linear motion binding

[tmbg]

I have a sliding block that's mounted under my table which the gantry will be attached to. The leadscrew is in the center, the rails are 1/2" thomson rail, unsupported. Overall unsupported length is 11". The rails are 9" apart and in the same plane as the leadscrew.

The slide block is 1" thick, 2" tall, and has oilite bushings for the rails.


When I assemble it, the slide block glides freely and effortlessly in one direction, but locks up like a bank vault in the opposite direction. You can kind of "walk" it back and forth down the rail in very very tiny steps, and if you hold your mouth just right, you can find the exact angle that breaks it free enough to slide, but it usually won't slide for long before locking up. At any point you can reverse directions and it'll slide smooth as butter.

What likely causes this? Is it a rail parallelism issue?

[tmbg]

here's video of the problem:

YouTube - X axis binding

YouTube - X axis binding

[Lucanus]

So, basicly the dimensional properties of the setup exacerbates even the smallest play in the mechanism. The block gets out of alignment, acts like a lever and the whole thing binds up.

Possible solution:
Add 2 more oilite bearings on a block parallel to the one you allready have, spaced at about 3 or maybe 4 inches.
Maybe even try to move the guide rails closer to the centre if the design permits. This would also help with the binding issue.
Or as a third option, change your bearings to 2 of them fancy linear bearing dohickeys, as those as much less prone to binding.

Good luck with your project

[ger21]

You need two blocks per shaft, the further apart the better. And if it's even slightly out of parallel, it'll still bind.

[tmbg]

I was trying to avoid making the block too long, as that'll significantly reduce my travel. I have a couple of thomson linear ball bearings coming today; I'll see how that works out.

[LeeWay]

That's called racking and it's pretty bad since it is so obvious in the video.
That isn't easy to get rid of especially with unsupported rails, but addition blocks would help as was suggested. If not that route, then pull the rails apart to the farthest extents it can go. Actually just an adjustment from where they are now, but a bit further out to help ease that racking action.

The other thing would be to use two screws. Such a small axis could be driven by a single motor on two screws with a belt drive.

[BillTodd]

If you don't want to lose travel, you could cross wire the two sides together using four pulleys in the corners (search the forum for wire & racking)

[tmbg]

Originally Posted by BillTodd If you don't want to lose travel, you could cross wire the two sides together using four pulleys in the corners (search the forum for wire & racking)

Yeah, that had occurred to me as well. I'm hoping the ball bushings will solve it; they allow for up to a degree of angular misalignment I believe. If I still have problems, I'll probably start looking at a cable solution. I'm not crazy about that solution though, because this machine is small and it'll be tough to shoehorn that all in.

[BillTodd]

Got room for a pair of scissors?

What sort of load is it going to drive? (i.e. well balanced, cutting force, etc.)

[tmbg]

Originally Posted by BillTodd Got room for a pair of scissors? What sort of load is it going to drive? (i.e. well balanced, cutting force, etc.)

It'll drive the gantry, which should have light to moderate cutting forces. I don't really have numbers, I'm doing all of this seat of my pants.

I'm having a hard time picturing your scissors arrangement. Got a visual for me?

[BillTodd]

I'm having a hard time picturing your scissors arrangement. Got a visual for me?

I don't know if these will help but, I had the same problem with a rack light unit I designed in the end I used a scissor mechanism to stop the thing racking and locking.

One of end of the scissor lever is fix (free to rotate only) the other end can slide (perpendicular to the axis)

Bill

[tmbg]

I'd be concerned that I wouldn't have enough overall width to handle the length of scissor needed for my travel. I'll look into it though