Turb Des

D
DESIGN
NAND
DANA
ALYSISSOFSMALLLSCALLE
W
WIND
DTURB
BINESSUPPO
ORTST
TRUCTTURESS
by
EmmaNel
Thessis presentedd in fulfilmen

nt of the reqquirements for
fo the degreee of Master of Science in
i the
F
Faculty
of Sttructural Enngineering at Stellenbosch Universitty
Supervisors:
Dr.JJ.A.v.B.Strassheim
Proof.P.E.Dunaiski
D
December201
12
Stellenbosch University http://scholar.sun.ac.za
DECLARATION
By submittingthis thesis electronically, I declare that the entirety of the workcontained therein is my own,
original work, that I am the sole author thereof (save to the extent explicitly otherwise stated), that
reproductionandpublicationthereofbyStellenboschUniversitywillnotinfringeanythirdpartyrightsandthat
Ihavenotpreviouslyinitsentiretyorinpartsubmitteditforobtaininganyqualification.
Date:
^h

i|DesignandAnalysisofSmallScaleWindTurbineSupportStructures
SYNOPSIS
Atechnologythathasadvancedimmeasurablyasaresultofthenecessityforgreenenergyproductionisthe
harnessing ofwind energy.One of the most importantaspects of a wind turbine is its supportingstructure.
Thetowerofawindturbineneedstobesufficientlyreliableandstructurallysoundtoensurethatthedesign
lifeofthewindturbinemachineisunaffected.Thetoweralsoneedstobeofthecorrectheighttoensurethat
thefullpotentialofenergycaptureisrealised.
The supporting structure of a wind turbine constitutes up to as much as 30% of the total costs of a wind
turbine. The most common wind turbine supporting structures seen worldwide today are Steel Monopole
Towers. The large cost proportion of the tower compels the industry to investigate the most feasible
alternativesupportingtowerstructuresandthuspromptedtheresearchdevelopedinthisthesis.Inthisthesis
the focus is on small scale wind turbines (<50kW), more specifically, a 3kW Wind Turbine. The proposed
alternativedesignthesupportstructuresofsmallscale windturbinestothepresentlyusedSteelMonopole
towerwasaSteelLatticetower.
Both a Steel Lattice and Steel Monopole Tower was designed for a 3kW Wind Turbine using rational design
methodsdeterminedfrompertinentsectionsoftheSouthAfricandesigncodes.TheTowerdesignsneededto
incorporatethedetailsoftheelementconnections,soastoencompassallofthecostparametersaccurately.
The foundation design of each of the towers was also required from the point of view of cost analysis
completeness,andendedupplayingacriticalroleinthefeasibilityanalysis.
Tovalidatethedesignmethods,thetwotowersweremodelledinthefiniteelementpackageStrand7anda
numberofdifferentanalyseswereperformedonthetwotowers.Theanalysesincludedlinearstatic,nonlinear
static, natural frequency and harmonic frequency analyses. The towers were assessed for a number of
differentloadcasecombinationsandwereexaminedintermsofstressstates,massparticipationfactorsand
deflections,tomentionafew,fortheworstloadingcombinationcasesthatwereencountered.
Once a final design was reached for both the Steel Lattice and Steel Monopole Towers, each element from
whichtheyweremadewasassessedfromastructuralviewpointtodeterminemanufacturingandconstruction
costs.
The cost analysis was conducted by means of asking a number of leading construction companies for unit
pricesforeachoftheidentifiedelementstobeassessed.
ThefabricationandconstructionofeachoftheTowerswasthencomparedtodeterminewhichonewasmore
feasible,intermsofeachdesignaspectconsideredaswellaslookingatthecompleteendproduct.
It was found that the Steel Lattice Tower was more feasible from the points of view of fabrication, and
construction,aswellashavingafarmorecosteffectivefoundation.Thiswasapositiveconclusionfromthe
perspectiveoftheproposalforamorefeasiblealternativetothepresentlyusedSteelMonopoleTowers.
Theoutcomeoftheresearchconductedherecouldcertainlyprovetobeworthconsideringfromawindfarm
developmentperspective,withparticularfocusontheupandcomingWindIndustrydevelopmentsinSouth
Africa.
ii|DesignandAnalysisofSmallScaleWindTurbineSupportStructures
AFRIKAANSSYNOPSIS
As gevolg van die noodsaaklikheid vir die produksie van volhoubare energie is tegnologie wat met rasse
skrede vooruitgegaan het die vir die benutting van windenergie. Een van die belangrikste aspekte van 'n
windturbineisdieondersteunendestruktuur.Dietoringvan'nwindturbinemoetfunksioneelenstruktureel
betroubaar wees om te verseker dat die ontwerpleeftyd van die windturbine masjien nie nadelig benvloed
wordnie.
Die toring moet ook die regte hoogte wees om te verseker dat die volle potensiaal van die wind energie in
meganieseenergieomgesitword.
Diekostevandieondersteunendestruktuurvan'nwindturbineverteenwoordigtot30%vandietotalekoste
van 'n windturbine. Die mees algemene vorm van ondersteunende strukture vir windturbines wat vandag
wreldwydtegekomword,isdievan'nenkelstaalbuisvormigetoring.Diegrootkostekomponentvandie
toring dwing die industrie om ondersoek in te stel na die mees koste effektiewe prakties uitvoerbare
alternatiefvirdieondersteunendetoringstruktuur.Hierdieaspekvandiestruktuurkonseptualiseringhetgelei
totdienavorsingwatinhierdietesisonderneemis.Diefokusvandienavorsingisopkleinskaalwindturbines
(<50kW),enmeerspesifiekop'n3kWwindturbinemodel.Diealternatieweontwerpwatontwikkelisvirklein
skaal wind turbines se ondersteunende structure, is 'n staal vakwerk toring as alternatief vir die staal
buisvormigetoring.
Beide'nstaalvakwerkenstaalbuisvormigetoringvir'n3kWwindturbineisontwerpdeurrasioneleontwerp
metodes.DietoepaslikegedeeltesvandieSuidAfrikaanseontwerpkodesishiervoorgebruik.Dieontwerpvir
die toring moet die besonderhede van die element verbindings in ag neem en die nodige koste parameters
moet akkuraat bepaal word. Die ontwerp van die fondament van elke toring is ook noodsaaklik vir die
volledigheidvandiekosteontledingenditspeelook'nkritiekerolindiegangbaarheidanalise.
Om die ontwerp metodes te bevestig, is die twee tipes torings in die eindige element pakket, Strand7,
gemodelleeren'naantalverskillendeontledingsvirdietweetoringsisuitgevoer.Dieontledingssluitlineren
nieliner statiese ontledings asook natuurlike frekwensie en dinamiese ontledings onder harmoniese
belastings in. Die torings is vir 'n aantal verskillende lasgevalkombinasies ondersoek en in die spannings
toestande, massadeelname faktore en defleksies vir die ergste laskombinasie gevalle wat ondervind is, is
geassesseer.
Sodra 'n finale ontwerp vir beide die staal vakwerk en staal buisvormige toring voltooi is, is elke element
beoordeeluit'nstruktureleenmateriaaloogpuntomdiekostesdaarvantebepaal.
Diekosteanaliseisbaseeropdatawatvoorsienisdeur'naantalvooraanstaandekonstruksiemaatskappyeop
'npryspereenheidbasisvirelkvandiegedentifiseerdeelementewatgeassesseermoesword.
Dievervaardigingenkonstruksievanelketoringisdanvergelykomtebepaalwattereendiemeeshaalbaaris,
intermevanelketoepaslikeontwerpsaspekendeurookdievolledigeeindprodukteevalueer.
Daarisbevinddatdiestaalvakwerktoringuitdieoogpuntvanvervaardigingenkonstruksie,asookasgevolg
van 'n meer kosteeffektiewe fondament, die voorkeur alternatief verteenwoordig het. Dit was 'n positiewe
gevolgtrekkinguitdieoogpuntvandiesoekena'nanderalternatiefasdiebuisvormigestaaltoringswattans
algemeeningebruikis.
Die uitkoms van hierdie navorsing verdien oorweging uit windplaas ontwikkelingsperspektief, met
spesifiekefokusopdieopkomendeontwikkelingeindiewindenergieindustrieinSuidAfrika.
iii|DesignandAnalysisofSmallScaleWindTurbineSupportStructures
ACKNOWLEDGEMENTS
IwouldfirstliketothankthelateProfessorDunaiskiforgivingmetheopportunitytodomyMastersdegree;if
itwerentforhisbeliefinmenoneofthiswouldhavecometobe.
ThankyoutoDr.StrasheimforbeingsuchakindandsupportiveStudyleader.
ThankyouEtiennevanderKlashorstforhelpingmewithsomanyhoursofpatience.
ThankyouDadforbeingmysoundingboardandmentoreveryday.
ThankyouJacquesLoubserforyourendlesssupportandguidancewiththisendeavour,Icouldnthavedoneit
withoutyou.
iv|DesignandAnalysisofSmallScaleWindTurbineSupportStructures
TABLEOFCONTENTS
Declaration..............................................................................................................................................................i
Synopsis..................................................................................................................................................................ii
AfrikaansSynopsis.................................................................................................................................................iii
Acknowledgements...............................................................................................................................................iv
ListofFigures.........................................................................................................................................................xi
ListofTables.........................................................................................................................................................xiii
ListofSymbols:.....................................................................................................................................................xiv
ListofAbbreviations:..........................................................................................................................................xviii
Chapter1:IntroductiontotheDesignandAnalysisofSmallScaleWindTurbines...............................................1
Chapter2:LiteratureReview..................................................................................................................................3
Introduction............................................................................................................................................................3
2.1TheUnderstandingofWindEnergy.............................................................................................................3
2.1.2MotivationforWindEnergy......................................................................................................................3
2.1.2HowWindisturnedintoEnergy...............................................................................................................4
Densityoftheair:..........................................................................................................................................4
RotorArea:.....................................................................................................................................................4
NumberofBlades:.........................................................................................................................................6
PitchversusStall:...........................................................................................................................................7
2.2LoadsInducedbytheWindonaWindTurbineSupportingStructure.........................................................7
2.3SupportingStructures..................................................................................................................................8
2.3.1TowerTypes..............................................................................................................................................8
FreestandingTowers:...................................................................................................................................9
LatticeStyleTowers:......................................................................................................................................9
MonopoleTowers........................................................................................................................................10
2.3.2TowerHeightConsiderations..................................................................................................................11
2.3.3ProsandConsofeachTowerType.........................................................................................................11
2.4TheDesignofSupportingTowers..............................................................................................................12
2.5TheAnalysisofSupportingTowers............................................................................................................12
2.5.1FiniteElementAnalysis...........................................................................................................................12
2.5.2FeasibilityAnalysis...................................................................................................................................13
Chapter3:Towers................................................................................................................................................14
Thedesignprocedure:......................................................................................................................................14
3.1TheGeometricLayout................................................................................................................................14
3.1.1TheGeometricLayoutoftheSteelMonopoleTower.............................................................................14
3.1.2TheGeometricLayoutoftheSteelLatticeTower...................................................................................16
v|DesignandAnalysisofSmallScaleWindTurbineSupportStructures
3.2TheBoundaryconditionsofTowerdesign.................................................................................................17
3.3LoadingConditions.....................................................................................................................................17
3.3.1LoadingConditionsfromtheWind.........................................................................................................17
3.3.2LoadingconditionsinducedbytheWindandtheWindturbinemachine..............................................22
3.4ValidationofStructuraldesign...................................................................................................................24
3.4.1DesignValidationoftheSteelMonopoleTowerGeometry...................................................................26
3.4.1.1AxialCompression................................................................................................................................26
3.4.1.2BendingLaterallysupportedmembers................................................................................................26
3.4.1.3Shear....................................................................................................................................................27
3.4.1.4InteractionEquations:Combinedbendingandaxialcompression......................................................27
3.4.2DesignValidationoftheSteelLatticeTowerStructure..........................................................................28
3.4.2.1Designofcompressionmembers:Effectivecrosssectionalpropertiesofcompressionmembers.....28
3.4.2.2Flexuralbucklingofaxiallycompressedmembers...............................................................................28
3.4.2.3Bendingandaxialcompression............................................................................................................32
3.4.2.4DesignofTensionmembers:................................................................................................................32
3.4.2.5Combinedbendingandaxialtensile/compressiveforces:...................................................................33
3.5DesignIteration..........................................................................................................................................33
3.6GeneralAspectsofTowerdesign:..............................................................................................................34
3.6.1Materials.................................................................................................................................................34
MaterialsfortheCircularHollowsections...................................................................................................34
MaterialsforAnglesectionsandPlates:......................................................................................................34
MaterialsforBolts:......................................................................................................................................34
3.6.2CorrosionProtection...............................................................................................................................35
3.6.3Access......................................................................................................................................................35
3.6.4Production...............................................................................................................................................36
3.6.5Fatigue.....................................................................................................................................................36
Chapter4:ConnectionDesign..............................................................................................................................37
4.1MonopoleRingFlangeConnections...........................................................................................................37
4.1.1BoltForceatseparation:.........................................................................................................................39
4.1.2Maximumbendingstrengthofflangefollowingplateseparation:.........................................................40
4.1.3Maximumbendingstrengthofflangewhenseparationoccursafteryieldingoftheflange:.................41
4.1.4EvaluationofthePryingForce:...............................................................................................................42
4.1.5Pryingforceactingbeforeseparation:....................................................................................................42
4.1.6Separationloadandfractureloadofthebolts:......................................................................................43
4.1.6.1SeparationLoad:..................................................................................................................................44
4.1.6.2FractureLoadoftheBolts:...................................................................................................................44
4.1.7MaximumStrengthofaJoint:.................................................................................................................44
vi|DesignandAnalysisofSmallScaleWindTurbineSupportStructures
4.1.8SuggestedDesignMethods:....................................................................................................................46
GeneralProcedure:......................................................................................................................................46
SimplifiedDesign:........................................................................................................................................46
4.1.9ConnectionDesignfrombasicPrinciples:...............................................................................................47
4.1.10ConnectionDesignsinFiniteElementProgram(Strand7):...................................................................48
MidsectionConnections:............................................................................................................................48
BaseConnection:.........................................................................................................................................49
4.1.10.1ConnectionCalculations:....................................................................................................................50
ValidationoftheBoltsizeandstrength:......................................................................................................51
Validationoftheflangesizeandstrength:..................................................................................................52
4.2SteelLatticeTowerConnections:...............................................................................................................54
4.2.1Connectiontype......................................................................................................................................54
4.2.2ConnectionDesign:.................................................................................................................................55
4.2.2.1Choosingtheappropriatebolts:...........................................................................................................55
4.2.2.2Connectiondetails:...............................................................................................................................55
4.2.3ConnectionCalculations:.........................................................................................................................56
4.2.4GeneralAspectspertinenttoallconnectiontypesdealtwithinthischapter:.......................................57
4.2.4.1Welds:...................................................................................................................................................57
4.2.4.2Bolts:.....................................................................................................................................................58
Chapter5:FoundationDesign..............................................................................................................................60
5.1Introductiontofoundations.......................................................................................................................60
5.2SteelMonopoleTowerFoundationDesign................................................................................................61
5.2.1FoundationType.....................................................................................................................................61
5.2.2FoundationDimensions..........................................................................................................................61
5.2.3FoundationCalculations..........................................................................................................................62
5.3SteelLatticeTowerFoundationDesign......................................................................................................66
5.3.1FoundationType.....................................................................................................................................66
5.3.2DesigningtheSteelBasePlates...............................................................................................................66
5.3.3Designingtheconcretefoundation:........................................................................................................67
5.4Conclusion..................................................................................................................................................70
Chapter6:FiniteElementAnalyses......................................................................................................................71
6.1Introduction...............................................................................................................................................71
6.2SteelMonopoleTower:FiniteElementAnalyses.......................................................................................71
6.2.1LinearStaticAnalysis...............................................................................................................................72
ThenodalreactionsattheBaseoftheTower:............................................................................................75
LinearStaticAnalysisstressstates:..............................................................................................................78
6.2.2LinearBucklingAnalysis..........................................................................................................................78
vii|DesignandAnalysisofSmallScaleWindTurbineSupportStructures
LinearBucklingSolver:.................................................................................................................................78
6.2.3NonlinearAnalysis...................................................................................................................................79
6.2.4DynamicAnalyses....................................................................................................................................81
NaturalFrequencySolverOverview:...........................................................................................................81
6.2.4.1TheEffectsofanOutofBalanceRotor:...............................................................................................83
HarmonicResponseSolverOverview:.........................................................................................................83
HarmonicResponseResults:........................................................................................................................84
Detailsoftheoutofbalancerotoranalysis:................................................................................................84
6.2.4.2EffectsofVortexSheddingduetowindaction....................................................................................85
Harmonicdisplacementresponsevs.timeanalysis:...................................................................................85
6.3StressStates...............................................................................................................................................87
HarmonicResponseStressAnalysis:............................................................................................................87
Vortexsheddingstressanalysis:..................................................................................................................87
6.4FatigueAssessment:...................................................................................................................................88
6.5MassParticipation:.....................................................................................................................................90
ResonanceEffects:.......................................................................................................................................91
6.6SteelLatticeTower:FiniteElementAnalyses.............................................................................................93
WindLoadModelling:..................................................................................................................................95
6.6.1LinearStaticAnalysis...............................................................................................................................96
BaseResponse:............................................................................................................................................98
LinearStaticstressanalysis:.........................................................................................................................99
6.6.2LinearbucklingAnalysis:.........................................................................................................................99
6.6.3NonlinearStaticAnalysis:......................................................................................................................101
6.6.4DynamicAnalyses:.................................................................................................................................102
NaturalFrequencyAnalysis:......................................................................................................................102
6.6.4.1DynamicEffectsofanoutofBalanceRotor:......................................................................................103
6.6.5MassParticipation:................................................................................................................................105
ResonanceEffects:.....................................................................................................................................106
6.7StressState...............................................................................................................................................107
Harmonicresponsestressanalysis............................................................................................................107
6.8FatigueAssessment..................................................................................................................................108
Chapter7:Feasibility..........................................................................................................................................109
7.1Introductiontofeasibility.........................................................................................................................109
7.1.1TowerSpecifications.............................................................................................................................109
7.1.1.1SteelMonopoleTower.......................................................................................................................109
7.1.1.2SteelLatticeTower.............................................................................................................................109
7.2Thestructuralrequirementsofthetowers..............................................................................................110
viii|DesignandAnalysisofSmallScaleWindTurbineSupportStructures
7.3Theaestheticrequirementsofthetowers...............................................................................................110
7.4Theconstructabilityofthetowers...........................................................................................................110
7.4.1Transportation:......................................................................................................................................110
7.4.2FabricationandConstruction:...............................................................................................................110
7.4.2.1Fabrication:........................................................................................................................................110
FirstFabricatorCostResponse:.................................................................................................................111
SecondFabricationCostResponse:...........................................................................................................111
7.4.2.2Construction:......................................................................................................................................112
SteelTowerFabrication:............................................................................................................................115
FoundationConstruction:..........................................................................................................................116
7.5Conclusion:...............................................................................................................................................118
Chapter8:ConclusionandRecommendations..................................................................................................120
ListofReferences...............................................................................................................................................121
RelevantSouthAfricanDesignCodes:.......................................................................................................121
OtherResources:.......................................................................................................................................121
AppendixA:TowerDesignCalculations.................................................................................................................A
A1:SteelMonopoleDesign................................................................................................................................A
A1.1DesignReferenceParametersandTowerResistanceCalculations...........................................................A
TowerGeometryandAxialCompressionResistance:...................................................................................A
FlexuralandAxialCompressionResistance:..................................................................................................B
ShearResistance:...........................................................................................................................................C
A1.2WindCalculationsonSteelMonopoleTower............................................................................................D
A1.3ActionsInducedontheTower...................................................................................................................F
A1.4InteractionEquations................................................................................................................................G
A2:SteelLatticeDesign.....................................................................................................................................H
A2.1DesignReferenceParametersandTowerResistanceCalculations..........................................................H
TowerGeometryandAxialCompressionResistance:..................................................................................H
A2.2WindCalculationsonSteelLatticeTower..................................................................................................K
ApplyingtheWindLoads:..................................................................................................................................M
WindLoadcases:...............................................................................................................................................M
A2.3ActionsInducedonthetopoftheTower.................................................................................................O
AppendixB:Connections........................................................................................................................................A
B.1SteelMonopoleConnections.......................................................................................................................A
RingFlangeConnections:Method1..................................................................................................................A
Method2:Generaltheoryofresistance:...........................................................................................................E
B.2SteelLatticeTowerConnections.................................................................................................................G
GussetPlateconnections..............................................................................................................................G
ix|DesignandAnalysisofSmallScaleWindTurbineSupportStructures
AppendixC:FoundationDesign..............................................................................................................................A
C.1SteelMonopoleTowerFoundationDesign..................................................................................................A
InputValuesforProkonFoundationDesign..................................................................................................A
OutputfromProkonFoundationDesign.......................................................................................................A
BendingScheduleOutputfromProkonFoundationDesign..........................................................................B
HandCalculationsforSteelMonopoleTowerFoundationDesign....................................................................C
C.2SteelLatticeTowerFoundationDesign........................................................................................................E
InputValuesforProkonFoundationDesign..................................................................................................E
OutputfromProkonFoundationDesign.......................................................................................................E
BendingScheduleOutputfromProkonFoundationDesign..........................................................................E
HandCalculationsDesigningtheSteelBasePlatesfortheSteelLatticeTowersFoundation.........................G
AppendixD:AttachedCDofDesignFiles...............................................................................................................A
x|DesignandAnalysisofSmallScaleWindTurbineSupportStructures
LISTOFFIGURES
Figure 1: Power coefficient curve, parametised in accordance with the blade pitch angle, image from
http://www.caspus.eclipse.co.uk/ah/publications/3dmwtucfd.pdf.......................................................................5
Figure2:TypicalWindTurbinePowerCurve.........................................................................................................8
Figure3:DifferentMonopoleProfiles..................................................................................................................15
Figure4:RingFlangeconnectionforaCHS..........................................................................................................16
Figure6:ActionsInducedonWindTurbineSupportTower................................................................................22
Figure7:FlexuralBucklingofanEqualLegAngle.................................................................................................29
Figure8:FlexuralTorsionalBucklingofanEqualLegAngle.................................................................................29
Figure9:DimensionsandCrossSectionalAxesofAngleSections.......................................................................31
Figure10:ModesofFailureforRingFlangeConnections....................................................................................38
Figure11:BendingMomentsdevelopedinRingFlangeConnectionFailures.....................................................38
Figure12:RingFlangeConnectionDetails...........................................................................................................39
Figure13:RingFlangeConnectionYieldLines.....................................................................................................40
Figure14:RingFlangeConnection:yieldlinebeforeseparation.........................................................................41
Figure15:PryingActionModel............................................................................................................................42
Figure16:PMAXversusTfrelation..........................................................................................................................46
Figure17:BoltSpacingandTributaryLength.......................................................................................................47
Figure18:DimensionalParametersofFlange......................................................................................................48
Figure19:MidsectionConnectionDesign...........................................................................................................49
Figure20:MidFlangeandWebStiffenerDetails.................................................................................................49
Figure21:BaseFlangeConnection.......................................................................................................................50
Figure22:BaseFlangeandWebStiffenerDetails................................................................................................50
Figure23:ReactionForcesatMonopoleTowerBase(SLS)..................................................................................51
Figure 24: Moments in plane 11 (direction of yaxis loading) for the base ring flange connection of the
Monopole.............................................................................................................................................................53
Figure25:Momentsinplane22(directionofxaxis)forthebaseringflangeconnectionoftheMonopole......53
Figure26:VonMisesMomentsinthebaseringflangeconnectionoftheMonopole........................................53
Figure27:TrescamomentsinthebaseringflangeconnectionoftheMonopole...............................................54
Figure28:EqualLegAngleSectionHoleDimensioning........................................................................................54
Figure29:LatticeTower,pinnedconnectionswithgussetplates........................................................................57
Figure30:Failedwindturbinefoundationdesign................................................................................................60
Figure31:MonopoleFoundationType................................................................................................................61
Figure32:MonopoleFoundationEquilibrium......................................................................................................62
Figure33:MonopoleFoundationequilibriumequivalent....................................................................................62
Figure34:MonopoleFoundationBearingPressurelayout..................................................................................63
Figure35:MonopoleFoundationBearingPressureDistributions.......................................................................64
Figure36:ProkonInputLayoutofMonopoleFoundation...................................................................................64
Figure37:ProkonOutputforMonopoleFoundation..........................................................................................65
Figure38:MonopoleSchematicBendingschedule..............................................................................................65
Figure39:EffectiveBasePlateArea.....................................................................................................................67
Figure40:SteelLatticeTowerFoundations.........................................................................................................68
Figure41:ProkonLatticeFoundationInput.........................................................................................................68
Figure42:ProkonLatticeFoundationOutput......................................................................................................69
Figure43:ProkonBendingscheduleLatticeFoundation.....................................................................................69
Figure44:MonopoleTowerWindTurbineConnectionSimulation.....................................................................72
Figure45:Monopole:LargestStaticdisplacementloadcombination.................................................................74
Figure46:Monopole,Lineardisplacementdistribution......................................................................................75
Figure47:Axessystemfornodalreactions..........................................................................................................75
xi|DesignandAnalysisofSmallScaleWindTurbineSupportStructures
Figure48:FX(N)BaseReactionForces,ULS(WorstCase)Combination3...........................................................76
Figure49:FY(N)BaseReactionForces,ULS(WorstCase),Combination3..........................................................77
Figure50:FZ(N)BaseReactionForces,ULS(WorstCase),Combination3..........................................................77
Figure51:Monopole,LinearBucklingFactors.....................................................................................................79
Figure52:MonopolenonlineardisplacementVs.Loading..................................................................................80
Figure53:InitialNonlinearAnalysis:MonopoleULS(WorstCase)......................................................................81
Figure54:Everysecondmodeshapeofthefirst20modesoftheSteelMonopole............................................82
Figure55:Monopole,NaturalFrequencyAnalysis...............................................................................................82
Figure56:WorstCasedisplacementoccurringforharmonicresponse...............................................................85
Figure57:Vortexsheddingharmonicresponse,worstcasedisplacement.........................................................86
Figure58:MonopoleStressresponsetothedynamicloadingasaresultoftheoutofbalancerotoreffects....87
Figure59:Vortexsheddingstressstate...............................................................................................................88
Figure60:Stressvs.NumberofCyclesforFatigueAnalysisofoutofbalancerotorinaccordancewithSANS
101621:2005Clause26.......................................................................................................................................89
Figure61:FatigueAnalysis,StressRangeversusNumberofcyclesforVortexsheddinganalysisinaccordance
withSANS101621:2005......................................................................................................................................90
Figure62:Resonanceevaluationofsteelmonopole...........................................................................................92
Figure63:ConnectionDetailsbetweentopparallelportionofthetowerandthetaperedportionofthetower
..............................................................................................................................................................................93
Figure64:LatticeTowerTurbineConnectionSimulation....................................................................................94
Figure65:WindLoadingCasesforsquareplanSteelLatticeTowers..................................................................95
Figure66:StresscomparisonfortwodifferentwindloadsonSteelLatticeTower.............................................96
Figure67:DisplacementcomparisonfortwodifferentwindloadsonSteelLatticeTower................................96
Figure68:DifferentwindloadcasesforSquareplanSteelLatticeTowers.........................................................97
Figure69:SteelLatticeTowerDisplacement(XY)StaticAnalysis,WindLoadcase1,SLS2.................................97
Figure70:BaseResponse,SteelLatticeTower,forWindloadcase1.................................................................98
Figure71:BaseResponse,Steellatticetower,forWindLoadcase2..................................................................98
Figure72:LinearBucklingFactors:SteelLatticeTower,WindLoadCase1.........................................................99
Figure73:LinearBucklingFactors,SteelLatticeTower,WindLoadcase2.......................................................100
Figure74:NonlinearDisplacement,InitialNonlinearAnalysis,LatticeTower...................................................101
Figure75:NonlinearDisplacement,InitialToweranalysis,LatticeTower.........................................................101
Figure76:NaturalfrequencyversusmodeforSteelLattice,WindLoadCase1................................................102
Figure77:NaturalfrequencycomparisonforMonopoleandSteelLatticeTowers...........................................102
Figure78:First10modeshapesforSteelLatticeTower....................................................................................103
Figure79:MaximumDisplacementeffectsfromdynamicanalysisofoutofbalancerotoreffects..................104
Figure80:Resonanceeffectsofsteellatticetower...........................................................................................106
Figure81:Stressstatefromoutofbalanceharmonicresponse........................................................................107
Figure82:StressRangeversusNumberofcyclesforfatigueanalysisinaccordancewithSANS101621:2005108
Figure83:CostComparisonforSteelTowersandFoundationscombined........................................................115
Figure84:CostDifferencebetweencompaniesandTowersforTowerandFoundationcostscombined........115
Figure85:SteelTowerQuotationComparison..................................................................................................116
Figure86:Differenceincostpercompanyandtowerdesignforthecostofthetowers..................................116
Figure87:CostComparisoninfoundationconstructionfordifferenttowersandcompanies..........................117
Figure88:Costdifferenceinfoundationconstructionfordifferenttowersandcompanies.............................117
Figure89:ComparativeEntityCostDifferences.................................................................................................118
xii|DesignandAnalysisofSmallScaleWindTurbineSupportStructures
LISTOFTABLES
Table1:ProsandConsofEachTowerType.........................................................................................................11
Table2:CostAnalysisParameters........................................................................................................................13
Table3:Table5SANS101603:2011....................................................................................................................18
Table4:Table1SANS101603:2011....................................................................................................................19
Table5:Table2SANS101603:2011....................................................................................................................20
Table6:Table4SANS101603:2011....................................................................................................................21
Table9:Equalleganglesectioncompressioncapacity........................................................................................33
Table10:SASCHTable6.1....................................................................................................................................34
Table11:Spreadsheetextractshowingflangedetails........................................................................................51
Table12:MomentResistanceoftheFlanges.......................................................................................................52
Table13:OrdinaryBoltstrengthcalculationsforSteelLatticeTowerConnections............................................56
Table14:BoltCapacityCalculationsforFatigueforLatticetowerConnections..................................................56
Table15:MonopoleLoadCaseCombinations.....................................................................................................72
Table16:MaximumdisplacementmagnitudesofMonopoleforServiceabilityLimitState................................74
Table17:Monopole,LinearBucklingFactors.......................................................................................................79
Table18:FatigueAnalysisofSteelMonopoleBase.............................................................................................89
Table19:FatigueAssessmentforSteelMonopoleandvortexshedding.............................................................90
Table20:LoadCaseCombinations,LatticeTower...............................................................................................95
Table21:WindloadingontheSteelLatticeTower,displacementandstressresults.........................................95
Table22:LinearBucklingFactorsSteelLatticeTower,WindLoadCase1...........................................................99
Table23:LinearBucklingFactors,SteelLatticeTower,WindLoadCase2........................................................100
Table24:FatigueAnalysisofsteellatticetower................................................................................................108
Table25:CostcomparisonofFabrication,Fabricator1.....................................................................................111
Table26:Fabricationcosts,Monopole,Fabricator2.........................................................................................111
Table27:FabricationCosts,LatticeTower,Fabricator2....................................................................................112
Table28:TableofItemdescriptionsforconstruction........................................................................................113
Table29:TableofComparativecostsforeachitemoftheMonopoleTower'sconstruction...........................113
Table30:TableofcomparativecostsforitemsassociatedwiththeLatticeTowers'construction...................114
Table31:SummaryTableofMonopoleCostsfromconstructiontofabrication...............................................114
Table32:SummaryTableofLatticetowerCostsfromconstructiontofabrication...........................................114
Table 33: Overall difference in costs between the monopole and steel lattice tower including construction
costs....................................................................................................................................................................114
Table34:SummaryTableofCostComparisonsbetweenallentitiesaveragedprices......................................118
xiii|DesignandAnalysisofSmallScaleWindTurbineSupportStructures
LISTOFSYMBOLS:
[K]globalstiffnessmatrix
[Kg]globalgeometricstiffnessmatrix
[M]globalmassmatrix
{x}vectorofthebucklingmodes;orvibrationmodevector
ainductionfactor
Arotordiskarea
A,B,C&Dtheterraincategories
Abcrosssectionalareaoftheunthreadedshankofthebolt
Aggrosseffectiveareaofasection
Amshearareaoftheeffectivefusionface
Aneneteffectiveareaofasection
Apeffectiveareaofthecompressedflangeplate
Awareaoftheeffectiveweldthroat,plugorslot
bdiameter
bPmmaximumtensilestrengthoftheconnection
c0(z)topographycoefficient
cf,forcecoefficient
CPpowercoefficient
cp,0,hdbasepressurecoefficient
cp,0minbvalueoftheminimumpressurecoefficient
cpeexternalpressurecoefficient
cpiinternalpressurecoefficient
Cprobprobabilityfactor
Crfactoredaxialcompressiveresistance
cr(z)roughness/heightcoefficient
CWwarpingconstantofasection
ddiameteroftheunthreadedshankofthebolt
Ddiameterofthewasherfaceoftheboltheadornut
xiv|DesignandAnalysisofSmallScaleWindTurbineSupportStructures
Diequivalentdiameterofthecirculartube
Dpboltpitchcirclediameter
dpdiameterofthebolthole
EModulusofElasticity
Ffrfrictionforce
futensilestrengthoftheparentmetal
fuultimatetensilestrengthofthebolts
FWwindforce
FW,eexternalwindforce
FW,Iinternalwindforce
FxThorizontal,perpendicularstaticdesignwindload
fyyieldstrength
FzTverticalselfweightdesignloadfromtherotorandthenacelle
GModulusofRigidity
G,Deadpartialfactorfortheultimatelimitstateofselfweightloading
Q,windpartialfactorregardingthelimitstate
Imomentofinertiaofasection
Jtorsionconstantforthesection
ksurfaceroughness
ltributarylength
lpgriplength,equaltotwicethethicknessoftheflange
mpfullplasticmomentperunitwidthoftheflange
Mrfactoredmomentresistance
Muappliedmoment
MxTorMyTdesignbendingmomentduetorotoroverhangandpseudostaticwindloading
MzTdesigntorsionmomentduetopseudostaticwindandselfweightloads
Ptensileforceappliedtothespecimen
Ppyieldloadoftheflange
PPmaximumloadoftheflange
xv|DesignandAnalysisofSmallScaleWindTurbineSupportStructures
Psboltseparationload
qp(z)peakwindspeedpressure
Rpryingforce;orradiusofgyration;orrotorradius
ReReynoldsnumber
Toboltpreload
Trtensileresistance
Tsseparationforce
Tuappliedtension
TYyieldstrengthofthebolt
vkinematicviscosityoftheair(v=15x106m2/s)
Vundisturbedfreestreamvelocityoftheairfarfromtheturbine
Vbbasicwindspeed
Vb,0fundamentalbasicwindspeed
vp(z)peakwindspeed
Vrshearresistance
Vturbinevelocityoftheairasitpassesthroughtheturbinebladesinmotion
Vuappliedshearforce
VWmeanfreestreamwindvelocity
W(z)distributedloadalongthetower
weexternalwindpressure
wiinternalwindpressure
Xutensilestrengthoftheweldmetal
Zheightabovegroundlevel
Zheightabovethegroundlevel
Z,elasticeffectivesectionmodulus
Z0heightofthereferenceplane,anddefinedintable1ofSANS101603:2011
Zcheightbelowwhichnofurtherreductioninwindspeedisallowedasdefinedintable1ofSANS10160
3:2011
zereferenceheight
xvi|DesignandAnalysisofSmallScaleWindTurbineSupportStructures
Zggradientheight,asdefinedintable1ofSANS101601:2011
Zplplasticeffectivesectionmodulus
exponentasdefinedintable1ofSANS101603:2011
A,positionoftheflowseparation
minpositionoftheminimumpressure,indegrees
bladepithangle;oristheangleoftheaxisoftheweld
tipspeedratio;ortheeffectiveslenderness;orbucklingloadfactor
densityoftheair
utensilestrengthoftheflangematerial
yyieldpointoftheflangematerial
materialfactor;oristhesolidityratio
bboltmaterialfactor
endeffectfactor
rotorspeed;oristhecircularfrequency
xvii|DesignandAnalysisofSmallScaleWindTurbineSupportStructures
LISTOFABBREVIATIONS:
BSBritishStandards
CHSCircularHollowSection
DLDeadLoad
EWEAEuropeanWindEnergyAssociation
FEAFiniteElementAnalysis
GRFGustResponseFactors
HAWTHorizontalAxisWindTurbine
LCLoadCase
SANSSouthAfricanNationalStandards
SASCHSouthAfricanSteelConstructionHandbook
SLSServiceabilityLimitState
SWSelfWeight
SWETStellenboschWindEnergyTechnologies
ULSUltimateLimitState
VAWTVerticalAxisWindTurbine
ZARSouthAfricanRands
xviii|DesignandAnalysisofSmallScaleWindTurbineSupportStructures
CHAPTER1:INTRODUCTIONTOTHEDESIGNANDANALYSISOFSMALLSCALE
WINDTURBINES
Therequirementforcleanenergyproductionworldoverhasincreasedsignificantlyinthelastfewyears.The
Kyoto Protocol which was introduced for enforcement on the 16th of February 2005 was a driving force for
suchrequirements.TheKyotoProtocolwasimplementedwiththeviewofreducinggreenhousegasemissions
producedbyleadingeconomiesworldwide,byatleast5percentbelowtheemissionlevelsof1990between
theperiods20082012.
Atechnologywhichhasadvancedimmeasurablyasaresultofthenecessityforgreenenergyproductionisthe
wind power industry. One of the most important aspects of a wind turbine is its supporting structure. The
towerofawindturbineneedstobesufficientlyreliableandstructurallysoundtoensurethatthedesignlifeof
thewindturbinemachineisunaffected.Thetoweralsoneedstobeofthecorrectheighttoensurethatthe
fullpotentialofenergycaptureisrealised.
The supporting structure of a wind turbine constitutes up to as much as 30% of the total costs of a wind
turbine. The most common wind turbine supporting structures seen worldwide today are Steel Monopole
Towers. The large cost proportion of the tower compels the industry to investigate the most feasible
alternativesupportingtowerstructuresandthuspromptedtheresearchdevelopedinthisthesis.Thefocusis
onsmallscalewindturbines(<50kW),morespecifically,a3kWWindTurbine.Theproposedalternativedesign
forthesupportstructuresofsmallscalewindturbinestothepresentlyusedSteelMonopoletowerwasaSteel
Latticetower.
Themeritofamorecosteffectivesupportingtowerforawindturbinecouldhavesignificanteffectsonthe
number of wind turbines that could be developed on a potential wind farm, as well as extremely positive
outcomesfortheWindindustryinSouthAfrica.
Theobjectivesofthisthesisare:
1.
2.
3.
4.
5.
ObtainingdesigncriteriaforthedesignofSteelMonopoleandSteelLatticeTowers
DevelopingrationaldesignmethodsforSteelMonopoleandSteelLatticeWindTurbineTowers
CreatingFiniteElementModelsofeachoftheaforementionedTowers
Assessingeachelementofdesignfromacostperspective
Developingafeasibilityanalysistocomparethefeasibilityofeachofthesupportingstructuresfrom
thefactorsdeterminedintheirdesign.
In order to develop an understanding of Wind Turbines, existing literature that described the way in which
wind was transformed intoenergy as well as the kindsof loadings that couldbe expected on the towers of
windturbineswasconsulted.InresearchingSteelMonopoleandSteelLatticeTowers,phasesofthelifecycle
ofsuchstructureswhichwouldpossiblyinfluencethedecisionmakingofwhichonetousewereconsideredby
investigatingtheProsandConsofeachtower.Thedevelopmentoftheliteraturewhichwasconsultedisdealt
withinChapter2.
ForathoroughcostevaluationofaSteelMonopoleandaSteelLatticetower,eachcomponentthatisrequired
fromadesignpointofviewneedstobereviewed.Becausemostoftherequiredinformationregardingwind
turbinetowersisproprietaryinformation,itwasdecidedupontodesigneachofthetowersfromscratchsoas
tohavethebestpossibleunderstandingofalloftheelementswhichwouldneedtobeevaluatedinthecost
analysis.
1|C h a p t e r 1 : I n t r o d u c t i o n
The South African National design codes (SANS 10160:2011, and SANS 101621:2005) were consulted in
conjunction with elements from the Eurocode (EN31) and the BS 8100 code in order to collectively form a
rationaldesignprocedureforthetwoSteelTowersdesignedinthisthesis.
ThedesignoftheSteelMonopolewasauniqueinterpretationoftheeverydaytaperedsteeltowersusedfor
windturbines,asitwasdesignedasanassemblyofpremanufacturedcircularhollowsections.Thepurposeof
thiswastodesignthemostcosteffectievepossibleSteelMonopoletocomparetotheproposedSteelLattice
tower.TheSteelLatticetowerwasasimpleandrobustdesign,suitingasmanyaestheticqualitiesaspossible
whilestilladheringtostability.ManyelementsofthedesignofaSteelLatticecommunicationtowerarethe
sameasforaSteelLatticeWindTurbineTower,whichimpliesthatonehascommunicationtowerpublications
to consult for design guidance. Chapter 3 leads one through the process of design for the Steel Lattice and
SteelMonopoleTowerswithalloftherelevantdesigncodeextracts.
The details of any steel structure culminate in their connections. For the steel Lattice Tower gusset plate
connectionsneededtobedesigned,andweredonesofromageometricperspectiveusingAutodeskInventor
ProfessionalandSANS101621:2005toverifytheirstrength.
The Steel Monopole towers circular hollow sections were connected by means of circular ring flange
connections. Because of the lack of South African literature detailing the design of circular ring flange
connections,muchresearchwasdoneintothemostlikelyyieldlineswhichwouldoccurindifferentpossible
modesoffailureinordertoassesstherationaldesignoftheSteelMonopolesconnections.
Chapter4dealswiththedetailsoftheconnectionsofthetwoSteelTowers.
An important aspect to consider ensuring completeness of cost analysis is the design of the foundations of
eachofthetwotowers.ThenatureofthedifferentbaseconnectionsoftheSteelLatticeandSteelMonopole
towers resulted in different foundation designs. The construction and building quantities of each of the
foundations were considered in the feasibility analysis. The designs of the foundations are covered in
Chapter5.
The design calculations executed in Chapter 3 needed to be verified and perhaps even altered through the
iterativedesignprocessofFiniteElementmodellingandcodeverification.BoththeSteelMonopoleandthe
SteelLatticeTowersweremodelledinthefiniteelementpackageStrand7.Allofthedifferentanalyseswhich
wereperformedonthetowersaredescribedandpresentedinChapter6.
In Chapter 7, the feasibility from fabrication to construction of each of the towers is analysed. Each of the
aspects which were identified through the design process from the structural design, construction and
manufacturingwhichwaspertinenttothecostanalysiswasconsidered.Theoutcomeofthefeasibilityanalysis
inChapter7providesworthwhileprospectsforthefutureofwindenergydevelopmentinSouthAfrica.
Afinalchapterwhichconcludestheresearchconductedforthisthesisandmakespossiblerecommendations
astowhereitmightleadnextisChapter8.
2|C h a p t e r 1 : I n t r o d u c t i o n
CHAPTER2:LITERATUREREVIEW
INTRODUCTION
Aliteraturereviewispresentedinordertovisittheliteraturethatpresentlyexistsonaparticulartopic.The
analysis and design of small scale wind turbines is a topic that has been researched in many parts, the
accumulationofwhichisdevelopedinthisthesis,howeverwhatissignificantisthatinitsentirety,ithasnot
beencoveredtotheextentthatitshallbecoveredhere.
Forthisreason,theliteraturethatexistsforeachportionofthisthesisshallbepresentedintheorderinwhich
thetopicsweredeveloped,asfollows:
1.
2.
3.
4.
5.
TheUnderstandingofWindEnergy
i)
Motivationforwindenergy
ii)
Howwindisturnedintoenergy
LoadsInducedbytheWindonaWindTurbineSupportingStructure
SupportingStructures
i)
Towertypes
ii)
Towerheightconsiderations
iii)
Prosandconsofeachtowertype
TheDesignofSupportingTowers
TheAnalysisofSupportingTowers
i)
FiniteElementAnalysis
ii)
Feasibilityanalysis
2.1THEUNDERSTANDINGOFWINDENERGY
In order to be able to adequately design a supporting tower for a wind turbine, it was essential to fully
understandWindEnergy.
2.1.2MOTIVATIONFORWINDENERGY
Thereisanurgentandcompellingcaseforthecleanproductionofenergy.Demandforenergyisgrowingas
the world population expands and industrialisation increases. Current production methods contribute to
pollution of the environment whether we consider nuclear generation or generation of power by coal fired
power stations. The latter have very large carbon footprints when seen in the context of global warming.
Following the Fukushima nuclear disaster in March 2011 and as a direct consequence of the disaster the
GermanGovernment,asoneexample,hasdecidedtophaseoutitsnuclearpowerplantsandtoconcentrate
insteadoncleanermethodsofproducingelectricitywithwindpowertakinganewlead.
Increasingglobalenergycostshavechangedcostdynamicsconsiderablymeaningthatalternativemethodsof
power generation that were previously thought to be financially not viable are now treated with greater
interest,nottheleastofwhichiswindpower.
ItiswellknownthattheKyotoProtocolthatbecameeffectiveonFebruary16,2005,placedanobligationon
industrialisedcountriestoreducetheiroverallgreenhousegasesatleast5%belowtheemissionlevelof1990
from2008to2012(Yeh,TandWang,L,2008:592).Todaystill,mostelectricalenergyisgeneratedbyburning
fossilfuels.Thisformofpowergenerationisthoughttobringaboutadversechangesinweatherconditions.
3|C h a p t e r 2 : L i t e r a t u r e R e v i e w
Furthermore,theburningoffossilfuelshasproducedsevereenvironmentalcontaminationintheformofacid
rain,urbansmog,andregionalhaze.
Although electricity can be generated in many ways using different kinds of energy, there is one common
featuretherotatingofaturbinegenerator.Ineachinstanceofenergyproduction,afuelisusedtoturna
turbine,whichdrivesageneratorthatfeedsagrid.
A turbine is a rotary mechanical device that extracts energy from a fluid flow (air, water, other fuel) and
convertsitintousefulwork.Movingfluidactsonthebladessothattheymoveandimpartrotationalenergyto
therotor.
Turbinesaredesignedtosuittheparticularfuelcharacteristicsusedtodrivethem.Thesameprincipalapplies
towindgeneratedelectricity.Althoughwindmaybeanintermittentsourceofenergy,unlikefossilfuels,itis
freeandcleanandthereisanabundanceofit.Whilepoliticalconsiderationsandeconomicshaveplayedan
important role in the development of the wind energy industry, and have contributed much to its present
success,engineeringstillremainspivotaltoitssuccess(EWEA,2009:31).
WindTurbinesaremostcommonlyutilisedasacollectionofunitsformingawindfarm.Theparticularsofthe
design strategies incorporating economics and site selection are critical to the success of wind farms. Many
systemintegrationstudiesthathavebeencompletedinrecentyearsshowthatthenumberofwindfarmsin
the USA and Europe have substantially increased in recent years. Due to the increasing number and size of
windfarmsacrosstheworld,thecostperkilowatthourofwindpowergenerationhasbeenreduced.Thishas
major implications for the energy producing entities such as Eskom in South Africa. It has even greater
implications for energy production in countries that have several forms of power generation, since the
productioncostsperkilowatthourhavebecomeincreasinglyattractive.(Yeh,TandWang,L,2008:592).
2.1.2HOWWINDISTURNEDINTOENERGY
A wind turbine obtains its power input by converting the force of the wind into torque acting on the rotor
blades. The amount of energy which the wind transfers to the rotor depends on the density of the air, the
rotorarea,andthewindspeed(Theenergyinthewind,2011).
DENSITYOFTHEAIR:
Thekineticenergyofamovingbodyisproportionaltoitsmass(orweight).Thekineticenergyinthewindthus
dependsonthedensityoftheair,i.e.itsmassperunitofvolume(Theenergyinthewind,2011).
Inotherwordstheheaviertheair,themoreenergyisreceivedbytheturbine.Airisdenserwhenitiscoldas
opposedtowhenitswarm.Athighaltitudes,theairislessdense.
ROTORAREA:
Therotorareadetermineshowmuchenergyawindturbineisabletoharvestfromthewind.Sincetherotor
areaincreaseswiththesquareoftherotordiameter,awindturbinewhichistwiceaslargecanharvestfour
timesasmuchenergy(Theenergyinthewind,2011).
Of the Important elements which should be examined regarding a wind turbine extracting energy from the
wind (some of which were mentioned briefly), are the basic components of the wind turbine, including the
brakes,hub,lowandhighspeedshaft,gearbox,generator,nacelleandtower.Duetothemotionofthewind
turbine as well as the components of which it is made up, a wind turbine may exhibit various different
movements,duetorotorandgeneratorrotation,(Balasetal.,2003:3781)suchvaryingpositionsoftherotor
areextremelyimportanttotakeintoaccountfordifferentloadingconditions.
Wind turrbines can bee designed as eitherupwinnd ordownwind machines.. In a downw ind machine, the wind
passesth
hetowerfirst,,beforeimpin
ngingontherrotor,withjussttheopposite
eforupwindmachines(Ba
alasetal.,
2003:3781).
There is a difference between using the lift orr drag force imposed
i
on the
t blades byy the wind in order to
generateepower.Sailorsdiscovered veryearlythaatitismoree
effectivetousetheliftforcee(perpendicu
ulartothe
wind)forreffectivepro
opulsion.How
weverindiffereentregions,theorientation
nofthebladeesneedstobe
evariedin
orderto harnesstheggreatestpowe
er.Fromthisttherearisetw
wodifferentkiindsofwindtturbines,onessinwhich
thebladeesareconnecctedtoaverticalshaft,nam
medverticalaxiswindturbiines(VAWT), andthoseinw
whichthe
bladesarreconnectedtoahorizonta
alshaft,HAWTT(Hansen,200
08:4).
Horizontalaxiswindturbinesareth
heonesthatsshallbeexam
minedfortherremainderof thisresearch;theyare
alsothem
majorityofwhatisfoundin
npractice(Haansen,2008:4).
Windturrbinescanbeclassifiedasffixedspeed(thherotorandggeneratorrotationalspeeddsareheldconstant)or
variable speed. One of
o the main objectives
o
forr wind turbine
e control is to maximize ppower. The amount of
powerprroducedbyaw
windturbinecanbeexpresssedas:
1
2
WherePisthepowerr,Aistherotordiskarea, theairden
nsity,andVwtthemeanfreeestreamwindvelocity
(Balasetal.,2003:378
81).
Cp is thee power coeffficient. In 191
19 Albert Bettz discovered that no wind
d turbine couuld convert more
m
than
59.3%offthekineticeenergyinwindtomechaniicalenergy.TThisbecamek
knownasBetzzLawandre
esultsina
maximum
m power coeefficient of Cp=0.593. This is a maximum, and to ca
alculate otherr values of th
he power
coefficien
nt,whichisa functionofth
hetipspeedrratio
aandtheblade
epitchangle
,whereis therotor
speedan
ndRistherottorradius,on
necalculatesandusesit todetermine
ethevalueoffthepowerccoefficient
fromtheeappropriate curve,seeFiggure1.Forannoptimumenergyproductiionstrategy,tthetipspeed ratioand
pitchanggleshouldbechosentogive
eanoptimum
mCp.
Figure1:Powercoefficientcurrve,parametissedinaccordancewiththe
ebladepitch angle,imagefrom
http:///www.caspus.eclippse.co.uk/ah/pub
blications/3dmwtu
ucfd.pdf
Theturbineshouldop
perateatthis tipspeedrat io,regardlesssofthewindspeed.Sincetthetipspeed
dratioisa
function ofbothrotorrspeedandw
windspeed,thherotorspeedmustbevariedasthewiindspeedvarries(Balas
etal.,2003:3781).Forthefixedspe
eedturbinethhisisnotposssible,although
hsomeattem
mptismadeto
ooptimize
energybychangingbladepitchtoadjustaerodynamictorqueasthewindspeedvaries.Inthevariablespeed
machine,rotorspeedcanbechangedbycontrollinggeneratortorque(Balasetal.,2003:3781).
The concept of a winddriven rotor is ancient, and electric motors were widely disseminated, both
domesticallyandcommerciallyinthelatterhalfofthe20thCentury.Makingawindturbinewiththehistorical
volumeofknowledgeandunderstandingofwindenergyharnessingseemssimplebutitisamajortechnical
challengetoproduceawindturbinethat:
1.
2.
Meetsspecifications(frequency,voltage,harmoniccontent)forstandardelectricitygeneration,with
eachunitoperatingasanunattendedpowerstation;
Copes with the variability of the wind (mean wind speeds on exploitable sites range from 5m/s to
11m/s,withsevereturbulenceintheearthsboundarylayerandextremegustsupto70m/s),and;
Competeseconomicallywithotherenergysources(EWEA,2009:63).
3.
The function of a modern powergenerating wind turbine is to generate steady quantities of network
frequency electricity. Each wind turbine must function as an automatically controlled independent mini
power station in order for it to fulfil its purpose effectively. The development of the microprocessor has
playedacrucialroleinenablingcosteffectivewindenergytechnology.Amodernwindturbineisrequiredto
workunattended,withlowmaintenance,continuouslyforatleast20years(EWEA,2009:63).
Following the determination of power generated by a wind turbine, a few important design elements that
needtobeconsideredarereviewedingreaterdetailhere.
NUMBEROFBLADES:
Thenumberofbladesisusuallytwoorthree.Twobladedturbinesarecheaperbecausetheyhaveoneblade
fewer;howeverasaresultofjusttwobladestheyrotatefasterandappearmoreflickeringtotheeyes.Three
bladedturbinesappearcalmerandarethereforeoftenpreferred(Hansen,2008:5).
Smallscale,multibladed(morethan3)turbinesarestillinuseforwaterpumping.Theyareofrelativelylow
aerodynamic efficiency but, with the large blade area, can provide a high starting torque. This enables the
rotortoturninverylightwindsandsuitsawaterpumpingduty(EWEA,2009:66).Ingeneraltherearesmall
benefitsforrotorshavinganincreasingnumberofblades.Thisrelatestominimisinglossesthatoccuronthe
tipsoftheblades.Theselossesare,inaggregate,lessforalargenumberofnarrowbladetipsthanforafew
wideones.
In rotor design, an operating speed or operating speed range is normally selected first, taking into account
issuessuchasacousticnoiseemissionaswellastheflickeringeffectontheeyementionedabove.Withthe
speedchosen,itfollowsthatthereisanoptimumtotalbladeareaformaximumrotorefficiency(EWEA,2009:
67).Thenumberofbladesis,inprinciple,open,butmorebladesimplymoreslenderbladeswouldberequired
forthefixed(optimum)totalbladearea.
It is hard to compare the two and threebladed designs on the basis of costbenefit analysis. It is generally
incorrecttosupposethat,inatwobladedrotordesign,thecostofoneofthethreebladeshasbeensaved,
andthisisasaresultofthepowerwhichisgeneratedbytwobladesofatwobladedrotorwhichdoesnot
equatewiththepowergeneratedbytwobladesofathreebladerotor.
Theimportantfactorintermsoftherotorsfeasibilityandhowmanybladestouseapartfromthenoiseand
visual effects is how cost effectively the different rotors can produce a kilowatt hour. Two blade rotors
generallyrunatamuchhighertipspeedthatthreebladedrotors,somosthistoricaldesignswouldhavenoise
problem.Thereishowevernofundamentalreasonforthehighertipspeed,andthisshouldbediscountedina
technicalcomparisonofthedesignmeritsofthetwoversusthreeblades(EWEA,2009:68).
PITCHVERSUSSTALL:
Thetwoprinciplemeansoflimitingrotorpowerinhighoperationalwindspeedsarestallregulationandpitch
regulation. The importance of having means by which to limit the power lies in the fact that turbines are
designedstructurallytowithstandhighwindsandstormsthataffecttheturbinestatically;whenthebladesare
notturningatthetimeofthehighwindsandstorms.Highwindshavethepotentialtodestroyawindturbine
ifitsbladesareturningandthespeedatwhichtheyarerotatingisnotcontrolled.Stallregulatedmachines
requirespeedregulationandasuitabletorquespeedcharacteristicintrinsicintheaerodynamicdesignofthe
rotor. As wind speed increases and the rotor speed is held constant, flow angles over the blade sections
steepen. The blades become increasingly stalled and this limits power to acceptable levels, without any
additionalactivecontrol.Installcontrol,anessentiallyconstantspeedisachievedthroughtheconnectionof
the electric generator to the grid. Stall control is a subtle process, both aerodynamically and electrically
(EWEA,2009:68).Stallcontrolenablesthewindturbinetoessentiallybecomelesseffectiveasthewindspeed
increasessoastoprotecttherotorfromrotatingtoofast.Aswasmentionedaboveitisameansofcontrol
thatisimplementedbymeansofdesigningthebladesinsuchawaythatthespeedcontrolisintrinsictotheir
design.
In summary, a stallregulated wind turbine will run at approximately constant speed in high wind without
producing excessive power and yet achieve this without any change to the rotor geometry, or the rotor
spinningoutofcontrolinamannerthatcouldbedetrimentaltotheturbineorturbinetower.
Thealternativetostallregulatedoperationispitchregulation.Thisinvolvesturningthewindturbineblades
about their long axis (pitching the blades) to regulate the power extracted by the rotor. In contrast to stall
regulation,pitchregulationrequireschangesofrotorgeometrybypitchingtheblades.Thisinvolvesanactive
control system, which senses blade position, measures output power and instructs appropriate changes of
bladepitch.Theintroductionoftheincreasednumberofpartsandtheactivepitchcontrolwhichwouldneed
moremaintenancethanastallcontrolledsystemincreasestheinitialandongoingcostsofthissystem.The
objectiveofpitchregulationissimilartostallregulation,namelytoregulateoutputpowerinhighoperational
windspeeds(EWEA,2009:68).
2.2LOADSINDUCEDBYTHEWINDONAWINDTURBINESUPPORTINGSTRUCTURE
Theloadsinducedonthesupportingtowersasaresultofthewindaredividedintotwodifferentclasses.First
thewindactingonthetowerasapressureforceisdeterminedinaccordancewithSANS101603:2011.Second
the overturning force that the wind causes from its interaction with the rotor of the wind turbine is
determined.
The Ultimate and Serviceability limit states for the aforementioned wind loads are determined by the load
factorsstipulatedinSANS101601:2011aswellasthewindspeed.
In examining the wind in greater detail, the following observations were made. The conversion of the wind
whichpassesthroughawindturbineintoenergyasapoweroutputisgreatlydependantonBetzLawaswas
discussedinsection2.1.2.Withthisinmind,itisonlyupuntilacertainwindspeedthattheefficacyofwind
energyproductionincreases.
InFigure2,atypicalPowerCurveforaWindturbineisshown.AtthecutinwindspeedofV1,powerstarts
beingproduced,inRegion2.Theamountofpowerthatcanbeproducedbyaparticularwindturbinereaches
its capacity at V2. V3 indicates the wind turbines cutout wind speed, which is the point at which the rotor
turnsoutofthewindtoprotectitselffromdamageinhigherwindspeeds.
Forthewindturbinedesignedinthisthesis,the3kilowattsofpowerdesiredareproducedatawindspeedof
11m/sandafrequencyof300rpm.Thecutoutwindspeedisat16m/s.Forthewindturbinedesignedinthis
thesis, th
he rotors effeective area perpendicular to the wind direction afte
er cut out is hhalf of the arrea which
facestheewindbeforeecutout.The rotorareaisanimportanttparameterinthecalculattionoftheovverturning
forcedevvelopedbyth
hewindonth
herotor.For thesevalues,theforcesacctingonthettowerscalcula
atedfrom
the diffeerent wind sp
peeds, result in the Servicceability Limitt state loads calculated wiith the wind speed of
16m/s, aand the Ultim
mate limit statte loads calcuulated with the peak wind
d speed desccribed by SAN
NS 10160
3:2011.
EachoftheaforementtionedService
eabilitylimitsstateandultim
matelimitstatteloadsarethhenmultiplied
dwiththe
loadingfactorsasdesccribedbySAN
NS101601:20011.
The otheer load that needs
n
to be ta
aken into connsideration is the massof the windturbbinenacelle and
a rotor,
whichin combination haveonecen
ntreofmass. Forthe3kW WindTurbine
eTowersdesi gnedinthistthesis,the
combinedmassoftheenacelleandrrotoris125kg .
Figgure2:TypicaalWindTurbinePowerCurrve
2.3SUPPORTINGSTRUCTURES
The focu
us of this thessis is the desiggn of two suppporting towe
ers fora3kW
W wind turbin e generator. The most
widely used tower tyype for this purpose
p
has bbecome the Steel
S
Monopo
ole Tower. It was decided
d upon to
investigate the feasib
bility of the equivalent
e
deesign of a Ste
eel Lattice To
ower to be ccompared to the Steel
Monopole. The existing literature pertaining too such design
n was difficult to obtain. Having deterrmined in
section22.2theforcesthatshallbea
actingonthe tower,furthe
erimportanta
aspectswhichneedtobeco
onsidered
indesign
ningtwosuchtowersareexxaminedhere .
2.3.1TOWERTYPPES
Towers ffall into categories: free standing
s
and guyed. Free
estanding tow
wers also knoown as selfsupporting
towersandarejustth
hat,freestanding.Theydeependonadeepormassivvefoundationntopreventtthetower
fromtop
pplingoverin highwinds,a
andtheymusttbestrongen
noughinterna
allytowithstaandtheforcesstryingto
bendtheetowertotheeground.Free
estandingtow
wersaremorreexpensivetthanguyedtoowersbutthe
eytakeup
lessspace,andinalmo
ostallinstancesarebetter suitedtowindturbinesupportingtowerrs.
FREESTANDINGTOWERS:
Therearetwotypesoffreestandingtowers.Themostcommonisthetrussorlatticetower,socalledbecause
it resembles the lattice work of trellis. The Eiffel tower is the bestknown example of a truss tower. The
monopoletowerisanotherformoffreestandingtower(Gipe,2004:151).Trusstowersaretypicallymorerigid
thanmonopoletowers.Towerscanbedesignedtowithstandanyload,butasthesizeofthewindmachine
increases,sotoo,dotheweightandcostofthetowersupportingit.Thesameistrueasthetowerincreasesin
height.Thecomponentsbecomeheavier,hardertomove,andmorecostlytoship(Gipe,2004:151).
LATTICESTYLETOWERS:
Truss towers for small wind turbines can be assembled from a series of 6m sections (suitable for regular
transportation).Forsmallwindpowermachines,thesectionsmaybepreassembledandweldedtogetherprior
to delivery. For household sized turbines, the tower is shipped in parts, and must be assembled on site.
Installation of truss towers usually requires a crane when they become large, and in order to lift the wind
turbine onto the tower one the tower is erected. The tower is assembled on the ground, then hoisted into
place and bolted to the foundation (Gipe, 2004: 152). Lattice towers that use welded steel profiles typically
representcosteffectiveandtestedsolutions.
Latticesteelstructureshavebeenwidelyusedformanylargeutilitiesincludingpowertransmissionstructures
andtelecommunicationtowers(Leeetal.,2006:709).
Conventionallatticesteeltowersusedforpowerlines,thedesignofwhichwouldbecomparabletothedesign
of a turbine tower, usually comprise of angle (Lsection) members. Typically there are bolted connections
between secondary horizontal or bracing members; with the main members which are eccentric, and the
secondary member is connected to one leg only. The towers are designed to resist designfactored loads at
failureduetoyielding,buckling,fractureandotherlimitbehaviours(Leeetal.,2006:709).Althoughthewind
turbine tower would be similar in appearance, the loadings on it would be unique as a result of the wind
turbinegeneratorattachedtothetopofit.
It has been noted that lattice tower structures with angle section members can be very difficult to analyse
because of the complicated threedimensional behaviour of the tower, in particular when considering large
deformationandinelasticmaterialresponse(Leeetal.,2006:709).Thisofcoursewouldalsodependonthe
kindofsoftwareavailableforanalysis.
Usingfiniteelementanalysis,trussmodelshavebeenfrequentlyadoptedforthelinearanalysisoflatticesteel
tower structures. However, it is well known that the truss model is not a good choice for obtaining the
accuratenonlinearresponseoftowerstructures.Theinelasticlargedeformationanalysisusingproperthree
dimensionalbeamcolumnfiniteelementsisnecessary(Leeetal.,2006:709).Andhenceasuitablechoiceof
finiteelementsoftwaretobeabletoperformsuchanalysesneedstobemade.
Forthereliablefailureanalysisoflatticesteeltowerstructures,thefollowingarenecessary:
1.
Use of reliable threedimensional Lsection beam finite elements considering the combination of
biaxialbending,axialstretchandshearingbehaviours;
2. Modellingeccentricitiesofconnectionsformembersconnectedonlyononeleg;
3. Considerationofgeometricalandmaterialnonlinearities;
4. Useofproperconnectionmodelsforthevarioustypicaljoints(Leeetal.,2006:710).
Theneedtodesignalatticetowerforresonantdynamicresponseduetowindloadariseswhenthenatural
vibrationfrequency(fundamentalfrequency)ofthestructureislowenoughtobeexcitedbytheturbulenceof
thenaturalwind(Harikrishnaetal.,1999:149).
The structural failure which is directly attributable to gust action emphasises the importance of these
parametersinarrivingatthegustwindload.Thegustresponsefactors(GRF)thatwillaccountforinfluenceof
theseimportantparameters,isameasureoftheeffectivedynamicloadproducedbygusts,andisintendedto
translatethedynamic response phenomenon producedby gust loading intoa simpler factored staticdesign
criteria(Harikrishnaetal.,1999:149).IntheSouthAfricanNationalStandards(SANS),thegustresponsefactor
has been accounted for in the above mentioned ways, in terms of having the gust loading being taken into
accountinthestaticpeakwindpressurecalculatedinaccordancewithSANS101603:2011.
Currently, the wind sensitive structures are designed using a semi analytical approach with a simple model
relating the upwind turbulent velocity fluctuations and fluctuating forces on the structure. In this approach,
the dynamic response is treated using random vibration theory and modal analysis (Harikrishna et al.,
1999:149). In most of the international design codes and standards, the gust response factor for the modal
coordinateiscomputedusingtheaboveapproachandthesamevalueisconsideredforallotherloadeffects
suchasbendingmoment,shearforceetc.Itisalsoassumedthatthefirstmodeofthestructurevarieslinearly
with height and the contribution of higher modes of vibration is neglected which in turn makes the gust
responsefactorconstantforthewholeheightofthestructure(Harikrishnaetal.,1999:149).
However, in the design of lattice towers, there is an apparent need for broadening the basis of design to
include more explicitly, the load effects, such as the largest deflection, bending moment and shear force.
TheseshallbeaccountedforintheaccuratemodellingcapabilitiesofStrand7Finiteelementpackage,aswell
asthroughhandcalculationsdoneinaccordancewithSANS101601:2011andSANS101621:2005.
MONOPOLETOWERS
Nearly all medium sized wind turbines are installed on monopole towers, though there are some notable
exceptions.DuringthegreatCaliforniawindrushofthe1980s,anequalnumberofturbineswereinstalled
onlatticeandmonopoletowers.
Consideringthescaleoftodaysturbines,however,aswellasaestheticdemands,thesehaveledtothealmost
exclusive use of gently tapered monopole towers. Many observers consider freestanding monopole towers
moreaestheticallypleasingthantrusstowers.
Thisiscertainlytrueincloseupviewsofthetowers,butitisntalwaysthecase.Surprisingly,monopoletowers
canbemorevisibleatadistancethanlatticetowers,especiallyinsilhouette.Inaridregionslatticetowerstend
to blend into the landscape more easily (Gipe, 2004: 154). Pole or tubular towers are more expensive than
lattice towers. The pole towers require a more substantial foundation than truss towers, which spread the
overturningforceoverawiderbase,andifdesignedcorrectly,shallnottransmitthelargeoverturnmoments
tothebase,throughtheappropriateuseofpinnedconnections.
The design considerations for monopole steel towers cannot be based on a particular design procedure
outlinedinaspecificcode.Ratherthroughrationaldesignmethodsandadaptingtheportionsoftheexisting
steeldesigncodestotheidentifiedelementsofthemonopoletowersthatcanbedesignedusingthecodes
provisions,adesignprocedurehastobeestablished.
This can prove very difficult and time consuming from the point of view of special load and response
phenomenawhichneedtobeconsideredandareparticulartosuchtypesofstructures.Anexampleofsucha
phenomenon is Vortex Shedding, which is pertinent for slender structures with circular cross sections
(Monopole towers). Design considerations such as these shall be examined in great detail in chapters to
follow.
2.3.2TOWERHEIGHTCONSIDERATIONS
Windspeedincreasessharplywiththetowerheight,causingamajorincreaseintheelectricityoutputofthe
system(TurbineTowers,2011).Thesamesmallturbinecanincreaseitspoweroutputby30%ormoreifthe
height of its tower is 30m instead of 18m (Turbine Towers, 2011). However, the additional cost of a higher
towerneedstobeconsidered.
Usuallytheheightofthetowersandthedimensionsandcapacityoftheturbinesareintrinsicallyassociated
(TurbineTowers,2011).
Theheightofasmallturbinetowershouldtakeintoaccounttheheightofthesurroundingobstacles:toattain
maximumefficiency,theheightofthetowershouldallowthebottomoftheturbinebladestobe10metersor
moreabovethetopofanyobstaclewithin100metersofthetower(TurbineTowers,2011).
However,thetowerheightdependsalsoontheturbinemodelandcharacteristics.Towersandwindturbines
are often supplied together, and manufacturers demand their turbines to be mounted on their towers
(TurbineTowers,2011).
AnothervariabletobetakenintoaccountaretheLocalBylaws,theymayimposeheightrestrictionstowind
towers: rules imposing maximums of 912 meters are common, particularly in residential areas which may
haveanegativeimpactontheelectricpowergeneratedbywindturbines(TurbineTowers,2011).
2.3.3PROSANDCONSOFEACHTOWERTYPE
In examining the life cycle of a wind turbine the execution phase incorporates a number of elements to
consider, such as the fabrication of the tower, how its transported, and the ease of installation and
maintenanceassociatedwiththetowertype.Table1showstheProsandConsofeachtowertypeinallthe
executionphaseelementstobeconsidered.
Table1:ProsandConsofEachTowerType
ExecutionPhaseElements
Fabrication
Transportation
Installation
SteelLatticeTower
The elements of a steel lattice
tower, typically angle sections are
easytofabricate,andthesections
of tower which are to be pre
assembled are done so through
simple methods of bolting or
welding.
Steel lattice towers can be
transported easily, and cost
effectively as large numbers of
elements can be stacked together
andtransportedatonetime.
Steellatticetowerscanbeerected
onsitebymeansofelementsthat
are bolted together. The level of
expertise required to assemble a
steel lattice wind turbine tower is
no more than would be required
for
the
everyday
telecommunication steel lattice
towersthatareerected.
SteelMonopoleTower
Tapered steel monopole towers
are specially fabricated from hot
rolled steel. They are far more
complex to fabricate than steel
lattice towers, and as a result of
thataremostcostlytofabricate.
Steelmonopoletowersneedtobe
specially transported in large
sections, or on one vehicle of
abnormal sized load. Great care
needs to be taken when
transporting large thin walled
sectionoftubulartower.
Steel monopole erection needs to
be highly supervised. The tubular
sectionsofhotrolledsteelneedto
be crane lifted into place and the
connections fixed in terms of
bolted ring flanges or friction grip
connections.
Maintenance
Steel lattice towers can be

