<< |
>>
#6915 Dodano: 27-06-2013 19:43. Głosów: 100
True Story: At our museum, back in the 1980’s our building elevator was all relay logic driven, much like the octal cube relays on top of the can crusher. There were 3 floors to the building, and on the second floor were some nature dioramas that included a fullsized stuffed moose. Well the moose was across the aisle facing the elevator.
For some reason, once in a blue moon, if you got in on floor 1, and pressed the button for floor 3, the elevator would start up – and even with no call on floor 2 – sometimes the elevator would stop on floor 2. The door would open, no one would be there but you would stare at the moose who was staring at you, and then the door would close and you would continue up to 3.
This was only once in a while, and most times without anyone else even being on the second floor. It was like the stuffed moose would move and sometimes run across the hall and press the button for the elevator. Door would mysteriously open… there was the stuffed moose…. door would close. We wound up calling it “Visiting the Moose” and it was local lore that the stuff moose knows how to work the elevator.
Elevator company could find nothing wrong- all buttons, switches, motors, sensors were working fine. There was a full printout of the relay control logic and wiring at the panel, and yes, it was all in Ladder Diagram. Dozens of relay coils and contact points, buttons, sensors, indicator lights, etc.
I spent a week reading that diagram and following the relay logic, and watching the relays (luckily with indicator lights on each) cycle as the elevator did its thing. I eventually found it. It was a race condition between a set of contacts closing on one relay and the elevator position sensor relay for floor 2. If one relay dropped out just a split second too slow before the other pulled in it created a false “stop at floor 2” condition. Since relays do take a finite amount of time to transition from one state to the other this can indeed cause issues in close timing applications. The fix here was to move 2 wires from the offending relay contacts to a relay that was actually controlled from that relay. The extra few milliseconds delay as one relay cascaded to the next was enough to prevent the race condition and the elevator stopped mysteriously stopping on floor 2.
I still swear the moose moves though while I’m not looking…
<< |
>>
For some reason, once in a blue moon, if you got in on floor 1, and pressed the button for floor 3, the elevator would start up – and even with no call on floor 2 – sometimes the elevator would stop on floor 2. The door would open, no one would be there but you would stare at the moose who was staring at you, and then the door would close and you would continue up to 3.
This was only once in a while, and most times without anyone else even being on the second floor. It was like the stuffed moose would move and sometimes run across the hall and press the button for the elevator. Door would mysteriously open… there was the stuffed moose…. door would close. We wound up calling it “Visiting the Moose” and it was local lore that the stuff moose knows how to work the elevator.
Elevator company could find nothing wrong- all buttons, switches, motors, sensors were working fine. There was a full printout of the relay control logic and wiring at the panel, and yes, it was all in Ladder Diagram. Dozens of relay coils and contact points, buttons, sensors, indicator lights, etc.
I spent a week reading that diagram and following the relay logic, and watching the relays (luckily with indicator lights on each) cycle as the elevator did its thing. I eventually found it. It was a race condition between a set of contacts closing on one relay and the elevator position sensor relay for floor 2. If one relay dropped out just a split second too slow before the other pulled in it created a false “stop at floor 2” condition. Since relays do take a finite amount of time to transition from one state to the other this can indeed cause issues in close timing applications. The fix here was to move 2 wires from the offending relay contacts to a relay that was actually controlled from that relay. The extra few milliseconds delay as one relay cascaded to the next was enough to prevent the race condition and the elevator stopped mysteriously stopping on floor 2.
I still swear the moose moves though while I’m not looking…