. .
Page 1 of 3 123 LastLast
  1. Hello All,
    So I started building a 1.2 c 2.4m bed CNC a long time ago. It's a bit unusual in a few ways, the gantry being a frankensitne mash-up of a scrapped stairlift extrusion, steel pipes, epoxy concrete, and home cast aluminium end plates. The machine build has not proceeded the way I see most builds happening (ie a sensible CAD design that gets tweaked BEFORE construction). Thats not to say I had no design, just it's morphed quite a bit during the build. I guess you could say I am quite a thick-headed so and so, that needs hands-on experience to learn anything. Ok that's enough background for now - the dilemma, and reason for the post...

    The x has two 2.4m long, 5mm pitch,32mm dia. ballscrews, which I have had for at least a year before finally getting round to fitting. They are from China (linearmotionbearings2008 on eBay) and are both a bit bent (approx 1mm off true bow). As I was going the rotating ballnut rout, I thought it wouldn't be such a problem, as the sideways force required to straighten them is not much over such a length (indeed the sag by gravity is noticeable), and I would be able to tension them a lot as the screws themselves are stationary and don't need to worry about over tight angular contact bearings.

    Here's one screw rigged up to the rotating ballnut assembly I have cobbled up:



    Things to note: I am testing this with just one screw - I haven't yet mounted the slave ballscrew on the other side... There is very little (no detectable by eye) racking in the gantry, even though I'm driving it from one side only, which seems promising rigidity wise.

    I think the 5mm pitch was a mistake! I thought that with the 400W servo (3,000 max RPM) and no gear reduction, I would get reasonable travel speeds (3000 x 5 = 9000mm/min). 9000mm/min sounded ok to me - keep in mind I'm a complete CNC newbe. What I didn't consider is the ballnut's limitations. From the word go the nut has felt a bit 'crunchy' - or at least not smooth, and also sounds a bit rattly - despite my greasing... I didn’t get them on video but, there have been a few ‘jams’ that have stalled the servo. When I turn by hand I can verify that yes, it is just the ballnut ‘catching’, and with a bit of ‘back and forthing’ I can sort of feel it past the jam. The catches have happened at relatively low speed. When going faster I guess the servo just powers through them.
    So options as I see them:

    1. Ditch the cheap ballscrews, bite the bullet and buy some good ones of 10 or 15mm pitch (suggestions for what would be ‘good’?).
    2. Continue on, mount the other x ballscrew. My feeling is this might hide the problem for a bit... In fact, writing it out here makes this seem a very poor choice?
    3. Try some ballnut fixes? push more grease in? attempt disassembly and look for swarf or other potential jamming culprits (how possible)?
    4. Something that I haven’t thought of that may seem obvious to you lot??

    Ok thanks a lot for reading, please share your thoughts :)
    Last edited by Bongo; 27-05-2021 at 09:06 AM.

  2. #2
    I would start with removing and cleaning all the nuts and linear bearings. It takes time and patience but worth a try.

  3. The Following User Says Thank You to A_Camera For This Useful Post:


  4. Hey A_Camera, thanks for this, seems a sensible start. Any idea how likely I am to be able to get the balls back in the nut or how difficult it is? Would love to hear if you (or anyone out there) has done it? Estimated time investment?
    Thanks again, B.

  5. #4
    Muzzer's Avatar
    Lives in Lytham St. Annes, United Kingdom. Last Activity: 15 Hours Ago Has been a member for 6-7 years. Has a total post count of 417. Received thanks 61 times, giving thanks to others 10 times.
    I had to do this on one of my ballscrews after The Stupid Fat Bloke managed to get into my workshop when I wasn't looking and rotated the ballscrew too far Wasn't quite as bad as I'd feared in the end.

    Best avoided if you can but there are videos out there that show how it's done. There are various methods suggested but this worked for me, as you can screw the ballscrew in as you fit the balls. You could probably do it in 20 mins or so. Or 2 minutes if you are a Chinese factory worker.



