At 0356 hrs. EDT, while operating at full power, the reactor automatically scrammed on a high signal on period channel 3. At that time, it was believed that the
scram was due to noise on the channel. All channels were checked and at 0449 hrs. it was decided that to
restart the reactor.
During the startup, the rods were pulled to the estimated critical position (ECP) without indication on any of the period channels (3 channels installed). It was decided that the xenon could be precluding the reactor so it was decided to bump the rods out. After bumping out the rods, no increase in period was indicated. During the ensuing discussion, the operator observed power increasing and, at approximately 4 Mw, decided to manually scram the reactor.
Investigation into this event is ongoing.