourse,beachangeinthesystem'sfrequencywitheverychangeincrankcasevolumeaswell,butsuch changeswillbetoosmalltoworryaboutinmostinstances. Forintakelengthexperiments,youwillneednotonlythestubexhaustsystem,butatachometeronyourmotorcycleandeitheralong,straightroad(ortrail)ora hill.Theselastarerequiredbecauseyoumustloadtheengineheavilyenoughtoallowyouagood,seatofthepantsreadingoftheenginespeedwherethepipehasits effectonpower.Andthateffectwillbeunmistakable,fortheenginewillpullverystronglywhenitcomesonthepipe.Youwillalsofindthatintakepipelengthcan movethestubexhaustequippedengine'spowerpeakoveraverywidespeedrange.MyfriendsatMcCulloch,whoacquaintedmewiththestubexhausttesting technique,tellofteststheyhaveperformedwithanenginehavinganintakeperiodof120degrees,andwithpipesrangingfrom5to9 1/2inchesinlengththeywereable tomovethepowerpeakanywherefrom3000rpmto8000rpm.The9 1/2inchlengthgavesubstantiallythesamepowerat3000rpmand4000rpmat5000rpm,an8

havepresentedhere. Backbeforetheunpleasantnessoftheearly1940s,anumberoftwostrokemotorcycleengineswerebuiltwithexternalscavengingairpumpsmuchinthe fashionoftoday'sGMCtruckenginesdespitetheresultingpenaltyinbulkandmanufacturingcost.Itwasthenthoughtthatnoenginerelyingonordinarycrankcase pumpingforscavengingairdeliverycouldhopetocompetewiththebetterfourstrokesintermsofspecificpoweroutput,andtherewaseveryreasonforthatpessimism: Foronething,nooneinhisrightmindwoulddesignapistontypeairpumpwithasmuchclearancevolumeasinevitablyexistsinanengine'scrankcase.Neithercouldit beconsidereddesirabletopreheatthescavengingairbeforeitisdeliveredtothecylinder,whichispreciselywhathappensinthecrankcasescavengedtwostroke engine.Finally,usingthetwostrokeengine'scrankcaseasascavengingpumpcondemnstheconnectingrodbearingstoadietoftoomuchrawfuelandtoolittleoil.Only thefactthatresortingtoaseparate,externalscavengingpumproughlydoublesthecostofasingleortwincylinderengine,makingitmoreexpensivethanafourstroke engineofequivalentpoweroutput,haskeptusatworkonthecrankcasescavengedtwostroke.Andovertheyears,outofnecessity,wehavelearnedtomakethe crankcasefunctionasapumpwithaneffectivenessthatwouldastonishtheengineersoffiftyyearsago. Muchoftheimprovementintheairdeliverycapabilitiesofcrankcasepumpingcanbetracedbackahalfcenturytoatwostrokestationaryengine(usedtodrive anelectricalgenerator)thatneitheremployedanexternalscavengingpumpnorrelieduponitscrankcasetodothejob.Instead,theworkofmovingairthroughthe cylinderwasperformedbytheeffectsofsonicwavesandinertiaintheengine'sintakeandexhaustpipes.Theformerwasconnecteddirectlytothecylinder'sscavenging portsthelatterwasaconventional,iflengthy,straightpipe.Thisenginehadtobemotoreduptoitsoperatingspeedandablastofcompressedairdirectedintoitsintake pipetostartitfiring.Butthenacombinationofwaveandinertiaactivityinitsintakeandexhaustpipeswouldtakeover,toscavengeandrechargethecylinder,andthe enginewouldthudthudmerrilyalongatthatspeeduntilitranoutoffuelorbroke.Beingutterlydependentontheresonantfrequenciesoftheattachedplumbing,itwould ofcourserunatonlyonespeedbutthatisavirtue,ratherthanadisadvantage,inanengineusedtoturnagenerator.Andwhiletheenginedescribedwasneithervery powerfulforitssizenorparticularlyefficient,itwassimpleandtroublefreeandpointedthewayforthehyperpowerfulracingenginesofanother,latertime. Obviously,enginescapableofrunningatonlyonespeed,andwhichmustbecrankeduptothatspeedbeforetheywillstart,areoflimitedusefulnessinthecontext ofmotorcycling(althoughacoupleofroadracingenginesapproachingthatconditioncometomind).Evenso,fewofthecurrentcropofhighperformanceengineswould performatallwellwithouthelpfromresonantpipes.Rotaryvalveenginesfunctionverynicelywithoutmuchassistancefromintaketractresonance,butthoseusing pistoncontrolledintakeportswhichnecessarilyhavetheirintakeopenperiodspreadsymmetricallybeforeandafterthepistonreachestopcentertendtoregurgitatea lotofthemixturedrawnintothecrankcasebytheascendingpistonunlessthisbackflowisinsomemannerprevented.Combinedinertiaandwaveactivity,whichmay attainpressurevaluesinexcessof1.5atmospheres,areenoughtobalanceanypressurescreatedbythedescendingpistonevenifportclosingisdelayeduntilasmuchas 100degreesaftertopcenter,asisthecaseinhighlydevelopedroadracingengines.Ofcourse,theseeffectsareextremelytransitory,andwillpreventbackflowonlyif theirarrivalisproperlytimed. RESONANCEEFFECTS Timingwavearrivalsonthetwostrokeengine'sexhaustsideisarelativelysimpletask,astheexhaustsystem'sresonatefrequencyisalmostexclusivelyafunction ofitslength.Afarmorecomplexsituationexistsontheengine'sintakeside,forthereyouhavenotstraightforwardorganpiperesonance,butaresonatingflask consistingofthecrankcaseandintaketract.Asithappens,thereisafairlysimpleformulafordeterminingtheresonantfrequencyofflasks,whichis

WhereV s issonicspeed(usuallyabout1100ft/sec) Aisthecrosssectionalareaoftheinlet Listheinletpipelength Vcistheflask(crankcase)volume Fromtheformula,youwillseethatresonanceinaflask,orinanengine'sintakesystem,(whichincludesthecrankcase),frequencyisdirectlyproportionaltothe squarerootofintakepipecrosssectionalareainverselyproportionaltothesquarerootofcrankcasevolumeandalsoinverselyproportionaltothesquarerootofintake pipelength.ResearcherFujioNagao,ofKyotoUniversity,whohasdonemuchofthebasicinvestigationinthisfield,hasestablishedthatintaketractlengthformaximum airdeliveryshouldbethatwhichprovides75pressurefluctuationscoincidentwiththeperiodofinletportopening.Thatissosay,ifanengine'sintakeperiodindegrees is120degreesanditstorquepeakisat6000rpm,thenitsintakeperiodintimewillbe0.003secondandtheintakesystem'sfrequencycoincidentwith75percentof that,or125cyclespersecond. Unfortunately,thisstraightforwardpicturebeginstogetverycomplicatedasonetriestoapplyanyoftheprecedingtotheconcreteexample.Wecan,forinstance, onlyguessatsonicspeed,asitisdependentontemperatureandonecanonlyguessatthetemperatureofastreamofgasessimultaneouslybeingcooledbyevaporating gasolineandheatedbysurroundingmetal.Butthatisnotthegreatestdifficultyonefacesincalculatingtheproperlength:Theflaskinquestionherehasavolumethat variescontinuouslywiththepiston'smotions,andtheintaketractisinmostenginesfarfrombeingasimpletube(notonlyareabutshapewillvaryfromthecarburetor intakebelltotheportwindow).Moreover,theinertiaofthefastmovingcolumnofgasesintheintaketractmustalsohaveaneffect.Inshort,calculatingintaketract lengthisahighlycomplexproblem,andwhileatrainedengineerwithaheavybackgroundinhighermathematicscouldprobablydothejobgivensufficienttime,itreallyis ataskforthecomputerandeventhenitwillbenecessarytouseabbreviatedformulaetoavoidaccumulatingtoomanyexpensivecomputerhoursontheproblem. Inallprobability,thosewhowillreadthisbookwillhaveneitheracomputernorthesparehoursforatheoreticaldeterminationofintaketractlength.Thus,it becomesnecessarytoarriveatthislengthexperimentally,whichfortunatelyisamuchshorter(andmoreaccurate)meansofaccomplishingthesamething.Thefirststepin thatdirectionistoisolatetheinfluenceofintaketractlengthontheengine,whichmeansremovinganyeffecttheexhaustsystemmayhaveontheresultsobtained.Todo this,youwillhavetofityourenginewithastubpipetooshorttobeeffectivewithintheengine'sprojectedoperatingrange,yetlongenoughtopreventthebackbreathing ofairinthroughtheexhaustporttoupsetmixturestrengthandthusmaskintakelengthinfluence.Forsmalldisplacementcylinders,atotalexhaustlengthof3 1/2inches (measuredfromportwindowtotheendoftheexhauststub)shouldbesatisfactory.Middledisplacementenginesshouldhavea4inchstubexhaust,andthosewitha 350400cccylinderdisplacementalengthof4 1/2inches.Quiteobviously,thesheernoiselevelproducedbystubexhaustswillmakesomeformofmufflinganecessityin manyareas,andwheresuchisthecaseyoucanaddmufflingwithoutupsettingtheexperimentbyintroducingalargevolumechamberintowhichtheexhauststubempties. Thischambershouldbeaslargeascanbefittedonthemotorcyclewithoutmakingitimpossibletoride,andtheattachedmufflershouldhavesufficientinternaldiameterto preventanybackpressurefromdevelopinginthesystem.Thehardwarethatmakesupthisnoeffectexhaustsystemneednotbebeautiful,nortuckedinonthe motorcycleneatlyenoughtopermitanythingmorethanstraightlineriding,butitshouldbeeasytoinstallbecauseyouwillhavetoretunetheintakelengthwithevery changeinporttimingand/orcarburetordiameter.Therewill,ofcourse,beachangeinthesystem'sfrequencywitheverychangeincrankcasevolumeaswell,butsuch changeswillbetoosmalltoworryaboutinmostinstances. Forintakelengthexperiments,youwillneednotonlythestubexhaustsystem,butatachometeronyourmotorcycleandeitheralong,straightroad(ortrail)ora hill.Theselastarerequiredbecauseyoumustloadtheengineheavilyenoughtoallowyouagood,seatofthepantsreadingoftheenginespeedwherethepipehasits effectonpower.Andthateffectwillbeunmistakable,fortheenginewillpullverystronglywhenitcomesonthepipe.Youwillalsofindthatintakepipelengthcan movethestubexhaustequippedengine'spowerpeakoveraverywidespeedrange.MyfriendsatMcCulloch,whoacquaintedmewiththestubexhausttesting technique,tellofteststheyhaveperformedwithanenginehavinganintakeperiodof120degrees,andwithpipesrangingfrom5to9 1/2inchesinlengththeywereable tomovethepowerpeakanywherefrom3000rpmto8000rpm.The9 1/2inchlengthgavesubstantiallythesamepowerat3000rpmand4000rpmat5000rpm,an8 1/ inchwasbestat6000rpmand7000rpmthesamepowercouldbehadwitheithera9 1/ or6 1/ inchlengthandthesamewastrueat8000rpm,butwithlengths 2 2 2 of5 1/ and8inches.Fromthesetests,itwasdeterminedthatthebestcompromiselengthfortheentirerange30008000rpmwas8inchesa5 1/ inchlengthwas

hill.Theselastarerequiredbecauseyoumustloadtheengineheavilyenoughtoallowyouagood,seatofthepantsreadingoftheenginespeedwherethepipehasits effectonpower.Andthateffectwillbeunmistakable,fortheenginewillpullverystronglywhenitcomesonthepipe.Youwillalsofindthatintakepipelengthcan movethestubexhaustequippedengine'spowerpeakoveraverywidespeedrange.MyfriendsatMcCulloch,whoacquaintedmewiththestubexhausttesting technique,tellofteststheyhaveperformedwithanenginehavinganintakeperiodof120degrees,andwithpipesrangingfrom5to9 1/2inchesinlengththeywereable tomovethepowerpeakanywherefrom3000rpmto8000rpm.The9 1/2inchlengthgavesubstantiallythesamepowerat3000rpmand4000rpmat5000rpm,an8 1/ inchwasbestat6000rpmand7000rpmthesamepowercouldbehadwitheithera9 1/ or6 1/ inchlengthandthesamewastrueat8000rpm,butwithlengths 2 2 2 of5 1/2and8inches.Fromthesetests,itwasdeterminedthatthebestcompromiselengthfortheentirerange30008000rpmwas8inchesa5 1/2inchlengthwas slightlybetterfortherange60008000rpm.Itshouldbeunderstoodthattheselengthsonlyapplytothisspecificexample,andwouldvarywithanychangesinintake timing,crankcasevolumeandintaketractdiameter.Theyaregivenhereonlytoillustratethat:A)itispossibletomakeagivenintaketimingworkoverawidespeed rangebyvaryingintaketractlengthandB)thatthesamelengthmayproducemorethanonepeak,orconverselythatmorethanonelengthmaybeeffectiveatanygiven powerpeak. CARBURETORLOCATION Whenperformingyourownexperimentswithintaketractlength,youwillbetemptedtomountthecarburetorasclosetothecylinderaspossible,andmake adjustmentsbyvaryingthelengthofanintakestackaddedonthecarburetor'smouth.Don'tdoit!Thatcertainlyisthemostconvenientmethod,butacarburetorplaced toofarbacktowardtheintakeportwindowisgoingtobesubjectedtoradicalpressurefluctuations,duetowaveactivityinthesystem,andthosepressurefluctuationsdo terriblethingstothecarburetor'sabilitytometerfuel.Worse,thelargestdeparturefromthedesiredmixturestrengthwilloccurpreciselyattheenginespeedswhereintake rammingisstrongest,whichmakesthetaskofselectingamainjetimpossible.Sothecarburetormustbelocatedoutattheendoftheintaketract,ratherthanclosetothe cylinderwhichisunfortunate,asmaximumairdeliveryintothecrankcaseisobtainedwhenthereverseistrue.Perhapssomeformoffuelinjectionistheanswer. Mixturestrengthproblemscanalsooccurduetotheplumbingbetweenthecarburetorandaircleaner,anditshouldgowithoutsayingthatsuchplumbingmayalso addtotheintaketract'seffectivetunedlength.Thecommonpracticeofconnectingcarburetormouthandaircleanerwithasectionofrubberhosemayhavemuchto recommenditfromthestandpointofconvenienceitalsoislikelytoestablisheitheranextensionthatbecomespartoftheeffectivetractlength,orasecondaryresonating systemthatheterodynesatsomefrequencieswiththemaintractandthusupsetsitsproperfunctioning.Therefore,itisgoodpractice,ifsometimesinconvenient,tomake connectingplumbingbothasshortandaslargeindiameterasispossible.Aircleanershavingplasticfoamelementsmaybemountedveryclosetothecarburetorwithout penalty,butthepapertypefiltermay,ifitistooclosetothecarburetormouth,becomesaturatedwithoilandfuelinwhichcaseitwillrefusetopassanyairatall. CRANKCASEVOLUME Atonetime,itwaseverytwostroketuner'shabittobeginanyseriousattemptatextractingmorethanstandardhorsepowerfromagivenenginebyincreasingthe engine'sprimarycompressionratiothatistosay,theratiobetweencrankcasevolumewiththepistonattopcenter,andatbottomcenter,asinthefollowingexpression: Or, Where,CR p isprimarycompressionratio V1iscrankcasevolumeatBDC V2ispistondisplacement Isuspectthatthepopularityofthispractice,stuffingcrankcases,wasderivedfromthefactthattheoldTcrankVillierswasthenthetwostrokeenginemost frequentlybeingmodified.Thisenginewasbuiltmorewithaneyetowardmanufacturingcostandlongtermreliabilitythanpoweroutput(ofwhichithadonlyextremely modestamounts)andtheconfigurationofitscrankshaftandcrankcaseprovidedonlyaverylowprimarycompressionratio.Toolow,infact,foranythingevenremotely approachinghighspeedoperation,sothatanyeffortsatincreasingitsprimarycompressionratiowereimmediatelyreflectedinapowerincrease.Butalotofpeople simplyassumedthatifincreasingtheVilliersengine'sprimarycompressionratiofrom,say,1.2:to1.4:1wasgood,thenraisingitevenfurtherwouldbebetterandthatall twostrokeengineswouldbenefitfrombeinggiventhesametreatment.There,theyfellintoerror. ReferringagaintotheworkofFujioNagao(andhisresultshavebeenverifiedbyotherresearchers)wefindsomeveryinterestingconclusionswithregardto primarycompressionratios:First,Nagaotellsusthatforgivenportareas,thereisanenginespeedatwhichmaximumairdeliverytothecylinderoccurs,andthatthis enginespeedisinverselyproportionaltocrankcasevolume,butthatthemaximumvaluechangesonlyslightlywithchangesincrankcasevolume.Toputitanotherway,the crankcasepump'svolumetricefficiencyisnearlyconstant,buttheenginespeedatwhichitattainsmaximumefficiencyrisesascrankcaseclearancevolumeisreduced. Significantly,too,Nagaogoesontosaythatanydeficiencyinairdeliveryduetoacrankcasevolumetoogreatforagivenenginespeedisfairlywellcompensatedby properlytunedintakeandexhaustpipes.Hesaysinconclusionthatlittleadvantageisobtainedbymakingthecrankcasevolumeexcessivelysmall. LaterworkinthesamefieldindicatesthatNagaounderstatedthecase,andthattheremaywellbepositivedisadvantagesinexcessivereductionsofcrankcase volume.HiroshiNaito,whoIbelievewasresponsibleforthefantasticYamahaGPracingengines,hasindicatedthatthereislittleimprovementtobehadevenfromultra highspeedengineswithprimary(crankcase)compressionratiosabove1.5:l.Thisapparentanomalycanbeexplainedbystudyingtheeffectsofthewholescavenging system,fromtheintaketractthroughthecrankcaseandscavengingportsandrightonoutintotheexhaustsystem.Andwemustthinkindynamic,ratherthanstaticterms. Startingwiththeexhaustsystem,wefindthatitispossibletoevacuatethecylindertowellbelowatmosphericpressurebyusingtheextractoreffectsoftheexpansion chamber'sdiffuser.Ifthelengthsandproportionsoftheexhaustsystemareproperlyestablished,thefreshchargemovingupthroughthetransferportswillnotonlybe pushedthroughbythepressurebelowbutwillgetadditionalaidfromwhatis,ineffect,anexhaustproducedvacuuminthecylinder.Further,thisvacuummaywellbe communicateddownintothecrankcase,viathetransferpassages,andcrankcasepressuretrappedatabelowatmosphericvaluewhenthetransferportsclose.This factorcanbeveryimportant,asitproducesanairdeliveryvolumegreaterthanwouldbepossiblewithcrankcasepressurealonetoimpelthefreshchargeintothe cylinder.Butthatisn'tthewholestory,obviously,forthepullontheexhaustsideofthecylinderismatchedbyapushwheremixturefromthecarburetorentersthe crankcasewithaforceequaltoabout1.5atmospheresinadvancedengines.Thus,wehaveevacuatedthecrankcasetosomethingbelowatmosphericpressurebefore refillingitwithafreshcharge,andthecrankcasewillhavebeenchargedtosomethingaboveatmosphericpressurebythepulsationsintheintaketract.Allthishas implicationsintermsofcrankcasevolume,forifweassumethatthepositiveandnegativepressuresappliedremainconstant,thenbulkflowthroughthecylinderwill increasewithincreasesincrankcasevolume.Doesthatseemobscure?Thenconsiderthatthereismoreairina1000ccflaskcompressedto1.5atmospheresthaninone of500cccapacity.Inlightofthis,youmightassumethatitistimetostartreducingcrankcasecompressionratiosespeciallyaspumpinglosses(horsepowerabsorbedin doingthiswork)riseastothethirdpowerofcompressionratio.Actually,thisisn'tquitetrueeither,forreductionsincrankcasecompressionratiocauseanequal reductionintheamplitudeofwaveactivityintheintaketract,whichinturnreducestherammingpressuresavailabletochargethecrankcase.So,crankcasecompression ratiosmustbeestablishedtobalancetheconflictingrequirementsofvolumeandwavestrength,andifitispure,peakhorsepowerthatinterestsyou,thenyouwillfindthat aprimarycompressionratioof1.5:lrepresentssomethingveryneartheoptimum.Ofcourse,thiswillonlybetrueifcarburetorsizealsoisselectedwithaneyetoward maximumhorsepower. Broadrangeperformanceisanothermatterentirely.Formotocross,youwillrequiremuchmoreflexibilityfromanenginethanwouldbetrueofroadracingoreven flattrack.Inthelargedisplacementclassesparticularly,wheresheerenginesizemakesitaltogethertooeasytogetmorehorsepowerthanamotorcycle'sreartirecan applytotheground,anengine'sabilitytopullstronglyoveraverywidespeedrangeismuchmoreimportantthananypeakreading.Forsuchapplications,thebestengine islikelytobeonewithaprimarycompressionratiowellbelow1.5:1,along,slowtaperexpansionchamberandarelativelylongintaketract.Withregardtothislastitem,

