DESIGNANDCONSTRUCTIONOFTHESPIRITTPCBySuwatTangwancharoenADISSERTATIONSubmittedtoMichiganStateUniversityinpartialentoftherequirementsforthedegreeofPhysics-DoctorofPhilosophy2016ABSTRACTDESIGNANDCONSTRUCTIONOFTHESPIRITTPCBySuwatTangwancharoenThenuclearsymmetryenergy,thedensitydependenttermofthenuclearequationofstate(EOS),governsimportantpropertiesofneutronstarsanddensenuclearmatter.Atpresent,itislargelyunconstrainedinthesupra-saturationdensityregion.ThisdissertationconcernsthedesignandconstructionoftheSˇRITTimeProjectionChamber(SˇRITTPC)atMichiganStateUniversityaspartofaninternationalcollaborationstoconstrainthesymmetryenergyatsupra-saturationdensity.TheSˇRITTPChasbeenconstructedduringthedissertationandtransportedtoRadioactiveIsotopeBeamFactory(RIBF)atRIKEN,JapanwhereitwillbeusedinconjunctionwiththeSAMURAISpectrometer.Thedetectorwillmeasureyieldratiosforpionsandotherlightchargedparticlesproducedincentralcollisionsofneutron-richheavyionssuchas132Sn+124Sn.Thedissertationdescribesthedesignandsolutionstotheproblempresentedbythemeasurement.ThisalsocomparessomeoftheinitialfastmeasurementoftheTPCtocalculationoftheperformancecharacteristics.ACKNOWLEDGMENTSThedesignandconstructionoftheSAMURAIPion-ReconstructionandIonTrackerTimeProjectionChamber(SˇRITTPC)involvedaninternationalcollaborationtostudyandconstrainthesymmetryenergyterminthenuclearequationofstate(EOS)attwicesupra-saturationdensity.TheTPCdesignandconstructionaswellasmanyadditionalaspectsoftheprojectweresupportedbytheU.S.andU.S.partofthecollaborationwhoobtaineda$1.2MgrantfromDOEtoconstructtheTPCandtoshipittoRIKEN.TheU.S.grantsalsoprovidedpartialtravelsupportfortheU.S.participantstogotoRIKENtoinstalltheTPCaswellastodocommissioningandphysicsexperiments.TheSAMURAImagnet,thelasercalibrationsystem,thegas-handlingsystem,themountingstructure,thebeamtracking,theTPCelectronics,dataacquisition,ancillarytrig-gerdetectorsandotherequipmentrequiredforthesemeasurementsweretobeprovidedbyourJapanesecollaboratorsintheoriginalproposal.TheJapangroupobtainedagrantfromtheMinistryofEducation,Culture,Sport,ScienceandTechnology(MEXT)topurchasetheGETelectronics.ThefollowingsgivearoughdescriptionofthesharedtasksbetweentheUSandJapangroup.Packing/Dismantlingandshipping(MSU):Toarrangepacking,dismantlingandshippingoftheSˇRITequipmentto/fromRIKEN.Housing(RIKEN):ProvisionofRoomtoinstalltheequipment.PreparationoftheSˇRITDevice(MSUandRIKEN):InstallationofGETelec-tronicsandtests.InstallationoftheSˇRITDeviceintheexperimentalarea(MSUandRIKEN):InstallationofthefullSˇRITDevice,includingtheTimeProjectionChamberintoSAMU-iiiRAIandcouplingtobeamline.Beamtracking(RIKEN:SAMURAICollaboration):Resolution,timing,PPACorMWPC,Spectrometers(RIKEN:SAMURAICollaboration):Installation,tests,calibra-tionandmaintenanceofalldetectionsystemsoftheapparatus.DataAcquisitionatRIKEN(RIKEN):CouplingwithRIKENDAQTestsCalibra-tions.Withtime,otherinternationalgroupsjoinedtheproject.Fortheancillarytriggerdetec-tors:TheChinesegroupprovidedtheActivecollimatorvetocounter,thePolishcollaboratorsprovidedtheKatanaforwardvetoforwardwallwhilethesidemultiplicityarrayswerecon-structedbytheKyotogroup.OurPolishgroupalsobuiltatriggerboxfortheexperiment.TheR3BgroupfromGSIinGermanyprovidedtheNeuLANDneutrondetectorarrayaswellastwohighlyenriched112Snand124Sntargets(600mg/cm2).ThestudentsintheKoreangroupmadeverytcontributionstosoftwaredevelopmentbothforonlineandanalysis.Tofacilitatecommunicationsandtoavoidmisunderstanding,aweeklyteleconferencemeetingontheprojectwasconvenedbetweentheUSandJapangroup.OthergroupssuchastheKoreangroupandoccasionallythePolishgroupalsojoined.TheTPCprojectwasfundedinOctober1,2010.TheenclosureandtheversionofthetargetmechanismwasconstructedatTexasA&MwhilethemainpartofthedetectorwasdesignedandconstructedatMichiganStateUniversity.AfterconstructionwasinMay,2013andinitialtestingatMSU,thedevicewasshippedtoRIKENinFebruary,2014.TheGETelectronicswasfullyinstalledinAugust,2015.TheTPCunderwentitsbeamtestinOctober,2015.ItwaslatercommissionedinsidetheSAMURAImagnetinivApril,2016.TheSˇRITTPCwasthensuccessfullyusedintwoexperimentsinconjunctionwiththeSAMURAIdipolemagnetatRIKEN,JapaninMay2016.Mymajorresponsibilityfortheprojectistodesignandtotestthegatinggriddriverwhichisusedtocontroltheoperationofthegatinggrid.Afterthecommissionrunminorchangesweremadetothedrivercircuit.Thedriverwasthenusedsuccessfullyinthesecondcommissionrunsandtwoexperiments(atotaltwoweeksofbeamtime).Aftersometuning,thenewdriverwasabletocontroltheoperationofthegatinggridasexpected.ThedesignandconstructionoftheTPCwasmainlyledbyJonBarney,JustinEstee,RebeccaShane,AlanMcIntoshandBobOlsenunderthesupervisionofProfessorsWilliamLynchandBettyTsang.WhenIjoinedtheMSUSˇRITTPCgroupintheFallof2011,IbecameinvolvedwithallphasesoftheconstructionoftheTPCasdescribedinthisthesis.IhelpedwheneverIwasneeded.Itooktheleadresponsibilityindesigninggatinggriddriverandtheconstructionofthevoltagestepdown(VSD).TheVSDcanstepdownthehighvoltagefromthecathode(20000V)togroundwithoutsparking.ThishighvoltagewouldberequiredforHydrogenorHeliumbasedcountergases.Untilnow,theTPChasonlybeusedwithP10gaswhichrequiresavoltageof7000Vorless.ItestedandplannedtheprocedurestobuildthevoltagestepdownasdescribedinChapter2.5.Additionally,toevaluatetheperformanceoftheTPC,Mycolleagues(JustinEsteeandYaofengZhang)andIranaseriesofsimulationsusingGARFIELD,aprogramforsimulatingdriftchamberandfunctionofthegatinggrid.Theseresultshelpusunderstandhowtotunethedetector.MyworkwiththeTPCprojectalsoinvolvedsolvingproblemsencounteredinotherareas.Subsequently,Idevelopedexpertiseinallgluingprocedures,(includinggluingofthecircuitboardsforthegatinggrid,groundandanodewires)andleakcheckingtheIalsolearnedhowtorunSpiRITROOTtoobtaintheresultswhenhalfoftheGETelectronicsv(6000channels)wasinstalled.IdidthedataanalysisandworkedwithProf.Lynchtodemonstratethattheinitialdesignofthe"ZAP"adapterboardthatwasusedtoconnecttheGETelectronicstotheTPCwouldproduceunacceptablylargernoiseduetoitslargecapacitance.Thisanalysisledustoabandonthatdesigninvolvingcircuitboardsinfavorofthepresentdesign,whichdoesnotaddtlytotheintrinsicnoiseoftheGETelectronics.Myanalysiswasalsoinstrumentalinsuggestingthattherewasnon-uniformityinthegainsoftheoriginalGETelectronicsastheywereshippedtoJapan.MyresultsonselectedchannelswerebyGenieJhang,aKoreangraduatestudentwhowasresponsibleforthesoftware,buttheFrenchengineersdidnotbelievetheresultsbecausenosuchlinearitywasreportedbyanyothergroup.JonBarneythendemonstratedthatthiswasnotduetoadefectintheconnectionsbetweenGETelectronicsandtheTPC.Tounderstandbetterthisproblem,weexploredthenon-uniformityinthepositiveinputsignalmode(usedbySˇRIT)andthenegativeinputsignalmode(usedbytheATTPC)anddiscoveredthatthenon-uniformitywasnotpresentinthenegativeinputmode.WhenGenieandtheRIKENgroupshowedthatallchannelsoftheGETelectronicsexhibitedthesamenon-uniformbehavioronpositive,wecouldconvincetheFrenchengineerstoseriouslylookfortheproblemandtheyfoundawinthedesign,thatrequiredmodifyigthemodule.ThusalltheAGETandADC(ASAD)boardswerereturnedtoFranceforrepairandcausedadelayoftheprojectby6monthsandrequiredustoabandonthestbeamtimeslotthatwehadpushedtoobtain.Theprojectwouldnotbesuccessfulwithoutthetremendousoftheteam.IwouldliketothankallmycolleaguesforthesupportduringmystayatMSUworkingonmyPh.D.thesisandvaluableadvicesfrommyadvisor,BillLynchandfromBettyTsang.ThenamesviofthosewhohavemadecontributionstotheSˇRITprojectarelistedbelow.ThePostdocswhoweresupportedbytheDOEgrants,graduatestudentsandundergraduatestudentswhohavemademajorcontributionstotheconstructionprojectwhichisthemaintopicofthethesisarealsolistedwiththedatesassociatedwiththeproject.Theinstitutionsarelistedintheorderofcontributionstotheproject.MichiganStateUniversity,USAWilliamG.Lynch(Co-PIandSpokesperson),ManyeeBettyTsang(PIandSpokesper-son),RebeccaShane(postdoc,2011-2014),GiordanoCerizza(Postdoc,2015-),JonBarney(undergraduatestudent,2011-2013,Graduatestudent,2013-),JustinEstee(undergraduatestudent,2011-2014,Graduatestudent,2014-),SuwatTangwancharoen(Graduatestudent,2011-2016),ZbigniewChajecki(postdoc),FeiLu(postdoc),PierreMorfouace,PrabiPalni,ClementineSantamaria(Postdoc,2015-),JohnYurkon),CorinneAnderson(under-graduatestudent,2012-),KraigAndrews(undergraduatestudent),BenBrophy(undergrad-uatestudent),PeterChan(Summerundergraduatestudent),JimmyDunn(undergraduatestudent),EdErsoy(undergraduatestudent),JonGilbert(undergraduatestudent),HanSetiawan(undergraduatestudent,2013-),DavidWitalka(undergraduatestudent),RachelHodges(Graduatestudent),JuanManfredi(Graduatestudent),JackWinkelbauer(Gradu-atestudent),MikeYoung(Graduatestudent).KyotoUniversity,JapanTetsuyaMurakami(JapanesePIandSpokesperson),NoriNakatsuka(Graduatestudent),MasanoriKaneko(Graduatestudent2014-)RIKEN,JapanHidetadaBaba),TadaAkiIsobe(Spokesperson),MizukiKurata-Nishimura(Post-doc2013-),HiroyoshiSakurai),DaisukiSuzuki),AtsushiTaketani)viiTexasA&MUniversity,USASherryYennello(Co-PI),BobOlsen),AlanMcintosh(Postdoc2010-2013),MattChapman(undergraduatestudent)KoreaUniversity,SouthKoreaByungsikHong(Faculty),GenieJhang(Graduatestudent,2013-),JungWooLee(Grad-uatestudent,2014-)InstituteofNuclearPhysics,PolandJerzyLukasik),PiotrPawloski),KristofPelzar),PawelLasko(Grad-uateStudent)TsinghuaUniversity,ChinaRenshengWang(Graduatestudent),ZhangYan(Graduatestudent2015-),ZhigangXiao(Faculty)WesternMichiganStateUniversity,USAMohamedElHossieny(Graduatestudent),MichaelFamiano(CoPI),CharlieSnow(un-dergraduatestudent),SteveDye(undergraduatestudent),BeijingNormalUniversity,ChinaYaofengZhang(Lecturer)RareIsotopeScienceProject,SouthKoreaHyosangLee),YoungJinKim)UniversityofLiverpool,UKMarielleChartier(Faculty),WilliamPowell(Graduatestudent),JaimeNorman(under-graduatestudent)viiiTABLEOFCONTENTSLISTOFTABLES....................................xiLISTOFFIGURES...................................xiiChapter1ParticleObservation..........................11.1Abriefhistoryofvisualparticledetectors...................11.2Proportionalcounter...............................51.3TimeProjectionChamber............................141.4SˇRITTimeProjectionChamber........................161.5Choiceofgas...................................20Chapter2DescriptionsofSPiRITTPC.....................302.1ThedesignandconstructionofthecageoftheSˇRITTPC.......302.1.1Conductivepaintedcomponents.....................312.1.2Sidewalls.................................332.1.3Frontwall.................................362.1.4Exitwindow................................392.1.5Cathode..................................422.1.6Fieldcageassembly............................432.1.7Electrostaticsofcage........................522.2ThedesignandconstructionofthepadplaneoftheSˇRITTPC......592.2.1Padplaneassembly............................592.2.2Padresponsefunction..........................682.3ThedesignandconstructionofthewireplanesoftheSˇRITTPC......742.3.1Wirewindingandtensionmeasurement.................742.3.2Wireplanecircuitboard.........................762.3.3Wireplaneassembly...........................812.4ThedesignandconstructionoftheenclosureoftheSˇRITTPC.......882.5ThedesignandconstructionofthevoltagestepdownoftheSˇRITTPC..892.6ThedesignandconstructionofthetargetmechanismoftheSˇRITTPC..942.7OveralldesignassemblyoftheSˇRITTPC..................972.8Electronics.....................................1042.9Plannedtriggersystem..............................1082.10Gashandlingsystem...............................115Chapter3Gatinggrid................................1193.1Purposeofthegatinggrid............................1193.2Simulationofgatinggrid.............................1243.2.1Singlewireplaneandaconductingplane................1243.2.2Superpositionoftheelectric....................127ix3.2.3Monopolargatinggrid..........................1293.2.4Bipolargatinggrid............................1343.3Gatinggriddriver.................................1373.3.1Designcriteria...............................1383.3.2Conceptualdesign............................1383.3.3Gatinggriddriverprototype1......................1403.3.4Gatinggriddriverprototype2......................1463.3.5Gatinggriddriverprototype3......................1503.3.6Gatinggriddriverprototype4......................1523.3.6.1Basicdesign...........................1523.3.6.2Testsandoptimizationofversion4gatinggriddriver....1533.3.6.3MeasurementsandoptiminizationfotheInducedsignalsonthepadsusingtheGETTPCreadoutelectronics......1633.3.7Gatinggriddrivertest..........................169Chapter4Conclusions................................1744.1Conclusions....................................174BIBLIOGRAPHY....................................176xLISTOFTABLESTable1.1:Listofvariables..............................5Table1.2:DesignparametersoftheSˇRITTPC.................17Table2.1:PropertiesofwireplanesoftheSˇRITTPC..............74Table2.2:Radiiofroundcornersofthevotlagestepdown............93Table2.3:onoptionsforAGET[1]..................107Table3.1:Dissociationandionizationenergyofgases[2]............122Table3.2:TheoftuningthegatinggriddriverwithCnandCp.Rp=0.95Rn=1.05..........................163Table3.3:Negativepeakheightoftheinducedsignalfromthetransitionofthegatinggriddriver............................164xiLISTOFFIGURESFigure1.1:Apositrontrackwasphotographedfromacloudchamberunderthemagneticof15000gauss[3].....................2Figure1.2:Abubblechamberpicture.Thedarklinesaretinybubblesformedalongthepathsofchargedparticlesunderastrongmagnetic[4].4Figure1.3:Landaudistribution...........................7Figure1.4:Basicstructureofaproportionalchamber[5].............8Figure1.5:Rangeofoperationfordetector[5]..............8Figure1.6:Simplestructureofmultiwireproportionalchambers[5].......10Figure1.7:CrossectiondiagramofMWPC.....................11Figure1.8:Electricldofamultiwireproportionalchamber.Allwireshaspotentialof1Vandtheouterelectrodeshasthepotentialof0V.Thecalculationshowstheresultford=40m,s=1mm,L=8mm[6]..................................12Figure1.9:Thetwocathodeplanesaredividedintostripsparallelandorthogonaltothewires;coordinatexparalleltothesensewiresandcoordinateyorthogonaltothesensewires.[7]...................13Figure1.10:AschematicshowstheoperationofPEP-4TPC[8].........14Figure1.11:Thereconstructedtracksfroma200GeVpernucleonAu-AucollisionusingtheSTARTPC[8]........................16Figure1.12:AnexplodedviewofSˇRITTPC.MoreinformationisavailableinSectionDesignandconstructionoftheSˇRITTPC.........17Figure1.13:OperationoftheTPC[9].......................18Figure1.14:TherstTownsendcotasafunctionofelectriceldstrength[10]....................................21Figure1.15:TheTownsendfornoblegases[11]................22xiiFigure1.16:Gas(gain)fortmixturesasafunctionofanodevoltage...................................24Figure1.17:DriftvelocityofanelectroninP-10gasasafunctionoftheratioofelectric[12]..............................25Figure1.18:DriftvelocityofanelectroninP-10gasasafunctionofthefractionofMethane[13]..............................26Figure1.19:LongitudinalforP-10[13]...................27Figure1.20:TransverseforP-10[13]....................27Figure1.21:(a)Electrondriftlinesintheplaneperpendiculartotheanodewire.(b)ASpreadingoutoftheelectronsalongtheanodewireduetotheExB...............................28Figure1.22:LorentzangleforP-10withmagneticof0.5T.Theanglesbe-tweentheelectricandmagneticare0,15,30,45,60,75and90degrees[13]............................29Figure2.1:3dimensionaldesignofthecageoftheSˇRITTPCwithouttheexitwindowpolyamidesheet.......................31Figure2.2:Preparethesidesupportforbeingpaintedwiththecopper-silverepoxypaint.Kaptontapeisusedtomasktheareasthatdowantthepainttogoover............................32Figure2.3:Thefrontwindowframewaspaintedwiththeconductingpaintwasanepoxywithsilvercoated.......................32Figure2.4:DrawingofasidewallPCBofthecage.ThereareholesfortheLASERcalibrationcomponents.Theuresforthelaserportsarenotshowninthis........................33Figure2.5:Afterthewallswereassembled,theinteriorhas50stripsand49stripsfortheexterior.Thestriponthetopsinksintothetopperimeterandtheoneatthebottomsinksintothecathode...........34Figure2.6:Sidewallofthecagewaslaidonthesurfacefor10hourstocure....................................35Figure2.7:SolderthincopperpiecesonthesidewallstoconnectthestripsonthePCBs.................................35xiiiFigure2.8:ThefrontsideoftheeldcageconsistsoftwoPCBsandthewindowframewhichisgluedtothetwowalls.Green:thewindowframegastwogaschannelswhichareusedtothegasintotheTPCfromthebottom.Purple:Theremovableinsertedwindowisscrewedonthewindowframefromtheinsideofthecage............36Figure2.9:Explodedviewofthefrontsideofthecage.ShowninthedrawingaretwogaschannelsforthegasintotheTPC..........37Figure2.10:Afterpositionthewindowframeandboards,theboardswereat-tachedtothewindowframewiththebrassscrews.......37Figure2.11:Insertedremovablewindowofthecage..............38Figure2.12:Theinsertedwindowisputonthefrontwallofthecage.Thecopperprovideelectricalconnectionstothestripsonthewalls.38Figure2.13:ThestripsonathinPPTAareconnectedtothestripsonthewindowframebycopper.........................40Figure2.14:ExitwindowoftheSˇRITTPCconsistsofthreepolycarbonateframes.41Figure2.15:Exitwindow...............................41Figure2.16:Roundaluminumedgewasattachedtotheedgeofthecathodetoreduceasharpangle...........................42Figure2.17:Circuitdiagramfortheelectricalconnectionsofthecathode....43Figure2.18:Assemblingthefrontandsidewallsofthecage..........44Figure2.19:Assemeblebothsidewallsofthecage...............45Figure2.20:ApplythebeadofAraldite2013epoxytothecornerpiecesofthebackwindowframe............................45Figure2.21:Backwindowframewasattachedtothewallofthedcage.....46Figure2.22:GlueexcessareaonthetopperimeterhasbeenmaskedwithplasticandaKaptontape............................47Figure2.23:PlacetothewallofthecageslowlyonthegroovewithbeadsofAraldite2013..........................47Figure2.24:Wallofthecagewasscrewedtothetopperimeter........48xivFigure2.25:Fieldcagewasplacedonthesupportforcheckingandcleaningtheepoxy...................................48Figure2.26:Movethecagetothesurfacetocurefor10hours......49Figure2.27:Preparethecathodeforgluing......................49Figure2.28:Tothepotentialoneachstrip,achainofresistorwassolderedbetweeninthestrips.....................50Figure2.29:Wiresweresolderedalongthestripatthecornerforbetterelectricalconnectionsbetweenthestripsoncornerpiecesandthewall.Silverepoxywasappliedonthewireatthemiddleofthestriponthecornerpiecestoprovideabetterconnection..................50Figure2.30:Insert1resistorsalongthestrips..................51Figure2.31:Coppererswereusedtoconnecttheconductivestripsontheexitwindowtoonesonthewall.......................52Figure2.32:DiagramshowstheresistornetworkoftheSˇRITTPC........53Figure2.33:ElectricinthedriftvolumeoftheTPCandRpasafunctionofVggVcath...................................56Figure2.34:PotentialcalculationofthecageisperformedbyANSYS.Rpis7fortheVcathandVggof-6kVand-110V,respectively.....57Figure2.35:Thestripsonthefrontwindowframeareslightlytfromtherestofthecade...........................58Figure2.36:ElectrondriftlinesnearthefrontwindowoftheSˇRITTPC.....58Figure2.37:Cross-sectionalviewofthepadplanecircuitboard[9].........60Figure2.38:Unitcellonthebeamrightsideisa180degreerotationofthatofthebeamleftside.NumbersindicatethechannelwhichthesignalfromthepadisregisteredtotheAGET...................61Figure2.39:Padplane................................62Figure2.40:Topplatewithinnersurfaceupwards.Thereare384rectangularholesforSamtecconnectorfeedthroughs................63xvFigure2.41:Ribsofthetopplate..........................63Figure2.42:Gasgetgluing..............................64Figure2.43:Stampingmetalpieceforgluinggasket................65Figure2.44:Completegasketgluingprocess.....................66Figure2.45:Gluingpatternforthepadplane....................66Figure2.46:Pad-planePCBisputtothepositionbythevacuumtable.....67Figure2.47:TheofthetopplateusingFAROlaserpositionscanner...68Figure2.48:ValuesofparameterK3asafunctionofananodewirepitchs,ananode-cathodeseparationhandtheradiusofananodewirera[14].71Figure2.49:ValueofK3forhs1:0[14]......................72Figure2.50:PadresponsefunctionforSˇRITTPCwithw=8mm(red)andw=12mm(blue).............................73Figure2.51:Wire-windingmachine..........................75Figure2.52:Theschematicofthesetupformeasuringawiretension.......77Figure2.53:measurementofthetensionofanodewires:(a)theanodewiresdonotvibrationwhenthefrequencydoesnotmatchthefundamentalfrequency.(b)Whentheapplyingfrequencymatchesthefundamen-talfrequencyoftheanodewire,thewirevibrateswithamaximumamplitude.................................77Figure2.54:Theprintedcircuitboard(PCB)foranodewires:26padsarecon-nectedtothehighvoltageby26of10resistors(blue).Eachpadisconnectedtoanexternalgroundby1nFcapacitor(brown)....78Figure2.55:Printedcircuitboardforgroundplane.................78Figure2.56:Printedcircuitboardforgatinggridplane..............79Figure2.57:Gluingthegroundboard........................79Figure2.58:Gluingtheanodeboardtoaacrylicspacer..............80xviFigure2.59:Fillthegapontheanodeboard....................80Figure2.60:AnodeboardsweremountedontheTPC...............81Figure2.61:Wirecombassembly...........................82Figure2.62:UsingEZpoxy83toholdthewiresinplaceandmaintainthetension83Figure2.63:Wireframewithlevelingblocks....................83Figure2.64:Solderwiresontotheconductivepadontheanodeboard......84Figure2.65:ConnecttheanodeboardtotheMHVconnector...........85Figure2.66:Insulatetheanodeplaneconnections.................86Figure2.67:Solderadjacentgroundboards.....................87Figure2.68:Layersofatransmissionline......................87Figure2.69:Transmissionlineisconnectedtothegatinggridboard.......88Figure2.70:EnclosureoftheSˇRITTPCwiththemotionchassis........89Figure2.71:DrawingofthebottomplateoftheSˇRITTPC...........90Figure2.72:PourEzpoxy84atthecenterofthebottomplate..........90Figure2.73:Polycarbonatesheetwasgluedtothebottomplatebyvacuuminggluingtechnique..............................91Figure2.74:After24hourscuringtime,thecovershavebeentakenandthebottomplatewascleaned........................92Figure2.75:Conductivepaintwasappliedatthecenterofthepolycarbonatesheet.92Figure2.76:Copperrodsandwereattachedtothepolycarbonatesheetby0-80screws...............................93Figure2.77:Aquarterroundcornerwasproducedbywelding...........94Figure2.78:DesignmodelofthetargetmechanismfortheSˇRITTPC......95Figure2.79:Atargetcanbeadjustedinydirectionbypositioningtheonthetargetframe..............................95xviiFigure2.80:AtargetframecanbemovedinzdirectionviatheZ-motioncontrol96Figure2.81:ThexpositionofthetargetcanbeadjustedfromtheX-motioncontrol.96Figure2.82:Actualtargetmechanism........................97Figure2.83:Topplatewiththecoverplatewasrotatedby90degrees.......98Figure2.84:Insertwindowandwereputonthewindowframeofthecage....................................99Figure2.85:Fourpeopleliftthecageandsetitonthesetscrewsonthetopplate....................................100Figure2.86:Springswereattachedtothebottomofthecagetoprovideanelectricalconnectionbetweenthecathodeandconductivesurfaceonthevoltagestepdown..........................101Figure2.87:Topplatewiththecageattachedhasbeenrotatedtothepositionforassemblingtotheenclosure.....................101Figure2.88:Topplatewasloweringwhilepeopleateachcorneralignthetopplateassemblywiththeenclosure.......................102Figure2.89:TheSˇRITTPCwithoutatargetmechanism............103Figure2.90:ThetargetmechanismisinstalledontotheSˇRITTPC.......104Figure2.91:Thelaseralignmenttechniqueswasusedtoaligntothecenterofatargettothecenteroftheentrancewindow..............104Figure2.92:GETconceptualdesign[1].......................105Figure2.93:GETsystemismountedontheSˇRITTPC.............106Figure2.94:AsAdmotherboard...........................107Figure2.95:PlannedcoolingsystemfortheSˇRITTPC..............108Figure2.96:Triggertimingstructureforthegatinggriddriver.Thediagramisnotinthecorrectscale..........................109Figure2.97:Plannedtriggertimingdiagram.ThemoduleslabeledFinthedia-gramaretheFIFO............................110xviiiFigure2.98:Connectionbetweenthescintillatorsandelectronicmodules.....111Figure2.99:Triggertiming..............................112Figure2.100:Avetosignal(blue)occurmorethan4sbeforetheT-StartScint.ThetriggersystemgeneratethenormalTTL-Open(yellow),TTL-Close(green)...............................113Figure2.101:ThereisnogenerationofTTL-Open(yellow)andTTL-Close(green)whentheT-StartScint(purple)andT-Veto(blue)aresimultaneouspresent...................................114Figure2.102:Whenavetosignalpresentsaftertheeventtrigger,TTL-Open(yel-low)closesandtheTTL-Closestarts..................114Figure2.103:ThedelayoutputfromG&D2passestothestartofG&D4toassurethattheTTL-Open(yellow)andTTL-Close(green)arenotsimulta-neouspresent...............................115Figure2.104:PositionofgasinputandoutputontheSˇRITTPC.........116Figure2.105:PlannedgashandlingsystemfortheSˇRITTPC..........117Figure2.106:PlannedgashandlingsystemfortheSˇRITTPC..........118Figure3.1:Ionizedelectronsdriftintothemultiplicationregionandmultiplyattheanodewires..............................121Figure3.2:Formationofpolymersfromfreeradicalmolecules[2]........122Figure3.3:Accumulationofnegativelychargedpolymersonthesurfaceofananodewire[15]..............................123Figure3.4:PolymersdepositonthesurfaceofananodewireinCO2/Isobuthane[15]....................................123Figure3.5:Gridofwiresparalleltoaconductingplane[11]...........125Figure3.6:WireplanetionoftheSˇRITTPC.Thepitchofanodeplane(s1)is4mm.Thepitchoftheground(s2)andgatinggrid(s3)are1mm.Thedistancefromtheconductingplane(padplane)totheanode(y1),ground(y2)andgatinggrid(y3)is4,8and14mm,respectively.Thedistancefromthecathodetothepadplaneyp=509.5mm.................................127xixFigure3.7:Awireispolarizedbyanexternalelectricld.Electricarepointingupward[11]...........................131Figure3.8:Driftlinesofionizedelectrons(y1=4mm,y2=8mm,y3=14mm,s1=4mm,s2=s3=1mm).....................132Figure3.9:Transparency,T,ofthegatinggridwithVcathode=6kV.Blacktriangle:simulation;Redline:AnalyticalsolutionusingEquation3.11...............................133Figure3.10:Operationofbipolargatinggrid(z1=4mm,z2=8mm,z3=14mm,s1=4mm,s2=s3=1mm).