584,800 active members*
4,822 visitors online*
Register for free
Login
Page 1 of 4 123
Results 1 to 20 of 62
  1. #1
    Join Date
    Jul 2020
    Posts
    36

    Torsion box router with a 4th axis.

    I have been working and tweaking this design in Fusion 360. I'm going to wait to for any feedback before doing the detailed connections from the Y rails to the spindle.. This is my 1st build and I know it will be a challenging learning experience. Here are some of my goals:

    * Ability to handle 24"x36" workpieces.
    * 8" vertical clearance.
    * Small enough to fit in the limited space in my shop. The longer X than Y-axis is being driven by this need.
    * Ridgid enough for light aluminum work. For things like making parts to add CNC to my engine lathe and mill.
    * 4th axis for round 3D work with a 7" swing. I'm considering side walls, partly so I could mount a lathe chuck and put the shaft through the wall.

    I'm looking at using a torsion box made from MDO moisture resistant plywood for the base and sealing that. I typically will only be able to access the router from the front, so I'm thinking of raising the base walls under the Y rails to provide for more gantry rigidity. I'll likely put a back on it to help with sawdust and ridigidity, but make it removeable should I ever want to cut stock bigger than the router dimensions.

    I'm looking for input as far as ideas or feedback you have regarding this initial design. Thank you!

  2. #2
    Join Date
    Jul 2018
    Posts
    6248

    Re: Torsion box router with a 4th axis.

    Hi Wire - Here's the quick calc on 12mm screws. Seems they will do the job. I based the whip and buckling on 39" long is that about right? Cheers Peter

    summary 10mm lead screw 12mm dia (so 16mm will work as well)
    using a 3Nm motor at stall produces 173kgf
    1000mm buckles at 400kgf so motor can't buckle screw
    1000mm whips or whirls at 1867rpm and motor can't go that fast cheers Peter (check the math I could be wrong)

    the ultimate speed of the motor comes down to the quality of the drive. If the N23 can get to 1000rpm and the lead is 10mm then it moves at 10m/min which is quite fast for a small machine. You have a large spindle on it and it will move the machine around (inertial forces) so you'll have to tie it down... even my small machines with 500W spindles have to be toned down as they can slide around on my plastic bench when I crank up the accel figures.

    I would change the bearing beam and make it as low as possible and make the walls a little higher to compensate. Will be stiffer and will contain dust better.
    Attached Thumbnails Attached Thumbnails 12x10mm.jpg  

  3. #3
    Join Date
    Jul 2020
    Posts
    36

    Re: Torsion box router with a 4th axis.

    I spent some time firming up the model, especially on the gantry. For the gantry beam, I'm looking at using a 7.5" x 1.5" LVL laminated beam along with some steel and aluminum. See the pics in the next post for more info. Any input is welcome!
    Attached Thumbnails Attached Thumbnails With New Gantry.JPG  

  4. #4
    Join Date
    Jul 2020
    Posts
    36

    Re: Torsion box router with a 4th axis.

    Here are details of my idea for a LVL (Laminated Veneer Lumber) beam capped with metal to allow for fasteners to grab in metal instead of wood. I got the idea from tinkering with the steel channel & LVL that I already had.

    The top and bottom of the LVL would be capped with steel channel. The ends of the LVL would be capped with 1" thick aluminum for bolting on the gantry uprights. On the forward facing side of the LVL a 1.5" x 1.5" 80/20 extrusion (heavier gauge than shown) would be bolted. This would provide lateral beam strength, a place to mount the X axis rails, an addional place to bolt the gantry uprights into. This is similar to the C shaped gantry beam some other builds use.

  5. #5
    Join Date
    Jul 2020
    Posts
    36

    Re: Torsion box router with a 4th axis.

    Hi Pete,

    Thank you for more great info. I updated the model of the router with more details on the gantry beam and mechanical connections. I also raised the walls about 1.5" higher which reduced the height of the gantry and hopefully increased rigidity. Let me know what you think.

    Thank you, Jerry

  6. #6
    Join Date
    Jul 2018
    Posts
    6248

    Re: Torsion box router with a 4th axis.

    Hi Jerry - I think your gantry has to be square. Gantries are loaded in all directions -up, left, right, back and front and torsion. Torsion especially is a problem. Rectangular beams are poor in torsion and being timber its shear modulus is down so everything counts. A square one does not impact your envelope or your footprint so you should go there. I also don't see the need to put the extrusion under the rail its local stiffness will be less then the solid timber. To accommodate the delta between the ballscrew bearings and the rails just route a pocket down the solid timber gantry. The metal bits at the ends could go, unless that's the strategy to align things with. The less things that strain has to flow into and out of (connections) the better (stiffer) the machine is...You can use normal screws into timber (what wood is it by the way) I use a taper tap and then when the screw is threaded it tightens at the bottom. Once everythings right I set the screws with PVA glue never comes undone and its easy to unscrew if needed. Peter

  7. #7
    Join Date
    Jan 2008
    Posts
    1523

    Re: Torsion box router with a 4th axis.

