ConceptualDesignofa DigitalControlSystem for NuclearCriticalityExperiments

bustlingdivisionΗλεκτρονική - Συσκευές

15 Νοε 2013 (πριν από 3 χρόνια και 11 μήνες)

113 εμφανίσεις

L -12758-T
Thesis
UC-714
Issued:April 1994
Conceptual Designofa Digital Control System
for Nuclear Criticality Experiments
St@zen Paul Rojas*
“GraduateResearchAssistantat Los AlamosGroupNE-6.
Los Alamos,New Mexico 87545
ACKNOWLEDGMENTS
The authcr gratefullyacknowledgesall staffandpersonnelof the AdvancedNuclear
Technologygroup,NIS-5,at Los AlamosNationalLaboratoryfor their guidance,
support,and assistanceincompletingthis study.
v
TABLE OF CONTENTS
List of Tables
...
00,.....0.........................................................................................
lull
List ofAppendix Tables
..................................................................................
xiv
List ofFigures
................................................................................................
xv
List ofAbbreviations
.......................................................................................
xvii
Chapter
1.INTRODUCTION
.....................................................................................
1
l.l Background.........
.-.--.~---~--~.~~-~~-~~-~-..-.....................1
l.l.l NuclearCriticality
.......................................................................
1
1.1.2MultiplicationFactor..-------ti--.--.-.--------....................1
1.1.3FissileMaterial......
.------------------------...............2
1.1.4PromptandDe1ayedNeutrons
....................................................
3
1.1.5 Cross Sections.....
.-...-----.--------..-.....................3
1.1.6ModeratorsandPoisons.............................................................3
1.1.70ver/UnderA40deratedSystems
.................................................
4
1.I.8PromptandDelayedCriticality
...................................................
4
1.1.9ReactiviQ----------- ------------------------................5
l.l.lOAtom andNumberDensities
......................................................
5
l.l.ll ScramSystem
............................................................................
6
1.1.12HistoricalPerspective:LACEF
.................................................
6
1.20bjeetives
............................................................................................
8
1.2.1MechanicalSystem
......................................................................
9
1.2.2HydraulicSystem
.........................................................................
10
1.2.2.1Valves
..................................................................................
11
1.2.2.2Accumulator
........................................................................
11
1.2.2.3PressureSwitches
................................................................
11
vii
1.2.2.4Pump
...................................................................................
11
1.3Scope ofStudy
....................................................................................
12
1.4ReportOutline
....................................................................................
12
2.THENUMERICALSTUDY.....................................................................13
2.1 Introduction
.........................................................................................
13
2.2The MonteCM1OMethod....................................................................13
2.3Los AiamosNationalLaboratoryandMonteCmlo...............15
2.41nput File Overview
..............................................................................
16
2.5 Uranyl NitrateSolutionSystem
...........................................................18
2.6 Results
.................................................................................................
21
2.7Summary
.............................................................................................
24
3.CONCEPTUALMECHANICALDESIGN..............................................25
3.1Basic MechanicalRequirements
..........................................................
25
3.2 options for Performingthe Slab
Tanks Experimenton Honeycomb.............................26
3.3 Mechanical DesignConcepts
................................................................
28
4.APPROACHESTOCRITICALEXPERIMENTCONTROL.............,....30
4.1 GeneraIModeofOperation
.................................................................
30
4~DigkalVs.An~ogHtitititio um.~.mm..~~..t~~ti~~~~ti~~..................31
4.3 CustomDigita!Systems
......................................................................
31
4.4 PurchasedSystems
..............................................................................32
5.CURRENTCONTROLSYSTEMHARDWARE
....................................,
33
5.i Introduction
.........................................................................................
33
5.2Test BenchControl System
..................................................................
35
5.3Configuringthe System
........................................................................
37
5.4ComponentSumm~
...........................................................................37
...
Vlll
5.4.11785 PLC-5115Processor
...................e.......................................37
5.4.21771-IBDDCinput Module.................................38
5.4.31771-OBDDC OutputModule...........................38
5.4.41771-IFEAAntdogInputModule.-..--------........................38
j.4n5120VACPowerSupply....~.......~~~..ti.......
..................38
5.4.6 CompumotorA.YLDrive
..............................................................
39
5.5 Control of DCStepper Motors.....................................39
5.6The StepperMotorDrive:Characteristicsand Selection......................41
6.CONTROLSYSTEMEXPANSION.........................................................43
6.1 PLCSystem ExpansionInto Kva I----------------....................43
6.1.11771-ASBRemotel/OAdapter.............................44
6.1.21771-MI StepperControlModule
...............................................
44
6.1.31771-OJPulse OutputExpander..........
.............................44
6.2 Control ofACSynchrcmous Constant SpeedMotom..........................45
7.USINGTHENUMERICALRESULTSTOSIZEHARDWARE...........48
7.1 Introduction.........
.---.--ti.n...------~-.~-...................-...48
7.2 HorizontalSolutionAssembly:the SlabTmks.......................48
7.2.1Requirements
...............................................................................
48
7.2.2A4aximumSpeedAllowable.................................50
7.2.3BasicFormulas....
.-----------------------...................51
7.2.4MaximumPulseRate..............................................52
7.2.5h4inimumSpeedPossible.....................................52
7.2.6Resolution
....................................................................................52
7.2.7RequiredOperating Torque
........................................................
53
7.2.8GearheadReduction
....................................................................
54
ix
7.3 Summary
.................................................................,...........................55
8.CONTROLSYSTEMPROGRAMMING
...............,................................57
8.1 Introduction
.........................................................................................
57
8.2 Basic ProgrammingConcepts.....................................59
8.3 Programmingthe Ana!ogInput Module
..............................................
59
8.4 Programingthe Stepper Motor Modules
.............................................
60
8.5Programmingan IntegratedDrive
........................................................
60
8.6StepperControl Via an IntegratedDrive
..............................................61
8.7 Sequential FunctionChart:File 001
.....................................................
63
8.8 Ladder Logic:Files 002-12
..................................................................65
8.8.1 Rung2:0--ti --------- -------------------------............65
8.8.2 Rung2:1
......................................................................................
65
8.8.3 Rung2:2
......................................................................................
65
8.8.4 Rung2:3
......................................................................................66
8.8.5 Rung3:0
......................................................................................66
8.8.6 Rung4:0
......................................................................................
66
8.8.7 Rung4:1
......................................................................................
66
8.8.8Rungs4:2
....................................................................................67
8.8.9 Rung4:3
......................................................................................
67
8.8.10 Rungs4:4-4:8
.............................................................................
67
8.8.11 Rungs4:9-4:10
..........................................................................
67
8.8.12 Rung6:0
....................................................................................67
8.8.13Rungs 7:0
..................................................................................
67
8.8.14Rung8:0
....................................................................................68
8.8.15 Rung9:0
....................................................................................68
3.8.16 Rung10:0
..................................................................................68
x
8.8.17Rung10:1
..................................................................................68
8.8.18Rung10:2
..................................................................................
68
8.8.19Rung10:3
.........................................................................
..........
69
8.8.20Rung10:4
...................................................................................69
8.8.21Rung10:5
...................................................................................
69
8.8.22Rung11:0
....,.............................................................................69
8.8.23Rung11:1..................................................................................
69
8.9
summary
.............................................................................................
70
9.USERINTERFACESOFIWARE
............................................................
71
9.1Ovexview
..............................................................................................
71
9.2 General Features
..................................................................................
71
9.3Uranyl NitrateExperimentinterface
....................................................
72
10.COSTESTIMATE
...................................................................................
75
!0.1 Labor Cos(s
.......................................................................................
75
10.2Material Costs
....................................................................................
75
10.2.1MechanicalHardware
...............................................................
75
10.2.2Controi
S).slemHardware
.........................................................
76
10.3cost summary
...................................................................................
77
11.CONCLUSIONS
~ 78
.....................................................................................
REFERENCES
...............................................................................................
80
APPENDICES
...............................................................................................
83
A.MCNPCARDSUMMARY
....................................................................
83
B.NUMBERDENSITYCALCULATIONS
..............................................
89
C.CONTROLPROGRAMLISTING
.......................................................
%
D.INPUTFILES
........................................................................................
117
xi
E.NUMBERDENSITYTABLES
..............................................................
125
xii
LISTOFTABLES
Table 2-1:AssumedValues for Number DensityCalculations.....................19
Table 2-2:Mdel Materials................ti.............--....20
Table 3-l:MechanicalFunctionand SolutionFromMostto
Least Essential
.............................................................................
28
Table 6-1:Control ComponentsforKIVA IonHand
.................................
44
Table 6-2:NecessaryControl SystemComponentsfol KJVA.....................44
Table 7-1:MechanicalRequirements
...........................................................
49
Table 7-2:Load Requirements
.....................................................................
49
Table 7-3:SizingCalculationsSummary
.....................................................
56
Table 8-1:S]abTank ProgramFiles....
............................63
Table 10-1:ApproximateHardwareCosts
...................................................
76
Table10-2:ApproximateControl HardwareCosts
......................................
76
Table 10-3:TotalEstimatedCost
.................................................................
77
...
Xlll
LISTOF APPENDIXTABLES
Table E-1:Number Densities
.......................................................................
127
Table E-2:.4tomFractions..........
..............................128
xiv
LIST OF FIGURES
Figure l-l:CriticalityWindow
...........
,,,,.0,..,................................................
4
Figure 1-2:PajaritoSite (TA.l8)..................................................................8
Figure l-3:SystemsOverview.........
...........................9
Figure l-4:HydraulicSystem
.....................................................................
10
Figure 2-1:ADart GameWitha HypotheticalFrequency13istribution.......l4
Figure 2-2:FissionCross Sections for SelectedIsotoPs....-.-...-..-l7
Figure 2-3:SlabGeomet~...........................................................................19
Figure 24:&ff Vs.Air Gap......................................................................2l
Figure 2-5:ReactivityVs.AirGap..............--.-..............22
Figure 3-1:Current ExperimentalConfiguration(Honeycomb)..........,.....26
Figure3-2:AdjustmentsIdeallyAvailablefor SlabTank Alignment
(S!abTankon MovableCart)
77
......................................................A-
Figure 3-3:MicrometerAdjustmentwithaLead ScrewConcept..................28
Figure 3-4:An Optionfor the Mechmic~Configuration...........-...29
Figure4-1:BlockDiagramoftheGeneral Control System......
.............30
Figure5-1:Typical CriticalAssemblyControlDeviceRequirements...........33
Figtire 5-2:Current Control RoomOne Configuration
................................34
Figure 5-3:Programand HardwareTest Station..........-..........35
Figure5-4:Control HardwareatPajarito
......................................................36
Figure 5-5:DCStepperControl Options
.....................................................
40
Figure 6-1:AnticipatedKIVAI Control System
..........................................
43
Figure 6-2:ACMotor Control Circuti.......................................46
Figure 7-1:Stepper Motor VelocityProfile..........................50
Figure 8-1:Signal FlowThroughthe System
...............................................58
Figure 8-2:13asicPLCProgrammingConcepts......................59
xv
Figure 8-3:AnalogInput Programming
.......................................................
60
Figure 8-4:General Featuresof the SFC
......................................................
64
Figure 9-1:ScreenShot of ComputerGeneratedControl Display...............74
xvi
LK)TOFABM.EVIATIONS
PLC:ProgrammableLogicController
SCRAM:quick
decreaseinsystemreactivityduetotheengagementofmechanical
motiondevices
PROM:ProgrammableRead OnlyMemory
PID:ProportionalIntegralDerivative
ITL:TransistorTransistorLogic
AIX:Analogto DigitalConvefier
MCNP:Monte CarloNeutronPhotonCode
LACEF:Los AlamosCriticalExperimentsFacility
DOE:Departmentof Energy
NEMA:NationalElectricalManufacturersAssociation
SHEBA:SolutionHighEnergyBurst Assembly
RAP:RemoteAutorangingPiccometer
SST:StainlessSteeI
BCD:BinaryCodedDecimal
xvii
CONCEPTUALDESIGNOFADIGITALCONTROLSYSTEMFOR
NUCLEAR CRITICALITY EXPERIMENTS
by
Stephen Paul Rojas
ABSTRACT
Nuclearcriticalityis a concernin manyareasof nuclearengineering
includingwastemanagement,nuclearweaponstestinganddesign,basic
nuclearresearch,andnuclearreactordesignandanalysis.As in manyareas
of scienceandengineering,experimentalworkconductedinthis fieldhas
provideda wealthof dataandinsightessentialto the fommlationof theory
and the advancementin knowledgeof fissioningsystems.In light of the
manydiverseapplicationsof nuclearcriticality,thereis a continuinginterest
to learnand understandmoreabout the fundamentalphysicalprocesses
throughcontinuedexperimentation.This thesisaddressesthe problemof
settingup andprogr
amminga microprocessor-baseddigital control system
(PLC)for a proposedcriticalexperimentusing.amongother devices,a
steppermotor,ajoystickcontro!mechanism,andswitches.This
experimentrepresentsa revisedconfigurationto test cylindricalnuclear
wastepackages.
AMonteCarlonumericalstudyfor the proposedcritical assemblyhas
been performedin order to illustratehowresultsfromnumerical
calculationsare usedin the processof assemblingthe control systemand to
corrobor:previousexperimentaldata.This studyinvolvesmodelinga
solutionsystemof uranylnitratein cylindricalgeometry(twocylindrical
slab tanks approximately28 inchesin diameterand4 inchesthick) with
the MonteCarloNeutronPhotoncode writtenat Los AlamosNational
Laboratory,NewMexico.The resultsof this studyyieldedthe sensitivity
effect of varyingthe distancebetweenthe tanks:informationusedas design
criteriato sizevariouscontrolsystemcomponents.In addition,the software
necessaxyfor experimentcontrolwas developed.
In summary,a controlsystemutilizingsomecommondevicesnecessary
to performa criticalexperiment(steppermotor,push-buttons,etc.) has been
assembled.Controlcomponentsweresizedusingthe resultsof a
probabilisticcomputercode (MCNP).Finally,a programwas writtenthat
illustratesthecouplingbetweenthe hardwareandthe devicesbeing
controlledin the newtest fixture.
xix
Chapter 1.
INTRODUCTION
1.1Background
1.1.1
NuclearCriticality
The termcriticality gerlerallyrefers to the studyof fissioningsystems that
approacha state of equilibriumbetweenthe number of neutronsbeing producedand
the number of neutrons dying. Neutrons are producedby the process of nuclear
fission.In a Los AlamosNational Laboratoryreport [15],HughC.Paxtondefines the
fissionprocess as:
The disintegrationof a nucleus (usually,Th,IJ,Pu,or heavier) into two
massesof similarorder of magnitude,accompaniedby a argereleaseof energy
and the emissionof neutrons.Althoughsome fissionstake place spontaneously,
neutron-inducedfissionsare of major interest in criticalitysafety....
Thus,the area of major interest in nuclearcriticalityis in the generationanddeath of
neutrons.The samediffusiontheorythat has successfullybeen appliedto heat transfer
and fluidmechanicshas alsobeensuccessfullyusedto model the process of nuclear
fission.In addition,as will be seen in chapter thee of this document,probabilistic
approacheshaveprovenextremelyuseful in the designand analysisof f~sioning
systems.
1.1.2MultiplicationFactor
The multiplicationfactor is denotedwiththe symbol k andis definedas
follows [9]:
k=
numberoffissions incurrent generation
numberoffissions inpreceding generation
So if the multiplicationfactor is less than one,thenthe systemis called subcritical
and the numberof fissionsoccurringin the systemis decreasingwithtime.On the
other hand,if the multiplicationfactor is greater thanone,then the sys:cmis said to be
supercritical and the number
offissionsincreaseswithtime.Ifthemi!hiplication
factor is equal to one,then the systemis said to be exactly critical ar:dthe number of
fissionsoccurringis constant withtime.In a critical assembly,the raticlof the number
of neutronsproducedto the number of neutronsdisappearingis commonlycontrolled
by the use of poison control rods (material that absorbsneutrons).Typicallythere
are two basic mannersin whichneutronsmayvanish:
1.Absorptionduringa nuclear reaction
2.Leakage fromthe surfaceof the reactor
There are many methodsused to model the fissionprocess:
.diffusiontheory(Fickslaw;differentialequations):a deterministicapproach
usingfinitedifferences
.MonteCarlo:a probabilisticapproachusing statistics

