Mechanism for automatic release at specified travel

[tinkertech]

Does anyone know of a mechanism that can release a one-way clutch automatically at a specified travel distance? My need is for a fixed reciprocating motion for both input and output, but the output needs to retract quickly under spring tension at the end of the stroke, independently of the input return stroke. Then the clutch needs to re-engage again when the input completes its return stroke.

A ratchet and pawl probably won't work for me because it's subject to wear and the force on this will be fairly high during the power stroke. I can't risk the tips of the ratchet teeth or pawl breaking under load.

 

[desertfox]

Hi

Any chance of a sketch of your device, makes it easier for us to see the problem.

desertfox

 

[zekeman]

A 1 way clutch is probably not the best solution. You probably need a cluch-brake package that will clutch in the forward motion and at the end of stoke sensed by , say a microswitch will release the clutch and momentarily actuate the brake. After a pause, the brake- clutch system is deactivated and the spring return will end the cycle.
If you have air available, then a pneumatic solution is a candidate.

 

[tinkertech]

We are early in the design phase and have no fixed design. That gives us flexibility, so the necessary mechanism can be linear, rotary, arcuate, or any other shape, as long as it accomplishes the goal. Criteria include simplicity, cost-effectivness and most critically failsafe robustness.

Attached is a simple graphic of what the mechanism should accomplish. Since we have nothing designed as yet, we don't have any drawings.

 

[tinkertech]

zekeman,

Unfortunately, no outside power source such as pneumatics, hydraulics or electronics will be available for this. It has to be simple, extremely reliable and strictly mechanical. Sorry, I should have specified that.

 

[desertfox]

hi tinkertech

theres no image with the link

 

[unclesyd]

As a WAG would a disk and pin (crank pin) work?

It could be done with a cam Circular. but would be a little more a little more complicated.

 

[unclesyd]

You might also want to look at non-circular gears.

http://www.cunningham-ind.com/typical.htm

 

[MikeHalloran]

The "extremely reliable" spec may rule out springs, unless you pay a lot of attention to related details, or the design lifetime is fairly short.

I'm thinking cams.



Mike Halloran
Pembroke Pines, FL, USA

 

[chicopee]

I think unclesyd has something-- elliptical gears may give you what you need.

 

[tygerdawg]

One of the things I don't see in the description is what is the nature of your prime mover? Is the input a constant RPM source of torque? Variable? Linear stroke input? What?

One thing that comes to mind is a mechanism sort of like a ratchet-n-pawl on the 2nd element that captures the 1st element at the beginning of stroke, then by virtue of a linear cam will release the 1st element at the correct point in downstream stroke.

The other idea was cams & cam followers.

There's a bunch of books out on the market for clever mechanisms...do a search on Amazon for them. That may give you the ideas you need. But ultimately you will have to synthesize the motion requirements and design mechanisms to produce the motion, which is an "old school" machine design task.

TygerDawg
Blue Technik LLC
Virtuoso Robotics Engineering

http://www.bluetechnik.com

 

[IRstuff]

Cams, or Geneva mechanisms, there are literally tons of mechanisms for this sort of thing.

Ingenious Mechanisms for Designers and Inventors dates back to 1935, and is no longer under copyright:
Ingenious Mechanisms

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[tinkertech]

Thanks to everyone for their input. The requirements have been refined a bit in the last few days. Basically, what the mechanism has to do is accept downward input from a horizontal lever for a short arc, only about 20º of travel. An output lever has to match that travel, driving a piston downward. At the end of that 20º, input and output levers must disengage and the piston should immediately and quickly retract upward without force from the input lever. Because of the free-return aspect of the retraction, something like the Geneva mechanism won't work. We are looking at something that uses a form of cam, but the limited travel makes cams rather impractical.

 

[zekeman]

Basically, a 20 degree segment gear engaging a spring loaded clutched gear could do it.

 

[HDS]

IRstuff the link you provided said the book was still copyrighted and only showed previews for all the versions. It looks like it is still in print and for sale from the publisher. (It is a 4 volume set.)

tinktech
A counter weight maybe an option.

 

[gadkinsj]

What about a bicycle freehub? I asked an engineer at Chris King (a high end bike hub manufacturer) how much torque one of their freehubs can handle and he said it could mate up to a Chevy 350 engine.

 

[unclesyd]

Here is the operation of a Whitworth mechanism as used on one type shaper.

http://www.youtube.com/watch?v=Dpw_xAvo5kk

http://shopswarf.orconhosting.net.nz/b/shc.html

Here is a very simple mechanism that may do the job.

http://robives.com/blog/quickreturn

I'm try to find a drawing of the mechanism used in high speed perforating machines, essentially he same as sewing machine head.

 

[tinkertech]

Thanks for the tips, but none of those will work. The sticking point has always been that the return MUST be independent of the input. However slowly or quickly the input is moving, the output must retract at a single, high speed. It's like pushing a rock toward a cliff. It doesn't matter if you push it quickly or slowly. The moment it gets over the edge, it will fall at the same rate. All of these mechanisms tie the output retraction to the speed of the input.

Again, thanks to everyone, but we've more or less figured out a mechanism that accomplishes what we need.

 

[cloa]

At least conceptually, what mechanism would that be?

 

[chicopee]

Rube Goldberg type