only 1 end of the worm gear would really be in a place to use a micro switch.

yes, but outside you seem to have a plenty of room ?yes ?no -- as there is (likely) the (manual) AB-trigger switch

 

Using 'NO' contacts for the end limits would not be fail-safe. A loose connection could prevent an end stop and cause mechanical damage.

Here's a fail-safe circuit employing limit switches / microswitches with 'NC' contacts. Control is through a 'centre off' DPDT toggle switch.

It's a 'H' bridge reversing motor control with end stops.


Source: https://www.robotroom.com/DPDT-Bidirectional-Motor-Switch.html

Regards,

Nandu.

 

We made a prototype at work with similar mechanics. On both ends we used FSR's that detect contact with lowering resistance. They are paper thin and fit almost anywhere. I used the ADC on a microcontroller to detect contact but all you would probably need is a transistor, perhaps a darlington.

 

On both ends we used FSR's that detect contact with lowering resistance

how the force will keep it's strength . . . over time = there must be an "impact signal" triggering the power off relay

 

If the leadscrew/nut combination has a hard stop at each end that can stand being run into without damage, then there is a fairly simple way to do what is asked with no switches or encoder. The sensor will monitor the current and as soon as it detects the current rise from a stall, it switches off the power, and then sets up the H-bridge driver for the opposite direction. Do a search for "screwdriver antenna controls" and you will find several circuits that can do what you ask for.

 

If the leadscrew/nut combination has a hard stop

or if your motor uses a clutch

 

or if your motor uses a clutch

It will also work with a clutch, BUT the current sensor must be adjus5ted much closer to the running current of the motor, since the current rise will be less.