R. H. M. R. D. Lag, T. G. S. R. L. Wong, J. P. N. L. D. P. W ......Figure 2. TF Coil for TPX (Room Temperature) In our design we use the formulation of Summers - et d.[SI to predict

UCRGJC-116947 PREPRINT

TPX Magnet System Status

R. H. Buher, M. R. Chaplin, D. D. Lag, T. G . O'Connor, D. S. Slack R. L. Wong, J. P. Zbasnik, N. Diatchenko, L. Myatt, R. D. Pillsbury,

J. H. Schultz, P. W. Wang, J. C. Citrolo

This paper was prepared for submittal to the 18th Symposium on Fusion Technology

Karlsruhe, Gemany August 22-26,1994

WSTRIBUTION OF THIS DOCUMENT IS UNLiMiTED

TPX Magnet System Status

€2. H. Bulme?, M. R Chapha, D. D. Lane, T. G. O'Conno?, D. S. Sad?, R L. Wonga, J. P. Zbasnika, N. Diatci~enko~, L. Mya&, I$ D. Pillsburyb, J. H. Schulbb, P. W. Wangb, J. C. Citroloc.

a w e t i c Fusion Energy Program, Lawrence Livermore National Laboratory, 7000 East Avenue, Livermore, CA 94550, USA

bPlasma Fusion Center, Massachusetts Institute of Technology, 185 Albany Street, Cambridge, MA 02139, USA

cF3inceton Plasma Physics Laboratory, Princeton University P.O. Box 451, Princeton, NJ 08543, USA

We present a status report on the magnet system for the Tokamak Physics eXpe.rimex WX), a machine with a major radius of 225 m and a minor radius of 05 m to be built at the Princeton Plasma Physics Laboratory. in which al l main coils will use cable-in-conduit superconductors. The 1 O i l toroidal field system must produce a 4T field at the plasma center (8.4 T peak field) and accommodate about 5 kW of steady-state heating from nuclear heating, eddy currents, and thermal radiation in the windings. The poloidal system provides a plasma initiation voltage of 20 V and a total flux swing of 18 Wb to ramp the plasma current to 2 MA and provide a short flat-top. The poloidal system consists of 14 individual coils arranged symmetrically above and below the machine midplane, connected to allow either double-null or single-null plasma configurations.

1. INTRODUCTION

Until now, the design of the superconducting magnet system for TPX has been carried out in a collaborative effort between Lawrence Livennore National Laboratory (JLNL,), the Plasma Fusion Center at the Massachusetts Institute of Technology (MIT), and the Princeton Plasma Physics Laboratory (PPPL). We have recently awarded the preliminary design and R&D contraa for the TF system to a team of Babcock & Wilcox and General Atomics and a similar contract for the PF system to a team of Westinghouse Electric Company, Northrup- Grumman, and Everson Electric. These two firms will be in competition for the final design and fabrication contracts of the entire magnet system. The conductor is being developed by the Laboratory team of LLNL, MIT, and PPPL and will be supplied to the eventual winner.

2. TF MAGNET DESIGN STATUS

Previous versions of the magnet design are described elsewhere [l-31; Figure 1 shows a CAD drawing of the complete magnet system. The 'IF coil pack, shown in Figure 2, will be pancake wound from a continuous 1.1 km length of cable-in- conduit-conductor to have a compact winding

package with the absolute minimum number of joints.

CENTRAL SOLENOID ASSEMBLY, Pfl-4 U8L

Figure 1. TPX Magnet System

The TF conductor, shown in Figure 3, bas a copper:noncopper ratio of 2.5 and an operating

DISCLAIMER

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, make any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

DISCLAIMER

Portions of this document may be illegible in electronic image products. fmages are produced from the best available originaf document.

Figure 2. TF Coil for TPX (Room Temperature)

In our design we use the formulation of Summers - et d. [SI to predict the critical current as a function of

field a d temperature; our parameters: T c h = 16 K, B c 2 h = 275 T, and Co = 10560 A-T/m2, describe a conductor that can hopefully be easily produced.

We use a'&d, then react process, with Incoloy 908 shims used in place of the insulation during reaction. Each coil has 12 pancakes in the toroidal direction, and each pancake has 7 turns. After reaction, the turns are separated, the spacers removed, the insulation applied, and the turns repositioned. The insulation is a 0.7-mm thick S-glass wrap around each turn with a 0.4-mm thick spacer between turns and pancakes. After impregnation under vacuum and pressure, the ground plane insulation is applied. The coils are shimmed into place in their 316LN coil cases, the cover pieces welded in place and the coils potted to ensure good mechanical contact between coil and the structure.

