Kaplan Turbine Improvement Thesis
Kaplan Turbine Improvement Thesis
Kaplan Turbine Improvement Thesis
turbine
Forbedre og vidreutvikle en Kaplan smturbin
Lars Fjrvold
Preface
During this thesis my knowledge about testing hydraulic machinery, error
analysis and CFD has increased drastically. Also getting the opportunity of
hands on experience helping Trygve Opland constructing the test rig in the
laboratory have been inspirational and eye opening. There are a lot of things
that can be taught in theory but there are so many things that require
experience in laboratory work. I would like to thank Trygve for all help and
guidanceduringthetestsandconstructionoftheturbine.
The efficiency tests and a lot of work have been done in collaboration with
fellow student Remi Andre Stople. The end result of this master thesis would
probablynothavebeenasgoodwithoutRemi.IwouldliketothankRemiforall
thehoursspenttogetherworkingandsocializing.
Next I would like to thank Bjrn Wither Solemslie for all the hours of help
received on the Matlab programs developed and guidance in calibration and
uncertaintycalculation.
TorbjrnK.NielsenhasbeenmyProfessorduringthisthesis.Duringtheweekly
meetingsinthebeginningofthesemesterhewasverygoodinguidingmeand
Remiinwhattodonextandwhatweshouldlookfor.Havingtheopportunity
ofalwaysbeingabletocomeintohisofficeandalwaysgethelpisfrommyside
veryappreciated,thankyouTorbjrn.
BrdBrandst,JoarGrilstadandHalvorHaukvikalsodeservethanksforhelping
throughouttheproject.
LastbutnotleastIwouldliketothankmyfamilyforsupportingmeduringthe
entireprocess.
LarsFjrvold
Trondheim,January20,2012
ii
Abstract
The goal with this master thesis was to establish Hill diagrams and improve a
KaplanturbineintendedforuseinAfghanistan.
Theturbineefficiencyhasbeentestedinsetting1and2.Turbineefficiencyin
setting 3 and 4 could not be tested because the runner blades interfere with
the housing making it impossible to rotate the turbine. The efficiency was
tested with an effective pressure head ranging from 2 to 8 meters. Best
efficiencypointwasnotreachedbecauseoflimitationsinthetestrigmakingit
impossibletoreachalowereffectivehead.Thebestefficienciestestedinthe
two different settings are presented in the table below together with the
uncertainty in the actual test point. All tests are done according to the IEC
standardformodeltestingofhydraulicturbines.
Setting 1
Setting 2
Efficiency
76.4 1.57%
83.8 1.59%
Rotational speed
552 rpm
602 rpm
Effective head
2.25 m
2.72 m
The computational fluid dynamics (CFD) simulations done on the inlet bend
indicates that the bend should be rounded and flow controllers should be
extendedovertheentirebend.Thisshouldbeconsideredtogetamoreeven
velocitydistributionattheinletoftheguidevane.
An alternative placement of the lower bearing was designed but is discarded
becauseofthedisadvantagesthemodificationleadsto.Highwearduetosand
erosion on the seals causing high maintenance and costly stops makes the
solutionnotoptimalforuseinwaterwithhighsandcontent.
Therunnerbladedesignischeckedagainstthedesignprocedurepresentedby
ProfessorHermodBrekkeinPumperogTurbinerandfoundtobesatisfying.Itis
concluded that time should rather be spent on optimizing the inlet of the
turbine.
Fluctuationsinthemeasurementsmakeitnecessarytochangethemeasuring
equipment or search for error in the existing equipment before further tests
can be carried out. In order to be able to test in setting 3 and 4 the runner
needstobeplacedwhilethebladesarefixedinsetting4.
iii
iv
Sammendrag
Mlet med denne masteroppgaven var etablere Hill diagram for en
Kaplanturbin designet for bruk i Afghanistan og se p muligheten for
forbedringer.
Virkningsgradentilturbinenerblitttestetiinnstillingenogto.Turbinenharfire
innstillinger,innstillingtreogfirelotsegikketeste daturbinenkiltesegfasti
turbinhuset allerede ved innstilling tre. Virkningsgraden ble bestemt for
effektive fallhyder fra to til tte meter. Best punktet til turbinen ble ikke
fastsattdatestriggengjordedetumuligoppnlavenokfallhyder.Denbeste
virkningsgraden testet i hver innstilling er presentert i tabellen under med
usikkerheten i det aktuelle testpunktet. Alle testene gjennomfrt er
gjennomfrt i henhold til IEC standarden for modelltesting av
vannkraftturbiner.
Virkningsgrad
Omlpshastighet
Innstilling 1
Innstilling 2
76.4 1.56%
83.8 1.60%
552 o/m
602 o/m
Effektiv
fallhyde
2.25 m
2.72 m
Svingningerimlingenegjrdetndvendigskiftemleutstyretellerfinnehva
somskapersigningeneimlingenefrnyetesterkanutfres.Lpehjuletm
installeresmensskovleneersattiinnstillingfireforkunnetesteiinnstilling
treogfire.
vi
Contents
Preface..........................................................................................................................i
Abstract......................................................................................................................iii
Sammendrag...............................................................................................................v
Contents....................................................................................................................vii
ListofFigures..............................................................................................................xi
ListofTables.............................................................................................................xiv
Listofsymbols...........................................................................................................xv
1
Introduction.........................................................................................................1
Prefacestudy.......................................................................................................3
2.1
RelatedworkattheWaterpowerlaboratory.............................................6
2.2
DesignofaKaplanrunner...........................................................................6
2.3
Maindimensions.........................................................................................9
2.4
Potentialflow............................................................................................12
TestingofKaplanturbine..................................................................................15
3.1
Efficiencytest............................................................................................15
3.2
Cavitationtest...........................................................................................16
3.3
Runawayspeed.........................................................................................18
3.4
Calibration.................................................................................................18
3.5
Pressuregauge..........................................................................................19
3.6
Torquegauge.............................................................................................19
3.7
Tripmeter..................................................................................................21
3.8
Flowmeter................................................................................................22
3.9
Frequencyanalysis....................................................................................23
vii
3.10
LabViewprogram......................................................................................23
3.11
Clearancewatertest.................................................................................23
3.12
RiskAssessment........................................................................................23
Uncertaintiesinmeasuring...............................................................................25
4.1
Spuriouserror............................................................................................25
4.2
Randomerror............................................................................................25
4.3
Systematicerror........................................................................................25
4.4
Totaluncertainty.......................................................................................27
4.5
UncertaintiesintheCalibration................................................................27
4.5.1
Uncertaintiesinthecalibrationoftheflowmeter...........................27
4.5.2
Uncertaintiesinthecalibrationofthepressuregauge.....................29
4.5.3
Uncertaintiesinthecalibrationofthetorquegauge........................31
4.5.4
Uncertaintyincalibrationofthethermometer................................32
4.6
4.6.1
Generaluncertaintyinthetests........................................................32
4.6.2
Uncertaintyinthepressuremeasurements......................................34
4.6.3
Uncertaintyinthetorquemeasurements.........................................35
4.6.4
Uncertaintyinthevolumeflowmeasurements...............................36
4.6.5
Uncertaintyintherotationalspeedmeasurements.........................36
4.6.6
Uncertaintyinthecalculationofdensityofwater............................37
4.6.7
Totaluncertaintyinthehydraulicefficiency.....................................37
Testrigsetup.....................................................................................................39
5.1
Uncertaintiesinthetests..........................................................................32
Detaileddescriptionoftherig...................................................................39
TheAfghaniKaplanturbine...............................................................................41
6.1
TheTurbinedesign....................................................................................41
viii
6.2
Specifications.............................................................................................42
6.3
Maindimensions.......................................................................................43
6.4
Runnerbladedesign..................................................................................43
Changesandlimitationsontherigandturbine................................................45
7.1
Pipedimensions........................................................................................45
7.2
Runnerbladefriction.................................................................................47
7.3
Upperbearing............................................................................................48
7.3.1
7.4
8
Bearingloadcalculation....................................................................51
Plexiglasscover..........................................................................................52
OptimisationsofinletbendusingCFD..............................................................55
8.1
CFDanalysisofinletbend.........................................................................55
8.2
Velocitymeasurementsininletbend........................................................58
8.3
Outlet.........................................................................................................61
Results...............................................................................................................63
9.1
Efficiencytests...........................................................................................63
9.1.1
Setting1.............................................................................................66
9.1.2
Setting2.............................................................................................68
9.2
Cavitationtests..........................................................................................70
9.3
Clearancewater.........................................................................................70
9.4
Mechanicalpower.....................................................................................71
9.5
Torque.......................................................................................................72
9.6
Fluctuationsinmeasurements..................................................................72
9.7
CFDofinletbend.......................................................................................73
ix
9.8
10
Velocitymeasurements.............................................................................77
Discussionofresults......................................................................................79
10.1
Efficiencytests...........................................................................................79
10.2
Clearancewater.........................................................................................79
10.3
Inletbend..................................................................................................79
10.4
Outlet.........................................................................................................81
11
Conclusion.....................................................................................................83
12
Furtherwork..................................................................................................85
12.1
13
Rigsetup....................................................................................................85
Bibliography...................................................................................................87
List of Figures
Figure21CrosssectionofaKaplanturbine,Voithhydro...........................................3
Figure22EfficiencycurvesforKaplanturbines.........................................................4
Figure23Sturbine.....................................................................................................5
Figure24Bulbturbine................................................................................................5
Figure25Vectorcomponentsofforcesactingonarunnerblade............................8
Figure26AxialcutofaKaplanturbinewithguidevanes..........................................9
Figure27Speednumberplottedvs.axialspeed.....................................................10
Figure28Relationbetweend/DandB0/Dplottedagainstns................................11
Figure29Suctionheadandnumberofblades........................................................12
Figure31Definitionofheights.................................................................................17
Figure32EfficiencycurvewithchangingThomanumber.......................................17
Figure33Thedeadweightcalibrationsetup...........................................................20
Figure34Calibrationcurveforthetorquegauge....................................................21
Figure35Tripmeterandreflexribbonreadytouse...............................................21
Figure36Theflowmeteruseinthetest.................................................................22
Figure41Calibrationcurvewitha95%confidenceinterval...................................29
Figure 42 Calibration curve for the pressure gauge with a 95% confidence
interval......................................................................................................................30
Figure 43 Calibration curve for the torque gauge with a 95% confidence
interval......................................................................................................................32
Figure44Testfordriftinmeasurements300rpm,setting1...................................33
Figure45Testfordriftinmeasurements300rpm,setting1...................................34
Figure51Testrigsetup............................................................................................40
Figure61Completeturbine,(Inventordrawing).....................................................41
Figure71Originalpipesection................................................................................45
Figure72Configuredpipesection...........................................................................46
Figure73Runnerbladewithbrokenbladesection,seenfrombelow....................48
Figure74Temperatureplottedvs.rpmat2.5meterpressurehead.Setting2......50
Figure75Temperatureplottedvs.pressurehead.Setting1,800rpm....................50
Figure76Temperatureplottedvs.pressurehead.Setting2,650rpm....................50
Figure77FlowaftertheturbineoutsideBEP..........................................................53
Figure81Velocityandpressuredistributioninpipebend......................................56
Figure82Originalgeometry....................................................................................57
Figure83Geometry1..............................................................................................57
Figure84Geometry2..............................................................................................57
xi
Figure85Geometry3..............................................................................................58
Figure86Pitottubemountedontheturbine..........................................................59
Figure87HeightdifferenceinaknifePitotmeasurement......................................60
Figure88Bearingwhenplacedinsidetheguidevanecentrepiece........................62
Figure912ndorderpolyfit.....................................................................................63
Figure923rdorderpolyfit......................................................................................64
Figure934thorderpolyfit......................................................................................64
Figure94Smoothingspline......................................................................................65
Figure95Hilldiagram,efficiencyplottedagainsteffectiveheadandrotational
speed.........................................................................................................................66
Figure96Rawdatameasurements.EfficiencycurvesatconstantRPMplotted
vs.effectivehead......................................................................................................68
Figure97Hilldiagram,efficiencyplottedagainsteffectiveheadandrotational
speed.........................................................................................................................69
Figure98Rawdatameasurements.EfficiencycurvesatconstantRPMplotted
vs.effectivehead......................................................................................................69
Figure99Clearancewateratconstantrpm.............................................................70
Figure910Clearancewateratconstantinletpressureheadinsetting2...............71
Figure911Mechanicalpowerat500rpm................................................................72
Figure912Torqueat500rpm..................................................................................72
Figure913Originalgeometrywithinletvelocityof1.0814m/s..............................74
Figure914Geometry1.Inletvelocity1.25m/s........................................................75
Figure915Geometry2.Inletvelocity2.387m/s......................................................75
Figure 916 Outlet velocity profiles with original geometry and different inlet
velocities....................................................................................................................76
Figur917Outletvelocityprofilesforgeometry3....................................................77
Figure131Inflationalongwalloftheinlet...............................................................R
Figure132Frontpanel...............................................................................................T
Figure133Dataacquisitionandtranslationofvoltsignals.......................................T
Figure134RPMsubVIforreadingsoftherotationalspeed.....................................U
Figure135BlockdiagramfortheRPMsubVI...........................................................V
Figure136Calculationnodeandfrontpanelvalues................................................V
Figure137Storage..................................................................................................W
Figure138Plottingofefficiencyheadgraph...........................................................W
xii
xiii
List of Tables
Table1Componenterrors........................................................................................27
Table2Uncertaintiesinthecalibrationoftheflowmeter.......................................28
Table3Uncertaintiesinthecalibrationofthepressuregauge................................30
Table4Uncertaintiesinthecalibrationofthetorquegauge...................................31
Table5Errorsinthetests..........................................................................................34
Table6Uncertaintiesinthetorquemeasurements.................................................36
Table7Uncertaintiesinthevolumeflowmeasurements........................................36
Table8Volumeflow..................................................................................................42
Table9Mainturbinecharacteristics.........................................................................43
Table10Actualandscaledoutletvelocity................................................................77
Table11Scaledvelocityvs.averagevelocity............................................................78
Table12Velocitymeasurementsatconstantpressurehead...................................78
Table13Weightcalibrationcalculation.....................................................................N
xiv
List of symbols
Symbol
Description
Unit
Lift
[N]
CL
Liftcoefficient
[]
Density
[kg/m3]
Velocity
[m/s]
Length
[m]
Drag
[N]
CD
Dragcoefficient
[]
Force
[N]
Turbinevelocity
Availablepowerinwater [W]
Gravityconstant
[m/s2]
Volumeflow
[m3/s]
Head
[m]
Efficiency
[%]
Numberofrunnerblades []
Distance
between [m]
runnerblades
xv
[m/s]
Speednumber
[]
Reducedangularvelocity [m1]
cm
Axialvelocityofwater
[m/s]
Turbinediameter
[m]
Hydraulicefficiency
[%]
Pm
MechanicalPower
[W]
Ph
HydraulicPower
[W]
Mechanicalefficiency
[%]
Tm
Torque
[Nm]
Specificenergy
[J]
Pressure
[Pa]
Height
[m]
Thomanumber
[]
qv
Volumeflow
[m3/s]
Massofwater
[kg]
f
sy
Standarddeviation
ey
B
Absolut
error
measurements
Arclength
Radius
[m]
Angle
[0]
xvi
[Varies]
in [Varies]
[m]
Measuredflowvelocity
[m/s]
Scaledflowvelocity
[m/s]
Heightdifference
[m]
Pitotcoefficient
[]
Re
Reynoldsnumber
[]
Dynamicviscosity
[kg/(ms)]
xvii
xviii
1 Introduction
Remote Hydrolight is a company designing turbines that aims to be cost
efficient and easy to produce. Remote Hydrolight represented by Anders
Austegr wanted to establish Hill diagrams for four runner vane settings on a
Kaplanturbinetheyhadproduced.
Austegrdalsorequestedcavitationandrunawayspeedtests.Iftimeallowedit
he also wanted suggestions on how the turbine could be improved. The
improvementsshouldnotmaketheturbinemoredifficulttoproduce.
2 Preface study
The Kaplan turbine was invented and developed by Austrian Victor Kaplan
around1913,andisdesignedtooperateatlowheadsandhighflowrates.The
turbine is an axial turbine, meaning that the direction of the water flow is
paralleltothebulbanddriveshaftthroughtherunnerblades.Itiscommonto
comparetheKaplanturbinewithapropellerduetoitsdistinctshape.
