Magnet Issues 1/19 Nov 2013 MICE MAGNET ISSUES Brief update on –Coupling Coils (CC) –Spectrometer Solenoids (SS) Both are US responsibilities Details of Focus Coil –Training FC #1 –Status FC #2 UK responsibility SS CC CC SS FC FC FC

Magnet Issues 2/19 Nov 2013 COUPLING COILS First CC cold mass recently tested in standalone cryostat Quenched at low current (62 A cf ~200 A) Heat load on cold mass ~ 75 W – too high  Improve MLI & thermal intercepts; expect to test again before end of year New gas return pipe routing New thermal intercepts

Magnet Issues 3/19 Nov 2013 SPECTROMETER SOLENOIDS Chequered history – refurbished and more cryocoolers added SS #2 trained to full current + 2% in February 3D mapped in June 2013 Arrived at RAL early October

Magnet Issues 4/19 Nov 2013 SS #1 Being prepared for testing / training in US HTS leads showed resistive behaviour – traced to mechanical problems at thermal intercepts 8 HTS leads replaced; joints strengthened + additional temperature sensors Training expected to start ~18 Nov; mapping in December

Magnet Issues 5/19 Nov 2013 SS #2 in R9 Arrived early October Thermal intercepts must be inspected

Magnet Issues 6/19 Nov 2013 FOCUS COILS FCs different in many details from SS and CCs OXFORDRALDARESBURY V. BlackmoreV. BaylissT. Hartnett J. CobbT. BradshawS. Griffiths W. LauM. CourtholdI. Mullacrane R. PreeceP. Owens M. TuckerC. White S. Watson

Magnet Issues 7/19 Nov 2013 FC Modules contain two coils which can be operated in: –‘Solenoid mode’ = same polarity –‘Flip mode’ = opposite polarity Flip mode is the more demanding: –2 x higher currents –2 x higher fields at conductors –4 x higher forces and stored energy Still responsibility of manufacturers Indirect cooling

Magnet Issues 8/19 Nov 2013 FC #1 TRAINING FC #1 trained easily to 114 Amps in Solenoid Mode Reached Amps in Flip Mode –Gnat’s whisker above 200 MeV/c operating current of A –Seems to have reached ~ its limit Run Current (Amps)

Magnet Issues 9/19 Nov 2013 FC #1 HISTORY 16 training runs –stopped to disconnect two weeks before FC #2 due to arrive Baseline design in specification was for 200 MeV/c –All design calculations – forces &c – for 240 MeV/c –Just reaching 188 A (200 MeV/c) leaves no overhead We know: –Quenches do not originate in HTS leads or LTS tails –15/16 times Coil 1 quenched first Curious Examine –Temperatures –Coil 1 – 2 differences –Voltages

Magnet Issues 10/19 Nov 2013 CRITICAL TEMPERATURES Cryocoolers have excess cooling power at 4.2K Temperatures stable until quench Temperature margins recently re-calculated in connection with PRY 200 MeV/c; 240 MeV/c – as expected Coils should operate at 200 MeV/c with overhead to 240 MeV/c He Channel 4.37 K Cold mass 4.57K Current (AC losses ~0.05K) Heater on/off

Magnet Issues 11/19 Nov 2013 COIL 1 – COIL 2 DIFFERENCES ? Coil 1 quenched first 15/16 times Any evidence of different thermal contact?  Monitor S/C transition during cool-down after quench Days after quench R(coil), Ohms Coil 1 Coil 2 Both coils become superconducting within 1 minute ~1.5 days after quench  No obvious evidence of different thermal properties

Magnet Issues 12/19 Nov 2013 FIELD MEASUREMENTS Basic jig + Hall probe to measure field on axis Measured field within ~1.5% of calculated field at 50 A ~0.5% agreement between coils at coil z-centres  Exclude short-circuited layers in one coil I = 50 Amps

Magnet Issues 13/19 Nov 2013 VOLTAGE MONITORING Q. P. system uses voltage differences to avoid noise triggers Voltage differences recorded with 16-channel ‘Picologger’ –10 mSec samples –Channels sampled sequentially Voltage differences somewhat difficult to interpret… Nevertheless…

Magnet Issues 14/19 Nov 2013 PRECURSORS Amplified difference voltages  Precursors to quench Picologger can’t see coincidences of short transients Average of 7 precursors / quench – true rate higher Frequency & amplitude increase towards time of quench in ( ~ last 5 Amps) Suggest ‘stick-slip’ movements of coils Quench Precursors ~15 minutes Time before quench (min) Amplitude (mV) (Largest precursor) Quench

Magnet Issues 15/19 Nov 2013 STICK – SLIP Quench Current Coils move against small irregularities – stick then slip Mechanical energy released in a slip may cause quench Irregularities slowly ironed out F ~ A; bobbin stretches by ~ 100 microns F Quench Fh Bobbin COIL

Magnet Issues 16/19 Nov 2013 NEW VOLTAGE LOGGING Picologger records showed something interesting –But difficult to interpret & too slow New voltage logging system built by Daresbury team –16 channels of National Instruments ADCs –Records voltages between adjacent voltage taps Not differences –1 mSec sampling –Stores ~ 8 minute window of data around time of quench Operational for last run of FC #1

Magnet Issues 17/19 Nov 2013 LAST QUENCH OF FC #1 Coil voltages run away over 10’s of mSec  Growth of normal (resistive) zone in coils – as expected Contactor opens Detector threshold ~0.6 V reached

Magnet Issues 18/19 Nov 2013 Opposite polarity precursors on each coil ~2.5 minutes before quench Estimate energy  U ~ 0.61 J ~ few 100 x MQE of 1 – 3 milliJoules Didn’t propagate as a quench  not resistive  movement of coils (?) Movement of 0.3 microns is enough May/may not cause quench depending on where heat deposited 100 mSec

Magnet Issues 19/19 Nov 2013 Temperatures, fields and thermal behaviour of coils show no obvious anomalies Evidence for small stick/slip movements of coils –Well-known training pathology Could also be cracking of epoxy matrix –Also a well-known training pathology Need destructive post mortem to confirm (?) Whole business is a Dark Art –Sensitive to things that are very difficult to control –No obvious simple fix External coil to pick up movements being considered WHAT HAVE WE LEARNT? “Premature quenching can also occur if the large forces between coil sections result in the motion of one coil with respect to another. This is most likely to occur in magnets having coils that are wound in opposition” American Magnetics (

Magnet Issues 20/19 Nov 2013 FC #2 STATUS FC #2 delivered to R9 on 4 October and now cooling down Anticipate starting to train at end of November –Faster turnaround Cryocoolers have better insulation between stages 1 & 2 After many discussions with manufacturers agree that the fate of FC #1 will depend on experience with FC #2

Magnet Issues 21/19 Nov 2013 THE END