[biojae]

From what I saw at several events (Arizona, Denver) I thought my team (399) was the only one doing a kicker like this, Good to see some people think alike!

The way that ours worked was similar to all above, but doesn't have a hard limit switch.

The hammer is tied to a CIM powered toughbox, geared down slightly with chain. On the toughbox there is a quadrature encoder going to the controlling jaguar.
When kicking there is no limit switch for stopping the motor (It is powered the whole way down, needless to say we broke several limit switches), only the encoder and the current sensor.

The encoder messes up sometimes when kicking the ball (chain stretch, etc.) so, the current sensor is a backup (saved many a mechanical stop, and properly ends the kick mode, returning to positon control mode (look at Az. regional webcast for vids without the current sensing in (all of them), wouldn't properly terminate the kick (the deflector's code would terminate it properly thus the many actuations) thus the weak kicks)

We have two modes for the kicker, SET and KICK!!.

When in SET mode, the jaguar uses PID in Position control mode to control the placement of the hammer (The higher it is set, the more power it can apply to the ball). There is a neodymium magnet in a plastic enclosure on the kicker arm that is used to zero out the encoder pulses. The reed switch that it passes by is connected to the digital sidecar, and on pulling the line low restarts the jag's position control.

When in KICK!! mode, the jaguar is placed in voltage control mode. full bus voltage is applied to the CIM and is continued until the encoder passes a set value, or the motor stalls. After either condition, the kicker is immediately retracted and returned to SET mode.

The hard stop we made is designed to absorb the excess energy (and there is a LOT of it) and dissipate it through the frame.