Redmond Plasma Physics Lab
seperator

The Translation, Confinement, & Sustainment (TCS) FRC Experiment

Alan L. Hoffman & John T. Slough
Presented to US-Japan Workshop-Physics Basis of D 3He Fusion, Niigata, Japan 1996

ABSTRACT

A new Translation, Confinement, and Sustainment (TCS) FRC facility is being constructed to investigate; 1) the effects of external field shaping on FRC confinement and stability and; 2) the ability to use Rotating Magnetic Fields (RMF) to both build up and sustain the flux of FRCs that have been formed in Field Reversed Theta Pinches (FRTP). TCS will utilize the modified LSX facility (LSX/mod) to form and translate low density, hot, mWb flux level FRCs into an LSX sized (80-cm diameter by 3-m long) quasi-steady confinement chamber, where the effects of field shaping can be studied, and 50 G level rotating fields applied.

 

Picture of TCS

 

I. INTRODUCTION

LSX/mod consists of a 2.5-m long, 40-cm plasma tube diameter FRTP formation chamber and a 2-m long, 27-cm diameter, 4-stage coaxial acceleration/translation section. It has been utilized for the past 2 years to perform Tokamak Refueling by Accelerated Plasmoid (TRAP) experiments, and has demonstrated the acceleration of high density, 0.7 mg FRCs to velocities sufficient for penetrating JET magnitude toroidal fields. FRCs formed in LSX/mod will be translated and expanded into TCS proper, which will consist of an approximately 3-m long, 90-cm diameter set of multi-turn quasi-steady magnet coils. The 80-cm diameter quartz plasma tube sections used previously for LSX will be used for the TCS vacuum chamber in order to facilitate antenna design for the RMF current drive system. The TCS axial magnetic field system will include a great deal of flexibility to allow for investigation of the effects of external separatrix shaping on FRC properties. The RMF system will employ SYLLAC RF tube technology to supply 5 kA currents to 3.5 mh antennas at 159 kHz frequencies (producing an w = 1 MHz, Bw = 50 G rotating magnetic field). The circulating antenna power will be about 100 MW, with up to 10 MW allowable for resistive plasma losses.

LSX/mod can form FRCs with about 1.5 mWb of poloidal flux at low (2 mTorr D2) fill pressures, and 3 mWb at high (10 mTorr D2) fill pressures. This will span the ne = 1-5x1020 m-3, Te+Ti = 200-500 eV (after expansion) operating range desirable for TCS. (By contrast, TRAP utilized 5 mWb FRCs formed at very high 20-30 mTorr fill pressures to maximize the final plasmoid kinetic energy density.) TCS will have an initial magnetic field of 1-2 kG, which can be raised to about 4 kG during the 5 msec RMF application to accommodate poloidal flux build-up. The initial FRC 's' values will be low (0.5-2), but will increase to about 3-7 if flux build-up is successful. At the lowest fill pressures, the resultant FRCs will be highly collisionless, with ratios of wce = eBw/me to classical nei of about 1000. Since the skin depth without synchronous electron rotation is only 1 mm, while theory predicts an increased RMF penetration by a factor of wce / nei due to synchronous rotation, this experiment should provide a good test of the RMF concept applied to hot FRCs.

II. EXPERIMENTAL DESIGN

Figure 1. TCS attached to LSX/mod

A sketch of TCS mounted on LSX/mod is shown on Figure 1. One of the accelerator coils will be removed to provide more room for TSC on the LSX/mod rail system. The transition from the 46-cm coil diameter formation section to the 30-cm diameter acceleration section was the most difficult obstacle to overcome in the TRAP experiments, especially at high flux levels which resulted in large diameter FRCs. However, with proper timing of the formation coils and correct matching of the external flux between the two sections, formation and translation became relatively routine. In fact, it was actually easier to produce translating FRCs than stationary ones due to the avoidance of the normally strong axial implosion.

The TCS chamber will consist of two of the old LSX 80-cm ID by 1.25-m long quartz plasma tubes, plus two conical stainless steel end flanges. The axial confinement field will be provided by (20) 90-cm diameter, 60-turn coils supplied in parallel in two groupings (the central fourteen and three coils at each end). This will allow for flexibility in adjusting the external field axial profile. Additional flexibility will be provided by the approximately 60-cm diameter transition coils, and by 30-cm diameter coils at the entrance and exit.

The rotating magnetic field on TCS will be provided by two sets of Helmholtz type coil antennas that are driven 90 out of phase. The antennas will be powered by Machlett ML-8618 triodes which were built for the old SYLLAC experiment. Each triode can operate at 14 kV and carry 220 amps. Several tubes in parallel, driving a high "Q" circuit, can supply the required 4-5 kA antenna current. A sketch of the circuit for one antenna is shown on Figure 2.

Figure 2. Vertical field antenna circuit

III. TRAP RESULTS

A MOQUI calculation of the formation, translation, and acceleration process in the TRAP experiment, which exactly matches experimental measurements of diamagnetism time-histories, is shown on Figure 3. Both flux and density profiles are shown at various times. The initial fill pressure was about 20-mTorr of D2, and the total FRC mass was about 0.8 mg. The FRC was ejected from the source section at a velocity of about 130 km/sec, and was accelerated up to 200 km/sec by the four-stage accelerator. The average acceleration of 0.5x1010 m/s2 was only limited by the relatively slow rise-time of the 3-turn acceleration coils (3-turn coils were needed to match the relatively high inductance of the original LSX capacitor banks). It can be seen from the calculations that the accelerated FRC is far from equilibrium, with the mass lagging the flux, and this is also very apparent from comparing excluded flux and interferometric or tomographic diagnostics.

Figure 3. MOQUI calculation corresponding to shot 1647

It was possible to produce FRCs with about 5 mWb of flux at the high TRAP fill pressures. TCS will require FRCs formed at 2-10 mTorr fill pressures, with the lower values resulting in the desirable collisionless conditions required for a good test of the rotating field current drive concept. It is hoped to be able to form FRCs with 1.5 mWb of flux at 2 mTorr, and with 3 mWb at 10 mTorr.

 

III. TCS DESIGN POINT

The primary TCS design point is based on a 2-mTorr fill pressure, which should result in a 500 eV total temperature plasma and 1.5 mWb flux levels. A strong preionization system will be built to permit operation down to 1 mTorr fill pressures (the same as achieved on LSX), which could result in a doubling of the final temperature. The highest anticipated operating fill pressure will be about 10 mTorr, which will result in relatively collisional plasmas, but higher FRC flux levels. TRAP was routinely run at 10 mTorr and above, so TCS will operate at the lowest TRAP fill pressure conditions, and below.

 

Table 1. TCS Design Points
table 1

 

Design point conditions at the two extremes of the TSC operating regime are shown on Table 1. The range of conditions in the source is shown in the first column, with the conditions upon translation and expansion into TCS shown to the right. The 'no RMF' conditions assume that the source temperature is recovered after capture in TCS. This preservation of energy could be aided if necessary by use of the acceleration coils, but only passive activation of the acceleration coil bias fields is anticipated. Where the separatrix length exceeds the TCS chamber length, the FRCs will be supported by the end mirrors, and will actually have a larger radius than shown.

The low density conditions shown on Table 1 are ideal for our test of RMF current drive. The plasmas will be very collisionless, and RMF penetration will only occur if the electron current can be driven synchronously with the rotating field. If the RMF current drive is successful, the FRC flux will be increased as shown in the table, and the external confinement field will need to be increased. Provision for this has been provided by the msec coil current rise-times in the TCS design.


RPPL Home

University of Washington

Contact Webmaster: rppl@u.washington.edu