Figure21CrosssectionofaKaplanturbine,Voithhydro
VerticalaxisKaplanturbinesareinmanywayssimilartoFrancisturbinesaswell
aspropellers.Besidestheshapeoftherunnerblades,theKaplanturbineuses
the same water way system and method to generate electricity as Francis
turbines. Like the Francis turbine the Kaplan have a spiral casing to distribute
the water around the turbine. Guide vanes are used to regulate the volume
flowthroughtheturbine.Theguidevanesarealsousedtoinduceaswirlinthe
water, so that the water hits the runner blades in the most efficient angle as
possible.Therunnerbladescaninmanycasesalsobeadjustedtomaintainas
3
Figure22EfficiencycurvesforKaplanturbines
Kaplanturbineshavealsotheopportunitytobemountedwithahorizontalaxis.
TheyarethenoftenreferredtoasSturbinesandBulbturbines.Sturbinesare
usedinthesamespectreofheadandflowratesasverticalaxisKaplans.Price
andavailablespacearefactorsthatgovernthechoicebetweenSturbinesand
verticalaxisKaplans.VerticalaxisKaplansrequiresasmallerlandareathanS
turbines. Sturbines have smaller hydraulic losses compared to a vertical axis
Kaplans, due to the fact that the water does not have to change direction
throughtheturbine.
Figure23Sturbine
The bulb turbine is only used in high energy sites with low head and high
volume flow. On the bulb turbine the generator and drive shaft is mounted
insidethebulbinfrontoftherunner.Afullgrownmanisabletostandupright
insidethebulb.
Figure24Bulbturbine
LiketheverticalaxisKaplan,Sturbinesandbulbturbineshaveguidevanesand
runnerbladeswiththepossibilitytobeadjustable.Theyallhaveadrafttubeto
regainpressureaftertherunner.Itisnormalforallthesolutionstobedivedto
preventcavitationattheendoftherunnerblades.
L CL
v2
l dr [ N ]
2
(2.1)
Equation(2.1)andequation(2.2)givesustheliftanddragforce.Tobeableto
calculate the forces the lift and drag coefficients have to be known. Lift and
drag coefficients can be found using certain programs, such as xfoil or it is
possibletofindliftanddragcurvesforcertainNACAandGttingenfoils.Toget
asaccuratecoefficientsaspossiblemodeltestsarenecessary.Especiallywhen
the blades are mounted in a cascade, which is the case for a Kaplan turbine,
model tests are important to calculate lift and drag coefficients. It is on the
otherhandpossibletousexfoilandcorrecttheresultsfromtheprogramwith
testdoneonothercascadesectionstogetreasonablecoefficients.Athickair
foilhaveagoodpeakperformancewhileamoreslenderairfoilhaveawider
spectre with high performance, but peak performance for a slender air foil is
not as good compared with a thicker air foil. Since a thick air foil have a high
peak performance they are also have a larger risk for cavitation and the
efficiencyfallsdrasticallyoutsidebestangleofattack.
D CD
v2
l dr [ N ]
2
(2.2)
When lift and drag is found, we can find the force acting in the rotational
direction.
Fu F cos( ) F sin( )[ N ]
(2.3)
2
Figure25Vectorcomponentsofforcesactingonarunnerblade
L
C v2
L
l dr
cos
2 cos
[N ]
(2.4)
AfterFuis found the energy output from the turbine can be determined. By
taking the sum of all forces acting on each small section on each blade, and
multiplythesumwiththerotationalspeedoftheturbine,theenergyoutputis
determined.
P Fu u
C L v2
sin( ) u dr W (2.5)
2 cos
Theenergytheturbineproducescannotbehigherthantheavailableenergyin
thewaterpassingtheturbine.Asimplewaytocalculatetheavailableenergyin
thewaterisbyusingequation(2.6).Hererepresentlossesinthewaterway,
valvesandotherfactorsthatreducetheeffectivehead.
P ' g Q H g 2 r dr
cz
H
z1
[W ] (2.6)
Figure26AxialcutofaKaplanturbinewithguidevanes
2 r
z1
[m] (2.7)
l 2 g H n cz cos
CL 2
t
v u sin( )
[] (2.8)
Qn (2.9)
Theaxialflowisdependentonhowfasttheturbineisrotating,socmisfound
from if experienced values if minor simplifications and linear relations are
accepted.InFigure27thelineardependenceandequationisshown.
Figure27Speednumberplottedvs.axialspeed
Fromthespeednumberthelargestrunnerdiameterisfoundthrough
D2
4 0 2
2 0.12 0.18 0
[m] (2.10)
SmallKaplanturbinescanbemadewithacylindricalhousingtoreducethecost
of manufacturing, but cylindrical housing lead to increased clearance loss
betweentherunnerbladesandthehousing.Whenasphericalhousingisused
the narrowest part on the draft tube is produced with a diameter 35%
narrowerthanD2.
10
Figure28Relationbetweend/DandB0/Dplottedagainstns
The diameter of the hub d and the height of the intake B0 also need to be
calculated.BothofthesetwoparameterscanbereaddirectlyofFigure28.B0
anddaredependentonthespecificrevolutionnumberns.
Number of runner blades and suction head is dependent on the pressure
distribution around the runner blade. Number of blades and suction head is
chosensothatthepressurearoundthebladeforawiderangeaspossiblenot
falls under the boiling pressure or under the critical cavitation number. The
pressure distribution around blades changes when blades are placed in a
cascadelikeinarunner.Suctionheadandnumberofbladescanbefoundby
using Figure 29 Suction head and number of blades and is a function of the
specificrevolutionnumber(1).
11
Figure29Suctionheadandnumberofblades
D
divV 0 (2.11)
Dt
TheNavierStokesequationdescribesthemotionofafluidsubstance.Itgives
thevelocityofafluidparticleatagivenpointinaflowatagiventime.When
12
NavierStokesisappliedtoaseriesofpointsinaflowoneobtainaflowvelocity
field.Afterthevelocityfieldisfoundthedragforcemaybefound.
NavierStokesequation:
DV
i j
g p
Dt
x j x j xi
ij divV (2.12)
Iftheflowisincompressibleandviscosityanddensityisconstantweget:
DV
g p 2V (2.13)
Dt
Whenfutherassumingthattheviscoustermsarenegligibleequation(2.13)is
reducedtotheEulerequationforinviscidflow.
Eulerequationforinviscidflow:
DV
g p (2.14)
Dt
The Euler equation for steady, incompressible and frictionless flow along a
streamlinebetweentwopoints1and2becomestheBernoulliequation.
13
14
Pm
(3.1)
Ph
P
(3.2)
Pm
Mechanicalefficiency:
Efficiency:
h m
P
(3.3)
Ph
To calculate the efficiency of the turbine torque (Tm), rotational speed (n),
volumeflow(Q)andpressurehastobemeasured.Alsothegravityconstant(g)
and the density () of the water have to be calculated in order to find the
efficiency.Themechanicalpoweriscalculatedwithequation(3.4).
Pm 2 n Tm
15
[W ] (3.4)
Ph E ( Q )1
pabs 1 pabs 2
[W ] (3.5)
v12 v22
( z1 z2 ) g
2
[ J ] (3.6)
Index2referstotheoutletandlowpressuresideoftheturbine(2),(3).
NPSE
(3.7)
E
In hydro power cavitation can be a big problem and lower the efficiency by
severalpercent.Lowpressurezonesarecreatedintheoutletofturbineswith
drafttubesandthatusepressuredifferencestoproduceenergy.
The Net Positive Suction specific Energy is defined according to the IEC
standard(3)asfollows:
p
C2 p
J
NPSE 2 gZ 2 2 v gZ ref [ ] (3.8)
2
kg
16
Figure31Definitionofheights
Bystudyingequation(3.8)and(3.6)itbecomesclearthattheThomanumberis
simplyrelatedtotheheightdifferencehsalsoreferredtoasthedivingofthe
turbine.
WhentestingforcavitationtheIECstandardsuggeststhatthespecificenergy
coefficient, see equation(3.9), is kept constant while the Thoma number is
slowlyreducedwhiletheefficiencyislogged.
Figure32EfficiencycurvewithchangingThomanumber
When lowering the Thoma number the efficiency will keep constant until it
reaches 0 is reached. Here the efficiency will increase because of the
17
lubricationeffectthecavitationbubbleshaveontherunnerbladesbeforethe
efficiencyfallsdrastically.
2 NPSE
(3.9)
2 R2
3.4 Calibration
Calibration of measuring equipment is done to minimize error in
measurements. Calibration of measuring equipment is done according to the
IEC(3).Whencalibratingequipmentthemeasuringrangeoftheequipmenthas
tobeknownandtheoperatingrangeoftheturbinehastobeknown.
The equipment is calibrated by measuring a number of values outside and
inside the range the actual testing will be carried out in against a known
quantity. The values measured are given as a Volt signal that has to be
correlated with the actual unit measured. By measuring values for the entire
range linear regression is used to find a calibration line that the Volt signal is
calibrated against to minimize the error. By increasing the number of values
measuredduringcalibrationintheupperandlowerrangetheuncertaintiesin
theequipmentisreduced.
18
The equipment calibrated in this case is pressure gauges, a flow meter and a
torquegauge.
19
Figure33Thedeadweightcalibrationsetup
Whencalibratingthetorquegaugethemeasurementstendedtodeviatefrom
eachotherevenwhenthesamedeadweightswereused.Thiswasdiscovered
when the measured value kept almost constant when weights were added or
removed from the weight bed. It was quickly discovered that this was caused
byfrictionbetweentherunnerbladesandthehousingandfixedbyrunningthe
turbine until there was clearance between the blades and the housing. To
secure that there were no friction affecting the measurements, values were
measuredfirstbyaddingweightsintheentirecalibrationrangeandthenvalues
weremeasuredwhenremovingweights.
When the weights were added to the weight bed the measured volt value
always was lower than the same volt value measured for the same weight
whenweightswereremovedfromtheweightbed.InFigure34thecalibration
curveandtheactualvaluesmeasuredcanbeseen.Thisphenomenoniscalled
hysteresis and is present in ferromagnetic materials. Ferromagnetic materials
havememoryduetomagneticpropertiesinthematerial(4).Hysteresisisa
commonphenomenonintorquemeasurements.
20
Figure34Calibrationcurveforthetorquegauge
The calibration was preformed two times both times resulting in the same
valuesandhysteresisinthemeasurementswasproven.Thecalibrationreport
ofthetorquegaugeisfoundinAppendixQ
Figure35Tripmeterandreflexribbonreadytouse
ThetripmeterismountedontheturbineshaftasdisplayedinFigure35Trip
meterandreflexribbonreadytouse.
21
qv
m2 m1
1 (3.10)
t
1 1
(3.11)
p
Thelastterminequation(3.10)isthecorrectionfactorforcorrectiondifferent
in buoyancy exerted from the atmosphere on the measured water and the
weightsusedduringcalibrationoftheweighingtank.a,pandisrespectively
thedensityoftheatmosphere,standardweightsandwater.
Figure36Theflowmeteruseinthetest
ALabViewprogramcreatedbyRemiAndreStoplewasusedtologthevolume
flow.Theprogramlogged1000valuesfromtheflowmetereverysecond.The
regression line was found using built in functions in excel while uncertainties
22
23
24
4 Uncertainties in measuring
All measurements have uncertainties attached to them. Uncertainties can be
related to three types of error in measurements, error caused by spurious,
randomorsystematicerror.
25
Thetotalsystematicerrorcanbefoundbycombiningthesystematicerrorfrom
eachmeasuringdevicewiththerootsumsquaremethod.Whencalculatingthe
uncertaintyconnectedtothemeasurementaconfidenceintervalof95%should
beusedaccordingtothe standard.Therootsumsquaremethodisexpressed
inequation(4.1).
f s fY (4.1)
Here fs is the combined uncertainty of each device in the system where the
uncertaintyineachcomponentisfY.Theuncertaintycaneitherbeknownfrom
the manufacturer of each device or it can be calculated. To calculate the
uncertaintynnumberofmeasurementsistakenataconstantoperationpoint
andthestandarddeviationisfoundwithequation(4.2)(6)and(7).
n
sY
Y
r 1
Yr
n 1
(4.2)
Yr =meanvalueofmeasurements
Yr =valueofrthmeasurement
n =numberofmeasurements
After the standard deviation is calculated the uncertainty in the standard
deviation can be calculated. Since the error have a normal distribution but
therearenotaninfinitenumberofmeasurementsstudenttdistributioncanbe
assumed.StudenttvaluescanbefoundintableL2inIEC60193(3).
eY
t sY
(4.3)
n
fY
eY
(4.4)
Y
26
ft f r2 f s2 (4.5)
Description
Systematicerroroftheprimarycalibrationmethod
fb
Randomerroroftheprimarycalibrationmethod
fc
Systematicerrorofthesecondaryequipment
fd
Randomerrorofthesecondaryequipment
fe
Physicalphenomenaandexternalinfluences
ff
Errorinphysicalproperties
Table1Componenterrors
27
Uncertainty
fQ ,a
f Q ,b
f Q , regression
Description
Systematic error in the
weighing tank system
Random error in the
weighing tank system
Systematic
and
random error in the
instrument
Value
0.072104%
0.0565366%
0.03404%
Table2Uncertaintiesinthecalibrationoftheflowmeter
The calibration curve with a confidence interval of 95% for the volume flow
meterisshowninFigure41.TheMatlabprogramsusedarefoundinAppendix
C.
28
Figure41Calibrationcurvewitha95%confidenceinterval
29
The calibration curve for the pressure gauge is shown in Figure 42. The
calibrationreportcanbefoundinAppendixL.
Figure42Calibrationcurveforthepressuregaugewitha95%confidenceinterval
ewhichistheerrorcausedbyphysicalphenomenaandexternalinfluencescan
for the pressure gauge be temperature changes within the instrument. Since
thepressuregaugehadbeenlocatedinthewaterpowerlaboratorymanydays
before the calibration was conducted the temperature within the instrument
was assumed to be in equilibrium with the surrounding temperatures. e was
thereforeassumedneglectable.
p,f,theerrorinphysicalpropertiesisforthecalibrationofthepressuregauge
assumed to be neglected. This is because the pressure gauge was calibrated
against air and the height difference between the pressure gauge and the
electronicallyinstrumentdonotcontributetopressuredifferences.
Uncertainty
p,a
p,regression
p,f
Description
Systematic error in the
electronic
pressure
calibrator
Systematic and random
error in the instrument
Systematic error in the
positioning of the
pressure gauge.
Value
0.010%
0.066751%
0.000%
Table3Uncertaintiesinthecalibrationofthepressuregauge
30
(4.7)
Description
Systematic error in
weights and the weight
bed
Systematic error in the
length of the arm
Systematic and random
error in the instrument
Value
0.0114325%
0.2%
1.235138%
Table4Uncertaintiesinthecalibrationofthetorquegauge
BycombiningthegivenerrorswiththeRSSmethodtherelativeuncertaintyfor
thecalibrationinbestefficiencypointforthetorquegaugewasfoundtobe
InFigure43thecalibrationcurveforthetorquegaugeispresented.Asseenin
thefigurethecalibrationisaffectedbehysteresis.Hysteresisisaphenomenon
whichisnormalfortorquegauges.Themeasuredvaluehasamemoryfrom
the previous measurement. This means that if for instance the torque gauge
measuresahighervalueforonegivenvalueifthepreviousmeasurementhada
highervaluethanifthepreviousmeasuredvaluewaslower.
31
Figure43Calibrationcurveforthetorquegaugewitha95%confidenceinterval
InFigure43youcanseethehysteresisbecausetherawdataseemstobegiven
in pairs. The actual case is that the raw data with the lowest volt value (here
low value is less negative than a high value) is measured with adding more
weights,hencefromalowervalue,andtherawdatawiththehighestvoltvalue
ismeasuredfromahighervalue.
of the runner blades when setting the position after it had been tested in
settingtwocausingthebladestohaveasteeperangleofattackinthelastrun
comparedtothefirstrun.
Comparisonoffirstandlastrun,300rpm
70
Firstrun
65
Lastrun
Efficiency
60
55
50
45
40
35
30
0
Pressurehead
Figure44Testfordriftinmeasurements300rpm,setting1
33
Volumeflow,300rpm
250
Volumeflow
200
150
100
50
Firstrun
Lastrun
0
0
Pressurehead
Figure45Testfordriftinmeasurements300rpm,setting1
Error
cal
h
j
ks
kr
l
Description
Systematic error in the calibration
Additional systematic error in the instrument
Error in physical properties
Systematic error due to physical phenomena and
external influences
Random error due to physical phenomena and external
influences
Random error in repeatability of secondary equipment
Table5Errorsinthetests
hispresentedaboveTable5andcanbepresentinallthemeasuringdevises.
