MECO Magnet

1
MECO Magnet

• Outline
• Final design plan
• Fabrication plan
• Fabrication costs
• Cost analysis
• Reply to specific MECO questions (Jim Miller)
• Magnet management requests

2
MIT Final Design Plan

• MIT Final Design Plan
• Brad Smith, MIT-PSFC
• Repeat of talk given at
• Informal MECO Meeting at BNL
• December 13, 2004

3
MECO Magnet Final Design Plan
Summary Roll-up of Magnet Plan

• NOTES
• This plan is has been proposed by MIT, but does
  not yet have the concurrence of MECO.
• PS, TS and DS fabrication durations are estimates
  presented at the Oct04 review, and are for
  reference only (not part of this Final Design
  Plan).
• The MIT plan intends to include all activities
  needed to complete a final design of the magnet
  system, even though some of those activities may
  be completed elsewhere (e.g., BNL).

4
RD Task Overview
5
RD Task Detail Sub-scale winding tests

• Provides structural modeling data for winding
  tensile and compressive modulus
• Provides modeling data for other insulation
  concepts (GUG, prepreg, wet winding)
• Contributes to the final selection of the
  insulation concept

6
RD Task Detail Conductor trials and testing

• Validates conductor properties assumed in
  analytical models for
• SSC cable as is
• De-keystoned cable used in large-build TS coils
  near APS window
• Cable-in-channel configuration of most coils

7
RD Task Detail Fabricate and Test Copper Winding

• Cu winding validates our baseline design for
• Impregnation quality
• Winding thermal properties (assessed to be
  similar for other candidate insulationsGUG,
  prepreg, and wet winding)
• Winding will have cooling detail based on a
  conduction cooled concept that can be tested
• Tests will validate thermal performance of
  insulation and conduction cooling with
  thermo-siphon, possibly after high
  electromagnetic pulse load.
• When testing is complete, coil will be cut open
  and inspected
• MIT test facility can be used to facilitate
  testing of a first article TS coil during
  production phase

8
RD Task Detail Joint Trials

• Joints will provide baseline concepts for joint
  clamping and soldering fixtures
• Samples will be dissected and visually inspected

9
Design Activities Summary
(Detail is rolled up)
10
Winding Tooling Design

• This task will produce a reference design for the
  winding tooling.
• The task is responsive to the MOG recommendation
  to design the tooling.
• Winding tooling must be available early to
  support a timely start of coil winding during the
  fabrication phase, otherwise the magnet delivery
  is delayed.
• The goal is to facilitate but not overly restrict
  the procurement process.
• Some bidding vendors may have tooling already.
• Others may not, in which case the existence of a
  reference design would facilitate their bid which
  will contain a manufacturing plan

11
Program Management Includes RFI/RFP Processes
NOTE Although the PS, DS and TS packages are
produced separately, this does not preclude a all
awards going to a single vendor.
12
Budgetary Cost Overview
Budgetary Cost for Magnet Plan
13
Estimated Cost by Month
Funding with contingency Flow per month required
to support the plan
14
Estimated Cumulative Cost
Cumulative funding with contingency
15
Conclusions

• An integrated program plan has been presented
  that is both responsive to the MOG
  recommendations, and recognizes the phased
  funding.
• A short period is included for plan negotiation
  and agreement.
• The plan leads to an early RFI (6/13/05) and
  includes all work leading to and including RFPs
  for the magnet parts.
• Existing and requested supplemental funds will be
  consumed quickly if the plan dates are to be met.
• Additional funding from the FY05 RSVP
  Congressional authorization will need to be
  committed soon to avoid delays.
• Additional funds to support the final design plan
  will be needed from future (FY06) Congressional
  authorizations to complete the plan.
• The plan can be adjusted for alternative funding
  profiles.
• The plan currently lacks the concurrence of MECO

16
Fabrication/Installation Schedule (As presented
at Oct04 ReviewNew FD Plan not yet merged here)

• Final design review releases fabrication
• Some tooling may require early release
• All activities are assumed to be single shift
  except for winding
• Conductor fabrication sequence PS , TSu, DS, TSd

17
PS Fabrication

• Winding tooling
• May require early release
• Two winding machines
• Multiple impregnation molds so as to not
  bottleneck production

18
TSu Fabrication
19
TSd Fabrication
20
DS Fabrication
21
Site Work at BNL
22
Magnet Cost Estimate

• Bases
• Initially, engineering estimates were made some
  have been updated, some remain
• Several larger cost items have been corroborated
  by industrial estimates
• GA estimates for PS, TSu, TSd, and DS magnets
  cryostats
• Tooling and tooling design
• Winding, assembly, mandrels, vessels, shipping
• Other industrial estimates
• Conductor (Outokumpu)
• Coil mandrels (DG Machine, MEJ-Inc, Millinocket
  Fab Machine, in addition to GA)
• Insulation (Cryogenic Materials, Inc, Can-Do
  National Tape)
• Refrigerator/liquefier (Air Liquide, Cryo
  Technologies, Linde)
• Return yokes (International Steel)
• Resource loading based on schedule yields the
  implied funding profile
• Costs from Oct04 review were escalated from FY04
  to FY05
• Escalation taken at 4 per annum, which is high
  compared to RSVP guideline
• Escalation rates will be corrected in baseline
  preparation
• Costs have been reviewed in Feb05 in preparation
  for this meeting
• Labor rates were adjusted prior to the Oct04
  review to reflect rates from an MIT ITER study.
• An oversight resulted in the application of these
  new rates to only the PS magnet costs during the
  Oct04 review. This has now been corrected.

