Title: Contents
1Superconducting LHC Magnets characteristics and
qualification in SM18 test station
Marco Buzio (AT/MTM)
- Contents
- Introduction overview of LHC magnet system
- Superconducting cables and magnets
- The LHC cryodipoles
- Cryogenic testing in SM18
- 4.1 Power tests
- 4.2 Field quality
-
2Acknowledgements
This talk gives some highlights on the work done
by many people in the AT and TS department over
many years. Special thanks to L Bottura, V
Chohan, A Masi, JG Perez, P Pugnat, S Sanfilippo,
A Siemko, N Smirnov, W Venturini Delsolaro
(AT/MTM) M Pojer, L Rossi, D Tommasini (AT/MAS) J
Axensalva, JP Lamboy, B Vuillerme, L Herblin
(AT/ACR)
3Overview of the LHC Magnet System
Magnet Type Function Field/Current/Magn. length N. Manufacturer
Main DipoleMB steer beams around the rings 8.33 T11850 A 14.3 m 1232 Alstom (F) Ansaldo (I) Noell (D)
Main Quadrupole MQ focalize beams 233 T/m 11850 A3.1 m 400 Accel (D)
Multipole Correctors spool pieces(e.g. MCS, MCDO) Compensate field errors in main magnets 0.1 3 T 50-550 A 0.15 1.5 m 3600 Antec (E)Tesla (UK)KECL, CAT (In)Compton Greaves (In)Sigmaphi (F)etc
Orbit correctors(e.g. MCBH, MCBV) Adjust beam orbit (hor/vert) 0.1 3 T 50-550 A 0.15 1.5 m 2800 Antec (E)Tesla (UK)KECL, CAT (In)Compton Greaves (In)Sigmaphi (F)etc
Lattice correctors(e.g. MQS, MQT, MS, MO) Adjust beam parameters (e.g. tune, chromaticity) 0.1 3 T 50-550 A 0.15 1.5 m 2800 Antec (E)Tesla (UK)KECL, CAT (In)Compton Greaves (In)Sigmaphi (F)etc
4Overview of the LHC Magnet System
Cryodipoles in SMA18
- Magnet lattice ½ cost of LHC, 10 yr RD
- Challenge large-scale, advanced technology
transfer to industriesextensive tests needed at
CERN (10 of magnet cost)
Short Straight Section in SM18
Different types of correctors
5Status
- Baseline includes cold tests for all main
magnets, details to be finalized - 1 octant of dipole cold masses delivered, 120
cold tested (only two rejected) - 6 Short Straight Sections assembled and tested
- cold test rate expected to ramp up from 8 to 14
magnets/week as soon as all 12 benches completed
(Q3 2004) - first dipole installation tests foreseen in June
- End of cold test phase expected Q3 2006.
6Why superconducting magnets ?
Copper NbTi
Resistivity 1.510-8 ?m 0 (DC)
Current density 5-10 A/mm² Pure NbTi _at_ 1.9K 2500 A/mm² LHC Cable 400 A/mm²
Cable cost 25 /kAm 25-50 /kAm
Max. temperature lt 90C(air or water cooled) lt 4.2 K(LHe cooled)
Max. magnetic field ? (stress-limited) 10 T _at_ 1.9K
Non-linearities Non-linear ferromagnetic hysteresis Diamagnetic effect, rate dependence
7Superconducting vs. resistive field quality
Field quality determined by coil geometry ?
geometry not accessible for direct measurements
at working temperature (cryostat/beam pipe) ?
results must be extrapolated ? coils must be
shimmed during production, errors extremely
difficult to correct ? conductor positioning
errors of 25 mm provoke relative field errors
10-4
Field quality determined by iron pole shaping ?
geometry accessible for direct measurements ?
Measurement/shimming can be iterated ? large
conductor positioning errors tolerable ? Field
depends on the homogeneity of material magnetic
properties
8Stability of superconductor quenches
- A quench is a sudden thermodynamic transition to
the normal resistive state, as the material
crosses the critical surface Bcr(Jcr,T) - Superconducting wire is intrinsically unstable
a quench can be triggered locally by the
deposition of a few mJ (very low heat capacities
at 1.9K !), which may be released by - cable
movements (few mm) ? magnetic friction, Lorentz
forces, mechanical friction - cracking of
resin - radiation from the beam - Quench stability is achieved by making the SC
into thin filaments embedded in a conductive
matrix. - The performance of a magnet is degraded w.r.t.
