# A linear motion mechanism equivalent to a clutch?

A (rotary) clutch is a simple device: if the torque between its two axes is greater than a threshold, it slips. Is there a linear motion equivalent?

I can think of a simple arrangement such as two cylindrical magnets within a tube, each tied to a cable. pulling the cables strongly enough will disconnect the magnets. However, this is probably difficult to make accurate and repeatable at a specific desired threshold force. I can add thin separators between the magnets until the separation force is near my target value, but because of the very strong dependency on distance it will be cumbersome and inaccurate. An alternative implementation would be something based on friction (change from static friction to sliding friction) but again, I'm not sure how to make this a repeatable over dozens or hundreds of usage cycles, and adjustable and calibrated to a specific threshold.

I'm looking for something that I can set to disconnect at, e.g., 100N +/-5% or at most +/-10%. Bonus points if the threshold force level can be calibrated within some reasonable range, e.g. by some kind of screw. A nice-to-have feature would be that it will not move much, or at all, until the threshold force is reached and it separates, but it's not a requirement.

This is probably a common mechanism and I just wasn't using the correct search terms... surely this is a somewhat common requirement for a mechanism? Is there a name for this? off-the-shelf products? if not, can you suggest a custom solution?

• a wheel with a brake ... think of an automobile and a road Oct 19, 2022 at 20:49
• There are magnetic door holders that can hold over 100Lbs that are readily available: i.stack.imgur.com/s1aAU.png - it sounds like you want a reusable solution; would you accept shear bolts?
– Rob
Oct 19, 2022 at 21:09
• A friction clutch works the same for rotary and for linear. Oct 19, 2022 at 22:23
• @jsotola this again depends on friction, do you feel this can be repeatable within +/- 5% or so? in what arrangement? Oct 20, 2022 at 21:50
• @Rob do you think the threshold force is adjustable and/or repeatable within 5% or so? my gut feeling is no, but I'm not sure Oct 20, 2022 at 21:51

You may be able to create a calibrated electromagnet solution in the manner you describe. Two cables pulling in opposite directions. One cable has a known mass of ferromagnetic material (steel) while the other cable carries power to an electromagnet.

One can adjust the power to the electromagnet to provide a repeatable force on the steel attached to the other cable. Calibration could be as simple as hanging a known weight from the steel-ended cable, dialing down the power on the electromagnet while observing a meter referencing the power being used.

Depending on the precision required, you should be able to duplicate this release point many times. The power source precision will determine your release value precision.

• Nice approach! active rather than purely mechanical but perhaps that's what needed to solve this cleanly. Oct 20, 2022 at 21:54

Figure 1. Image source: Industrial Clutch.

Consider keeping the rotary clutch as these will be commonly available in a wide variety of specifications. I think most will use a cam-release like a torque wrench and should have long-term repeatability. Convert the rotary motion to linear using a rack and pinion.

I don't actually know if the device pictured (found in an image search) will do the job. I searched for 'torque clutch with rack and pinion'.

• interesting approach. I was hoping for something more minimalist and compact but perhaps it's not possible, while getting adjustability and repeatability, so maybe this is the way to go. Oct 20, 2022 at 21:53

Why don't you use to the sliding friction of each of the slopes with each other. One of the slopes could have a spring that could adjust the preload. It would always be repeatable, adjustable and self centering.

• thank you for this answer. However, while this will probably produce consistent force, it does not have the clutch behavior I was looking for in my original question. I want something that doesn't move, or barely moves, until the force on it is above a threshold, and then it moves freely. The threshold value should be repeatable, and in a perfect world, easily modifiable. Nov 2, 2022 at 11:15

The only simple mechanical solution I could come up with is a kind of linear captive ball-detent force limiter. I don´t know if something like this is on the market already.

It does use the basic principles of a already proven and repeatable torque limiting methods,and it is easy the manufacture and ideally adjustable. for calculating the force you can look at ball-detent torque limiters already on the market.