23
Overall Cost(FY05)

• Final design costs are taken from the final
  design plan as presented here.
• OH and profit rates
• Material 15
• Labor 100
• Note The 100 overhead rate is a hold-over from
  earlier MIT costing and is only relevant here in
  that it determines the equivalent total rate
  used, which is based on an MIT labor rate survey
  done for the ITER project.
• Profit 10

24
Labor Rates
25
PS Cost(Excluding Contingency)
26
TSu Cost (Excluding Contingency)
27
TSd Cost (Excluding Contingency)
28
DS (Excluding Contingency)
29
Funding by FY Profile (Assumes FY05 Start)

• Actual billings may vary due to accounting lag
  time
• Unspent but committed funds need to be carried
  over from one year to the next

30
Cost Analysis(overview)
31
Cost AnalysisPS
32
Cost AnalysisTSu
33
Cost AnalysisTSd
34
Cost AnalysisDS
35
Cost Analysis Summary(Reasons for, Amounts and
Percentages of Cost Growth from 2002)
36
New Changes from Magnet Review(Oct04) and ROM
for Cost Savings

• Lengthen the TS coils in the 90 bend regions
  from 20 cm to 40 cm. This reduces the number of
  coils that have to be manufactured in TSu from 32
  to 24 and from TSd from 30 to 22. Note This
  needs field verification!
• Change the PS from bath cooling to conduction
  cooling with a thermo-siphon loop.

37
Cost Savings ROM for Design Changes
38
Specific Questions(Received from Jim Miller)

• Where are the boundary lines between the magnet
  system and the rest of the experiment?
• The numbers you are sure about and the numbers
  you are not sure about. Without completing the
  RFP process, how much faith should we have in the
  magnet costs?
• Proposed approach and schedule for RFI, RFP,
  procurement, construction, etc.
• Is further scrubbing of costs planned? Who will
  do it?
• What is your understanding of the management of
  the magnet? That is, what is the line of
  responsibilities above the sub-system manager?
• Where are the main risks?

39
Answers to Questions

• 1. Where are the boundary lines between the
  magnet system and the rest of the experiment?
• The boundary lines will ultimately be defined
  through the interface documents, whose creation
  has been started by Mike Hebert. To my
  understanding, these will be further developed by
  the MECO Chief ME when he comes on board. This is
  a Systems Engineering function, which is
  generally handled at the project level.
• The gray areas at present are mostly in the
  facility interface area. This interface should
  become more clear in coming weeks as RSVP, MECO,
  MIT, and BNL meet to further define BNL roles.

40
Answers to Questions

• 2. The numbers you are sure about and the
  numbers you are not sure about. Without
  completing the RFP process, how much faith should
  we have in the magnet costs?
• Confidence level in the cost sheets is generally
  reflected by the size of the contingency.
• We have had some form of external (non-MIT)
  confirmation on about 68 of the total estimated
  cost. This includes cost information as follows
• GA on PS, TSu, TSd and DS cryostat costs
• Refrigerator/liquefier vendors on R/L costs
• Steel supplier on return yokes
• The largest uncertainty is with installation
  costs, as it is difficult to characterize the
  installation environment at BNL. The assigned 50
  contingency here is only a guess.

41
Answers to Questions

• 3. Proposed approach and schedule for RFI, RFP,
  procurement, construction, etc.
• The baseline approach leading to the RFI/RFP has
  been covered in the discussion of the Final
  Design Plan.
• RSVP has been queried regarding the
  procurement/construction approach. No decision
  has yet been made.
• The baseline approach reflected in the plan is to
  obtain RFP responses from industry to a
  relatively small set of large procurement
  packages. A package could be an individual
  cryostat (PS, TSu, TSd, DS) or some combination
  of two or more. The R/L, power supplies and
  magnet steel would also probably be purchased
  through separate contracts.
• Jon Kotcher states that, through discussions with
  Mike Harrison, he has learned BNL-SMD is
  interested in performing a magnet integration
  task where integration would be required in
  particular should the magnet be provided in
  smaller components. In that instance, SMD would
  assemble the components at BNL, test them, and
  install them in the AGS. This will be explored in
  coming weeks, starting perhaps next week.

42
Answers to Questions

• 4. Is further scrubbing of costs planned? Who
  will do it?
• Bill Molzon indicated recently in an email that
  RSVP plans to review the magnet costs prior to
  baselining.
• MECO looked into having a separate cost estimate
  done in the late 2002 timeframe, but that was
  superseded by the GA estimate followed by lack
  of sufficient funds to do anything more.

43
Answers to Questions

• 5. What is your understanding of the management
  of the magnet? That is, what is the line of
  responsibilities above the sub-system manager?
• Mike Hebert received some clarification to these
  questions in a recent discussion with Jon
  Kotcher.
• Magnets are apparently a cost center outside of
  MECO, but the magnet SSM is a L3 manager under
  the MECO PM.
• The MIT final design plan (presented above) has
  been accepted as the default baseline.
• Jon Kotcher has requested that Tom Taylor be on
  distribution for everything regarding the magnet
• Magnet management seems to still be unfolding.
• My opinionMECO should propose a management plan
  that works cooperatively with RSVP to achieve the
  MECO needs and objectives. I would be happy to
  help.

44
Answers to Questions

• 6. Where are the main risks?
• Unsettled approach to procurement
• Uncertain nuclear heating in the PS (could be
  offset with a cold compressor)
• Large size may lead to lengthy leak checking
• Installation environment is not well defined
• Ensuring BNL safety requirements are properly
  reflected in the magnet RFP.
• Coordination of magnet-facility interface with BNL

45
Magnet Management Requests

• Memo on UCI tolerance study approach and results
• Updated written field specification
• Latest UCI field (TOSCA) model and field
  verification points file/algorithm

46
Success

• The magnet team wants MECO to be successful
• MECO and magnet teams should work cooperatively
  together and with RSVP to achieve MECO objectives.