transport theory (integralequations):a deterministicapproachusing discrete
ordinates
The simplest,most fundamentalapproachand the approachused earlyon in nuclear
systemdesignis diffusiontheory.However,the MonteCarlo metho~ioffers improved
accuracyin modelingcomplexgeometriesand it has been the adaptedmethodin the
current study.
1.1.3FissileMaterial
Amaterialis said to be fissileif it is capableof fissionat lowenergylevels (i.e.,
slowneutronswith lowkineticenergy).238U,which is abundant on the planet,is
usedto breed fissileplutoniumby bombardingthe 238Uwith neutrons.Another
2
methodusedto create fissilematerial is the refinementprocess usedto enrich the
percentage
of
235U
innaturallyoccurringuranium.
1.1.4
Prompt and Delayed Neutrons
More than 99 percent of the neutronsemittedin a fissioningsystemare emitted at
the instant fissionoccurs;these neutronsare called prompt. That fractionof a
percent of neutronsthat are emitted at a relativelylong time after the initial fission
event are called delayed neutrons.The averagenumber of prompt and delayed
neutronsreleasedper fissionevent is giventhe symbol
V.
1.1.5Cross Sections
Across sectionis an experimentallydeterminedparameter with units of cm2 which
indicatesthe probabilityof a certainevent occurring.Differenttypes of cross section
data used in nuclear engineeringincludescattercross sections,absorptioncross
sections,or fissioncross sections.Cross sectionstake on the units of barns where
1barn = 1X10-24cm2.In essence,the nuclearcross sectionis the effective cross
sectionof the nucleus that a neutronsees whenit is travelingnear the nucleus.The
total cross sectionis the combinationof the fissioncross section,absorptioncross
section,scatter cross section,etc.,and is a measureof the probabilitythat any type of
interactionoccurs whena beamof neutronsimpingeson a target composedof many
nuclei.
1.1.6ModeratorsandPoisons
Amoderatoris a substancewhich tends to slowdown (thermalize) neutrons.
TypicaJmoderatorsincludewater and polyethylene.Apoisonis a substancewhich
tends to absorbneutrons.Typical poisons includeboron and cadmium.Poisons may
be of the burnable type [14] which meanstheir absorptioncross sectiondecreasesas
time progresses(thus increasingthe reactivityof the system).
3
1.1.7 Over/Under ModeratedSystems
A
systemissaidtobeovermoderatedif themultiplicationfactordecreases(i.e.,
critical mass increases)withdecreasingdensity(i.e.,increasethe amount of
moderator).On the othel hand,if as the density is decreasedthe multiplicationfactor
increases(i.e.,critical mass decreases),then the systemis said to be
undermoderated, This informationis typicallyillustratedin a plot of muhiplication
factor vs.density(or equivalently,a plot of critical mass vs.hydrogento uranium
ratio).
1.1.8Prompt andDelayed Criticality
Delayedcriticalityis usedto describethe stateof a fissilematerial ic which
the multiplicationfactor is unity fromthe contributionof both delayedand prompt
neutrons.Prompt criticalityis a termusedto describethe state of a fissile material in
whichthe multiplicationfactor is unitysolelyfromthe contributionof prompt
neutrons.Thus there is a window in betweendelayedcriticality(the steady-state
condition) and prompt criticality. This windowis giventhe symbol ~ and it follows
that the fractionof fission neutrons that are prompt is 1-$ This can be seenby
consideringthat a k of unity is due to both prompt and delayed neutrons;therefore,
+
)
p<o (Do
k=l
k = 1/(1-~)
DelayedCritical
Prompt Critical
p=o
P=P
Figure 1-1:CrMcality Window
IIf notfor thjswindow,bombswouldbe
rathereasytobuild
whilenuclearreactorswouldbe
more
difficult;asit is,thereverseistrue.
4
to get
rid of the delayed neutrons,we subtract the reactivityamount P.One may
questionthe validityof subtractingthese two values sinceat first sight they appear to
be different units;however,at closer inspectionit is apparentthat the units are dentical
since the reactivity~ is simplythe change in k whichhas beennormalizedin ccordance
with valueof unityat delayedcritical.Therefore,the multiplicationfactor considering
only prompt neutrons is ( l-~)k.Whenthis value is unity,the systemis said to be
prompt critical since a multiplicationfactor of unityis reachedwithonly prompt
neutrons.
1.1.9Reactivity
Reactivityis definedas the percentagethe systemis abovedelayedcritical:
k1
P~
=
Thus,negativereactivityindicatesa systemthat is belowdelayedcritical whilepositive
reactivityindicatesa systemthat is abovedelayedcritical.Typically,the reactivityis
expressedin dollars (or fractionsof a dollar:cents) by usingthe conversionfactor:
~ reactivity= 1dollar.The value ~ is the differencein reactivitybetweendelayed
critil dity and prompt criticality.Thus,if a systemis promptcritical,then p=~
(remember,if the systemis delayedcritical,thenthe multiplicationfactor is one,and
the reactivityis zero).Typically,the texmadding reactivity is usedwhen the system
is alreadyat delayedcritical (i.e.,k=l,
p=O).
1.1.10
AtomandNumberDensities
Typically,number densitiesare used for MonteCarloinput files to define the
materialcharacteristics.The numberand atomdensitiesare definedas follows:
5
N -
~NA
= atomdensity = atomsI cm3
M
atoms
numberdensity= (N)(1x10-24cm2/barn)=
barn--cm
where:
NA= AvogadrosNumber
M= molecularweight
1.1.11ScramSystem
Ascramsystemrefers to an electro-mechanicalsystemwhichproduces a prompt
decreasein reactivitydue to physical movement.For example,a scramfor the uranyl
nitrateexperiment wouldconsist of quicklymovingthe two fissileslab tanks apart
fromone another to quicklydecrease the multiplicationfactor.TypicaI1yboth
automaticand manual scramsystems are designedintocritic?l experiment apparatus
(the automaticmechanismsare coupled to particle detectors located aroundthe
experiment).In addition,for the experiment proposedin this study,an additional
gravityassistedscrammechanismmay be incorporated.Althoughthis document
focusesonly on the primarymanual scrammechanism(a hydrauliccylinder),it should
be notedthat two such redundant scramsystems will also be incorporated:one
automatic,and one gravityassisted.
1.1.12HistoricalPerspective:LACEF
The urgencyof World War 11that spurredthe ManhattanProject also demanded
that a site be establishedat the Los Alamos National Laboratorywhich wouldserve as
an area for experimentalwork as well as isolatethe populationfromradiationin the
event of a criticalityaccident [12].The area chosen was in Los Alamos Pajarito
Canyonand came to be knownas Pajarito Site. Before 1947,critical experiments
were performedat the site by hand.This changed when,in 1946,Louis Slotinwas
6
killedas a result of a componentof an assemblyslippingintoa more reactiveposition
producinga superprornpt-critical
pulseofradiation.
As
aresult,thesiteestablished
muchmoreexactingrules governingthe operationof critical assemblies,one of which
was the policyof performingmost critical experimentsremotely.Suchexperimentsare
nowperformedin what are called kiva..:2
buildingshousingcritical experimentsthat
are controlledfroma remote location.The control systemoutlined in this document
will serve as the maincontrol systemfor the original kiva which is nowreferredto as
KIVAI.Figure 1-2displaysa plan viewof PajaritoSite.Today LACEF(Los Alamos
Critical ExperimentsFacility)housesthe most significantcollectionof critical
assembliesin the westernhemisphem.The assembliesthat may be operatedat LACEF
can be dividedinto threecategories:
.BenchmarkAssembliesare configurationscontainingpreciselyknown
componentsthat have interchangeableor adjustablefissilecores and reflectors.
.AssemblyMachinesare generalpurposeplatfoms intowhichfissile,
moderating,reflecting,and control componentsmaybe loaded for short range
studiesof the neutronicpropertiesof the materials.The assemblymachine
describedin the followingsectionfalls into this category.IKis worth noting
that assembiymachinesdo not actuallycontainfissilematerial;they only
manipulateit.