Figure 4 shows the present mechanical configuration of a right hand and a left hand 2-coil

module separated by an eddy current break at the wedging face in the nose region and a bolted eddy current break at the outer radius. The 4 - d modules are assembled into the tokamak configuration by welding; the magnet weld prep anxis allow access to the sections of the vacuum vase1 that need to be welded during machiie assembly.

Figure 3. TPX Conductors

The intercoil structure consists of 316 LN shell plates and toroidal and poloidaI ribs to react the Out of p h e loads. A 3 d global analysis 161 was performed for the worst normal operating point and the stresses satisfied the static allowables for welded 316LN. A detailed 2 4 calculation performed on the winding pack in the nose region reveals regions of high stresses in the insulation regions between corners of the conduits that appear to be due to t h e m 1 contraction mismatches [7]. This problem is presently being addressed.

For maintainabilty, all the supercritical helium cooling connections are all on the outer radius, so each individual helium circuit is a double pancake.

Each circuit has an electrical isolator, a flowmeter, and a Ne, the flow is outwards in the outer pancakes since the heating is highest for these outer pancakes and we also want to flush away the joint heating. The design thetmal load on the TF system is about 12.1 kW, including 10.7 kW from nuclear heating and 1.4 kW from eddy current heatihg from the vertical stability coils. Of the 10.7 kW of nuclear heating, 4 kW is deposited directly in the windings and another 0.7 k W is deposited indirectly from thermal diffusion of the heating deposited in the coil structure.

EDDY CURRENT BREAK @ TF WEDGING FACE g INSULATION

RIGHT HAND -y- 2-COIL MODULE

M O L T E D EDDY CURRENT BREAK

t..J

L E F T HAND 2-COIL MODULE

Figure 4. >Coil Modules for TPX

Our thermal analysis 181 shows that for steady state 0perati0~. with a total TF flow of 450 g/s of helium at an inlet temperature of 5 K inlet and an outlet pressure of 3 atm the peak helium t emptu re will be 6 K which provides a calculated temperature margin (TCS - Tbath) of 2 K, which exceeds the temperature margin of 1 K required in the TPX Structural and Cryogenic Design Criteria 191. We also find that we satisfy our other criteria of:

Fraction of critical current < 0.6, Effective heat transfer coefficient

< 1000 W/&K for ~ b 3 ~ n < 600 W/m2-K for Nb-Ti,

Temperature headroom (Ta - Tin) > 2 K. We plan to connect the TF coils electrically in

series so we won't have unbalanced currents, but with two pairs of Ieads to allow us to discharge with

two interleaved dump resistors in case of quench. This allows a peak terminal voltage of 7.5 kV, with a peak voltage to ground of 3.75 kV. Our quench detection methods are sti l l in the development stage, but we will choose from the suite of sensors which include cowound voltage leads in the cable space, fiber optic sensors to measure temperatwe rise in the cable, and flowmeters on the helium inlets and outlets to look at flow reversal and enhancement IlOJ. Initial calculations 1111 show that for a 0.1 m initial quench zone, a 1 second delay time, and a system discharge voltage of 15 kV, the peak cable temperature is 151 K, which meets our design criterion For a 4-m long quench zone, we have a peak pressure of 42 am.

3. PF MAGNET DESIGN STATUS

The PF coil positions and dimensions were determined by an optimization which minimized a PF cost function subject to a set of contraints. The cost function depends on coil properties, volume, field, and current, and is changed by varying the position, size, and shape of the winding set. The constraints are field contours which satisfy previous MHD equiIibrium analysis, physical s t aysu t zones dictated by other features of the machine, and the TPX magnet design criteria 1121. Engineering detail was added to the resulting configuration, and computer runs were made to ensure that the MHD equilibria were satisfied over the intended operating mge.

The inner leg of the TF system was shifted outward by about 0.1 m in a recent design iteration to reduce the peak field on the TF windings and the outer diameter of the central solenoid was increased which resulted in a more efficient, i.e. thinner, central soIenoid. This, coupled with the shape changes of the other PF coils which tended to reduce the peak conductor fields, results in a vastly improved PF coil set over that which was originally proposed The PF magnet conductors are also shown in Figure 3.

