EEVblog Electronics Community Forum

Electronics => Projects, Designs, and Technical Stuff => Topic started by: rhb on March 09, 2018, 03:41:43 am

Title: Sticky linear rail bearing
Post by: rhb on March 09, 2018, 03:41:43 am
I'm building a CNC XY stage using linear rail and stepper motors.  Rails were purchased on eBay, mostly on price with bearing blocks.  These are supported 16 mm rails.  The blocks were rather roughly milled and bead blasted to improve the appearance.

The blocks vary significantly in stick slip and one was sticking so badly it would push the rail across the table rather than slide.  After discovering that I went through and systematically oiled and tested all the blocks.  I backed off the preload screws and ran them in to just make contact.  But the one block persisted in jamming.

I hosed the problem block down with WD40 and lubricated it with oil.  That made it run  almost as freely as the other blocks.  This behavior makes me think that the sticking was caused by dirt in the bearing.  A second shot of WD40  and oil improved performance more.

However, this is my first time playing with these.  The other 7 blocks contact the support rail when rocked from side to side. The problem block does not without significant force.  So there is still something wonky about it. 

Is this something I can reasonably expect to cure myself or do I need to return it?  If I can cure it what is the procedure?
Title: Re: Sticky linear rail bearing
Post by: Gregg on March 09, 2018, 04:10:52 am
Assuming the linear bearing is made in China and the type with lots of little balls inside, there is probably machining debris behind the balls that may or may not wash out with repeated applications of solvents like WD40 etc. Compressed air between solvent washes will help dislodge gunk. 
As a last resort you can probably take it apart to clean it, many have circlips holding them together but watch out for the deluge of tiny ball bearings.  The trick to getting them all back is to use some thick grease to stick them to the carrier sub assembly and then wash the grease out after it is all together. 
For Chinese mechanical stuff I always buy at least one extra of each part or 20% extra if I’m buying volume.  If you don’t need it right away, you will be sure to need it later.
Title: Re: Sticky linear rail bearing
Post by: rhb on March 09, 2018, 02:17:06 pm
The rocking anomaly suggests that the plastic ball race is elliptical which could easily happen if the preload screws were tightened a bit too much. Some of the horizontal preload screws were completely loose.

It now moves freely in the X direction as long as the transverse pitch of the block is constant, but will jam if I rock it while sliding it.

Deformed thermosetting materials will revert to their original if heated gently.  Any ideas what the ballraces are made of?  Thermosplastic materials don't take heating well.
Title: Re: Sticky linear rail bearing
Post by: raptor1956 on March 10, 2018, 01:43:00 am
Is this a single rail system or is it a dual rail/rod system?  If the latter then if the spacing between the rods is incorrect there can be enormous side forces that will dramatically increase the force needed to move them.  In fact, the general approach is to fix the block for one rob and let the block for the other rod float side to side.  Remember, any change in temperature will alter the spacing so the system needs to be able to float one side to compensate.

Otherwise check for a smooth finish on the rail/rods as well as any debris that might effect the bearing race.


Brian
Title: Re: Sticky linear rail bearing and a ballscrew question
Post by: rhb on March 10, 2018, 02:50:55 am
The block now moves freely.  A couple of heats to ~ 100 C with a hot air gun and it rocks almost as well as the others, and no longer sticks.  I'm considering submerging all of the blocks in mineral spirits, moving them back and forth and then relubricating them.

I got my ballscrews today.  The block that goes on the ballscrew is not a slip fit on the ballscrew.  Are these supposed to be pressed on?  I can run a cylinder hone through the block to open it up.  Given the 6 bolt holes it doesn't seem to me this should be a press fit.  Everything came from the same supplier.  I've got a 20 ton press but I'm leery of distorting the raceway. I measure ~0.004" interference with a dial caliper.  I'll put the mikes on it in the morning to get a proper measurement.

Edit:  This is a double rail, single ballscrew setup.   I plan to scrape all the bearing blocks to a common plane before mounting and aligning the rails using a  0.0001" dial indicator.  I've considered using a grade B surface plate for the mounting, but as these are supported rail I'm skeptical that is warranted.  All my testing so far has been individual blocks on a single rail.
Title: Re: Sticky linear rail bearing
Post by: raptor1956 on March 10, 2018, 04:18:17 am
Aligning the rails isn't the only thing to consider as a temperature change, particularly if you use aluminum mounting, will cause a change in distance due to thermal expansion effects.  So, even if you set it up perfectly with very little force required to move the blocks if the temp changes by even a few degrees it is possible the relative distance will change enough to apply side forces of tens, hundreds or even thousands of newtons.  Again, the typical way to deal with that is to have one block rigidly mounted to moving table with the second block permitted a degree of freedom side-to-side so that even with substantial change in temperature there is no significant increase in side forces.  The second block should be used to confine the motion of the table vertically and fore/aft but allowing that side of the table to expand/contract with temp change.


Brian
Title: Re: Sticky linear rail bearing
Post by: rhb on March 10, 2018, 02:11:53 pm
Very valid point.  As an exercise when I got my Clausing lathe I tried machining a test bar 16" long with a set of 1/2" wide collars to the tightest tolerances I could.  I quickly concluded that a coolant system was needed in addition to long waits between cuts to let the bar equilibrate.

There's a shop near me that makes stone counter tops so I had in mind a scrap sink cutout  for the base.

This project got started conceptually 7-8 years ago.  At the time I did a bunch of analysis and research.  Then things got stalled by other things and the project lay dormant until a few months ago. My sister and her husband have been staying with me while they get the house ready.  He has Parkinson's.  That and several close friends dying suddenly in the past few years loosened my purse strings.

I had given thermal expansion some passing thought, particularly the mixture of steel and aluminum, but no analysis.  Clearly I need to estimate expected movements and address them.  Thanks for the reminder.  For now I  think I'll focus on controlling a single screw and rail while I do a complete analysis.  Much of my interest is the gratuitous pursuit of precision.