Quoted from JKnPA:
I didn't see anything broken on the 10K stepper that would actuate that switch for the Motor. Do you ???
Maybe. There's a spot on the rim of the 10k stepper gear at about 8 o'clock. Could that be a pin that has sheared off?
Here's a theory that's plausible but a little far fetched. It's more confusing since I think the schematic calls the 10k stepper a 10M stepper. Anyway, there's a Normally Open switch on the 10k/10M stepper that causes the score motor to turn when it's closed:
Ballerina motor (resized).jpg
It's labeled as "-1" which makes me think that when this thing resets it resets to a -1 position (like some Gottliebs do) and must take one step forward to the zero position as part of the reset sequence. When the stepper is in the -1 position the switch closes, calls for the score motor and presumably the score motor makes it take one step forward to its reset position. (There's a 2nd switch on the schematic also labeled the same way which would account for the second switch in the switch stack.)
The two switches at the bottom of the gear appear to be Normally Closed, which keeps your score motor running, right? Could it be that they should both be Normally Open instead? The photo in reply #10 seems to show a piece of insulating fish paper on the left side of the switch stack which would indicate that the switch stack is intended to be activated from the left side. (Is there fish paper on the left side? And does it have a wear spot on it where a pin might have been?) The spacer between the two switches also seems to be mounted on the right switch which seems to confirm that the switch wasn't intended to be activated from the right.
Putting it all together here's a scenario for how it might be intended to work:
Ballerina 10k unit annotated (resized).jpg
At reset the 10k/10M stepper gear rotates counter clockwise until the (missing) pin is at the 6 o'clock position. In that position the pin *closes* the two switches in the switch stack which starts the score motor. The score motor eventually causes the 10k/10M stepper to advance one step, which moves the (missing) pin to the left and allows the two switches top *open*. The 10k/10M stepper is then ready for a new game.
Or, did I totally misunderstand the problem?