SolutionAssembliesallowcritical operationswithfissilesolutions.The
experimentproposedin this study is a solutionexperimentmountedon an
assemblymachine.
2~c HoPi Indianname
for
a round
ceremonialchamber
7
1.2Objectives
The main goal of this project is to create a numericaImodel of a fissioningsystem
(critical assembly) using a well-establishedcomputercode and then bringtogether a
systemfor controllingthe assemblybasedon the resultsof the numerical study.
Althoughthe numericaIresults will be specificto a certaincriticalexperiment,the
control systemwill be inherentlygeneral and mayreadilybe usedto control other
experiments(specifically,an experimentinvolvinga stepper motor and hydraulic
system).In order to address the specificdetails that must be consideredwhen sizing
and selectingcontrol components,a completesizinganalysis for the proposedsystem
is givenin chapter seven.An introductionto the proposedexperiment follows.
T
-.
..
SJEbo
..”’
Kivoi..
Bldg
.PL-23

.~ ‘ T
\
.
‘,-..,
RYjcxito
,
/‘,”.,.,.::-,;.,,.
.
.“..
‘-
.
Rood
..,.
..
.
...
..
./!?‘,~,““.::-,,.
.
..-
..

‘.
Fuprito
-n “.
“..
...
.,..
..,.
..
...
“----- y$yx..” -,,,.
\
..
.....

.* -..
,\
~.,
.
..
,.
.’
....,.
-.
.,
.
.
iiiiii
-~ 1
.
.,
..
.-,
....,.
.
,.
.,.
....

Q
,.
..
....
..:
.’
.-..
:.
.
.
,,
.
=?%*
,,
,::
::
.....,.
:
..:j/-
.
● ✎✎ ❞✎✎✛✌ ✞ ✌ ✎✌ ✌ ✚
8
Figure 1-2:Pajatio Site (TA-18)
1.2.1MechanicalSystem
Ihesystemchosenfor studyis a uranyl nitratesolutionsystemin cylindrical
geometry.The experimentconsistsof two slab tanks filledwithhighlyenriched
uranyl nitrate,U02(N03)2,that must be pushedtogether remotely.The general
mechanicalsetupfor achievingthis is shownbelow[11].Note that the detaileddesign
of mechanicalcomponents(e.g.,support brackets or translationtables) has been
omitted in order to focus attentionon the two systems of primaryinterest froma
control systempoint of view:the stepper motorfleadscrewcombinationand the
hydraulicsystem.
PoisonMakrial (orGravityAssisted-
Movabkslabrank
sWr8mrorilrad~
o
Ilydradic cylidcr
(Supplrrtfnma Orniaedfwchrily)
(SupyrortfmrresomiucdfordAy)
, 3EE5L
SideView
Front View
Figure 1-3:Systems Overview
Ahydrauliccylinder is used to pushthe movablecart towardthe stationarytank.
Once the air gap betweenthe two tanks is decreasedto a preset distance,the hydraulic
cylinder is shut down and the final approachto critical is made with a stepper motor
and lead screwthat drive a linear translationtable (uponwhichthe movingtank sits).
The stepper motorlleadscrewcombinationis used in favor of the hydraulicsystemfor
final closure in order to increaseresolution(as will beeomeevident in the following
chapters,sucha systemis extremelysensitiveto small changesin the air gap).This
document focusesonly on the control of the stepper motorfleadscrewand hydraulic
systems.Acompletespecificationof all the syste~~ necessaryto performthe
experimentwouldinvolvedesigningthe frameworkfor the secondarygravityassisted
scramsystem(general concept illustratedin Figure 1-3)as well as all of the detail
designfor componentssuchas mountingbracketsor mechanicalinterfaces(e.g.,the
leadscrewh.ranslationtable interface).
1.2.2HydraulicSystem
Figure 1-4belowshowsthe simplifiedhydraulicsystemcircuit that is used to
control the hydrauliccylinder portionof the assembly[10].Basically,the pressure
differentialmthe cylinder is controlledby runningline pressurethrougha series of
threenormallyopen or normallyclosedcontrol valves.s
v
I
No.
ACC.
1
N.C.1

IN
Figure 1-4:Hydraulic System
3Nom~]y open:whenpowerisofc,
thewdveisopen
Normatlyclosed:whenpowerisoff,thevalveisclosed
10
1.2.2.1Valve$
Atotal of three valves are used to regulatethe pressure throughout the system.
When N.O.is open and N.C.1and N.C.2 are closed,the systemis scrammed.When
N.O.is closed,either N.C.1or N.C.2 can be openeddependingon the speeddesired
(speedis set by needle valves associatedwith N.C.1and N.C.2).When all of the
valves are off,the hydrauliccylinderis inactive.
1.2.2.2Accumulator
The accumulatorserves as a power source to the hydrauliccylinder in the case of
power loss.If power is lost,the normallyopen valveconnectingthe accumulatorto
the scramline will open andN.C.1and N.C.2will close,causingthe assemblyto
automaticallyscram.
1.2.2.3Pressure Switches
Two pressure switches,identifiedby PS1 and PS2,are used to check the
accumulatorpressurerange and the scramline pressurerange.Whena Iimit is
reached,the switchis activatedon.
1.2.2.4PumD
The pump providesthe pressurenecessaryto move the hydrauliccylinder and
pressurizethe accumulator.For the purposesof the control systemthat will be
discussedin the followingchapters,it is assumedthat control of all hydraulicsystem
componentsrequires 10-60volt DCpower.Therefore,as will be made clear in the
followingchapters,the only type of control systemoutput deviceneeded is a DC
output module to send the appropriateDC voltageto the desired valve or pump.
11
1.3
scow
of Study
This studycompletesthe preliminaryconceptualworknecessaryfor conductinga
criticalexperimentinvolvinguranyl nitratein cylindricalgeomet~ usinga horizontal
split tableand,in essence,proves that suchan experimentis feasibleby outliningthe
hardwareand softwarenecessaryfor conductingthe experiment.The scope of this
study includesthe conceptualdesignof the main mechanicalcomponentsas well as the
systemneededto control these components.Detaileddesignof mechanical
componentsis outside the scopeof this studyand is takenas a given.
1.4 Report Outline
This report begins with a numerical studyof the physical systemto be controlled.
Next,the mechanicaldesignrequirementsare outlinedand the conceptual mechanical
designthat fulfillstheserequirementsis illustratedandbrieflydiscussed.The
hardwareneeded to control this physical systemis the next general topic addressed.In
these chapters ( chapters four,five,and six),the approachto control and the control
systemcomponentsare discussedand introduced.In chapter seven,the results of the
numerical study are used to size hardwarecomponentsfor the experiment.In chapters
eight andnine,the softwareused in conjunctionwith the control hardwareis
discussed.Finally,in chapter 10,a cost analysisfor performingthe experiment is
presentedand discussed.
12
Chapter2.
THENUMERICALSTUDY
2.1 Introduction
Numericalstudies(as well as any availableexperimentalresults)will give an idea
of the characteristicsof the physicalsystembeingcontrolledand yield suchdesign
informationas:howfast the systemcan be movedtogether,what kind of torque is
needed,and what kind of control systemhardwarek needed to satisfy all of the
specifications.Used for this purpose,the MonteCarlo method(MCNP) is briefly
discussedhere and numerical results for the experiment introducedin chapter one are
generated.The data generatedhere is used to size a stepper motor and the peripheral
electrical&vices neededfor the uranyl nitrateexperiment.An admittedlysimple
modelof a uranyl nitratesolutionsystemin cylindricalgeometryhas beencreated;
however,at this level of design,it is sufficientto corroborategeneral trends in data
froma previouslyconductedexperimentinvolvinguranyl nitrateand similar
geometry.This numerical studyis the fmt step in the followingprogressionof events:

