Anyone Using An Ajax Dm45nc Mill?

Yeah - I still can't quite believe it - I have been playing with the on board programming and have actually made a few cuts with it, but I need to make a fixture to hold the part I want to start with - this is just hobby. I'd love to have a plasma table, but the only CNC I have is the mill right now. My other CNC work has been at my work with a couple wood routers. The BT30 tooling is bit pricey but I have enough to get me started for now. I still need a couple drill chucks.
 
on crap sorry, got you mixed up with someone else! that'll teach me to post on meds :) either way, keep up with the pics and keep an eye on cl/ eBay for the tooling.
 
Ok - so I have the mill set up and have ran a couple test parts. On the round pockets, it looks like I have some backlash in the x axis - as near as I can tell, about .014. I know nothing about ball screws and have not started looking for this yet, but was hoping someone might give me some idea of what I might be looking for. With that much, I'm not sure the backlash compensation is adequate. The first part that I want to make has two curved profiles and I'm concerned about the surface finish and accuracy of the radius. The Y axis seems very tight. The gibs are tight and I can't find any slop in the table. So my question is - assuming the ball screw itself isn't shot, my first place to look would be the support bearings. Is there usually some adjustment to take care of backlash at the bearings? I'll have to tear the servo and mount off to see anything on this machine. It looks like there might be a belt drive, since the servo is offset from the leadscrew.
 
Let's start here, a little primer on ball screws http://www.barnesballscrew.com/how-a-ball-screw-works/

0.014 backlash is a lot for a ball screw, normally a ballscrew should zero backlash, backlash compensation in the software will not compensate when cutting an arc. There are three places where the backlash can come from. The ballnut, the support bearings, and the motor coupling.

I would first confirm the the motor coupling has no play and is tight on the shafts, we found this in alloy's machine the other day.

Next is the support bearings, leadscrews are normally tensioned in one of two ways, by putting the entire screw under tension with a hard bearing stop on each end and a nut that just puts tension on the entire screw, or one end is allowed to float in the bearing housing, and the tensioning is all done at the motor end by preloading the support bearings against the housing.

The final place that backlash can be generated is the ballnut. The ballnut could be loose in it's housing (loose bolts), or the housing may be loose on the saddle mount(again loose bolts). If both of these are tight the the last place is in the ballnut/ballscrew interface.

If the backlash is coming from the ballnut, then at 0.014, there is something drastically wrong. If you have a single ballnut then there is no hope short of maybe larger balls. If you have a double ballnut, then there is a an excellent chance of fixing it. With a double nut system, the nuts will have some provision for tightening against each other, thus preloading them. There are a number of different systems used to preload double ballnuts, normally involving a threaded coupling of some kind. In one case I recently worked on, there was a shim pack between the ballnuts, that got changed over to a threaded system because the shim pack proved to be impossible to adjust correctly.

I hope this helps
.
.
.
 
Jim - That does help - a lot! I need to tear it apart and start looking. I was pretty sure that much backlash meant I had something wrong! I'll post what I find out. It does have a handwheel on the ball screw, so that may help me isolate where the problem is (motor / drive or ball screw / mount, anyway).
 
Made it to the shop for a few minutes this evening and did some troubleshooting. I started by locking the gibs on the table and rotating the hand wheel. I could feel a definite click at each end of the backlash. I removed the cover from the drive and confirmed that the drive components were all tight. I removed the servo motor and placed a dial indicator on the end of the lead screw. I measured right at .014" end play. It looked like the play was in the support bearings. I started to take the retainers of on the drive end to see if I could see any numbers. The bearings did not want to push out easily, so I went to the other end and gave it a light tap. It felt solid, so I went ahead and pulled the end plate to see if there was a retainer of some sort and found a adjusting nut with a lock tab. I released the lock tab and found there was quite a bit of slop. I started tightening the adjusting nut and was able to adjust out all but about .002 to .003 of the backlash!
 
Awesome! If you have a single ballnut then that's about as good as it gets unless you want to re-stuff the ballnut with larger bearings. 0.123 balls are common, replacing those with 0.125 balls will give you near 0 backlash.
 
Wow- I missed an Ajax/Centroid CNC post. Man - I am busy!!! What you have there is a very sweet deal. I paid abour $3K for my AjaxCNC upgade. you have the DC-AIO (all in one) module as the guts . I have the same thing. This PLC is run from the Centroid system on a PC. There were early Linux versions but they have discontinued them. Do you know what version you have? So far- after using this for 6months - I love it. Just intuitive and fun. Let me know if you have any issues. I do have a print for the DC-AIO and yo ucan download the whole print. Keep us posteded. :)
 
OK- I poked thru the posts a bit slower. If the PC is running the Centroid Linux version - You should still be fine. Just that you need to know Linux (Unix) to navigate around the OS outside of the canned Centroid platform. I wonder if you can upgrade for a cost to the newer PC version.
 
I have version 1.02 and would like to at least get it up to the last version of the Linux software - 2.72, I believe (working from memory here). I have read the upgrade info and downloaded the install disk, but have not yet tried to make a new flash card or boot drive. I have a couple bootable flash card readers on the way off of ebay. I will probably add more RAM to the motherboard and have even considered trying to use a SATA solid state drive in place of the flash card. It looks like there are some pretty significant differences in the hardware between the Linux and Windows versions, so I doubt if I will try to go to Windows. I'd love to get to where I can use a flash drive to transfer files, but my version only supports floppy discs. CNC was one of those things I always wanted to do, but kept running into too many roadblocks - so when this thing showed up - even my wife said to jump on it! The on board CAM software seems to work pretty well and is pretty easy to use, so far (at least for 2.5 D work). This winter I hope to start learning Autodesk's Fusion 360 for more advanced projects. Oh - one of the programmers where I work is a real Linux lover, so I do have some support on learning what I need to there!
 
You must log in or register to reply here.
Back
Top