Joseph F Muratore (Magnet Division)
6-June-2017
Revised 20-June-2017
Revised 7-10-July-2017
Revised 11-July-2017
Revised 2-Aug-2017
sPHENIX Solenoid Magnet High Field Test Plan Summary
INTRODUCTION
The sPHENIX (formerly BaBar) magnet is a solenoid consisting of two layers (inner and outer) wound inside an aluminum support cylinder with a Rutherford style Cu/NiTi superconducting cable embedded in an ultrapure aluminum matrix/stabilizer. The cable consists of 16 strands with Cu: SC ratio >1:1. Both layers have graded current density in order to supply field uniformity in the tracking region of± 3%. This is done by using cable thickness of 8.49 mm in the central region and 4.93 mm in the ends. Otherwise the cable is the same with a width of 20 mm. The magnet is assembled into its own cryostat and can be cooled down to 4.5 K by conduction with piping filled with liquid He supplied by the Building 912 cryogenics facility.
This magnet has been tested in Building 912 at 4.5 K with no flux return iron and to a current of 100 A. The results were successful and can be reviewed in the technical note “Low Field Test of sPHENIX Solenoid Magnet” by J. Muratore et al. For this second test, also in Building 912, the magnet will be encased in flux return iron slabs so that this test can be done to a maximum current of 4830 A (operating current 4596 A + 5% margin).
As for the previous low field test, the solenoid coil and the leads are equipped with voltage taps to monitor the voltages for the purpose of quench detection and quench analysis (if necessary). There are also temperature sensors strategically placed for cryogenics monitoring purposes. The locations of the voltage taps and also the temperature sensors are shown in the following schematic. There are also installed strain gauges and linear potentiometers to monitor changes to the structure during cooldown/warmup and powering ramps. A Hall probe system will be used to measure the central field.
From C. Schultheiss
As can be seen in the power supply/electrical system flow chart below, the magnet will be protected by a 68 mΩ dump resistor, with which at least 50% of the magnet stored energy (27MJ at 4596 A) can be extracted externally in the case of a quench. Magnet excitation at 4.5K is done by a 5000 A/ 20 V power supply. The center point of the dump resistor is tapped to ground, and this decreases the total voltage by half what is seen to ground. A maximum internal voltage of about 340 V may be generated between any two voltage taps, but only 170V will be seen with respect to ground. The schematic shows that this ground current will pass through a 100 Ω resistor across which the ground current may be measured.
From C. Schultheiss
At quench, switches will open, allowing the current to pass through only coil and dump resistor. This results in a fast discharge, compared to a slow discharge, which is attained by allowing the current to pass through a free-wheeling diode and not the dump resistor. The latter is more suitable for the various safety interlocks that could trip but where the magnet has not quenched. The fast discharge, which switches in the dump resistor, is appropriate for a quench event and is tripped by a quench detector that has two modes of detection, depending on which of two signals pass a set voltage threshold (50 – 200mV) after a specified validation time interval (1-10ms). One signal is the voltage difference between the inner and outer layers, after applying a small scaling factor to the inner layer to compensate for its slightly smaller inductance. On the unlikely but not impossible event of a simultaneous quench in both layers so the difference signal does not reach the threshold, a second signal is also monitored by the quench detector, and this is the difference between the total coil voltage and the calculated inductive voltage ½ L(dI/dt), where L is the inductance and dI/dt is the ramp rate.
Note that in the event of a fast discharge, even without a quench having occurred, as in the case of a manual trip of the power supply at a specified current, the coil may experience quenching due to heat generated by eddy currents induced in the aluminum support cylinder. Such fast discharge tests will be performed in order to validate the operation of the quench detection system so heating is expected. Such quenches are over large percentage of the magnet coil and will not be a risk to magnet safety. This type of a quench is called a quench-back, and does provide a further protection for the coil in addition to the dump resistor energy extraction. We can expect a quench-back will result in a 37 K uniform temperature increase in the coil and several hours will then be needed to cool the magnet back to 4.5K.
As has been discussed by C. Schultheiss at the safety review for this test, the following safety controls have been put into place:
•ELECTRONIC EQUIPMENT RACKS ARE OUTSIDE THE 5 G LINE.