maintained on site. If an element
ofthetowerneedsreplacingitcan
be done so on site. Steel lattice
towers can be protected against
corrosion by means of hot dip
galvanisationduringfabrication.
Steelmonopoletowersneedtobe
returned to their place of
fabrication if structural repairs
needtobedone,whichisacostly
exercise. Steel monopole towers
can be galvanised during
fabrication, however it is a rather
skilledprocessthatneedstomake
surethantheinsidesofthetubular
sections are also well galvanised.
Alternatively the steel monopoles
can be painted and regular
maintenance of the painting is
doneonsite.
ItisclearinTable1thatfromtheelementscommentedon,steellatticetowershavemoreprosthanthesteel
monopoletowersdo.Thisishoweverseeminglystrangewhenthemostwidelyusedtowersaremonopoles.
The one aspect that was not covered in Table 1 which most commonly governs the choice of wind turbine
supportingtowersindevelopedcountriesistheiraestheticappraisal.Itisforthisreasonthattheprosofsteel
lattice towers versus the cons of the steel monopole were presented in Table 1 to exclude the subjective
elementofaestheticpreference.
2.4THEDESIGNOFSUPPORTINGTOWERS
Aswasmentionedintheintroductiontothischapter,thereisnoreadilyavailablecompletedesignmethodfor
thedesignofwindturbinesupportingtowers.Muchoftheinformationthatisneededfromcompaniesthat
presentlymanufacturewindturbinesisproprietary.
ThedesignoftheSteelMonopoletowerforthisthesiswasdoneasanassemblyofthreepremanufactured
circularhollowsectionswhichwereconnectedtooneanotherbymeansofcircularringflangeconnections.
ThedesignoftheSteelLatticetowerforthisthesiswasinspiredbyexistingsteellatticetowers,withallowance
madeforastraightportionatthetopofthetowerfortherotatingbladesofthewindturbinemachine.
ThedesigncriteriawhichareprescribedbytheSouthAfricandesigncodes,inparticularSANS101621:2005,
and SANS 10160, parts 1 and 3: 2011 were examined to design the towers. The extensive sections of the
designcodeswhichwereutilisedfordesignarepresentedinChapter3.
Otherelementsofdesignwhichwereexaminedwerethedetailsoftheconnectionsrequiredforeachofthe
towers,aswellasthefoundationsrequiredforthetwodifferenttowers.
2.5THEANALYSISOFSUPPORTINGTOWERS
Theanalysisofthetwosupportingtowersdesignedforthisthesisincorporatedafiniteelementanalysiswhich
playedanintegralpartofthedesignprocess.Apartfromanalysingthetowersforstructuralstability,acost
analysiswasperformedonthetwotowerstoasseswhichoneofthetwowasmorefeasible.
2.5.1FINITEELEMENTANALYSIS
ThefiniteelementpackageStrand7waschosentoanalysethetwosteeltowers.Strand7hasthecapabilityof
modellingawiderangeofstructures,whichwasattractiveforthedesignofboththeSteelMonopoleaswell
astheSteelLatticeTower.
Strand7 makes it possible to construct models, run analyses and investigate results simultaneously using a
seamless interface. As well as its userfriendliness and extensive operating manual and theory manual
available online, Strand7 has an extensive element library, the ability to model standard and specialist
materialsaswellasarangeofcomprehensivesolversjusttomentionafewaspects.
TheanalyseswhichwereperformedonthetwotowersusingStrand7werethefollowing:
1.
2.
3.
4.
LinearStaticAnalysis
LinearBucklingAnalysis
NonlinearAnalysis
DynamicAnalyses
i)
NaturalFrequencyAnalysis
ii)
Effectsofanoutofbalancerotor
iii)
Effectsofvortexsheddingduetowindaction
iv)
HarmonicFrequencyAnalysis
v)
ModalMassParticipationEvaluation
2.5.2FEASIBILITYANALYSIS
Theimportanceofhavingdesignedeachofthetowersfromfirstprinciplesandhavingdesignedthedetailed
connections required for each tower as well as their foundations was so that each element could be
thoroughlyunderstoodwhentakingintoaccounthowtodoarealisticandthoroughcostanalysis.
Theelementsoftheexecutionphasewhichwerementionedinsection2.3.3andwereconsideredinthecost
analysiswerethefollowing:
1.
2.
Fabrication
Construction
A list of elements pertainingto both of the above wasgiven to a number of different companies who were
asked for a unit price for each of the elements described in Table 2. From this it was possible to do a cost
analysisforthequantitiesofeachitemthatwererequiredforeachofthetowersthathadbeendesignedin
theprecedingchapters.
TheoverallfeasibilityoftheSteelLatticetowerincontrasttotheSteelMonopoletowerwasthenexamined.
Table2:CostAnalysisParameters
Itemdescription
EarthWorks
1.Excavationofthefoundations,includinglevelingandstockpilingofselectedmaterialforreuseinfilling
2.CompactionofexistingsoiltoStandardProctorOptimumDensity
3.Bitumouspaintingandsealingonconcretesurfacesincontactwithearth
ConcreteWorks
4.ReinforcedConcrete
5.Steeltrowelfinishforslabs
6.Highyieldsteelbarreinforcement
7.Castinanchorboltsuptp20mmdiaincludingtemplates
8.Castinanchorboltsbiggerthan20mmdiaincludingtemplates
SteelWork
9.Preparationofshopdetaildrawing,fabrication,painting,deliveryanderectionoflightsteelstructuralsteelwork,
completewithallthenecessarycleats,shopbolts,brackets,gussets,packsandbaseplates.
CHAPTER3:TOWERS
This chapter deals with the design procedure for the two steel towers considered for this thesis. It also
examinesallotheraspectscommontobothsteeltowers.
Thedesignandanalysisoftowersarealwaysintegrated,andobtainingtheoptimaldesignmostoftenentails
aniterativeprocess.Thedesignprocedurebeginswithaninitiallayoutofthegeometryofthetower,which
canbeanextremelychallengingtask.Thetaskofthegeometriclayoutentailsincorporatingallofthespecial
needsofthetower,aswellaskeepingaestheticsandeconomicsinmind.Manyofthesedesignparameters
canconflictoneanother,andfindingabalancebetweenthedesign,planningandconstructioniskey.
THEDESIGNPROCEDURE:
1.
2.
3.
4.
5.
GeometricLayout
i)
Monopoletower
ii)
Latticetower
ExaminingtheboundaryconditionsoftheProblem
Loadingconditions
i)
FromtheWind
a) Onthetowers
b) Asoverturningforcesactingontherotorofthetowers
ii)
WindTurbinemachineeffects
a) Massofthenacelleandrotor
b) Dynamiceffects
ValidationofGeometricdesign
i)
FiniteElementAnalysis(FEA)
ii)
UsingtheFEAresultsincalculationsspecifiedbySANS101621:2005tovalidatethechosen
steelelements.
Iterativedesignbetweenstepsi)andii)inpart4ofthedesignprocedure
Thereafterothergeneralaspectsoftowerdesignshallbediscussed.
3.1THEGEOMETRICLAYOUT
3.1.1THEGEOMETRICLAYOUTOFTHESTEELMONOPOLETOWER
For steel monopole towers, such as the ones shown in Figure 3 below, there are either monopoles with a
constantdiameter,oroneswithvaryingdiameteruptheheightofthetower.
For either a constant diameter monopole or a tapered monopole, the main load that needs resisting is the
wind,eitherasadistributedwindloaduptheheightofthetower,orasanoverturningforcefromthewind
actingontherotor.Theappliedloadsimposedbywindpressuresonthetowerneedtobecarefullyconsidered
for the effects of vortex shedding which may occur as a result of the circular crosssection associated with
monopolesandtheslendernessofwindturbinesupportingtowers.
Thepurposeofvaryingthediameterisbecausethelargestmomentsoccuraboutthebaseofthetower,where
thestabilityisofimportance.Becausethemotionsofthetowerincludehorizontal,verticalandeventwisting
motions, tensions arise at the base of a tower, where it connects with a relatively rigid earth through its
foundation.
14|C h a p t e r 3 : T o w e r s
In order to take thesse tensions in