material properties due to manufacturing process,
non-uniform field and current distributions - Quench detection and active magnet protection are
necessary to avoid excessive localized heat
deposition (global margin for LHC dipoles 85) -
Critical surface of NbTi
9LHC superconducting cable
- Total 400 tons NbTi, 7000 km
- Rutherford type structure
- 28 or 36 strands per cable, twisted to minimize
linked flux during field ramps - up to 8800 7 mm Ø filaments per strand, embedded
in Cu to achieve thermal stability(minimize
Joule heating maximize heat transfer after a
quench) - SC cross-section as high as 60 of the total to
increase current density ? relatively low
stability - insulated with barber-pole wrapped polymide to
allow for high LHe penetration (90 filling
factor) - keystoned a different design for each coil
layer (to allow current grading)
Cu
NbTi
10LHC superconducting cable
One LHC superconducting cable carries up to 13000
A ..
which is equivalent to about 10 conventional
power lines
or this thick bunch of resistive copper cables.
11Cryodipole overview
Instrumentation connector
12Dipole coils
Cu spacer blocks
Beam pipe
Quench Heaters (outer)
ideal cos(q) current distributionover a circular
profile giving anuniform field inside the circle
NbTi cable
Quench Heaters (inner)
and the optimized approximationmaking use of a
discrete conductor
Outer layer (lower B, higher J)
Inner layer (higher B, lower J)
13Lorentz forces
- Magnetic coils tend to expand under the effect
of self-forces - Coils in the two apertures are attracted
- Stress levels depend on field strength, thermal
contraction, level of pre-compression and width
of any gaps
14Energy stored
The total energy stored in the magnetic field of
each dipole is
7 MJ
that is equivalent to the mechanical energy
necessary to lift a massof 32 tons to the height
of 22 m
or the thermal energy sufficient to melt 5 kg
of steel
or the electrical energy needed to light a
100W bulb for 20 hours
or the chemical energy contained in about 500
gr of delicious Swiss muesli !
However WARNING electrical insulation can be
irreversibly damaged by sparks containing less
than a J !!
15Dipole cold mass
Austenitic (non-magnetic) steel collars
Shrinking cylinder (welded under compressionto
confer curved shape)
Alignment pins
Beam pipe kapton insulation
20K GHe cooling conduit
Corrector spool pieces
Pseudo-random cryopumping holes
Lyre-shaped current leads
Beam screen
16Two-in-one concept
- Part of the magnetic flux returning through the
yoke goes to enhance the field in the other
aperture (10) ? Magnetic symmetry is broken?
Field errors are coupled
collars
iron yoke
- Number of magnets to build is halved? Space
occupied in the tunnel is considerably reduced ?
Difficulty, cost and risk of construction are
increased
Overall higher cost effectiveness
- Distance and parallelism of the two beam rings
are guaranteed? Construction errors cannot be
corrected ? tighter tolerances
17Magnet test sequence (essential steps)
- SM18/SMI2
- Warm HV insulation- Fiducialization
(geometry)- Special magnetic measurements(polarit
y, field direction)- Inserion of beam screen-
Preparation for storage
- SM18
- Cold HV insulation - Quench protection system-
Training- Magnetic measurements- Special tests
(short sample limit, geometry )
- SMA18
- Electrical leak tests- Instrumentation
setup- Cryostating- Preparation for cold
tests
Cold mass arrives at SMA18
Long-term storagein Prevessin
6 6 cryogenic test benches
Cryogenic Feed Box
13kA power converters
Scanning machine for SSS
Long coil shaft system for MBs
VME acquisition and control racks
18Thermal cycling on cryogenic test benches
GHe _at_ 80K
LHe _at_ 4.2K
SF LHe _at_ 1.9K
Cool-down
- cool-down and warm-up greatly accelerated (2
weeks in LHC) - LHe bath at ambient pressure is made inside cold
mass (40 kg LHe/dipole) - Heat exchanger carries innovative bi-phase
superfluid He flow to subcool down to 1.9K
warm-up with He _at_ 320K
19Power tests
- HV insulation tests up to 3kV between coils,
ground and quench heaters - Instrumentation tests DAQ system and magnet
instrumentation (voltage taps, T sensor) verified
and calibrated - Quench protection system test quench heaters
are discharged in various combinations at the 1.5
kA level to verify time constants, max. induced
voltages and currents, etc - Training field is repeatedly ramped up, and
possibly quenched, until the ultimate field level
is reached (9T).The location of the quench is
systematically measured to assess quality of
construction. - Special tests e.g. provoked quenches at 4.2K to
assess the current-carrying capability of the
cable (short sample limit) - Provoked quench at 7T empty He before warm-up
Quenches are normally originated in the
heads(complex 3D geometry looser
tolerances)Quenches originating in the straight
part may point to fabrication defects
20Training example good dipole
Ultimate field
detraining
Memory effect
Nominal field
21Training example bad dipole
Ultimate field
No memory effect
Nominal field
22Magnetic measurements
why magnetic measurements ? To qualify/accept
magnets - minimum level of performance must be
guaranteed- harmonics measurement may spot
construction errors For the machine reliable
operation requires detailed knowledge of -
transfer function to within 200 ppm- harmonics
up to decapole to within 50 ppm- field direction
to within 1 mrad- dynamic effects modelling
database
- Loadline local and integrated transfer function
and harmonics as a function of steady-state
excitation level - LHC cycle quantify dynamic effects as they will
occur in the real machine cycle (magnet is put in
a reproducible clean state by means of a previous
quench and current pre-cycle) - Coupling currents systematic study of dynamic
effects as a function of ramp rate and powering
history - Field direction average direction of the field
w.r.t. cryostat fiducials - Magnetic axis position of the axis (locus of
B?0) w.r.t. cryostat fiducials, used to define
the precise position of the magnet in the tunnel
and to assess the geometry in cold conditions.
