584,860 active members*
5,252 visitors online*
Register for free
Login
Page 1 of 6 123
Results 1 to 20 of 116
  1. #1
    Join Date
    Feb 2007
    Posts
    75

    Very Large Gantry Router Questions

    Hullo...

    I'm a rank newbie here, and I've got a few questions, regarding the design of a 66"x 40" x 10" gantry router... I'm looking to build a design that's accurate (inaccurate?) within ~0.015" (obviously, [much] more accurate would be better) for milling foam, and possibly some balsa. (Very light duty, sideloads measured in ounces, around 10 oz. max..) A friend of mine is building the driving hardware, so I'm all set in terms of interface/driving board.

    <>

    Attached are a couple sketches of my current design... (Dimensions in inches.)

    I'd be using 25mm/1" open linear ball-bearing units at all bearing points (http://www.vxb.com/page/bearings/PRO...ystems/Kit7445 or similar) and would place supports under the X-axis (66") rails, every 12" or so. (Alternatively, could I use closed bearings, and allow the gantry support columns to rest on the router bed, taking the weight loading off the rails?)

    The spindle would be driven by a Dremel, with a flexible pen-tool attachment, though I'm definitely open to other (more powerful) low-dollar solutions. (As currently designed, the Dremel would actually hang on a slide above the router, with the pen-tool clamped to the Z-axis, to minimize weight loading on the gantry assembly.)

    For axis driving, I'd be using four stepper motors; one driving each gantry column (to prevent skewing) and one each on the Y and Z axii, with Acme leadscrew and dumpsterCNC.com anti-backlash leadnuts on all axii.

    As for questions:
    Where does one purchase shaft/rail material in 6' lengths, at "reasonable" cost? I know these things are expensive, but I would like to do this as cheaply as possible. (Obviously, it needs to have precise OD, if I'm going to use the ball-bearing units. I am tempted to use one of the adjustable rollerblade-bearing designs, but the amount of machining required puts me off a bit.)

    All structures shown would be built from MDF, with a few pieces of plywood here and there. Again, I'm open to alternatives. (I have access to a manual milling machine, lathe, etc.) I'm aware that I need a 'box' type structure to better support the Z-axis. Would MDF be sufficient, or should I redesign the gantry columns etc. to be fabricated from aluminum plate?

    If I were to drill both of the supports for the X-axis at the same time (clamping both blanks in a milling machine, and drilling away) would it be fair to say that I could count on the rails being parallel? If not, what sort of rail adjustment mechanism would be recommended? Also, how wide should the supports be, in order to achieve some approximation of cantilevered shaft/rail mounting? (Right now, I'd be using several layers of MDF, for a total width of ~2".)

    I'm sure I've made some gross oversight somewhere, so any input would be much appreciated.

    ~Luke
    Attached Thumbnails Attached Thumbnails Router.jpg   RouterDim.jpg  

  2. #2
    Join Date
    Dec 2006
    Posts
    252

    The flex in the Y axis...

    I am using the 20mm linear setup from VXB I love that stuff BUT!!!! the to Y rails you have will flex even at the 25mm size so what you need to do is support them vertically some way. I love the design but I am thinking that 2 flat bars sitting vertially and joined to the points at each end just below the y rails will give you more support and stop the flex you will get the beaings need to be the "c" style ones or open rather than full circle ones. this will stop the issue cold I think.

    I can whip out a sketch for you and would direct toward my thread and the pics I post mostly the last one dealing with this flex issue.

    http://www.cnczone.com/forums/showthread.php?t=29877

    Let me know on the sketch

    Coog
    Building Stage:[xxxxxxxxxxxxxxxxxxxx] 105%
    Finishing Stage:[xxxxxxxxxxxxxxxxxxx-] 95%

  3. #3
    Join Date
    Feb 2007
    Posts
    75
    Ok. Oh, golly, so the post went through, huh?

    That means there could (must?) be about 5 copies of this thread...! (If so, VERY SORRY! Mod, can you clean that up?)

    It kept telling me "the thread ID was invalid, please contact the admin"....

    ~Luke

  4. #4
    Join Date
    Feb 2007
    Posts
    75
    So, you're saying I should tie the two columns together with something flat and rigid, and then place supports between that and the Y-axis rails, right?

    Again, instead of using the C-bearings, can I allow the Y-axis carriage to slide on the flat supports? (I'm just tempted by the VXB 20mm "4-closed-bearings-for-$25" set.)

    I'm thinking I could put some round-headed bolts on the bottom of the carriage, so they would be adjustable height, and could slide along some steel square stock, or similar?

    (See attached.)

    ~Luke
    Attached Thumbnails Attached Thumbnails RouterMod.jpg  

  5. #5
    Join Date
    Dec 2003
    Posts
    24216
    Check out the Pacific Bearing products Supported rail and lube free bearings.
    http://www.pacific-bearing.com/produ...oduct.cfm?ID=1
    There is an application mentioning a Router Manuf.
    Also Misumi have similar products.
    Alos ITEM Extrusion Products sell a profile that stainless bearing rod clips into the profile.
    Al.
    CNC, Mechatronics Integration and Custom Machine Design

    “Logic will get you from A to B. Imagination will take you everywhere.”
    Albert E.

