U.S. ATLAS HL-LHC Upgrade Level 2 and Deputies-NSF only Management Mailing List

[David W Miller <davemilr AT uchicago.edu>]

Hi all,

Here is the draft response for WBS 6.5 (Tile). I’ve basically stolen the second paragraph from John’s report.

Cheers, 

Mark & David

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

Narrative: 

WBS 6.5 covers the pre-production and production of the MainBoard (MB) system, the Low-Voltage Power Supply (LVPS) system and its housing (LVBox), as well as the monitoring mother board (ELMB-MB) for the hadronic Tile Calorimeter. Delays of the start of the MREFC funding will prevent the construction of the pre-production and production systems despite the plans and design for these systems being already complete or nearly complete. 

Assuming a 1-year delay in the beginning of the MREFC funding, it would be essential for WBS 6.5 (Tile) to avoid losing the key technical staff required to make the upgrade project successful, since many (most) of those staff would not be replaceable.  While technicians could, in some cases (but not all), be assigned to other projects at the relevant universities, this is not true in essentially all cases for the engineers.  Therefore, as an estimate of the standing army cost for the 12-month period starting April 1, 2020, we take 100% of the engineering manpower planned for that period of the MREFC ($375k) plus 50% of the technician manpower planned for the same period (~$48k), for a total cost to maintain the key technical manpower for the TileCal L2 subsystem of ~$425k.

Specifically, delays to the pre-production of version 4 of the MB (MBv4) will delay the integration tests planned for the non-US Scope system that interfaces to the MB, the so-called Daughter Board (DB). Version 6 of the DB (DBv6) is planned for testing with the 100 pre-production MBv4 boards in Spring/Summer. One more full integration of DB+MB is required before production is launched. Therefore, a 1 year delay to pre-production of MBv4 schedule will also directly delay the DBv6 schedule. 

Similarly, the production of the required number of LVPS boxes relies on an international partner (Prague) to produce half of the components. Delays to the US schedule will render the project unable to meet its international obligations. 

In addition, delays in ordering components may find that some have reached end of life and therefore replacements will need to be found. This incurrs 1 month of EE personpower to redesign the circuit and additional time (8 weeks) and cost (~$100k) for radiation certification.

In the case of WBS 6.5.1, two senior people will retire in CY2023, including the person with the most experience debugging the system. Emeritus status of the 6.5.1.1 CAM will eliminate base support of scientific personnel, so a cost of approximately $200k will be realized in replacing the scientific personnel with engineers.

Milestones:

30-Jul-2020: ELMB1198EX -- EXTERNAL MILESTONE -- ELMB PRR

25-Sep-2020: LVP523052EX -- EXTERNAL MILESTONE -- ATLAS Tile LVPS PRR

23-Dec-2020: MAB1360EX -- EXTERNAL MILESTONE -- ATLAS Tile Main Board PRR

29-Apr-2021: LVP540140EX -- EXTERNAL MILESTONE -- ATLAS LV Box PRR (CERN)

Each of these ATLAS production readiness reviews (PRR) will be delayed.

Costs:

The attached charts show the costs expected in each of the years of the MREFC. To estimate the cost to WBS 6.5, we have taken the full engineering FY20 cost (MREFC work) + half of FY21 (since the delay would be from April 2020 — April 2021). We have taken half of the technician cost for this same period.

The Engineering cost is ~$375k (~2.4 FTE), the technician cost is $47k (~0.5 FTE). 

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

On 12 Dec, 2019, at 15:23, John Parsons <parsons AT nevis.columbia.edu> wrote:

Hi all,

Here is some draft text for LAr.  Comments are welcome.
Regards,
John


Assuming a 1-year delay in the beginning of the MREFC funding, it would be essential for LAr to avoid losing the key technical staff required to make the upgrade project successful, since many (most) of those staff would not be replaceable.  While technicians could, in some cases (but not all), be assigned to other projects at the relevant universities, this is not true in essentially all cases for the engineers.  Therefore, as an estimate of the standing army cost for the 12-month period starting April 1, 2020, we take 100% of the engineering manpower planned for that period of the MREFC (~$1.56M) plus 50% of the technician manpower planned for the same period (~$0.16M), for a total cost to maintain the key technical manpower for the LA2 L2 susbystem of ~$1.72M.
Should the delay in the MREFC funding approval preclude work on the MREFC tasks during the (assumed) 12-month delay, the result would propagate into a 1-year delay on the entire international LAr schedule. For example, the custom ADC development in WBS 6.4.1 at Columbia+UT Austin would pause after the current Pre-prototype 3 stage and subsequent ATLAS Preliminary Design Review (PDR), delaying the Prototype stage and subsequent ATLAS Final Design Review (FDR) by 12-months. Since the ADC is on the critical path for the FEB2 board being developed at Columbia in WBS 6.4.2, the 1-year ADC delay would propagate into the FEB2 schedule and carry through until the eventual FEB2 production and delivery to CERN.  The FEB2 delay would also delay system tests required as part of the approval process for other frontend deliverables from our international partners, such at the Calibration board, power supplies, LATS Timing/Control board, etc. Similarly, a 12-month delay is producing and testing the v1 SRTM board at SBU in 6.4.3, would propagate throughout the international LAr schedule for the off-detector electronics development, production, and delivery to CERN.

