INVESTIGATIONS ON ThE MODElED ShRINK-FITTED JOINTS OF …jmcm.itee.radom.pl/images/pe_2018/pe_1_2018/PE_1_2018_s... · 2018-04-16 · the marine engine, its failure, especially during

QUARTERLYISSN 1232-9312 12018(108)

Journal of MachineC o n s t r u c t i o n and Maintenance

p 33ndash44

Zbigniew SIEMIąTKOWSKI Mirosław RUCKIKazimierz Pulaski University of Technology and Humanities in Radom Poland Faculty of Mechanical Engineeringzsiemiatkowskiuthradpl mruckiuthradpl

Sergiy lAVRyNENKONational Technical University ldquoKharkov Polytechnic Instituterdquo Ukraine lavrkpikharkovua

INVESTIGATIONS ON ThE MODElED ShRINK-FITTED JOINTS OF ASSEMBlED CRANKShAFTS

Key words mechanical engineering shrink fittings assembled crankshafts large size engine

Abstract The paper describes the results of investigations on shrink-fitted joints Models of the marine engine cranks and pivots were made in the scale of 15 considering real conditions of the large-sized crankshafts fabrication The results of ultrasonic measurements of stresses were compared to the temperature distribution during the cooling process of the assembled couplings The conclusion was that the unsteady stress distribution might be caused by the unsteady cooling process The proposed model of the cooling process provided an explanation on how the stresses and deformations appear As a consequence more tight joints may not reveal a better ability to bear a torque load

Badania modeli połączeń skurczowych składanych wałoacutew korbowych

Słowa kluczowe budowa maszyn połączenia skurczowe składane wały korbowe silniki wielkogabarytowe

Streszczenie Przedstawiono wyniki badań nad połączeniami skurczowymi Pomniejszone modele korb i czopoacutew w skali 15 zostały wykonane z zachowaniem rzeczywistych warunkoacutew technologicznych dla okrętowych wałoacutew korbowych Wyniki pomiaroacutew naprężeń metodą ultradźwiękową zostały poroacutewnane do rozkładoacutew temperatur w czasie chłodzenia złożonych połączeń Wykazano że nieroacutewnomierny rozkład naprężeń może być wynikiem nieroacutewnomiernego przebiegu chłodzenia Zaproponowano model procesu chłodzenia wyjaśniający sposoacuteb powstawania naprężeń i odkształceń w połączeniu skurczowym ktoacutere skutkują tym że zwiększenie wcisku nie przekłada się na zwiększenie przenoszonego momentu skręcającego

copy 2018 Zbigniew Siemiątkowski Mirosław Rucki Sergiy Lavrynenko This is an open access article licensed under the Creative Commons Attribution International License (CC BY)

httpscreativecommonsorglicensesby40

Introduction

There are three common technologies for the marine crankshaft fabrication Unlike the fully forgedcrankshaftsproductionoftheassembledandsemi-builtcrankshafts that may be considered as the most common forlargemarinemainenginesinvolvestheshrinkfittingmethod[4]Forexamplethecrankshaft6S35MCtype

isof4700mmlongitscrankradiusis700mmandtheweightis11730kganditisassembledintheverticalposition All of its elements eg crankpin journalmain journal and crankwebs are prepared separatelyand joined together in an appropriate angle positionwith a shrink-fitting technology Figure 1 presents theelementsofsuchacrankshaftandFig2showshowitlooksafterassembling

34 Journal of Machine Construction and Maintenance | 12018

Fig 1 Elements of the assembled crankshaft a) supporting journal b) crank c) main journal d) journal with flange [16]

Fig 2 Scheme of the assembled crankshaft

The crankshaft is a very responsible element ofthe marine engine its failure especially during theoperationinthestormmayleadtoathreattowardsthelivesofshipboardpersonnelandpassengersMoreoverthe industrial design requires knowledge based on the investigationintheareaofnewmaterialsandmethods[1012]andknowledgeconcerningsurfacemorphology[7]BecauseofsafetyreasonsthejointsmustbeabletobeartheincreasedloadTocalculatetheloadingcausingtheshaft-hubdetachmentRadietal[13]proposedananalyticalmodelbasedupontheLameacuteequationandthebeamtheoryIntheirsolutiontheshaft-hubcontactwasmodelled in terms of two elastic Timoshenko beams connected by a distributed elastic spring Its stiffnesswasevaluatedanalyticallyItisknownthatfracturesincrankshaftscanoccurbybending fatigueby torsionalfatigue or a combination of both [5] Moreover inthe assembled crankshafts the hub must be axiallypositioned on the shaft and resist the moments and forces generated bymisalignments [9]However it isknown that the residual stresses and their distribution in mechanical components may cause them to fail at a load level significantly lower than expected [1]Siemiatkowski [16] demonstrated experimentally thatthe relative interference above 22permil did not improvethe bonding strength of the shrink-fitted crank-pivotjoint and Burenin et al [2] proved theoretically thatplasticdeformation in the shrink-fitted couplingundertemperaturestresssignificantlyreduceditstightnessTheeffectoftheshrinkfittingratioswasinvestigatedbyLee

etal[8]andKimetal[6]proposedatechniquebasedonthe3DFiniteElementsmethodfortheshrink-fittingprocedureoptimizationStrozzietal[17]reportedtheresultsofastudyonelasticstressconcentrationZhangetal[20]studiedtheinitiationcharacteristicsoffrettingfatiguecracks in theshrink-fittedshafts toanalyse thewearmechanism

