1

4M07

[(]

Q1 for JLAB’s 12 Gev/c Super High Momentum Spectrometer

Steven R. Lassiter, Paul D. Brindza, Michael J. Fowler, Steve R. Milward, Peter Penfold, and Russell Locke

Abstract— The reference design for the first Quadrupole magnet of TJNAF’s Super High Momentum Spectrometer (SHMS), Q1, is presented. The SHMS is a dQQQD design that will be capable of resolve particles up to 11 Gev/c in momentum. Q1 follows the successful design of the High Momentum Spectrometer’s (HMS) Q1, that of an elliptically shaped super ferric yoke, conformal mapped window frame coil, and helium bath cooled coil design. The primary differences between the two designs being in the choice of superconducting cable and an overall longer magnet length. A single stack of surplus SSC Rutherford NbTi cable replaces the original four stack copper stabilized conductor used in the HMS’s Q1. The SHMS Q1 will have a warm bore diameter of 400 mm, produce field gradients up to 9.1 T/m with an effective length of 2.14 m. Test coil windings progress will be given as well as reports on forces, conductor stability and energy margins.

Index Terms— Superconductor Magnets, Detector Magnets, Cold Iron Magnets, Quadrupole Magnets

I. INTRODUCTION

T

HE planned upgrade of the Thomas Jefferson National Accelerator Facility (TJNAF) to 12 Gev/c calls for a Super High Momentum Spectrometer (SHMS) to be located within the experimental area of Hall C [1]. The SHMS is a dQQQD design using all superconducting magnets. The maximum momentum to be delivered to Hall C will be 11 Gev/c. The angle range of the SHMS is from 5.5° to 40° with a solid angle of >4.5msr. The addition of a small bender magnet and a narrow width design for the first quadrupole, ‘Q1’ was required to facilitate the SHMS reaching small angles in tandem with the existing High Momentum Spectrometer (HMS) at its smallest angle. Fig. 1 is a top view of the two spectrometers at their minimum angle.

The successful performance of the HMS’s Q1 magnet [2]-[6], an elliptical shaped super ferric quadrupole magnet based upon the conformal mapping of a window frame dipole, was chosen as the basis for the SHMS’s Q1 design. Much of the design and tooling can be reused in the manufacture of the magnet, leading to significant cost savings. Larger field gradients and availability of materials lead to a few changes to the original design that include: the length of the magnet increasing by 15%, using surplus Rutherford cable instead of the original copper stabilized superconducting cable, the elimination of the correction coils and slight increase in the thickness of the return yoke without increasing the overall magnet width. The magneto-static solver, TOSCA®, was used to model the magnetic performance as well as the forces and store energy calculations.

Fig. 1. SHMS and HMS at their most forward angles of 5.5° and 12.5° respectively.

II. Q1 Optics and Spatial Requirements

Optical and spatial requirements require a 9.1 T/m gradient with an effective length of 2.14 m and a warm bore diameter of 0.4m for the Q1 magnet. Resolution requirements require well understood and highly reproducible magnetic characteristics. Integral field harmonics up to 2.1% of the quadrupole field can be tolerated and still provide an acceptable resolution for the SHMS [7] without the need for any correction coils.

Q1’s cryostat cutout for the exit beamline will be increased to accommodate the small forward angle requirement and its longer length. To minimize the stray fields along the exit beam line requires sufficient return yoke while at the same time, allowing for clearance to the beam line at small forward angles. The majority of the flux is return in the upper portion

Fig. 2. The elliptical shaped SHMS Q1 magnet showing the notch in the cryostat for the exit beamline and the cryogenic service can on top.

of the yoke. Table 1 lists relevant magnet parameters for the Q1 quadrupole magnet.