(b)Electronsterminateatthepositivewiresofthegatinggrid.....................135Figure3.11:(a)Transparencyofbipolargatingwiththepresenceofmagnetic(b)Theclosingvoltageofagatinggridincreaseslinearlywiththemagneticeld.............................137Figure3.12:conceptualdesignofthegatinggriddriverforSˇRITTPC.....140Figure3.13:Circuitdiagramofthegatinggriddriverprototype1.........141Figure3.14:Dischargingsignalofthegatinggriddriverprototype1onthebreadboard.Therearesomeslowoscillationonthesignal..........141Figure3.15:Testingtheprototype1PCBwithastandardcapacitorof11.6nFandanoperatingvoltageof30V.Bluesignalisadischargingsignalfromthepositivesideofthecapacitor.Thereisanegativelobeafterthedischargewhichindicatesthatthecircuitisunderdamped....143Figure3.16:MatchingthedatatothesimulationfromPSPICE..........145Figure3.17:Dischargingsignalsofthegatinggridusingthegatinggriddriverprototype1................................145Figure3.18:Dischargingsignalfromtheoscilloscope(operatingvoltageof75V,resistanceofthesystemof4.8...................146Figure3.19:ReadoutsignalfromtheprototypeofGETElectronics........146Figure3.20:Circuitdiagramofthegatinggriddriverprototype2.........147xxFigure3.21:Signalsfromtheoscilloscopeshowsthedischargingfromthepositiveside(blue)andnegativeside(purple)ofthestandardcapacitorof22nF.Thetwosignalsdischargeatatrateandhaveaatthebeginningasindicatedinredcircle...............148Figure3.22:Adjustthedischargingrateonthenegativesidetomatchthesignalonthepositiveside............................149Figure3.23:Green:theTTLtriggersignal;Purple:GatesignalfordrivingN-mosfetswitch;Blue:GatesignalfordrivingP-mosfetswitch....149Figure3.24:Circuitdiagramofthegatinggriddriverprototype3.........150Figure3.25:Thesignalfromtheoscilloscopeshowsthetimeofcharginga22nFcapacitortotheoriginalvoltageof12V.Blue:signalfromthepositivesideofthecapacitor;Purple:signalfromthenegativesideofthecapacitor..............................151Figure3.26:Thegatinggriddriverprototype3hasbeentestedwithastandardcapacitorof16.5nF.Theoperatingvoltagesare-40and-180V.Ithasaturn-ondelaytimeof100ns.Whenthegatinggridisopen,thepositive(yellow)andnegative(blue)sidesofthecapacitorareshortedthroughthemosfetswitchesandprovidetothecommonvoltageof-110V...................................152Figure3.27:Circuitdiagramofthegatinggriddriverprototype4.........154Figure3.28:Thecircuitboardistestedwithoutconnectingtoastandardcapac-itor.Thegatinggriddrivershortstwopowersuppliestogetherat45nsafteritistriggered.Blue:dischargingsignalfromthepositiveside.Green:dischargingsignalfromthenegativeside.........155Figure3.29:Printedcircuitboardofthegatinggriddriverprototype4.Thecol-oredcirclesindicatetheconductivepadswhichcanputtheadjustingcomponents,Rp,Rn,CpandCnon.Rnismountedontheothersideoftheboardsoitisnotvisibleinthispicture...........155Figure3.30:Testthegatinggriddriverprototype4withoperationvoltagesof-40and-180V.thevaluesofCpandCnare100pF.(a)Whenthegatinggridisclosed,alternatingwireshavethevoltagesof-40and-180V.Onceitopens,thetwopowersuppliesareshortedtogetherandgivethecommonvoltageof-110V.(b)theopeningtimeofthegatinggridis250ns...............................158Figure3.31:Thereisainducedsignalonthepadwhenthegatinggridopens...159xxiFigure3.32:TransitionofthegatinggridfromclosedtoopenandbacktoclosedstateinSPICEsimulation.Cp=Cn=100pFandRp=0.95Rn=1.05.................................160Figure3.33:(a)TransitionofthegatinggridfromclosedtoopenstateinSPICEsimulation,Cp=600nFandCn=100pFandRp=0.95Rn=1.05(b)Thesameas(a)withentcapacitorvaluesCp=100nFandCn=100pFandRp=0.95Rn=1.05........161Figure3.34:ThedependenceofXandBasafunctionofCn-Cp.........162Figure3.35:Testthegatinggriddriverwithastandardcapacitorof26nFandvaryCpandCntoseetheefromtheSPICEsimulation.Thegreenlineindicatesthecommonvoltagelevel.............163Figure3.36:TestthegatinggriddriverwiththeSˇRITTPCbyvaryingCpandCn.ThesizeofthenegativepeakisincreasingwithCn.......164Figure3.37:UsingCp=1000pFandCn=440pF,thenegativepeakisreducedfrom800to105ADCchannels.....................165Figure3.38:InducedsignalsarereadoutbyAGETElectronics.Inthistion,CpandCnare1000and490pF,respectively...........166Figure3.39:Toppanel:theinducedsignalcomesfromthetransitionofthegatinggrid.Bottompanel:theexternalsignalwhichisusedtocancelthenoiseissentthruthegroundplane...................167Figure3.40:Useanexternalsignal(red)whichhastheoppositepolarityifthenoise(black)tocancelit.Thebluelineshowsthesuperpositionofthosetwosignal.Ithasasmallnegativepeakof50ADCchannels.168Figure3.41:AschematicofthecommissioningexperimentsetupoftheSˇRITTPC[74].................................170Figure3.42:TransparencyofthegatinggridoftheSˇRITTPCasafunctionofthecommonvoltageofthegatinggridforVcathode=-6632V....171Figure3.43:TransparencyofthegatinggridoftheSˇRITTPCasafunctionofthevoltageVg)ofthegatinggridforVcathode=-6632V..173xxiiChapter1ParticleObservation1.1AbriefhistoryofvisualparticledetectorsIntheworldofsub-atomicparticles,physicistsputtremendoustimeandtostudypropertiesofparticlesandtheinteractionsbetweenthem.Manytheorieshasbeenproposedtoexplainthephysicsofparticlesandtheirinteraction.Individualsub-atomicparticlesaretoosmalltodirectlyobservedbyussophysicistsdetectthemwithvarioustypesofdetectorssuchasscintillators,semiconductordetectorsandgasdetectors.Intheirsimplestform,noneofthosetechniquesenablesscientiststodirectlyseethepathsofallparticlescomingoutfromasourceorareaction.Ifthenumbersofparticlesarelargeorthereactioniscomplex,themissinginformationistooimportanttoallowscientiststomakesolidconclusionsaboutthephysicsthatisbeingstudied.In1912,acloudchamberwasinventedbyCharlesThomsonReesWilson[16].Thedeviceisasealedchambercontainingavaporofwateroralcohol.Theideaistoallowwateroralcoholvaportoreachasaturatedpointinthecontainerandthenlowerthepressure.Thiswillproduceasupersaturatedcondition.Ifachargedparticletransversesthechamber,thevaporwillcondense,followinginteractionsofenergeticparticleswiththecloudchambergas,intodropletsalongthetrackofachargedparticle.Theparticletrackcanbephotographed.Thisisabreakthroughofscieninstrumentbecausethepathofsubatomicparticlesfromareactioncanbeobservedforthetime.Inthepresenceofamagnetic,positively1Figure1.1:Apositrontrackwasphotographedfromacloudchamberunderthemagneticof15000gauss[3].andnegativelychargedparticlescanbedistinguishedastheycurveinoppositedirections.In1932.CarlAndersonfoundaparticlethathasapositivechargebuthasthemassofafreeelectronbyusingacloudchamberunderthemagneticof15000gauss,leadingtothediscoveryofapositron[3].InFigure1.1,thepositrontrackinthemiddlewasphotographedfromacloudchamber.Thistypeofchamberalsoallowsphysiciststostudyelectromagneticshowersandnuclearreactions.Theperformanceofthecloudchamberhastwomainlimitations.Fewpathsofparticleswereobservedduetothelowdensityofgasandthedetectionrateofparticleswastoolowtouseinconjunctionwiththenewacceleratorsconstructedin1950s.In1952,DonaldGlasertriedtoimprovetheperformanceofthecloudchamber[17,18].Glaser'sideaistoreplacethegaswithliquid.ThechamberwaswithliquidslightlybelowitsboilingpointatacertainpressureunderaconstantmagneticThen,thepressurewasreducedbelowthevaporpressureoftheliquid.Ifchargedparticlestransversesthechamber,theyionizeatomsandcausetheliquidtoboilalongtheirpaths.Therefore,onecanseethe2bubblesalongtheparticles'strailsasdemonstratedinFigure1.2.Inventionofthecloudandbubblechambersplayedanimportantroleinhighenergyphysics.Manyparticleshasbeendiscoveredfrombubblechamberpicturessuchasdiscoveryof(1964)[19]anddiscoveryof"charmed"quark(1974)[20,21].However,theperformancewaslimitedtoonlyvisibletrailsofparticles.Somerareprocessesthatoccurwithlowprobabilitiesmaybemissedduetothelongtimeframeofdetectionwhichlimitsthetotalnumberofeventsthatcanbeobserved.Inthelate1960s,GeorgeCharpakdevelopedmultiwirechambers[22,23](seeSection1.2)whichallowsphysiciststopreciselydetecthighenergyparticlereactionsatamuchhigherrateandalsomeasuretheenergyofaparticleatthesametime.Withthehelpofreadoutelectronics,physicistsareabletorecordhundredsofeventswithinasecond.Typically,multiwirechambersconsistsofseveralwireplaneswithtorientationswhichareusedtodeterminethetrailsofparticleswithhighprecision.Inmultiwirechamber,Eachwireworksasadetector.Whenreadoutwithhighspeedelectronicsandanalyzedwithhighspeedcomputers,thedatahandlingcapacitytremendouslyincreased.In1974,DavidNygrenintroducedanewgasdetectorcalledaTimeProjectionChamber(TPC)[24]whichthemeasurementofthousandsofsub-atomicparticlesfromasingleeventwhilemeasuringtheirpropertieswithhighaccuracy.Itemploysmanyofthetechniquespioneeredwiththemultiwirechamberandalsohasanelectronicreadoutsystem.TheTPCisexplicitlydesignedtoreconstruct3-dimensionaltracksofparticles.ThedetailsabouttheTPCwillbediscussedinSection1.3.3Figure1.2:Abubblechamberpicture.Thedarklinesaretinybubblesformedalongthepathsofchargedparticlesunderastrongmagnetic[4].41.2ProportionalcounterTounderstandthepropertiesofanygasdetector,oneneedstoconsiderhowafastchargedparticleinteractswiththegas.ThemostimportantinteractionsaretheCoulombinteractionsbetweenthechargeofthefastparticleandthechargeoftheelectronsboundtotheatomsinthegas.Theseinteractionstransferenergytotheelectronsandknocksthemoftheiratoms.TheCoulombforcebetweentheincidentchargedparticleandtheatomicelectronsdecreasesinverselyasthesquareoftheimpactparameter.IntegratingovertheimpactparameterleadstotheBethe-Blochequation,dEdx=Kz2ZA1212ln2mec222TmaxI22()2:(1.1)Table1.1:Listofvariables.SymbolMIncidentparticlemassEIncidentparticleenergyTKineticenergymeElectronmassreClassicalelectronradiusNAAvogadro'snumberzeChargeofincidentparticleZAtomicnumberofmediumAAtomicmassofmediumK4ˇNAr2emec2IMeanexcitationenergyDensitycorrection1=p12ThelistofvariablesisshowninTable1.1.isthespeedoftheprojectileintheunitofthespeedoflightc.HereTmaxisthemaximumkineticenergywhichcanbepassedona5freeelectroninasinglecollision.TmaxisgivenbyTmax=2mec2221+2me=M+(me=M)2:(1.2)TheBethe-Blochequationcorrelatesthemeanenergylossperunitlengthor"stoppingpower"oftheparticleinthedetectorgastoitschargeanditsvelocity[25].TheoftheenergylossbyionizationofachargedparticleswastheoreticallydescribedbyLandauin1944[26].Thedistributionofenergylossinthinmediaisgivenbyf()=1p2ˇexp12(+e):(1.3)isthereducedenergyvariablewrittenby=EEmp˘;where˘=KZAˆ2X:(1.4)Eistheactualenergyloss.Empisthemostprobableenergyloss.˘istheaverageenergyloss.Xisthethicknessofthemedia.Figure1.3showsthecharacteristicshapeofLandaudistribution.Thelongtailatverylargeenergylosscorrespondstoaneventwhichoneormoreenergeticelectrons,usuallycalleddeltaelectrons,havebeenproduced[27].Combiningthiswithmeasurementsofotherquantities,suchastheradiusofcurvatureoftheparticleinamagneticortheparticlestotalenergycanallowonetodeterminewhattypeofparticleitis,suchasapion,orthenucleusoflightnuclearisotope.Thus,akeyquantityonewantstomeasureistheenergylossinthedetectorgas.Particleswithenergieslessthanthatforminimumionizationcanbedistinguishedbytheirenergylossingas.Thisenergylossisdepositedintothegas,wheresomeofitgoestounbindingtheelectronfromthe6Figure1.3:Landaudistributiongas(ionizationpotential)andtherestgoestothekineticenergyoftheelectrons.Thesefastprimaryelectronsloseenergyinthegas,followingtheBethe-Blochequation,andeventuallymostoftheenergylossoftheincidentparticleisconvertedtototalionization.AttypicalvalueoftheenergyperionizedelectroninP-10gasisabout26eV[28,29].ForatotalenergylossofE,thenumberofsecondaryelectronsisE26eV.Onecanthereforedeterminetheenergylossbycountingthesesecondaryelectrons.Thebasicideaofamultiwireproportionalchamberwasbasedonproportionalcounters,whichhasbeendevelopedaround1940s.Figure1.4showsabasicstructureofasinglewireproportionalcounter.Ithasacylindricalgeometry.Avoltageisappliedbetweenthemorenegativecathodetubeandthemorepositiveanodewirelocatedatthecenterofthetubetocreateahighelectricwheregasoccurs[5].ThemodesofoperationoftheproportionalcounterisshowninFigure1.5.Atlowvaluesofthevoltage,thestrengthisnotttopreventtherecombinationoftheoriginal7Figure1.4:Basicstructureofaproportionalchamber[5]Figure1.5:Rangeofoperationfordetector[5].pairs.Therefore,thecollectedchargeislessthanitshouldbe.Asthevoltageincreases,therecombinationprocessissuppressedandthesystemisintheionsaturationstatewheretheconstantisachieved.Inthesaturationrange,thecollectedchargesisnearly8equaltothechargeoftheoriginalionpairs.Ifthevoltageisfurtherincreased,thecollectedchargebegintomultiply.Atacertainrangeofvoltages,thegasmultiplicationislinear.Thisregioniscalledproportionalregion.Here,thecollectedchargeswillbeproportionaltotheoriginalionpairs.Ifwekeepincreasingthevoltage,thennon-linearityofgasmultiplicationwilloccur.Thisnon-linearisrelatedtopositivechargesproducedinthesecondaryionizationprocess.Thecloudofthesepositivechargesmoveveryslowlycomparingtothespeedofelectronstowardstothecathode.Ifthedensityofthepositivechargesistlyhigh,theyformaspacechargethatcantlydistorttheelectricinthevolume.Sincethegasmultiplicationdependsonthemagnitudeoftheelectricthenon-linearstartstobeobserved.Fromthispoint,thecollectedchargesincreasenon-linearlywithincreasingnumberoftheoriginalionpairs.Thisregioniscalledlimitedproportionality.Forfurtherincreasingappliedvoltage,theGeiger-Muellerregionisreached.IntheGeiger-Muellermode,thespacechargecreatedbythepositiveionsbecomesdominant.Avalancheswillproceedalongthewireuntiltnumberofpositivechargeshasbeenreachedtoreducetheelectricbelowthepointthatthegasmultiplicationcanhold.Therewillbenoinformationthenumberoftheoriginalionpairs[10].Inthiscase,themodeofoperationthatisusefultodeterminetheenergylossaretheionsaturationandproportionalmodes.Botharewidelyusedinnuclearandparticlephysics.Ifonewantstotrackparticlesthroughthegas,however,oneneedstosampletheionizationinsmallvolumesofthegas,wherefewelectronsareproduced;thusthesignalsinionchambermodeprovetobetoosmalltomeasureaccurately.Forthisreason,theproportionalmodeispreferredfortracking.Toachieveproportionalmode,oneneedstohavethelargeelectricpossiblenearasmallradiuswireatanelevatedelectrostaticpotentialasinaproportionalchamber.Forsmallwiresoforderof20umdiameter,onecanmultiplyeachoftheelectronsproducedin9theincidentionizationbyafactorof1000-10000[30],increasingtheratioofthesignalfromtheionizationrelativetotheelectronicnoiseintheelectronicssystemthatsamplesthisionization.Touseaproportionalcountertotrackatrajectoryofaparticle,itispossibletostackproportionalchambersbut,itwouldbeachallengemechanically.Also,therewasabeliefthathavingmultiwireinthesamegaschambermaynotworkproperlyduetoalargecapacitanceinthestructure.Moreover,non-screenedwiresmaycausethesignaltospread[27].In1968,Charpakandcollaboratorsintroducedamultiwireproportionalchamber(MWPC)whichconsistsofthinequallyspacedwiressandwichedbetweentwocathodeplanes.Figure1.6demonstratesaschematiccross-sectionofsuchastructure.Typically,thedistancebe-tweenaplaneofwirestothecathodeisaboutthreeorfourtimeslargerthanthewirespacingforproperoperation[27].IntheMWPC,eachwireworksasanindividualdetector.Thedetectedsignalswerereadoutbyanelectronicsystem.TheMWPCisthevisualparticledetectorthathaselectronicreadout.Thisallowedthedatatakingcapacityofexperimentstogreatlyincrease.Figure1.6:Simplestructureofmultiwireproportionalchambers[5]Figure1.7showsanplaneofwiresofradiusa,spacingsandthedistancefromthewireplanetocathodeL.Thecoordinatesystemhasbeencenterononewire.Weassume10thatallwireshavethesamechargeqperunitlength.Thepotentialforathinwireswithequalspacing[27]canbeexpressedbyV(x;y)=CV04ˇ0ˆ2ˇLsln4sin2(ˇxs)+sinh2(ˇys)(1.5)E(x;y)=CV0201+tan2(ˇxs)tanh2(ˇys)1=2tan2(ˇxs)+tanh2(ˇys)1=2:(1.6)V0isthepotentialonthewireandcapacitanceperunitlengthCisgivenbyC=2ˇ0ˇLsln(2ˇas):(1.7)Figure1.7:CrossectiondiagramofMWPCEarlymultiwireproportionalcounterswereconstructedwiththincathodewindowsaboveandbelowtheanodewireplane.Theywereusedtodetectchargedparticlesthatpassedthroughonecathode,throughtheanodewireplaneandouttheothercathode.Whena11Figure1.8:Electricldofamultiwireproportionalchamber.Allwireshaspotentialof1Vandtheouterelectrodeshasthepotentialof0V.Thecalculationshowstheresultford=40m,s=1mm,L=8mm[6].chargedparticletransversesthechamber,itionizesthegasatomormoleculeandaprimaryionpairisproduced.Theelectronsdriftalongtheelectriclinesintheoppositedirectionofthetowardsthenearestwirewherethemultiplicationoccurs.ThevoltagethatappliestothewireiscrucialasdiscussedinSection1.2.Byputtingaoneachwire,onecandeterminewhichwireisclosesttotheionizationallowingthepositionoftheionizationtobemeasuredwithanaccuracycomparabletothewirespacing.However,athighanodevoltages,sparksonthewireoftendamagedthepreamps.ToimprovethespatialresolutionoftheMWPC,Breskinet.al.introducedthecathodeplanethatisequippedwithstripsparallelandorthogonaltothesensewiresinsteadofasinglecathodeplane[7].Withthisdetector,eachcathodestripisconnectedtotheinputofa,whichholdsthecathodeatavirtualgroundwhilemeasuringtheimage12Figure1.9:Thetwocathodeplanesaredividedintostripsparallelandorthogonaltothewires;coordinatexparalleltothesensewiresandcoordinateyorthogonaltothesensewires.[7].chargesinducedonitbythemotionsoftheelectronsandionsneartheanodewires.Sparkingtypicallydidnotdamageattachedtothecathode.Also,theionizationistypicallyononeortwoofthewires.Withthisdesign,onecanapplythecentralofgravitytechniquetoobtainthelocationofinteractionintwodimensionswithanaccuracybetterthanthespacingoftheanodewires.InFigure1.9,thesignalisthelargestonthecathodestripsnearesttowheretheavalancheoccursanddecreaseswithdistancefromtheavalanche.Thepointofinteractioncanbeobtainedbycalculatingthecenterofgravityofthesignalwhichisexpressedbyx=Qib)xiQib):(1.8)Qiisthechargecollectedontheithstrip.xiisthecoordinateoftheithstrip.bisa13smallcorrectiontothenoise[5].TherearealsofurtherdevelopmentsonacathodedesigntoimprovespatialresolutionoftheMWPC[31]anddataacquisitiontechniquestoachievehighqualityresults[32,33].1.3TimeProjectionChamberATimeProjectionChamber(TPC)isagas-detectorthatprovidesa3-dimensionaltrackofparticlemovingthroughthevolume.TheTPCplaysacrucialroleinthestudiesofhightrackdensityenvironmentandparticleidenbyenergyloss[34].TheTPCwasinventedbyDavidNygrenattheLawrenceBerkeleyLaboratory(LBL)inthelate1970s.Theapplicationwasthestudyof29GeVpositron-electroncollisioninPEP-4detectorasseeninFigure1.10[8].Figure1.10:AschematicshowstheoperationofPEP-4TPC[8]WhenachargedparticletransversetheTPC,itionizesthegasatomsormoleculesalongthetrajectory.Theliberatedelectronsdriftintheelectrictowardsthedetectionregion14whichconsistsofwireplanesandpadplaneasshowninFigure1.10.Tomeasurethepositionoftheparticletrailwithhighprecision,theelectricneedstobeuniformthroughoutthedriftingregion.Inaddition,thereisahighmagneticappliedinparallelwiththeelectricld.WiththepresenceofmagneticitprovidesthepossibilitytoobtainthemomentumoftheparticlefromthecurvatureandalsominimizesthelateralofelectronsindirectionsperpendiculartotheelectricandmagneticForatypicalTPC,suchasSˇRITTPC,isreducedbymorethananorderofmagnitude.Attheanodewires,electronsaremultipliedandinduceimagechargesonthepadplanenearbywhichisatgroundpotential,andsegmentedintoelectricallydistinctpads,eachconnectedtoitsownTheseinducedsignalsarereadoutviatheandtherestofelectronicreadoutsystem.Thepositionoftheinducedimagechargesonthpadplanedeterminetheparticle'strajectoryintwodimensions.Thethirddimensionisobtainedfromthedrifttimetothepadplanefromwheretheionizationoccurstotheanodewires.Oneofthemajordetectorsforrelativisticheavy-ioncollisionsisaTPC.TheTPCcanprovidethe3-dimensionalpictureofhundredsorthousandsofparticletracksfromasinglecollision.Fewotherdetectorcanhandlesuchahugemultiplicity.InFigure1.11,thousandsofparticletracksarereconstructedfromtheSolenoidalTrackeratRHIC(STAR)TPC[8].15Figure1.11:Thereconstructedtracksfroma200GeVpernucleonAu-AucollisionusingtheSTARTPC[8]1.4SˇRITTimeProjectionChamberTheSAMURAIPion-ReconstructionandIon-TrackerTimeProjectionChamber(SˇRITTPC)[9]hasbeenconstructedatMichiganStateUniversityaspartofaninternationalcol-laborationstoconstrainthesymmetryenergyterminthenuclearequationofstate(EOS)attwicesuprasaturationdensityregion.Thedetectorwillbeusedinconjunctionwithsu-perconductingSAMURAIdipolemagnetoftheSAMURAIspectrometer[35,36]atRIKEN,Japan[37,38,39].Figure1.13illustratestheoperationoftheTPCforapositiveparticletraversingthechamber.WhenachargedparticlepassingthroughtheactivevolumeoftheTPC,itionizesthegas.TheionizedelectronsdriftalongtheelectricasshowninblackarrowsintheFiguretotheanodewiresandgetmultipliedthere.Theimagechargesinducedonthepadsprovideaprojectionoftheparticletrajectoryonthehorizontalplane(x;z).Herethepositiveparticlebendstobeamleft(counter-clockwise).Anegativeparticlewouldbendto16Figure1.12:AnexplodedviewofSˇRITTPC.MoreinformationisavailableinSectionDesignandconstructionoftheSˇRITTPC.Table1.2:DesignparametersoftheSˇRITTPC.Pad12mmx8mmGasgain1270Numberofpads12096(112x108)Electric142V/cmlength50cmDriftvelocity5.5cm/sPressure760TorrMultiplicitylimit200GasmixturesP-10(90%Ar+10%CH4)Rangeofparticledetectionˇ,Z=1-8beamright(clockwise).Ifthedirectionofthemagneticisreversed,asinthecaseoftheSˇRITTPC,thedirectionsofthetracksarereversedfromcounterclockwise-toclockwiseforpositiveparticles.Theverticalcomponents(y)oftheparticletrajectoryisobtainedfrom17Figure1.13:OperationoftheTPC[9]thearrivaltimeofelectrons.TheSˇRITTPChasadoptedsomedesignparameterfromtheEoSTPC[40],whichoperateswithsimilarmagnetgeometryandhasrequirementsforpion-trackreconstruction.Figure1.12showsanexplodedviewoftheSˇRITTPC.Thedetectorisarectangularbox.Theouterenclosurehasthedimensionof2.06mlongx1.50mwidex0.74mhigh(seeSection2.4).Thedcageisdesignedtohaveadriftlengthof50.9cm.Thewallsofthecagehave6mmwidecopperstripswith4mmgapbetweenthem.OneimportantfeatureoftheSˇRITTPCistohaveathinwall.Thisallowslightchargedparticlestoexitthegasvolumeandinteractwithascintillatortriggerarraythatsamplesfrommultiplicityoftheparticlesthatexitthecage(seeSection2.9).TheconstructiondetailsandpropertiesofthecageisavailableatSectionFieldcageoftheSˇRITTPC.Thearrangementof18thepadandwireplanesresemblethatoftheEoSTPC[40](seeSection2.3).SignalsfromthepadsintheSˇRITTPCarereadoutbytheGenericElectronicforTPCs(GET)whichwillbediscussedinSection2.8.TheP-10gasmixtureswillbeusedintheSˇRITTPCexperiment.ThegaspropertiesisdiscussedinSection1.5.InthedesignoftheSˇRITTPC,wedecidedtoemploythemultiwiredriftchambertechnologyinsteadofmorerecentlydevelopedtechnologiessuchasGasElectronMultipliers(GEMs)[41]andMicromeshgaseousstructurechambers(Micromegas)[42].Micromegasuseathinmetalmeshinsteadofanodewires.Themeshissupportedasmalldistanceabovethepads.ThereisasimplewireplaneabovetheMicromegasprovidingstrongelectricinwhichtheavalancheforms.GEMsaremadeofplasticfoilswhicharemetalcoatedonbothsideswith50-100mdiameterholesinthem.Themetalcoatingsarebiasedtoafewhundredvoltstocreateastrongelectricintheholes.Electronsdriftingthroughtheholescreateavalancheasmuchasthatofaroundanodewires[8].WhenaMICROMEGASgasisemployed,theelectronsaredirectlydepositedonthesensepadsofTPC.ThisisoftenthecasewithaGEMgastechnologyisused.However,thereareexamplesofwhereaGEMisusedasanadditionalgasstagebeforeamultiwireproportionalcounterreadoutasusedintheSˇRITTPC.Barringthislatteroption,bothGEMSandMICROMEGAStechnologieshaveanintrinsicspatialresolutioninthepadplanecoordinatesthatislimitedtothepitchofthepadsonthepadplane,whichistypicallyoftheorderof5-10mm.AsdiscusedinSection1.3,muliwireproportionalcounterreadouttechnologiesasemployedintheSˇRITTPChaveinducedchargesinmanypadsthathavebeusedtointerpolatethetracktoafactorof10lessthanthepadpitch,whichismuchmoreprecise,asdiscussedinSection2.2.191.5ChoiceofgasAvalanchemultiplicationcanoccurinanygas[27].Somegasesorgasmixturesmayhavetpropertiessuchaslowworkingvoltage,highgainoperation,goodproportional-ity,fastrecovery,etc.Therefore,thechoiceofgasdependsontheindividualexperimentrequirement,whichtheTPCisdesigned[43].Afterionization,electronandionscreatedbyamovingchargedparticledriftintheop-positedirectionstotheanodeandcathoderespectively.Manysubsequentcollisionsbetweenelectronsandthegasatomsormoleculesoccurduringthedriftofthesecharges.Electronsandionshavedramaticallytmobilitiesinthegas,inparticularionshavedriftve-locitiesinthegasthataretypically3ordersofmagnitudesmallerthanaretheelectrondriftvelocitiesinthesameregionofthedetector[44,45].Iftheelectricisverystrong,asinthevicinityoftheanodewiresoftheTPC,freeelectronsthatareacceleratedbytheappliedelectricbetweencollisionswiththegas.Forlowtheelectronicvelocityisrandomizedbetweencollisions.Inanalogywiththedriftofelectronsthroughconductors,theelectronsonaveragedriftinadirectionsoftheelectricwithadriftvelocitythatisproportionaltotheelectricandinverselyproportiontothegasdensityoressentiallythegaspressure.Incaseoftlyhighelectrictheelectronsmayachievetlyhighkineticenergybetweencollisionswiththegasmoleculessothatthehaveenoughenergytoionizeneutralgasmoleculesduringthenextcollisions,increasingofthelocaldensityoffreeelectronsandions.Typically,thethresholdforthissecondaryionizationisoftheorderof106V/mfortypicalgases[10].Allelectrons,includingthoseproducedbythesecondaryionizationwillbesubsequentlyacceleratedbytheelectricandfurtherionizethemoleculesinthegas,leadingtoa20rapidincreaseinthenumberofelectronsmovingtowardstheanode.Ifnisthenumberofelectronsatagivenposition,afterthepathdx,theincreaseinnumbercanbeexpressedbydn=dx(1.9)whereistheTownsendcoientwhichhastheunitof1/[Length].Theinverseofisthemeanfreepathforanelectroninthisgasandthiselectrictoionizingcollisionproducingoneadditionalelectron.Thepositiveioncreatedintheionizationmaycollidewiththecathodesurfaceandliberateanelectronfromthesurfaceofanelectrode.ThesuccessrateofemittinganelectronfromthesurfaceofanelectrodeisdescribedbythesecondTownsendcot[46].TheTownsendcontiszerofortheelectricbelowthethresholdandgraduallyincreaseswithincreasingstrengthabovethethresholdasshowninFigure1.14.Figure1.14:TheTownsendcoentasafunctionofelectricstrength[10].Figure1.15showstheTownsendcotforvariousnoblegases.Normally,com-21Figure1.15:TheTownsendfornoblegases[11].plexmoleculeshaveahigherthreshold(E=p10)fortheavalanchemultiplicationtooccurthannoblegases[27].Therefore,noblegasesareoftenusedasamaincomponentindetectorwhenhighgasaisdesired.XenonorKryptonismuchmoreexpensive.Ingeneral,Argonisamorechoice[47].However,usingapurenoblegascanbeaproblem.Theexcitednoblegasescanonlyreturntogroundstatebyradiationprocess.ForArgon,theminimumenergyforanemittedphotonis11.6eV[48].FortheSˇRITTPCthisenergyiswellabovetheworkfunctionofthecathodemadeofaluminum(4.08eV)[49].Therefore,thesephotonscanexciteelectronsfromthecathodeandsidewallsleadingtotheemissionofelectronsfromremotesurfacesofthegascontainmentvesselandadelocalizationoftheionization.Thisistypicallycounteredbytheadditionofmoleculargasadditivestothecountergastoquenchtheseemittedphotonsasdiscussedbelow.22WhenArgonisionized,theelectronswilldrifttotheanodeandtheArgonionswillmigratetothecathodewheretheionsareneutralized.Theneutralizationcanleadtotheemissionofphotonsortotheemissionofanotherelectronfromthemetalsurface.Indeed,ifthephotonemissionisnotsuppressed,itcanresultinapermanentdischargeinthegasdetectorevenatamoderategasgain.Complexmoleculescontaininglargeamountofnon-radiativeexcitedstatesallowsawiderangeofphotonabsorption.Forexample,methaneisantabsorberintherangeof7.9to14.5eV[50]whichcoverstherangeinenergyofthephotonsemittedbyArgon.However,theadditionofexcessivequenchinggasesorpoorchoiceofquenchinggasescanbeaproblem.Inparticular,largemoleculestendtoformpolymerswhichionizedandthesepolymerstendtoattachtothesurfaceofanodewiresorthesurfaceofthecathode.Thiswillreducetheperformanceofthedetector(seeSection3.1).Thefactor,G,canbeobtainedbyintegratingEquation1.9betweenSminwheretheavalanchestartsandthewireradiusa:G=nn0=exp[ZaSmin(s)ds]=exp[ZE(a)Emin(E)dE=dsdE]:(1.10)nandn0aretheandinitialnumberofelectronintheavalanche[11].InFigure1.16,gasgainfortmixturesareobtainedfromGARFIELDsimulation[51].TheP-10gas(Argon90%+Methane10%)willbeusedfortheSˇRITTPCexperiment.LaterinthisSection,wewilldiscussthepropertiesoftheP-10gas.Inthepresenceofelectricelectronsdriftalongthewithadriftvelocity,vd.Thedriftvelocitydependsonthepressureandtemperature.Thecontaminationinthegassuchaswateroroxygencanthedriftvelocityaswell[52].Figure1.17demonstratesthe23Figure1.16:Gason(gain)fortmixturesasafunctionofanodevoltage.driftvelocityofelectroninP-10gasasafunctionoftheratioofelectricEandpressurep.ThereisapeakofthedriftvelocityatE=p=0.14V/cm/mbarwhichcorrespondstotheelectricof142V/cmforanatmosphericpressure.Thedriftvelocityatthepeakis5.5cm/s.Ifoneoperatesagasdetectorsothatthedriftvelocityisnearthispeak,thedriftvelocitywillbelesssensitivetochangesinthegasdensity,whichisgivenbythetemperatureandpressure,andtovariationsintheelectricIntheSˇRITTPC,thedriftvelocityisimportantfortheoperationofthegatinggrid.SincetheSˇRITTPChasthedriftlengthof50.9cmwhichisthedistancefromthecathodetopadplane.Ifthedriftvelocityis5.5cm/s,thegatinggridneedstoopenfor9.25saftertriggeredbyacandidateeventtoallowallionizedelectronspassthroughthegridtotheanodeplanelocatedat1cmabovethegatinggrid.Thecurrentdesignofthegatinggriddrivercanopenthegatinggridin350nscorrespondingto1.9cmofdriftlength.Someionizedelectronscreatedfromthe24candidateeventlocatedwithin1.9cmbelowthegatinggridwillnotpassthroughthegridafterthegridisopened.Figure1.17:DriftvelocityofanelectroninP-10gasasafunctionoftheratioofelectric[12].ThedriftvelocityincreaseswiththepercentageofMethanefortheP-10mixturesasshowninFigure1.18.Inaddition,thealsoplaysanimportantroleinthespatialresolution.Electronsandionsmovingalongtheelectricscattertheatomsandmoleculesofthegas.ThiscausesvariationinthevelocityalongtheelectricwhichleadstoaspatialinthedirectionparalleltotheelectricandalateralperpendiculartotheelectricThelongitudinalandtransverseforP-10asafunctionofelectricareshowninFigure1.19and1.20.Inthecalculation,theelectric25Figure1.18:DriftvelocityofanelectroninP-10gasasafunctionofthefractionofMethane[13].EisparalleltomagneticB.ThetransversedecreaseswithincreasingmagneticInthetrackreconstructionoftheTPC,thelongitudinalcausesanuncertaintyofthepositionwheretheionizationoccursiny-direction(vertical)whilethetransversecausesanuncertaintyofthepositionin(x,z)plane.ThelattercanbesuppressedbyintroducingamagneticparalleltothemainelectricldoftheTPC.26Figure1.19:LongitudinalforP-10[13].Figure1.20:TransverseforP-10[13].27TheangleofthedriftingelectronmakeswiththeelectricisasLorentzangle.TheLorentzangleisoneofimportantpropertiesofP-10.ItcanhelptounderstandtheofEBonthetransportofelectronsneartheanodewires.Themisalignmentoftheelectricandmagneticneartheanodewirescanresultindistortionofelectrondriftlineswhichthepositionresolutionandarrivaltime[53,54,55].ConsiderthedriftlinesintheplaneperpendiculartotheanodewiresasshowninFigure1.21(a).Figure1.21:(a)Electrondriftlinesintheplaneperpendiculartotheanodewire.(b)ASpreadingoutoftheelectronsalongtheanodewireduetotheExBTheelectrontrajectoriesthatstartatzvaluesfurtherfromthewirehaveportionswheretheyarenearlyhorizontal,i.e.perpendiculartothemagneticThesetrajectoriesaremorebytheExBdriftvelocitythanthoseclosertotheanodewire.TheoftheExBdriftvelocityorLorentzangleareclearlyseeninFigure1.21(b)whichshowstheprojectionofthesetrajectoriesontheplaneperpendiculartothemagneticThehorizontallineistheionizationtrackoftheoriginalparticle.Thedashedlineshowtheprojectionoftheelectrondriftlines.Trajectoriesoriginatingfartotheleftare28upwardsandthoseoriginatingfurthertotherightareecteddownwards.TheExBspreadsoutthechargefromtheoriginaltrackalongthewireworseningthespatialresolutionofthechargedistributionalongthewire.ThedriftlinescomingfromregionsalongthetrackhaveasmallerLorentzanglecorrection.Therefore,thechargewillnotbeshiftedasmuchalongthewire.Alternatively,onecanunderstandthisbyconsideringtheelectronsoriginatingfurtherawayfromthewireexperienceamuchweakerelectriccomparingtothoseclosertothewire,theelectriclinesforthoseelectronsoriginatingfurtherawayarenearlyperpendiculartothemagneticandtheLorentzangleislarge.Whenthoseelectronsdriftclosertothewire,theoftheExBdriftvelocityandtheLorentzangledecreasesingthehighelectricasshowninFigure1.22.Figure1.22:LorentzangleforP-10withmagneticof0.5T.Theanglesbetweentheelectricandmagneticare0,15,30,45,60,75and90degrees[13].29Chapter2DescriptionsofSPiRITTPC2.1ThedesignandconstructionofthecageoftheSˇRITTPCThecageoftheSˇRITTimeProjectionChamber(TPC)isarectangularboxwiththeinteriordimensionsof144.64cminthehorizontalzdirectionofthebeam,by96.61cminthehorizontalxdirectionperpendiculartothebeamand51.10cmintheverticalydirectionperpendiculartothebeam.Thecageisagas-tightvolumewithP-10gas.Figure2.1is3-dimensionaldesignofthecagewithoutapolyamidesheetontheexitwindow.Thecageconsistsofemainpartswhichareatopperimeter,sidewalls,afrontwall,anexitwindowandacathode.Thesideandfrontwallsaremadeof1.6-mmthickhalogen-freeG10printedcircuitboards.G10isaerglass-epoxylaminateconsistingoflayersoferglasswithanepoxybinder.Theexitwindowisconstructedofapolycarbonateframeanda125-mthickpolyamidesheetwiththedimensionsof39by81cm.Theentrancewindowhasa4-mPPTA(polyp-phenyleneterepthalamide)foilwiththedimensionsof6by7cm.Thefourverticalcornerswhicharemadeofhalogen-freeG10andconstructedfromquartersectionsofa4-inchdiameterG10tube.Thismakesthecornersroundedavoidingthehighelectricgradientsthatwouldbeassociatedwithangleatthecorner.Thedetailsofeachpartofthecagewillbediscussedlaterinthissection.30Figure2.1:3dimensionaldesignofthecageoftheSˇRITTPCwithouttheexitwindowpolyamidesheet.2.1.1ConductivepaintedcomponentsToprovideauniforminsidethecage,thecornerpieces,sidesupports,exitandfrontwindowframesneedtohavestripsthatextendovertheinsulatingsurfacesonthesepiecestomatchtheconductingequipotentialsurfacesonthewalls.WeusetheCHO-SHIELD610conductivespraycoatedepoxypainttoprovideaconductivesurfaceonthosecomponents.Itisatwo-componentepoxyspraypaintwithsilvercoatedcopperesembeddedinit,whichformedaconductiveepoxysurfacewhencured.InFigure2.2,weshowtheinsulatingpolycarbonatesidesupportbarsjustbeforetheywerepainted.Here,theclearpolycarbonatesurfacesarecoveredby4mmwideKaptontape,whichwasusedtomasktheinsulatingsurfacesthatshouldnotbespraypaintedbytheconductingepoxy.ThisKaptontapehadaacrylicadhesivethatcouldbereadilyremovedbycleaningwithethylalcohol.Similartechniqueswereusedforalloftheotherinsulatingsurfaces.InFigure2.3,onecanseehowthefrontwindowframewaspaintedwiththeconductive31Figure2.2:Preparethesidesupportforbeingpaintedwiththecopper-silverepoxypaint.Kaptontapeisusedtomasktheareasthatdowantthepainttogoover.Figure2.3:Thefrontwindowframewaspaintedwiththeconductingpaintwasanepoxywithsilvercoated.epoxypaint.Aftertheepoxypaintcured,theframewasunmaskedandcleanedcarefullywithethanol.Thesameprocedureswereusedtocreatenecessaryconductivesurfacesonthethreeothersurfaces.Allconductivesurfaceswerecheckedforcontinuity.Thepaintedconductivestripsmatchedtheelectrodesurfacesontheneighboringcircuitboardsonthecagewallstobetterthanamm.322.1.2SidewallsFigure2.4:DrawingofasidewallPCBofthecage.ThereareholesfortheLASERcalibrationcomponents.TheforthelaserportsarenotshowninthisMostofthesidewallsofthecageareformedof1.6-mmhalogen-freeG10printedcircuitboards.Eachsidewallwasformedof3PCBs;thesidewallcontained6suchPCBintotal.ThedimensionsofeachPCBare46.3by49.6cm.Thereare6-mmwidecopperstripsand4mmgapsbetweenthestripswhichcorrespondtoa1-cmpitchontheinteriorandexteriorofthecageasshowninFigure2.5.The4mmgapbetweenconductingstripswaschosentoallowtheoperationathigherpossibleElectrostaticdriftTherearevias,shownlargerthanscaleonthedrawing,whichconnecttheinsideandoutsidestripsthatareatthesameelectrostaticpotential.InFigure2.4,thereareholesformountingtheLASERcalibrationcomponents.Theselaserholeshavebeenequippedwithquarzwindowsandmountingtoholderopticsconnectionsforacalibrationlaser.Whilethelasersystemshasbeentested,theyhavenotyetbeenmountedontheSˇRITTPC.33Figure2.5:Afterthewallswereassembled,theinteriorhas50stripsand49stripsfortheexterior.Thestriponthetopsinksintothetopperimeterandtheoneatthebottomsinksintothecathode.Thetechniqueforgluingthecageisthefollowing.Toassembledthesidewalls,theregionswherethewallsweregluedtothesidesupportswererstsandedbyusinga220-gritsandpaper.AKaptontapewithacrylicadhesivewasusedtomaskareastowhereglueshouldbeapplied.Inaddition,a6-inchwideplasticpiecesweretapedonbothsidesofthewallstoprovidemoreprotectionforthePCBs.Then,3-mmbeadsofAraldite2013epoxywereappliedonthesidesupports.Then,2-56brassscrewswereinsertedthroughtheboardsandintothesidesupports.Positioninghasbeendonecarefullytopreventexcessmovementandsmearingoftheglue.Then,thePCBsandsidesupportswerepressedtogetherandthescrewsweretightedthatbothpositionedandsecuredthePCBsandcircuitboardswhilethegluehardened.Then,excessglueonbothsidesiscleanedwithethanol.Finally,thesidewallwasundisturbedonthesurfacefor10hourstocureasshowninFigure2.6.EachsidewallconsistsofthreePCBs.Tominimizehandlingwhiletheglueishardening,weglueed34eachsideintwosteps,PCB1and2weregluedandafterthegluesets,thethirdPCBwasattachedtohaveacompletesidewall.Figure2.6:Sidewallofthecagewaslaidonthesurfacefor10hourstocureAftertheepoxycured,wesolderedthincopperpiecesontheinteriorofthecagetoconnectallthestripsontheneighboringPCBs.Ontheexteriorofthecage,wireswerealsosolderedonthewallstoensureawellelectricalpotentialontheoutsideofthesupportbarsandaroundthecurvedcornersofthecage.ThisalsoensuredelectricalcontinuitybetweenthepaintedstripsonthesidesupportsandPCBs.ThisprocesswillbediscussedindetailsonSection2.1.6.Figure2.7:SolderthincopperpiecesonthesidewallstoconnectthestripsonthePCBs.352.1.3FrontwallInthemechanicaldrawinginFigure2.8,thefrontwallofthecageisshown.ItconsistsoftwoPCBwalls,awindowframe(green)andaremovablewindow(purple).TheremovablewindowisscrewedonthewindowframeasshowninFigure2.9.Onthewindowframe,therearetwogaschannelswhichareusedtoreroutethegasfromtheconnectionatthetopofthecagethroughthewindowframetothebottomofthecage.Thisisdesirablebecausethegasconnectionscanpresentapointwheresparkscanoccur,thusitisbetterthattheseconnectionsaremadeatthetopwherethevoltageislow.Thesegaschannelsweremadegastightduringthegluingprocess.Theprocedureforassemblingthefrontwallissimilartothatofthesidewall.Firstofall,allglueregionswerepreparedbyusinga220-gritsandpaper.TheglueexcessareasweremaskedThegluebeadswereappliedalongsideofthegaschannelsonthewindowframe.Thegluebeadontheinsideofthechannelsneedstobemuchsmaller(1mm)thanthatoftheoutsideofthechannels(3mm)sothattheexcessgluewillnotthegaschannel.Figure2.8:ThefrontsideofthecageconsistsoftwoPCBsandthewindowframewhichisgluedtothetwowalls.Green:thewindowframegastwogaschannelswhichareusedtothegasintotheTPCfromthebottom.Purple:Theremovableinsertedwindowisscrewedonthewindowframefromtheinsideofthecage.36Figure2.9:Explodedviewofthefrontsideofthecage.ShowninthedrawingaretwogaschannelsforthegasintotheTPC.InFigure2.10,theboardswereslowlypositionandloweroverthewindowframe.Thewindowframewasscrewedtothewalls.Aftercleaningtheexcessglue,thefrontwallwascarefullyedoverandlaidonthesurfacefor10hourstocure.Figure2.10:Afterpositionthewindowframeandboards,theboardswereattachedtothewindowframewiththebrassscrews.ThepurplepieceinFigure2.9representstheremovablewindowofthecagetowhichtheactualwindowisglued.Ithasthedimensionsof9.9cmwide,16.9cmlong37(a)Frontsideofaninseretedwindow(b)BacksideofaninsertedwindowFigure2.11:InsertedremovablewindowofthecageFigure2.12:Theinsertedwindowisputonthefrontwallofthecage.Thecopperprovideelectricalconnectionstothestripsonthewalls.and0.9525cmthickandisconstructedfrompolycarbonate.InFigure2.11(a),thethin(4m)PPTAfoilformingthebeamentrancewindowisshown.Ithasthedimensionsof5.73by7.0cm.Thisallowsthebeam,producedandscatteringparticlestopassintothe38cagefordetectionbytheTPC.ThecopperasseeninFigure2.11provideelectricalconnectionstotheconductivestripsonthewallasdemonstratedinFigure2.12.ThestripsontheinsertedwindowontheexteriorwerealsoconnectedtothestripsonathinPPTAfoilbycopperasshowninFigure2.13.AluminumconductivestripswereevaporatedontothewindowusingthelargeNSCLevaporator.Theexteriorsideofthewindowhasano-ringseal,whichiscompressedbyscrewsthatarescrewedintothewindowfromoutsidetheldcage.Shouldthewindowbebroken,thewindowframecanberemovedandthewindowreplaced.Oneshouldnotethattheconductivestripsontheinteriorofthewindowarealignedwiththeinteriorcagestrips,butare8mmwideandseparatedby1mmfromtheneighboringstrips,unlikethecaseofthecagestrips,whichare6mmwidewith4mmgaps.ThischoiceisoptimizedfortheP10gasofinitialTPCoperation,wherethecageelectricwillhavealowvalueof130V/cm.Also,oneshouldnotethattheconductivestripsontheoutsideofthewindowarealignedwiththoseontheinside,unlikethecaseoftheldcage.Thusthereisashiftintheseexternalstripsontheoutsideofthewindow.Thischangeisrequiredtoensurethattherearenosparksthroughthewindowtoavoltagetialbetweeninsideandoutsidestripsthataretelectrostaticpotentialwhilebeingonlyseparatedbythewindowthickness4m.2.1.4ExitwindowTheexitwindowandwindowframeoftheSˇRITTPCwereconstructedoutofthreepolycar-bonateframesasshowninFigure2.14.Thepurpleframewasgluedandscrewedpermanentlytothecage.Theouterdimensionsofthisframe(purple)are49.6by90.96cmandtheinnerdimensionsare43.45by84.76cm.Onthissideoftheexitwindowframe,thereare39Figure2.13:ThestripsonathinPPTAareconnectedtothestripsonthewindowframebycopperholesfor2-56screws.Theseareusedtoattachedtheframetothecornerpieces.Theothertwopieces(gray)werecombinedtomaketheremovabledownstreamwindow.Thesetwopiecesweregluedtogether.The125mkaptonwindowwasstretchedandgluedtotheinnersurfaceoftheoftheremovablewindowframeasseeninFigure2.15.ThealuminumconductivestripsonthePPTAwereevaporatedontothemwithalargeevaporatorsetupontheNSCLdetectorlaboratory.Aluminumconductingepoxywassprayedontheouterwindowframestocontinuetheconductingstripsontheoutsideoftheremovablewindowtotheedgeofthewindowframe.Smallcircuitboardsthatmatchedthe6mmstripsand4mmgapswerepressedontotheoutsideoftheexitwindowasapressurecontact.WiresleadingfromthesecontactsweresolderedtothesidepanesoftheTPCcagetomakeelectroniccontinuitybetweenallofthestripsthataresupposetobeatthesamepotential.Ontheinsideofthecage,continuityoftheinnerconductingstripswasensuredbyberylliumcopperToensurethatthatthesemakeapositivecontactonthestripsontheinnerwindow,copperstripsweregluedwithsilverepoxytothestrips40ontheinnersideofthewindow.Thecontactsmadecontactwiththesecopperstripswhentherearwindowisscrewedontothecagesandtheo-ringiscompressedthatsealsthewindow.Figure2.14:ExitwindowoftheSˇRITTPCconsistsofthreepolycarbonateframes.Figure2.15:Exitwindow412.1.5CathodeThecathodeoftheSˇRITTPCisconstructedofthealuminumhoneycombwhichisbondedtothecagebyAraldite2013epoxy.Thecathodeisarectangularsolidwiththedi-mensionsof101.6cmwide,149.86cmlongand1.54cmhigh.Theuppersurfaceofthecathodeisasolidaluminumsheet,whilethelowerisperforatedsoastoallowthecathodetobeplacedinavacuum,shouldtheTPCbeevacuatedandlaterwithanexplosivecountergas,whenusedasanactivetarget.Thefourcornersofthecathodeareroundedtoreducesparkingatahighvoltage(5-20kV).InFigure2.16,analuminumroundedgewasattachedtotheedgeofthecathodetofurtherreducethepossibilityofsparking.SincetheworkfunctionofaluminumcanbelowerthantheenergyoftheLASERoriginallypurchasedforcalibration,theinnersurfaceofthecathodewascoatedbygraphiteatthecenterareaof96.4by144.45cmtominimizetheproductionofphoto-electronsbyscatteredlaserlight.Figure2.16:Roundaluminumedgewasattachedtotheedgeofthecathodetoreduceasharpangle.ThecathodeoftheSˇRITTPCisconnectedtotheexternalpowersupply,voltagestep42downandcage.Therearetwosetsof50resistorswiththeresistanceof100usedasvoltagedividertothevoltageinthecage.Betweeneachneighboringpairofstrips,therearetworesistors,eachwith100Mohm,makingtheresistancebetweenneighboringstripstobe50Mohms.Thus,thetotalresistanceofthecageisaboutRFC=2.5Thevoltagestepdownhas7resistorswithindividualresistanceof100foratotalresistanceofRVSD=700InFigure2.17,thecircuitdiagramshowshowthecomponentsareconnected.Thebeginningoftheresistancenetworkisconnectedtoaconductivesurfaceofthevoltagestepdown.TheVoltagestepdowndesign,discussedinsection2.5,hasanupperconductivesurfacewhichisshortedbyaspringloadedcontacttotheundersideofthecathode.ThisuppersurfaceiscapacitivelycoupledthroughapolycarbonatedinsulatingsheettotheTPCenclosure.ThecorrespondingcapacitancetogroundisCVSD=7.09nF.WhencombinedwithRLine,thiscapacitancemakesalowpassthatvelyoutnoisefromthecathodepowersupply.ThecathodeisconnectedtothepowersupplyviaaresistorofRLine=10Figure2.17:Circuitdiagramfortheelectricalconnectionsofthecathode2.1.6FieldcageassemblyInthissection,thedesignandassemblyoftheTPCcagearemoreclearlydescribed.Carefulplanningisrequiredastheepoxysetsuponabout2-3hoursdependingontemper-43Figure2.18:Assemblingthefrontandsidewallsofthecage.ature.Soallassemblyprocedureswerepracticedwithoutglueandtimedbeforetheassemblywasundertaken.Priortogluing,regionsthatshouldnotbegluedonthesideandfrontwallsaremaskedbytheplasticsheetsandKaptontape.First,thesideandfontwallswereassembledasdiscussedinSections2.1.2and2.1.3.ThenacornersectionwasattachedtoeachofthesewallsandtothebackwindowframeasshowninFigure2.20.Afterthesepieceswereglued,thefullassemblyofthecagewasperformed.InFigure2.18,theprocessillustratedfortheTPCsidepanels.Herethisworkisinitiallydoneonatableandthentransferredtoprecisionsurfaceplate.Blocksareusedtosupportthewalls,allowingaccesstothebottomofthewallstoremovetheexcessepoxy.InbothFigures2.18and2.19,thepersonintheforegroundisapplyingtheAraldite2013epoxy,whiletheotherwereapplyinga3-mmbeadofAraldite2013epoxyonthecornerpiece.Onceitisdone,thethirdpersonbroughtthesidewallandattachtothefrontwall.Afterallstripsonthefrontandsidewallshavebeenaligned,thefrontandsidewallswerescrewedtogetherandthenwerepeatthesameprocessfortheothersidewallasseeninFigure2.19.Thenextstepistogluethebackwindowframetothecagewall.InFigure2.20,44Figure2.19:Assemeblebothsidewallsofthecage.Figure2.20:ApplythebeadofAraldite2013epoxytothecornerpiecesofthebackwindowframe.a3-mmbeadofAraldite2013epoxywasappliedtothecornerpieceofthebackwindowframewhereitattachestothesidepanels.Aftertheepoxybeadswereapplied,thebackwindowframewasassembledtotherestofthewallasshowninFigure2.21.Wecheckedthealignmentofallstripsandcleantheexcessglue.Thenthewallwaslefttocurefor1045hours.Figure2.21:Backwindowframewasattachedtothewallofthecage.Aftertheepoxyonthewallcured,thenexttaskistogluetothewalltothetopperimeter.Theregionsonthetopperimeterthatwillbegluedwerepreparedbya220-gritsandpaperandcleanedwithEthanol.AllareathatshouldnotbegluedbutcouldaccumulateexcessgluewasmaskedwithplasticandaKaptontape.Then3-mmbeadsofAraldite2013wereputonthegrooveasshowninFigure2.22.Thecathodewasscrewedtothewallswithoutepoxybecauseitiseasiertomovethewallandpositionittothegrooveonthetopperimeter.Inaddition,thecathodeprovidesrigidityandpreventsthewallfromdeforming.InFigure2.23,thewallwiththecathodewaslowerslowlyoverthegrooveofthetopperimeter.Itisimportantthatthewallwentintothegrooveproperly.ThenthecagewasmovetothesideofthetablesothatwecanscrewthewalltothetopperimeterfromthebottomasseeninFigure2.24anditistruefortheotherside.Afterthewallwasscrewedtothetopperimeter,thecagehasbeentransfertothesupportsforcheckingandcleaningtheepoxyfrominsideandoutsideasshowninFigure46Figure2.22:GlueexcessareaonthetopperimeterhasbeenmaskedwithplasticandaKaptontape.Figure2.23:PlacetothewallofthecageslowlyonthegroovewithbeadsofAraldite2013.2.25.Onceitisdone,thecagehasbeenplacedonthesurfaceplateagaintocurefor10hoursasshowninFigure2.26.Aftertheepoxycured,thenextstepistogluethecathode.Firstofall,thecathode47Figure2.24:Wallofthecagewasscrewedtothetopperimeter.Figure2.25:Fieldcagewasplacedonthesupportforcheckingandcleaningtheepoxy.wasremovedfromthecageandpreparedforgluing.3-mmbeadsofAraldite2013wereappliedintothegrooveonthecathodeasshowninFigure2.27.Thentherestofthecagewasintothegrooveandscreweddown.Theexcessgluewascleanedandthecagewaslefttocurefor10hours.Oncethemechanicalassemblyofthedcageisdone,thechainofresistors,48Figure2.26:Movethecagetothesurfacetocurefor10hours.Figure2.27:Preparethecathodeforgluing.describedinSection2.1.6wassolderedonthewalltothepotentialoneachstripasseeninFigure2.28.Fortheinteriorofthecage,inFigure2.29,eachstripatthecornerwasconnectedacrossthecornerviaawire.Silverconductiveepoxywasappliedbetweenthewireandpaintedconductiveepoxystriponthecornertoprovideabetterelectrical49Figure2.28:Tothepotentialoneachstrip,achainofresistorwassolderedbetweeninthestripsFigure2.29:Wiresweresolderedalongthestripatthecornerforbetterelectricalconnectionsbetweenthestripsoncornerpiecesandthewall.Silverepoxywasappliedonthewireatthemiddleofthestriponthecornerpiecestoprovideabetterconnection.connection.Then,1resistorswereinsertedonthefrontcorneronthebeamleftsidealongthestripsbyusingsilverepoxyasshowninFigure2.30.These1resistorsareused50fortwomainreasons.First,wecancheckwhetherwehaveagoodconnectionforinsideandoutsideoneachstripwhentheexitwindowisattachedtotheTPC.Thetotalresistanceoneachstripbetweentheexitwindowandwallshouldbeintheorderofifthestriphasagoodelectricalconnection.Inaddition,ifthemagneticsuddenlyquenches,havingaresistorcanminimizethemagneticforceonthecagebecauseitwillreducetheinducedcurrentonthestriploop.InFigure2.31,copperwereusedtoconnecttheexitwindowtothecage.TheassemblyoftheTPCwillbediscussedinSection2.7.Figure2.30:Insert1resistorsalongthestrips.51Figure2.31:Copperwereusedtoconnecttheconductivestripsontheexitwindowtoonesonthewall.2.1.7ElectrostaticsofcageIftheelectricinsidethedriftvolumeisdistorted,itwillelectrontrajectoriesfromthepointofionizationtothemultiplicationregionoftheTPC.ThetrackingalgorithmsassumetheelectrictobeparalleltothemagneticWhenthatisnottrue,thetrajectoriesmaynotbeverticalasassumedinthetrackingalgorithmandalsothedrifttimes,whichareproportionaltotheelectricmaybetthanexpected.Eithercanresultinthecomplicatedreconstructionofthetrackandlargepossibleuncertaintiesinthereconstructedmomentaoftheproducedparticlesinthenuclearreactions.Toachievethebestreconstructionofaparticletrack,havingauniformelectricinsidethedriftvolumeisimportant.AsmentioninSection2.1.6,weusearesistorchaintograduallystepdownthevoltagefromthecathodetothegatinggridandmaintainauniformverticalelectricinthedriftregion.AsshowninFigure2.32,10resistors,R,weresolderedbetweenthe52stripfromthecathodetothetopperimeteronbeamleftandbeamrightsidesoftheTPC.Asallstripsatthesameheightareelectricallyconnected,thetworesistor10resistorchainsareinparallelleadingtoaresistanceof5betweentwoneighboringstripsontheinsideofthecage.Anotherresistor,Rp,wasconnectedfromthetopperimetertoground(Topplate).Figure2.32:DiagramshowstheresistornetworkoftheSˇRITTPC.TheeresistancebetweenthecathodetogroundcanbewrittenasReff=49R+Rp:(2.1)ThevoltageoneachstepcanbeobtainedbyVn=Vcath(Rp+(50n)R)Reff;n=1;2;3;:::;50;(2.2)whereVcathisthevoltageofthecathodeandVnisthevoltageonthenthstep.Forn=1,53VnisequaltoVcath.ThevalueoftheRpisadjustabletoensurethattheelectricfromcathodetogatinggridisuniform.IntheSˇRITTPC,thedistancebetweencathodeandgatinggrid,jyggycathj,is49.614cmandthedistancebetweentopperimeterandgatinggrid,jyggytpj,is0.614cm.Therefore,thedistancebetweentopperimeterandcathode,jytpycathj,is49.0cm.Anelectriccanbedividedintotworegion.E1isanelectricbetweentopperimeterandgatinggridandE2isanelectricbetweencathodeandtopperimeterwhichcanbeexpressedbyE1=VggVtpjyggytpj(2.3)E2=VtpVcathjytpycathj(2.4)Ingeneral,E1andE2arenotthesamesincethedistancebetweenthetopperimeterandgatinggriddoesnothavethesamepitchastherestofthecage.Therefore,anadjustableresistor,Rp,maynothavethesamevalueasonesonthestrips.Theelectricintheseregionshavetobematchedtoachieveauniformelectricfromthecathodetothegatinggrid.WeneedE1=E2.Therefore,theexpressionforRpcanbewrittenasRp=49Rf(Vgg;Vcath)1(2.5)wheref(Vgg;Vcath)=yggycathyggytp+(ytpycath)VggVcath(2.6)InEquation2.5,RpdependsontheratioofVgg=Vcath.Figure2.33(a)showstheelectric54inthedriftvolumeoftheTPCasafunctionofVgg=VcathforVcath=-8kV.ThevaluesofRpneededtoadjustthevoltagesonthestripsasafunctionofVgg=VcathforVcathisshowninFigure2.33(b).OnceweknowthestrengthoftheelectricwhichcanbeexpressedbyEquation2.7,wecanusetheratioVgg=VcathforVcathtodeterminethevalueofRp.ETPC=VggVcathjyggycathj(2.7)OnceRpisknown,allvoltagesonthestripscanbedetermined.Figure2.34showstheequipotentiallinesinsideandoutsidethecageoftheSˇRITTPCperformedbyANSYSr"Maxwell".Thecalculationshowsthatadjustingthevoltagesaccordingtotheformulagivesauniformelectricinthedriftvolume.ThevoltageoneachstripisobtainedfromtheEquation2.2.IntheSˇRITTPCthestripsontheupstreamwindowareslightlytfromtherestofthecageasseeninFigure2.35,describedinSection2.1.3.Figure2.36illustratestheelectrondriftlinesnearthefrontwindowoftheSˇRITTPC.Ideally,theelectricinsidetheTPCshouldbeuniformeverywhere.Inreality,theelectricoftheTPCisbythevoltagesonthecathode,gatinggridplaneandthestripsonthewallofthecage.IntheareaclosetothewallorfrontwindowoftheTPC,theelectricdeviatesfromuniformityduetothetransversecomponentoftheelectricThedeviationofelectriccausethedistortionofelectrondriftlines.Typically,theverticalcomponentofthepositionwheretheionizationoccursisobtainedfromthetimethatanelectrondriftsupwardtotheanodewire.Thedistorteddriftingelectronstakelongertimetoreachtheanodewireandthepositioninthehorizontalplanethattheseelectronsreachistfromtheoriginalposition.Theseleadtoverycomplicatedcorrectionofthetrack55(a)ElectricinthedriftvolumeasafunctionofVggVcathforVcath=-8kV(b)RpasafunctionofVggVcathFigure2.33:ElectricinthedriftvolumeoftheTPCandRpasafunctionofVggVcath.56Figure2.34:PotentialcalculationoftheldcageisperformedbyANSYS.Rpis7fortheVcathandVggof-6kVand-110V,respectively.reconstructionprocedures.Inparticular,theelectriclinesinthiscalculationarebentawayfromthefrontwindowandendupdisplacedabout1mmtowardstheinteriorofthecage.However,thistransverseelectricdisturbancedecreasesexponentiallywiththedistancefromthewindow[11].FortheSˇRITTPC,thecalculationfrom[51]showsthatatthedistanceof2.3cmfromthewindow,theelectricisuniformandtheelectrondriftlinesbecomeastraightline.Oneshouldnotthatnearthewindow,theionizationfromacentralheavyioncollisionwillbeverylargeandthetrackdensitywillbeveryhigh.Fortypicalevents,itislikelythatthisregionmaybeverytoanalyze,whichmaylimittheimpactofthisregionontheexperimentalanalyses.57Figure2.35:Thestripsonthefrontwindowframeareslightlytfromtherestofthecade.Figure2.36:ElectrondriftlinesnearthefrontwindowoftheSˇRITTPC.582.2ThedesignandconstructionofthepadplaneoftheSˇRITTPC2.2.1PadplaneassemblyThemotionofionsclosetotheanodewiresandtherequirementthatthepadplaneanditspadsremainatgroundpotentialcausesimagechargestomovebetweentheconductingpadsandtheattachedGenericElectronicsforTPCs(GET)electronics.ThepadplanefortheSPiRITTPChas12096rectangularcopperpadsofx=8mmbyz=12mmarrangedin112rowsand108columns.Heretherowsgointhexdirectiontransversetothebeamandthecolumnsgointhezdirectionparalleltothebeam.Thelongdimensionofthepadarealignedwiththebeamdirection[9].Inatypicalexperiment,thebeamliesalongthez-axisandideally,theelectricandmagneticlieexactlyalongthey-axis.Thepadplaneliesinthex-zplane.Thecircuitboardsforthepadplaneisa6-layeredboardinwhichgroundlayersseparatethecharge-sensitivepads,tracesandreadoutconnections.InFigure2.37,weillustratethePCBconstruction.InthisFigure,thegraylayersillustratethevariousG10thickness.Thepads,thegroundlayersandsignallayersareindicatedintheFigure.AsshownintheFigure,thesignallayersareshieldedfromthechargesensitivepadsbythegroundlayers.Thisisdonetominimizecross-talkbetweenlargesignalsinonepadintoothersignallinewheretheinterestingsignalsfrompions,forexample,maymuchsmaller.Thepadplaneismadefromfourseparatehalogen-freeG10PCBs.Thisisimportantbecausehalogenatomsareextremelyelectronegativeandwouldcaptureionizedelectronscomingfromthetracks.EachPCBcontains3024pads,whichweregroupedinto48unitcellsof63padseach.Signalsfromthese63padsweresentby63tracesto63electronic59Figure2.37:Cross-sectionalviewofthepadplanecircuitboard[9].readoutchannelsoftheGETreadoutelectronicsviathetracesinthepadplaneandsomeshortcablesthattakethesignalstotheASADreadoutboards,discussedinSectionGETelectronics.InFigure2.39,aunitcellisconnectedtotwoseparate44-pinSamtechconnectorsontheothersideofthePCB.OneoftheseSamtecconnectorshandles32ofthe63signalsandtheotheronecarriestheother31ofthesignals.Oneoftheconnectorhas32channelsconnectedtothepadwhiletheotherconnectorhas31channelsconnectedtothepads.Theunitcellonthebeamrightsideisa180degreerotationofoneonthebeamleftsideasshowninFigure2.38.ThePCBscollectthechargefromtheregion.Theyalsotransmitthechargetotheelectronics.Inaddition,theyformpartofthegascontainmentbarrierbetweenthecageandtheairoutsideoftheTPC.Itisextremelyimportantthatthepadplaneistowithinabout100m,rigidandleaktight.ThesePCBsaregluedonthetopplatebyamulti-stagegluingprocess.Duringthisgluingprocess,thetopplateisremovedfromtheTPCandturnedupsidedown.TheTPChasarotationmechanismthatallowsittoberotatedaboutitscenterofgravity.Holdthetopplateintopositionwhilethepadplaneisgluedandlaterwhilethewireplanesareattached.Thetopplatewasboltedtotherotationmechanism,upside-downwithitsinnersurfaceupwards.InFigure2.40,thecenterareaofthetopplatehas384rectangularconnectorfeedthroughholesfortheSamtechconnectors60Figure2.38:Unitcellonthebeamrightsideisa180degreerotationofthatofthebeamleftside.NumbersindicatethechannelwhichthesignalfromthepadisregisteredtotheAGET.61Figure2.39:PadplanewhichwillbeattachedtothePCBsontheotherside.Oneunitcellwilluse2Samtechconnectors.Thetopplatehasribsontheopposite(outside)surface,showninFigure2.41,thatlargelypreventthetopplatefrombendingalongtheedgesofthetopplate.Theseribsalsolargelyforceallpointsonthetopplateatagivenvalueofztolieinastraightline.Thisdesign,however,doesallowopposingcornersofthetopplatetobothbedisplacedupordownrelativetotheothercorners.Inpractice,thesecornerscouldbeoutofplanebyasmuchas2mmdependingontheoftheorwhichsupportedthetopplateandrotationmechanismassembly.Toavoidcompilationsfromthis,wejackedupthecornersoftherotationmechanismtomakethetopplateasaspossible.Duringthegluingprocess,wehavetopreservetheofthepadplaneandthepadplaneneedtoberigidlyglued.Inaddition,weneedtoavoidtheglueontheelectricalconnectorsontheothersideofthepadplane.Ifthereisaleak,weneedsomesolutionstoit.Therefore,wedecidedto62Figure2.40:Topplatewithinnersurfaceupwards.Thereare384rectangularholesforSamtecconnectorfeedthroughs.Figure2.41:Ribsofthetopplateemploygasketsthatwillbegluedtothetopplateandtothepadplaneandholdthepadplaneandrigid.Thenweneedawaytoinjectgluetocorrectforleaks.Afterthetopplatewasreasonablyleveled,thepolycarbonategasketswillbegluedonthetopplate.AsshowninFigure2.42,eachconnectorfeedthroughholeonthetopplatehas2gasketpiecesgluedaroundthehole.Thebiggergaskethastheouterdimensionsof3.000"by1.375"and63theinnerdimensionsof2.500"by1.125".Thesmallergaskethastheouterdimensionsof1.75"by0.80"andtheinnerdimensionsof1.25"by0.55".Thebiggergasketislargeenoughtogoaroundtheareaforoneconnectorwhilethesmallergasketislargeenoughtogoaroundtheconnectorfeedthroughholebutnotcoverthescrewholesforholdingtheconnector.ThestampingmetalpieceasseeninFigure2.43isusedtoputthegaskettothecorrectposition.Thebigandsmallgasketswereholdtothemetalbyalittledropofwater.Oneneedstocheckthatthegasketwillnotdropwhenweturnthemetalpieceupsidedown.ContinuousthreadsofEZpoxy83wereappliedalongthemiddleofgasketedges.Then,themetalpiecewiththegasketsattachedtoitwasplacedonthetopplaceandputalittlebitpressureonthemetalpiece.Afterliftingthestampingpieceup,thegasketsshouldstayonthetopplatearoundthehole.Werepeatedtheprocedure24timesasseeninFigure2.42(a)tomakeonesectionofgasketgluing(Figure2.42(b)).Then,putthe14"x9.5"metalplatewithneoprenefoamandTliningontopofthegluedpieces.Weputsomeweighttoprovideauniformpressuretothemetalplateandleavefor24hoursfortheepoxytocure.(a)Useastampingpiecetoputthegaskettothecorrectposition(b)AcompletesectionofgasketgluingFigure2.42:GasgetgluingAfterthepolycarbonategasketsweregluedonthetopplateasseeninFigure2.44,eachofthefourPCBsformingthepadplanewereseparatelyglued.Wheninplace,thehorizontal64Figure2.43:StampingmetalpieceforgluinggasketpositionofeachPCBwasbytwodowelpinsinthetopplate.Thefollowingprocedurewasusedtogluethepadplanetothetopplate.First,thepadplanewasputintoitcorrectposition,asbythedowelpins.Thenaprecisionvacuumplatewasputontopofthepadplaneandthevacuumwasapplied.Theprecisionvacuumtablewasattachedtoaprecisionalignmentjigthatcouldthehorizontalandverticalpositionandorientationofthevacuumtable,butallowedittoberemovedandreplacedaccurately.AprecisionvacuumtablewasmountedonanalignmentandthenloweredontothePCBwhichpullitThen,thePCBplustablewasraisedtothecorrectheight.AftercheckingatthepositionofthePCBwascorrect,bothtableandPCBwereremoved.ThePCBheldinplacebyvacuumpressure.Then,thinextrudedlinesofAraldite2013epoxywereapplytothegasketsurfacesasshowninFigure2.45thatshouldlieunderthePCB.Thesewereappliedwithtthicknesssothatsomegluewouldbondthetopplatedirectlytothegasketandpadplane,buttgluesoastothespacebetweentheinnerandoutergasketssurroundingaconnectorfeedthroughhole.ThisisbelievedtobeimportantbecausetheAraldite2013displaysagreateradhesionbetweenthealuminumtopplateandthepolycarbonategasketthandoestheEZpoxy83.ThenthevacuumtablewithaPCBwere65returnedtothealignmentabovethegluingsurfacesandloweredtothecorrectheight.Theywerethenheldbyavacuumtableduringthegluingprocess.Itwasnecessarytoapplysomeweighttothevacuumtabletogetthepadplanetothecorrectheight(seeFigure2.46).Thisneedforextraweightwasnotanticipated.Figure2.44:CompletegasketgluingprocessAfterthePCBswereglued,thetopplateattachedtotherestoftheenclosureandthetopFigure2.45:Gluingpatternforthepadplane66platewassubsequentlytestedtoseewhetheritwasgas-tightbyapplyingasmallpositivepressuretotheinsidesurfaceofthetopplate.MethanegaswasinjectedintotheenclosureandtheleakrateofMethanegaswasdetectedwiththeINFICONhydrocarbonleakdetector.Aboutahalfofthe384feedthroughholeshadadetectableleak.Mostwereminor.TheseleakswerepluggedbyinjectingEZpoxy83throughoneofthescrewholeandtherebythevolumebetweenthetwogasketssurroundingeachfeedthroughholes.Thepossibilitythatthiswouldbeneededwasanticipatedandmotivatedtotwogasketdesign.Afterthisgluingoperation,theTPCwasgas-tight.Figure2.46:Pad-planePCBisputtothepositionbythevacuumtable.AFAROlaserpositionsensorwasusedtodeterminetheofthetopplatepriortothemountingthewireplanes.Theresultfromthelaserpositionmeasurementindicatesthatthedistancebetweenthepadplaneandtheanodewiresisconstantwithin125m.Themeasurementsindicatedthatthepadplanewasslightlyclosertotheanodewiresatthecenterofthepadplane.Thismaybeduetotheweightappliedduringthegluingofthepadplanetothetopplate,whichmayhavedistortedthetopplatewhilegluing.InFigure2.47,67themeasurementofthetopplateaftergluingthepadplaneisshown.Thetopplateiswithin5mils(0.13mm).Thisdiscrepancywaswithinourdesignsptions.Figure2.47:TheofthetopplateusingFAROlaserpositionscanner.2.2.2PadresponsefunctionFortheTPC,dE/dxandparticletrackingarethemainmeasurements.TheaccuracyofdE/dxdependsonthegainstabilityandtheenergyresolution.Thedistributionofinducedchargeonthepadplaneandelectroninthedriftregiontheresolutionoftheparticletracking.Thepadresponsefunction(PRF)includestherelationshipbetweentheinducedsignalonthepadandthepositionofatracktravelingparalleltothelength68ofthepad.ThewidthofPRFisanimportantparameterthattheresolutionoftheTPC[56].TheinducedchargedistributiononthepadplanecanbeexplainedbyGattichargedistribution[57,58].Itisasemi-empiricalformulacalculatingthechargedistributiononthecathodeplaneaccordingtothegeometry.Thearrangementofthesystemisapproximatedbyanenclosedcathode-anode-cathodeconsistingofthegroundplane,anodeplaneandthepad-planewhichisalsoatgroundpotential.Withinthisregion,theinitialunmultipliedsecondaryelectronicchargeissmallandthemuchlargerchargesoftheelectronicsandionsproducedbytheavalancheareequalandopposite.Theimagechargesonthegroundandanodeplanearelargelyinresponsetothisavalanche.Therefore,thesumofallsignalswithinthegasandontheelectrodesurfacesatanygiventimeiszero.Thesumofcathodesignalsarethereforeequaltothesumofnegativesignalsoftheanode.ThiscanbeexpressbyIc1(t)=12Ic(t)=12Iw(t):(2.8)WhereIc1isthechargesinducedontheoneofthecathodeplane(padplane),Icisthechargesinducedonbothcathodeplanes(i.e.groundplaneandpadplane)andIwisthechargesinducedontheanodeplane.Thechargesonancathodestripcanbeobtainedbythecathodechargedistribution:dIc1(t;)=Ic(t)(2.9)Z1)=12(2.10)Where=xhisthedistanceofthestripfromtheavalanchepositionin69thecathodeplanexwhichisnormalizedtotheanode-cathodeseparationh.Thechargedistributioncanberepresentedbyasingle-parametersemi-empiricalexpressionknownasGattifunction:)=K11tanh2K21+K3tanh2K2(2.11)TheparameterK1andK2arebyK3:K1=K2pK34arctanpK3(2.12)K2=ˇ21pK32(2.13)ThevaluesofK3areillustratedinFigure2.48forhs>1:0.Here,h=4mmisthespacingbetweentheanodewireplaneandthenearbygroundplaneorpadplanecathodesands=4mmisthespacingbetweenanodewires.Forthesmallervaluesofhs(1:0),K3hasatvalueforthethattheanodewiresareparallelorperpendiculartothelengthofthepadsonthepadplane[58,14]asseeninFigure2.49(a)and(b).Notethat)representsthechargedistributiononthesinglecathodeplanefromasingleavalanche.ForSˇRITTPCwithhs=1:0,thecorrespondingvaluesofK3are0.625and0.550forparallelandperpendicularrespectively.NotethatthevaluesofK3frombothshouldbeofassistanceofestimation.Theparallelisusefulfordeterminingthecoordinateofthetrackrelativetothelong12mmdimensionofthepadsandtheperpendicularisusefulfordeterminethecoordinateofthetrackrelativetotheshorter8mmdimensionofthepad.ThevalueofK3usedintheSpiRITROOTwasobtainedfromextrapolatingthefunction70inFigure2.48withtheexpression:K3=(Ash+B)(Csra+D+Es2r2a)(2.14)Wheresistheanodewirepitch,histhedistancebetweentheanodeplaneandthepadplaneandraistheradiusoftheanodewire.TheEquation2.14givesthevalueofK3=0.7535.Figure2.48:ValuesofparameterK3asafunctionofananodewirepitchs,ananode-cathodeseparationhandtheradiusofananodewirera[14]Forthestripwithwidthwcenteredatpositiononecancalculatethesignalbytheexpression:I(t;w)=Z+w=2w=2dI(t;0)0=Ic(t)Z+w=2w=20)0=Ic(t)P0()(2.15)71(a)ValuesofparameterK3asanodewiresareperpendiculartothepadplane.(b)ValuesofparameterK3asanodewiresareparalleltothepadplane.Figure2.49:ValueofK3forhs1:0[14]72Figure2.50:PadresponsefunctionforSˇRITTPCwithw=8mm(red)andw=12mm(blue)P0()=K1K2pK3arctanhpK3tanhK2+w2harctanhpK3tanhK2w2h(2.16)P0()isthepadresponsefunction(PRF).Thepadresponsefunctionprovidesthein-formationonwhatfractionoftotalcathodesignalsisinducedinthestrip.AsthewidthwincreasesthePRFapproachesthemaximumvalueof12.Thismeansthatanypadcanreadupto50%ofthetotalcharge.ForSˇRITTPC,thewidthofthepadis8mmisinthedirectionparalleltothewireand12mmisinthedirectionperpendiculartothewire.ThePRFsofbothwidthsareshowninFigure2.50.ThePRFofthewidthof8mmshowsthat39%ofthetotalchargeiscollectedatthecentralstripwhileitis45%forthewidthof12mm.Thetwoadjacentpadsforthewidthof8mmcollect5.5%ofthetotalchargeeachanditis2.4%forthewidthof12mm.Inbothcases,thesumofthecollectedchargesfromthecentralandadjacentpadsare50%.Forthetrackmeasurement,anepadresponsefunctionPeff()canbeusedaswell.Peff()canbeconstructedbyasuperpositionofGaussiancurvesanditist[11].732.3ThedesignandconstructionofthewireplanesoftheSˇRITTPCThemultiplicationregionoftheSˇRITTPCconsistsofthreewireplaneswhichareanode,groundandgatinggridplanes.ThearrangementofthethreewireplanesareresembledthatofEOSTimeProjectionChamber(EOSTPC)[40].2.3.1WirewindingandtensionmeasurementAllthreewireplanesoftheSˇRITTPChavebeenfabricatedfromthewire-windingma-chineinthecleanroomattheNationalSuperconductingCyclotronLaboratory(NSCL)atMichiganStateUniversity(seeFigure2.51).Themachineprovidesaprecisepitchandawtensiononeachwire.ThemachinefabricatestwoframesofwiresatatimeasseeninFigure2.51(a).ThetensionandpropertiesofeachwireplaneareshowninTable2.1.Table2.1:PropertiesofwireplanesoftheSˇRITTPC.AnodeGroundGatinggridMaterialGold-platedtungstenBeCuBeCuDiameter(m)207676Tension(N)0.51.21.2Max.Current(mA)60010001000Toperformthewire-windingprocess,ofall,twoidenticalwirearemlymountedonthewire-windingmachine.Then,awireisrunthroughthepulleysystemasshowninFigure2.51(b)andattachedtothemiddlebarbelowthewireframes.Thetensionforthewiresisnedbythecombinationofsprings.Notethatoneneedstorun74awire-windingmachinewithaproperspeedsothatthewiredoesnotbreakduringtheprocess.Afterthewirewindingprocessiscomplete,allwiresarehelpinplacewithfastcuringHardman04001epoxy.Thenthewiresbetweenbothendsofthewireframesarecut.Finally,thewireframehasdetachedfromthemachineandmovedtoacleanboxandtransportedtotheTPC.