Welcome to Our Community

Some features disabled for guests. Register Today.

Super Strengthened Axis?

Discussion in 'CNC Mills/Routers' started by drxlcarfreak, Apr 24, 2017.

  1. drxlcarfreak

    Builder

    Joined:
    Apr 18, 2017
    Messages:
    32
    Likes Received:
    12
    Hi all, first post here. I have been reading up on this exciting open source product for a few weeks now since I have discovered the capabilities. I am going to be working on a design here soon based on the awesome designs that I have seen (my hopes probably exceed what is "cheaply" possible, but I can expand on that later). First I had a couple questions about strengthened beams.

    I saw some people using 2 c-beams back to back to create an I-beam to vastly strengthen unsupported spans which seems great for the Z axis where the force is more of a cantilever and the forces from the router are more linear in the X and Y directions. Vertical deflection is just as important on the X&Y axes as well, but there are also more torsional forces involved from the router, the gantry in my head would be the biggest one since the (I think we are calling the bottom axis Y) axis rails are sort of tied together and the cutting forces would be turned into more of a linear horiztonal force.

    For these cases, wouldn't it make more sense to tie the C-beams together to make a box beam? This may be done elsewhere, but I have not been able to find anyone else doing it this way in my searches. Even for lateral forces, the moment of inertia is 46% higher than a dual C-Beam I-Beam (Sketchup is telling me 703,646.9 mm^4 vs 1,031,956.1 mm^4). Granted, this does assume full rigidity between the pieces in both cases. Unfortunately, it has been a few years since I have graduated and have not been using my engineering degree enough to remember how to calculate the torsional constant, but from "eye-balling it" calculations it definitely would be stronger torsionally. I wish I could put a number to this though.

    Granted, the downside is you can't use the recess of the C-beam for v-slot wheels, so it makes the machine a bit bigger.

    And adding onto this tangent, to add even more strength to the dual C-Beam Box-Beam, why not slide a 40x40 channel in the useless void in the middle to stiffen up the beam even more? Granted, Sketchup is telling me the moment of inertia increase is only 7-8% in each direction since all of the added support is so close to the centroid, but added stiffness is added stiffness provided that it doesn't increase the weight too much for the motors to push! At this point, I think we are approaching the size of a 3030 piece of 8020, and that is used on aluminum cutting routers with 48" gantries.
    Beam.png
    Anyways, these are just some questions/thoughts I have had trying to implement the design I am looking to create. I look forward to your comments!
     
  2. Kevon Ritter

    Kevon Ritter Veteran
    Builder

    Joined:
    Apr 30, 2015
    Messages:
    597
    Likes Received:
    294
    Just get bigger extrusion that is made as one piece already. If you plan to bond it all with epoxy (or similar), you can even consider filling in the larger extrusion At the end of the day, the Open builds hardware is probably limited to the linear motion which would be the wheels. Stronger this means stronger that. You'll have supported shaft at the least with a heavier build.
     
  3. drxlcarfreak

    Builder

    Joined:
    Apr 18, 2017
    Messages:
    32
    Likes Received:
    12
    I was under the impression that the C-beam is the largest beam. Wheel strength and wear on aluminum extrusion is next on my list to figure out.
     
  4. Kevon Ritter

    Kevon Ritter Veteran
    Builder

    Joined:
    Apr 30, 2015
    Messages:
    597
    Likes Received:
    294
    Haha, far from it. C-Beam is the largest for the V-Wheel linear motion. The V-Wheel system is based off of a 20x20mm grid. Commonly available extrusion is available up to 45x45mm. But like I said, you probably should not be using any kind of wheel setup on anything bigger.
     
  5. drxlcarfreak

    Builder

    Joined:
    Apr 18, 2017
    Messages:
    32
    Likes Received:
    12
    Ahh, that is what I meant. The biggest available for the V-wheel system. What kind of system would you use for a larger setup if you aren't using wheels?