MCNPstudies to determinevalues neededfor the sizingof basic control
systemcomponents

control systemcomponent sizing:stepper motor,stepper motor drive

componentinstallation

programmingand debugging
2.2 The Monte Carlo Method
The methodof solvinggoverningequationsby statistical accumulation(playinga
game) is usedin manyareasof scienceand engineeringincludingconductiveand
radiativeheat transfer,turbulence,and most pertinent to this paper,neutronphysics.
The MonteCarlomethodinvolvesa physical processthat inherentlyexhibits some
formof randomness.The terms random walk or Markov chain often arise in the
discussionofsuchprocesses.Strictlydefined,a
Markovchain is a series
of
sequential
events for whichthe probabilityof each succeedingevent is uninfluencedby prior
events [16].Fromthis definitionarose the termrandomwalk:an expression
describingthe randomnesswith whicha dmnk man ambles down the street.This
randomwalkphenomenonis present nearlyeverywherein nature:

the directiona bundle of photons is emitted duringa radiativeheat transfer
process can be modeledas a randomprocess

the generationand death of vortices in turbulent fluid flowcan be seen as a
randomprocess
.the life and deathof a neutronduringthe fissionprocess can be modeledas a
randomprocess
Thus,any seeminglyrandomprocesscan be modeledas long as one critical pieceof
experimentaldata is available:the frequencydistributionof the event or events.For
example,in a game of darts,the frequencyof a dart hittingat some radial positionon
the dart board maybe graphicallydisplayedby plottingfrequency(i.e.,numberof
times the dart hits) vs.radial position on the dart board.
14
Figure 2-1:A Dart Game Witha Hypothetical FrequencyDistribution
Typically,thefrequencydistributionismathematicallymanipulatedintermsofmore
convenientfunctionssuchas the probabilitydensityfunction or the cumulative
distributionfunction. For instance,if the frequencydistributionis denotedby f(~)
then the probabilitydensityfunctionis foundby normalizingthe frequencydistribution
(i.e.,dividingby the area under the j(<) curve):
p(g)=,
f(~)
Jf(E)4
u
So if randomnumbersare chosenfor& the resultingdistributionmust resemblethat
definedby the equationshownabove.In other words,we may create a probabilistic
model that repeatedlyplays the same game utilizingrandomnumbersand physics.
However,those randomnumbs must agreewith the probabilitydensityfunctionthat
is observedin physicalrealityanddefinedby the generalequationshown.
2.3 Las Alamos National Laboratoryand Monte Carlo
The Monte Carlo methodemerged fromwork done at Los AlamosNational
LaboratoryduringWorldWar II and the inventionof the methodin general is
attributedto Fermi,Von Neumann,and Uhun.This initial work on the Monte Carlo
methodeventuallyled to what is nowknownas MCNP:the Monte Carlo Neutron
Photoncomputer code [6].MCNPis a general purpose Monte Carlo code that can be
used for neutron,photon,or coupledneutron/photontransport and is generally
recognizedas one of the best codes in its class since it incorporatesstate-of-the-art
physics,data,and mathematicalmethods.
MCNPfollowsthe entire lifeof manyparticlesfromlife to death;the game
15
(fissionprocess) is startedby a sourceof freedstrengthspecifiedby the user.When
run in neutrontransport only mode,there are four possibleevents a neutroncan see
duringitslifetime:
1.Neutron scatter
2.Fission
3.Neutron capture
4.Neutron leakage
MCNPsimplyfollowsthe entire lifeof each particleby randomlyselectingone of the
possibleevents (and,if scatter is selected,a rdndomdirection) basedon a set of rules
(physics)and probabilities(transportdata) governingthe processesand materials
involved.As the lifetimehistoryof more and more neutronsis followed,the
distributionof neutrons is better known.Typical fissioncross sectiondata for 235U
and other fissilematerialsis shownon the followingpage [15].In additionto fission
cross sections,other cross sectiondata is used by MCNPincludingscatter,absorption,
and capture data.As seen on the plot,235Uhas a muchhigher probabilityof fission
occurringwhen the neutronsare in the thermal (i.e.,ambient temperature)region
rather than the fast (i.e.,greater than ambient temperature)region.In betweenthe
thermal and fast regions,the probabilityfor fissionfluctuatesgreatly;becauseof this,
nuclear systemsare commonlyreferredto as being in one of two distinct states:fast
or thermal (the systemis forcedto be thermal or fast by design).
2.4 Input File Overview
Atypical MCNPinput file is composedof four major sections;each sectionbeing
composedof a number of input cards (horizontal rows of data).The four major
sections are:
.Geometryspecificationcards
.Surfacespecificationcards
F----:
//
.’
.’
17

Importancecards

Materialspecificationcards

h4CNPmode cards

Tallycards
Eachof these sectionsis discussedin moredetail in Appendix A.
2.5 Uranyl Nitrate Solution System
Asimplemodel of a uranyl IIitratesolutionsystemin cylindricalgeometryhas been
created.The geometryused in the study is shownin Figure
2-3
and the [naterialsused
in the model are definedinTable 2-2.This geometrywas createdby defininga total
of 13surfaces,nine of whichwereplanes normal to the Yaxis;three were cylinders
centeredon the Yaxis,and two were spheres centeredabout the origin.The cells
were createdby definingthe appropriateintersectionand unionof surface senses as
explainedin AppendixA.The twocylindricalslabtanks were sumoundeciby a
sphericalshell of six inchconcrete;air was placed insidethis sphere and aroundthe
tanks.The importanceof the spherical regionof interest was assigneda value of one
v$ile everythingoutsidethat regionwas assignedan importanceof zero.The numtxr
densitiesused for the uranyl nitratesolutionwerecalculatedassumingthe data shown
in Table 2-1 (see AppendixBfor the calculations)[4] while the number densities for
the remainingmaterialswere takenfrompublishedliterature[17] (in reality,atom
fractionswere entered;but as explainedin AppendixA,this is equivalent to number
densities).While not to be used in practice,the concrete shell was used in order to
crudelymodel anyreflectioneffects fromsurroundingwalls.
18
28.5”
DIA.
I
27.ti”DIA.
Spherical shell made from two
spheres
H
w+
4.48”Lh’i@ Nitmc.487”30%
Bond
Po]y
H
Figure2-3:Slab Geometry
Table2-1:Assumed Valizesfor Number Density Calcukztions
?-
SolutionConcentration:405.2 gll
NominalOveidl Densityof Uranyl Nitrate:1.558g/cc
SolutionAcid Content:.32Molar (HNO~)
Enrichment:93.1 % 235U
5.9%
23%
1%
23%
8 for 2%J =.M7
2(I
Fissionproducts such as 236Uwere not includedin the model since the affect on
the fissionprocess has beenassumednegligible.In addition,other extraneous
elements that might be present in solutionsuch as Fe,Na,or Al were not modeled.
The general goal of this studywas to determinethe systemsensitivityby varyingthe
air gap betweenthe two tanks,therebyrevealinglimitationsand characteristicsthat
must be consideredwhenassemblingthe control system.
Table2-2:Model IUatcwiaic
-  -..- -.-------.......-
Material Composition
Number Density
(atoms/barn-cm)
UranylNitrateSolution
234
1.W33X1O-5
235;
9.67I9x1O-4
238u
6.052
1X
10-5
H
.054439385
0
.03600846
N
.00226872
StainlessSteel
c
.000317
Cr
.016471
Mn
.001732
Fe
.06036
Ni
.006483
Si
.001694
Concrete
H
1.4868x10-2
c
3.814x10-3
o
4.15I9X1O-2
Ca
1.1588x10-2
Si
6.037x10-3
Mg
5.87x10-4
Fe
1.968x10-4
Al
7.35X1O-4
Na
3.O4X1O-4
(table continues)
Air
Ni
.784
0
.211
Ar
.005
30 wt%Berated Poly
H
5.19X10-2
c
2.O6X1O-2
B
3.54x1O-2
2.6 Results
Figures 2-4 and 2-5 showthe results of the air gap study.All values of I+ff shown
are at the 68%confidencelevel.The ~ff valuesshownhave beencalculatedby
combiningthree separateestimationtechniquesthat MCNPemploys (collision,
absorption,and track lengthestimates).The followinggenera!characteristicsof the
model are noteworthywhen a comparisonwithexperimentaldata is made:

model does not include fissionproductsor other elements such as iron,
aluminuin,or sodiumthat couldbe present in solution

modelgeometrymaydiffer slightlydue ambiguitiesin someexperimental
dimensions
Keff
Vs.Alr Gap
9
9
-00~
MONP
Rooulm
flmsnt
Rooults
0.0s0
}
r
~
9
-:----
7
-----.+...
0.980
1
.--%:+..-
l
....>.>=
0.070
0900
L-m-T&A
0.ss o.4a 0.6s o.ea o.7a
o.8a o.sa t.oa I.w
AIR OAP (Inohoo)
Figure2-4:Keg VS.Air Gap
21
I?eactlvlty Vs.Alr
C3ap
0 MONP ROSU180
-
•lmll~f espcrtmon8 nDowlts
4.QOOO----
,
4

%
- 1.
a4a - 2/808
r-2 - 0 000
0.0000 - -
.i
-=-...,,
.=
-1.0000
}
..,
““”-\-.,
f
-2.0000
‘=...
---:--
..>..
=
E-
,L------ ‘-
-s.0000 -d
I
‘\
““-..........
4
/?,
‘\\:;K-.
-4.0000 -=
b’ ‘-”- ‘1
-i
y -
-o.B2e - 2.890X r-2 - 0.6s4
-6.0000
!
I
I
1
I
I
,
1
r
I
1
0.s3 0.43 0.63 0.03 0.7s O.ea 0.0s
1.0s 1.13
AIR ~AP (Inuhoo)
Figure 2-5:Reactivity vs.Air Gap
The results shownare withintwo percent of similarexperimentalresults for uranyl
nitratein slab geometry.Althoughthis error might
ii~kkdly
appear small,it is
enormous in terms of criticality.For example,a 1.426percent emor in ~ff translates
into approximatelytwo dollarsdifferencein re~ctivity:a differencebetweenthe system
being well subcriticalVS,the systcmbeingdelayedcritical;refer to air gap=.45 on
Figures 24 and 2-5.Whilethe experimentalreactivityat this separationdistance is
ze]o,the reactivityfromthe MCNP study is calculatedas:
= -2.07 dollars
TJ
Duetotheextremeeffectsofsmallchangesin!+~,anymodelthatwillbeusedto
predict criticalitymust preciselyaccount for the effect of each material in the general
vicinityof the fissioningsystem.This was not the case for the exjxximentaldata
referencedhere;the model does not preciselyaccount for all materialspresent in the
actual experiment.So why model the systemin the first place?Althoughthe
informationshown in the two figurescannot (and absolutelyshouldnot) be used for
criticalityprediction,it canbe used for sizingthe control system.This is because,
:d[hw]~k
the MCNPresultr arc offset h:-I!heexperimentalresults.the slope oi the
MCNPdata is in approximateagreement withthe experll.i~,....!ts.In fact,the
Monte Carlo results yield a 4.5 %conservativeestimate for the slope of the plot.This
meansthat the maximumrate at whichthe assemblywill be allowedto move (as
definedby the conservativeestimate)will be slowerthan the actual maximumspeed
allowable(as definedby the experimentalresults).In this case,we were luckysince
there were experimentalresults with whichto compare the numerical results.If an
experiment were to be performedwithout such a luxury,we would need to
painstakinglyensure that the model resembledthe physicalsystemas accuratelyas
possibleby modelingeach materialpresent in the experimentexactly(e.g.:exact
dimensions,exact solutioncomposition).
Nominallyfour to five millioncollisionswembankedwithapproximately140,000
neut.mnsgeneratedper run.In assessingthe results of these Monte Carlo calculations,
a distinctionis madebetweenprecisionand accuracy.Precisionis the uncertaintyin
the average valuecalculatedby the programitself.Accuracy,on
the
other hand,is a
measureof howclose the calculatedresult is to truth.Jr other WO-k,results may be
precise and not veg accltr:‘f r.’onv.
~!’iymi,- be:’ ‘( TaItZ’and ‘0[
very
prcCi~.In
Ilnc]udingthe~ssiblc presenceof unvented
radiolyticgws
-)3
thiscase,sinceweknowthatthetruthfulvalueisapproximately10.89centshmbased
on previousexperimentswith the uranyl nitrate packages,we may concludethat the
model predictionof 11.38centshmn is very accurate(.49cents/mmconservative)
despitethe lackof highlydetailedmodeling.The precisionof the data shownis at the
68%confidencelevel;in other words,if 100runs are performed,then 68 of the values
(one standarddeviation)will fall withinthe precisionbars shownon the plot.While
the data shownis not extremelyprecise,the numberof cycles performed(50cycles
with a nominal 3000 neutronsgeneratedper cycle) is consideredlarge enoughto result
in adequateaccuracy.Therefore,the numerical results shownhere corroboratethe
slope measuredfromexperimentand suggest a moreconservativeestimatefor sizing
the control components.This is the estimate that will be used in chapter eight.
2.7 Summary
The Monte Carlo studies performedindicate that approximately 11.38cents of
reactivityare added for each millimeterof closure betweenthe tanks.This value
conservativelycorroboratesthe experimentallymeasuredvalueof 10.89cents per
millimeterand,in general,providesan estimate for howmuchexcess reactivityis
present in the system(i.e.,howsensitivethe systemis to smaJldisplacements).As will
be shownin chapter seven,these valuesset the maximumallowablevelocity,the
stepper motor and drive type,the lead screwpitch,and the gear reductionratio.
24
r
Chapter3.
COWEPIUAL MECHANICAL DESIGN
3.1 Basic Mechanical Requirements
While it is not the goal of this studyto providethe detaileddrawings for
manufacturingexperimentalapparatus.the mechanicaldesignaspect of the project
must neverthelessbe addressedon a conceptuallevel.The fundamentalmechanical
requirementsfor the experiment includethe following:
.Two frames must be de~ignedthat will holdeaehof the twocylindrical tanks.
Reflectionfromthese framesmust be heldto a minimum;therefore,a minimum
amount of materialtiwt s!ill providesthe greatest stabilityand reliabilitymust
be used.Sincethe meanfreepath of aluminumis relativelylarge,this material
is somewhat transparent to neutrons and would serve well for the application.
.Atranslationdevice must be designedor purchasedthat will be drivenby a
stepper motor upon whichthe movingslabtank will rest.
.Adjustmentdevices must be designedor purchasedin order to adjust the
relativeslab tank positions.
The mainrequirementscan be summarizedby the needto secureboth slab tanks on
eachcart withthe maximumadjustmentcapability(positionalfine tuning) and the
minimumneutronreflection.Sincethese goals are contradictoryin nature,a number
of design iterationswill be necessary;a conceptualdesignto beginthe process is
offeredin this chapter.
As seen in Figure 3-1 on the followingpage,the honeycomb structure is a lattice
of extrudedaluminumtubes that were placedtogether for a previousexperiment on
the horizontal split table referredto as Honeycomb.
25
26x 243.in.quxm Al Ida
h
Figure3-I:Current ExpenmentalConfiguration(Honeycomb)
This is the cumentconfigurationof the experimentalassembly;as shown,the extruded
aluminum tubes are held in place by four clamp %x.The next section discusses the
options availablein fit~.ingthis split table for the slab tanksexperiment.
3.2 Options for Performing the Slab Tanks Experiment on Honeycomb
There are two options that may be consideredwhen approachingthe conceptual
mechanicaldesign.The first is to mount the tankson the existing split table with the
honeycombstructures in place.The second option (Figure 1-3)is to take the
honeycombstructureoff of both tables and designspace frames for each tank from
scratch (as opposed to retro-fittinga design to the existing honeycombstructure).The
latter of these two options is preferablesince positioningof the slab tanks maybe
accomplishedin a much moreprecisemanner usingthis approach.As the numerical
studyclearlyindicated,the systemis exfremelysensitiveto small changesin tank
position;therefore,relativelysmall toleranceson the order of.001 inchmust be
imposedon the mechanicaldesign.Ahhoughthis avenue is more costly,it provides
for greater experimentalaccuracysince positionaladjustmentsmayreadilybe designed
intothe structurefromthe outset.The advantageof removingthe Honeycomb
material is particularlyevident whenconsideringthat locatingany singlepoint within
the lattice is achievedat best with large uncertaintiesdue to the structures
26
constructionandthe originalexperimentalintent:to mockuprelativelylargecritical
systemswith inherentlyloosetolerances.As seen in Figure 3-2below,the tank would
ideallyhaveadjustmentcapabilitiesintheX,Y,Z,THETAX,andTHETAZ
directionsin order to ensureproper tank alignment and increaseexperimental
flexibility.
While
this goal is ideal,cost and fabricationconstraintspracticallylimit the
adjustmentfeatunx to a minimum:the Xand Ydirections.
z
THETA Z
-x-
x
t
z
I
Figure 3-2:Adjustments IdeallyAvaikble for Sikb TankAlignment (Sikb Tankon
Movable Cart)
Adjustmentin the Ydirectionallows for final closure via a stepper motorfleadscrew
attachedto a translationtable while adjustmentin the Xdirectionoffers fine
adjustmentto ensure the tanks are not offset with one another.The remaining
adjustmentaxes shownmust be fixedaccuratelyby the mechanicaldesignitself or
adjustedwithshims.
27
3.3
MechanicalDesign Concepts
Table 3-1 displays the generaldesign philosophyfromthe most essential,
basic
functionsat the top,down to the less essential but no less desirabledetailed functions
at the bottom.Includedin this table are likelyhardwaresolutionsto the desired
functions.
Table3-I:Mechanical Functionan
FUNCTION
Tank securityand stability
Roughtank movementover a relatively
Ionzdistance
Fine tankmovementover a relatively
short distance
Tank alignmentadjustmentin the X
direction
Tank alignmentadjustmentin the Z
direction
Rotational adjustments
SolutionFromMost to Least Essential
SOLUTION
Rigidaluminumspaceframe
Hydrauliccylinder
Stepper motor/leadscrew
Micrometerhead/leadscrewdevice
(figure 3-3)
Micrometerhead/Ieadscrewdevice
(figure 3-3)
Rotationaltable/micrometerhead device
Micrometer
x
Figure 3-3:MicrometerA~ustment witha Lead ScrewConcept
28
Figure3-4 belowillustratesone possibIemechanicalconfigurationfor achievingthe
minimumrequirements.In this option,two translationtablesare essentiallystackedon
top of one another in order to provide for the Xand Ytranslation.Amicrometer head
is used as the means for adjustment in the Xdirectionwhile a stepper motor is usedto
achievefinal closure
intheYdirection.Figure
3-4presentsone feasibleoptionfor the
mechanicalhardwareconfigurationand is not intendedto be exclusiveof other
configurationsthat maybe equallyviablesuch as different spaceframedesignsor slab
mountingmethods.The followingchaptersdiscusshowsucha systemwill be
controlledremotelyand assumea givenmechanicalconfiguration.
XTr+ioII
Table
)
,
,
I
(
I
Imuknl
r@bMAd.saEw
\
r
7
———
lid
pJ
I
0
1
[
MOVABLE CART
1
I
~...
1
Front View
SideView
Figure 3-4:An @ion for the Mechanical Con#igumtion
29
Chapter4.
APPROACHES TO EXPERIMENT CONTROL
4.1 General Modeof Operation
After determiningthe basicsystemparameters(namely,the systemsmechanical
designand changein reactivitywith linearposition),we are in a positionto consider
the control system.Typically,the control systemusedfor critical experimentswill not
operate using PIDtype automaticcontrol and will not require the extremelyfast
response found
inhighperformance
servo-typecontrol systems [8,19].Instead,the
systemwill incorporatesimplefeedbackto verifythe state and positionof output
devices;this is the simplestsystemthat achievesconsistentand reliablemechanical
control.Althoughfeedbackis present in suchsystems (optical encoder,
thermocouples,etc.),it will generallynot be used for proportional type control of an
experiment (as indicatedby the dashed line in Figure4-1 below).This restrictionis
dictatedby current DOEenforcedtechnical specifications.
t
I
I
I
I
I
!
L
————————.
--Eiizl-----

I
I
I
I
I
I

Figure4-1:Blbck Diagramof the General Control System
30
4.2
Digital Vs.Analog
In the past,control of
criticalexperimentremoteassemblymachineshasbeen
achievedthroughthe use of hard wiredcontrol systems.Althoughsuchsystems have
provenreliable,the advent of the powerful,dependable,lowcost,microprocessorhas
madedigital systemsa verylucrativeoption.Uniikehardwiredsystems,a digital
control systemoffers the flexibilityof quicklyandeasilychangingthe controller
characteristicsby simplyre-writingthe control program.For example,if the estimate
for the slopecalculatedin the previous chapter is later foundto be too conservative,
the closurevelocitymaybe easilychangedsimplyby re-writinga fewlinesof code.
This flexibility,combinedwithincreasedpower andreliability,has propelledthe digital
control systempast its hardwiredcounterpart for critical experimentcontrol
applications.There are two avenuesthat might be pursuedwhencontrollinga system
digitally:a customdesignedsystem,or an off-the shelf purchasedsystem.
4.3 CustomDigital Systems
The first option involvesdesigningthe entirecontrol systemarounda single
microcontrollerchip.Typically,suchchips containon-boardmemo~,timers,ports,
and other support functionsthat would normallyrequire separateICchips.
Customizedmicrocontroller-basedsystemsoffer the followingadvantagesto the
potential user:
.control of the systemand softwareat the machinelanguagelevel
.increasedflexibilityto meet exoticdemands
On the other hand,customizedsystems involvethe followingdrawbacks:
.the systemis harder to maintaindue to its increasedcomplexity

to construct such a systemrequires a PROMburner and other additional
peripheralhardwareinvestments
31