But for the tests conducted in this report drift is not present and the
uncertaintyisthereforeneglectedinthefollowingcalculations.
pressure transducer was calibrated against air but during tests the operating
fluidiswater.Thiscreatesanuncertaintyinthemeasuredinletpressurevalue.
The systematic error due to an offset of 4.65cm in the pressure transducer
placement relative to the centre of the pipe is denoted as fp1,offset. The
systematic uncertainty is calculated to be with an offset of 4.65cm while the
error in the ruler used was 0.2cm which give an fp1,offset of 0.7853% for the
pressure measured at the given operation conditions of 500rpm and
0.25468meterpressurehead.
p1,l is calculated with a studentt confidence interval on the measured data
from the tests. A Matlab program was created to do the calculations and is
foundinAppendixC.Therandomuncertaintyfortestseries500rpminsetting2
wascalculatedtobe10.6552%.Thisvalueisextremelyhighandiscausedby
severe fluctuations in the pressure readings. This will be discussed in chapter
9.6.
The total relative uncertainty in the pressure measurements was calculated
withtheRRSmethodtobe
(4.9)
Itisclearthattherelativeuncertaintyintheoutletpressuremeasurementsis
muchhigherthantherelativeuncertaintyintheinletpressuremeasurements.
Thisisbecauseoftheequipmentusedtocalculatethepressure.Whiletheinlet
pressure is calculated with costly equipment the outlet pressure is measured
withcheaphomemadeequipmentwithahighlevelofuncertainty.
35
Uncertainty
,cal
Description
Systematic error in the
calibration
,l
Value
1.25128%
0.0312%
Table6Uncertaintiesinthetorquemeasurements
BycombiningthetwouncertaintieswiththeRRSmethodatotaluncertaintyin
thetorquemeasurementsisfound:
It is clear that the uncertainty linked to the calibration of the torque gauge is
dominantinthetestuncertainty.
Description
Systematic error in the
calibration
Q,l
Value
0.097745%
0.0394%
Table7Uncertaintiesinthevolumeflowmeasurements
WhencombinedtheQ,landQ,calgiveatotaluncertaintyinthevolumeflowof:
randomuncertaintyintherpmmeasurementswascalculatedtobe0.0363%.
Thisgivesatotaluncertaintyintherotationalmeasurementsof
2
f rot f IEC
f rot2 ,l 0.061787% (4.12)
f h
(e h )
f f f
2
(4.13)
Tofindthedifferentuncertaintiesinequation(4.13)theequationsfollowingis
used and explained. Q is found in chapter 4.6.4. Uncertainties in the density
andthegravitationconstantareassumedtohaveasosmallvaluethattheycan
beneglected.Uncertaintyingis0.01milliGalstatedinthecalibrationreportof
measuredgravityinthelaboratory.
2
ev12
2
2
e p1 ep 2
g
e
g
e
z1
z2
eE
2
fE
p1 p2
v 2 v22
E
g ( z1 z2 ) 1
ev22
2
(4.14)
The uncertainty in energy in equation(4.14) is given as absolute uncertainties
whileinthisthesisrelativeuncertaintiesareused.Inequation(4.15),(4.16)and
(4.17)therelationbetweenabsoluteandrelativeuncertaintyisgiven.
37
e p p f p (4.15)
ev2
v 2 fv (4.16)
e
f v fQ2 2 r (4.17)
r
Nexttheuncertaintyinpressureisfoundbyusingequation(4.18).Whenthisis
donethetotaluncertaintycanbecalculatedusingequation(4.13).
f P ( f ) 2 ( f ) 2 (4.18)
f f rot (4.19)
Thetotalrelativeuncertaintyinthehydraulicefficiencyiscalculatedwithdata
from setting 2 at 500rpm. The reason for choosing values from 500rpm and
setting2isbecausethisisthesettingclosesttothebestdesignsettingwhichis
accordingtoAndersAustgrdsetting3and490rpm.Foradetailedcalculation
ofthetotalrelativeuncertaintyseeAppendixA.
Thetotalrelativeuncertaintyinthehydraulicefficiencyiscalculatedtobe
h=2.2145%
38
Therigiscontrolledandrunwithonelaptopwhichcontrolsthepumpandapc
thatlogdataandaboxthatcontrolstherotationalspeedofthegenerator.This
control station is located next to the turbine. The pumps are originally
controlled from the control room located on the second floor in the water
powerlaboratory.TheKaplanturbineandthecontrolstationarelocatedonthe
firstfloor.Remotedesktopenablescontrolofthepumpsviathelaptopinthe
controlstation.
During calibration of the volume flow meter the 200mm pipe is connected
directlytothe600mmpipethatleadstotheweighttank.Thisimpliesthatthe
watercircuitdonotrunthoughtheturbine.
Figure51Testrigsetup
The under water level isdetermined with the use of afloating cylinder which
follows the under water level. The floater is placed some distance from the
outlettonotbedisturbedbyairinthewatercomingoutofthedrafttube.
Thetripmeterismountedonthetopoftheturbinewhilethereflexribbonis
placedontheshaft.
Thetorquegaugeisplacedinthemiddleoftheturbineandthegeneratoron
theshaft.
40
Flowcontrollers
Guidevanes
Runner
Lower bearing with
airandgreaseintake.
Figure61Completeturbine,(Inventordrawing)
The inlet pipe has a 90 degree bend in front of the guide vanes, this is
unfortunate and can cause turbulent and a chaotic flow right in front of the
41
6.2 Specifications
According to Anders Austegrd the turbine is designed for an effective head
rangingfrom1.5metersto6metersandavolumeflowfrom0.09to0.41m3/s.
Hehasalsogivenanestimatedflowtableandestimatedthecavitationnumber
tobe0.7.
Head [m]
1.5
2
3
4
6
8
10
0.09
0.11
0.13
0.15
0.18
0.21
0.23
0.18
0.21
0.26
0.30
0.36
0.42
0.47
Table8Volumeflow
42
0.21
0.24
0.29
0.34
0.41
0.47
0.53
The minimum volume flow is estimated volume flow when the runner blades
are fixed in setting 1, maximum volume flow is in setting 4 while optimal
volumeflowisobtainedinsetting3.
44
Figure71Originalpipesection
Whenthecalibrationstarteditquicklybecameobviousthatthelargediameter
reduction created heavy cavitation. This was problematic because the flow
45
meter is sensitive to air bubbles and noise. Cavitation creates both of these
problems.
Tobeabletocontinuethecalibrationteststheareareductionhadtobedone
smoothertoavoidcavitation.Therewerenoavailablepipeconesavailablewith
therightdiametersinthewaterpowerlaboratory.Anewconethereforhadto
bemadereducingthediameterfrom600mmdownto300mm.Aconereducing
thediameterfrom300mmto200mmwasavailable.
Platesteelwithathicknessof3mmwasusedtocreatethecone.Equation(7.1)
withr1,r2,B1andB2wasusedtocalculatethedimensionsofthecone.Bisthe
arclengththathadtobecuttogivethegivenradius,r.
B1 r1
1800
B2 r2
1800
(7.1)
The finished cone mounted in the rig setup is shown in Figure 72. After the
installation of the new cone the pressure head to be able to reach desired
volumeflowwasdrasticallyreducedandcavitationwaseliminatedinthispipe
section.
Figure72Configuredpipesection
Eventhoughthecavitationwaseliminatedfromthefirstpipesectionwithan
area change cavitation still occurred during tests in another pipe section. The
46
inlet on the turbine has a pipe diameter of 400mm. Since the pipe diameter
from the first area reduction is 200mm it demands a new area change to be
able to connect the two diameters together. This area expansion is done
immediatelyaftera90degreebendafewmetersinfrontoftheturbine.The
cavitationbecamesosevereinsettingtwothattheinletpressurecouldnotbe
increased as desired. The cavitation bubbles created in the expansion got so
severe that they entered the turbine, making the tests invalid. Some of the
measuring series preformed in setting two is therefor stopped before a
pressureheadof6metersisreached.
47
Figure73Runnerbladewithbrokenbladesection,seenfrombelow
AsseeninFigure73therearandoftherunnerbladeistouchingthehousing
markedwithareadcircle.Bylookingcloselyitispossibletoseethatapieceof
thebladeisbrokenoffinsidetheredcircle.Whiletherearendisjammedinto
thehousingthefrontparthasalargegapbetweenthebladeandthehousing
marked with a green circle. In the area of the yellow circle erosion from the
bladeonthehousingcanbeseen.
An attempt to adjust the blades to setting three was made but was not
managed.A pieceofone ofthebladesbrokeoffin theattempt toadjustthe
blades into setting three. This made it impossible to preform test in setting
threeandfour.
49
Temperature
50
45
40
35
30
25
300
400
500
600
700
800
rpm
Temperature
Figure74Temperatureplottedvs.rpmat2.5meterpressurehead.Setting2
57
52
47
42
37
0
Pressurehead[m]
Figure75Temperatureplottedvs.pressurehead.Setting1,800rpm
Temperature
52
47
42
37
0
Pressurehead[m]
Figure76Temperatureplottedvs.pressurehead.Setting2,650rpm
As seen in the three figures above the temperature increase with revolutions
and pressure head. In the figures where temperature is plotted vs. pressure
head the revolutions per second is kept constant and when temperature is
plotted vs. rpm the pressure head is kept constant. The operating conditions
50
werekeptconstantineachmeasuringpointuntilthetemperaturewasalmost
constant.
These measurements were done to find if there was an upper limit for the
temperature increase and where it may be. It is clear that the temperature
wouldincreaseifthepressureornumberofrevolutionswereincreased.
Sincethetemperaturealsoincreasewithpressureheadthebearingmostlikely
heatsupduetofrictioninthebearingitself.
The situation was discussed with professor Ole Gunnar Dalhaug, what should
be done regarding the temperature to be able to continue the tests. To
dismount the entire turbine, change the bearing and straighten the shaft was
consideredbywasfoundtobewaytotimeconsuming.Theconclusionwasto
constantly measure the temperature on the bearing house and stop
measurements if the temperature approached temperatures that could cause
the bearing to malfunction or jam. The critical value of the temperature was
assumedtobearound600CbasedonpreviousexperiencesoftechnicianTrygve
Oplandand engineerBrdBrandst.Thesetemperaturesettclearboundaries
forthetestsconducted.
P Fa 1, 2 Fr (7.2)
51
Staticequivalentaxialloadisahypotheticalloadwhichwouldcausepermanent
deformationonthebearingatthepointonthebearingundermoststresswhen
both axial and radial force is applied. For a spherical roller bearing the static
equivalentaxialloadiscalculatedwithequation(7.3)providedthatFr/Fa0.55.
P0 a Fa 2.7 Fr
[ N ] (7.3)
P0aisthestaticequivalentaxialload,FaandFrareactualaxialloadandactual
radialloadrespectively.
The operating life time of a spherical roller thrust bearing can be calculated
usingequation(7.4).
10/3
Lnm
C
a1askf
P
(7.4)
52
Figure77FlowaftertheturbineoutsideBEP
Figure77showshowtheflowlookslikeoutsidebestpoint.Asseentheflow
almostappearsasamistflowingthroughtheturbine.Airbubblesfromtheair
intakeinthelowerbearingandcavitationcontributetocreatethemist.The
mistmadeitimpossibletodocumentcavitationontherunnerbladesbecauseit
was impossible to see the blades. When the water had no spin the flow was
clearer hence less bubbles but it was still impossible to document cavitation
withahighspeedcamera.
53
54
55
Figure81Velocityandpressuredistributioninpipebend.
In Figure 81 the velocity and pressure distribution is shown right after the
bend.Thedifferenceinvelocityandpressuredistributionwillincreasewiththe
velocityintheflow.
Theturbinetestedinthisthesisisplacedafterasharpcornerbendasdescribed
above.ACFDanalysiswastherefore conductedtocheckifthebend could be
optimized. A wide range of different bend solutions were tested in the CFD
analysis and compared against the geometry of the actual inlet bend of the
Afghaniturbine.
The geometry of the actual inlet bend is complex and parts of the bend are
difficulttomeasure.Measuresandthegeometryusedtocreate themodelin
FLUENT are therefore based on measurements taken on the physical turbine
and measures found in the Inventor drawings created by Anders Austegrd.
ThereasonfornotbaseallmeasuresontheInventordrawingsisthatthereare
significantdifferencesinthedrawingsandtheactuallyproducedturbine.
TheintendedvolumeflowrangefortheKaplanturbineis0.09m3/sto0.53m3/s
whichgivesaninletvelocityinaverageof0.71619m/sand4.2176m/s.Optimal
designinletvelocityis1.6711m/s.CFDsimulationsofthedifferentgeometries
wasdoneatinletvelocitiesrangingfrom1.0814m/sto2.387m/s.
Four different geometries are tested, the original geometry plus three new
geometries.Thefourgeometriesarepresentedinthefiguresbelow.
56
Figure82Originalgeometry
Figure83Geometry1
Figure84Geometry2
57
Figure85Geometry3
InGeometry1thebendisroundedoneflowcontrollerisplacedinthemiddle
ofthebendspendingfromtheinlettotheoutletofthebend,thesectionafter
thebendisextendedwith50cmcomparedtotheoriginalgeometry.
Geometry2hasthesamedimensionsasgeometry1andtheonlydifferenceis
thatgeometry2hastwoflowcontrollersinthebend.
Geometry 3 has the same bend as geometry 2 but has the same length after
thebendastheoriginalgeometry.
58
Figure86Pitottubemountedontheturbine
h 2 g
[m / s ] (8.1)
h is here the height difference between the hydraulic pressure and the
stagnationpressurewhileisthePitotcoefficientgivenbyKjlle(2)torange
from0.98to1.00.ThemeasurementsareonlyvalidiftheReynoldsnumberis
above100.
Re
cd
(8.2)
WhenmeasuringthevelocityaPitottubewiththreepressureholeswasused.
The centre hole measures the stagnation pressure and the two holes on the
side of the tube, in this case the side of the knife measures the hydraulic
pressure. The height of the two water columns leading from the hydraulic
pressure measurements have to be levelled in order to have a valid
measurement. In Figure 87 the two columns measuring dynamic pressure is
levelled and the measurement is valid. The velocity in the flow can then be
calculated using equation(8.1). In real measurements it is difficult to get the
twohydruliccolumnstobeexactlylevelledduetofluctuationsintheflowand
59
timelaginthemeasurements,meaningthatwhenthePitotistwistedinorder
tolevelthewatercolumnsittakestimefromwhenthetubeistwistedtowhen
thewaterlevelisstable.
In the tests performed in this thesis an allowed height difference in the two
dynamic measurements is set to 5mm. A 5mm height difference is found
acceptable since the goal with the measurements is to confirm CFD results
which have a high level of uncertainty. The velocity measurements also have
uncertainties linked to them, they are not considered in this thesis since the
uncertaintiesinthemeasurementsaresmallcomparedtotheCFDanalysis.
Ph
P0
Ph
Figure87HeightdifferenceinaknifePitotmeasurement
Thevelocitydistributionintheoutletmayvarywithdifferentpressureheads.
TobeabletocomparethemeasuredvelocityprofilewiththeCFDresultswhere
the inlet and the outlet pressure is governed by the volume flow the actual
velocitymeasurementshavetobescaledagainstadimensionlessfactor.
c' c
2 gH *
2 gH
[m / s] (8.3)
Inequation(8.3)H*andHistheheightdifferencethemeasuredstaticpressure
intheinletandoutlet.H*isthereferencevaluekeptconstantandhastobeset
atanappropriatevalue.Thevaluecanbesetwhenforinstance foradesired
60
volume flow. If the pressure is kept constant during the test this is not
necessary.
8.3 Outlet
Theoutletoftheturbineconsistsofadrafttube,threefinsandahubunderthe
turbine. The draft tube and the hub help improve the performance of the
turbine. The fins on the other hand do not necessary give a positive
contributiontotheperformanceoftheturbine.Frequencyanalysisconducted
byRemiAndreStoplealsoshowthatfluctuationsinthemeasurementscanbe
causedbythebladespassingthethreesupportfins.
The water coming out of the runner have in most cases swirl because the
turbineisnotoperatingatbestefficiencypoint.Thefinswilldisturbthewater
swirlandmaycausepressurefluctuationswhentherunnerbladespassthefins.
Toavoiddisturbanceoftheswirlaftertheturbinethebearingplacedunderthe
turbine can be moved into the already existing guide vane housing directly
abovetherunner.
61
Figure88Bearingwhenplacedinsidetheguidevanecentrepiece
AcylindricalrollerbearingischosenwithanNUdesign.NUdesignlettherebe
minormovementsintheaxialdirection.Thistypeofbearingdoesnottakeany
loadintheaxialdirection.InFigure88thebearingisindicatedwithindex3.
