Trigger Implementation for DESPEC and FRS

Dzenana BuljubasicUniversity of Tuzla, Faculty of Science and Mathematics, Department of Physics,

dzenana.buljubasic@gmail.com

Fragment Separator (FRS) represents a system of linked detectors, each deliveringimportant information on secondary particles; mass, charge, half-life.. for later com-plete particle identification. Considering the amount of data recieved during the beamtime, it is necessary to find a method with which we will be able to filter those in-formation as required, without burdening data acquisition system. Two concepts forrunning a data acquisition system are widely used: triggered and free-running. Amodern general-purpose data acquisition should be able to accommodate both modesand also to deal with legacy electronics if needed.

1 Introduction

The region south of 208Pb nuclei along theN = 126 line has been identified as a regionof key interest for the NUSTAR collaborationat GSI/FAIR; it is of high interest for nuclearstructure studies and nuclear astrophysics ap-plication. We are aiming to extend the frontiersof known isotopes and to obtain nuclear struc-ture data on gross properties. NUSTAR experi-ments address front-line physics questions at thelimit of feasibility, and reach a level of complex-ity that requires not only new and innovativedetector concepts, but also calls for an adequateintegrated data acquisition system.

Fig. 1: The region of interest in the chart of nuclei

2 Experimental setup

The main goal of the experiment is to produceand identify new neutron-rich isotopes in the el-

ement range between terbium and rhenium andto measure their production cross section. Thiswill be achieved by extraction of 1GeV 208Pb67+

primary beam from the SIS-18 and focusing iton the production target at the entrance of theFRS. A target of 2.5g/cm2 Be with 223mg/cm2

Nb backing, will be used to produce the isotopesof interest via projectile fragmentation.

Fig. 2: Schematic view of the FRS setup

New NUSTAR equipment, at least as proto-types or start versions, are used for experimentin phase-0. In the FRS, new TPC and MUSICdetectors are used for beam monitoring and par-ticle identification. At the final focus (S4 area),the FRS Ion Catcher and detectors of the DE-SPEC collaboration (AIDA+FATIMA) are be-ing used for the first time. It is the combined useof these novel instruments, which allows thesepioneering studies and the rich nuclear data har-vest in short time.[1]

1

2 3 NDAQ and triggers

2.1 Equipment

Two types of experiments are planned by theHI/DESPEC collaboration, both of which areplanned to be located in the low energy branchof the FRS. A typical in-flight spectroscopyexperiment employs tracking detectors aftera secondary target at the final focal plane ofthe FRS. The secondary target is surroundedwith gamma ray detectors. At a later stage ofthe project, the addition of a particle trackingdetector around it to perform particle detectionin knockout studies, is planned.

The other part of the collaboration is dedi-cated to investigate decay spectroscopy of exoticnuclei produced by the FRS. Here, depending onthe decay modes of the studied nuclei and theobservable of interest, different detectors can beused, sometimes in combination.

2.1.1 AIDA

AIDA (Advanced Implantation Detector Array)is a silicon array for implanting exotic ions thatwill be positioned at the center of the DESPECsetup. The ASICs connected too the silicondetectors are equipped with fast-recovery pre-amplifiers able to identify implantation eventsand the subsequent beta decay. The ASIC isread out with the same system as the R3B sil-icon tracker and is integrated into NDAQ viawhite rabbit time stamping and data flow cou-pling. The AIDA array is located at the centerof the DESPEC setup.

2.1.2 FATIMA

FATIMA (Fast TIMing Array) made of LaBr3is optimized to perform gamma-gamma or beta-gamma measurements to determine life times ofnuclear levels. The timing requirement for theelectronic are quite stringent.

2.1.3 DEGAS

DEGAS will be the work horse germaniumdetector array of DESPEC. In phase-1 it willuse the Euroball detectors in a new geometricconfiguration with updated electronics, runningas a hardware-triggered system.