ratiosmustbeestablishedtobalancetheconflictingrequirementsofvolumeandwavestrength,andifitispure,peakhorsepowerthatinterestsyou,thenyouwillfindthat aprimarycompressionratioof1.5:lrepresentssomethingveryneartheoptimum.Ofcourse,thiswillonlybetrueifcarburetorsizealsoisselectedwithaneyetoward maximumhorsepower. Broadrangeperformanceisanothermatterentirely.Formotocross,youwillrequiremuchmoreflexibilityfromanenginethanwouldbetrueofroadracingoreven flattrack.Inthelargedisplacementclassesparticularly,wheresheerenginesizemakesitaltogethertooeasytogetmorehorsepowerthanamotorcycle'sreartirecan applytotheground,anengine'sabilitytopullstronglyoveraverywidespeedrangeismuchmoreimportantthananypeakreading.Forsuchapplications,thebestengine islikelytobeonewithaprimarycompressionratiowellbelow1.5:1,along,slowtaperexpansionchamberandarelativelylongintaketract.Withregardtothislastitem, Imightaddthatitispossiblewithintakelengthsofabout30inchestoboostanengine'scrankcasedeliveryratio(volumeofairpumped,percycle/pistondisplacement)to asmuchas1.2:l.However,duetorestrictionsinavailabletime,theseultralongpipeswillnotworkexceptatverylowenginespeedsand,moreover,tendtoworkonly overaverynarrow:peedrange.Indeed,alltunedintakepipeseffectivelyreduceanengine'soperatingrange,thoughthisiscompensatedbytheirprovidingapowerboost atsomeenginespeeds,anditmightbethatanearzerolengthwouldprovidethebestspreadofpower.Thedifficultyhereisthatsomesortofsmoothpassagemustbe providedforthetransitionfromthesubstantiallysquareintakeportwindowtotheroundcarburetorthroat,andifyouaddthelengthofthatpassagetothecarburetor's length,thenyouhavearesonatingintakesystemevenifyoudon'twantone.Thechoiceisthusreducedtoselectingalengththatprovidesthebestresultsoverall. Thereisonemeansofhaltingresonanteffectsintheintaketract,andtoaccomplishthatyouhaveonlytoreducethecarburetorthroatindiameteruntilitscross sectionalarearepresents35percent,orless,oftheintakeportarea.Curiously,acarburetorofthatsmallsizeplacedattheportwindow,behindtheintakepipe,willnot stallwaveactivityinthesystembutonelocatedoutattheendoftheintakepipeactsineffectasawavedamper.Imentionthisonlyforitsvalueasacuriosity,asthereis nopointinfittingsuchasmallcarburetor except,perhaps,onatrialsmotorcycleandiffittedattheportwindowitwouldbesubjecttothesamemixturestrength instabilityasanyothercarburetorprecededbyalengthofresonatingpipe. REEDVALVES Another,indirectmeansofgainingthebenefitsofaresonating,ramintaketract(aswellascertainothers)istointerposeacheckvalvebetweencarburetorand cylinderwhichwillinsureaonewayflow,andtrapinthecrankcaseanythingthatpassesthroughthevalve.Nowasithappens,theonlycheckvalvecurrentlycapable offunctioningfastenoughtokeeppacewithahighrevvingtwostrokeengineistheonegenerallycalledareedvalve.Thenamefairlyaccuratelydescribesthevalve, whichhasathin,broadmetalorphenolicreedseatingagainstanapertureandclampedatoneendsothatitmaybendawayfromsaidaperture.Thereusuallyaremultiple reedsandaperturesinanyreedvalveassemblyalsoinmostinstancesthesewillbesetatanangleobliquetothedirectionofgasflow,soastominimizeflowlossespast thereeds.Usually,too,therewillbefourseparatereeds,clampedagainstthesidesofeitherapyramidorawedge,withthecarburetorfeedingtheinteriorandthepoint beingaimedintothecrankcase.Atonetimemanyattemptsweremadetousecomparativelystiffreeds,toovercomeatendencyforthereedstofractureanddropoffat highenginespeeds,butnoweveryoneutilizesthinreedsbackedbystops.Thestops,whicharesimplythick,curvedstripsofmetalclampedoverthereeds,controlboth

thereeds'travelandtheshapeofreed bending.Youwillappreciatethata reedheldonlyatitsendwillbendquiteabruptlyrightnexttotheclampthatholdsitinplace,andthatitmaywellbendtoofarforitsownwellbeingundersome conditions.Reedstopspreventbothofthosethingsfromhappening,forcingthereedtocurveveryevenlyaroundthestopandallowingittobendonlyasfarasthe fatigueresistingpropertiesofthereedmaterialenableittowithstand. Onlyrarely,today,doreedsfractureattheirrootanddropintoanengine'scrankcasebutitstillhappens,andforthatreasonmanymanufacturerswhobuild reedvalveenginesprefertoemployplastic(usuallyafiberreinforcedphenolicsandwich)reeds.Steelreedscandoterribledamagetoanengine'sinteriorwhenthey breakoffandareaspiratedupthroughitstransferportsthephenolicreedsimplygetsgnawedintofragmentsandexpelledouttheexhaustport.Thisdanger,withsteel reeds,apparentlyisveryreal,fortherecanbelittledoubtbutthatengineperformanceisimprovedbytheuseofsteeloverphenolicinthereedmaterialsimplybecause steelisthemoreflexiblematerial. Actually,thefailureofacompletesteelreedisanextremelyrareevent.Muchmorefrequentlythefailurewillbeconfinedtoapiecebreakingawayatthereed's tipwhichiscausedbytherepeatedimpactsofthereedagainstthereedblock.Formanyyearsthisproblemseemedtodefysolution,asverythinreedswerenotstrong enoughtoresistthesefrettingfracture,andthickerreedswereshatteredbytheirowngreaterinertia.Yamahafoundthesolution,whichisbeautifulinitssimplicity.The 1972Yamahareedvalve(theycallit"TorqueInduction")engineshavereedblockscoatedwithneoprene,andwhilethecoatingisverythin,ithasenoughresilienceto cushiontheimpactofthereedsandpreventfrettingfailures.IunderstandthattheYamahareedsareofstainlesssteel,withthicknessesof0.008and0.006inchforthe 250cc/360ccand100cc/125cc/175ccengines,respectively.Theuseofastainlesssteelasareedmaterialwillbeexplainedifyouconsiderthedevastatingeffectofeven averylittlerustonsuchthinstripsofmetal. Iammuchinclinedtodoubtthatreedvalveswillbecomepopularinroadracingengines,astheydointroduceaflowrestrictionintotheintakesystemthatmust haveadepressingeffectonpeakhorsepoweratultrahighenginespeeds.ButthereiseveryreasontoassumethatTorqueInductionwillcomeintoverywideusefor everykindofoffroadmotorcycle:Thereis,forexample,thereedvalve'sabilitytomakethemostoutofpositiveintakeresonanceeffectswhilestallingoutofphase resonance.Also,thereedvalvehas,byitsverynature,theabilitytoeffectivelyadjustintaketimingtosuitallenginespeeds.Finally,itdoesseemthatreedvalvingeases

Iammuchinclinedtodoubtthatreedvalveswillbecomepopularinroadracingengines,astheydointroduceaflowrestrictionintotheintakesystemthatmust haveadepressingeffectonpeakhorsepoweratultrahighenginespeeds.ButthereiseveryreasontoassumethatTorqueInductionwillcomeintoverywideusefor everykindofoffroadmotorcycle:Thereis,forexample,thereedvalve'sabilitytomakethemostoutofpositiveintakeresonanceeffectswhilestallingoutofphase resonance.Also,thereedvalvehas,byitsverynature,theabilitytoeffectivelyadjustintaketimingtosuitallenginespeeds.Finally,itdoesseemthatreedvalvingeases

problemswithbroadrange carburetion,endingtheoftnotedtendencyfor carburetorstoproducewildlyvariedmixturestrengthsatdifferentenginespeeds. Anotherbenefitthataccrueswithreedvalvingisthatwithallpossibilityofbackflowingouttheintakeremoved,itbecomespossibletousetherearcylinderwall forsomethingotherthanasupportforthepiston.Yamaha,longanadvocateofmultipletransferports,hasaddedanotherport,oppositetheexhaustport,initsreedvalve engines.Thisportisashighastheother(four)transferports,butisverynarrow.Significantly,thisextratransferportisnotfedmixturefromthecrankcaseitrelies entirelyupontheextractoreffectoftheexhaustsystemtopullmixturefromthecarburetorthroughthereedsandreedcavity,andupthroughtheportintothecylinder.My friendsatMcCulloch whohaveanenormousaccumulatedexperiencewithreedvalveengines,tellmethatpowercharacteristicscanbealteredveryradicallyby wideningandnarrowingthisboostport,and/orbyarrangingformixturecompressedinthecrankcasetoflowupthroughtheboostportinsteadofrelyingpurelyonthe depressioncreatedinthecylinderbytheexhaustsystemtodothejob.Buttheyalsoseemtothinkthattheboostport'swidthissomethingonedeterminesalmostentirely throughexperimentation,soitwouldappeartobewisetostartwithaverynarrowportwindowandwidenitgraduallyuntilthedesiredpowercharacteristicsareobtained. THEROTARYVALVE Betterthanreedvalvesinfunction,ifnotinmechanicalconvenience,istherotaryvalve.Specifically,thedisctyperotaryvalve,whichistheonlyvarietycurrently inuseinmotorcycleengines.Thereislittlepointindwellingonthiskindofintakevalvingovermuch,becauseitreallyisonlypracticalonaGProadracingengine. Touringtypeengines,iftheyhavemorethanasinglecylinder(andthetrendobviouslyistowardmultis)becomeinconvenientlywidewhenadiscvalveassemblyisadded ateachendofthecrankshaft,andthistypeofintakevalvingisallbutimpossibletoapplytoaninlineenginewithmorethantwocylinders.Therotaryvalve'sextrawidth is,ofcourse,anembarrassmentevenonsinglecylinderengineinoffroadapplications. Ifyouignorethemechanicaldisadvantagesoftherotaryvalve,itbecomeshighlyattractivefromthestandpointofnotonlypeakpower,butbroadrangepower. Thereasonforthisisthatitdoesnotleavetheenginetorelyuponpressurewavesmarchingbackandforthinapipetopreventchargelossbackthroughthecarburetor. Itispossibletoextractjustasmuchhorsepower,atthepeak,fromagivenenginewitheitherrotary

valveorpistonportinduction,butthelatterwilllosepowerveryrapidly eithersideofthebhppeakwhiletherotaryvalvecontinuestofunctionandcrankcasepumpingremainseffective. INTAKEPORTSHAPE Intakeporttimingiscoveredelsewhereinthisbook,butthiscertainlyistheplaceforadiscussionofintakeportshape whichhasadefiniteinfluenceon crankcasepumpingefficiency.Ingeneral,thebestflowcoefficientforanygiventimingareavaluewillbeobtainedwiththewidestpossibleport.Thatistosay,aportthat wrapsaroundthecylinderasfarasispermittedbymechanicalconsiderations(studplacement,transferportlocation,etc.).Ofcourse,withaverywideportthereisa

eithersideofthebhppeakwhiletherotaryvalvecontinuestofunctionandcrankcasepumpingremainseffective. INTAKEPORTSHAPE Intakeporttimingiscoveredelsewhereinthisbook,butthiscertainlyistheplaceforadiscussionofintakeportshape whichhasadefiniteinfluenceon crankcasepumpingefficiency.Ingeneral,thebestflowcoefficientforanygiventimingareavaluewillbeobtainedwiththewidestpossibleport.Thatistosay,aportthat wrapsaroundthecylinderasfarasispermittedbymechanicalconsiderations(studplacement,transferportlocation,etc.).Ofcourse,withaverywideportthereisa tendencyfortherearedgeofthepistonskirttosnagatthebottomoftheportwindow,whichmeansthatitmaybenecessarytouseawindowshapemorenearlyround thansquaretopreventrapidwearatthebottomofthepistonskirt.Whichisoftengoodpracticeinanycase.Aroundedportwindow,oronewithaVshapetoitslower edge,provideswhateffectivelyisaslowerrateofportopening,whichisveryusefulinreducingintakeroar apointthatmustbe

consideredinatimewhenstatutorynoiselimitsarepoppingupalloverthe place.Also,themoregradualopeningoftheporttendstoextendthedurationofthesonicwavethatisused,onitsreturntrip,tosuperchargethecrankcase,andthat hastheeffectofbroadeninganengine'spowerband.Finally,aportwithroundedcornershasamuchbetterflowcoefficientthanonethatissquare.Thesamemaynot besaidforroundingbacktheloweredgeofthepistonskirt,asthatextendstheintaketiming topermitbackflowasthepistondescendswithoutproducingany measurableimprovementinflowcoefficient.Itis,however,possibletoimproveflowwithadownturnedlipatthetopedgeoftheintakeportwindow.Buttheprimary thingonemustrememberwhencarvingawayatanintakeportisthatripplesintheportwalls,oranysuddenchangeincrosssection,haveafarmoredamagingeffecton flowcoefficientthanaslightlyroughfinishintheport.Therefore,itisvastlymoreimportanttosmooththeportthantogiveitamirrorfinish.Anditshouldbeobviousthat theportface,thegasketsandheatblock(ifany)andcarburetorshouldallalignveryneatly,withoutanystepsbetweenpartsevenifthatmeansdowelingeverythingin placetoassurealignment.

thingonemustrememberwhencarvingawayatanintakeportisthatripplesintheportwalls,oranysuddenchangeincrosssection,haveafarmoredamagingeffecton flowcoefficientthanaslightlyroughfinishintheport.Therefore,itisvastlymoreimportanttosmooththeportthantogiveitamirrorfinish.Anditshouldbeobviousthat theportface,thegasketsandheatblock(ifany)andcarburetorshouldallalignveryneatly,withoutanystepsbetweenpartsevenifthatmeansdowelingeverythingin placetoassurealignment. Scavenging,inthecontextofpistontypeinternalcombustionengines,istheprocessinwhichtheproductsofcombustionareclearedfromacylinderattheendof thepowerstrokeandafreshair/fuelchargeisintroducedinpreparationforthecompressionandpowerstrokestofollow.ThisprocessiscommontoallOttocycle engines,butitcanbeaccomplishedintwoentirelydifferentways:Inthefourstrokecycleengine,itoccupiesatminimumafull360degreesofcrankshaftrotation,with onepistonstrokebeingdevotedtopushingexhaustproductsfromthecylinder,pastavalveinthecylinderheadthereturnstrokeaspiratesafreshchargeinthrough anotherportandpastanothervalve.Thus,thereisafairlycompletemechanicalseparationofthegasesinvolved,andwhilevalvetimingwillcommonlybearrangedsothat thereissomeoverlappingoftheexhaustandintakephasesofthisoperation,littledilutionofthefreshchargebyexhaustgasesispossible,andanyshortcircuitingofthis air/fuelmixtureouttheexhaustportduringtheearlystagesoftheintakeopenperiodhasnoadverseeffectontheweightofthechargeultimatelytrappedinthecylinderat intakeclosing.Obviously,themechanismrequiredtooperateafourstrokeengine'svalvesaddsconsiderablecomplicationtothebasiccrankrodpistonassembly,but theveryefficientscavengingobtainedis,formostapplications,consideredtobeworththevalvegearcomplexities.Andtheveryhighbrakemeaneffectivepressures resultingfromthishighscavengingefficiencyabmepintheorderof200psiforhighlytunedexamplesofthetypeoffsetthescheme'ssingledisadvantage,whichisthat powerstrokesoccurat720degreeintervals. Twostrokecycleenginesdeliverpowerstrokestwiceasoften,at360degreeintervals,andinthoseintendedforindustrial(trucks,electricalgenerators,etc.)or marineapplicationsessentiallythesamebmepasareobtainedfromfourstrokeenginesaretobeexpected.Butinsuchenginesonefindsanevengreatermechanical complexitythaninthefourstrokedesign,forinalltwostrokeenginesthescavengingprocessoccursintimeborrowedfromthecompressionandpowerstrokes.In effect,thismeansthatallofthecylinderclearingandrechargingforwhich360degreesofcrankrotationarereservedinthefourstrokeenginemustoccurwhilethepiston ishaltedatthebottomofitsstroke.And,lackingtimeforamoreleisurelyexchangeofgases,theprocessmustbehelpedalongbyextremelylargeportareasandhigh scavengingpressures.Usually,enginesofthetypebeingdiscussedwillhavemultipleexhaustvalvesintheircylinderheads,andaringofwindowsaroundthecylinder's basethroughwhichscavengingairisforcedbyanenginedrivenpump.Suchenginesalmostinvariablyarediesels,inwhichfuelisinjectedonlyafterallthevalvesand portsareclosed(injectionbeginningatTDCandcontinuingforperhaps70degreesofcrankangle)andthescavengingpumpdeliversairinexcessofwhatisrequiredto fillthecylinder,sothereisnolossofchargenoranydilutiontocausealossofpower.Unfortunately,notonlyisthistypeengineverycomplicatedandexpensive,itcannot bescaleddowntoasizeusefulinmotorcycletermssimplybecausetheexhaustvalveswillnotopenandclosefastenoughtokeepupwiththekindofcrankshaftspeeds neededforthepoweroutputswehavecometoexpectfromoursmalldisplacementengines. Intheend,onlythefamiliarpistonporttransferandexhaustvalvingarrangementissuitableformotorcycleengines,andthatiswithanexception,ofsorts, existingintheuseofdiscorreedtypeintakevalvingpreciselywhathascometobeuniversallyapplied.Presentmotorcycleenginesareallscavengedthroughwindows intheircylinderwalls,withscavengingairbeingsuppliedfromtheircrankcases.Thissystemisbeautifulinitssimplicity,butitdoeshaveseriousshortcomings:First,there istherelativeincapacityofthecrankcaseasascavengingairpump,whichpreventseventhehopeofhavingexcessairtouseinclearingthecylinder.Secondly,theuseof thepiston'smotionstoopenandclose(actually,touncoverandcover)theexhaustandtransferportscreatesenormousdifficultiesinanumberofareasrelatedtoclearing andrechargingthecylinder.Thelowscavengingpressureavailablemakesitabsolutelyessentialthatpressureinthecylinderbenomorethanslightlyhigherthan atmosphericwhenthetransferportsopen,whichmeansthattheexhaustphasemustbeginwellinadvanceoftheuncoveringoftransferports.And,becausethepiston controlledexhausttimingisnecessarilysymmetrical,theexhaustportwillremainopenlongafterthetransferportscloseleavinganunobstructedopportunityforthefresh chargetoescapethecylinder.Indeed,thechargeinjectedintothecylinderhaseveryreasontoescape,astheupwardmotionofthepiston,movingtoclosetheexhaust portandbegintheeffectivecompressionstroke,isdisplacingthegasesaboveitscrown.Gaspressuresalwaystrytoequalize,andthoseinthecylindercanonlydothat bymovingbackintothetransferports,whilethesearestillopen,andouttheexhaustports.Thus,itisvirtuallyinevitablethatsomeportionofthefreshchargewillbelost intotheexhaustsystem,andthattheupwardstrokeofthepistonwillalsotendtoaspiratesomeofthechargebackdowninthecrankcase.