Thebearingshouldbeplacedasclosetotherunneraspossibletopreventcast
intheshaftandrunner.
To prevent leakage from the water way into the bearing a viper seal can be
chosen, index 4. AHPseals (11) offers a wide range of seals for rotating
equipment. The Rotaflon series is a high performance series of seals for
rotatingshafts.TheRBseriesischosenforthisparticularcase.TheRBseriesis
chosen because it can tolerate the pressure, temperature and rpm of the
turbineshaftproducesandoperatingconditions.Thelifetimeofthesealisnot
given because it is highly dependent on the working conditions. An identical
sealisnecessaryunderthebearing.Thissealisnotincludedinthedrawing.
Index 5 is the grease intake from outside the turbine. A small tube has to be
insertedintotheguidevanesconnectingtheoutsidewiththebearing.
ThemechanicaldrawingofthebearingisfoundinAppendixP.
62
9 Results
9.1 Efficiency tests
To create the hill diagram curve fitting was performed in Matlab to create
smooth lines. Four different curve fitting functions were tried in order to find
the function that created the best curve for the Hill diagram. The functions
usedwerea2nd,3rdand4thdegreepolynomialcurvefittingfunctionsaswellas
asmoothingsplinefunction.Anuncertaintyof2%ineachpointwasselected
to evaluate if the fitted curve is within an acceptable range of the actual
measurements.
Figure912ndorderpolyfit
63
Figure923rdorderpolyfit
Figure934thorderpolyfit
64
Figure94Smoothingspline
Above the four curve fitting functions are presented used on a measured
setting2serieswitharotationalspeedof687rpmandvaryingeffectivehead.
Thebluedotsarethemeasureddatawhiletheredlineisthefittedcurve.The
verticallinesareerrorbarswithavalueof2%relativetothemeasureddata.
ThecurvefittingforallthemeasuredseriesarefoundinAppendixH.
SmoothingofcurvesisdoneinordertoobtainaHilldiagramwhichiseasyto
read and to find values between the measured values. The fitting function
createsafunctionwhichallowstheuseroftheMatlabprogramdevelopedto
findtheefficiencyatanygivenoperationalpoint.
The smoothing spline function is the function which results in the lowest
deviation between the fitted data curve and the measured data while the 2nd
order poly fit function gives the highest deviation. Even though the 2nd order
poly fit function results in the highest deviation only one point on the fitted
curve lies outside the 2% uncertainty value. The 2nd order function is the
functionwhichgivesthemostrealisticcurvewhenevaluatingallthelinesforall
the functions. The 3rd and 4th order functions creates unrealistic gains in
efficiencyatlowheads.Thisiscausedbyhighuncertaintyinthemeasurements
whilethesmoothingsplinecreatesunevencurves.
65
Theefficiencytestwerestronglyaffectedandlimitedbythetestrigsetupand
the turbine. For a detailed description of the limitations and challenges
encounteredwhiletestingseechapter7.1,7.2and7.3.
9.1.1 Setting 1
TheHilldiagramspresentedinFigure95andFigure97arepresentedasHevs.
rotational speed diagram. The reason for not presenting the Hill diagram asa
QEDvs.nEDdiagramisthattheturbineisnotamodeloraprototypesothatthe
flow,headandrotationalspeeddoesnothavetobescaledwhentheturbineis
taken into production to find the efficiency and power output. The turbines
performanceisgovernedbytheheadandtherotationalspeed.Therotational
speedgovernsthevolumeflow.A2ndorderpolyfitfunctionisusedtocreate
thecurvesusedtocreatetheHilldiagrams.
Figure95Hilldiagram,efficiencyplottedagainsteffectiveheadandrotationalspeed
TheHilldiagraminFigure95indicatesthatthereisabestefficiencypointwith
efficiencyof76.4%ataround575rpmand2.25metereffectiveheadinsetting
1.ButwhenstudyingFigure96itbecomesclearthatthebestefficiencypoint
66
most likely lies outside the tested range. BEP is most likely found with an
effectiveheadlessthan2metersandwithanrpmaround500.
ThereasonwhytheHilldiagramproducesaBEPat575rpmand2.3meffective
headisbecauseofthe2ndorderpolyfitfunctionusedtogeneratethediagram
does not match the measured data completely. The Hill diagram is therefor
onlyshowingtrendsintheefficiency.
Figure95hasasaddlepointaround675rpmand3metereffectivehead.Thisis
most likely caused be uncertainties in the measurements and the fact that
moretestsshouldbecarriedoutaround675rpm.Thecontourfunctionusedto
create the Hill diagrams interpolates values between the fitted curve values.
More measurement series would mean that the uncertainty in the
interpolationwouldbereduced.
The highest efficiency point is not tested due to limitations in the rig. See
chapter7.1.
Thehighestefficiencymeasuredwas76.4%ataneffectiveheadof2.25meters
and552rpmbutthegraphsinFigure96showsthatlineshavenotreachedthe
topefficiencyaround550rpm.
67
Figure 96 Raw data measurements. Efficiency curves at constant RPM plotted vs.
effectivehead.
9.1.2 Setting 2
Due to the same limitations in the rig as in setting 1 the BEP is not found in
setting2.
TheHilldiagraminFigure97makeitappearliketherearethreedifferentbest
efficiency points. Since not all rotational speeds are measured with different
heads the interpolation between each measured series can make the Hill
diagramappearedgyandnotcontinuous.Iftooserieswithrpmkeptconstant
at525and675weremeasuredthepeaksaround450and600wouldprobably
disappearandthediagramwouldlookmoresimilartotheoneinsetting1.
The Hill diagram shows BEP at 725rpm at an effective head of 5 meters. The
highest efficiency measured is 83.8% at an rpm of 602 and effective head of
2.72 meters. As seen in Figure 98 the best efficiency point in setting 2 is
probablynotreached.
68
Figure97Hilldiagram,efficiencyplottedagainsteffectiveheadandrotationalspeed
Figure 98 Raw data measurements. Efficiency curves at constant RPM plotted vs.
effectivehead
69
Clearencewater[l/s]
Clearencewaterat750rpm
0.1
0.08
0.06
0.04
Setting2
0.02
Setting1
0
0
Inletpressurehead[m]
Figure99Clearancewateratconstantrpm
70
Clearencewater[l/s]
Clearence waterH=2.5m
0.0535
0.053
0.0525
0.052
0.0515
0.051
300
400
500
600
700
800
rpm
Figure910Clearancewateratconstantinletpressureheadinsetting2
71
Mechanicalpower[kW]
Mechanicalpowervs.Heat500rpm
11
9
7
Setting1
Setting2
3
1
1
He[m]
Figure911Mechanicalpowerat500rpm
9.5 Torque
Thetorquegeneratedat500rpmispresentedinFigure912.Asseenfromthe
figurethemeasurementsareperformedcloseuptothemaximumvalueofthe
torquegaugeof200Nm.
Torquevs.Heat500rpm
Torque[Nm]
180
130
Setting1
Setting2
80
30
0
10
He[m]
Figure912Torqueat500rpm
anearlystageinthetestingofturbinebecausevalueswereplottedintheuser
interphase in LabView. Stople performed a frequency analysis of the
fluctuationsusingaMatlabprogramdevelopedbystudentAndersTrklep.
AmeasurementseriesStoplestudiedshowedfluctuationintorquewithamean
value of 43Nm of 7.5Nm. Stople found that the dominating factors when
analysing the frequencies most likely are blade frequency, electric noise,
rotationalfrequencyandbladepassingfrequencyfromthelowerbearing.
Thefrequencyanalysisshowsthatthereisapeakinthefrequencyeverytimea
blade passes a guide vane. This frequency varies with the rotational speed.
Electric noise occurs with a frequency around 50Hz. This electric noise does
mostlikelyoriginfromtheasynchronousgenerator.Rotationalfrequenciesdo
most likely origin from the cast in the shaft with an increase in friction once
everytimetheshaftturns(12).
A frequency occurs every time a blade passes the fins that support the lower
bearing.
Stople also did a frequency analysis of the pressure and volume flow
fluctuations. The analysis did not give clear results in why the pressure and
volume flow fluctuate at operation conditions that should give stable
measurements. No dominating frequencies where found and there were no
directcorrelationbetweenthefrequenciesinpressureandvolumeflow.Since
the measuring equipment used to measure pressure and volume flow are
independent and the analysis show no correlation Stople concludes that the
errormostlikelyliesintheloggingcard,theanaloguedigitalconversionorthat
thetwoequipmentaremalfunctioning.
73
AseriesofdifferentgeometrieshavebeenusedintheCFDanalysisoftheinlet
bend. None of the geometries stand out performing better than the other
geometriesatallconditionstested.
Figure913Originalgeometrywithinletvelocityof1.0814m/s
Withaninletvelocityof1.0814m/stheoriginalgeometryisthegeometrygiving
the most uniform outlet velocity. The difference in outlet velocity between
inner and outer corner is 0.3m/s. The outlet velocity profile can be seen in
Figure916totheleft.
Whentheinletvelocityisincreasedto1.25m/sgeometry1givethebestresult.
At 1.25m/s the geometry gives the most uniform velocity distribution at the
outletGeometry1canbeseenwithvelocitystreamlinesinFigure914withan
inletvelocityof1.25m/s.
74
Figure914Geometry1.Inletvelocity1.25m/s
Figure915Geometry2.Inletvelocity2.387m/s
75
When the velocity is further increase to 2.378m/s geometry 2 with two flow
controllers performs best. Geometry 2 gives a completely uniform outlet
velocitywhentheinletvelocityis2.378m/s.
Allgeometriestestedhaveauniqueinletvelocitywhichgivesauniformoutlet
velocity.Whentheinletvelocityisincreasedordecreasedoutsidethisunique
flowratethetendencyofallthetestedgeometriesisthattheoutletvelocityat
theinnercornerincreaseshencetheoutercornervelocitydecreases.
Vinlet=1.0814m/s
Vinlet=1.25m/s
Vinlet=2.387m/s
Figure916Outletvelocityprofileswithoriginalgeometryanddifferentinlet
velocities.
76
Vinlet=1.0814
Vinlet=1.2
Vinlet=2
Figur917Outletvelocityprofilesforgeometry3
Average Outlet
Velocity [m/s]
Scaled Outlet
Velocity [m/s]
1.1264
Measured
Outlet Velocity
[m/s]
0.9631
0.9885
1.0814
1.2327
1.1136
1.2623
1.4385
1.2405
1.80446
1.09206
Table10Actualandscaledoutletvelocity
When the scaled velocities are divided on the average velocities a value of 1
give that the outlet velocity is uniform across the outlet area. A value above
77
oneresultinanoutletvelocityprofileasintherighthandsidegraphinFigure
916. A value under one result in an outlet velocity profile as seen in the left
hand side graph in the same figure. The average values are found by dividing
thevolumeflowontheoutletareaandisthereforeonlyatheoreticalvalueand
does not change with the pressure and does therefore not have to be scaled
againstareferencevalue.
Inlet Velocity [m/s]
0.9885
1.2623
1.2544
Table11Scaledvelocityvs.averagevelocity
In the second measurement series the pressure head were kept constant to
checkifthepressurehadtheimpactontheresultsassuggestedintheprevious
section.
Measurement
Inlet
velocity
[m/s]
Avg. Outlet
velocity
[m/s]
Measured Deviation
outlet
from avg.
velocity
in %
[m/s]
2.423435
89.52631
1.122049
1.27868
1.241416
1.414709
2.635236
86.27407
1.018597
1.160787
2.163922
86.41852
0.946977
1.079169
2.011873
86.42798
Table12Velocitymeasurementsatconstantpressurehead
Thevelocitydifferencebetweentheaverageoutletvelocityandthemeasured
isalmostconstantforallmeasurementswithmaximumvariationofonly3.25%.
78
10 Discussion of results
10.1 Efficiency tests
The Hill diagrams created are based on a 2nd order poly fit function. When
performing the curve fitting the uncertainty was kept constant at a value of
2%.Thisisdoneinordertocheckifthecurvecreatedbythepolyfitfunction
lieswithintheuncertaintyinthemeasurements.Whenperformingcurvefitting
the individual uncertainty in each point should be used and not a constant
uncertainty.
The2ndorderpolyfitfunctionwasusedtocreatetheHilldiagrams.Thecurve
the 2nd order poly fit function creates has the highest deviation between
measured data and the curve value. The function is still chosen because it
creates the most realistic curves when comparing the curves with curves in
Figure22.
79
Even though the trend in the outlet velocities in the first series matches the
trendintheCFDwithaswitchinoutletvelocityataround1.085m/sthesecond
measuringserieshavevelocitiesclosertothevelocitiesobtainedfromtheCFD
analysis.IntheCFDanalysistheoutletvaluesareclosetoorabove2m/swhile
allmeasuredvaluesinthesecondseriesliesabove2m/s.Thefirstseriesonly
correlatewiththeCFDafterscalingthevalues.Thescalingmethodusedcanbe
questionedandmaynotbethebestwaytoevaluatethevelocitiesagainsteach
other.
Since the velocities in the second series are overall higher than in the first
series.Thefirstseriesisfoundtobeinvalid.Thereasonforthisisbecausethe
equipment used is old and the Pitot tube clogs up fast. When performing the
test the Pitot tube was changed between the two series because one of the
holes on the first Pitot had clogged itself during a two week period. The
possibilityofthetwoothersbeingalmostcloggedisthereforereasonablyhigh.
WhattheresultsfromtheCFDanalysisandthesecondmeasuringseriesshows
isthatthebendmakesthevelocitydistributionunevenwhichisnotoptimalfor
theperformanceoftheturbine.
Theabsoluteoptimalsolutionwouldhavebeentoplaceaspiralcasingwhere
the bend is today. This would definitively make the velocity and pressure
distribution before the runner even and frequencies from the guide vanes
woulddisappear.Aspiralcasingwouldalsogivetheopportunityofadjustable
guide vanes. Adjustable guide vanes does in most cases make the turbine
efficiency higher for a wider flow range because it is easier to obtain optimal
angle of attack for the runner. Adjustable guide vanes would not higher the
BEP.
Thechallengewithaspiralcasingisthattheyaredifficulttoproducebecause
oftheircomplexgeometry.Spiralcasingisnotconsideredasanoptionforthis
turbinebecauseofthis.
CFD analysis
When a spiral casing is out of the question the three other geometries
simulated has to be considered. Geometry 1 and 2 are extended with 50cm
frominlettooutlet.Thiscancauseaproblembecausetheminimumeffective
80
10.4 Outlet
When moving the lower bearing above the runner the bearing is placed in a
highpressurezone.Thehighpressuremakesthesolutionofsuckingairintothe
bearing house impossible. If air with higher pressure should be used to keep
waterleakingintothebearingairhastobepumpedinwithacompressor.This
is not an ideal solution since the turbine is meant to be cost efficient and
shouldbeeasytomaintain.
Inthesuggestionmadeforadesignofabearinghouseabovetheturbinehigh
performance seals are used to keep water from leaking into the turbine. This
typeofsealshasahighlifetimeandhightearresistance.Theproblemisthat
81
thewaterinAfghanistancanhaveaveryhighconcentrationofsand.Thesand
would most likely tear the seals down causing leaks within 100 hours of
operation. This implies that the turbine had to be taken apart and seals and
bearingshadtobereplacedveryoftencausingproductionstop.Newpartsare
costlyandproductionstopwouldaddtothecostofreplacingtheparts.
82
11 Conclusion
TheKaplanturbinetestedhasareasonablyhighefficiencytakenthedesigninto
account. The best efficiency point is not tested and configurations have tobe
madeintherigandtherunnerpositionhastoberaisedinordertoreachbest
efficiency point. The uncertainty in the measurements does not lie within the
IECstandardlimit.Thehighuncertaintyinthemeasurementsdoesoriginfrom
thepressuremeasurements.
TheCFDanalysisoftheoriginalbenddoesnotshowanyseparationorbackflow
inthebend.Thesimulationscannotbefoundvalidsincetheoutletvelocitiesdo
notmatchmeasuredvelocities.
The results from the CFD analysis performed on the new geometries can be
used as guidance for further work. The conclusion that can be drawn from
thesesimulationsisthattwoflowcontrollersisbetterthanoneforthedesired
flow rate and that the rounded bend give more uniform outlet velocity
distribution that the original bend. A rounded bend should therefore be
simulated in a three dimensional CFD analysis with two flow controllers
spanningthroughtheentirebend.