    Before I was turned onto this system, I was set on going with one of the CNCRouterParts setups, but they are a bit too expensive for me for my first go at a CNC.
     
  6. Rodm

    Rodm Well-Known
    Builder

    Joined:
    Jan 3, 2017
    Messages:
    121
    Likes Received:
    42
    There are alot of great ideas and designs here and more showing up all the time. What do you plan to do with your cnc? How big does machine need to be to do that well? Design for your use. You may not have to beef up as many things as your thinking now.
     
  7. drxlcarfreak

    Builder

    Joined:
    Apr 18, 2017
    Messages:
    32
    Likes Received:
    12
    There are almost an overwhelming amount of machines and ideas on this site!

    The base plan for my machine is to cut foam and bondo for plugs to make fiberglass molds from. In looking at efficiencies of layering and piecing milled parts vs cost of a larger machine, it looks like the sweet spot for me is a cutting area of a 25"x25"x6" cutting area. I know full cutting depth of 6" will be impacted by the spindle and the mount, but there wont ever be just a full 6" plunge cut, it will be more of slopes over the 24". In looking at models of completed builds, it looks like I can position the cutting pieces to adequately clear the mounts.

    Due to lack of budget and room, I also would like the machine to be able to mill some wood/MDF/small bits of aluminum at times. A way out of left field goal here would be to have a gantry width that can clear a sheet of MDF. This is a secondary goal though, and probably not really achievable with this system.

    I know that cutting harder pieces and a long Z axis dont mix. My thought that I have not been able to model out yet, is a recessed bed to allow the z-axis to extend down 6" to the spoil board cutting area of about 24"x24", then have a sort of a drop in secondary bed of 48"x24" to bring the bed level up to the same level as say the Ox CNC for cutting harder materials. I have seen builds with 48" and even 1500mm gantries, but have not seen any real conclusion on them to see how stiff they actually are, so I am guessing that I will loose quite a bit of rigidity going with that large of a gantry.
     
  8. Giarc

    Giarc OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Jan 24, 2015
    Messages:
    3,018
    Likes Received:
    1,688
    You could build everything flat like a normal ox and just build a box (or use solid layers-much heavier) with a spoil board on top to bring it up to about 1.5 inches from the bottom of the spindle mount when cutting aluminum and hardwoods. That way you can feed through longer materials. On mine, the main work surface is never touched unless I am cutting things 2-3.5 inches thick. When milling anything else, I have extra layers of spoil board so the z does not have to drop that much. This also helps keep the main surface looking pretty, and that is what really matters. :thumbsup:
     
  9. drxlcarfreak

    Builder

    Joined:
    Apr 18, 2017
    Messages:
    32
    Likes Received:
    12
    That is similar to what I am thinking of, just a lot simpler! I took a quick look at your build. Looks like an awesome machine! Have you been able to cut aluminum with it with the bed raised up? It looks like you used dual 20x80s and 838mm long, a bit of an upgrade from the Ox! How rigid is it for you?
     
  10. Giarc

    Giarc OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Jan 24, 2015
    Messages:
    3,018
    Likes Received:
    1,688
    I do cut aluminum. Here is a video of it cutting aluminum plates for my eggbot.

    [​IMG]
    I am getting ready to cut a plate for a downrigger mount for my kayak and a better rod holder plate for when trolling for salmon in my kayak.

    The X axis seems to be very rigid. The two beams are bolted together. I also always cut close to one side rather than in the middle. Usually starting at the 0,0 point. The Y axis 20x80 and is supported the full length by an aluminum plate which is then fastened to a 20x40 piece which is in turn fastened to the torsion box table. The Y axis beams can not flex (at least not noticeably) vertically or horizontally. My goal was to be able to cut 26 inch wide pieces (cabinet sides). I could have made it 1000mm instead of the 838mm, but it was not necessary for my needs, and it would have taken too much floor space. Therefore, I opted for less flex in the X and more available floor space in my small shop. Technically, the X axis was supposed to be 850mm, but in my efforts to get it perfectly square, I kept shaving off a little more until I figured I had better stop. :) It can cut 667mm with the limit switches (670mm without) in the X axis, which is a hair over 26.25 inches.

