[thecook]

hi,
I salvaged some linear rails with integrated ball screws which seem to be extremely robust. It feels like the X-Y axes, which are bolted together, must weigh 60kg although a lot of that is housing.
I'm wondering if I could make a CNC mill with them, a vertical mill (similar to the Tormach 440) to cut aluminum, harder metals if possible (but not a show-stopper if harder metals not possible). Unfortunately this is my first CNC machine d.i.y. build and despite reading about them for years, I feel like I could be missing something obvious why these rails would not be usable. The only thing I can think of so far is that the separation between the linear ways inside either X or Y axis is small so I might have quite limited usable milling space due to the rails not being able to withstand much torque.
Does anybody have any thoughts on if they'd make for good rails for the intended machine?
regards,
Randy



Here's some additional info I found from the company making these rails:

MOLS 100.500/100.150 (x&y rails connected together)
Recirculating linear ball bearings
Covered by bellows for protection
0,05 mm auf 300 mm precision
Travel 500mm on the long rail and 150mm on the short rail ?
spindle "KGT 12x5mm"
built-in inductive end switches on both ends
~$6200 new


ERO-SPEED 2000S/100 (was on z-axis)
Ball screw
Covered by bellows for protection
Travel speed at max: 20 m/min.
Repeat exactness: ± 0,01 mm
spindle "KGT 16x5mm"
built-in inductive end switches on both ends
Bearing number dyn.: 10700 N
Dyn. Load moment Mx: 90 Nm
Load moment My: 44 Nm
Load moment Mz: 44Nm
Travel 10cm I think.
(more specs here, p.18 under "ERO SPEED 2000SF", Page Not Found)
~$1500 new

[awerby]

Those are called "automation actuators". They can be used in the construction of CNC machines, and have some advantages in that the screws are already mounted parallel to the rails in a fairly rigid structure and the whole thing is sealed up against chips. It looks like those are good rails, and the screws are probably of fairly high quality as well. Even if you can't figure out how to run the motors they come with, you can usually substitute servos or steppers you can drive and end up ahead of people starting with loose rails and screws. But how well the machine you build from them does at cutting metal will depend on how you deploy them. If you attach them to a very rigid frame, and restrict the Z axis movement so as not to get leverage working against you, it should be possible to cut aluminum. I built a machine from actuators but went with a deep Z axis, so I've just used it for wood, which works pretty well.

[thecook]

Thanks a lot for that input. It sounds like it might be worth investing some more time into trying to make something out of these rails.
Do you know if there are any tests I can do to know if the ball screws are still good, that they're not worn out or otherwise not really usable? They are reasonably hard to turn by hand but I guess maybe that is normal?
The rails did have some servo motors attached, Kollmorgen-Seidel, 2x 6SM 37L-4000 (1.5 Nm) on X&Y and 3SM 37M-4000-G (1.3 Nm) on Z.
They do not look to be easy to use and the only driver I found, the ServoStar, was extremely expensive and I believe I’d need three of them. I am thinking of using those Teknic all-in-one servos, like https://www.teknic.com/products/clea...c-servo-motors. At least, they look nice to my untrained eye.
It seems to me the only thing I need for a frame in this case is a largish, flat-bottomed “C” shaped piece of metal where the X-Y rails attach to the inner bottom part of the C and the Z axis attaches to the front of the top part of the “C”. I am restricted in terms of using heavy metal pieces for the frame since I have to carry everything down a flight of steps to my basement, so, nothing over about 60-70kg. I thought of making this C-shaped mounting part out of multiple plates that have the same C-like cross-section then bolt together to form the same shape but much thicker, heavier and hopefully more rigid.
I think my next step is to figure out how to make the frame and if that works, then buy some motors & controllers to test it out.

[awerby]

Ballscrews should turn easily. But yours are attached to a stage that runs on rails, so there could be issues in there someplace. Most likely, if these actuators have been lying around unused for some time, is that the grease has frozen up and is causing them to act sluggish. Regreasing the rail trucks and ball nuts and running them back and forth should help, in that case.

Those Teknic motors look interesting. The problem with an "all in one" system like that is heat management - usually you want to be able to cool your drives, but having them stuck to the motors makes that difficult. If you get them, go with the step-and-direction version, so you can control them with Mach3 or similar CNC control software.

It should be possible to fabricate your frame in sections, and bolt them together in your basement. First figure out the optimum structure you want, and then break it down into pieces you can handle.

[thecook]

Ok, I will try re-greasing, maybe work it in by turning the spindle with a hand drill. Any hints on what kind of grease to use?

As for the heat management of the Teknic motors, this guy used them on his CNC and says that after 49min. of running continuously they were still felt like room temperature:

https://www.youtube.com/watch?v=Z2kAcz-rG5s&t=681


(link goes to timepoint where he says that but rest of video is also an interesting description of the servos)
I don't know for sure if that means they won't overheat since it probably depends on the cuts being made, but it seems to be a good sign.

I think I will try making the bolted-together "C" fame out of plywood or even MDF first. I wouldn't expect rigidity but I can get a feel for how it might work as a design and won't make an expensive mistake written in iron (or stone...why not?).