« iepriekšējāTurpināt »
A LOOK AT PULSED POWER RESEARCH
Observations from a recent visit to several Japanese industrial companies
summarized. The number of Japanese organizations pursuing this line of
States, but in a few cases there were some unexpected and impressive
performance parameters (e.g., v > 7 km/s for a rail gun and very high performance solid state power switches). The reason for the high Japanese
interest in pulsed power research remains somewhat of an enigma.
by M. Kristiansen
them, whereas Japanese firms want to visited them. Representatives from many
make sure that they do not overlook a of these institutions have visited our I was recently invited to present a new field or opportunity. The lectures laboratories at Texas Tech over the 3-day series of lectures (14-16 November were attended by 59 people from 25 past few years, some for extended peri1991) on pulsed power technology at different organizations.
ods of time, so I was already reasonably Aso-Ikoinomura, Kumamoto Prefec- I was also surprised to learn that the familiar with the main theme of their ture (Japan's largest national park). Institute of Electrical Engineers of Japan work. Before the lectures I had the opportu- has an active subcommittee on Applied nity to visit some Japanese universities Technology for Electromagnetic Mass Toshiba Corporation, and industrial laboratories engaged in Acceleration, and I had the pleasure to Fuchu Works related work. It is interesting that the be hosted by them one evening in Tokyo. lectures, which were sponsored by the This committee has members from This is, of course, one enormous Institute of Electrical Engineers of Japan industry, universities, national labora- corporation with a totalofsome 70,000 and organized by Kumamoto Univer- tories, and the Japanese Defense Agency. employees in the main company and sity, were also supported by 13 large There are strong Japanese interests and over $30B in annual sales. The Fuchu industrial firms. These were: Asahi activities in various aspects of electro- Works employs some 7,500 people, with Chemical Industry, Fuji Electric Cor- magnetic launchers (EMLs), and much another 5,500 in the company's subsidporation, Hitachi, Ishikawajima-Harima of this report will be related to EML iaries. The staggering fact is that 60% Heavy Industries, Japan Steel Works, research. This research is mostly related of the employees are engineers and Kawasaki Heavy Industries, Kobe Steel, to armor testing by companies, such as 25% are assembly workers. I do not Mitsubishi Electric Corporation, Japan Steel Works, in the hope that the have the facts from similar industries Mitsubishi Heavy Industries, Nichicon Japanese Defense Agency will initiate in the United States or Europe, but I Co., Nissin Electric Co., Sumitomo a major program in this field.
find it hard to believe that any of them Electric Industries, and Toshiba Cor
has this high a percentage of engineers. poration. I believe that it is very unlikely TECHNICAL VISITS
I had, on a previous occasion, visited that any such set of lectures could be AND OBSERVATIONS
Toshiba's Hamakawasaki Works and sponsored by such an impressive and
was familiar with some of its circuit varied group of U.S. industrial organi- In the following I will summarize breaker and vacuum interrupter work. zations, no matter who gave the lec- the technical observations I made at I was met by representatives from both tures. U.S. firms would decide up front some Japanese industrial and univer- places and it is difficult for me to sepathat this was a topic of no interest to sity laboratories in the order in which I rate the exact responsibilities of each place. My thanks go to Drs. K. Sugi, loss, but lower saturation magnetic field, machines, bridges, etc. My visit was to E. Kaneko, H.Ohashi, K. Hoshi, than Metglas (registered trademark of
than Metglas (registered trademark of its research institute. Of particular I. Ohshima, and K. Okamura.
Allied Chemical). The new material is interest to me was the arcjet and EML Toshiba Corporation is a leading presumably available in 25-um-thick work. The configurations of the magmanufacturer of high voltage systems foils. The greatly reduced losses in this neto plasma dynamic (MPD) and arcand switch gear and I was especially new material may make it very attractive jet thrusters for electric space propulimpressed by the latter aspect. Suffice for many repetitive pulsed power sion were fairly conventional. The it here to say that Toshiba's continued applications, such as isotope separation institute's published work on arcjet effort of reducing the size of the equip- and NOX/SOX removal from flue stacks. thrusters at recent international conment is truly remarkable. Its vacuum
ferences has been concerned with low interrupter work is also outstanding Tokyo Institute of Technology power (0.5-1 kW)H,N, and simulated and it has been the leading organiza- at Nagatsuta
hydrazine (N, + 2H) devices. The tion in increasing the interruption
reported performance has not been current and voltage of these devices Dr. Koichi Kasuya has for many anything unusual. The researchers claim, while continually reducing their sizes. years worked on ion beam generation however, that a barium oxide impregThis is achieved by clever electrode from cryogenically cooled (liquid nitro- nated tungsten cathode has less eroconfigurations and special electrode gen) anodes. The institute has a series sion than other materials, e.g., thorimaterials. The production of these of PICA (particle beam inertial con- ated tungsten, in their 8-KA, 600-us, 1devices increased by a factor of 14 in finement fusion apparatus with cryo- to 2-pps MPD generator. This is quite the period 1970-86, whereas U.S. genically refrigerated anode driver) interesting since we have the opposite industry basically dropped out of the machines in use for this work. The goal experience in continuous arcjets at competition.