 Scavenging,inthecontextofpistontypeinternalcombustionengines,istheprocessinwhichtheproductsofcombustionareclearedfromacylinderattheendof thepowerstrokeandafreshair/fuelchargeisintroducedinpreparationforthecompressionandpowerstrokestofollow.ThisprocessiscommontoallOttocycle engines,butitcanbeaccomplishedintwoentirelydifferentways:Inthefourstrokecycleengine,itoccupiesatminimumafull360degreesofcrankshaftrotation,with onepistonstrokebeingdevotedtopushingexhaustproductsfromthecylinder,pastavalveinthecylinderheadthereturnstrokeaspiratesafreshchargeinthrough anotherportandpastanothervalve.Thus,thereisafairlycompletemechanicalseparationofthegasesinvolved,andwhilevalvetimingwillcommonlybearrangedsothat thereissomeoverlappingoftheexhaustandintakephasesofthisoperation,littledilutionofthefreshchargebyexhaustgasesispossible,andanyshortcircuitingofthis air/fuelmixtureouttheexhaustportduringtheearlystagesoftheintakeopenperiodhasnoadverseeffectontheweightofthechargeultimatelytrappedinthecylinderat intakeclosing.Obviously,themechanismrequiredtooperateafourstrokeengine'svalvesaddsconsiderablecomplicationtothebasiccrankrodpistonassembly,but theveryefficientscavengingobtainedis,formostapplications,consideredtobeworththevalvegearcomplexities.Andtheveryhighbrakemeaneffectivepressures resultingfromthishighscavengingefficiencyabmepintheorderof200psiforhighlytunedexamplesofthetypeoffsetthescheme'ssingledisadvantage,whichisthat powerstrokesoccurat720degreeintervals. Twostrokecycleenginesdeliverpowerstrokestwiceasoften,at360degreeintervals,andinthoseintendedforindustrial(trucks,electricalgenerators,etc.)or marineapplicationsessentiallythesamebmepasareobtainedfromfourstrokeenginesaretobeexpected.Butinsuchenginesonefindsanevengreatermechanical complexitythaninthefourstrokedesign,forinalltwostrokeenginesthescavengingprocessoccursintimeborrowedfromthecompressionandpowerstrokes.In effect,thismeansthatallofthecylinderclearingandrechargingforwhich360degreesofcrankrotationarereservedinthefourstrokeenginemustoccurwhilethepiston ishaltedatthebottomofitsstroke.And,lackingtimeforamoreleisurelyexchangeofgases,theprocessmustbehelpedalongbyextremelylargeportareasandhigh scavengingpressures.Usually,enginesofthetypebeingdiscussedwillhavemultipleexhaustvalvesintheircylinderheads,andaringofwindowsaroundthecylinder's basethroughwhichscavengingairisforcedbyanenginedrivenpump.Suchenginesalmostinvariablyarediesels,inwhichfuelisinjectedonlyafterallthevalvesand portsareclosed(injectionbeginningatTDCandcontinuingforperhaps70degreesofcrankangle)andthescavengingpumpdeliversairinexcessofwhatisrequiredto fillthecylinder,sothereisnolossofchargenoranydilutiontocausealossofpower.Unfortunately,notonlyisthistypeengineverycomplicatedandexpensive,itcannot bescaleddowntoasizeusefulinmotorcycletermssimplybecausetheexhaustvalveswillnotopenandclosefastenoughtokeepupwiththekindofcrankshaftspeeds neededforthepoweroutputswehavecometoexpectfromoursmalldisplacementengines. Intheend,onlythefamiliarpistonporttransferandexhaustvalvingarrangementissuitableformotorcycleengines,andthatiswithanexception,ofsorts, existingintheuseofdiscorreedtypeintakevalvingpreciselywhathascometobeuniversallyapplied.Presentmotorcycleenginesareallscavengedthroughwindows intheircylinderwalls,withscavengingairbeingsuppliedfromtheircrankcases.Thissystemisbeautifulinitssimplicity,butitdoeshaveseriousshortcomings:First,there istherelativeincapacityofthecrankcaseasascavengingairpump,whichpreventseventhehopeofhavingexcessairtouseinclearingthecylinder.Secondly,theuseof thepiston'smotionstoopenandclose(actually,touncoverandcover)theexhaustandtransferportscreatesenormousdifficultiesinanumberofareasrelatedtoclearing andrechargingthecylinder.Thelowscavengingpressureavailablemakesitabsolutelyessentialthatpressureinthecylinderbenomorethanslightlyhigherthan atmosphericwhenthetransferportsopen,whichmeansthattheexhaustphasemustbeginwellinadvanceoftheuncoveringoftransferports.And,becausethepiston controlledexhausttimingisnecessarilysymmetrical,theexhaustportwillremainopenlongafterthetransferportscloseleavinganunobstructedopportunityforthefresh chargetoescapethecylinder.Indeed,thechargeinjectedintothecylinderhaseveryreasontoescape,astheupwardmotionofthepiston,movingtoclosetheexhaust portandbegintheeffectivecompressionstroke,isdisplacingthegasesaboveitscrown.Gaspressuresalwaystrytoequalize,andthoseinthecylindercanonlydothat bymovingbackintothetransferports,whilethesearestillopen,andouttheexhaustports.Thus,itisvirtuallyinevitablethatsomeportionofthefreshchargewillbelost intotheexhaustsystem,andthattheupwardstrokeofthepistonwillalsotendtoaspiratesomeofthechargebackdowninthecrankcase. Difficultiesinherentinthepistonportscavengingsystemarenotconfinedtochargeloss,orbackflowintothecrankcase.Oneofthegreatproblemsiscreatedby thelackofmechanicalseparationoftheexhaustgasesandtheincomingfreshcharge.Weexpectthattheengine'sexhaustgaseswillchoosetoescapefromtheexhaust port,andthatthechargecominginthroughthetransferportswillpushtheresidualexhaustproductsaheadofittocompletelyclearthecylinder,buttheactualprocessis bynomeansthattidy.Thecylinderpressuremaydropveryneatlytoatmospheric,orevenbelow,butitstillwillbefilledwithexhaustgases,andthesewillnotnecessarily besweptouttheexhaustportmerelybecauseothergaseshaveenteredthecylinder.Inpointoffact,itispossibletoshortcircuitthescavengingflowstraightfromthe transferportstotheexhaustportandleavetheexhaustresidualsintheuppercylinderentirelyundisturbed.Thispossibilityhashauntedthedesignengineerthroughoutthe twostrokeengine'slonghistory,andmanyanelaboratesystemofportsandpistoncrownconvolutionshasbeencreatedtoconfoundthisworstofalldemons. Formanyyearscrossflowscavengingwaspreeminent,principallybecauseitmakesmaximumuseofcylinderwallarea.Inthecrossflowengine,portsring virtuallytheentirelowercylinder,withhalfbeingexhaustandthoseoppositebeingtransfer.Giventhedirectionoftheflowemergingfromthetransferports,thecharge wouldshootstraightacrossandouttheexhaustsidebutfortheshapeofthepistoncrownwhichlipsuponthetransfersidetoformadeflectorthatredirectsthetransfer flowupward.Crossflowscavengingisstillemployedinoutboardmarineandmodelairplaneengines,inthelatterbecauseitisamanufacturingconvenienceandinthe formerbecauseitsinsensitivitytoscavengingpressuresandvolumeofflowprovidesuperiorlowspeed,partthrottlerunningcharacteristics.Thecrossflowsystemis, however,handicappedintermsofmaximumpowerbythelargesurfaceareacreatedinallthatlumpinessinthe

wouldshootstraightacrossandouttheexhaustsidebutfortheshapeofthepistoncrownwhichlipsuponthetransfersidetoformadeflectorthatredirectsthetransfer flowupward.Crossflowscavengingisstillemployedinoutboardmarineandmodelairplaneengines,inthelatterbecauseitisamanufacturingconvenienceandinthe formerbecauseitsinsensitivitytoscavengingpressuresandvolumeofflowprovidesuperiorlowspeed,partthrottlerunningcharacteristics.Thecrossflowsystemis, however,handicappedintermsofmaximumpowerbythelargesurfaceareacreatedinallthatlumpinessinthe

piston'scrown,whichverygreatlyincreasestheheatflowintothepiston andlowersthecompressionratioagivencylinderwilltolerateascomparedtoapistonwithaflatorslightlydomedpistoncrown.Apartfromthisthermalproblem,there ismuchtorecommendcrossflowscavenging,butthethermalproblemisofsufficientmagnitudeeveninwatercooledenginestoremoveitfromseriousconsideration foranyhighoutput,twostrokeengine. Numerousscavengingsystemsnotrequiringadeflectortypepistonhavebeentried:ThepreWW2Villiershadexhaustportsonoppositesidesofitscylinderand fourtransferports,inpairs,betweenthem.ABarnesandReineckedesignhadaringofexhaustportslocatedabovearingoftransferportsandapartconicalpiston crown,allofwhichsentthescavengingflowinanarrowcolumnupthemiddleofthecylinder,andforcedtheexhaustoutflowtofollowapathdownthecylinderwalls. Curtissemployedmultipletransferandexhaustportsonoppositesidesofacylinder,andbiasedthedirectionofthetransfersupwardandtooneside,sothatthefresh chargespiraledupintothecylinder.ButthebestofthescavengingsystemswasonedevisedbyaDr.Schneurle,ofGermany,inwhichapairofmirrorimagetransfer portsflankedasingleexhaustport,directingthescavengingflowtowardthecylinderwalloppositetheexhaust,andupward,toloopoverandthusclearthecylinder. Schneurle'sloopscavengingmethodwaspatentedbyhim,in1925,andthishadtheeffectofsimultaneouslyelevatingGermanindustry'sfortunesinthetwostrokeengine fieldwhileforcingpracticallyeveryoneelsetoseekalternativeandlessefficientsystems.Ofcourse,nowthattheSchneurlepatentshaveexpiredeveryoneemployssome formofhisscavengingmethod,althoughbestresultsarebeingobtainedwithmorethanSchneurle'soriginalpairoftransferports. Onlyadecadepast,EastGermany'sMZwasconsideredtobetherepositoryofreallyadvancedresearchinhighspeedtwostrokeenginedesign,andoneWalter Kaadencouldbesaidtohavethebestgraspoftheintricaciesofscavengingsystemsofanyoneworkinginthefield.Today,nodiscussionoftwostrokeenginescavenging ispossiblewithoutconcentratingalmostexclusivelyondevelopmentinJapan.Japaneseengineersdidnotinventthetwostrokeengine,norhavetheyemployedany systemofscavengingportsthathasnotseenearlierserviceelsewhere.Buttheyhavedoneanenormousamountofbasicresearchdirectedatquantifyingwhatpreviously hasbeenknownonlyintermsofgeneralitiestheyhaveestablishedveryfirmdesigncriteriaforthemanagementoffactorsthatonceweredecidedalmostpurelythrough cutandtryexperimentation.Ofcourse,noneofthiswouldbeofmorethanincidentalinterestbutforthefactthatsomeoftheJapanesefirmshaveabandonedtheironce absolutepolicyofsecrecyandaresharingwhattheyhavelearnedwiththerestoftheworld.Yamaha,particularly,hasmadeavastcontributiontotheoverallstateofthe artbypublishingfairlyspecificcriteriafortheporttimingsandareasrequiredforenginesofanygivencylindervolumeandoperatingspeed.Likemanyothers,Iknewthat porttimingandareawereinterrelatedfactors,butthejobofobtainingandsortingthroughdataonawiderangeofenginestoestablishapattern,andthenexperimentally verifyingconclusionswastootimeconsumingandexpensivetoevencontemplate,asanindividual.Yamahahasdonethatworkforus,andpublishedenoughinformation onthesubjecttocompleteatleastmyunderstanding(adetaileddiscussionispresentedelsewhereinthisbookasachapter,headed,PortTiming).Fromanumberof SAEpapersfromJapanaswellasexamplesfromGermanyandtheUnitedStatesandmyownexperience,Ihavealsoaccumulatedmuchincidentalinformationrelated totheshapes,numberanddispositionofports.Thesefactorsprofoundlyinfluencescavengingflow,whichinfluenceshorsepowerverygreatly,andwewillforthemoment concentrateonthemalone. EXHAUSTPORT Thedifferencebetweensuccessandfailurewithamodifiedenginecanbethetreatmentoftheexhaustport.Evenassumingthatnochangeismadeinexhaustport

onthesubjecttocompleteatleastmyunderstanding(adetaileddiscussionispresentedelsewhereinthisbookasachapter,headed,PortTiming).Fromanumberof SAEpapersfromJapanaswellasexamplesfromGermanyandtheUnitedStatesandmyownexperience,Ihavealsoaccumulatedmuchincidentalinformationrelated totheshapes,numberanddispositionofports.Thesefactorsprofoundlyinfluencescavengingflow,whichinfluenceshorsepowerverygreatly,andwewillforthemoment concentrateonthemalone. EXHAUSTPORT Thedifferencebetweensuccessandfailurewithamodifiedenginecanbethetreatmentoftheexhaustport.Evenassumingthatnochangeismadeinexhaustport timing,simplywideningtheportwindowwillresultinapowerincreaseitalsocanresultindrasticallyshortenedringlife,amountinginextremeexamplestooutrightand nearlyinstantbreakageoftherings,and/orsevereoverheatingofthepistoncrown.Therearereasonsfortheseproblems:Atwostrokeengine'spistonringsalwaysbulge outintoanyportwindowtheypass,andwhiletransferportwindowsseldomarewideenoughtopermitthistoanextentsufficienttocausedifficulties,thesamecertainly maynotbesaidoftheexhaustport.Arelativelymildlytunedenginewillhaveanexhaustportwidthequaltoatleast50percentofitsborediameter(whichistosay,a3 inchcylinderborewouldhavea1.5inchexhaustportwidth)andthatisenoughtoallowtheringtospringoutintotheportwindowveryperceptibly.Makethatport square,withsharpcomersandsharpedges,andtheringwillbedestroyedveryquickly.Andifyouenlargetheportsothatitswidthrepresents70percentofcylinder borediameter,ringfailurewouldalmostcertainlyoccurduringthefirstrevolutionofthecrankshaft.Yet,racingengineshavebeenrunquitesuccessfullywith70percent portwidths,andwhileIcannotrecommendthatkindofextremityformodifiedproductiontypeengines,themerefactthatithasbeendonetellsusthattechniquesexistto makeitpossible. Basically,ringlifeisimprovedatanygivenportwidthbyA)roundingtheshapeoftheportwindow,andB)breakingthewindow'ssharpedges.Bothofthese measuresareemployedinallengines,buttheyreachspecialandsomewhatexaggeratedforminracingapplications.Thetraditionalportwindowshapeissquare,or rectangular,withitscornersroundedtohelppreventringsnagging.Assumingthattheportwidthdoesnotexceed60percentofborediameter,theradiiattheport window'scornersshouldbeabout15or20percentoftheportwidth,andthatisjustwhatyouwillfindinmostengines.Butastheportiswidened,thosecornerradii havetobemadelargertoabout2spercentofportwidthwhenthelatterapproachesbeing70percentofcylinderdiameter.Actually,eventheseverylargeradiiwillnot completelypreventringsnaggingiftheyarenotjoinedbystraightlineedges.Theupperandloweredgesoftheportwindowshouldbearched,onaradiusequaltoabout twicetheportwidth,inportshavingawidththatis60percentofborediameterorless.Unfortunately,thesesimplerulesofthumbarenotadequateatportwidthsabove the60percentlevelandsuchwidthsarebecomingverycommon.Withtheringswehadadecadepast,whichwererelativelythick,axially,andquitenarrow,radially, (nottomentionbeingmadeofordinaryandratherbrittlecastiron)theupperlimitforportwidthwas62percentofcylinderborediameternow,withourvastlyimproved ringsithasbecomepossibletowidentheportoutto70percentofbore,orslightlymore,ifweareverycarefulinshapingtheportwindow. Justasthereisnomeansofpredicting,withanygreataccuracy,whatkindofcamandtaperapistonwillrequiretofitcloselyinacylinderwhenbothareat operatingtemperature,neitheristhereanyfirmruleforshapingultrawideexhaustportwindows.Bothareestablished,initially,onthebasisofpastexperience,andthen modifiedaccordingtotestresults.Ithasbeendemonstrated,inpractice,thatamodified

ellipseisthebasicshapeofportwindowsinthe62 to70percent(of bore)range.Thus,whiletheringmayactuallybulgeoutintotheportwindowenoughtocauseitsinstantdestructioninasquareport,orinonewithstraightlineedgesare joinedwithsimpleradii,thecontoursofanellipticalportwindowwillsweeptheringgentlybackintoitsgroove.Then,theonlyproblemthatwillbeencounteredisthatthe ringmaybulgeout,andbepushedback,unevenlywhichmaydriveoneendoftheringsharplyagainstitslocatingpinandeventuallycausethepintoloosenandcome adrift.Itshouldbeobviousthatthislastdifficultywillbemostpronouncedwhentheportwindowisnotperfectlysymmetrical,asanydeparturefromsymmetrywillresultin theringbeingdisplacedtoonesideasitispushedbackintoitsgroove. Carefulcraftsmanshipwillpreventthisasymmetricaldisplacementoftheringitwillnot,ofitself,forestallotherproblemsassociatedwithverywideexhaustport windows.Inmyopinion,oneshouldneversimplyandarbitrarilywidenaportouttothe70percentlimit.Instead,thesafeandsensibleapproachistobeginat62 percent,withashapethatisasnearlyanellipseasispossible.Quiteobviously,sharplimitationsaregoingtobeimposedbytheshapeoftheexistingportwindowthe ideaistoprovidethemostgenerousradiipermittedbythebasicshapewithwhichonemustbegin.Obviously,too,thisreshapingoftheexhaustportwindowwillbe easierifyouhaveoptedforincreasingtheexhausttiming,asthatwillgiveyouroomtoworkabovetheexistingport.Then,havingestablishedtheinitialshape,youwill havetoinspecttheringsandtheedgesoftheportwindowforevidenceofscuffingorsnagging.Seldomwilltherebeanyproblemaroundtheloweredgeoftheport,as thepistonslowsconsiderablynearthebottomofitsstroke.Mostofanyscuffingthatappearswillbearoundthecomersoftheportoutrightsnaggingwillmakeits presenceknownintheappearanceofscratchesleadingupwardfromthecenteroftheportwindow. PORTEDGECHAMFERS Ihavealreadyindicatedthatthekindofreshapingpossibleislargelyafunctionofthestockportwindow'sshape,butalterationsinshapearenottheonlycurefor scuffingandsnaggingavailabletous.Practicallyeveryoneknowsenoughtoroundofftheportwindow'ssharpedgeswhatmostdonotknowisthatasimpleroundoffis notwhatisneeded.Thepurposeinbreakingthoseedgesatallistopreventringsnaggingbyeasingtheringbackintoitsgroove,andthisjobisdonebestnotbyasimple radius,butbysurroundingthewindowwithaveryslightbevel,about0.080inchinwidth,andtaperingintowardthewindowtoadepthofperhaps0.015inch.Itisof coursenecessarytoworkaslightradiuswherethebevelreachestheportwindow,justtobesafe,buttherealjoboftuckingtheringsafelyawayinitsgrooveis performedbythebevel.Youwillappreciatethatthesamekindofbevelisneededatalltheportwindowsthoseatthetransferportsneednotbequiteaslarge. Asregardstheexhaustport,asecondaryfunctionisservedbyprovidingabevel,

radius,butbysurroundingthewindowwithaveryslightbevel,about0.080inchinwidth,andtaperingintowardthewindowtoadepthofperhaps0.015inch.Itisof coursenecessarytoworkaslightradiuswherethebevelreachestheportwindow,justtobesafe,buttherealjoboftuckingtheringsafelyawayinitsgrooveis performedbythebevel.Youwillappreciatethatthesamekindofbevelisneededatalltheportwindowsthoseatthetransferportsneednotbequiteaslarge. Asregardstheexhaustport,asecondaryfunctionisservedbyprovidingabevel,

andradiusededges,aroundtheportwindow.Thereisavery considerablycontractionofflowthroughanysharpedgedorifice,andsuchorificesmaybemadeeffectivelylargerbyprovidingthemwitharoundedentry.Improvements inflowintheorderof30percentcouldbehadwereitpossibletogivetheportwindowedgesaradiusof,say,inch.Unfortunately,todothiswouldmeanadvancing thepointofexhaustopeningalikeamount,whichinmostengineswouldresultinaveryradicalexhausttimingindeed.Itis,ontheotherhand,oftenpossibletocarvejust sucharadiusatthesidesofanexhaustportalthoughitisquestionablethatthisradiuswouldbeaseffectiveassimplywideningtheporttothesameextent.Theradius approachdoeshavetheadvantageofleavingintactmuchofthemetalaroundtheport,whichcanbeimportant:Thicksectionsofmetaltendtoequalizecylinder temperaturesandpreventthekindoflocaldistortionthatissuchapotentcauseofpistonseizure.Also,inengineshavingexhaustportscloselyflankedbycylinderhold downstuds,theremaynotberoomenoughtowidentheportasmuchaswouldotherwisebedesirable,andinthateventthesideradiusedportsbecomeanecessity. Inthevastmajorityofenginestherewillbeafairlylargeincreaseincrosssectionalareabetweenthestockexhaustwindowandtheactualexhaustoutlet.Indeed, thisincreaseoftenistoolargetogivebestresultswithexpansionchamberexhaustsystems:Whatmayseemtobenomorethanaflowimprovingenlargementinarea leadingintotheexhaustpipeduringtheoutflowphaseofscavengingbecomesasuddenconstrictionforwavesreturningtothecylinderfromtheexpansionchamber.In fact,ifthedifferenceintheareasattheportwindowandtheouterendoftheportbecomesasgreatas1:3,virtuallyalloftheexpansionchamber'sresonanteffectswillbe lost.Whathappens,insuchcases,isthatthewavesreturningtothecylinderarereflectedbackintothechamberbytheabruptconstrictionoftheport.Maximum transmissionofthesewavesintothecylinderwill,ofcourse,beobtainedwitha1:lportwindow/portoutletratio,butthatkindofstraightthroughpassagerepresents somethinglessthantheoptimuminminimizedflowresistanceduringtheblowdownphaseofscavenging(theperiodbeginningwhentheexhaustportcracksopenand endingwiththeopeningofthetransferports).Thus,thewallsoftheexhaustportshoulddivergesomewhat,givingaprogressivelyincreasingcrosssectionalareaoutto theexhaustflange.Themostimportantthingtoremember,here,isthatsuddenchangesinsectionshoulddefinitelybeavoided.Neithergasflownortheeffectsofsonic wavesintheexhausttractareservedbyabunchoflumpsandjogsthisbeingfarmoreimportantthanamirrorfinishontheportwalls.Givenanabsolutelyfreehand withanexhaustport(whichseldomispossible,whenyou'remodifyinganexistingengine)Iwouldbeinclinedtomaketheexhaustpassage'sareasequaltothoseinan8 degreecone.However,unlessyouhaveadynamometeravailablefortheverificationofexhaustpassageexperiments,thebestapproachtothiswholematteristoleave unalteredtheexhaustflangeendoftheport,andcarefullyblendtheenlargedexhaustwindowintotherestoftheport. Thereisonemoreitemtobecoveredhere:thebridgedexhaustport.ThebridgetowhichIreferisaverticalribthatdividestheexhaustportandcreatestwo separateexhaustwindows.Fewengineshavebridgedexhaustports,butifyoursdoes,itshouldneitherberemovednornarroweddespitetheobviousadvantagein gasflowtobegainedthereby.One,andperhapsthemostimportant,reasonwhywedonotseemorebridgedexhaustportsisthatthebridgeisinevitablypoorlycooled. Inconsequence,ittendstodistortbackintothecylinderslightly,andthatoccurrencebecomesthemoreunfortunatebecausethepiston/cylinderpressurepointthus createdisverypoorlylubricatedandlocalizedseizingalongthebridgeiscommon.Narrowingthebridgereducestheheatpathintothemetalsurroundingtheport,making thebridge'stemperaturejustthatmuchhigher,whileweakeningitandmakingseriousdistortionallthemoreprobable.Despitetheseconsiderableshortcomings,the bridgedexhausthasbeenemployedandwillcontinuetobeincertainapplications,foritoffersanopportunitytomaketheexhaustportwindowareaextremelylarge

withouttheusualattending difficultieswithringsnagging.Ontheother hand,itmakesworseoneoftherealproblemsinherentinoversizeexhaustwindowswhichispistoncrownoverheating.Atthemomentofexhaustportopening,agreat blastoffiregoesjettingdownthesideofthepiston,andthisputsalotofheatintothepistonskirt.Clearly,anyenlargementofthewindowexposesmoreofthepistonto

