Welcome to Our Community

Some features disabled for guests. Register Today.

Bearings on the lead screw loose?

Discussion in 'CNC Mills/Routers' started by Paul Pridday, Aug 12, 2016.

  1. Paul Pridday

    Builder

    Joined:
    Jul 17, 2016
    Messages:
    67
    Likes Received:
    23
    Hi all. Finally made it home and am starting my C Beam Machine build. I am just getting to the Z axis lead screw install and am finding that the ID of the bearings have a fair amount of slop on the lead screw. In the build vid that Mark did it shows that they have to be sanded to get a good fit. Am I using the wrong bearings? They fit into the end plates and seem to be the only ones in the Z axis pack.

    Thanks,

    Paul
     
  2. UltiBots

    UltiBots Well-Known
    Builder

    Joined:
    May 8, 2014
    Messages:
    57
    Likes Received:
    76
    Paul - I too found my leadscrews had considerable play in the bearing. This has contributed to vibration noise. I am considering using PTFE tape to provide a better fit and reduce noise.

    Regards - Brad
    www.UltiBots.com
    ==================
    CB3030 CNC
    Dewalt 611
    Elaire 1/8" Collet
    1/8" Single Flute Carbide End Mill
    Workflow = Fusion 360 & Cam > Chilipeppr > Tinyg
    Feeds & Speeds For 6.5mm 6061 T6 Aluminum Plate
    - All operations = Conventional Milling
    - Cutting = 25"/min (635mm/min), Ramp = 10"/min (254mm/min), Plunge = 8"/min (203.2mm/min)
    - Adaptive (bearing recess) = .4mm multiple depths, 0.2 deg ramp
    - Contour cut (bearing finish) = 0.1mm stepover finish and repeat @ 254mm/min
    - Bore cut (screw holes) = 0.25mm pitch, 0.1mm finish pass and repeat
    - Contour cut (perimeter) = 0.05 deg ramp angle, 0.1mm finish and repeat @ 254mm/min
    - Pocket cut (not used) = 0.25mm rough & 0.2mm finish, 3mm max stepover
     
  3. Paul Pridday

    Builder

    Joined:
    Jul 17, 2016
    Messages:
    67
    Likes Received:
    23
    I'm wondering if I should try to source out some bearings with tighter tolerance before continuing with the build?? PTFE tape should work but would it not require refitting every now and then??
     
  4. Rick 2.0

    Rick 2.0 OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Dec 20, 2013
    Messages:
    2,892
    Likes Received:
    1,551
    More than likely it is the screw that is undersized as this is something that has been noted by several others lately. Searching for alternate bearings probably won't turn up anything any smaller. Bearings are generally dead on when it comes to size.

    As far as the PTFE tape, it really has no where to go and if fitted up tightly enough, it should be moving with the screw and the inside rim of the bearing as one.
     
  5. Paul Pridday

    Builder

    Joined:
    Jul 17, 2016
    Messages:
    67
    Likes Received:
    23
    I'll give it a try. I noticed though that it didn't seem too bad going through the lead screw block. Is that because the blocks now have the anti backlash in them?
     
  6. Rick 2.0

    Rick 2.0 OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Dec 20, 2013
    Messages:
    2,892
    Likes Received:
    1,551
    Yes, the anti-backlash piece helps the lead screw blocks grip the threads and not just the outer diameter like the bearings do.
     
  7. Paul Pridday

    Builder

    Joined:
    Jul 17, 2016
    Messages:
    67
    Likes Received:
    23
    Thanks guys. The PTFE tape seems to do the trick just fine. I'm a few days away from completion (honey do's getting in the way) and my Dewalt 611, 1/8" collet and 3D printed shim are almost here. I know this has been asked before but I am thinking that I should pick up a couple 10 packs of the cheap eBay mills. A pack of 2 flute flat end and a pack of 2 flute bullnose. I am going to do test runs on foam (styro or blue cor or whatever Menards has) and then futz around on some plywood scrap I have. I know the cheapo mills will not give a great finish or last like a nice Whiteside etc., but I'm sure I will kill it with stupidity well before it gets dull. Sound logical??

    Cheers,

    Paul
     
  8. Joe Santarsiero

    Joe Santarsiero OB addict
    Staff Member Moderator Builder

    Joined:
    Oct 30, 2014
    Messages:
    902
    Likes Received:
    196
    I am confident those tools will be satisfactory with your foam testing. :D
     
    Paul Pridday likes this.
  9. Paul Pridday

    Builder

    Joined:
    Jul 17, 2016
    Messages:
    67
    Likes Received:
    23
    Did you ever wish you could roll back the clock and 'unpost' something? Looking at the question now it does look quite idiotic. However, in my defense I did warn that I would have a lot of dumb questions.
    I'm going to blame it on being excited to finish the build and get cutting. You'd think at 53 a guy would be over the 'Christmas Morning' syndrome.

    Paul
     
    GrayUK likes this.
  10. Rick 2.0

    Rick 2.0 OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Dec 20, 2013
    Messages:
    2,892
    Likes Received:
    1,551
    I have to disagree. If you hadn't asked the world wouldn't know about the PTFE tape trick.
     