    I recently ordered a blower nozzle to attach to my air compressor for clearing the aluminum chips.
     
  11. drxlcarfreak

    Builder

    Joined:
    Apr 18, 2017
    Messages:
    32
    Likes Received:
    12
    Nice! I may steal a thing or two from your build if that is okay.

    I like the connector pieces and bolting for the X axis. I haven't gone all the way through the Ox build, but it didn't look like they are connected in that build, I know its not shown in the model. It really makes sense to tie them together so they have to twist as a single beam instead of being able to twist independently and rack.

    Your Y axis looks like a good upgrade and a great way to drop the spoiler board down a bit lower. As rigid as your supported X rail and spoiler board table looks, without securing it to a torsion box, I bet it is as solid as an Ox router bolted to a torsion box. I may actually stick with around an 850mm granty for now. While a 1250mm cutting width gantry would be awesome to be able to push full sheets through, it may be overkill for now, and honestly to expand the X later I really only need a wider table and gantry.

    I guess my next step is looking into cutting forces. Is there any data out there on a typical cutting force of aluminum, vs mdf vs foam?
     
  12. Giarc

    Giarc OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Jan 24, 2015
    Messages:
    3,018
    Likes Received:
    1,688
    Steal away! That is what this site is all about. At one point in time, I wanted to push full sheets through in case I got the wild idea to build kayaks again, However, I never really needed anything wider than 8-12 inches for that, so that helped in the decision making. Honestly, other than a couple of strangely shaped cabinet doors, and some foam Halloween tombstones, I do not think I have ever cut anything from stock wider than 8 inches. I seem to mostly wear out the same 12 inch x 12 inch spot on my spoil board. I would like to master the pin hole registration technique for longer goods.
     
  13. drxlcarfreak

    Builder

    Joined:
    Apr 18, 2017
    Messages:
    32
    Likes Received:
    12
    Awesome! That's good to know. Haha fair enough. A smaller machine will tuck away easier in my garage too! Are you able to tell me about what your machine cost all in, just so I have an idea of what I am in for?
     
  14. Giarc

    Giarc OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Jan 24, 2015
    Messages:
    3,018
    Likes Received:
    1,688
    I am scared. I have never done this before. :nailbite:

    $133.63 for my first OpenBuilds order (linear rail and screws)
    $124.49 for another OpenBuilds order via Amazon for 4 of the 20 x 80 v-slot
    $62.99 for a third OpenBuilds order via Amazon for 3 of the 20 x 40 v-slot
    $160 for DQ542MA drivers (+ a another optional $40 to replace a driver I spilled beer on while powered. Not recommended.)
    $114.57 for three 269 oz Nema 23s for X and Y, and one 178.5 oz Nema 23 for Z, and a 36 V power supply. OpenBuilds no longer sells the 269 oz (or was that 270 oz Nema 23s)
    $210 in miscellaneous supplies and tools from Aliexpress (12 mm x 6.5 mm shaft couplers, pillow block bearings (8 and 12 mm) , metric drill bits, tap sets, step bits, shrink tube, cooling fans, limit switches, etc...
    $200 - ish for OB wheel kits
    $24 for genuine Arduino

    Ugh... I have never added it up until now...it is starting to hurt.