The lower bearing should be kept as it is today when the turbine is used in
waterwithhighsandcontent.Ifthewaterhasalowsandcontentitshouldbe
considered to move the bearing above the runner, this would cancel the
frequencyfromthebladespassingthesupportfins.
83
84
12 Further work
Inordertocompletetheefficiencyteststherunnerneedstobeadjustedand
placedafractionhigherinthehousinginordertoperformtestsinsettingthree
andfour.Thepipingsysteminfrontoftheturbineneedstobereconfiguredin
ordertoobtainalowerinletpressurethantherigallowstoday.
The runner shaft should be replaced or the existing shaft should be
straightenedinordertogetridofcast.
Anewsphericalbearingshouldbeinstalledintheupperbearing.Thebearing
installed today should be able to withstand the forces acting in the axial
direction. The bearing could be damaged so a new one with the same
dimensionsisrecommendedtoavoidbreakdownduringtests.
To be able to test for cavitation a new section may be designed under the
turbine housing where the bearing is today. A straight section made of plexi
glassisrecommendedtogivesufficientlightconditionsforahighspeedcamera
test. A straight section would also open the possibility for outlet pressure
measurements directly after the runner eliminating the high uncertainty the
methodusedtodaygives.
When bearing and shaft problems have been solved runaway tests can be
carriedout.
The new geometry derived from the CFD analysis should be further
investigated. A three dimensional CFD analysis is recommended in order to
check for three dimensional effects. More simulations can also be performed
fordifferentinletconditions.Theoutletpressurecanbecontrolledtocheckif
separation occurs at the bend in the existing geometry with outlet velocities
foundinthevelocitymeasurements.
86
13 Bibliography
1. Brekke, Hermod. Pumper og Turbiner. Trondheim: Vannkraftlaboratoriet,
NTNU,2003.
2.Kjlle,Arne.HydrauliskMleteknikk.Trondheim:ArneKjlle,2003..
3. INTERNATIONAL STANDARD. IEC 60193 Hydraulic turbines, storage pumps
and pumpturbines. Model acceptance tests. : Norsk nasjonalkomite for
InternationalElectrotechnicalCommision,IEC,1999.
4.Sethna,Jim.CornellUniversityLaboratoryofAtomicandSolidStatePhysics.
[Online]
30
June
1994.
[Cited:
11
12
2011.]
http://www.lassp.cornell.edu/sethna/hysteresis/WhatIsHysteresis.html.
5.ISO4185.
6. Solemslie, Bjrn Winther. Optimalisering av ringledning for peltonturbin.
Trondheim:NTNU,2010..
7.RonaldEWalpole,RaymondMyers,SharonMyers,KeyingYe.Probability&
StatisticsforEngeneers&Scientists.London:Pearson,2007.01302047675.
8.Storli,PlTore.Modelltestavfrancisturbin.NTNU,Trondheim:s.n.,2006.
9. SKF. www.karbtech.hu. www.karbtech.hu. [Online] 2003. [Cited: 28 10
2011.]
http://www.karb
tech.hu/letoltesek/skf%20axialis%20beallo%20gorgoscsapagy%20katalogus%20
5104_.pdf.
10. H K Versteeg, W Malalaseekera. An introduction to Computational Fluid
Dynamics, The Finite Volume Method, second edition. s.l.: Pearson Prentice
Hall,2006.9781930934214.
11.APHSEALS.AmericanHighPerformanceSeals.AmericanHighPerformance
Seals. [Online] APH Seals, 10 1 2012. [Cited: 10 1 2012.]
http://www.ahpseals.com/products/rotaflon_rb.php.
12.Stople,RemiAndre.TestingefficiencyandcharacteristicsofaKaplantype
smallturbine.Trondheim:RemiAndreStople,2011.
87
88
Appendix A
A.1 Flow meter
Systematicerrorintheweighingtanksystem:
fQ , isthesystematicuncertaintyinthedensityinthewaterandmay
beassumedaccordingtoIEC60193(3)tobe 0.01%
WhencombiningtheuncertaintieslistedabovewiththeRSSmethoditresults
inatotalsystematicerrorintheprimarycalibrationmethodof
f Q ,w istherandomuncertaintyoftheweightcellsandthecalibration
ofthemandisfoundbyPlToreStorlitobe 0.00072% in(8).
WhencombiningtheuncertaintieslistedabovewiththeRSSmethoditresults
inatotalrandomerrorintheprimarycalibrationmethodof
f Q ,c ,the systematic error in the instrument, here being the volume flow
meter. When calibrating the goal is to minimize the uncertainty in the signal
givenbytheinstrumentbycalibratingitagainstagivenphysicalvalue,herethe
weighingtank.Sincetheflowmeternotiscalibratedagainstallpossiblevolume
flowsthiscreatesanuncertaintylinkedtovaluesinvolumeflownotincludedin
the calibration. This relative uncertainty is referred to as f Q , regression . Also the
randomerror fQ ,d isincludedinthe f Q , regression .
Itisimportanttomentionthatthevolumeflowmeteriscalibratedoutsideits
guaranty range. This is done since the test require that the volume flow
exceedstheguarantiedvolumerangeoftheflowmeter.Therewerenoother
flow meter available at the waterpower laboratory and the results have been
discussedwithProfessorTorbjrnNielsenandfoundacceptable.
f Q ,e ,physicalphenomenaandexternalinfluencesareassumedtobenegligible
sincethecalibrationwasdoneunderthesameconditions.
fQ , f ,theerrorinphysicalpropertiesisalsoassumedtobenegligible.
f,arm,withmeasuredlengthofthearmof0.5meterandauncertainty
on the ruler of 0.001 meter the systematic uncertainty is found to be
0.2%
f,w, the systematic uncertainty in the weights and the weight bed is
calculated to have a maximum uncertainty of 0.0114325%.
DocumentationofuncertaintiesintheweightsisdonebyJustervesenet
(13).
f,w,theregressionuncertaintyiscalculatedinthecalibrationprogram
createdbyBjrnWintherSolemslieandisfoundtobe1.235138%.See
Calibrationreport.
P2:
Intheoutletpressuremeasurementsthequantitiescausinguncertaintiesareas
follows:
Uncertaintyintheradiusoftheoutlet,r2.Thisuncertaintyisassumed
tobesmallbutcannotbeneglected.
Theuncertaintyinthevelocityofthewatercalculatedtobeequalforv1
andv2.v=0.10852%.
The measured length from the bottom edge of the draft tube to the
watersurface.H=0.001m/0.5m=0.2%
Theuncertaintyinthewaterlevelfoundbytheflotationdevicecreated
to read the water level in the lower reservoir. Due to friction in the
deviceandtheuseofanormalruler,theuncertaintyisassumedtobe
flot=2%. It would have been difficult to calculate a correct value for
thisuncertaintyandanassumedvalueisthereforeaccepted.
ByusingtheRSSmethodthetotalrelativeoutletpressureuncertaintyisfound
tobe
1)
4)
1.
1.
2.
3.
4.
5.
Eisdependentonthefollowingquantities:
e p1
and
ep2
3)
ev2
1
ev2
2
v12 v22
ep1/=p1*fp1*g=0.25468*9.82146516*0.106843=0.267249681m2/s2
ep2/=p2*fp2*g=0.411998*0.020158*9.82146516=0.0815678m2/s2
ez2 is zero because z2 is set as reference level ez2 calculated to be
0.019643m2/s2whentheerrorinthemeasurementsare2mm.
ev1/2 and ev2/2 are calculated to be 0.221578m2/s2 and 0.06199 m2/s2
whenerisassumedtohaveavalueof0.1mm.
p1/=2.044633m2/s2,p2/=4.04642m2/s2
Iscalculatedtobe21.0965m2/s2,z1z2=2.148.
p1 p2
5) g ( z1 z 2 )
6)
6.
v12
v2
and 2 Iscalculatedtobe1.1134m2/s2and0.31148m2/s2whenQis
2
2
0.18752[m3/s].
Thisresultinatotalrelativeuncertaintyinenergyof:
fE
Appendix B
Appendix C
C.1 Matlab file Random Uncertainty
%% Leser inn raadatafilene for loggeverdier under mlinger--clear all
clc
temp=rawdata_import();
lengde=length(temp);
t=1.960;
%% Finner summen av alle voltverdiene ved alle mlepunktene
og hvor mange
% punkter det i hver mleserie
m=0;
for i = 1:lengde
nan_locations = find(isnan(temp{2,i}));
temp{2,i}(nan_locations) = 0;
n_rows(i) = size(temp{2,i},1);
x_values(i) = temp(2,i);
x = x_values{1,i};
for j = 1:n_rows(i)
m = m+1;
AMIOMq(m,1)=x(j,1);
%All Matrix In One
Matrix
AMIOMp(m,1)=x(j,2);
AMIOMm(m,1)=x(j,3);
end
end
% Finner s standardavvik til trykk-, moment- og
volumstrmsmlingene, for
% s renge ut usikkerheten til hver enkelt strrelse.
n=sum(n_rows);
avrq = mean2(AMIOMq);
avrp = mean2(AMIOMp);
avrm = mean2(AMIOMm);
Sxq = std2(AMIOMq);
Sxp = std2(AMIOMp);
Sxm = std2(AMIOMm);
randomunc_q = (t*Sxq)/sqrt(n);
randomunc_p = (t*Sxp)/sqrt(n);
randomunc_m = (t*Sxm)/sqrt(n);
uncq=randomunc_q*100/avrq
uncp=randomunc_p*100/avrp
uncm=randomunc_m*100/avrm
f_y19=(e_y19/Avg)*100
xlswrite('ABC',e_y19,1,'B20')
num=size(select,1);
for l = 1:num
if nhelp(l,k) == 0
alskj = 1;
else
n2(l)=nhelp(l,k);
end
end
n(k)=mean(n2);
end
%% Kjrer en lkke for finne en finere linje og for f
punkter mellom de
% mlte verdiene
for i = 1:ops
select = temp{2,i};
num=size(efficiency,1);
for j = 1:num
if efficiency(j,i) == 0
A(j,i) = NaN;
B(j,i) = NaN;
Lars(j,i) = NaN;
Lars2(j,i) = 0;
D(j,i) = 0;
E(j,i) = 0;
else
A(j,i) = Q(j,i);
B(j,i) = efficiency(j,i);
Lars(j,i) = he(j,i);
Lars2(j,i) = he(j,i);
D(j,i) = Q(j,i);
E(j,i) = efficiency(j,i);
end
end
[p,h] = polyfit(Lars2(:,i),E(:,i),4);
z=min(Lars(:,i)) : ((max(Lars(:,i))min(Lars(:,i)))/24999) :
max(Lars(:,i));
%
z=min(A(find(~isnan(A(:,i))),i)) :
(max(A(find(~isnan(A(:,i))),i)
%)-min(A(find(~isnan(A(:,i))),i)))/24999
:
max(A(find(~isnan(A(:,i))),i));
hill.etha(:,i) = polyval(p,z);
polys{i}=p;
hill.he(:,i)=z;
hill.n(1:25000,i)=n(1,i);
i
end
%% Sorterer Hill-data
[hill.he,i]=sort(hill.he);
for u=1:25000
for v=1:11
n_temp(u,v)=hill.n(i(u,v),v);
etha_temp(u,v)=hill.etha(i(u,v),v);
end
end
hill.n=n_temp;
hill.etha=etha_temp;
%% Plotter hilldiagram
figure
grid on
[C,h]=contour(hill.n(1:100:end,:),hill.Q(1:100:end,:),hill.
etha(1:100:end,:),
[50:5:65 67:2:71 72:1:75 75.1:0.5:82
82.1:0.2:84],'linewidt',1.5);
grid on
xlabel('n [rpm]')
ylabel('Q [m^{3}/s]','rotation',90)
set(gca,'fontSize',12)
clabel(C,'fontsize',12)
figure
surf(hill.n(1:100:end,:),hill.Q(1:100:end,:),hill.etha(1:10
0:end,:))
xlabel('n [rpm]')
ylabel('Q [m^{3}/s]','rotation',90)
zlabel('{\eta} [%]')
figure
hold on
farge={'b' 'k' 'r' 'g' 'm' 'y' '*-b' 's-k' '*-r' 'o-g' 'sb'};
for i=1:ops
plot(A(:,i),B(:,i),farge{i},'linewidt',1.5);
leg(i)={strcat('n=',num2str(n(i)),'rpm')};
end
xlabel('Q [m^{3}/s]')
ylabel('{\eta} [%]')
grid on
legend(leg,'orientation','horizontal','location','northouts
ide')
for i = 1:lengde
x_temp = x_values(i);
for j = 1:m_rows(i)
for k = 1:n_cols(i)
y = yi(i);
y_temp2(j,k) = (y-yavg)^2;
x = x_temp{1,1};
x_temp2(j,k) = (x(j,k)-xavg)^2;
xy_temp2(j,k) = (x(j,k)-xavg)*(y-yavg);
end
end
y_temp3(i)=sum(sum(y_temp2));
x_temp3(i)=sum(sum(x_temp2));
xytemp3(i)=sum(sum(xy_temp2));
maxtemp(i)=max(max(x));
mintemp(i)=min(min(x));
i
end
Sxx = sum(x_temp3)
Syy = sum(y_temp3)
Sxy = sum(xytemp3)
b=Sxy/Sxx
%% Setter inn kalibreringsligningen som er funnet ved
benytte excel:
plot(Y)
xlabel('Volt [V]')
ylabel('Volume flow [l/s]')
Appendix D
Table13Weightcalibrationcalculation
Weightnumber Actualweight[kg]
Uncertainty[kg] Uncertainty[%]
Uns^2
24
1,997513
0,000062
0,00310386 9,63394E06
40
4,99809
0,00015
0,003001146 9,00688E06
44
4,99924
0,00015
0,003000456 9,00274E06
45
4,99816
0,00015
0,003001104 9,00663E06
51
1,999092
0,000065
0,003251476 1,05721E05
52
1,998907
0,000089
0,004452433 1,98242E05
53
1,999424
0,00006
0,003000864 9,00519E06
54
1,999742
0,00006
0,003000387 9,00232E06
55
1,999887
0,00006
0,00300017 9,00102E06
56
1,999224
0,000059
0,002951145 8,70926E06
57
1,999551
0,000062
0,003100696 9,61432E06
58
1,999752
0,000059
0,002950366 8,70466E06
59
1,998864
0,000062
0,003101762 9,62093E06
Totaluncertainty[%]
0,011432591
Appendix E
Screenshotofbearinglifetimecalculation.Dateenteredwww.skf.com15.10.2011.
Appendix F
The results generated by a CFD analysis is depending on the settings listed
below. More parameters can influence the final result, but they have only a
smallimpactandareconsideredtoneglectableinthisanalysis.
Mesh
A mesh is generated by small nodes placed inside the geometry to be tested.
Thenumberofnodeswilldecidehowfinethemeshis.Afinemeshisnecessary
to pick up boundary layer effects and other flow effects such as separation.
Boundarylayereffectscausedbywallshareareamajorfactorinflowanalysis.
Tobeabletopickuptheseeffectsafinemeshnearwallsurfacesinnecessary.
Anextremelyfinemesh,above20.000.000nodes,isinmostcasesnotpossible
do analyse on an ordinary computer it is common to use inflation along the
walls.Inflationrefinesthemeshnearthewallsandmakesthemeshgrowwith
a factor towards the normal mesh size. This reduces the number of nodes
neededinordertoobtainagoodmesh.
Boundary layers are present in all fluid flows and can be divided into three
separatelayers:
1. Viscoussublayer:Viscousshearisthedominantfactor.
2. Bufferlayer:Velocityandturbulencearedominantfactors.
3. Overlaplayer:Bothviscousandturbulentshearsareimportant.
There are several turbulence models to choose from when solving the fluid
flow.Themodelusedinthesimulationswillbedescribedlater.
Whensimulatingitisnormaltocheckforgridindependence.Thismeansthat
whenchangingthemeshtheresultdoesnotchange.
Boundaryconditions
When the geometry and the mesh have been created initial boundary
conditionshavetobesettostartthesimulations.Theboundaryconditionsgive
theinitialconditioninthefirstsetofnodes.Simulationisaniteratingprocess
solved numerically where the state in one set of nodes is dependent on the
state in the previous set of nodes which changes until the simulation has
converged,ifsteadystatehasbeenchosen.
Q
Figure131Inflationalongwalloftheinlet
Theinletconditioninthesimulationsdoneinthisthesiswasinletvelocity.Inlet
velocity was chosen on recommendations of Ph.D. Mette Eltvik and Martin
Holst.Theoutletconditionwassettooutflow,meaningthatallflowinhasto
flow out of the specified boundary without any predetermined pressure or
velocity.Thesimulationisfreeandtheflowisnotforcedintoapredetermined
direction.TheoutletconditionwaschosenafterconsultingProfessorTorbjrn
Nielsen.