is to obtain high purity, high current -200 A. This could be due to variations Of special interest was the develop- ion beams for inertial confinement in the impregnation process, but it needs ment of a wide range of high power fusion (ICF). The PICA-4 machine, for further study. semiconductor switches. Among these instance, is a 3-12, 120-ns, 1.5-MV They have also used the arcjet to devices were 3-KV, 9-KA; 5-KV, 4-KA; machine. The anode is covered with D, produce diamond film with a very clean, and 6-KV, 3-KA gate turn off (GTO) or ammonia ice, and the best results narrow Raman spectrum. In this case transistors. For the first device a di/dt seem to be a few kA current at a few they inject H, + Ar in the conventional of 8 kAlus was reported. A nationally 10s A/cm2 current density. The insti- way from the rear of the device and sponsored (Ministry of International tute also does laser development work inject CH, in the nozzle section. The Trade and Industry (MITI)) project on with equipment provided by industry.
with equipment provided by industry. film is deposited on a Mo (1,000 °C) metal oxide semiconductor (MOS) In all cases the work was sufficiently substrate. The deposition rate is assisted gate triggered (MAGT) far from my own area of expertise that 1.5 um/min for an arcjet power level of thyristors has a design goal of 2.5 kV it was difficult to judge its quality. 2-5 kW. The back pressure in the and 10" A/s-cm2. They have achieved a
chamber is 20-100 Torr and the plasma current density of 2 kA/0.14 cm 2 and Ishikawajima-Harima Heavy jet then does not expand enough to operated at 5 kHz. These devices have Industries (IHI) Co., Ltd., produce large area films. Five Japanese low losses and require low trigger power. Yokohama
companies are apparently engaged in Figure 1 shows a convenient diagram
similar work. for comparing the operating regimes My hosts at IHI were Drs. IHI's EML (rail gun) was of a conof various types of high power K. Uematsu, S. Morimoto, T. Majima, ventional design, superficially similar semiconductor switches. Toshiba has and T. Tagaeto. This is again a very
to the one described in the next section. done considerable, successful work on large company with many employees in both cases the entire gun assembly series-parallel connection of these (*15,000 in 1988) and numerous activ- was submerged in vacuum. Apparently devices. The company also reported ities in the heavy construction field, the results were not as good as those the development of a new amorphous such as nuclear power plants, ships, oil reported in the next section of this (cobalt based?) metal with much lower drilling platforms, manufacturing report, and the reason for this was not
Figure 1. Operating regions for various semiconductor devices (reprinted with permission from
K. Okamura, E. Kaneko, et al., "Pulsed power and industrial application," J.I.E.E. Plasma
The Institute of Space and they disassemble and rebore the gun opening switch is of considerable interAstronautical Science,
after each shot. Each time they bore off est both in the United States and in Kanagawa
0.5 mm. Extreme care in assembly and Russia, where the latest nuclear weapons
rebore seems to be the only explana- effects simulators under construction Professor N. Kawashima was my host tion they could give.
depend critically on the success of such during this visit. This institute has a The system operates at 2 shots/wk switches. These switches are equally very interesting rail gun in terms of its and is mostly used for micrometeorite important to inertial confinement fusion performance. The basic system appears and space debris impact simulation devices. Akiyama and his group have fairly conventional, as shown in Figures 2 studies.
shown that it is possible to do meaningand 3. The 2-meter-long gun has a
ful work in this field with modest, nominal bore of 1.3 cm. The rails are Kumamoto University
university-size machines. They have, made of copper and the insulator of
for instance, demonstrated a factor of polycarbonate. The whole rail system Dr. Hidenori Akiyama has cooper- 12 voltage gain (from 30 to 400 kV) at is enclosed in a vacuum tank evacuated ated with our research group for sev- 80 kA with 10-ns risetime using a twoby a rotary vacuum pump. No preinjec- eral years and has spent considerable stage opening switch with a first-stage tion is used. The interesting claim is time in our laboratory. We have co- fuse followed by a plasma opening switch that the researchers achieve 6 km/s authored several publications and he (POS). velocity for a 0.9-g projectile “stably” was the main organizer of my lectures Avery interesting concept is the use and 7.5 km/s in “some cases.” This is, and laboratory visits. He has an excel- of a laser-produced plasma, on a rotatindeed, quite impressive. In the United lent, small research group that utilizes ing graphite target, to achieve high States 6 km/s is considered a “velocity space and facilities to the fullest. His reproducibility and long life. They are barrier,” which is under extensive interaction with students and staff is in the process of upgrading the laser investigation. There have been a few, most cordial and there is evidence of from 1 J to 30 J/pulse to get higher unverified reports of a few shots with excellent group spirit and enthusiasm. . currents than the 20 kA they get with v> 7 km/s, but not on a repeatable The university's ASO I and II the 1-J laser. High reproducibility and basis.
machines have produced very interest- long life are critical parameters for all It is not clear why this device has ing plasma opening switch results with the intended applications of such such excellent performance except that very modest size facilities. This type of switches.
Figure 2. Schematic layout of ISAS rail gun "HYPAC" system. Adapted and reprinted with permission
from A. Yamori et al., "Rail gun experiment (HYPAC) at ISAS,” IEEE Transactions on Magnetics 27(1), 120-129 (1991). © 1991 IEEE.