back to the main page

REMOTE FEB TEST INSTRUCTIONS (currently discarded: 12 Dec 2005)

Started on 28 November 2005 by H.Wilkens and J. Hoffman.
Contributions by: P. Gorbounov, E.Rauter... (insert your name here)
back to the main page

Starting the DCS application

Use one of Point 1 gateways (atlasgw01-gpn or atlasgw) to start a remote session as user lardcs :

-> ssh -X -C lardcs@atlasgw01-gpn
password: (ask Henric)
Host: (Windows)

Confirm (or ignore) the warning that a console session will be abandoned (you will take over the console PVSS session, if it was running). However, you may wish to check with the expert(s) that were working in USA15 before starting a remote session.

Log on DCS Windows XP, using the same username and the password.

If the PVSS is not running (no "Power Supply" panels are displayed), start the PVSS session (see how in Appendix A )

Turning on the power

  • on Vision-2 (small!) "Power Supply" panel: power on (left switch)
  • WAIT till voltage of 280V is displayed
  • open output relay (right switch ON)
  • check LV values

TIP Known problems

  • The main panel may happen to have displayed ELMB UNOPERATIONAL (in red, you cannot miss it). When it happens you don't see the values of low voltages and temperatures in the panel. But if the high voltage (280) and the current (5) are displayed you may try to open on PVSS console Windows User Interface (with Device Monitor) and press green bulb to turn it on. Then open +ELMB->expand CANBUS ELMB. If wanted crate is unoperational choose it and send RESET signal, or send RESET signal to all nodes. If it doesn't work, consult Henric and/or Stefan if it's ok to run (cycle power in canvas crate may be needed).

Starting the DAQ application:

Connect to DAQ machine via the Point 1 gateway (atlasgw01-gpn or atlasgw):

-> ssh -X -C lardaq@atlasgw01-gpn
password: (ask Henric)

STEP 1 - Hostname: pc-lar-mbl-ro-01 (or )

Accept the defaults at "STEP 2" and check "STEP 3" to be lardaq ! prompts; a linux session starts.
Launch Larg Online:

~ > playme&
Enter correct side (A or C) and the FT number, confirm. Then, on the right part of the "Graphical User Interface"
  • select "InstallationPanel" tab and use the control buttons in that tab (note the comments in the right part of the tab).
    • activate button "boot/ LED order", WAIT till it turns green
    • activate button "Configuration", WAIT till it turns green
    • activate button "DCU check", WAIT till it turns green. If it gets red , then inspect carefully red error messages at the bottom (log) part of the panel. Ignore the voltage alarm if the value is within the range +/- 1V (?) from the nominal. However, take the DCU temperature alarms very seriously, especially the values above 39 degrees - report to experts before continuing! . Note down the anomalies in runs_log.txt .
    • Locate the current time position on the 'gnuplot' temperature graph. Try to determine its trend and make sure that all FEBs are represented. This is particularly important if the DAQ was not running for a while and there was a gap in the trends. Never rely on the 'continuity' of the trends (cooling may fail anytime), therefore watch continuously the temperature graph during the first 10-15 minutes of the data taking!
    • Run the Pedestal check. It consists of a series of 3 short runs of ~1000 events (one run per gain).
    • Run the Calibration check. It consists of 3 ramp runs of 12800 events and one delay run of 38400 events. Watch the trigger rate, it should be about 39-40 Hz (otherwise, report to experts).
    • Do not run TriggerSum tests!
ALERT! It is very important to watch the FEB temperature graphs. Call experts, should the temperature fail to stabiaking and turn the LV off if the temperature starts increasing towards 40 degrees.

  • Write down the run numbers and all observations in ~/Desktop/runs_log.txt

  • Do a quick data analysis, note down the observations in runs_lof.txt :
~ > sod
~ > cd /data/MobileDaq/root
~ > nice -n20 plots <run-number>
where the run-number is the first run number of the pedestal/ramp series, or the delay run number.
nice must be used if the analysis is run in parallel with the data taking.

---++ Possible problems during data taking and remedies... ...suggested by Stefan Simion:

  1. red error message _"Unable to read size of evt fragment".
    Remedy: start over ("stop" on Run Control, "shutdown", "exit", then playme& etc. )
  2. no error message, but event counter is stuck at a multiple of 100 (TBM problem) or
    no error message, but event counter is stuck at and arbitrary number. Remedy: call experts, otherwise
    • print out the TTCrx status, attach the hardcopy to runs_log.txt and start over:
~ > ssh vmelar03
~ > sod
~ > cd ~/.FEB_Jun2005/feb2001
~ > read-TTCrx-status -M1 -c -e
~ > read-TTCrx-status -M2 -c -e

Shutting Down the system

  • DAQ: on "Graphical User Interface"
    • in run control section hit 'unload'
    • in DAQ supervisor section hit 'shutdown' confirm
    • hit 'exit' to exit the interface

  • DCS: on "Power Supply" panel
    1. FIRST, close output relay (right switch to "0")
    2. THEN, power off (left switch to "0")

TIP Hints and tricks

  • Be patient with DAQ/Run controls: DAQ and run state transitions may take a while (up to 2 minutes).

The "end of commented-out section".

Edit | Attach | Watch | Print version | History: r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r1 - 2005-12-11 - PetrGorbounovSecondary
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Sandbox All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2020 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback