A Co-57 based XRFA system has been installed on-line at ICPP to monitor uranium concentrations after first cycle decontamination. A small, medium, or large collimator is used to restrict fission product (fp) background count rates. Using two 50 mCi sources, 1 gU/1 can be measured to +-2% in 10 min with the medium collimator. A computer-based MCA runs the system automatically, controls an insertable U foil (allows calibration anytime) and an automatic W shutter (permits removal of fp background); prints hourly mean U concentrations and a daily log of past 24-h means. Co-57 sources are changed annually in 1 min without disassembly of a lead-steel enclosure that surrounds the de-entrainment tank and XRFA assembly.

A study is made of the transients caused by the fast dump of large superconducting coils. Theoretical analysis, computer simulation, and actual measurements are used. Theoretical analysis can only be applied to the simplest of models. In the computer simulations two models are used, one in which the coil is divided into ten segments and another in which a single coil is employed. The circuit breaker that interrupts the current to the power supply, causing a fast dump, is represented by a time and current dependent conductance. Actual measurements are limited to measurements made incidental to performance tests on the MFTF Yin-yang coils. It is found that the breaker opening time is the critical factor in determining the size and shape of the transient. Instantaneous opening of the breaker causes a lightly damped transient with large amplitude voltages to ground. Increasing the opening time causes the transient to become a monopulse of decreasing amplitude. The voltages at the external terminals are determined by the parameters of the external circuit. For fast opening times the frequency depends on the dump resistor inductance, the circuit capacitance, and the amplitude on the coil current. For slower openings the dump resistor inductance and the …

The Supervisory Control and Diagnostics System (SCDS) of MFTF is a multiprocessor computer system using graphics oriented displays with touch sensitive panels as the primary operator interface. Late in the calendar year 1981 the system was used to control an integrated test of the vacuum vessel, vacuum system, cryogenics system and the superconducting magnet of MFTF. Since the completion of those tests and starting in early calendar 1983 the system has been used for control of the neutral beam test facility at LLNL. This paper presents a short overview of SCDS for the purpose of orientation and then proceeds to describe the difficulties encountered in these preliminary encounters with reality. The band-aids used to hold things together as disaster threatened as well as the long-term solutions to the problems will be discussed. Finally, we will present some comments on system costs and management philosophy.

For MFTF-B, the plasma diagnostics system is expected to grow from a collection of 12 types of diagnostic instruments, initially producing about 1 Megabyte of data per shot, to an expanded set of 22 diagnostics producing about 8 Megabytes of data per shot. To control these diagnostics and acquire and process the data, a system design has been developed which uses an architecture similar to the supervisory/local-control computer system which is used to control other MFTF-B subsystems. This paper presents an overview of the hardware and software that will control and acquire data from the plasma diagnostics system. Data flow paths from the instruments, through processing, and into final archived storage will be described. A discussion of anticipated data rates, including anticipated software overhead at various points of the system, is included, along with the identification of possible bottlenecks. A methodology for processing of the data is described, along with the approach to handle the planned growth in the diagnostic system. Motivations are presented for various design choices which have been made.

Extensive use of electron cyclotron resonant heating (ECRH) in the Tandem Mirror Experiment-Upgrade (TMX-U) requires continuous development of components to improve efficiency, increase reliability, and deliver power to new locations with respect to the plasma. We have used rectangular waveguide components on the experiment and have developed, tested, and installed circular waveguide components. We replaced the rectangular with the circular components because of the greater transmission efficiency and power-handling capability of the circular ones. Design, fabrication, and testing of all components are complete for all systems. In this paper we describe the design criteria for the system.

The electron-cyclotron-resonant heating (ECRH) systems of rectangular waveguides on Tandem Mirror Experiment-Upgrade (TMX-U) operated with a overall efficiency of 50%, each system using a 28-GHz, 200-kW pulsed gyrotron. We designed and built four circular-waveguide systems with greater efficiency and greater power-handling capabilities to replace the rectangular waveguides. Two of these circular systems, at the 5-kG second-harmonic heating locations, have a total transmission efficiency of >90%. The two systems at the 10-kG fundamental heating locations have a total transmission efficiency of 80%. The difference in efficiency is due to the additional components required to launch the microwaves in the desired orientation and polarization with respect to magnetic-field lines at the 10-kG points. These systems handle the total power available from each gyrotron but do not have the arcing limitation problem of the rectangular waveguide. Each system requires several complex components. The overall physical layout and the design considerations for the rectangular and circular waveguide components are described here.