The Japanese government has decided to postpone the launch of the nation’s next reconnaissance satellite by six months or more following the discovery of potentially defective integrate-circuits in the satellite, a government official said August 26.

The Prime Minister’s Cabinet Office, which is in charge of the nation’s Information Gathering Satellite (IGS) program, decided Aug. 25 to postpone the launch of what would be the nation’s third reconnaissance satellite in orbit after deciding it was necessary to replace a number of field programmable gate array (FPGA) chips made by Actel Corp. of Mountain View, Calif., according to Yasuhiro Itakura, research officer at Japan’s Cabinet Satellite Intelligence Center, which is part of the Cabinet Office.

The satellite, which carries an optical sensor, was to have been launched by a Japanese H-2A rocket from the Tanegashima Launch Center before the end of March 2006, but it will take about six months to replace the potentially faulty chips and test the satellite to prepare it for flight, Itakura said in an August 26 telephone interview.

Some 10 chips need to replaced, he said. Details about when the problem was discovered were not available at the time of the interview.

Problems with Actel’s earlier version of its FPGA were discovered in autumn 2003, after more than 1 million of the devices were shipped to various vendors.

Ken O’Neill, director of military and aerospace product marketing for Actel, said after news of the defect became known, Actel supplied the Japanese government with the latest version of the company’s FPGA, which the company has the option to install in place of the old version. Since then, the government has been doing reliability testing of both the old and new product, though Actel had not received official word that the company would be replacing the chips as of press time, O’Neill said.

Actel believes the new version of the FPGA should not cause any further problems, O’Neill said.

“They have been tested pretty extensively, and clearly show a very high level of reliability,” O’Neill said. “We have confidence that the reliability of the earlier version is high, but the latest version of the software does offer a higher level of reliability,” he said.

FPGAs contain hundreds of thousands of programmable elements, according to O’Neill, and the defect found in the old version of the chips affected one antifuse within the design, causing it to fail. O’Neill said the chips that do fail usually do so early in the lifetime of the part.

As a supplier, Actel is not directly involved with the rebuilding process, O’Neill said.

The other scheduled flight of a radar-type satellite, which is due for launch sometime in the Japanese government’s 2006 fiscal year (April 2006-March 2007), is not affected by the problem with the Actel chips, and its launch schedule has not been altered, Itakura said.

Each of the the next information-gathering satellites to be launched will have the same capabilities as the original satellites launched by an H-2A rocket in March 2003. One type of satellite has an optical sensor capable of 1-meter resolution, while the radar-type satellite has a resolution of 1-3 meters.

The IGS program was developed in response to an August 1998 incident when North Korea launched a missile that overflew Japanese territory and landed in the Pacific. Two more satellites were slated to join the first pair in orbit in November 2003, but those satellites were destroyed when the H-2A rocket carrying them failed.

A graduate of the Columbia Graduate School of Journalism, where he won the Horgan Prize for Excellence in Science Writing, Paul Kallender-Umezu is co-author of “In Defense of Japan: From the Market to the Military in Space Policy” (Stanford University...