Defining a STAR Trigger setup

last update 17-Jan-2015: je

You don't have to understand anything in this trigger area if you just use the RUN CONTROL GUI and select canned files. The RUN CONTROL GUI is intended to be your connection to the trigger. The information contained here is intended to give you a simple overview and to act as a reference area for discussions.

Trigger overview

The STAR trigger consists of detectors (BBC, BEMC, EEMC, FPD, FMS, MTD, TOF, VPD, ZDC), various VME boards developed by the trigger group (e.g. RCC2, DSM, QT, TCU, STP-pci, etc), VME control processors, linux-based PCs (e.g. startrg.starp.bnl.gov, l2ana01.trg.bnl.local, trgscratch.starp.bnl.gov, scaler48.scaler.bnl.local, etc.) and code. The trigger system can communicate through TCD (Trigger Clock Distribution) boards with any of the active STAR detectors (BSMD, BTOW, ESMD, ETOW, FMS, GMT, IST, MTD, PXL, SST, TOF, TPC, VPD). It relies on connections to SLOW CONTROLS for hardware monitoring and to DAQ for non-trigger-detector data taking. It relies on connections to RHIC for beam related data, but can run by itself for cosmic ray events or for susbsystem testing.

Selecting Detectors

You first decide whether you are setting up a trigger for a physics event or for a special event, like a TPC laser event. Special events are taken in priority order and involve directing a trigger command to only one detector subsystem for any event. Physics events always take priority over special triggers. First we discuss physics events, then special triggers.

To define a trigger To run the trigger, you must decide which detectors are involved (see right-most section of the Run Control pallette). You have to specify what action to take for each detector, whether to include it in the event or not. The default is that any detector not explicitly excluded will be included when it is alive. The selections define the action word, basically a sequence of bits that DAQ uses to decide which detectors to read for this event type.

Trigger Selection bits

You then have to specify what you want from the trigger detectors. This is presented as a sequence of bits to the TCU. You define a trigger by specifying the bit pattern or by selecting a predefined trigger type. The bit patterns are established at configuration time, and, for the coming run will be relatively stable. We use multiple bit sets for the current run.

The TCU input bits define the physics input, and the LIVE BITS define the state of STAR. These 20 bits define the trigger word. When you select the trigger type, you specify what detectors you want, what TCU bits you want, and what detectors you don't care about. Then, if the detectors are alive, an event of that specific type is formed and they are included in the DAQ readout of the event.