Turbulencemodel
The turbulence model used in the simulation is the SST komega turbulence
model.Themodelisgoodintheviscoussublayerandisalsogoodinthefree
stream. The model also behaves well in adverse pressure gradients and in
separatingflow(14).Thesimulationsareperformedinordertofindoutifthe
sharpbendcausesbackflowandseparation.
Residuals
Residualsarethedifferencebetweentheiteratedvalueandtheexactsolution.
Since FLUENT does not know the real value the residuals gives the value
betweentwoiterations.Theresidualsshouldconvergewithavaluelowerthan
R
10E4.Thesimulationcanconvergewithavaluelowerthan10E4withoutthe
solutiontobecorrect.Thisisameshproblemandagridindependencetestis
needed.SeeMeshchapterabove.
Yplus
They+valueiscalculatedasinequation(8.6):
u* y
(8.6)
Whereu*isthefrictionvelocityatthenearestwall,yisthedistancefromthe
nearestwalltothefirstnodeandisthekinematicviscosityofthefluid(15).
The y plus value should not exceed a value of 5 when using the SST komega
turbulencemodelinordertopickuptheeffectsintheviscoussublayer.They
plus value should have a value close to one for the turbulence model to
performoptimal.
Appendix G
Three different LabView programs were used during test and calibration. The
program used to calibrate pressure and torque is developed by Hkon Hjort
Francke and further developed by Bjrn Winther Solemslie. The calibration
programforthevolumeflowandtheloggingprogramusedtologandcollect
datainthetestisdevelopedbyRemiAndreStople.
Figure132Frontpanel
Figure133Dataacquisitionandtranslationofvoltsignals
A DaqMX package in LabView is used to read the Volt signal from each
measuringdevice.ThevaluesfromtheDaqMXpackagearesendtoaforloop
where the signals are translated into physical values. Output values from the
looparesendtorawdatastorageandtoanewforloopwhichcalculatedthe
meanvalueandthestandarddeviationinthemeasurements(12).
Figure134RPMsubVIforreadingsoftherotationalspeed
RotationalspeedisreadbyaprogrammadebyJoarGrilstadandimplemented
in the main program with help from Bjrn Winther Solemslie. The program
continuouslyreadstherotationalspeedanddeliversthelastreadrpmvalueto
themainprogram.ThisisdonetoavoidthedelaythesubVIcreatesinthemain
program.Delayoccurssincetheopticaltripmeteronlyproducesasignalevery
time thereflexbandpassestheopticalreaderand istherefore dependenton
therotationalspeed.Themainprogramontheotherhandcreatesvaluesata
predetermined fixed rate (12). In Figure 135 the block diagram of sub VI is
shown.
Figure135BlockdiagramfortheRPMsubVI
Afterthecalibrationforloopthevaluesaresenttoacalculationnode.Allthe
calculations in the program are done in the calculation node. The calculated
values are bundled together in a long array. Mean values from the measured
calibratedparametersareaddedtothesamearray.Thearrayisthensendto
meanvaluestorage.
Figure136Calculationnodeandfrontpanelvalues
In Figure 137 raw data storage, mean value storage and rpm storage are
shownrespectivelyfromlefttoright.Therpmandmeanvaluestoragechecksif
thereexistanrpmormeanvaluefileeverytimetheprogramissaved.Ifafile
existsanewlineisaddedtothefilecontainingthelastsavedvalueselsethey
create an rpm file and a mean value file. Raw data storage creates a new file
everytimetheprogramissaved.
V
Figure137Storage
Remi Andre Stople implemented a function seen in the Figure below that
plottedtheefficiencyvs.effectiveheadintheuserinterphase.Thiswasdonein
ordertodetectanyspuriouserrorsinthetests.
Figure138Plottingofefficiencyheadgraph
Appendix H
H.1 2 nd order poly fit
Appendix I,
CalibrationcertificateDruckDPI
Appendix J
Weightcalibration
Appendix K
Riskassessment.
Risikovurderingsrapport
Kaplanrigg
Prosjekttittel
Prosjektleder
Enhet
HMSkoordinator
Linjeleder
Riggnavn
Plassering
Romnummer
Riggansvarlig
Risikovurdering
utfrtav
TestavKaplanturbin
TorbjrnNielsen
NTNU
BrdBrandstr
OleGunnarDahlhaug
Kaplanrigg
Vannkraftlab
42
LarsFjrvoldogRemiAndrStople
LarsFjrvoldogRemiAndrStople
INNHOLDSFORTEGNELSE
1
INNLEDNING....................................................................................................................1
2
ORGANISERING................................................................................................................1
3
RISIKOSTYRINGAVPROSJEKTET......................................................................................1
4
TEGNINGER,FOTO,BESKRIVELSERAVFORSKSOPPSETT..............................................1
5
EVAKUERINGFRAFORSKSOPPSETNINGEN...................................................................2
6
VARSLING.........................................................................................................................2
6.1
6.2
Frforskskjring............................................................................................................2
Vedunskedehendelser.................................................................................................2
7
VURDERINGAVTEKNISKSIKKERHET...............................................................................3
7.1
7.2
7.3
7.4
7.5
7.6
7.7
Fareidentifikasjon,HAZOP...............................................................................................3
Brannfarlig,reaksjonsfarligogtrykksattstoffoggass....................................................3
Trykkpkjentutstyr.........................................................................................................3
Pvirkningavytremilj(utslipptilluft/vann,sty,temperatur,rystelser,lukt)...........4
Strling.............................................................................................................................4
Brukogbehandlingavkjemikalier..................................................................................4
Elsikkerhet(behovforavvikefragjeldendeforskrifterognormer)............................4
8
VURDERINGAVOPERASJONELLSIKKERHET....................................................................4
8.1
8.2
8.3
8.4
8.5
8.6
8.7
8.8
ProsedyreHAZOP............................................................................................................4
Driftsogndstoppsprosedyre........................................................................................5
Opplringavoperatrer................................................................................................5
Tekniskemodifikasjoner..................................................................................................5
Personligverneutstyr......................................................................................................5
Generelt...........................................................................................................................5
Sikkerhetsutrustning.......................................................................................................5
Spesielletiltak..................................................................................................................5
9
TALLFESTINGAVRESTRISIKORISIKOMATRISE.............................................................5
10
KONKLUSJON...................................................................................................................6
11
LOVERFORSKRIFTEROGPLEGGSOMGJELDER............................................................7
12
VEDLEGG..........................................................................................................................8
13
DOKUMENTASJON...........................................................................................................9
14
VEILEDNINGTILRAPPORTMAL......................................................................................10
1
INNLEDNING
Beskrivelse av forsksoppsetningen og formlet med eksperimentene. Hvor er riggen
plassert?
2
ORGANISERING
Rolle
LabAnsvarlig:
Linjeleder:
HMSansvarlig:
HMSkoordinator
HMSkoordinator
Romansvarlig:
Prosjektleder:
Ansvarligriggoperatrer:
3
NTNU
Sintef
MortenGrnli
HaraldMhlum
OlavBolland
MonaJ.Mlnvik
OlavBolland
MonaJ.Mlnvik
ErikLangrgen
HaraldMhlum
BrdBrandstr
BrdBrandstr
TorbjrnNielsen
LarsFjrvoldogRemiAndrStople
RISIKOSTYRINGAVPROSJEKTET
Hovedaktiviteterrisikostyring
Prosjektinitiering
Veiledningsmte
Innledenderisikovurdering
Vurderingavteknisksikkerhet
Vurderingavoperasjonellsikkerhet
Sluttvurdering,kvalitetssikring
Ndvendigetiltak,dokumentasjon
Prosjektinitieringmal
DTG
X
X
X
X
X
4
TEGNINGER,FOTO,BESKRIVELSERAVFORSKSOPPSETT
Vedlegg:
ProsessogInstrumenteringsDiagram,(PID)
Skalinneholdeallekomponenteriforsksoppsetningen
Komponentlistemedspesifikasjoner
Tegningerogbildersombeskriverforsksoppsetningen.
Hvoroppholderoperatrseg,hvorergassflasker,avstegningsventilerforvann/luft.
Annendokumentasjonsombeskriveroppsettogvirkemte.
1
5
EVAKUERINGFRAFORSKSOPPSETNINGEN
Sekapittel14Veiledningtilrapportmal.
Evakuering skjer p signal fra alarmklokker eller lokale gassalarmstasjon med egen lokal
varslingmedlydoglysutenforaktuellerom,se6.2
Evakuering fra riggomrdet foregr igjennom merkede ndutganger til mteplass, (hjrnet
GamleKjemi/Kjelhusetellerparkeringsplass1ab).
Aksjon p rigg ved evakuering: Trykke ndstopp for stopp av pumper og ndstopp for
generator.
6
VARSLING
6.1 Frforskskjring
Varslingperepost,medopplysningomforskskjringensvarighetoginvolvertetil:
x HMSkoordinatorNTNU/SINTEF
HaraldStein.S.Mahlum@sintef.no
Erik.langorgen@ntnu.no
Baard.brandaastro@ntnu.no
x Prosjektlederepnaboriggervarslesforavklaringrundtbrukavavtrekksanlegget
utenfareellerforstyrrelseravnoenart,seriggmatrise.
All forskskjringen skal planlegges og legges inn i aktivitetskalender for lab. Forsksleder
mfbekreftelsepatforskeneerklarertmedvriglabdriftfrforskkaniverksettes.
6.2 Vedunskedehendelser
BRANN
Vedbrannenikkeselveristandtilslukkemedrimeligelokalttilgjengeligeslukkemidler,
skal nrmeste brannalarm utlses og arealet evakueres raskest mulig. En skal s vre
tilgjengeligforbrannvesen/bygningsvaktmesterforpvisebrannsted.
Ommuligvarsless:
NTNU
SINTEF
LabsjefMortenGrnli,tlf:91897515
HMS:ErikLangrgen,tlf:91897160
Instituttleder:OlavBolland:91897209
GASSALARM
Vedgassalarmskalgassflaskerstengesumiddelbartogomrdetventileres.Klarermanikke
innen rimelig tid f ned nivet p gasskonsentrasjonen s utlses brannalarm og laben
evakueres. Dedikert personell og eller brannvesen sjekker s lekkasjested for fastsl om
determuligtettelekkasjeoglufteutomrdetpenforsvarligmte.
Varslingsrekkeflgesomioverstendepunkt.
PERSONSKADE
x FrstehjelpsutstyriBrann/frstehjelpsstasjoner,
x Ropphjelp,
2
x Startlivreddendefrstehjelp
x Ring113hvisdeterellerdetertvilomdeteralvorligskade.
ANDREUNSKEDEHENDELSER(AVVIK)
NTNU:
Rapporteringsskjemaforunskedehendelserp
http://www.ntnu.no/hms/2007_Nettsider/HMSRV0401_avvik.doc
7
VURDERINGAVTEKNISKSIKKERHET
7.1 Fareidentifikasjon,HAZOP
Sekapittel14Veiledningtilrapportmal.
Forsksoppsetningendelesinniflgendenoder:
Node1 Rrsystemmedpumpe
Node2 Roterendeturbin
Node3 Generatoroppsett
Vedlegg,skjema:Hazop_mal
Vurdering:
Node1:
Rrelementerereksterntlevertoggodkjentforaktuelttrykk
Node2:
Roterendedelererfordetmesteikketilgjengelig.Roterendedelerifrilufterlettsynlig,og
utenfornormalarbeidssone.
Node3:
Generatoroppsetterforsvarligmontert,vanskeligtilgjengeligfragulv.
7.2 Brannfarlig,reaksjonsfarligogtrykksattstoffoggass
Sekapittel14Veiledningtilrapportmal.
Inneholderforskenebrannfarlig,reaksjonsfarligogtrykksattstoff
Ja Trykksattvann
Vurdering:Arbeidsmediumervann.Allerrsomerlevertaveksterntleverandrer
trykktestet.
7.3 Trykkpkjentutstyr
Inneholderforsksoppsetningentrykkpkjentutstyr:
JA
Utstyrettrykktestesihenholdtilnormogdokumenteres.
Trykkutsatt utstyr skal trykktestes med driftstrykk gange faktor 1.4,for utstyr som har
usertifiserte sveiser er faktoren 1.8. Trykktesten skal dokumenteres skriftlig hvor
fremgangsmteframgr.
Vedlegg:Sertifikatfortrykkpkjentutstyr.
Vurdering:
3
7.4 Pvirkningavytremilj(utslipptilluft/vann,sty,temperatur,rystelser,lukt)
Sekapittel14Veiledningtilrapportmal..
NEI
Vurdering: Vil eksperimentene generere utslipp av ryk, gass, lukt eller unormalt avfall.?
Mengder/konsistens.Erdetbehovforutslippstillatelse,ekstraordinretiltak?
7.5 Strling
Sekapittel14Veiledningtilrapportmal.
NEI
Vedlegg:
Vurdering:
7.6 Brukogbehandlingavkjemikalier
Sekapittel14Veiledningtilrapportmal.
NEI
Vedlegg:
Vurdering: Inneholder eksperimentene bruk og behandling av kjemikalier Hvilke og hvilke
mengder? Hvordan skal dette avhendes, oppbevares?, risikovurder i henhold til
sikkerhetsdatablad Er det behov for beskyttelses tiltak tillegges disse i operasjonell
prosedyre.
7.7 Elsikkerhet(behovforavvikefragjeldendeforskrifterognormer)
NEI
Herforstsmontasjeogbrukiforholdtilnormerogforskriftermedtankepberringsfare
Vedlegg:
Vurdering:
8
VURDERINGAVOPERASJONELLSIKKERHET
Sikrer at etablerte prosedyrer dekker alle identifiserte risikoforhold som m hndteres
gjennom operasjonelle barrierer og at operatrer og teknisk utfrende har tilstrekkelig
kompetanse.
8.1 ProsedyreHAZOP
Sekapittel14Veiledningtilrapportmal.
Metodenerenunderskelseavoperasjonsprosedyrer,ogidentifisererrsakerogfarekilderfor
operasjonelleproblemer.
Vedlegg:HAZOP_MAL_Prosedyre
Vurdering:
4
8.2 Driftsogndstoppsprosedyre
Sekapittel14Veiledningtilrapportmal.
Driftsprosedyrenerensjekklistesomskalfyllesutforhvertforsk.
Ndstopp prosedyren skal sette forsksoppsetningen i en harmls tilstand ved uforutsette
hendelser.
VedleggProcedureforrunningexperiments
8.3 Opplringavoperatrer
DokumentsomviserOpplringsplanforoperatrerutarbeidesforalleforksoppsetninger.
x Kjringavpumpesystem
x BrukavLabVIEWprogram
x Kjringavgenerator
Vedlegg:Opplringsplanforoperatrer
8.4 Tekniskemodifikasjoner
Vurdering:ModifikasjonergjresisamrdmedTorbjrnNielsen,BrdBrandstrogAnders
Austegrd
8.5 Personligverneutstyr
Vurdering:Vernebrillerpkrevd
8.6 Generelt
Vurdering:Alleforskkjresmedoperatrtilstede.
8.7 Sikkerhetsutrustning
Vernebriller
8.8 Spesielletiltak
9
TALLFESTINGAVRESTRISIKORISIKOMATRISE
Sekapittel14Veiledningtilrapportmal.
Risikomatrisenvilgienvisualiseringogensamletoversiktoveraktivitetensrisikoforholdslik
atledelseogbrukerefretmestmuligkomplettbildeavrisikoforhold.
IDnr Aktivitethendelse
FrekvSans Kons
RV
1
Roterendeaksling
2
B
B2
2
Fremmedelementerivannet
1
A
A1
3
Rrbrudd
1
A
A1
5
Vurdering restrisiko: Deltakerne foretar en helhetsvurdering for avgjre om gjenvrende
risikovedaktiviteten/prosessenerakseptabel.Avsperringogkjringutenomarbeidstid
10
KONKLUSJON
Riggenerbyggettilgodlaboratoriumpraksis(GLP).
Hvilketekniskeendringerellerendringeravdriftsparameterevilkrevenyrisikovurdering.
