-- PierreAlainLoizeau - 16 Apr 2009

This page is mean as a user guide for the different daqs used during the test Beam at GSI in April 09 for the Heidelberg+Chinese setup

DAQ list and characteristics

Every daq is located in the ~/nxm/hdrpc/April09 folder from the RIO prompt.
  • "rpc":
    • Located in the "rpc folder
    • Version of the MBS with only RPCs and VULOM: tacquila/Sam3 + Vulom3 + v775a/v792a
    • This is the third main DAQ, which should be used in case of a full SSD crash

  • "all detectors, raw SSD"
    • Located in the "all_raw" folder
    • Version of the MBS with RPCs, VULOM and the 4 single side silicon detectors (P side): tacquila/Sam3 + DL702 + Vulom3 + v775a/v792a
    • The SSD are read in raw mode => no zero-supp => size of events is the biggest => ~94k events for 1.3 GO
  • "all detectors, silitune (SSD tuning test)"
    • Located in the "all_silitune" folder
    • Version of the MBS with RPCs, VULOM and the 4 single side silicon detectors (P side): tacquila/Sam3 + DL702 + Vulom3 + v775a/v792a
    • The SSD are read in raw mode and their frontend timings are changed every 30k events, 10 values are sweeped
      • => 300k events in total = 1 file of ~ 4 GO or maybe 2-3 files
      • Should be operated only by me (Pierre-Alain)
      • The file HAS to be started BEFORE the acquisition is launched to get all 30k events
  • "all detectors, 0-supp SSD"
    • Located in the "all_supp" folder
    • Version of the MBS with RPCs, VULOM and the 4 single side silicon detectors (P side): tacquila/Sam3 + DL702 + Vulom3 + v775a/v792a
    • The SSD are read in the hardware zero-supp mode, so the size of the events should shrink quite a lot in comparison with the "all detectors, raw SSD" version
      • => Expected ~300K to 400k events per 1 GO
      • This is the secondary main DAQ in case the SSD vme board 2 crash

  • "all detectors, raw SSD, 2 DL702 boards"
    • Located in the "2dl702_raw" folder
    • Version of the MBS with RPCs, VULOM and the full silicon setup (4 P+ 3 N sides): tacquila/Sam3 + DL702#1/DL702#2 + Vulom3 + v775a/v792a
    • The SSD are read in raw mode => no zero-supp => size of events is the biggest + 7 SSD => ~50k events for 1.2 GO
  • "all detectors, silitune (SSD tuning test, 2 DL702 boards
    • Located in the "2dl702_silitune" folder
    • Version of the MBS with RPCs, VULOM and the full silicon setup (4 P+ 3 N sides): tacquila/Sam3 + DL702#1/DL702#2 + Vulom3 + v775a/v792a
    • The SSD are read in raw mode and their frontend timings are changed every 30k events, 10 values are sweeped
      • => 300k events in total = 1 file of ~ 8 GO!! or more probably 5-10 files
      • Should be operated only by me (Pierre-Alain)
      • The file HAS to be started BEFORE the acquisition is launched to get all 30k events
  • "all detectors, 0-supp SSD, 2 DL702 boards"
    • Located in the "2dl702_supp" folder
    • Version of the MBS with RPCs, VULOM and the full silicon setup (4 P+ 3 N sides): tacquila/Sam3 + DL702#1/DL702#2 + Vulom3 + v775a/v792a
    • The SSD are read in the hardware zero-supp mode, so the size of the events should shrink quite a lot in comparison with the "all detectors, raw SSD, 2 DL702 boards" version
      • => Expected ~150K to 200k events per 1 GO
      • This is the main DAQ that should be used if no silicon crash happens.