1 Internal stress measurement in the models

There are many methods of the the analysis of shrink fit couplings in different applications [19]but in case of such a large-sized elements empiricalinvestigationandmeasurementisextremelydifficultandexpensive In particular fabrication of the assembledcrankshaftjustfortheresidualstressmeasurementwiththelayerremovaltechnique[3]wouldbefinanciallytooexpensive

Therefore 15 scale models were made in orderto make laboratory measurements possible [16]The shrink fitted couplings were fabricated keepingtechnology as close to the technical conditions of large-sizedcrankshaftsproductionaspossible

Surface topography has a crucial effect on theprecision of the obtained parts [11] Thus roughnessand form deviations of the models were measured inorder to ensure the appropriate relative tightness ofthecouplingsintherangebetween20and25permilThe

Journal of Machine Construction and Maintenance | 12018 35

roughnessofthesurfaceswaskeptbetweenRa=032and040micromExamplesoftheexaminedmodelsofthecrank and the roundness measurements are shown in Fig3Thesurfacesdesignedtobeincontactweredriedanddegreasedbefore theoperationbutno substancesincreasing friction were applied To complete shrink-fittedjointseachcrankwasheatedintheareaarounditsorificeuptothetemperatureof350degCwhilepivotwasleft at room temperatureAfter the pivotwas insertedintothecrankthecouplingwaslefttocoolinambientconditions

Fig 3 An example of the investigated model of the crank (left) and the out-of-roundness of its orifice (right) [11]

The shrink-fitted elements generate internalstresses which were measured using an ultrasonicmeasurement device (Debro) in the laboratory of theInstitute of Fundamental Technological Research(PolishAcademyofSciences)Foreachcouplecrank-pivot and ring-pivotmeasurementsweremadebeforetheywerejoinedtogetherandafterwardsExamplesofmeasurement results for radial stresses in the assembled modelsareshowninFig4

Fig 4 An example of the measured radial stress distribution in the shrink-fitted crank and pivot along the axes [16]


The stress measurement results should be treated asthestressesaveragedalongthematerialthicknessInthe case when they were performed close to the border betweenthedetailstheyindicatedtheaveragealongthejointBothdigitalsimulationandthehole-drillingstrain-gagemethodconfirmedtheunsteadystressdistribution[15]Theextensionof theconcentratedstressover theyield point is the factor most probably responsible for thereducedtightnessoftheshrinkfit

2 Calculations of the shrink fitted joints

The interferences between the pivot and crankwere calculated with two methods analytically as athick-wallpipewithconstantinnerpressureandusingFinite Elements Method (FEM) as a pressure steadily appliedtotheinnersurfaceofthecrankboreTheresultswere obtained in form of stress distribution graphs and deformationdiagrams

The Lameacute equations for the pressure p could be written for the known interference w as follows

(1)

wherew ndash radial interference between the crank bore and

pivotrwcrzc ndashinner and outer radius of the shrink-fitted

orificeandpivotrespectivelyRwkRzk ndash inner radius of the crank bore and outer radius

ofitscylindricalpartrespectivelyE ndashYoungrsquosmodulus

Assuming that the interference w is the difference betweenthedisplacementsofthepivotucz and the one of crank ukitcouldbewritten

w=ucz ndash uk (2)

Thusitcanbewritten

(3)

(4)

where νndashPoissonrsquosconstant

The stresses both radial sr and tangential st in thepivotandcrankarecalculated from the followingequationsa)Forthepivot

(5)

(6)

where risthecurrentradiusofthepivotfromtherangeltrwcrzcgt

b) For the crank

(7)

(8)

where R is the current radius of the crank from the range ltRwkRzkgtwhichenables to calculate the stressdistribution

Based on the vonMisesndashHuberrsquos hypothesis thereducedstressesforthepivotsred(cz) and the ones for the crank sred(k)canbecalculatedfromtheEquations(9)and(10)respectively

(9)

(10)

Intheaboveequationsthefollowingvalueswereinput E = 21times105 MPa rwc = 115 mm rzc = Rwk = 570mmRzk=1100mmν =03andtheinterferencew =01275mm

