Page 3 - Plasma and Fusion Research,ISSN 1880-6821

Plasma and Fusion Research: Regular Articles
Volume 13, 3405062 (2018)
Fig. 3 The diagram of the experimental setup with the discharge circuit. This cross-section corresponds to the dashed line in Fig. 1.
trode is φ 16 mm. The miniature gun is installed radially at the same cross-section as the gas injection ports. The discharge method is similar to MCPG’s; neutral gas flows into the gap between the electrodes, and a plasma gener- ated by a high-voltage breakdown in the miniature gun is accelerated/ejected by Lorenz self-force. A bias field is also applied to assist breakdown between miniature gun’s electrodes. The inverter circuit, consisting of 8 IGBTs in series, switches the discharge current on the miniature gun. The inverter unit is protected from the surge voltage using a charging-type snubber circuit. A diode and inductor also installed for protection from high voltage and large cur- rent of the MCPG. The charging voltage of the 52 μF ca- pacitor bank is 3.3 kV, and the inner electrode is positively charged. Typical waveform of the discharge current has a flattop with approximately 550 A at the peak, and the rise time is ∼ 14 μs. The miniature gun starts to discharge at 40 μs before the main discharge of MCPG, and the dura- tion of current is 30 μs; the duration can be adjusted ar- bitrarily. The impurity influx should be negligibly small because of its low current. By adopting the miniature gun, MCPG breakdown can occur at lower gas pressures than that of without the PI system.
3. Experimental Results
Figure 4 shows the variation of CT parameters while changing iron-core length (dFe) and coil position (dCoil) without the PI system, where dFe is the length between the edge of the coil and the end of iron-core, and dCoil is the distance between the edge of the coil and the gas port. Here, the bias current is 1.85 A and the gas pressure is 0.276 MPa.
By changing dCoil and dFe, the bias magnetic flux and the bias field distribution in the area where the plas- moid accelerates can be controlled; consequently various CT parameters can also change. According to the experi- mental result seen in Fig. 4, the CT velocity can be affected by both the coil position and the iron-core length. While, Bz depends more strongly on the coil position than on the iron-core length. In the cases of dFe = 0 and 40.6 cm for
Fig. 4
Dependency of CT parameters on iron-core length (dFe) and coil position (dCoil) without the PI system.
3405062-3
example, Bz is higher at dCoil = 35.6 and 5.08 cm, respec- tively. As dCoil becomes longer, the interlinkage mag- netic field distributes wider in the axial direction in the area where the plasmoid accelerates. The plasmoid accelerates while gathering more magnetic field so that the poloidal magnetic field (Bz) of CT’s can be affected by the bias coil position. By changing the bias flux and the bias filed dis- tribution, the amount of poloidal flux and the velocity of the generated CT can be controlled.
Figure 5 shows the evaluation of the CT parameters at lower bias current and gas pressure with the PI system. Here, the bias current is 1 A and gas pressure is 0.172 MPa. The iron-core is fully inserted and fixed in each case, and only the bias coil is moved. The PI system can efficiently produce breakdown, thus the MCPG can operate at a lower gas pressure, reduced by approximately 40 %. By reducing the amount of neutral gas, the generated/ejected CT has a faster velocity; more than 100 km/s in dCoil = 0, 5.08, and 45.7 cm cases. It also reached approximately 140 km/s in dCoil = 0 case. From edge probe measurements, the electron density is lower than the typical value (∼ 5 × 1015 cm−3) of the CT injector developed for the C-2U FRC [3].