(a)Wire-windingmachine(b)SpringtensionsystemFigure2.51:Wire-windingmachine75Tensiononthewiresiscrucial.Toavoidthegravitationalsaggingattensionisnecessary.FortheSˇRITTPC,thetensionforanode,groundandgatinggridwiresare0.5,1.2and1.2N,respectively.Tomeasurethetensionofthewiresontheframes,thestringresonancetechniqueisusedtoverifythetensionofeachwireplane.ForthesystemthatbothendsofastringarethefundamentalvibrationmodeofthestringoflengthLcanbeexpressed[59,60]byf=12LsT:(2.17)Wherefisaresonancefrequency,ismassperunitlengthandTisawiretension.ThetensionofawirecanbecalculatedbyT=42f2(2.18)ForthetestsetupinFigure2.52,afrequencyoscillatorisusedtofeedthesinusoidalcurrenttothewire.Astrongearthmagnetisplacedat1cmunderthewire.Theforcegeneratedbetweenthecurrentandthemagneticpullthewirebackandforthinthehorizontalplanewithafrequencyf.Thefundamentalvibrationloopisobservedwhilethefrequencyischanging.Whentheamplitudeofthewireloopreachesthemaximum,thefrequencyisrecorded.ThetensionofthewirecanbeobtainedfromEquation(2.18).2.3.2WireplanecircuitboardThePCBsforallthreewireplanesweremadeofRoger4003.Theareaofthewireplaneisdividedinto14sections.Thisallowssomesectionsoftheanodetooperateatthereducedgasgainforananalysisofheavierparticles.EachPCBfortheanodehas26conductive76Figure2.52:Theschematicofthesetupformeasuringawiretension.(a)anodewireswhentheyarenotresonance(b)anodewiresataresonancefrequencyFigure2.53:measurementofthetensionofanodewires:(a)theanodewiresdonotvibrationwhenthefrequencydoesnotmatchthefundamentalfrequency.(b)Whentheapplyingfrequencymatchesthefundamentalfrequencyoftheanodewire,thewirevibrateswithamaximumamplitude.77padswiththepitchof4mmforthecenterofthepad.Eachpadisconnectedtothehighvoltageviaa10resistorandtoexternalgroundviaa1nFcapacitorasdemonstratedinFigure2.54.ThePCBsforthegroundplanehas104conductivepadswiththepitchof1mmeachboard.Allthepadsareconnectedtothepadnamed"GND"ontheboardasshowninFigure2.55.ThePCBsforthegatinggridplanehasthesamenumberofpadsandpitchasthegroundboard.Thepadsonthegatinggridboardareconnectedtothepadnamed"POS"onthetopsideand"NEG"onthebottomsideoftheboardalternativelyasseenonFigure2.56.Figure2.54:Theprintedcircuitboard(PCB)foranodewires:26padsareconnectedtothehighvoltageby26of10resistors(blue).Eachpadisconnectedtoanexternalgroundby1nFcapacitor(brown).Figure2.55:PrintedcircuitboardforgroundplaneAllboardsweregluedtothealuminumspacertobringupthetopsurfaceoftheboardto78Figure2.56:Printedcircuitboardforgatinggridplanetheheightof4mm,8mmand14mmfromthepadplanefortheanode,groundandgatinggridboards,respectively.EachspacerandPCBwerealignedbythedowelpins.Gluingthegatinggridandgroundboardshasthesameprocedure.Firstofall,weputthekaptontapeonthePCBstoprovideaprotectionfromtheepoxy.Then,EZpoxy83wasapplyonthealuminumspacer.Whenitisdone,thePCBswereattachedtothespacerasseeninFigure2.57.Notethatweneedtomakesurethatthedowelpinsdonotextentpastthetopofthecircuitboard.AfterallPCBswereattachedtothespacers,Auniformpressurewasappliedtotheboardsandleavefor24hourstocure.Figure2.57:GluingthegroundboardGluingananodeboardthespacerhasafewmorestepsthanthegatinggridandground79boardgluing.Sincetheanodeboardhaselectricalcomponentsonit.Weneedtopreventthecomponentfromshortingtothespacer.Therefore,wegluedtheanodeboardwithresistorsandcapacitorsonittotheacrylicspacerwithEZpoxy83asseeninFigure2.58toprovideaninsulation.Theacrylicspacerhastheholewheretheelectricalcomponentswereplacedon.Aftertheepoxycured,wecoveredallelectricalcomponentswithEzpoxy83toprovideanotherinsulationasseeninFigure2.59.Finally,theanodeboardweregluedtothealuminumspacer.Figure2.58:GluingtheanodeboardtoaacrylicspacerFigure2.59:Fillthegapontheanodeboard802.3.3WireplaneassemblyThewireplanehasbeenassembledfromanode,groundandgatinggridplanes,respectively.Notethatallprocesshasbeendoneinthecleanenvironment.Fortheanodeplane,ofall,thePCBsgluedtothealuminumspacerswereputonthetopplateonbothsidesasseeninFigure2.60.AllPCBswerealignedtothepositionbythedowelpinsandtotheTPCwithnon-magneticscrewssincetheTPCtypicallyrununderastrongmagneticTheheightfromthetopsurfaceoftheanodeboardtothepadplanehasbeencheckedtobe4mm.AfterallanodeboardsweremountedtotheTPCproperlyandaccurately,thewirecombassemblywasmountedontothetopplateandalignwirecombsbyusingsquareholesasshowninFigure2.61.Theheightofthewirecombscanbeadjustedbyusinganadditionalspacerwhichcouldbringthecombsuptotheheightofthegatinggridandgroundplanes.Figure2.60:AnodeboardsweremountedontheTPCThen,thewireframewithanodewireshasbeentransportedtotheTPC.NotethatthemaximumnumberofwiresthatthewireframecanhaveishalfofthewholewireplaneoftheTPC.Thewireframehasthelevelingblocksattachedtothefourcornersoftheframewithscrewsprotruding2-3cm.TheseblockwereusedtoadjusttheheightofthewireframeasseeninFigure2.63.Before,wetransportthewireframetothetopplate.Oneneedstocheckthatthelevelingscrewsontheframeextendenoughtobewellabovethewirecombs81Figure2.61:Wirecombassemblyassembly.Afterthat,thewireframewasmovedontothetopofthetopplate.Theheightofthewireframewasadjustedtoberoughlythesameandthenalignthewirestothecombs.Thewiresshouldberoughlyverticallycenteredinthewirecombvalleysandslightlytoonesideofthesolderingpads.Inthiscase,whentheframeispusheddownlaterally,thenthewiresarecenteredinthepad.Afterwecheckthealignment,thewireframearelowereddownslowlyuntilthewiresareinthecombswiththecorrectpitchandcenteredinthepad.Notethatweneedtomakesurethatallwirestouchthepad.Tochecktheconnection,wecanuseavoltmetertocheckwhetherthepadisshorttotheframe.Ifonlysomepadsshorttotheframe,thewireframeshouldbelowereddownslowlyuntilallpadsshorttotheframe.Beforesolderthewires,weneedtoholdallwiresinplacebyapplyingEZpoxy83onthewhiteareaofthecircuitboardasshowinFigure2.62.Inthisprocedure,itisagoodideatohavetwopersonworkingtogether.Thepersonusesasyringelayepoxybeadoverthewhiteareaonthecircuitboard.Thesecondpersonusesthesyringewithanangledtipandremovesexcessepoxyandmakesurethattheepoxydoesnotgooverthesolderpad.Oncetheepoxyhascured,thewireswillbesolderedontothepadasdemonstratedinFigure822.64.Figure2.62:UsingEZpoxy83toholdthewiresinplaceandmaintainthetensionFigure2.63:WireframewithlevelingblocksOncesolderingisdone,theexcesswirewillbecutTopreventthecutwirefromswingaroundthewireplane,weplaceapieceoftapeoverthewiresabout3"fromthesolderjointsandcutfromtheframe.Then,thewireframeiscarefullyliftedandputthetapeoverbothsidesofthecutwires.Theprocedureforthewireplaneassemblyuptothispointwill83Figure2.64:Solderwiresontotheconductivepadontheanodeboardbethesameforthegatingandthegroundplanes.Notethatthepitchofthegatinggridandgroundwiresare1mm.Thedistancefromthepadplaneforthegroundandgatinggridplanesare6and14mm,respectively.Afterthatwireswerecutweneedtotesttheelectricalconnectiononthecircuitboard.Foranodeboard,everywireisconnectedto"HV"pad.Ifweuseamultimeter,itshouldread10betweenthe"HV"padandtheotherendofthewire.Forthegroundwires,oneneedtoensurethatallwiresareconnectedtothegroundpad.Forthegatinggridwires,therearetwosetsofwiresinterlacedwitheveryother.Ensurethatonesetofwiresisconnectedtothe"POS"padandtheothersetofwiresisconnectedtothe"NEG"pad.Oncetheconnectionsaretested.Thewireswerecutasclosetothesolderjointsaspossible.Then,theelectricalfeedthroughconnectorsweremountedontothetopplate.Anode84Figure2.65:ConnecttheanodeboardtotheMHVconnectorboardsareconnectedtoMHVconnectorswhichgoneartheboardwithBNCconnectorsinbetweenasseeninFigure2.65.Topreventtheanodesolderjointsfromsparkingtothegroundwires,anotherlayerofEZpoxy83wasappliedoverthesolderjoints.Also,someEZpoxy83wasputovertheMHVconnectorsandsomeAraldite2013onthesolderedwirestoprovideanadditionalinsulationasshowninFigure2.66.Notethatthewireplaneassemblyprocedureistodooneplaneatatime.Forgroundboards,the"GND"padsareconnectedontheadjacentboardsandthe"GND"padonthemostdownstreamboardwassolderedtotheBNCconnectornearthedownstreamendofeachrowasseeninFigure2.67.Forgatinggridboards,wesolderedthe"POS"to"POS"and"NEG"to"NEG"padsontheadjacentboards.Therearetwotrans-missionlinesaddedtothegatinggridsectioneachsideofthepadplane.Thetransmission85Figure2.66:Insulatetheanodeplaneconnections86Figure2.67:SolderadjacentgroundboardslineismadeofalternativelayersbetweenEZpoxy83andCopperribboncables.EachribboncableinsidethetransmissionlinehastheCopperwireextendstoconnecttothepositiveornegativepolaritiesofeachgatinggridboardasseeninFigure2.68and2.69.Onceallgatinggridboardsareconnectedtothetransmissionline,thetwoendsofthetransmissionlineweresolderedtotheduallemoconnectornearthedownstreamendofeachrow.Figure2.68:Layersofatransmissionline87Figure2.69:Transmissionlineisconnectedtothegatinggridboard2.4ThedesignandconstructionoftheenclosureoftheSˇRITTPCTheenclosureoftheSˇRITTPCisarectangularboxwiththedimensionsof206cmlong,150cmwideand74cmhigh.Thewallsontheenclosureweremadeofthinaluminumsothattheparticlesfromnuclearreactionscanpassthrough.WecanusetheparticlemultiplicitiesthatgothroughtheenclosureonthesideanddownstreamoftheTPCtogenerateatriggerfortheelectronics.InFigure2.70,themotionchassiscanbeattachedtotheenclosureontheupstreamanddownstreamsides.ThisallowstheTPCtobemovedandrotated.Therearetwoclearpolycarbonatewindowsontheleftandrightsides.Thewindowswereputonthetheenclosurewiththeo-ringsothatthevolumeremainsgas-tightandwecanseethecageandthetargetmechanismthroughthiswindows.88Figure2.70:EnclosureoftheSˇRITTPCwiththemotionchassis2.5ThedesignandconstructionofthevoltagestepdownoftheSˇRITTPCIntypicalexperiment,thecathodecanbebiasedupto20kV.Weneedtopreventitfromsparkingtothecomponentsnearby.Thevoltagestepdownwasassembledonthebottomplateoftheenclosure.Itallows20kVfromthecathodeatthebottomofthecagetostepdowntogroundover8copperringsonthebottomplate.Thebottomplateisanaluminumplatewiththedimensionsof142.24cmwide,195.90cmlongand1.27cmhigh.Theplatehasbeencutdownatthecenterby126.87cmwide,174.88cmlongand0.63cmdeepasshowinFigure2.71.Fortheassemblyofthevoltagestepdown,ofall,thebottomplatewaslaidonthe89Figure2.71:DrawingofthebottomplateoftheSˇRITTPCFigure2.72:PourEzpoxy84atthecenterofthebottomplatesurfaceandglueexcessareawasprotectedwithplasticandaKaptoptape.Toprovideaninsulationsurface,apolycarbonatesheetof1/4"(0.63cm)thicknesswasusedtocoverthebottomplate.InFigure2.72,weusedEzpoxy84toglueapolycarbonatesheettothe90bottomplate.TheEzpoxyhasbeenpouredatthecenteroftheplateandthenweslowlyputapolycarbonatesheetoverthearea.Toglueevenlyoveralargesurfacearea,weusevacuuminggluingtechnique.Theprocedureistoputthesealanttapearoundtheareathatweneedtoglueandcoveritwithabreatherclothoverthepolycarbonateandavacuumplasticoverthesealanttape.ThenwepumpeditdownsothatthepolycarbonatesheetwaspresseddownwithevenpressuretotheplateasshowninFigure2.73.Thenweleftittocurefor24hours.Afteritcured,thecovershavebeentakenforcleaning.Figure2.74showsthebottomplateaftergluingprocess.Figure2.73:Polycarbonatesheetwasgluedtothebottomplatebyvacuuminggluingtech-nique.InFigure2.75,aftersuccessfulgluing,aconductivepaintwasappliedatthecenterofthepolycarbonatewhichwillbeatthesamevoltageofthecathodeonthecage.Thenextstepistoinstallcopperringsontothethepolycarbonate.Thevoltagestepdownconsistsof8copperrings,eachofwhichwasformedfrom4straightcopperrodsonstandThelongandshortrodsare145.31and97.31cmlong,respectively.Therodsandwere91Figure2.74:After24hourscuringtime,thecovershavebeentakenandthebottomplatewascleaned.Figure2.75:Conductivepaintwasappliedatthecenterofthepolycarbonatesheet.onthepolycarbonatesheetby0-80screwsasshowinFigure2.76.4quarterroundcornerswereproducedbyweldingasshowninFigure2.77.TheradiiofringscanbeseeninTable2.2.Once8copperringswascomplete,100resistorsweresolderedbetweentherings.Thisallowsthevoltageonthecathodetostep92Table2.2:Radiiofroundcornersofthevotlagestepdown.RingRadii(cm)1(innermost)4.5525.9737.3948.81510.24611.56712.88814.20Figure2.76:Copperrodsandwereattachedtothepolycarbonatesheetby0-80screws.downgraduallytogroundontheenclosure.Thenthebottomplatewiththevoltagestepdownwasmountedtotheenclosure.Totestthevoltagestepdown,drynitrogengaswasintroducedtotheenclosureand20kVwasslowlyintroducedtotheinnermostring.Thereisnosparkingobservedduringthetest.93Figure2.77:Aquarterroundcornerwasproducedbywelding.2.6Thedesignandconstructionofthetargetmecha-nismoftheSˇRITTPCAtargetpositionfortheSˇRITTPCisplacedoutsidethedetector.InFigure2.78,thetargetmechanismfortheSˇRITTPChasbeendesignedtobeabletoadjustthetargetpositioninx,yandzdirections.ThetargetframehaseavailableslotsallowinganexperimenttohavemultipletargetsandalsohastwoforadjustingthepositioninydirectionasshowninFigure2.79.InFigure2.80,theZ-motioncontrolhastwoanglemotioncontrolsasindicatedinyellowandgray.Insidetheboxonthetargetframestructure,thegraygearisdesignedtohaveathread.Therefore,whentheyellowgearisrotated,itdrivesthegrayoneandtheboxwillmovebackandforthinthezdirection.Tokeepthetargetframerigidandavoidbendingjustacenterpartoftheframe,analuminumsupportbarisintroducedonthestructureto94Figure2.78:DesignmodelofthetargetmechanismfortheSˇRITTPC.Figure2.79:Atargetcanbeadjustedinydirectionbypositioningtheonthetargetframe.assurethatthewholetargetframeismovingatthesametime.AsimilarideawasappliedtotheX-motioncontrolaswell.InFigure2.81,whenthe95Figure2.80:AtargetframecanbemovedinzdirectionviatheZ-motioncontrolFigure2.81:ThexpositionofthetargetcanbeadjustedfromtheX-motioncontrol.96anglemotioncontrolisturning,itdrivesthethreadedspline(yellow)andthetargetframewillmoveinthexdirection.TheactualtargetmechanismcanbeseeninFigure2.82.Figure2.82:Actualtargetmechanism2.7OveralldesignassemblyoftheSˇRITTPCTheassemblyoftheTPCwillstartfromattachingtheeldcagetothetopplate.Thereisalexanringbetweenthecageandthetopplate.TheringhasO-ringsonbothsidestoensurethatthevolumeremainsgas-tightafterassembly.Toprovideaprotectionforthewireplanes,acoverplatewasputoverthelexanringwithsetscrewsandwastightenedwithwingnuts.ThetopplatewithacoverwasrotatedbycranetothepositionthatthepadplaneisperpendiculartothegroundasseeninFigure2.83.ClampswereinsertedaroundalexanringtosecuretheO-ring.Oneshouldnotethatthecoverplatemustbeloosenedslightlytointheclamps.Afterinsertallclamps,allbutsixwingnutswereremovedfromthesetscrews.Topreparethecagefortheassembly,thecagewassetonastablesupportsothatupstreamanddownstreamendsareinacorrectorientationwithrespecttothetopplate.Topperimetershouldrestonacleansurface.Priortoattachingthecagetothe97Figure2.83:Topplatewiththecoverplatewasrotatedby90degrees.98Figure2.84:Insertwindowandwereputonthewindowframeofthecage.topplate,theupstreamwindowandcopperelectrodewereputonthewindowframeasshowninFigure2.84.Toattachthecagetothetopplate,theremainingwingnutsandcoverplatewereremoved.Thisoperationmustbedoneinthecleanroom.AsinFigure2.85,twopeoplewerestationedattopperimeterandothertwopeopleatthecathode.Allfourpeopleliftedthecage.Thetwopeopleonthetopperimeterguidedtheperimeterontothesetscrewswhiletheothertwoprovidesupport.Theguildersattached2or3wingnutstosetscrewsatthetopwhiletheothertwocontinuetosupport.Foreachhole,thesetscrewswereremovedandanylonspacerwasinserted.Thenscrewonascrewandanylonwashertotightenthetopplateontothecage.InFigure2.86,springswereattachedtothebottomofthecage.Thisprovidesanelectricalconnectionbetweenthecathodeandconductivesurfaceofthevoltagestepdownandmakethesurfacetohavethesamepotentialasthecathode.99Figure2.85:Fourpeopleliftthecageandsetitonthesetscrewsonthetopplate.Afterthecagewasattachedtothetopplate,theexitwindowwasputonthecage.ThenthetopplatewiththecageattachedonithasbeenrotatedtothepositionasshowninFigure2.87.Theenclosurewasmovedunderneaththetopplatewithacorrectorientation.Someadditionalweightsmayputonthetopplatetoensurethatitislevel.ThenslowlylowerthecraneandpeoplewerestationedateachcornerstoalignthetopplateassemblywiththeenclosureasdemonstratedinFigure2.88.Oncethetopplateandenclosurewerealigned,themotionchassiswereremovedfromthetopplate.Thenwescrewedthetopplatetoenclosureproperly.Twopeopleonoppositecornersweregraduallytighteningscrewsalittleeachtimearoundthetopplateuntilthegapbetweenthemislessthan5mils.Onceitisdone,theSˇRITTPCwithoutatarget100Figure2.86:Springswereattachedtothebottomofthecagetoprovideanelectricalconnectionbetweenthecathodeandconductivesurfaceonthevoltagestepdown.Figure2.87:Topplatewiththecageattachedhasbeenrotatedtothepositionforassemblingtotheenclosure.101Figure2.88:Topplatewasloweringwhilepeopleateachcorneralignthetopplateassemblywiththeenclosure.mechanismisshowninFigure2.89.Finally,atargetmechanismwillbeinstalledontothedetector.AlmostallcomponentsofthetargetmechanismstructurewasassembledtogetherasdemonstratedinSection2.6.ThestructureofthetargetmechanismwasattachedtoaluminumbarsofthemotioncontrolfeedthroughsfromsquarengesonthetopplateasshowninFigure2.90.AlaseralignmenttechniquewasusedtoalignthecenterofatargettothecenteroftheentrancewindowoftheSˇRITTPC.Thelasercrosswillbepointingatthecenteroftheentrancewindow.InFigure2.91,agridpaperwasputonthetargetframe.Thepositionofthelasercrossonthegridpaperisusedtoshowhowmuchwehavefromthecenteroftheentrancewindow.102Figure2.89:TheSˇRITTPCwithoutatargetmechanism103Figure2.90:ThetargetmechanismisinstalledontotheSˇRITTPC.Figure2.91:Thelaseralignmenttechniqueswasusedtoaligntothecenterofatargettothecenteroftheentrancewindow.2.8ElectronicsSignalsfromthepadsintheSˇRITTPCarereadoutbytheGenericElectronicSystemforTPCs(GET),whichisaandscalablemediumsizedsystemupto30k104Figure2.92:GETconceptualdesign[1]electronicchannel[1].Figure2.92showstheGEThardwarearchitecture.Signalfromthepadsaresentviatheprotectioncircuitcards,ZAPandshortcablestotheAGETchipsonanAsAd(AsicandAdc)motherboard.EachAGETchipcanservice64padsandcontainsaPreamp(PA),aSwitchedCpacitorArray(SCA)withamaximumof512timebucketsat1to100MHz,multiplexerandinspectionfunctions.ThephysicalAsAdmotherboardcanbeseeninFigure2.94.DatafromanAsAdboardisreadoutbyaConcentrationboard(CoBo),whichresidesinaTCAcrate.EachCoBoreceivesinputsfrom4AsAdboardsorupto1024pads.Therefore,theSˇRITTPCneedsatotalof12CoBosand48AsAdboardstoreadoutallpadsasshowninFigure2.93.EachAGETallowsthegainandtobeTheoptionsofanAGETareshowninTable2.3.Currently,thegain/channelfortheSˇRITTPCissettobe0.12pC,whichprovidesasimilargainsettingfortheFrontEndElectronics(FEE)cardsoftheSTARTPC[61].Eachchannelhasadiscriminatorwhichallowsforselectivereadout105Figure2.93:GETsystemismountedontheSˇRITTPCoflivechannels.The12bitADCsamplesontheAsAdboardsgiveane10.5bitresolution.EachCoBoisfurnishedwithaXilinxVirtex5FPGAchip,whichiscoupledtofastmemorywithadoublearchitecture.ThefunctionsoftheCoBoaretoits4AsAdcards,(2)collectADCoutputsandperformdatareduction,timestamping,formattingfunctionsand(3)transferdataat1Gb/stothe10Gb/sTCAswitch.GETusesGigabitEthernetviaTCP/IPandembeddedLINUXandVxWorks.Themaximumeventrateforthesystemcanbe500eventspersecond.FortheSˇRITTPCexperiment,theanticipatedrateislessthan100eventspersecond.Theplanforthecoolingsystemfortheelectronicsistousetheairattheroomtemper-atureandblowovertheAsAdboardsviatheplastictunnelsasshowninFigure2.95.TheplastictunnelhasbeenpunchedholesatwhichthepositionofeachholeisrightovereachAsAd.Notethatthesizeoftheholeisimportant.Ifitistoosmall,thetemperatureof106Figure2.94:AsAdmotherboardTable2.3:optionsforAGET[1]ParameterValuePolarityofdetectorsignalPositiveorNegativeNumberofchannels64ExternalYesInputdynamicrange120fC;1pC;10pCGainAdjustable/channelOutputdynamicrange2Vp-pI.N.L<2%Peakingtime69,117,232,501or1024nsPowerconsumption<10mW/channelNumberofSCATimebins512Samplingfrequency1MHzto100MHzReadoutfrequency20MHzto25MHzSCAReadoutmode512cells;256cells;128cellsthewingairwillbehighduetotheadiabaticprocess.Figure(result)showsthatwithouttheaircoolingthetemperatureoftheelectronicsrisesupto40C.Withtheaircooling,thetemperatureisslightlyincreaseandstableat37C.107Figure2.95:PlannedcoolingsystemfortheSˇRITTPC.2.9PlannedtriggersystemThestructureofthetriggertimingfortheSˇRITTPCisasfollows.Therewillbeatleastonebeamtimingscintillatorupstream.Thetpathbetweenthetimingscintillatorandtargetis215cm.Thedistancebetweenthetargetandvetoscintillatorsis187cm.AtdownstreamoftheTPC,therewillbeawallofmultiplicityscintillatorsasthevetoscintillators(Krakowarrays).Thiswallis150cmwide,whichisthesamewidthoftheTPC.AlongbothsidesoftheTPC,therearerowsofscintillatormultiplicitypaddles(Kyotoarray).Inaddition,thereisanactivecollimatorthatis10cmupstreamofthetarget.Eachoftheseelementsofthetriggerhasadelaytoproducealogicsignalandittakestimeforthesesignalstopropagatealongcablestothetriggerlogic.ThesesignalswillbecombinedtomakethesubsequentTTLsignalthatstartsthegatinggriddriver.Eachscintillatortakes3nstogeneratethelightandanother3nsforthelighttopropagatetothephotomultipliertube(PMT),avalanchephotodiode(APD)orMulti-PixelPhotonCounter(MPPC).The108Figure2.96:Triggertimingstructureforthegatinggriddriver.Thediagramisnotinthecorrectscale.startsignalisreadoutbyaPMT,whichtakes25nsanddiscriminatedbyaconstantfractiondiscriminator,whichtakes15ns.Therecanbealogicsignalfromthestartin46nsafterthebeamhitsthestartscintillator.Addingtimesfromthelightgenerationandpropagationtimestothewavelengthshifter(12ns),anddiscriminationtimes,alogicsignalfromthevetoscintillator,activecollimatorormultiplicitypaddlesinabout40nsafterchargedparticleshitthesescintillators.Wehavetoaddthetimeinpropagationtimesfortheparticlestohitthescintillatorsandtheelectronicpropagationdelaytimesfromthescintillatorstothetriggerelectronics.Thelatterdependsontheplacementoftheelectronics.InFigure2.96,oncethelogicsignalreachesthegatinggriddriver(GGD),thechipontheboardtakes10nstogeneratetheTTLsignal(TTL-Open)toopenthegating109gridfor10s.AttheendoftheTTL-Open,anotherTTLsignal(TTL-Close)willbesenttotheGGDtoclosethegatinggridwithin2.5s.ClosingtimeofthegatingcanbeinorderofmillisecondwithouttheTTL-Closesignal.Figure2.97:Plannedtriggertimingdiagram.ThemoduleslabeledFinthediagramaretheFIFO.Figure2.97showatriggerdiagramandVETO,MultandCollunitsaredemonstratedinFigure2.98.Accordingtothediagram,ifawideNIMsignalfromtheactivecollimatorwaspresent,itwillresultinrejectionofthisevent.Therefore,thereisnopremaster,TTL-open,TTL-closeandtrigger.InFigure2.99,thetimesignalforthemultiplicityscintillator,T-Multandstartscintillator,T-StartScinthavewidthsof10ns.Thesignalfromtheactivecollimator,T-Collhaveawidthof100nswhichwouldpreventthegenerationofthepremas-ter,TTL-Open,TTL-Closeandtriggersignals.ThepremastersignalcomesfromthelargeANDinthemiddleofthedigram,whichisshownwithepossibleinputs.Itislikelythat110Figure2.98:ConnectionbetweenthescintillatorsandelectronicmodulestheT-MultandT-StartScintwouldneverbeusedatthesametime;eitherT-StartScintorT-Multwouldbeused.IfT-StartScintisused,thenonegetsapremasterifthecircuitisnotbusyandthereisT-StartScintwithnoT-VetoandT-Coll.IftheT-Vetoispresent,4-swideNIMsignal,itwouldpreventthegenerationofthepremaster,TTL-Open,TTL-Closeandtriggersignals.Also,itwouldkeepthegatinggridcloseuntilthechargefromthisbeampulseisabsorbedonthegatinggrid.ThepremasteroutputservesasthestartforG&D1andG&D2.TheTTLsignalsfromthesegategeneratorsaregeneratedabout22-25nslater,whichisapproximately102nsafterbeamvelocityparticleshitthevetowallwhenthetriggeristhestartscintillator.G&D2generatestheTTL-Opensignal.TheTTL-OpengeneratedbyG&D2requiresboththepremasterasastartandthORofG&D1delaysignaloradelayedT-Vetoasastop.WesetG&D1torequireonlyastartfromthepremasteranditwillgenerateadelayedoutput11slater.IfthereisnoT-Vetofromarandomeventinthemeantime,thisdelayedstopfromG&D1willstopG&D2andthewidthoftheTTL-Openwillbeabout11sandthegatinggridwillopen.AtthesametimethattheTTL-Openisstoppedbythedelayedsignal111Figure2.99:TriggertimingpassingthroughtheOR,thedelayedoutputoftheG&D2willtriggerG&D4andgeneratetheTTL-Closesignal,whichwillclosetheswitchestorechargethegatinggrid.ThedelaytimebetweenthedelayedoutputofG&D2andthestartofG&D4isnecessarytoassurethattheTTL-OpenisreturnedtozerobeforetheTTL-Closestarts.Otherwise,thiswillshorttwopowersuppliesthroughfourmosfetswitcheswithpotentiallydisastrousresults.Ifthereisabeamparticlefromthelaterbeampulsethathitthevetolessthan11saftertheeventthattriggeredthepremaster,theionizationofthebeamparticlewillreachthegatinggridin2.27-4.09s.Ifthisionizationisnotblockedbythetimeitreachesthegatinggrid,itmaycausetheanodewirestospark.Therefore,theT-VetofromthebeamparticlewillpasstoG&D3,whichwillgenerateadelayedoutput2.27sorlater.ThisdelayedoutputgoesthroughtheORandstopsG&D2earlier.Thegatinggridwillstartthecloseatabout2.27safterthearrivaloftheT-Vetosignal.This2.27sdelaymayneedtobeadjustedtomakesurethatthegatinggridisfullyclosedbeforetheionizationfromthis112beampulsepassthroughthegatinggrid.AprototypetriggercircuitwasconstructedtotestthecircuitinFigure2.97.ApulserwasusedtogeneratetwoNIMsignalswhichwereusedtosimulatetheresponseofthecircuittoaneventbyaT-StartScintNIMtimingsignalof10nslengthfromthestartscintillatorandaT-Colltimingsignalof4.09slengthfromtheORofthevetopaddles.Thistestcircuitwasgeneratedusing1leveladapterNIMmodule,one4-foldlogicFIFO,onePhillipScien5-foldcoincidencemoduleandtwoLecroy222gategenerators.Figure2.100:Avetosignal(blue)occurmorethan4sbeforetheT-StartScint.ThetriggersystemgeneratethenormalTTL-Open(yellow),TTL-Close(green).Figure2.100showsthecaseforavalidevent.AT-Vetosignal,showninblue,occursmorethan4sbeforetheT-StartScint,(toplineinpurple).Therefore,thetriggergeneratorsgeneratethenormalTTL-Open(yellow),TTL-Close(green).ThelengthoftheTTL-OpenisbythewidthofthetiminggatefromG&D1.WhentheT-VetoandT-StartScintaresimultaneous,itpreventsapremasterandthesubsequentgenerationofeitherTTL-Open,TTL-CloseortriggerasshowninFigure2.101.Figure2.102showsthecasewheretheprojectile-likeresiduearrivesthevetowhilethegating113Figure2.101:ThereisnogenerationofTTL-Open(yellow)andTTL-Close(green)whentheT-StartScint(purple)andT-Veto(blue)aresimultaneouspresent.Figure2.102:Whenavetosignalpresentsaftertheeventtrigger,TTL-Open(yellow)closesandtheTTL-Closestarts.gridisopen.AfteranadjustabledelaybythearrivalofthevetosignalandbythelengthofG&D3,theTTL-Open(yellow)closesearlierthan11sandTTL-Close(green)starts.InFigure2.103,thereisnoTTL-Close(green)generatedforthisevent.ThelengthofthedelayfromG&D3needtobeadjustedwiththebeamforsucheventssothattheTTL-114Figure2.103:ThedelayoutputfromG&D2passestothestartofG&D4toassurethattheTTL-Open(yellow)andTTL-Close(green)arenotsimultaneouspresent.Openclosesattherighttimetopreventbeamionizedparticlesfromenteringtheanodeplane.Inthemeantime,itallowssometimeforthepreviouseventtobefullycollectedbeforethegateisclosed.ThedelayoutputfromG&D2passestothestartofG&D4sothattheTTL-OpenandTTL-Closeareneversimultaneouspresent.2.10GashandlingsystemTheSˇRITTPCusesP-10gas(90%Ar,10%methane).ThepropertiesoftheP-10gasisdiscussedinSection(Choiceofgas).Figure2.105showstheplannedgashandlingsystemwhichwillbeusedintheexperimentoftheSˇRITTPC.Thedetectorisoperatedattheroomtemperature(20C)and1atm.Theprocedureforhandlingthegasisasfollows.ThegasfromtheP-10cylinderisfedtothecageoftheTPCviathemasswcontroller.SincetheexitwindowoftheSˇRITTPCismadeofa75micron-thickKaptonsheet,thetialpressurebetween115Figure2.104:PositionofgasinputandoutputontheSˇRITTPC.thevolumeinsidethecageandtheoutsideshouldnotexceed1.2atm.Therefore,themaximumwratethatwecouldrunis1000cm3/minute.Beforethegasinputofthecage,thereisapressuregaugeformonitoringthepressureofaninputgas.InFigure2.104,thegaswsintothetopchannelsonthefrontwindowandthevolumefromthebottom.Theexcessgaswillexitthecageonthedownstreamside.Fromthegasoutputofthecage,weeitherconnecttothegasinputoftheenclosuresothatthewholevolumeofthedetectoriswithP-10gasorconnectatgaslinesuchasNitrogentothegasinputoftheenclosure.ThelatterallowstheTPCtohavetwontgases.TheschematicdiagramfromusingtwotypesofgasisshowninFigure2.106.Thelastsectionistoconnectthegasoutputoftheenclosuretothebubblertopreventthecontaminationofgas.Atthegasoutputoftheenclosure,therearehumidityandoxygenmonitorstochecktheconditionofthevolumeinside.116Figure2.105:PlannedgashandlingsystemfortheSˇRITTPC117Figure2.106:PlannedgashandlingsystemfortheSˇRITTPC118Chapter3Gatinggrid3.1PurposeofthegatinggridAgatinggridisusedtocontrolthepassageofelectronsandionsintheTPC.Inatypicalexperiment,thegatinggridiskeptclosedexceptwhenexternallytriggeredbyaninterestingevent.Whenthegatinggridisclosed,noelectronorpositiveionspassbetweenthedriftvolumeandtheregioncontainingtheanodeplanewheregasoccurs.Whenitisopened,thegatinggridwillbetransparenttodriftingelectronsallowingthemtopassessentiallywithoutlossfromthedriftregiontotheanodewires.Thegatinggridservestwofunctions:1.Itpreventsunwantedelectronsgoingintotheavalancheregionandthebackwofthepositiveionsfromtheavalancheregionintothedriftvolume[62].Preventingthesepositiveionsfromenteringthedriftregionwillhavetheofminimizingthespace-chargefromsuchpositiveions,whichcandistortthedriftinthedetectorandadversely[63,64].2.Anadditiongatinggridminimizesthedepositionofpolymersontheanodewiresandpreventsagingoftheanodewires[65,66].IntypicaloperationoftheSˇRITTPC,chargedparticlesproducedinheavy-ioncollisionsionizethegasinsideareactionchamberwithP-10gas(90%Argon,10%Methane).Ideally,theionizedelectronsdriftalongtheparallelelectricandmagnetictowardsamulti-layeredsetofwireplanesasillustratedinFigure3.1.AsdiscussedinSection1.4,thedrifttimeoftheseelectronstotheanodeplaneprovidesthevertical(y)locationof119theionizationtrackandthehorizontal(xandz)locationsoftheionizationatthepadplaneprovidesthehorizontallocationsoftheionizationtrack.Unfortunately,therecanbehorizontalcomponentsoftheelectricormagneticthataddtotheverticalcomponents,discussedinSection2.1.7.ThesecandistorttheCartesianreconstructionofthetracksfromtheelectronicsignalsmeasuredatthepadplaneoftheTPC.Suchelectricdistortioncanariseifthepositiveionsproducedneartheanodewiresareallowedtodriftalongthedominantlyverticalelectricandmagneticbackintothedriftvolume.Thepositiveionsdriftslowlyanditispossiblethatthebuild-upofthosespacechargesinthedriftregionsoftheTPCcanleadtohorizontalcomponentsoftheelectricthatcouldtlychangethedirectionoftheelectronicdriftvelocityawayfromverticalandleadtoratedependentproblemsinthereconstructionofthemomentaofthedetectedparticles[67].Thiscanbeavoidedifpositiveionsarecapturedonthegatinggridbeforetheycangotothedriftregion.Asecondproblemthatthegatinggridcanreduceistheagingoftheanodewiresduetodepositionofpolymersonthewires,causedbyimpuritiesinthecountergasorbynegativelychargedpolymersproducedfromthemethaneduringtheprimaryionizationorfromtheavalanche[15].Ionizationofsuchmoleculesinthegascanresultintheremovalofelectronsfromthemoleculesorinthebreakupofcovalentbondsofthegasmolecules.Table3.1showsthattheionizationenergyforsimpleelectronremovalfromamoleculeistypically2-5timeslargerthanacovalentbond.Whenacovalentbondbreaks,thereisapossibilitytoformachargedradicalmolecule.Ifsuchmoleculesareclosetotheanodeandarenegativelycharged,theycanbeattractedtowardstheanodesurfaceduetothehighelectricneartheanodewireasillustratedinFigure3.2.Thesemoleculescanbedepositedtoformanirregularpolymercoatingontheanodesurface.Blinovetal.[15]performedatestonthepolymerdepositonthesurfaceofananodewirewithvarioustypesofgasmixtures.The120Figure3.1:Ionizedelectronsdriftintothemultiplicationregionandmultiplyattheanodewires.studyshowsthatchargedpolymerstendtoaccumulateonthetipsoftheroughsurfaceoftheanodewireasshowninFigure3.3.Oncethelayerstickstothemetalsurface,polymerscancontinuetoaccumulate,preferentiallyonthehighestpoints,wheretheelectricisgreatest.Thepossibilityofbuildingupofthepolymersincrease;thegrowthatthehighpointslookslikethegrowthof"hair".AnexampleforpolymeraccumulationcanbeseeninFigure3.4.Thispolymercoatingincreasestheewirediameter,reducingthegasgain.Ifthesepolymersarenotcontrolled,theperformanceoftheTPCwilldeteriorateafterthewirechambershavebeenusedforsometime.121Table3.1:Dissociationandionizationenergyofgases[2]Dissociation(eV)Ionization(eV)Ar-15.8Xe-12.1H24.515.6N29.715.5O25.112.5Ethanol3.210.5Iso-propanol3.210.2DME3.29.98C6H63.611.5H2Ovapor4.812.6Methylal3.210.0CO27.813.8Iso-buthane3.210.6CH44.312.6Figure3.2:Formationofpolymersfromfreeradicalmolecules[2]122Figure3.3:Accumulationofnegativelychargedpolymersonthesurfaceofananodewire[15]Figure3.4:PolymersdepositonthesurfaceofananodewireinCO2/Isobuthane[15]123Apossiblesolutiontotheseproblemsistouseagatinggrid.Thiswillminimizetheaccumulationofthechargedpolymersandalsopreventthepositiveionsfromdriftingintothedriftvolumeanddistortingtheld.3.2Simulationofgatinggrid3.2.1SinglewireplaneandaconductingplaneTostudyabehaviorofagatinggrid,weneglectanyofthemagneticandconsiderhowtocontrolthepassageofchargesthroughthegatinggridwiththeuseofelectricalone.Westartwithagridofwiresparalleltoaconductingplane.InFigure3.5,thexzplaneiscoincidentwithaconductingplane.Thewiresisorientedalongtheyaxisandthezaxisisperpendiculartotheplane.Neighboringwiresinthegatinggridareseparatedfromeachotherbythepitchs.Theelectrostaticpotentialmainlydependsonyandzcoordinatesduetothesymmetryalongthezdirection.Usingcomplexvariabletechniquesinwhichthebothelectrostaticpotentialandtheelectricareanalyticfunctionsinthecomplexplane.ThestandardcoordinatesystemforTPChasthezaxisalongthebeamaxis,thexaxisishorizontalthatyaxisisvertical.Thiswouldmakethe(y;z)planetobetheplaneinwhichcomplexvariabletechniqueswouldbeapplied.Sinceyandzarenottheusualvariablesforcomplexvariables,itisconvenientforthecalculatingthepropertiesofthegatinggridtomakeachangeofcoordinatesystemasfollows.Forthefollowingdiscussion,weretaintheydirectiontobevertical,butmakethexdirectiontoliealongthebeamaxis.Thezdirectioninthisnewcoordinatesystemishorizontalperpendiculartothebeamaxis.Inthisnewcoordinatesystem,thepotentialdependsonlyonxandy;itdoesnotdependonz.Theconducting124planehaszeropotential(y=0).Inthesecomplexvariabletechniques,thepotentialinthechargefreeregionisananalyticfunctionofthecomplexpositionx+iycorrespondingtotheCartesiancoordinate(x;y).TheelectricatcomplexpositionUforalinesourceatcomplexpositionU0runninginthezdirectionperpendiculartotheplaneisE=2ˇ0(UU0).Duethepresenceoftheconductingsurfacerepresentedbythepadplane,forfullelectricincludesthefromthesurfacechargeonthegroundplane,whichcanberepresentedbyanimagechangeofatpositionU[11].IfonefurthersolvesforthecomplexpotentialofasinglewirewithlinechargedensityofatU0=x0+iy0,oneobtains˚(U)=2ˇ0ln(UU0)(UU0)(3.1)whereU=x+iyisthecoordinateofageneralpointandUisthecomplexconjugateofU0.Byaddingthecontributionofeachwire,thetotalpotentialcanbewrittenas˚(U)=2ˇ0k=+1Xk=ln(UU0k)(UU0k):(3.2)Figure3.5:Gridofwiresparalleltoaconductingplane[11]Thecorrespondingrealpotential[11],V(x;y)=Re˚(U),is125V(x;y)=4ˇ0lnsin2[(ˇ=s)(xx0)]+sinh2[(ˇ=s)(yy0)]sin2[(ˇ=s)(xx0)]+sinh2[(ˇ=s)(y+y0)](3.3)wherex0andy0correspondtothecoordinatesofthemidpointbetweentwowiresatthecenterofthewireplane.TheelectriccalculatedfromtheEquation3.3isgivenby:Ex(x;y)=201A11A2sin2ˇs(xx0);Ey(x;y)=20sinh[(2ˇ=s)(yy0)]A1sinh[(2ˇ=s)(y+y0)]A2;whereA1=cosh2ˇs(yy0)cos2ˇs(xx0);A2=cosh2ˇs(y+y0)cos2ˇs(xx0):(3.4)Thepotentialofawiregridatadistancethatismuchlargerthanthepitchcanbeapproximatedbythefollowingequations.V(x;y)=y0sfory<y0;y0y˛s2ˇV(x;y)=y00sfory0<y;yy0˛s2ˇ(3.5)Thus,thepotentialcreatedbyoneplaneofwireinEquation3.3atadistanced˛s=2ˇbehaveslikeasheetofchargewithasurfacechargedensity˙=.Nearthesurfaceofthewire,thepotentialisevaluatedat(xx0)2+(yy0)2=r2.Then,thepotentialofthewiregridcanbeapproximatedbyV(wiregrid)=00s1s2ˇy0ln2ˇrs:(3.6)1263.2.2SuperpositionoftheelectricFigure3.6:WireplanetionoftheSˇRITTPC.Thepitchofanodeplane(s1)is4mm.Thepitchoftheground(s2)andgatinggrid(s3)are1mm.Thedistancefromtheconductingplane(padplane)totheanode(y1),ground(y2)andgatinggrid(y3)is4,8and14mm,respectively.Thedistancefromthecathodetothepadplaneyp=509.5mm.InSˇRITTPC,therearethreeplanesofwiresasshowninFigure3.6.y1,y2andy3representthedistancefromtheconductingplane(padplane)totheanode,groundandgatinggridplanes,respectively.Thegroundandgatinggridplaneshavethepitchof1mmandtheanodeplanehasthepitchof4mm.TocalculatethepropertiesoftheTPC,wesuperimposethecontributionfromallwireplanesbyusingEquation3.5and3.6.Then,127thepotentialofthewireplanecanbeexpressedbyVa=a0s1y1s12ˇln2ˇras1+zy10s2+gy10s3+˙py10Vz=ay10s1+z0s2y2s22ˇln2ˇrzs2+gy20s3+˙py20Vg=ay10s1+zy20s2+g0s3y3s32ˇln2ˇrgs3+˙py30Vp=ay10s1+zy20s2+gy30s3+˙pyp0:(3.7)Va,Vz,VgandVparethepotentialoftheanode,ground,gatinggridandcathode,respec-tively.aisthechargeperunitlengthoftheanodewire.zisthechargeperunitlengthofthegroundwireandgarethechargeperunitlengthofthegatinggridwire.raistheradiusoftheanodewire.rzistheradiusofthegroundwire.rgistheradiusofthegatinggridwire.˙pisthesurfacechargedensityofthecathode.TosimplifytheEquation3.7,weasurfacechargedensity˙foreachwireplaneas.Then,wecanrewritetheseequationsas0BBBBBBBBB@VaVzVgVp1CCCCCCCCCA=A0BBBBBBBBB@˙a˙z˙g˙p1CCCCCCCCCA(3.8)whereAisthematrixofthepotentialcots.TheinverseofAisthecapacitancematrix.128Onceitisknown,wecanobtaintheinducedchargeoneachelectrodeasinEquation3.9.0BBBBBBBBB@˙a˙z˙g˙p1CCCCCCCCCA=A10BBBBBBBBB@VaVzVgVp1CCCCCCCCCA(3.9)AccordingtoEquation3.7,AcanbewrittenasA=100BBBBBBBBB@y1s12ˇln2ˇras1y1y1y1y1y2s22ˇln2ˇrzs2y2y2y1y2y3s32ˇln2ˇrgs3y3y1y2y3yp1CCCCCCCCCA(3.10)Fromthismatrixand,moreaccuratelyfromitsinverse,wecandeterminethetransmissionofelectronsandionsthroughvariousgrids.3.2.3MonopolargatinggridWithintheactivevolumeoftheSˇRITTPCthereisaconstantelectricdirecteddownwardstowardsthecathode.Theelectronsionizedinthegasdriftupwardsalongelectricandmagnetictowardsthegatinggrid.Dependingonthevoltagesonthegridwires,theseelectronswilleitherstoponthewiresorpassthroughthegridtowardsthegroundandanodewireplanes.Thoseelectronsfollowingelectriclinesthatterminateonagatinggridelectrodewillstoponthiselectrode.Thosethatdonotwillpassthroughthegatinggrid.FromGauss'slaw,thesurfacechargeonaconductorisgivenbytheelectriceldnormaltoitssurface.129Thus,theratioofthesurfacechargedensityonthegatinggridtothesurfacechargedensityonthecathodeisequaltothefractionofelectriclinesthatterminateonthegatinggrid,andconsequentlytothefractionoftheelectronsproducedbyionizationwithintheactivevolumethatarecapturedonthegatinggrid.Thus,fromsection3.2.2,iftheionizedelectronsfollowtheelectriclines,thetransparencyofthegatinggridTcanbeobtainedbyT=1˙+gj˙pj(3.11)where˙pisthenegativesurfacechargedensityofthenegativelybiasedcathodeand˙+gisthepositivechargedensityofthegatinggrid.Figure3.7illustrateshowawireresidinginanexternalelectricEwillbepolarizedbytheAsurfacechargedensity˙Dwillbepolarizedbytheexternalelectricnormaltothesurfaceofthewire.Thedependenceoftheresultingchargedensityonthewireasafunctionofanglecanbeexpressedas˙D=2E0cos;(3.12)whereistheanglebetweentheelectricEandtheradiusvectorfromthecentertothesurfaceofthewire.Ingeneral,thewirecanalsohaveatotalnonzerolinearchargeinadditionto˙D.Therefore,thetotalsurfacechargedensityis˙w=2ˇr+2E0cos;(3.13)whereristheradiusofthewireasshowninFigure3.7.Thetotalchargeperunitlengthcanbothnetpositiveandnetnegativecontributiondistributednon-uniformlyoverthesurface130ofthewire.Thepositivechargedensityonthegatinggridis˙+g=+s3(3.14)Figure3.7:AwireispolarizedbyanexternalelectricElectricarepointingupward[11].InordertocalculatethetransparencyofthegatinggridT,weonlycountthepositivechargesuponwhichtheelectriclinestransportingchargeterminate.Theselinechargescanbeobtainedbyintegratingthesurfacechargeovertheregionwherethechargeispositive.Thesecanbeobtainedfromthefollowingequations.Whenthechargedensityremainsnegativeoveralltheentiresurfaceofthewire,wehave:+=0when2ˇr<2E0:(3.15)Whenthechargedensityispositiveovertheentiresurfaceofthewire,wehave:+=when2ˇr>2E0:(3.16)131Whenthechargedensitypositiveatsmall0,andbecomesnegativeatlarger0,theintegralovertheangleswerethesurfacecharge˙>0yields:+=0ˇ+4Er0sin0when2E0<2ˇr<2E0;0=arccos4ˇ0Er:(3.17)(a)Gatinggridopened(b)GatinggridclosedFigure3.8:Driftlinesofionizedelectrons(y1=4mm,y2=8mm,y3=14mm,s1=4mm,s2=s3=1mm)Figure3.8illustratestheoperationofthegatinggridintheabsenceofamagneticThewireplanesfromthetoptothebottomarethegatinggrid,groundandanodeplanes,respectively.Theionizedelectronsaredriftingdownwards.Whenthegatinggridisopened,asshowninFigure3.8(a),allionizedelectronpassthroughtheanodeplane.Noelectronspassthegridwhenitisclosed,asillustratedinFigure3.8(b).FortheclosestateofthegatinggridinFigure3.8,thewiresarentlypositivelybiasedsothatthereisnetpositivechargedensityontheentiresurfaceofthewire.ThesecalculationswereobtainedfromGARFIELD,agasdetectorsimulationprogram,132whichcalculatesthemotionofelectronsandionssubjecttotheelectricandmagneticandtothescatteringwithgasatomsandmolecules[51,68].Accordingtothecalculation,thevoltageofthegatinggridneedstobenegativeenoughdependingonthecathodepotentialtoachieveafulltransparency,whichoccurswhentherearenoregionsofnetpositivechargedensityonthegatinggridwires.Figure3.9showsthecomparisonbetweenthecalculationofthetransparencyofthegatinggridfromGARFIELDandtheanalyticalsolution,givenbyEquation3.11.ThetransparencyfromGARFIELDcalculationagreeswiththeresultfromtheanalyticalsolution.Figure3.9:Transparency,T,ofthegatinggridwithVcathode=6kV.Blacktriangle:simulation;Redline:AnalyticalsolutionusingEquation3.11.AsshowninFigure3.9,itispossibletoregulatethechargetransmittedthroughthegatinggridbyvaryingthegridvoltage.This"mono-polar"gatinggridschemecanbeemployedtocontrolelectronsandionspassingthroughthegatinggrid.Thedeviceoperatedinthismode133canbefoundinReference[63].However,asdiscussedinSection3.2.4,abipolargatinggridismoresuitablefortheSˇRITTPCbecauseitallowsthegridtoclosewithminimumnoiseintroducedtotheTPCelectronics.3.2.4BipolargatinggridSection3.2.3iscalledmonopolargatinggrid,becauseallwiresonthegatinggridarebiasedtoacommonpotential.However,fewgatinggridsoperateinthatmodeduetothespeedrequired.FortheSˇRITTPC,thetransitionofthegatinggridfromclosetoopenstatehastooccurwithinatimeintervalT1.ThecurrentsrequiredtochangethevoltagesonthegatinggridwithinthisTintervalcancausealargedisturbanceontheanodewires.Theinducedsignalfromapoorlydesignedgatinggridcanbeenormouslylargerthanthedatasignal.Havingthegroundgridbetweenthegatinggridandtheanodeplanepartiallyshieldstheanodeandpadplanefromthegatinggridtransitionnoise.However,thisshieldingisttoallowtheuseofamonopolargatinggridfortheSˇRITTPC.AnimprovedesignforthegatinggridinvolvesclosingthegatinggridbybiasingadjacentwiresonthegatinggridtotwotpotentialsVaverage+VgandVaverageVgontheadjacentwires.Astheaveragevoltageofthegatinggriddoesnotchange,thenetchargeonthegatinggriddoesnotchangeforthismethod.Intheclosedthegridhaspositivesurfacechargedensityeveryotherwireandnegativesurfacechargedensityonthewireinbetween.Ifonesetsthevoltagebetweenthewires,theelectrondriftlineswillterminateatthepositivewiresclosingthegate.ThetransparencyofthebipolargatinggridcanbeobtainedfromEquation3.11.Thepositivechargedensityinducedonthewireisexpressedby134˙+=0s3ˇlnˇrg2s3:(3.18)Thetotalchargevariationonthegatinggridis0.Therefore,thenegativechargedensityis˙=˙+.Onthepositivewire,thetotalsurfacechargedensityisgivenby˙+bipolar=˙++˙g2:(3.19)The˙g2meansthatonlyhalfplaneofthegatinggridcontributesthepositivecharge.ThetransparencyofthebipolargatinggridcanbeobtainedfromT=1˙++˙g=2j˙pj:(3.20)(a)Gatinggridopened(b)GatinggridclosedFigure3.10:Operationofbipolargatinggrid(z1=4mm,z2=8mm,z3=14mm,s1=4mm,s2=s3=1mm).(b)Electronsterminateatthepositivewiresofthegatinggrid.InFigure3.10(b),theclosestateofthebipolargatinggridistfromthe135rationinFigure3.8(b).InFigure3.8,whenthegridisclosed,alldriftingelectronsterminateonthewiresthatareabovethetrack.However,intheclosedstateofbipolargatinggrid,driftingelectronsterminateatthenearestadjacentpositivelybiasedwire,thuspreventingthetransportofelectronsthoughthegatinggrid.Likewiseanypositiveionproducedattheanodewillbedriventowardsthenegativelybiasedgatinggridwires,stoppingthere.Next,wewillconsidertheofhavingamagneticthatisparalleltothemainelectricasisthecaseintheSˇRITTPC.ThebehaviorofdriftingelectronsandionsisstronglybythemagneticTheofmagneticischaracterizedbytheparameter!˝where!isacyclotronfrequencyoftheelectronand˝ismeantimebetweencollisionsoftheelectronswiththegasmolecules.Itgovernsthedegreetowhichthemagneticcanthetrajectorybetweencollisions.Whenitissmall,themotionthermalizesbeforetheofthetrackcurvaturebecomeimportant.Thus,smallvaluesof!˝leadtosmallmoofthetrajectoriesbythemagneticThishastheconsequencethattheiontrajectoriesaresimilartowhattheywerewithoutmagneticWhen!˝islarge,chargedparticlescantravelalongwaysalongthecurvedtrajectorydictatedbythemagneticThemeanfreepathforionisgenerallysmallerthanthatforelectronsas!˝isverysmallforionsandcanbeverylargeforelectronsinsomegases.Largercorrectionsoccurfortheelectronsinthepartsoftheirtrajectorieswheretheelectricactstobendtheelectrontracktowardsthewire,whichisinadirectionperpendiculartomagneticInthispartofthetrajectory,theelectronictrackscanbebentbythemagneticinthedirectionparalleltothewire,diminishingthenumberofelectronscaptureonthewire.IfoneincreasestheVgtoclosethegate,thevoltagerequiredtoclosethegateforionsislargelyunchanged,buttheelectronwillrequirealargevoltagetoclosethegate[11].Figure3.11(a)showstheelectrontransparencyofagatinggridwiththe136presenceofmagneticInthesimulation,theelectricld~Eisparalleltothemagnetic~B.ThevoltagerequiredtoclosethegateincreasesroughlylinearlywiththemagneticasseeninFigure3.11(b).Amendoliaetal.studytheofthemagneticontheperformanceofthegatinggridandalsoseethatthevoltagesofthegatinggridareincreasinglinearlywiththemagneticstrength[69].IntheexperimentfortheSˇRITTPC,thedetectorwillbeoperatedunderthemagneticof0.5T.Tocompletelyclosethegatinggrid,weneedtoapplyVg50V.TheplannedVgwillbe75Vtoassurethatthegatinggridiscompletelyclosed.