The obtained results are as follows a) For the maximal radial interference w =01275mm

therelativetightnesswasWr=224permilthepressureinthejointp=167MPathecrankradiusincreaseuk = 0092 mm and the pivot radius decreaseucz=-0035mm

b) For the minimal radial interference w =0117mmtherelativetightnesswasWr=205permilthepressureinthejointp=153MPathecrankradiusincreaseuk = 0084 mm and the pivot radius decreaseucz=-0033mmThegraphofcalculatedpressureintheshrink-fitted

pivot and crank is presented in Fig 5 and the stressdistributionispresentedinFig6

It can be also calculated that the maximal bearable torquefortheshrink-fittedcouplingisasfollows

Msmax=2mpRpLp(11)

whereμ ndashfrictioncoefficientRp ndashnominalradiusoftheshrink-fittingL ndashaxiallengthoftheinterferenceequaltothecrank

thickness

2


Intheexperimentalmodelthetheoreticalpressurewas p = 167 MPa nominal radius Rp = Rwk = rzc = 57 mm contact length L = 48 mm and frictioncoefficientwasmicro=02Thusthemaximaltorquewascalculated as Msmax=327kNm

Figure 7 presents the examples of the stressescompared to the deformations calculated with the Finite ElementsMethod(FEM)andFig8showstheexamplesofthecomponentsofstressesindifferentdirections

Fig 5 Theoretical (Lameacute) pressure distribution in the pivot and crank for the interference w = 01275 mm

Fig 6 Theoretical (Lameacute) stresses in the shrink-fitted pivot and crank for the radius R = 57 mm and interference w = 01275 mm

Fig 7 Reduced stresses sred(k) (left) and deformations (right) in the crank


Fig 8 Components of the stress tensor in the crank in different directions a) XX b) XY c) XZ d) YY e) YZ f) ZZ

FromFigures7and8itcanbeseenthatthecrankgeometrygeneratesaspecifickindofstressduringtheshrinkage and the rise of respective pressure on thecrankboreAsaresultofunsteadyandirregularstressesdeformation takesplace so the axisof the crankboremovesalongtheX-axisThemaximalvonMisesstressescan be seen in the front side of the cylindrical part of thecrankInFig7(left)theplacewherethemaximalreduced stress σredmax = 4673MPa appears ismarkedSimilarlythemaximaldisplacement0122mmisseenin the front side where the maximal stress appeared (Fig 7 right) while the minimal one takes place onthe back side The abovementioned phenomenon iscausedbythevaryingstiffnessofthecrankintheradialdirectionfromthecrankbore

3 The experimental rig for the shrink-fitted joint temperature measurement

Apart from the stress measurements the localtemperaturewasmeasuredintheareaofjointbetweencranks and pivots or rings and pivots respectively In

order to perform the measurement and distribution analysis of the temperature during the cooling process of the shrink-fitted joint a dedicated experimental rigwasmadeTheequipmentmadeitpossibletosimulatethe full assembling cycle of the pivotwith crank ieheatinginsertofthepivotintoboreandthecoolingofthecouplingTheexperimentalsetupisshowninFig9

During the measurement the model crank (1) isplaced on the supports (2) of height projected in thescale15relatedtotherealdimensionsoftheassembledmarinedieselenginecrankshaftFromthebottomsidethe burner is placed to heat the examined element up to 350ordmC

In the special measurement ring (5) eightindependentthermocouplesPTT-K(NiCrndashNiAl)wereplaced (6) They were distributed at 45ordm increments (Fig4 above) and theirmeasurementdiameterswere10mmandthickness1mmThemeasurementringwasplaced in a way that ensured a small pressing force on thecranksurfacearoundtheboreThespecialpadsandsiliconlubricatorprovidedappropriateheattransferandelectricisolationbetweenthermocouples

The thermocouples were connected with the measurement card (DAC-pad 71B) produced by


ldquoAdvantechrdquo(7)Thecardwasinstalledinacomputer(Aristo FT-6000E) equipped with a Pentium 120processor(8)ThemeasurementcardmadeitpossibletoperformtemperaturecompensationThethermocoupleswereadjustedatthetemperatureof0ordmC

Themetalplatecover(9)eliminatedthenoisesthatmay affect the measurement results especially thosecausedbytheairswirlingaroundthemeasurementarea

Fig 9 The experimental rig for the temperature measurement (details explained in the text)

Thesignalsamplingwasmadewithafrequency30kHzThemeasurementinaccuracywasestimatedatthelevelof2onthebasisofthecomparativetestmeasurement

The locations of the temperature measurements and an example of the graphs obtained for each sensor duringthecoolingareshowninFig10

Fig 10 The temperature registered by the sensors T1-T8 during the cooling in the air

4 Measurement results and discussion

The temperature is closely related to the dimensional changes of the heated crank and its bore diameterTheoreticalcalculationwasmadefortheidealassumptionofproportional lineardiameterchangeΔD at the mean temperature obtained from the opposite thermocouples eg T1 and T5 or T3 and T7 Thefollowing formula was applied

