Anyone tried using linear bearings?

Hi,

I have a Harbor Freight X2 that I converted to stepper motor CNC operation a year or so ago. I haven't used it much because of constantly losing steps and scrapping my projects. I finally turned my velocities and accelerations way down, but I am still unhappy with how the gibs move. I have to choose between tight and veeerrrryyyy slow or fast and loose.

Anyhow, I have been researching linear bearings and am curious whether anyone has used them on a milling machine in place of gibs. I was thinking round rails with continuous support mounted on an extruded aluminum T slot type frame. I see tolerances around .001" on the suppliers' websites, but I wonder how they would stand up to chatter and general milling forces.

My thought is to mount the table on the Y axis, then mount this all on the X axis, which would be on the bottom, mounted to the t-slot tubing. I think this would allow me to mount the Y axis with the motor in back, and allow me to mount the X so as to make use of the entire table. I would also like to extend the spindle head out a tad and increase my Y while I am at it. I would also mount the Z on linear bearings attached to rails on T-slot framing.

I am currently using 5/8" Thomsen ball screws and nuts, but am considering using an Ebay kit to replace all 3 axes' ballscrews, since I've lost several balls out of all 3 nuts while messing with them. Does anyone know if you can buy replacement ball bearings for them?

Yes you can buy just the balls and reload the nut. Mic one of the original bearings. You can order the next size up and eliminate a little more backlash.

I have seen linear rails used on milling machines. Never the tubular kind (not that they won't work) but the solid rail kind with the counterbore every few inches. The people I seen use them generally cut the base down some and mount them on the original ways or close to it.

It would be a LOT cheaper, easier, and more sensible to fix your steppers. They are obviously undersized, or are being under-powered. Lots of us have built X2s that worked just fine. Mine would do rapids at almost 300 IPM, and never lost steps, even with no springs or counter-weights on the Z axis..

Regards,
Ray L.

How big are your stepper motors? I am running a 400 in/oz on my X and 270 in/oz on the other two axes. I have a counterweight arrangement for my Z. I bought a kit from Probotix for it. I have a 24V power supply and I am running max voltage to the steppers with 1/8 steps.

Here is the kit I bought for the steppers: 3-Axis Monster Mill Stepper Motor Driver Kit - 3-Axis Monster Mill Stepper Motor Driver Kit

I am currently sitting here with Draftsight and BobCAD designing better stepper brackets, and I have new thrust bearings on order as well. I switched to Oldham couplers a while back and that also saved me a lot of headaches.

I am also moving my X ball nut to the outside of the saddle so nut clearance will be slightly easier to manage. lol I have had a devil of a time keeping the X ball nut secure where it is in the saddle. I also plan to put an end support on the X screw as well. My X screw is a bit over long, so I want to make stand-offs to maximize my table usage while I am at it.

Quote:

---

Originally Posted by SCzEngrgGroup

It would be a LOT cheaper, easier, and more sensible to fix your steppers. They are obviously undersized, or are being under-powered. Lots of us have built X2s that worked just fine. Mine would do rapids at almost 300 IPM, and never lost steps, even with no springs or counter-weights on the Z axis..

Regards,
Ray L.

---

Those motors need to be run at 48V to get best performance. So, no doubt a big part of your problem is the 24V power supply, which I'd guess is dictated by the (poor) unipolar drivers. A 48V power supply, and good BIPOLAR drivers capable of supporting that voltage (think Gecko G540....), would make a huge difference.

Regards,
Ray L.

Mine are little PCB drivers. I guess you can see them in the link...

I will research the Gecko drivers. I'm sick of crappy performance.

Think about what is going to happen to your max spindle to table measurement once you put a set of linear rails under both x and y axis. I can't believe you will succeed without losing a minimum of 2.2". Staring with 10.5", that takes you down to 8.3". Vise plus chuck, arbor, and drill wouldn't all fit anymore.

If you are going to spend the money to add linear rails to something, I would start with a more capable machine.