Beam Test Trigger and electronics
Triggers in Trigger Supervisor
On August 21, David, Elliott and Zisis discussed the desired triggers in the Trigger Supervisor (TS). TS can handle 10 triggers, the first four with 24-bit prescaling, the next four with 16-bits and the last two without prescaling. We agreed that having a separate north and south trigger for the BCAL was desirable, to allow for the study of events near the ends, where the far PMT may not have receive enough light to cross the threshold and generate a signal. As a result, we decided that the optimal trigger setup should look like (+ means OR, x means AND):
- Tagger Master OR (MOR)
- BCAL-R (right/north)
- BCAL-L (left/south)
- MOR x BCAL-R
- MOR x BCAL-L
- BCAL-R + BCAL-L = BCAL-OR
- BCAL-R x BCAL-L = BCAL-AND
- MOR x BCAL-OR
- MOR x BCAL-AND
- Veto or Pulser
Feel free to discuss this and bring forth your own ideas.
TDC Channel Arrangement
The F1 TDCs have two banks of 16 channels and we have three such modules. The optimal arrangement appears to be:
- 1st F1: 16 north in upper bank, 16 south BCAL PMTs in lower bank. As the BCAL has 18 PMTs on each end, it seem reasonable to exclude the two PMTs on each end that receive the lease amount of light and threfore are least likely to participate in a north or south trigger from the BCAL.
- 2nd F1: 10 outputs of the TS in upper bank, Four corner PMTs of BCAL, Veto and two cosmics in lower bank (7 channels).
- 3rd F1: Spare.
Electronics Cabling 25 July, 2006
Here is a diagram made by Chris Cuevas of the cabling for the BCAL test. This focuses mainly on what is needed in addition to the standard CLAS setup for the Data Acquistion.
The original PDF files sent out by Chris can be found here:
Electronics and Trigger
<*>preliminary, submit changes to Blake L .
David, Elliott, Simon and Zisis discussed electronics, trigger and data issues on Friday, July 14.
- Cables: the recent emails on cables were discussed. Form Richard we hear that there are 80ns, 170ns and 250ns cables available from RadPhi. Details of their connector ends and accessability (they are in cable trays) will be investigated by Elliott. We will attempt an analytical calculation on the cable lengths and unit delays to see whether we will need further cables to delay the ADCs or TDCs. This will also depend on precisely what delays we get from the Tagger. We might have access to connectors, etc, from the Test Lab. Elliott will post timing information to the Wiki soon.
- Electronics: the diagram below is obsolete. The LRS 2313 is now replaced by a Caen C207 (equivalent Leading Edge), the LRS1877 by F1 TDCs (see GlueX-doc-142) and the LRS1881 by Caen V792. Zisis will update the diagram. We will also eventually indicate cable types and lengths on this diagram. Zisis will investigate using Alberta CFDs and the dynamic range of the V792 versus the LeCroy 2249A (the latter shows limitations from the cosmics runs at Regina). For specs of the CAEN devices see the CAEN website. For convenience, the relevant manuals have been download to the GlueX Portal .
- Trigger: A brief discussion took place on the configuration of the CAEN C207 for the energy sum from the BCAL. Currently, we are planning to have the 2nd and 3rd layer PMTs in the trigger (ie 12 units) and decide to form a trigger based on this input. We have flexibility to interrogate other cells of the Module. We also have flexibility in using the Trigger Supervisor. Elliott and David have experience in using this.
- Tagger: There are standard procedures for activating the H-B Tagger. We may need to run through checks and plateau the counters. There are a programming manuals and scripts, and Franz and Eugene are the experts. Franz will provide us with a tutorial before the Beam Tests.
- BPM: It is a slow control (EPICS). We have plans to readout the scalers from the BPM (~every minute) and enter these in the data stream. We will also look into reading other EPICS information. David and Elliott both have familiarity with this.
- Spares: We seem to have adequate (~60) channels of HV. For other spares, we may use the Primex electronics, although we decided at this stage to identify components and leave them in their crates until needed.
- Data: Assuming 120 hours of beam (may be a bit optimistic), 4 bytes/word, we have ~320 bytes/event from the BCAL and another ~80 bytes/event from the Tagger (if we run it in 2us window), ie about 0.5 kbytes/event. With the Tagger at 1 MHz (ie 20 hits/event in20 us window) and a 1kHz trigger rate, this comes to a bit over 200 GB for the entire run. This can be handled either with the automated system of sending data to the Silo or even by hand. We will choose the simpler of the two.
- Analysis: David will work on getting ROOT histos in memory (for online checks) and with Regina to get the offline analysis working.
Drawing from Beam Time Request