545,670 active members*
2,468 visitors online*
Register for free
Login
Page 4 of 4 234
Results 61 to 73 of 73
  1. #61
    Gold Member
    Join Date
    Sep 2006
    Posts
    6462

    Re: Ball screw question

    Hi no matter what the cost of a hobby machine, if it has backlash due to cheap single ball nut screws from day one it will only get worse, and as backlash is not tolerable in any form whatsoever.......who ever heard of programming out .005" of backlash due to a worn screw.

    That a single ball screw will wear is the same as using a radial ball race and subjecting it to axial loads instead of using angular contact bearings designed specifically for axial loading.

    I assume that the tracks of a high precision ground ball screw are super hard and so can better resist the end loads generated when the screw is driving against a load, so as the difference is then only the hardness of the track ways, making a cheap rolled ball screw and having the ball track ways super hard too without grinding them should prolong the life.....theoretically......the difference in quality is the rolling process that injects a tolerance factor for pitch accuracy over a distance.

    A single nut ball screw will eventually have backlash no matter how expensive it started out at.......we are really using the same principle of the deep groove radial ball race to resist axial loading and as soon as a smidgeon of wear occurs, as Thomas showed in his drawing on post # 58, you get backlash.

    Having a "bit of backlash" is like saying "I'm a bit terminally sick, and that cannot be tolerated in any form ever..

    Applying the same design as a deep groove ball race, you can to some degree use them as replacements for angular contact bearings provided you have two that have some means to press them apart to simulate angular contact bearing characteristics........they won't be good for radial forces after that, but for a while they will track accurately and act as radial/angular contact bearings lightly loaded.

    That being said, the same applies to the ball screw........have two ball nuts with a means to press them against one another and you have a means to eliminate backlash........it just depends on the hardness of the tracks during manufacture for longevity of duty cycle.

    If the tracks of both rolled and ground screws are of equal hardness then you will have equal wear rates and so can use double ball nuts to eliminate backlash.

    When wear occurs it's not for the full length of the screw, but probably more in the centre or wherever the most work is done, so you have a screw, rolled or ground that will have backlash in the middle and none at the ends etc.

    Having two ball nuts with a resilient coupling will take care of the tight and loose spots without jamming dead tight if they were merely backed against one another with hard screw pressure.

    For hobby use, ground screws are like "Gilding the Lilly" without realising the full potential of the grinding process, but rolled screws with simple resilient anti backlash measures applied can make a mongrel into a thoroughbred without breaking the bank or spending the kid's inheritance.

    BTW, if undersized balls were fitted to the rolled nuts in the beginning you would have a ball screw with undue backlash from the beginning, but progressively occurring backlash over time is an indication of either wear in the trackways (most probable) both for the screw and the nut or the balls.
    Ian.

  2. #62
    Banned
    Join Date
    Jan 2015
    Posts
    12

    Re: Ball screw question

    How many hours do your double ballnut ballscrews have on them handlewanker? What is the lead error of them and do you compensate for that in your software? did you map your ballscrew with your software?

  3. #63
    Registered
    Join Date
    Feb 2009
    Posts
    179

    Re: Ball screw question

    [QUOTE=handlewanker;1630692]Hi no matter what the cost of a hobby machine, if it has backlash due to cheap single ball nut screws from day one it will only get worse, and as backlash is not tolerable in any form whatsoever.......who ever heard of programming out .005" of backlash due to a worn screw.



    Having a "bit of backlash" is like saying "I'm a bit terminally sick, and that cannot be tolerated in any form ever..

    I congratulate you on your dedication to zero lash, but if we all insisted on that kind of accuracy, there would simply be no such thing as a hobby machine and companies like NOOK and THOMSON would be out of business. Sort of like saying " Anyone who cannot afford a Rolls Royce should just walk to work" As far programming out backlash, for straight linear moves it works fine. For circular interpolation it's trickier at the quadrants, but I have been able to keep my circles round and to within 0.001". If you want to get to 0.0001", then you will have to buy that Rolls Royce machine.
    Getting back to the original poster's question, I think most of these companies producing low cost rolled screws probably use undersized balls to be sure that they will fit within the manufacturing tolerances of the screws and nuts. For an owner with a little time to spare, trying larger balls may solve the problem, but if you feel any tightness or binding you will need to try a smaller ball until you find the proper size. It's the kind of fiddling that an owner can afford to do, but would be prohibitive for a manufacturer. The double nut modification can solve the problem as well, but there are usually issues of space, mounting and loss of travel to contend with.