Inconsequence,ittendstodistortbackintothecylinderslightly,andthatoccurrencebecomesthemoreunfortunatebecausethepiston/cylinderpressurepointthus createdisverypoorlylubricatedandlocalizedseizingalongthebridgeiscommon.Narrowingthebridgereducestheheatpathintothemetalsurroundingtheport,making thebridge'stemperaturejustthatmuchhigher,whileweakeningitandmakingseriousdistortionallthemoreprobable.Despitetheseconsiderableshortcomings,the bridgedexhausthasbeenemployedandwillcontinuetobeincertainapplications,foritoffersanopportunitytomaketheexhaustportwindowareaextremelylarge

withouttheusualattending difficultieswithringsnagging.Ontheother hand,itmakesworseoneoftherealproblemsinherentinoversizeexhaustwindowswhichispistoncrownoverheating.Atthemomentofexhaustportopening,agreat blastoffiregoesjettingdownthesideofthepiston,andthisputsalotofheatintothepistonskirt.Clearly,anyenlargementofthewindowexposesmoreofthepistonto thiskindofheating,whiledeprivingtheskirtofpartofitscontactwiththerelativelycoolcylinderwall.Raisingtheexhaustportheighthasthesameeffect,andeitherof thesepowerenhancingmodificationsmay,astheyarecarriedprogressivelyforward,requireacompensatingreductionincompressionratiotoprotectthepiston. FLOWPATTERNS Someenginesdonotrespondnearlyaswelltoincreasedexhaustportwidthastheyshould,whichbringsustotheoverallproblemofflowinthecylinderduring thescavengingoperation,andthetransferports.Aswasstatedearlier,wemayimaginethemixtureflowingfromthetransfersandneatlysweepingawayresidualexhaust products,butitdoesnotreallyhappeninquitethatway.Foronething,theresimplywillnotbeavolumeofgasesdeliveredupfromthecrankcasesufficienttoclearallthe exhaustproductsfromthecylinder.Seldomwillthedeliveryratio(theratioofpistondisplacementtoairactuallypumpedintothecylinder)bemuchbetterthan1:0.8,or 80percent.Therefore,acylinderhavingapistondisplacementof250ccwillonlyhave200ccofair/fuelmixturecomingintoitthroughthetransferportleaving,by implication,atleast50ccofexhaustgasestrappedinthecylinderevenifweassumeanearperfectseparationofexhaustproductsandtheincomingcharge.Actually, therewillbesomemixingofthetwoduetoturbulence,withtheresultthatsomepartofthechargeislostouttheexhaustportandthereisagreaterdilutionofthefresh charge,withexhaustproducts,thanwouldbeassumedfromthedeliveryratioalone. Deliveryratioisalmostentirelyafunctionofcrankcasepumpingefficiency,andthetransferports'timeareafactorwhichistosay,thevolumeofthecharge deliveredintothecylinderisentirelyindependentofthenumberanddispositionofthetransferports.Thosethingsaredealtwithelsewhereinthisbookweareconcerned herewithmakingthemostofthemixtureactuallydelivered,andinthatregardtheimportanceofthetransfers'shapesandplacementcannotbeexaggerated.Often,the mostsubtlechangesyieldverylargedifferencesnotonlyinpeakpower,butintheshapeoftheentirepowercurve,anditisalltooeasytodealanengineaconsiderable injurywhileperformingsomeminoralterationwithasteadyhandandthebestofintentions.Inthisrespect,Ithinkitmostunfortunatethatthetwostrokeenginecannotbe drivenbelowaminimumlevelofoperatingefficiencybyeventhemostawfulbutcheryofitstransferports,asanenginethusservedwillcontinuetorun,andrunfairly cleanly,afterthatkindofsurgery.Howmuchbetteritwouldbeifaseriousdeparturefromtheoptimumwouldproduceagreatfitofmisfiring,orsomeotherobvious evidenceofdistress.Unhappily,itwillnot,whichmeansthatanengine'stransferportsshouldbeleftstrictlyaloneunlessyouhaveboththeknowledgeandthetoolsto makeanymodificationsproperly.ThetoolsyouwillhavetoacquireonyourownknowledgeiswhatIhopetoprovidewiththisbook. PerhapsthemostvaluablebitofinformationIcansupplyisthatunlessyouplantoalterfairlyradicallythespeedatwhichyourenginemakesitsmaximumoutput, thereisnoneedtodoanythingbeyondsmoothingthecastingflawsoutofthetransferportsandeventhatshouldbeapproachedwithsomecaution.Why?Becausein scavengingefficiencysoverymuchdependsonsymmetryofflow.Getonetransferportflowingconspicuouslybetterthanitsmateontheoppositecylinderwall,andwhile youmayhaveimprovedthedeliveryratioslightly,thescavengingpatternwillhavebeenupsetandpoweroutputwilldrop.Asamatteroffact,Iaminclinedtothinkthat theupperreachesofthetransferpassagesshouldbeleftentirelyalone,unlesstoremovesomeconspicuouscastingdefect.Itisinanycaseextremelydifficulttodo

thereisnoneedtodoanythingbeyondsmoothingthecastingflawsoutofthetransferportsandeventhatshouldbeapproachedwithsomecaution.Why?Becausein scavengingefficiencysoverymuchdependsonsymmetryofflow.Getonetransferportflowingconspicuouslybetterthanitsmateontheoppositecylinderwall,andwhile youmayhaveimprovedthedeliveryratioslightly,thescavengingpatternwillhavebeenupsetandpoweroutputwilldrop.Asamatteroffact,Iaminclinedtothinkthat theupperreachesofthetransferpassagesshouldbeleftentirelyalone,unlesstoremovesomeconspicuouscastingdefect.Itisinanycaseextremelydifficulttodo

preciseworkupneartheport window,andalmostimpossibletoraisethe port'sroofwithoutalteringitsangle whichisthewrongthingtobedoing.Allthingsconsidered,itisprobablyeasiesttoraisethetransferports,whenyouwantto increasethetransfertiming,byraisingtheentirecylinder.Aspacerunderthecylinderwillaccomplishthis,anditisusuallyasimplemattertotrimtheloweredgesofthe transfersandexhaustporttoalignwiththeedgeofthepistoncrownatbottomcenter.Ofcourse,thismethodshortenstheintaketiming,andsoyouwillhavetodoabit oftrimmingthereaswell,butanythingiseasycomparedwithtryingtocarvehighertransferportswiththeportroofsheldtotheiroriginalconfiguration.Unlessyou discoverthatliftingthecylinderraisestheintakeporttothepointwherethepistonringendsspringoutintoitwhenthepistonmovesdowntothebottomofitsstroke,or unless,forsomereason,itisnotpossibletomachineathicknessequaltothatofyourspacerfromthetopofthecylindertoreturntotheoriginalcompressionratio.When eitherofthosethingspresentaproblem,changesintransfertimingshouldbeeffectedbycuttingshallowtroughsinthepistoncrownwhichisameasurethatcanbeused ontheexhaustside,too,andshouldbeusedasapreliminaryexperimenttoseewhethertheporttimingyouthinkyouwantiswhatyoureallyneed. AllpresentloopscavengedmotorcycleenginesfollowDr.Schneurle'soriginalpatternfairlyclosely,andmanyperhapsmostaremoreorlessexactlythesame:

Twin streamsofincomingchargeemergefromtwintransferportsflanking theexhaustport,andanglebackacrossthepistoncrownandslightlyupward,joiningintoasinglestreamatapointapproximatelytwothirdsofthewaybackfromthe exhaustport.Thisstreamisdeflectedupwardbytherearcylinderwall,andthenitsweepsuptothetopofthecylindertobedirectedbackdowntheforwardcylinder wallmovingtheresidualexhaustgasesoutoftheexhaustportasitadvancesinthatdirection.Thereis,asyoumightexpect,someturbulencegeneratedbythisactivity, whichisunfortunatebecauseturbulencepromotestheverykindofchurningandmixingthatshouldbeavoided.Buttheturbulenceisminimizedwhentheflowis symmetrical,andtherewillthereforebelessdilutionofthefreshchargetrappedinthecylinderatexhaustportclosing.Skewingeithertransferporttooneside,orlifting theupperedgeofoneslightlyhigherthantheotherwillbadlyupsetthescavengingpattern Curiously,theilleffectsofsuchmisalignmentsaremorestronglyfeltatenginespeedsbelowthatwheremaximumhorsepowerisproduced.Onlyaslight weakeningofanengine'speakoutputisnotedwhenmodificationstoitstransferportsleavethescavengingflowbadlyaimed,butthereappearsamarkedinstabilityatthe pointofmaximumtorqueandbelow.Thisphenomenonwouldbearinvestigation,butIsuspectthatitreflectsthefactthatthereisanalmostexplosiveentryofthe scavengingstreamsintothecylinderathighspeeds.Also,itisamatterofrecordthatthemodern,shortstrokeengineisgenerallylesssensitivetoscavengingpatternsthan thelongstrokeenginesofyearspastthisbecausesheerbulkflow,evenifbadlydirected,doesamuchbetterjobofscavengingthekindoflow,largediametercylinder presentedbytheshortstrokelayoutthanataller,moreslendercylinder.Still,anddespitethefactthathighhorsepowernumbersmakegoodconversation,powerrangeis goingtobeanextremelyimportantconsiderationuntilsuchtimeaswehavetransmissionsprovidinginfinitelyvariableratios.Sothebestscavengingsystemisonethathas goodbulkflowcapabilitieswhilemaintainingahighdegreeofflowcontrol. Thislastitem,theproperdirectionofthescavengingstreams,isimportantforreasonsbeyondthereductionofturbulenceandfreshcharge/exhaustproducts mixing:Cylindersforhighperformanceenginesneedverywideportstoavoidexcessivetimingdurations,whichmeansthattheportsmustbecrowdedtogethertooclosely toentirelyavoidthedangersofshortcircuitingthecharge.Havingahighdeliveryratioavailsanenginenothingifthemixturestreamsemergingfromitstransferportsare allowedtodivertfromtheirintendedpathanddisappearouttheexhaustport.Obviously,thisdangerincreasesasthetransferportsarecrowdedclosertotheexhaust

presentedbytheshortstrokelayoutthanataller,moreslendercylinder.Still,anddespitethefactthathighhorsepowernumbersmakegoodconversation,powerrangeis goingtobeanextremelyimportantconsiderationuntilsuchtimeaswehavetransmissionsprovidinginfinitelyvariableratios.Sothebestscavengingsystemisonethathas goodbulkflowcapabilitieswhilemaintainingahighdegreeofflowcontrol. Thislastitem,theproperdirectionofthescavengingstreams,isimportantforreasonsbeyondthereductionofturbulenceandfreshcharge/exhaustproducts mixing:Cylindersforhighperformanceenginesneedverywideportstoavoidexcessivetimingdurations,whichmeansthattheportsmustbecrowdedtogethertooclosely toentirelyavoidthedangersofshortcircuitingthecharge.Havingahighdeliveryratioavailsanenginenothingifthemixturestreamsemergingfromitstransferportsare allowedtodivertfromtheirintendedpathanddisappearouttheexhaustport.Obviously,thisdangerincreasesasthetransferportsarecrowdedclosertotheexhaust portobviously,too,adegreeofcrowdingisalmostinevitable.Ontheavailableevidence,itwouldseemthatthepointatwhichshortcircuitingbecomesaproblemis

whentheseparation betweentheexhaustport'ssidewallandtheforward edgeofthetransferportisdecreasedbelow0.350inch thatminimumdimensionholdingtrueforcylindersof125ccandup.Butitshouldbeunderstoodthatthis proximityisacceptableonlywhendeterminedefforthasbeenmadetodirectthescavengingstreamssharplytowardtherearcylinderwall. MULTIPLETRANSFERPORTS Somemanufacturers,notablySuzuki,continuetoemployandgetgoodresultsfromasinglepairoftransferports.Nevertheless,themainthrustofdevelopment todayistowardincreasinguseofmultipletransferports.Thistrendbeganinthedesignofengineshavingotherthanpistoncontrolledintakeports,simplybecausesuch designshaveanunclutteredrearcylinderwallanditwasinevitablethatsomeonewouldaddathirdtransferportinthisotherwiseunemployedspace.MZ'sKaaden incorporatedathirdtransferportinhis1962GPengine,fedfromawindowhighonthepistonskirt,justtocoolthepistoncrownandcarryoiluptothewristpinbearing. Butwhileimprovedreliabilitymayhavebeenthemotivationbehindthischange,italsoresultedinapowerincrease.Othersfoundmuchthesameresult,andby1965the tripletransferlayoutwasstandardforengineshavingrotarydiscintakevalves. Isolatedexamplesofpistonportinductioncombinedwithmultipletransferportsaretobefoundthroughoutthetwostrokeengine'shistory.Still,Iaminclinedto assigncreditforstimulatingrealinterestinthelayouttoHermannMeier.MeierwashiredbyRoyalEnfieldtodesignanddeveloptheengineforthatfirm'sshortlived productionracer,butwasnotallowedanythingapproachingcompletefreedominthisendeavor.ManagementtoldhimtoworkaroundthethenpopularAlphacrank assembly(itwasamorereliablereplacementfortheVillierscrank,rodandbearings)andforbadeanythingastricky(andexpensive)asrotaryvalveinduction.Inshort, hewasaskedtodesignacylinderthatwouldmagicallyproduceallthepowerofagenuineGPenginewithoutanyofthelatter'scomplication.Understandably,Meier's besteffortscouldnotovercometheobstaclesplacedinhispathbyRoyalEnfieldsmanagement,buthediddevelopanenginethatwasanotchup,intermsofpower, overthetwincylinderYamahaproductionracersofthattimewhichwasquiteaneattrick.AndhisRoyalEnfieldenginedidhavemorethantheusualpairoftransfer ports.ThepresenceofalargeintakeportontherearcylinderwallpreventedMeierfrompositioningathirdtransferportthere,sohesimplycrowdedtheprimarytransfer portsforward,closertotheexhaustport,andaddedapairofsmaller,auxiliarytransferportsbetweenthemaintransfersandtheintakeport.Theauxiliaryportswerefed inpartdirectlyfromthecrankcase,butapairofwindowsadjacenttoandslightlyhigherthanthewristpin,inthepiston,alsofedmixturethroughmatchingholesinthe cylinderintothetransferpassages. The250ccGrandPrixRoyalEnfield,handicappedasitwasbymarginalreliabilityandwretchedhandling,soonpassedintooblivion,buthardonitsheelscame thenowfamousfiveportYamahawithavirtuallyidenticaltransferportsystem.Initially,theYamahafeditsauxiliarytransferportsentirelythroughwindowsinits pistons,andthetransfersthemselvesweremerelyapairofgroovesflankingtheintakeports.Thisarrangementsoon,however,gavewaytothenowestablishedYamaha practiceofprovidingalloftheflowfortheauxiliaryportsfromthecrankcaseitisonlyintherecentreedvalveenginesthatwindowsonceagainappearinYamaha pistons.Withoutreallyknowing,IwouldsuspectthatYamahadroppedthewindowedpistonconceptbecauseitaddsacomplicationinmanufacturingandweakensthe piston andbecauseYamaha'sneedlerollerwristpinbearingssurviveverynicelytheconditionsofscantylubricationandhightemperaturesthatcausedproblemsfor MeierandKaaden. Inanycase,multipletransferportsaremuchinstyle,anddoappeartogivesuperiorresultsifproperlymanaged.Butbeadvisedthatsimplyhackinganextrapair ofholesupthroughthecylinderdoesnotconstitutepropermanagementofthemultipletransferconcept,fortheauxiliarytransferportsmustactinconcertwiththemain t r a n s f e r s o r t h e y w i l l o n l y c o n t r i b u t e t o t u r b u l e n c e whichmeansanincreaseincharge

MeierandKaaden. Inanycase,multipletransferportsaremuchinstyle,anddoappeartogivesuperiorresultsifproperlymanaged.Butbeadvisedthatsimplyhackinganextrapair ofholesupthroughthecylinderdoesnotconstitutepropermanagementofthemultipletransferconcept,fortheauxiliarytransferportsmustactinconcertwiththemain t r a n s f e r s o r t h e y w i l l o n l y c o n t r i b u t e t o t u r b u l e n c e whichmeansanincreaseincharge

dilution and/oroutrightshort circuiting.Unlessyouhavethe equipment,anddexterity,tocutextraportswithgreataccuracy,youwouldbebetteroffconfiningyoureffortstoimprovingyourengine'ssinglepairoftransferports ifa singlepairiswhatithas. SUBTLETIES Thebestbasicshapeforatransferportisasweepingarc,butmanyare(forreasonsofmanufacturingcostorbecauseofspaceconsiderations)substantially straight,withasharpturnwheretheymeettheportwindow.Thereislittledifferenceinthesetwoapproachesintermsofbulkflow,butscavengingstreamcontrolis superiorwithtransferpassagesthatsweepwidetojoinwiththeportwindowslessabruptly.Consequently,thepeakpowerobtainedwillbemuchthesamewitheither transferpassageconfiguration,butthelessabruptshapegivesbetterperformancebelowthepowerpeak.Obviously,itismucheasiertotellaboutthedifferencethanto makeanyrealchangeintheports'shapes especiallywhentheengineinquestionisatwin,withitscylinderscrowdedtogether but

italsoobviouslyisgoodpracticetodoeverythingpossibletomakethe straightportalittlelessstraight.Thiscanbedonebycarvingawaymetalfromtheport'souterwall,andaddingasimilarthicknessofplasticDevconfilleronthewall nearestthecylinderbore.Rememberthatthetransferpassageshouldtaperdownfromitsentranceatthecylinderbasetotheportwindowitself,andthatitisvirtually impossibletomaketheentranceareatoolarge.Enlargingthetransferpassagesreducesanengine'sprimarycompressionratio,ofcourse,butwehavealreadyestablished thatcrankcasecompressionisnotterriblyimportantandmixturedeliveryintothecylindermostdefinitelyis! If,orwhen,youperformthemodificationjustdescribed,becarefulnottoaltertheportwindow'sdirection,eitherupwardortotheside.Youmightimproveupon themanufacturer'shandiworkinthisregardthechancesarethatanychangewillbefortheworse.Inpractice,theeffectiveareaofthetransferportsbecomesgreater whentheyareangledmoretowardthemiddleofthecylinderandlesstowardtherearcylinderwall,butwhileachangeintheorderof10degreeswillsometimesyielda boostinpeakpower,thisincreasewillbepurchasedatsomeconsiderablecostinpowerrangeandIcannotrecommendittoanyonewhoisunwillingtoA)buildaflow

If,orwhen,youperformthemodificationjustdescribed,becarefulnottoaltertheportwindow'sdirection,eitherupwardortotheside.Youmightimproveupon themanufacturer'shandiworkinthisregardthechancesarethatanychangewillbefortheworse.Inpractice,theeffectiveareaofthetransferportsbecomesgreater whentheyareangledmoretowardthemiddleofthecylinderandlesstowardtherearcylinderwall,butwhileachangeintheorderof10degreeswillsometimesyielda boostinpeakpower,thisincreasewillbepurchasedatsomeconsiderablecostinpowerrangeandIcannotrecommendittoanyonewhoisunwillingtoA)buildaflow

patterntestbenchandB) dynamometertestanumberofcylinderswith incrementalrealignmentsoftheirtransferports. Traditionally,thebasementworkshopexperimenter'sscavengingflowtestmethodhasbeentofeedanengineplentyofoil,andinspectthepatternonthepiston crownafteraperiodofrunning.Thatisbetterthannothing,butitprovidesonewithonlyaverycrudepictureofwhatishappeninginsidethecylinderwhiletheengineis running.Bettermethodshavebeendeveloped:Oneisthewateranalogytest,inwhichacylinderisfittedwithaLucitecylinderheadandwaterforcedupthroughits transferportsandouttheexhaustport.Airisinjectedintothewater,insmallbubbles,justbeloweachtransferpassageentrance,andthesebubblesmaketheflow patternsinsidethecylindervisible.Another,lessmessymethodalsoentailsmakingaclearLucitecylinderhead,butinsteadofwaterbeingused,airisaspiratedfrom the exhaustport(withalargevacuumcleaner,ifyouhaveone)andsmokeorfinedyedropletsareinjectedintotheairstreamasitenterseachtransferport.Onlybysuch meanscanyouactuallyseeasymmetricalscavengingstreams,andshortcircuiting,andsuchtechniquesareessentialifyouplananymajoralterationintheportingyour engine'smakerhasprovided. Thereare,asithappens,somebroadrulesconcerningthesizeanddirectionofmultipletransferports.Oneoftheseisthatinafiveport'cylindertheauxiliary transfersshouldhaveanoutletareabetween0.33and0.40thatofthemaintransferports.Unfortunately,nothingasclearlydefinedexistsasregardstheupwardtiltofthe transferports.Inmostengineshavingrotaryintakevalvesandanauxiliaryportorportsontherearcylinderwall,themaintransferswilldischargeatanglesrangingfrom straightacrossthecylinder,90degreesfromthecylinderwall,toasmuchas15degreesofupwardtilt.Thethirdport(orports)willalwaysbetiltedupward,butthiscan rangefromaslittleas45degreestoasmuchas60degrees.Maximumpowerisobtained,inmostengines,withtheleastamountofupwardtiltingasthismakestheport windowseffectivelylargerbutpowerrangetendstosufferastheupwardbiasinthescavengingstreamflowisreduced.Thus,inamotocrossengine,oneshouldstart withaconsiderableamountofupwardbiasandreducethisuntiltheengine'spowerrangevergesonbecomingtoonarrowtheoppositeapproach,thatofstartingwith flatportsandaddingepoxyfillertotheportroofstoprovideupwardbias,mightmoreprofitablybeappliedinroadracingengines.Ineitherinstance,changesshouldbe madeinonedegreeincrementsuntiltheoptimumisfound.

