Latest revision 26 October 1999

          Bad LAD 5 Data on TJD 11465

 TJD 11465 = 14 October 1999

During the last approximately 3 hours of TJD 11465, the count rates in
both DISCLA and CONTINUOUS data for LA Detector 5 became much too high
during 12 separate occurrences.  In each case, the rates returned to
normal by themselves.  The high voltage was never off (due to SAA
passage or by special command) during the entire time.  The reason for
the high rates is still unknown.

The spurious high count rates were strongest at the lower energies.
They appeared in all CONTINUOUS data channels, 0-15.
They appeared primarily in DISCLA data channels 1 and 2, with only
narrow spikes appearing in channel 3 and nothing apparent in channel 4.

The incident followed similar occurrences on TJDs 11459, 11460, 11462.

The approximate time ranges when the spuriously high rates occurred are
 77200 - 77654
 77814 - 77871
 77947 - 77980
 78095 - 78179
 78371 - 78398
 78719 - 78992
 79244 - 79559
 79911 - 79973
 80564 - 80630
 81126 - 81208
 84241 - 84316
 85871 - 86115

Quality Flags 75 (indicating high count rates in LAD 5) were set for
the following SoD intervals on TJD 11465:
 77198 - 79650 (covers 7 separate high-rate intervals, ends after 
                particle events end)
 79908 - 79976
 80561 - 80633
 80950 - 81211 (includes smaller background variation before big rise)
 84238 - 84319
 85868 - 86118

Trigger No. 7803 (a GRB) occurred at SoD 78754.58, i.e., during an
interval when the count rate in LAD 5 was too high; this has probably
caused the location determined by the LOCBURST program to be invalid.
The trigger detectors (DSELB) were 7 and 3.  The brightest detectors
(DSELH1,2,3,4) were 7, 3, 5, 0.