  6. #6
    Join Date
    Dec 2006
    Posts
    252

    sorta...

    in your last drawing you show the support but the gantry is wrapped around the rails insted remove the bottom portion where it sits between the flat bar and the round bar and go to a open bearing like in my thread.

    VXB link
    http://www.vxb.com/page/bearings/PRO...Systems/kit989

    then you can use 1" of flat standing vertical directly underneath.

    pics attached.

    Coog
    Attached Thumbnails Attached Thumbnails RouterMod.jpg   supported.png  
    Building Stage:[xxxxxxxxxxxxxxxxxxxx] 105%
    Finishing Stage:[xxxxxxxxxxxxxxxxxxx-] 95%

  7. #7
    Join Date
    Feb 2007
    Posts
    75
    That's what I was asking, though... The closed 20mm bearings are 50% less expensive than the open 20mm/25mm bearings. [cheapskate grin]

    I'm guessing that I'd be playing with fire (friction) by allowing the carriage to be directly supported by the anti-flex member, the way I sketched it?

    ~Luke

  8. #8
    Join Date
    Dec 2006
    Posts
    252

    friction bad...

    you want as smooth as possible movement so friction is bad BAD IAD BAD!!! dont cheapskate when it comes to the rails and motors you'll only pay twice insted of 1.25 to begin with.

    the open bearings are your ticket and that design is nice with a double bar gantry and some verticle supports it will be solid and run reall good plus your z-car sits on 2 sides of each of the bearings.

    I am a long way into mine and I would (if money permitted) change to this design of yours in a heart beat! In fact I might have to as I am not liking my dual rail vertical solution I get horizontal flex where you would get vertical.

    Anyway AWESOME design and yes I am contemplating a change....

    Coog
    Building Stage:[xxxxxxxxxxxxxxxxxxxx] 105%
    Finishing Stage:[xxxxxxxxxxxxxxxxxxx-] 95%

  9. #9
    Join Date
    Dec 2006
    Posts
    252

    in looking again...

    why do you have 2 rails per x axis? if they are going to be supported (I suggest the same as the Y axis chat) then one would be fine and it would be easier to build and align.

    Thoughts?

    Coog
    Building Stage:[xxxxxxxxxxxxxxxxxxxx] 105%
    Finishing Stage:[xxxxxxxxxxxxxxxxxxx-] 95%

  10. #10
    Join Date
    Dec 2006
    Posts
    73
    Looks very nice IAD, I'm planning for the same thing with my design.
    But:
    Which way does the gravity work?
    I would suggest to "flip" the C-bearing blocks so that the opening is facing downwards. Then you could add some kind of supportblock if needed, and the weight distribution around the bearing should get a lot better... Just simple thougths and physics thou, I still haven't recieved mine...
    Building Stage: [///////------]45% Now we are getting somewhere...
    Ordering & planning: [///////////---] 80% Aaaalmost done...

  11. #11
    Join Date
    Feb 2007
    Posts
    75
    The paired X-axis rails were used to helped reduce the sagging at the center, by halving the weight load that each side had to support.
    Changing the design to use the C-bearings, the only reason I can think of for leaving them is to make each gantry column a rotationally rigid structure, which would support those massively overbuilt gantry columns nicely. But you're right, there's really no need, and the ~6' rails are quite expensive.

    I would definitely orient the opening of the C-blocks downwards. The design shown was originally designed based on 'tube-running-on-tube' bearings, but common sense gained the upper hand.

    I learned about the going-cheap-costs-more-in-the-end thingy... Both by observation and experience! Before I start buying, I just want to make sure I've explored all options.

    ~Luke

  12. #12
    Join Date
    Feb 2007
    Posts
    75
    I just had a rather crazy idea, but I guess it's worth asking about.

    What if I added a 2-3 rollerblade bearings as shown, to support the weight? Added friction should be negligible, and the columns don't have to be made any taller, to allow the 10" clearance I want underneath. Of course, how straight is ~0.25" x 1.25" steel/aluminum flat stock?

    Anyway, the long axis will use the C-bearings no matter what, for simplicity's sake.

    By the way, the CAD drawings I've been posting are really just sketches I've done, to try and get a feel for how all this will work, so many things are not shown in any sort of build-worthy form. (Just my disclaimer, for component-size errors in the pictures, and what-not.)

    ~Luke
    Attached Thumbnails Attached Thumbnails RouterMod2.jpg  

  13. #13
    Join Date
    Dec 2006
    Posts
    252

    I like it but..