    Others suggest holding the balls in place with grease or plastic foam and then screwing the ballscrew in place afterwards, which sounds like a recipe for frustration and disaster unless you like that sort of thing.

    ######

    Incidentally, at 3000rpm, those balls would be flying through the ballnuts. Try to imagine them being diverted by the plastic guides and ask how long those guides will last. There are max rpm specs for ballscrews and I don't imagine those ones would be very high. Most of the manufacturers publish application notes. Looking at p33 of this one, I'd be surprised if even a top end 32mm ballscrew would be happy at 3000rpm. https://www.hiwin.com/pdf/ballscrews.pdf

    A source for better quality parts might be one of the machine breakers. I've bought ballscrews from FA Parts in Korea. https://www.ebay.co.uk/str/industrialpartsshop They stock new old stock (NOS) and used parts from dismantled machines. You might find a decent quality Japanese, Korean etc ballscrew, possibly ground rather than rolled, possibly never used for the same price as a "new" Chinese rolled ballscrew. I got ground Korean 2005, 1090mm long for £109 and a 1205, 500mm long for £51 inc carriage, both delivered within a week - no comparison with the Chinesium ones I got for my Bridgeport conversion.
    Last edited by Muzzer; 12-04-2021 at 06:03 PM.

  6. The Following 2 Users Say Thank You to Muzzer For This Useful Post:


  7. #5
    Quote Originally Posted by Bongo View Post
    Hey A_Camera, thanks for this, seems a sensible start. Any idea how likely I am to be able to get the balls back in the nut or how difficult it is? Would love to hear if you (or anyone out there) has done it? Estimated time investment?
    Thanks again, B.
    I don't know how long time it takes for you, it depends on many things, but no doubt that it can be done. I did clean sixteen linear bearings, and it took me a while, but the results were well worth the effort. I am sure that ball bearings are not more different, especially the larger ones.

  8. The Following User Says Thank You to A_Camera For This Useful Post:


  9. #6
    A ballscrew I had which sounded like that had very slightly undersized balls in it. If you can determine how well they fit and can find a slightly larger diameter set of replacements that could help. However I’ve searched and not had any luck other than in large quantities.
    But given that they very occasionally get stuck suggests a good clean would be a good start. I did post a YT video in my drill press series on the grease method but I do like Muzzer’s quick method above !

    https://m.youtube.com/watch?v=QbvH5dnUF1Q&t=3s
    Starts about 26:00
    Last edited by routercnc; 12-04-2021 at 10:07 PM.
    Building a CNC machine to make a better one since 2010 . . .
    MK1 (1st photo), MK2, MK3, MK4

  10. The Following User Says Thank You to routercnc For This Useful Post:


  11. Thanks so much for sharing your thoughts guys, appreciate that. @Muzzer: great link and info thanks. I will try pulling off the nut and see how I go, though it sounds like to get good rapids out of the machine I will ultimately want to change the pitch to something like 15mm and get a smaller toothed pulley for the servo...
    Good call on the Place in Korea BTW, I actually got my servos, drivers, and the ballscrews for the y and z axis from that e-bay shop, and have been well pleased. I have looked a few times, and if they had anything in the right length, dia, pitch, I'd be going that route... I will keep checking and might shoot them a message. Though I don't relish the extra cost, I feel the hours I've put into the machine build would justify it...

    @Routercnc Thanks for this - an interesting watch... I'll see what I get and how it goes tomorrow. I must admit, my first thought was that the balls might be a bit undersized, or possibly just unevenly sized. I can't detect any backlash (by hand).
    Thanks again guys, I'll report back with the findings.

  12. #8
    From the little I know about ballscrews it would take a lot to persuade me to dismantle them.I think I would first pump quite a bit of grease through them to flush out anything that shouldn't be there and I might do it more than once and in slightly different positions.If it didn't improve things I think I would watch several youtube videos before even contemplating dismantling and I would take careful note of any warnings.I would only do the job on a large and clear bench with some kind of rail surrounding it for restraining escaped components.The sequence of balls may be important as I have read accounts of sizes being intentionally varied along the assembly and the variations are tiny,but necessary so that you don't get a situation akin to a line of balls trying to move but effectively having their neighbours walking on their heels and impeding free motion.