    Get rid of the aluminium extrusions.

    Aluminium extrusions are often used because they are "easy" to build a machine frame of.

    You are not using the "easy" aspect of them so I don't see any benefit to them for you.

    They are not necessarily flat or stiff.
    7xCNC.com - CNC info for the minilathe (7x10, 7x12, 7x14, 7x16)

  8. #8
    Join Date
    Aug 2014
    Posts
    232

    Re: Torsion box router with a 4th axis.

    How about a square section gantry with a round tube inside it screwed where it touches, with one linear rail mounted at the bottom of a vertical face and the other rail on the top back corner as far from the first as possible.

    The Y carriage will have to be L shaped, but the room on top of it can house the motor for the rack and pinion Y drive and ballscrew Z drive.

    Similar to my AXYZ machine:

    ( That's not my machine ). It works well.

  9. #9
    Join Date
    Jul 2020
    Posts
    36

    Re: Torsion box router with a 4th axis.

    Quote Originally Posted by pippin88 View Post
    Get rid of the aluminium extrusions.

    Aluminium extrusions are often used because they are "easy" to build a machine frame of.

    You are not using the "easy" aspect of them so I don't see any benefit to them for you.

    They are not necessarily flat or stiff.
    Pippin,

    My thinking on the extrusions was to allow an adjustable interface between the laminated wood walls of the torsion box to the linear Y rails (See the circled area in the pic). This may or may not be valid.
    I currently have my side walls within 0.004" of each other. But I wanted the extrusion or other non glued down part to allow me to make finer adjustments to the height and parallel of the rails. It gives me the ability to loosen bolts and place shims. I'm a little hesitant to completely rely on the glued up wood to be within the tolerance I may want to get to. I also thought it would be easier to adjust the rails and give them metal to fasten to instead of the wood. The extrusions would also allow me to shift the Y rails to the top instead of the sides on this newbie's 1st router.

    I have suggestions to redesign my gantry beam and the extrusion may not be needed for adjustability on what I & the great minds here come up with. What are your thoughts on using the extrusion on the lower rails to allow finer adjustments?

  10. #10
    Join Date
    Jul 2020
    Posts
    36

    Re: Torsion box router with a 4th axis.

    I wanted to verify my axis labels so I'm not causing confusion. On my 24"x36" machine in the pics above, the machine will only have access from the front 36" side which is facing the spindle. I'm thinking the 2 shorter rails on top of the torsion box would be Y and the rails along the gantry beam would be the X axis. Is that correct?

  11. #11
    Join Date
    Jul 2020
    Posts
    36

    Re: Torsion box router with a 4th axis.

    Quote Originally Posted by Zorbit View Post
    How about a square section gantry with a round tube inside it screwed where it touches, with one linear rail mounted at the bottom of a vertical face and the other rail on the top back corner as far from the first as possible.

    The Y carriage will have to be L shaped, but the room on top of it can house the motor for the rack and pinion Y drive and ballscrew Z drive.

    Similar to my AXYZ machine:

    ( That's not my machine ). It works well.
    Zorbit,

    Are you thinking of using a section of a round steel pipe/tube as the strength member inside a wood box beam? That's an interesting idea. If the forces of expanding foam could be controlled, you could fill in the 4 sections between the pipe and wood with expanding foam and really lock it together.
    I have also thought about the advantage of having 1 rail on the side of the gantry beam and the other on top. I was currently looking at using a ball screw and using the top of the gantry beam for the drag chain. I don't have a lot of room to add extra depth to the machine to hang the drag chain off the back of the gantry beam.

  12. #12
    Join Date
    Jul 2020
    Posts
    36

    Re: Torsion box router with a 4th axis.