Annetmedium,trykk,mekaniskeinngrep
ApparaturkortetfrengyldighetpXXmneder
ForskpgrkortfrengyldighetpXXmneder
6
11
LOVERFORSKRIFTEROGPLEGGSOMGJELDER
Sehttp://www.arbeidstilsynet.no/regelverk/index.html
x Lovomtilsynmedelektriskeanleggogelektriskutstyr(1929)
x Arbeidsmiljloven
x Forskriftomsystematiskhelse,miljogsikkerhetsarbeid(HMSInternkontrollforskrift)
x Forskriftomsikkerhetvedarbeidogdriftavelektriskeanlegg(FSE2006)
x Forskriftomelektriskeforsyningsanlegg(FEF2006)
x ForskriftomutstyrogsikkerhetssystemtilbrukieksplosjonsfarligomrdeNEK420
x Forskriftomhndteringavbrannfarlig,reaksjonsfarligogtrykksattstoffsamtutstyrog
anleggsombenyttesvedhndteringen
x ForskriftomHndteringaveksplosjonsfarligstoff
x Forskriftombrukavarbeidsutstyr.
x ForskriftomArbeidsplasserogarbeidslokaler
x ForskriftomBrukavpersonligverneutstyrparbeidsplassen
x ForskriftomHelseogsikkerhetieksplosjonsfarligeatmosfrer
x ForskriftomHytrykksspyling
x ForskriftomMaskiner
x ForskriftomSikkerhetsskiltingogsignalgivningparbeidsplassen
x ForskriftomStillaser,stigerogarbeidptakm.m.
x ForskriftomSveising,termiskskjring,termiskspryting,kullbuemeisling,loddingog
sliping(varmtarbeid)
x ForskriftomTekniskeinnretninger
x ForskriftomTungtogensformigarbeid
x ForskriftomVernmoteksponeringforkjemikalierparbeidsplassen
(Kjemikalieforskriften)
x ForskriftomVernmotkunstigoptiskstrlingparbeidsplassen
x ForskriftomVernmotmekaniskevibrasjoner
x ForskriftomVernmotstyparbeidsplassen
Veiledningerfraarbeidstilsynet
se:http://www.arbeidstilsynet.no/regelverk/veiledninger.html
7
12
VEDLEGG
8
13
x
x
x
x
x
x
x
x
x
x
x
DOKUMENTASJON
Tegninger,foto,beskrivelseravforsksoppsetningen
Hazop_mal
Sertifikatfortrykkpkjentutstyr
HndteringavfalliNTNU
SikkerbrukavLASERE,retningslinje
HAZOP_MAL_Prosedyre
Forsksprosedyre
Opplringsplanforoperatrer
Skjemaforsikkerjobbanalyse,(SJA)
Apparaturkortet
Forskpgrkort
9
14
VEILEDNINGTILRAPPORTMAL
Kap5Evakueringfraforsksoppsetningen
Beskrivihvilkentilstandriggenskalforlatesvedenevakueringssituasjon.
Kap7Vurderingavteknisksikkerhet
Sikreatdesignavapparatureroptimalisertiforholdtilteknisksikkerhet.
Identifisererisikoforholdknyttettilvalgtdesign,ogeventueltinitiereredesignforsikre
atstrstmuligandelavrisikoelimineresgjennomteknisksikkerhet.
Punktene skal beskrive hva forsksoppsetningen faktisk er i stand til tle og aksept for
utslipp.
7.1 Fareidentifikasjon,HAZOP
Forsksoppsetningendelesinninoder:(eksMotorenhet,pumpeenhet,kjleenhet.)
Ved hjelp av ledeord identifiseres rsak, konsekvens og sikkerhetstiltak. Konkluderes det
medattiltakerndvendiganbefalesdissepbakgrunnavdette.Tiltakenelukkesnrdeer
utfrtogHazopsluttfres.
(eks No flow, rsak: rr er deformert, konsekvens: pumpe gr varm,
sikkerhetsforanstaltning: mling av flow med kobling opp mot ndstopp eller hvis
konsekvensen ikke er kritisk benyttes manuell overvkning og punktet legges inn i den
operasjonelleprosedyren.)
7.2Brannfarlig,reaksjonsfarligogtrykksattstoff.
I henhold til Forskrift om hndtering av brannfarlig, reaksjonsfarlig og trykksatt stoff samt
utstyroganleggsombenyttesvedhndteringen
Brannfarlig stoff: Fast, flytende eller gassformig stoff, stoffblanding, samt stoff som
forekommerikombinasjoneravsliketilstander,somikraftavsittflammepunkt,kontaktmed
andrestoffer,trykk,temperaturellerandrekjemiskeegenskaperrepresentererenfarefor
brann.
Reaksjonsfarlig stoff: Fast, flytende, eller gassformig stoff, stoffblanding, samt stoff som
forekommerikombinasjoneravsliketilstander,somvedkontaktmedvann,vedsitttrykk,
temperaturellerandrekjemiskeforhold,representererenfareforfarligreaksjon,eksplosjon
ellerutslippavfarliggass,damp,stvellertke.
Trykksattstoff:Annetfast,flytendeellergassformigstoffellerstoffblandingennbranneller
reaksjonsfarlig stoff, som er under trykk, og som derved kan representere en fare ved
ukontrollertutslipp.
Nrmere kriterier for klassifisering av brannfarlig, reaksjonsfarlig og trykksatt stoff er
fastsatt i vedlegg 1 i veiledningen til forskriften Brannfarlig, reaksjonsfarlig og trykksatt
stoff
http://www.dsb.no/Global/Publikasjoner/2009/Veiledning/Generell%20veiledning.pdf
http://www.dsb.no/Global/Publikasjoner/2010/Tema/Temaveiledning_bruk_av_farlig_stoff_Del_1.p
df
RiggogarealskalgjennomgsmedhensynpvurderingavExsone
10
x
x
x
Sone0:Alltideksplosivatmosfre,foreksempelinneitankermedgass,
brennbarvske.
Sone1:Primrsone,tidviseksplosivatmosfreforeksempeletfylletappe
punkt
Sone2:Sekundertutslippssted,kanfeksplosivatmosfreveduhell,for
eksempelvedflenser,ventilerogkoblingspunkt
7.4Pvirkningavytremilj
Medforurensningforsts:tilfrselavfaststoff,vskeellergasstilluft,vannellerigrunnen
styogrystelserpvirkningavtemperaturensomerellerkanvretilskadeellerulempefor
miljet.
Regelverk:http://www.lovdata.no/all/hl19810313006.html#6
NTNUretningslinjerforavfallse:http://www.ntnu.no/hms/retningslinjer/HMSR18B.pdf
7.5Strling
Strlingdefineressom
Ioniserende strling: Elektromagnetisk strling (i strlevernsammenheng med blgelengde
<100 nm) eller hurtige atomre partikler (f.eks alfa og betapartikler) som har evne til
ionisereatomerellermolekyler
Ikkeioniserende strling: Elektromagnetisk strling (blgelengde >100 nm), og ultralyd1,
somharlitenelleringenevnetilionisere.
Strlekilder:Alleioniserendeogsterkeikkeioniserendestrlekilder.
Ioniserende strlekilder: Kilder som avgir ioniserende strling, f.eks alle typer radioaktive
kilder,rntgenapparater,elektronmikroskop
Sterke ikkeioniserende strlekilder: Kilder som avgir sterk ikkeioniserende strling som
kan skade helse og/eller ytre milj, f.eks laser klasse 3B og 4, MR2systemer, UVC3kilder,
kraftigeIRkilder4
1 Ultralyd er akustisk strling (lyd) over det hrbare frekvensomrdet (>20 kHz). I strlevernforskriften er
ultralydomtaltsammenmedelektromagnetiskikkeioniserendestrling.
MR (eg. NMR) kjernemagnetisk resonans, metode som nyttes til avbilde indre strukturer i ulike
materialer.
3UVCerelektromagnetiskstrlingiblgelengdeomrdet100280nm.
4IRerelektromagnetiskstrlingiblgelengdeomrdet700nm1mm.
2
Forhverlaserskaldetfinneseninformasjonsperm(HMSRV3404B)somskalinneholde:
x Generellinformasjon
x Navnpinstrumentansvarligogstedfortreder,oglokalstrlevernskoordinator
x Sentraledataomapparaturen
x Instrumentspesifikkdokumentasjon
x Referansertil(evtkopierav)datablader,strlevernbestemmelser,o.l.
x Vurderingeravrisikomomenter
x Instruksforbrukere
x Instruksforpraktiskbruk;oppstart,drift,avstenging,sikkerhetsforholdsregler,
loggfring,avlsing,evt.brukavstrlingsmler,osv.
x Ndprosedyrer
SeellersretningslinjentilNTNUforlaser:http://www.ntnu.no/hms/retningslinjer/HMSR34B.pdf
11
7.6Brukogbehandlingavkjemikalier.
Herforstskjemikaliersomgrunnstoffsomkanutgjreenfareforarbeidstakerssikkerhet
oghelse.
Seellers:http://www.lovdata.no/cgiwift/ldles?doc=/sf/sf/sf200104300443.html
Sikkerhetsdatablar skal vre i forkenes HMS perm og kjemikaliene registrert i
Stoffkartoteket.
Kap8Vurderingavoperasjonellsikkerhet
Sikrer at etablerte prosedyrer dekker alle identifiserte risikoforhold som m hndteres
gjennom operasjonelle barrierer og at operatrer og teknisk utfrende har tilstrekkelig
kompetanse.
8.1ProsedyreHazop
ProsedyreHAZOPgjennomfressomensystematiskgjennomgangavdenaktuelle
prosedyrenvedhjelpavfastlagtHAZOPmetodikkogdefinerteledeord.Prosedyrenbrytes
nedienkeltstendearbeidsoperasjoner(noder)oganalyseresvedhjelpavledeordenefor
avdekkemuligeavvik,uklarheterellerkildertilmangelfullgjennomfringogfeil.
8.2Driftsogndstoppprosedyrer
Utarbeidesforalleforsksoppsetninger.
Driftsprosedyrenskalstegvisbeskrivegjennomfringenavetforsk,inndeltioppstart,under
drift og avslutning. Prosedyren skal beskrive forutsetninger og tilstand for start,
driftsparameteremedhvorstoreavviksomtillatesfrforsketavbrytesoghvilkentilstand
riggenskalforlates.
Ndstoppprosedyrebeskriverhvordanenndstoppskalskje,(utfrtavuinnvidde),
hvasomskjer,(strm/gasstilfrsel)og
hvilkehendelsersomskalaktiverendstopp,(brannalarm,lekkasje).
12
Kap9Risikomatrise
9Tallfestingavrestrisiko,Risikomatrisen
Forsynliggjresamletrisiko,jevnfrskjemaforrisikovurdering,plotteshverenkeltaktivitets
verdiforsannsynlighetogkonsekvensinnirisikomatrisen.BrukaktivitetensIDnr.
Eksempel:HvisaktivitetmedIDnr.1harfttenrisikoverdiD3(sannsynlighet3xkonsekvensD)
settesaktivitetensIDnririsikomatrisensfeltfor3D.Sliksettesalleaktivitetenesrisikoverdier
(IDnr)innirisikomatrisen.
Irisikomatrisenerulikegraderavrisikomerketmedrd,gulellergrnn.Nrenaktivitetsrisiko
havnerprd(=uakseptabelrisiko),skalrisikoreduserendetiltakgjennomfres.Nyvurdering
gjennomfresetterattiltakeriverksattforseomrisikoverdienerkommetnedpakseptabelt
niv.
KONSEKVENS
Svrt
alvorlig
Alvorlig
E1
E2
E3
E4
E5
D1
D2
D3
D4
D5
Moderat
C1
C2
C3
C4
C5
Liten
B1
B2
B3
B4
B5
Svrt
liten
A1
A2
A3
A4
A5
Svrtliten
Liten
Middels
Stor
Svrt Stor
SANSYNLIGHET
Farge
Rd
Gul
Grnn
Beskrivelse
Uakseptabelrisiko.Tiltakskalgjennomfresforredusererisikoen.
Vurderingsomrde.Tiltakskalvurderes.
Akseptabelrisiko.Tiltakkanvurderesutfraandrehensyn.
13
Vedleggtil
Risikovurderingsrapport
Kaplanrigg
Prosjekttittel
Prosjektleder
Enhet
HMSkoordinator
Linjeleder
Riggnavn
Plassering
Romnummer
Riggansvarlig
TestavKaplanturbin
TorbjrnNielsen
NTNU
BrdBrandstr
OleGunnarDahlhaug
Kaplanrigg
Vannkraftlab
42
LarsFjrvoldogRemiAndrStople
INNHOLDSFORTEGNELSE
x
VEDLEGGAHAZOPMAL..................................................................................................1
x
VEDLEGGBPRVESERTIFIKATFORLOKALTRYKKTESTING.............................................1
x
VEDLEGGFHAZOPMALPROSEDYRE.............................................................................1
x
VEDLEGGGFORSKSPROSEDYRE...................................................................................1
x
VEDLEGGHOPPLRINGSPLANFOROPPERATRER......................................................3
x
VEDLEGGISKJEMAFORSIKKERJOBBANALYSE..............................................................4
x
VEDLEGGJAPPARATURKORTUNITCARD........................................................................6
x
VEDLEGGKFORSKPGRKORT..................................................................................7
Consequences
None
None
None
None
Turbinecanbe
overloaded
Safetyvalvecan
beopened
Causes
Toohighpump
speed
Toolowpump
speed
Toohighpump
speed
x VEDLEGGAHAZOPMAL
Project:
Node:1
Ref
Guideword
#
Noflow
Reverseflow
Moreflow
Lessflow
Morelevel
Lesslevel
Morepressure
Emergencyshut
down
Safeguards
Recommendations
Reducepump
speed
Increasepump
speed
Reducepump
speed
Action
Page
Date
Sign
Project:
Node:1
Ref
Guideword
#
Lesspressure
Abnormal
operation
Project:
Node:2
Ref
Guideword
#
Noflow
Reverseflow
Moreflow
Consequences
None
None
Consequences
None
None
Loadmaybe
toohigh
Cavitationmay
occur
Causes
Faultinsystem
Faultinsystem
Causes
Toohighpump
speed
Safeguards
Safeguards
Recommendations
Recommendations
Reducepump
speed
Action
Page
Checksystem
Checksystem
Action
Page
Date
Sign
Date
Sign
Project:
Node:2
Ref
Guideword
#
Lessflow
Morelevel
Lesslevel
Morepressure
Lesspressure
Abnormal
operation
Project:
Node:3
Emergencyshut
downbutton
Turbinecanbe
overloaded
None
None
Toohighpump
speed
Faultinsystem
Faultinsystem
None
Toolowpump
speed
Safeguards
Consequences
Causes
Recommendations
Page
Checksystem
Checksystem
Reducepump
speed
Increasepump
speed
Action
Page
Date
Sign
Ref
#
Faultinsystem
Abnormal
operation
Lossof
connectiontothe
electricalgrid
Turbinegoesto
runawayspeed
Causes
Guideword
Turbinecanbe
overloaded
None
Consequences
Emergencyshut
downswitch
Safeguards
Recommendations
Checksystem
Checksystem
Action
Date
Sign
x
VEDLEGGBPRVESERTIFIKATFORLOKALTRYKKTESTING
TrykktestenskalutfresIflgeNSEN13445del5(Inspeksjonogprving).
SeogsprosedyrefortrykktestinggjeldendeforVATLlab
Trykkpkjentutstyr:
.
Benyttesirigg: .
Designtrykkforutstyr:
..bara
Maksimumtillatttrykk:
..bara
(i.e.burstpressureomkjent)
Maksimumdriftstrykkidennerigg:
..bara
Prvetrykketskalfastleggesiflgestandardenogmedhensyntilmaksimum
tillatttrykk.
Prvetrykk: ..bara
Testmedium:
Temperatur:
Start:
Tid:
Trykk:
Slutt:
Tid:
Trykk:
(.xmaksimumdriftstrykk)
Iflgestandard
C
bara
bara
Eventuellerepetisjonerfraatm.trykktilmaksimumprvetrykk:.
Test trykket, dato for testing og maksimum tillatt driftstrykk skal markers p
(skiltellerinnsltt)
Stedogdato
Signatur
1
Trinn p feil
plass
Feil handling
Uriktig
informasjon
Trinn utelatt
Trinn mislykket
Pvirkning og
effekter fra
andre
Prosedyre er laget
for ambisis eller
preget av forvirring
Prosedyren vil lede
til at handlinger blir
gjennomfrt i feil
mnster/rekkeflge
Prosedyrens
handling er feil
spesifisert
Informasjon som er
gitt i forkant av
handling er feil
spesifisert
Manglende trinn,
eller trinn krever for
mye av operatr
Trinn har stor
sannsynlighet for
mislykkes
Prosedyrens
prestasjoner vil
trolig bli pvirket av
andre kilder
Consequences
Causes
VEDLEGGFHAZOPMALPROSEDYRE
Project:
Node:1
Ref
Guideword
#
Uklar
Safeguards
Rec#
Recommendations
Page
Action
Project:
Node:1
Ref
Guideword
#
Consequences
Causes
Safeguards
Rec#
Recommendations
Page
Action
x
VEDLEGGGFORSKSPROSEDYRE
Experiment,name,number:
TestavKaplanturbin
ProjectLeader:
TorbjrnNielsen
ExperimentLeader:
LarsFjrvoldandRemiAndrStople
Operator,Duties:
LarsFjrvold:Operationoftherig
RemiAndrStople:Operationoftherig
Conditionsfortheexperiment:
Experiments should be run in normal working hours, 08:0016:00 during
wintertimeand08.0015.00duringsummertime.
