Hubs

[m3ch4num470r]

Team 449 competed in Trenton last week and we felt we did pretty well, but one of our hubs kept coming loose. The hub connected to an arm of our forklift, so the 5 or so degrees it could move let the end of the arm move several inches. We use AndyMark 500 series hubs which we have drilled and tapped for 8-32 set screws on a 1/2 inch keyed shaft. We tried tightening and applying lock-tite several times, but one hub kept coming loose after every match. Does anyone know a good way to get those hubs to stay tight or know where we can buy better hubs that are still light? Thanks.

[Richard Wallace]

Maybe a picture of your hub installation would help. I'm having trouble understanding what is getting loose. Is the hub slipping axially (in the direction of the shaft's axis) or is it rotating relative to the shaft? Are you using a key, or does your set screw extend into the shaft key slot?

[m3ch4num470r]

Here's the best picture of it that I can get. My hand is conveniently over the area of interest.

The hub is rotating relative to the shaft. We are using standard key stock which one set screw pushes against (the other is 90 degrees over, pushing against the shaft itself). We think that a combination of the set screw managing to come loose and the fact that AndyMark doesn't perfectly machine their hubs is causing this motion

[DonRotolo]

Hmmm. Is it possible that there is some movement in that connection? In theory, if the screw is loc-tite'd it won't back out or loosen, so the hub stays tight on the shaft. But, th the hub can expand a little (or the shaft shrink a little), allowing some loseness, you'll see it loosen up a lot more very fast.

I can't imagine the shaft getting smaller. If the hub were cracked, it would come loose, but I assume you looked for that. Foreign matter interfering with the fit?

In the past, in the non-robotics world, I had a similar problem. Never figured it out, but replacing the sprocket fixed it - maybe that will do it, especially since the other side isn't coming loose.

By the way, you guys were awesome at NJ.

Don

[Martijn]

we actually have the same problem, the hubs are allowing our entire arm to rotate several degrees. we're thinking of just drilling a hole all the way through the hub and the axle once we get to competition, and then stick a bolt thru it. That should keep the whole thing from moving. Moral of the story, i think we'll try using hex axles and hubs next year so we dont have to deal with keys and set screws anymore..

[Richard Wallace]

Sorry, I had to go out for a couple of hours. Thanks for the picture and for clarifying how you're using the hub.

I agree with Martijn, hex bore hubs and hex bar shafts distribute the load much better than keys, and they make it easier to get a tight fit.

Also, dynamic loading on arms can be much higher than what the same hub would see in a drivetrain, because wheel loading is limited by the robot's weight and the wheel's CoF on carpet.

Have you calculated the static loading on the key slot of your shaft? The dynamic loading will be a multiple of that, due to bouncing and rocking when your arm handles the ball, strikes things, etc. High dynamic loads could have loosened the fit of the key by expanding the key slots in the hub and/or the shaft.

If you are correct about the hub rotating 5 degrees relative to the shaft, that's quite a bit. A little trig shows that a 5 degree shift on a half-inch circle is about 0.022" displacement. That amount of slop is ten times the "looseness" that I would typically allow on the fit of a key. Does the keyslot in the hub look deformed? Was the key filed too much? I'm guessing the key is steel and I know the hub is aluminum -- is the shaft steel or aluminum?

[m3ch4num470r]

Thanks for advice, everyone. We'll be replacing the hub with a freshly machined one tomorrow. Hopefully it'll last us through the competition. Next year we'll use hex shafts.