∆Rwk=Rwk a ∆T (12)

where Rwkndashcrankboreradiusa ndashlinear thermal expansion coefficient (for steel

S34MnVa=14times10-6[1ordmC])∆T ndash absolute increase of the temperature on the crank

surface


The differences between the theoretical and measured values ofΔD did not exceed 002mmTheresults for the same crank (relative tightness Wr = 225permil) are presented in the Table 1 The problemhoweverappearsbecausetheaxisofthecrankboreis

movingduringtheheatingprocessForexampleforthesamecranktheradiusesrelatedtothethermocouplesT1andT5increaseforthevalues∆Rc(T1)=0226mmand∆Rc(T5)=0265mmrespectivelyTheillustrationsofthethermaldeformationresultingintheaxismovementareshowninFig11

Fig 11 Thermal deformation of the crank bore immediately before insertion of pivot a) for the relative tightness Wr = 225permil b) for the relative tightness Wr = 200permil

Table 1 Increase of the bore diameter during the crank heating process

Heatingtime

t [min]

Mean absolute increase of the temperature around the bore in the crank

Measured increase of the diameter between two

corresponding thermocouples

Calculated increase of the diameter between two


(T1+T5)2[ordmC]

(T2+T6)2[ordmC]

(T3+T7)2[ordmC]

(T4+T8)2[ordmC]

DD1[mm]

DD2[mm]

DD3[mm]

DD4[mm]

DD1[mm]

DD2[mm]

DD3[mm]

DD4[mm]

0 0 0 0 0 0 0 0 0 0 0 0 010 93 98 100 97 014 015 015 015 015 016 016 01520 162 166 168 165 025 025 025 025 026 026 027 02630 207 210 213 211 033 032 032 032 033 034 034 03440 240 243 245 243 039 038 038 038 038 039 039 03950 263 267 268 267 042 041 041 041 042 043 043 04360 279 283 283 282 045 044 044 044 045 045 045 04570 290 294 296 293 047 047 047 046 046 047 047 04780 299 301 303 302 048 048 048 048 048 048 048 04890 304 306 307 307 049 049 049 049 048 049 049 049

As reported in other papers eg Sun et al [18]there is a difference between the heating time of the pin sideandroundsidearoundtheboreInthecurrentstudythemostinterestingobservationofthecoolingprocessis the slight increase of the crank temperature near the boreaftertheinitialfalltothevalueofca210ordmCAsit

isseeninFig10abovesuchanincreaseisindicatedbythe thermocouplesT1T2 andT8placedwhere thereis much more material to be cooled than in case of the otherthermocouplesNophenomenonofthatkindtakesplace in the case of the ring-type detail as shown in Fig12


Fig 12 Graph of the temperature during the shrink-fitting process for the ring and pivot of relative tightness Wr = 225permil

Moreover Fig 12 indicates another problemwith unsteady coolingAfter the pivot is inserted thethermocoupleT2indicatesevidentlyquickercoolingtoca300ordmCthantheotheronesAtca280ordmCthecoolingspeed becomes almost equal but again T5 reaches235ordmC more quickly than other thermocouples Thismaybetheevidenceofnon-axialinsertionofthepivotwhich resulted in increased heat transfer along the main contactlinebetweenthepivotandbore

From Fig 10 an additional explanation can bederived for the deformation mentioned aboveAt thetime when the pivot is to be inserted into the crankbore the difference between the temperature at thepointsT1andT5isca55ordmCThisfactresultsfromthedeformationofaxiseventhoughthemeantemperatureforΔD1isalmostthesameasforotherdiameters(304ordmCcomparedto306and307)Sunatal[18]reportedthatthe displacement of centre of the boremay reach 05mmcoaxiallyitmaybedistortedby04ndash09mmandarotationof0044degmayoccur

The obtained measurement results led to conclusion thattheinevitablyunsteadycoolingprocessisthemainsource of irregular shrinkage and subsequent unsteady distribution of the residual stresses that also result with thedeformationsof theassembledcrankshaftTounderstanditbetterthecoolingprocessmaybedividedtofourstagesasfollows ndash Immediately after the pivot is inserted to the hotcrankboretheheattransfergoesfasterontheoutersurfaces of crank than along the surface inside the boreAsaresult theupperandloweredgesof thebore are shrinking faster and getting in contact with thecoldpivotasshowninFig13a

ndash The contact circles around the pivot providemoreintensive heat transfer than the rest of jointwhereair remains between the hot crank bore surface and thecoldpivot surfaceThepivot receivesheatandexpands so its length and diameter increase andsimultaneouslyitissqueezedbytheshrinkingcrankboreThecontactcirclesgrowintotightringsintheupperandlowerpartsoffitting(Fig13b)