TherewouldseemtobenoruleatallindeterminingtheupwardanglesforthekindofpairedtransfersYamahacurrentlyemploys.TheTD2roadracingengine hasitsmaintransferportsangledupward15degrees,whilethesmaller,auxiliaryportsdischargestraightacrossthepistoncrown.Buttheotherwiseverysimilar,350cc TR2engine'smaintransfershaveazeroupwardangle,andtheauxiliariesangledupwardverysharplyat45degrees.Yamaha'slatestroadracingengine,the350ccTR3, combinesthese,withitsmaintransfersdirectedupwardabout15degreesandtheauxiliariesevensteeperatabout45degrees.Inallcases,bothmainandauxiliaryopen atthesametimeintheYamahaengines,althoughthisisnotuniversalbyanymeans.Somedesignershaveobtainedbestresultsbyeitheradvancingordelayingslightlythe openingoftheauxiliarytransfers,withrespecttothemaintransferporttiming,andtheoldGPSuzuki50ccsinglehaditsthirdportopenbeforeitsexhaustportwhich causedcrankcasepressuretoriseverysharplyjustbeforetheexhaustblowdownandsubsequenttransferphase.Ofcourse,thisratherradicalmethodofimproving mixturedeliverywaseffectiveonlyoveranextremelynarrowpowerrange,andeventhenitmusthavehadotherseriousshortcomingsforneitherSuzukinoranyoneelse

hasitsmaintransferportsangledupward15degrees,whilethesmaller,auxiliaryportsdischargestraightacrossthepistoncrown.Buttheotherwiseverysimilar,350cc TR2engine'smaintransfershaveazeroupwardangle,andtheauxiliariesangledupwardverysharplyat45degrees.Yamaha'slatestroadracingengine,the350ccTR3, combinesthese,withitsmaintransfersdirectedupwardabout15degreesandtheauxiliariesevensteeperatabout45degrees.Inallcases,bothmainandauxiliaryopen atthesametimeintheYamahaengines,althoughthisisnotuniversalbyanymeans.Somedesignershaveobtainedbestresultsbyeitheradvancingordelayingslightlythe openingoftheauxiliarytransfers,withrespecttothemaintransferporttiming,andtheoldGPSuzuki50ccsinglehaditsthirdportopenbeforeitsexhaustportwhich causedcrankcasepressuretoriseverysharplyjustbeforetheexhaustblowdownandsubsequenttransferphase.Ofcourse,thisratherradicalmethodofimproving mixturedeliverywaseffectiveonlyoveranextremelynarrowpowerrange,andeventhenitmusthavehadotherseriousshortcomingsforneitherSuzukinoranyoneelse hasemployeditinrecentyears. Nowthatreedintakevalvinghascomeintofashion,therewillbealotofsixthandseventhportingdonetovariousengines.Therearetwodifferentapproaches toaddingextrarearcylinderportsinreedvalveengines:ThatchosenbyYamahaissimplytocarveaslotleadingupfromtheintakeporttolinktheportcavitywiththe cylinderwhenthepistonisdownatthebottomofitsstroke.Thissixthportisnotfedfromthecrankcasemixturemovesupthroughitandintothecylinderduetothe rammingactionofgasinertiaintheintakemanifold,andalsobecauseofthepartialvacuumformedinthecylinderbyexhaustsystemdynamics.Yamahaplacesthetopof thisextratransferportsonalevelwiththefouralreadyfoundintheirfiveportengines,andtheroofoftheportisangledupwardverysharplyabout70degrees,infact. Apparently,suchtuningasisdonewiththiskindofportismostlyamatterofvaryingitswidth,butnotenoughexperimentalworkhasbeendonewithittoprovideus withanyguidelines.Theotherapproachtothisparticularkindoftransferportistocutwindowsintherearofthepistontopermitpartofthechargecompressedinthe crankcasetoblowupintotheintakeport,fromwhereitisthenabletoescapethroughtheboostportandintothecylinder.Fromearlyreports,theeffectofthis modificationistoimproveverysubstantiallyanengine'spowerrangeatsomeexpensetopeakpower.Thisis,ofcourse,preciselytheeffecttobedesiredinalarge displacementscramblerormotocrossengine,buthardlywhatonewouldbelookingforfromaroadracingengine.Myownexperiencewithreedvalvesisstilltoolimited topermitmeanyfirmconclusions,andthepotentialbenefitsoftheportingtheymakepossible,butIaminclinedatpresenttobelievethattheyfall

somethingshortofbeingtheanswertothetwostroketuner'sprayers. Indeed,itmaybethattheirprincipalcontributionistokeepcarburetioncleanoverawiderspeedrangethanispossiblewithpistoncontrolledintakeporting.Iverymuch doubtthatreedvalvingwilleverequaltheordinarypistoncontrolledinductionintermsofmaximumhorsepowerthoughIrecognizethatreedshavetheirusesinengines intendedforgeneral,allpurposemotorcycles. Farmorepromisingasameanofcompletingtheringofportsaroundthebaseofthetwostroke'scylinderisamethodbeingtriedbyAermacchi,inItaly. Engineersthesehavebeenexperimentingwithatallerthanusualpiston,whichprovidesthemwithenoughspaceabovetheintakeportforatrioofveryshorttransfer passagesfedfromthreeholesdrilledthroughthebackofthepiston,justbelowthering.Togetroomfortheseholes,withoutpushingthepistoncrownupabsurdlyhigh abovethewristpin,Aermacchi'sengineershavegonetoaflatcrownpiston,withasingleDykespatternringlocatedatitsupperedge.Asimilarmodificationmightbe possibleinotherpistonportengines,withoutgettingintotoomuchspecialmachiningorprohibitivecosts. Nomatterwhatkindoftransfersystemyouultimatelyemploy,rememberthatwhileaddingportareawillimprovebulkflowoftheair/fuelchargeupfromthe crankcase,itwillallbefornothingunlesstheflowiscontrolledafteritentersthecylinder.Inacylinderwithmainandauxiliarytransferports,theupwardpitchofthe auxiliaryportsisusedtoliterallyherd themainscavengingstreamstogetheranddirectthemupintothecylinder.Thiscanbeaccomplishedbyshootingtheauxiliary streamsunderthemainflow,andbendingitupwardorbyaimingtheauxiliarystreamsupwardandusingthemtodragthemainflowalong.Thelatteristobefavored whenitispossible,simplybecauseasIhavestatedrepeatedlytiltingaportupwardreducesitseffectiveareaandreducingtheareaoftheauxiliaryportshaslesseffect onthetotalflowcharacteristics.Ontheotherhand,ifyourenginehasitsexistingtransferportsangledupwardat,say,15degrees,andisconstructedsothatonlyvery limitedspaceisavailableforanextrapairofports,thenitmightbebettertoaimtheauxiliariesstraightacrossthepistoncrownaswasdonebyYamahainitsearly experimentswiththefiveportconcept. Nogreaterfutilityexiststhantoinvestheavilyofone'stimeandmoneyinbuildinganenginewithmirrorsmoothports,thebestavailablepistons,rings,etc.,and thenapproachtheproblemsofcarburetionandignitionarmedonlywithenthusiasm.Itisanobservablefactthatsomeinherentlymediocreengineshavebeenmadeto

onthetotalflowcharacteristics.Ontheotherhand,ifyourenginehasitsexistingtransferportsangledupwardat,say,15degrees,andisconstructedsothatonlyvery limitedspaceisavailableforanextrapairofports,thenitmightbebettertoaimtheauxiliariesstraightacrossthepistoncrownaswasdonebyYamahainitsearly experimentswiththefiveportconcept. Nogreaterfutilityexiststhantoinvestheavilyofone'stimeandmoneyinbuildinganenginewithmirrorsmoothports,thebestavailablepistons,rings,etc.,and thenapproachtheproblemsofcarburetionandignitionarmedonlywithenthusiasm.Itisanobservablefactthatsomeinherentlymediocreengineshavebeenmadeto prevailinracingsimplybecausethemenchargedwithselectingjettingandsparktimingknewtheirbusiness,butneverinthehistoryofthesporthastherebeenanengine sosplendidinitsinternalarrangementthatatuner'sineptitudecouldnotrenderitentirelyineffectual.Asarule,twostrokeenginesarebothmoredifficulttotuneandmore sensitivetofinetuningthantheirpoppetvalvecousins especiallyasregardssparktiming,withsometwostrokeenginesbeingsoextraordinarilysensitivetoignition advancethatashiftofonlyasingledegreeeithersideoftheoptimumwillresultinameasurablelossofpower.Yamaha'sTDseriesracingengines,forexample,mustbe timedplusorminusonly0.0012inchofpistontravelfromthespecifieddistanceofTDC.Theseenginesarealsoverysensitivetosparkintensity,andmagneto maladjustmentsorotherdefectstooslighttoproduceperceptiblemisfiringwillcauseamarkeddeteriorationofperformance.Thetwostrokeengine'sair/fuelmixtureisa somewhatlesstouchymatter,asoverrichmixturestendtocompensatefortheirreducedeffectivenessinburningbyimprovinginternalcooling,butthereisanoptimum heretoo.Findingtheseoptimaforsparkandmixturecanbeanexasperating,timeconsumingchore,butthereisnoescapingthenecessityforputtingforththeeffort. Whilemosttunerswouldagreethatmuchdependsonthebasictuners'arts,theyarenonethelessinclinedtoseekmagicalsolutionstoanyproblemsnotinstantly resolvedbychangingamainjetorreplacingasetofpoints.Sadly,seldomdoesmechanicalexoticamagicboxignitionsandtheliketrulysolveaproblem.Moreoften theymerelyreplacetheexistingdifficultywithanother,ormultiple,evenmoremysteriousproblems.Forinstance,mostengineshaveignitionsystemsentirelycapableof producingsparksatarateadequateforthespeedrangeanticipatedbytheengines'makers,butmaycollapseintoafitofmisfiringathigherrevs.Mosttunerswillassume thattherootcauseofthisdistressisthattheyhavebuiltthemselvessuchanincrediblyvigorousenginethatnothingweakerthanGod'sOwnLightningisenoughtomakeit runproperly,whichsendsthemscurryingawayinsearchofsometransistorized,magneticallytriggeredsystemwithenoughsheerzaptofractureatoms.Oftenasnot,the systemtheygetwillclearupthemisfireandconvincethemthattheyhavedoneaGoodThingeventhoughthenewsystemisimpossibletotime,accuratelyandisproneto sudden,inexplicablefailures.Actually,thefundamentalreasonforthemisfiringprobablywasthatatthehigherspeeds,thepointcamprofilewastoosuddenforthe breakerpointsprings,leadingtopointfloat.Thepropercurewouldhavebeentomodifythecamprofiletoopenandclosethepointsmoregently,and/orsubstitutethe standardpointssetwithonehavingstrongersprings,insteadofabandoningthestraightforwardmechanicalprobleminfavorofablackboxoftroublesbestleftto electronicsengineers. Detailsvaryaccordingtothefashionofthemoment,butmuchthesameseriesofeventsoftenoccursubsequenttothediscoveryofacarburetionproblem.A persistentmisfirefrequentlyleadspeopleintotheerrorofassumingthatthecarburetorathandmustbereplacedwithsomethingmoresplendid.Today,themost frequentlychosenreplacementisaMikunicarburetor,andforoncethedictatesoffashionaresubstantiallycorrect,astheMikuniisaveryfineinstrumentandwhenused intheappropriatethroatsizeitcanbemadetogiveexcellentresults.Unfortunately,itisdistinctlypossibletoemployeventheMikunicarburetorbadly,unlessyouknow whatyou'redoingandarepreparedtoworkatitdiligently. THEBASICCARBURETOR Thereisnothingquitesosimple,andatthesametimecomplicated,asacarburetor.Yourdictionarywilltellyou,correctly,thatitismerelyadeviceformixingfuel (inthiscontext,gasoline)withair,andalthoughearlyexamplesemployedseveraldifferentmeanstothisendithaslongsincebeenproventhatthejobishandledbestby directingtheairthroughaventuriinwhichafuelentrynozzleisincorporated.Pressurewithintheventuriisreducedbelowatmosphericindirectproportiontoairvelocity, whichappliesasuctiontothefuelnozzleinthesameproportion.Thus,fuelisdrawnfromthenozzleataratecloselytiedtoairvelocity,andthebasicproportionsofthe combustibleair/fuelmixtureproducedbythecarburetormaybeestablishedbyvaryingthecrosssectionalareaofthefuelentryascomparedtothatoftheventuri.And, inpointoffact,earlycarburetorswerelittlemorecomplicatedthanthedevicejustdescribed,withonlyacontrolledlevelfuelreservoir(thefloatchamber)andsomeform ofthrottlevalveadded. Moderncarburetorsarestillbuiltaroundthebasicventuri,fuelnozzle,reservoir,andthrottle,butvariousdetailshavebeenadded.Themostimportantoftheseis thecorrectionairfeature,whichisneededtocompensatefortheventuri/fuelnozzle'sgreatinherentdefect:Uncompensated,thebasiccarburetorwilldeliveranair/ fuelmixtureinwhichevenhigherproportionsoffuelappearasairvelocitythroughtheventuriincreases.Thereasonforthisisthatthepressuredropwithintheventuriis accompaniedbyareductioninairdensity,andwhileflowfromthefuelnozzleisindirectproportionwithairvelocity,theactualmassofairpassingthenozzledoesnot remaininproportion.Inconsequence,mixturestrengthriseswithincreasesinvelocityunlessmeasuresaretakentopreventthatfromhappening.

 Nogreaterfutilityexiststhantoinvestheavilyofone'stimeandmoneyinbuildinganenginewithmirrorsmoothports,thebestavailablepistons,rings,etc.,and thenapproachtheproblemsofcarburetionandignitionarmedonlywithenthusiasm.Itisanobservablefactthatsomeinherentlymediocreengineshavebeenmadeto prevailinracingsimplybecausethemenchargedwithselectingjettingandsparktimingknewtheirbusiness,butneverinthehistoryofthesporthastherebeenanengine sosplendidinitsinternalarrangementthatatuner'sineptitudecouldnotrenderitentirelyineffectual.Asarule,twostrokeenginesarebothmoredifficulttotuneandmore sensitivetofinetuningthantheirpoppetvalvecousins especiallyasregardssparktiming,withsometwostrokeenginesbeingsoextraordinarilysensitivetoignition advancethatashiftofonlyasingledegreeeithersideoftheoptimumwillresultinameasurablelossofpower.Yamaha'sTDseriesracingengines,forexample,mustbe timedplusorminusonly0.0012inchofpistontravelfromthespecifieddistanceofTDC.Theseenginesarealsoverysensitivetosparkintensity,andmagneto maladjustmentsorotherdefectstooslighttoproduceperceptiblemisfiringwillcauseamarkeddeteriorationofperformance.Thetwostrokeengine'sair/fuelmixtureisa somewhatlesstouchymatter,asoverrichmixturestendtocompensatefortheirreducedeffectivenessinburningbyimprovinginternalcooling,butthereisanoptimum heretoo.Findingtheseoptimaforsparkandmixturecanbeanexasperating,timeconsumingchore,butthereisnoescapingthenecessityforputtingforththeeffort. Whilemosttunerswouldagreethatmuchdependsonthebasictuners'arts,theyarenonethelessinclinedtoseekmagicalsolutionstoanyproblemsnotinstantly resolvedbychangingamainjetorreplacingasetofpoints.Sadly,seldomdoesmechanicalexoticamagicboxignitionsandtheliketrulysolveaproblem.Moreoften theymerelyreplacetheexistingdifficultywithanother,ormultiple,evenmoremysteriousproblems.Forinstance,mostengineshaveignitionsystemsentirelycapableof producingsparksatarateadequateforthespeedrangeanticipatedbytheengines'makers,butmaycollapseintoafitofmisfiringathigherrevs.Mosttunerswillassume thattherootcauseofthisdistressisthattheyhavebuiltthemselvessuchanincrediblyvigorousenginethatnothingweakerthanGod'sOwnLightningisenoughtomakeit runproperly,whichsendsthemscurryingawayinsearchofsometransistorized,magneticallytriggeredsystemwithenoughsheerzaptofractureatoms.Oftenasnot,the systemtheygetwillclearupthemisfireandconvincethemthattheyhavedoneaGoodThingeventhoughthenewsystemisimpossibletotime,accuratelyandisproneto sudden,inexplicablefailures.Actually,thefundamentalreasonforthemisfiringprobablywasthatatthehigherspeeds,thepointcamprofilewastoosuddenforthe breakerpointsprings,leadingtopointfloat.Thepropercurewouldhavebeentomodifythecamprofiletoopenandclosethepointsmoregently,and/orsubstitutethe standardpointssetwithonehavingstrongersprings,insteadofabandoningthestraightforwardmechanicalprobleminfavorofablackboxoftroublesbestleftto electronicsengineers. Detailsvaryaccordingtothefashionofthemoment,butmuchthesameseriesofeventsoftenoccursubsequenttothediscoveryofacarburetionproblem.A persistentmisfirefrequentlyleadspeopleintotheerrorofassumingthatthecarburetorathandmustbereplacedwithsomethingmoresplendid.Today,themost frequentlychosenreplacementisaMikunicarburetor,andforoncethedictatesoffashionaresubstantiallycorrect,astheMikuniisaveryfineinstrumentandwhenused intheappropriatethroatsizeitcanbemadetogiveexcellentresults.Unfortunately,itisdistinctlypossibletoemployeventheMikunicarburetorbadly,unlessyouknow whatyou'redoingandarepreparedtoworkatitdiligently. THEBASICCARBURETOR Thereisnothingquitesosimple,andatthesametimecomplicated,asacarburetor.Yourdictionarywilltellyou,correctly,thatitismerelyadeviceformixingfuel (inthiscontext,gasoline)withair,andalthoughearlyexamplesemployedseveraldifferentmeanstothisendithaslongsincebeenproventhatthejobishandledbestby directingtheairthroughaventuriinwhichafuelentrynozzleisincorporated.Pressurewithintheventuriisreducedbelowatmosphericindirectproportiontoairvelocity, whichappliesasuctiontothefuelnozzleinthesameproportion.Thus,fuelisdrawnfromthenozzleataratecloselytiedtoairvelocity,andthebasicproportionsofthe combustibleair/fuelmixtureproducedbythecarburetormaybeestablishedbyvaryingthecrosssectionalareaofthefuelentryascomparedtothatoftheventuri.And, inpointoffact,earlycarburetorswerelittlemorecomplicatedthanthedevicejustdescribed,withonlyacontrolledlevelfuelreservoir(thefloatchamber)andsomeform ofthrottlevalveadded. Moderncarburetorsarestillbuiltaroundthebasicventuri,fuelnozzle,reservoir,andthrottle,butvariousdetailshavebeenadded.Themostimportantoftheseis thecorrectionairfeature,whichisneededtocompensatefortheventuri/fuelnozzle'sgreatinherentdefect:Uncompensated,thebasiccarburetorwilldeliveranair/ fuelmixtureinwhichevenhigherproportionsoffuelappearasairvelocitythroughtheventuriincreases.Thereasonforthisisthatthepressuredropwithintheventuriis accompaniedbyareductioninairdensity,andwhileflowfromthefuelnozzleisindirectproportionwithairvelocity,theactualmassofairpassingthenozzledoesnot remaininproportion.Inconsequence,mixturestrengthriseswithincreasesinvelocityunlessmeasuresaretakentopreventthatfromhappening.

