Description: | ts4 went offscale and sm4 dropped to a constant 16% when a "new" program
was installed on 3/30/05. The wrong program was installed.
Because ts4 did not change with time, it did not mess up the calculation
of e and h. e and h are slightly overestimated, but only by about 1%,
far less than the system error. This is a rare type of instrument problem where the
concequences are minimal.
Therefore, e and h are not listed as incorrect in this DQR and are still usable during the
stated period. |