debugging,maintenanceandconstructionrequiresspecializedknowledgeand
experience
Thus,for specializedapplicationsdemandinga largedegreeof flexibilityin control,a
customizedsystemmaybe appropriate.
4.4 PurchasedSystems
The other alternativeis to purchasea pre-manufacturedmicroprocessorbased
system,typicallyreferredto as a ProgrammableLogicController (PLC) [13],froma
vendor.This alternativeis preferredin the nuclearcriticalityarena becausein-depth
documentationand verificationof control systemreliabilityis greatlysimplified.
Unlikemost customizedsystems,pre-purchaseddigital systemsoffer relativelysimple
programmingsoftwareand allowfor moreefficient andthoroughmaintenance.For
these reasons,it was decidedto purchasea PLCsystemfromthe Allen-Bradley
corporationrather th creatinga customizedcontrol system.This control systemis
introducedin the followingchapter.
32
ChapterS.
CURRENT
CONTROLSYSTEMHARDWARE
5.1
Introduction
Currently,Control RoomOne at LACEFis fittedwith a digital control system
that was originallyinstalledto control the SHEBA(SolutionHigh EnergyBurst
Assembly)experiment(see Figure 1-2).Specifically,the systemis manufacturedby
the Allen-Bradleycorpmtion and incorporatesconvenientsystemmodularitywith a
simplegraphicalprogramnu
nglanguage.An overviewof the typicalphysicalsystem
to be controlledis seenbelowin Figure 5-1.
MICROPROCESSORCONTROLPROGRAM
HYDRAULICSYSTEMS
SCRAMSYSTEMS
MECHANICALSYSTEMS
Valves
Motor
Limit switches
Pressureswitches
Control switches
Pneumaticcylinders
Valves
Stepper motor
Encoder/Resolver
Table
Mountingbrackets
Figure5-1:T~ical Cri&al Assembly ControlDevice Requirements
The main goal of this sectionof the studyis to assemblea functionalcontrol systemon
a test bench that will allowfor programdevelopmentand hardwaretesting without
intrusionon the current systemin Control RoomOne at LACEF.With such a system,
the followingtypes of devices maybe tested:
.DCstepper motors (with two different approachesto their control as discussed
later)
.digitallycontrolledday contact devices
.ACsynchronousccnstant speed motors
The devices shownin Figure 5-1are basic to controllingmanytypes of critical
assemblies.An assemblythatrequiresrotationalortranslationalmotionwillemploy
one or all of these devices in additionto the peripheral componentsthat formthe
backboneof the control system;these peripheralcomponents makeup the test bench
control systemthat is describedin the followingpages.In chapter seven,expansionof
law
control
Plonselectcx
Modulerack
/
Closeddrcuit
TV
and inct&tmGghts
Additic)nd
1
.
Ma
(X@smOn
34
I-4
I
I
!
t
!
I/
I
I
I
v
L
I
I
RAPunit
-1
JoysWk
\
\
Con?ldlel
v.-
I
Nw
Ci%%2.s
am
\
Cmlputergenefoted
stat-up
controlcS@oy
pink%
L
\
Figure 5-2:Current Control RoomOne Configuration
the control system
thatisalready
in use in Control RoomOne to includeexperiments
in IUVAI is discussed.This chapter,as well as the next,formthe groundworkfor the
eventual expansioninto KIVAI (note in Figure 1-2that the SHEBAbuildingis
separatefromIUVAI;the SHEBAbuildingcurrentlyemploysan Allen-Bradley
systemwhileKIVA1doesnot).
Figure 5-2
onthepreviouspageshowsthecurrent
configurationof Control RoomOne.
5.2
Test Bench Control System
Shown in Figure 5-3 is a schematicof the control systemtest station that has been
setup at Pajaritosite.This systemhas beencreated fromspare parts availablefromthe
SHEBAsystem;when it is requiredto expandthe current Control RoomOne system,
parts fromthis systemmay be
used.The
pwpose of the setup that currentlyexists is to
provide a platformfor on-line progr
amrningand testingthat can be used to write and
debug ladder logic programswhichwill eventuallybe uploadedto the processor
UJCALW3CNA!LW 8
SUJTS.4GROWS(1~1.A2B)
REMOTE W2CNASW%
12SWTS.4GROUPS(IT31-A3BIA)
o
1
2 3 4 5 6 7 8
9
10 I J2
O- 1771ASBREMOTElK3ADAPTER
!-l!=>WY IKJMODULESWTS
J2->120VACKJWER
SUPPLY
0-178
I=>177LJBDIX JNPUlMODULE(I030V)
2* 1771OBDm~~ MODULE
(I06W)
o
6!&swmsmFMm0R
3-> t731-lFEAANALIXJ JNPUTMODULE
46=> EMPTY
10MmuLEsm
7* 120VACRXVSRSUPPLY
=“ REMOTEmCHAssJsJSNOTwlREDTOLKALCNASSIS
SINCE NO
LX OU13Wl MODULES ARECURRWLY AVAILABLE FORTNLSPURPOSE
Figure5-3:Programand Hardware Test Station
currently
used in Control RoomOne.In addition,such a test platformserves asa
center for the test and evaluationof hardwarethat might eventuallybe used for the
KIVAI system.This separatesystemallows for on-lineprogr
ammingand hardware
.
,!,
familiarizationwithoutintrudkgonthesystemcurrmtlyinoperationinControlRoom
One.AsseeninFigure5-3,theassemblyusesadriverwhichcontrolsasteppermotor
attachedtoa turntable;thisisoneof twooptionsfordrivinga steppermotor.
5.3
ConfiguringtheSystem
Systemconfigurationinvolvedsettingdipswitchesonthetopandbottomofthe
PLC-5/15processormoduleand theI/Ochassisbackplane[1.2,3].Theseswitch
settings
deftnedparameterslike:Aeracknumber,theI/Ogroup(usedwhen
programming
thePLCsystem),thetransmissionrate,andthepowersuppliespresent.
TheCompumotorAXdrivewasconfiguredbysettinga singledipswitchassembly
nearthebottomof thecase[7].Thisswitchassemblydefinedwhatkindof motor
(i.e.,currentrequirement)wastobedrivenandtheaddressof theAXdrive(tobe
usedwhenwritinga programtobestoredinthedrive).
5.4ComponentSumrn8ry
5.4.11785PLC-5/15
Processor
Thisis theheartof thecontrolsystem.Thecontrolprogram(writteninladder
logic usingthePLCsoftware)is uploadedtothePLC-5PROMfromthepersonal
computerviathe 1784Kl/Bcommunicationboard.Ashorttestprogramutilizingtwo
switchesandLEDindicatorsavailableontheDCoutputmodulewaswrittento
validatethateachcomponentseeninFigure5-3isoperable[1].ThePLC-5module
seinesasthemaincontrolmodule;thesolepurpw a;thePCis asa programming
terminalanddatamonitorstation(usingControlViewsoftware).ThePCdoesnot
directlycontrolanything.
37
38
h
5.4.21771
IBDDC InputModule
TheDCinputmoduleacceptsinputfromcommonDCdevicessuchasswitchesor
pushbuttonsandcommunicateswiththeprocessorviathebackplane(printedcircuit
board)located
onreartheI/Ochassis.DCinputandoutputmodulesrequirea
separateDCpowersourceas seeninFigures5-3and5-4.Forthe KIVAI expansion
discussedinchzptersix,a DCinputmodulehasbeenplacedintheremoteI/Orackin
ordertoprovidefor theoptionof a scrammechanismlocatedinsidetheKIVAitself.
5.4.3177]
OBDDC OutputModule
TheDCoutputmoduleisthe logicalcounterparttotheDCinputmoduleand,like
all othermodulesinthechassis,communicateswiththeprocessorviathebticlcplane.
Theoutputmoduleis wiredtodevicesrequiring10-60VDCpowersuchas lights,
seven-segmentdisplays,horns,or valves.
5.4.41771
IFEAAnalogInputModule
Asitsnameimplies,theanalogmoduleacceptsanalogsignalsfromrealworld
devices.Forcontrolof acriticaiassembly,thismodulewillacceptanalogsignakfrom
thepotentiometersthatmakeupthejoystickcontroldevice(atypicaljoystickis made
usingoneor morepotentiometers).Theladderlogicprogramwilldictatethedetails
of theinterfacebetweenthejoystickandtheoutputdevice.Thismoduleis simplyan
analogtodigitalconverter.
5.4.5120
VACPower Supply
Thisprovidespower(convertingvoltagefromACto DC)viathebackplanetothe
moduleslocatedintherack.Apowersupplyof thistypeis requiredforeachJ/O
chassis.
5.4.6 CompumotorAXLDrive
ThecompumotorAXLdrivecombinestwofunctionsinonepackage.First,it acts
asa pulsesource.Thisisbasicallya sourceof squarepulsewavesthatis usedto tell
thesteppermotorhowfar andfastit shouldrotate.Thesecondfunc!ionof the
Compumotorpackage
is asa driveor translator.
Atranslatortakesthepulsetrain
generatedbytheprogrammingcommandsandtranslatesi:intothevoltagenecessq
forenergizingthemotorwindings.Themotordrivemaybeprogramn,sdviaanRS-
232 port
froma regularPC.Theprogramiswrittenina simplelanguagedeveloped
k
y compumotor and is storedin the AXLSresident PROM(up to seven such
programs may be stored).
AstoredprogrammaybeexecutedviatheRS-232
connection,or,it maybeexecutedviathePLCcontrolsystem.To runtheprogram
viathePLCsystem,onesendsa pseudoBCDnumber(3bitsinsteadof four)tothe
SEQI,SEQ2,andSEQ3linesof theAXLdrive.Dependingonthecombinationof
thethreesequences,oneof thesevenprogramswillberun.Althcughthismethodof
steppermotorcontroloffersa greatdealof powerandflexibility,thereis another
optionto steppercontrolthat ispreferred.Thisoption(a separatepulsesource)is
di.scusscdinthenextsection.Theprogrammingthataccompaniestheintegrateddrive
isdiscussedinchaptereightaswell.Inthefuture,detaileddevelopmentsinclude
addingfollowingcomponentstothesystem:
.encoderfeedbackmodule
.powercontactrelaymodule