•CRYOGENICS SYSTEM INTERLOCKS
–Cryo system is ready and at temperature
–Gas-Cooled Lead Flow interlock - GCL Voltages
•QUENCH SYSTEM INTERLOCKS
–Quench System not in Quench Mode
–Energy Dump Switch is closed
–Quench Detector Rack is closed
•POWER SUPPLY INTERLOCKS
–Water Flow Interlock
–SCR Fuse Interlock
–SCR Over-Temperature Interlock
–Transformer Temperature Interlock
–Cabinet Temperature Interlock
–Control Power Interlock
–AC Phase Loss
–AC Input Over-Current
–AC Current Imbalance
–AC Sequence Interlock
–DC Output Over-Current
–Output Filter Fuse Interlock
–Power Supply Enclosure Door Interlock
–Water Leak Detection
–Ground Fault Detection
–DCCT Fault Detection
All interlock indicators must be validated before the test and monitored during the test.
Information from previous tests.
SLAC acceptance testing (4826 A, with iron flux return):
At 1.7 A/s, 0.1 K temperature rise
At 2.7 A/s, 0.3 K temperature rise
At quench-back, 37 K temperature rise.
At 4826 A, current held constant for 14 hours without incident.
Inductance 2.573 H (measured at 1 A/s); computed 2.56 H
Low Field Test (100 A, no iron flux return):
Inductances measured at 3 A/s:
Outer Coil L = 1.080 H
Inner Coil L = 1.057 H
Total magnet L = 2.137 H
Quench Detector (voltage difference) threshold = 100 mV (at 3 A/s)
Validation Time Period = 5 ms
sPHENIX Solenoid Parameters
Winding Axial Length 3512 mm at room temp
Winding Mean Radius 1530 mm at room temp
Operating Current 4596 A
Operating Central Field 1.5 T (at 4596 A)
Design Current 4826 A (1.58 T) (5% margin)
Operating Stored Energy 27 MJ (at 4596 A)
Winding structure: 2 layers with graded current density
Uniformity in tracking region ± 3%
Dump resistor 68 mΩ
TEST LIMITS
No polarity change.
Maximum test current ≤ 4850 A
Maximum test ramp rate < 3.3 A/s
Maximum coil internal quench voltage (between any two voltage taps) 326 V (@4800 A)
Maximum quench voltage to ground 163 V (@4800 A)
TEST PLAN
A. Electrical Checkout at Room Temperature
1. Measure lead resistances to ground with meter. (Verify R ≥ 20 MΩ)
2. Hipot coil off ground at 520 V (magnet leads). (Verify Ignd ≤ 10 μA)
3.Impulse test at 400 V of separate layers and full coil.
4. DC voltage series resistance measurements of voltage taps with 1 A.
5. 1 A level shift test to check data acquisition system.
6. Perform R-L-Q test at room temperature.
7. Connect magnet to power supply.
8. Ramp magnet to 2 A at 0.1 A/s and measure magnet total voltage, and inner and outer
layer voltages in order to determine the inductances. Determine scaling factor to balance
inner and outer inductances
9. Checkout of quench detector (QD)system - validate stop signal
a. Half Coil Difference (Δ)– inner and outer layer voltage difference
b. Current Derivative (Idot)– total coil voltage and L(dI/dt) difference
10. Vacuum leak check of cryostat.
11. Set slow logger for 1 min sampling intervals.
Monitor LHe level, temperatures, and strains during cooldown.
12. Notify cryogenics personnel to start cooldown.
13. Disconnect magnet from power supply.
B. Cooldown to 4.5 K
Monitor LHe level, temperatures, and strains during cooldown. Slow logger at 1 min.
C. Electrical Checkout at 4.5 K
NOTE: First 5 tests can be done when temperature < 7 K. Hipot must be done at 4.5 K.
1. Measure lead resistances to ground with meter. (Verify R ≥ 20 MΩ)
2. Check voltage taps continuity at patch panel. (Each tap should read nominally 200 Ω)
3. 1 AAC measurements of total coil, inner layer, and outer layer.
4. Impulse test at 400 V of separate layers and full coil. Compare to room temperature
impulse test.
5. Perform R-L-Q test at 4.5 K.
6. Hipot coil off ground at 520 V (magnet leads). (Verify Ignd ≤ 10 μA)
7. Connect magnet to power supply. Use inductance results of Part A.8 as starting point for
power supply regulation and initial quench detector settings.
8. Set fast data logger at 1 kHz sampling rate (1 ms intervals).Set slow data logger at 1 s
intervals.During the tests, monitor LHe level, temperatures, strains, and voltages of
magnet, coils, SC leads, and gas-cooled leads. Adjust lead flow if necessary.