On 12/12/19 4:02 PM, John Parsons wrote:

OK, I will take FY21 + 0.5*FY21 as some ballpark estimate, assume we need ALL the engineering manpower and 50% of the technician manpower, and nothing else (ie. no M&S or travel or ...).  That will give at least some approximation of what it would take to not lose the people.
On 12/12/19 3:58 PM, David W. Miller wrote:

Good point, I’d forgotten that. Indeed, should take the full FY20 number then indeed.

Thanks!
David

On Dec 12, 2019, at 2:57 PM, Stephanie Majewski <smajewsk AT uoregon.edu <mailto:smajewsk AT uoregon.edu>> wrote:

Hi David,

I thought in the cost books FY20 only included MREFC...? So no need to take the 1/2.

Stephanie

On Thu, Dec 12, 2019 at 12:53 PM David W. Miller <David.W.Miller AT uchicago.edu <mailto:David.W.Miller AT uchicago.edu>> wrote:

    Hi John,

    What I did (in a private mail sent to Mark Oreglia for comment)
    was just to average FY20 and FY21 since Apr2020—Mar2021 spans half
    of each.

    Cheers,
    David

    > On Dec 12, 2019, at 2:44 PM, John Parsons
    <parsons AT nevis.columbia.edu <mailto:parsons AT nevis.columbia.edu>>
    wrote:
    >
    >
    > Hi all,
    >
    >       I was looking back at the documentation from the time of
    FDR, such as the L2 Cost Book.   However, everything is rolled up
    according to FY, so I don't see a simple way with those docs to
    get the costs for Apr. 1/2020 - Mar. 31/2021.
    >       Mike/Jon, will you be able to get that from the Cost
    Controls people?
    >
    >               John
    >
    > On 12/12/19 1:14 PM, Michael Tuts wrote:
    >> Hi NSF scope L2 and deputy,
    >> Despite knowing the LS3 would be delayed by a year, NSF is now
    asking on a very short timescale for an assessment of the impact.
    There is one question posed where we can use your input on the
    impact of a 1 year delay of MREFC funding.
    >> “Please also comment on the impacts, by Level 2 WBS, of the
    impacts if there were to be a one-year delay in the start of MREFC
    funding. Include the risks or standing army costs of engineer and
    technician labor, or other negative impacts of NSF funding delay.
    Im assuming the risk of losing key technical staff in your
    engineering pool that would otherwise move to other projects at
    the various universities is a big factor.”
    >> Can you please provide an initial assessment for your L2 asap –
    note we are requested to provide an answer by 9am tomorrow (Friday)!
    >> Thanks, PO
    >> *From:* Coles, Mark W. <mcoles AT nsf.gov <mailto:mcoles AT nsf.gov>>
    >> *Sent:* Thursday, December 12, 2019 12:23 PM
    >> *To:* Michael Tuts <tuts AT pmtuts.net <mailto:tuts AT pmtuts.net>>;
    Anders Ryd <anders.ryd AT cornell.edu <mailto:anders.ryd AT cornell.edu>>
    >> *Cc:* Gonzalez, Saul <sgonzale AT nsf.gov
    <mailto:sgonzale AT nsf.gov>>; Caldwell, C. Denise <dcaldwel AT nsf.gov
    <mailto:dcaldwel AT nsf.gov>>
    >> *Subject:* VERY IMPORTANT
    >> *Importance:* High
    >> Dear Mike and Anders
    >> Would each of you please send me, by 9AM Friday morning, your
    immediate assessment of the impacts on the ATLAS or CMS MREFC
    program of the CERN schedule delay? Please provide a summary, by
    level 2 WBS element, of what you believe this impacts to be, or if
    there are none.
    >> Please also comment on the impacts, by Level 2 WBS, of the
    impacts if there were to be a one-year delay in the start of MREFC
    funding. Include the risks or standing army costs of engineer and
    technician labor, or other negative impacts of NSF funding delay.
    Im assuming the risk of losing key technical staff in your
    engineering pool that would otherwise move to other projects at
    the various universities is a big factor.
    >> Please provide a summary description of the process that ATLAS
    and CMS, at the international level, will implement (or that you
    anticipate they will implement) to realign their construction
    programs, and any knock-on effects on NSF.
    >> Please try to keep this at a concise summary level so that the
    whole response is about one page for each detector.
    >> I expect to get a thousand hand wringing questions from various
    high officials within NSF about this, as they hate surprises.
    Obvviously you are having to react quickly to new information, so
    it will be important to emphasize where the NSF program is
    decoupled from other agency activities or the international
    schedule, minimizing impacts, and where a delayed start by NSF
    would nevertheless be damaging to NSF-supported scientists and
    their opportunities to participate in the HL LHC program.
    >> Email if you have more questions. Our goal here at NSF is to
    put out a three page summary by COB Friday, one page from NSF
    internal evaluation, one page from ATLAS, and one from CMS.
    >> The delay information we’ve seen here at NSF was blamed on the
    need for a long detector fabrication schedule. What is driving it?
    >> Thanks
    >> Mark
    >> _______________________________________________
    >> Usatlas-hllhc-l2deputymgmt-nsf-l mailing list
    >> Usatlas-hllhc-l2deputymgmt-nsf-l AT lists.bnl.gov
    <mailto:Usatlas-hllhc-l2deputymgmt-nsf-l AT lists.bnl.gov>
    >>
   https://urldefense.com/v3/__https://lists.bnl.gov/mailman/listinfo/usatlas-hllhc-l2deputymgmt-nsf-l__;!!C5qS4YX3!Xev7P0lg11r1uuKZqpmcW-3ngbM0I423yHYW1d8Xc-gz37u_ym5WF8_friwwxsRb8T8$