The other goal is to measure masses in a newregion with FRS Ion Catcher setup: this combi-nation permits to implant the ions in the Cryo-

genic Stopping Cell (CSC) and to extract themto the Multi-Reflection Time-Of-Flight MassSpectrometer (MR-ToF-MS), where their masscan be measured with high (60keV ) accuracyeven for very low yields (10 ions in total).[2]

3 NDAQ and triggers

All NUSTAR experiments use radioactivebeams and rely on tracking; Time-of-Flight(ToF) and energy-loss information delivered bydetectors of the FRS. Therefore, the ability tocorrelate data streams of the FRS with thoseof the experiments is crucial. Thus a commondata acquisition infrastructure is critical foran optimal operation of NUSTAR experiment.There are a number of characteristics such anintegrated data acquisition system (NUSTARData AcQuisition NDAQ) has to have.

The most important one is the necessary flex-ibility to handle different detectors and read-out electronic type. NDAQ also needs to beable to incorporate future developments. It hasto provide not only a maximum interoperabilitybetween the readout systems of different detec-tor systems, but also of different detector gen-erations. This means handling of different op-erating modes, the ability to cope with largedata rate while assuring data integrity, and morecomplex control and command tasks.

3.1 Hardware-triggered mode

In the hardware-triggered mode, local triggersignals, generated from individual detectors,are combined logically via a coincidence matrixin a Signal Loging Box (SLB) to generate andto dispatch a Master Start (MS) to set ofdetectors (e.g. FRS/FATIMA). The detectorsrecieving the MS can depend on the triggercondition that generated it; the trigger isdefined according to the physics requirementsof the given experiment.

The transmission times of signals betweendifferent subsystems is, especially for large dis-tances, a critical issue. For correct separationand identification of secondary beams, datafrom the FRS is required by all NUSTAR exper-iments using NDAQ. The ions pass through theFRS detectros a relatively long time before thetrigger decision can be drawn. A system with

3.2 Software triggered free-running mode 3

digitization after reception of trigger-based gaterequires all analog FRS signals to be delayedfor several µs to several ns until the arrival ofthe trigger.

The solution is to have a buffer chain for everychannel in the front-end electronics, where alldigitized signales are stored with time stamps.Data transmission waits for a validation signalwhich causes the extraction of the data fromthe readout within the time window based on atrigger provided by the experiment.

3.2 Software triggered free-runningmode

The free-running operating mode, together witha software-triggered event selection, constitutesan advantage because it is independent ofsignal latencies and allows for a significantreduction of the global dead time. All digitizedsignals are read out and trigger coincidencesare constructed in software or firmware.

In the free-running mode, subsystems areworking asynchronously and every channel isself-triggered: each analog signal passing a localtreshold is digitized and time stamped. Theevent is generated later in the DAQ recordingchain by a more advanced trigger logic. But,the recording of all uncorrelated signals in afree-running experiment leads to very largeamounts of data that waste computing anddata storage resources, and is not favoured.

The requirements to handle this operatingmode are to provide properly scaled bandwidthaccording to the requirements of the subsys-tems, a back pressure mechanism which is ableto detect and manage any saturation of thedata flow from individual detectors or detectorscomponents.[2]

3.3 TRLO II

The VULOM-based trigger logics (firmware) re-places several crates of NIM and CAMAC elec-tronics used for trigger decisions, counting anddead-time locking and open up many new pos-sibilities for handling pending and calibrationtriggers. As the TRIDI modules need similarkinds of logics, and actually could serve as smalllocal trigger systems for stand-alone test opera-tions, the same code is used for both, with minor

tweaks for the different inputs and outputs. Alllogics run with one 100 MHz clock. The boldintention is that this firmware should be suit-able for (almost) any triggered nuclear physicsexperiment.[3]

4 Results

The goal of the project was to implement thetrigger system allowing to run alternatively bothsetups at the end of the FRS: (i) active stopperand CSC cell to perform mass and beta life timemeasurements or (ii) the DESPEC setup withAIDA, DEGAS and FATIMA. Figure 3 shows aschematic view the different systems and theirinformation used to build this general triggersystem.