    It would be much easier to clean the threads themselves very thoroughly prior to any form of dismantling and this might pick up any particles that have been attracted to the protective grease that would have been applied for initial assembly and transportation.If you are lucky it might be enough to help but as you have discovered a 5mm pitch is a bit fine for the intended use.

  13. The Following User Says Thank You to routerdriver For This Useful Post:


  14. #9
    Don't waste your time with greasing as it will make no difference and at best only mask the problem, at worst it will mix with what's grinding now and turn the grease into a very nice grinding paste.!! . . Stripping and cleaning ball-screws is a piece of cake and doesn't take long. BUT you must be set up and prepared.

    It's a must that you take them apart in a container with reasonably high sides and if the container is steel place something like a kitchen roll in the bottom to stop the balls from bouncing because they have a habit of shooting off like bullets if dropped from any height. DONT put a magnet near them because you'll just magnetize them and make them attract every speck of metal dusty within a 4mile radius...

    Regards the rotating nut then I've done a few now and you need at least 25mm pitch and then put a ratio on the ball-nut of 3:1 or 2:1 to get the nut rotation speed down. If using 32mm screws then I'd go with 32mm pitch as it's a common pitch size.

    I wouldn't continue with those 5mm pitch screws as you will destroy the ball nut within weeks if you try to spin at anything like 3000rpm, even 2000rpm will eat the plastic tubes away in no time which then causes the balls to jam and stick turning them into 50p pieces as they slide rather than rotate. Cut your losses and sell them on because you will destroy them.
    -use common sense, if you lack it, there is no software to help that.

    Email: [email protected]

    Web site: www.jazzcnc.co.uk

  15. The Following User Says Thank You to JAZZCNC For This Useful Post:


  16. Hey JazzCNC. That sounds like very good advice. I plunged on and took the nut off!

    Click image for larger version. 

Name:	VideoCapture_20210413-204701.jpg 
Views:	187 
Size:	218.5 KB 
ID:	29808

    Click image for larger version. 

Name:	20210413_204826.jpg 
Views:	168 
Size:	184.8 KB 
ID:	29807

    I actually managed not to lose any ball bearings. Good call on the non-bouncy surface ;)
    Didn't have a lot of time today but will make a small vid of the process - once I have successfully put it back together, that is!
    Before Taking it off, I could feel with my hand that there was a little side to side play in the nut, though I couldn't detect any axial play that would obviously translate straight into backlash. Anyway, it made me suspect the balls are slightly undersized.

    Agree, I probably am better off selling them and going to a higher pitch. What is your experience with cheap Chinese ones vs the branded,JazzCNC? 32mm pitch seems a lot, but I see that it does seem popular to match pitch to dia...

    As a side note, the ball diverters are actually metal.
    Thanks again for all your thoughts, keep em coming.

Page 1 of 3 123 LastLast

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. Rotating Ball nut
    By drumsticksplinter in forum Lead Screws, Nuts & Supports
    Replies: 30
    Last Post: 12-05-2020, 08:01 PM
  2. BUILD LOG: Rotating Nozzle
    By Peter3102 in forum DIY Router Build Logs
    Replies: 0
    Last Post: 25-11-2019, 11:09 AM
  3. WANTED: Rotating Nut Fixture x2
    By PaulP in forum Items Wanted
    Replies: 13
    Last Post: 02-01-2018, 10:50 PM
  4. Rotating ball nut
    By D-man in forum Projects, Jobs & Requests
    Replies: 0
    Last Post: 21-03-2015, 08:39 AM
  5. BUILD LOG: New build.Stell frame cnc 1500X1000.Ball screw and gearing calculations
    By ba99297 in forum DIY Router Build Logs
    Replies: 12
    Last Post: 03-09-2013, 10:37 PM

Tags for this Thread

Bookmarks

Posting Permissions

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