9. Ramp to 50 A at 1 A/s. Determinethe total, inner, and outer inductances and adjust Idot
QDand Delta QD scaling factor as necessary to balance for the difference in the inner and
outer layer inductances; determine the Idot QD voltage threshold. Adjust lead flow if
necessary.
10. Ramp to 100 A at 1 A/s.Adjust lead flow if necessary.Ramp back down to 0.Verify results
from C.9.
11. Ramp to 100 A at 1 A/s. Verify results from C.8. Shut off power supply (manual QD trip)
andanalyze voltage tap data signals for proper operation of instrumentation.
D. Power Supply Shutoffs
1. Ramp to 1000 A at 2 A/s in 100 A steps. Hold at 1000 A for 5 min. Monitor LHe level,
temperatures, strains, and voltages of magnet, coils, SC leads, and gas-cooled leads.
Adjust lead flow if necessary.
Shut off power supply (manual QD trip) and analyze voltage tap data signals for proper
operation of instrumentation and hardware.
Adjust sampling rate if necessary.
2. Ramp to 1000 A at 2A/s. Verify inductances. Adjust QD thresholds if necessary.
Ramp to 2000 A at 2 A/s in 100 A steps.Hold at 2000 A for 5 min. Monitor LHe level,
temperatures, strains, and voltages of magnet, coils, SC leads, and gas-cooled leads.
Adjust lead flow if necessary.
Shut off power supply (manual QD trip) and analyze voltage tap data signals for proper
operation of instrumentation and hardware.
Adjust sampling rate if necessary.
E. Ramp Testing
Set the Delta QD at 50 mV. Set Idot QD at previously determined threshold.
Fast data at 1 KHz. Slow logger at 1 s. 2 A/s ramp rate.
Monitor LHe level, temperatures, strains, and voltages of magnet, coils, SC leads, and gas-
cooled leads. Adjust lead flow when necessary.
If QD trip occurs, analyze voltage tap signals to determine nature of trigger event.
If signals show no anomalies, repeat test.
1. Ramp to 2000 A at 2A/s. Verify inductances. Monitor SC and gas-cooled leads. Adjust lead
flow if necessary.
Ramp to 3000 A at 2 A/s in 100 A steps.Monitor SC and gas-cooled leads. Adjust lead
flow if necessary. Adjust QD thresholds if necessary.
Ramp to 4000 A at 2 A/s in 100 A steps.Monitor SC and gas-cooled leads. Adjust lead
flow if necessary. Adjust QD thresholds if necessary.
Ramp to 4830 A at 2 A/s in 100 A steps.Monitor SC and gas-cooled leads. Adjust lead
flow if necessary. Adjust QD thresholds if necessary.
2. Ramp to 2000 A at 2.5 A/s. Adjust QD thresholds if necessary.
Ramp to 3000 A at 2.5 A/s. Adjust QD thresholds if necessary.
Ramp to 4000 A at 2.5 A/s. Adjust QD thresholds if necessary.
Ramp to 4830 A at 2.5 A/s. Adjust QD thresholds if necessary.
3. Five (5) power cycles from 0 A to 4830 A at 2.5 A/s. Verify slow logger at 1 s.
4. Ramp to 4830 A at 2.5 A/s. Hold for 12 hours at 4830 A. Monitor leads and coil voltages,
among the other parameters listed above.
5. (Optional) Repeat ramp and power cycles to 4830 A at 3.0 A/s. Increase QD thresholds
accordingly.
F. Discharge Tests
Fast logger at 1 kHz or faster. Slow logger at 1 s.
1. Ramp magnet to 4830 A at 2.5 A/s. Hold for 15 min.
Manual power supply trip with dump resistor out of the circuit (slow discharge through
free-wheeling diodes). Analyze fast and slow logger data.
2. Ramp magnet to 4830 A at 2.5 A/s. Hold for 15 min.
Manual power supply trip with dump resistor in circuit (fast discharge through
free-wheeling diodes). Analyze fast and slow logger data. Note that magnet will develop
quench-back during fast discharge.
3. Ramp magnet to 4830 A at 2.5 A/s. Hold for 15 min. ramp back down to 0.
Shut down power supply.
4. When tests have been completed, set slow logger for 1 min sampling intervals.
Monitor LHe level, temperatures, and strains during warmup.
5. Notify cryogenics personnel to start warmup to room temperature.
G. Warmup to Room Temperature
Monitor LHe level, temperatures, and strains during warmup. Slow logger at 1 min.