ndash The heat transfer and dynamic gradient of temperature cause deformations of both coupled elementsTheaxialextensionofthepivotafter theappearanceoftight rings results in different contract forces between the pivot and crank boreThemost important anddominant force is the radial interference force pbutaxial friction forces F1 and F2 directed against one anotherinevitablyappearasshowninFig13c

ndash The last stage starts when the pivot reaches itshighest temperature and the coupling is cooled downtogetherTheshrinkingprocessisfinishedandinterferencefitisformedwithitsresidualstressthatprovides the desired bonding strength Howeverthe resulting boundary surface is formed in the way presentedideallyinFig13dhoweverinrealityitis more deformed and also has gaps as reported by otherresearchers[18]These4stagesof thecoolingprocessprovidethe

explanation on how the stresses are being generated in the large size assembled crankshaft and why their concentrationleadstothereducedbondingstrengthThedeformations foundafter the reporteddestructive tests[16]stayinconformitywiththeaboveexplanation


Fig 13 Important stages of the cooling process a) first ldquocontact circlesrdquo between hot crank bore and cold pivot b) appearance of the tight rings c) growth of the shrinking force p and friction forces F d) resulting boundary surface

Conclusions

The present study enabled us to indicate some important problems concerning the shrink-fittingtechnology applied in assembling large-size marine diesel engine crankshafts First of all temperaturemeasurements provided a better understanding of theprocess Secondly residual stress calculations wereperformedaccordingtotheLameacutemethodandusingtheFinite Elements Method Thirdly deformations wereanalysed and theoretical expectations were comparedwith experimental resultsAnd finally amodel of theprocesswasproposedbasedontheobtainedresults

Itwasfoundthatinanyanalysisandprojectworkstheshrink-fittingprocedureshouldbedividedintothreestages heating insertion and cooling In the cooling

process unsteady temperature distribution is causedby the different thickness of the material around the boreTheregisteredtemperaturedifferenceat theboreoppositesidesreached55ordmC

Next the insertion of the pivot into the crankbore causes the rapid fall of the temperature whichmay end with a slight temperature increase before the proper cooling process startsNo such increase of thetemperature takesplace in the ring-pivot shrink-fittingprocessInsomecaseswhenthepivotisinsertednon-axially the temperature may decrease more quicklyalongthemaincontactline

The last phase of the procedure the cooling isresponsiblefortheappropriatejointbetweenthecrankbore and the pivot squeezed in it During the coolingprocessthestressescausethedisplacementofthecrank


boreaxisTheanalysisofstagesofthecoolingprocessgave some explanation on why the resulting surfacebetweenshrink-fittedcrankandpivotisnotcylindricalandtheresidualstressdistributionisunsteady

The presented researches can be the basis for some practicalrecommendationsasfollows

ndash It is not advantageous tomake the joint too tightbecause the unsteadiness of process itself will generate larger deformations making the jointweaker

ndash The deformation and stress distributions should be calculated taking into consideration unsteady heat transferindifferentdirections

ndash Simplified calculations may result in much largerdeformations than expected including the axisdisplacementThe abovementioned processes should be taken

intoconsiderationfromtheverybeginningoftheprojectstageinordertokeeptheaccuracyandreliabilityoftheshrink-fittingsintheassembleddieselmarinecrankshaft

Acknowledgement

Thearticleistheextendedversionofthepaperthatwas originally presented at the conference M2D2017(ref number 6538) AlbufeiraPortugal 11ndash15 June2017

Funding sources

This research did not receive any specific grantfromfundingagenciesinthepubliccommercialornot-for-profitsectors

References

1 Baldi A 2016 Using Optical Interferometryto Restart the Ring-Core Method ExperimentalMechanics561191ndash1202

2 BureninAADatsEPTkachevaAV2014OnthemodelingoftheshrinkfittechnologyJournalofApplied and IndustrialMathematics 8 493 DOI101134S199047891404005X

3 CarpenterHW Reid RG Paskaramoorthy R2014 Extension of the layer removal techniquefor the measurement of residual stresses in layered anisotropic cylinders International Journal ofMechanicsandMaterialsinDesign10(3)269ndash280DOI101007s10999-014-9245-2

4 FonteMDuartePAnesVFreitasMReisL2015On theassessmentof fatigue lifeofmarinediesel engine crankshafts Eng Fail Anal 56 51ndash57DOI101016jengfailanal201504014

5 JadhavAChalwaVGaikwadP2013FatigueFailure Analysis Of Marine Engine CrankshaftInternational Journal of Engineering Research ampTechnology2614ndash621

6 KimHYKimCBaeWBHanSM2007Development of optimization technique of warmshrink fitting process for automotive transmissionparts (3D FE analysis) Journal of MaterialsProcessing Technology 187ndash188 458ndash462 DOI101016jjmatprotec200611107

7 Krolczyk GM Maruda RW Nieslony PWieczorowski M 2016 Surface morphologyanalysisofDuplexStainlessSteel (DSS) inCleanProduction using the Power Spectral DensityMeasurement 94 464ndash470 DOI 101016jmeasurement201608023

