Title: MuCOOL and MICE Coupling Magnet Status
1MuCOOL and MICE Coupling Magnet Status
- Michael A. Green
- Lawrence Berkeley Laboratory
- Berkeley CA 94720
2The ICST Coupling Solenoid Design
Coupling Magnet Cryostat
Coupling Magnet Cold Mass
3Magnet Progress to Date
- A portion of the superconductor has been
delivered. The rest of the conductor is being
fabricated. - Engineering design of the coupling magnet is over
90 percent complete. - The engineering design of the two ICST test coils
is complete. - The coil winding tooling is complete.
- The first dummy test coil is being wound at ICST.
- The first test coil joints have been fabricated.
The joints will be tested in a magnetic field at
Berkeley.
4ICST Coil Winding Facility
Coil Winding Apparatus
De-spooling and Tensioning Apparatus
Winding a Copper Test Coil
5Purpose of Test Coil 1
- This coil is the first superconducting solenoid
wound with the coupling coil conductor. - The joint design and the coil taps for the quench
protection system is verified in this magnet. - This coil will be part of the crew training
process. - Test coil 1 will produce enough field so that the
magnet can be used for cold diode tests and other
sorts of tests to be done in a magnetic field at
300 K, 77 K and 4.2 K. - Both of the superconducting test coils will be
cooled using the ICST refrigerator.
6Purpose of Test Coil 2
- The winding parameters for the coupling coil will
established by winding a full diameter test coil. - The 2nd test coil will be cooled down using the
ICST refrigerator in a cryostat suitable for the
training the large test coil and the coupling
magnets. - Test coil 2 will be strained by the cool down and
the magnetic field so that the worst case strain
seen by the MICE coupling coils is seen in the
test coil. - There will be a full scale test of the coupling
coil cold diode and resistor quench protection
system with the test coil sub-divided into four
circuits.
7Parameters for the Test Coils and the Coupling
Coil
Coil Small Small Prototype Prototype Coupling 96 layer
Coil 24 layer 12 layer 96 layer 80 layer Coupling 96 layer
Coil Length (mm) 285 285 72 72 285
Coil Inner Radius (mm) 175 175 750 750 750
Coil Thickness (mm) 25.5 12.8 102.5 85.4 102.5
Number of Layers 24 12 96 80 96
No. Turns per Layer 166 166 42 42 166
Magnet J (A mm-2) 114.6 114.6 114.6 114.6 114.6
Magnet Current (A) 210 210 210 210 210
Self Inductance (H) 4.529 1.11 50.059 35.88 592.5
Peak Induction in Coil (T) 2.79 1.43 3.91 3.56 7.44
Magnet Stored Energy (KJ) 99.867 24.4 1104 791.2 13069
4.2 K Temp. Margin (K) 3.3 3.9 2.5 2.9 0.8
Conductor Len/coil (km) 4.701 2.27 20.299 16.735 80.118
The small test coil will have 24 layers. The
large prototype test coil will have 96 layers.
8Magnet Load Lines
9Magnetic Field in the Small Test Coil at a
Current of 512 A
10Von Mises Stress of the Small Test Coil after
Cool Down and Charging at 512 A
After cooling to 4.2 K
After charging to 512 A
11Magnetic Field in the Large Test Coil at a
Current of 401 A
12Von Mises Stress in the Large Test Coil
After Coil Banding
After Coil Winding
After Charge to 401 A
After Cool Down to 4 K
13Stress and Deflection Simulations for the 2nd
Test Coil (the full diameter test coil)
- Stress and deflection simulations were done on
the large test coil. Stress and deflection
calculations were done after the magnet was
cooled to 4.2 K and powered to 401 A. - The hoop stress and Von Mises stress is almost
the same for the large test coil at 401 A when
compared to the coupling coil at 210 A. The
radial deflection is also similar. If the large
test coil can operate at 400 A, the coupling coil
should operate to at least 200 A.
14Von Mises Stress in the Coupling Coil
Winding banding
Charging
Winding coil
Cooling
The stress and deflection calculations include
the winding pre-stress in the coil and banding,
which keeps the coil from lifting off the mandrel
when it is fully charged to a current of 210 A.
The boundary conditions are different for
the coupling coil than they were for the large
test coil. The minimum winding Pre-stress is 70
MPa. The aluminum banding pre-stress is 30 MPa.
15Modeling the Temperature Distribution in the
Coupling Coil during Normal Operation
The heat flow into the magnet as shown is 1.5 W
from radiation and conduction.
DT 0.074 K
The heat flow due to AC losses in the magnet coil
during to a rapid discharge from 210 A to no
current in 5400 seconds has also been modeled
using ANSYS. The coupling magnet can be
discharged in 5400 seconds without quenching.
16A Summary of Quench Protection Simulations for
the 2nd Test Coil and the Coupling Coil
Results for Test Coil 2
Parameter Parameter 0 mW 60 mW
2-Section Hot Spot Temperature (K) 136 136
2-Section Max Internal Voltage (V) 6043 6006
4-Section Hot Spot Temperature (K) 129 129
4-Section Max Internal Voltage (V) 3112 3074
6-Section Hot Spot Temperature (K) 127 126
6-Section Max Internal Voltage (V) 2077 2043
Results for Coupling Coil
Parameter Parameter 0 W 5 W
8-Section Hot Spot Temperature (K) 135 88
8-Section Max Internal Voltage (V) 2669 1391
17The ICST set up allows the coupling magnet to be
trained rapidly to its full current.
Refrigerator
1500 mm ID Magnet Coil
18ICST MuCOOL Milestones
- Complete the large coil test stand 6/15/08.
- Complete the two test coils 6/30/08.
- Finish winding the MuCool Coil 7/31/08.
- Complete the MuCool coil Cold Mass 9/20/08.
- Train MuCool coil cold mass assembly 9/30/08.
- Finish Mucool magnet assembly 12/02/08.
- Mucool magnet acceptance test finished 1/01/09.
- Ship Mucool magnet to Fermilab 1/10/09
19ICST MICE Magnet Milestones
- Finish winding 1st MICE coil 10/09/08.
- Finish and train 1st MICE coil assembly 11/30/08.
- Finish winding 2nd MICE coil 12/18/08.
- Finish 1st MICE magnet assembly 2/09/09.
- Finish and train 2nd MICE coil assembly 2/17/09.
- 1st MICE magnet acceptance test finished 3/02/09.
- Ship 1st MICE magnet to LBNL 3/15/09.
- Finish 2nd MICE magnet assembly 4/28/09.
- 2nd MICE magnet acceptance test finished 5/29/09
- Ship 2nd MICE magnet to LBNL 6/10/09.
20Concluding Comments
- The coupling magnet engineering is nearly done,
but there may be a few changes in the final
design based on what is learned from the two test
coils. - The schedule for the Mucool and MICE magnets is
based on things going well. The schedule doesnt
consider unforeseen problems that may be
encountered when the MuCool magnet is run with
the 1.5 W coolers for the first time.