nto account, the strength of the towe
ers lower co mponents ne
eed to be
adjusted insuchawaaythattheycanresistthossemomentsaandforces.Be
ecausethefoorcesandmom
mentsare
notasgrreatinmagnittudehigherup
ponthetoweer,andinordertopreserve
ematerialsanndonlyprovid
dewhatis
requiredforresistancee,thetowers(mostoftenhheights>20m)havetapered
dprofiles.
Fortheccaseofthe16
6mhigh3kWw
windturbine towerdesign
nedforthisthesis,itwasdeecidedtouse
estandard
circularh
hollowsection
ns(CHS)that areregularlyymanufacture
ed(asoppose
edtocustom madesection
ns).Itwas
decided to use CHS of the same diameter ovver the heigh
ht of the tow
wer. The chooices for the standard
manufacttured CHS were governed
d by the requuirement of cost
c
efficacy. Although thee weight and
d material
usageof standardCHSSishighertha
anthatoftheesameheightttaperedmon
nopole,thecoostofmanufa
acturinga
specialiseedtowerdesignasopposedtousingreaadilymadesecctionsissignifficantlyhigherr.
Figure3:DiffferentMono
opoleProfiles
The 16m
m high tower was divided into 3 sectioons of length smaller than 6m each. Eaach CHS then
n had ring
flangesw
weldedontoeeitherendbyffullpenetratioonwelds.Figu
ure4illustrate
estheringflanngeconnectio
ontype.
Figure4:Ring Flangeconne
ectionforaCH
HS
3.1.2THEGEOMEETRICLAYOU
UTOFTHES TEELLATTICETOWER
Theprim
maryloadthattalatticetow
werofawind turbineissu
ubjectedtois thewind;whhetherasadistributed
loadontthetoweritseelf,oractinga
asanovertur ningforce.Th
hegeometrica
alconfiguratioonofthetow
weristhus
muchsub
bjectedtotheeinfluencethatthewindw
willhaveonit.Differentgeometricallayoouts,suchassquareor
triangulaar cross sectio
onal layouts, will
w have diffeerent wind re
esistance. The
e choice of ellements, such
h as angle
sections,orcircularseectionscreatin
ngthelatticew
willalsoaffectthewindressistanceoftheetower.
For a tow
wer, which sttands alone without
w
any ssignificant maasses or wind
d resisting obbjects attache
ed to it, a
triangulaarcrosssectio
oncomparedttoanequivaleentsquarecrrosssectionhasfarlesswinndresistance.Interms
oftheintternalelemen
nts,rounded elementsarefarlesswind
dresistanttha
anflatanglessections.Soin
norderto
create a free standing tower of the nature deescribed abovve, with the lowest
l
wind resistance, one would
chooseatriangulartow
werwithtubu
ularelements .
Howeverr,roundsectio
onshavecom
mplicatedend connections,andarealsomoreexpensiivethanthee
equivalent
anglesecctions.Also,ttheoptimum geometricalllayout(mentionedinthep
paragraphpreecedingthiso
one)isfor
the instaance in which
h wind resista
ance is the foocus of the laayout optimissation, in an instance in which
w
the
attachedobjectsand massesmayccreatefargreaaterwindresistanceincom
mparisontothhatofthetow
wer,using
a squaree cross sectio
on, and using angle sectio ns may fulfil other criteria. Factors too be considered in the
selection
nprocessofellementsinclud
dethefollowiing:
Practicalproffilesizes
Theslendernessofindividu
ualmembers
Thepriceand
ddeliverytime
e
Rationalandcosteffective
eproduction
Uniformconn
nectiontypesresultingfrom
mmemberchoice
Facilitiesofhotdipgalvaniising
Transportatio
onanderectio
on
The man
nufacturing off angle sectio
ons has relat ively low cossts and the jo
oints resultinng from angle
e sections
consisto
ofboltsandplatesandnow
weldingisreq uiredonsite..Ontheotherrhandthemaanufacturingo
oftubular
sectionsismorecostly,fromthepointofviewofpriceperkilosteelfortubescomparedtoothersections,and
theconnectionsrequiredfortubular/roundsectionsaremorecomplicatedandtimeconsuming.
From the above mentioned reasons and other research, it was decided upon to have a square plan lattice
tower,madefromequalleganglesections.Thetowerwithsquareplantapersinwardfromlevelgroundupto
theheightof13.5m,atwhichpointithasparallelsidesuptothetopat16mabovegroundlevel.
3.2THEBOUNDARYCONDITIONSOFTOWERDESIGN
Both the steel monopole and the steel lattice towers were required to be 16m in height. They were both
requiredtohaveadequateconnectionsavailableatthetopofthetowerforthe3kWwindturbinemachine,
andthetowerneededtohavespatialallowancefortherotationofthe4mdiameterbladesoftherotor.
TheSteelMonopoletower,beingofconstantdiameterhadnoproblemallowingspaceforthebladerotations
of the rotor. Due to the nature of the ring flange connection on the base of the monopole tower it was
designedtohaveafixedbase,whichwouldtransferaxialforcesandmomentstothefoundation.
TheSteelLatticetoweraswasmentionedinthegeometricaldesignhadaparalleltopsectiontoallowforthe
rotation of the blades. The lattice design of the tower, allowed for equal leg angle sections to have pinned
connectionsatthebaseofthetowerwhichtransferrednomomentstothefoundation,onlyaxialforces.
3.3LOADINGCONDITIONS
AlthoughtherewillbemanyelementsofthedesignprocedurefortheSteelMonopoleandtheSteelLattice
Towerswhichwilloverlap,suchasloadings,therewillalsobevastdifferencesintermsofhowthoseloadsare
appliedtothetwodifferentstructures.Thedesignaspectsmostcommontobothdesignsaretheappliedloads
associatedwiththe3kWwindturbinemachine.
3.3.1LOADINGCONDITIONSFROMTHEWIND
Intermsofcalculatingthepeakwindpressureactingonthetowerfordifferentheights,andalsotodetermine
thewayinwhichthewindpressurewasappliedtodifferentmembers,SANS101603:2011wasconsulted.For
thesakeofconvenience,thesectionsoftheSANS101603:2011thatwereusedforcalculationsarerepeated
inthischapter.AllSANSextractsareinitalics.
Firstasummaryofthecalculationprocedurefollowedisaddressed,whereaftereachofthestepsneededin
theprocedureareexamined.
SANS101603:2011
7.5WindActions
7.5.1Calculationprocedure
A summary of the calculation procedures for the determination of wind actions is given in table 5 of SANS
101603:2011.
SANS101603:2011
Table5Calculationprocedure
Description
Symbol
Reference
Fundamentalbasicwindspeed
vb,0
Figure1
Basicwindspeed
vb
Equation(1)
Terraincategory
A,B,C,D
Table2
Referenceheight
ze
7.5.2.2
Topographycoefficient
c0(z)
7.3.3
Roughness/Heightcoefficient
cr(z)
7.3.2
Peakwindspeed
vp(z)
equations(3)and(4)
Peakwindspeedpressure
qp(z)
Equation(6)
Internalpressurecoefficient
cpi
8.3.9
Externalpressurecoefficient
cpe
8.3.2to8.3.8
Internalwindpressure
wi
Equation(7)
Externalwindpressure
WindForcecalculatedfromforce
coefficient
we
Fw
Equation(8)
Equations(9)and
(10)
InternalForces
Fw,i
Equation(11)
ExternalForces
Fw,e
Equation(12)
FrictionForces
Ffr
Equation(13)
Table3:Table5SANS101603:2011
Section7:windspeedandwindpower
SANS101603:20117.2BasicValues:
7.2.1 The fundamental value of the basic wind speed, vb,0, is the characteristic 10min mean wind speed,
irrespective of wind direction and time of the year, measured at 10m above ground level in open country
terrain with low vegetation, such as grass and isolated obstacles, with separation of at least 20 obstacle
heights.
7.2.2Thebasicwindspeedshallbecalculatedusingthefollowingequation:
,
Wherevb,isthebasicwindspeeddefinedat10mabovegroundterraincategoryB;
Vb,0isthefundamentalvalueofthebasicwindspeedcorrespondingtothespecificgeographicallocation,which
shallbetakenfromfigure1inSANS101603:2011.
Cprob,isdefinedinsection7.2.3.
7.2.3 The basic values are characteristic values having an annual probability of exceedance of 0.2, which is
equivalent to a mean return period of 50 years. The probability factor, cprob, is given in equation 2 which
follows:
1
1
1
0.98
Where K, is the shape parameter depending on the coefficient of variation of the extreme value distribution
withavalueof0.2,andnistheexponentwithavalueof0.5.
Note;thereturnperiodmaybetakenasthedesignworkinglifeofthestructure(seeSANS101601).
7.3Peakwindspeed
7.3.1Variationwithheight
7.3.1.1 the peak wind speed, vp(z) at a height, z, above the terrain, depends on the terrain roughness and
topographyaswellasbasicspeed,vb,andshallbedeterminedusingthefollowingequation:
,
Wherevb,peak=1.4vb,inthisequation,aconversiontakesplacefromthe10minwindspeed,intermsofwhichthe
basicwindvelocities,vb,0andvb,aredefinedtoa3sgustwindspeed.
Cr(z)istheroughnessfactor,givenin7.3.2;
C0(z)isthetopography(orography)factor,takenas1.0unlessspecifiedotherwisein7.3.3.
7.3.2Terrainroughness
7.3.2.1Theterrainroughnessfactor,cr(z),accountsforthevariabilityofthemeanwindspeedatthesiteofthe
structuredueto
a)theheightabovegroundlevel,and
b)thegroundroughnessoftheterrainupwindofthestructureinthewinddirectionunderconsideration.
Thefactor,cr(z),shallbedeterminedusingthefollowingequation:
1.36
Where
Zistheheightabovethegroundlevel;
Z0istheheightofthereferenceplane,anddefinedintable1ofSANS101603:2011;
Zgisthegradientheight,asdefinedintable1ofSANS101601:2011;
Zcistheheightbelowwhichnofurtherreductioninwindspeedisallowedasdefinedintable1ofSANS10160
3:2011;
istheexponentasdefinedintable1ofSANS101603:2011.
SANS101603:2011
Table1Parametersofwindprofile
1
TerrainCategory
2
Height(zg)
3
Height(z0)
Height(zc)
5
Exponent()
250
0.07
300
0.095
350
0.12
400
10
0.15
7.3.2.2Variousterraincategoriesarespecifiedintable2ofSANS1016003:2011.
7.3.2.3Atlowelevationsabovethegroundlevel,thewindprofile(i.e.magnitudeofthecr(z)factor)isstrongly
influencedbylocalsurroundings,whicharesitespecificandwhichmayintroduceaccelerationofthewindflow.
This is especially relevant with in developed areas i.e. rough terrain categories. No further reductions in the
windspeedbelowcutoffheights,zc,whicharestipulatedintable1above,arepermitted.
7.3.2.4Thevariationoftheroughnessfactor,cr(z),withheight,isgivenintable3ofSANS101601:2011.
7.3.2.5 The terrain roughness to be used for a given wind direction depends on the distance of the terrain
coveredwithauniformroughnesswithangular+15osectorofthisdirection.Smallareas,withadeviationin
theroughness,whichconstitutelessthan10%oftheoverallarea,canbeignored.
7.3.2.6 When there is a choice between adopting two or more terrain categories for a given area then the
terraincategorywiththelowerroughnessshallbeused.
7.3.2.7Thesmothertheterraincategoryintheupwinddirectionshallbeadoptedifastructureissituatednear
achangeofterrainroughnessatthefollowingdistancesandinthefollowingcategories:
a)Lessthan2kmfromthesmoothercategoryA;or
b)Lessthan1kmfromsmoothercategoriesBandC.
InothercasestheproceduredescribedinA.2ofSANS101603:2011maybeused.
SANS101603:2011
Table2TerrainCategories
3
Illustratio
n
Category
Description
Areawithlowvegetationsuchasgrassandisolatedobstacles
(forexampletreesandbuildings)withseparationsofat
canbe
seen
incodes
least20obstacleheights.
7.3.3Terraintopography
7.3.3.1Wheretheterraintopography(forexamplehillsorcliffs)increaseswindspeedsbymorethan5%,these
effectsshallbetakenintoaccountbyusingthetopographyfactor,c0(z).
NotetherecommendedprocedureisgiveninA.3ofSANS101603:2011.
7.3.3.2Theeffectsoftopographymaybeneglectedwhentheaverageslopeoftheupwindterrainislessthan
3o.Theeffectsoftheupwindtopographyhavetobeconsideredtoadistanceoftentimestheheightofthe
isolatedtopographicalfeature.
Sections7.3.4and7.3.5ofSANS101603:2011donotneedtobeconsideredhere.
7.4Peakwindspeedpressure
The peak wind speed pressure, qp(z) at height, z, which includes the mean and short duration wind speed
fluctuations,shallbedeterminedusingthefollowingequation:
0.5
Where,istheairdensity,expressedinkilogramspercubicmetre(kg/m3).
Therecommendedvaluesoftheairdensityasafunctionofaltitudeabovesealevelaregivenintable4ofSANS
101603:2011.
SANS101603:2011
Table4Airdensityasafunctionofsitealtitude
1
Sitealtitudeabovesealevel
(m)
Airdensity(kg/m3)
1.20
500
1.12
1000
1.06
1500
1.00
2000
0.94
Note 1. A temperature of 20o has been selected as appropriate for South Africa and the variation of mean
atmosphericpressurewithaltitudeisallowedforintheabovetable.
Note2.Intermediatevaluesofmaybeobtainedfromlinearinterpolation.
Having determined the peak wind pressure, the details of how the wind loads are applied to the Steel
Monopoleareexaminedfurtherbydeterminingthepressurecoefficientsforcircularhollowsections.
Theforcecoefficientfromsection8.10.2.1asdescribedbyfigure30inSANS101603:2011shallbeusedinthe
determinationofthedistributedwindload.Thisisdiscussedinfurtherdetailinsection3.3.2ofthisthesis.
The way in which the peak wind pressure calculated in SANS 101601:2011 section 7.4 is applied to a steel
latticetowerisuptothedesigner.Thewindcouldbemodelledasapressureforceontheelementsfacingthe
winddirectionorasaseriesofpointorlineloads.AswillbeseeninChapter6,themannerinwhichthewindis
modelledforasteellatticetowerhasaninsignificanteffectontheoverallresponse.
3.3.2L OADINGCO
ONDITIONSINDUCEDBYYTHEWIND
DANDTHEWINDTURBBINEMACHINE
The calculation proceedure for dettermining thee wind inducced loading as
a a pressuree force has now
n
been
examined
d. Next the other action
ns induced oon the towerrs are examined. Figure 4 shows a schematic
s
represen
ntation of thee loads imposed on a windd turbine sup
pporting towe
er. The loads shown in Figure 5 are
pertinentttotheSteelMonopoleTo
oweraswella stheSteelLatticeTower.
Figure5:ActionsInduccedonWindT
TurbineSupportTower
FxT,istheehorizontal,p
perpendicularstaticdesign windloadonrotorandnaccelleathubh eight,andisccalculated
asfollow
ws(Peterson2010,Report,p
page40).
4 1
Where
Q,windisthep
partialfactorrregardingtheelimitstate,q
qp(z)isthewiindpressureaathubheightt,Aisthe
sweptrotorareaofth
hebladesofth
hewindturbinne,andaisth
heinductionfactorofthetuurbine.Theva
alueofa
approach
hes0.33forawelldesigned
dwindturbinee,thisisthevvaluewhichis knownastheeLanchesterBetzlimit,
andisth
hevalueatwh
hichtheenerrgyabsorbed fromthewin
ndturbineism
maximised.Siincetheobjecctiveisto
create a welldesigned
d wind turbin
ne, the value of a shall be
b used as 0.3
33 here. (Peteerson, 2010, page 40).
Although
haisregularlycalculateda
asfollows:
Where,V
Vistheundisturbedfreestreamvelocittyoftheairfaarfromthetu
urbine,andVVturbine,isthevvelocityof
theairassitpassesthro
oughtheturb
binebladesin motion,(roto
orsweptarea)).(Peterson,22010,page40
0).
ThepartialfactorfortheultimatelimitstateistakenasQ,wind=1.5,inaccordancewithSANS101601:2011,
Table 3Partial factors for actions for the ultimate limit state. The value of 1.5 is in accordance with the
ultimatelimitstatepartialfactorforwindactiononslendernonredundantstructuresthatexhibitsignificant
crosswindresponse.Thepartialfactorusedfortheserviceabilitylimitstatethatistobeusedinaccordance
withtheFxTforce,isQ,wind=0.6,inaccordancewithSANS101601:2011Section8.3.1.1Thecombinationof
actionsforirreversibleserviceabilitylimitstates.
FzT,istheverticalselfweightdesignloadfromtherotorandthenacelle.Forthisselfweightcomponent,the
weightoftherotorandnacelleareaddedtogetherandmultipliedwiththeappropriatepartialfactor.Asfaras
theliveloadingintheverticaldirectionisconcerned,thereisntone.Thepartialfactorfortheultimatelimit
state of selfweight loading is a factor of G,Dead=1.2, thisis in accordance with SANS101601:2011, Table3.
ThepartialfactorfortheserviceabilitylimitstateofselfweightloadingisG,Dead=1.1.Thispartialfactorforthe
serviceabilitylimitstateisinaccordancewithSANS101601:2011Section8.3.1.1Thecombinationofactions
forirreversibleserviceabilitylimitstates.
MxT,orMyTisthedesignbendingmomentduetorotoroverhangandpseudostaticwindloadingontherotor
andnacelle.Thismomentissimplytheverticalselfweightdesignloadmultipliedwiththedistancebywhichit
is eccentric, to determine the overturn moment that would cause. The moments are calculated for both
instancesof ultimate limit state and serviceability limit state, with thepartial factorsused in calculating the
selfweightloads.
MzT, is the design torsion moment due to pseudo static wind and selfweight loads from the rotor and the
nacelle.Thismomentholdsnovalueforaparkedwindturbine.Alsoforthepurposeofthisdesignprocedure,
thesmoothbearingsonwhichthewindturbinemachineshallbemountedwhichnaturally(notbymeansof
mechanicalinstruction)turnthenacelleinandoutofthedirectionofthewindprovidelittleornoresistance
to the turning motion. Thus for a running turbine too the torsional moment shall not be considered in this
designprocess,asitisbelievedtohaveverylittleornoeffect.Thiswasinaccordancewiththemanufacturing
specifications of the wind turbine machine, however as a precaution, not shown in this thesis, a torsional
momentwasappliedtothetower,ofmagnitudeequaltotenpercentoftheweightofthenacelleandrotor
withaleverarmequaltotheeccentricitycausedbythecentreofmassandtheeffectswereindeednegligible.
W(z)isthedistributedloadalongthetower,whichisasaresultofthepeakwindpressureandtheexternal
pressurecoefficientcpe.Theloadisappropriatelymultipliedwiththepartialloadingfactorsforultimatelimit
stateandserviceabilitylimitstate,andthefactorsarethesameasthoseusedforthehorizontalstaticdesign
loadFxT.Thedistributedloadiscalculatedasfollowsinaccordancewith SANS 101603:2011 section 7.5.2.4
thewindpressureactingontheexternalsurfaces,we,shallbeobtainedusingthefollowingequation;
Where,qp(z)isthepeakwindspeedpressure;
Z,isthereferenceheightrelevanttotheexternalpressure;
Cpe,isthepressurecoefficientfortheexternalpressure.
As it was discussed in the previous section, the distributed load from the wind shall be calculated with the
forcecoefficientasopposedtotherangeofdifferentexternalpressurecoefficients.
Inordertomaketheabovedistributedpressureforceauniformlydistributedlineloadonewouldmultiplyit
with the reference area upon which it is acting, in accordance with SANS 101603:2011 section 7.5.3 Wind
Forces. Onewouldmultiplythepeakwindpressurebythediameterofthetower,orwidthorangularsteel
elementtoobtainthedistributedwindloadinN/m.
Thustheequationusedtocalculatethedistributedwindloadwilllookasfollows:
Where, the Q,wind partial factor is taken as is was described earlier, and the value of qp(z) is also as it was
before, only the fore coefficient cf, is inserted here, and is calculated as described above. D is the outside
diameter of the tower for the steel monopole and the width of a steel equal leg angle section for the steel
latticetower.
3.4VALIDATIONOFSTRUCTURALDESIGN
ThepartsofthesteeldesigncodeSANS101621:2005thatareusedtovalidatethestructuraldesignofthe
twosteeltowersareexaminednow.Thefirstfewsectionsareutilisedforcalculationsthatconcernboththe
steel monopole and the steel lattice towers. For both towers, the validation of the geometric design begins
withexaminingtheeffectivecrosssectionalareasofmembersintermsoftheirslendernessratiosintension
andcompressionaswellastheclassclassificationoftheelementsineachtower.
Thevalidationofdesignthenexaminesthefollowing:
1.
2.
DesignValidationofMonopoleTower
i)
AxialCompression
ii)
Bending
iii)
Shear
iv)
Combinedbendingandaxialcompression
DesignValidationofSteelLatticeTower:
i)
Effectivecrosssectionalpropertiesofcompressionmembers
ii)
Flexuralbucklingofaxiallycompressedmembers
iii)
Designoftensionmembers
iv)
Combinedbendingandaxialtensileorcompressiveforces
SANS101621:2005
Section10.4:SlendernessRatios
10.4.1General
The slenderness ratio of a member in compression shall be taken as the ratio of effective length, K.L, to the
correspondingradiusofgyration,r.Theslendernessratioofamemberintensionshallbetakenastheratioof
theunbracedlength,L,tothecorrespondingradiusofgyration.
10.4.2Maximumslendernessratio
10.4.2.1Theslendernessrationofamemberincompressionshallnotexceed200
10.4.2.2Theslendernessratioofamemberintensionshallnotexceed300.Thislimitmaybewaiveredifother
means of flexibility, sag, vibration and slack in a manner commensurate with the service conditions of the
structure,orifitcanbeshownthatsuchfactorsarenotdetrimentaltotheperformanceofthestructureorof
theassemblyofwhichthememberispart.
Section11:Widthtothicknessratios:elementsincompression
ClassificationofSections
11.1.1Forthepurposesofthisstandard,structuralsectionsshallbedesignedasclass1,2,3or4dependingon
themaximumwidthtothicknessratiosoftheirelementssubjectedtocompression,andasspecifiedin11.1.2
and11.1.3.Theclassesaredefinedasfollows:
a)class1sectionswillpermitattainmentoftheplasticmomentandsubsequentredistributionofthebending
moment;
b) class 2 sections will permit attainment of the plastic moment but need not allow for subsequent moment
redistribution;
c)class3sectionswillpermitattainmentoftheyieldmoment;and
d)class4sectionswillgenerallyhavelocalbucklingofelementsincompressionasthelimitstateofstructural
resistance.
11.1.2 Class 1 sections, when subject to flexure, shall have an axis of symmetry in the plane of loading and,
whensubjecttoaxialcompression,shallbedoublysymmetric.
11.1.3Class2sections,whensubjecttoflexure,shallhaveanaxisofsymmetryintheplaneofloadingunless
theeffectsofasymmetryofthesectionareincludedintheanalysis.
Inordertoclassifythesectionasaspecificclass,oneseeswhichcategoryitsparameterssatisfywithregardto
section11.2Maximumwidthtothicknessratiosofelementssubjecttocompression.
Table 3 inSANS101621:2005givesthemaximumwidthtothicknessratios,W,ofelementssubjecttoaxial
compression. Table 4 in SANS 101621:2005 gives the maximum widthtothickness ratios, W, of elements
subjecttoflexuralcompression.Sectionswithwidthtothicknessratiosthatexceedthemaximumvaluesin
table3ortheclasslimitsintable4shallbedesignedasclass4sections.
Table3SANS101621:2005Maximumwidthtothicknessratios:elementsinaxialcompressionstatesthatfor
circularhollowsections:
23000
Wherefyistheyieldstrengthofthesteelusedtoconstructthetower,distakenastheouterdiameter,andtis
thethicknessofthesection.
ForLegsofangles,thiswouldbeusedinthelatticetowerclassification:
200
Wherefyistheyieldstrengthofthesteelusedtoconstructthetower,bistakenasthewidthofthelegofan
anglesection,andtisthethicknessofthesection.
Table4SANS101621:2005Maximumwidthtothicknessratios:elementsinflexuralcompressionstatesthat
forcircularhollowsections:
13000
18000
66000
3.4.1DESIGNVALIDATIONOFTHESTEELMONOPOLETOWERGEOMETRY
3.4.1.1AXIALCOMPRESSION
SANS101621:2005Section13.3AxialCompression:
13.3.1FlexuralBuckling
Thefactoredaxialcompressiveresistance,Cr,ofdoublysymmetricsectionsconformingtotherequirementsof
clause11forclass1,2or3sectionsshallbetakenas:
1
Where n=1.34, for hotrolled, fabricated structural sections, and hollow structural sections manufactured
accordingtoSANS6571.
Thiscompressiveresistanceonlycatersforloadscausedbypseudostatic
Orquasistaticactionsonthesupporttowerstructureduetowindloadsand/orselfweight.
Torsionalortorsionalflexuralbucklingisnotexamined,duetothegeometricparameterswhichsatisfysection
13.3.1ofSANS101621:2005,andnot13.3.2.
Next, the bending resistance of the tower needs tobedetermined.When initiallylooking at thetower as a
freestanding cantilevered column, one might be quick to assume that the bending resistance would be
calculatedinaccordancewithsection13.6ofSANS101621:2005,whichdescribestheprocedureforlaterally
unsupportedmembers,however,althoughthetowerislaterallyunsupported,inpartc,ofSANS101621:2005
itisstatedthatforclosedsquareandcircularsections,Mrshallbedeterminedinaccordancewithsection13.5.
3.4.1.2BENDINGLATERALLYSUPPORTEDMEMBERS
SANS101621:2005
13.5BendingLaterallysupportedmembers
Thefactoredmomentresistance,Mr,developedbyamembersubjectedtouniaxialbendingmomentsabouta
principalaxisandwherecontinuouslateralsupportisprovidedtothecompressiveflangeshallbetakenas
a) Forclass1andclass2sections
b) Forclass3sections
c)
Forclass4sections,furtherdetailsareprovidedwhichshallbecloselyexaminedintheslightlymore
unlikelyeventthataclass4sectionariseshere.Thisissaidinlightofthefactthatclass4sectionsshall
beavoidedasfarasispossibleintermsofdiameterandthicknessconsiderations.
Where,ZplandZe,aretheeffectivesectionmoduli,takenastheplasticeffectivesectionmodulusforclass1
and class 2 sections, and the elastic effective section modulus for class 3 sections. The equations for the
modulirespectivelyarethefollowing:
1
32
3.4.1.3SHEAR
Thecapabilityofthesectiontoresistshearisslightlymoreintuitive,ascanbeseenbythefollowingextract
fromtheSANS.
SANS101621:2005
13.4Shear
13.4.2Websofflexuralmembersnothavingtwoflanges
Thefactoredshearresistanceforcrosssectionsnothavingflanges(e.g.solidrectangles,roundsortees)shall
bedeterminedbyrationalanalysis.Theshearstressatultimateloadatanylocationinthecrosssectionshall
notexceed0.66fyandshallbereducedwhereshearbucklingisaconsideration.
Andso,byrationalanalysisitisfairtoassumethatthemaximumshearstressresultingonthetowerneedsto
belessthanthefollowingresistingshearstresscalculatedas:
0.66
Intermsofcalculatingthemaximumshearstressoccurringinthetowertocheckagainsttheaboveequation,
forcircularhollowsections,themaximumshearstresswhichoccursontheneutralaxisofthecrosssectionis
twicetheaverageshearstresswhichoccursonthecrosssection,calculatedas:
2
3.4.1.4INTERACTIONEQUATIONS:COMBINEDBENDINGANDAXIALCOMPRESSION
Nowthatthecapabilityofthesectiontoresistaxialcompressionaswellasbendinghasbeendeterminedthe
combinationofaxialcompressionandbendingneedstobeexamined.
Section13.8.3MemberstrengthandstabilityAllclassesofsectionsexceptclass1andclass2Ishapedsections
Membersrequiredtoresistbothbendingmomentsandanaxialcompressiveforceshallbeproportionalsothat
1.0
ThefollowinginteractionequationhasbeenadaptedfromtheinteractionequationobtainedinSANS10162
1:2005section13.8.3.
1.0
Theinteractionequationhasbeenadaptedbecauseinthisinstanceofdesign,MUXandMUYisthesamething.
Thisrelationshipshallbeassessedforboththeultimatelimitstateandtheserviceabilitylimitstate.Thevalue
ofU1x,istakenas1inaccordancewithSANS101621:2005section13.8.2.
Shearcapacitycheckatthebaseofthetower:
Fromrationalanalysisthecheckperformedontheshearatthebaseofthetowerisinaccordancewiththe
following:
1.0
3.4.2DESIGNVALIDATIONOFTHESTEELLATTICETOWERSTRUCTURE
3.4.2.1DESIGNOFCOMPRESSIONMEMBERS:EFFECTIVECROSSSECTIONALPROPERTIESOF
COMPRESSIONMEMBERS
Toverifythestrengthofthecompressionmembers,eachelementseffectivecrosssectionisexamined.
InaccordancewithSANS101621:2005clause10.4Slendernessratios,theparametersoftheeffectivecross
sectionandlengthofeachmemberareexaminedtoensurethattheslendernessratioofthememberdoesnot
exceed200.
Themomentofinertiaofthecrosssectionforeachelement,whichshallbecalculatedaboutbothaxesismore
criticalfortheaxisaboutwhichthemomentofinertiaissmaller,sincethatistheaxisaboutwhichbucklingwill
occur,andtheradiusofgyrationisdeterminedaccordingtothebucklingdirection.
3.4.2.2FLEXURALBUCKLINGOFAXIALLYCOMPRESSEDMEMBERS
Membersoflatticetowersmaybeconsideredasstraightandaxiallycompressed,andneedtobecheckedfor
theirfactoredaxialcompressiveresistance.
The axial compressive resistance of each member can be calculated in accordance with SANS 101621:2005
clause13.3AxialCompression.
Inthecaseofamemberbeingsubjecttoacompressionload,whichwouldfailduetoinstability,theresisting
capacityofthatmemberisdependentonnotonlythematerialvalues,butalsothegeometricdimensionsof
themember(Kiesslingetal,2003:400).Itshallbemadeclearinthedesignprocedurethatfollowshowthe
geometricdimensions(suchaseffectivelength,KL)aswellasthematerialvalues(suchasyieldstrength,fy)are
integratedintotheresistancecalculations.
Dependingontheshapeofthecrosssection,beingsingly,doubleorasymmetric,thestabilitybehaviourof
thesectionisinfluenced.Alsoimportantisthepositionofthegravitationalcentreandtheshearingcentre,as
well as the point of load action. Depending on the above mentioned criteria, flexural buckling, flexural
torsional buckling or torsional buckling can occur (Kiessling et al, 2003: 406). Flexural torsional buckling or
torsionalbucklingwillbethemorelikelytobeprevailinginmembersinwhichtheslendernessratioislower.
Equallegganglesaresiinglysymmetriccrosssecti ons,thegravitationalcrossssectionofw
whichdoesnotcoincide
withthe shearingcen
ntre(Kiessling etal,2003:4406).Thesecctionsareassumedascenttrallycompressedsuch
thatthe gravitationalcentreisthepointatwhic htheloadisaapplied.Ame
embercanfaillinthedirectionofthe
axisofsyymmetryinfleexuralbuckling,orotherwissebyflexuraltorsionalbuckling,seeFiguure6.
anEqualLegA
Angle
Figure6:Flexural Bucklingofa
Figure7:FlexuralTorssionalBucklinggofanEqualLegAngle
In the instance of flexxural bucklingg, the cross seection movess rectangularly to the axis of minimum radius of
gyration,,Figure6.Inttheinstanceo
offlexuraltorrsionalbucklin
ngtheequalleganglemovvesrectangula
arlytothe
axisofsyymmetry,whilsttranslatinggalongtheotther(longitud
dinal)axis,see
eFigure7.Of thetwo,failingdueto
torsionalflexuralbuckklingwillbed
dominantinthheinstanceo
ofangleswith lowslendernnessratios(Kiiesslinget
al,2003:406).
13.3.1FleexuralBucklin
ng
The facttored axial compressive
c
resistance,
r
Cr is calculateed as follow
ws for sectionns conformin
ng to the
requirem
mentsofclausee11forclass1
1,2or3sectioons:
1
Where, n
n=1.34 for ho
otrolled, fabricated structtural sectionss, and hollow
w structural ssections manufactured
according
gtoSANS657
71(coldform
mednonstresssrelieved).
13.3.2TorsionalorTorsionalflexuralbuckling
Thefactoredcompressiveresistance,Cr,ofasymmetric,singlysymmetric,andcruciformorotherbisymmetric
sections not covered under 13.3.1 shall be computed using the expressions given in 13.3.1 with a value of
n=1.34andthevalueoffetakenas
b)forsinglysymmetricsections,withtheyaxistakenastheaxisofsymmetry,thelesseroffexandfeyz,where:
1
Where x0 and y0 are the principal coordinates of the shear centre with respect to the centroid of the cross
section.
For equal angle sections, 13.3.2 b) is the governing clause to follow, and the angle sections axes shall be
defined where the UU axis is taken as the YY axis of symmetry for use in these calculations, and so by
implicationtheVVaxisbecomestheXXaxis.SeeFigure8.
Thevaluesfortheradiusofgyrationaswellasthemomentofinertiaaboutaspecificaxiscanbecalculatedas
describedabove,orlookedupintheSouthAfricanSteelConstructionhandbookforaparticularsection.
Figure8:Dim
mensionsanddCrossSectionalAxesofAngleSections
s
Inthepo
ossibleinstancceinwhichon
nehasanuneqqualanglesecction,thefollo
owingwouldaapply.
c)Forasyymmetricsecttions,feisthesmallestroottof
Where
Intheinsstanceofan unequalangle
esection,sinccetherearen
noaxesofsym
mmetry,thexxandyaxisa
asdefined
horizontaallyandverticcallyinFigure8wouldremaainasisforcaalculationpurposes.
Thefollowingcodeexttractisapplicableforclass 4sections.
13.3.3Cllass4membeersincompresssion
Thefacto
oredcompresssiveresistance,Cr,forsectiionsthatare designatedassclass4sectiionaccordingtoclause
11 shall be determineed by calculatting the slendderness ratios of members using their ggross section properties
p
andanefffectiveareaw
whichiscalculatedasfollow
ws:
a) When
Where