8 Lee HC Saroosh MA Song JH Im YT2009Theeffectofshrinkfittingratiosontoollifein bolt forming processes Journal of MaterialsProcessing Technology 209 3766ndash3775 DOI101016jjmatprotec200808032

9 MancusoJR1999CouplingsandJointsDesignSelectionampApplicationMarcelDekkerIncNewYorkndashBasel

10 Maruda RW Krolczyk GM Nieslony PWojciechowskiSMichalskiMLegutkoS2016The influence of the cooling conditions on thecuttingtoolwearandthechipformationmechanismJournal of Manufacturing Processes 24P1 107ndash ndash115DOI101016jjmapro201608006

11 Nieslony P Krolczyk GM Wojciechowski SChudy R Zak K Maruda RW 2018 Surfacequality and topographic inspection of variablecompliance part after precise turning AppliedSurface Science 434 91ndash101 DOI 101016japsusc201710158

12 Nieslony P KrolczykGM ZakKMarudaRWLegutko S 2017 Comparative assessment ofthe mechanical and electromagnetic surfaces of explosively clad Ti-steel plates after drillingprocessPrecisionEngineering47104ndash110DOI101016jprecisioneng201607011

13 RadiELanzoniLStrozziABertocchiE2017Shaft-hub press fit subjected to bending couplesAnalytical evaluationof the shaft-hubdetachmentcoupleAppliedMathematicalModelling50135ndash ndash160DOI101016japm201705018

14 SiemiatkowskiZ 2017Experimental evaluationof the shrink-fitted joints in the assembledcrankshafts Journal of Engineering Technology6(2)832ndash841

15 SiemiątkowskiZRuckiMKudlacekJ2018Internal Stresses Analysis in the Shrink-FittedJointsof theAssembledCrankshafts InHamrolA Ciszak O Legutko S Jurczyk M (Eds)Advances in Manufacturing Springer Poznań pp633ndash640DOI101007978-3-319-68619-6_60


16 SiemiatkowskiZRuckiMLavrynenkoS2017Investigationsoftheshrink-fittedjointsinassembledcrankshafts In Silva Gomes JF Meguid SA(Eds) Proceedings of the 7th InternationalConferenceonMechanicsandMaterialsinDesignAlbufeiraPortugalpp1155ndash1158

17 StrozziABertocchiEBaldiniAMantovaniS2016NormalizationofthestressconcentrationsattheroundededgesofaninterferencefitbetweenasolidshaftsubjectedtobendingandahubMechanicsBased Design of Structures and Machines 44(4)405ndash425DOI1010801539773420151086274

18 SunMYLuSPLiDZLiYYLangXGWangSQ2010Three-dimensionalfiniteelementmethodsimulationandoptimizationofshrinkfittingprocess for a large marine crankshaft Materialsand Design 31 4155ndash4164 DOI 101016jmatdes201004027

19 WangXLiuBGaoKMengQSunY2016Analysis of thermal deformation and influencingfactorsinshrink-fittingassemblyofaluminumalloydrill pipe Advances in Mechanical Engineering8(10)1ndash15DOI1011771687814016674099

20 ZhangY Lu LGongY Zhang J ZengD2017 Fretting wear-induced evolution of surfacedamage in press-fitted shaftWear 384-385 131ndash ndash141DOI101016jwear201705014


Fig 1 Elements of the assembled crankshaft a) supporting journal b) crank c) main journal d) journal with flange [16]

Fig 2 Scheme of the assembled crankshaft

The crankshaft is a very responsible element ofthe marine engine its failure especially during theoperationinthestormmayleadtoathreattowardsthelivesofshipboardpersonnelandpassengersMoreoverthe industrial design requires knowledge based on the investigationintheareaofnewmaterialsandmethods[1012]andknowledgeconcerningsurfacemorphology[7]BecauseofsafetyreasonsthejointsmustbeabletobeartheincreasedloadTocalculatetheloadingcausingtheshaft-hubdetachmentRadietal[13]proposedananalyticalmodelbasedupontheLameacuteequationandthebeamtheoryIntheirsolutiontheshaft-hubcontactwasmodelled in terms of two elastic Timoshenko beams connected by a distributed elastic spring Its stiffnesswasevaluatedanalyticallyItisknownthatfracturesincrankshaftscanoccurbybending fatigueby torsionalfatigue or a combination of both [5] Moreover inthe assembled crankshafts the hub must be axiallypositioned on the shaft and resist the moments and forces generated bymisalignments [9]However it isknown that the residual stresses and their distribution in mechanical components may cause them to fail at a load level significantly lower than expected [1]Siemiatkowski [16] demonstrated experimentally thatthe relative interference above 22permil did not improvethe bonding strength of the shrink-fitted crank-pivotjoint and Burenin et al [2] proved theoretically thatplasticdeformation in the shrink-fitted couplingundertemperaturestresssignificantlyreduceditstightnessTheeffectoftheshrinkfittingratioswasinvestigatedbyLee

etal[8]andKimetal[6]proposedatechniquebasedonthe3DFiniteElementsmethodfortheshrink-fittingprocedureoptimizationStrozzietal[17]reportedtheresultsofastudyonelasticstressconcentrationZhangetal[20]studiedtheinitiationcharacteristicsoffrettingfatiguecracks in theshrink-fittedshafts toanalyse thewearmechanism