23Non-linear effects
- Superconducting filament magnetization
(persistent eddy currents) - large hysteresis with relative errors of the
order of 10-3 at low field (injection) - hysteresis depends on temperature, current and
current history - main field and multipoles affected in different
ways
- Linear regime (geometric contribution)
- field is proportional to the current (can be
computed with Biot-Savarts law) - the T.F. depends only on the coil geometry
injection
- Iron saturation
- affects only small area in the collar (Bgt2T)
- relative errors of the order of 10-2 at high
field - additional multipoles generated
24Dynamic superconductor effects
- Coupling currents effects
- finite inter-filament and inter-strand
resistance (RC) gives rise to loops linked with
changing flux - multipole errors ? ?, RC-1
- hysteresis depending in a complex way upon field
level, temperature and powering history
- Decay and snap-back
- superconductor magnetization and coupling
currents interact in a complex way to give
long-term logarithmic time dependence effects
(field decay) - hysteresis branch switching may occur at the end
of a decay phase to cause sudden current
redistribution and additional multipole errors
(snap-back)
25Magnetic measurement equipment twin 15m harmonic
coil shafts
coil
SiN flange bronze roller
Micro-cableconnector
shafts
Ti bellows
Twin Rotating Unit
- 2?13 Al2O3 sectors carrying 3 rectangular coils
each - cover the full length of the magnet to obtain
integral in a single rotation ( few seconds) - stiff torsionally, bending at joints to follow
the curvature of the magnet - doubles up as an antenna for quench localization
- resolution 0.1 mT, 50 mrad accuracy 100 ppm
26Magnetic measurement equipment harmonic analysis
A coil
V(t)
GV(t)
F(q)
?
G
FFT
A
A-C
Programmable amplifier
C coil
An,Bn
Compensated (A-C)
Absolute (A)
27Magnetic axis measurement in dipoles
Quadrupole Configured Dipole(ugly quad)
Dipole axis measurement with a warm rotating coil
mole telescope tracker
28Laser Tracker Survey
Taylor-hobson 3.5 tooling ball
Rotating head
Retro-reflector
- Laser tracker to detect 3D position of spherical
optical targets (retro-reflector corner cubes)
mounted on standard conical receptacles - Interferometer and phase modulation detector for
distance measurement2 ? 16000 pt. incremental
angular encoders for direction measurement - Nominal accuracy 5 ppm 50 mm _at_ 10
mReal-world repeatability 150 mm _at_ 10 m - Measurement modes static, cyclic, dynamic (lt
1kHz) - Used for- measurement of magnetic axis and
field direction of dipoles and quadrupoles, warm
and cold (moles, Chaconsa, 15m shafts benches)-
geometric survey of auxiliary equipment
(reference magnets)
LEICA Tracker unit
29References
L Rossi, Superconducting magnets for the LHC main
lattice, Presented at 18th International
Conference on Magnet Technology MT 18 , Iwate,
Japan , 20 - 24 Oct 2003 M Allitt, R Wolf et al
Status of the Production of the LHC
Superconducting Corrector Magnets Presented at
18th International Conference on Magnet
Technology MT 18 , Iwate, Japan , 20 - 24 Oct
2003 L Rossi, Superconducting Cable and Magnets
for the Large Hadron Collider Presented at 6th
European Conference on Applied Superconductivity
EUCAS 2003 , Sorrento, Napoli, Italy , 14 - 18
Sep 2003 W Scandale, E Todesco et al, Influence
of Superconducting Cable Dimensions on Field
Harmonics in the LHC Main Dipole, LHC Project
Report 693 CERN Accelerator School on
Superconductivity in Particle Accelerators,
Hamburg 1995, CERN Yellow Report 96-03 The Large
Hadron Collider Conceptual Design, Report
CERN/AC/95-05