    $20.94 2 pieces from Amazon 15mm x 30mm Black Plastic Semi Closed Drag Chain Cable Carrier 1M
    $64.85 Hopkins 90164 2x4 basics Workbench Amazon.com
    $54.99 HTC-2000 adjustable mobile base,
    $27.14 HTC K1218 Extension Rail Kit for HTC2000/PM1000 Universal Mobile Base
    $40 for shielded cable from Home Depot
    $20 for the computer case
    $10.58 for aviation connectors (Amazon)
    $99.99 Makita router Home Depot
    $56- 4x8 sheet 3/4 inch particle board (2 layers spoil board and cross members of torsion box) and 4x8 sheet of 1/2 inch particle board (torsion box top and bottom)
    9.99 Mist Coolant Lubrication Spray System (just bought) from Amazon.com
    $40 for 1/4" and 1/8" aluminum plates and bars ($3.50 per pound) from a local metal dealer. This is an estimate, I still have quite a bit left which I have used for other projects.
    $20 (approximately) from various sources for other hardware (3mm set screws and M3, M4 socket cap screws) mostly from amazon, but some from Home Depot.

    Time to break out the Knappogue Castle bottle to help ease the pain. :eek:

    MCMaster Carr
    99030A703 1018 Carbon Steel Precision Acme Threaded Rod, 3/8"-8 Size, 1/2" Travel/Turn, 6' L, Right-Hand Thread, 4 Starts 1
    16.86
    8573K125 White Delrin (R) Acetal Resin Sheet, 3/4" Thick, 6" X 6" 1
    18.93
    91290A272 Black-Oxide Class 12.9 Socket Head Cap Screw, Alloy Steel, M5 Thread, 70MM Length, 0.80MM Pitch, Packs of 25 1
    2.03Per Pack
    91290A274 Black-Oxide Class 12.9 Socket Head Cap Screw, Alloy Steel, M5 Thread, 75MM Length, 0.80MM Pitch, Packs of 25 1
    3.77Per Pack
    99030A704 1018 Carbon Steel Precision Acme Threaded Rod, 1/2"-10 Size, 1/2" Travel/Turn, 6' L, Right-Hand Thread, 5 Starts
    2 total 81.04
    Merchandise 122.63
    Shipping 22.35
    Total $144.98

    OUCH!

    Approximate total:

    $1639.14

    Yowza! I wish I would not have done that. If I had to guess, I would have said about $1000. I guess all those 20-60$ amazon/Home Depot purchases really add up! Oh, and I spent about $100 on cutters.

    I should mention that I have lots of screws and other parts left over for other projects and I made an eggbot with left over v-slot. Also, the taps and drill bits are still being used for other projects.I have enough 20x80 left to make 2/3rds of a Voxel Ox, which I may do. And enough delrin for a lifetime supply of anti -backlash nuts.

     
    GrayUK and Joe Santarsiero like this.
  15. drxlcarfreak

    Builder

    Joined:
    Apr 18, 2017
    Messages:
    32
    Likes Received:
    12
    Hahaha. Much appreciated even though it was painful! I'll have to look at this in more detail, but I was actually expecting worse. The Ox kits I've seen are about $1,100 base for the larger foot print, and that doesn't include the spoilerboard, mobile work station or torsion box. Your design has a larger work area and looks much stiffer! I'd say well worth the extra coin.
     
  16. Kevon Ritter

    Kevon Ritter Veteran
    Builder

    Joined:
    Apr 30, 2015
    Messages:
    597
    Likes Received:
    294
    Once you ditch the 20x80 with belts and go with the CBeam with screws, the price jumps up, but so does the machine as a whole in my opinion. I haven't tried the belts, but I have no confidence in their long term precision and reliability. I know the belt is reinforced, but it's still made of rubber. I wouldn't even try to put it on anything that will see continuous cutting loads. I dont know, it's just one if those things for me.
     
  17. Giarc

    Giarc OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Jan 24, 2015
    Messages:
    3,018
    Likes Received:
    1,688
    I really like the acme lead screws, too. Oh, I also bought a laser for it. Since I am mounting it on this one, I will have two machines in one thereby cutting the price in half. I feel better now.