13.3.3
3
0.6
644
f is the calcculated comp

pressive stresss in the elem
ment (<=fy), ussing ultimate
e limit loads and
a gross
sectionpropeerties;
k=4.0foreleementssuppo
ortedalongboothlongitudin
naledges;or
0.43forelementssupportedalongonelongitudinaledgeonly.
WhenW>Wlim,
Aefshallbedeterminedusinganeffectiveareacalculatedwithreducedelementwidthsmeetingeither
the maximum widthtothickness ratios of class 3 sections or, where applicable, an element width
equaltobtobetakenas:
0.95
Whichresultsin
0.208
3.4.2.3BENDINGANDAXIALCOMPRESSION
Bending about both principal axes combined with axial compression forces represents the most general
loadingofamemberinalatticetower(Kiesslingetal,2003:408).Inordertoverifythedesignofmembersfor
these conditions, one would verify that the interaction equations were satisfied as was done in accordance
withclause13.8ofSANS101621:2005.
3.4.2.4DESIGNOFTENSIONMEMBERS:
Inordertoverifythechoiceofthesectionofamemberintermsofitscapabilitytoresisttensileforces,one
will need to determine the tensile resistance of each member to a tensile force which should be relatively
largeincomparisontoanycompressiveforcesthatmemberisrequiredtoresist.Thetensileresistingcapacity
ofamemberisdependentonthenatureoftheendconnectionsofthemember.
InaccordancetotheSANS101621:2005clause12and13.2Axialtensionmemberandconnectionresistance,
thetensileresistanceiscalculatedasfollows.
12Grossandnetareas
12.1Application
Membersintensionshallbeproportionedonthebasisoftheareasassociatedwiththepotentialfailuremodes.
Membersincompressionshallbeproportionedonthebasisofthegrossarea.
12.2Grossarea
Grossareasshallbecalculatedbysummingtheproductsofthethicknessandthegrosswidthofeachelement
(flange,web,legorplate),asmeasurednormaltotheaxisofthemember.
12.3Netareaandeffectivenetarea
12.3.1Theeffectivenetarea,Ane,shallbedeterminedbysummingthecriticalnetareas,An,ofeachsegment
alongapotentialpathofminimumresistancecalculatedasfollows:
a)forasegmentnormaltotheforce(i.e.,indirecttension)
b)forasegmentinclinedtotheforce
12.3.2 In calculating wn, the width of the bolt holes shall be taken as 2mm larger than the specified hole
diameter. Where it is known that drilled or punchedand drilled holes will be used, this allowance may be
waived.
13.2Axialtensionmemberandconnectionresistance
Thefactoredtensileresistance,Tr,developedbyamembersubjectedtoanaxialtensileforceshallbetakenas
a) Theleastof
i)
ii)
0.85
iii)
0.85
Andforpinnedconnections,
0.75
3.4.2.5COMBINEDBENDINGANDAXIALTENSILE/COMPRESSIVEFORCES:
This was done in accordance with SANS 101621:2005 section 13.8.3, detailed in section 3.4.1.4 above. The
equationisrepeatedhereforconvenience.
1.0
3.5DESIGNITERATION
For the instances in which unsuitability of structural members were determined through the geometrical
validation process, those members were improved andthe process was runthrough againuntil the optimal
membershadbeenselected.
ForthedesignoftheSteelMonopoleTower,thesmallest,thinnestCHSswerechoseninaccordancewiththe
boundaryconditionsofslenderness,andthe508x12.7mmCHSswerefoundsuitableforalldesignparameters
oncemodelledinStrand7.
TheSteelLatticeTower,constructedofequalleganglesectionsofvaryingsizes,hadallmembersexaminedfor
the design parameters stipulated in section 3.4 above. It was found through iterative design that the initial
50x50x3mm Angles (the smallest used in the tower design) were inadequate and were ungraded to
50x50x5mmAngles.ThecalculationresultsforthisareshowninTable7.
Table7:Equalleganglesectioncompressioncapacity
TestingofSectionCapacity
Section
check
Cu
Cr
50x50x3 61865.54 59542.47 NOTOK

50x50x5 61865.54 101739.8 OK
3.6GENERALASPECTSOFTOWERDESIGN:
FortheSteelMonopoleandtheSteelLatticeTowersdesignedforthisthesisdesignconsiderationswhichare
relevanttobothtowersingeneralarethefollowing:
1.
2.
3.
4.
5.
TheMaterials
MaterialProtection
TowerAccess
TowerProduction
Fatigue
3.6.1MATERIALS
MATERIALSFORTHECIRCULARHOLLOWSECTIONS
ThemonopoledesignforthisthesisincorporatedtheuseofCircularHollowSections(CHS).InSouthAfricathe
SteelInstituterecommendsthatGradeS355JRsteelbeutilisedforsteeldesign.
MATERIALSFORANGLESECTIONSANDPLATES:
The angles used in lattice towers need to be sufficiently strongto withstand anumber ofdifferent weather
elements,aswellasdynamicandvibrationalloadingsandalsostaticloads.Asuperiorcapacitytodeformation
reducesthelocalstresspeaksintheelementswhichaidsinavoidingthedevelopmentofcracks(Kiesslingetal,
2003:379).
The angle sections shall be made out of Structural steel. The term structural steel may be used to define
steelforelementswhoseprimarypurposeistosupportloadsorresistforceswhichactonastructure(SASCH,
2008: 2.3). Both SANS 10162: Part 1 and SANS 2001CS 1 require that steel for structural applications must
complywiththerequirementsofSANS1431.
Thepresent situation in SouthAfrica is that steel sections are produced toEN10025Hot rolled products of
structuralsteelsPart2:Technicaldeliveryconditionsfornonalloystructuralsteels,andspecificallytoGrade
S355JRofthestandard(SASCH,2008:2.3).Theexceptiontotheruleisthecaseofequalanglesectionsupto
50x50mmthatarecommonlymadefromcommercialqualitysteel,andequalleganglesectionsthatarebigger
than50x50mmuptothesizeof80x80mmthataremadeineithercommercialqualityor355JRsteel(SASCH,
2008:2.3).
The mechanical properties of the different types of steel are given in Table 2.1 of the South African Steel
ConstructionHandBook.
MATERIALSFORBOLTS:
BoltsoftwopropertyclassesarecommonlyusedforstructuralpurposesinSouthAfrica:Class4.8andClass
8.8bolts.BothclassesofboltsarecommonlyusedwithClass8nuts,althoughClass5nutsarepermissiblefor
class 4.8 bolts. The mechanical properties of bolts are given in Table 6.1 of The South African Steel
ConstructionHandbook(SASCH).Table6.1alsocontainsinformationonpropertyClass10.9,whichis,forman
economic standpoint not recommended for standard use, but rather for friction grip bolting. See Table 8
below.
Table8:SASCHTable6.1
SASCH
Table6.1
MechanicalPropertiesof
Bolts
Property
PropertyClass
TensileStrength(MPa)
Nominal
Minimum
400
420
800
800
800
830
1000
1040
YieldStress(MPa)
Nominal
Minimum
320
340
0.2%ProofStress(MPa)
Nominal
Minimum
640
640
640
640
900
940
Elongationatfracture(%)
Minimum
14
12
12
4.8
8.8
10.9
d<=16mm d>=16mm
Table6.2intheSouthAfricanSteelconstructionhandbookshowsthedimensionsofthedifferentsizedbolts.
3.6.2CORROSIONPROTECTION
Structuralsteelissubjectedtocorrosioninthepresenceofmoistureandoxygen.Inacorrosiveenvironment,
thesteelcanbeprotectedbypainting,hotdipgalvanisingormetalspraying(SASCH,2008:2.8).Thecorrosion
of a metal may be encouraged through contact with another metal in moist conditions; this is known as
bimetallic,galvanicorelectrolyticcorrosion.Itisthusadvisedthatoneinsulatesteelfromothermaterialsto
preventsuchacceleratedcorrosionintheareasofcontact(SASCH,2008:2.8).
Virtuallyallofthesteellatticetowerstodayarecomprisedofelementsthathavebeenhotdipgalvanised.Hot
dipgalvanisingincorporatescreatingaprotectivelayerofironzincalloywhichisformedonthesteelduring
thehotdipgalvanisingandabovethatlayeronemadeofpurezincisconstituted.
Afterapproximatelytentofifteenyearsthezinclayerwillbeweatheredtosuchanextentthatanalternative
corrosionprotectionwillbenecessary.Inareasofparticularlycorrosiveatmospheres,thenumberofyearsin
which the protection needs renewing may be vastly reduced (Kiessling et al, 2003: 378). There is however
another method, in which together with the hotdip galvanisation, a paint coating is applied directly to the
elements.Thegalvanisingtogetherwiththepaintlayeractsinasynergeticmannerresultinginamorethan
doubledlifecycleasthatachievedbyonlyonecorrosionmethodalone(Kiesslingetal,2003:379).
Obtaininganaestheticallypleasinghotdipgalvanisingfinishisafunctionoftheappropriatesteeldesign,good
fabrication and correct steel chemical composition. The chemical composition encompasses the amount of
silicon(Si)andPhosphorus(P).ForminingapplicationstheSishouldbebetween0.15to0.3%withPlessthan
or equal to 0.02%. For architectural purposes the Si should be 0.03% with P less than or equal to 0.01%
(SASCH,20082.12).
3.6.3ACCESS
Steellatticetowersneedtobeascendedbyexperiencedstaffduringconstructionandmaintenance.Inorder
toaccommodatethis,accordinglydesignedaccessneedstobeprovided.Ladderswithinthefacesofthebody
ofthestructurecanbeinstalled(Kiesslingetal:2003,377).Accessshouldbeinstalledonselfsupportingsteel
structures,startingataheightof3mabovethegroundtodiscourageunauthorisedaccessofthestructureby
strangers as well as vandalism. The access ladders can be a permanent fixture on the towers, or temporary
attachmentsattimesofmaintenance(Kiesslingetal:2003,377).
3.6.4PRODUCTION
In the manufacturing process, the manufacturing of holes for bolts is a vital step in the production of steel
latticetowers.Thereareanumberofwaysinwhichtheholescanbemade,theycanbedrilledorpunchedor
subpunchedandreamed.Punchingfullsizeisafast,economicalwaytogetpreciseholesize,butthematerial
thicknessthatcanbepunchedislimited(Kiesslingetal,2003:378).
Inthemanufacturingofsteelmonopoles,itiscrucialtoensurethatthefabricatoriscapableofhighquality
weldingseemssuitabletothehotrolledsteelsections.
3.6.5FATIGUE
Theeffectofcyclicloadingisimportanttoexamineindesignsoastoavoidfailurebyfatigue.Thefollowing
extractfromtheSouthAfricandesigncodesisforfatigueconsiderations.
SANS101621:2005
26Fatigue
26.1General
Inadditiontocomplyingwiththerequirementsofclause26forfatigue,anymemberorconnectionshallalso
complywiththerequirementsforthestaticloadconditionsusingthefactoredloads.Specifiedloadsshallbe
usedforallfatiguecalculations.Aspecifiedloadlessthanthemaximumspecified,butactingwithagreater
numberofcycles,maygovernandshallbeconsidered.Membersandconnectionssubjectedtofatigueloading
shallbedesigned,detailedandfabricatedsoastominimizethestressconcentrationsandabruptchangesin
crosssection.Thelifeofthestructureshallbetakenas50years,unlessotherwisestated.
26.2Proportioning
Intheabsenceofmorespecificinformation,whichissubjecttotheapprovaloftheowner,therequirementsof
clause 26 in its entirety provide guidance in proportioning members and parts. Fatigue resistance shall be
providedonlyforthoseloadsconsideredtoberepetitive.
InChapter6,thedetailsofthedynamicanalyseswhichwereconductedwithrespecttoanypossiblerepetitive
loadsaredealtwith.Thepossibilityoffatiguefromrepetitivecyclesisminimisedtothelargestpossibleextent
byusingholdingdownboltsinaccordancewiththestrengthrequirementsinsection13.12.1.3.
Thestressrangestowhichtheelementsareexposedarealsovirtuallynegligibleintermsoffatigueandthe
numberofrepetitiveloadsthatmaybeappliedinthetowersdesignlife.
Allofthedetailedcalculationsdonewithrespecttothedesignprocedurecoveredinthischapterareshownin
AppendixA.
Furthermore the Foundations of the two steel towers as well as the connection details for the two towers
needtobedesigned,andaredonesointhefollowingtwochapters.
CHAPTER4:CONNECTIONDESIGN
ThischapterservestoexplaintheprocessusedtodesigntheconnectionsrequiredinboththeSteelMonopole
andSteelLatticetowers.
TheSteelMonopoletowerrequiredringflangeconnectiondesignforthepointsatwhichthethreedifferent
sectionsoftowerwerejoined,aswellasaconnectionatthebaseofthetowerwhichexperiencedthegreatest
momentsandaxialforcesthroughoutthetower.
TheSteelLatticetowerrequiredthedesignofpinnedconnectionsfortheendsofeachmemberinthetower.
As shall be discussed in this chapter under the pinned connections topic, the requirement of gusset plate
designasaresultofspacelimitationsandredundancydesignforboltswasalsointroduced.
4.1MONOPOLERINGFLANGECONNECTIONS
Researchinganumberofdifferentpublicationstofindasuitabledesignmethodforthedetaileddesignofring
flanges connected by a number of tension bolts revealed that there was an immense amount of literature
coveringanumberofdetailedfiniteelementanalysesofthemanycomplexaspectsthatariseinringflange
connections.Furthermorenoimmediateclearandconcisedesignmethodwasavailable.
InanarticlepublishedbythedepartmentofStructuralEngineeringattheUniversityofTokyo,entitledBolted
Tension Flanges Joining Circular Hollow Section Members written by B. Kato and R. Hirose in 1985, a
theoreticalanalysisofringflangeconnectionswasdeveloped.
Thefollowingdesignprocedurewasadoptedfromthetheoreticalanalysisdevelopedintheabovementioned
article.
Thefirstimportantelementthatwasmentionedinthearticlewasthatforallringflangedjoints,highstrength
boltsweretightenedtoproduceapreloadof60%oftheproofload.
Asshallbethecasewiththedesignoftheringflangesfortheturbinetower,thewallthicknessofthetubesis
thinnerthanthecorrespondingflangethickness.
Theoreticalanalysissuggeststhatfailurecouldoccurbecauseofoneofthreemodesoffailure:
1.
2.
3.
Flexuralfailureoftheflangeplate
Fractureofthehighstrengthbolts,or
Fractureofthetubeintension.
Theabove3mentionedmodesoffailureareillustratedinFigure9.Thebendingmomentsthatdevelopasa
resultofsuchfailuresareillustratedinFigure10.
37|C h a p t e r 4 : C o n n e c t i o n D e s i g n
Figure9
9:ModesofFFailureforRingFlangeConn
nections
Figurre10:BendingMomentsddevelopedinR
RingFlangeCo
onnectionFaiilures
Theoreticcalanalysisexxaminestheb
bendingstren gthoftheflangeplatesby
yusingtheyieeldlinetheoryy,andthe
tensile fo
orce acting on
o the highsstrength boltss is evaluated
d considering
g the prying force acting from the
oppositesideoftheflaangethroughthecontactaarea.
Theflanggestrengthandtheboltfo
orcewillreacchdifferentvaaluesdependingonwhethherornotthe
eultimate
stateisreeachedbeforeeoraftertheoccurrenceoofseparation((perfectreleasseofboltpreload).
Figure11:Ring
F
gFlangeConn
nectionDetails
4.1.1B OLTFORCEEATSEPARAATION:
Firsttob
beevaluatedisstherelationshipbetweenntheforceacttingonthebo
oltsattheinsttantofsepara
ation,aso
calledseparationforcee,Ts,andtheboltpreload To.Therelatio
onshipisgivenas:
1
Intheab
boveequations,
Ab=theccrosssectionaalareaoftheunthreadedsshankofthebolt
Ap=theeeffectiveareaaofthecomprressedflange plate
From thee research co
onducted by the
t Universityy of Cardiff as
a well as the
e University oof Tokyo, the equation
adoptedforthedeterm
minationofA
Apisthefollow
wing:
diameterofth
hewasherface
eofthebolthheadornut,and
D=thed
lp=thegrriplength,equ
ualtotwiceth
hethicknessooftheflange
dp=thed
diameterofth
hebolthole.
Using th
he above form
mulae, the different Separration force versus
v
Bolt preload (Ts vss. To) relation
nships are
defined:

Ty=theyyieldstrengthofthebolt
d=thediameteroftheunthreaded
dshankofthebolt.
4.1.2MAXIMUMBENDINGSTTRENGTHO FFLANGEFO
OLLOWINGPLATESEPA
ARATION:
Duetoth
hefactthattheformation ofacollapseemechanism intheflangeoccursaftertthereleaseo
ofthebolt
pre load,, the constraint around th
he bolts needd not be conssidered. A collapse mechaanism is considered by
looking aat the possib
ble patterns of yield linees along which the flange might collaapse, and within
w
this
examinattion the existtence of boltt holes is ignnored. The su
uggested yield
d lines, takenn from the th
heoretical
analysisaarethefollow
wing:
Figgure12:RingFFlangeConne
ectionYieldLines
The equaations that fo
ollow will nee
ed slight adapption, since th
hey are develloped for thee above show
wn figures,
where6boltsareconssidered.Thea
analysisproceeedswiththefollowingstep
ps:
1.
2.
3.
First, as show
wn in Figure 12, the circuular tube (flan
nge) is approxximated by taaking a portion of the
section;atweelfthofthefla
angeisanalys edtakingsym
mmetryintoacccount.Ofcouursethetwelffthhereis
specifictotheeconnectionhaving6boltss,soingeneraalonewouldttake:
1

2
Next, different points on the section are examined, and different yield linees are determ
mined for
differentpositionsofthatpoint,andthheresultisth
hatofdifferentyieldlineppatterns,whicchcanbe
seeninpartb
bandcofFigu
ure12.
Next the virtual plastic wo
ork done by tthe different points chosen
n in step twoo are examine
ed for the
differentpattternsshowna
above,andtheelocationoftthepointatw
whichtheleasttvirtualworkisdoneis
4.
theonewhichoccursatth
heintersectio noflinesP1aandP2inparta)ofFigure 12,whichressultinthe
boltpitchcircclesdisplayedinpartd)ofFFigure12.
Finally,byap
pplyingtheyie
eldlinetheoryytotheyieldpatterninpa
artd)ofFigurre12,theyieldloadof
the flange, Pp, and the maximum
m
loadd of the flan
nge, Pu is obttained. The w
way in which they are
calculatedissshownintheequationsbellow(Kato&H
Hirose,1985):
1
4
4
48tan
12
omentperuniitwidthoftheeflange
mp=theffullplasticmo
Dp=theb
boltpitchcirclediameter
Di=theeequivalentdiameterofthecirculartube
y=theyyieldpointofttheflangema
aterial
u=thettensilestrengtthoftheflanggematerial.
The equiivalent diameeter (Di) for riing flanges deefines the loccation of the plastic hinge line around the tube,
whichforthecaseof aringflange whichisjoineedbyafillet weld,happen
nstoarbitrariilyliehalfwayythrough
the width of the fillett weld, leavin
ng the equivaalent diamete
er equal to the outer diam
meter of the tube,
t
plus
effectively,oneweldw
width(Kato& Hirose,1985 ).
NoteherrethatthedevelopmentofftheequationnforUinstep
p4aboveisalsoparticular totheexamp
pleshown
inFigure11for6boltts.TheequationforUwouuldneedtobe
eadaptedtotthenumberoofboltsused aswellas
theportionoftheflan
ngeexaminedforvirtualplaasticwork.Ap
proposedgeneralequationforUisthefo
ollowing:

tan
8

2
4.1.3MAXIMUMBENDINGSTTRENGTHO FFLANGEW
WHENSEPAR
RATIONOCCCURSAFTER
YIELDIN GOFTHEFLANGE:
Where w
washers are used at the bo
olt heads andd nut faces, a rigid zone ca
an develop beefore separattion takes
place.Assimpleyieldp
patternwhichtakesthiszonneintoaccoun
ntisdepictedbelow:
RingFlangeCoonnection:yie
eldlinebefore
eseparation
Figure13:R
Sincetheelimitoftheyyieldlineistakentotheouuteredgeofth
hewasherface,theouterddiameterofth
hewasher
needsto
obetakeninttoaccountin calculatingthheyieldload oftheflange,Pp,andtheemaximumlo
oadofthe
flange,Pu,whichisreachedbeforeseparationocccurs.Theupd
datedequatio
onsare:
48
8
12
It is noteed that strictlly speaking th

he yield line theory is onlyy applicable to
t the evaluaation of the full
f plastic
strength (yield strenggth) of the plate. Howeverr, it is assum
med that the loadcarrying capacity will increase
almostlin
nearlyupuntilitreachesamaximumva lueofmu,and
d(Kato&Hiro
ose,1985);
1
4
4.1.4EVALUATION
NOFTHEPRYING
R
FORCCE:
Ifaflanggeisflexible, whenthecon
nnectionissuubjectedtoa tensileforce,,itwillbend, andtheflanggewillbe
subjecttoareactionfforcefromthe
eoppositefla ngethroughtthecontactfa
acetheysharee.Thepryingfforcesare
examined
dseparatelyfforbeforeand
daftersepara tion.
4.1.5PRYINGFORRCEACTINGBEFORESEPPARATION:
Whenafflangeyieldsb
beforeseparationtakesplaace,thepryinggforceisdistrributedoverttheshadedarreashown
intheFiggure12orFiggure13.Inorrdertodeterm
mineanestim
mateoftheprryingforce,bbecausetheexxamplein
thearticllehas6boltss,atwelfthoffthecircularfflangeismodelledasarigiidbodyonannelasticfound
dation,as
shownbeelow:
Figure144:PryingActio
onModel
In the ab
bove twelfth of the circular flange, thee arc of the section is approximated bby a linear se
egment. A
bending moment,ofm
magnitudebetweenmpanddmu,isapplie
edperunitlengthoftheyiieldlineGH.A
Areaction
forceperrunitlengthaalonglineIKisscalculatedsuuchthatthere
eisequilibrium
mofthebenddingmomenta
aboutline
GH.Itisaassumedthattthepryingfo
orceislinearlyydistributedttowardstheedgeoftheflaangeasdepicttedbythe
shadedaareainFigure14.
Theaforeementionedb
bendingmome
entequilibriu mabouttheyyieldlineGHiswrittenas:
. 2. .
1
6
12
12
12
Theaboveequationcanbesolvedforr:
12. .
12
Thenthetotalreactionactingthroughoutthecircularflange,R,is:
48.
12
2
3
Takingnotethatthisisthepreseparationstate,A=Do/2andB=Dp/2a/2.IntroducingAandBintotheabove
equation for the reaction R, a ratio of prying force to external load at this level of hinge moment m is
calculated.TheexternalloadP(m)=Umisobtainedbyreplacingthempwithm.
2
3
Itisassumedthattheaboveratioisalsoapplicabletothestatewhentheflangeiselastic.
Whentheflangehasyieldedandseparationtakesplace,thereactionforcewillbedistributedalongtheedge
lineoftheflange,lineIKinFigure14.InordertocalculatethereactionperunitlengthoflineIK,callitrs,the
equilibriumofthebendingmomentaboutlineGHiscalculatedas:
. 2. .
12
12
2. .
12
Andthenrscanbesolvedfor:
.
12
Andthecorrespondingtotalreactionactingthroughoutthecircularflange,Rs,is:
24
12
Takingnotethatthisisthepostseparationstate,A=Do/2andB=Dp/2.IfAandBareintroducedintotheabove
equation for Rs, then once again the ratio of the prying force to the external load at this level of hinge
moment,m,canbedetermined:
Nopryingactionwillactwhentheflangeremainselasticatthetimeofseparation,butareactionwilloccuras
soonasyieldingoftheflangebeginsduetofurtherincreasesoftheload(Kato&Hirose,1985).
4.1.6SEPARATIONLOADANDFRACTURELOADOFTHEBOLTS:
4.1.6.1SEPARATIONLOAD:
Separation of the bolts in the connection takes place when the sum of the tensile force applied to the
specimen,P,andthepryingforce,Rreachtheseparationforceofthebolts,Ts,definedintheabovesection
onboltforcesatseparation(4.1.1).Thisequationformisthefollowing:
1
Or
1
WherePsistheboltseparationload.
4.1.6.2FRACTURELOADOFTHEBOLTS:
Fractureoftheboltsoccurringafteryieldingoftheflange:
Here,themaximumtensilestrengthoftheconnection,bPm,isderivedasfollows:
1
Or
InwhichTuisthetensilestrengthofabolt,anditisnotedhere,thatthisprocessisnecessarilyaccompanied
byboltseparation.
Fractureoccurringbeforeyieldingoftheflange:
Here,themaximumtensilestrengthoftheconnection, bPom,isgivenasfollows,sincenopryingactionwillbe
imposedonthebolts:
b
4.1.7MAXIMUMSTRENGTHOFAJOINT:
Having viewed all of the important factors which influence the maximum strength of a joint, a summary is
presentedhere:
1.
Criticalloadscorrespondingtoflangefailure
i.
When the flange yields and reaches its maximum strength before separation takes place,
yieldloadoftheconnection:
Maximumloadoftheconnection:
ii.
Whentheflangeyieldsafterseparationtakesplace,theyieldloadoftheconnection:
Maximumloadoftheconnection:
2.
Thecriticalloadscorrespondingtoboltfailure
i.
Boltseparationload(yieldload)
ii.
iii.
1
Maximumloadofthejointwhenyieldingofflangeoccursfirst
1
Maximumloadofjointwhentheboltsyieldandfracturebeforeyieldingoftheflangeoccurs,
Depending on the relative magnitudes of the above values, the maximum strength of the connection is
determinedasfollows:
1.
2.
3.
Connectionmaterialfractureoccurringbeforeseparation
1i.FailureoftheflangetakesplacewhenPu<Ps.Thusthemaximumconnectionstrengthis:

