The W-2 SLSF (Sodium Loop Safety Facility) experiment was an instrumented in-reactor test performed to characterize the failure response of full-length, preconditioned LMFBR prototypic fuel pins to slow transient overpower (TOP) conditions. Although the test results were expected to confirm analytical predictions of upper level failure and fuel expulsion, an axial midplane failure was experienced. Extensive post-test analyses were conducted to understand all of the unexpected behavior in the experiment. (1) The initial post-test effort focused on the temperature oscillations recorded by the 54 thermocouples used in the experiment. In order to synthesize the extensive data records and identify patterns of behavior in the data records, a computer-generated film was used to present the temperature data recorded during the experiment.

NOVANET is a control system oriented fiber optic local area network that was designed to meet the unique and often conflicting requirements of the Nova laser control system which will begin operation in 1984. The computers and data acquisition devices that form the distributed control system for a large laser fusion research facility need reliable, high speed communications. Both control/status messages and experimental data must be handled. A subset of NOVANET is currently operating on the two beam Novette laser system.

Most neutral beamlines contain an iron-core ion-bending magnet that requires shielding between the end of the neutralizer and this magnet. This shielding allows the gas pressure to drop prior to the beam entering the magnet and therefore reduces beam losses in this drift region. We have found that the beam losses can be reduced even further by eliminating the iron-core magnet and the magnetic shielding altogether. The required bending field can be supplied by current coils without the iron poles. In addition, placement of the baffles and apertures can affect the cold gas entering the plasma region and the losses in the neutral beam due to re-ionization. In our study we varied the placement of the baffles, which determine the amount of pumping in each chamber, and the apertures, which determine the beam loss. Our results indicate that a baffle/aperture configuration can be set for either minimum cold gas into the plasma region or minimum beam losses, but not both.

The parameter B/sub e//nr/sub p/ (here, B/sub e/ is applied magnetic field strength, nr/sub p/ is the plasma density-radius product) is proposed as a key parameter for spheromak heating studies. If B/sub e//nr/sub p/ is too large, increased magnetic fluctuations limit heating; low B/sub e//nr/sub p/ value results in excessive radiation losses. An optimum range appears to be B/sub e//nr/sub p/ approx. 1 to 5 x 10/sup -20/ Wb.

The Tandem Mirror Experiment-Upgrade (TMX-U) at Lawrence Livermore National Laboratory (LLNL) uses 24 neutral-beam injectors to heat and fuel the experimental plasmas. This system is unique because TMX-U operates four times more injectors than any other fusion experiment. These injectors deliver an average of 50 A (accel) at 17 keV for 75 ms. Source conditioning time has been reduced to approximately four days for the entire system after extended machine air cycles. TMX-U is also unique because it has 35 usable injector assemblies for the 24 power systems. This quantity of injectors makes possible the development of new hardware and injector modifications, and the reconditioning of damaged sources without affecting machine operation. Efficient operation of a system of this size requires coordinated interaction between the injector service groups and the physics organization. We describe the current state of TMX-U performance and the aspects of group interaction essential to a project of this size.

The Plasma Current Sensor (PCS) diagnostic includes large diamagnetic loops (DL) that fully encircle the plasma as well as small multi-turn pickup coils (PCs) located between the plasma and the superconducting magnets. Both types of sensors respond to changing magnetic flux linkages caused by plasma currents and are used to measure plasma diamagnetism, from which estimates of temperature and density can be made. The DLs are used in the central cell and Axicell regions, while the PCs are used in the Yin-yang regions where DLs are impractical. Other PCs are used in the central cell to detect axial plasma currents, to help tune trim coils in the transition cell and confirm theoretical estimates of radial diffusion limits. This paper describes the PCS diagnostic and presents the detailed mechanical and electrical designs.

The definition of a time-dependent spectrum registered by an idealized spectrometer responding to a time-varying electromagnetic field as proposed by Eberly and Wodkiewicz and subsequently applied to the spectrum of laser-induced fluorescence by Eberly, Kunasz, and Wodkiewicz is here extended to allow a stochastically fluctuating (interruption model) environment: we provide an algorithm for numerical determination of the time-dependent fluorescence spectrum of an atom subject to excitation by an intense noisy laser and interruptive relaxation.

Hopkinson split-bar tests were performed on Type 21-6-9 austenitic stainless steel from ambient temperature to 1023 K. These high-strain-rate tests (10/sup 2/ to 10/sup 4/ s/sup -1/) were compared with lower-strain-rate tests (10/sup -4/ s/sup -1/). The results indicate that over this temperature range, the strain-rate sensitivity of 21-6-9 is not strongly dependent on the strain rate. This suggests that the mechanism(s) of plastic flow at the higher rates is similar to that at the lower rates. This contention was corroborated by transmission electron microscopy.

Performance of the MFTF-B tandem mirror reactor is critically dependent upon precise alignment of the superconducting magnets in the transition region. Given the size and nature of these magnets, mechanical alignment under operating conditions to the tolerances required is an impossible task, as placement must anticipate deflections due to vacuum conditions, cooling to cryogenic temperatures, and magnetic forces. A cost-effective solution to the alignment problem is presented here. In each transition region, a set of eight trimming magnets capable of introducing dipole- and quadrupole-field components allows the field curvatures to be finely tuned to offset the effects of mechanical alignment errors. Trimming current values for correcting representative single- and multiple-coil misalignments, as well as an analytic method for determining them are given. Finally, design considerations that will minimize the need for trim coils in future tandem mirror devices are suggested.