Allcarburetorsexceptthosewithdiaphragmcontrolledmeteringmakethenecessaryadjustmentfordecreasingairdensitybymeansofacorrectionairsystem,in whichairatatmosphericpressureisdeliveredtothefuelnozzleviadrilledorcastpassages.Initsmoreprimitiveforms,theairissimplyledtoanannularopeningaround thespraynozzleasmaybeobservedinAmalcarburetors.Aspressurewithinthecarburetorthroatdrops,thedifferencesinviscositycauseairflowfromthecorrection airsystemtorisemorerapidlythanfuelflowfromthespraynozzle,andthisdoesmuchtostabilizemixturestrength.Butmuchbettermixturestabilityisobtainedby directingthecorrectionairintoawellbelowthespraynozzle,fromwhichitispulledthroughaseriesofholesinanemulsiontube.Thisemulsiontube,whichmaybe incorporatedaspartofthespraynozzle,takesinfuelatitslowerend(usually)andadmitsthecorrectionairthroughapatternofholesdrilledinitssides.Atlowratesof flow,thewellisalmostentirelyfilledwithfuel,andairpassesonlythroughthetopmostholesintheemulsiontube.Butasoverallflowincreases,thefuellevelinthewell

whichairatatmosphericpressureisdeliveredtothefuelnozzleviadrilledorcastpassages.Initsmoreprimitiveforms,theairissimplyledtoanannularopeningaround thespraynozzleasmaybeobservedinAmalcarburetors.Aspressurewithinthecarburetorthroatdrops,thedifferencesinviscositycauseairflowfromthecorrection airsystemtorisemorerapidlythanfuelflowfromthespraynozzle,andthisdoesmuchtostabilizemixturestrength.Butmuchbettermixturestabilityisobtainedby directingthecorrectionairintoawellbelowthespraynozzle,fromwhichitispulledthroughaseriesofholesinanemulsiontube.Thisemulsiontube,whichmaybe incorporatedaspartofthespraynozzle,takesinfuelatitslowerend(usually)andadmitsthecorrectionairthroughapatternofholesdrilledinitssides.Atlowratesof flow,thewellisalmostentirelyfilledwithfuel,andairpassesonlythroughthetopmostholesintheemulsiontube.Butasoverallflowincreases,thefuellevelinthewell dropstouncovermoreholesandthecorrectionaircomponentofthefrothdeliveredtothespraynozzlebecomeslarger. Byalteringtheholepatternintheemulsiontube,andwithadjustmentstothetotalamountofairbeingadmittedtothecorrectionairsystem,thecarburetor's mixturedeliverycharacteristicsaretailoredtosuitaparticularengine.Itmustbenotedherethatanabsolutelyevenmixturestrengthmaynotbedesirable:Somehigh outputenginesrequirearichermixtureattheirtorqueand/orpowerpeaksthanatotherspeeds,whichmeansthatthecorrection airsystemshouldbeadjusted accordingly.Generally,alargemainjetandsmallcorrectionairjetdeliveramixtureincreasinginrichnesswithenginespeed,(asyoumightexpect),withtheopposite beingtrueofsmallmainjetsandlargecorrectionairjets.Butwhiletheoverallslopeofmixturedeliveryisdeterminedbythemainjet/correctionairjetproportions,slight periodsofrichnessandleannessmaybecreatedbyalterationsintheemulsiontube'sholepattern.Forexample,anemulsiontubewithlargeholesatitsupperextremity andsmalleronesfartherdownwilltendtodeliverarichermixtureathigherrevstheoppositeconditionisfoundwhentheupperholesaresmallerthanthoseintherestof theholepattern.And,whentheemulsiontubeisaclosefitinsidethefuelwell,itispossibletomakeafurtheradjustmentwithdifferencesindiameterdownthelengthof thetube:acollar,midwaydowntheemulsiontube,canconstitutearestrictionwithinthefuelwelllargeenoughtobecomeakindofsecondary

correctionairjet.Withmaximumapplicationofthesemixture compensatingtechniques,itbecomespossibletoemployverylargecarburetorthroatsizesrelativetocylinderdisplacement,whichiswhythesophisticatedMikuniisa betterchoicethantheAmalGPpatterncarburetordespitethelatter'sunquestionedadvantageinairflowcapacity,sizeforsize.A35mmAmalGPwillflowmoreairthan a35mmMikuni,butyoucanfita40mmMikunionanenginethatwoulddeveloptheblindstaggerswithanAmalGPlargerthan35mminthroatsize. Othercomplicationsincarburetorconstructionhavebeenintroducedtocopewithpartthrottleconditions.Atverylowenginespeeds,forinstance,airvelocity throughthecarburetorwillnotbehighenoughtopickupfuelfromthemainspraynozzleandthemixturemustbeprovidedbyothermeans.Typically,idlespeedmixtures willbeproducedbyakindofcarburetorwithinthecarburetor:Underthefloorofthecarburetorthroat,wherethethrottlevalveseats,youwillfindasmallchamber suppliedwithfuelandair,andwithentrytothethroatthroughasingleholedownstreamfromthethrottlevalve,orthroughoneholedownstreamandoneormore upstreamfromthethrottle.Theleastcomplicatedarrangementisoneinwhichthereisonlyasingleholebehindthethrottlevalve,andthechamberisfedairpastan adjustableneedlevalvewhilefuelissuppliedthroughanorificeoffixedsize.Here,theidlespeedmixture(whichisafrothemergingfromtheentryholeinthecarburetor throat)iscontrolledbyvaryingtheamountofairadmittedtothechamber,andtheidlemixturewillbefullrichwhentheneedlevalveisclosed.Othersystemshaveafixed pilotairjetandanadjustablefueljet,whichreversestherich/leanpositionoftheneedlewhilestillothershavefixedairandfueljets,andbulkflowfromtheidlemixture chamberiscontrolledbytheadjustableneedlejet.Thesedetailsarerelativelyunimportanttothetunerwhatisimportantisthattheidlemixturesystemnotonlykeepsthe enginerunningatlowspeedsbutalsohandlesthetransitionbetweenclosedthrottlerunningandthepointatwhichenoughairisflowingthroughthethroattoinitiatethe movementoffuelupfromthemainspraynozzle.

upstreamfromthethrottle.Theleastcomplicatedarrangementisoneinwhichthereisonlyasingleholebehindthethrottlevalve,andthechamberisfedairpastan adjustableneedlevalvewhilefuelissuppliedthroughanorificeoffixedsize.Here,theidlespeedmixture(whichisafrothemergingfromtheentryholeinthecarburetor throat)iscontrolledbyvaryingtheamountofairadmittedtothechamber,andtheidlemixturewillbefullrichwhentheneedlevalveisclosed.Othersystemshaveafixed pilotairjetandanadjustablefueljet,whichreversestherich/leanpositionoftheneedlewhilestillothershavefixedairandfueljets,andbulkflowfromtheidlemixture chamberiscontrolledbytheadjustableneedlejet.Thesedetailsarerelativelyunimportanttothetunerwhatisimportantisthattheidlemixturesystemnotonlykeepsthe enginerunningatlowspeedsbutalsohandlesthetransitionbetweenclosedthrottlerunningandthepointatwhichenoughairisflowingthroughthethroattoinitiatethe movementoffuelupfromthemainspraynozzle. Thetransitionalperiodismanagedbestbyhavingfeedholesbothinfrontofandbehindthethrottlevalve.Withthethrottleagainstitsstop,someairpassesunder thevalveandpicksupthemixedfuelandbubblescomingfromtheidlefeedhole,whileairisdiverteddownthroughtheholeinfrontofthethrottlevalvetomixwiththe fuelintheidlemixturechamber.Butasthethrottleopens,thedepressionexistingbehindthevalvemovesforwardtocovertheupstreamfeedhole,whichmeansthat insteadofairenteringthehole,fuelispulledfromit,andthenetresultisthatthecarburetorisdeliveringenoughadditionalfueltocompensatefortheincreaseinairmoving pasttheopeningthrottle.Thatis,ofcourse,onlythecasewhenallthefuelandairpassagesaretherightsizeandwhentheyare,themixturewillremainattheproper proportionsuntilenoughvelocityisestablishedpastthemainfuelnozzletorelievetheidlesystemoffurtherduties. Carburetorswithbutterflytypethrottlesoftenhaveapatternofholesaheadofthethrottlevalve,andthesearecalled"progression"holes.Asthevalvedisc pivots,anditsloweredgeswingsforward,itmovesaheadoftheprogressionholesinorderofplacement,andeachholethenswitchesfrombeinganairbleedtobecome afueljet.Withtherightpatternofprogressionholes,evenaverylargecarburetor(largeintermsofthroatsizerelativetocylinderdisplacement)canbemadetokeepthe enginerunningwithoutstumblingwhilethetransitionismadetofuelflowfromthemainnozzle.Rarelyareprogressionholes,inthemultiple,foundinslidetype carburetors.Inthese,thetransitionalperiodishandledbytheslidecutawayandthehigherthecutaway,theleanerthetransitionalperiod'smixture. ADJUSTINGMIXTURE Almostcertainly,thecarburetoryouwillbeusingwillhaveacircularslidethrottle,becausethisisthetypemostcommonlyemployed,andmostsuccessful,inthe

fieldofhighoutputmotorcycleengines.Almostcertainly,too,the carburetoryoubuyforyourracingenginewillhavebeenjettedandgivenaslidecutawaysuitableforasomewhatlargerstockengine,whichshouldwarnyouthatamore orlesscompleteretuningoftheinstrumentwillbenecessary.Manytunersbegintheretuningprocessbyfindingthecorrectmainjet,andthatisagoodbeginningunless thereisanundiscoveredproblemwiththemidrangemeteringsystem thelong,taperingneedleclippedtothethrottleslide,andtheneedlejetitself.These,in combination,constituteavariablefuelmeteringvalve,andiftheflowisrestrictedbetweentheneedleandneedlejettoanextentgreaterthantherestrictionprovidedby themainjetrequiredtofeedtheengineatfullthrottle,thennoamountofswitchingmainjetswillgettheenginerunningproperly.Sotheretuningprocessmustalways beginbydeterminingifthereissufficientfuelflowpasttheneedlejettofeedtheengine.Ihavefoundthatthismattercanbesettledverysimplybyloweringtheneedleto itslastnotch,whichmaximizestheflowrestrictionattheneedlejet,andthenremovingthemainjetentirelyfromthecarburetor.Theengineshouldthenrun,ifnone too well,onpartthrottle,butfloodasthethrottleisopenedfully.Shouldtheenginebewillingtorunonfullthrottle,youmaybesurethatalargerneedlejetisrequired. Aftertheselectionofaneedle/needlejetcombinationthatwillpassmorefuelthantheenginecandigest,youthenproceedtotheproblemoffindingthecorrect mainjet.Untilyoubecomereallyexpertintheartof"reading"sparkplugs,therightapproachistostartwithahugemainjetandthenreducethesizeuntiltheenginewill justbarelyrun,onfullthrottle,without"fourstroking".Becauseofthebenefitsincoolingthatareobtainedwithveryrichmixtures,youwillgetverynearthemaximum powertobehadfromahighoutputtwostrokeenginewithamixturethatvergesonbeingsorichthatmisfiringoccurs.Theoptimumusuallywillbefoundwithaslightly leanermixturethanthatborderingonfourstroking,butasthepotentialgainisrathersmall,andtheriskofmeltingapistonisverylarge,leanermixturesshouldbetried verycautiously. Whileyouareexperimentingwithmainjets,themeteringneedlewhichcontrolsthemixturestrengthwhenthethrottleisfromaboutaquartertothreequarters openshouldbeinstalledwithitsclipinthemiddlegroove,orhalfwaythroughitsadjustmentrange.Youultimatelywillprobablychangethissetting,andperhapsswitch toadifferentneedle,butyou'llneedastartingpointfortheadjustmentsrequiredtothethrottlecutawayandtheidlesystem.Starttheseadjustmentsbybackingoutthe throttlestopscrewuntilthethrottleiscompletelyclosed,andthenturnitbackinuntilthethrottleisjustbarelycrackedopen.Havingdonethat,youalsoclosetheidle mixturescrewcompletely,andthenopenittwoorthreeturnsbeforestartingtheengine.Theobject,injugglingthetwoadjustments,istokeepworkingwiththeidle mixtureadjustmenttoincreasetheidlespeedwhiledroppingtheidlebybackingoffonthethrottlestop.Eventuallyyou'llarriveatthelowestthrottlesettingatwhichthe enginewillidlesatisfactorily,andthecorrectmixtureatthatthrottleopening,unless,ofcourse,theidlesystemiswronglyjetted.Oncarburetorswithan"air"adjustment, andafixedjetfeedingfuel,you'llknowthatthefixedjetistoosmallifenginespeedcontinuestorise(atafixedthrottlestopsetting)untiltheadjustmentscrewisturnedall thewayintotheclosedpositionatsomepointinthisprocesstheidlemixtureshouldbecomeoverrich,andifitdoesnotthefueljetistoosmall. Theoppositeis,ofcourse,truewhenenginespeedcontinuestoclimbastheidlemixtureadjustmentscrewisopened,withoutanoptimumeverappearing.A similar,butopposite,ruleappliesforidlesystemswithfixedairjetsandanadjustmentforfuelflow.Icannotgiveyoualistingofwhichmanufacturersusewhattypeofidle system,asthisvariesevenbetweencarburetorsofasinglemake.Tobecertain,you'llhavetotakeyourcarburetorapart(orcheckthemaker'sliterature,ifavailable)to seewhetherthemixtureadjustmentisforfuelorair. Throttlecutawaywillbethenextmatterforyourattention,andthisfactoralmostexclusivelyconcernswhathappensduringthefirsteighthofthrottleopening.Itis

similar,butopposite,ruleappliesforidlesystemswithfixedairjetsandanadjustmentforfuelflow.Icannotgiveyoualistingofwhichmanufacturersusewhattypeofidle system,asthisvariesevenbetweencarburetorsofasinglemake.Tobecertain,you'llhavetotakeyourcarburetorapart(orcheckthemaker'sliterature,ifavailable)to seewhetherthemixtureadjustmentisforfuelorair. Throttlecutawaywillbethenextmatterforyourattention,andthisfactoralmostexclusivelyconcernswhathappensduringthefirsteighthofthrottleopening.Itis

possibletohaveatoolowcutawayonthefrontedgeofthethrottleslide, butacarburetorintendedforsomebigenginewillalmostalwayshavetoomuchcutawayforonewithasmallercylinderdisplacement.Thetoohighcutawayproblemwill bemanifestedinatendencyfortheenginetocoughanddiewhenthethrottleisopened,andthecuresimplyisanewthrottleslidewithlesscutaway.Whendealingwith asinglecylinderengine,youcanbuyareplacementslidewiththelowestcutawayoffered,andfilethebevelhigheruntiloffidlerunningisclean.Thesameapproachmay alsobetakenwithmultiplecylinders,butitisverydifficulttogetthecutawaysmodifiedexactlythesameandiffinancespermityoushouldjustbuyaselectionofslides. Incidentally,atoolowcutawaywillmaketheenginesurgeandburbleatoneeighththrottleorlessandifyouareforcedtomakeaverylargechangeincutaway,you'll havetostarttheidlesystemtuningprocessoverfromthebeginning.Probablythebestsequenceinoverallcarburetortuningistobeginwiththemainjet,thenrough adjusttheneedleandneedlejet,afterwhichthecutawayandidlejettingaremanagedmoreorlesssimultaneouslyasitisalmostimpossibletoseparatethemcompletely. Factoriesdevotemonthstofindingpreciselytherightmeteringneedle,becauseatouringmotorcyclespendsmostofitslifebeingrunsomewherebetweenone andthreequarterthrottle.Fortunately,considerationsoffueleconomyarestrictlysecondaryinracing,soyouneednotspendmonthsswitchingneedlesandneedlejets, butaracingmotorcycleismucheasiertoridewhenitatleastrunscleanlyonpartthrottleandyou

willhavetomakesomeeffortinthisdirection.Ifthereissurgingand stutteringatsteadythrottlewithintherangecontrolledbytheneedle,thenthemixtureistoorichandtheneedleshouldbelowered.Anenginethatrunsfairlycleanlyat steadythrottlebutstumblesandhesitatesasthethrottleisopenedfartherissufferingfrommidrangeleannessandthecarburetor'sneedleshouldberaised.Sometimes youwillgetbothsymptomswiththesameneedle,withanoverleanconditionatonequarterthrottlechangingtobecomeoverrichasyouapproachthreequarters throttle.Thatshouldtellyoutheneedle'staperiswrong,beingtoosteep,whichmeansthatnneedlewithamoreshallowtaperwillberequired.Obviously,theopposite mayalsobeencountered.Unhappily,theseneedlesareexpensive,butthereisnosatisfactoryalternativetobuyingaselectionandtryingthemuntiltherightoneisfound. Atonetimearemotemountedfloatchamberwasthesinequanonofracingcarburetors.Commonly,carburetorswereboltedrigidlytotheenginestheyfedand floatchambersweremountedinrubbertoisolatethemfromtheworsteffectsofenginevibrationwhichcancausesuchextremefrothingofthefuelthataccuratemetering isimpossible.Butwhileremotemountingmayhavesolvedoneproblem,itintroducedanother:Accelerationandbrakingcausedasurginginthelongfuellineslinkingthe carburetorandfloatchamberwhichalsoupsetmetering.Theanswer,allalong,wastoincorporateeverythinginonecastingandmountthewholecarburetor/float chamberassemblyinrubber.Thedifficultyinthatwasfindingarubbercapableofwithstandinga

floatchambersweremountedinrubbertoisolatethemfromtheworsteffectsofenginevibrationwhichcancausesuchextremefrothingofthefuelthataccuratemetering isimpossible.Butwhileremotemountingmayhavesolvedoneproblem,itintroducedanother:Accelerationandbrakingcausedasurginginthelongfuellineslinkingthe carburetorandfloatchamberwhichalsoupsetmetering.Theanswer,allalong,wastoincorporateeverythinginonecastingandmountthewholecarburetor/float chamberassemblyinrubber.Thedifficultyinthatwasfindingarubbercapableofwithstandinga

lotofheat,andconstantexposuretogasoline,withoutdisintegratingand arubberthatcouldbebondedtoasteelmanifoldflangetoavoidtheinconvenientspacerequirementsoftheoldhoseandclampsassembly.Advancesinsynthetic rubberseventuallybroughtustherightmaterial,withtheresultthatnearlyallJapanesemademotorcyclesnowhavetheircarburetorsheldinrubbersleevesthatdoubleas stubmanifolds,andtherestoftheworld'smotorcyclemakersfollowJapan'sleadwhenpossible.Withthesemanifoldsleevesnowavailableinavarietyofsizes,thereis littlereasonforanyoneusingarigidcarburetormounting,andtheequallywideavailabilityoftheMikunicarburetorallowsevenlessreasonforemployingsuperannuated instrumentsliketheAmalGP.Diaphragmtypecarburetors,likethoseuniversallyfoundonkartengines,arerelativelyindifferenttovibrationandneednotberubber mounted,butalltherestbenefitfrombeingisolated.Evenwhennoobvioussymptomsofmixtureinstabilityappear,youmaybesurethatthetendencyisthereandof coursevibrationwillcausearemarkablyrapidwearingofthrottleslides,needleandneedlejet,andeventhefloatvalve.IwouldalsounhesitatinglyrecommendtheMikuni carburetoras,forthemomentatleast,itdoesthebest,mostdelicatemeteringofanyreadilyavailablemotorcyclecarburetor.AnotherattractionoftheMikuniisthatitis suppliedinmanydifferentsizes,andbackedbyanexcellentselectionofjets,slides,needles,etc.Finally,theMikuniismuchbetterstreamlined,internally,thanitappears, andhasagreaterairflowcapacity,sizeforsize,thanalmostanyothercarburetor.Consequently,thereislittlereasonforchoosinganythingbuttheMikuniunlessyouare veryshortofmoneyandforcedtotakewhatyoucanget. Itispossiblethatoneofthediaphragmtype"kart"carburetorsmayofferadvantagesinsomespecializedapplications,andIknowpeoplewhoclaimthatgood resultsareobtainablewiththingsliketheCVseriesKeihin(the"constantvacuum"carburetorfoundonHonda'sCB350andCB450).However,oneshouldremember thattheKeihinCVisdesignedtoimprovethebroadrangeperformanceoffourstrokeenginesandinherentlypoorlysuitedtothetwostrokeengine'squickgulpintake characteristicswhichmeansthatitissomethinglessthanaperfectchoiceforthelatter,intermsofsheerpower,eventhoughitmightshowsomeadvantageonatrials engine.Similarly,thediaphragmtypecarburetorwasinvented,andstillismostwidelyused,toovercometheordinaryfloatchamber'sinabilitytofeedfuelatsteeply angledorinvertedpositions.Thereisnothinginitsmakeuptorecommenditwhenyouarelookingforpurehorsepower,andIwouldconsidermostdiaphragmtype carburetorsagoodchoiceonlyinapplicationswherethereissomuchjoltingandjouncinginvolvedthataconventionalfloattypeinstrumentcannotfunctionnormally.The singleexceptionhereisthepressurepulsecarburetordevelopedbyMcCulloch,whichemploysthecrankcase'spumpingactiontometerfuelflowinsteadofaventuri. Thepressurepulsecarburetoristhuscapableofmeteringsatisfactorilyevenatextremelylowairflowrates,andprovidesgoodthrottleresponseandabroadrangeof powerwhenusedinrelativelyenormoussizes.Intheoriginalapplication,McCulloch's100ccand125cckartengineswerefittedwithapressurepulsecarburetorhaving a1.375inchthrottlebore,whichismuchlargerthanwouldhavebeenpossiblewithaconventionalcarburetor.Subsequentdevelopmentworkshowedthataslight constrictionwasnecessarytocreateaventurieffectthatwouldcompensateforthiscarburetor'stendencytowardhighspeedleanout,butitstilliscapableofcombining remarkablepeakpowerwithabroadeffectiverange.Thesingledifficultywithpressurepulsecarburetorsisthattheyareextremelysensitivetobothcylinderdisplacement andcrankcasecompressionratios,workingwellonlyonenginesforwhichtheyarespecificallydesigned.Theirairpassages,whichbleedcrankcasepressureintoand awayfromachamberbehindthemeteringdiaphragm,havecalibratedorificesandanyvariationintheconditionsanticipatedbythemakerrequireacompleterecalibration. Itisnotajobforamateurs. Selectingcarburetorthroatsizeisenormouslydifficult:Thefourstrokeengine'scarburetormaybechosenthroughrelativelyuncomplicatedconsiderationof cylinderdisplacementandoperatingspeed,butinthetwostrokeengine'scasethereisanaddeddifficultyintroducedbythequickgulpintakecharacteristicandbythe overridingimportanceofpulsationsintheintaketract.Engineswithsmallishintakeportsandrelativelylongintakeperiodsrespondbesttosmallcarburetorsthosehaving verywide,lowintakeportswillhaveshorterintakeperiodstoprovidethesamespecifictimeareavalueandneedabiggercarburetorthroatsizeifthrottlingistobe avoided.Frankly,unlessyouintendraisingyourengine'speakingspeedveryconsiderably,itiswisetoretainthestockcarburetorsizeeventhoughyoumaywantto exchangetheoriginalcarburetorforaMikuni.Switchingtoalargecarburetorwillaltertremendouslytheintaketract'sresonantfrequencyandusuallywillrequirethatthe intaketractlengthbechangedtobringthesonicwavemotionsbackintophasewiththeintakeperiod.Otherwise,thelossofintakerameffectwillmorethanoffsetany gainsobtainedthroughthelargercarburetor'saddedflowcapacity. Despiteallthedifficultiesintroducedwithamajorchangeincarburetorsize,thechangebecomesanecessityifotherenginealterationsworktoraisepeaking speedsbymorethan,say,15percent:Acarburetorchosenwithapowerpeakat6000rpminmindiscertaintocausesomethrottlingwhenaskedtodelivermixtureat 8000rpm.Theproblemhereis,howdowedeterminethemostsuitablethroatsize?Everyoneknowsthatwhileabiggercarburetorissometimesindicated,"bigger"does notautomaticallybecome"better".Again,Yamaha'sNaitohandNomuraprovideananswer,ofsorts,withahandyformulafordeterminingthrottlebore: WhereDisthrottlebore,inmillimeters Kisaconstant Ciscylinderdisplacement,inliters Nisrpmatpeakpower Forracingengines,theconstant,K,isgivenas0.8to0.9,andifyouknowhowtoworksquarerootproblemsitiseasyenoughtoestablishthroatdiametersfor allengines.ThecatchhereisthatYamaha'sformulaseemsapplicableonlytoroadracingengines.Forexample,theYamahaTR3hasaunitcylinderdisplacementof 173ccanddevelopsmaximumpoweratabout10,000rpm,andifweassumethatK=0.85,then