1ii.Failureoftheboltsdoesnotoccurbecauseboltfailuremustprecedeboltseparation.
Connectionfailureoccurringattheinstantofseparation
2i. Failure of the flange takes place when Pu<Ps<bPm or when Pu<bPm<Ps. The maximum connection
strengthis:
2ii. Failure of the bolts occurs when bPm<Pp<Ps<Pu or Pp<bPm<Ps<Pu or bPm<Pu<Ps. The maximum
connectionstrengthis:
Connectionfailurewheretheflangeyieldsbeforeboltseparationoccurs
3i.Failureoftheflangeisreachedbeforefractureoftheboltsoccurs,whenPs<Pu<bPm.Themaximum
connectionstrengthis:
3ii. Bolt failure occurs immediately after yielding of the flange, when bPm<Ps<Pp<bPom or
Ps<bPm<Pp<bPom.Themaximumconnectionstrengthis:
3iii. Failure of the bolts occurs between the yielding of the flange and flange failure, when
Pp<Ps<bPm<PuorPs<Pp<bPm<Pu.Themaximumjointstrengthis:
4.
5.
Connection strength occurs due to rupture of the bolts occurring before the flange yields, when
o
bP m<Pp.themaximumconnectionstrengthis:
When the tensile strength of the tube is lower than any of the above mentioned connection
strengths,failureinthetubewilloccurfirst.
Because all of the above equations and strength conditions are functions of the bolts diameter, tube
diameter,boltpitchcirclediameter,flangediameter,andflangethickness,thefailuremodeofaconnectionis
determinedbytherelativesizesoftheabovementionedgeometricalparameters(Kato&Hirose,1985).
4.1.8S UGGESTEDDESIGNMETHODS:
GENERALPROCEDUREE:
Thestren
ngths,Pp,PuaandPuincreassewithdecreeaseofboltpitchcirclediam
meter,Dp,asccanbeobservvedinthe
aboveeq
quations.The distanceofb
boltholefrom thetubewall,however,is limitedbythhepracticalitie
esofbolt
tightenin
ng. The minim
mum Dp is sett to Dp=Di+3dd, assuming that the minim
mum feasiblee bolt distancce is 1.5d.
Other ob
bservations su
uch as how the ratios 1 aand 2 decreaase with an increased flannge diameter also may
influenceethechoiceo
offlangediam
meter,howeveerforpracticalpurposes, italsolooksuuglyhavingto
oolargea
flange.TheflangediameterissetttoDo=Dp+4daasanoptimum
msize,assum
mingthedistanncefromthe centreof
thebolth
holetotheed
dgeoftheflan
ngeistobe2dd.
SIMPLIFI EDDESIGN:
The relattionship betw
ween the maxximum strenggth of a connection and th
he flange thicckness is show
wn in the
followinggfigure:
Figure15 :PMAXversusTfrelation
The above values arre obtained using
u
the lim
mited connecttion geometrry stipulated above unde
er general
procedurre (4.1.8). It can
c be seen that the connnection stren
ngth can be in
ncreased con siderably by using the
larger prreload for th
he bolts for a certain regioon of flange thickness.
t
Wh
hen the flangge thickness exceeds
e
a
critical vvalue, bolt failure occurs before yield of the flange
e. In this casse the conneection strengtth can be
attributeedtothestren
ngthofthebo
oltssincetheppryingforcew
willnotinfluen
ncethebehavviour.Suchad
designcan
be consid
dered to be most
m
effective if the increease of weigh
ht of the jointt is not substtantial. If this design is
adopted,,themaximum
mstrengthof thejointsho uldbeproporrtionedtobelargerthanthheyieldstrenggthofthe
tube,sinccesufficientd
deformationccapacitycannootbeexpectedinthejointitself(Kato& Hirose,1985)).
4.1.9CONNECTIO
ONDESIGNFFROMBASICCPRINCIPLEES:
From thee knowledge gained throu
ugh extensivee research do
one into the design of cirrcular ring flanges, the
followinggdesignmeth
hodisdevelop
pedfromthe basicprincip
plesofequilibrium,andassimilarapproa
achtothe
designm
methodforpryyingactionpre
esentedintheeSouthAfricaanSteelConstructionHandbbook(SASCH)).
Forthed
designoftherringflangeconnections,froomthefinitee
elementanaly
ysisdoneinSt
Strand7,thete
ensileand
compresssiveforcesattheconnectio
onpointswerreknown,asw
wellastheap
ppliedmomenntthatneeded
dresisting
duetoth
heimposedloads.
Howeverr,thedimensiionsofthefla
angesandtheenumberof boltsandtheintervalsatw
whichtheyw
weretobe
spacedw
wereinitiallyeestimated.Iterativecyclesoofdesignand
dcalculationw
werethendonneuntilanap
ppropriate
designcaapableofresisstingtheappliedloadswassdefined.
Thefirststepwasdeciidinguponthe
enumberofbbolts,andthespacingbetw
weenthem.
Figgure16:BoltSSpacingandT
TributaryLeng
gth
helayoutofb
boltschosenttobeginthed
designprocedure.Theboltsswerechosen
nasM20,
Figure166illustratesth
Class 10..9S bolts in accordance
a
with recomme ndations from
m the SANS design
d
codes, as stipulated
d in SANS
101621:2002section22.Thebolts werespacedradiallyat22
2.5degrees,w
whichallowedfor4boltspe
erquarter
oftheflaange.
Thenextstepinthedesignwastocchecktheres istingcapabiliityofthebolttscollectively aswellasthe
eresisting
capacityoftheflange,whichstillne
eededdimenssionalvalues.
Figgure17:DimeensionalParam
metersofFlan
nge
Here,theereareanum
mberofdifferentsequenceesinwhichth
hevariousdim
mensionscan bedetermine
ed,orthe
resisting capacitiescalculated.Eitherway,onew
willneedthe assistanceofafiniteelem
mentanalysis, eitherfor
therequiredforcesocccurringatthespecifiedcoonnectionpoiintsinorderttodetermine aflangethickknessand
width,orrotherwiseto
odeterminea
anappropriattelysizedflangefromiterativemodellinnginthefinite
eelement
program,,inwhichcasetheresistinggcapacityoftthedesignedfflangecouldb
becheckedaggainsttheapp
pliedloads
andmom
ments.
Itwasop
ptedheretom
modelaflange
econnection inStrand7withaninitially
yarbitraryflanngethicknessandthen
runanan
nalysis.TheSttrand7model couldthenb ealteredinth
hetermsofth
heflangethickknessuntilthestresses
developeedintheflanggeconnection
nswereaccepptable.Thediistancesbetw
weenthecentrreofthebolttsandthe
wallofth
hetowerasw
wellasthedistancesbetw
weenthecenttreofthebolttsandtheouuteredgeofttheflange
werecho
oseninaccord
dancewithth
hedistancesooutlinedintheSouthAfricanSteelConsstructionhand
dbookfor
theminim
mumallowabledistancesw
withregardtomachiningan
ndboltinstalla
ationtools.
The follo
owing section handles the connection ddesigns that were develop
ped using Strrand7, followe
ed by the
calculatio
onsdoneinorrdertoensure
etheresistan cecapacityofftheconnections.
4.1.100CONNECTIONDESIGN
NSINFINITEEELEMENTPROGRAM(S
( TRAND7) :
MIDSECCTIONCONNEECTIONS:
Theconn
nectionsatth
hemidpoints withinthetoower,atthevverticalheighttsof6mand 11mrespectivvelywere
designed
dtobe20mm
mthick,withw
webstiffenerssat22.5degrreeradialspacings,which lieexactlyinbetween
eachboltt.Thewebstiiffenerswerechosenat100mmthickand
d100mmhigh
halongtheheeightoftheto
ower,and
triangulaarlystretchedtotheedgeo
ofthe100mm wideflanges..Allsteelused
dinthisdesiggnwasS355JR
R.
Figure18:Mid
F
dsectionConnectionDesig
gn
Figure19:MidFlaangeandWebStiffenerDe
etails
nectionswereemodelledsu
uchthatthereewasa1mm gapbetweenthetwocon nectingflange
es,andat
Theconn
theplaceementofthe bolts,rigidlin
nkswereuseddtoconnecttthetwoflange
es,soasbest tosimulatethebolted
connectio
onforanalysis.
BASECO
ONNECTION:
Theflanggeconnection
ndesignatthebasehada 40mmthickfflange,andwebstiffeners thatwas200
0mmhigh,
20mmth
hickandstretchedtriangularlytotheeddgesoftheflangesthroughadistanceoof130mm.Onceagain
the webstiffeners weere spaced att 22.5 degreees radially, an
nd were so placed exactlyy in between the bolts
aroundtheflange,whichareM20C
Class10.9Sboolts.
Fromtheemodellingacccuracyofthe
ebaseconnecction,thepointsatwhichtheboltswerremodelledw
werefixed
foralldeegreesoftransslationalandrrotationaldispplacement.
Figure20: BaseFlangeC
Connection
Figure21:BaseFllangeandWe
ebStiffenerDe
etails
4.1.100.1CONNECCTIONCALCCULATIONS:
Thevalueesgiveninthespreadshee
etextractinTTable9canbe
ecorrelatedvvisuallywithFFigure16.The
etributary
th
length off each bolt which
w
is noted
d as L in Figu re 16 was taken as one 16 of the extternal diametter of the
tower.Fiigure16also illustratesthedistancebeetweentheboltcentreandthetoweraaswellasthe
edistance
betweenboltcentreaandedgeoffla
angenotedassbandarrespectively.
Table9:Spreadsheeetextractshowingflangedetails
Trributarylengtthl
l
0.10 m
0.90
fla
angestrength
h
fy
3
3.35E+08
Pa
M
Maximumdesig
gnstrengthoffbolttoavoid
dfatigue
fu
2
2.62E+08
Pa
seeSA
ANS101621:2005clause2
26
bo
oltstrengthca
alculatedwith
h262MPainaccordancew
withSANS101
1621:2005cla
ause13.12.1.3
3
2
m
314.16 mm
Ab
M20
0
0.80
b
Tr
4
49385.84
N
AttBaseConnecction:
Atmiidandtopcon
nnections
FrromDiagram
FromDiagram
0.05 m
0
0.05
m
0.08 m
0
0.05
m
0.04 m
0
0.02
m
Thebolt strengthinth
heabovespre
eadsheetwa scalculatedinaccordance
ewiththeafoorementionedequation
fromSAN
NS101621:20
005.
VALIDATTIONOFTHEBOLTSIZEANDSTRENGTH::
Because withapryinggactionthete
ensileforceofftheboltsin aconnection subjecttoa tensileloadin
ngmaybe
significan
ntlyincreased
d,itisnecessa
arytocheckiifthecapacityyofthebolt iscapableof handlingthe potential
increaseinloaddueto
otheprying.
Figure22
2:ReactionFoorcesatMono
opoleTowerB
Base(SLS)
Figure222fromthefiniteelementanalysisinStraand7,showsth
hereactionfo
orcesatthepoointsofboltp
placement
alongtheebaseflangeconnectionfo
ortheSLSloa dcase.Incom
mparingthem
magnitudeofttheforcestha
atneedto
beresisted,thetensileforcethattheM20boltscanresistascalculatedinTable9(Tr=49kN)ismostcertainly
adequate.
ThereasonthattheSLSwastheloadcasechosentoexaminethereactionforcesrequiredtoberesistedwas
forthedesigninstanceoffatigue.Theclauseoutlinedin13.12.1.3ofSANS101621:2005isdiscussedatthe
endofthischapterandexplainsthisinfurtherdetail.
VALIDATIONOFTHEFLANGESIZEANDSTRENGTH:
The theoretical design magnitude which the flanges can resist through calculations incorporates the steels
yieldstrengthandtheflangethicknessaswellasthetributarylengthofeachbolt.Thedesignstrengthofthe
flangesoncetheoreticallycalculatedwerecomparedtotheforcesandstressesthattheyarerequiredtoresist
inaccordancewiththeoutputfromStrand7.
Themomentswhichtheflangeswereabletoresist,forplasticityandelasticityrespectively,arecalculatedin
accordancewiththefollowingequations:
In Table 10 the moments were calculated in accordance with the elastic moment of resistance, for a more
conservativeapproach.
Table10:MomentResistanceoftheFlanges
Momentresistanceoftheflangesrespectivelyare:
AtBaseConnection
Mr
8.02E+6 Nmm/mm
Atmidandtop
connections
Mr
2.00E+6 Nmm/mm
ThesemaybecomparedtothemomentsthatarerequiredtoberesistedthatareobtainedfromStrand7.
Forarangeofdifferentmomentsthatariseinthebaseflangeindifferentplanes,showninFigure23through
to Figure 26, the moments that the flanges are able to resist far outweigh the magnitude of the imposed
loads,andthusvalidatingthatthegeometricdimensions,andrelatedstrengthsaremorethanadequatefor
theconnectiondesign.
ntsinplane11
1(directiono fyaxisloadin
ng)forthebaseringflangeeconnectionofthe
Figurre23:Momen
Monopole
ebaseringfla
angeconnectiionoftheMo
onopole
Figure24:Momentssinplane22(directionofxxaxis)forthe
5:VonMisesMomentsintthebaseringflangeconnectionoftheM
Monopole
Figure25
omentsintheebaseringfla
angeconnectiionoftheMoonopole
Figure26:Trescamo
4.2STEEELLATTICEETOWERCONNECTION
O
NS:
4.2.1CONNECTIO
ONTYPE
Bolted co
onnections arre the most widely
w
used i n element co
onnections for steel latticee towers. Thiss element
connectio
ontypeissuitableforhotdipgalvaniseedmembersaandboltsthat aregalvaniseedandthencconnected
ds.Atolerancceofbetween
afterward
n1and2mmiisprovidedbe
etweenthediameteroftheeboltandthediameter
oftheho
ole.Memberssareoftenco
onnectedwithhoneboltonly,inparticularallredundaantmemberssinlattice
framewo
orksandbracingsoflessheavilyloadedttowers(Kiessliingetal,2003
3:376).
Nomoreethan6bolts shouldbeussedinalinefooraconnectio
on,sothatth
heindividualsstrengthofea
achboltis
utilised. Where the fo
ore mentione
ed connectionn type is inad
dequate, one may design aa connection with two
parallelliinesoffasteners.
Thewidtthoftheangllesectionsch
hosenwillby implicationalsodetermine
ethelargest possiblebolt diameter
allowed.Thearrangem
mentofthech
hosen/requireedboltsshouldbedoneinsuchawaythhatthearranggementof
thebolth
holesshouldp
preventcorrosionandloca lbucklingand
dfacilitatethe
einstallationoofbolts.
Figure27:EqualLeggAngleSectionHoleDimen
nsioning
Theparametermeasurementsthatarerecommeendedfortheaboveconnectionareasfoollows:
1.2
1.5
min 1.4 ; 200
2.5
Inaccordancewiththerecommendationsforcorrosionprevention:
&
40
Wheret,isthethicknessofthethinneroftheconnectedparts(Kiesslingetal,2003:376).
4.2.2CONNECTIONDESIGN:
4.2.2.1CHOOSINGTHEAPPROPRIATEBOLTS:
The connection design for Steel Lattice towers is determined by the necessity of bolts and or welds. The
connectionsbetweenbracingmembersandouterlegmembersintheSteellatticetowerweredesignedtobe
pinnedconnections.Inordertocreateanactualpinnedconnectionthebracingmembersweretobeboltedto
theouterlegmembers,andwhereappropriatetooneanother.
The initial calculations to determine what size bolts to use for the connections, tested the shear bearing
capacityofdifferentsizedboltsandcomparedtheirbearingcapacitytothemaximumaxialforcesdevelopedin
themembers.Itwasdeterminedthatinfactasinglebolthassufficientstrengthtoresistthelargestapplied
axialforce,howeveritwasdecideduponforthefollowingreasonstousetwoM16boltsateachconnection
insteadofjustone.
Ifoneweretoonlyuseoneboltattheendofeachmemberasaconnection,theconnectionwouldmostlikely
behavewellintension;howeveritwouldcauseasignificantlossinflexuralstiffnessattheconnectionlocation
intheinstanceofcompression.Flexuralstiffnessisimportantwhenthemembercarriescompressionbecause
Euler buckling is dependent on such stiffness. Because of the Kbracing configuration in a number of the
bracingmembers,itisacertaintythatsomeofthememberswillbeincompression,andsotheneedformore
thanoneboltwouldalreadybejustified.
Another reason to use more than one bolt is to introduce redundancy. Redundancy significantly increases
systemreliability,andsoimplicitlydecreasingriskassociatedwithfailurebyintroducingmorethanoneboltat
eachconnection.
Lastlyanotheradvantagefoundinusingmultipleboltsateachendisforpracticalconstructionpurposes;when
therearetwoboltholes,theoneholecanbeusedtoaligntheotherwithaspudwrench.
4.2.2.2CONNECTIONDETAILS:
Inorderforaconnectiontobetrulypinned,thelinesalongwhichthecentroidsofeachmemberlieneedtoall
meet at a point. If this is the case the pinned connection shall have no moments created by means of any
eccentricitiesatthepointoffixity.
Inorderforthethreeandinsomeinstancesfivemembersofeachnodalpointinthelatticetowertobepinned
connections,weldedplategirderconnectionshadtobedesignedbecauseofspaceconstrictions.
Theplatesofdeterminedstrengthandthicknesswerethentobeweldedontotheouterlegmembersatthe
appropriatepointsbymeansoffilletwelds,andthebracingmembersthenboltedontothoseplatesbymeans
oftwoM16boltswiththeappropriatespacing.
4.2.3CONNECTIONCALCULATIONS:
InTable11andTable12thecalculationsthatdeterminedthenumberofboltsofspecificsizefordifference
capacityrequirementsweredetermined.
The dimensions of the gusset plates were determined through three dimensional modelling of the pinned
connectionsonAutodeskinventorProfessional2011.Figure19showstheplacementofsuchaconnection.
AttachedintheConnectionsAppendixareisometricdetaileddrawingsofthegussetplateconnections.
Table11:OrdinaryBoltstrengthcalculationsforSteelLatticeTowerConnections
ShearBoltResistance
SANS101621:2005
Clause13.12.1.2
OrdinaryGrade8.8boltyieldstrength
70x70x6
100x100x8
g1
40 mm
g1
50 mm
dmax
16 mm
dmax
24 mm
0.8
0.8
Ab
452.3893 mm
fu
830 Mpa
Ab
201.0619 mm
fu
n
Vr
830 Mpa
2
56072.15 N
V r,n
112144.3
62016 N
Vu
Vu/V r
check
0.553002
OK
Vr
126162.3 N
Vu
62016 N
Vu/V r
check
0.491557
OK
Table12:BoltCapacityCalculationsforFatigueforLatticetowerConnections
70x70x6
100x100x8
g1
40 mm
g1
50 mm
dmax
20 mm
dmax
24 mm
0.8
0.8
Ab
452.3893 mm
Ab
314.1593 mm
fu
n
Vr
262 Mpa
3
27656.07 N
fu
n
Vr
262 Mpa
2
39824.74 N
V r,n
82968.21
V r,n
79649.48
Vu
V u/Vr
check
62016 N
0.747467
OK
Vu
V u/V r
check
62016 N
0.778612
OK
Table 12 shows the effects

e
of redu
uced bolt streength having been taken into account for fatigue, and
a hence
increasin
ngtherequirednumberofb
bolts.
Figure28:La
atticeTower, pinnedconnectionswithg
gussetplates
4.2.4G ENERALASPECTSPER
S
TINENTTOA
ALLCONNEC
CTIONTYPESDEALTWITTHINTHIS
CHAPTEER:
4.2.4.11WELDS:
FortheM
MonopoleTow
wer,theweld
dingrequired infabrication
n,wasthatofftheringflanngesweldedo
ontoeach
CHSofth
hetower.Theeweldsweresstipulatedas fullpenetrationwelds.The
eweldmetalw
wasdetailedasE70XX,
whichhaasatensilestrengthof480
0MPa,andbeecausethebassematerialoffthetowerissS355JR,with
hatensile
strengthof470MPa,theweldeffecctscanbeignooredandthettowercanbeanalysedasiffthereweren
nowelds.
Forthessteellatticeto
ower,thegusssetplateswhiichwereillusttratedinFigurre28needed tobewelded
dontothe
outerleggsofthesteeltowerinthefabricationp rocess.Thew
weldswerestip
pulatedasfill etweldsinacccordance
with SAN
NS 101621:2005. The top
p 2.5m of thee tower where the eleme
ents were 50xx50x5 equal leg angle
sections,werealsoprreweldedtoggetherinthe fabricationphasebyfillet weldsdesignnedinaccorda
ancewith
SANS1011621:2005.
SANS1011621:2005
13.13.2.22FilletWelds
Thefacto
oredresistancceforshearandtensionshhearorcomprressioninduce
edshearshalllbetakenas thelesser
of
a) Forthebasem
metal

b) Fortheweldm
metal

0.67
0.67
;or
1.00
0.50
Whereistheangleoftheaxisoftheweldwiththelineofactionoftheforce(zerodegreesforalongitudinal
weld and 90 degrees for a transverse weld), and the other terms are defined in 13.12.2.1. Conservatively,
(1+0.5sin1.5)canbetakenas1.
Extractsfrom13.12.2.1:
Amistheshearareaoftheeffectivefusionface(mm2),fuisthetensilestrengthoftheparentmetal(MPa),Aw
is the area of the effective weld throat, plug or slot (mm2) and Xu is the tensile strength of the weld metal
(MPa).
4.2.4.2BOLTS:
For a structure like the wind turbine supporting towers in this thesis, it is recommended that pretensioned
bolts are used in the design of the connections. In accordance with the South African Steel Construction
handbook(SASCH),thefollowingisdetailedwithregardtoHighStrengthfrictiongrip(HSFG)bolting.
Frictiongripboltedconnectionsareconnectionsinwhichshearforceistransmittedbythefrictiondeveloped
betweenthefrayingsurfacesoftheconnectedparts,whichareclampedtogetherbyhighpretensionedforces
inthebolts.Theycontinuebystating,Theboltsareofhighstrengthmaterialandarepretensionedduring
installationtoaforceofatleast70%oftheirtensileresistance.Furthermore,ascanbeseeninSANS10162
1:2005clause22.2.2partdwhichisdescribedbelow,foranystructureswhichincludethepossibilityofcyclic
loadingHSFGboltsmustbeused.
Inafrictiongripboltedconnection,itisnecessarytoensurethataftertheboltshavebeentightenedtheyare
alltensionedtoaminimumvaluegiveninthetablebelowwhichoutofSANS10094:
SANS10094
Table1MinimumBoltTension
Nominalsize
MinimumbolttensionT(kN)
ofBolt
Class8.8S
Class10.9S
M16
91
114
M20
M24
M30
M36
142
205
326
475
178
257
408
595
6.2.2.2Washersusedwithclass10.9Sbolts
Whenclass10.9Sboltsandclass10Snutsareusedtojoincomponentsofanygradeofsteelofan
ultimatetensilestrengthoflessthan585MPa,hardenedwashersshallbefittedunderbothbolt
headandnutor,inthecaseofcomponentsofhigherstrength,fitawasherunderonlythemember
thatisrotated.Ensurethatbearingsurfacesarenormaltotheaxisofthebolttowithin1,and
whereataperedwasherisusedtoachievethis,provideaflat,hardenedwasherbetweenthe
taperedwasherandtheboltheadornut.
Note,onlyClass10.9nutsmaybeusedwithclass10.9Sbolts.
SANS101621:2005
13.12Bolts
13.12.1.3Boltsintension
Thefactoredtensileresistancedevelopedbyaboltinajointsubjectedtotensileforce,shallbetakenas:
0.75
Whereb=0.8,andthecalculatedtensileforce,Tu,isindependentofthepretensionandshallbetakenasthe
sumoftheexternalloadplusanytensioncausedbypryingaction.
Ahighstrengthboltsubjectedtotensilecyclicloadingshallbepretensionedasforfrictiongripconnections(see
clause22).Connectedpartsshallbearrangedsothatpryingforcesareminimised,andsoinnocaseshallthe
pryingforceexceed30%oftheexternallyappliedload.Thepermissiblerangeofstressunderspecifiedloads,
basedontheshankareaofthebolt,shallnotexceed214MPaforclass8.8Sboltsor262MPaforclass10.9S
bolts.
InLieuofcalculatingtheactualfatiguestressrange,whichrequirestheeffectofboltpretensiontobetaken
intoaccount,thestressrangemaybesimplytakenasthecalculatedstressbasedontheshankareaofthebolt
underspecifiedloads,includingthepryingforce,andindependentofthepretensionforce.
SANS101621:2005
Clause22.DesignandDetailingofboltedconnections.
22.1General
Thisclausedealsprimarilywithclass4.8,8.8,8.8S,10.9and10.9Sboltassembliescomplyingwiththerelevant
parts of SANS 1700, of equivalent fasteners. The bolts might or might not be required to be installed to a
specificminimumtensiondependingonthetypeofconnection.
22.2Designofboltedconnections
22.2.2Useofhighstrengthfrictiongripbolts
Pretensionedhighstrengthfrictiongripbolts(class8.8Sor10.9S)shallbeusedin
a)frictiongrip connections where slippage cannot be tolerated (such connections include those subject to
fatigueortofrequentloadreversal,orthoseinstructuressensitivetodeflection),
d)connectionssubjecttoimpactorcyclicloading
e)connectionswheretheboltsaresubjecttocyclictensileloading(see13.12.2.3).
Thedesignproceduredevelopedfortheringflangeandgussetplateconnectionshavebeenhandledinthis
chapterandthedetailedconnectioncalculationsareshowninAppendixB.
CHAPTTER5:FOU
UNDATION
NDESIGN
This chap
pter examinees the details of the two ffoundations designed
d
for the Steel Moonopole, and the Steel
latticeto
owerrespectivvely.
Firstthe designoftheeSteelMonopoleTowers Foundationsshallbereviewed,followeedbythedesiignofthe
Steellattticetowersfo
oundation.
5.1INTTRODUCTIONTOFOUNDDATIONS
The purp
pose of a foundation is to transfer the loads from the structure to the groundd, without ca
ausing the
ground tto fail in sheaar or to allow
w excessive seettlement of the structure
e to occur (M
Marshall and Robberts).
R
These reequirements are
a met by ensuring
e
the bearing presssure below the
t foundatioon does not exceed a
specified
dpermissibleb
bearingpresssure.ThisisdooneinaccordaancewithSAB
BS0161(19800).
AccountingforthedesignofthefoundationinteermsoftheSStrand7modelisalsoimporrtantfromthepointof
viewoftthestiffnesso
oftheground,,foundationaandthewayiinwhichthettowerinteracctswiththefo
oundation
and stiffn
ness of the su
upporting gro
ound. The impportance in th
he stiffness arises due to tthe possible effect
e
the
stiffnessofthebaseoffthetowerw
willhaveontheeresultsobtaainedfromdyn
namicanalyseesdoneonthetower.
The foun
ndation desiggn is also exttremely spec ific to the siite of installa
ation. The stiiffness of the
e ground,
mentioneedabove,isrelatedtotheelasticmodu lusofthegroundatthesite.Thusonenneedstoknow
wthetype
of ground, the densityy of the grou
und as well ass the other so
oil propertiess such as Poisssons ratio and elastic
modulus.Inordertoo
obtaintheabo
oveparamete rs,suitableso
oiltestsmustbedone(MarrshallandRob
bberts).
Forthep
purposesofco
ompletenesso
ofdesignand alsoamorecompletecostanalysisitw
wasdecidedupontodo
aprovisio
onalfoundatio
ondesignforaselectedgrooundtype.
windturbinefoundationdesign
Figurre29:Failedw
60|C h a p t e r 5 : F o u n d a t i o n D e s i g n
5.2STEEELMONOPPOLETOWERFOUNDATTIONDESIG
GN
5.2.1F OUNDATIO
ONTYPE
Inaccord
dancewiththeeCementand
dConcreteInsstitutesAnallysisandDesig
gnofConcretteStructures,asimple
footingw
wasdesignedfortheMonopoleTowerfooundation.
Figure30:M
MonopoleFoundationType
Figure330showsthellayoutofthessimplefootinggwithacolum
mnprotruding
goutofitwithhanapplieda
axialforce
andbend
dingmoment..Itwasdecide
edupontohaaveafoundationthatlooke
edgeometricaallyexactlythesameas
Figure 330, where the tower wou

uld lie on topp of the column. This deccision was baased on the calculated
c
moment applied to th
he foundation
n, which indu ced such a laarge propensitty for overturrn that the fo
oundation
designw
wouldcomply withtheneed
dtobeburieddacertainde
epthbelowthegroundtohhelpresistthe
eoverturn
forces.
5.2.2F OUNDATIO
ONDIMENSIONS
The dimeensions of th
he concrete fo
oundation weere decided upon
u
by means of a staticc equilibrium equation
whichtooktheapplieedmomentoffoverturn(M )equatedto theconcretefoundationaandtopsoils capability
n (MR) about the most likeely point (O), see Figure 31
3 . This resullted in the ap
ppropriate
to resist that overturn
dimensio
onalvolumes ofconcrete,a
aswellastheedepthtowh
hichthesoilw
wouldneedtoobeexcavate
edtobury
thefound
dation.InacccordancewiththeSouthAffricanNationaalStandards(S
SANS),thesaffetyfactorforroverturn
of1.5waasapplied,MR/M=1.5.
Figgure31:MonoopoleFounda
ationEquilibriium
5.2.3F OUNDATIO
ONCALCULA
ATIONS
Once thee appropriatee foundation dimensions had been ob
btained, the next
n
step waas to determiine which
category of design the foundation fitted intoo to determin
ne whether the
t induced bearing presssure was
acceptab
ble. The desiggn category was
w based on the ratio bettween the ap
pplied momennt divided byy the axial
load,whichwasthencomparedtoatheoretical ratioofD/6,whereDisthebasewidth ofthefounda
ation.The
ratioisth
henusedtottransformthe
ecombinedm
momentandaxialforceinto
oasingleforccetoconsider,anaxial
load,wittheccentricityye.Thisratio (D/6)andth euseofitwh
hichfollowsw
wereprescribeedbytheAnalysisand
DesignoffConcreteStrructuresbyJo
ohnM.Robbe rtsandVerno
onMarshall.
Figure32
2:Monopole Foundatione
equilibriumeq
quivalent
Intheprocessoftranssformingthecoactingapppliedmomenttandaxialforcetoactasaasingleaxialfforcewith
aneccen
ntricity(which
hwasdeterminedfromtheeratiobetweenmomenta
andaxialforcee),oneneedsstomake
suretotakeintoacco
ounttheincreasedmagnituudeinaxialfo
orceasaresultoftheselfw
weightofthe
econcrete
foundatio
onfootingaswellastheso
oilthatistob eplacedontopofit.Thisttendstovastlyyreducetheratio,and
distanceeiin Figure32 wouldbe

fromasiituationwherrewithanexttremelylarge momenttheeccentricityd
extended
d beyond thee distance of the base dim
mension, it iss brought to a reasonablee distance within that
dimensio
onwhichthen
nallowsoneto
ocalculatetheeappliedbearingpressure.
The applied bearing pressure is calculated neext which in the instance
e of the aboove design ca
ategory is
matedasalineeardistributio
onbeneathth econcretefooting.ThiscanbeseeninFFigure33.
approxim
Figure33
3:MonopoleFFoundationB
BearingPressu
urelayout
Thedesiggnequations usedtocalcu
ulatethemaxximumandm
minimumbearingpressuressbelowthefo
oundation
dependo
onthedesigncategoryofD
D/6andwhetthereisbigggerorsmallerthanthisrattio.Forthefo
oundation
designed
dfortheMono
opoleinthistthesis,itwasddesignedforb
boththeratio ofe>D/6andde<D/6.Thed
difference
in bearin
ng pressure distribution be
elow the founndation for th
he two differe
ent ratios aree shown in Figure 34,
theseweerebroughtab
boutbychanggingthebaseddimensionofthefoundatio
onfrom3m,(ee>D/6)to4m(e>D/6).
Mono
opoleFoun
ndationBearingPre
essure
D
Distributio
on
120
Pressure(kN/m3)
100
80
60
e>D/6
e<D/6
40
20
0
0
Baseof Foundation((m)
Figure34:M
MonopoleFouundationBearringPressureDistributions
Due to tthe overall maximum bearring pressure of the widerr footing (D=4m) being siggnificantly sm
maller and
havinga moreevenpressuredistrib
butionitwas decidedupon
ntousethefo
oundationwitthabasedim
mensionof
D=4m.
The desiign moment for such a foundation
f
w
was calculated
d as can be seen in the Monopole fo
oundation
calculatio
ons Appendixx C. The theo
oretical desiggn moment was
w closely comparable too the design moment
calculateedinthefiniteeelementpacckageProkon,,anditwasde
ecideduponttodotherebaarquantitycalculations
andlayou
utusingProko
on,seeFigure
e35andFiguure36.
Figure35
5:ProkonInpputLayoutofMonopoleFoundation
Figure
e36:ProkonO
OutputforMo
onopoleFoun
ndation
put,Figure37
7showsthebbendingscheddulewhichwaasgenerated, whichforcosstanalysis
FromtheeProkonoutp
purposessenablesonetocalculatettheamountoffreinforcingssteelrequired inafoundatioonofthisnature.
Figure37:MonopooleSchematiccBendingsch
hedule
5.3STEELLATTICETOWERFOUNDATIONDESIGN
5.3.1FOUNDATIONTYPE
ThefoundationdesignfortheSteellatticetowerwasmuchsimplerthanthedesignprocessnegatedbythe
largeoverturnmomentsoftheMonopoleTower.Thesteellatticetowerbynaturereducesthemagnitudeof
theforcesatthefoundationbecauseofitsstructureandabilitytospreadloadsthroughoutitsmemberswith
pinnedendconnections.
The Steel lattice towers outer supporting legs, being equal angle sections, needed to be welded onto
appropriatelysizedsteelbaseplateswhichwouldthenbeboltedontotheconcretefoundationbymeansof
twoboltstomimicapinnedconnectionwhichwouldtransfernomomentstothefoundation.
5.3.2DESIGNINGTHESTEELBASEPLATES
Duetothenatureofapinnedconnection,whichwastheconnectionforwhichthebasepointsofeachlegwas
designedandmodelled,axialloadsaretheonlyforcesthatneedtobetakenintoaccountfromadesignpoint
ofview.Theloadingonawindturbinesupportingtowerneedstobeexaminedfortheinstancesofthelargest
loadsbeingmovedfromonelegtotheotheralongwiththemotionsofthewindturbine.
For this reason the largest axial forces developed in a single leg for the ultimate limit state were utilised in
designingtherequirementsofthebaseplateforeachofthelegs,whichwerealltobeidentical.
The Ultimate Limit state loads from Load Case two, where the wind loading was applied to the tower
diagonallywereutilisedsincetheydevelopedthelargestaxialforcesinthelegs.
The steel base plates were designed in accordance with the South African Steel Construction Handbook,
section4.2.2Concentricallyloadedbases,derivedfromBS5950:2000.
a)
Firstly the required steel area of the base plate required in terms of the applied axial force is
calculated, which can then be used as a lower bound parameter as a minimum area for the base
plate.
0.6
,
b) Secondlytherequiredeffectiveareaaroundthecolumnbaseneedstobedetermined,anareawhich
isprojectedatleastadistancecawayfromthecolumnbaseallthewayaroundthecolumn,see
Figure 38. In order to determine the value of parameter c the following quadratic equation is
solved:
5

.

Figure38:EEffectiveBase
ePlateArea
c)
Lastly,thethiicknessofthe
eplateneedsttobedetermiinedfromthefollowingequuation:
3
0.9
9
Wherew=preessureunderttheplate(<=00.6fCU),andfY istheyieldstressoftheplaate.
Onemustjusstcheckthattthebaseplateesthickness isnotlesstha
anthethickneessoftheflan
ngeofthe
column.
Oncetheeabovethreestepshavebeendone,arrationalcalculationcanbemadetakingiintoaccountttheabove
proposed
d minimum dimensional parameters as well as the minimum spacing aallowances fo
or bolted
connectio
onssuggested
dintheSouth
hAfricanSteeelConstructio
onHandbook todetermineethemostap
ppropriate
platesizeeforthebaseplates.
5.3.3DESIGNINGTHECONCRRETEFOUNDDATION:
In order to obtain a concrete rein

