Linear
Collider Forum of America
Laboratory/
Magnet Manufacturers Industrialization Workshop
Held at Fermi
National Laboratory, Batavia, IL
Oct 18, 2006
Objective:
Brief U.S. industrial magnet designers and manufactures on the quantity,
designs, costing status and issues of the approximately 16,500 magnets
required for the baseline configuration of the ILC. Provide industry feedback
to the ILC program on issues raised during the workshop. The industrialization
workshop was held following the Oct 16 and 17 LCFOA general membership meeting
at Argonne and Fermi National Laboratories. The workshop was sponsored by the
LCFOA. The principal attendees at the meeting are listed in the below table.
There were a few additional observers from FNAL.
|
Process
Equipment Co. |
Richard
Schafer |
Tipp City, OH |
937-667-9322 |
rschafer@processeq.com |
|
Quality
Transformer and Electronics |
Gene Clift |
Milpitas, CA |
408-263-8444 |
gcc@qte.com |
|
Advanced
Magnet Lab, Inc. |
Rainer Meinke |
Melbourne, FL |
321-728-7543 |
rbmeinke@yahoo.com |
|
Everson
Tesla, Inc |
Greg
Naumovich |
Nazareth, PA |
610-746-1521 |
gnaumovich@eversontesla.com |
|
Everson
Tesla, Inc. |
Bill
Umbenhauer |
Nazareth,
PA |
610-746-1522 |
bumbenhaur@eversontesla.com |
|
New England
Techn1-Coil, Inc. |
Daniel
Marisseau |
Tuftonboro,
NH |
603-569-3100 |
info@technicoil.com |
|
SLAC |
Cherrill
Spencer |
Menlo Park,
CA |
650-926-3474 |
cherrill@slac.stanford.edu. |
|
FNAL |
John Tompkins |
Batavia, IL |
630-840-5260 |
jct@fnal.gov |
|
LCFOA |
Ken Olsen |
Washington,
DC |
202-222-8315 |
ken.olsen@lcfoa.org |
Background:
The meeting was chaired by Cherrill Spencer of the Stanford Linear Accelerator
Center (SLAC). Cherrill first presented and described a preliminary inventory
of the room temperature and superconducting magnets required by styles and
quantities for the ILC baseline configuration as shown in the table at the end
of this report. In summary the present design requires 166 individual styles
of magnets with a total quantity of 16,556 magnets, 2484 of which are
superconducting.
Cherrill noted in
her presentation that these quantities will probably be reduced, although not
by a significant amount, due to potential cost reduction design modifications
currently being considered by the Global Design Effort (GDE). Cherrill’s
presentation has been posted on the FNAL web site at
http://ilc.fnal.gov/documents/lcfoa_10_06.html.
Both Cherrill
Spencer and John Tompkins (FNAL) answered clarification questions from
industry on the ILC magnet requirements and program schedule. The workshop
then proceeded into a discussion of topics prepared by the ILC magnet systems
group.
Discussion
Topics: Cherrill next presented the topical issues of concern to the
ILC magnet working group. This presentation is also posted at the above site.
The following five general topics were then discussed by the group:
-
Standards for materials and hardware
-
Cost estimating techniques
-
Preferred magnet procurement processes
-
U.S. magnet manufacturing capacity
-
Other issues of concern to industry
The results of
there discussions are provided below:
Standards for
materials and hardware: A general review of U.S. industry practices and
specifications for materials and fabrication on items such as copper, epoxy,
connectors, conductor wrapping techniques, etc. concluded that current
state-of-the-art high end manufacturing processes can meet the ILC’s
recommendations without any significant issues or cost increases. Industry
strongly recommended that the ILC specify square conductors for the ILC
magnets and said that there are no issues with power and water connectors on
square conductors with proper installation techniques and choice of brazing
fluxes. Industry also stated they have worked with the type of several part
epoxies described in the SLAC presentation and they will not present any
problems in the manufacturing process, although they were becoming more
expensive as the resin was an oil byproduct and ILC might want to consider a
one-part epoxy.
The major concern
brought up by industry was the hardware reliability criteria proposed for the
magnets in the ILC program. The ILC has established an overall machine
availability criteria. Because of the large quantity of magnets in the system
and the assumption by the ILC availability group that if any magnet fails the
ILC would not operate, each magnet must operate without failure for 20,000,000
hours to meet the overall criteria. Industry commented that this requirement
exceeds anything they manufactured for DOD such as for shipboard radar systems
that must operate under combat conditions. Therefore, this requirement will
significantly increase the cost of the ILC magnets. Industry recommended the
program revisit this criteria and the trade-offs between reducing the mean
time to repair and reliability criteria.