    This idea with the flat to the sides works except you now sit on the skate bearings (very worthy items) and you cannot guage whether they are holding the gantry or the round stock is so you might end up with over built and the skate bearings at the items doing all the work. as for the possible alternative and cost savings route since we open the skate bearing door, think this way...


    http://www.vxb.com/page/bearings/PRO...esale/100Skate


    lets roll this around no pun intended


    Coog
    Attached Thumbnails Attached Thumbnails skate1.png   skate2.png   skate3.png  
    Building Stage:[xxxxxxxxxxxxxxxxxxxx] 105%
    Finishing Stage:[xxxxxxxxxxxxxxxxxxx-] 95%

  14. #14
    Join Date
    Feb 2007
    Posts
    75
    I had definitely thought that setup, as I mentioned in my first post, BUT... Now I have to machine something that can mount the bearings in those orientations. This may be cake for some, but... Not me, per say. Especially if I want to make them adjustable, to reduce side-to-side slop. Also, I'd sort of like an extra bearing on the bottom, so the thing can't lift.

    However, it would eliminate the cost (and sourcing) precision shafts. I'm assuming you'd just be using some sort of flat-stock for the rails, at that point? (Which leads to the question of just how straight that stuff really is.)

    ~Luke

  15. #15
    Join Date
    Feb 2007
    Posts
    75
    On the other hand, the bearing mount could be quite simple... Something like this, perhaps? It seems a design like this would be tolerant of small alignment errors, without significant increase in friction/slop?

    Of course, there's still no lower bearing, but I imagine the several pounds of steel rail and MDF on the gantry would be enough to keep things in place?

    This design is beginning to grow on me... Any idea what sort of angular/eccentric slop do those bearings have?

    ~Luke
    Attached Thumbnails Attached Thumbnails RouterBearingMod.jpg  

  16. #16
    Join Date
    Dec 2006
    Posts
    252

    no slop really..

    If you look at other cnc designs here you will see many people using these skate bearings and if you went to a "c" channel 2" with 1" legs you can leave the flex behind and make a bearing on top bottom and side no problem.

    sets of skate bearings one front one rear and a set on top side and bottom with angles and bolts with jam nuts for adjustment.

    maybe this setup its cheap quick easy and seen it done here already...


    Coog
    Attached Thumbnails Attached Thumbnails skate4.png  
    Building Stage:[xxxxxxxxxxxxxxxxxxxx] 105%
    Finishing Stage:[xxxxxxxxxxxxxxxxxxx-] 95%

  17. #17
    Join Date
    Dec 2006
    Posts
    252

    like this only channel not pipe

    you should do these plans
    http://www.mechmate.com/

    and look at this for inpiration
    http://www.cnczone.com/forums/forumdisplay.php?f=300


    http://www.solsylva.com/cnc/cnc5.html

    I think this adapted to channel steel facing toward each other or inward would give you the adjustment the stability and remove the flex.

    Coog
    Building Stage:[xxxxxxxxxxxxxxxxxxxx] 105%
    Finishing Stage:[xxxxxxxxxxxxxxxxxxx-] 95%

  18. #18
    Join Date
    Feb 2007
    Posts
    75
    Thanks! I like the Mechmate design. That thing is HUGE. (Or at least, quite large... Still trying to get a feel for what passes as 'huge' when it comes to these things.)

    Anyway, here's my latest proposal for a bearing mounting structure. Construction would be 0.25" thick steel throughout. The bottom bearing is adjustable, as are the 'outer' horizontal bearings. All machining would be performed on a milling machine, to ensure that the holes on each component are parallel.

    Obviously, there's a lot of flat surfaces, which would allow it to be bolted to the MDF carriage assembly pretty easily.

    Any thoughts?

    ~Luke
    Attached Thumbnails Attached Thumbnails RouterBearingMod2.jpg   RouterBearingMod2a.jpg   RouterMod3.jpg  

  19. #19
    Join Date
    Jul 2005
    Posts
    668
    You have to constrain all 3 axes. The gantry will rise without constraint.
    Steve
    DO SOMETHING, EVEN IF IT'S WRONG!

  20. #20
    Join Date
    Feb 2007
    Posts
    75
    The latest design I posted is constrained in all directions, isn't it?

    It can't move up, down, left or right... Granted, a single unit would be able to rotate slightly about the long axis of the rail, but two of them attached on either side of the carriage will not be able to rotate.

    Or am I missing something?

    ~Luke

Page 1 of 6 123

Similar Threads

  1. North Phoenix Large CNC router
    By AZKick in forum Mentors & Apprentice Locator
    Replies: 4
    Last Post: 12-23-2011, 05:54 PM
  2. Vacuum Table for large CNC router
    By stairgod in forum Commercial CNC Wood Routers
    Replies: 5
    Last Post: 07-15-2011, 06:16 AM
  3. large(ish) 5-axis router project
    By vwdevotee in forum DIY CNC Router Table Machines
    Replies: 34
    Last Post: 10-30-2007, 12:04 PM
  4. Flex compensation devices in large gantry routers?
    By Splint in forum Commercial CNC Wood Routers
    Replies: 5
    Last Post: 07-18-2006, 12:59 AM

Posting Permissions

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