Aiph5u,
This would be a thrust bearing adjustment, Steve has already determined there is no play in the bearing, there is no adjustment that can be made to the ball nut that I am aware of.
Pat
Adjustment before replacement would be my preferred first course of action. However, I didnât see how to do that in what you posted. Could you please provide info on how to adjust ball nut backlash.
Hey Pat, hey Steve,
you are right, sorry, I had overseen that. So do we have to assume that there is play in the ball screw itself. Steve @Will.Sign.It, you did not provide info on how long the ball screw was used and how much dust it has drawn in that time period.
In use since May of 2022. Not sure how many hours on it. Usage since then several days per week at several hours per day. I have excellent dust collection always on and have only cut a variety of soft materials, never any types of metals. I have meticulously kept machine clean and lubricated with 3 in 1 oil as required. Has been stalled only twice in workpieces due to RRF (router rotational failure)⌠Not enough dirt or stresses to account for existing backlash in my opinion.
1 Like
Hey Steve,
my machine is still unused and not assembled and I plan to measure this on the brand new ball screws. But at the moment, doing this is not in sight due to time and health issues, but itâs promised to publish it in the forum then.
Tom @TMToronto already mentioned somewhere that there are ball screws with no backlash, that simply have two ball nuts that you adjust one to the other. But the Onefinity machine is not prepared for that.
To remove backlash and obtain the optimum stiffness and wear characteristics for a given application, a controlled amount of preload is usually applied. This is accomplished in some cases by machining the components such that the balls are a âtightâ fit when assembled, however this gives poor control of the preload, and cannot be adjusted to allow for wear. It is more common to design the ball nut as effectively two separate nuts which are tightly coupled mechanically, with adjustment by either rotating one nut with respect to the other, so creating a relative axial displacement, or by retaining both nuts tightly together axially and rotating one with respect to the other, so that its set of balls is displaced axially to create the preload.
â Source: Ball screw #Operation â Wikipedia
I remember also there was this mysterious issue here that was related to the seating of the ball screw.
1 Like
Interesting discussion on measuring ball screw backlash.
On my previous Onefinity machinist, I had no reason to check it because I was repeatedly getting X & Y accuracy of +/- .001" [.03mm] when measuring 2D profile cut parts. Even my previous Chinese machine with 12mm ball screws and 15mm linear rails was good to .002" [0.05mm] repeatability.
I have not yet checked my new machine, but I am now curious to do so. Not that I expect to see anything major, but I have not yet qualified the machine for accuracy & repeatability. I have no reason to doubt it, but it will be good to know before I do my next parts for Xometry.
1 Like
Excellent info gentleman. Thanks. I most certainly and highly recommend checking backlash on all axis before cutting anything on a new machine. I did not do that. I trusted it was good. Wish I hadnât. Only found my issue recently while cutting hard oak for the first time. Had to incorporate my experiences machining hardened tool steels and adjust my appropriate feeds speeds d.o.c way down just to reduce chatter to reasonable amount.
1 Like
I have Onefinity Woodworker that was upgraded to the Journeyman with about 2500 hours on it and zero measurable backlash in the ball screw/ball screw nut. Similar to what you state, machine in operation several hours a day, excellent dust collection with shrouds protecting the Y rail ball screws, and lubricated with 3 in 1 oil regularly (usually daily).
1 Like
For those interested, when I did my machine deflection testing, I wanted to see what was actually happening during machining.
I mounted the base of a dial indicator (0.01 mm increment) onto the frame of the Z20 assembly, and indicated off the bottom of my ATC spindle. I tried my best to keep the contact point on a tangent line perpendicular to the dial indicator travel. It was set up to measure deflection in the X axis travel.
The short video below shows the dial movements during a facing operation on roasted ash. I am using a 1" 3 blade surfacing endmill, with a shallow 0.5 mm DOC and 80% WOC.
Using double ball nuts had crossed my mind a while back.
When I upgraded to my X50 rail on my Woodworker, I noticed that there was a machined ball nut hole and 4 threaded holes on both sides of the X gantry block. This is not the case for the Y axis blocks (but perhaps on newer or larger models like the foreman?). I got me thinking that perhaps one could purchase and install a second ball nut on the opposite side, then use the mounting screws to carefully adjust/add preload.
It would be a fun experiment, but not one I think is required on these machines.
Glad to hear that your Journeyman has held up so well for you.
Hey Tom,
I think if you could machine some kind of aluminium âringâ to mount it here, one could add such a zero-backlash functionality. The ring would have a few more holes with threads in a circular way around the ball screw opening and allow the second ball nut to be attached at a specific set of two (or four on X-50) holes to implement the fine adjustment for such a zero-backlash screw. The adjustment would be made by loosening and then rotating the nut and attaching it again.
I really whish I could work on metal at the moment.
1 Like
That would definitely be a solution. I have also seen some builds where they just experiment with metal shims of varying thicknesses sandwiched and tightened between the flange and the ball nut gantry housing. Compression springs have also been used, as long as they are supported and are not allowed to rub on the ball screw itself.
2 Likes
You could also use a âcrush washerâ in this application, although they have disadvantages to a shim or shim stack. When torqueing them down, you have to be careful to not overtighten because once that has been done, itâs no longer usable, and they are not reusable.
Another option would be to assemble it using Plastigauge to determine your finished shim size. Plastigauge is basically a crushable wax type material used typically in engine assembly to gage crank bearing clearances. But the same idea applies here. It would get you close to the final shim size.
Honestly, if I were to pursue something like this, I would definitely go with a spring-preload method rather than a âhardâ assembly, which could be prone to introducing accelerated wear to the ball screw/nut assembly. A hard assembly would also not be advised to use on a less accurate grade of ball-screw. A spring preload however, will give the desired results without the concern of âgnats assâ accuracy required with shims. I believe it would also make for a smoother rotation of the ball screw. I may even look at solid rubber to use in place of a spring in this application.
2 Likes
I agree. I have seen springs often and effectively used with Delrin components on lead screws.