nforced founddation compaarable to the
e kind of foottings designe
ed for the
Monopole Tower, Pro
okon was utilised. Howeveer, the design
n was far morre simple, sin ce it consiste
ed of four
smaller ssquare foundaation blocks that
t
were plaaced on top of
o the ground
d, which also contained co
omparably
littlesteeelreinforcing..Theonlyloadappliedtoeeachoftheco
oncretefound
dationswasannaxialload,w
whichwas
appliedo
ontopofthessteelbaseplate.
Fiigure39:SteeelLatticeTow
werFoundations
In Prokon thefoundationfootings shown in Figgure 39 were modelled,ho
owever, in ordder to model the steel
baseplattessothatth
heanalysisinccorporatedthheloadsapplie
edtothem,theyweremoodelledasinfinitelyflat
columnstubs.BecauseProkonha
asbeenprogrrammedtosu
uggeststeelre
einforcingint hecolumnsectionof
thebase,itisshowniintheoutput fortheelem entsofthem
modeltheba
modelwhichw
wereusedtom
aseplates.
Howeverr,sincetheyaareinfinitelytthinandmakkenophysicalallowancefo
orreinforcinggthatcanofccoursebe
ignoredaasnosuchthingexists,thecolumnstubbsarepurelyytheretobetttersimulatettheappliedloadsinthe
foundatio
on.
Figure 400 shows the input diagram
m for design in the finite element pacckage Prokon . Figure 41 shows
s
the
output d
design from Prokon,
P
and the bending sschedule for the
t steel rein
nforcement reequired in each of the
concretefoundationfo
ootingsisshowninFigure442.
Fiigure40:ProkkonLatticeFo
oundationInp
put
Figgure41:ProkoonLatticeFou
undationOutput
Figure4
42:ProkonBeendingschedu
uleLatticeFou
undation
5.4CONCLUSION
AllofthequantitativeinformationgatheredfromthefoundationdesignsfortheSteelMonopoleandtheSteel
Lattice towers is used in the feasibility chapter to provide greater insight into the effects the different
foundationdesignshaveontheoverallfeasibilityofthedifferenttowertypes.
The design procedure utilised in designing the foundations was handled in this chapter, and the detailed
calculationsareshowninAppendixC.
CHAPTER6:FINITEELEMENTANALYSES
6.1INTRODUCTION
ThischapterexaminesthefiniteelementmodellingofthetwoSteelTowersdesignedforthisthesis.Firstthe
wayinwhichthesteelMonopoleTowerwasmodelledwillbedescribed,followedbytheresultsobtainedfrom
the analyses performed on the tower. Thereafter, the modelling procedure for the steel lattice tower is
discussedandtheresultsoftheanalysesperformedforthattower.
Theanalysesthatwereperformedoneachofthetowerswerethefollowing:
1.
2.
3.
4.
LinearStaticAnalysis
LinearBucklingAnalysis
NonlinearAnalysis
DynamicAnalyses
i)
NaturalFrequencyAnalysis
ii)
Effectsofanoutofbalancerotor
iii)
Effectsofvortexsheddingduetowindaction
iv)
HarmonicFrequencyAnalysis
v)
StressstatesandFatigueassessment
vi)
ModalMassParticipationEvaluation
6.2STEELMONOPOLETOWER:FINITEELEMENTANALYSES
The Steel Monopole Tower designed for this thesis was modelled in the finite element analysis package
Strand7.Themonopolewasmodelledwithplateelementsofvaryingthickness;itwasmodelledasbeingfixed
atthebaseandwiththeloadingeffectsofasimulatedwindturbinemachineactingonthetopofthetower.
Theeffectsofthewindturbinemachineweresimulatedbymeansofcreatingaconicalformationofrigidlinks
at the tower top to which a further rigid link was attached in order to make allowance for the simulated
eccentric centre of mass of the combined nacelle and rotor of the wind turbine machine, modelled as a
lumpedmassattheendoftheextendedrigidlink(seeFigure43below).Theforcethatshallbedescribedas
theoverturnforceinthefollowingsectionsaccountedfortheactionofthewindonthebladesoftheWind
TurbineMachine. Thecalculation details of the overturn force weredescribed in Chapter 3 in section 3.3.2.
Theoverturnforcesimulatedtheactionofthewindonthebladesoftherotorasafunctionofhubheightand
windspeedandwasappliedatthepeakoftheconicalformationofrigidlinksinStrand7,seeninFigure43.
InStrand7differentaspectsofthetowerweremodelledasseparateloadcasesandthencombinedbymeans
of linear load case combinations which accounted for the actions defined for a number of different
ServiceabilityandUltimateLimitstatesinaccordancewithSANS101601:2011.
71|C h a p t e r 6 : F i n i t e E l e m e n t A n a l y s e s
Figure43:Mo
onopoleTow
werWindTurb
bineConnectio
onSimulationn
edasseparateeloadcases;theselfweig
ght(SW)ofth etower,thed
deadload
Thefollowingaspectsweremodelle
w
action oon the tower, and the
(DL) weigght of the simulated equipment (windd turbine machine), the wind
UltimateLimitState(U
ULS)andServviceabilityStatte(SLS)overtturningaction
ns.Theovertuurningactionisdefined
by a horizontal force applied at hu
ub height andd is dependen
nt on the swe
ept rotor areaa and the approaching
wind speeed, wanting to push the tower over. The namingg of the overturning forcees as the ULSS and SLS
overturningforcesresspectivelywasdonesoin accordancew
withtheexpla
anationgiven insection2.2
2forhow
theservicceabilitylimittstateoverturrnforcefrom thewindisdefinedupuntil16m/sandttheultimatellimitstate
isdefined
dfrom16m/sswindupwarrds.Theyare loadscasestobeusedintheloadcom
mbinationsstip
pulatedin
Table13.IncludedinttheSLSovertu
urnandULSooverturnloadcasesarethe effectsoftheewindactingalongthe
tower.
Theoverrturningaction
nswerecalcu
ulateddifferenntlyfortheULSandSLSbe
ecausetheopeeratingwindspeedsof
the3kW windturbineisupto16m//s.Atspeeds beyond16m//stheturbinerotorturnsooutofthewindsoasto
avoiddam
magetotheu
unitathigherwindspeeds. Forthisreaso
onthewindsp
peedof16m//swasusedto
ocalculate
the SLS fforces acting on the towerr induced by wind, and the peak wind speed as speecified by SAN
NS 10160
3:2011w
wasusedforttheULSforcesactingonthhetowerindu
ucedbywind.Table13illuustratesthelinearload
combinattionsthatareeinaccordanccewiththeloaadfactorsforULSandSLSiinSANS1016001:2011.Inco
orporated
intheOvverturn(ULS&
&SLS)Loadca
asesrespectivvelyarethew
windactionsa
actingonthettowerforthe
edifferent
windspeeedsaswasdeescribedabovve.
Tab
ble13:MonoopoleLoadCaseCombinations
LoadCases
ULS1(SW)
ULS2(No
oWind)
ULS3(W
WorstCase)
SLS1(SW)
SLS2(N oWind)
SLS3(W
WorstCase)
1.2
1.2
1.2
EquipmentD
DL
1.2
1.2
Overturn(ULS),windspeed>16
6m/s
1.5
Overturn(SLLS),windspeed<16m
m/s
0.6
SW
ontheMonoppoletowersothatitcouldbeassessedinntermsofsta
abilityand
Thefollowinganalysessweredoneo
deflections,aswellashowitwouldrespondtoddynamicloadin
ngeffects.
6.2.1L INEARSTATTICANALYSSIS
Thelineaarstaticanalysisisutilisedttoexaminethhedeflectionscausedbythelinearloadccombinationssshownin
Table13..
AlinearstaticsolutionbyStrand7isobtainedassumingthatthebehaviourofthestructureislinearandthat
theloadingonthestructureisstatic(staticloadsdonotchangeinmagnitudeordirectionwithtime).Forthe
response of a structure to be linear, the mechanical behaviour of all materials in the model must follow
Hookeslaw;thatis,elementforcesarelinearlyproportionaltoelementdeformationandwhentheloadingis
removed,thematerialreturnstoitsoriginalshape.Inaddition,thestructuresdeformationmustbesosmall
afterloadingthatthedeformedgeometryisundistinguishablefromtheoriginal,unloadedgeometry.Because
ofthesetwoassumptions,solutionscanbearbitrarilycombinedtoconsidermorecomplexloadingconditions,
suchastheloadcombinationsshowninTable13forthemonopolemodelledhere(Strand7theorymanual).
Allthedeflectionsthatwereobtainedforthedifferentloadcaseswereexamined.Thedirectioninwhichthe
towerwasloadedinthemodelwasintheYdirectionwithregardtohowthewindandoverturnforceswere
imposedonthetower.
Figure 44 illustrates the load combination for which thedisplacementof the top of the towerfor the linear
staticanalysis(intheYdirection)isassessed,whichistheServiceabilityLimitstateloadcaseforwhichallloads
areincorporated.
The output from Strand7 comes in the form of the visual displays in Figure 44, which shows the deformed
towersdeflectionsexaggeratedbyafactorof10%forvisualeffect.Strand7alsodisplaysalogfile,whichwas
usedforthemaximumdisplacementsoftheserviceabilitylimitstatetabulatedinTable14.
Figure44:Mon
nopole:LargeestStaticdisplacementloadcombinatioon.
Table14:M
Maximumdisp
placementmaagnitudesofM
Monopolefor ServiceabilityyLimitState
Case
e
DX(mm)
DY(mm)
DZ(mm)
D
SW
0.000
0.01
0.06
EquipmentDL
0.000
0.39
0.02
Overrturn(ULS)
0.022
33.23
0.98
Overrturn(SLS)
0.000
4.23
0.13
Thedefleectionsthrougghouttheheiightofthetoowerareshow
wngraphically
yinFigure45 fortheanalyysisofthe
Ultimatelimitstate(W
Worstcasesce
enario)inwh ichallloadsaareappliedwiithmaximumloadingfacto
orsaswas
shownin
nTable13.
Lin
nearDisp lacementMonopole,
ULSS(WorstCa
ase)
DisplacementDY(mm)
60.00
50.00
40.00
30.00
20.00
10.00
0.00
10.00
11
16
m)
HeighhtofTower(m
Figure4
45:Monopolee,Lineardispllacementdisttribution
THENODDALREACTION
NSATTHEB ASEOFTHE TO
OWER:
Thelineaarstaticanalysisresultswerealsoutiliseedtodetermin
nethenodalrreactionsatthhebaseofthe
etowerin
orderto verifythecho
oiceofholdin
ngdownboltssaswellasob
btainingtheb
baseforcesanndreactionsn
neededto
designth
hefoundation.
Figure477,Figure48aandFigure49 illustratetheeplateforced
distributionsa
asresultantfoorcesintheX
X,YandZ
direction
nsrespectivelyy,undertheB
Baseofthetow
wer.Figure47
7,Figure48an
ndFigure49w
wouldbeapplicablefor
anypointsaboutthe verticalzaxissfortheinstaancesinwhichthewindtu
urbinerotatesswiththewin
ndandso
thedesiggn ofthebaseflange need
ded to be donne in such a way
w as toressist the largesst possibletension and
compresssionforcesth
hatweredevelopedhere.
Thereacttionsareshow
wnrelativetotheaxessysttemshowninFigure46.
Figure46:Axeessystemforn
nodalreactions
Figure47:FX(N) BaseReactioonForces,ULSS(WorstCase
e)Combinatioon3.
Figure48:FY(N)BaseReactioonForces,ULSS(WorstCase
e),Combinatioon3
Figure49:FZ(N)BaseReactioonForces,ULSS(WorstCase),Combinatioon3
LINEARSTATICANALYSISSTRESSSTATES:
InordertoassessthestressstateoftheSteelMonopoleforthelinearstaticanalysis,thestressdistributions
fortheloadcombinationULS3asdescribedinTable13wereexamined.IntermsofthevonMisesStressstate
oftheplateelements,aswellastheTrescaStressstateoftheplateelementsnoneofthestressstatesexceed
theelasticmateriallimit(355MPa)oftheS355JRgradesteelthattheSteelMonopoleTowerwasdesignedto.
6.2.2LINEARBUCKLINGANALYSIS
BeforepresentingtheLinearBucklinganalysisresults,anoverviewofthelinearbucklingsolverisaddressed.It
isnotedherethatthelinearbucklinganalysisisanupperboundsolution.
LINEARBUCKLINGSOLVER:
TheLinearBucklingSolverinStrand7calculatesthebucklingloadfactorsandcorrespondingmodeshapesfora
structureundergivenloadingconditions.Itisbasedontheassumptionsthatthereexistsabifurcationpoint
wheretheprimaryandsecondaryloadingpathsintersect,andbeforethispointisreached,allelementstresses
changeproportionallywiththeloadfactor.
Alinearbucklingsolutionisobtainedfromsolvinganeigenvalueproblem:
Where;
[K]istheglobalstiffnessmatrix;{x}isavectorofthebucklingmodes,isthebucklingloadfactorand[Kg]is
theglobalgeometricstiffnessmatrix.
The geometric stiffness matrix, also known as the initial stress stiffness matrix, is a symmetric matrix
dependent on the element stress level. It reflects the effect of the geometric change on elements from a
known stress value. For beam and plate bending structures, the geometric stiffness matrix represents the
stiffeningeffectduetothetensileaxial/membranestresses.
Thebucklingsolutionisonlypossiblewhenanexistingsolutionisavailablefordeterminingthecurrentstress
state of the structure, which is required for the calculation of the element geometric stiffness matrix. In
Strand7,boththeLinearStaticSolverandtheNonlinearStaticSolversolutionscanbeusedtostarttheLinear
BucklingSolversolution(Strand7Software).
TheLinearBucklinganalysisdeterminesabifurcationpoint,wheretheprimaryandsecondaryloadingpaths
intersect. At the bifurcation point more than one equilibrium position is possible. The primary path is not
usuallyfollowedafterloadingexceedsthispointandthestructureisinthepostbucklingstate.Theslopeof
thesecondarypathatthebifurcationpointdeterminesthenatureofthepostbucklingbehaviour.Apositive
slope indicates that the structure will have post buckling strength whilst a negative slope means that the
structurewillsnapthroughorsimplycollapse.
Realstructureshavegeometricandloadingimperfections,oftencausingtheprimarycurveandthebifurcation
pointtodisappear.
Forrealstructures,linearbucklinganalysesarebestforpreliminarydesign,andstudyingtheeffectsofvarious
parameters.
There are many factors present in real structures that have a large influence on the stability and critical
bucklingload.TheLinearbifurcationanalysisneglectsalloftheseandassumesthatthestructureisperfect.
Consequently,thepredictedbucklingloadwillalwaysbeanoverestimateofthatforarealstructure(Strand7
Software).
Strand7calculatesthebucklingfactorsforeachloadcombinationseparately.Figure50illustrateshowtheload
bucklingfactorsdifferinmagnitudeforthedifferentloadcombinations.
BucklingFactors(20Modes)
3.50E+03
BucklingFactor
3.00E+03
2.50E+03
ULS(SW)
2.00E+03
ULS(NoWind)
1.50E+03
ULS(WorstCase)
SLS(SW)
1.00E+03
SLS(NoWind)
5.00E+02
SLS(WorstCase)
0.00E+00
5.00E+02
11
16
Modenumber
Figure50:Monopole,LinearBucklingFactors
Table15:Monopole,LinearBucklingFactors
Mode
SW
1
2
3
4
5
6
ULS(SW)
135.5
136.0
914.8
918.1
2313.4
2319.1
ULS(NoWind)
90.3
90.6
609.8
612.1
1542.3
1546.0
99.6
100.0
678.0
680.4
1686.3
1689.5
ULS(WorstCase) SLS(SW)
51.2
53.0
53.6
55.0
55.5
57.3
135.5
136.0
914.8
918.1
2313.4
2319.1
SLS(NoWind)
SLS(WorstCase)
119.5
83.2
119.9
86.4
813.6
87.9
816.4
89.1
2023.5
90.9
2027.4
92.1
In Figure 50, the ULS and SLS worst case scenario load cases both stopped converging after two modes,
whereasthereststoppedconvergingafter6modes.Thiscanmostlikelybeexplainedbythecomplexityofthe
algorithmsusedinsolvingthebucklingloadfactorsinStrand7.Table15showsthebucklingfactorsforthefirst
6modesofeachloadcombinationthatwasdefinedinTable13.Anybucklingfactorlargerthan1impliesthat
buckling is not expected but ultimate limit state load factors apply. Where negative buckling factors are
shown,theyindicatealoadreversal,forexampleratherthantheloadbucklingfactorapplyingfortension,it
occurswiththemagnitudeshownbutforcompression.
6.2.3NONLINEARANALYSIS
In thenonlinear analysis, due to the incremental nature of the solution,a number of load stepsneed to be
defined, for which a number of equilibrium iterations is done. It is no longer possible to have linear load
combinationsinanonlinearanalysis,becausethemannerinwhichthedifferentloadingcomponentsaretobe
appliedtothestructurefollowstherealworldloadingsequence.
Inthenonlinearanalysis,theselfweightofthestructurealongwiththedeadloadoftheappliedequipmentis
imposedandanalysedinitially.Theresultsofthisinitialanalysisarethenutilisedastheinitialconditionsfora
secondnonlinearanalysisinwhichthewindloadsareapplied.Theloadfactorsaretakenintoaccountatthe
beginning of each of these analyses by means of multiplying the overall applied loads by the appropriate
factors.
Althoughthemajorityofengineeringstructuresoperatewithinthelinearregimeandtheassumptionsmade
for the linear solution are valid, there is a wide class of problems that do exhibit nonlinear behaviour. The
nonlinearstaticsolverpredictsthebehaviourofstructureswithnonlinearbehaviourbeingtakenintoaccount.
In Strand7, three different types of nonlinearities can be included: geometric, material and boundary
nonlinearity.
For the nonlinear analysis performed on the Monopole tower, the loading was applied in 20 increments of
equalsize,startingfromazeroloadandendingupattheappropriatelyfactoredload.
Forthesamedesigncaseaswasillustratedaboveinthestaticlinearanalysis,thestaticnonlinearanalysisof
themonopolewasdoneasfollows.Firstastaticnonlinearanalysiswasdonewheretheselfweightanddead
loadoftheequipmentweretheonlyappliedloadswithultimatelimitstateloadingfactorsof1.2.Asecond
nonlinearanalysiswasdone,wheretheresultsfromtheinitialanalysiswereusedastheinitialconditions.In
thesecondnonlinearanalysis,onlythewindandoverturningforceswereappliedwithaloadfactorof1.5,for
theultimatelimitstate.Figure51illustratestheprogressionofthemaximumdisplacement,whichoccurredat
the top of the tower, relative to the load increments applied. The Strand7 Nonlinear Static Solver uses an
algorithmbasedonNewtonRaphsonmethod.
DisplacementDY(mm)
NonlinearDisplacementVsLoading,Monopole,
ULS(WorstCase)
50.00
45.00
40.00
35.00
30.00
25.00
20.00
15.00
10.00
5.00
0.00
SecondNonlinear
Analysis
InitialNonlinear
Analysis
1
11
16
LoadIncrements
Figure51:MonopolenonlineardisplacementVs.Loading
Because the scale of displacements for the initial and second nonlinear analyses is so different, the initial
analysis displacement where only the selfweight of the tower and the dead load of the equipment were
appliedisplottedseparatelyinFigure52:
InitialNonlinearAnalysisDisplacementVs
Loading,Monopole,ULS(WorstCase)
DisplacementDY(mm)
0.50
0.40
0.30
0.20
0.10
0.00
1
11
16
LoadIncrememts
Figure52:InitialNonlinearAnalysis:MonopoleULS(WorstCase)
ThedisplacementsillustratedinFigure52essentiallyactasinitialgeometricimperfectionsinthesecondstatic
nonlinear analysis that is done, and so the results obtained from the final nonlinear analysis are less
conservative in terms of the bucking potential and the degree of deflection to be concerned with in
comparisonwithalinearbucklinganalysis.
6.2.4DYNAMICANALYSES
Thedynamicforcesinvestigatedherewerenotnecessarilythegoverningloadcasesintheoveralldesign;they
werehoweverassessedasaccidentalloadcases.
ThedynamicanalysesperformedonthetowersinthisthesiswereprecededbyaNaturalFrequencyanalysis,
performedbytheNaturalfrequencysolverinStrand7.Followingthenaturalfrequencyanalysis,thefollowing
aspectswereexamined:
1.
2.
TheeffectsofanOutofBalanceRotor
TheeffectsofVortexShedding
NATURALFREQUENCYSOLVEROVERVIEW:
The natural frequency solver is used to calculate the natural frequencies (or free vibration frequencies) and
corresponding vibration modes of an undamped structure. The natural frequency analysis problem is
formulatedasthefollowingeigenvalueproblem:
Where[K]istheGlobalstiffnessmatrix,[M]istheglobalmassmatrix,{x}inthevibrationmodevector,and
isthecircularfrequency(radians/sec),naturalfrequency=/2[Hz].
Theresultsobtainedfromanaturalfrequencyanalysisincludeasetofeigenvalues,andtheircorresponding
eigenvectors.
Theeigenvaluesobtainedfromtheeigenvalueproblemrepresentthenaturalfrequenciesforeachvibration
mode.Theeigenvectorsrepresentavectorofnodaldisplacementsthatdefinethefreevibrationmodeshapes
ofthestructure.Foreachofthecalculatednaturalfrequencies,thereisonecorrespondingeigenvector.Note
herethatthemodeshapes(eigenvectors)aresetsofrelativedisplacementsforthestructurevibratinginthat
particular mode undeer no externa

al excitation. TThese eigenvvectors can be
e shown in t he same wayy as node
displacem
ments,buttheeabsolutesca
aleisnormalissedtoamaxim
mumdisplacementof1.0.
Themod
deshapesofeeverysecondm
modeforthe first20mode
esoftheStee
elMonopoleaareshowninFigure53,
onlyeverrysecondmo
odeisshown becauseofthheaxisymmettricgeometryofthestruct ure;everyothermode
shapeistthesame.
53:Everyseco
ondmodeshaapeofthefirsst20modeso
oftheSteelM
Monopole
Figure5
Thenatu
uralfrequencyyforthefirst 20modeswaasplottedfor differentdam
mpingratiosrrangingbetwe
een5and
10%,whichdidntalteertheresponseandisshoowninFigure 54.Thedamp
pingratiosweerechosenasstheyare
typicalru
uleofthumbrratiosforstee
elstructures.
Frequency(Hz)
NaaturalFreq
quency
16
60.00
14
40.00
12
20.00
10
00.00
8
80.00
6
60.00
4
40.00
2
20.00
0.00
dampingrratio0.1
dampingrratio0.05
10
15
20
Modes
Figurre54:Monoppole,NaturalFFrequencyAnalysis
6.2.4.1THEEFFECTSOFANOUTOFBALANCEROTOR:
The 3kW wind turbine machine, for which the supporting structures in this thesis were designed, was
stipulated to not induce any out of balance rotor disturbances, such that the effects of a possible out of
balancerotorcouldessentiallybeignoredforthepurposesoftowerdesign.
However, it was decided to investigate the effect of accidental out of balance rotating forces on the tower,
fromtheperspectiveofapotentialaccidentalloadcase,andtheunderstandingofwhattheeffectsmightbe.
Therefore the investigation done on the effects of the out of balance rotor was for the purposes of
conservativedesignaccountingforaccidentalloadcases,andtoexaminewhatpossiblestressescouldariseas
aresultofsuchrepetitiveoutofbalanceforcesthatcouldresultinfatigueinthebaseconnection,fromthe
stressesoccurringintheweldandflangeplate.
In order to determine the possible effects of an out of balance rotor on the tower, a harmonic response
analysiswasdone.Harmonicresponseisexaminedforthepurposeofassessingthehumanresponsecriteriaas
wellasthefatigueanddisplacementstatesofthestructureifexposedtorepetitiveloading.
For wind turbine towers, the human response to dynamic loading does not need to be considered, as the
machinewouldnotbeoperationalwhenanyhumanascendsthetower.
Theimportanceofexaminingthefatigueofthetowerforthisaccidentalloadcaseisbecausesuchanoutof
balanceforcecouldoccurwithoutbeingvisibletothenakedeye,andsobesubjectingthetowertodynamic
loadingwithouttheeffectsbeingnoted.Ifthiswasthecasefatiguecausingstructuralfailurecouldoccurand
thismustbeavoidedasfarasispossible.
HARMONICRESPONSESOLVEROVERVIEW:
Harmonicanalysisisusedtopredictthesteadystatedynamicresponseofastructuresubjectedtosinusoidally
varyingloads.Astructurewillinitiallyvibrateinanirregularmannerafterasinusoidalloadisapplied.Thisis
referredtoasthetransientstage.Afterthetransientstage,allpointswithinthestructurewillvibrateinthe
same sinusoidal fashion with a frequency identical to the forcing frequency, , but generally with different
amplitudes and different phase angles. This part of the response is known as the steady state stage. The
harmonic response solver calculates the maximum values of the steady state response and the associated
phaseangles(Strand7Software).
Strand7allowstwotypesofharmonicloadingofwhichonlyonetypemaybeincludedinagivensolution.
Theharmonicresponsesolutionreliesonnaturalfrequencies,andmustthereforebegiventhecorrectnatural
frequencyfileinordertocompletethesolutioncorrectly.
ThesamenaturalfrequencyresultfilecanbeusedtoruntheHarmonicSolverasmanytimesasneededwith
differentharmonicsolutionsettingssuchasfrequencyrangeanddampingvalues.
The response of the full structure is calculated by superposition of the modal response considering the
magnitudesandphaseanglesofthemodalresponses.Thisisbasedontheclosedformsolution.
FortheResponsevs.Frequencysolutiontype,anenvelopeofthemaximumvaluesoftheresponseisgivenin
theresultfileateachforcingfrequencystep.Whendampingisincluded,thesemaximumvaluesmaynotoccur
simultaneously and for this reason, the results for the phase angles are given. Output data includes the
maximumvaluesofnodalreaction,displacement,velocity,accelerationandphaseangleaswellasnodeand
elementforces,momentsandstress.
FortheResponsevs.Frequencysolutiontype,thesolutioniscalculatedatequallyspacedstepsovertheuser
specifiedfrequencyrange.Additionalstepsareautomaticallyintroducedatthenaturalfrequencieswithinthe
specifiedrangetocapturepeakresponses.
It is however very important to assess the Response vs. Time solution as well to understand the transient
phase affects that could cause anunwanted response on the tower inthephase through which the turbine
startsrotatinguntilitisatitsfullyoperationalspeed.
HARMONICRESPONSERESULTS:
Theresultsoftheanalysisareabsolutemaximumenvelopevaluesofthesteadystatedynamicresponse.The
maximumvaluesdonotoccuratthesametime.Consequentlytheresultsdontpresentanequilibriumstateof
thestructureandtheydonotcorrespondtothemaximumstressresults.Thestressatoneparticularpointis
just the maximum value that occurred at that particular point during the steady state stage of the dynamic
response.
DETAILSOFTHEOUTOFBALANCEROTORANALYSIS:
Foraconservativesimulation,10%ofthetotalweightoftherotorandnacellecombinedwascalculatedand
appliedtothetowertoaccountfortheoutofbalanceforcesthatcouldpossiblyarisebecauseoftherotornot
beingperfectlyinbalance.The10%ofthetotalweightoftherotorandnacellecombinedwhichwascalculated
and applied was done so as two vectors of equal magnitude, one in the vertical direction, and one in the
horizontaldirectionattheappropriatepointofapplication.Thepointatwhichthevectorswereappliedwas
thesamepointatwhichtheoverturningforceswereapplied,thehubheightofthewindturbinemachine.The
twovectorswerethengivenaloadingparticipationfactorofone,suchthattheywerebothimposed,andthen
were applied with a phase angle of 90 degrees between them so as to simulate the rotating motion of the
rotor.Thevectorswereappliedwithafrequencywhichwasequaltotheequivalentoperatingfrequencyofthe
windturbinerotorinrevolutionsperminute,whichequatestoa5Hzfrequency.Amodaldampingratioof5%
wasspecified.
Theobservedresponseofthetowertotheoutofbalancerotorwasnegligible,intermsofthedegreetowhich
itdisplacedthetower,seeninFigure55.
Fiigure55:WorrstCasedisplaacementoccu
urringforharm
monicresponnse
Because oftherepetittivenatureoffadynamiclooad,thetowerbasestillneedstobeasseessedforfatigue,even
withsuch
hlowdisplaceements,thisissdoneissectiion6.3.
6.2.4.22EFFECTSO
OFVORTEXSHEDDING DUETOWIN
NDACTION
To examine the effeccts of vortex shedding, thee harmonic response solve
er in Strand77 was used, where
w
the
chosenaanalysistypeo
oftheharmon
nicresponsevvs.timeanalyysiswasusedtoexaminethheeffectstha
atshallbe
discussed
dhere.
HARMON
NICDISPLACE MENTRESPONSEVS .TIMEEANALYSIS:
Inorderttoanalysetheedifferentpo
ossibleeffectsthatthewind
dcouldhaveonthetower ifitcoincided
dwiththe
vortex sh
hedding frequency of the
e tower, the wind was ap
pplied as a series
s
of noddal loads of calculated
c
magnitud
deattheapp
propriatenode
es,ontwooppposingsides ofthetower.Thewindlooadwasthen giventhe
wind loads were
same freequency as the vortex shed
dding frequenncy, 11.2Hz, and
a each of the opposingw
w
given
loadingfactorsof1,an
ndaphasean
ngledifferenceebetweenthe
emsetat180 degrees.
Theeffecctsasadisplacementrespo
onseversusti mewereexam
minedtoseethekindsofppossibleexcita
ationsthe
towermiightexperienccehavingthewindloadapppliedatthesameharmonicfrequencyaasthecalculattedvortex
sheddinggfrequencyw
withthefollow
wingresults.
Figure566showsthem
maximumdisp
placementsw
whichoccursduringtheposssibleresonannceeffectoftthevortex
sheddinggfrequencyco
oincidingwith
hthenaturalffrequencyoftthesupportingstructure,t heeffectsofwhichare
negligiblee. However, as
a a result off the repetitiive loading co
ondition the vortex sheddding could ind
duce, it is
importan
nt to still examine the stre
ess range thaat the tower is subjected to under thiss loading condition for
fatigue;tthisIsexamineedinsectionss6.3and6.4.
Figu
ure56:Vortexxsheddinghaarmonicrespo
onse,worstca
asedisplacem
ment
6.3STRRESSSTATESS
HARMON
NICR ESPONSESTRESSANALYSIS:
Thestresssrangewhichastructureissubjectedttointheharm
monicresponsseanalysisiseessentialtoexxamineto
determin
ne the effectss the repetitivve loading m ay have had on the fatigu
ue state of thhe structure. Figure 57
shows th
he base of th
he steel mono
opole tower, and the rangge of stressess it is subjectted to under harmonic
loading.TThestressran
ngeisextreme
elylow.
Figure557:Monopole
eStressrespo
onsetothedyynamicloadin
ngasaresultoftheoutof balancerotoreffects
VORTEX SHEDDINGSTTRESSANALYSSIS:
Asaresu
ultofthewayinwhichthevvortexsheddiingwasanalysedasadynamicload,the stressrangewhichthe
structuree was subjectted to under the repetitivee loading of the
t wind as the
t vortex shhedding frequ
uency was
examined
d as a precaution to avoiid fatigue. Figgure 58 show
ws the stress range whichh this dynamiic loading
subjects the supportin

ng structure to
t and it is off extremely lo
ow stress, to such an exteent that fatigu
ue can be
ignored.
Figure58:Voortexsheddin
ngstressstate
e
6.4FATTIGUEASSEESSMENT:
Followingg the assessm
ment of the stress
s
range w
which the outt of balance rotor would ssubject the supporting
towertoasthiswould
dbeapossiblecontinuoussloading,the
efatigueeffecctsthatsuchlloadingwould
dhaveon
thebaseofthesteelm
monopoletow
werareexami ned.Inaccord
dancewithSA
ANS101621:22005Clause2
26.3.2,the
datashow
wninTable16wasanalyse
ed.Thetowerwasanalysed
dfortheinstanceinwhichaanoutofbala
ancerotor
was runn
ning 24 hours of the day 365 days a year. It was determined that
t
the fatiggue effects of such an
accidentaalloadcasearenegligiblea
andsowillnottbethecause
eofstructuralfailure.
Table
e16:FatigueA
AnalysisofSteelMonopole
eBase
StressRange,fsr,MPa
11000
A
100
B
B1
C
D
10
E
E1
Value
V
1
1E+4
1E+5
1E+6
1E+7
1E+8
1E+9
1
1EE+10
es,N
NumberoofStressCycle
Figure559:Stressvs.NumberofCyyclesforFatiggueAnalysiso
ofoutofbalancerotorinaaccordancewithSANS
1016 21:2005Clau
use26
The reason that no blue dot indica
ating a value on Figure 61 appears is as a result of it not occurriing in any
visible m
margin of the graph on the
e scale which is shown, as it representss such small sstress and succh a large
numbero
ofcycles.
Table17:Fatigu
ueAssessmenntforSteelMonopoleandvortexsheddding
StressRange,fsr,MPa
11000
A
100
B
B1
C
D
10
E
E1
Value
V
1
1E+4
1E+5
1E+6
1E+7
1E+8
1E+9
1
1EE+10
es,N
Figure600:FatigueAnaalysis,StressRangeversussNumberofccyclesforVorttexsheddingaanalysisinacccordance
withSSANS101621
1:2005
6.5MAASSPARTICIPATION:
Inorderttoassessifthemassparticipationoftheesteelmonop
poletowerwassufficient,inntermsofrep
presenting
themasssdistribution throughmodellingthetow
werascloselytohowitwouldbehavinggbeenconstru
ucted,the
massparrticipationforrtranslationalandrotationnalexcitationsswereexamin
ned.Fromtheeoutputofth
heNatural
frequenccysolver,itwaasobservethatafterjust220modesthemassparticip
pationwasat over90%,ind
dicativeof
anexcellentrepresenttationofmasssdistribution throughmodelling.
MODE PARTICIPATION FOR TRANSLATIONAL EXCITATION

Mode
Frequency
Modal Mass
Modal Stiff
PF-X
PF-Y
PF-Z
(Hz)
(Eng)
(Eng)
(%)
(%)
(%)
1
1.6794E+00
7.8453E-01
8.7349E+01
0.554
59.116
0.000
2
1.6830E+00
7.8348E-01
8.7607E+01
59.069
0.553
0.000
3
1.0184E+01
1.0255E+00
4.1986E+03
0.148
18.628
0.005
4
1.0202E+01
1.0133E+00
4.1637E+03
18.633
0.148
0.000
5
2.7414E+01
1.2355E+00
3.6656E+04
5.816
0.018
0.000
6
2.7451E+01
1.2853E+00
3.8237E+04
0.018
5.824
0.019
7
4.3932E+01
5.0112E-01
3.8183E+04
0.112
0.000
0.000
8
5.9145E+01
1.0872E+00
1.5014E+05
0.000
3.228
0.395
9
5.9677E+01
9.6044E-01
1.3503E+05
3.252
0.000
0.000
10
6.8618E+01
1.4149E+00
2.6301E+05
0.000
0.028
77.324
11
8.8989E+01
1.1548E+00
3.6102E+05
0.000
1.995
0.202
12
8.9360E+01
1.0495E+00
3.3085E+05
1.983
0.000
0.000
13
1.2614E+02
6.9072E-01
4.3388E+05
0.774
0.000
0.000
14
1.2861E+02
1.0833E+00
7.0734E+05
0.000
1.013
0.043
15
1.3188E+02
7.6377E-01
5.2442E+05
0.002
0.012
0.000
16
1.3188E+02
4.5242E-01
3.1065E+05
0.000
0.001
0.000
17
1.3320E+02
9.8624E-01
6.9078E+05
0.002
0.009
0.000
18
1.3320E+02
6.1918E-01
4.3370E+05
0.000
0.001
0.000
19
1.3534E+02
3.4988E-01
2.5300E+05
0.000
0.000
0.000
20
1.3534E+02
3.5463E-01
2.5645E+05
0.000
0.001
0.000
---------------------------------------------------------------------------TOTAL MASS PARTICIPATION FACTORS
90.362
90.577
77.988
MODE PARTICIPATION FOR ROTATIONAL EXCITATION

Mode
Frequency
Modal Mass
Modal Stiff
PF-RX
PF-RY
PF-RZ
(Hz)
(Eng)
(Eng)
(%)
(%)
(%)
1
1.6794E+00
7.8453E-01
8.7349E+01
78.274
0.733
0.000
2
1.6830E+00
7.8348E-01
8.7607E+01
0.733
78.243
0.000
3
1.0184E+01
1.0255E+00
4.1986E+03
10.035
0.080
0.000
4
1.0202E+01
1.0133E+00
4.1637E+03
0.080
10.102
0.002
5
2.7414E+01
1.2355E+00
3.6656E+04
0.001
0.412
0.275
6
2.7451E+01
1.2853E+00
3.8237E+04
0.354
0.001
0.001
7
4.3932E+01
5.0112E-01
3.8183E+04
0.000
0.715
67.860
8
5.9145E+01
1.0872E+00
1.5014E+05
1.797
0.000
0.000
9
5.9677E+01
9.6044E-01
1.3503E+05
0.000
1.496
2.429
10
6.8618E+01
1.4149E+00
2.6301E+05
0.004
0.000
0.000
11
8.8989E+01
1.1548E+00
3.6102E+05
0.766
0.000
0.000
12
8.9360E+01
1.0495E+00
3.3085E+05
0.000
0.750
1.433
13
1.2614E+02
6.9072E-01
4.3388E+05
0.000
1.256
5.151
14
1.2861E+02
1.0833E+00
7.0734E+05
1.068
0.000
0.000
15
1.3188E+02
7.6377E-01
5.2442E+05
0.015
0.003
0.000
16
1.3188E+02
4.5242E-01
3.1065E+05
0.002
0.000
0.000
17
1.3320E+02
9.8624E-01
6.9078E+05
0.012
0.002
0.000
18
1.3320E+02
6.1918E-01
4.3370E+05
0.002
0.000
0.000
19
1.3534E+02
3.4988E-01
2.5300E+05
0.000
0.000
0.000
20
1.3534E+02
3.5463E-01
2.5645E+05
0.001
0.000
0.000
93.145
93.792
77.153
RESONANCEEFFECTS:
In having determined the natural frequencies of the tower, as well as what proportion of mass that
participateswitheachmode,thepossibleeffectsofresonancecanbeassessed.
Becausethemachinesoperatingfrequencyis5Hz,modes1withnaturalfrequency(1.67Hz)and2withnatural
frequency(1.68Hz)arethemostcrucialtoexamine,astheyshallbeexperiencedinthetransientphasefrom
therotorbeingstationaryat0Hzuntilitisrunningatafullyoperationalspeedof5Hz.Examiningarangeof
frequencies that encompass the operating frequency, such as the following: 0.6(operating frequency) to
1.6(operatingfrequency),thenaturalfrequenciescanbeevaluatedforresonantpeaksinaccordancewiththe
proportionofmodalmassparticipatingwithinthatrange.Figure62isillustrativeofthisforthesteelmonopole
tower,howeverthereisnotasignificantresonantpeakwithinthepertinentoperatingfrequencyrange,andso
thetuningofthesteelmonopoleisadequate.
100
90
80
Mass
Participation
(%)
70
HorizontalMass
Participation
VerticalMass
paricipation
LT
60
50
40
30
20
10
0
50.00
100.00
Frequency(Hz)
Figure61:Resonanceevaluationofsteelmonopole
150.00
6.6STEEELLATTICEETOWER:FINITEELEM
MENTANALYYSES
TheSteeellatticesupportingtower forthe3kWw
windturbine wasdesigned
dandthenannalysedusing thefinite
elementpackageStran
nd7.
TheanalyysesthatwerreperformedontheSteelLLatticeTowerrwereoutline
edanddescribbedintheintroduction
tothisch
hapterandareerepeatedhe
ere:
1.
2.
3.
4.
LinearStaticA
Analysis
LinearBucklin
ngAnalysis
NonlinearAnalysis
DynamicAnalyses
i)
Natu
uralFrequencyyAnalysis
ii)
Effecctsofanouto
ofbalancerottor
iii)
Effecctsofvortexssheddingdue towindaction
iv)
Harm
monicFrequencyAnalysis
v)
StresssstatesandFatigueassesssment
vi)
Mod
dalMassParticipationEvaluuation
The toweer was design

