Next European Dipole (NED) Status Report
Arnaud DevredCEA/DSM/DAPNIA/SACM & CERN/AT/MAS
on behalf of the NED Collaboration
CARE Steering Committee Meeting5 September 2005
Some very good
news…
• The DOE has agreed to fund the US-LHC Accelerator
Research Program (LARP) with a budget of 11 M$ for FY06
(approved by Congress but not yet signed by the President.
• This budget level should be kept constant for a few years
(until 2009?)
• For FY06, it will be divided up into 5 M$ for magnets, 4 M$ for
accelerator-related R&D and 2 M$ for management (shared
between FNAL, BNL, LBNL and SLAC).
• The goal of the magnet part of LARP is to build by 2009 one
or two 4-m-long, 90-mm-aperture, 200 T/m quadrupole magnet
prototypes.
…for our American
colleagues!
Some pretty sad
news…
• The EUROMAG NEST Adventure proposal has been turned
down by the EU on the ground that it was not “adventurous”
enough…
• Hence, we are back to square one regarding the funding of
the model magnet manufacturing and test…
…for the NED
collaboration!
NED Programme
• The NED Programme is articulated around four Work
Packages and one Working Group
1 Management & Communication (M&C),
2 Thermal Studies and Quench Protection (TSQP),
3 Conductor Development (CD),
4 Insulation Development and Implementation (IDI),
5 Magnet Design and Optimization (MDO) Working
Group.
• It is carried out by a collaboration made up of 8 institutes:
CCLRC/RAL, CEA, CIEMAT, CERN, INFN/Genova and INFN/Milano,
Twente University (TEU) and Wroclaw University of Technology
(WUT).
M&C Work Package
• We have held three Steering Committee (SC) meetings since
the beginning of the year– 20 January at CERN– 14 April at CERN– 7 July at WUT
• Next SC meeting will be held at CERN during the CARE
general meeting; next ESAC meeting will be held at CERN
before or after the planned HHH/WAMDO (April 2005).
• Second quarterly report will be completed by the end of the
week.
• All relevant documents are stored into EDMS and posted on
the NED website
http://lt.tnw.utwente.nl/project.php?projectid=9
TSQP Work Package
• The TSQ Work Package includes two main Tasks
– development and operation of a test facility to
measure heat transfer to helium through conductor
insulation
(CEA and WUT; Task Leader: B. Baudouy, CEA),
– quench protection computation
(INFN-Mi; Task Leader: G. Volpini).
Heat Transfer Measurement (1/3)
• The first part of the Task was
to design and build a new double
bath cryostat.
• CEA wrote detailed
specifications that were handed
out to WUT in June 2004.
• WUT performed a call for
tender in the Summer of 2004
and selected Kryosystem in
Poland to manufacture the
cryostat.
Radiation shields
Vacuum container
Heat exchanger piping
Heat exchanger
Expansion valve
Pumping
He IIp
He IIs
LHeGHe
Insert
Cryogenic vessel
He I
Experimental volume
Schematic of NED cryostat(courtesy F. Michel, B. Baudouy and B. Hervieu, CEA)
Heat Transfer Measurement (2/3)
• A first reception test of the cryostat was carried out on
Kryostem’s premises the 3rd week of April 2005, which
revealed a few problems.
• WUT reacted very promptly and worked in close collaboration
with CEA to correct these problems.
• A second reception test was carried out the 2nd week of July
2005 (including thermal and leak tests in liquid helium at 4.2
K), which was deemed successful.
• The cryostat will be delivered to CEA/Saclay on 19
September, for final implementation and commissioning.
• The 6-month delay with respect to the initial schedule is not
expected to any deleterious impacts on the overall NED
Programme.
Heat Transfer Measurement (3/3)
Lambda plate
He II heat exchanger
Cryostat with thermal
shields
Inner view of cryostat with Instrumentation (Courtesy M. Chorowski, WUT)
Quench Computation (1/3)
• INFN-Mi has undertaken a detailed analysis of the thermal
and electrical behaviors of NED-type accelerator magnets
during a quench.
• The computation was started considering the “conservative”,
88-mm-aperture, cos, layer design developed by D. Leroy.
• It studied the influence of
– magnet length (1, 5 and 10 m),
– operating current (15, 22 and 29 kA),
– external dump resistance (15, 25 and 35 m),
– quench detection delay (30, 40 and 50 ms),
– quench protection heater length.
• It also compared the results obtained by two different codes:
QLASA at INFN-Mi and QUABER at CERN.