Operating the MBS DAQ system

  1. Go to the HD-RPC computer and login as fopiuser if it's not already the case
  2. Launch 4 RIO windows (our RIO is r3-42), either via the telnet r3-42 command or via the out r3-42 command. In the first case you will need to login as "fopi" with the normal fopi password. In the second case the login is automatic.
  3. Prepare your windows on the screen so that you have 1 windows on one side and 3 on the other (see picture for an example)
  4. In the big console, enter cd nxm/hdrpc/April09/__the_daq_you_want_to_run___ with "__the_daq_you_want_to_run___" replaced by one of the folders listed in previous part.
  5. In the upper small console, enter cd nxm/hdrpc/dsp, and then fopideb 4. You can then resize this console to the size of the printout if you want. Let's call this windows 'MAIN'.
  6. In one of the two remaining console, enter cd nxm/hdrpc/dsp. Let's call this windows 'A'.
  7. In the last console, enter cd nxm/hdrpc/triglog. Let's call this windows 'B'.
  8. In 'MAIN', enter resl to reset the daq system and the mbs to start it again.
  9. In 'MAIN',on the MBS prompt, enter @startup and wait until all message appear.
  10. In 'A', enter startrpc and wait until the prompt come back.
  11. In 'MAIN' :
    1. If you want to take all events from the first one, enter open file ./data/__filename__.lmd -disk size=__wanted_Size__ , replacing "__filename__" by the name of your file (usually voltageIngoRpc_VoltageSSD1_VoltageSSD2_VoltageSSD3_VoltageSSD4_VoltageChineseRPC_Date_Time) and "__wanted_Size__" by the maximal size in MO you want for your raw data file. This will start you first data file.
      This is the advised way of doing in case of high rate (~100-200 Hz).
    2. If you don't care of the first event, enter directly sta acq to start the acquisition. Then wait until the MBS prompt come back before going on with the next step.
  12. In 'MAIN' : (letters correspond to previous step)
    1. Once the MBS prompt is back, enter sta acq to start the acquisition.
    2. Enter open file ./data/__filename__.lmd -disk size=__wanted_Size__ , replacing "__filename__" by the name of your file (usually voltageIngoRpc_VoltageSSD1_VoltageSSD2_VoltageSSD3_VoltageSSD4_VoltageChineseRPC_Date_Time) and "__wanted_Size__" by the maximal size in MO you want for your raw data file. This will start you first data file.
      This step can be a bit tricky because of the scalers slow control output which flush the screen quite quickly in high rate case.
  13. In 'B', enter triglog -f=triglog.txt to reload the trigger setup in the trigger box board.
  14. In 'A', enter rate to see get an actualized printout of the last second rate/file status/event number status.
  15. If you want to close a file before its end (or if you started a file without size limit), just enter close file in 'MAIN'.
  16. To stop the acquisition, enter sto acq in 'MAIN'.
  17. If you want to change the MBS version, in 'MAIN', enter @shutdown, then once no message comes anymore (should be 5-7 lines), enter quit and go back to step 3 .

Special cases

  1. To recompile the daq program, just enter make after step 3. You should do this ONLY AFTER speaking with Norbert or me (Pierre-Alain).
  2. The source file is "f_user.c". Please ALWAYS speak with me (Pierre-Alain) before ANY modification!!!!
  3. To convert a "*.ntu" file (NTuple) to a "*.root" file (TTree), you can either use :
    1. The command h2root filename.ntu filename.root 1 1 16384 (replacing filename by the name of your file) => Convert a single file!!!
    2. The "ntuple_conv" script that you can find in "/lynx/Lynx/fopi/usr/fopi/nxm/hdrpc/April09". For this copy the file in the folder where your .ntu files are. Then just enter ./ntuple_conv and the conversion will start => Convert all .ntu file in a folder!!!
    3. If a message appear saying that "h2root" is a unknown program/command, enter ". rootlogin new" to reload current GSI root version.
Topic revision: r4 - 2009-12-15, ChristopherHuhn
 
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding GSI Wiki? Send feedback
Imprint (in German)
Privacy Policy (in German)