repairing ballscrew?

Hey Rick,

I think you best make sure what part of this assembly is worn. If it is, as you say others suggest (and I agree with), that it is the bearings that go first, then that reference IS NOT to the ball-nut. It is the bearings that the ball screw shaft ends turn in, one pair of which will be trapped thrust bearings, and the other end a looser fit lengthwise to allow some clearance for temperature changes. Not familiar at all with the Shizouka brand though.

Take care.

Brian Laws>I think the X axis ballscrew on my Shizuoka B-3V cnc mill is worn, or at least

Suggest check / replace your thrust bearings first, make sure your couplings are tight,etc

For ballscrew info google on :

< alt.machines.cnc > -- there was a quite recent thread covering them.

Hey thanks! I read your thread on a.m.c. about how you cured 0.002" slop by replacing your thrust bearings. This *could* be my problem, since perhaps it would explain the thud noise on part of the revolution (if the thrust bearing had a flaw in it). The ballscrew noise and backlash seem the same over the length of the screw. That would tend to point to the thrust bearing or the ballnut. Here is hoping it is the thrust bearing...now to figure out how to get to the thrust bearings and how to check them for wear.

Rick

"rick" wrote in news:2zbad.2079$Mh7.336@trnddc04:

Rick, Normally, 0.0015" of backlash is not an issue, provided you have compensated for it in the control. There should be a backlash parameter in the control. On 99% of the controllers I've dealt with, you can compensate for up to 0.25 mm/0.0098" of lash. As for the noise, I would not expect this from a screw with only 0.0015" of lash. Either you have a lubrication problem to the nut and have ruined the bearings, or you have a thrust/radial bearing issue on the screw ends. If there is no brinelling of the screw, chances are the bearings in the nut are just fine. I would definately check the screw mounting bearings though.

Put a test indicator on the free end of the screw, if it jacks in and out in relation to the housing or table when you repeatedly reverse the screw then you got slop in your thrust bearings.

Thanks Andy. The screw shaft looks perfect, and althought I can't (without taking the 400lb+ table off) check for lubrication, the whole thing drips constantly, so something must be getting lube from the autoluber. I am now thinking (with some relief) that it is indeed a problem in the mounting/thrust bearings. When I finish a job that I have mounted on the machine, I am going to take off the x axis motor and bearing plate and see if I can get to the thrust bearings without taking the table top off.

Rick

Doh! now why didn't I think of that!

Sigh...I will never be a true machinist...

Thanks.

Rick

Rick, the first thing you need to check is wear on the thrust bearings on the end of the ball screw for endplay. They often fail before the ballscrew and will give you endplay.

As far as screw wear, I normally replace a ballscrew if there is more than .0008 backlash on the screw itself, in production shop environments.

A quick way to check for ballscrew wear, is to put a magbase upside down on the bottom of your ways, and indicate to the side of a lead. Push and pull the table by hand and see how much movement you are getting. Thats your backlash. Then put the indicator on the knee and indicate the table. Push and pull. This will show how much endplay you have in the thrust bearings. Get a total reading, and subtract the screw backlash for bearing backlash

.0009 +/- screw .0015 +/- total = .0006 bearing endplay and so forth.

You may..may have to loosen up your gibs a bit to get a good movement. Lots of guys tighten them down to the point us old farts cant push and pull them by hand

Gunner

Confronting Liberals with the facts of reality is very much akin to clubbing baby seals. It gets boring after a while, but because Liberals are so stupid it is easy work." Steven M. Barry

Tempting for me to take credit, but taint my origonal ider neither......

A true machinist never stops asking questions and learning new things.....I got you to thinking about it all and took away much of the mystery here and so my job is pretty much done.....

=======

Dont take out the bearings unless you absolutely have to, usually these are angular contact sets and they tend to readily fall apart on dis-assembly......while you *can* carefully re-grease and re-assemble for limited use while waiting for new bearings and seals to arrive, dont expect them to last for very long afterwards.

List the bearing number, manufacturer and size info here and someone will probly be able to help in finding a low cost replacement source or perhaps even a compatible substitute off ebay--these are generally not cheap at retail pricing levels beings they are a sort of specially designed bearing.......

