IT WORKS! (simple clutch mechanism)

[dbb-the-bruce]

I started a thread about this style clutch a week or two ago. After a few iterations, I've ended up with what's shown here.

This photo clearly shows the ramped / teeth of the driving/driven part of the clutch. When it spins CW, the balls get pinched against the outer ring and drive it CW also. When the inner clutch stops, the outer ring continues to spin freely.

First to be absolutely clear - I didn't "invent" or come up with this on my own. I found it in a book of mechanical mechanisms (yet another thread). In my case I was looking for some sort of one way drive that I could reasonably fabricate on a small scale.

Here is the clutch assembled into the hub of a flywheel. I've got a clock work gear drive that winds up the RPMs on the flywheel until the spring runs down. At that point the flywheel is free to continue spinning.



I made the steel carrier in the first photo so I could see what was going on with the different versions and fine tune the number of balls and the ramp design. It turned out to be -REALLY- useful to keep the balls in place while inserting the clutch into the center of the flywheel.

I'm pretty impressed with how well this works as is right now. The flywheel currently just spins on the center shaft and there is quite a bit of friction - surprisingly it still continues to spin quite well after the drive stops, but it has a lot of drag that back drives the gear train. I've got a really small ball bearing that I'm going to use to mount it to the shaft that should help a lot.

You can also see little indents in the brass clutch where balls jam. With steel against brass, the brass is already losing. I'm going to try cutting steel on my little desktop CNC (another learning project). That will allow me to make the drive clutch out of steel and use the carrier as the outer race in the flywheel.

I'll post a video of the windup gear box in action when I figure it out.
-Dave

 

[dbb-the-bruce]

Like a turbine spooling up for take off!

View attachment Flywheel.mp4

 

[Superburban]

Thats just cool. I would wear it out in a few days. Any idea what the speed is? Reminds me of the old airplane inertia starters, I have two that I play with when I run across them.

 

[brino]

Fantastic!

-brino

 

[dbb-the-bruce]

Any idea what the speed is?

Each gear stage is 12:1, so if you can time the slowest one and multiply by 144 that would provide a rough number.

I haven't done that.... yet. I've been meaning to get one of those point and measure RPM guns, I think it might now be time. I'm hoping that when I put the flywheel on the ball bearing it will take a really long time to run down.

I'm also in love with how long it takes to accelerate and that it just continues to do so. Pretty cool to have a mechanical system that efficiently converts the statically stored energy in the spring into kinetically stored energy of the flywheel.

 

[nnam]

Reminded me of lawn edge trimmer starter clutch, skip to 5:00 mark

I think the problem with this is that it doesn't drive if the outer part is moving, since the balls are pushed out, so it would slow slip.

 

[extropic]

The lateral acceleration holds the balls against the flywheel (outer body) which is one reason it has so much residual torque and back-drives the gear train.

I'm wondering why you didn't use a commercially available Sprague clutch like . . . ? The triple row type would solve all the issues.

If you "just gotta try", try putting the ramps on the ID of the flywheel rather than the OD of the drive.

 

[DavidR8]

That is exceedingly cool!

 

[dbb-the-bruce]

The lateral acceleration holds the balls against the flywheel (outer body) which is one reason it has so much residual torque and back-drives the gear train.

Not clear what you mean by "lateral acceleration" here, maybe centrifugal force? Regardless, the "back drive" is due to the friction of the flywheel on the shaft. I've played with it without installing the ball bearings and it does the same thing. The gearing in reverse provides the flywheel with a 1:144 multiplier - it takes very little force at the flywheel to start to wind the gear back up.

Someone else pointed me at commercial sprague clutches and other one-way bearings or drives. I did take a look at McMaster-Carr but most of the stuff was way to big and way to expensive!

One of the main goals this ongoing project was to learn the skills and gain the confidence to build stuff like this from design conception to a functioning mechanical / clockwork assembly. Specifically, making as much as reasonably possible myself.

 

[RandyWilson]

Pretty much every automatic transmission uses one-way clutches of some sort; be they ramp clutches like what you have or sprags. Sprags use elements like a lopsided figure 8. Turned one way, they rock in away from the races and slip. Turned the other, they rock into the races and wedge. The roller clutches use wide rollers rather than balls, and have accordion springs pushing the rollers into the ramp. This is so all engage at the same time for even loading.

You can greatly reduce your local load by using a roller rather than a ball.

A bit of trivia. In the never ending quest for more gears and tighter packaging, the new(ish) ZF FWD 9 speed used by several has a two-way sprag clutch. They somehow made it reversible. It locks one way until the computer flips some linkage, then it locks the other way. Hopefully I'm retired before those start showing up in the shop.