My PM-940 CNC Modifications

I've got some good news... ok, so it is a bit of both, good and a far bit less bad news. If you are like me, its good news, because I found what I believe to be the reason my mill has more backlash than it should and it is fixable. The bad news is, it probably will cost 6 angular contact bearings and milling some new bearing / stepper motor mounts to do this the "right" way however there may be a way to do it inexpensively too ( see below ). In explaining this I am going to focus just on the X axis but this should apply to all three axis. On my PM 940 CNC the X axis stepper motor is on the left side of the "table". A nut is used to tighten a set of bearings that fix the end of the ball screw directly to the "table". This nut, when tightened should apply enough force to remove backlash but the bearings should allow the ball screw to rotate without any friction unless the nut is overly tight. I had tightened mine up sufficiently to ensure I *should* have a minimum amount of backlash defined by the balls and the screw itself. The ball screws in the PM940 are supposed to be be of a good grade, so I was hoping for the best. Sadly, I'm seeing a minimum of .002 of backlash. I kind of put it out of my mind as other things needed attention but it always bothered me that my mill had a significant deficiency. HOWEVER, I remembered something I read about ball screw ends from china sometimes are designed poorly. Basically they use two bearings but stupidly the inner races are allowed to touch each other. Basically the two bearings are back to back, acting as one bearing. So even if they were two sets of angular contact bearings, if the inner races are allowed to tighten against each other they have no value other than allowing low friction when turning. The bearings side play is then what defines the amount of backlash. Check out my little video that shows the two bearings spinning together with just a small amount of force squeezing them together by hand.
Also, they don't appear to be angular contact...

One additional detail is that the right side of the table also has a nut assembly that holds the hand cranks. When I first got the mill I removed all the hand cranks and nuts to ensure that the ball nut assembly would not be bound up with opposing force.

There is one possible way to do this on the cheap and use the existing castings and possibly the existing bearings. Just find a sufficient washer and place it between the large outer races. I haven't tried to remove the existing bearings because they appear to have been pressed in with no way to press them out without damaging ( pressing via the inner races is not good, but maybe necessary ), I haven't tried. I'm going to order replacement bearings before I try to remove them. Then I'l at least have a way to machine something that willl replace the castings, etc, if needed.

sleepy... gnite

Jake
 
Last edited:
Great post!

Careful with those bearings you spec'ed... doesn't look like they are sealed.

PZ
 
Great post!

Careful with those bearings you spec'ed... doesn't look like they are sealed.

PZ

Good point. I wasn't thinking about the exposed side ( toward the table ) !

btw, I was thinking that maybe you might need to remove the hand cranks as they might be causing some additional drag. If you have little backlash it might be that both sides are opposing each other, giving you very little backlash, but could wear out your ball screws quickly... if that is indeed what is happening. Hopefully, yours is just fine the way it is though...

Jake
 
Last edited:
I defiantly want to dig into mine. When I bought this I bought it as something that I hoped I could upgrade to be on par of very near a Tormach 1100... So motor/ball screw mounts are in my future for sure.

PZ
 
Ok, some rather good news related to the backlash fix. As noted above I dug into my X table ball screw end bearings on the left side. Tonight I pressed out the existing bearings using a harbor freight hydraulic press. Added a McMaster 35mm ID 45mm OD spring steel shim in **between** the two bearings Large OD races of the existing angular contact bearings ( a side note... one had its shield removed so it was exposed so it was pretty dirty ). Then I pressed the old bearings back in sandwiching the single shim in between the two bearings and voila... I didn't do much of a test due to lack of time but my X axis is now below .001 for sure and could very easily be at or below .006 , whereas it was clearly .002 before. So I'm pretty happy now. New bearings on the way and I'l do the Y and Z when they get here.

Note, I still believe it may be necessary to remove the "other side nut and hand cranks or at least loosen it to allow it to have lots of slop. Otherwise you could end up ruining the ball bearings or the ball screw itself.

and here is the link to the the 35mm ID x 45mm OD x 1mm shim I used :

https://www.mcmaster.com/#98055A346

Cheers,

Jake
 
Well this is good news! Did you order the bearings from VXB you linked above?
 
Well this is good news! Did you order the bearings from VXB you linked above?
Yes, hopeful they will get me by until I have time to make new bearing holders, then i’ll fork out the money for higher quality AC bearings with shields.
 
TomS,
Not much to see really, its fairly simple. My first posts on this thread actually outline what I used to implement the pulley and belt system. The only "tricky" part was to create a key for the splines on the spindle. ( Whoa ! splines and spindle use the same letters !!! Spooky ! )

Here are a few pictures. The secondary pulley ( red 3d printed part with a small belt ) is for the encoder for rigid tapping. I'm still working on that... so stay tuned....

Nothing sexy yet... once I get to rigid tapping I'll be cranking more stuff out to make the mill look better.

Cheers,

Jake

IMG_0997.JPG
IMG_1003.JPG
IMG_0999 2.JPG


IMG_1001.JPG
 
TomS,
Not much to see really, its fairly simple. My first posts on this thread actually outline what I used to implement the pulley and belt system. The only "tricky" part was to create a key for the splines on the spindle. ( Whoa ! splines and spindle use the same letters !!! Spooky ! )

Here are a few pictures. The secondary pulley ( red 3d printed part with a small belt ) is for the encoder for rigid tapping. I'm still working on that... so stay tuned....

Nothing sexy yet... once I get to rigid tapping I'll be cranking more stuff out to make the mill look better.

Cheers,

Jake

View attachment 257212View attachment 257214View attachment 257215

Thanks for the pictures. Looks like you have eliminated the spindle drive sleeve with your setup and "locked" it in position. I'm still using the drive sleeve because I need about an inch of spindle travel to actuate my tool changer. Guess I'll have to live with the spline noise.

Not my mill but the configuration is the same.
pulley3.jpg
 
Back
Top