Kisaconstant Ciscylinderdisplacement,inliters Nisrpmatpeakpower Forracingengines,theconstant,K,isgivenas0.8to0.9,andifyouknowhowtoworksquarerootproblemsitiseasyenoughtoestablishthroatdiametersfor allengines.ThecatchhereisthatYamaha'sformulaseemsapplicableonlytoroadracingengines.Forexample,theYamahaTR3hasaunitcylinderdisplacementof 173ccanddevelopsmaximumpoweratabout10,000rpm,andifweassumethatK=0.85,then D=35.4mm Thus,thecalculatedthrottleborediameterisonlyslightlylargerthanthatactuallyusedontheTR3engine,whichindicatesthattheformulaprobablyisvalidfor mostroadracingpowerunits.ButwhenweapplythesameformulatoYamaha'sDT1,a250ccsingledevelopingmaximumpoweratabout6000rpm,thecalculated throttleborediameteris31mmevenifweuseK=0.8.Inreality,theDT1isfittedwitha26mmcarburetoranddynamometertestshaveshownthatthisengineresponds badlytoa30mmcarburetor,whichmeansthatfortheYamaha"enduro"model,thebestresultsareobtainedatK=0.64.Ontheotherhand,thesimilarDT1MXdoes performwellwitha30mmcarburetor,andthatworksouttobeK=0.69.AndtheSachs125'sdisplacement,powerpeakandcarburetorsizegivesusK=0.91,which demonstratesthattherearefactorsherenotgivenproperrecognitionintheNaitohNomuraformula.Iamconvincedthatthereisa"gulpfactor",consistingofallthe thingsthatinfluencetheshapeoftheintakepulse,(timing,portshape,connectingrodlength,etc.),complicatingthismatter.Someday,Imayhavetimetosolvethe problem.Forthemoment,IcanonlytellyouthatK=0.8isasafevalueforroadracingengines,whilemotocrossandenduroenginesmayneedanythingfromK=0.65 toK=0.9,withrelativelylow,wideintakeports(whichgiveabrief,strongintakegulp)favoringthehighervalues. Afinalnoteoncarburetion:Inalltwostrokeenginesintakepulsesareverystrongandthesonicwaveactivityconsiderable,whichhaseffectsbothgoodand unpleasant.Onthecreditsideisthatthehighamplitudepulsationsdomakeitpossibletoobtainveryhighspecificpowerfromthemechanicallysimplepistonportengine byblockingblowbackduringthesecondhalfoftheintakeperiod.Butthesesamepulsationsalsohaveaterribleeffectonthecarburetor'sabilitytoaccuratelymeterfuel, byleadingalargepartoftheairdrawnintotheenginepastthespraynozzlethreetimes:Airpassesthenozzlemovingintotheintaketract,thenreversesdirectionasa resultofthepulsegeneratedwhentheintakeportchopsshut,andpassesthespraynozzleathirdtimeasthenextintakeperiodbegins.Thismaysoundslightlyimprobable butthereisabsoluteevidenceitishappeninginthefogoffueloneseesdancinginfrontofthetwostrokeengine'scarburetor.Nowifthistriplepassageoccurredatall enginespeeds,noproblemwouldariseunhappily,intaketractresonancesthefundamentalvibrationanditsharmonicsslideinandoutofphasewithchangesinengine speed.Relativelysmallcarburetorstendtodamptheseresonances,andthereforearelesssubjecttovibrationsinmixturestrength,whichmakesthemparticularlysuitable inanyapplicationwhereabroadpowerrangeismoreimportantthanmaximumpower.Thisproblemwithmixturedeliverycaneasilybecomesoseverethatitwillbe impossibletoobtaincleanrunning,especiallyifthecarburetorisinsertedatthewrongpointintheoverallintaketract(whichconsistsoftheport,manifoldifanyandthe pipeconnectingthecarburetorwiththeaircleaneraswellasthecarburetoritself).Maximumairflowwillbeobtainedwiththecarburetorcrowdedclosetotheport window,andanextensiononthecarburetor'sinlettoprovidethecorrecttractlength,butthatarrangementalsogivestheworstconditionsformixturedelivery.Positioning thecarburetorattheintaketract'souterendreducesvolumetricefficiencysomewhat,butprovidesthebestmixturestrengthstability.Connectionstotheaircleanershould beasshortaspossible,butifitisnecessarytoseparatethecarburetorandaircleanerbymorethanacoupleofinches,thepassagelinkingthemshouldbeeitheracone (divergingatleast15degrees)oraparallelwalltubehavingabout400percentofthethrottlebore'scrosssectionalarea.Bothofthesewillprovideessentiallythesame conditionasapure,atmosphericinlet,andpreventsecondaryresonancesthatalsocanupsetfuelmetering. IGNITIONFUNDAMENTALS Ignitionsystemsalwaysseemtocreatedifficultiesforthosewhotunehighoutput,twostrokeengines,andforgoodreason:First,thetwostrokeengine'signition systemmarchesdoubletime,ascomparedwithafourstrokeoperatingatthesamecrankspeed,whichnotonlydoublestherateofwearatthebreakerpointsbutalso raisestheheatinputtothesparkplugs.Secondly,theplugsaresubjectedtoafogofoilbetweenfiringsthatbakesintoalayerofsemiconductivematerialovertheplug insulator.Therefore,thetunerfacesasituationinwhichtheignitionsystem'sabilitytogeneratesparksdeterioratesrapidly,whilethevoltagerequiredtoactuallyfirethe plugsisontheincrease.Difficultiesinevitablyoccur,butthesecanberesolvedusuallywithoutresortingtodraconianmeasuresifthereisaclearunderstandingof ignitionfundamentals. Centraltothefunctionofallignitionsystemsisthehighratiotransformerwecalla"coil",whichcomprisestwosetsofwindingsaroundalaminated,softironcore. Inductioncoilsofthistypeemploythemutualinfluenceofmagnetismandelectricalcurrent:passacurrentthroughacoilofwireandyoucreateamagneticfieldsweepthe magneticfieldthroughacoilandyouwillinduceacurrentinthewire.Sointhecommonignitioncoil,youwillfindprimarywindingstocreateanelectromagneticfield,and secondarywindingsinwhichanelectricalpotentialisinduced.Thearrangementconstitutesakindofelectricaltransmission,inwhich,forexample,12voltsareappliedto about400turnsofprimarywindingsandamplifiedbytheapproximately15,000turnsofwireinthesecondarywindings(astepupratioof375:l)tothe20,000plusvolts neededtoproduceasparkattheplug.With100percentefficiency,thecoilwouldincreasetheprimaryvoltageto45,000voltsinthesecondarywindings(375x12= 45,000)butthatisapracticalimpossibility.Otherpracticaldifficultiesincludethelimitationontheamountofcurrenttheprimarywindingswillacceptwithoutmelting,and thevoltagethatcanbe.inducedonthesecondarysidebeforetheresistanceoftheinternal

insulationisovercomeandshortcircuitsparkingoccursinsidethecoil. Also,therearelimitstotheelectricalloadtheignitionsystem'scontactpointswillcarry,andextremelyhighsecondaryvoltagescancauserapidelectricalerosionofthe

about400turnsofprimarywindingsandamplifiedbytheapproximately15,000turnsofwireinthesecondarywindings(astepupratioof375:l)tothe20,000plusvolts neededtoproduceasparkattheplug.With100percentefficiency,thecoilwouldincreasetheprimaryvoltageto45,000voltsinthesecondarywindings(375x12= 45,000)butthatisapracticalimpossibility.Otherpracticaldifficultiesincludethelimitationontheamountofcurrenttheprimarywindingswillacceptwithoutmelting,and thevoltagethatcanbe.inducedonthesecondarysidebeforetheresistanceoftheinternal

insulationisovercomeandshortcircuitsparkingoccursinsidethecoil. Also,therearelimitstotheelectricalloadtheignitionsystem'scontactpointswillcarry,andextremelyhighsecondaryvoltagescancauserapidelectricalerosionofthe sparkplug'selectrode. Allofthepresentlyusedignitionsystemsfeaturethehighratiotransformer,orcoil,justdescribedthereareseveralwaysoftriggeringthatcoilintoactionand thesecanhaveaconsiderableeffectonignitionperformance.Thesystemmostcommonlyemployedontouringenginesisoneinwhichcurrentisdrawnfromabattery, passesthroughtheprimarywindings,andthenthroughanenginedrivenswitch(the"points")togroundwhichcompletesthecircuit,asthebatteryhasoneterminal connectedtothemotorcycle'sframe,orground.Itshouldbeunderstoodthattheignitionpointsarenormallyclosed,andthatasparkisobtainedwhenthepointsare bumpedopenbythecontactbreakercam.Thereisavoltageriseinthesecondarywindingswhenthepointscloseandcurrentstartstoflowthroughtheprimarysideof thecoil,butduetoaphenomenoncalled,"impedance",themagneticfieldbuildstooslowlytoinducemaximumvoltage.Onlywhenthepointsopen,andtheprimary currentdisappears,doesthefieldmovequicklyenoughincollapsingtoinduceignitionvoltage.Becausetherearesofewprimarywindingturns,comparedwiththe secondaries,thefieldcollapsecreatesaverymuchlowervoltagethere,butitwouldbehighenoughtocauseanelectricalarctoformbetweenthepointcontacts,which haveonlybarelyopened,exceptfortheactionofthecondenser.Thislastitemactsasakindofelectricalreservoir,anditreceivestheinducedprimarycurrentand preventsvoltagefromrisinghighenoughtoformanarcacrosstheopeningpoints.Shouldthecondenserfail,anarcwillbeestablished,andtheeffectisratherlikehaving thepointsclosed,withthecurrentintheprimarywindingsreachingalevelsufficienttochokebackthemagneticfield'scollapseandthusdrasticallyreducethevoltagein thesecondarysideofthecoil. Viewedstrictlyasregardsitssparkproducingability,theoldbatteryandcoilignitionsystemislimited,butisnotnearlyasbadaspresentdayenthusiasmforthe moreelaboratealternativesmaymakeitappear.Itrequiresthatyoucarryabatteryonthemotorcycle,andthatbatterymustberubbermountedtopreventvibrationfrom shakingitsplatesapart,butabatterylargeenoughtoenergizetheignitionsystemforthedurationofaraceislittleheavierthanamagnetoand,moreover,doesnotborrow fromtheengine'spoweroutput.Youmust,ofcourse,keepthebatteryonalowratechargerbetweenracemeetingstopreventsulfationofitsplates,andifyouarethe forgetfultypethatmaybecountedashortcoming.Still,thebatteryandcoilsystem'ssingleseriousdeficiencyissimplythatitdoesnotproducesparksquicklyenough. Bythat,Idonotmeanthatitisslowintermsofthetotalnumberofsparksitwillyieldinagiventimeperiod(itiscapableofmakingatleast10,000sparksperminute), butsimplythatthevoltagetoproduceasinglesparkbuildsratherslowly.Thatcharacteristicismostunfortunatewhencombinedwiththeperpetuallysemifouled conditionofthetwostrokeenginessparkplugs,forifsparkvoltagerisestooslowly,theelectricalpotentialinthecoil'ssecondarywindingswillleakawayacrossthe depositsonthepluginsulatorandvoltagewillnotclimbtothelevelrequiredtoionizethegasesinthegapbetweentheelectrodeandthegroundwire.Inthatinstance,no sparkoccurs,eventhoughthebasicsystemmaybecapableofdeliveringvoltagesof30kv(30kilovolts,or30,000volts),whichcertainlywouldbeenoughtoproduce ignitionwithacleanplug.Indeed,touringenginesrunningonrelativelyleanmixtures,withonlyminimalamountsofoilpassingthrough,andwith"hot"plugsperformvery satisfactorilywhensparkedbythehumblebatteryandcoilsystem.Racingengines,however,mustbefedrichmixtures,lotsofoil,andbefittedwithdecidedlycoldplugs whichmeansthattheyoftenwillrunwithouttheslightestsignofdistressonwideopenthrottle(aconditionthatkeepstheplugsburnedclean)beingsuppliedsparksfrom abatteryandcoilsystem,butlapseintoafitofmisfiringthefirsttimethethrottleisclosed. Amagnetoisthemostcommonlyappliedalternativetothebatteryandcoilignitionsystem.Manypeopleassumethatthesinglevirtueofthemagnetoisthatit

satisfactorilywhensparkedbythehumblebatteryandcoilsystem.Racingengines,however,mustbefedrichmixtures,lotsofoil,andbefittedwithdecidedlycoldplugs whichmeansthattheyoftenwillrunwithouttheslightestsignofdistressonwideopenthrottle(aconditionthatkeepstheplugsburnedclean)beingsuppliedsparksfrom abatteryandcoilsystem,butlapseintoafitofmisfiringthefirsttimethethrottleisclosed. Amagnetoisthemostcommonlyappliedalternativetothebatteryandcoilignitionsystem.Manypeopleassumethatthesinglevirtueofthemagnetoisthatit

eliminatesanyneedforcarryingabatteryonthemotorcycle,butinthe contextofracingthataspectismerelyaminorconvenience.Afarmoreimportantjustificationformagnetoignitionisthatitprovidesafarmorerapidvoltagerisethana batteryandcoilsystem.Racingplugshaveaveryshortinsulatornose,whichmeansthattheyalsohavepoorresistanceto"wet"foulingthekindofsparkvoltageshort circuitingthatoccurswhentheplugnoseiscoatedwithsootandrawgasoline.Thismixtureisn'twhatyou'dcallagoodconductorofelectricity,butitisgoodenoughto drainofftheslowrisevoltagedeliveredbybatteryandcoilignitionsystem.Thecriticalvoltagerisetimeseemstobeabout45microseconds:iftheelectricalpotential acrossthesparkgapreaches15kvin40microsecondsorless,misfiresduetowetfoulingrarelyoccur.Andagoodmagnetodeliversmaximumvoltagetotheplug(a maximumthatmayeasilyexceed30kv)inonly3540microsecondsafterthesystemistriggered. Theexplanationforthisfasterriserateisfoundinthemannerinwhichmagnetosmakesparks:Energyissuppliedbyanenginedriven,rotatingmagnet,which sweepsitsfieldthroughacoil,orcoils,togeneratetheoperatingvoltage.Insomemagnetosallofthecoilsarecontainedwithinonehousing,butmostmotorcycleshave "energytransfer"magnetos(originallydevelopedforhighaltitudeaircraftengines).Thesehavelowtensiongeneratingcoilsconnectedinparallelwiththebreakerpoints andthesparkcoil.Thepoints,andthephasingoftherotatingmagnetsrelativetothegeneratingcoils,aretimedtogether,forthepointsmustopenwhenthemagneticflux isstrongest.Withthepointsclosed,andthemagnetorotorswingingintopositionnexttothegeneratingcoil,therotor'smagneticfieldsweepsthroughthecoilandinduces astrongcurrent.Thiscurrentis,ineffect,shortcircuitedthroughthepoints,sothatthewireinthecoilpassestheelectricityaroundinacontinuousloop,andthatflowof currentmakesthecoilintoanelectromagnetwithafieldpolarityoppositethatoftherotor.Soyouhavetwomagneticfieldspushingagainsteachother,andthatcondition persistsuntilthepointsopen.Thenthecurrentinthegeneratingcoilisinterrupted,anditsmagneticfieldcollapses,whichpermitstherotor'sfieldatthatmomentforced backlikeaspring tosnapthroughthecoil.Theextremelyrapidfluxchangeinducesanextraordinarilystrongvoltagesurgeinthegeneratingcoil,andthisis communicatedtotheprimarywindingsofthesparkcoilwhichrespondsbyproducingenoughvoltagetofirethesparkplug.Substantiallythesameprocessoccursinthe traditionalmagneto,inwhichthegeneratingcoilisalsothesparkcoil'sprimarywinding,andisoverlayedbythehightensionwindings. TheprincipalreasonsforusingtheE.T.magnetoarethatA)thesparkcoilisremovedfromdirectproximitywiththeengine,whereitfrequentlybecomestoohot foritsownsurvival,andB)itallowsyoutokeepthelengthofthehightensionsparkleadtoaminimum.Thislastisespeciallyimportant,forwhileelectricalresistanceis

unimportanthere,ignitionleadshavecapacitanceandinductance propertiesthatadverselyeffectvoltagedeliveredtothesparkplugandshouldthereforebekeptasshortaspossible.Byandlarge,magnetosarcveryreliable,butwhen troubleoccurs,lookforeitherpitted(ordirty)points,afaultycondenser,ormorerarelyaweakcoil.Especially,makeeveryefforttokeepthepointsclean,smooth, andadjusted.Thecontactsurfaceshavetocarryaconsiderableelectricalload,andiftheybecomeburned,orfilmedwithoil,sparkvoltagedropsverysharply.Afterall, thevoltageinducedinthesparkcoilisinthefinalanalysisaproductofthefluxchangedownatthefieldcoil,whichisinturnlargelyafunctionofthecurrentstrengthinthe fieldcoiljustpriortothepoints'opening.Oilfilmsandpittingincreasetheresistanceacrossthepointsurfacesandreducethecurrentstrength.Condensersfailmostly becausethecondenserusuallyistuckedawayinahousingontheenginenexttothepoints.There,theybecomeoverheated,theirinternalinsulationsoftens,andshort circuitingfollows.Inmyopinion,aracingengine'scondenser(s)shouldalwaysbemountedexternally,eventhoughthatarrangementmaynotbevisuallytidy.Iprefer placingthecondensersnexttothe,sparkcoils.Allofthemagneto'selementsareconnectedasaparallelcircuit,soyoucansimplyattachthecondenserleadstothespark coil'sinputterminal.Imightmentionhere,too,thatitisunlikely'thatyouwillfindacondensertoolargetogivegoodresultsonthemagneto.IhaveusedthebigMallory condensersinmanyapplicationsandfindthatthereoftenisanimprovementinignitionperformancewiththeaddedcapacitance.Why?Probablybecausethelarger