Quench Computation (2/3)
• The results show that, for the entire parameter space, the
magnet is quite safe to operate, thereby justifying the choice of
wire and cable parameters made early on.
0
50
100
150
200
250
300
Rd
=15
mO
hm
Rd
=25
mO
hm
Rd
=35
mO
hm
Rd
=45
mO
hm
De
lay=
30
ms
De
lay=
40
ms
De
lay=
50
ms
I=15
kA
I=22
kA
I=29
kA
Ho
t sp
ot
tem
pe
ratu
re (
K)
QLASA
QUABER
0
200
400
600
800
1000
1200
1400
Rd=
15 m
Ohm
Rd=
25 m
Ohm
Rd=
35 m
Ohm
Rd=
45 m
Ohm
Del
ay=3
0 m
s
Del
ay=4
0 m
s
Del
ay=5
0 m
s
I=15
kA
I=22
kA
I=29
kA
Tota
l vol
tage
(V)
QLASA
QUABER
Quench simulation results on 10-m-long, 88-mm-aperture, cos, layer design
(Courtesy M. Sorbi INFN-Mi)
Quench Computation (3/3)
• Similar computations have now been started on the more
“innovative”, 160-mm-aperture, slot design also proposed by D.
Leroy.
• The quench computation task is near completion.
88-mm-aperture, layer design
(Courtesy D. Leroy, CERN)
160-mm-aperture, slot design
(Courtesy D. Leroy, CERN)
CD Work Package
• The CD Work Package includes two main Tasks
– conductor development
(under CERN supervision; L. Oberli has now taken over
D. Leroy as the official Task Leader),
– conductor characterization
(CEA, INFN-Ge, INFN-Mi, and TEU; Task Leader: A. den
Ouden, TEU).
• It is the core of the programme and absorbs about 70% of
the EU allocated funding.
Conductor Development (1/2)
• As a conclusion of preliminary design studies carried out in
2003 and 2004 under the supervision of D. Leroy, the following
specifications were derived for NED Nb3Sn strands
– diameter 1.250 mm,
– eff. filament diameter < 50 m,
– Cu-to-non-Cu ratio 1.25 ± 0.10,
– filament twist pitch 30 mm,
– non-Cu JC 1500 A/mm2 @4.2 K & 15
T,
– minimum critical current 1636 A at 12 T,
818 A at 15 T,
– N-value > 30 at 4.2 K and 15 T,
– RRR (after heat treatment)> 200.
(It is also requested that the billet weight be higher than
50 kg.)
Conductor Development (2/2)
• Based on these specifications, a call for tender was issued by
CERN in June 2004 and two contracts were awarded in
November to 2004 to Alstom/MSA in France (“Enhanced
Internal Tin” process) and SMI in the Netherlands (“Powder in
Tube” Process).
• After discussion with CERN, the two companies agreed to
work out their development program into two successive RD
Steps (referred to as STEP 1 and STEP 2) followed by final cable
production.
• A tentative schedule was established as follows
– STEP 1: Summer 2005,
– STEP 2: Summer 2006,
– Final production: December 2006.
Conductor Characterization (1/2)
• The NED-type conductor characterization represents a real
challenge, given the unprecedented performances that are
expected (e.g., a critical current of ~1600 A at 4.2 K and 12 T
on a 1.25-mm wire, to be compared to the timid ~200 A
presently achieved on 0.8 mm ITER wires).
• To validate sample preparation and measurement processes,
the laboratories involved (CEA, TEU and INFN) have launched a
cross-calibration program reminiscent of the ITER/EDA cross-
calibration program carried out in the mid-1990’s.
• Since the Summer of 2004, three rounds of calibration wires
have been prepared and circulated among the various partners.
Conductor Characterization (2/2)
SMI (billet 179)
500
700
900
1100
1300
1500
1700
1900
6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 16.0
B [T]
Cri
tica
l cu
rren
t [A
]
TU, sample 1
TU, sample 2
LASA, sample 1
LASA, sample 2
SMI/Toshiba Test Wire
(results are within 2%)
• TEU and INFN have now achieved a good convergence.
• The problems at CEA have been identified and are being
solved.
• All 3 partners should be ready when the first wires become
available.
(Courtesy T. Boutboul,
CERN)
FE Wire Model
• In Parallel, INFN-Mi has started to develop an ANSYS model of
an “un-reacted,” Alstom/MSA-type wire so as to simulate
cabling effects.
• Running such a computation requires a detailed knowledge of
the mechanical properties of the materials making up the wire
(in the cold work state where they end up prior to the cabling
operation).
