On 14 January 1959, two helicopters, as the first flight of the 3rd Japanese Antarctic Research Expedition (JARE), landed at Syowa Station with the aim of conducting geo-scientific research for the International Geophysical Year (IGY). Syowa Station had been uninhabited for one year. The research team immediately announced sensational news to the world: they found two living dogs that had survived through the whole winter in Antarctica without being fed.
As a member of this expedition, I started work on the resumption of the main power plant of this station. I participated in this party as a geo-scientist, as well as an electrical engineer. Soon, we set up a new diesel electric generator of 40 kW, 3 phase, 100 and 200V. After finishing the mechanical set up on the engine mounts, I was confused about how to connect the center point of the star circuit to a grounding post. The Syowa Station was built on rocks with very poor conductivity – eastern Antarctic pre-Cambrian Granuate (granite geneiss). The rock conductivity in this area was approximately 10-6 S/m.
I had to change the circuit connection of the generator output from star to delta in order to avoid the grounding problem, and this was just the beginning of a long struggle against numerous and serious EMC problems in our Antarctic expedition. For example, when the HF 1 kW output radio transmitter was keyed down, or the ionosphere bottom-side sounder (10 kW peak output, sweep frequency range from 0.5 MHz to 17MHz) was turned on every 15 minutes, the data recorded on high-sensitivity geophysical observation instruments in the station showed very strong electromagnetic interference based on the poor grounding condition.
During the whole austral autumn of 1959, I had to work everyday to solve these interference problems by trial and error. Finally, we reached a tentative best solution, although several issues still remained. Syowa Station was built on a small island named East Ongul Island. I extended a 600 m long and one square cm cable from the generator hut to the shore and terminated the cable with 2 m by 1 m copper plate, which was sunk into the sea water as the common mode grounding center after very hard work of thick sea ice boring.
I once again changed the generator wiring from delta back to star and connected the center point of the star circuit to the grounding cable connected to the sea water. This grounding approach, however, did not dramatically reduce the level of interference caused by the radio transmitters. The interference level was reduced by about 6 to 10 dB, but considerable interference still remained. Next, I extended the grounding cable from the generator hut to the radio communication hut and the ionosphere observatory hut, and connected the ground terminals of the radio transmitter and the ionosonde to the grounding cable. We found that this arrangement did not reduce interference levels encountered in magnetometers and other sensitive instruments.
After many trial-and-error steps, I put a new sea ground plane into another ice hole, which was separated by 400 m from the common mode grounding point for the power plant, and extended this new grounding cable to a new radio hut and ionosphere observatory. Finally, I connected the grounding post of the RF antenna circuit to this grounding cable. A dramatic improvement was achieved. Today, this ground system, which was separated from the common mode grounding line, is called the normal mode grounding line.
After my work on the grounding problem in the third (1959) expedition, later Japanese Antarctic research expeditions did not attempt any further improvements in the field of EMC techniques. In 1976, I returned to Syowa Station again, as the leader of the 17th Japanese Antarctic Expedition, and I faced another new interference problem during the construction of an inland station. We had a plan to extend the upper atmosphere multipoint observation network around Syowa Station. As a part of this network, we selected a new observation site located 300 km southeast of Syowa Station. We called this site "Mizuho" Station.
The Mizuho Station was built on the snow surface of the 2200 m thick Antarctic ice sheet. The electrical characteristics of the Antarctic ice sheet are as follows: the dielectric constant is 1.05 at the surface and 3.8-3.95 at a depth of 150 m from the bottom of the ice sheet, and its conductivity is 10-7 S/m or less. The internal structure of the ice sheet is Polycrystalline, including high pressure air cells. This situation is quite similar to the grounding condition of space satellites. We set a diesel engine generator in the hole of an ice tunnel. The hole sunk deeper and deeper year by year with new snow accumulation.
After rebuilding the station, we had to solve two large difficulties. The first was an unbelievably great natural electrostatic charging and discharging phenomenon. The meteorological conditions at Mizuho Station are as follows: the average temperature is -20°C in summer and -50 to -60°C in winter; relative humidity is 1 to 6 % (water vapor) throughout the year; wind velocity is between 10 m/s and 30 m/s from the southeast.
The noise level of electrostatic charge started and quickly reached to over-scale of receiver's "S" meter and finally made a strong electric discharge sound somewhat like a lightning strike. The noise level then returned to normal, but increased again soon thereafter. We found that the noise showed a good synchronization to snow cloud packet drifting. Outside, there was always a katabatic wind blowing at 10 m/s to 30 m/s, and very dry, tiny snow powders were carried by this wind. Gusty wind brought a mass of snow cloud, producing a large amount of electric charge. Electrostatic noise started to increase with the approach of a snow cloud and continued to increase during the passing of strong drifting snow mass until a discharge took place, like a lightning strike. During the passing of the drifting snow, this charging-discharging took place several times, and the noise stopped shortly after the snow clouds passed.
In order to reduce the buildup of electrostatic charge, we tried to bury all metallic materials completely under the snow surface. By this method, we obtained a perfectly successful result in decreasing the electrostatic noise. If any portion of metallic material became exposed outside the snow surface, strong electrostatic charging would come back quickly. So we buried all metallic materials, including HF and riometer antennas, under the snow, and the all-sky camera and exhaust pipe of the generator were covered by a thick plastic board.
The second difficulty was the ground system of this station. When the radio transmitter (50W) was keyed down, all the observation equipment and sensitive coupling electric facilities experienced heavy interference. The power cable between the generator and the observation cabins ran beside and across the cabin wall. All the equipment was connected to this power cable through the capacitive coupling. We connected grounding points of all the equipment and machines with a thick power cable, and this cable was extended about 40 m, where we built a counter-poise consisting of 20 radial wires, each 20 m long. This counter-poise was set at 0.5 m under the snow surface. After the setting of the counter-poise, we connected common mode grounding cable to the center of the counter-poise. This counter-poise type grounding system worked perfectly to eliminate the natural and man-made noise in our base. It could work as both common mode and normal mode ground. The RF system, including antennas and sensors, was also connected to this ground as single point grounding, and we obtained very satisfactory results.
When we were faced with strong interference at Syowa Station in the third expedition in 1959, the results of our improvements were very similar to the modern system. Of course, we had not known of this new technique at that time. I applied this grounding technique later in the design of rocket and satellite instrumentation, after leaving the Antarctic. And I have encountered several similar cases of noise and interference reduction technologies in rocket, balloon, and satellite experiments in Japan. The experience obtained at least 30 years ago in grounding and interference reduction is still of great use in my EMC work even today.
|
Takeo Yoshino is a professor at Fukui University of Technology in Japan. He is a Life Member of IEEE, and a member of the Board of Directors of the IEEE EMC Society. Professor Yoshino was chairman of the IEEE EMC-S Tokyo chapter from 1990 to 1994. For details on his very interesting career and achievements, see "Personality Profile" in the Fall 1999 EMCS Newsletter (pages 32-33).