TABLE I MAGNET PARAMETERS

Parameter / Quantity
Pole Radius / 0.250 m
Warm Bore / 0.402 m
Axial Cryostat Length / 2.44 m
Yoke Length / 2.32 m
Current Density / 18,100 A.T/cm2
Kilo Amp Turns /Pole / 255 A.T
Turns / pole / 80
Operating Current / 3188 A
Stored Energy / 0.629 MJ
Inductance / 123.7 mH
Magnet Weight / 18 tons

III.  Q1 Magneto-Static design

The yoke steel is 1006. Table II gives the magneto-static results. The large integral field gradient requirement was met by increasing both the overall length of the magnet and raising the central gradient by means of increased current. The cross sectional area of iron in the magnetic was essential unchanged. Only the width of the narrow “leg” was increased, utilizing residual space from the change in the choice of conductor. Saturation of the iron was in the range of 2.3 T within the body of the magnet. The highest field saturation occurs at the pole edges and along the end chamfering, with fields reaching up to 4.61T. The largest field within the coil, 2.78 T, occurs along the inner radius of the end turns. Iron saturation leads to no uniform properties but with field maps generated with Tosca and used in the program Snake it was verified that the optical requirements would be satisfied and understood. Fig. 3 is a plot of the field gradient at the maximum current, starting at the center of the magnet. The gradient is taken at the warm bore radius of 0.20 m. The integral multipole harmonic is given in Fig. 4, over the whole momentum range of the SHMS. The position of the single stack of conductor relative to the yoke was used to optimize the harmonic content towards the higher momentum settings.

TABLE II MAGNET RESULTS

Parameter / Quantity
Gradient Max / 9.105 T/m
Effective Field Length / 2.136 m
Peak Yoke Field / 4.61 T
Peak Coil Field / 2.78 T
Field at Pole (R=0.25 m) / 2.276 T
Momentum Range / 2 to 11 Gev/c
Integral Harmonic N=4
% of N=2 / -.04 to -1.02 %
Integral Harmonic N=6 % of N=2 / -2.21 to 0.21%
Integral Harmonic N=10 % of N=2 / -0.32 to -.10 %

Fig. 3. Field Gradient of SHMS Q1 at maximum current. Plot starts at the center of the magnet and extends out beyond cryostat. Cryostat, yoke and coil lengths are shown at the top to give perspective.

Fig. 4. Integral Field Harmonics at warm aperture over the momentum range of 2 Gev/c to +11 Gev/c. 2xn= multipole.

IV.  Conductor and stability

Surplus high current Superconducting Super Collider (SSC) outer 36 strands Rutherford cable replaces the original low current copper stabilized conductor used in the HMS’s quadrupole magnets. The SSC cable was originally key stone to an angle of 1.01°. The cable has been successfully re-flattened to within 80% of its width. Post flattened short sample test showed no signs of degradation. Table III gives the conductor characteristics. The conductor is stacked into a single layer coil consisting of 80 turns per pole. Each turn is wrapped with a 50% overlapped Kapton film followed by B-stage epoxy-glass tape to bond the turns together. The coil is wound unto its own support structure, providing a fully clamped system that also provides passages for bath cooling of liquid helium. The coil ends where wound using a near constant perimeter configuration, with the conductor being allowed to deform as it will around a cylindrical shape at the ends. Trail winding of the coil has been contracted to Scientific Magnets and the results have been successful with 1 test coil assembled to date. The cryostable design has an operational overhead for the conductor of 3.89 K, 6784 Amps and 2.91 T. The Stekly parameter has been calculated to be 0.57. The load line for Q1 is given in Fig. 5.

TABLE III CONDUCTOR PARAMETERS

Parameter / Quantity
Conductor Dimensions / 11.688 x 1.093 mm
Filament size / 0.402 m
Cu:SC Ratio / 1.8 : 1.0
Ic (4.42K and 5.69T) / 9972
Ic / Io (4.42K and 5.69T) / 3.13
Kilo Amp Turns /Pole / 255 A.T
Critical Current Margin / 6784 A
Temperature Margin / 3.89 K
Kapton Thickness / 0.10 mm
B-stage Epoxy Thickness / 0.11 mm

V. Mechanical

Mechanical design implications from increasing the cold mass length of the magnet by 15% have been studied by Scientific Magnetics [8]. Their analysis included assessing the effect of the increased cold mass on: the cold mass supports, implications for yoke build up, yoke packing density and yoke pre-load, magnet sag issues, radiation shield support, increase in cryogenic heat loads as well as manufacturing and cost implications. Their study concluded that sufficient margins were found to exist with the original design to safely handle the expected mechanical loads. Their report also concluded that the cryogen safety relief devices need to be scaled or be capable to accommodate a pressure 15% higher then original design in the HMS magnet. The 5KAmp “No Burnout” current leads [10] are expected to be the largest single source of heat for the magnet at 9W at full current. The heat load to 4K is estimated to be less then 20W and less then 30W to the LN2 shields. The report also found no difficulties with the manufacturing aspects due to lengthening of the cold mass.