  4. #64
    Gold Member
    Join Date
    Sep 2006
    Posts
    6462

    Re: Ball screw question

    Hi, I quite agree with the notion that you only get what you pay for, but with a leetle bit of cleverness perhaps the cheapie rolled ball screw can produce better results than just mounting it on the machine as is and hoping it will perform.

    A lot depends on the original mountings that held the nut as this will dictate the travel from end to end which will be approx. 60mm or 70mm less with a second ball nut on the screw.

    For the previous poster....Vingchoochoo.....I don't have a machine YET with any ball screws on to make any comparison tests and will not, as that would mean I'm just a tester of ball screws whereas I'm building a bridge mill/router to do CNC machining and to that end will attempt to be economical with the hardware to some degree and see what comes out in the wash.

    I do have a 1605 rolled ball screw and two nuts to do a test on that I bought specifically for that purpose, but am now going to utilise it as part of the build as I believe you should put your money where your mouth is.

    There's nothing new about fitting double nuts to Acme threads to take up the backlash on manual mills..... after a fashion.....as the double acme compensating nuts only take up some of the slack in a screw/nut combo and you still have a degree of backlash, but manual mills are not expected to be backlash free....ever.

    My manual Bridgeport had .25mm backlash (on the dial) and I used it like that for several years, a situation that had it been CNC orientated would never have worked, so fitting double ball nuts to a rolled screw can have benefits if it's possible to fit them.

    The problem with retrofitting double ball nuts is that the screws become too short or there is no room to extend the screw mountings to fit longer screws.

    In my opinion, double ball nuts, either bought in as such or fitted to be such, are by comparison to drum brakes as opposed to disk brakes all round on a car.

    I'm in the fortunate position with the build I'm currently involved in, that from the outset the double ball nut on all axis is a possibility that won't break the bank........why eat sandwiches when with a bit of salad dressing you can have a banquet.

    I doubt wether anyone with hobby machines as the main work horse would anticipate producing a circle by CNC means that had a tolerance of +- a few thou at best.....ball screw tolerance being only one link in the chain.
    Ian.

  5. #65
    Registered
    Join Date
    Jan 2008
    Posts
    458

    Re: Ball screw question

    Quote Originally Posted by handlewanker View Post

    For the previous poster....Vingchoochoo.....I don't have a machine YET with any ball screws on to make any comparison tests and will not, as that would mean I'm just a tester of ball screws whereas I'm building a bridge mill/router to do CNC machining and to that end will attempt to be economical with the hardware to some degree and see what comes out in the wash.

    I do have a 1605 rolled ball screw and two nuts
    HILARIOUS- I am laughing at myself as well as his post. All along I thought he was some expert in machines and screws, and it turns out he doesn't have a machine at all, just a ball screw he bought off E-Bay. All this the time he's telling everyone to buy only the best, while he is " being economical with his hardware" .

  6. #66
    Banned
    Join Date
    Jan 2015
    Posts
    12

    Re: Ball screw question

    handlewanker, i'm not understanding your poor attitude towards other members choices since you do not have your double ballnut ballscrews in operation on a machine. you have nothing factual to relate yet i mean as in personal usage. double ballnut ballscrews have been shown many times on the forums by other hobby members in many different forms and yes can make for zero backlash but it does not compensate for lead errors or other sources of backlash in the system especially as wear increases over time. you failed to answer my question about using software but i see you would not have progressed to that point yet to be able to make an informed response. several means are available to increase a machines accuracy using software and yes even the rolls royce machines do this too. with these tools available to the hobbyist it should not be hard to accept that using a single nut with very minimal backlash is certainly not something worthy of your disparaging comments.

  7. #67
    Gold Member
    Join Date
    Sep 2006
    Posts
    6462

    Re: Ball screw question

    Whatever........I stand by my "remarks".....if whatever you're using pleases you, be happy.

    My last post on this thread.
    Ian.