(a)Electrontransparencyofagatinggrid(b)ClosingvoltageasafunctionofthemagneticFigure3.11:(a)Transparencyofbipolargatingwiththepresenceofmagnetic(b)Theclosingvoltageofagatinggridincreaseslinearlywiththemagnetic3.3GatinggriddriverGatinggriddrivercontrolsthetransitionbetweentheclosedandopenstatesofthegatinggrid.InSˇRITTPC,theoperationofthegatinggridhasbeendesignedforabipolargridinwhichthegatinggridwillhavetosamepotentialfortheopenstateandisbiasedupordownbyVginadjacentwiresfortheclosedstate.1373.3.1DesigncriteriaInatypicalexperiment,thegatinggridiskeptclosedmostofthetimeunlessitisexternallytriggeredbyaninterestingevent.Thereisa"dead"regionbelowthegatinggridattheedgeoftheactivevolume,inwhichionizedelectronswilldriftintothegridbetweenthetimethattheeventtriggerisandgateisfullyopened.Thesizeofthedeadregionisgovernedbytheelectrondrifttimeandthetimeneededtoopenthegate.TheformerisdeterminedbythegaspressureandelectricItispossibletochangetheelectrondrifttimebychangingthecountergas,butdoingthatcanchangetheoperatingparametersoftheTPCantly.IfonedesirestouseP-10gas,whichconsistsof90%argonand10%methaneandhasadriftvelocityofabout5.5cm/s,then,thegatinggridneedstoopenasfastaspossible(500ns)toreducethedeadregion.AsdiscussedinSection3.2,largeamountofchargesbroughtontothegatinggridwithinashorttime(200-300ns)canpotentiallycausealargeinducedsignalonthepadswhichcanbeconsiderablylargerthanthechargeionizedbyaweaklyionizingparticlesuchasanenergeticpion.Toavoidthisproblem,thegatinggridhastodrainthechargeatequalrateonbothpositiveandnegativesidessothattheaveragepotentialofthegatinggridremainsconstant.3.3.2ConceptualdesignFigure3.12showsaconceptualdesignofthegatinggriddriverfortheSˇRITTPC.Thegatinggridisconnectedtothegatinggriddriverviatwolowimpedancetransmissionlines(4whichsupplyvoltagetoalternatingwires.IntheSˇRITTPC,thetransmissionlineshasbeendesignedandcustommadetominimizethenoisefromthetransitionofthe138gatinggridfromopentoclosestateandviceversa.ThevalueoftheimpedanceofthetransmissioncableandthusitssizeisdictatedbytheavailablespaceintheTPC.Inthetransmissionlines,thecurrentsfromthepositiveandnegativepolaritiesofthegatinggridwinoppositedirectionsalongadjacentconductorsinthesetransmissionlines.Thenoisefromthetransitioncanbereducedastheinducedsignalsfromthecurrentswhichhavetheoppositepolaritiesarelargelycanceled.Therearetwopowersuppliesconnectedtothecircuit.Oneofthepowersourcesgiveshigherpotential+HVwhiletheothergivesthelowerpotential-HV.Theyareconnectedthroughtwotypesofmosfetswitches(NandPtype).NandP-mosfetswitchedareconnectedtothe+HVand-HV,respectively.Theadvantageofusingamosfetswitchisthatwhentheswitchisclosed,theinternalresistanceislowercomparingtoothertypesofswitches.Thishelpsinallowingonetotunetheresistanceacrosstheswitchestoalowertotalresistancevalue.Also,amosfetswitchhasashortturn-ondelaytime(<100ns).Itallowsthegatinggridtomakeatransitionquickly.Fortheclosedstate,bothmosfetswitchesareopened.Thealternatingwiresofthegatinggridarebiasedby+HVand-HV.Whenthegatinggridisopened,thetwomosfetswitchesareclosed.Thetwopowersuppliesareconnectedtogetherbythecombinedresistanceofthetwoswitches.Thecurrentwsbetweenthetwosidesofthegatinggriduntilbothsidesofthegridreachtheaveragevoltage.Thegatinggridiskeptopenedlongenoughtoletallelectronsfromtheinterestingeventtopassthroughthegridandthenthetwosidesofthegatingarerestoredtotheirorignalvalues,closingthegate.Thekeychallengeistoopenthegatequicklywithouttheaveragevoltageofthetwosideofthegatinggriddeviatingtlyfromtheaveragevalue.139Figure3.12:conceptualdesignofthegatinggriddriverforSˇRITTPC3.3.3Gatinggriddriverprototype1Intheprototype,twoofBEHLKEmosfetswitches(HTS21-14)areusedtocontrolthegatinggridoperation.Figure3.13showsthecircuitdiagramofthegatinggriddriverprototype1.TheBEHLKEswitchesneedaDCvoltageof5VtostabilizethemanditiscontrolledbytheexternalTTLsignal.TheswitchesremainopeneduntilitistriggeredbytheTTLsignal.TheclosingtimeintervalisequaltothewidthoftheTTLsignal.Therefore,weareabletodecidehowlongweneedtoopenthegatinggrid.R1andR2inthecircuitdiagramareusedtoadjusttheimpedancebetweenthegatinggriddriverandthegatinggrid.Thisprototypewasassembledonthebreadboardandhadaresistanceof6Itwasconnectedtothegatinggridvialowimpedancecables.Thecapacitanceofthegatinggridwasmeasuredtobe26.5nF.Avoltageof75Vwasappliedforthetest.Thetimefordecaystartsatthetimewhenthegatestartstoopen.InFigure3.14,weshowthegatinggridsignals.Thegreensignalshowsthepositive140Figure3.13:Circuitdiagramofthegatinggriddriverprototype1Figure3.14:Dischargingsignalofthegatinggriddriverprototype1onthebreadboard.Therearesomeslowoscillationonthesignal.141voltageonthegatinggrid.Thepinksignalshowsthenegativevoltageonthegatinggrid.Thescopesensitivityofthepinkandgreentracesissetto2V/division,sothe75Vinitialvoltagesonthegatinggridarescale.TheyellowlogicsignalshowstheTTLtriggersignalwhichtriggerstheclosingofthetwoBEHLKEswitches.Theseswitchesbegintoclose150nsaftertheleadingedgeofthislogicsignal.Theleadingedgeofthelogicsignalisabout600nsbeforethetwogatingvoltagesignalscomewithin0.4Voftheminimumvoltage.Thisrapiddischargeofthegatinggridoccurswithinabout600-150=450nsoftheleadingedgeofTTLtriggersignal.However,thereareslowoscillationsthatcontinueonaftertheopeningofthegatinggriditself.FromFigure3.11(a),thetransparencyofthegatinggridwillbeabout80%bythetimethegatinggridvoltageshavedecreasedby1/eoftheoriginalvoltageof75V,sothegatewillbemostlyopenabout300nsafterthegatinggridswitchesbegintoclose.Thisslowoscillationindicatesthatthecircuitisunderdamped.Onecanalsoseethattheoscillationonboththemorepositiveside(greentrace)andthemorenegativeside(pinktrace)ofthegatinggridareinphase.Thismeansthatthereisacommonmodeoscillation,whichmustbesuppressed.Ifthepositiveexcursionofthevoltageoscillationislarge,itmaythereadoutdatabymakingnoisesignalthatiscomparabletothepionsignals,makingittosetthethresholdsinthereadoutelectronicstlylowtotriggeronthepionparticles.Oneperiodoftheslowoscillationcomesfromthecapacitanceandinductanceofthegatinggridandgatinggriddrivercircuit.Itisimportanttotrytominimizethisinductance,inparticularthat,whichcanbeattributedtolongwiresandtracesandtheutilizationofcomponentswithlargeinductances.Therefore,aprintedcircuitboard(PCB)wasdesignedtominimizetheinductanceandimprovegroundingoftheboard.However,itshouldbenotedthattheinductanceofthegatinggriditselfisnolongerinthecircuitforthistest.142InFigure3.15,thePCBwastestedwith11.6nFstandardcapacitorandanoperatingvoltageof30V.Theresistanceofthecircuitis4.8Inthiscase,ittakes400nsbetweentheleadingedgeoftheTTLtriggertodischargetotheaveragevoltageforthegatinggrid.Consideringthat120nsofthistimeoccursbetweenthearrivalofthegatinggridtriggerandtheclosingofthegatinggridswitches,theactualdischargetakesabout280nsandthelowfrequencyoscillationistlydecreased.Itshouldbenotedthattheinductanceofthegatinggridisnotinthecircuitduringthistest.Inaddition,thereisanegativelobeonthesignalafterthedischarge.Thisindicatesthatthecircuitisstilltlyunderdamped.Toinvestigatetheslowoscillation,weuseasimplemodelofRLCseriescircuittoestimatetheeinductanceofthecircuit.Toachieveacriticallydampedconditiontogetridofthenegativelobe,R,LandCarerelatedbyR=2qLC.Figure3.15:Testingtheprototype1PCBwithastandardcapacitorof11.6nFandanoperatingvoltageof30V.Bluesignalisadischargingsignalfromthepositivesideofthecapacitor.Thereisanegativelobeafterthedischargewhichindicatesthatthecircuitisunderdamped.WeusetheSPICEcircuit-analysisprogramtosimulatethecircuit.SPICEwasoriginally143developedattheUniversityofCalifornia,BerkeleybyLauranceNagel.TheprogramprovidesDC,ACandtransientanalysisandisusedtochecktheintegrityofthecircuitdesignsandtopredictcircuitperformance[70].Forthesimulationsdoneinthiswork,weusetheversionOrCADEEPSPICEdownloadedfrom[71].Inthesimulation,aresistor,acapacitorandaninductorwereconnectedinseries.Theresistorrepresentstheresistanceofthegatinggriddriver.ThecapacitorrepresentsthecapacitanceoftheSˇRITTPC.NotethattheactualTPChasmuchmorecomplicatedstructure.Theinductorrepresentstheinductanceofthesystem.Bymatchingthesimulationtothedatawiththeactualgatinggrid,wedeterminetheeinductanceofthePCBtobearound160nH.ThecomparisonbetweendataandcalculationisshowninFigure3.16.Therefore,witharesistanceof4.8ohmsandacapacitanceof27nF,oneshouldachieveacritically-dampedcondition.Afteradjustingthecapacitancevalueinabenchtestingtomatchthiscondition,thenegativepeakdisappearsandthedischargetimeincreasesto600ns.Thecapacitanceofthegatinggridis26.5nFwhichisclosetothecritically-dampedcondition.TheresultoftestingthePCBwiththeactualgatinggridwithanoperatingvoltageof30VisshowninFigure3.17.Asexpected,thereisnonegativepeakandittakesabout600nstodischarge,however,thevoltagedeclinesby1/3inabout160ns,atwhichtimethegateisnearly80%open.InthetestwiththebreadboardandthePCB,thereisa120nsdelaytimeinopeningthegate.ThegatinggriddriverhasbeenconnectedtotheSˇRITTPCandtestedwiththevoltagesof75V.TheinducedsignalonthepadsarereadoutbytheprototypeoftheGeneralElectronicsfortheTPC(GET).Figure3.19showstheinducedsignalfromthepadswhichisbyGETelectronicsasafunctionoftimebucket.Sincethedatawastakenbyusingapeakingtimeof232nsandwritingtimeof50MHz,onetimebucketcorresponds144Figure3.16:MatchingthedatatothesimulationfromPSPICEFigure3.17:Dischargingsignalsofthegatinggridusingthegatinggriddriverprototype1to20ns.Theinducedsignaloccurringbetweentimebucket125thto200thcomesfromthedischargingofthegatinggridwhenitisopenedwhiletheothernegativepeakindicateswhenthegateisclosed.145Figure3.18:Dischargingsignalfromtheoscilloscope(operatingvoltageof75V,resistanceofthesystemof4.8Figure3.19:ReadoutsignalfromtheprototypeofGETElectronics3.3.4Gatinggriddriverprototype2Theprototype1gatinggriddriverhastheturn-ondelaytimeof120nswhichisratherlong.Ideally,thegatinggridwouldopeninstantaneouslyaftertriggered.Wethereforetriedtoreducetheturn-ondelaytimeasmuchaspossible.Intheprototype,theBEHLKEswitcheswerereplacedbyN-mosfet(IRF640)andP-mosfet(IRF9640)switcheswhichhavethesameturn-ondelaytimeof14nsasshowninFigure3.20.Todrainchargesfromthegatinggrid146asfastaspossible,weneedtodrivethemosfetswitchesatthesaturatedregionwheretheinternalresistanceoftheswitchesissmallest,0.18forIRF640and0.5forIRF9640.ToachievethesaturatedregionoftheMOSFETswitchesthatweuse,theamplitudeoftheTTLsignalsshouldbegreaterorequalto10VforN-MOSFETand-10VforP-MOSFETswitches.Tosafelyoperatetheseswitches,theamplitudeoftheTTLsignalcannotexceed18V.Thesemosfetswitchesaredrivenbythegatedriver(MCP14E11)whichcansupplythegatesignaltocontroltheoperationofNandP-switches.Figure3.20:Circuitdiagramofthegatinggriddriverprototype2Theprototype2hasbeentestedwithastandardcapacitorof22nFandanoperatingvoltageof12V.Onesideofthecapacitorwasconnectedtothepositivesideandtheothersidewasconnectedtothenegativesideofthegatinggriddriverlabeled"GatingGrid(+)"and"GatingGrid(-)"inthecircuitdiagram.ThevalueofR1andR2are5each.InFigure3.21,thedischargingsignalfromthepositivesidewasinblueandthesignalfromthenegativesidewasinpurplewhichisinverted.Atthebeginningofdischarging(inredcircle),147thereisahighfrequencynoiseanditdisappearswhenthecapacitorisfullydischarged.Inaddition,thedischargerateofthesignalsfrombothsidesaret.Thiscomesfromthebetweentwoswitcheswhicharerelatedtotheinternalresistanceandcapacitanceoftheswitch.Inthiscase,thedischargingonthenegativesideisslowerthanthatofthepositiveside.Figure3.21:Signalsfromtheoscilloscopeshowsthedischargingfromthepositiveside(blue)andnegativeside(purple)ofthestandardcapacitorof22nF.Thetwosignalsdischargeatatrateandhaveaatthebeginningasindicatedinredcircle.Toadjustthepurplesignaltomatchtheblueone,a10resistorwasaddedinparalleltoR1(5sotheresistanceonthenegativesideis3.33ThisbroughttwosignalstothesamedischargingrateasshowninFigure3.22.Thegatinggridwasopenedat100nsaftertriggered,slightlyfasterthantheprototype.TheswitchesfullyclosewhenthevoltagebetweenthegateandsourceofthemosfetreachesVGS=10V.Figure3.23showsthat,thisgatedriverchip(MCP14E11)requires100nsforthegatesignaltogofrom0to12Vwhichistoolong.Tooptimizethis,weneedcircuitcomponentsthatareabletogeneratea10Vsignalfaster.Inaddition,theclosingtimefortheprototype1and2isaround50s.Accordingtothecircuitdiagram,whentheswitchesareopened,thecapacitor148willbechargedviaa200resistoroneachside.ThislongclosingtimewilldecreasestheperformanceofTPCandsomeofunwantedionizedelectronscangothroughthegatinggridbecausethepotentialonthegridhasnotreachedthefullyclosestate.Figure3.22:Adjustthedischargingrateonthenegativesidetomatchthesignalonthepositiveside.Figure3.23:Green:theTTLtriggersignal;Purple:GatesignalfordrivingN-mosfetswitch;Blue:GatesignalfordrivingP-mosfetswitch1493.3.5Gatinggriddriverprototype3Todecreasetheclosingtimeofthegatinggridfromtheprototype1and2,thegatinggriddriverhasbeenmobyintroducinganotherpairofNandP-mosfetswitchestoclosethegatinggridfaster.Figure3.24:Circuitdiagramofthegatinggriddriverprototype3InFigure3.24,thegatinggriddriverconsistsof2N-mosfetand2P-mosfetswitches.TheP-mosfetistheIRF9640andtheN-mosfetistheIRF640.Thedriverisconnectedtothegatinggridviathetransmissionlinewhichislabeled"GatingGrid(+)"and"GatingGrid(-)."Thepairoftheswitches,N1andP1,areclosedtoshortthepositiveandnegativesidesofthegatinggridandopenthegate.Toopenthegate,onemustapplyapositiveTTLsignaltotheconnectorlabeledTTL-1.ThiswillcloseN1andP1switchesthatbridgethe150gatinggridandshortittogether.Thenthegatinggriddischarges.AttheendoftheTTL-1signal,thetwoswitchesopenandthetwosidesarenolongerconnectedtogether.Thentheychargebacktotheiroriginalvoltage.HowfastthegatinggridrechargedependsonwhetheronesuppliesaTTLsignalthroughtheTTL-2input.IfaTTLsignalissenttotheTTL-2rightaftertheendoftheTTL-1signal,thetwoswitches,N2andP2,areclosed.Theyconnectthe+HVand-HVsuppliestotheirrespectivesidesofthegatinggridthroughaswitchanda10resistor.IfonedoesnotsupplyaTTLsignaltotheTTL-2input,theseswitchesremainopenandthegatinggridswillrechargemoreslowlythrough1resistorsthatconnectacrossthepositiveandnegativepolesofeachswitch.A22nFcapacitorisconnectedtothegatinggriddriverandtestattheoperatingvoltageof12V.Theclosingtimeofthegatinggriddriveristhetimetorechargethecapacitortotheoriginalvoltage(12V).Ittakes2stochargethecapacitorto12V.Inthepreviousprototype,ittakesaround50storechargeit.Figure3.25:Thesignalfromtheoscilloscopeshowsthetimeofcharginga22nFcapacitortotheoriginalvoltageof12V.Blue:signalfromthepositivesideofthecapacitor;Purple:signalfromthenegativesideofthecapacitor.InFigure3.26,wetestedthetransitionoftheprototypebyconnectingittoastandard151capacitorof16.5nF.Theoperatingvoltagesare-40and-180V.WhenthegatinggriddriverwastriggeredbytheexternalTTL(purple),thepositiveandnegativesidesofthecapacitorwereshortedtogetherandhadthecommonvoltageof-110V.Thisprototypehastheturn-ondelayof100nswhichisverysimilartothepreviousones.Eventhoughweareabletodecreasetheclosingtimeofthegatinggridbyintroducinganotherpairofmosfetswitches(N2andP2),theopeningtimeisstilltoolong.ThisissueisthatthegatedriverchipwhichreceivestheTTLtriggersignalandgeneratesagatesignalof10Vfortheswitchesdoesnotrespondfastenoughforourpurposes.Toshortenthisopeningtime,thegatedriverchipneedstorespondquickerandtakelesstimetogeneratethe10-12Vgatesignal.Figure3.26:Thegatinggriddriverprototype3hasbeentestedwithastandardcapacitorof16.5nF.Theoperatingvoltagesare-40and-180V.Ithasaturn-ondelaytimeof100ns.Whenthegatinggridisopen,thepositive(yellow)andnegative(blue)sidesofthecapacitorareshortedthroughthemosfetswitchesandprovidetothecommonvoltageof-110V.3.3.6Gatinggriddriverprototype43.3.6.1BasicdesignToaddresstheslowrampoftheMC14E11triggerchipsinthepreviousprototype,animprovedcircuitwasdeveloped.InFigure3.27,thecircuitdiagramofthisnewprototypeis152shown.Itusesthecircuitdiagramoftheprototype3butthegatedriverchips,MC14E11werereplacedbyMIC4420/4429whichcangeneratethevoltagefrom0Vto12Vin20ns.Inaddition,theN-mosfetandP-mosfetswitchesaredrivenbytheMIC4420(ND)andMIC4429(PD),respectively.Thegatedriverchipischangedtoreducetheturn-ondelaytimeofthegatinggriddriver.Inaddition,toavoidanunwantedinducedsignalwhenthegatinggridisopened,thepositiveandnegativewiresofthegridneedstobedischargedatthesamerate.Inthepreviousprototypes,thetimesrequiredtoclosetheP-mosfetandN-mosfetswitcheswhichallowpassageofcurrentfrombothsidesaret.ThedischargingrateofthegatinggridcorrespondstotheRCtimeconstant˝whichisbyRp,Rn,CpandCninthecircuitdiagram.Thediodeswhichareconnectedinparalleltothe1kresistorsinthemiddleofthecircuitdiagramareusedtoholdthegatedrivingsignalstableforalongopeningtime(500s)ofthegatinggrid.InatypicalexperimentoftheSˇRITTPC,theopeningtimeofthegatinggridisaround10s.Therefore,thediodesarenotnecessaryfortheseriesofexperimentswiththeP10gas.Thecouldbeusefulforamuchslowergassuchashydrogenorhelium.3.3.6.2Testsandoptimizationofversion4gatinggriddriverInFigure3.28,thegatinggriddriverprototype4hasbeentestedwithallpowerandtriggersystemwithoutconnectingtoastandardcapacitororthegatinggrid.WhentheTTLsignal(yellow)issuppliedtotheTTL-1input,thetwoswitches,N1andP1areclosed.The+HV(-20V)and-HV(-160V)suppliesareconnectedthroughthoseswitches.Theswitches,N1andP1,wereclosedat45nsaftertheyweretriggered.Inthisprototype,wethereforereducedtheturn-ondelaytimeofthecircuitfrom120ns(prototype1)to50ns,however,wecanseethatthenegativeswitch(ingreen)opensabout2nsfasterthanthepositiveswitch153Figure3.27:Circuitdiagramofthegatinggriddriverprototype4(inblue).Afterclosingthereisanoscillationwithaperiodofabout3ns,whichisoutofphasebetweenthepositiveandnegativesides.Suchafastoscillationwouldbetoofastforthepadelectronicstoamplify.ThepadelectronicswouldonlyamplifyslowerFouriercomponentsofthesignalwithperiodsatleastanorderofmagnitudelonger.Figure3.29showsthePCBofthegatinggriddriverprototype4.ThereareplacesforadjustingcomponentswhichareRp,Rn,CpandCn.Theyareusedformatchingthedischargingrateofthetwosidesofthegatinggrid.Forthetesting,weconnectthegatinggriddrivertotheSˇRITTPCandrunwithoperatingvoltagesof-40and-180V.ThevaluesofCpandCnare100pF.TheresistanceoftheIRF9640inserieswithRpandIRF640inserieswithRnare1.6and1.7respectively.154Figure3.28:Thecircuitboardistestedwithoutconnectingtoastandardcapacitor.Thegatinggriddrivershortstwopowersuppliestogetherat45nsafteritistriggered.Blue:dischargingsignalfromthepositiveside.Green:dischargingsignalfromthenegativeside.Figure3.29:Printedcircuitboardofthegatinggriddriverprototype4.Thecoloredcirclesindicatetheconductivepadswhichcanputtheadjustingcomponents,Rp,Rn,CpandCnon.Rnismountedontheothersideoftheboardsoitisnotvisibleinthispicture.155ThetransitionofthegatinggridfromclosedtoopenstatecanbeseeninFigure3.30.Whenthegatinggridisclosed,thewiresarebiasedto-40and-180V,alternatively.OncetheTTLsignalissenttotheTTL-1input,thetwopowersuppliesareshortedtogetherandprovideacommonvoltageof-110V.AttheendoftheTTL-1,another2-sTTLsignalissenttotheTTL-2inputtoclosethegatinggridandittakes2.5storechargeeverywirebacktotheoriginalvoltages.InFigure3.31,fromthereadoutoftheGETElectronics,thereisaninducedsignaldetectedonthepads.Thissignaloccurduringthetransitionofthegatinggridfromclosedtoopenstate.Thesignalhasanegativepeakduringthe450nsofthetransitionandthengoestothepositiveside.Thenegativesignalhasthepeakheightof800ADCchannels.Thisinducedsignallast1.6s.ThemaximumofADCchannelsthattheAGETElectronicscanholdis4096channels.Thenegativeexcursionofthissignalisinthedirectionoftheexpectedsignalfromrealevents,whicharealsonegative.Wewouldliketokeepthesizeofthenegativepeakoftheorder1%ofthemaximumADCwhichcorrespondsto50ADCchannelsorless.Inthetypicalexperiment,thesignalfromaninterestingeventalsohasthenegativepolarity.Itmightbepossibletoaccommodatethepresenceofanegativesignalonthepadsofamagnitudegreaterthan50channelsifwedecidetoskippartialreadoutandrecordallofthedataonallofthepadsforeveryevent.Inthatcase,onecouldsimplyrecordthelineshapeofthegatinggridsignalandperformasoftwaresubtractiononthedata.Thisisnottheidealoption.TheGETelectronicscanbereadoutinapartialreadoutmode,whichsuppressesthereadoutofchannelsbelowapresetthreshold.Ifonecouldholdthenegativeexcursionsofthegatinggridpickupsignaltolessthan50channels,thethresholdcouldbesetaboveitandonlythepadswithrealdata(notgatinggridnoise)wouldbereadout,reducingthereadouttimeandthesizeofthedataThismotivatedextensiveandfocusedtominimizethenegativeexcursions.Which156suchaminimizationhasthebeneofminimizingtheamountofdatatobereadout,thepositiveexcursionsarealsoimportanttocontrol.Toeliminatetheofthegatinggridpickupnoiseonthedata,however,onewouldneverthelesswanttosubtractitfromtherecordeddata.Theaccuracyofthissubtractionisnaturallyimprovedifboththenegativeandthepositiveexcursionsofthegatinggridnoisesignalarereducedasmuchaspossible.Thisoptionofpartialreadoutmodemotivatedoptimizingthegatinggridnoisesignal.Tostudythedischargecharacteristicsofthemosfetswitchesforthecircuitoftheprototype4andguideouttoreducethegatinggridnoise,weuseaSPICEmodelofthemosfetswitchesprovidedbythemanufacturerinthesimulation.SPICE(SimulationProgramwithIntegratedCircuitEmphasis)isacomputerizedcircuitsimulationprogramthatcanbeobtainedfrom[71].N-mosfetmodelisavailableat[72]andP-mosfetmodelisavailableat[73]http://www.vishay.com/mosfets/list/product-91086.Inthesimulation,thegatinggridisreplacedbyacapacitorof26.5nFwhichapproximatesthecapacitanceoftheSˇRITTPCgatinggrid.Figure3.33showstheSPICEsimulationsofthegatinggridtransitioningfromtheclosedtoopenandthenbacktoclosestate.Atthebeginningofthesimulations,thegatinggridisclosed.Alternatewiresrepresentedbythetwosidesofthecapacitorarebiasedto-40and-180V.TheN-andP-mosfetswitchescloseat0.5s.Ittakes350nsforthegatinggridtoreachthecommonvoltageof-110V.Inthisparticularsimulations,thegatinggridcloses4safteropeningandittakes3storechargethewirebacktotheoriginalvoltages.WestartwiththedefaultvaluesofCp=Cn=100pFandRp=Rn=1Figure3.32illustratestheSPICEsimulationofanidealcircuit.Inreality,circuitboardsaremadeofcomponentswithimprecisespvalues.Resistorscanhavesmallinductancevalues,forexample.Inaddition,thegatinggridismorecomplicatedtheapproximationofasinglecapacitanceofabout26nF.ThisdemonstratedbythecomplexnoisesignalsshowninFigure157(a)Transitionofthegatinggridfromopeningtoclosingstate(b)Dischargingsignalswhenthegatinggridisopened.Figure3.30:Testthegatinggriddriverprototype4withoperationvoltagesof-40and-180V.thevaluesofCpandCnare100pF.(a)Whenthegatinggridisclosed,alternatingwireshavethevoltagesof-40and-180V.Onceitopens,thetwopowersuppliesareshortedtogetherandgivethecommonvoltageof-110V.