• To determine these properties, CERN has carried out a series
of nano-indentation and micro-hardness measurements on
various wire samples, and compared the results with available
literature data.
• The next step is to apply this model and the appropriate
mechanical properties to the wire layouts presently considered
by Alstom/MSA.
IDI Work Package
• The IDI Work Package includes two main Tasks
– studies on “conventional” insulation systems relying
on ceramic or glass fiber tape and vacuum-impregnation
by epoxy resin
(CCLRC; Task Leader: E. Baynham),
– studies on “innovative” insulation systems relying on
pre-impregnated fiber tapes and eliminating the need
for a vacuum impregnation
(CEA; Task Leader: F. Rondeaux).
Conventional Insulation (1/2)
• CCLRC and CEA have developed in collaboration an
engineering specification (issued in July 2004) and a
coordinated test programme (issued in October 2004).
• Since then, CCLRC has carried out a number of screening
tests of candidate materials.
• The tests are applied to standardized laminates
representative of inter-turn insulation and include
– electrical breakdown test,
– short beam shear test,
– inter-laminar fracture test.
Precrack (release film)
Crack growth from test
Example of Double Cantilever Beam (DCB)
fracture test(courtesy S. Canfer, CCLRC)
Conventional Insulation (2/2)
• CCLRC has also investigated the issue of “sizing” (a lubricant,
usually organic, coated onto the fibers of tapes, that need to be
removed prior to conductor wrapping and winding, thereby
rendering the fiber tape fragile and easy to tear off).
• Very promising results have been obtained with an improved
polyimide sizing, produced by Hydrosize, NC, and applied by
JPS, SC, which seems to be able to sustain the required Nb3Sn
heat treatment without carbonization (thereby eliminating the
need for “de-sizing”).
• More complete evaluation tests are underway.
Innovative Insulation
• The work on innovative insulation has not started yet,
pending the hiring of a technician at the CEA chemistry
laboratory, which has been delayed until early next year.
• To compensate for this lack, it was decided last spring to
reallocate the EU funding of this task to hire a postdoc at CEA.
• A candidate has been identified, who is expected to start
working this fall.
• The timing of this task is now becoming critical with respect
to the overall NED program.
MDO Working Group (1/3)
• The MDO Working Group is made up of representatives from
CCLRC, CEA, CERN and CIEMAT, under the Leadership of F.
Toral, CIEMAT.
• Its main charge is to compare different magnet
configurations so as to assess their efficiency in terms of
manufacturability, performance and cost.
1.599814 MN/m
1.759376 MN/m
1.366123 MN/m
1.104555 MN/m
0.522062 MN/m
1.443344 MN/m
1.721069 MN/m
Coslayer design(courtesy D. Leroy,
CERN)
Intersecting-Ellipses design
(courtesy H. Felice,CEA)
Motor-type design(courtesy F. Toral,
CIEMAT)
MDO Working Group (2/3)
-80.00
-60.00
-40.00
-20.00
0.00
20.00
40.00
60.00
80.00
100.00
0.00 5000.00 10000.00 15000.00 20000.00 25000.00 30000.00
Shims
Variation Shims
Variation Without
Persistent Currents Without
Snapshot after 6 days of calculation
• In parallel, work is pursued at CERN so as to optimize the
baseline, 88-mm-aperture, cos layer design with respect to
– conductor geometry,
– iron shape (to reduce saturation effects),
– ferromagnetic shims (to compensate magnetization
effects).
(courtesy N. Schwerg, CERN)
MDO Working Group (3/3)
• CCLRC/RAL is also developing a 2D ANSYS model of the 88-
mm-aperture, cos layer design so as to optimize mechanical
support.
• This model includes “sub-models” of individual coil turns to
compute peak stresses in cable strands and cable insulation.
Von-Mises Stress in Cable
0 MPa < σVon-Mises < 180 MPa
Circumferential Stress in Insulation
-105 MPa < σθ < -35 MPa
Plane-stress with ½ unit thickness
Generalised plane-strain
Contact elements
(courtesy P. Loveridge, CCLRC)
Conclusion
• A great deal of progress has been made since my last
presentation (at the CARE general meeting in Hamburg last
year), leading to a number papers presented at various
conferences (1 at CEC/ICMC, 1 at EUCAS and 4 at MT).
• The cryostat for heat transfer measurements is completed
and will be delivered to CEA next week.
• The next few months will be critical for the Conductor
Development program with the results of the STEP 1 wires.
• The only Task that has not started is the Innovative Insulation
Task at CEA, but the hiring of a Postdoc should help.
• The funding of the model magnet manufacturing remains an
open question.