Site: Matsushita Central Research Laboratories
Corporate Research Division
3-1-1 Yagumo-Nakamachi, Moriguchi
Osaka 570, Japan
Tel: 81-6-906-9029; Fax: 903-0996
Date Visited: January 30, 1997
WTEC Attendees: F. Patten (report author), M. Beasley, G. Gamota, H. Morishita, M. Nisenoff, R. Ralston, J. Rowell
Hosts: Dr. Koichi Kugimiya, Executive Engineer; Director of Operations, Corporate R&D
Dr. Kentaro Setsune, Senior Scientist, Central Research Laboratories, Kyoto
Dr. Hidetaka Higashino, Senior Researcher, Central Research Laboratories, Kyoto
Dr. Akira Enokihara, Senior Researcher, Central Research Laboratories, Kyoto


Overall, Matsushita Electric Industrial Co., Ltd., has annual revenues of about $60 billion, employs roughly 230,000 people, and produces almost every item of household electronics as well as specialized products. There are about 4,500 researchers in the corporate research divisions. The Central Research Laboratories and 3 other research laboratories in the Corporate R&D Division have about 500 employees.


Within Matsushita's Central Research Laboratories there are about 50 researchers involved with materials, thin-films, and devices; of these, about 10 people are actively involved with high-temperature superconductors (HTS). The focus of the effort is on mobile communications with multichannel access (MCA). The specific orientation is towards developing high-power filters for transmission, with power requirements on the order of 40-50 watts. The Central Research Laboratories receive more than half of their funding from MITI, with a 4-year contract that finishes in 1999. They are allied with Sumitomo, which fabricates 3-inch double-sided films, and Kyocera, which provides the cryocooler package system design. Sumitomo and Kyocera also receive more than half their support in this area from MITI, and each company provides the remaining support from internal funds.

Matsushita's initial proposal in this effort was to fabricate disk-like planar film structures that could mitigate the current-crowding responsible for the reduction of power handling capability. Early modeling studies have shown that the stripline techniques for the fabrication of high-power filters could support up to about 1,000 watts of power, but may easily yield hot spots at edges and result in unpredictable early breakdown. Each of these disks would be a pole in a multiple pole planar film. Each pole can accommodate multiple resonant electromagnetic modes, which requires effort in designing the feed structure for the disk resonators; the separation of the modes is determined by the ellipticity of the disk. Matsushita researchers have fabricated single disks that have unloaded Q of from 2,000 to 3,000 and about 0.5 dB of insertion loss. When double-sided films are employed, Q improves to the level of 6,000 to 8,000 with about 0.25 dB insertion loss. The researchers observe no change in the insertion loss up to 15 watts of incident power (about 41.5 dBm). With regard to intermodulation distortion, the third-order intercept is projected to be at about 300 watts (about 55.1 dBm). They have modeled the surface charge density on the disk resonators at the center frequency and found it to be rotating in time and well distributed. This is true also for the surface currents, i.e., there is a very uniform distribution of current, and therefore, a capability of handling high rf power. Transmission and reflection measurements were shown for a single disk structure at a frequency of 5.1 GHz, and about 100 MHz bandwidth.

In evaluating the progress of Matsushita's research effort, it seems that the Central Research Laboratories lack the simulation tools to progress quickly enough to a multipole resonant thin-film filter that could handle the power requirement of about 50 watts. Since performance is dependent upon HTS materials parameters and specific disk shape, spacing, and orientation of the feed lines, generalized electromagnetic codes cannot be expected to produce accurate results. The complexity is such that mechanical "tweaking" will be of no use without more accurate guidance from the codes. In this regard, the Matsushita researchers are making concerted efforts to extend the work to more than 1 pole, and they maintain they are able with projecting several designs to determine how they would tune or trim such filters, although the means were not disclosed. The panel's hosts were aware of U.S. work on higher order multipole filters, for example at Conductus, and inquired about the type of simulation used.

The WTEC panelists' impressions of work at Matsushita were consistent with impressions from other corporate laboratories. The goals of the long-term development effort were not shown clearly. Although cellular/mobile communication was given as the overall target, Matsushita chose to work on the problem of transmit filters rather than receive filters, which were felt to be easier to fabricate; our hosts offered no systems analysis to validate this approach. A complete cryogenics package for an MCA system might include transmit and receive filters, as well as a cryocooler and thermal isolation packaging. AMTEL is a possible supplier of filters and refrigerators, and the user might be Matsushita Communications Industrial Co.

Published: August 1998; WTEC Hyper-Librarian