stepperpositioningmodules,translator,andnewsteppermotor
5.5Controlof DCStepperMotors
Often,an
experimentalapplication
requiressmall,precisechangesinmotion;thisis
typicallyachievedwitha DCsteppingmotor.ADCmotorusesdirectcurrentpower
‘#
inordertoinduceacurrentontherotorof themotorthatcausestheoutputshaftto
rotate [5].This
is accomplishedbytheuseof a commutatorandbrush.
The
:ommutatorisacylindricalobjectuponwhichthebrushesridetochangethedirection
of
current
inthewindingsinordertokeeptherotorrotating.Abrush is a pieceof
conductivematerialridingonthecommutatorwhichconductscurrentfromthepower
supplyto therotorwindings(stator).ApermanentmagnetDCmotorusesa
permanentmagnetasthestatorinsteadof windings;windingsareusedfortherotor.
Therearetwomainoptionsavailablefor thecontrolof a standardpermanent
magnethybridsteppermotor.Thefirstoptioninvolvespurchasinganintegratedunit
froma manufacturer(liketheCompumotorAXLdrivediscussedintheprevious
chapter).Thesecondoptioninvolvespurchasinga standalonepulsesourcefromthe
El
PLC
Chassis
L
Pulse/kmslator
StepperMotor
source
OptionI:CombinedPulseSource/Iranslatof
ProgrammingLoadSplitBetweenthe
PLCandPulse/hanslator
PukeSource
TranslatorStepperMotor
PLC
Chassis
Option2:SeperatePulse
Source,Translator
ProgrammingLoadConsolidated
tothePLC
Figure5-5:DCSte~er ControlOptions
40
PLCvendorandsendingthatpulsetrainthroughaseparatetranslator(i.e.,splitup
thetranslationandpulsegenerationfunctions).DCsteppermotorsarecontrolledby
varying
thepulseratesenttothemotortranslator(ordrive).Thetranslatorand
motoraresizedaccordingtothespecifictaskandthepulserateis controlledthrough
software.Asdescribedbefore,a translatorbasicailytakesthepulserate(generated
extemailythistime)andtranslatesthepulserateintotheappropriatevoltage
necessaryto movethemotor.Thefirstof thesetwooptionswasrealizedonthetest
bench;however,it is preferabletoconsolidateallof theprogrammingandcontrolto
onlytheremotePLCchassis(insteadof progr
amrningbothanintegrateddriveandthe
PLCsystem).Forthisreason,a stepperpositioningassemblythatsplitstranslationand
pulsegenerationfunctionshasbeenusedintheControlRoomOneexpansionlayout
discussedinchaptersix.
5.6The StepperMotorDrive:Characteristicsandselection
Thesteppermotortranslatorcontainsthelogicnecessaryto translatethepulses
generatedbysteppermotorcontrolassembly(indexer)intothecorrectvoltageneeded
bythesteppermotorforshaftrotation.Thesteppermotorwillrotateonestepfor
eachpulsereceivedfromthecontrolleranditsperformanceisve~ closelycoupledto
thetranslatorsperformance.If a lowperformancetranslatoris usedinconjunction
witha highperformancesteppermotor,theoverallperformanceisdefinedbythe
lowestperformingdeviceanda costlymismatchexists.Translatorsmaybe purchased
tooperateina numberof
ferentmodes;typically,full,half,or micro-stepping
mode.Inthefullstepmode,thetranslatorwillstepthemotor
onefullstepforeach
pulsereceivedwhileinthehaifstepmodethemotorwillbesteppedone-halfofitsfull
step(0.9degreesif themotorusedhas1.8degredstep).Microsteppingreferstoa
featurtfoundonmoreexpensivetranslatordeviceswhicharecapableofsteppingthe
motorbyas littleas 1/125of thefull stepperpulsereceived.Thetranslatoralso
containstheciruitrynecessarytoopticallyisolatethehighervoltageareafromthe
controlsystem.
42
Chapter6,
CONTROL SYSTEM EXPANSION
6.1PLCSystemExpansionIntoKIVAI
Figure6-1showsacontrolsystemdesignthatmaybe usedfor thecurrentPLC
controlsystemexpansionintoKIVA1.Inparticular,theconfigurationshownmaybe
usedtocontroluptothreesteppermotorsandwouldbeidealfora varietyof
experimentalsetupsincludingtheuranylnitrateexperimentonthehorizontalsplittable
honeycomb.
LOCAL
REMOTE
\
\
REMOTEIK3CHASSIS:12SKY2S.6GROUPS(1771.A3BI)
kpm
qf-k
AC
Powerin
Iwiw&
1
k!?
\
\
LOCALUOCNA.SSIS:6SLOTS,4GROUPS(1771-A2B~
o I 2 3 4 5 6 7 8 9 10 1} )2
b 4
t
I
,
ACPawt
o=> 1771-ASBREM(YIELIOADAPTER
AL.
G!!9
I => 1771-IBDDCINPUT MODULE(IW30V)
2=> 1771-0BDCCOLIlT~MODU3X (10.6OV)
3=> 1771-MI
STEPPERCONTROLLERMODULE
4 -6=> 1771-OJPULSEOU1PLtrEXPANDER
7-II m EM37Y LfOMODULESIXJIS
12=> 120VACPOWER SUPPLY
C!i!!9
I
iha!!
\
\
\
I
\>
‘lo SsEm BUILDINGREMOIERACKS
0=> 1785flc-~is PROCFAWOR
\
1e IT?l-IBD DCINP~MODULE(10.30V)
\
2=> 1771OBDDCOWPW
MODULEIKMOV)
3= > 1711-lFEA ANALOGINPUl MODULE\
4.60 EMPTYUOM.C;JULESLOTS
7=> 120VAC POWERSUPPLY
\
Figure6-1:AnticipatedKWAI ControlSystem
This layoutcontainsthe followingadditionalmodulesthat havenot beenusedin the
43
testbenchsetupshowninchapterfive.
6.1.1 1771-ASBRemoteI/OAdapter
The remoteI/Oadapterservesas a communicationlinkwiththe localPLC-5
processor and communicateswith all of the other modules in its rack via the
backplane.
6.i.2 1771-A41
Stepper ControlModule
Thismodulecontrolsthepulseoutputthat is sentout tothesteppermotor.
6,1.31771-OJ
PulseOutputExpander
Onepulseoutputexpanderis requiredforeachsteppermotortobecontrolled.One
steppercontrolmodulemayaccm.modatea maximumof threesuchmodulesand
thereforethreesteppermotors.Thepulseoutputexpanderis responsiblefor
communicatingthepulseinformationfromthecontrolmoduletothesteppermotor
itself.InordertofabricatethesystemshowninFigure7-1,thefollowingpartsmust
beusedfromtheteststationthat iscurrentlysetupinbuilding30:
Table6-1:ControlComponentsfor KWAIon Hand
ITEM FUNCTION
1771-IBDDCinputmodule
provideforlocalDCinput
177I-OBDDCoutputmodule provideforlocalDCoutput
1771-A3B112slotI/Ochassis houseremotel/Omodules
1771-ASBremoteI/Oadapter providecommunicationforremoterack
1771-A2B8 slotI/Ochassis
houselocalI/Omodules
4
?naddition,thefollowingcomponentswhichare
notinstockareneeded:
Table6-2:NecessaryControlSystemComponentsfor KIVAI
ITEM
FUNCTION
1771-OBDDCoutputmodule
ProvideforremoteDCoutput
PLC-5/15to 1771ASBtwinaxial Enableconnectionbetweenlocaland
cable(either1770-CDor custom remoterack
made)
(maximumdistanceis 10,000feet)
(tablecontinues)
44
11771-QAstepperpositioning 
assembly
Slo-Synsteppermotor
S1o-Syntramlatorlpowersupply
(steppermotordrive)
Generatepulsesourcefor steppermotor
control
Convertelectricalenergyintouseful
mechanicalenergy
Translatedigital informationfrompulse
source to motor shaft rotation;provide
powerandisolation
The remainingitemsnecessaryto realizethe systemshowninFigure6-1 are in stock
onsite(i.e.,PC,powersupplies,andconnectors).
6.2Controlof ACSynchronousConstant SpeedMotors
Whenanapplication
requireslessdemandingprecisioninsmallmovements,an
AC
constantspeedmotorisoftenusedasanonloffdevicetofulfilltherequirements.For
example,whilea steppermotormightbeusedtopreciselypushtwotankstogether
througha leadscrewinoneexperiment,anACconstantspeedmotormightbeusedto
rotatesafetydrums(containingpoison)inandout by 180°inanotherexperiment.An
ACmotorisdifferentiatedfroma DCmotorbythefactthat,for ACmotors,rotor
currentsarenot inducedbya commutatorandbrushes.Instead,currentis inducedin
therotorconductorsbythestatorschangingmagneticfield(sometimesreferredtoas
inductionaction);thus,thistypeof motoroperatesbysettingupa rotatingmagnetic
fieldona rotor.Typically,therotorisof the squirrelcage typeandtherotating
magneticfieldissetupbysendingsinglephasesof muhiphaseACpowerintospecific
polewindings(woundupconductivewire).Wheneachphaseof thepolyphaseAC
poweriscorrectlydirectedto itspolewinding,a magneticfieldis setupwhichtendsto
rotatethesquimelcage.Bycontinuallykeepinga rotatingmagneticfieldappliedin
thisway,thecage(rotor)is keptcontimmllyrotating.Typically,one,two,or three
phaseACpoweris usedto rotatethecage.Thedifferencebetweenthespeedof the
rotatingmagneticfieldandthecageitselfistermedtheslip. Whenthedesignof the
45
motorallowsthecageto lock intothespeedof therotatingmagneticfield,theslip
iszeroandthemotoris saidtobeoperatingat synchronousspeed. Whenthismode
of operationisreached,themotoroperatesat a constantspeedthat isdirectly
proportionaltothefrequencyof thepowersupply.
ACsynchronousconstantspeedmotors,specificallySlo-SynSSseriesmotors,may
becontrolledwithaPLC systembyutilizingtheRCcircuitshowninFigure6-2[1$].
Figure6-2:ACMotorControlCircuit
Theconstantspeedmotoris turnedonandoff andchangesdirectionbyselectingthe
appropriateswitchshowninFigure6-2.Thepurposeof theresistorandcapacitoris
tochangethesinglephaselinevoltageintothetwophasevoltagethat is appropriate
forenergizingthemotorcoils.Althoughanactualthreewayswitchcouldbeusedto
controlthemotorstatus,criticalexperimentsmustbeoperatedremotely;therefore,a
powercontactoutputmodulethatcanopenandcloserelaysmaybeusedto select
fromamongthethreeswitchpositions.Basically,thisty-pcof moduleallowsremote
relaystobeturned
on
andoff viathecontrolprogramstwed inthemicroprocessor
46
memory(locatedina localI/Ochassis).Oncethemotorhasreacheditsfinalposition,
a limitswitchcanbe usedtoturnthemotortoitsoff position.Inthisway,useof
opticalencoderf=dbackiseliminatedwithoutlosingfinalpositionalinformation.
47
Chapter
7.
USINGTHE NUMERICALRESULTSTOSIZE HARDWARE
7.1Introduction
Nowthatthegeneralcontrolsystemcomponentshavebeenestablished,thenext
stepis tosizea motorandtranslatordrivetothespecificjob of runningtheuranyl
nitrateexperiment.Typically,thisinvolvesperfoting thefollowingtasks:
.determinewhattypeof motionisdesired:typicallyeitherlinearor rotational

approximatethemaximumandminimumallowablevelocitiesbasedon
criticalitycalculationsor measurements

determinewhattypeof translatorwillbeused
.determinethemotortorquerequired
Thisgeneralprocesshasbeenperformedfortheuranylnitrateslabtankexperiment.
Thegenera!approachis todefineallof therequirementsbasedonthesystem
informationandthenmatchthemanufacturedhardwaretothosespecifications.
7.2HorizontalsolutionAssembly:theSlabTanks
Recallthegeometryandphysicalsetupdiscussedinthepreviouschapters.The
mechanicalgoalis topushtwocylindricaltankstogetherwithoutaddingmorethan
fivecentsof reactivitypersecond.Thiswillbeachieved,inpart,throughtheuscofa
steppermotorandleadscrewthatwillpushthetablesupportingoneof thetanks
!owardtheotherstationarytank.
7.2.1
Requirements
Inorderto preciselychoosethecorrectsteppermotor,it is firstnecessarytodeeide
whatgeneralperformancecharacteristicsmustbesatisfied.Theblocksof information
onthefollowingpageserveas a datasummaryareafora steppermotorthatwillbe
drivinga 61kgmassthrougha leadscrew.Notethatcertaincharacteristicsarelisted
48
as
to becalculated.Thesevalueswillbedeterminedlater.Figure7-1showsa
graphicalrepresentationofthedesiredvelocityprofile.
MECHANICAL
REQUIREMENTS
Application
Weight
Mounting
Resolution
Accuracy
Environmental
Damper
(toreducesettling
timeforeachstep)
bads orTerminals
Gearing/Lead
Screws
Table7-1:MechanicalRequirements
VALUE
TranslationalTable
Weight=61 kg
NEMAtype
flange
200stepsper revolution(1,80/step)
.0540/step(3%of resolution)
Becauseof lowpoweroperation,therearenospecial
environmentalconsiderations