    >
    > --
    >
   ______________________________________________________________________
    >
    > John Parsons
    > Nevis Labs,           Email: parsons AT nevis.columbia.edu
    <mailto:parsons AT nevis.columbia.edu>
    > Columbia University   Phone: (914) 591-2820
    > P.O. Box 137          Fax: (914) 591-8120
    > Irvington, NY 10533   WWW:
   https://urldefense.proofpoint.com/v2/url?u=http-3A__www.nevis.columbia.edu_-7Eparsons_&d=DwID-g&c=aTOVZmpUfPKZuaG9NO7J7Mh6imZbfhL47t9CpZ-pCOw&r=Z0wuTxQYnTTMlYma4lGZlO3skeCGVBrGGurYkWbhJsJuG5HaY82rmCNSdFOV7C4Z&m=iGs3hVDY7I8x8431V5EUhE-U8067DiiBCuW0wQcPfNA&s=IVwIexqfUeKKouYh65z7vIsZw2E6e_D-oxa6lK9qwlc&e=

    >
    >
   ______________________________________________________________________
    > _______________________________________________
    > Usatlas-hllhc-l2deputymgmt-nsf-l mailing list
    > Usatlas-hllhc-l2deputymgmt-nsf-l AT lists.bnl.gov
    <mailto:Usatlas-hllhc-l2deputymgmt-nsf-l AT lists.bnl.gov>
    >
   https://urldefense.com/v3/__https://lists.bnl.gov/mailman/listinfo/usatlas-hllhc-l2deputymgmt-nsf-l__;!!C5qS4YX3!Xev7P0lg11r1uuKZqpmcW-3ngbM0I423yHYW1d8Xc-gz37u_ym5WF8_friwwxsRb8T8$


    _______________________________________________
    Usatlas-hllhc-l2deputymgmt-nsf-l mailing list
    Usatlas-hllhc-l2deputymgmt-nsf-l AT lists.bnl.gov
    <mailto:Usatlas-hllhc-l2deputymgmt-nsf-l AT lists.bnl.gov>
   https://urldefense.com/v3/__https://lists.bnl.gov/mailman/listinfo/usatlas-hllhc-l2deputymgmt-nsf-l__;!!C5qS4YX3!Xev7P0lg11r1uuKZqpmcW-3ngbM0I423yHYW1d8Xc-gz37u_ym5WF8_friwwxsRb8T8$

--
______________________________________________________________________

John Parsons
Nevis Labs, Email: parsons AT nevis.columbia.edu
Columbia University Phone: (914) 591-2820
P.O. Box 137 Fax: (914) 591-8120
Irvington, NY 10533 WWW: http://www.nevis.columbia.edu/~parsons/

______________________________________________________________________