Student Intern Helps to Unravel Ion BeamsSeptember 14, 2017
The ion beams that FF-1 produces during a shot are an important feature of its operation. In the present experimental phase of our work, the energy of the ion beams give one measurement of the energy transferred into the tiny plasmoid that produces the fusion reactions. The current of the ion beams is also a measure of how many particles are confined in the plasmoid. In the longer term, the ion beam will be one of two ways to derive energy from a Focus Fusion generator, with direct conversion of the beam energy into electricity in a circuit.
However, there are a number of puzzles we need to solve about our measurements of the ion beams. We measure the energy of the beams by measuring the velocity of the ions—the greater the velocity, the greater the energy. We do this by using two coils, called “Rogowski coils” or RC for short. When the beam passes through a coil, it induces a current in the coil that we can measure—a current proportional to the rate of change in the ion beam current. By measuring the time that the beam passes the upper RC or URC and then the lower RC or LRC we can calculate the velocity.
The puzzle arises because, for some shots, the energy we measure seems too high—more than 5 MeV. Such high-energy deuterons (deuterium nuclei) can undergo reactions with the steel at the end of the drift tube that create short-lived radioactive elements. However, we have observed only a very tiny increase in radiation—far less than we expect if the ions were really 5 MeV energy. (Induced radioactivity would not occur in a fusion generator, since the energy of the ions would be extracted by a different sort of coil before they could hit anything.) In addition, our JavaFusion program for automatically finding and measuring the peaks in the output of the RC and other instruments was missing many of the peaks.
Figure 2. A typical ion beam (from May 12, shot 3) as recoded by the URC (blue line) and the LRC (red line). The LRC current is enlarged by 5 to compensate for the greater distance from the source. The time between the first blue and red peaks indicates a puzzling high energy of 10 Mev, while the longer time between the second blue and red peaks show a more-easily-understood 1 Mev.
So Chief Scientist Eric Lerner worked with Student Intern Justin Cohen (a junior nuclear engineering major at North Carolina State) this summer to start to unravel the puzzles. Cohen looked at the outputs of the RCs for the shots FF-1 fired in May of this year. He measured and analyzed the currents and timings observed, comparing the results that he obtained by hand with those of the JavaFusion program. Based on his analysis, he and Lerner arrived at new criteria for JavaFusion to distinguish the real signals (sharp peaks as in Figure 3), from the electromagnetic noise that the RCs pick up.
LPPF’s Electrical Engineer Fred Van Roessel is now working to implement the new criteria in Java Fusion. Once that is done, hundreds of shots from this year and last year can be analyzed automatically and quickly. We expect this new analysis to give us the clues that we need to resolve the puzzle of the high ion beam energies.