    The only place I used Cbeam was on the Z. Originally, I planned to use it but they ran out of stock so I modified my design. I am glad I did.
     
  18. Kevon Ritter

    Kevon Ritter Veteran
    Builder

    Joined:
    Apr 30, 2015
    Messages:
    597
    Likes Received:
    294
    Speaking for two in ones, I was talking with a buddy of mine and one of the ideas that came up was a printer/router combo. The printer would just dump material where it's needed, and the a router would shape it. They would both have independent Z movement.
     
  19. Giarc

    Giarc OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Jan 24, 2015
    Messages:
    3,018
    Likes Received:
    1,688
    Sounds interesting. It would be cool to see it in action.

    I had a debate with a friend regarding printing vs milling with regards to making plates and brackets for mounting fishing stuff on our kayaks. Printing a part takes a couple hours and uses $11 per pound plastic. I get aluminum plate for $3.50 per pound and can mill a piece in less than an hour easily. I think designing and printing the prototype for fitting and testing then milling the final part is probably the way to go.
     
  20. drxlcarfreak

    Builder

    Joined:
    Apr 18, 2017
    Messages:
    32
    Likes Received:
    12
    I definitely agree about the belts. I think that for my purposes, screws are the minimum. I am a fan of the rack and pinion setup even moreso, but I believe that is because if I want to extend an axis in the future, I can just get some more rack and add on instead of having to buy a new screw and have a junk one sitting around. I don't feel like I am wasting parts, or at least wasting less so. Even still, the more I think about it, if I were to expand the machine, I would probably be more likely to build a completely second machine in tandem.

    Haha, I like that idea. One thing that would be beneficial in my garage is a plasma cutting table. It would be nice to be able to combine those machines, but without some heavy work I don't see getting a two for one special like that!

    Could you elaborate more on the 20x80 vs. C beam a bit more? You guys mentioned the price jumping up, but to get the same footprint of the C-beam requires 2 20x80 beams.

    The printer/router combo sounds pretty cool. I would be interested to see how that would be used. In my limited experience, it seems like 3d printing is great for rapid prototyping to make sure that the part fits before committing to cutting it in the final material. I have been wanting one for years, but when it is time to break out the checkbook, I just can't come up with a great list of things that I would make.
     
  21. Giarc

    Giarc OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Jan 24, 2015
    Messages:
    3,018
    Likes Received:
    1,688
    I think the C-beam would be fine. It wasn't available in 1500 mm lengths when I made my purchases. It is not that much more expensive. I am very satisfied with my doubled up 20x80 X axis and I believe it would be stiffer than a single c-beam the same length. But, I am no engineer and there may be engineery magic which occurs because it is a single extrusion vs two bolted together. On the Y axis, I added 20 x 40 perpendicularly to my 20 x 80 beams to stiffen them and to use the v slot for clamping (which I, I have found, hardly ever use:rolleyes:). With the Cbeam, this "stiffener" is built in.
     
  22. GrayUK

    GrayUK Openbuilds Team Elder
    Staff Member Moderator Builder

    Joined:
    May 5, 2014
    Messages:
    1,668
    Likes Received:
    627
    Giarc:
    "OUCH!
    Approximate total:
    $1639.14"


    Craig, I can tell you liked that bit of financial pain really!! :D
    Would I ever do that? No, no, no, never ever! No Way!! :eek:
    What if my wife saw it?! I would lose sleep over something like that! :nailbite:
    Stuff like that needs to be left alone, and I will live in sweet ignorance. :thumbsup:
    Gray
     
    Giarc likes this.
  23. Giarc

    Giarc OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Jan 24, 2015
    Messages:
    3,018
    Likes Received:
    1,688
    I don't think she will ever look here. Also, I forgot the $30 65mm spindle mount. :) But, I made a custom aluminum fishing rod holder mount. It would have cost me $25 for one, so by using scrap aluminum someone gave me I am chipping away at that $1639.14...uh, I mean $1669.14.
     