None
8leads
Thompsonleadscrew
Table7-2:LoadRequirements
LOADREQUIREMENTS
VALUE
Torqueat speed
tobecalculated
Inertia
tobecalculated
MaximumSpeed
tobecalculated
49
(seefigure1)
Acceleration/DecelerationRequired
.5secto maxvelocity
.5sectostandstill
Singlesteptimenxponsedesired
2.5ms
Pulse
Rate
Max,Velocity/PulseRate
.5
sec
Position
Figure7-I:StepperMotorVekbcityI+ofle
7.2.2MaximumSpeedAllowable
Inchapterthree,theabsolutevalueof theslopeof thereactivityvs.air gapplot
wasfoundtobeapproximately11.38centshm,whichequals2.89$/inch.The
maximumreactivityinsertionratefora Class11reactor(reactorthatgoesdelayed
critical)asspecifiedbythetechnicalspecificationis fivecentshec.Withthis
information,wemaycalculatethemaximumpermissiblelinearspeedof thesystemas
follows.Thisvaluewillbeusedwhenwritingthecontrolprograminchaptereight.
(maxah~wablereactivity)(constant)
maxlinearspeed = 
slope
(5
=)( 1$ )
00 cents
=
2.89~
inch
= 1.038=
50
Duetothisextremelyslowspeed,a gearheadreductionwillbe neededinorderto
increasethenominaloperatingspeedofthemomandavoidlowpowerresonance
eff=ts.
7.2.3
Basic Formulas
Therearethreebasicformulasthatareoftenusedinchoosingcontrolsystem
componentsfora translationaltypesystemusinga leadscrew.Inthecasepresented
here,
equation(/) maybeusedtosolveforthegearheadoutputspeed,equation(2)
yieldsthemotor
shaftspeed,andequation(3) maybeusedtocalculatethepulserate.
G/f =
(P)(hnearSpeetf)[=]qvm
(1)
where:
GH=gearheadoutputshaftspeedinrpm
P= leadscrewpitchinthreaddkch
LinearSpeed=speedinincheshinute
Next,thespeedat whichthemotoris rotatingis foundwith:
AUS= (x)(
GH)[=]rpm
(2)
where:
MS=
motorshaftspeedinrpm
x=theplanetarygearratiox:1
Finally,themaximumpulserateallowablecanbefoundusingtheequation:
PSEC=
~sp:;s)[=$r:::
(3)
51
where:
PSEC=pulseskc
output by the
stepperpositioningassembly
SP= steps/pulse
DS=
degreeshtep
7.2,4MaximumPulseRate
Usingthevaluedeterminedforthemaximumspeedallowable,themaximumpulse
rateof thesteppercontrolassemblymaybe calculated
maximumgearheadoutputshaftspeedis foundtobe:
GH= IOrpm
Usingequation(1),the
Usingequation(2),themaximummotorspeedis foundtobe:
MS=
520rpm
Finally,usingequation(3),themaximumpulseratethatthestepperpositioning
assemblymaysendoutwithoutexceedingthereactivityinsertionlimitis:
PSEC=~60
pUk/H
7.2.5MinimumSpeedPossible
Theabsoluteminimumspeedat whichtheunitcanberuncanbedeterminedby
consideringthesmallestpulseratethecontroldeviceiscapableof generating.
Assumingtheminimumis 1pukkc,useof equations(1) throttgh(3) yielda
minimumspeedof.0025irhin.Sincethispulserateisextremelylowandmight
resultinmotorvibrations,a minimumpulsemteof 500pulseskc isassumed.With
thisvalue,theminimumlinearspeedis foundI.Obe.15inhnin.
7.2.6Resolution
Theresolutionmaybecalculatedbyconsideringtheminimumnumberof single
pulsesthestepperassemblycansendout inaonesecondperiod.Thustheresolution
isjust theminimumvelocity(inkc) multipliedbythetotaltimetogeneratethefinite
52
numberof pulses.Fora minimumpulserateof 1pulsehec,the
resolution
is.00004in.
Fora minimm pulserateof
500pulseshec,theresolutionis.0025in.
7.2.7RequiredOperatingTorque
Thelast stepis tocalculatetheamountof torquerequiredtomovethe61kgmass
at the
desiredvelocitybyappiyingthebasicequation:
(4)
EM oments=
Ia
where:
1=totalsystemmassmomentof inertia
a = angularacceleration
Thetotalsystemmassmomentof inertiais foundbyaddingeachseparateinertia
component:
Itotal = Ieq + IsCrew+ I rotor
(5)
Where Ieq
istheequivalentinertiaofthemassbeingmoved.Therotorinertia
iSfound
frommanufacturespublishedMeratursfor a selectedsteppermotorto be2.5lb-in2
whilethescrewandequivalentinertiasarefoundwiththefollowingequations[18]:
It~(lb- in2) =
weightx.025
screwpitch
where:
weightis inpounds
screwpitchisinthreadsperinch
IK,W(lb- in2) =diameter
x
lengthx.028
(6)
(7)
when:
screwdiameterandlengtharebothininches
.028
isthenominaldensityofsteellb/in3
Next,thetotaltorquerequiredis foundusingthefollowingequation:
~O,u,(oz- in)= ~Cc,p,+T,,iC,iO~
= I,da,xe*ciV Xconversionfactor
time
force
+
screwpitch
x q x
conversionfactors
(8)
where:
velocity[=]rad/sec
time[=]sec
force[=]in/lb
q =screwefficiency
Notethevelocityusedhereis thevelocityinradkx of thegearheadoutputshaft.An
additional10oz-inof torquehasbeenaddedtoaccountformiscellaneoustorques
fromgearheadinertkoroffset loads.AssumingtheconstantsshowninTable7-3,
equations(4) through(8) yielda requiredtorqueof 8.05in-lb.
7.2.8
GearheadReduction
Thefinalstepis tochoosethetranslator-motor-gearheadcombinationthat will
satisfj all of therequirementssummarizedabove.Thetotaltorquerequiredto
accelerate!heloadwithouta gearhcadattachedwascalculatedas 8.05in-lb.!Vhenthe
gearheadisadded,it willhavetheeffectof multiplyingthenominalmotortorqueto
increasetheoutputtorque:
O~tputTorque= Motor Torquex Eficiency x Ratio
(9)
54
I
Soforamotorwitharatedtorqueof 1.87in-lbat 750rpm,theactualoutputtorque
(k
to
thegearheadis approximately79.475in-lb(assuminga gearheadefficiencyof
.&f);this
obviouslyexceedstherequiredamountandwouldsufficeforaccelerating
ittt
loadtothemaximumvelocityin.5 seconds.Pull-in torqueisthemaximum
to;
que thatcan acceleratea load
withoutlosingsynchronismwiththepulserate.The
~.otorchosen musthavea pull-intorquethatis greaterthanthecalculatedvalueof
38.97Oz-in.in additionto
providinga torque multiplication,the gearhcadreduction
was used
inordertoreducethespeedof thesteppermotorandavoidpossible
vibration
problemsthat maybe present at timlowor high
endof thetorquevs.speed
(steps/see)curve.Thetranslatorchosenforthisapplicationis theSlo-Syn430-PT
packagedtranslatordrive.Thisdriveoperatesinhalfstepmode(0.9 degktep)and
utilizesa bipolarchopper,2phasesteppermotordrive.
7.3Summary
Allof thecalculationsdiscussedhavebeenplacedina spreadsheettoexpeditethe
processof parametricvariation.TheresultsareshowninTable7-3onthefollowing
page.Thecalculationsperformedinthissectionhavedefinedthebasicvelocityprofile
andmotortorquetobe usedforthesteppermotorintheuranylnitratesolution
experiment.In.5 seconds,the 17.85inllb(minimum)steppermotorwillacceleratethe
masstoa maximumlvelocityof 1.038irthnin.Thiscorrespondstoanaccelerationof
17.3rev/sec2toa pulserateof 3461pulsedsec.Withknowledgeof thenumbers
summarizedinTable7-3,alongwiththegeneralprofileshowninFigure7-1,weare
nowinthepositiontoprograma systemutilizingeitheranintegrateddriveor a
separatepuke source.
55
Table
7-3:
SizingCalcuk#ionsSummary
]INPUT PAJ?AMTERS
I
I
1
1
2
3
4
5
6
7
8
1?3
11
12
13
14
15
16
17
18
OUTPUT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15-
16
17
18
TcXdtmuera
red(hlb)
:17.85CD7{
Slcpec#rea3ivity@cjt($fin) ~
\
~
2.89
Mm.
lnsdkn 17de(omtshw),
.Lea3Sc.mwPitch(th/ln)
15
10
.Mcx Pdse Rde(pha&x) ~
~ mm
.Plmetay Gea Reckticn
f?cilo~
50
.Trcmldx St@ngh&&(st~@se) ~
0.5
.st~ng Ma
Resd@cn
(ct@t@:1.8
.DesircdR~dutkn @n).
O.ml
.Tdd W@@t (M)!:
4347
,Tknetor~vdodty(s ec);~ 0,5
.Saavdawta
(h),:~ 1.5
.SmV1.=@h On )...~.
48
.Scxmvf3tch(thjn)..
5
,Screweffidw
0,9
.~dg
rdcf Inertla(lbin%?):
2.5
:Frldicn facatosli&w@@t(@ ~
6
.Mdrwn
PUseRate,@ sesAmcj.~
.MSC
Taw(cx-ln)..
5
FfWAJvE”TERS
,fvb.
Lima
Spedkwed(ln~n) 11,038062
jhkx.G@ i+edOu@t Shcft Spged(rp)!10.38062
~hkx.,Ivk3tcxSpe@(r~)!
I 519.OJ1l
NIx Ma Speed(cb@sec).
] 3114,187
.MCX.Pdse Rc#e@ses~@..3~”2~~
.Mn.tvkia Speeg(r~)
,Mn.
Ivlda Sp3ed(d3gk) ~:m
,Mn.G&x Hea50ut@ Shd Sp=d(rm).
;.5
,Mn.Lima Spf3edQnmn).
.0.15
.Mn.Linea Spe@(in&c)
~
o.Qq?5
,Eqjvcht
hhs
InErtia(lblnW):;
4.-&l7
!s~~ IMO fl~jnq)
6:804
;Accd=cikm&M~@m
(rw,$&2)
j i7.3olo4
:Totd
Inertiaflbjw).
I
j
30.952g4
.Taqmto-qerqfijdt~(~_-in) ~0,21-~
.—-..
.Ta~tocqx@@O syst~ (gzfln)
~
280.3891
.T@d tgcper-~red(a-in)