1 Internal stress measurement in the models

There are many methods of the the analysis of shrink fit couplings in different applications [19]but in case of such a large-sized elements empiricalinvestigationandmeasurementisextremelydifficultandexpensive In particular fabrication of the assembledcrankshaftjustfortheresidualstressmeasurementwiththelayerremovaltechnique[3]wouldbefinanciallytooexpensive

Therefore 15 scale models were made in orderto make laboratory measurements possible [16]The shrink fitted couplings were fabricated keepingtechnology as close to the technical conditions of large-sizedcrankshaftsproductionaspossible

Surface topography has a crucial effect on theprecision of the obtained parts [11] Thus roughnessand form deviations of the models were measured inorder to ensure the appropriate relative tightness ofthecouplingsintherangebetween20and25permilThe











(1)






w=ucz ndash uk (2)

Thusitcanbewritten

(3)

(4)



(5)

(6)


b) For the crank

(7)

(8)



(9)

(10)







Msmax=2mpRpLp(11)


thickness

2




























surface






Heatingtime

t [min]






(T1+T5)2[ordmC]

(T2+T6)2[ordmC]

(T3+T7)2[ordmC]

(T4+T8)2[ordmC]

DD1[mm]

DD2[mm]

DD3[mm]

DD4[mm]

DD1[mm]

DD2[mm]

DD3[mm]

DD4[mm]

0 0 0 0 0 0 0 0 0 0 0 0 010 93 98 100 97 014 015 015 015 015 016 016 01520 162 166 168 165 025 025 025 025 026 026 027 02630 207 210 213 211 033 032 032 032 033 034 034 03440 240 243 245 243 039 038 038 038 038 039 039 03950 263 267 268 267 042 041 041 041 042 043 043 04360 279 283 283 282 045 044 044 044 045 045 045 04570 290 294 296 293 047 047 047 046 046 047 047 04780 299 301 303 302 048 048 048 048 048 048 048 04890 304 306 307 307 049 049 049 049 048 049 049 049














Conclusions













Acknowledgement


Funding sources


References
































(1)






w=ucz ndash uk (2)

Thusitcanbewritten

(3)

(4)



(5)

(6)


b) For the crank

(7)

(8)



(9)

(10)







Msmax=2mpRpLp(11)


thickness

2




























surface






Heatingtime

t [min]






(T1+T5)2[ordmC]

(T2+T6)2[ordmC]

(T3+T7)2[ordmC]

(T4+T8)2[ordmC]

DD1[mm]

DD2[mm]

DD3[mm]

DD4[mm]

DD1[mm]

DD2[mm]

DD3[mm]

DD4[mm]

0 0 0 0 0 0 0 0 0 0 0 0 010 93 98 100 97 014 015 015 015 015 016 016 01520 162 166 168 165 025 025 025 025 026 026 027 02630 207 210 213 211 033 032 032 032 033 034 034 03440 240 243 245 243 039 038 038 038 038 039 039 03950 263 267 268 267 042 041 041 041 042 043 043 04360 279 283 283 282 045 044 044 044 045 045 045 04570 290 294 296 293 047 047 047 046 046 047 047 04780 299 301 303 302 048 048 048 048 048 048 048 04890 304 306 307 307 049 049 049 049 048 049 049 049














Conclusions













Acknowledgement


Funding sources


References



























(1)






w=ucz ndash uk (2)

Thusitcanbewritten

(3)

(4)



(5)

(6)


b) For the crank

(7)

(8)



(9)

(10)







Msmax=2mpRpLp(11)


thickness

2




























surface






Heatingtime

t [min]






(T1+T5)2[ordmC]

(T2+T6)2[ordmC]

(T3+T7)2[ordmC]

(T4+T8)2[ordmC]

DD1[mm]

DD2[mm]

DD3[mm]

DD4[mm]

DD1[mm]

DD2[mm]

DD3[mm]

DD4[mm]

0 0 0 0 0 0 0 0 0 0 0 0 010 93 98 100 97 014 015 015 015 015 016 016 01520 162 166 168 165 025 025 025 025 026 026 027 02630 207 210 213 211 033 032 032 032 033 034 034 03440 240 243 245 243 039 038 038 038 038 039 039 03950 263 267 268 267 042 041 041 041 042 043 043 04360 279 283 283 282 045 044 044 044 045 045 045 04570 290 294 296 293 047 047 047 046 046 047 047 04780 299 301 303 302 048 048 048 048 048 048 048 04890 304 306 307 307 049 049 049 049 048 049 049 049