Fig. 3: Schematic representation of logic gates forthe FRS

The triggers were encoded in the VULOM-TRLO II confiugration using the configurationfile based on the scheme in Figure 3. A print-out of the implemented configuration is shownin Figure 4 as a trigger matrix. To do this 7inputs were used, namely: (i) ”or” signal fromAIDA electronic, (ii) ”or” signal from the beta-plastic, (iii) ”or” signal from the FATIMA detec-tors, (iv) ”or” signal of DEGAS, (v) scintillator”41” (SC41) of the FRS (first in the final focalplane) which means an ion passed throught theFRS, (vi) scintillator ”43” (SC43) of the FRS(last in the final focal plane) and (vii) ”or” sig-nal of all active stopper silicons. With those,fourteen triggers were built, as listed below:

(i) AIDA trigger; used for calibration purposesand as a counter

(ii) Beta-plastic with a veto on incoming par-ticle; this trigger is used as the main beta-trigger with the DESPEC setup

(iii) Implantation trigger for the main DESPECsetup which is the beta-plastic in coinci-

4 References

dence with the SC41 to confirm the ionpassing through the FRS is really reachingAIDA

(iv) Beta-plastic alone; for calibration purposesor test measurement (not disciminating im-plantation and decay)

(v) As Trigger 3, but with a coincidence withFATIMA; for prompt beta-gamma coinci-dence, used for specific settings without iso-mers or for calibration

(vi) FATIMA trigger alone; for calibration pur-poses

(vii) As Trigger 3, but having any of the twogamma detectors (FATIMA or DEGAS) incoincidence; for pure prompt beta-gammatrigger

(viii) ”or” of DEGAS and FATIMA; for gammacalibration purposes

(ix) DEGAS trigger alone; for calibration pur-poses

(x) SC41 alone (main FRS trigger) to detectparticles incoming in the FRS; for FRS cal-ibration and tuning

(xi) SC41 and SC43; which is particle passingthrough the active stopper and being im-planted in CSC (Cryogenic Stopping Cell)to have the identification of them

(xii) SC41 in coincidence with the active stopperand anti-coincidence with SC43; for activestopper implantation trigger

(xiii) Active stopper trigger, both implantationand decay; for calibration purposes

(xiv) Active stopper and anti-coincidence withSC41; for beta-decay for the active stopperruns

Because of the delay in the beam time,scheduling active stopper, AIDA, DEGAS andFATIMA, were not mounted at S4 by the endof the project. Nevertheless the logic was testedfor trigger 10, 11, 12 and 13 using pulser signalssimulating the detectors and was shown to work.The delay to implement for the coincidence (oranti-coincidence) in the module was equivalentto the one added intentionally on the pulser sig-nals. Trigger 14 is defined with a bit more re-finement than trigger 4. This is due to the pres-ence of the scintillator SC43 in the setup with

Fig. 4: Trigger matrix

the active stopper. Indeed the SC41 thresholdcould be high and it is selected to assure the besttime of flight for the FRS. The SC43 plastics canhave a really low threshold. This means that re-ally light particles (protons) passing through theFRS could trigger the active stopper, but wouldpass through SC41 and the active stopper andwould then trigger SC43 and the active stopper.So in this case they will not induce fake beta-trigger in the active stopper. For the DESPECsetup we can not protect from such trigger, butthose spurious events have to be removed in theanalysis.

Acknowledgments

Many thanks to dr. Stephane Pietri, for his pa-tience and guidence throughout this experienceand also to dr. Hans Tornqvist, for his assis-tance.

References

[1] S. Pietri, Proposal: Search for new neutron-rich isotopes and exploratory studies in theelement range from terbium to rhenium

[2] H. Simon, Technical Report for the Designof the NUSTAR Data AcQuisition System,(2016)

[3] H.T. Johansson, TRLO II - flexible FPGAtrigger control, (2010)