  11. Paul Pridday

    Builder

    Joined:
    Jul 17, 2016
    Messages:
    67
    Likes Received:
    23
    OK, so I am just at the point of installing the X/Z gantry on to my C Beam Machine and I am having problems with the top corner brackets. As you can see in the photos, the X axis beam doesn't appear to be long enough to mate up properly to the frame uprights. Looking at Marks build video, it looks like the X axis end plates are at the outside edge of the upright columns. Mine are about 1/4" shy. As you can see in the second photo, this doesn't allow the 10mm bolts and tee nuts to line up. Have I got something built wrong here?

    Thanks,

    Paul IMG_2663.JPG IMG_2664.JPG IMG_2663.JPG IMG_2664.JPG
     
  12. Rick 2.0

    Rick 2.0 OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Dec 20, 2013
    Messages:
    2,892
    Likes Received:
    1,551
    This is a problem that pops periodically due to fabrication tolerances. The side-to-side framing members in the base are too long. You'll need to shorten them to the same length as the C-beam extrusion on the X-axis.
     
    Mark Carew likes this.
  13. Paul Pridday

    Builder

    Joined:
    Jul 17, 2016
    Messages:
    67
    Likes Received:
    23
    Aaaaarrrrrghhhhh. Really?!? Will the extrusions cut with a mitre saw blade? I'll pick up a cheapy for my saw if it will. Glad my routers not here yet.

    I'm sure I'll feel much better when I win the new controller.
     
    Mark Carew likes this.
  14. Rick 2.0

    Rick 2.0 OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Dec 20, 2013
    Messages:
    2,892
    Likes Received:
    1,551
    Far as I know, pretty much everybody cuts their extrusions with a power miter saw. A non-ferrous cutting blade with lots of carbide teeth is best. I use a 100-tooth Oshlun, works well but wasn't cheap. Truly for the 2 cuts you need to make, save the money on a blade and just use a hacksaw.
     
    Paul Pridday likes this.
  15. Mike Piechowski

    Builder

    Joined:
    Jun 11, 2016
    Messages:
    33
    Likes Received:
    22
    I cut extrusion on my 10" miter saw all the time. Sometimes I even swap to the special triple-chip blade I have that is designed to cut aluminum.
     
  16. Paul Pridday

    Builder

    Joined:
    Jul 17, 2016
    Messages:
    67
    Likes Received:
    23
    I went ahead and cut them with a hacksaw. Had to do a little filing on them after. I'm sure it would be the same with using the miter saw. If I was doing a bunch I would definitely go miter saw. (Getting old and lazy)

    The front and rear frame pieces are now exactly the same length as the X axis C beam. Funny thing is I still had to suck the top of the columns in with a bar clamp to get everything to line up. Loosened the column bolts and tee nuts off and retightened after to relieve the tension. Will see how it runs on the initial startup.

    Paul
     
  17. Mark Carew

    Mark Carew OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Jun 28, 2013
    Messages:
    2,758
    Likes Received:
    2,438
    The PTFE tape is a great idea to keep any rattle down. In the past we have had to sand the leads down on the ends because they were too exact on the 8mm and so would not fit in the 8mm ID bearings. This did work but it was more of a pain to try and sand to a proper fit. They were also too tight for the Nut Blocks causing the lead to dig its way though the Nut Blocks.
    We have since retooled the leads to be smoother and they now have a slightly smaller diameter to fit the Nut Blocks and bearings more freely. Any side to side play from the lead in the bearing does not effect the systems linear precision as the gantry is locked inline and there is no load on the lead bearings. That being said there is still noise created there so this is something that are looking to correct moving forward.
    Since doing this retool the new leads work very nicely and in combination with the new Anti-Backlash Nut Blocks creates a great tight precise system.
    On the V-Slot being over sized this was actually done intentionally with about 3 to 5mm over the lengh. The reasoning was that one could cut a 1000mm piece of V-Slot in half to get two 500mm pieces. We see that with the bundles this can be a pain so we are looking at making all sizes of V-Slot exact sizes in the future.
     
  18. Paul Pridday

    Builder

    Joined:
    Jul 17, 2016
    Messages:
    67
    Likes Received:
    23
    Everything is up and running mechanically. :):):):):)

    I am having bit of trouble with the GRBL panel setting. I have to set jog at 10 to get a 10mm jog. I went through the settings page in the C Beam Controller Software folder from the build vid and copied them over. Is this possible due to a new Rev of GRBL Panel?

    Also I get my zero point set up on the front left corner, zero everything out in the software and by disconnecting and reconnecting. I then ran the sample pen drawing as Mark did in the vid, stopped it part way through and then hit the go to special position 1. In both cases it never went back to the zero point. I was also wondering if it is possible to move the Y axis C Beam back to attach at the center rail on the front frame rail. This would allow the Y axis to travel further back that distance. As it is my table will only go back far enough to allow the router max Y travel to go to 2" from the front of the table...