    Quote Originally Posted by peteeng View Post
    Hi Jerry - I think your gantry has to be square. Gantries are loaded in all directions -up, left, right, back and front and torsion. Torsion especially is a problem. Rectangular beams are poor in torsion and being timber its shear modulus is down so everything counts. A square one does not impact your envelope or your footprint so you should go there. I also don't see the need to put the extrusion under the rail its local stiffness will be less then the solid timber. To accommodate the delta between the ballscrew bearings and the rails just route a pocket down the solid timber gantry. The metal bits at the ends could go, unless that's the strategy to align things with. The less things that strain has to flow into and out of (connections) the better (stiffer) the machine is...You can use normal screws into timber (what wood is it by the way) I use a taper tap and then when the screw is threaded it tightens at the bottom. Once everythings right I set the screws with PVA glue never comes undone and its easy to unscrew if needed. Peter
    Pete, you may be right about the vertical nature of my lvl based gantry beam. I thought the aluminum extrusion may help resist side movement. I have read that a lvl is 4 times as strong as a similarly sized sawn timber beam. There is a lot of info on the vertical deflection of lvl, but not too much on side deflection, which as you mentioned, I need. Googling "cnc router" images shows most gantry beams higher than the wide, but mine is a lot higher than wide if you don't count the aluminum extrusions. Here are a couple of brain stormy options:

    Go with a full extrusion beam. (expensive)
    Go with a square steel tube. (hard to get a flat surface on)
    Build a box beam from 3/4" MDO plywood. Possibly with a steel tube in it as Zorbit mentioned. (But is this as strong as the 2 laminated LVL idea below?)
    Bolt steel plate on the top and bottom of the LVL to act as flanges of an I-beam
    Laminate 2 lvls together and cut in a notch for the screw. That would give me a beam about 7.5" x 3.5" which seems a better proportion. And like an I beam, a notch in the center of it probably isn't going to affect strength much.
    I'd need to think about how I mount the rails to the LVL allowing for fine tuning.
    Also, it the notch is just big enough for the screw, there is no way I can get part of the motor tucked under the Z axis plate. That would mean the motor sticks all the way out, I flip it using a belt or switch to rack & pinion.


    As I already have access to LVLs, what are the thoughts about the doubled up LVL with the notch for the screw? The weight of the 2 LVLs would be about 25lbs. LVL is pretty stable already, but 2 glued to each other would be even more stable.

  13. #13
    Join Date
    Jul 2018
    Posts
    6248

    Re: Torsion box router with a 4th axis.

    Hi Jerry - There are plenty of successful 100% timber routers out there. I think you should not mix your media. Stay in timber world or AL world etc. A small bit of metal to augment timber is fine but thats it. Timber is strong enough if stiff enough. Forget about strength you need to define how stiff the machine has to be to do what you want it to do and start there. All of the materials will have "adjustment issues" each needs its own solution. With care you can make the timber flat. Steel surfaces are hand scraped or lapped flat so timber is similar. So:
    1) what you want to do with the cnc defines its stiffness (cutting timber, plastic, aluminium etc)
    2) Then what tolerance you want the machine to work to defines the processes and care that is required to achieve that tolerance, including the quality and tolerance of the bought mechanical parts
    3) with a timber machine it needs to be in a stable environment & all surfaces need to be sealed with resin or deep paint.

    https://www.cnc-holzfraese.de/

    Cheers Peter

  14. #14
    Join Date
    Jan 2008
    Posts
    1523

    Re: Torsion box router with a 4th axis.

    Quote Originally Posted by TheWire View Post
    I wanted to verify my axis labels so I'm not causing confusion. On my 24"x36" machine in the pics above, the machine will only have access from the front 36" side which is facing the spindle. I'm thinking the 2 shorter rails on top of the torsion box would be Y and the rails along the gantry beam would be the X axis. Is that correct?
    Yes.

    Standing in front of the machine, facing it as you will use it:
    X is left to right
    Y is front to back
    Z is down to up vertically

    This is based on the convention of drawing and graphs, where X is the horizontal axis, Y is perpendicular to X and Z is the third dimension, perpendicular to both.

    My machine is setup with X across the gantry.
    7xCNC.com - CNC info for the minilathe (7x10, 7x12, 7x14, 7x16)

  15. #15
    Join Date
    Jul 2018
    Posts
    6248

    Re: Torsion box router with a 4th axis.

    Hi Jerry - Regarding XYZ
    Z's easy, universally +Z is up and -Z is down (I think). Now X & Y just have to follow the right hand screw rule. I use a laptop on my machines and I orient the X&Y in the same orientation as UCCNC has it on the screen so its WYSIWYG. This means it's different on two machines because on one I have the laptop facing the machine and on the other its to the side of the machine. So its up to you. But convention has it that standing at the front of the machine as Pippin says X positive is to the left, Y negative is away from you and Z positive is up. Peter
    Attached Thumbnails Attached Thumbnails RHS Rule.jpg  

  16. #16
    Join Date
    Jul 2020
    Posts
    36

    Re: Torsion box router with a 4th axis.

    RNFEA
    Red Neck Finite Element Analysis

    I'm trying to figure out the deflection on my proposed LVL gantry beam.
    Ger21 (Gerry) told me 50 lbs of force to mill aluminum on a small router would be a figure to start out with. I was looking at using 7.5" x 1.75" LVL as a primary member in my gantry beam (above pics). Pete had expressed concern about the performance of the vertically oriented beam. So I did an experiment using a piece of LVL supported on each side to represent the length of my gantry beam. I set-up a dial indicator to measure deflection and placed a 50lb battery in the center of the beam.

    Results:
    1 LVL on its side: 0.010" deflection
    2 LVL stacked on their sides but not fastened in any way 0.008" deflection
    2 LVL on their sides 0.008" Surprising as this is the way they are supposed to be strongest.

    Obviously not a scientific test. It doesn't take into account the more complex loads that a spindle would exert on a beam. I have to seek out what other's are seeing for deflection, but 0.01" seems like a lot for machine that can hopefully handle light aluminum work. Once I find out more about acceptable amounts of deflection, I can figure out the deflections of various other gantry beam options.

    Does anyone have any deflection information on their builds or a place I might look for more info?

    Thank you,

    Jerry

  17. #17
    Join Date
    Jul 2018
    Posts
    6248

    Re: Torsion box router with a 4th axis.

    Hi Jerry - Small benchtop mills are 1-2N/um (1um=0.001mm) serious VMC are 100N/um plus. I've seen figures of up to 650N/um for very serious machines. Torsion is going to be the weakest link so you need to set up something that will torque the trial gantry. There is a thread in mycncuk that quotes various machine stiffnesses but I can't seem to find it. I shot at 10N/um for small machines... Peter attached is some info on mills

  18. #18
    Join Date
    Jul 2020
    Posts
    36

    Re: Torsion box router with a 4th axis.

    Quote Originally Posted by peteeng View Post
    Hi Jerry - Small benchtop mills are 1-2N/um (1um=0.001mm) serious VMC are 100N/um plus. I've seen figures of up to 650N/um for very serious machines. Torsion is going to be the weakest link so you need to set up something that will torque the trial gantry. There is a thread in mycncuk that quotes various machine stiffnesses but I can't seem to find it. I shot at 10N/um for small machines... Peter attached is some info on mills
    Pete,

    Thank you for the info. If my calculations are correct, the stiffness I seen on the single horizontal LVL was about 0.9N/um. I'd expect that to triple if I glued 2 LVLs together. As I have them, I may do that and try the test again. I might also build a torsion box beam to see how that performs with the simple deflection test.

    It seems the gantry beam is going to be the part of the router that needs the most design attention. I have the torsion box base together and its extremely rigid, but getting something that stiff in the air above the table is another story.

  19. #19
    Join Date
    Jul 2020
    Posts
    36

    Re: Torsion box router with a 4th axis.

    I made good progress with constructing the torsion box. I'm fairly certain about this part of the design so I moved ahead with it. I used MDO water-resistant plywood used for boats and outdoor signs. MDO has the flatness of MDF with the rigidity of plywood. Its great to work with. The base torsion box is 38.5" x 24.375" x 5.5" I have a ~6" grid of intersecting dado'd ribs inside. Of course, I forgot to take a picture of all the ribs in place. For the assembly, I laid a piece of MDF on a floor that was poured with self leveling gypsum and used that as my reference for flatness. I used urethane glue and 200+ screws to hold it together. I'm sealing it with several coats of primer/sealer. I hope to get the sidewalls, which support the 2 Y rails on this weekend.
    This box is really rigid & so flat I can't see light under a straight edge. As my Dad used to say... You could breed elephants on that thing

  20. #20
    Join Date
    Jul 2018
    Posts
    6248

    Re: Torsion box router with a 4th axis.

    Hi Jerry - The Z axis is the thing that requires the most attention to get stiffness, then the gantry itself. Your torsion box points the way for both. Geometry is your friend, make things as big as possible. THINK BIG, BIG IS BEAUTIFUL etc etc Peter

Page 1 of 4 123

Similar Threads

  1. Torsion Box Materials
    By Ranga in forum WoodWorking Topics
    Replies: 1
    Last Post: 03-13-2018, 04:48 PM
  2. Torsion Box Construction
    By TommyGunn in forum DIY CNC Router Table Machines
    Replies: 8
    Last Post: 01-14-2013, 09:35 PM
  3. Torsion box vs Laminating MDF.
    By kiwichris in forum DIY CNC Router Table Machines
    Replies: 6
    Last Post: 06-27-2011, 10:26 AM
  4. Torsion box question
    By chuckknigh in forum DIY CNC Router Table Machines
    Replies: 13
    Last Post: 12-27-2010, 04:47 AM
  5. 80/20 torsion box and router design
    By KTP in forum 80/20 TSLOTS / Other Aluminum Framing Systems
    Replies: 1
    Last Post: 03-01-2006, 04:24 AM

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
  •