Project Status and Reviews

1
Project Status and Reviews

• Michael Hebert
• UC Irvine
• Feb 6, 2005
• MECO Collaboration Meeting

2
MREFC Program Status

• In October Congress approved 15M for RSVP in the
  FY05 NSF budget
• These funds will only be released if we have
  successfully baselined the experiment. The goal
  is to achieve that in late April, i.e. before the
  May National Science Board meeting.
• It now appears that passing this baseline review
  is a political necessity. To quote Jon Kotcher,
  there really is only one message we do
  whatever we must do to get this project baselined
  in the next few weeks - nothing else matters. If
  we don't make this, we indeed may have nothing.
  People should be heavily sensitized to this fact,
  and the situation these projects are in
  politically. Personnel must focus solely on this
  goal, putting aside arguments and considerations
  not directly related to a successful outcome
  here, if this is going to succeed. That's the
  long and the short of it. Do with it what you
  will, but the fate of RSVP is in the balance of
  the next few weeks.
• The FY06 NSF request will be known on Monday as
  part of the Presidents Budget Request to Congress

3
MREFC Total Budget

• RSVP Management has set the following cost
  targets (burdened FY05)
• KOPIO - 55M
• MECO magnet - 56M
• MECO - 24M
• AGS - 40M
• Project Office - 20M
• The total is 195M, to which is added 46M in
  RSVP Management Contingency for a grand total
  MREFC target cost of 241M.
• Exactly how these totals were arrived at is not
  clear, but the statement is that it was by
  scaling June 2001 costs for MECO and KOPIO by a
  common factor. Given that MECO presented costs of
  85M and KOPIO 75M (less 20M for foreign
  contribution) last month, there is a clear
  disconnect that we need to resolve.
• This also emphasizes that there has been a
  (poorly communicated) fundamental shift in the
  status of the magnets vis-à-vis the rest of MECO
  (next slide).

4
Magnet Management

• I have been informed that the MECO magnets are a
  fundamentally different object and as such the
  money for the magnets will be handled separately.
  
• Operationally this means that we cannot recoup
  savings in the magnets elsewhere in MECO, nor
  bolster the magnets if save money in our detector
  cost scrubbing efforts. Any savings we realize
  will revert into the RSVP contingency pool.
• Management of the magnets has also changed, but
  the details are vague. On one hand, Brad is still
  reporting to me as the subsystem manager of this
  MECO component. On the other hand, Brads
  guidance on procurement activities is coming from
  the RSVP Project Office.
• Brad and I are working together to understand
  what all of this means exactly. The only thing
  that gives me hope is that we been able to work
  well with Brad on everything to date and he
  understands that we need to be intimately
  involved in the magnets, regardless of what the
  org chart looks like.
• We may gain some additional insight late next
  week as there will be a meeting with BNL to look
  at the role SMD and C-AD can (and want to) play
  in the magnets.

5
Construction Schedule

• It now appears likely that RSVP will be cast as a
  five-year construction program, not six as one
  thought clearly an improvement
• Begin construction in Fall of this year,
• Wrap up installation and pre-flight efforts no
  later than the Fall of 2010
• Engineering running beginning in 2011
• This is consistent with our original MECO plan,
  however that plan envisioned 30M in the first
  year. We will need to determine the impact of
  receiving half of that in the near future. More
  on that in my next talk.

6
RD Money 1

• All pre-FY04 funds have been expended
• UCI has received spending authority via a
  subcontract with NYU for the 2M (less NYU
  support) MECO portion of the nominal 6M in FY05
  RRA money allotted to RSVP by Congress.
• There are a total of nine subcontracts possibly
  coming out of that money
• MIT 516k In place Magnet Subsystem Manager
  support, several small tests of prototype
  components aimed at reducing cost risk in the
  magnet procurement
• BNL 171k Stalled Nominally for beamline
  transport design, shielding layout, liaison staff
  support, detector installation and support
  structure designs, neutron absorber designs. We
  have been trying for four months to get BNLs
  Work for Others office to understand that this is
  a NEW contract and handle it as such. Recent
  KOPIO experience suggests that we reconsider how
  we handle this. KOPIO funds finally reached the
  Lab last week, only to withdrawn within days.
  This because the money was going to be redirected
  towards baselining efforts in C-AD in lieu of
  KOPIO engineering work. Our money will likely
  suffer a similar fate if we succeed in getting it
  in the door.
• Houston - 88k In Place Additional tracking
  chamber prototype construction and testing,
  chamber front-end electronics design and
  prototyping, Education/Outreach support

