CNC – Z Axis rebuild

After setting up my CNC and running some tests and jobs I was quickly realizing that I was having problems with my Z axis losing steps. I was working on a carving and noticing the cuts were not consistent, my first clue. The image below was from a sample I ran, as it got further into the job I was noticing that the Z axis was retracting correctly and hence ruining the job.

bad_z

After running some more tests I finally decided to rebuild my Z axis and would try to use the CNC to do it. Since it wasn’t going to be highly detailed I wasn’t as concerned about the Z problem and losing steps.

So, I went looking around the internet and found a ball screw off of E-Bay from China.. go figure. Anyway it was extremely affordable, right at $100 shipped so how could I go to wrong. I found the specs and dimensions ans started drawing it up in AutoCad

dfu  sfu  SNAG-0045

SNAG-0054  SNAG-0052  SNAG-0052

7-2   7744d5ef   bfc71eb4

Since I was retrofitting the new Z I needed to make sure it would fit so measure, measure, measure.. I needed to make sure the ball screw would fit between the back plate and the Z plate since I wanted to continue to use the linear slides that I currently had. From what I figured out, the only way that was going to happen was if I shimmed the linear rails off of the back plate, the ball screw was simply too large to tfit.

IMG_2437   IMG_2440   IMG_2441

IMG_2443   IMG_2447   IMG_2444

IMG_2447   IMG_2457   IMG_2456

IMG_2449   IMG_2444   IMG_2445

Here are the CAD drawings if your interested in them..(New Z Plate Download)

   SNAG-0062   SNAG-0063   SNAG-0064

SNAG-0065  

Well the Z axis parts finally arrived, actually they arrived faster than I thought they would. Everything looks good, 2 pillow blocks, 2 couplers, jam nut, anti-backlash nut and ball screw.

IMG_2888   IMG_2887   IMG_2886

I had to make 1 modification the the ball screw and that was to grind a flat spot on it for the coupler to lock against, for a more secure set. I just pulled out my hand grinder and went at it.

IMG_2889

I wanted to verify the measurements matched up to what the spec sheet had and in general all looked good except for one of the pillow blocks.

IMG_2427   IMG_2426   IMG_2455

IMG_2429   IMG_2431   IMG_2430

IMG_2436   IMG_2433

The top pillow block mounting holes were not drilled correctly, so I modified my drawing accordingly.

SNAG-0034

I created my tool paths in MasterCam and generated the G-Code then ran a test piece on some 1/4″ plywood just to be sure I drew it correctly. All looked good and off to the CNC I went with a piece of 3/8 aluminum and my drawings. Sorry, no video or drawings of the milling process since it was first attempt at a larger aluminum cut.

Since my table is made of MDF and I needed some cutting fluid for cutting aluminum I came up with an idea to use plastic to cover the surface. Lowe’s sells a roll of sticky backed plastic sheeting for covering flooring when doing remodeling, etc. and that was perfect for this job. I covered the top, bolted down the aluminum plate, set the 0.0 reference and the Z height and sprayed on some lubricant, WD-40.The CNC did a great job at cutting the plate but there are some improvements that need to happen, chip removal and a misting system of some kind. It was quite a pain in the rear to keep spraying WD-40 and the cutter was re-cutting old chips, not optimal.

IMG_2869   IMG_2872   IMG_2873

IMG_2874

I then cut the anti-backlash nut holder out of 3/4″ aluminum using the same procedure above with WD-40. I fit perfectly, very cool…

IMG_2870

I had to tap some of the holes in the back plate and then it was ready for a trial fit. I can’t tell you how easy this was to assemble with the parts cut accuratly as opposed to had cutting them on the original build, very cool!

IMG_2890   IMG_2885  

IMG_2891   IMG_2882

Slotted holes for adjusting were an excellent idea for removing any major play when bolting the plate back on the machine.

IMG_2898

After the Z was fully assembled I tested the thickness needed between the anti-backlash nut holder and the movable plate on the Z axis and it was within .005 of an inch accurate, not bad for a home made CNC.

IMG_2883

All back together…

IMG_2897   IMG_2917   IMG_2918

And the final test…

IMG_2916

If I was able to help you figure out and fix your problem, buy me a beer! and Thanks..

btn_donate_LG

Leave a Reply

Your email address will not be published. Required fields are marked *