The block diagram of the smart electro-mechanical actuator, redrawn with the rod in place of the motor box. Follow a command down both ARINC 429 lines, through every check, into the supply — and watch the rod move.
What the manual says. For inconsistent APM or sensor data it says only:
a signal is sent to all the series actuators to freeze them
. It does not name that
signal, and it describes the discrete switching from valid to invalid
only for the case
where the rod position is inconsistent with the position called up by the control.
What the drawing suggests. BACON enters the actuator from outside, so the ON discrete is held up by the AFCS side, not by the actuator. Dropping it would stop all three actuators from acting on commands and would remove their validated signal at the same time — exactly the effect the manual describes. A worm screw does not back-drive, so the rod stays where it is as soon as the motor loses its supply.
Where to check it. Whether the freeze is the ON discrete being removed or a separate signal, this page cannot say. The AFCS architecture in 17.12.3 shows discretes marked BATON 1 and BATON 2 running from APM 1 and APM 2 to the series actuators — that is the page to compare against.
Arrangement, labels and wording follow the training material: 17.9 SEMA "series" actuators — operating principle and operation in the event of failures. Three actuators are mounted in series in the pitch, roll and yaw channels and receive their flight control and stabilization commands directly from the APMs.