7
RD Money 2

• BU 91k Expected in the next week Conceptual
  design of Trigger and DAQ electronics boards and
  ASICs.
• Syracuse 20k Expected in the next week
  Transverse Tracker mechanical design and testing
• LBL 125k No SOW Yet Six months of an
  estimated one year effort to redesign the BaBar
  "Elefant" readout chip
• LLNL 40k On Hold Indefinitely This is for
  magnet contracting support. Although consultation
  with this group has been vastly more useful than
  similar discussions with BNL procurement, there
  remains the strong possibility that the latter
  will handle this procurement. Until we have a
  plan, there are no plans to go forward with work
  at LLNL.
• TBD 150k On Hold Indefinitely This is a
  contract for the conceptual design of the RF
  Modulated Magnet. We have not identified a vendor
  yet, so this one is on hold
• TBD 18k On Hold Indefinitely This is a
  contract for the conceptual design of the Heat
  and Radiation Shield. With the departure of Jon
  Hock from the Lab, this activity halted at BNL.
  The hope is that a new collaborator might take
  this on.

8
Supplemental RD Money

• RSVP has put in a proposal for additional
  emergency funding this year
• The only items included were RSVP PO and MECO
  Magnet RD, no other MECO tasks, nor anything for
  KOPIO or the AGS
• Note that the magnet RD task list differs
  significantly from our request RSVP Management
  prerogative were told

9
Recent Reviews

• October 04 MECO Magnet Review
• November 04 AGS Review
• January 05 RSVP Simulations Review
• January 05 BNL LOG Cost Review
• Upcoming Reviews
• Early April RSVP Internal Baselining Readiness
  Review (next talk)
• Late April NSF RSVP Baseline Review (next talk)
• TBD NSF RSVP Operations Cost Review - could be
  as early as April but we can hope reason prevails

10
Magnet Review

• MECO (primarily Brad) presented a very detailed
  exposition on the current state of the magnet
  design to a committee consisting mostly of ATLAS
  magnet experts
• Magnet layout
• Component masses
• Coil operating point summary
• Field spec and tolerances
• Conductor, joints, leads and buses
• Electrical insulation
• The report of the review committee was generally
  favorable. There were, however, a large number of
  recommendations, on both technical and management
  / procurement issues. The following slides
  summarize these points.
• One benefit of the review is that allowed us to
  include funds for magnet studies (the only MECO
  money that made the cut) in the supplemental
  proposal to NSF that appears likely to be funded
  soon.

• Power supplies
• Quench detection and protection
• Structural design criteria
• Cryogenic design
• Costs and Construction Schedule

11
Magnet Recommendations 1

• Reduce the number of coils to minimize number of
  connections / interfaces
• Conduction cooling of the PS coils to eliminate
  the need for 6700 liters of LHe
• Gaseous He cooling of the internal heat shields
  and warm-to-cold supports
• Simplified analog quench detection / protection
  system with far fewer channels
• There is clearly insufficient input for single
  vendor bid package. In our opinion there needs to
  be model coil input for a final engineering
  design. This is required in order to qualify the
  technology adopted for winding and bonding, the
  indirect cooling and thermal performance, and in
  addition to perform the operations and thereby
  train the core team for following up the
  production.
• Our rough estimate is that the overall project
  might take about a year more than that
  presented.
• Cost comments
• 10 profit appears optimistic it may be 15
• Labor rates look low (in particular for
  qualified welders)
• Cost is based on 04 prices. Should escalate to
  06 prices
• The most important comment is that we would
  expect the final price coming from a single
  vendor to be considerably higher than that
  quoted. This is primarily because certain
  features are either underestimated or omitted,
  e.g. 15 purchasing overhead appears low. Our
  experience is 20-25
• 0 risk included. Once again, our recent
  experience suggests 20-30

