# Converting rotation to reciprocating movement; long stroke with quick return stroke

1. Aug 24, 2011

### Sethodine

I don't know if I'm using the terms properly, since I am no mechanical engineer, but I am attempting to find the best way to convert rotational work from an electric motor to draw a long, high-torque linear stroke, followed by a very fast return stroke.

To clarify my question, I trying to build a mechanism by which to draw a bowstring on a crossbow, then quickly release the bowstring once full-draw as been reached.

A possible solution was posted in https://www.physicsforums.com/showthread.php?t=139122" some years ago, which I found while searching Google. However, I do not believe in thread necromancy, so I thought I would post a new thread with my specific application in mind.

Currently, I'm thinking of some sort of grooved cylinder attached to the motor. A sleeve with followers in the grooves would be pulled by the rotation of the cylinder. Attached to the sleeve would be the bowstring. When the sleeve reaches the end of the cylinder (maximum bowstring draw), the followers would be lead into a new set of grooves which would run nearly straight along the axis of the cylinder. This in turn causes the sleeve to quickly return to the "beginning" of the cycle, spring-assisted by the tension of the bowstring. Ideally, this would result in the bowstring being "released" with enough force to launch a projectile.

The problem is, I don't think this could be accomplished in a single rotation of the cylinder. I would need something more like a screw, but designed in such a way that the sleeve could return quickly once it has reached the end of the screw. I suppose this might be accomplished by using multiple, "staggered" followers, such that the sleeve will always be following the screw until it reaches the right pattern of return-grooves.

Does this approach seem practical?
If not, I would gladly accept suggestions for an alternate approach.

(Please don't discuss the practical application of an automatic crossbow. I'm a tinkerer, how useful a device is does not factor into my judgment of how worthy it is.)

Last edited by a moderator: Apr 26, 2017
2. Aug 24, 2011

### Unrest

Re: Converting rotation to reciprocating movement; long stroke with quick return stro

It seems like a waste of energy launching the sleeve along with the arrow.

Couldn't you use a rod with a hook on the end that moves slowly in both directions. It can go forward to hook onto the bowstring, then pull back, then using another kind of actuator, or keep pulling the string over some inclined ramp to unhook it and release it freely.

3. Aug 25, 2011

### Sethodine

Re: Converting rotation to reciprocating movement; long stroke with quick return stro

I was really tired when I first posted this; with a clearer head I can see that my original idea simply wouldn't work (not to mention being far too complicated). However, I think I might be able to come up with a similar mechanism.

Perhaps, rather than a "sleeve", the screw has some sort of hooked incline (like a dull drill bit) which can grab the bowstring and carry it to full draw. Upon reaching the end of the draw, the string simply falls off the screw and into the loaded position behind the bolt (arrow). A cam or peg attached perpendicularly to the end of the screw then fires the triggering mechanism during the next rotation.

I'd considered a piston-like device to simply draw the bowstring using the conventional crossbow draw mechanism (like the hook-and-rod you mentioned), but it seems like maximum draw would require a rather large cam/disc to be spinning perpendicular to the axis of the crossbow. I am afraid that such a cam would make the crossbow nearly impossible to aim (let alone hold comfortably) due to vibration and/or gyroscopic forces. And then you need to incorporate some sort of timing device so that the bow is triggered to fire only during the piston's down-stroke.
If the mechanism was moving slow enough these points wouldn't be so problematic, but I'm hoping to eventually achieve a system which can be scaled up to around 1 round per second.

However, if a screw is used to simultaneously draw and fire the bow, then it could be a rather compact device that could fit right under or above the "barrel" (you know, the plane on which the bolt travels along when the bow is fired). The axis of rotation would be parallel to the crossbow, which might cause some slight twisting in the direction of rotation, but will not effect aim as much. Furthermore, I *think* any gyroscopic twist inflicted on the bolt upon launch would only help to improve the aim of the projectile in flight, somewhat like the rotation of a bullet caused by a rifled barrel.

Hmmm... I think I have a lot to think about. Too much thinking, in fact. It might be time to start tinkering again.