CODA Users Workshop Data Acquisition at Jefferson Lab Newport News June 7, 2004 DAQ Systems in Hall A Overview of Hall A Standard Equipment: HRS, Beamline,... Parity Experiments Third Arms: BigBite, RCS Calorimeter, Neutron Array 12 GeV: MAD Summary Bodo Reitz Jefferson Lab
Personnel Access Tunnel 3 rd Arm Truck Ramp
Hall A HRS: Standard Detector Package left arm (hadron-stack) 960 Ch multihit-tdc (VDC) 128 Ch highres-tdc (trigger) 192 Ch ADC (PID) 240 Ch Scalers right arm (electron stack) 960 Ch multihit-tdc (VDC) 128 Ch highres-tdc (trigger) 256 Ch ADC (PID) 240 Ch Scalers 640 Ch multihit-tdc (FPP)
Standard HRS DAQ Setup typical: coincidence experiment involving both HRS event-by-event readout (plus scalers for normalization) level-one trigger: NIM/CAMAC electronics no higher level triggers, (no fast-clears) DAQ: running CODA 2.5/VXWorks 5.4 VME crate with TS (and scalers) in detector hut 2 Fastbus-crates per arm also in detector hut PCs and disk-array in counting house (RHEL)
Performance of the Hall A DAQ Coincidence Setup (2 HRS) single arm rates (raw): 1-300 khz trigger rate after prescaling: <3 khz event size: ~1.5 kb (with pedestal suppression on) computer deadtime: 10% per 1 khz (unbuffered) Single Arm Setup both HRS run independently redistribution of Fastbus-modules avoiding slow modules (LeCroy 1875) utilizing buffered readout: 2.5% per 1 khz (up to 4 khz)
Issues with Present Hall A DAQ Fastbus is obsolete problems with: repairs, spares, additional channels,... switch to VME based modules but: even for VME not many commercial suppliers left rate limitations: to get reasonable deadtimes: CODA-rate <2 khz CODA issues: time-loss for downloading and rebooting after crashes time-loss for ending and starting runs (Hall A specific) only marginal help features and online-documentation within CODA strip charts of CODA are not working
DAQ for Beam- and Target Diagnostics polarized Helium-3 target: NMR and EPR polarimetry: PC/LabView based ARC beam energy, beam position and beam current monitors: included in EPICS environment EP beam energy, Compton and Moller Polarimeter: CODA based (various versions) single ROC in VME Crate custom made VME modules for EP and Compton (Micro-Strip detectors) usually are working fine, as long as the experts are around, and no upgrade is needed various test setups for future experiments (EEL, Test-Lab) CODA based single Crate setups (Fastbus or VME) CODA light would be helpful there will be multi-crate test-setups soon (3 rd arm experiments)
Hall A Parity Experiments (Happex) parity experiments requiring high statistics and therefore high rates (>1MHz) cannot be handled with the standard (event-by-event) Hall A DAQ independent DAQ system required Happex: Integrating DAQ customized ADCs (Princeton) integrating over one helicity pulse (33ms) high resolution (18 bit) high demands on linearity and noise pseudo-randomly adding of a voltage ( DAQ-noise ) to reduce non-linearity present ADC module ( Princeton ) is obsolete, JLab Electronics groups was asked to design and build a follow-up module other PV experiments: require counting based DAQ for 100 MHz rates would need customized electronics for trigger/pid decision within 10 ns
3 rd Arm Experiments Real Compton Scattering (experiment run in 2001) HRS(left) lead glass calorimeter with ~700 blocks (ADCs, some TDCs, some Scalers) DVCS in Hall A (scheduled for summer/fall 2004) HRS(left) proton array (100 PMTs) lead glass calorimeter with 132 blocks some customized electronics (Flash-ADCs)
Experiments Using a 3 rd Arm II BigBite Experiments: SRC scheduled for Jan. 2005 uses both HRS BigBite spectrometer for protons, using scintillators with 152 PMTs scintillator array for neutrons ~100 bars plus ~50 veto detectors DAQ with 400 Ch. of ADCs and TDCs (VME+Fastbus)
Experiments Using a 3 rd Arm III Hall A GEN (end of 2005???) uses BigBite as electron arm BigBite detector package: + MWDC with ~2000 wires (multi-hit TDCs) + scintillators with 32 PMTs (high-res TDCs) + lead glass shower detector (~200 PMTs) (ADCs) uses plastic scintillators as neutron detector + 240 neutron bars (=480 Ch. high-res TDC + ADC) + 200 veto detectors (=200 Ch. multi-hit TDC + ADC)
3 rd Arm Experiments: demand for additional ADC+TDC (and scaler) modules significant fraction of the modules will be (has to be) in VME TDC: CAEN V767, V775, V1190, JLab-F1 ADC: CAEN V792, JLab??? frequent requests for TIR and TriggerSupervisors for help to setup CODA/DAQ systems (good documentation could reduce some of those) radiation damage 3 rd arm electronics will not be as well shielded as the ones in the HRS detector hut noticed a lot more problems with DAQ system in the RCS setup and the ones located along the beam line
JLab @ 12 GeV: MAD A totally new system, with strong requirements!
Comparison MAD vs HRS MAD HRS central momentum: 6.0 GeV /c 4.0 GeV/c configuration: QD QD QQDQ bend: 32 deg 45 deg mom. acceptance: +- 15% +- 4.5% mom. resolution: ~1E-03 1E-04 solid angle: 28 msr (6 msr) 6 msr >35deg (12deg) >12.5deg ext. target: +-6 cm +-5 cm y0 resolution: 2.6 mm (4.6mm) 1.5 mm
MAD: Detector Package
MAD: Some Specs! detector channels for MAD (MAD+1 HRS) 450(700) Ch. ADCs 200(300) Ch. TDCs (high-res., multi-hit) 1700(2700) Ch. TDCs (med-res, multi-hit) 400 Ch. scalers DAQ for MAD: rate capability has to exceed 5 khz, 20 khz desirable
Flash-ADCs and Pipeline-TDCs higher rate capability (>20 khz) avoids delay cables Flash-ADC: no extra TDC for that channel needed significant increase in amount of raw-data network and computer upgrades online data-reduction possibility of software second level trigger Flash-ADC: pulse form discrimination Is the necessary hardware available? pipeline-tdc: CAEN V1190 and JLab F1 first modules have been seen. experience? availability? is there a Flash-ADC in sight???
Summary at present: several CODA 2.x systems with VME/Fastbus issues: minor bugs, documentation near future several 3 rd arm systems under construction replacement of Fastbus which VME modules are available and recommended? 12 GeV new big detector setup, requiring higher rate capabilities upgrade of HRS DAQ necessary at the same time approach with PipelineTDC/FlashADC very desirable