thevoltageinducedinthesparkcoilisinthefinalanalysisaproductofthefluxchangedownatthefieldcoil,whichisinturnlargelyafunctionofthecurrentstrengthinthe fieldcoiljustpriortothepoints'opening.Oilfilmsandpittingincreasetheresistanceacrossthepointsurfacesandreducethecurrentstrength.Condensersfailmostly becausethecondenserusuallyistuckedawayinahousingontheenginenexttothepoints.There,theybecomeoverheated,theirinternalinsulationsoftens,andshort circuitingfollows.Inmyopinion,aracingengine'scondenser(s)shouldalwaysbemountedexternally,eventhoughthatarrangementmaynotbevisuallytidy.Iprefer placingthecondensersnexttothe,sparkcoils.Allofthemagneto'selementsareconnectedasaparallelcircuit,soyoucansimplyattachthecondenserleadstothespark coil'sinputterminal.Imightmentionhere,too,thatitisunlikely'thatyouwillfindacondensertoolargetogivegoodresultsonthemagneto.IhaveusedthebigMallory condensersinmanyapplicationsandfindthatthereoftenisanimprovementinignitionperformancewiththeaddedcapacitance.Why?Probablybecausethelarger capacityreducesevenfurthertheslightarcingatthepoints,andgivesacleanerterminationofthefieldcoilcurrent. Sheermismanagementprobablyisthemostpotentsourceofallproblemsencounteredwithmagnetos.Asnotedearlier,themagneto'sperformanceisstrongly dependentonhavingthepointsbreakrightatthemomentofmaximumflux,whichmeansthatwhenyouaretimingthepointstodeliverasparkatthecorrectmomentin termsofcrankshaftrotation,youmustalso"time"themagnetosrotorandstator.Usuallytherewillbeanindexingmarkontherotortobealignedwithasimilarmarkon thestatorotherwiseyouwillneedamanufacturersmanualdealingwiththemagnetosconstructiontogiveyouinstructionsasregardstheorientationofrotorandstator. Youcanadaptamagnetofromoneenginetoworkonanother,buttheconversionisn'talwayseasy:Onedifficultyarisesbecausetheairgapbetweentherotorandstator (thestatoristhelaminatedcoreforthegeneratingcoil)isverysmall,andwhiletherotorwillcentermoreorlessautomaticallyonthecrankshaftunlesstheremachiningof tapersisincrediblysloppy,gettingthestatormountingconcentricwiththerotorcanbequiteajob.Whateverelseyoudo,justdon'tcompensateformisalignmentby trimmingawayatthestator'spoles,asaverysmallincreaseintheairgapwillcauseamarkedsaggingofthemagneto'soutput.Infact,youshouldn'tevendrillthestator's mountingholesoversizedwhenmakingtheinstallation.Asithappens,themagneticfieldshiftsthatinduceanelectricalcurrentinthemagneto'swindingsdothesamejobin thestator,andthereisthesamecreationofanopposingmagneticfield,onlythisonecannotbecontrolledbythecontactbreakerpoints.Forthatreasonthestator whichcouldmoreconvenientlybemadeinonepiece isassembledfromanumberofthinlaminations.Withtheinducedcurrentdividedandconfined,themagnetic effectsareminimized,butifyoudrillthroughthestatorlaminations,orgrindawayatitspolesurfaces,youbreakthroughthecoatingsofvarnishthatinsulatebetweenthe laminations,establishcontact,andconvertthestatorintoanotherelectromagnet.Ofcourse,thesparkcoillaminationsaresimilarlyinsulated,forsimilarreasons,and cuttingthroughthemalsoispoorpractice.Thesethingsarecommonlydone,outofignoranceoftheconsequenceswhichusuallyarenotsosevereastoberesponsible foroutrightmisfiringbutshouldbeavoidedintheinterestofgettingthebestperformancepossiblefromanygivenignitionsystem.Finally,Iwouldcautionagainstsimply assumingthatboltingacoiltoamotorcycle'sframeautomaticallyinsuresaproperground.OnmanyoccasionsIhavebeenaskedtohelpsolveapersistentand mysteriousignitionproblem,onlytofindthatthesingledifficultywasanincompletecircuit:thecoilhadbeenboltedtotheframewithafinedisregardfortheinsulating propertiesofpaint.Persistentcoilfailureisanothercomplaintfrequentlyheard,particularlywithreferencetotheoldinternalcoilmagneto.Overheatingofthecoilusually isatfaulthere,butyoushouldbeawarethatthosewhomadethecoilplannedforthetemperaturesinvolvedandinternalshortcircuitingusuallywillnotdevelopunless thereisanelectricaloverload.Theoverloadcomeswhenasparkleadislost,andthevoltagefromthecoilnolongerhasapathtogroundviathesparkplug.Ifconditions aresuchthattheelectricalpotentialbetweentheplug'selectrodeandgroundwireionizesthegasestherewhencoilvoltageisat15kv,thenthatisashighasthevoltagewill go,becauseallthecoil'senergywilldischargethroughtheplugandmostofthedischargeperiodwillbeatavoltageverymuchlowerthanthatpresentnsthesparkwas formed.Butifthesparkleadisdisconnectedforanyreason,thecoil'sinternalvoltage,willclimbextremelyhigh,andmaybecomehighenoughtoforcethroughthe insulation.Whenthatoccurs,apermanentpathtogroundisestablishedandmuchofeverysuccessivevoltagesurgewillburnitswayacrossthesamepath.Theresultisa sharplossofvoltageattheplug,andtheurgentneedforareplacementcoil.Heatwillsoftentheinsulationinsidethecoilandmakeshortcircuitingeasier,butmoreoften thannotthetroublewouldhavebeenavoidedhadthetunerbeencarefultokeeptheleadsinplaceandtogroundtheminsomemannerwhenmotoringtheengine(by pushingthebike)withthesparkplugsremoved. "Transistorized"ignitionsarebecomingmorepopular,andinsomecirclesareregardedasthesovereignremedyforallignitionailments.Theyaresomethingother thantheirreputationsuggests.Ordinarybatteryandcoilsystemscanbetransistorized,byusingthepointstovalve"trigger"currenttoatransistorwhichactuallycarries theelectricalload.Thiskindofsystemoffersnoparticularadvantageformostmotorcycleignitionsystems,asitsprincipalvirtueisthatitwillhandletheveryhighprimary currentsencounteredinmulticylinderautomobileengines.Also,thepointswilllastalmostindefinitely,astheyareunderonlyamicroampload.Thebasictransistorized ignitionsystemcanbeembellishedtoincludeamagnetictriggeringdevicethatreplacesthemechanicalcontactbreakerpointsentirely,andtheadvantagehereisthatall problemswithpointfloatareeliminatedandthesparksmaybedeliveredwithgreataccuracy.Adisadvantageisthatthecircuitrybecomesrathercomplicated,andan externalpowersource(abattery)isrequired.Thetransistorizedbatteryandcoilignitionsystemwithmagnetictriggeringiscapableofdeliveringanextraordinarynumber ofsparksperminute,butthecharacterofindividualsparksremainsthesameasisobtainedwithconventionalpointsandIamnotpersuadedthatthistypesystemoffers anyadvantagesintermsofhighspeed,highoutputtwostrokeengines. Thereareadvantageswithothertypesoftransistorizedignitions,ifotherofthesystem'selementsareproperlyarranged.Asinthecaseofthecapacitordischarge system,whichcanbemadetobebatteryoperated,orasamagnetotypedevice.Eitherway,thereisafairlyconventionalsparkcoil,andthesparkisproducedbythe dischargeofthecapacitor'selectricalcontentsintothecoil'sprimarywindings.Usually,thecapacitorwillbechargedtoabout400volts,whichisenoughtoproducea. morethanadequatevoltageatthecoil'soutputlead.Butthecapacitordischargesystem'sbestfeatureisthatitistremendouslyfast,fasterthaneventheconventional magneto:Somethingintheorderof50100microsecondsarerequiredforasingleelectricalpulsefromaconventionalbatteryandcoiltoreachsparkingstrengtha magnetorequires3540microseconds(whichisjustbelowthewetfoulinglevelof45microseconds)butagoodcapacitordischargesystemtakesonlyacoupleof microsecondsinclimbingtofullvoltage,whichmeansthatitissofarbelowthewetfoulinglevelthatitmaybeconsideredvirtuallyimmunetothatdisorder.Moreover,its sparkisdeliveredsorapidlythatfiringwilloccurevenwhentheplugisverybadlyfouledwithoiloranythingelse.Thesingledisadvantageofthecapacitordischarge ignitionsystemisthatitcandeliverthisterrificsparksobrieflythatignitionoftheair/fuelmixtureinthecombustionchamberdoesnotfollow.Magnetos,forinstance, keeptheirsparksgoingforabout0.51.0millisecond,whilethecapacitordischargesystem'ssparkmaylastlessthanatenthofthattime.Ithasbecomecommonpractice toaddaresistorinthecircuittopreventthecapacitorfromdumpingtoofastanddeliveringatoobriefsparktotheplug,buttherestillaresystemsbeingmadewithoutthis featurewhichistoberegretted,ascoldstartproblemswilloccurevenwitheverythingelseinorder,includingafresh,cleanplug.Atoobriefsparkisalso,likelytobe unabletofirethekindofrichmixturesneededbytwostrokeracingengines. Sports/touringmotorcycles,likethe750cc,3cylinderKawasakiusuallyhaveabatterypoweringtheircapacitordischargeignitionsystem.Thebattery's12volt potentialismultipliedtotherequired400voltsbyaspecialoscillatingcircuitandtransformer,andthecapacitorsdumpedintothesparkcoilsviatransistorsofthetype calledSCR(siliconcontrolledrectifiers).TheSCRsareactuatedbyseparatecircuits,whichareinturncontrolledbymagnetictrigger.Conventionalpointswillalsowork here,butmostmanufacturersseemtofeelthataslongastheyhavetoresorttosolidstateelectronics,theymightaswellgoallthewayanddispensewiththebreaker points.Amoresuitablesystem,forracing,isoneinwhichthecapacitor(orcondenser)ischargeddirectlyfromamagnetotyperotor/statorarrangementwithan appropriategeneratingcoil.Byinsertingarectifiertypetransistorbetweenthegeneratingcoilandthecapacitor,voltageistrappedinthelatter,tobeusedinactivatingthe sparkcoil.Withalittlejigglingofcircuitry,itispossibletousethesamerotatingmagnetstochargethecapacitor,andtriggeranSCRintoreleasingthechargetothecoil, andthisisthearrangementyouwouldfindinsidetheencapsulatedwiringofmostcurrent"CDmagnetos"ifyoucouldgetthemapart.Thereisn'tmuchIcantellyouabout theseignitionsystemsexceptthatwhentheyworktheyworkverywell,andwhentheystopworkingyouareobligedtostartreplacingmagicboxesbecauseeverythingin theirinnerworkingsisA)sealedoffsoyoucan'ttouchthetransistors,resistors,etc.,andB)youhavetoknowmoreaboutelectronicstofixthethingsthanmostpeople haveattheircommand.Ihavedonesomeworkinthisarea,butIwouldn'tfeelcomfortabletryingtoinstructyouintheintricaciesofsolidstateelectronicsevenifthatdid notrequiremorespacethancanbegiventhesubjectheremyignoranceistoogreat.OnethingIcantellyouisthatallthemagneticallytriggeredsystemsshouldbe timedusinga"strobe"lightitisimpossibletosetthemaccuratelybyanyothermeans,andthebestofthesesystems,badlytimed,isinferiortotheconventionalbattery andcoil,ormagneto.Propertimingisthefirstrequirementwithanyignitionsystem. Findingtheexactignitionadvanceyieldingbestresultswithagivenengineis,asnotedatthebeginningofthischapter,atimeconsumingchore,butoneworth pursuingdiligently.Actually,unlessyouhavechangedyourengineveryconsiderablyfromthestockspecificationsyouprobablywillfindthatmaximumperformanceis obtainedatasettingverynearthatrecommendedbytheengine'smanufacturer.Generallyspeaking,modificationsthattendtoraiseengineoutputwithoutincreasingthe crankspeedatwhichmaximumpoweroccurswillrequirethatthesparkberetardedslightlyfromthestandardsetting.Majorupwardrelocationsofthepowerpeak usuallyrequireamoreadvancedspark.Iwouldbedelightedtoprovideyouwithauniversallyapplicableruleforpredictingtheexactsparktimingfortwostrokeengines unfortunatelynosuchruleexists.Itisknownthatabout75percentofthecombustionprocessshouldbecompletedbythetimethepistonhasreachedTDC,andthat

andcoil,ormagneto.Propertimingisthefirstrequirementwithanyignitionsystem. Findingtheexactignitionadvanceyieldingbestresultswithagivenengineis,asnotedatthebeginningofthischapter,atimeconsumingchore,butoneworth pursuingdiligently.Actually,unlessyouhavechangedyourengineveryconsiderablyfromthestockspecificationsyouprobablywillfindthatmaximumperformanceis obtainedatasettingverynearthatrecommendedbytheengine'smanufacturer.Generallyspeaking,modificationsthattendtoraiseengineoutputwithoutincreasingthe crankspeedatwhichmaximumpoweroccurswillrequirethatthesparkberetardedslightlyfromthestandardsetting.Majorupwardrelocationsofthepowerpeak usuallyrequireamoreadvancedspark.Iwouldbedelightedtoprovideyouwithauniversallyapplicableruleforpredictingtheexactsparktimingfortwostrokeengines unfortunatelynosuchruleexists.Itisknownthatabout75percentofthecombustionprocessshouldbecompletedbythetimethepistonhasreachedTDC,andthat whentheaveragerateofflamefronttravelcanbefound,calculatingsparkadvancefromtheflamepropagationrateandthedistancefromthesparkplugtotheremote endofthecombustionchamberisasimplearithmeticalproblem.Butconfusionarisesbecauseflamepropagationratesvarysoenormously.Boththetypeandlocationof thesparkplughasaninfluenceinthis,asdoprecombustionturbulenceandtheshapeofthecombustionchamber.Enginespeedalsohasaneffect,andasyourmodified engineprobablydepartsfromtheoriginalspecificationsinalloftheseparticulars,itisessentialthatyoudothecarefultestingnecessaryinfindingthatelusiveoptimum. Starttestingwiththeignitiontimingretardedabout5degreesfromthestocksetting,andthenadvancethesparkintwodegreeincrementsuntilbestresultsareobtained. Dynamometertestingisthebestmeansofverifyingresults,butyoucanalsouseadragstrip,forithasbeenfoundthatwhilesheerridertechniqueisthedeterminingfactor inelapsedtime,themotorcycle'sspeedattheendofthequartermileisalmostexclusivelyafunctionofenginehorsepower.Thus,ifyoufindthatmaximumspeed(which maybetakenfromanaccuratetachometer)isreachedwithasparkadvanceof,say,27degreesBTC,thenthattimingwillbetheoneprovidingmaximumhorsepower. Interestingly,theinfluenceofmixtureturbulence whichincreaseswithenginespeedissostrongthatthesparkadvanceoptimumdoesnotchangemuchevenovera fairlywidespeedrange.Indeed,fromallavailableevidence,theabilityofcombustionchamberturbulencetospeedburningissogreatthatincreasesinpeakingspeeds frequentlyrequireaslightretardingofignition,whileadvancingthesparkmaywellgiveasmallimprovementinpoweratlowerthanpeakenginespeedsattheexpenseof maximumpower. SPARKPLUGS Peoplewhoarerelativelyinexperiencedintheartoftuningracingtwostrokeenginesseemalwaystofallintothetrapoftryingtouseasparkplugthatistoo "cold",and/oronehavingthewrongnoseconfiguration,whichtheythencompensatewithatooleanmixture.Thiskindoferrorprobablystemsfromabasic m i s u n d e r s t a n d i n g o f s p a r k p l u g h e a t r a n g e , a n d t h e r e a s o n f o r h a v i n g m o r e t h a n o n e t y p e o f p l u g . T h e r e a s o n ?

Asithappens,thetemperatureoftheburningmixtureinthecombustion chamberishighenoughtomelttheengine,anditfailstodosoonlybecauseheatiscarriedawayfromthecombustionchamber'swallsfastenoughtopreventthemfrom reachingflametemperature.Thesparkplugreachesmuchhighertemperaturesthanthecylinderheaditself,becauseheatmovingawayfromtheplug'snosemustcrossthe jointformed,bythethreadsontheplugandintheplughole.And,ofcourse,theexposedplugelectrodesareseparatedfromthecylinderhead'scoolingfinsbyavery longheatpath,andinthecaseofthecenterelectrode,bytheplug'sinsulator.Consequently,theelectrodesgetextremelyhot,andthatisverymuchamixedblessing: Fairlyhightemperaturesarerequiredtokeeptheplug'selectrodesandinsulatortipburnedclean,topreventtheformationofsootydepositsthatcanshortcircuitthe spark.Butiftheelectrodesareallowedtobecometoohot,theywillconstituteanindependentanduncontrolledignitionsource.Thatistosay,theelectrodesmay becomehotenoughtoignitetheair/fuelmixtureinadvanceofthespark.Whenthatoccurs,combustionchambertemperaturesclimbevenhigherwhichcausesauto ignitionstillearlierandthatyieldsyethighertemperaturesinanascendingspiralthatquicklyleadstooutrightenginefailure. So,thecorrectsparkplugforagivenengineisonethatstayshotenoughtoburnitselectrodesandinsulatortipclean,butdoesnotreachtemperatureshigh enoughtocauseautoignition(morecommonlytermed,"preignition").Unfortunately,allsparkplugstransferheatfromtheirelectrodestothecylinderheadintowhich theyarethreadedataratefixedbytheirconstruction,whileengineoperatingtemperaturesvaryenormously."Correct"selectionofasparkplugisthereforeamatterof choosingonehavingaheattransferratecompatiblewiththeapplicationathand.Theprocessofselectionislessexactingtodaythanintheearlydaysoftheinternal combustionengine,principallybecausethemicaandporcelainusedasinsulatorshavebeenreplacedbyaluminumoxideceramicsthatarestrongerandmuchbetter conductorsofheat.Otherimprovementsinsparkplugconstructionhavefurtherwidenedplugheatrange,hutnottosuchextentthatanyplugwillperformsatisfactorilyin anyengine.Touringplugshaveextendedtips,whichburncleanbuthaveexposedgroundelectrodesandcenterelectrodes,andtheserespondtoprolongedfullthrottle operationbybecomingincandescentracingplugshaveshort,shieldedelectrodesandwilltoleratehighercombustionchambertemperaturesbutfoulveryquicklyunder startingandidlingconditions.

conductorsofheat.Otherimprovementsinsparkplugconstructionhavefurtherwidenedplugheatrange,hutnottosuchextentthatanyplugwillperformsatisfactorilyin anyengine.Touringplugshaveextendedtips,whichburncleanbuthaveexposedgroundelectrodesandcenterelectrodes,andtheserespondtoprolongedfullthrottle operationbybecomingincandescentracingplugshaveshort,shieldedelectrodesandwilltoleratehighercombustionchambertemperaturesbutfoulveryquicklyunder startingandidlingconditions.

Themostcommonsparkplugconfigurationistheoneinwhichthegroundelectrodeextendsovertheendofthecenterelectrode.Amuchbettertype,inanytwo strokeengineapplication,hasthegroundelectrodereachinginslightlylowertoapproachthecenterelectrodefromitsside.Theimprovement,here,isthattheground electrodeissomewhatshorter,andbeingshorteroffersamoredirectheatpathtotheplugbodywhichmeansthatitislesslikelytobecomewhitehot.Arefinementof thisdesignhasashortbitofplatinumwireinsertedinthetipofthegroundelectrode.Thiswire'sdiameterisquitesmall,anditisusedinconjunctionwithacenter electrodealsoreducedindiameteratitstip,anapparentlyminordifferenceunlessyouconsiderthatmuchlessvoltageisrequiredtoformasparkbetweentwopointsthan betweentwoflatsurfaces.Anordinaryirongroundelectrodecannotbemadepointed,becauseitstipwouldinstantlyoverheatandmelt,butplatinumwithameltingpoint of1774 o C,ascomparedwith1535 o Cforironislesslikelytomelt,hasathermalconductivity18timesgreaterthaniron,andwillnotoxidize.Asaresult,theplatinum electrodesurvivesitshostileenvironmentverywellevenwhenusedinsmalldiameters,andIwouldnothesitatetorecommendtheplatinumtipplugformostracing applications.Atanygivenheatrange,theplatinumplugwillprovidelongerlifeandlesstendencytofoulthananyconventionaltypeplug.Pureracingplugs,whichhave deeplyrecessedinsulatorsandcenterelectrodes,withanextremelyshortgroundelectrodebridgingstraightacrossfromaholethroughthesideoftheplugbody,should beusedonlyasalastresortintwostrokeengines.Racingplugsofthetypedescribedarevery,verypronetowetfouling,oilfoulingandeveryotherkindoffouling possible.Theyareanecessaryevilinhyperhorsepowerroadracingengines,butshouldneverbeusedwhereextendednoseplugswillserve.Notunlessthereissome kindofmajorbreakthroughineithersparkplugorignitionsystemdesign,whichalwaysisapossibility.Iwouldrecommendthatallwhoareseriousaboutkeepingahead ofthepackkeepabreastofdevelopmentsintheseareas.Allthesparkplugmanufacturersareworkingconstantlytogiveusimprovedfoulresistanceandreducedspark voltagerequirements,andtheyareveryaccommodatingtoanyonewhotakesthetroubletowriteandaskforliterature. "Reading"sparkplugs,andtheprocessofselectingcorrectheatrange,fallsmuchmoreintotherealmofartthanscience,anditisanartinwhichonebecomes reallyproficientonlyafterlongandsometimespainfulexperience.Butthereareafewrulesthatmaybeusedforguidancebythosewhohaveyettoacquireexperience orbythemanywhoseexperiencehaslefttheirignorancelargelyundiluted:First,youshouldknowthatitisallbutimpossibletoreadanythingintheappearanceofaspark plugunlesstheenginehasbeencutcleanafterhavingbeenbroughtuptooperatingtemperatureandgivenalongburstofwideopenthrottle.Veryexperiencedtunerswill seethesignsthey'relookingforunderthelayerofsoot,oilandfuelthat'accumulatessoquicklyatidle,buteventheyvastlyprefertoworkwithcleancutplugs.Second. gettherightheatrangebeforeyoutrytorendmixturestrength,andmyrecommendationisthatyoualwaysusethehottestplugtheenginewilltolerate.You'llknowaplug istoohotwhenyouobservesignsofblisteringaroundtheinsulatornose(whichwillalsobescorchedwhite)andontheelectrodes.Atooleanmixturewillalsogiveyoua whitishinsulator,butwillnotusuallyproducetheburned,pittedappearanceoftheelectrodesthatischaracteristicofatoohotsparkplug.Alsolookforsignsofmelting alongthesharpedgesatthegroundelectrode'sendanysharpcornerwillgethotterthanotherareasalongtheelectrode,andtroublewillfirstberevealedthere.Aplug thatistoocoldsimplylooks,andis,wet.Plugsofthecorrectheatrangegethotenoughtoburnawayoil,andsoot,andwillhaveonlydry,browntotandepositsontheir insulatorsafterahardrun.Asnotedbefore,thecorrectmixturestrengthwillbewryslightlyleanerthanthatwhichisjustleanenoughtokeeptheenginefromfour stroking.Howmuchleaner?Notverymuch,anduntilyouhavegainedconsiderableexperiencewithaparticularengineyoushouldnotreducestrengthbelowthejetting thatprovidescleanrunning.Togetabitclosertotheoptimum,Iwatchthefaint,almostinvisibleringofsootthatformsaroundtheelectrodeontheinsulator'snose,and thelightdustingofsootovertheexposedendoftheplugbody.ThereisapointatwhichIsee"justenough"soot,andifthereismoreorlessthanthatIinterpretthe signstomeanamixturethatistoorichortoolean,respectively.AndIcouldn'tbegintotellyouhowmuchsootis"justenoughthatwouldbeliketryingtoexplaina taste,orsound,orsmell.Withexperience,youlearntorecognizewhatitisyou'relookingfor,andthereisnosubstituteforthatexperience.Youwillalsolearnifyou knowwheretolookthatthefaintlightspotontheelectrodesatthesidesofthesparkgapareanimportantcluetoignitionsystemperformance.Whenyouhaveaspot ofaboutthesamediameterasthegroundelectrode'sendshowingonthecenterelectrode,orviceversa,thenyoumaybesurethemagnetoisdoingitsjob.Whenthat spotbeginstofade,orbecomeraggedaroundtheedgesandshrinkingindiameter,theignitionsystemisn'tperformingasitshould.Finally,withexperienceyou'lllearnto giveyourverycloseattentiontoallaspectsofthemundanetaskofselectingjetsandplugs,andsparktiming,becauseinthesethingsyouultimatelysucceedorfailasa tuneralltherestismeremathematics,surgeryandwrenchtwirling.

spotbeginstofade,orbecomeraggedaroundtheedgesandshrinkingindiameter,theignitionsystemisn'tperformingasitshould.Finally,withexperienceyou'lllearnto giveyourverycloseattentiontoallaspectsofthemundanetaskofselectingjetsandplugs,andsparktiming,becauseinthesethingsyouultimatelysucceedorfailasa tuneralltherestismeremathematics,surgeryandwrenchtwirling.