The room temperature dimensions of the PF system coils are presented in Table 1, where 4 and ZC are the radial and vertical centerline dimensions and the radial and vertical thicknesses include the 8- mm thick ground plane insulation.

We intend to wind each coil pancake-fashion using a continuous length of conductor to conserve space and minimize the number of joints. This may not be practical, however, for PF6 and PF7 where the conductor lengths are 2.83 km and 2.27 km, respectively and we may have cabler limitations.

The central solenoid assembly is preloaded with the stainless steel structure to ensure that the coils remain under axial compression. Links from the preload structure support the coils from the top of the TF inner leg and allow radial displacements and rotation of the coil stack. We assume that the PF5, PF6, and PF7 mils will be mounted on slip planes to alleviate stress buildup due to differences in thermal contraction behveen thesecoils and the supporting TF Structure.

A variety of startup scenarios have been considered thus far, including 12 V and 20 V breakdown voltages with field nulls at 2.47 m and 2.25 m [13]. The resulting maximum coil operating conditions for these are shown in Table 2.

Table 2. Peak Conditions During Startup /scenario I 1 2 v , I 2ov. I 12 v. I

2.47 m 2.25m I 2-25 A* I, kA 24.1 24.5 1 27.5

IPF2,PF3 I PF3 1 PF3 PairV, 1 -5.31 I -8.91 I -6.13 I

I fc I 0.59 I 0.59 I

I PF3 dB/dT, I 10.4 1 16.1 I 19.4

I PF3 I PF2 1 PFI I * This does not include the cryostat.

The present design satisfies the TPX design criteria for our reference operating scenario 1141, except that when the specified PF coil nuclear heating is included, we violate the temperature headroom requirement for PF6 by about 0.1 K. P; there is no problem with PF7 if it is cooled in a single-pancake fashion.

From a protection standpoint, PF5 is the coil expected to have the highest hot spot temperatun?; assuming a 1-sec delay time, simulations result in a peak cable temperature of 64 K for a 0.1 m long quench zone and a peak .pressure of 32 atm if the complete high-field turn quenches[151.

REFERENCES 1.

2. 3.

4.

5.

6.

7.

8.

9.

W.V. Hassenzahl et al, IEEE Trans Magnetics, vol30, No. 4, July, 1994, p. 2058. Joel H. Schultz et al., Proc. 1993 SOW p. 788. DwightB. Lang et al, "The Superconducting Magnet System for the Tokamak Physics Experiment", Proc 1994 ANS, New Orleans,. H. Tsuji et al., proc. of 1 lth International Conf on Magnet Technology, Aug 1989, p.806. LT. Summers et aL, IEEE Trans on Magnetics, vol27, No. 2, Mar 1991, p 2041. RL. Myatt, "Structural Evaluation of TPX TF Coil System Modified Case Transition Region",

RL. Myaa, "Effects of a Slip Plane Around TF Conduit on Stresses in Glass-Epoxy Corners," TPX No.l3-940803-MlT/L,Myatt-01.

the New 1.25 Aspect Ratio DPC-U1 Conductor,"

P. Heibenroeder, "TPX Structural and Cryogenic Design Criteria," TPX No. 91-921012- PPPLiPHeitzenroeda-01.

TPX N0.13-940708-MlT/LMyatt-01.

P.W. Wmg, "TF'X Steady-State Heat Removal of

TPX NO. 13-940601-MlT/PWang-01.

IO. J. H. Schultz, and S. Poutrahimi et al., " Novel Quench Detection Methods for the Superconducting Magnets in ITER and TPX," this conference.

1 I. J. H. Schultz, " QuencNDump Simulations of ECT TF Magnet," TPX No. 12-940715-

12. D. S t r i d e r and R. Buulmer, "PF Coil Optimization", TPX No. 14-940311- ORNL/DStriCkler-Ol.

13. P. Wang, presentation at TPX Technology Transfer Meeting, LLNL, Aug 9-1 1,1994.

14. J. H. Schultz, "HC Scenario Dynamic Simulation

15. J. H. Schultz, "QuencNdump Simulations of ECT

MIT/JHSchultz-0 1.

-IV", TPX NO. 14-940518-MIT/JSch~ltz-01.

PF Magnets", TPX No. 14-940721- MIT/JSChuIb-Ol.

* This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under contract number W-7405- Eng-48, and Princeton Piasma Physics Laboratory under contract number DE-ACO2-76-CHO3073.