    Thanks for help with all the dumb questions... :confused:

    Interface.jpg Settings.jpg Zeropoint.JPG
     
  19. Mark Carew

    Mark Carew OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Jun 28, 2013
    Messages:
    2,758
    Likes Received:
    2,438
    Congratulations @Paul Pridday !!! :thumbsup:
    You could always move it back a bit, but I use this area for a safe spot for my front hold down clamps without fear of ever hitting them
     
    GrayUK and Paul Pridday like this.
  20. Paul Pridday

    Builder

    Joined:
    Jul 17, 2016
    Messages:
    67
    Likes Received:
    23
    That makes sense for sure to have area at the front for hold downs.

    Any idea what is going on with my GRBL settings and why it's not responding correctly?

    Thanks,

    Paul
     
  21. Mark Carew

    Mark Carew OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Jun 28, 2013
    Messages:
    2,758
    Likes Received:
    2,438
    I think when I change out the spoiler on my C-Beam Machine I will make the front and back Build table (first MDF board under the spoiler board) a little longer to allow for bestsy style toggle clamps to be directly mounted to it. :)
    Just with a quick look everything looks ok. But I would take a look at the grbl Panel page and see if GerritV (creator of grbl panel) can see what may be happening here. Be sure to read over a little as it may have been already been covered.
    Grbl Panel Control
     
  22. dean knipping

    Builder

    Joined:
    Mar 24, 2015
    Messages:
    84
    Likes Received:
    16
    also make sure you have the current version from github and not what's in the C-Beam folder as that version is old. I also saw that a new version just came out last week so give that one a shot
     
  23. Mark Carew

    Mark Carew OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Jun 28, 2013
    Messages:
    2,758
    Likes Received:
    2,438
  24. Paul Pridday

    Builder

    Joined:
    Jul 17, 2016
    Messages:
    67
    Likes Received:
    23
    Thanks guys. I'll give it a try when I get home. Not sure it is GRBL panel though as I get the same with Universal G Code Sender.
     
  25. Paul Pridday

    Builder

    Joined:
    Jul 17, 2016
    Messages:
    67
    Likes Received:
    23
    I'm back!! Looks like I am having the same issue with the new GRBL panel so I am thinking it's something to do with my settings with the CNC Shield or something. I do not have any microstepping enabled. Would that have anything to do with it? Ie. there are no jumpers at all on my CNC shield.



    Here is my first C Beam Machine vid showing what it's doing.

    Thanks.

    Paul
     
  26. Giarc

    Giarc OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Jan 24, 2015
    Messages:
    3,010
    Likes Received:
    1,677
    As long as your motors are 1.8 degree step motors (200 steps per rev) they should work the same. I think your steps/mm is wrong since you have no jumpers. You are using 200 steps per mm. With no jumper and 8 mm per screw revolution (I believe that is what the openbuilds screw does), you should have 25 steps/mm. The steps/mm can be calculated like this: steps per mm = (steps per revolution * microsteps)/mm per rev. For you, with no microstepping, that is X= (200 * 1) / 8. So, X = 25. This is probably why your travel distance is about 10 times too great. Not sure about the return to special position issue. I have never tried that. I am lazy. My special position is 0,0. I just send it there. For Z I just jog it up high enough to change the bits.:)

    PS. I am not known for my math skills, so please check my work. https://github.com/grbl/grbl/wiki/Configuring-Grbl-v0.9
     
    #26 Giarc, Aug 23, 2016
    Last edited: Aug 23, 2016
  27. Paul Pridday

    Builder

    Joined:
    Jul 17, 2016
    Messages:
    67
    Likes Received:
    23
    Thank you sir. That will be my next try.

    Cheers

    Paul
     
  28. Mike Piechowski

    Builder

    Joined:
    Jun 11, 2016
    Messages:
    33
    Likes Received:
    22
    That is most likely the issue. As I recall, the stock settings are for 1/8 microstepping. Any reason you're not microstepping?

    To really dial things in, run a pointer adjacent to a ruler, send a 100 mm move and measure how far you went, then adjust your steps per mm to dial it in.

    The motors sounded a bit "off" in your video. Might be that you're not microstepping, might just be the camera. Have you checked your current on your stepper drivers? If you're missing steps, it wouldn't return to the home position.

    When I used some inexpensive drivers that came with my clone CNC shield, things didn't work so well, they got hot, I let the magic smoke out of a few, and I could stall the motors easily, even at a claimed 2A. That's not the case with the Panucatt drivers I am using now. I'm using the genuine Protoneer Pi shield now (older version), it works very well so far.
     
  29. Paul Pridday

    Builder

    Joined:
    Jul 17, 2016
    Messages:
    67
    Likes Received:
    23
    Thanks for all the help guys. Sorry for taking so long getting back but as usual... life gets in the way. BUT! take a looky here!

    Hello World.JPG

    It was all down to the microstepping as suggested. Sounds nice and smooth. Even my wife was impressed!! I started on the printed side of the Styrofoam as I thought I might botch it but it went smooth. So I went back and dropped the Z by 3mm and manually jogged the face clean. Next time I'll input a raster code to clean it up. Well... I'll try anyway.

    Thanks for all the help everyone. :):):):):)

    Paul
     
    Giarc and GrayUK like this.

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice