excessive Z axis slop. Help!!

[msomerville]

I have a VM30 from 1997 that has excessive slop in the Z axis. I first put the indicator on the flange of the spindle and I am getting about .015 of movement! Then I moved the indicator off of the flange and onto the head itself to make sure the spindle itself wasn't moving. Same results. I have checked the endplay of the ballscrew and I get .0005 difference on the indicator. I put the indicator on the ball nut itself and I do get .0033 of movement. I then moved the indicator right behind the motor to make sure that it wasn't moving between the two head pieces and there wasn't. I have checked all of the hardware for tightness and I am kind of at a loss right now.

Does anyone have any suggestions on where to look?

[single phase]

I take it that you are getting 15 thou backlash when you switch directions?

Record the backlash settings for all the axis and zero out the Z.

Why is there 3 thou movement in the nut? It is just that worn?

[msomerville]

I am getting the .015 movement when I use a pry bar to do the "push pull test". I think the nut is that worn. I am hoping by replacing the ballscrew that this will fix the problem.

[M_Controls]

I service CNC machinery. It is very unlikely that your problem is the wear in the ballscrew. More likely that it is the belt or pulleys, if it is belt-driven. I have had trouble in the past with Milltronics pulleys rusting, causing the belt to decay rapidly. Also suspect is the ways, perhaps they have come a bit loose? And have you checked the endplay between the actual spindle shaft and the spincle housing?

[msomerville]

I have checked the bearing blocks and everything else I could think of, but it didn't even occur to me to check the ways. I did notice that when I was checking it the ballscrew turned ever so slightly so I think that is where some of the movement is coming from. I will check the ways and see if they are tight.

[M_Controls]

FYI: To check the ways tightness, set up an indicator base on the flat area of the ways, and indicator on the housing near where it rides on the ways. Then (carefully) pry upward near the spindle area (or on the spindle itself, but not hard enough to hurt it) and see what the indicator reads. Duplicate for all four corners of the ways; and you may desire to try something similar for side-to-side movement (separately) as well. But if you had side-to-side movement you most likely would have some Y axis errors in your work, but some would be adjusted with your milling tool dia comp. Naturally, the top corners are the most likely to give you a bad reading if you do not have a counterweight, but in certain cases the bottom corners might read bad, telling you of another unlikely problem, such as a badly warped taper.

Problems I have seen in the past: Problems with way adjustment jam screws (or jam nuts) coming loose (not Milltronics). Some don't even have jam screws (nor jam nuts). Broken way tapers: It looks like everything is fine, but the taper inside is actually moving independantly of the adjustment end. In this case, the ways might check tight until the one time you move the axis in the right direction, and the way taper moves in the loose direction, then it checks loose! In the case of a broken way taper, when you remove the way wipers you can sometimes measure the distance of the way taper from the end, on the opposite end of the adjuster, and see that sometimes that distance changes when you reverse the axis movement direction. It does not always move when you reverse direction though; it usually moves more the drier the ways are. I've seen the ways lock up tight with this problem.

[msomerville]

Thanks for the info. I will do some more checking and post what I found out.

[neilw20]

On my X3 machine I had what appeared to be bad backlash in Z.
First found stepper coupling on screw driven by flats and grubscrew.
Grubscrew had been tightened, but not onto the bottom of the flat.
First loss fixed.
When the screw pushed up to move the head, the force increases, causing the Z to go DOWN before stiction in slide is over come. Same thing happened when going down. The screw pulled, the spindle nose went up a bit, before moving down.
Careful adjustment of the gibs, has decreased this error to 0.02mm
When I need an accurate Z for finish, I overshoot down by 0.02mm then move back up to allow things to settle. This works quite well. If you use backlash compensation, this causes all sorts of pauses, as each axis corrects itself on a change of direction, as the slowest axis is stopped while it changes direction.
Backlash compensation cure is worse than the disease.
Bottom line: Adjust all slideways to minimize problems.
I have experienced similar feature on 10 a ton machine.