Conclusions













Acknowledgement


Funding sources


References

















































surface






Heatingtime

t [min]






(T1+T5)2[ordmC]

(T2+T6)2[ordmC]

(T3+T7)2[ordmC]

(T4+T8)2[ordmC]

DD1[mm]

DD2[mm]

DD3[mm]

DD4[mm]

DD1[mm]

DD2[mm]

DD3[mm]

DD4[mm]

0 0 0 0 0 0 0 0 0 0 0 0 010 93 98 100 97 014 015 015 015 015 016 016 01520 162 166 168 165 025 025 025 025 026 026 027 02630 207 210 213 211 033 032 032 032 033 034 034 03440 240 243 245 243 039 038 038 038 038 039 039 03950 263 267 268 267 042 041 041 041 042 043 043 04360 279 283 283 282 045 044 044 044 045 045 045 04570 290 294 296 293 047 047 047 046 046 047 047 04780 299 301 303 302 048 048 048 048 048 048 048 04890 304 306 307 307 049 049 049 049 048 049 049 049














Conclusions













Acknowledgement


Funding sources


References











































surface






Heatingtime

t [min]






(T1+T5)2[ordmC]

(T2+T6)2[ordmC]

(T3+T7)2[ordmC]

(T4+T8)2[ordmC]

DD1[mm]

DD2[mm]

DD3[mm]

DD4[mm]

DD1[mm]

DD2[mm]

DD3[mm]

DD4[mm]

0 0 0 0 0 0 0 0 0 0 0 0 010 93 98 100 97 014 015 015 015 015 016 016 01520 162 166 168 165 025 025 025 025 026 026 027 02630 207 210 213 211 033 032 032 032 033 034 034 03440 240 243 245 243 039 038 038 038 038 039 039 03950 263 267 268 267 042 041 041 041 042 043 043 04360 279 283 283 282 045 044 044 044 045 045 045 04570 290 294 296 293 047 047 047 046 046 047 047 04780 299 301 303 302 048 048 048 048 048 048 048 04890 304 306 307 307 049 049 049 049 048 049 049 049














Conclusions













Acknowledgement


Funding sources


References


































surface






Heatingtime

t [min]






(T1+T5)2[ordmC]

(T2+T6)2[ordmC]

(T3+T7)2[ordmC]

(T4+T8)2[ordmC]

DD1[mm]

DD2[mm]

DD3[mm]

DD4[mm]

DD1[mm]

DD2[mm]

DD3[mm]

DD4[mm]

0 0 0 0 0 0 0 0 0 0 0 0 010 93 98 100 97 014 015 015 015 015 016 016 01520 162 166 168 165 025 025 025 025 026 026 027 02630 207 210 213 211 033 032 032 032 033 034 034 03440 240 243 245 243 039 038 038 038 038 039 039 03950 263 267 268 267 042 041 041 041 042 043 043 04360 279 283 283 282 045 044 044 044 045 045 045 04570 290 294 296 293 047 047 047 046 046 047 047 04780 299 301 303 302 048 048 048 048 048 048 048 04890 304 306 307 307 049 049 049 049 048 049 049 049














Conclusions













Acknowledgement


Funding sources


References



























Heatingtime

t [min]






(T1+T5)2[ordmC]

(T2+T6)2[ordmC]

(T3+T7)2[ordmC]

(T4+T8)2[ordmC]

DD1[mm]

DD2[mm]

DD3[mm]

DD4[mm]

DD1[mm]

DD2[mm]

DD3[mm]

DD4[mm]

0 0 0 0 0 0 0 0 0 0 0 0 010 93 98 100 97 014 015 015 015 015 016 016 01520 162 166 168 165 025 025 025 025 026 026 027 02630 207 210 213 211 033 032 032 032 033 034 034 03440 240 243 245 243 039 038 038 038 038 039 039 03950 263 267 268 267 042 041 041 041 042 043 043 04360 279 283 283 282 045 044 044 044 045 045 045 04570 290 294 296 293 047 047 047 046 046 047 047 04780 299 301 303 302 048 048 048 048 048 048 048 04890 304 306 307 307 049 049 049 049 048 049 049 049














Conclusions













Acknowledgement


Funding sources


References

































Conclusions













Acknowledgement


Funding sources


References
























Conclusions













Acknowledgement


Funding sources


References





























Acknowledgement


Funding sources


References




























Documents

INVESTIGATIONS ON ThE MODElED ShRINK-FITTED JOINTS OF …jmcm.itee.radom.pl/images/pe_2018/pe_1_2018/PE_1_2018_s... · 2018-04-16 · the marine engine, its failure, especially during