  8. #68
    Registered
    Join Date
    Feb 2008
    Posts
    153

    Re: Ball screw question

    I bought some screws from Chai of linearmotionbearings2008, two 1605 screws with ends machined, "anti-backlash" nuts, and end support bearings/blocks. They were cheap enough that it won't break me if they end up in the garbage, and most reviews are good. I measure about .0015 backlash with a quick bench setup. Here are a few pictures showing some quality issues. The screw has an obvious defect along the entire length that was there before the threads were rolled. You can see the defect in the ball groove and on the ridge, and it's the entire length of the screw. It also has a good bit of wobble. The support block mounting holes are all out of place. I was surprised to see NSK Japan bearings in the floating end, unfortunately the angular contact bearings in the fixed end are Chinese.

    Attachment 265306

    Attachment 265308

  9. #69
    Registered
    Join Date
    Oct 2014
    Posts
    77

    Re: Ball screw question

    Well so everyone knows I got the result I was expecting. First many people here assumed I had never worked on this before and every time I tried to indicate what exactly I was doing it went back to you will wreck it or you have uneven wear, sorry but that wasn't the case. All I wanted from my post was what size the originals were. Moving in .0001 increments I now have silk smooth action across the entire screw and have reduced my backlash to under .001, better than the .003 I was seeing. Note that the sizes of a 2014 shopmaster varied between the z axis .144 and x .148 as measured on start average . I got a great deal on the balls and sold the ones not used. So thanks to the constructive compliments and for the others, I got to where I wanted

  10. #70
    Registered
    Join Date
    Oct 2014
    Posts
    77

    Re: Ball screw question

    Also for the wreck it catagory, don't install in machine before you completely inspect movements across all of the screw, with zero force. The second you force something even a few oz then the damage is done.

  11. #71
    Registered
    Join Date
    Mar 2012
    Posts
    90

    Re: Ball screw question

    That was a great result.

    Did you use all the same size balls per axis and correct size was different for each one? How much of a change was there from the original ball size? Were the original balls all over the place as far as diameter for each axis? How does one check for correct operation without doing any damage. Do you just turn by hand and if you feel binding stop and repeat using smaller balls?

  12. #72
    Registered
    Join Date
    Oct 2014
    Posts
    77

    Re: Ball screw question

    Hello,
    On the z axis I started with .144 plus or minus .0002 and was able to go to .1456 with smooth action prior to install, by placing screw vertical the weight from the nut allowed it to turn on own down the length of the screw. For y axis started at .148 plus or minus .0004 and went to .150 and same test, placing screw vertical and again nut turns on its own by its own weight. Still waiting to do x axis till I can get it out with machine tilted.

  13. #73
    Registered
    Join Date
    Oct 2014
    Posts
    77

    Re: Ball screw question

    If anyone was to do this I recommend not using my values specifically as you need to have multiple size balls to start because it was a trial and error moving up sizes slowly and taking time to ensure no binding under no presure. As then like mentioned damage can be done if forced.

  14. #74
    Registered
    Join Date
    Jan 2009
    Posts
    8

    Re: Ball screw question

    Does anyone have drawings or schematics of the ends of the acme lead screws (x, y, z axes) which came standard on the Shopmaster Eldorado Bridgemill 3-in-1 as manufactured circa 2004. I am interested to upgrade this unit to ball screws; unfortunately, JT at Shopmaster no longer offers these as upgrades. Although there are many alternatives available for online purchase, the ends would need to be machined so as match the factory-installed screws. I am trying to avoid disassembling my machine to make these drawings before ordering and assume others have faced this problem in the past.

    Thanks for any assistance!

Page 4 of 4 234

Similar Threads

  1. Ball screw question.
    By Nth in forum Uncategorised MetalWorking Machines
    Replies: 6
    Last Post: 08-25-2010, 07:48 PM
  2. Ball screw question from a new guy!
    By ben_kapitan in forum Community Club House
    Replies: 2
    Last Post: 04-19-2009, 04:19 AM
  3. Ball screw Question
    By rodzilla in forum Linear and Rotary Motion
    Replies: 3
    Last Post: 03-02-2007, 11:16 PM
  4. acme screw to ball screw question
    By Billw in forum DIY CNC Router Table Machines
    Replies: 9
    Last Post: 07-18-2005, 06:10 AM
  5. Ball Screw Question
    By jjwl89 in forum DIY CNC Router Table Machines
    Replies: 13
    Last Post: 07-20-2004, 01:12 PM

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •