What's new
What's new

How to align linear sleeve bearings?

dbooksta

Plastic
Joined
Sep 28, 2016
Location
Philly suburbs
I have an assembly I need to ride on two linear motion rails, so two sleeve bearings per rail. I have the assembly tapped for the bearing blocks, but if I tighten all blocks the bearings bind. If I let one of the blocks on the rail "float" by loosely attaching it with only one screw then the rails move smoothly (see photo below). How can I secure all the blocks without binding?

I'm assuming I have to shim the floating blocks, and then I'll be able to screw them tight. But I've never tried this before. What tools, materials, and tricks can I use to accomplish this?

IMG_0644.jpg
 
In my experience, the first thing you need to know is the geometric conditions of all the elements, in terms of flatness and coplanarity of mounting surfaces. Typically I do that sort of work with a surface plate and indicator to determine where things are. Then you have a starting guess as to how to shim the bearing blocks, or conversely, how to "correct" the part to which the bearing blocks get fastened.
 
How are the two cross-pieces attached to that piece in the center? My guess is there is twist between the two cross-pieces.
 
The cross-pieces are grooved and screwed to the center piece, which is rabbetted to fit the grooves. There is undoubtedly some twisting from that. If I could somehow plane the whole assembly perfectly flat I would (though I have no access to machines large enough to do that), but even then I don't know if the bushing blocks are perfectly identically dimensioned....

Suffice it to say: I'm stuck working on the assembly I have, as this is a one-off I picked up already done to this stage.
 
How are the two cross-pieces attached to that piece in the center? My guess is there is twist between the two cross-pieces.


I agree
Its a pour design
The 4 positions of the bearings need to stay in the same plane with all forces involved
You can machine the assembly in one go to get it all alligned
If you cant do that because your machine is not big enough you can drill and tap some holes just beside the mounting holes in the frame
With those you can adjust that last bearing in such a way it all runs smootly
Then you can fill the gap between bearing and mounting surface with a hard epoxy suited for that purpose

You also have to make sure all the mounting holes are alligned properly

Peter
 
In my experience, the first thing you need to know is the geometric conditions of all the elements, in terms of flatness and coplanarity of mounting surfaces. Typically I do that sort of work with a surface plate and indicator to determine where things are. Then you have a starting guess as to how to shim the bearing blocks, or conversely, how to "correct" the part to which the bearing blocks get fastened.

IIRC the alignment sensitivity of the bearings is less than half a degree. Even if I could draw it out I don't think I can drill and tap a layout like this to that tolerance, so I'd still have to "hand-shim" in the end.
 
If you cant do that because your machine is not big enough you can drill and tap some holes just beside the mounting holes in the frame
With those you can adjust that last bearing in such a way it all runs smoothly
Then you can fill the gap between bearing and mounting surface with a hard epoxy suited for that purpose

"Shimming" with epoxy is a good idea!

I'm not able to understand what you're describing as far as tapping adjacent adjustment holes. Can you elaborate?
 
You have 4 holes in each bearing to mount it to the assembly

So you also have 4 holes in your baseplate for each bearing
(I would drill holes through the baseplate and use a nut and bolt to hold the bearing in place if you cannot drill to the exact rigt position You have more sideplay that way as with a threaded hole in the baseplate)
Now drill and tap 4 holes just beside those holes and put some screws in from below
With those srews you can push against the bearing lifting it up
No you have 4 bolts to pull the bearing to the baseplate and 4 to push the bearing off the baseplate
With this you can adjust that bearing in the right position

Peter
 
Are those countersink head machine screws, never had any luck with using them in aligned uses? CTB for socket head screws tighten as you move carriage back and forth may give the result you need. This design is not the best but it might be salvaged.
 
Since the mount surfaces are fixed by your middle bar you just need a plate, dti and surface gage and the ability to use those picture below.
(Picture shamelessly copied from member Mcgyver's post's here).
Doing so allows you to make anything fit.
Yes the bearing housings are the same.
Also at some point those bars are going to get mounted, that parallel and twist must also be held.

Easier way is to make sure everything involved is machined nice and square and same size first but sometimes you are refitting onto places on big machine castings and you have to hand work in mounting areas where only a fool would try to attach a precision slide.

I'm not so much for epoxy shimming as it ages and I want my machines to outlive me.
Have used Devcon, did not like the idea and only as a filler so it was not under load.
When done this way I put jacking set screws in the block.

As mentioned what are the mounting bolts? Flat contact only here, no angled c-sinks. Holes are designed with clear to allow a mounted assembly to align to the rod mounts. A SAE flat washer also helps prevent tightening torque and friction from moving the block around.
Bob
 

Attachments

  • Scarpers.jpg
    Scarpers.jpg
    95.8 KB · Views: 104
Anytime i've ever done a linear slide used a floating pair and a ridgid pair of bearings. For instance for each style of Pacific Bearing they have one that is fixed and one that is on o-rings. Its hard to get two rails perfectly parallel. Using the fixed bearings on one rail and the compliant ones on the other is what they are made for. Maybe this help...?
 
If you can't provide machined mounting surfaces that are in the same plane or in parallel planes, I would go with Peter from Holland's approach. Three or four grub screws from below the bearing blocks would act as levelling screws, allowing you to adjust the vertical position and tilt of each bearing blocks. The normal mounting screws would hold the block against the levelling screws. I suggest adding blocks beside the bearing blocks with grub screws so you can fine tune the horizontal position of the bearing blocks.
 
I specifically used countersunk screws thinking that I would want those heads to counteract any slipping force. But yeah, now I realize that requires hopelessly high precision, so I'll go with flat/pan-head screws and washers. (Also peak lateral forces really shouldn't be that high in this application.)

Grub screws sound like the best solution, and using through-bolts instead of threading the base to hold the blocks should make that easier.

(And yes, I still have to get this thing to roll with the rods mounted, but I figured if I can't align them separately I'd have no hope of straightening them them out with both rods locked down!)
 
Yep,

I agree with art, CarbideBob, and wawoodman. Ditch the countersunk screws, and use SHCS. Those blocks need to float to find a natural alignment as they are torqued down, those CS screws are forcing the blocks into misalingment as they are tightened. Now, if your assembly has twist in it, or your blocks are not all on a reasonably flat plane, you will still be out of luck until you fix that.
 








 
Back
Top