Attendees:

Action Items:

DescriptionDue dateAssigneeTask appears on
  • Require concept for toggling power on/off to small motors (CosyLab contacted)
MEC-U Collective Team Meetings
  • Seeking existing SLAC solution for closed loop pointing stability loop (vision/motor)
MEC-U Collective Team Meetings
  • Require concept for toggling power on/off to GigE cameras (CosyLab contacted)
MEC-U Collective Team Meetings
  • Omar Quijano To work on documenting SLAC's IT infrastructure design including hardware specs and key information.
Omar Quijano2023-02-21 MEC-U Project Controls and Data Systems Team Meetings
  • Terzi, Ken To adjust some items on the presentation then send out to the group and Omar. Mitchell Cabral Will upload this to the Sharepoint.
Terzi, Ken2023-02-21 MEC-U Project Controls and Data Systems Team Meetings
Barry FishlerSet Point API WG pt.2
  • Jing Yin TPR Card Specification Stub (Specific question from Ken: Number of outputs for TTL? LVTTL or TTL? Both?)
Jing Yin2023-02-07 MEC-U Project Controls and Data Systems Team Meetings
Tyler Johnson2022-11-14 MEC-U Project Controls and Data Systems Team Meetings
  • SLAC to provide GUI standards for laser systems (standards in development.). Follow up. Mitchell Cabral 
Mitchell Cabral2022-11-03 MECU LLNL/SLAC Meeting
  • LLNL to share C++ standards for SLAC review. Follow up. Vinod Gopalan
Vinod Gopalan2022-11-03 MECU LLNL/SLAC Meeting

Topics:


Agenda Item

DAQ

Minutes:

DAQ

Retention:

  • Disc:
    • immediately available
    • retained for 4 months
      • Cases can be made to retain longer
  • Tape:
    • takes 1-2 days to be made available
    • retained for 10 years
  • Collection rates can be specified by the DAQ.
  • Only diagnostic cameras as of now. No GigE cameras yet.  

Laser System 

  • Cameras:
    • triggered during experiment.
    • recorded during experiment. 
  • Operation
    • May need data from ramp up/down.
    • Need timestamping
      • To be meticulous: Trigger then readback
      • Rough estimate: Just readback 
  • LLNL would like a DAQ for their system for testing. 
    • What does the DAQ provide and what does LLNL need to provide in order to keep the data structure. 
  • EPICS collection vs DAQ collection
    • EPICS is great until about 120 Hz timestamp.
    • DAQ for LCLS2 is built to be able to handle 1MHz timestamping. 
      • Reproduce everything that happened, shot by shot.
      • Not recording every single event. 
      • EPICS doesn't play nice with this.
        • Doesn't readout the same way, doesn't like timestamping this way
      • IOC timestamp data via TPR
      • Need to start the DAQ then press record. 
      • DAQ stores EPICS images. 
        • using PV access. 
        • Question from Jana: Should DAQ be ingesting the EPICS information and storing it along side the science information or create a new category of diagnostic data for laser systems. 
      • EPICS archiving engine is totally unrelated to DAQ
        • Limitations of EPICS achiever that would push to DAQ 
          • All the acruchema of the DAQ
          • DAQ integration with e-log and analysis system
          • Operationally the laser systems should be able to use the DAQ subsystem.
          • Something would need to be build to grab the info from the EPICS achiever to the e-log. 
          • DAQ can take in a lot of data and do a lot of computing but this introduces a lot of complexity.
          • DRP in the DAQ allows us to dial back the data rate, choose events.
        • There is a new higher-performance EPICS archiver (what is the name?)
    • The current DAQ will work if we avoid from downloading too many files. 

DAQ Tools

  • Small data tools:
    • Is a current software to process data files.
    • Can store image as XTCs and have tiff conversions on demand. 
    • Currently we create one tiff file per image as well as an hdf5. Capturing this all day for a 12 hour shift at 10hz would break the file system. 
      • Potential to break file system if we blindly create tiffs per image. 
  • AMI (A Monitoring Interface)
    • SLAC in-house tool.
    • AMI is written to be much simpler. 
    • AMI is data visualization.
    • Main real-time DAQ to give you a live-view. 
    • LCLS2 DAQ seems like too much, too big.


  • DAQ Supported Device List
    • CamLink Cameras
    • SLAC built cameras
    • If there's an epics PV then the DAQ will pull from it. 


  • LLE Laser 
    • The DAQ is currently used to get data from the current MEC Long Pulse.
    • The DAQ is always recording, but every time it gets a trigger it records it. 


Time frame to cycle back:

  • Jana is away for the next three weeks. 
  • Barry Fishler Write down requirements for Data Archiver System. Jana will be back Sept 12th  
  • Pull Chris O'Grady, Dan, Murali, Matt Weaver on possible EPICS/DAQ solutions. 
  • Contact Amedeo for IT Infrastructure build for MEC-U Mitchell Cabral 

Action Items:





  • No labels