Cost
estimating: The ILC magnet group has a short time frame to estimate the
total costs of all magnets for the machine. Cherrill proposed a cost
estimating technique that develops a formula which relates the cost of winding
and potting a coil to its finished weight. One adds the cost of insulated
copper to the labor cost of winding and potting for the variety of magnet
coils and finds a relationship between coil weight and cost, which then would
be used for predicting the cost of other coils. One could follow the same
method to arrive at a cost per weight of finished steel core. Industry stated
that this cost by weight method is too risky for the magnets required for the
ILC and would lead to erroneous estimates. While this may work for the simpler
dipoles, it should not be used for quadrupoles and special magnets. Industry
made two suggestions for the program to consider between now and the end of
the year on the costing issue.
1.
Hire a
senior engineer with cost estimating experience as a consultant to do a best
estimate of the inventory of ILC magnet requirements. A name of a recently
retired person who may be available was provided to the program for the next
to months to help with this effort.
2. In
addition, industry recommended the program provide the next level of detail
on ILC magnet requirements in prioritized order to industry. Industry will
provide rough order of magnitude (ROM) budget estimates to the program at no
cost for those magnets in the upper cost categories, the so called “cost
drivers” due to quantity and/or complexity. This will be provided on a
non-proprietary basis based on available time and resources.
Procurement
processes: The group discussed the following three alternatives for
procuring the ILC magnets which were presented by the program.
-
ILC does
magnetic design, top assembly layout drawings and all detailed part
drawings; and provides detailed written fabrication specs.
-
ILC does
magnetic design, does top assembly layout drawings, completely specifies
conductor, pole shape, # water circuits, provides written fabrication specs,
but does NOT provide detailed part drawings
-
ILC provides
detailed fabrication specs, magnetic parameters, overall mechanical
dimensions, required field quality and provides NO drawings at all.
There was considerable discussion on the advantages and
disadvantages of each and there was a general consensus among the industry
present that alternative 3 would provide the lowest price to the ILC program.
It allows companies the flexibility to use proprietary fabrication techniques
and best practices. But at least two of the companies stated they would only
be able to provide magnets under alternative 1.
The ILC proposed a potential cost saving by having the
program procure the copper conductor required for the magnets and providing it
to the manufacturers. This buying in bulk should result in a lower price per
unit weight. The copper would then be shipped directly from the conductor
vendor to the manufacturers. The companies thought this would be acceptable.
Production schedules were also discussed. Magnet
manufacturers will view the ILC magnets as a large order to be scheduled in
with other orders. Therefore to minimize overtime and other additional costs,
the longer the lead times for production and the lower the annual production
rates, the lower the price to the program.
Other cost-reducing techniques were discussed. If some of
the many different styles could be merged into one style; that would reduce
the overall cost per unit because there would be less tooling variation, less
magnetic modelling, and drawing time.
U.S. manufacturing capacity:
The magnet manufacturers did not see this as an issue given the quantity,
variety and schedules of the ILC magnet requirements. This was based on the
assumption that about one third of the magnets would be fabricated in the U.S.
Key Issues and Recommendations:
The following recommendations were made by the 5 companies present during the
meeting:
Near term cost estimate: The
ILC program should immediately hire a consultant with considerable magnet
manufacturing experience to prepare a budget estimate for the projected ILC
magnet inventory. This would add considerable credibility to the preliminary
cost estimate due by the end of 2006. A name of a recently retired person was
recommended during the meeting.
In addition, the magnet companies present would provide a
non-proprietary budget estimate of the more complex magnet components
providing the ILC program supplies the detail design information required.
Overall Magnet Program Cost Reduction:
Industry representatives suggested the ILC program review the number of magnet
styles and designs to reduce the number of individual designs if possible.
Production costs per unit will be lower for longer production runs. More
standardization would also benefit the operation and maintenance of the ILC.
Magnet Reliability: Review
the effects of radiation on the proposed resin systems with prototypes and
life cycle testing to understand the impact based on the MBTF requirements.
Development and life cycle testing on braze joint techniques should be
performed to reduce potential failure regions as these were cited as the more
likely regions for failure over the life of the magnets.
2007 Industrial Cost Study:
The ILC should contract with industry in 2007 to prepare a non-proprietary
magnet cost study similar to the one being performed on the RF units. This
study would also estimate cost savings that may be achieved by modifying the
design criteria and through model standardization.
2007-2009 Magnet Prototypes:
The ILC should review the magnets and list which ones have difficult criteria
or very tight tolerances or will be in very high radiation areas and develop a
plan to selectively prototype and test these units with industry over the next
three years.
Ken Olsen
Oct 31, 2006
