[Dsg-hallc_nps] [EXTERNAL] Re: DSG NPS slides

[Mark Jones]

Aaron,
       It ends up that THia cancelled tomorrow's meeting. So it will  probably be next week,
 when I will present. If you have updates to the slides regarding Brad's comments,
then you can send updated slides before next Tuesday

Cheers,
          Mark
________________________________
From: Brad Sawatzky <brads at jlab.org>
Sent: Monday, October 31, 2022 3:17 PM
To: Aaron Brown <ambrown at jlab.org>
Cc: Mark Jones <jones at jlab.org>; Patrizia Rossi <rossi at jlab.org>; Dsg-hallc_nps at jlab.org <dsg-hallc_nps at jlab.org>
Subject: [EXTERNAL] Re: [Dsg-hallc_nps] DSG NPS slides

Hi Aaron,

Just a few comments/corrections.

The LED and chiller/thermal systems are *way* behind schedule.  I just
want to reiterate that this needs to be an immediate priority (ie.
target of 2 weeks + 1 week contingency at worst, not a best-effort
4+ weeks or 'unknown' as noted in the slides).

LED boards were delivered to the DSG back in April with the plan that
the controls would be ready by July/August.  Chiller controls were on
a similar timeframe.  We need to test the bleaching function of the LED
system as well, so the both the chiller and thermal readbacks will be
needed soon.

The DSG is also needed to support the chiller controls hook-up, LED
controls cabling, and connection to the thermal/humidity readbacks and
associated troubleshooting.  This work will happen in EEL/108 and should
be added to your list/ schedule.  NOTE: at this point, such work will
interrupt ongoing detector checkout, and will require coordination.

Do you have all of the cables, connectors, power supplies, etc needed
for those systems?  The 5V supply required for the humidity sensors (and
added cabling + power breakout box) was a surprise a couple weeks back.
I know the missing cable and breakout connectors for that power rail
were ordered a few weeks ago.  What is the plan for the 5V power supply
itself?  Installing these items will require modifications to the NPS
frame and routing cables internally -- that should be added to the
list/schedule.


I certainly do understand that the DSG shifted focus from the initial
timeline when NPS was busy working the noise issue, but I am *very*
concerned that the combination of Thanksgiving at the end of November
and the inevitable inefficiencies in December mean a near-term delay
will result in nothing being delivered until mid-January.

NPS installation is approaching very rapidly.  This needs to be an
*immediate* and direct priority if we want to actually meet the
Experimental schedule and avoid chaotic last-minute scrambling.


I do understand that we are all oversubscribed.  So, if the above
timeframe is not possible due to EIC, MOLLER, CLEO/SOLID, tasks (etc)
that are taking precedence over the upcoming experimental program, then
we need to know that ASAP so we have a chance to respond and/or realign
priorities.


On slide 2:
- The DSG will not be working under the production accounts.
  - Do not wait for access to a particular account.  Please add
    start-up instructions for all systems on the wiki here:
    https://wiki.jlab.org/cuawiki/index.php/NPS_Controls
    - For example, the /home/cdsg/phoebus-4.6.10-SNAPSHOT/phoebus.sh
      'HV' script noted on the wiki (or an equivalent, as noted in last
      week's email thread) should be adapted to point at relevant paths
      for the Phobeus install and the Config files for a given
      application.
      - The tarball 'phoebus_screens.tar.gz' I received from Brian
    - That script can then be trivially adapted to run wherever it needs
      to run within the Hall C cluster.
  - Similar 'launch' scripts will be needed for each Phoebus application
    you note.


On slide 3:
- The VME64x crate shown does not have the LED driver modules noted.
  - The two modules there are intel SBCs that were just being stored in
    that crate a long time back.
  - Wrong picture?


The associated Phoebus config files for all systems really need to be
version controlled and made available on Github by the DSG.
  - We need the backup/reference that will automatically provide
  - We will also need to track and merge changes made by both the DSG
    and Physics people on their respective systems -- something git
    is designed to manage well.

Startup/Restart instructions need to be documented for all DSG supported
systems.  (Add them to the above wiki link, for example.)

EPICS PVs need to be added to the Archiver and verified.  (Please give
them good names and put them under 'NPS' or 'HALLC/NPS' in the Archiver
hierarchy.)

While I'm running through things in my head, I am reminded of the
required alarm handlers for all of these systems.  We've discussed
trip points, visual displays, etc in the past, but I'd like some
clarifications on the low-level implementation.

Are these alarms (including an audible alarm) integrated into the
respective Phoebus interface?  Do they rely on BEAST (commonly part
of the CSS/Phoebus setup)?  Do the relevant PVs have the appropriate
alarm-handler fields attached?


-- Brad


On Mon, 31 Oct 2022, Aaron Brown via Dsg-hallc_nps wrote:

> Hi Mark,
> Here are a few slides detailing the work DSG is doing for NPS.
> Let me know if there's anything else you need.
> Thanks,
> Aaron


--
Brad Sawatzky (he/him), PhD <brads at jlab.org> -<>- Jefferson Lab/SciComp/F272
Ph: 757-269-5947  -<>-  Fax: 757-269-5235  -<>- Pager: brads-page at jlab.org
The most exciting phrase to hear in science, the one that heralds new
  discoveries, is not "Eureka!" but "That's funny..."   -- Isaac Asimov
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/dsg-hallc_nps/attachments/20221031/7940f435/attachment-0001.html>