12
Magnet Recommendations 2

• In total, the real cost may be 25 higher when
  the single vendor option is chosen. The estimate
  of cost of system integration at BNL requires
  further study. Our proposal would be to get
  independent budget cost estimates based on the
  purchase of major components.
• We consider the risk involved in going to a
  single vendor to be too high. You may pay twice -
  now for risk, later for cost overruns (and have
  to do some of the work yourself anyway).
  Consequently we strongly advise to go for a
  Buy-in-Parts scenario. This may save 25 in
  cost, as well as increase the likelihood of
  getting a working magnet on time. This cost
  saving will be more than sufficient to finance
  the core magnet team required for production
  follow-up, trouble-shooting and assembly.
  Important side note, ATLAS got burned on a bad
  procurement which sensitized 4 of the 5 committee
  members to vendor reliability
• Coming back to the magnet itself, it would be
  helpful to get feedback from industry in
  particular on coil winding, segmentation and
  impregnation. With this in mind it would be
  useful to make a draft RFI as soon as possible
  (within 3 months, say) for the magnet, to be
  followed by a draft RFP/SOW to get some answers
  on cost, schedule and technical feasibility from
  a few companies within the next 6 to 9 months. In
  addition, and most important, this process should
  be exploited in the short term to permit visits
  to potential suppliers for technical discussions
  that will hopefully lead to improvements in the
  final engineering design.

13
Magnet Recommendations 3

• We advise that in order to reduce project risk
  the effort should now be concentrated on the
  magnet itself. The provision of immediate funding
  should enable the achievement of the following
  critical studies and preparatory work in the
  course of the next 6 months, which are essential
  in order to keep the magnet project on track
• Re-design of cooling of the PS, replacing bath by
  conduction cooling,
• Simplification of the magnet by reducing the
  number of coil segments,
• Make test windings to validate conductor, winding
  and insulation technologies,
• Do the industrial studies through the RFI, and
  prepare the RFP,
• Strengthen the team at MIT to speed up design
  work,
• Design and start construction of tooling for a
  full scale model test coil.
• The Reports conclusions
• We are confident that the conceptual design which
  has been developed is a sound basis to move
  forward to an RFP for the magnet.
• The groups that have carried out this work should
  be supported and strengthened.
• To this end we recommend sufficient intermediate
  funding, following a cost estimate, for the high
  priority studies listed above.
• Equipment that will have to be operated and
  maintained by BNL should be specified with their
  collaboration, and procured through BNL.
• These recommendations are made with the objective
  of reducing technical, financial and schedule
  risk, and minimizing the overall cost of the
  magnet.

14
AGS Review

• Technical and Cost Review in November
  presentations on
• Booster
• AGS
• Electrical
• Mechanical
• Switchyard
• The committee noted the significant cost growth
  although they acknowledged that much of it was
  due to costs originally covered in pre-ops.
  They nonetheless questioned the need for the full
  scope of the AGS upgrade effort
• The committee generally recommended addition cost
  scrubbing, resource loaded schedule development,
  and searching for major savings if possible
• They also made a specific recommendation to
  reduce beam development from five years to three
  in order to save 14M in support of operations
  personnel and power costs.