This is all assuming the problem actually turns out to be the bearings in the first place, of course....

Cheers, and good luck.

Gunner,

A short piece of 4x4 lumber or stout aluminum bar, along with a largish short handled hammer......

( and a helper to hold the 4x4 so YOUR fingers aint the ones get smashed )

Ayup. Sounds like a thrust bearing. You might be able to localize it by simply using a stethoscope on each end of the table over the bearings while jogging the table back and forth. Simple program.

Gunner

Confronting Liberals with the facts of reality is very much akin to clubbing baby seals. It gets boring after a while, but because Liberals are so stupid it is easy work." Steven M. Barry

Have you checked to make sure the ways are getting way lube. I used a Shizouka B-10V and had to manually squirt way lube on them daily or else it would vibrate. Mike H

Yep, the ways are getting plenty of lube. They drip it when you run the x axis way out on either end. I can also see the lube channels full in the top of bottom dovetail ways.

Rick

Hey, it is the bearings after all! Well, I assume it is.

I made the measurement you suggested, ie, a dial indicator on the far end of the ballscrew, with the indicator on a mag base attached to the table surface. I wrote a quick program to jog x back and forth 0.1". The dial indicator was showing the ballscrew moving in and out 0.0014" relative to the table. (This is a 0.0005" indicator, so I had to estimate). I remeasured the total backlash on x by indicating from the head to the table and it was just a shade under

0.0015". Thus my ballscrew probably has less than 0.0001" backlash?

I can't get to the thrust bearing yet to read the part #, since I would have to take the x axis motor and mount off and I need to use the mill today :-) I will post the part # when I am able.

Thanks for the advice!

Rick

Greetings Rick, Someday you may well be a true machinist. Machinist status will occur when someone you know, who is either a good machinist, or knows what a good machinist is, describes you to someone else and says something like: "Yeah, I know Rick, he's a good machinist". Of course, you will not be aware of the conversation, but you will be a "True Machinist" Eric

[ ... ]

That depends. If it is set up like the X-axis ballscrew in the BOSS-3 CNC Bridgeport which I have, the ball screw is firmly captive (and non-rotating) in one end of the table. The ball *nut* rotates where it is mounted to the side of the saddle -- in a pair of opposed thrust bearings.

So -- for that situation, the test would be two part:

1) Set up an indicator to measure the free play of the table when you push and pull on it. This is the total backlash. 2) Dig into the stuff under the table (perhaps removing the drive motor and its belt) until you can measure the end of the ball nut itself -- or the end of the bearings if you can gain access without undoing the nut which secures the ballnut in the bearings. Measure how much play you get while pulling and pushing on the table. (You might need someone else to read the dial while you push and pull.) This is the play in the bearings only. 3) Subtract this figure from the first. The result of that subtraction will be the backlash in the ball nut's fit on the ballscrew. And *that* should vary a bit, between the middle of table travel and the ends (and will represent the wear in the ball screw itself). If you *do* wind up rebuilding the ball nut, be careful that you don't put in balls enough oversized to cause it to be difficult to turn at the ends of the screw, or you will wear out the new balls, and the ball nut rather quickly.

Good Luck, DoN.

Hi, DoN.

All true, but the OP had mentioned his machine as being a Shizuoka BV-3 and my recollection was of this model being the more typical system, and having the motor mounted to the saddle on the right hand side.......

In fact, here is a photograph of what I would assume to be the same or very similar base machine as his :

A nice looking mill, at least for hobby and toolroom use........but alas, for my purposes full chip guarding is an absolute necessity these days.

Yep, thats my baby. It had the partial chip guarding like in that pic, but I have removed (and stored) even that as I like to be able to get myself around the part (spindle off!) when working on prototypes and quickie things. It has box ways for Y and Z, and dovetail ways for X, and all the ballscrews rotate, not the nuts.

Pretty good 7000lb hobby machine, especially for the $300 I paid for it. :-)

I have spent several times that in BT40 tooling, fixtures and cutters, even on ebay!

Rick