Repacking ballscrews/ballnuts

I have an NSK ground ballscrew that somebody (not me, honestly) was clever enough to run the ballnut past the end of the threads and allow
some of the balls to escape. The ballscrew is a ground screw, and nice condition, but a bit dirty, so I am thinking about taking the nut completely off, cleaning everything out with a light solvent (WD40?) and then getting some new balls and repacking the nut. I've never repacked a ballscrew before. Questions:
-What type of balls are typically used (I'm sure it probably depends on the screw material, but I forgot to bring the model # with me) on NSK precision (C5) ground ballscrews? Chrome plated? Stainless? Carbon steel? ...
-What is the technique for actually loading the balls?
-How should I clean the ballnut out?
-What type of lubricant should I use during/after packing the nut?
Thanks! -Ralph
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Balls: not likely to be chrome plated, most probably 52100 carbon steel but could be stainless. Check what you have: put one in some water overnight and it will rust stain. Original will have been matched for size, I doubt if you'll ever get the precision of a factory build again. Lubricant, well tell us more about the application and environment.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 22 Jun 2006 14:13:26 -0700, snipped-for-privacy@gmail.com wrote:

Why not just use the original balls? IIRC The threads on the ends of the screws are not hard enough to have damaged them ... Otherwise, if you have lost some of the balls (how?) a bearing supply house could probably measure one & replace ....
--
Cliff

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Very doable if you keep it clean.
Take all the balls out and mic them for size. Keep a count on them too. They usually come in standard sizes, and over/under sizes. Put aside all the old ones and buy all new ones from most any bearing supplier. They are not usually stainless, but just relatively hard steel, and about $3-$25 for 100. Most screws we had done take about 100-200 balls.
Clean the screw well. Pull any seals and soak the nut in kerosene overnight, then alcohol for a few hours. Shake and stir occasionally to get the little bits and old oil/grease out.
For assembly use grease to hold the balls in place as you wind the nut and fill return tubes, winding the screw a bit as you go. If you are using an oil-lubed nut/screw, then use any light grease that cuts well with Vactra 2 oil, as you want it to wash out later. If it is a grease packed nut/screw use red axle or whatever other grease you are told to lube the nut and linear ways of the machine with. It is a relatively low speed bearing so fancy grease is not usually necessary.
The big trick here is when you start winding it on, to make sure you understand how to wind it on and off with most of the balls in place, and realize that there are two separate ball tracks opposing each other for preload. Use the grease to hold them in their tracks as you wind it on. It's easy to tell: One way works, and all the balls stay in the tracks and feed into the return tubes as you wind. The other way, and they all fall out on the floor. Pack the return tubes as you finish.
In reality you'll do it a few times (on and off, "dang-it", on and off, "dang-it again", and on...) then you'll understand how it works.
We have reloaded many screws for various reasons, including the one you mentioned, where some bone-head unwound it off the end of the screw, and then tried to put it back on with not enough balls.
If the screw and nut bearing surfaces are in good shape, they usually turn out beautifully smooth again. (However, if the tracks are rough and pitted, just throw it away.)
--
Steven Haerr
CNCTRADER CORPORATION
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Thu, 22 Jun 2006 15:36:34 -0700, "Steven Haerr - CNCTrader Corp"

LOL .... Sounds like you've done this.
--
Cliff

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Cliff,
You still working on that sample 2 axis profile? How long does it take an egg-spurt like you to figure out a simple profile and post it?
Been asking for what near a month now for you to post this profile so you can prove your right and I am wrong. You have managed to post over 1,000 newsgroup responses and most are off topic. How about you focus and post a simple profile to prove your point. Afraid you can't stand up to pier review?
Have you ever programmed a cnc lathe? You have stated you programmed thousands of controllers so how long can one simple profile take you?
Tom
Cliff wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

http://www.cargolaw.com/images/disaster2001.pier5.GIF
http://www.cargolaw.com/images/disaster2001.pier2.GIF
--
Cliff


Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Cliff,
That's the best you can do, you really are an inept troll.
You can post 1,000 items off topic but you can't answer direct questions or post a simple profile? You represent yourself as an expert, a professional programmer that programs multi-axis cnc machines but you can't produce a simple 2 axis lathe profile to prove your point? Why can't you?
You can post dribble but you can't/won't post a sample of your work to prove your point? Coward, post this profile and stand up to peer review.
Damn spell checker, Tom
Cliff wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Pear? BB has the apples & oranges problems <G>.
--
Cliff

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Cliff,
Apples, oranges or pears, stop messing around and post that sample of your work to prove your point.
How do ya like them apples? Tom
Cliff wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

It's "spelling checker" <G>. I don't use one .... sometimes it shows ...
--
Cliff

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Cliff,
Good one, shows your age, working on that sample profile yet?
Tom
Cliff wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
I am reminded of the source of the term "shagging", after abrading my knees and elbows looking for those darned little thingies in the carpet pile.... /mark
Steven Haerr - CNCTrader Corp wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Assuming its the type that has a pair of "keys" holding the nut halves together.......
Nsk ships these with an extra ball--and it falls out as soon as you separate the 2 nut sections.
Mike it...( but actually the ball dia is not especially critical with this design anyways...
Re-assemble with a good grease--I personally like to use the mobil shc-32
Leave the spacers out to check your backlash--proper spacer thickness will be exactly one pitch minus any measureable backlash...go a tad bit skinnier than that if you want to add some preload.
--
SVL



Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
So, I mic'ed out 4 of the balls, and it appears that there might be 2 different size classes. I made the mistake of measuring in inches instead of mm, as an NSK screw is almost certainly metric.
Anyhow, here's what I got:
3.971mm (0.15635 inches) 3.969mm (0.15625 inches) 3.933mm (0.15485 inches) 3.936mm (0.15495 inches)
So, my guess is that there are smaller balls spaced in between the larger balls to prevent jamming.
Thanks! -Ralph
snipped-for-privacy@gmail.com wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Ralph, Yep the alternating small balls keeps them rolling instead of skidding under load.
Gary H. Lucas
wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 22 Jun 2006 16:51:40 -0700, snipped-for-privacy@gmail.com wrote:

Get some known certified standards. Mark the balls with a magic marker. Now measure them at random about 10 times each ... AND your known standard. Have someone else write down your measurements AND look at the items measured when recording -- YOU DO NOT LOOK AT WHICH ONES THEY ARE AS YOU MEASURE THEM TIME AFTER TIME. Find the SD of your measurement using the standards measurements from what was recorded for them. Do the same for your actual balls.
You may be a bit shocked ....
BTW, The real SD is probably the squareroot of (SD[balls]^2 - SD[standard]^2).
Those balls are probably within millionths of being the same .... in reality ...
--
Cliff

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Cliff wrote:

Ummm... I don't think so. I will grant that I may mismeasure 0.0001" or so, but I seriously doubt I am off by 0.001" or more very often.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 23 Jun 2006 09:55:27 -0700, snipped-for-privacy@gmail.com wrote:

Run the tests I suggested. You might be shocked. So might be others that try similar. I know that I once was (not me doing the measuring). And that excludes thermal growth effects .... they may exist as well.
They may be right about the two sizes of balls though ... I don't know (or recall) ...
--
Cliff


Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

At that point, they might just as well to be several thou undersize then....
Regardless, they'll still be in contact with one another along a long 'chain'--AND at a third contact point someplace on the ballscrew or nut groove.
--
SVL






Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.