  24. drxlcarfreak

    Builder

    Joined:
    Apr 18, 2017
    Messages:
    32
    Likes Received:
    12
    Haha, well that money is an "investment". After a few jobs it will begin to pay for itself. Well that's the hope anyways!

    Actually the dual 2080 beam has a higher moment of interia vertically, so from direct sagging due to weight it would be stronger. In civil engineering, we can put two C beams together to create an equivalent I beam (I have one in my living room, 2 150lb beams were easier to lift and bolt to each other than one 300lb beam). They need to be bolted together at a certain spacing, which I can't remember off the top of my head. The only concern I'd have would be the potential twisting of each beam independently between bolts, but it would be tiny, tiny, tiny twisting. I like your stiffner idea, I'm trying to figure out a way to use something like that to be able to raise the Y rails up another 3-4 inches or so to get some clearance to the table and still be rigid.
     
  25. drxlcarfreak

    Builder

    Joined:
    Apr 18, 2017
    Messages:
    32
    Likes Received:
    12
    This may be a dumb question, but what is the typical length of the bit sticking out of the collet on something like the Makita RT0701C? I searched quickly and couldn't find much, most of the models I have looked at show about 1" under the collet.

    I am not concerned with the cutting depth so much, just looking at clearances with the tip of the bit, the Z axis and clearing the material below. Is it 1"? Are there "cheap" options for longer bits for say 2-3"? It would be just for cutting foam and other soft materials, so strength isn't much of a concern with longer bits. Thanks!
     
  26. Giarc

    Giarc OpenBuilds Team
    Staff Member Moderator Builder Resident Builder

    Joined:
    Jan 24, 2015
    Messages:
    3,018
    Likes Received:
    1,688
    Kodiak Cutting tools has extra long end mills that are affordable and I have had good luck with them Long Length Carbide End Mills | Kodiak Cutting Tools . I had to but them to drill a hole pattern in my lower-most spoilboard since a standard length would not drill through all the way. They work good on foam too. At least the kind of foam I used. (2" thick Owens corning Pink foam board).
     
    #26 Giarc, May 5, 2017
    Last edited: May 6, 2017
  27. Kevon Ritter

    Kevon Ritter Veteran
    Builder

    Joined:
    Apr 30, 2015
    Messages:
    597
    Likes Received:
    294
    Also checkout Lakeshore Carbide.
     
  28. drxlcarfreak

    Builder

    Joined:
    Apr 18, 2017
    Messages:
    32
    Likes Received:
    12
    Thanks guys! So I started playing around with trying to get 6" of usable travel (the model is just under 7" of clearance) without the vslot interfering with the model too much, and came up with the below. Looks like if I use a bit that sticks out about 3" from the collet, and extend the z-carriage down about 40mm past the wheel, I can get the bit to touch the spoil board below.

    I have a few things here kind of extended to their max cantilever-wise (a 4" bit, the router is slid all the way down in the mount, the mount is attached below the bottom wheel), but are any of these deal breakers for cutting foam?

    I am hoping with this design for a rare aluminum or hard wood job, I can slide in a reinforced spoil board on that sort of shelf you can see behind the gantry, swap in a much shorter bit, slide the router up a bit in the mount for less of a moment arm, and be pretty much the same dimensions and strength as a regular ox.

    Josh Progress Design.png
     
  29. drxlcarfreak

    Builder

    Joined:
    Apr 18, 2017
    Messages:
    32
    Likes Received:
    12
    Here are a couple renders of what I am trying to convey. The blue is a sample of a foam block that I would be cutting. Granted, I could turn the piece around and there wouldn't be a clearance issue due to the vslot, but I believe that there will be a few pieces where there may not be that option, like a void in the middle, so if possible I would like to build for the flexibility!

    Z beam no clearance.png Z Clearance.png
     

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