Question / Problem:
Missed Shoe Alarm for a sliding Shoe sorter
Environment:
Sliding Shoe sorter
BOSS / MC4 system
Answer / Solution:
Background:
The first divert photoeye on BOSS / MC4 / ICW controlled sliding shoe sorters is used as an encoder to drive the tracking and sortation logic. Because of this, logic exists to ensure that every shoe bearing is detected at the first divert photoeye, and alarms are generated if a bearing is missed. This is done by anticipating the typical interval from lead edge to lead edge of the bearings. (about 5", or 40-50ms at 500-600fpm). If a new lead edge is not detected within a small window after the expected interval, the alarm is set.
Possible causes:
- Missing pin/bearing - The pins and bearings are pressed in to the plastic sliding pushers with a high friction fit. As such, it is possible for a pin/bearing to become loose and fall out of the pusher. This is rare, but has been documented. The alarm should indicate a number as part of the alarm text, which should represent how many shoes / pushers have passed the first divert since the detected missing pin. The pushers / shoes at that point should be checked for missing pins. If a pusher is found with a missing pin, replace it, and make sure the pin and bearing are found inside the sorter before it is run again at speed.
- Problems with the first divert photoeye - as noted above, because motion on the sorter is being tracked based on the first divert photoeye detecting pins. Debris in this divert switch can cause either extra cycles or missed cycles on the photoeye input, causing incorrect tracking of motion. There have been particular problems in the past with cellophane wrap falling inside the sorter and collecting in the divert switch/es.
- Problems with the I/O system - There can be problems in the I/O system that prevent the state of the first divert photoeye from reaching the system controller (BOSS RTC). These can include wiring problems related to the photoeye, DCM and/or input module. Problems have also been documented with OPTO-22 systems with out-of spec 5V power. Any of these can have similar effects to 2. above.
- Problems with the control system - QNX is by design a real-time operating system, and as such should be capable of managing processes at specified intervals. However, reads and writes of the hard disk drive are time-bound events and because they are hardware-level processes, they automatically get a top-priority status. Long read/write events will interrupt high speed processes, preventing scans of the I/O by the software and setting up the same dynamic as 2. or 3. above. Historically, this has happened because one of the BOSS system log files has reached an excessively large size.