• KOPIO, MECO infrastructure
• Beam development
• Schedule Personnel
• Operations cost schedule
• Management Issues

15
Simulation Review Recommendations

• While MECO has done a good job addressing many
  issues via simulation, a clear inadequacy of the
  simulations is the lack of detector digitization
  and realistic event reconstruction. An increased
  level of manpower devoted to software and
  simulations is clearly needed. We recommend
  that, as soon as possible, MECO
• Develop a detector Monte Carlo with full
  digitization of detector signals, and realistic
  event reconstruction starting from raw data to
  validate background estimates.
• Perform tests with the AGS to demonstrate the
  required extinction when running at 8 GeV with
  intensity of at least 20 Tp/spill.
• Continue to develop more realistic simulations of
  the full target/transport channel/detector chain
  to further explore exotic background scenarios.
• In addition, the committee recommends that both
  KOPIO and MECO
• Develop deliverables-based schedules with
  performance milestones, which include the steps
  recommended in this report planned for completion
  as early as possible.
• Finally, the scenario described to us for AGS
  running during RHIC data-taking, while perhaps
  conservative, would provide only a marginal
  amount of RSVP running per year. We urge BNL to
  make it a high priority to maximize AGS slow
  spill running during RHIC operation. Also, both
  KOPIO and MECO should benefit if data-taking is
  structured to provide the longest possible runs
  in alternate years, rather than shorter runs
  based on an equal division of available running
  each year.

16
LOG Cost Review

• An attempt was made to understand RSVP costs at
  level 3 over the course of a three day review
  under the auspices of the Laboratory Oversight
  Group (LOG) for RSVP
• Each level 3 system (e.g. Tracker, Calorimeter,
  etc.) prepared a single sheet summary of their
  costs broken down by materials, labor,
  contingency, total in addition to a brief
  presentation that tended to be a technical
  presentation owing to the way the review charge
  was written.
• Given the difficulty of digesting this
  information with little material in advance
  (there were zero working days between the
  Simulations and LOG Reviews), it is not
  surprising that we did not get a lot of
  substantive feedback. The conclusions (verbal
  only) were
• The AGS and the MECO Magnets appear ready to
  baseline costs
• Contingencies appear to be systematically
  underestimated
• No technical show stoppers were noted by the
  committee, although they did endorse early beam
  testing to validate the MECO beam extinction and
  KOPIO neutral beam

17
MECO Cost Estimate (FY05 )
WBS Task Cost (M) with academics Cost (M) without academics Cont without academics
1.3 MECO 94.52 85.83 28.8
1.3.1 Extinction 2.62 2.28 43.1
1.3.2 Production Target Shield 2.83 2.76 40.9
1.3.3 Superconducting Solenoids 55.20 55.20 28.6
1.3.4 Muon Beamline 3.35 3.25 33.6
1.3.5 Straw Tracking Chamber 5.50 3.78 24.5
1.3.6 Electron Calorimeter 7.98 6.30 25.6
1.3.7 Cosmic Ray Shield 2.47 1.67 13.3
1.3.8 Data Acquisition and Trigger 3.38 2.47 25.4
1.3.9 Simulations and Analysis 4.00 0.93 46.0
1.3.10 Installation and Integration 3.07 3.07 27.6
1.3.11 Project Management 4.13 4.13 28.7
18
AGS Cost Cutting Measures

• The AGS cost target is 40M. The Jan 04
  estimate was 82M including upgrades to the AGS,
  Booster, Switchyard, and A and B Lines plus a
  five year beam development plan.
• Now, ALL beam development efforts have been
  eliminated
• C-AD questions whether we can exceed 20 Tp/s
  proton throughput, let alone reach 40 Tp/s due to
  ALARA limits. Losing these studies means we
  wont start trying to understand what that factor
  of two requires until 2008 when operating funds
  start
• The same statement applies to testing of
  extinction and extraction at 7.5 GeV/c
• Other changes
• Using existing shielding blocks for the concrete
  portion of the cosmic ray shielding. Measured
  activity rates are low so this looks like a
  painless change.
• Existing equipment removal will be on the DOEs
  dime, as it should be
• Reusing exists (ancient) power supply controls,
  which might lead to reliability problems but
  there is a large stock of working spares
  available
• Reuse existing vacuum pumps likely not a
  problem
• Reuse existing counting house (trailer) quality
  of life issue