ned with a topportion wh ich had paralllel sides whicch fitted into a bottom secction that
taperedggraduallydow
wntolevelgro
ound;seeFiguure62forcon
nnectiondetailsbetweent hetwotowerrsections.
The strucctural membeers chosen fo
or the design varied in sizze from 50x50
0x5mm equa l angle leg se
ections to
70x70x6m
mmsectionsttothelargesttsectionswhi chwere100xx100x8mmsectionsatthe bottom.Thestructural
memberssconnectionssweremodelledaspinnedd.
Figuree62:ConnecttionDetailsbe
etweentopp arallelportionofthetowe
erandthetapperedportionofthe
tower
Theeffecctsofthewind
dturbinemacchineactingoonthetopofthetowerwerresimulatedbbymeansofaprismatic
assemblyyofrigidlinkssthatmetata
acentralposittionandthen
nhadanexten
ndingrigidlin kthatcouldccreatethe
appropriatepointatw
whichtheecce
entriccentre ofmassoftherotorandnacellecouldbbeappliedasalumped
mass. Th
he model of the
t top of the tower is shhown in Figurre 63. The fo
orce that shaall be described as the
overturn force in thee following se
ections accou nted for the action of the
e wind on thhe blades of the Wind
TurbineM
Machine. Theecalculation details
d
of thee overturn forrce weredesccribed in Chappter 3 in secttion 3.3.2.
Theoverrturnforcesim
mulatedtheactionofthew
windontheblladesofthero
otorasafuncctionofhubh
heightand
windspeeedandwasappliedatthepeakofthecoonicalformationofrigidlin
nksinStrand7,,seeninFigure63.
werTurbineC
ConnectionSim
mulation
Figure63:LatticeTow
werthathasa squareplanssection,twow
windloadingdirectionsweereconsidered
d.Forthe
Forasteeellatticetow
twodiffeerentwindloaadcases,twoseparatemoddelswerecreatedandeach
hoftheloadccombinationssspecified
in accord
dance with SA
ANS 101601:2011 were a pplied to the
e models. Figu
ure 64 shows the two win
nd loading
casesinp
planview.
F
Figure64:Win
ndLoadingCaasesforsquarreplanSteelLatticeTowerrs
Table18shows,foreaachoftheabo
ovewindload ingcases,the
eloadcombina
ationsthatweereappliedto
othesteel
latticeto
ower.
Table
e18:LoadCa seCombinatiions,LatticeT
Tower
LoadCases
ULS1
ULS2
ULS3
SLS1
SSLS2
SW
1.2
1.2
0.9
EquipmentD
DL
1.2
1.2
0.9
Overturn(ULS)
1.5
1.5
Overturn(SLLS)
0.6
WindNormaal
1.5
1.5
0.6
WINDLO
OAD MODELLLING:
The desiggn implications of the wayy in which th e steel lattice
e tower was loaded for thhe wind actions on the
tower waas investigateedto determiine theexten t of effect that different lo
oadings may have on the results of
deflectionandstress.
Twoextrremeloading optionswereexamined,1 .Thefrontan
ndbackmemberswereloaadedwithadistributed
windload
dand2.Onlythefrontmem
mberswerelooadedwithadistributedwindload.
The first of the two options
o
equates to twice the total app
plied load of the second ooption. The re
esults are
tabulated
dinTable19,andgraphedinFigure65aandFigure66..
Table19
9:Windloadin
ngontheSteeelLatticeTow
wer,displacem
mentandstreessresults
Option
1(Backaandfront)
2(Frontonly)
difference
D
Displacement(
(DX)SLS2(mm
m) MaxAxiaalStressULS3(MPa) TotaalAppliedLoad(X),(N)
30247.9
9.99
89.68
15215.9
8.73
84.49
15032
1.26
5.19
MaximumAxialStressVs.WindLoad
MaximumAxialStress(Mpa)
90
89
88
87
86
85
84
15000
20000
25000
30000
TotalWindLoad(N)
Figure65:StresscomparisonfortwodifferentwindloadsonSteelLatticeTower
ServiceabilityDisplacement(mm)
ServiceabilityDisplacementVs.WindLoad
10.2
10
9.8
9.6
9.4
9.2
9
8.8
8.6
15000 17000 19000 21000 23000 25000 27000 29000 31000
TotalAppliedWindLoad(N)
Figure66:DisplacementcomparisonfortwodifferentwindloadsonSteelLatticeTower
FromtheresultsinTable19itcanbededucedthatthewindloadmodellingparticularsonthemembersofthe
SteelLatticeTowerwouldnotinfluencetheendresultssignificantly.Forconservativedesign,option1ofthe
twooptionswaschoseninthefinalmodel.
6.6.1LINEARSTATICANALYSIS
Thefirstanalysisofthesteellatticetowerwasthelinearstaticanalysis.
InFigure68,thedisplacementresultfortheloadcombinationSLS2asdefinedinTable18isshownforwind
loadcase1.Themagnitudeofdisplacementdifferenceforthetwoloadcases1and2whenexaminediswithin
0.01mmfromoneanother.Figure67belowshowsthedirectionsinwhichthetwodifferentwindloadcases
weremodelled.
Figu
ure67:Differe
entwindloaddcasesforSquareplanSte
eelLatticeTow
wers
Figure68:SSteelLatticeT
TowerDisplaccement(XY)SStaticAnalysiss,WindLoad case1,SLS2
BASERESPONSE:
Thebaseenodalreactio
onforcesandmomentsweerealsoobtain
nedfromthelinearstatica nalysis,andu
utilisedfor
thedesiggnofthesteellbaseplatesa
andfoundatioonsrequiredfo
orthistower.
FortheSSteellatticeto
owerthebase
ereactionforccesareshown
ninFigure69
9andFigure770forthetwo
odifferent
windload
dcases.Thefforcesaredep
pictedforthe ULS2,seeTable18,(WorstCase)scena riosforeachwindload
caseresp
pectively,sothatthehighestloadscanbbeillustrated. Theloadsare
eshowninvecctorformactiingonthe
fourbaseepointsoftheelatticetowerinFigure69 andFigure70
0.
F
Figure69:Bas
seResponse,SSteelLatticeT
Tower,forWiindloadcase 1
F
Figure70:Bas
seResponse, Steellatticettower,forWin
ndLoadcase 2
LINEARSTATICSTRESSANALYSIS:
InordertoassessthestressstateoftheSteelLatticeTowerforthelinearstaticloading,thestressdistributions
fortheloadcombinationULS2asdescribedinTable18wereexamined.IntermsoftheAxialStressstateof
thebeamelements,aswellastheTotalFibreStressstateofthebeamelementsandtheTorsionalStressstate
noneofthestressstatesexceedtheelasticmateriallimit(355MPa)oftheS355JRgradesteelthattheSteel
LatticeTowerwasdesignedto.
6.6.2LINEARBUCKLINGANALYSIS:
ThelinearbucklingfactorsareplottedinFigure71forwindloadcase1oftheSteellatticetowerandFigure72
for wind load case 2. The buckling factors are shown for all of the different linear load combinations which
weredefinedinTable18.InTable20andTable21thefirst6bucklingfactorsforeachofthewindloadcases
areshown,thenegativevaluesinTable20andTable21arethebucklingfactorswhichdidntconverge.
LinearBucklingLoadFactors:SteelLatticeWindLoad
Case1(first20modes)
1000.00
LoadBucklingFactors
800.00
SW
600.00
ULS2
400.00
ULS1
200.00
ULS3
0.00
200.00
SLS1
1
11
400.00
16
SLS2
Modes
Figure71:LinearBucklingFactors:SteelLatticeTower,WindLoadCase1
Table20:LinearBucklingFactorsSteelLatticeTower,WindLoadCase1
Mode
SW
1
2
3
4
5
6
ULS1
241.1
242.5
243.3
245.1
421.1
447.8
ULS2
152.4
153.6
154.7
156.0
274.5
280.6
ULS3
21.4
23.3
24.1
24.1
31.7
31.8
SLS1
21.6
23.1
24.9
24.9
30.5
30.6
SLS2
228.7
230.4
232.0
234.0
411.8
420.8
50.9
53.2
53.3
61.0
92.0
93.2
LinearBucklingLoadFactors:SteelLatticeWindLoadCase2
(first20modes)
1000.00
LoadBucklingFactors
800.00
600.00
SW
ULS1
400.00
ULS2
200.00
ULS3
SLS1
0.00
1
11
SLS2
16
200.00
400.00
modes
Figure72:LinearBucklingFactors,SteelLatticeTower,WindLoadcase2
Table21:LinearBucklingFactors,SteelLatticeTower,WindLoadCase2
Mode
SW
1
2
3
4
5
6
ULS1
ULS2
241.1
242.6
243.4
245.1
421.2
447.9
152.2
154.3
154.5
155.9
274.5
279.9
ULS3
20.4
20.8
28.0
28.8
31.8
32.7
SLS1
19.9
21.4
28.4
29.2
31.4
32.2
SLS2
228.3
231.4
231.7
233.8
411.8
419.9
51.8
65.3
83.2
84.9
88.1
100.8
6.6.3NONLINEARSTATICANALYSIS:
NonlinearDisplacement,InitialNonlinear
Analysis,LatticeTower
DisplacementDX(mm)
0.35
0.3
0.25
0.2
0.15
0.1
WindLoadCase1
0.05
0
1
11
16
LoadIncrement
Figure73:NonlinearDisplacement,InitialNonlinearAnalysis,LatticeTower
DisplacmentDX(mm)
NonlinearDisplacement,InitialNonlinear
Analysis,LatticeTower
30
25
20
15
10
WindLoadCase1
5
0
1
11
16
LoadIncrement
Figure74:NonlinearDisplacement,InitialToweranalysis,LatticeTower
Theintensionherewastoplacethedisplacementvs.loadingincrementforthetwowindloadingcasesonthe
samegraphs;however,thesteellatticetowerforwindloadcasetwowasunabletobeanalysedasaresultof
Softwarelimitations.
6.6.4DYNAMICANALYSES:
The first analysis which was done to precede and enable the other analyses in this section was a natural
frequencyanalysis.
NATURALFREQUENCYANALYSIS:
Frequency(Hz)
NaturalFrequencyofSteelLatticeTower
50.00
45.00
40.00
35.00
30.00
25.00
20.00
15.00
10.00
5.00
0.00
1
11
16
Mode
Figure75:NaturalfrequencyversusmodeforSteelLattice,WindLoadCase1
NaturalFrequencycomparisonofthetwo
supportingstructures
140.00
Frequency(Hz)
120.00
naturalfrequency
lattice
100.00
80.00
naturalfrequency
monopole
60.00
40.00
20.00
0.00
1
11
16
Mode
Figure76:NaturalfrequencycomparisonforMonopoleandSteelLatticeTowers
Figure77showsthefirst10modesshapesforthenaturalfrequencyanalysisfortheSteelLatticeTower.
Figure7
77:First10moodeshapesfo
orSteelLatticceTower
6.6.4.11DYNAMICCEFFECTSO
OFANOUTO
OFBALANCEEROTOR:
The 3kW
W wind turbin
ne machine, for which thhe supportingg structures in this thesiss were desiggned, was
stipulatedto not indu
uceanyout ofbalance
o
rottor disturbancces,that the effects of a ppossible out ofbalance
o
rotorcou
uldessentiallyybeignoredfo
orthepurpos esofTowerd
design.
Howeverr,itwasdecid
dedtoinvestiigatetheeffeectsofaccidentaloutofba
alancerotatinngforcesontthetower
top,from
mtheperspecctiveofapote
entialaccidenttalloadcase,andtheunde
erstandingof whattheeffe
ectsmight
be. Therrefore the invvestigation do
one on the eeffects of the
e out of balance rotor waas for the purposes of
conservaativedesignacccountingforaccidentalloaadcases,notloadingsthatwereexpecteedtooccur.
Foracon
nservativesim
mulationtheoutofbalance forcesthatco
ouldpossiblyarisebecauseeoftherotornotbeing
perfectlyyinbalance,1
10%ofthetottalweightofttherotorand
dnacellecomb
binedwascallculatedanda
appliedas
two vecttors of equal magnitude, one in the vvertical directtion, and one
e in the horizzontal directio
on at the
appropriatepointofaapplication.Th
hepointatwhhichthevecto
orswereappliiedwasthesaamepointatw
whichthe
overturning forces weere applied, the
t hub heighht of the wind turbine machine. The ttwo vectors were
w
then
givenalo
oadingparticipationfactorrofone,suchhthattheywe
erebothimpo
osed,andtheenwereappliedwitha
phasean
ngleof90deggreesbetween
nthemsoasttosimulateth
herotatingmo
otionoftherrotor.Thevecctorswere
appliedw
withafrequencywhichwa
asequaltoth eequivalent operatingfrequencyoftheewindturbinerotorin
revolutio
onsperminutee,whichequa
atestoa5Hzffrequency.Am
modaldampin
ngratioof5%
%wasspecified
d.
Figure788showsthem
maximumdisp
placementsw
whichoccurdu
uringtheharm
monicresponsseversustimeanalysis
asaresultoftheouto
ofbalancerotoreffects,anddascanbese
ee,thedisplaccementsarennegligible.
Figgure78:MaximumDisplaccementeffecttsfromdynam
micanalysisofoutofbalanncerotoreffe
ects
6.6.5MASSPARTICIPATION:
InordertoassessifthemassparticipationoftheSteelLatticeTowerwassufficient,intermsofrepresenting
themassdistributionthroughmodellingthetowerascloselytohowitwouldbehavingbeenconstructed,the
massparticipationfortranslationalandrotationalexcitationswereexamined.FromtheoutputoftheNatural
frequencysolver,itwasobservethatafterjust20modesthemassparticipationwasbetween70%and90%,
indicativeofanexcellentrepresentationofmassdistributionthroughmodelling.
MODE PARTICIPATION FOR TRANSLATIONAL EXCITATION

Mode
Frequency
Modal Mass
Modal Stiff
PF-X
PF-Y
PF-Z
(Hz)
(Eng)
(Eng)
(%)
(%)
(%)
1
7.7548E+00
2.0912E-01
4.9646E+02
3.304
15.304
0.005
2
7.8928E+00
2.1544E-01
5.2985E+02
15.709
3.355
0.024
3
1.1388E+01
2.5417E-01
1.3014E+03
0.000
0.000
0.000
4
1.5882E+01
1.2459E-01
1.2408E+03
0.044
8.541
0.000
5
1.5883E+01
1.2453E-01
1.2402E+03
8.428
0.043
0.000
6
1.6527E+01
2.3053E-01
2.4857E+03
0.000
0.000
0.000
7
1.9010E+01
6.0814E-01
8.6759E+03
1.060
34.465
0.003
8
1.9159E+01
6.5590E-01
9.5051E+03
34.816
1.198
0.082
9
2.0017E+01
2.4401E-01
3.8598E+03
0.000
0.000
0.140
10
2.2359E+01
1.3305E-01
2.6260E+03
0.083
0.639
0.001
11
2.2380E+01
1.3408E-01
2.6514E+03
0.468
0.061
0.010
12
2.3772E+01
1.9633E-01
4.3799E+03
0.000
0.000
0.000
13
2.7884E+01
2.1761E-01
6.6795E+03
0.000
0.000
0.137
14
3.1875E+01
1.4212E-01
5.7005E+03
0.833
3.614
0.013
15
3.1990E+01
1.3954E-01
5.6378E+03
3.180
0.735
0.062
16
3.4517E+01
1.3979E+00
6.5754E+04
0.000
0.000
0.000
17
3.6158E+01
1.0903E-01
5.6275E+03
0.000
0.000
0.000
18
3.8583E+01
7.4552E-02
4.3813E+03
0.496
2.040
0.088
19
3.9078E+01
6.0927E-02
3.6732E+03
1.637
0.563
0.508
20
3.9579E+01
5.9716E-02
3.6931E+03
0.003
0.001
0.361
70.061
70.560
1.434
MODE PARTICIPATION FOR ROTATIONAL EXCITATION

Mode
Frequency
Modal Mass
Modal Stiff
PF-RX
PF-RY
PF-RZ
(Hz)
(Eng)
(Eng)
(%)
(%)
(%)
1
7.7548E+00
2.0912E-01
4.9646E+02
41.837
8.801
0.671
2
7.8928E+00
2.1544E-01
5.2985E+02
9.400
41.542
0.001
3
1.1388E+01
2.5417E-01
1.3014E+03
0.000
0.000
0.000
4
1.5882E+01
1.2459E-01
1.2408E+03
1.217
0.006
0.073
5
1.5883E+01
1.2453E-01
1.2402E+03
0.006
1.184
0.061
6
1.6527E+01
2.3053E-01
2.4857E+03
0.000
0.000
0.000
7
1.9010E+01
6.0814E-01
8.6759E+03
35.962
1.122
0.059
8
1.9159E+01
6.5590E-01
9.5051E+03
1.281
36.459
0.363
9
2.0017E+01
2.4401E-01
3.8598E+03
0.000
0.000
0.000
10
2.2359E+01
1.3305E-01
2.6260E+03
0.106
0.015
0.011
11
2.2380E+01
1.3408E-01
2.6514E+03
0.026
0.196
0.012
12
2.3772E+01
1.9633E-01
4.3799E+03
0.000
0.000
0.000
13
2.7884E+01
2.1761E-01
6.6795E+03
0.000
0.000
0.000
14
3.1875E+01
1.4212E-01
5.7005E+03
0.803
0.182
0.057
15
3.1990E+01
1.3954E-01
5.6378E+03
0.141
0.628
0.101
16
3.4517E+01
1.3979E+00
6.5754E+04
0.000
0.000
0.000
17
3.6158E+01
1.0903E-01
5.6275E+03
0.000
0.000
0.000
18
3.8583E+01
7.4552E-02
4.3813E+03
0.800
0.190
0.340
19
3.9078E+01
6.0927E-02
3.6732E+03
0.204
0.599
0.274
20
3.9579E+01
5.9716E-02
3.6931E+03
0.002
0.000
0.000
91.783
90.926
2.026
RESONANCEEFFECTS:
In having determined the natural frequencies of the tower, as well as what proportion of mass that
participateswitheachmode,thepossibleeffectsofresonancecanbeassessed.
Because the machines operating frequency is 5Hz, and the first natural frequency is 7.75, there will be no
problemsinthetransientoranyotherphaseofthemachinesoperationwithrespecttoresonance.Thesteel
latticetowerisperfectlytunedinthisrespect.
70
60
50
HorizontalMass
Participation
VerticalMass
paricipation
LT
Mass
Participation
(%)
40
30
20
10
0
10.00
20.00
30.00
40.00
Frequency(Hz)
Figure79:Resonanceeffectsofsteellatticetower
50.00
6.7STRRESSSTATE
HARMON
NICRESPONSEESTRESSANA
ALYSIS.
Thestressseswhichweerethemostsignificantas aresultofth
heharmonicresponsetothheoutofbala
ancerotor
effectsoccurredintheeconnectionofthetoppaarallelportionofthesteelllatticetower tothebottom
mtapered
section.TTheStressrangetowhichthissection ofthetower wassubjecte
edisshowninnFigure80.T
Thestress
rangeiseextremelylow
w.
nicresponse
Figure80:SStressstatefrromoutofbalanceharmon
6.8FATTIGUEASSEESSMENT
Followinggthestress state
s
analysis in section6.77, thefatigue
e state of the elements in the steel latttice tower
wasassessedintermssoftheeffectarepetitivelooadingofano
outofbalance
erotorwouldhaveonthestructure.
AscanbeeseeninTable22thetowerwasanalyssedfortheinsstanceinwhicchtheturbineemachinewa
asrunning
24hourssaday365daysayearwiithanoutof balancerotor,andtheefffectsoffatiguueinthemem
mbersare
negligiblee.
Table22:FatigueeAnalysisofssteellatticeto
ower
StressRange,fsr,MPa
11000
A
100
B
B1
C
D
10
E
E1
Value
V
1
1E+4
1E+5
1E+6
1E+7
1E+8
1E+9
1
1EE+10
es,N
Figure881:StressRan
ngeversusNumberofcycleesforfatigueanalysisinacccordancewitthSANS101621:2005
CHAPTER7:FEASIBILITY
7.1INTRODUCTIONTOFEASIBILITY
Promptedbyenquiriesfromindustry,itwasdecidedtoinvestigatethefeasibilityofdesigningasteellattice
alternative to the most commonly used steel monopole tower as a supporting tower for a small scale wind
turbine.
Thepurposeofthischapteristodevelopanunderstandingandaninsightintothedesignaspectsthatwere
takenintoaccountforthecostanalysisofthetwodifferentsteelsupportingtowers.
Theitemsdealtwithinevaluatingthecostsofthetwosteeltowersdesignedinthisthesiswerethefollowing:
1.
2.
3.
4.
5.
Thetowerspecifications
i)
SteelMonopoleTower
ii)
SteelLatticeTower
Thestructuralrequirementsofthetowers
Theaestheticrequirementsofthetowers
Theconstructabilityofthetowers:
i)
TransportationConsiderations
ii)
FabricationConsiderations
iii)
ConstructionConsiderations
Overallcostcomparison
7.1.1TOWERSPECIFICATIONS
7.1.1.1STEELMONOPOLETOWER
The tower specifications for the steel monopole tower designed for this thesis were prescribed by the
StellenboschWindEnergyTechnologiesPtyLtd(SWET)Company.SWETisaspinoffcompanystartedbythe
ElectricalEngineeringdepartmentatStellenboschUniversity.TheUniversityhasa50%shareofthecompany.
SWET has produced a new small wind turbine technology which it wants to thoroughly test and eventually
commercialise. A 3kW prototype machine was built in November of 2011. SWET needed a turbine tower
designwhichitcouldusewiththewindturbineprototype.TohelpSWETwiththisdesigntheCivilEngineering
Departmentwasapproached.
ThedesignparameterssetbySWETwerethatthetowerneededtobe16mhigh,withsectionsnolongerthat
6m, and it was to be as cost effective as possible. From the requirement of cost efficacy, it was decided to
examine the possibility of designing a monopole that was constructedfrom existing circular hollow sections
(CHS)toreducefabricationcosts.
Toestablishthemostrealisticcostbenchmarkforcomparativepurposesitwasdecidedtodesignthelightest
possible option within the bounds of regularly manufactured and fabricated circular hollow sections. The
circularhollowsectionsneededringflangeconnectionsweldedontoeachendofthetubularsections,which
alsohadwebstiffenersweldedonatspecifiedpoints,ofspecifiedsize.Itmustbeemphasisedthattheonly
objectiveherewasthecheapestsolutionforaneffectivemonopoletower.
7.1.1.2STEELLATTICETOWER
109|C h a p t e r 7 : F e a s i b i l i t y
To design a cost effective alternative to the Monopole Tower, with comparable strength and reliability, the
SteelLatticeTowerdesignedinthisthesishadstilltobemorecosteffectivethanthemonopoledesignedout
ofexistingCHS.
IntermsofdesignboundariesgivenbySWET,theonlyboundariesspecifiedforthedesignofthesteellattice
towerweretheheightof16m,andthattowersectionsbenolongerthat6m.
7.2THESTRUCTURALREQUIREMENTSOFTHETOWERS
Boththesteellatticetowerandthesteelmonopoletowerwererequiredtohaveadequatestrengthinorder
towithstandtheloadingsuniquetothe3kWwindturbinetobemountedontopofeitherone.Thestructural
requirementsanddesignweredescribedindetailinChapter3.
7.3THEAESTHETICREQUIREMENTSOFTHETOWERS
Themostdominatingaspectinpresentdaychoiceofsupportingtowerforwindturbinesistheaestheticsof
the tower. An overwhelming opinion is that the steel monopole is more aesthetically pleasing than a steel
latticetower.
TheSteelLatticeTowerdesign,whilehavingtobestructurallysound,sufferedthecomplicatingrequirement
ofhavingtobeaestheticallypleasingwhencomparedtoSteelMonopoletowerdesign.Themannerinwhich
thatwasapproachedincludedconsideringthebalanceofthenumberofmembersusedtotheavailableopen
spacebetweenthem,aswellassymmetryandproportion,andanoverallshapethatwassufficientlyfamiliar
andpleasing.
7.4THECONSTRUCTABILITYOFTHETOWERS
Whiletheaestheticsandpracticalitieswereconsidered,itwasimportanttobearinmindfabricationmethods
thatwouldavoidcomplexdetails.Specialconsiderationwasgiventothesizesofthefabricatedelementssince
thiswouldhaveacriticalbearingontheeaseoftheirtransportationtosite.
7.4.1TRANSPORTATION:
The Steel Monopole tower and the Steel Lattice tower were both designed so that they didnt have any
componentslongerthan6m.Dissembledpartscouldthenbetransportedonregularfreightvehicles,thereby
eliminatingtheneedforcostlyabnormalfreightvehicles.
7.4.2FABRICATIONANDCONSTRUCTION:
Obtaining sufficient information to make valid comparisons of costs presented severe challenges.
Manufacturersandcontractorswerereluctanttodiscloseproprietaryinformationforresearchpurposesand
thosewhodidprovidedinformationonconditionofstrictanonymity.Thismeantthatconsiderablecarehadto
be taken to compare like with like and in order to reach the best outcome the comparisons were done in
stages.
Thefirstsetsofcostsobtainedwerebasedonfabricationalone.Thesecondsetsofcostsobtainedwerefor
constructionofthetowersandincludedfabricationcosts.Thiswasdoneasanadditionaltestofthevalidityof
thefabricationinformationgiveninthefirstsetofcosts.Ofgreatsignificanceinthesecondsetsofcomparison
weretheconsiderablyhigherfoundationcostsofthemonopoledesign.
Thecostaspectsoffabricationofthetwotowersshallbediscussedfirstandthereaftertheconstruction.
7.4.2.1FABRICATION:
Toassessthefabricationcostsofthetwodifferenttowers,thedesigninformationfortherequiredassemblyof
thetwodifferenttowerswassenttothreedifferentsteelfabricatorsinSouthAfricaforquotations.Ofthese
tworesponded.
Theresponsesdifferedsignificantly.
FIRSTFABRICATORCOSTRESPONSE:
The first fabricator used themasses of thetowers as itsmost significant variable. The reason wasthatthey
haveaflatfeeofZAR30/kgforthefabricationoftheSteelLatticeTowers,includingthesteelsupply.Limited
considerationisgiventoexactdesignsincelatticedesignvariationshavesmalleffectonthefinalpriceandare
simply accounted for in the price. Their manufacturing price was ZAR48/kg for the Steel Monopole Tower
fabrication.
Table23illustratesthecomparativedifference.
Table23:CostcomparisonofFabrication,Fabricator1
TowerStyle
Cost
(ZAR/kg)
Mass
(kg)
TotalCost
(ZAR)
SteelMonopole
48
2800
134400
SteelLattice
30
2500
75000
ThedifferenceincostbetweenthetwotowersisR59000whichisconsiderableespeciallywhenexpressedasa
percentage.TheMonopolewas79.20%moreexpensive.
SECONDFABRICATIONCOSTRESPONSE:
ThesecondfabricatordealtwiththeelementsoffabricationingreaterdetailasshowninTable24.
Table24:Fabricationcosts,Monopole,Fabricator2
DESCRIPTION
QTY
UNIT
SUPPLY AND
FABRICATE
Amount
Steel Work
Cicular Section - 5m (2 Off)
1.6
R 77 096.13
Cicular Section - 6m
1.0
R 46 257.68
32.0
No
R 55.20
Mid type flange
3.0
No
R 12 661.50
Base type flange
1.0
No
R 3 627.10
Corner Stiffeners (8 Off)
0.013
R 543.79
Side stiffeners (6 Off)
0.003
R 125.66
Rectangular Section
0.03
R 1 598.88
Bolts - M20
TOTAL EXCL. VAT

TheSteellatticeTowersquotationisshowninTable25.
.
R 141 965.92
Table25:FabricationCosts,LatticeTower,Fabricator2
DESCRIPTION
SUPPLY AND
FABRICATE
Amount
QTY
UNIT
Angle 50x50x5 (33.2m)
0.13
5 060.27
Angle 100x100x8 (157.50m)
1.92
74 740.91
Angle 70x70x6 (39.94m)
0.30
11 678.30
Gusset plates (60 plates)
0.34
18 968.83
Base Plate
0.02
1 335.60
111 783.91
Steel Work - Parallel tower

Bracing
TOTAL
ThedifferenceincostbetweenthetwotowersfromthesecondfabricatorsquotationwasR30182.Whilethis
difference is less than that for the first quotation it is still significant, especially when expressed as a
percentage.InthiscasetheMonopolewas27.00%moreexpensive.
7.4.2.2CONSTRUCTION:
Threeleadingcompaniesprovidedconstructioncosts.Theywereaskedtoshowthecostofsteelfabrication
separately to enable validation of the information contained in the other quotations. As mentioned above
these companies all wished to remain anonymous to protect their proprietary information, and are for the
purposesofthenextsectionshownasCompanyA,CompanyBandCompanyCrespectively.
Inadditiontothesteelfabricationcostsitwasrequestedthatthequotationsfortheconstructioncostsshould
includetheconstructionofthefoundations,fromexcavationtoquantitiesofconcreterequiredtobuildthe
foundations.
Table26indicateswhichcostswereconsideredinthedetailedpricingofthetwodifferenttowersthatwere
designedforthisthesis.
Table26:TableofItemdescriptionsforconstruction
Itemdescription
EarthWorks
1.Excavationofthefoundations,includinglevelingandstockpilingofselectedmaterialforreuseinfilling
2.CompactionofexistingsoiltoStandardProctorOptimumDensity
3.Bitumouspaintingandsealingonconcretesurfacesincontactwithearth
ConcreteWorks
4.ReinforcedConcrete
5.Steeltrowelfinishforslabs
6.Highyieldsteelbarreinforcement
7.Castinanchorboltsuptp20mmdiaincludingtemplates
8.Castinanchorboltsbiggerthan20mmdiaincludingtemplates
SteelWork
9.Preparationofshopdetaildrawing,fabrication,painting,deliveryanderectionoflightsteelstructuralsteelwork,
completewithallthenecessarycleats,shopbolts,brackets,gussets,packsbaseplatesetc.
Table27showsthetotalpriceforeachitemconsideredintheMonopoledesignwithinthecategoriesofEarth
Works,ConcreteWorksandSteelWork.
Table27:TableofComparativecostsforeachitemoftheMonopoleTower'sconstruction
Monopole
Itemdescription Unit
EarthWorks
1
2
m
3
m
27.2
71
1939
111
3030
78
18
108
12
207
13
242
43.8
60
2621
183
8004
80
3503
m
2
m
t
nr
nr
9.4
4980
46811
2417
22720
1764
16580
1
0
0
16
27
14253
200
200
27
21
3969
21
11721
97
112
21
3200
21
13070
215
248
1791
2.8
45254
126711
33594
94064
42777
119776
m
3
ConcreteWorks
4
5
6
7
8
SteelWork
9
CompanyA
CompanyB
CompanyC
Quantity UnitPrice(ZAR) Total(ZAR) UnitPrice(ZAR) Total(ZAR) UnitPrice(ZAR) Total(ZAR)
2132
Table 28 shows the total price foreach item consideredin the design for theSteel Lattice tower within the
categoriesofEarthWorks,ConcreteWorksandSteelWork.
Table28:TableofcomparativecostsforitemsassociatedwiththeLatticeTowers'construction
LatticeTower
Itemdescription Unit
EarthWorks
1
2
m
3
m
2
m
3
ConcreteWorks
5
6
7
8
SteelWork
9
m
t
nr
nr
CompanyA
CompanyB
CompanyC
Quantity UnitPrice(ZAR) Total(ZAR) UnitPrice(ZAR) Total(ZAR) UnitPrice(ZAR) Total(ZAR)
0
71
111
78
12
13
60
239
183
731
80
320
4980
9960
2417
4834
1764
3528
4
0
0
0
27
14253
5981
5981
2.5
45254
21
13070
215
248
113135
21
11721
97
112
33594
83986
42777
106943
Inordertocorrectlycomparethevalues,thetotalsforeachcategory,EarthWorks,ConcreteWorksandSteel
WorkareshowninTable29forthesteelmonopoletower,andTable30fortheSteelLatticetower.
Table29:SummaryTableofMonopoleCostsfromconstructiontofabrication
Company
EarthWorks(ZAR)
4669
11242
5877
ConcreteWorks(ZAR)
101519
26710
18392
SteelWork(ZAR)
126711
94064
119776
GrandTotal(ZAR)
232898
132016
144045
Table30:SummaryTableofLatticetowerCostsfromconstructiontofabrication
Company
EarthWorks(ZAR)
239
731
320
9960
4834
3528
SteelWork(ZAR)
113135
83986
106943
GrandTotal(ZAR)
123334
89551
110790
ConcreteWorks(ZAR)
Table31showsthetotalcostdifferencebetweentheSteelMonopoletowerandtheSteelLatticetower.The
coststhatwereassessedtoarriveatsuchamonetarydiscrepancywerethetotalcostsofcompaniesA,BandC
forallcategorieswhichincludedfoundationconstructionaswellassteelfabricationanderection.
Table31:Overalldifferenceincostsbetweenthemonopoleandsteellatticetowerincludingconstruction
costs
Company
Overalldifference(ZAR)
109564
C
42465
33255
ThecostcomparisonsmadeinTable29andTable30areshowngraphicallyinFigure82.
CostofMonopoleandSteelLatticeTowers
includingtheirfoundationcosts
250000
Cost(ZAR)
200000
150000
MonopoleTowerand
FoundationCosts
100000
LatticeTowerand
FoundationCosts
50000
0
A
Company
Figure82:CostComparisonforSteelTowersandFoundationscombined
ItcanbeconcludedthatconsistentlyhigherpricesarefoundforSteelMonopoletowersascomparedtoSteel
Lattice towers. The differences in cost between the Steel Monopole tower and the Steel Lattice tower have
beenplottedinFigure83.Figure83byimplicationshowsnotonlythecostdifferencebetweeneachkindof
tower but also the difference as quoted by the different companies, resulting in companies A, B and C in
ascendingorderasbeingthemostcosteffective.
CostdifferenceforTowerandFoundations
combined
120000
Cost(ZAR)
100000
80000
60000
Costdifferencefortower
andfoundations
combined
40000
20000
0
A
Company
Figure83:CostDifferencebetweencompaniesandTowersforTowerandFoundationcostscombined
STEELTOWERFABRICATION:
Figure84showsthecostoffabricatinganderectingthesteeltowers,exclusiveoffoundationconstruction.Itis
abundantlyclearthatthecostoftheSteelLatticetowers,regardlessoftheCompanyfromwhichthequotation
came,wastheleastcostlyineachofthe5instances.
Cost(ZAR)
SteelTowerQuotations
160000
140000
120000
100000
80000
60000
40000
20000
0
SteelMonopole
SteelLattice
Company
Figure84:SteelTowerQuotationComparison
In Figure 85, the differences between the cost of the fabrication of the Steel Lattice tower and the Steel
Monopoleareshown.
Cost(ZAR)
CostDifferencebetweenTowerTypes
70000
60000
50000
40000
30000
20000
10000
0
CostDifference
Company
Figure85:Differenceincostpercompanyandtowerdesignforthecostofthetowers
In order to account for the discrepancies in cost from the different companies, the average cost of a steel
tower was taken from the Monopole quotations, and an average cost was taken from the Steel Lattice
quotations.Theaveragecostforeachrespectivelywasthefollowing,R123383.35forMonopoles,andR98
169.43forSteelLatticetowers.TheaveragecostdifferencebetweenthetwokindsoftowersamountedtoR
25213.92whichwasthenexaminedasanoverallpercentageoftheaveragecostoftheMonopoletowersand
theSteelLatticetowersrespectively.Asapercentageofcosttheaveragedifferencewouldaccountfor20.45%
inpricereductionforMonopoletowersanda25.68%increaseinLatticetowersrespectivelyforthenumbers
toevenout.Theseamountsaccountforasignificantpercentageofthetotalcostofeachtower,whichshallbe
discussedagainlater.
FOUNDATIONCONSTRUCTION:
As was specified earlier in the chapter, the foundation construction includes the excavation of the ground
necessarytoconstructthefoundation,thetreatmentoftheconcretethatisexposedtotheground,andthe
amountofbackfillrequiredaswellasthevolumeofconcretenecessaryforconstruction.
Figure86showsthecostcomparisonofthefoundationconstructioncostsasaseparateentity.
CostComparisoninfoundationconstructionof
differenttowers
120000
Cost(ZAR)
100000
80000
MonopoleFoundation
Costs
60000
40000
LatticeFoundationCosts
20000
0
A
Company
Figure86:CostComparisoninfoundationconstructionfordifferenttowersandcompanies
Aswasthecasewiththecostsofthecombinedtowerandfoundationdesignandthetowerfabricationwhen
lookedatseparately,theMonopolesfoundationconstructioncostsareconsistentlyhigherthantheyarefor
theSteelLatticetowersfoundation.Onceagainitisconcluded,inascendingordercompaniesA,BandCare
themorecosteffectiveoptions.
CostDifferenceinfoundationconstruction
120000
Cost(ZAR)
100000
80000
60000
CostDifferencein
foundationconstruction
40000
20000
0
A
Company
Figure87:Costdifferenceinfoundationconstructionfordifferenttowersandcompanies
Figure88showseachofthecostdifferencesbetweenthequotationpricesfortheSteelLatticetowerandthe
Steel Monopole drawn on the same graph to illustrate which of the entities, foundation costs or steel
fabricationcostswasadominatingfactorinthepricediscrepancies.
ItwasdeterminedthatinaccordancewiththeparticularquotesgivenfromthecompaniesA,BandCthatthe
foundationconstructioncostswerethedrivingforceinpricediscrepanciesbetweenthedifferentcompanies,
in the most dominating instance in Company A, which had the most expensive price per unit for reinforced
concretepercubicmetre.
ComparitiveEntityCostComparisons
120000
Cost(ZAR)
100000
80000
FoundationCost
Comparison
60000
40000
SteelTowerCost
Comparison
20000
Foundationandsteel
TowerCostComparison
0
A
Company
Figure88:ComparativeEntityCostDifferences
7.5CONCLUSION:
The clear conclusion is that regardless of the steel fabricator or Construction Company, the overall costs of
constructing the Steel Lattice tower designed for this thesis as compared to the Steel Monopole tower
designedforthisthesisarelower.
WhencomparingaveragepricestheMonopoletowerandtheSteelLatticetowerwerecomparedseparately
fortheSteelFabricationcosts,thefoundationconstructioncostsandthentheeffectsofthecombinedcostsas
showninTable32
Table32:SummaryTableofCostComparisonsbetweenallentitiesaveragedprices
SummaryTable
Monopole
Lattice
Difference
SteelFabricationCost(ZAR)
123383.35
98169.43 25213.92
FoundationConstructionCost(ZAR)
56136.31
6537.33 49598.98
TotalConstruction,FabricationandErectionCost(ZAR)
109564.21
42465.32 33255.00
It is likely that wind turbines of the scale designed in this thesis, namely 3kW, would be used in larger
numbers, for small wind farms. Bearing this in mind reducing the cost of the tower construction and
foundation construction would be ever more appealing the larger the number of turbines. By reducing the
towerandfoundationconstructioncostsbyatleast30%,wherethetowerproportionofawindturbineasa
wholeconstitutesasmuchasupto30%oftheoverallcosts,theimplicationsarethereductionofthetotalcost
ofeachwindturbinebyasmuchas9%,whichonamultipleeffectscaleismostsignificant.
For the steel monopole towers seen constructed for wind turbines world over, considering the optimised
designtheyhave,thefabricationrequirementswouldbefarmorecostlythanthoseofthemonopoledesigned
for this thesis out of readily available Circular Hollow Sections. From the estimated price increase in
commercialmonopoletowersseeninindustry(ascomparedtothemonopoleforthisthesis),theinevitable
conclusion can be drawn that Steel Lattice towers are the most cost effective solution to wind turbine
supportingtowers.
CHAPTER8:CONCLUSIONANDRECOMMENDATIONS
It was evident from the outset of this research that the topic at hand provided many scattered pockets of
insightful information, information whichhadnotbeenaccumulated and arranged in thecontext that is set
outhere.Althoughtheresearchconductedforthisthesisisextremelyusefulintermsofpresentdevelopments
inthewindindustryinSouthAfrica,inthesensethatitisrelativelynewandundeveloped,theinfancyofthe
windenergyindustryinSouthAfricawasahindrancebecauseofalackofinformationthatcouldcontributeto
thedesignsofthewindturbinetowers.
Europeanpublicationsprovedmostusefulintermsofthevolumeofknowledgethattheycouldprovideonthe
windindustryingeneralaswellasthedevelopmentofenergycapturingbymeansofwindturbines.
It was certainly essential to do complete designs of each tower in order to accurately assess the structural
designprocessanddifferencesbetweentheprocessesinvolvedwitheachofthetowers.Thefullextentofthe
connection and foundations designs was also necessary as it assisted in understanding factors that could
possiblyinfluencethedesignandconstructionprocessesofeachofthetowers.
HavingdoneextensiveFiniteElementAnalysesofthesteellatticeandthesteelmonopoletowers,iscanbe
concludedthatthesteelmonopolehasmorecomplicatingdynamiccomponentstoconsiderinassessingthe
differenteffectsofvortexshedding.However,thesteellatticetowerhadtobemodelledmoreextensivelyfor
thedifferentpossiblewindloadcases,sooveralleachofthemhadcertainmodellingprosandcons,neitherof
whichstoodoutfromtheotherasmoretimeconsumingtoanalyseinitsfiniteelements.
It can be concluded from the findings in Chapter 7 that the overall feasibility of a steel lattice tower as a
supporting structure for a small scale Wind Turbine is significantly greater than that of a steel monopole
tower.
Thesteelmonopolewasfoundtobemorecostlytofabricate,morecostlytoconstructandhadmorecostly
foundationsandusedmoresteelthatthesteellatticetower.
From the prerequisite of a proposed alternative that provides the same stability, the same height
requirementsastheexistingsolutionbutatareducedcost,theresearchconductedinthisthesiswascertainly
successfulinproducingsuchanoutcome.
In terms of further development in this topic, it would be an interesting extension to create a design
optimisationinterfacebetweenStrand7andMicrosoftExcelwhichwouldallowonetohaveanoptimiseduse
ofsteel.TheStrand7softwaremakesallowanceforsuchinterfaces.
It would be recommended that a tapered steel monopole tower also be considered in the cost analyses of
furtherstudyaswellasdifferentcrosssectionaldesignsofthesteellatticetower,andperhapsahybridofthe
steellatticeandsteelmonopoletowershouldbeinvestigatedtoharnessthebestaspectsofeachofthemin
onedesign.
120|C h a p t e r 8 : C o n c l u s i o n a n d R e c o m m e n d a t i o n s
LISTOFREFERENCES
RELEVANTSOUTHAFRICANDESIGNCODES:
SANS101621:2005
SANS101601:2011
SANS101602:2011
SANS101603:2011
OTHERRESOURCES:
Balas,M.J&Wright,A&Hand,M&Stol,K.2003.DynamicsandControlofHorizontalAxisWindTurbines,in
ProceedingsoftheAmericanControlConference:Denver,Colorado:37813793.
Gipe,P.2004.WindPower.WhiteRiverJunction,Vermont:ChelseaGreenPublishingCompany.
Graham Kelly, S. 2000. Fundamentals of Mechanical Vibrations. Singapore: McGrawHill Higher education
companies.
Hansen,O.L.M.2008.AerodynamicsofWindTurbines.UnitedKingdom:Earthscan.
Harikrishna, P & Shanmugasundaram, J & Gomathinayagam, S & Lakshmanan, N.1999. Analytical and
experimentalstudiesonthegustresponseofa52mtallsteellatticetowerunderwindloading.Computersand
Structures70(1999):149160.
Kato, B., Hirose, R. 1985. Bolted Tension Flanges Joining Circular Hollow Section Members. Journal of
ConstructionalSteelResearch5(2):79101.
Kiessling, F., Nefzger, P., Nolasco, J.F & Kaintzyk, U. 2003. Overhead Power Lines, Planning, Design,
Construction.Germany:SpringerVerlagBerlinHeidelberg.
Lee, P & McClure, G. 2006. Elastoplastic large deformation analysis of a lattice steel tower structure and
comparisonwithfullscaletests.JournalofConstructionalSteelResearch63(2007):709717.
Marsh,G.2005.Windturbines,howbigcantheyget?Refocus,March/April2005.
Peterson, B. E. 2010 Evaluate the effect of turbine period of vibration requirements on structural design
parameters.AppliedPhysicalSciencescorp.
ProkonSoftwareconsultants.ProkonStructuralAnalysisandDesignsoftware. Copyright19962012Prokon
SoftwareConsultants.
Robberts,J.M,Marshall,V.2009.AnalysisandDesignofConcreteStructures.CementandConcreteInstitute.
SmallWindTurbines.2011.[Online].Available:www.windenergythefacts.org[2011,Ferbruary17].
Sorensen, J.D & Toft, H.S. 2010. Probabilistic Design of Wind Turbines [Online]. Available:
www.mdpi.com/journal/energies,[2011,March03].
SouthAfricanSteelConstructionHandbook.2008.SouthAfrica:PaarlPrint.
121|R e f e r e n c e s
StottrupAndersen,U.(1997).MastsandTowers.Tech.Rep.,DenmarkUniversityIASSWorkingGroup.
Strand7SoftwarePtyLtd.
Theenergyinthewind.2011.[Online].Available:www.windpower.org[2011,February16].
TheEuropeanWindEnergyAssociation(EWEA):2009.WindEnergyTheFacts.Earthscan:London.
TurbineTowers.2011.[Online].Available:www.houseenergy.com[2011,February18].
Yeh,TandWang,L.2008.AStudyonGeneratorCapacityforWindTurbinesUnderVariousTowerHeightsand
RatedWindSpeedsUsingWeibullDistribution.IEEEtransactionsonEnergyConversion,23(2):592602.
122|R e f e r e n c e s
APPENDIXA:TOWERDESIGNCALCULATIONS
A1:STEELMONOPOLEDESIGN
A1.1DESIGNREFERENCEPARAMETERSANDTOWERRESISTANCECALCULATIONS
TOWERGEOMETRYANDAXIALCOMPRESSIONRESISTANCE:
SlendernessRatio
SANS101621:200510.4
Monopoles:Constantcrosssection
Innerandouterdiameters
do
0.508 m
di
0.4826 m
t
0.0127 m
Lengthoftower(height)
L
16 m
Effectivelengthfactor
K
2 **cantilevercolumn
inertia
4
I
0.0006064 m
Crosssectionalarea
2
A
0.0197616 m
Radiusofgyration
r
0.1751726 m
Slendernessratio
KL/r
182.67702
Check
COMPRESSIONSLENDERNESSOK
SANS101621:200511
Widthtothicknessratios:elementsincompression
Classificationofsections
Table3:Maximumwidthtothicknessratios:
elementsinaxialcompression
23000
necessaryparameters
do
508 t
*mm
12.7 fy(MPa)
*mm
E(GPa)
CLASS3SECTION
A|A p p e n d i x A : T o w e r D e s i g n C a l c u l a t i o n s
300
200
FLEXURALANDAXIALCOMPRESSIONRESISTANCE:
Table4:Maximumwidthtothicknessratios:
elementsinflexuralcompression
13000

