Quadrant Selector "Stuck" Anomaly
Background:
A condition has been occurring throughout the first few months of observations where the quadrant selector (QS) would become off by a position or two. This creates a "stuck" situation when commanding a new position. It is not known whether it is a problem with the control algorithm or the motor electronics but the motor becomes confused and leaves the current applied to the motor while trying to move. The condition occurs predominantly after 180° moves. In June ’98, a patch was created to the control algorithm which causes the quad selector to move through the encoder reference position more frequently in hopes that the electronics would have better knowledge of the QS position. The QS has since become stuck several more times however. An additional action point (#43) and RTS (#53) were created to reset the QS when this condition occurs. AP #43 now executes prior to the existing Motor OI AP #21 which will reset the QS and resume normal science observing.
Key Telemetry:
Event #1043 - Quadrant Selector Reset (002B, 0035, 0033) - where
002B is the number of the actionpoint that failed (x2B = 43 decimal)
0022 is the number of the RTS that executed (x35 = 53 decimal)
0033 is the number of counts (x33 = 51 decimal)
Motor Current (IIAEMTR) ~ 0.63 Amps
QS Position (IKQSPOS) - will got to an invalid position. Valid positions are:
Quadrant Position
171/A 28/29
UV/B 20/19
195/C 10/11
284/D 2/1
Recovery Procedure:
A recovery procedure exists (TI_qs_reset.proc) but is not necessary since the QS_reset RTS will perform similar steps:
00:00 ICSQPAUS pause the sequence
00:00 ICDPFLGS INDEX=1, MASK=255, VALUE=0 reset the picture state machine
00:00 ICQSRS reset the quadrant selector
00:03 ICQSSP POSNO=33 load position 20/UV
00:00 ICQSCW command the move
00:10 LCAPRESET AP=43 reset AP 43 statistics
00:00 LCAPSTATE AP=43, AP_STATE=ACTIVE reactivate AP 43
00:00 ICSQRESM resume the sequence
References:
http://tracedata.nascom.nasa.gov/~trace/cdhsw/trace/design/sc/pre_rts.htm