[LooseCannon]

Our makerspace has a Asian Bridgeport clone which was converted into a CNC mill. I knew nothing of CNC when I got there and neither did anyone else. As I have learned more about machining more and more problems are becoming apparent on the mill. I was hoping I could post here for some advice. There are two problems. The first of which may be able to be fixed. The second I will post a new thread to explain. The first issue is backlash. The conversion did use ball screws which should give little backlash. However, we are getting between .007 and .009 on each axis. When we tighten the ways a little the backlash gets even worse up to .025 and it is variable. Sometimes it is .019 and sometimes it is .025. If I had to make a wild guess I'd have to say something is flexing but I can't base that on any observation. Is this enough information to get advise? If there is anything else anyone out there needs I will be happy to find it out.

[123CNC]

Was it a new mill will converted to CNC? If well used, the ways may have greater wear in the highest use areas, typically near center. The other thing you may want to check is the preload on your AC bearings of the screws drive end. Tighter gibs may be loading this joint more than when running a little looser and thus revealing itself, or as you aptly stated more flex in this area. Depending on quality of conversions parts, care, and hours of use, the bearings may be worn as well. Of course you could also have variable wear in the screw as well, if frequently ran in the 'sweet spot' or poor lubrication to screw and or ways.

Adjusting the gibs is a good first pass in trying to isolate and identify the problem. Check screw repeatability along its length and try to mechanically force and try to measure any displacement at the ball nut to screw and separately at the screw to AC bearings.

Are you having any backlash issues with z?

[LooseCannon]

It was a new (maybe 10 - 20 hours on it) mill when we purchased it. I must admit I don't know what the AC bearing is or how to find it. Once I do find it how do I measure the preload? I have not measured Z. I didn't even think to do so. I will go into the lab in the next few days and give that a try. I will also try to grab some pictures. Is there any part I should focus on?
BTW- The mill is a converted RF-31 drill mill

[123CNC]

AC or angular contact bearing, used on shafts and spindles to help resolve forces along the axis of the shafts. The bearings are typically mounted on the ball screws drive end and often have spacers and a locknut to help load the bearing to eliminate its' contribution to free play or backlash. You may have a plain bearing on the opposite end of the screw or typical in short screws like your BF-31 it may be free floating. Here is a youtube example of a guy measuring backlash in his bearings with mounts at both ends. Yours' are mounted on your mill, but it will help you visualize what to look at and how to try and measure. https://www.youtube.com/watch?v=3aYmsXQegZQ

In the youtube example he is measuring play in the mounts by thrusting or applying force parallel to the axis of the screw and the fixed point of the gauge is common to the bearing mount. You should be able to do this on your mill as well, perhaps a magnetic base for the gauge will be handy. To isolate measuring any play in the ball-nut to screw interface will be more difficult. You will need to have the gauge mount and measure tip between the screw and nut, e.g. the table and screw, or the saddle and screw, not between table and saddle or between saddle and bed or knee. The later would be cumulative backlash, and your'e trying to find your worst actor(s), AC/mount bearing or ball-nut.

In a single ball-nut application, preload is typically accomplished during the manufacturers' assembly with properly sized balls or alternating 'properly' sized balls. Double ball-nuts may have a mechanical means to preload the axial load between the two nuts. Not very likely you have a double ball nut, but if so it may allow some adjustment to reduce backlash. If single nut, most likely, you can live with it, try to measure the balls and replace with properly sized (not recommended for novice), or replace ballscrew and nut.

The only reason I asked about z, is based on an assumption that the ballscrews are of the same manufacture and quality, it may help you gauge quality of the screws and nuts relative to the machines bearing surfaces. Where on the table and saddle you have adjustable gibs, the z has no adjustment.

Although your very low run hours almost rule out any wear patterns, make sure your lubrication system is functioning properly and getting everything needed wet, i.e. machine bearing surfaces, ball-nut, and possibly your AC or mount bearings, if they are not sealed bearings.

[LooseCannon]

I finally ran into the guy who did the conversion and last reassembly. He said he did check the screws when he put them back on. How it was done is he put the dti on the wooden table the mill was sitting on. He put the business end on the screw and pushed the table. Nothing moved. Then he put the business end on the mill table itself and pushed the table. Nothing moved. Far from perfect so it isn't a test negative. At least it gives us more information. I also tested the z backlash. It is about 2.5 thousandths. Much better than the table but still there is backlash. That too may give us information.