E/M forces were also calculated independently at JLAB using Tosca and then loaded into a two dimensional FEA model. Stresses within the yoke were found to be below 41MPa with maximum deflections less then 4x10-5 m and are shown in Fig. 6.

Fig. 5. Load Line data for the SHMS Q1. BI curve is nonlinear due to saturation of iron. Measured data is from the flattened SSC cable.

Fig. 6. Stress and deflection on one quadrant of the iron yoke due to magnetic forces.

VI.  Conclusion

The first quadrupole magnet of the SHMS, Q1, has undergone and passed several in house and DOE technical reviews [11]. The continuing successful operation of the HMS’s Q1 magnet, along with the longer magnet length proves that the current design can easily satisfy the tight spatial constraints required by the SHMS as well as achieving the optical requirements over the whole momentum range. The Rutherford cable design indicates that it will remain cryostable and that training is unlikely provided that adequate mechanical support is provided. Trail windings using the flattened Rutherford cable are ongoing at Scientific Magnetics.

Fig. 7. Trail winding setup. Picture courtesy of Scientific Magnetics.

Fig. 8. End turn geometry, showing a ten turn trail winding layup. Picture courtesy of Scientific Magnetics.

Acknowledgment

The authors gratefully acknowledge the assistance of Dr.
Bruce Strauss (USDOE/OHEP), Dr. Ron Scanlan (LBNL) and Dr. Dan Dietderich (LBNL) for providing the superconductor from the USDOE/Office of High Energy Physics equipment surplus.

References

[1]  The Science and Experimental Equipment for the 12 GeV Upgrade of CEBAF, Jan. 2005, http://www.jlab.org/12GeV/development.html.

[2]  L. H. Harwood et al, “A Superconducting Iron-dominated Quadrupole for CEBAF”, IEEE Transaction on Magnetics, vol. 25, Mar 1989, p. 1910.

[3]  S. R. Lassiter et al, “Large Aperture Superconducting Cryostable Quadrupoles for CEBAF’s High Momentum Spectrometer”, IEEE Transaction on Magnetics, vol. 27, Mar 1991, p. 118.

[4]  S. R. Lassiter et al, “Final Design and Construction Progress for CEBAF’s Cold Iron Quadrupoles”, IEEE Transaction on Applied Superconductivity, vol. 3, Mar 1993, p. 118.

[5]  S. R. Lassiter et al, “Magnetic Measurements of Large Aperture Superconducting Magnets for TJNAF’s High Momentum Spectrometer”, IEEE Transaction on Applied Superconductivity, vol. 7, June 1997, p. 614.

[6]  P. D. Brindza et al, “Commissioning the Superconducting Magnets for the High Momentum Spectrometer (HMS) at TJNAF”, IEEE Transaction on Applied Superconductivity, vol. 7, June 1997, p. 755.

[7]  John J. LeRose, JLAB, Newport News, VA, “Optics Study of SHMS using Raytrace and TOSCA fields”, private communication, January 2007.

[8]  Study of the SHMS Q1 with 15% longer cold mass. Technical Report Ref No. E165-01 SW1, Sept. 2006.

[9]  P.B. Brindza et al, “The Cosine Two Theta Quadrupole magnets for the Jefferson Lab Super High Momentum Spectrometer”, IEEE Magnetic Technology, submitted for publication.

[10]  12 GeV Upgrade Project Conceptual Design and Safety Review of Superconducting Magnets, JLAB, Newport News, VA, Sep. 2006.

[(]Manuscript received August 30, 2007.

Authored by Jefferson Science Associates, LLC under U.S. DOE Contract No. DE-AC05-06OR23177. The U.S. Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce this manuscript for U.S. Government purposes.

Steven R. Lassiter, Paul D. Brindza and Mike Fowler are with Jefferson Science, Newport News, VA. 23606 USA (phone: 757-269-7129; fax: 757-269-5520; e-mail: ).

Steve R. Milward, Peter Penfold and Russell Locke are with Scientific Magnetics, Abingdon, OX14 3DB, UK (phone: +44 (0) 1865 409200; e-mail: )