(b)theopeningtimeofthegatinggridis250ns.158Figure3.31:Thereisainducedsignalonthepadwhenthegatinggridopens.3.35thatthespicemodeldoesnotreproduce.Theideaofincludingtheextracapacitorsandresistors,Cn,Cp,RnandRpassociatedwiththeN1andP1switchesallowsadjustmenttoaccommodaterealoperationoftheTPCgatinggrid.Typically,RnandRpcontrolthedischargetimewhileCnandCpareadjustedtobalancethepositiveandnegativecharge.BychangingthevaluesofCp,Cn,RpandRn,onecansimulatethesituationwhenthedischargefrombothpositiveandnegativesidesofthewiresarenotsymmetric.Figure3.33(a)showstheSPICEcalculationsofthesignalfromthegatinggridtransitioningfromtheclosedtoopenstatewithCp=600nF,Cn=100pFandRp=0.95Rn=1.05Herethegreenlinedepictsthevoltageonthepositivesideofthegatinggridandtheredlinedepictsthevoltageonthenegativesideofthegatinggrid.Inthisidealizedcircuit,theasymmetryofthedischargevoltageobtainedbyaddingthepositiveandnegativevoltages159Figure3.32:TransitionofthegatinggridfromclosedtoopenandbacktoclosedstateinSPICEsimulation.Cp=Cn=100pFandRp=0.95Rn=1.05orobtainthelightbluetrace.IntheSPICEcalculationsforthiscircuit,theasymmetryisverysmallcorrespondingtoanexcursionofX=-13Vfromthecommonvoltage(-110V)attheminimum.At0.25safteropeningthegate,thesumofthevoltagesisB=-2V(1%)fromthecommonvoltageof-110V.Toshowtheoftuningthegatinggriddriver,wevarythevaluesofCnandCpinTable3.2.TheplotsshowninFigure3.34arethedependenceofXandBasafunctionof(Cp-Cn).Asexpectedintheidealenvironmentofthissimulation,thebestiswhenbothCpandCnhavethedefaultvaluesof100pFandRp=0.95Rn=1.05AsshowninFigure3.33(b)withthebesttunedvalues,thedischargerateisnownearlysymmetricandboththeXandBvaluesaresmallat-1and-0.1VasshowninTable3.2.NotethatthesignalsfromtheSˇRITTPChavenegativepolaritysoitismoreimportanttominimizethenegativechargeinthetuning.160(a)(b)Figure3.33:(a)TransitionofthegatinggridfromclosedtoopenstateinSPICEsimulation,Cp=600nFandCn=100pFandRp=0.95Rn=1.05(b)Thesameas(a)withtcapacitorvaluesCp=100nFandCn=100pFandRp=0.95Rn=1.05161(a)X(V)(b)B(V)Figure3.34:ThedependenceofXandBasafunctionofCn-Cp.162Table3.2:TheoftuningthegatinggriddriverwithCnandCp.Rp=0.95Rn=1.05Cp(pF)Cn(pF)Cn-Cp(pF)X(V)from-110VB(V)1001000-10.2200100-100-3-0.37300100-200-6-0.93400100-300-9-1.48500100-400-12-2.2600100-500-13-2.581002001004.840.751003002007.61.3110040030010.651.8910050040012.782.4810060050014.473.1(a)CpislessthanCn(b)CpislargerthanCnFigure3.35:Testthegatinggriddriverwithastandardcapacitorof26nFandvaryCpandCntoseethefromtheSPICEsimulation.Thegreenlineindicatesthecommonvoltagelevel.3.3.6.3MeasurementsandoptiminizationfotheInducedsignalsonthepadsusingtheGETTPCreadoutelectronicsItisimportanttoassesstheinofthegatinggridopeningonthereadoutelectronics.Todoso,weattachedthegatinggriddrivertothegatinggridontheSˇRITTPCandthenweopenedthegatinggridandreadoutthesignalthatwasinducedonthepadsduringtimethatthegatinggridwasopening.163Thesetestswereperformedshortlyafterthereadoutelectronicsbegantofunctionprop-erly.Thus,theanalysiscapabilitieswerelimitedtheexplorationoftheresponseofspselectedpadsonthepad.Thoseresultsareshowninthefollowing.Table3.3:NegativepeakheightoftheinducedsignalfromthetransitionofthegatinggriddriverCp(pF)Cn(pF)Peakheight(ADC)100100800100220131410027017243302707513902704151000270264100033021110004401051000490182Figure3.36:TestthegatinggriddriverwiththeSˇRITTPCbyvaryingCpandCn.ThesizeofthenegativepeakisincreasingwithCn.Wetestedthegatinggridprototype4withtheSˇRITTPCbyvaryingCpandCn.Theoperatingvoltagesare-40Vand-180V.InFigure3.36,weshowthepulseinducedonthepad164Figure3.37:UsingCp=1000pFandCn=440pF,thenegativepeakisreducedfrom800to105ADCchannels.forthishighestAGETgainrangeof120fC.Forresultsshownbelow,weshowthedigitizedvaluefortheinducedpulse.Inthisrepresentation,thesignalinducedbymultiplicationofelectronsneartheanodewireswouldbenegative.TheamplitudeofthenegativepulseincreaseswithsignalandwouldbeproportionaltothedE=dxoftheionthatproducedthispulsebyionizingthegassomewhereabovethepad.Thegatinggridproducesanegativepulsethatcouldbeconfusedwithanionizationevent.ThesizeofthenegativepeakofthisgatinggridnoisepulseisincreasingwithCn.MoredetailedresultcanbeseeninTable3.3.ThepeakheightofthenegativesignaldecreaseswhenCpislargerthanCn.ByadjustingthecapacitanceofCpandCn,thenegativepeakat800ADCchannelswasreducedto105ADCchannelsforCp=1000pFandCn=440pF.InFigure3.37,weshowGETdatafromthepadsforthischoiceofCp=1000pFandCn=440pF.Asdiscussedpreviously,thenegativepeakhasbeenreducedto105ADCchan-nels.Therearetwoplacesthatthissignaloccur.Itwouldbeanaddedcomplicationto165cancelbothnegativesignalsatthesametime.Ontheotherhand,theinducedsignalforCp=1000pFandCn=490pFhaspositivepeakatthebeginningandthengotoanegativepeakasseeninFigure3.38.Wechosetotestwhetherthistypeofpulsecouldbecanceledbyinjectingaexternalpulseonthegroundplane.Giventheavailableequipment,thepulseshowninFigure3.38waseasiertoreplicateandtrytocancelthenegativesignalwithanexternalpositivesignal.Figure3.38:InducedsignalsarereadoutbyAGETElectronics.InthisCpandCnare1000and490pF,respectively.Totestthismethod,thethatCpandCnare1000and490pFwaschosen.Anexternalopposite-polaritysignalwasgeneratedusingatimingamwithaninvertedtransformer.InFigure3.39(top),theinducedsignalfromthetransitionofthegatinggridhasthenegativepeakof182ADCchannels.Therearesomeationsafterthegatinggridisclosed.Forreference,eachtimebucketoutofthe512showncorrespondsto40nsforthis.AsshowninFigure3.39(bottom),theexternalsignalsendingthroughthegroundplanedoesnotperfectlymatchtheinducedsignalfromthegatinggridduetothelimitationofanelectronicmodule(TennelecTC-241S).Thepurposeofthismethodistocancelthatinduced166Figure3.39:Toppanel:theinducedsignalcomesfromthetransitionofthegatinggrid.Bottompanel:theexternalsignalwhichisusedtocancelthenoiseissentthruthegroundplane.signalifitispossibleormakeittobeasmallpositivesignalsothatitwillnottriggerthereadoutelectronics.InFigure3.40,thesuperpositionsignalbetweentheinducedsignalofthegatinggrid(black)andanexternalsignal(red)usedtocancelthenoisehasasmallnegativepeakof50ADCchannelswhichisaround1.2%ofthemaximumADCchannel(4096channels).Intheoriginalsetup,thegroundplaneisconnectedtothebodyoftheTPCviaa50resistor.Weexaminedatthesignalfromthegroundplaneduringthetransitionofthegatinggridontheoscilloscopeandobservedanumberofquickoscillations.Clearly,thegroundplaneisstronglycoupledtothegatinggrid,withanestimatedcapacitanceof1.5nFbetweenthepositivewirestogroundplaneand1.5nFtogroundplane.Toreducethecoupling,weshortedthegroundplanetogroundwithadeadshort.Thisbringsthenegative167Figure3.40:Useanexternalsignal(red)whichhastheoppositepolarityifthenoise(black)tocancelit.Thebluelineshowsthesuperpositionofthosetwosignal.Ithasasmallnegativepeakof50ADCchannels.peakdownto30channels.WithoutopeningtheTPC,however,wecanonlydothisatthedownstreamendofthegroundplane.Alaterandmoredetailedexaminationshowedthatthereductionofnoisebyshortingthegroundplanewasmosteforpadsnearthedownstreamendofthepadplane.Itwouldbeinterestingtoallowthegroundplanetobeshortedalsoattheupstreamend,butthatwouldrequireopeningtheTPCandphysically168modifyingthetheconnectionstotheexternalgroundandforthatreason,thishasnotyetbeenattempted.3.3.7GatinggriddrivertestInOctober2015,thecommissioningexperimentoftheSˇRITTPCwasperformedwith200AMeV79SebeamoutsidetoSAMURAImagnet.InFigure3.41,theTPCwascomplementedbyarraysoftriggerdetectors.Toprovidecentralityselection,weusedMultiplicityTriggerArray(MTA)coveringtheleftandrightsidesoftheTPCwith30modulesinstalledoneachsidesandtheKATANA-Multiplicityarrays,consistsof5paddlesonthedownstreamleftsideand7paddlesonthedownstreamrightsideoftheTPC.TheKATANA-VetoarraysegmentmadeofthreethinscintillatorsdetectorswasplacedupstreamoftheKATANAMultiplicityarray.Theasymmetryofthearrayisdesignedforthecurvedtrajectoriesofthebeamandpositivelychargedparticles.Forthecommissioningexperiment,theKATANAarraywaspositionedsothatthecentralofKATANA-Vetointerceptthebeampath.TheActiveVetowaspositionat22cmupstreamofthetargetsothat4scintillatordetectorssurroundingthetargetgeneratethevetosignalwhenthebeamdeviatesfromthetarget.AllthescintillatordetectorswerewithMulti-PixelPhotonCounters(MPPC).ThesignalsfromtheSˇRITTPCarereadoutbytheGETelectronics(seeSection2.8).Thegatinggridtesthasbeendoneasthefollowing.ThebeamenterstheTPCvolumeandionizedthegasatomsormolecules.Theionizedelectronsdriftupwardtothegatinggrid.Inthistest,theanodeplanewasbiasedto680Vsothatthenumberofinducedchargesdonotsaturatethereadoutelectronics.Thevoltageofthecathodewassetto-6632V.Tostudythetransparencyofthegatinggridasafunctionofthecommonaveragevoltageofthegatinggrid,webiasthegatinggridtothisvalueandmeasurethesignalonthepadsofthepad169Figure3.41:AschematicofthecommissioningexperimentsetupoftheSˇRITTPC[74].planeintheSˇRITTPCwiththeGETelectronicsreadout.Thesesignalsareproportionalthenumberofsecondaryelectronscomingthoughthegatinggridtotheanodewires.Then,wemeasurethegatinggriduntiltheinducedsignalsstopincreasingandattainaconstantcorrespondingtotheconditionwherethegatinggridistransparenttotheelectrons.Ifweplottheratioofsignalonthepadsdividedbythemaximumsignalattainedforacompletetransparentgatinggrid,weobtainthetransparencyofthegatinggridasafunctionofthegatinggridaveragevoltage.Weshouldnotethatwehavealsotochangethevoltagesappliedtothecageforeachnewaveragegatinggridaveragevoltage,asdescribedininSection2.1.7inordertoobtainthecorrectdriftforeachaveragegatinggridvoltagesetting.ThetransparencyofthegatinggridasafunctionofthecommonvoltageofthegatinggridisshowninFigure3.42.Theperformanceofthegatinggridhasbeenpredictedbyusingandanalyticalsolution(seeSection3.2).BothGandtheanalyticalsolutioncanpredictthetransparencyofthegatinggridaccurately.wereiteratethatifthevoltage170ofthecathodeissettoatvalueasmightbeexpectedforatcountergas,wewouldalsoneedtoadjustthevoltageofthegatinggridtoachieveafulltransparencywhenthegatinggridisopened.Inthiscase,wecanusetheresultfromthecalculationtoprovidethevoltageneededtoapplytothegatinggrid.Figure3.42:TransparencyofthegatinggridoftheSˇRITTPCasafunctionofthecommonvoltageofthegatinggridforVcathode=-6632V.Oncethevoltageofgatinggridthatcangiveafulltransparencyisknown,wehavetestedtheperformanceofthegatinggridinbipolarmode(seeSection3.2.4)whichisusedtooperatetheSˇRITTPC.Thetesthasbeenperformedasthefollowings.Firstofall,thevoltagesofthecathodeandgatinggridweresetto-6632and-114.8Vwhichprovideafulltransparencyforelectrons.Then,weappliedthevoltageVg)tothealternatingwiresonthegatinggridandmeasuredthesignalsonthepadplane.TheperformanceofthebipolargatinggridisshowninFigure3.43.Notethatallthegatinggridtestshavebeen171performedintheabsenceofamagneticWithVg=10V,wecan50%closethegatinggridandthegatinggridtendstobecompletelyclosedatVg20V.ThisresultrelativelyagreeswiththecalculationbyinFigure3.11(Section3.2.4).Wenotethatthecalculatedvoltagerequiredtoclosethegatinggridislargerforthesimulationthanfortheanalyticsolution.Thisoccursbecausetheelectronsdonotpreciselyfollowtheaveragetrajectorygivenbytheelectriclines.Thisisparticularlyrelevantathighermagneticvaluesbecausetheelectriclinesnearthegatinggridwiresarenotparalleltothemagneticandthistendstobendtheelectrontrajectoriesindirectionsparalleltothegatinggridwires.Thenrandomaboutthemeantrajectoriescanallowtheelectronstoavoidthegatinggridandtraveltotheanodewires.ThiscanbemodeledbydoingMonteCarlosimulationsoftheelectrondrift.SuchsimulationsareshowninFigure3.8.IntheSˇRITTPCexperiment,thedetectorwillbeoperatedinthemagneticof0.5T.ThevoltageVg)requiredtoclosethegatinggridis50Vfor0.5T.Typically,onewouldchooseavalueforVgthatissomewhatlargertocompensateforimperfectionsinthegatinggridfabrication.WenotethattheSTARTPCusealargervalueofVgof75Vtominimizeleaksinthegatinggrid.Thuswehavetestedtheperformanceofthegatinggridanditsdriverat75Vanditperformedsatisfactorily.However,thevalueofVgisproportionaltotheinducednoisesignalwhenthetransitionfromclosedtoopenstateofthegatinggridoccurs.SothereisanincentivetokeepVgassmallaspossible.172Figure3.43:TransparencyofthegatinggridoftheSˇRITTPCasafunctionofthevoltageVg)ofthegatinggridforVcathode=-6632V.173Chapter4Conclusions4.1ConclusionsTheSAMURAIPion-ReconstructionandIon-TrackerTimeProjectionChamber(SˇRITTPC)hasbeensuccessfullyconstructedatMichiganStateUniversityaspartofaninterna-tionalcollaborationstoconstraintthesymmetryenergyatsupra-saturationdensityregion.ThepropertiesoftheSˇRITTPChasbeensimulatedbyGARFIELDandANSYSrMaxwell.Theresultfromthesimulationshowsthattheelectricinthedriftvolumebecomesuniformafter2.3cmfromthewallofthecage.Inhighdensitytrackenvi-ronment,thedistortionoftheelectricclosetothewalldoesnottlythetrackreconstructionofaparticle.ThepropertiesofthegatinggridoftheSˇRITTPChasbeenstudiedbyusingGARFIELDandelectrostaticanalyticalsolutions.BothGARFIELDandanalyticalsolutionsareinareasonableagreementtodescribethetransparencyofthegatinggridasafunctionofthecommonaveragevoltageformono-polarmodeandasafunctionofthevoltageVg)forbipolarmode.Anewgatinggriddriverforatimeprojectionchamberhasbeendesignedtooperatethegatinggridwiresinbipolarmode.TheperformanceofthegatinggriddriverhasbeencalculatedbyPSPICEcicuitanalysisprogram.Accordingtothesimulation,itopenandclosethegatinggridoftheSˇRITTPCin0.3and2s,respectively.Thecircuitconsistsof1742pairsofN-andP-MOSFETswitchesandcanbeadoptedtoanyTPCwiththegatinggridoperatedinbipolarmode.WehaveshownhowthePSPICEcanbeusedtounderstandthepropertiesofthecircuit.ThegatinggriddriverhasbeenusedsuccessfullyinthebeamtestoftheSˇRITTPCinOctober,2015.ItwillbeusedonthecommissioningexperimentoftheTPCinsidetheSAMURAIdipolemagnetaswellastheseriesoftheexperimentstostudyheavyioncollisionsusingtheSˇRITTPCattheRIBFinRIKEN,Japan.175BIBLIOGRAPHY176BIBLIOGRAPHY[1]E.Pollacco,S.Anvar,H.Baba,P.Baron,D.Bazin,C.Belkhiria,B.Blank,J.Chavas,P.Chomaz,E.Delagnes,F.Druillole,P.Hellmuth,C.Huss,E.Galyaev,B.Lynch,W.Mittig,T.Murakami,L.Nalpas,J.-L.Pedroza,R.Raabe,J.Pibernat,B.Raine,A.Rebii,A.Taketani,F.Saillant,D.Suzuki,N.Usher,andG.Wittwer,GET:AGenericElectronicSystemforTPCsforNuclearPhysicsExperiments,PhysicsProcedia37(2012)1799{1804.[2]J.Va'vra,ReviewofWireChamberAging,Nucl.Instrum.Meth.A252(1986)547{563.[3]C.D.Anderson,ThePositiveElectron,Phys.Rev.43(1933)491{494.[4]G.T.Jones,Positronannihilationin,PhysicsEducation34(1999)no.5,276.[5]H.Hunteler,DevelopmentandEvaluationofaMultiwireProportionalChamberforaHighResolutionSmallAnimalPETScanner.PhDthesis,WestfalisheWilhelms-UniversitatMunsterInstitutfurKernphysik,2007.[6]G.Charpak,D.Rahm,andH.Steiner,Somedevelopmentsintheoperationofmultiwireproportionalchambers,NuclearInstrumentsandMethods80(1970)no.1,13{34.[7]A.Breskin,G.Charpak,C.Demierre,S.Majewski,A.Policarpo,F.Sauli,andJ.Santiard,High-accuracy,bidimensionalread-outofproportionalchamberswithshortresolutiontimes,NuclearInstrumentsandMethods143(1977)no.1,29{39.[8]S.Klein,TheTimeProjectionChamberturns25,CernCourier44(2004).[9]R.Shane,A.McIntosh,T.Isobe,W.Lynch,H.Baba,etal.,SˇRIT:Atime-projectionchamberforsymmetry-energystudies,arXiv:1409.6343[physics.ins-det].[10]G.F.Knoll,RadiationDetectionandMeasurement.JohnWiley&Sons,Inc.,1999.[11]W.Blum,W.Riegler,andL.Rolandi,ParticleDetectionwithDriftChambers.Springer,2008.177[12]A.Andronic,S.Biagi,P.Braun-Munzinger,C.Garabatos,andG.Tsiledakis,Driftvelocityandgaininargon-andxenon-basedmixtures,NuclearInstrumentsandMethodsinPhysicsResearchSectionA:Accelerators,Spectrometers,DetectorsandAssociatedEquipment523(2004)no.3,302{308.[13]R.Bossingham,GasProperties,tech.rep.,STARcollaborations,BrookhavenNationalLaboratory,1999.[14]E.Mathieson,InducedChargeDistributionsinproportionaldetectors,tech.rep.,BrookhavenNationalLaboratory,1991.[15]V.Blinov,I.Popkov,andA.Yushkov,Agingmeasurementsinwirechambers,NuclearInstrumentsandMethodsinPhysicsResearchSectionA:Accelerators,Spectrometers,DetectorsandAssociatedEquipment515(2003)95{107.[16]N.N.D.GuptaandS.K.Ghosh,AReportontheWilsonCloudChamberandItsApplicationsinPhysics,Rev.Mod.Phys.18(Apr,1946)225{290.[17]D.A.Glaser,SomectsofIonizingRadiationontheFormationofBubblesinLiquids,Phys.Rev.87(Aug,1952)665{665.[18]D.A.Glaser,Thebubblechamber,Sci.Am.192(1955)no.2,46.[19]V.E.Barnes,P.L.Connolly,D.J.Crennell,B.B.Culwick,W.C.Delaney,W.B.Fowler,P.E.Hagerty,E.L.Hart,N.Horwitz,P.V.C.Hough,J.E.Jensen,J.K.Kopp,K.W.Lai,J.Leitner,J.L.Lloyd,G.W.London,T.W.Morris,Y.Oren,R.B.Palmer,A.G.Prodell,D.Radojicic,D.C.Rahm,C.R.Richardson,N.P.Samios,J.R.Sanford,R.P.Shutt,J.R.Smith,D.L.Stonehill,R.C.Strand,A.M.Thorndike,M.S.Webster,W.J.Willis,andS.S.Yamamoto,ObservationofaHyperonwithStrangenessMinusThree,Phys.Rev.Lett.12(Feb,1964)204{206.[20]J.J.Aubert,U.Becker,P.J.Biggs,J.Burger,M.Chen,G.Everhart,P.Goldhagen,J.Leong,T.McCorriston,T.G.Rhoades,M.Rohde,S.C.C.Ting,S.L.Wu,andY.Y.Lee,ExperimentalObservationofaHeavyParticleJ,Phys.Rev.Lett.33(Dec,1974)1404{1406.[21]J.E.Augustin,A.M.Boyarski,M.Breidenbach,F.Bulos,J.T.Dakin,G.J.Feldman,G.E.Fischer,D.Fryberger,G.Hanson,B.Jean-Marie,R.R.Larsen,V.Luth,H.L.Lynch,D.Lyon,C.C.Morehouse,J.M.Paterson,M.L.Perl,B.Richter,P.Rapidis,R.F.Schwitters,W.M.Tanenbaum,F.Vannucci,G.S.Abrams,D.Briggs,W.Chinowsky,C.E.Friedberg,G.Goldhaber,R.J.Hollebeek,J.A.Kadyk,B.Lulu,178F.Pierre,G.H.Trilling,J.S.Whitaker,J.Wiss,andJ.E.Zipse,DiscoveryofaNarrowResonanceine+eAnnihilation,Phys.Rev.Lett.33(Dec,1974)1406{1408.[22]G.Charpak,R.Bouclier,T.Bressani,J.Favier,andC.Zupancic,Theuseofmultiwireproportionalcounterstoselectandlocalizechargedparticles,NuclearInstrumentsandMethods62(1968)no.3,262{268.[23]G.CharpakandF.Sauli,Multiwireproportionalchambersanddriftchambers,NuclearInstrumentsandMethods162(1979)no.1,405{428.[24]J.MarxandD.Nygren,TheTimeProjectionChamber,Phys.Today31N10(1978)46.[25]R.M.Barnett,C.D.Carone,D.E.Groom,T.G.Trippe,C.G.Wohl,B.Armstrong,P.S.Gee,G.S.Wagman,F.James,M.Mangano,K.onig,L.Montanet,J.L.Feng,H.Murayama,J.J.andez,A.Manohar,M.Aguilar-Benitez,C.Caso,R.L.Crawford,M.Roos,N.A.ornqvist,K.G.Hayes,K.Hagiwara,K.Nakamura,M.Tanabashi,K.Olive,K.Honscheid,P.R.Burchat,R.E.Shrock,S.Eidelman,R.H.Schindler,A.Gurtu,K.Hikasa,G.Conforto,R.L.Workman,C.Grab,andC.Amsler,ReviewofParticlePhysics,Phys.Rev.D54(Jul,1996)1{708.[26]L.Landau,Ontheenergylossoffastparticlesbyionization,J.Phys.(USSR)8(1944)201{205.[27]F.Sauli,Principlesofoperationofmultiwireproportionalanddriftchambers,LecturesgivenintheAcademicTrainingProgrammeofCERN(1977).[28]S.N.Ahmed,3-leddetectors,inPhysicsandEngineeringofRadiationDetection(SecondEdition),S.N.Ahmed,ed.,pp.157{231.Elsevier,secondeditioned.,2015.[29]R.S.WMeasurementofthegasconstantsforvariousproportional-countergasmixtures,NuclearInstrumentsandMethods115(1974)no.2,461{463.[30]R.Hendricks,Thegasationfactorinxenonledproportionalcounters,NuclearInstrumentsandMethods102(1972)no.2,309{312.[31]R.AllemandandG.Thomas,Nouveaudtecteurdelocalisation,NuclearInstrumentsandMethods137(1976)no.1,141{149.179[32]B.Duval,J.Barth,R.Deslattes,A.Henins,andG.Luther,Position-sensitiveX-raydetector,NuclearInstrumentsandMethodsinPhysicsResearch222(1984)no.1,274{278.[33]G.Luther,P.Cowan,A.Henins,andS.Brennan,Newtwodimensionalpositionsensitiveproportionaldetectorsusingchargedivision,NuclearInstrumentsandMethodsinPhysicsResearchSectionA:Accelerators,Spectrometers,DetectorsandAssociatedEquipment246(1986)no.1,537{540.[34]H.J.Hilke,Timeprojectionchambers,ReportsonProgressinPhysics73(2010)no.11,116{201.[35]H.Sato,T.Kubo,Y.Yano,K.Kusaka,J.Ohnishi,K.Yoneda,Y.Shimizu,T.Motobayashi,H.Otsu,T.Isobe,T.Kobayashi,K.Sekiguchi,T.Nakamura,Y.Kondo,Y.Togano,T.Murakami,T.Tsuchihashi,T.Orikasa,andK.Maeta,SuperconductingDipoleMagnetforSAMURAISpectrometer,AppliedSuperconductivity,IEEETransactionson23(June,2013).[36]T.Kobayashi,N.Chiga,T.Isobe,Y.Kondo,T.Kubo,K.Kusaka,T.Motobayashi,T.Nakamura,J.Ohnishi,H.Okuno,H.Otsu,T.Sako,H.Sato,Y.Shimizu,K.Sekiguchi,K.Takahashi,R.Tanaka,andK.Yoneda,SAMURAIspectrometerforRIbeamexperiments,NuclearInstrumentsandMethodsinPhysicsResearchSectionB:BeamInteractionswithMaterialsandAtoms317,PartB(2013)294{304.[37]T.Kubo,RIbeamseparatorBigRIPSatRIKENandelsewhereinJapan,NuclearInstrumentsandMethodsinPhysicsResearchSectionB:BeamInteractionswithMaterialsandAtoms204(2003)97{113.14thInternationalConferenceonElectromagneticIsotopeSeparatorsandTechniquesRelatedtotheirApplications.[38]O.Kamigaito,M.Kase,N.Sakamoto,Y.Miyazawa,E.Ikezawa,N.Fukunishi,S.Kohara,M.Fujimaki,M.Hemmi,T.Chiba,Y.Chiba,H.Ryuto,A.Goto,andY.Yano,ConstructionofaboosterlinacfortheRIKENheavy-ionlinac,ReviewofScienInstruments76(2005)no.1,.[39]Y.Yano,TheRIKENRIBeamFactoryProject:Astatusreport,NuclearInstrumentsandMethodsinPhysicsResearchSectionB:BeamInteractionswithMaterialsandAtoms261(2007)no.1-2,1009{1013.[40]G.Raietal.,ATPCdetectorforthestudyofhighmultiplicityheavyioncollisions,NuclearScience,IEEETransactionson37(1990)56{64.180[41]F.Sauli,GEM:Anewconceptforelectronationingasdetectors,NuclearInstrumentsandMethodsinPhysicsResearchSectionA:Accelerators,Spectrometers,DetectorsandAssociatedEquipment386(1997)no.2,531{534.[42]Y.Giomataris,P.Rebourgeard,J.Robert,andG.Charpak,MICROMEGAS:ahigh-granularityposition-sensitivegaseousdetectorforhighpenvironments,NuclearInstrumentsandMethodsinPhysicsResearchSectionA:Accelerators,Spectrometers,DetectorsandAssociatedEquipment376(1996)no.1,29{35.[43]O.Kiselev,O.ProkandA.Vorobyov,Proceedingsofthe7thInternationalWireChamberConferenceMeasurementofelectrondriftvelocitiesandLorentzanglesinfastgasmixtures,NuclearInstrumentsandMethodsinPhysicsResearchSectionA:Accelerators,Spectrometers,DetectorsandAssociatedEquipment367(1995)no.1,306{310.[44]Schultz,G.,Charpak,G.,andSauli,F.,Mobilitiesofpositiveionsinsomegasmixturesusedinproportionalanddriftchambers,Rev.Phys.Appl.(Paris)12(1977)no.1,67{70.[45]T.Yamashita,H.Kobayashi,A.Konaka,H.Kurashige,K.Miyake,M.Morii,T.Nakamura,T.Nomura,N.Sasao,Y.Fukushima,M.Nomachi,O.Sasaki,F.Suekane,andT.Taniguchi,Measurementsoftheelectrondriftvelocityandpositive-ionmobilityforgasescontainingCF4,NuclearInstrumentsandMethodsinPhysicsResearchSectionA:Accelerators,Spectrometers,DetectorsandAssociatedEquipment283(1989)no.3,709{715.[46]K.T.A.L.Burm,CalculationoftheTownsendDischargeCoandthePaschenCurveCo,ContributionstoPlasmaPhysics47(2007)no.3,177{182.[47]R.Veenhof,ChoosingagasmixturefortheALICETPC,tech.rep.,CERN,2003.ALICE-INT-2003-29.[48]J.Gawlowski,J.Maurin,andJ.Niedzielski,Argonionizationdetectorsensitivetohydrogen,JournalofChromatographyA168(1979)no.1,1{7.[49]E.-G.KimandJ.-L.Bredas,Thenatureofthealuminum-aluminumoxideinterface:Ananoscalepictureoftheinterfacialstructureandenergy-levelalignment,OrganicElectronics14(2013)no.2,569{574.181[50]F.Sauli,GaseousRadiationDetectors:FundamentalsandApplications.CambridgeMonographsonParticlePhysics,NuclearPhysicsandCosmology.CambridgeUniversityPress,2014.[51]R.Veenhof,recentdevelopments,NuclearInstrumentsandMethodsinPhysicsResearchSectionA:Accelerators,Spectrometers,DetectorsandAssociatedEquipment419(1998)no.2-3,726{730.[52]I.Lehraus,R.Matthewson,andW.Tejessy,MATERIALSTESTFACILITYFORTHEALEPHTPC,tech.rep.,CERN,1989.ALEPH-89-089.[53]U.J.Becker,Y.H.Chang,J.F.Donahue,M.E.Sarakinos,G.E.Sasser,andB.R.Smith,Calibrationsetupfordriftchambergases,NuclearInstrumentsandMethodsinPhysicsResearchSectionA:Accelerators,Spectrometers,DetectorsandAssociatedEquipment306(1991)no.1,194{199.[54]T.Kunst,B.Gtz,andB.Schmidt,Precisionmeasurementsofmagneticctionanglesanddriftvelocitiesincrossedelectricandmagnetic,NuclearInstrumentsandMethodsinPhysicsResearchSectionA:Accelerators,Spectrometers,DetectorsandAssociatedEquipment324(1993)no.1,127{140.[55]U.J.Becker,Y.-H.Chang,J.P.Rodin,andB.R.Smith,Driftvelocityandctioninnon-perpendicularelectricandmagnetic,NuclearInstrumentsandMethodsinPhysicsResearchSectionA:Accelerators,Spectrometers,DetectorsandAssociatedEquipment360(1995)no.3,476{480.[56]W.Bettsetal.,StudiesofServeralWirenadPadationsfortheSTARTPC,Tech.Rep.263,STARCollaboration,1996.[57]E.Gatti,A.Longoni,P.Semenza,andH.Okuno,OptimumgeometryforstripcathodesorgridsinMWPCforavalanchelocalizationalongtheanodewires,Nucl.Instrum.Meth.163(1979)83{92.[58]E.MathiesonandJ.Gordon,CathodeChargeDistributionsinMultiwireChambers.2:ApproximateandEmpiricalformulae,Nucl.Instrum.Meth.A227(1984)277{282.[59]W.Baldinietal.,ALaserBasedInstrumentforMWPCWireTensionMeasurement,tech.rep.,CERN,2007.CERN-LHCb-2007-120.[60]T.Liu,C.Horn,A.Junique,C.Lippmann,M.Mager,L.Musa,andT.Richert,Proceedingsofthe2ndInternationalConferenceonTechnologyandInstrumentationin182ParticlePhysics(TIPP2011)Trigger-InducedMechanicalResonancesofGatingGridWiresintheMulti-WireProportionalChambersoftheALICETPC,PhysicsProcedia37(2012)472{477.[61]M.Anderson,J.Berkovitz,W.Betts,R.Bossingham,F.Bieser,R.Brown,M.Burks,etal.,TheSTARtimeprojectionchamber:auniquetoolforstudyinghighmultiplicityeventsatRHIC,NuclearInstrumentsandMethodsinPhysicsResearchSectionA:Accelerators,Spectrometers,DetectorsandAssociatedEquipment499(2003)no.2-3,659{678.[62]H.Sako,H.Sugimura,J.Ahn,Y.Han,S.Hasegawa,S.Hwang,Y.Ichikawa,K.Imai,R.Kiuchi,K.Ozawa,S.Sato,andK.Shirotori,DevelopmentofaprototypeGEMTPCwithagatinggridforanH-dibaryonsearchexperimentatJ-PARC,NuclearInstrumentsandMethodsinPhysicsResearchSectionA:Accelerators,Spectrometers,DetectorsandAssociatedEquipment763(2014)65{81.[63]D.Bryman,M.Leitch,I.Navon,T.Numao,P.Schlatter,M.Dixit,C.Hargrove,H.Mes,J.MacDonald,R.Skegg,J.Spuller,R.Burnham,M.J.-M.Poutissou,G.Azuelos,P.Depommier,J.-P.Martin,R.Poutissou,M.Blecher,K.Gotow,andA.Carter,Gatedgridsystemusedwithatimeprojectionchamber,NuclearInstrumentsandMethodsinPhysicsResearchSectionA:Accelerators,Spectrometers,DetectorsandAssociatedEquipment234(1985)no.1,42{46.[64]F.Bohmer,M.Ball,S.Dorheim,C.Hoppner,B.Ketzer,I.Konorov,S.Neubert,S.Paul,J.Rauch,andM.Vandenbroucke,Simulationofspace-chargectsinanungatedGEM-basedfTPCg,NuclearInstrumentsandMethodsinPhysicsResearchSectionA:Accelerators,Spectrometers,DetectorsandAssociatedEquipment719(2013)101{108.[65]J.Adam,C.Baird,etal.,Astudyofageingctsinwirechamber,Nucl.Instrum.Meth.217(1983)291{297.[66]H.SipilaandM.Jarvinen,Extendedlifetimeofawirechamber,Nucl.Instrum.Meth.217(1983)298{300.[67]P.Nemethy,P.J.Oddone,N.Toge,andA.Ishibashi,GatedTimeProjectionChamber,Nucl.Instrum.Meth.212(1983)273.[68]S.Biagi,MonteCarlosimulationofelectrondriftandincountinggasesundertheeofelectricandmagnetic,Nucl.Instrum.Meth.A421(1999)234{240.183[69]S.Amendolia,W.Blum,R.Benetta,M.Cherney,F.Fidecaro,J.Froberger,B.Hubbard,R.Jared,I.Lehraus,F.Liello,P.Marocchesi,R.Matthewson,J.May,T.Meyer,E.Milotti,F.Nanni,A.Peisert,M.Price,F.Ragusa,J.Richstein,R.Richter,L.Rolandi,W.Schlatter,J.Sedgbeer,R.Settles,U.Stierlin,M.Takashima,W.Tejessy,G.Tromba,W.Witzeling,S.L.Wu,andW.Wu,eofthemagneticonthegatingofatimeprojectionchamber,NuclearInstrumentsandMethodsinPhysicsResearchSectionA:Accelerators,Spectrometers,DetectorsandAssociatedEquipment234(1985)no.1,47{53.[70]L.W.NagelandD.O.Pederson,SimulationProgramwithIntegratedCircuitEmphasis(SPICE),tech.rep.,ElectronicsResearchLaboratory,CollegeofEngineering,UniversityofCalifornia,Berkeley,CA,1973.[71]CadenceDesignSystems,PSPICE,http://orcad.com,June,2014.[72]VishayIntertechnology,IRF640MOSFETSwitch,http://www.vishay.com/mosfets/list/product-91036/,2015.[73]VishayIntertechnology,IRF9640MOSFETSwitch,http://www.vishay.com/mosfets/list/product-91086/,2015.[74]G.Jhangetal.,Comissioningexperiment,Paper,October,2015.184