1
18000
66000
necessaryparameters
do
508 t
12.7 fy(MPa)
*mm
d0/t
class1
class2
class3
*mm
300
E(GPa)
200
40
43.333333
60
220
CLASS1SECTION
AxialCompressionresistance:
SANS101621:2005Section13.3
FlexuralBuckling13.3.1
1.34
0.9
2.2520586
factoredaxialcompressiveresistance
Cr
970903.31 N
970.9033113 kN
B|A p p e n d i x A : T o w e r D e s i g n C a l c u l a t i o n s
SANS101621:200513.5BendingLateralysupportedmembers
**usingclassclassificationfromflexuralcompressionclassification
Class1and2sections
Class3sections
Plasticandelasticeffectivesectionmoduli
Zpl
0.0031163
Ze
0.0023874
factoredmomentresistance
841393.8
Mr
Nm
**takesintoaccountwhichclasssectionitis
SHEARRESISTANCE:
SANS101621:200513.4.2Shearresistance
ofwebsofflexuralmembersnothavingtwoflanges
***maxatultimateloadmaynotexceed0.66f y
V r(MAX)
178200000 N
Vr=0.66(0.9A)fy
Vr
check
3169364.1 N
SHEAROK
C|A p p e n d i x A : T o w e r D e s i g n C a l c u l a t i o n s
A1.2WINDCALCULATIONSONSTEELMONOPOLETOWER
SANS101603:2011
ULTIMATELIMITSTATE(ULS)
7.2BasicValues
location:universityofstellenbosch:engineeringfaculty
7.2.2
fundametalvalueofthebasicwindspeed(fromfigure1inSANS101603)
28 m/s
Figure1inSANS101603:2011
vb,0
terraincategory
B
Table2inSANS101603:2011
returnperiod
50 years
thisreturnperiodmaybetakenasthedesign
designlifeofcomponents
workinglifeofthestructure
20 years
altitude(abovemeansealevel)
100 m
hubheight(referenceheight,abovegroundlevel)
16.2 m
(heightoftowerplusonemeter)
towersurfacefinish
galvinisedsteel
roughnesscoefficient
k
0.2 mm
CalculationprocedureinaccordancewithTable5SANS101603:2011
ULS
Fundamentalbasicwindspeed(SANS101603:2011Figure1)
28 m/s
vb,0
probabilityfactorSANS101603:20117.2.3
K
0.2
p
0.02
(annualprobabilityofexceedanceformeanreturnperiodof50years)
n
0.5
cprob
Basicwindspeed(SANS101603:2011Equation1)
28 m/s
vb
SANS101603:20117.3.1
Variationwithheight
terrainroughnessfactorSANS101603:2011
z
16.2 m
0
z0
zg
300
zc
cr(z)
5
0.095
1.0323091
c0(z)
peakwindspeedSANS101603:20117.3
39.2 m/s
vb,peak
vp(z)
40.466516 m/s
SANS101603:20117.4Peakwindspeedpressure
1.184 (Table4SANS101603:2011)
2
969.42302 N/m
qp(z)
D|A p p e n d i x A : T o w e r D e s i g n C a l c u l a t i o n s
SERVICEABILITYLIMITSTATE(SLS)
7.2BasicValues
7.2.2
vb,0
16 m/s
terraincategory
B
returnperiod
50 years
20 years
100 m
16.2 m
towersurfacefinish
galvinisedsteel
k
0.2 mm
SLS
16 m/s
vb,0
K
0.2
p
0.02
n
0.5
cprob
16 m/s
vb
SANS101603:20117.3.1
Variationwithheight
z
16.2 m
z0
0
zg
300
zc
cr(z)
5
0.095
1.032309
c0(z)
22.4 m/s
vb,peak
vp(z)
23.12372 m/s
1.184 (Table4SANS101603:2011)
qp(z)
316.5463 N/m2
E|A p p e n d i x A : T o w e r D e s i g n C a l c u l a t i o n s
A1.3ACTIONSINDDUCEDONTTHETOWERR
Distrib
butedwind
dloadupth
heheightofthetower
ULS
w(z)
SLS
w(z)
beyond16m
m/swindspeeed(28m/s)
286.65218
8 N/m
w(z)
564.275
59 N/m
w(z)
184.253
34 N/m
upto16m/sswindspeed
93.600712
2 N/m
Axialdrragforce,inxxdirectionexxertedonturrbineduringoperation
a
0.33
in
nductionfacttorofthewin
ndturbine
Dblades
4m
A
12.56637
7m
Ultimattelimitstate::
Q,wind
FxT,ULS
1.5 SANS10160
01:2011tablee3
8080.406
6N
FxT,Unnfactored
538
86.937 N
Servicea
abilitylimitsstate:
Q,wind
0.6
6 SANS10160
01:2011secti on8.3.1.1
FxT,SLS
2110.8
8N
FxT,Unnfactored
Verticallselfweightloading,inzzdirection
weightss(kgs):
120
total
0 kg
Ultimattelimitstatee:
G,Dead
1.2
2 SANS10160
01:2011tabl e3
FZT,ULS
1412.64
4N
Servicea
abilitylimitsstate:
G,Dead
1.1
1
FZT,SLS
1294.92
2N
3518 N
F|A p p e n d i x A : T o w e r D e s i g n C a l c u l a t i o n s
A1.4INNTERACTION
NEQUATION
NS
G|A p p e n d i x A : T o w e r D e s i g n C a l c u l a t i o n s
A2:STEELLATTICEEDESIGN
A2.1DESIGNREFERENCEPARRAMETERSAANDTOWERRESISTANCECALCULAATIONS
TOWERGEOMETRYAANDAXIALCOMPRESSION
O
RESISTANCE:
Memb
bercapaccitycheckks:
Attributtesofeachm
member
50x50x5
parame
eter
m(kg/m
m)
70x70x6 100x100x8
1
3.77
6.38
12.2
A(mm )
ax(mm)
480
14
813
19.3
1550
27.4
ay (mm))
14
19.3
27.4
v1(mm))
19.9
27.3
38.7
v2(mm))
17.6
24.6
35.2
I(10 mm
m)
0.11
0.369
1.45
3.05
15.1
7.27
21.3
19.9
30.6
0.174
0.585
2.3
4.92
19
11.8
26.8
32.5
38.5
0.046
0.153
0.599
Ze(10 mm
m )
r(mm)
3
4
m)
J(10 mm
Cw
fy (Mpa))
2.29
9.73
5.6
13.7
15.5
19.6
4.58
0
355
11.2
0
355
37.6
0
355
fu(Mpa))
470
470
470
Aboutxxxandyy
6
Ze(10 mm
m )
r(mm)
Aboutu
uu
6
I(10 mm
m)
3
Ze(10 mm
m )
r(mm)
Aboutvvv
6
I(10 mm
m)
3
H|A p p e n d i x A : T o w e r D e s i g n C a l c u l a t i o n s
Slenderness
50x50x5
K
L
KL/r
70x70x6
K
L
KL/r
1
583.095
59.9275
CompressionCheck OK
TesionCheck
OK
CompressionCheck OK
TesionCheck
OK
ClassClassification
Table3SANS101621:2005elementsinaxialcompression
50x50x3
70x70x6
Legsofangles
Legsofangles
b
50
b
t
5
t
b/t
10
b/t
200/fy
class
0.5
100x100x8
K
L
KL/r
1
2605.52
190.1839
10.6149
200/fy
class
class3
1
3658.04
186.6347
CompressionCheck OK
TesionCheck
OK
100x100x8
Legsofangles
b
t
b/t
70
6
11.66666667
0.5
10.6149
200/fy
class
class4
100
8
12.5
0.5
10.6149
class4
CompressionTests
FlexuralBuckling
Clause13.3.3Class4membersincompression
50x50x5
W
10
k
0.43
0.9
61865.54 N
Cu
f
1.43E+02 MPa
checkwhichftouse
1.43E+02 MPa
E
200000 MPa
1.58E+01
Wlim
70x70x6
W
k
Cu
f
checkwhichftouse
E
Wlim
check
check
Aef=A
beff
57.07 mm
effectivecrosssectionalarea
Aeff
11.6667
0.43
0.9
61865.54 N
8.46E+01 MPa
8.46E+01 MPa
200000 MPa
2.06E+01
Aef=A
beff
check
78.42 mm
480 mm
Aeff
100x100x8
W
k
Cu
f
checkwhichftouse
E
Wlim
12.5
0.43
0.9
61865.54 N
4.43E+01 MPa
4.43E+01 MPa
200000 MPa
2.84E+01
Aef=A
beff
89.44 mm
813 mm
Aeff
1550 mm
Determiningf e
fey
2095.8409
fey
208.8385
fey
218.6524
fex
32938.4580
fex
10379.0092
fex
10576.3877
x0
2
y0
2
rx
2
ry
396.01 mm
0 mm
94.6729 mm
361 mm
x0
2
y0
2
rx
2
ry
745.29 mm
0 mm
187.69 mm
718.24 mm
1651.22 mm
1497.69 mm
2
y0
2
rx
2
ry
0 mm
384.16 mm
1482.25 mm
3364.1 mm
x0
r0
851.6829 mm
r0
fez
784.2316
fez
584.0104
fez
504.7604
0.5350
0.5486
0.5548
r0
feyz
648.9137
feyz
175.0291
feyz
176.4395
fe
648.9137
fe
175.0291
fe
176.4395
0.7396
1.4242
1.4185
1.34
1.34
1.34
I|A p p e n d i x A : T o w e r D e s i g n C a l c u l a t i o n s
AxialCompressionSANS101621:200513.3
Cr
101739.8005 N
Cr
97492.64562 N
Cr
186686.297 N
check
SatisfactorySection
check
SatisfactorySection
check
SatisfactorySection
SF
check
0.608076089
SF
check
0.634566224
OK
OK
SF
check
OK
0.331387686
TensileCapacityofMembersSANS101621:2005
clause13.2a)
i)
Tr
ii)BlockTearout
clause13.2a)
i)
Tr
ii)BlockTearout
153360 N
2
400 mm
Ane
clause13.2a)
i)
Tr
ii)BlockTearout
259753.5 N
2
717 mm
Ane
Tr
143820 N
Tr
257797.35 N
Tr
62000 N
Tu
62000 N
Tu
SatisfactorySection
0.404277517
OK
PinnedConnections
clause13.2b)
Tr
check
OK
check
Tu/Tr
check
clause13.2b)
Tr
check
126900 N
SatisfactorySection
0.238687833
OK
227468.25 N
OK
1422 mm
Ane
Tu
check
Tu/Tr
check
495225 N
511280.1 N
62000 N
check
Tu/Tr
check
SatisfactorySection
0.125195618
OK
clause13.2b)
Tr
check
OK
451129.5 N
LoadBearingCapacityofMembersSANS101621:2005
clause13.10LoadBearing
br
t
d
n
a
Br,i
0.67
5
16
2
40
151152 N
br
t
d
n
a
Br,i
0.67
6
16
2
40
181382.4 N
br
t
d
n
a
Br,i
0.67
8
16
2
40
241843.2 N
Br,ii
125960 N
Br,ii
151152 N
Br,ii
201536 N
check
OK
check
OK
check
OK
J|A p p e n d i x A : T o w e r D e s i g n C a l c u l a t i o n s
A2.2WINDCALCULATIONSONSTEELLATTICETOWER
SANS101603:2011
ULTIMATELIMITSTATE(ULS)
7.2BasicValues
7.2.2
vb,0
28 m/s
terraincategory
B
returnperiod
50 years
20 years
100 m
16.2 m
towersurfacefinish
galvinisedsteel
k
0.2 mm
28 m/s
vb,0
K
0.2
p
0.02
n
0.5
cprob
28 m/s
vb
SANS101603:20117.3.1
Variationwithheight
z
16.2 m
0
z0
zg
300
zc
5
0.095
cr(z)
1.032309
c0(z)
39.2 m/s
vb,peak
vp(z)
40.46652 m/s
1.184 (Table4SANS101603:2011)
2
969.423 N/m
qp(z)
K|A p p e n d i x A : T o w e r D e s i g n C a l c u l a t i o n s
SERVICEABILITYLIMITSTATE(SLS)
7.2BasicValues
7.2.2
16 m/s
vb,0
terraincategory
B
returnperiod
50 years
20 years
100 m
16.2 m
towersurfacefinish
galvinisedsteel
k
0.2 mm
**seecodeextractssheet**
16 m/s
vb,0
K
0.2
p
0.02
n
0.5
cprob
16 m/s
vb
SANS101603:20117.3.1
Variationwithheight
z
16.2 m
0
z0
zg
300
zc
0.095
cr(z)
1.032309
c0(z)
22.4 m/s
vb,peak
vp(z)
23.12372 m/s
1.184 (Table4SANS101603:2011)
316.5463 N/m2
qp(z)
L|A p p e n d i x A : T o w e r D e s i g n C a l c u l a t i o n s
FactorredWindLo
oad:
Ultimate
eLimitState
w(z)
ServiceabilityLimitStatte
26 N/m
1454.13452
189.9278 N/m
m
w(z)
APPLYINNGTHEWIN
NDLOADS:
WINDL OADCASESS:
Figure35
5SANS101603:Forcecoeffficientsaccord
dingtothecro
osssectionalsshapeandwin
nddirectiononthecrossseection
SquareLLatticeCrossSSection
StraightPart
cf,0
cf
case:
(iv)
(v)
0.573238
0.553312
TaperedPart
case:
(iv)
(v)
1.8
2.5
cf,0
0.202345
0.638628
0.925
0.93
2.36
2.55
1.6
665
2.3
325
cf
0.97
0.945
2.28
892
2.409
975
SolidittyRatio
Straight
t Part
Loadcase(iv)
0.573238
StraightPart
Loadccase(v)
0.553312
ProjecteedArea
ProjecctedArea
2
A
0.75 m
AreaofSSteel
2
Ac
0.429929 m
A
6m
1.06066
Areao
ofSteel
2
Ac
6m
0.586876
TaperedPart
Loadcase(iv)
TapereedPart
Loadccase(v)
Ac
A
6.282819 m
31.05 m
0.202345
Ac
5m
23.33925
36.5459
9m
0.638628
8
M|A p p e n d i x A : T o w e r D e s i g n C a l c u l a t i o n s
SlendernessRatio
TaperedPart
L
13.5 m
b
2.2997 m
b
L/b
0.3 m
5.870331 m
L/b
45 m
*averagecrosssectionatmidheightoftaperedpart
**smallestpossiblecrosssection
Inaccordancewithtable22,SANS101603No.1includeslatticestructures
L
13.5 m
forL<15m,=2L/bor70,whicheverissmaller
90
check
70
StraightPart
L
b
2.5 m
0.3 m
L/b
8.333333
Inaccordancewithtable22,SANS101603No.1includeslatticestructures
L
2.5 m
forL<15m,=2L/bor70,whicheverissmaller
16.66667
check
16.66667
2
WindLoadCases(Factored)(N/m )
LC(iv)
LC(v)
ULS
SLS
ULS
SLS
StraightPart
StraightPart
2421.134 316.2297 2421.134 441.5821
TaperedPart
TaperedPart
3328.805 434.7827 3504.101 457.6785
2
WindLoadCases(Unfactored)(N/m )
LC(iv)
LC(v)
Straight Tapered Straight Tapered
1614.089 2219.203 2253.909 2336.067 ULS
527.0496 724.6378 735.9701 762.7974 SLS
height
LC(iv)
LC(v)
0 2219.203 2336.067
13.5 2219.203 2336.067
13.5 1614.089 2253.909
16.2 1614.089 2253.909
N|A p p e n d i x A : T o w e r D e s i g n C a l c u l a t i o n s
A2.3ACTIONSINDDUCEDONTTHETOPOF THETOWER
Axialdrragforce,in
nxdirectionexertedontturbineduringoperation
a
0.33
inductionf actorofthewindturbin
ne
Dblades
4m
2
12.56637 m
A
Q,wind
1.5 SANS1016
601:2011tab
ble3
FxT,ULS
8080.40
06 N
Servicea
abilitylimitstate:
Q,wind
601:2011se ction8.3.1.1
1
0.6 SANS1016
FxT,SLS
2110.8 N
Verticallselfweightloading,in
nzdirection
weightss(kgs):
120 kg
total
G,Dead
1.2 SANS1016
601:2011tab
ble3
1412.64 N
FZT,ULS
Servicea
abilitylimitstate:
G,Dead
1.1
FZT,SLS
1294.92 N
O|A p p e n d i x A : T o w e r D e s i g n C a l c u l a t i o n s
APPEN
NDIXB:CONNECTIO
O
ONS
B.1STEEELMONOPPOLECONNECTIONS
RINGFLANGECON
NNECTIONS:METHOD1
1
Parametters
d
Di
20 diameteroffthebolt
508 isequaltotheoutertubediameter++thewidthto
othecentreo
ofthefilletswelds.
D0
648 Do=Dp+4d
Dp
568 Dp=Di+3d
Tt
12.7
Tf
n
25
16 numberofb
bolts
perimetterofmidflaange
rati ooftubethi cknesstoflangethicknesss:
pm
1784.425 mm
0
0.508
**articl e'slowest0.21
Tt/TTf
minimumspacingbe
etweenboltssalongperim
meter
30 mm
smin
maxNo.Boltsinspacing
No.maxx 59.48082 bolts
spacingbetweenboltsalongperimeter(actuaal)
111.5265 mm
s
A|A p p e n d i x B : C o n n e c t i o n D e s i g n
Boltfo
orceatsepa
aration:
Therelationshipbetw
weentheforceactingonth
heboltsandth
heinstantofsseparation
Separati onForce
Ts=
To
Ts
k
223.3182 kN
Boltpreload
To
k
178 kN
Tab
ble1SANS100
094
crossse ctionalareao
oftheunthreaadedshankoffthebolt
2
m
Ab
314.1593 mm
**areaausingdiameeterofboltbeeforethreadin
ng
effectiveeareaoftheccompressedfllangeplate
2
m
Ap
1233.948 mm
*eq
qn
D
lp
37 mm
m
50 mm
m
diameterofwasherfaceofbo
olt
grip
plength=2Tf
dp
22 mm
m
diameterofthebolthole
Ty
d
254.2857 kN
k
20 mm
m
1.254597
*
**Table6.4Re
edBook
**+2mm
yie ldstrengthoffthebolt
tensilestreengthofboltTTable6.1RedBook
diameteroftheuntreadedsh
hankofthebo
olt
Yieldloa
adoftheflan
nge
Pp
6.433458 kN
Pp=Ump
**ch
heck1/4
mp
53.90625 kNm
2
mp=(1/4)Tf y
0.119345
U==(48tan(/12))*(Dp/(DpDi)
umloadoftheflange
maximu
Pu
**fullplasticmomeentperunitw
widthofthefflange
Pu=(u/y )Pp
8.764422 kN
U=(128tan(/32))*(Dp/( DpDi))
FlangeSttrengths
y
M
345 MPa
u
470 MPa
M
Maximu
umbendingstrengthoffla
angewhensep
eparationoccu
ursaftermaxximumyieldin
ngoftheflange
U'
P'p
9
0.2910549
4.0489417
7 kN
P'u
5.5159496
6 kN
B|A p p e n d i x B : C o n n e c t i o n D e s i g n
Evalua
ationofprryingforcee
Thepryiingforceactin
ngbeforesep
paration
Parameters:
B
265.5 mm
Dp/2a/2
A
324 mm
37 mm
43.67438 kNm
Do/2
2
mp=(1//4)Tf y
(m
mp<m<mu)
mp
mu=(1/4)Tf u
13911.26 mu
73437.5 Nm
m
Pryingfforceactingb
beforeseparation
Reaction
nforceperun
nitlengthoflineIK:
r
0.088466 kN/m
Totalreaactionactingthroughouttthecircularfl ange
1844.621 kN
R
RatioofPryingForce toexternall oad
0.145113
1
Pryingfforceactingaf
aftertheseparation
Reaction
nforceperun
nitlengthoflineIK:
rs
0.961693 kN/m
284 Dp/2
Totalreaactionactingthroughouttthecircularfl ange
324 Do/2
1249.435 kN
Rs
RatioofPryingForce toexternalLLoad
0.375
2
a
m
37
43.67438 kNm
k
C|A p p e n d i x B : C o n n e c t i o n D e s i g n
Separationloadandfractureloadofthebolts
Separationload
P'
Tensileforceappliedtospecimen
3573.092
Ts
Ps
3120.297 Boltseparationload
P'=Ps
Fractureloadofthebolts
Fractureoccursafteryeldingoftheflange
4068.571 Tensilestrengthofbolts
Tu
bPm
2958.961 Maximumtensilestrengthofaconnection
P=bPm
Fractureoccursbeforeyieldingoftheflange
4068.571 Tensilestrengthofbolts
Tu
bPm
4068.571 **sincenopryingactionneedstobetakenintoaccounthere.
MaximumStrengthofaConnection:
1.Criticalloadscorrespondingtoflangefailure:
i)flangeyieldsandreachesitsmaxstrengthbeforeseparationtakesplace
maximumloadofjoint:
5.5159496 kN
P'u
ii)flangeyieldsafterseparationtakesplace
maximumloadofjoint:
Pu
8.7644217 kN
D|A p p e n d i x B : C o n n e c t i o n D e s i g n
METHO
OD2:GENERALTHEORY
YOFRESISTTANCE:
FromStraand7
48.70697 kN
P
N
Q
4.870697 kN
N
T
49.39 kN
N
higheestsingularnodaltensionaro
oundbasecirccumference
limiteedto0.1P,toccreateathicke rmoreconserrvativeflangetthickness
Boltrresistance
Forasimp
plestresscalculaationinthesinglleboltinthesecctionoftubulartension"P"
A
314.1
1592654 Areaofaboltinmm2
=P/A
15503
39100.8 Pa
155.0
0391008 Mpa
oltinaccordanceewithhighestteensileloadonperimeter.
stressperbo
wouldbeevvenlessthanthiisfortheboltintheflange,sinccetheforceisfurtherawayfrom
mthecentre.
Maximum
mallowablestressssineachbolti naccordancewiithclause13.12.1.3is262MPafo
ora10.9Srpreten
nsionedbolt.
E|A p p e n d i x B : C o n n e c t i o n D e s i g n
Tributarylengthl
l
0.099745567 m
0.9
flangestrength
fy
3.35E+08 Pa
Maximumdesignstrengthofbolttoavoidfatigue
2.62E+08 Pa
seeSANS101621:2005clause26
fu
boltstrengthcalculatedwith262MpainaccordancewithSANS101621:2005clause13.12.1.3
2
Ab
3.14E04 m
M20
b
0.80
49385.84 N
Tr
AtBaseConnection:
FromDiagram
a
0.05 m
b
0.08 m
t
0.04 m
Momentresistanceoftheflangesrespectivelyare:
AtBaseConnection:
Mr
8019.543567 Nm/mm
Atmidandtopconnections
FromDiagram
a
0.05 m
b
0.05 m
t
0.02 m
Atmidandtopconnections
Mr
2004.886 Nm/mm elasticmoments
ForEquilibrium:Themaximumallowabletensileforcewouldbe:
P
80683.34917 N
P
44741.78 N
FromequilibriumthemaximumPryingforcewouldbe:
Q
31297.51266 N
Q
4644.059 N
check!
Q
60146.58 N
Qmustbesmallerthanthefollowing:
240586.307 N
F|A p p e n d i x B : C o n n e c t i o n D e s i g n
B.2STEELLATTICETOWERCONNECTIONS
GUSSETPLATECONNECTIONS
LoadBearingCapacityofGussetPlateSANS101621:2005
clause13.10LoadBearing
br
t
d
n
a
Br,i
0.67
8
16
2
45
241843.2 N
Br,ii
226728 N
check
OK
BoltResistance
ShearBoltResistance
SANS101621:2005
Clause13.12.1.2
OrdinaryGrade8.8boltyieldstrength
70x70x6
100x100x8
g1
40 mm
g1
50 mm
dmax
16 mm
dmax
24 mm
0.8
0.8
Ab
452.3893 mm
fu
830 Mpa
Ab
201.0619 mm
fu
n
Vr
830 Mpa
2
56072.15 N
V r,n
112144.3
Vu
V u/V r
check
62016 N
0.553002
OK
Vr
126162.3 N
Vu
62016 N
V u/V r
check
0.491557
OK
G|A p p e n d i x B : C o n n e c t i o n D e s i g n
TakingFFatigueintoa
accountwithreducedyielldstressforB
Boltsandhenceincreasing
gtheirnumbeer
70x70x6
6
00x100x8
10
g1
40 mm
g1
50 mm
dmax
20 mm
dmax
m
24 mm
0.8
0.8
Ab
314.1593 mm
Ab
45
52.3893 mm
fu
n
Vr
262 Mpa
3
27656.07 N
fu
n
Vr
262 Mpa
2
39
9824.74 N
V r,n
82968.21
V r,n
r
79
9649.48
Vu
V u/V r
check
62016 N
0.747467
OK
Vu
V u/V r
ch
heck
62016 N
0.778612
OK
K
Spacing
gandEdged
distancesforrmultiplebo
oltsinarow
w
d
16 mm
Table6.16SSACSH
4 mm
45
smin
dedge
4 mm
40
dend
4 mm
40
H|A p p e n d i x B : C o n n e c t i o n D e s i g n
I|A p p e n d i x B : C o n n e c t i o n D e s i g n
APPEN
NDIXC:FOUNDATIO
O
ONDESIGN
N
C.1STEEELMONOPPOLETOWEERFOUNDATTIONDESIG
GN
INPUTVALUESFORPROKONFOUN
NDATION DESSIGN
OUTPUT FROMPROKONFOUNDATTIONDESIGN
A|A p p e n d i x C : F o u n d a t i o n D e s i g n
BENDINGGSCHEDULEOUTPUTFROMPROKONF OUNDATIONDESIGN
B|A p p e n d i x C : F o u n d a t i o n D e s i g n
HANDCALCULATIO
ONSFORSTEELMONOPPOLETOWEERFOUNDATIONDESIG
GN
SSLS:
A
Axialload:
N
3
31.83345
kN
M
Moment:
M
4
48.53834
kNm
m
B
BaseShear:
V
3
3.009367
kN
ULS:
Axiallo
oad:
N
34.76136 kN
Momen
nt:
M
241.400
08 kNm
BaseSh
hear:
V
14.9600
06 kN
Eccentri
ricity
e
6.944516 m
1
1.524759
m
VariablleParameteers
Dimenssionsoffoun
ndationdeteerminedthroughequilib
briumofoveerturn
h1
0.5 m
h2
1.2 m
h3
UpperB
Block
D
0.2 m
BottomBlocck
B
B
1m
4m
1.Staticcequilibrium
mwithasafe
etyfactor: ULS
U
Overturrnmomentaaboutbottomcorneroffoundationfooting
Mo
171.8781 kNm
Resistin
ngMomentaaboutbottomcorneroffoundationfooting
MR
976.6556 kNm
check:
MR/Mo
MR/Mo>
1.5
5.682258 OK
C|A p p e n d i x C : F o u n d a t i o n D e s i g n
2.Maximumbearingpressure:
Newaxialloadincludingownweightoffoundation:
ULS
SLS
soilforce
N1
soilforce
508.5504 kN
N1
381.4128 kN
concreteforce
N2
265.5763 kN
toweraxialforce
N3
34.76136 kN
concreteforce
N2
243.445 kN
toweraxialforce
N3
31.83345 kN
NTOT
NTOT
656.6912 kN
M
e new
48.53834 kNm
0.073913 **neweccentricity
808.8881 kN
M
e new
check
241.4008 kNm
0.298435 **neweccentricity
e<D/6
BearingCapacityoftheGround
Bearingcapacityofground:
p
250 kN/m
Inaccordancewith12.3.3oftheCementandConcreteinstituteBook
Maximumbearingpressure:
Wheree>D/6
79.22983 kPa
pmax
56.82431 kPa
pmax
Wheree<D/6
pmax
73.18683 kPa
27.92418 kPa
OK
pmin
check
check
OK
BearingPressureDistribution
Pressure(kN/m3)
120
100
80
60
e>D/6
40
e<D/6
20
0
0
BaseofFoundation(m)
D|A p p e n d i x C : F o u n d a t i o n D e s i g n
C.2STEEELLATTICEETOWERFOUNDATION
O
NDESIGN
INPUTVALUESFORPROKONFOUN
NDATION DESSIGN
OUTPUT FROMPROKONFOUNDATTIONDESIGN
BENDINGGSCHEDULEOUTPUTFROMPROKONF OUNDATIONDESIGN
E|A p p e n d i x C : F o u n d a t i o n D e s i g n
F|A p p e n d i x C : F o u n d a t i o n D e s i g n
HANDCALCULATIONSDESIGNINGTHESTEELBASEPLATESFORTHESTEELLATTICETOWERS
FOUNDATION
Forces at Base of each Leg
Worst Case Scenario
Fxyz
Fz (Comp and Tension)
72413 N
68850 N
In accordance with 4.2.2 Concentrically loaded bases SASCH

1. Determining the size of the base plate
c
fcu
Abaseplate
0.6
30 MPa
2
0.004023 m
2
4022.944 mm
for a square base plate:
b,d
63.42669 mm
NOTOK
2.Determiningthevalueofc
pcolumnperimeter
400 mm
check
Acolumn
forquadraticequation:
a
b
c
x
x
Valueofc
c
ThicknessofBaseplate
fy
w
t
tcommercial
tmin,flangethickness
1550 mm
4
400
2472.944
5.841169
105.8412
5.841169 mm
355 MPa
18 MPa
2.161244 mm
5 mm
8 mm
checkbaseplatedimensions
107.6823 mm
bmin,candanglesize
M24
24 mm
spacing
40 mm
bmin
315.6823 mm
350 mm
G|A p p e n d i x C : F o u n d a t i o n D e s i g n
APPENDIXD:ATTACHEDCDOFDESIGNFILES
Attached CD of Microsoft Excel design calculations as well as the Strand7 designs of each of the Towers
designedinthisthesis.
A|A p p e n d i x D : C D A t t a c h e d o f D e s i g n F i l e s

Turb Des

Enviado por

Dados do documento

Título original

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Turb Des

Enviado por

Direitos autorais:

Formatos disponíveis

D

Thessis presentedd in fulfilmen

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Steel lattice towers can be

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

In order to take thesse tensions in

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

f is the calcculated comp

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

50x50x3 61865.54 59542.47 NOTOK

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za

Stellenbosch University http://scholar.sun.ac.za