Experimentsoutsidenormalworkinghoursshallbeapproved.
One person must always be present while running experiments, and should
beapprovedasanexperimentalleader.
An early warning is given according to the lab rules, and accepted by
authorizedpersonnel.
Besurethateveryonetakingpartoftheexperimentiswearingthenecessary
protectingequipmentandisawareoftheshutdownprocedureandescape
routes.
Preparations
PosttheExperimentinprogresssign.
StartLabVIEWprogramandgenerator
Runthegeneratorto100rpm
Startuppump
Adjustpumpandgeneratortowantedoperationpoint
Duringtheexperiment
LogdatawithdesignatedLabVIEWprogram
Endofexperiment
Decreasegeneratorandpumpspeedstepwiseto100rpm
Shutdownpump,thenthegenerator
Removeallobstructions/barriers/signsaroundtheexperiment.
Tidyupandreturnalltoolsandequipment.
Tidyandcleanupworkareas.
Returnequipmentandsystemsbacktotheirnormaloperationsettings
(firealarm)
Toreflectonbeforethenextexperimentandexperienceusefulforothers
Wastheexperimentcompletedasplannedandonscheduledinprofessional
terms?
Was the competence which was needed for security and completion of the
1
Date/
Sign
Completed
Carriedout
experimentavailabletoyou?
Do you have any information/ knowledge from the experiment that you
shoulddocumentandsharewithfellowcolleagues?
2
x
VEDLEGGHOPPLRINGSPLANFOROPPERATRER
Experiment,name,number:
TestavKaplanturbin
ProjectLeader:
TorbjrnNielsen
ExperimentLeader:
LarsFjrvoldogRemiAndrStople
Operator
LarsFjrvold
RemiAndrStople
KjennskaptilEPTLABgenerelt
Lab
adgang
rutiner/regler
arbeidstid
Kjennertilevakueringsprosedyrer
Aktivitetskalender
Kjennskaptilforskene
Prosedyrerforforskene
Ndstopp
Nrmestebrann/frstehjelpsstasjon
Date/
Sign
X
X
X
X
X
X
3
x
VEDLEGGISKJEMAFORSIKKERJOBBANALYSE
SJAtittel:
Dato:
Kryssavforutfyltsjekkliste:
Deltakere:
Sted:
SJAansvarlig:
Arbeidsbeskrivelse:(Hvaoghvordan?)
Risikoforbundetmedarbeidet:
Beskyttelse/sikring:(tiltaksplan,senesteside)
Konklusjon/kommentar:
Anbefaling/godkjenning:
Dato/Signatur:
Anbefaling/godkjenning: Dato/Signatur:
SJAansvarlig:
Omrdeansvarlig:
Ansvarligforutfring:
Annen(stilling):
4
HMSaspekt
Ja
Dokumentasjon,
erfaring,
kompetanse
Kjentarbeidsoperasjon?
Kjennskap til erfaringer/unskede
hendelserfratilsvarendeoperasjoner?
Ndvendigpersonell?
Kommunikasjonogkoordinering
Mulig
konflikt
med
andre
operasjoner?
Hndtering av en evnt. hendelse
(alarm,evakuering)?
Behovforekstravakt?
Arbeidsstedet
Uvantearbeidsstillinger?
Arbeiditanker,kummerel.lignende?
Arbeidigrfterellersjakter?
Rentogryddig?
Verneutstyrutoverdetpersonlige?
Vr,vind,sikt,belysning,ventilasjon?
Brukavstillaser/lift/seler/stropper?
Arbeidihyden?
Ioniserendestrling?
RmningsveierOK?
Kjemiskefarer
Brukavhelseskadelige/giftige/etsende
kjemikalier?
Bruk
av
brannfarlige
eller
eksplosjonsfarligekjemikalier?
Mkjemikalienegodkjennes?
Biologiskmateriale?
Stv/asbest?
Mekaniskefarer
Stabilitet/styrke/spenning?
Klem/kutt/slag?
Sty/trykk/temperatur?
Behandlingavavfall?
Behovforspesialverkty?
Elektriskefarer
Strm/spenning/over1000V?
Stt/krypstrm?
Tapavstrmtilfrsel?
Omrdet
Behovforbefaring?
Merking/skilting/avsperring?
Miljmessigekonsekvenser?
Sentralefysiskesikkerhetssystemer
Arbeidpsikkerhetssystemer?
Frakoblingavsikkerhetssystemer?
Annet
Nei Ikke
Kommentar/tiltak
aktuelt
5
Ansv.
x
VEDLEGGJAPPARATURKORTUNITCARD
Apparatur/unit
DettekortetSKALhengesgodtsynligpapparaturen!ThiscardMUSTbepostedonavisibleplaceontheunit!
FagligAnsvarlig(ScientificResponsible)
Telefonmobil/privat(Phoneno.mobile/private)
TorbjrnNielsen
91897572
Apparaturansvarlig(UnitResponsible)
Telefonmobil/privat(Phoneno.mobile/private)
BrdBrandastr
91897257
Sikkerhetsrisikoer(Safetyhazards)
Rotatingequipment(covered)
Sikkerhetsregler(Safetyrules)
Usesafetygoogles
Ndstoppprosedyre(Emergencyshutdown)
Pushemergencystopbutton
Stopgeneratorbyturningtheemergencyswitchonthepanel
Herfinnerdu(Hereyouwillfind):
Prosedyrer(Procedures)
Bruksanvisning(Usersmanual)
Nrmeste(nearest)
Brannslukningsapparat(fireextinguisher)
Frstehjelpsskap(firstaidcabinet)
NTNU
Instituttforenergiogprosessteknikk
Dato
Signert
Mainentrance
Mainentrance
SINTEFEnergi
Avdelingenergiprosesser
Dato
Signert
6
x
VEDLEGGKFORSKPGRKORT
Forskpgr!
Experimentinprogress!
DettekortskalsettesoppfrforskkanpbegynnesThiscardhastobepostedbeforeanexperimentcanstart
Ansvarlig/Responsible
Telefonjobb/mobil/hjemme
LarsFjrvold/RemiAndrStople
90863846/48496748
Operatrer/Operators
Forsksperiode/Experimenttime(startslutt)
1.okt20111.feb2012
LarsFjrvold/RemiAndrStople
Prosjektleder
Prosjekt
TorbjrnNielsen
TestavKaplanturbin
Kortbeskrivelseavforsketogrelatertefarer
Shortdescriptionoftheexperimentandrelatedhazards
TestofcharacteristicsofaKaplanturbine,whichincludesefficiency,cavitationandrunawayspeed.
Possiblehazardsisrotatingequipment.(covered)
NTNU
SINTEFEnergi
Instituttforenergiogprosessteknikk
Avdelingenergiprosesser
Dato
Dato
Signert
Signert
7
Appendix L
Calibrationreportpressure
AA
CALIBRATION REPORT
CALIBRATION PROPERTIES
Y= -999.01794012E-3X^0 + 500.28151634E-3X^1
CALIBRATION SUMARY:
_______________________________________
Remi Andr Stople
CALIBRATION VALUES
Value [bar g]
Voltage [V]
Best
Uncertainty
Poly Fit Deviation [bar g] Uncertainty [% ]
[bar g]
[bar g]
0.020000
0.041000
0.060000
0.080000
0.100000
0.120000
0.140000
0.161000
0.180000
0.200000
0.250000
0.300000
0.350000
0.400000
0.449000
0.501000
0.550000
0.600000
0.650000
0.700000
0.750000
0.800000
0.820000
0.840000
0.860000
0.880000
0.900000
0.920000
0.940000
0.960000
0.980000
0.999000
1.020000
1.040000
1.060000
1.080000
1.100000
2.037372
2.078004
2.116075
2.155628
2.196224
2.237098
2.276441
2.319024
2.356898
2.397896
2.496808
2.597481
2.697051
2.796616
2.894809
2.997955
3.096235
3.195993
3.296806
3.395519
3.496163
3.596134
3.635903
3.676856
3.715379
3.755681
3.795988
3.835699
3.875976
3.916669
3.955765
3.994152
4.034591
4.075705
4.115461
4.155641
4.195495
0.020242 -0.000242
0.040569 0.000431
0.059615 0.000385
0.079403 0.000597
0.099712 0.000288
0.120161 -0.000161
0.139843 0.000157
0.161147 -0.000147
0.180094 -0.000094
0.200605 -0.000605
0.250089 -0.000089
0.300454 -0.000454
0.350267 -0.000267
0.400077 -0.000077
0.449201 -0.000201
0.500804 0.000196
0.549971 0.000029
0.599878 0.000122
0.650313 -0.000313
0.699697 0.000303
0.750048 -0.000048
0.800061 -0.000061
0.819957 0.000043
0.840445 -0.000445
0.859717 0.000283
0.879880 0.000120
0.900044 -0.000044
0.919911 0.000089
0.940061 -0.000061
0.960419 -0.000419
0.979978 0.000022
0.999182 -0.000182
1.019413 0.000587
1.039982 0.000018
1.059871 0.000129
1.079973 0.000027
1.099911 0.000089
COMMENTS:
0.837284
0.398230
0.265707
0.194368
0.151557
0.123050
0.102859
0.087052
0.076068
0.066751
0.050335
0.039451
0.032351
0.027185
0.023731
0.020863
0.018401
0.018059
0.017224
0.016760
0.016477
0.016396
0.016386
0.016414
0.016433
0.016459
0.016503
0.016562
0.016011
0.016706
0.015872
0.015458
0.016178
0.016971
0.017056
0.016984
0.017197
0.000167
0.000163
0.000159
0.000155
0.000152
0.000148
0.000144
0.000140
0.000137
0.000134
0.000126
0.000118
0.000113
0.000109
0.000107
0.000105
0.000101
0.000108
0.000112
0.000117
0.000124
0.000131
0.000134
0.000138
0.000141
0.000145
0.000149
0.000152
0.000151
0.000160
0.000156
0.000154
0.000165
0.000177
0.000181
0.000183
0.000189
The uncertainty is calculated with 95% confidence. The uncertainty includes the randomness in the calibrated instrument during the calibration,
systematic uncertainty in the instrument or property which the instrument under calibration is compared with (dead weight manometer, calibrated
weights etc.), and due to regression analysis to fit the calibration points to a linear calibration equation.The calculated uncertainty can be used as
the total systematic uncertianty of the calibrated instrument with the given calibration equation.
Appendix M
Torquegaugecalibrationreport
BB
1 av 2
file:///C:/Users/fjarvold.WIN-NTNU-NO/Documents/My Dropbox/Kap...
CALIBRATION PROPERTIES
Type/Producer: Deadweights
SN: Uncertainty [%]: 0
POLY FIT EQUATION:
Y= + 1.18875679E+0X^0 -40.47256213E+0X^1
CALIBRATION SUMARY:
_______________________________________
Remi Andr Stople
26.11.2011 17:13
2 av 2
file:///C:/Users/fjarvold.WIN-NTNU-NO/Documents/My Dropbox/Kap...
CALIBRATION VALUES
Value [Nm]
Voltage [V]
29.506137
39.322023
49.141284
58.961531
68.780167
78.600366
103.150300
127.694920
152.239210
162.056210
171.872310
181.689970
191.510880
201.320130
201.320130
191.510880
181.689970
171.872310
162.056210
152.239210
127.694920
103.150300
78.600366
68.780167
58.961531
49.141284
39.322023
29.506137
-0.697598
-0.927063
-1.173654
-1.409864
-1.657529
-1.902043
-2.506997
-3.098922
-3.701927
-3.954640
-4.196823
-4.437190
-4.617651
-4.875829
-4.950892
-4.727244
-4.507636
-4.271881
-4.031786
-3.791524
-3.166679
-2.544969
-1.922779
-1.689544
-1.437925
-1.188835
-0.941347
-0.695748
Deviation
[Nm]
0.083793
0.612637
0.451744
0.711983
0.506949
0.431061
0.496964
1.084855
1.223980
0.813032
0.827389
0.916746
3.433960
2.794062
-0.243930
-1.001539
-1.934359
-2.210411
-2.309264
-2.402256
-1.657430
-1.039870
-0.408188
-0.788744
-0.423727
-0.162654
0.034555
0.158664
Uncertainty
[%]
3.132472
2.181250
1.606597
1.235138
0.973467
0.787094
0.519086
0.417186
0.401435
0.408801
0.418339
0.429651
0.433284
0.448794
0.459752
0.449321
0.440058
0.429225
0.419470
0.412721
0.421142
0.516460
0.782114
0.963144
1.223328
1.598337
2.171092
3.134673
Uncertainty
[Nm]
0.924272
0.857712
0.789503
0.728256
0.669552
0.618659
0.535439
0.532725
0.611142
0.662487
0.719009
0.780632
0.829786
0.903512
0.925574
0.860499
0.799542
0.737718
0.679777
0.628323
0.537777
0.532730
0.614745
0.662452
0.721293
0.785443
0.853717
0.924921
COMMENTS:
The uncertainty is calculated with 95% confidence. The uncertainty includes the randomness in the calibrated instrument during the calibration, systematic
uncertainty in the instrument or property which the instrument under calibration is compared with (dead weight manometer, calibrated weights etc.), and due to
regression analysis to fit the calibration points to a linear calibration equation.The calculated uncertainty can be used as the total systematic uncertianty of the
calibrated instrument with the given calibration equation.
26.11.2011 17:13
Appendix N
Austegrdcalculations.
CC
Appendix O
Plexiglas
DD
Appendix P
Bearingdrawing
EE
Appendix Q
Torquegaugecalibrationreport
FF
10
11
12
13
14
15
16
17
Qvir
2.643
52
31913
36616.5
90.11
16.62
4703.5
52.09
52.314
90110
2.729
59.4
36616.5
41964.1
90.11
16.62
5347.6
59.23
59.478
90110
2.808
59.4
41964.1
47860.4
90.11
16.65
5896.3
65.30
65.581
90110
2.874
70.9
41860.5
54256.4
90.099
16.67
2.938
75.9
54256.8
61107.1
90.11
16.64
6850.3
75.87
76.962
90110
3.149
93.7
17034.5
25456.5
90.1
16.68
8422
93.29
94.631
90100
3.258
102
25456.4
34659.2
90.109
16.65
9202.8 101.93
102.358
90109
3.775
144.2
34659.2
47669.4
90.1
16.68
13010.2 144.11
144.721
90100
4.251
183.1
32850.5
49351.4
90.11
16.7
16500.9 182.75
183.53
90110
5.086
250.9
49351.7
64443.5
60.106
16.75
15091.8 250.58
251.652
60106
5.898
316.4
38477
57538.9
60.117
19061.9 316.45
317.797
60117
7.215
422.7
42530.3
51044
20.109
16.72
8513.7 422.53
424.327
20109
7.558
452
19294.5
37433.8
40.197
16.85
18139.3 450.36
452.281
40197
8.323
514
17044.2
37683.2
40.108
17.06
20639 513.56
515.771
40108
8.484
527
25517.2
41394.8
30.103
17.19
15877.6 526.39
528.668
30103
8.072
493
16807.1
31697.9
30.104
16.95
14890.8 493.66
495.774
30104
8.168
12395.9 137.31
90099
18
500.8
27528
42623.4
30.103
17.05
15095.4 500.46
502.613
30103
8.285
509.7
42623.4
55457.4
25.106
17.08
12834 510.17
512.37
25106
31086.6
47066.9
30.104
17.27
15980.3 529.77
532.357
19
8.418
521
19
8.530
530
Qvir
2.6
2.7
2.8
2.9
3.1
3.3
3.8
4.3
5.1
5.9
7.2
7.6
8.3
8.5
8.1
8.2
8.3
8.5
52.314
59.478
65.581
76.962
94.631
102.358
144.721
183.53
251.652
317.797
424.327
452.281
515.771
528.668
495.774
502.613
512.37
532.357
600
y=81.49352283x 162.87533823
R=0.99999506
500
400
300
200
100
0
0
10