584,808 active members*
5,362 visitors online*
Register for free
Login
Page 1 of 4 123
Results 1 to 20 of 74
  1. #1
    Join Date
    Jul 2009
    Posts
    60

    nickswimsfast's CNC Build Log

    About me:
    I'm a senior studying Mechanical Engineering at OIT in Oregon. So I've taken some CAD/CAM classes and enjoying it thoroughly. I'm fortunate that my school teaches us welding and machining in full fledged classes with proper equipment. In 2009 I started on the project to build a budget CNC machine with a friend.

    Project Background
    So I started off on a 1000$ budget and very little equipment access, with the intent to cut costs as much as possible to simply get our first machine made. The machine was originally intended to cut both metal, and wood. The idea was to have a huge work range, but the importance of frame stiffness had eluded me at that time (young, nieve, etc..)

    Summer ended, after some substantial research, design, and purchasing (but not a enough.) The project has been on the backburner ever since, as school came first. I got a high paying internship (20K over 6 months) and now have the financial backing to "do it right."

    I'm caught in a quandry of design directions. I can either backyard engineer up the final 40%, or take a step back. Go back to the drawing board, shore up the details, and have a decent machine (or so I think.)

    Current Hardware:
    • Antek Transformer - Toroid Transformer (currently wired @ 24VDC) wasn't aware of Vrms at the time... I think i can rewire it for 48VDC, but my control board can't take more than 36VDC. Used this to make a powersupply with help from my EE friend.


    • Control Board - HobbyCNC 36VDC board/drivers only. Soldered this badboy up. Added .125" aluminum angle iron as heatsink for drivers.


    • 3 Keling 425 oz-in Steppers - Wired these in unipolar mode running at 2 amps. I'm concerned about power over lines because we used the unshielded RJ-45 cable i had (8x24AWG wire).


    • Homebuilt computer - This was my lanning computer from high school, I have mach installed on it. It can supply a reasonably clean 25Khz pulse stream.

    • Ballscrews - Purchased set of 3 via linearmotionbearings from ebay. Custom sized to match the linear bearings geometry. Slightly oversized compared to linear bearings to make life easy. C7 Tolerance. Got tired of trying to buy ground ballscrews off ebay. Got the complete set for 375 shipped. Came with everything I needed for the ballscrews. Not bad.

    • Profile Rails - Purchased via lots of patience on ebay. THK rails. From memory, they are SRS blocks for the X and Y axis. SHS for the Z-axis.



    • Work Range With the components above and saddle plates of ranging 8-10" in height the effective work range is about:
      1. X-axis = 22.28 in
      2. Y-axis = 19.88 in
      3. Z-axis = 12.38 in








    Designs

    The old/original design.


    The Wood Frame components- as it is right now. If you are wondering why i did this... 1) I don't fear messing up. 2) I had 1 free week to work on this with only a skill saw, and harbor freight hand power drill. 3) I knew I'd learn something for about less than 20$ of 2x4's. 4) Completely reversible.



    Due to the realization, I can't go any further without a solid plan, I started modeling everything I could. Here is just one axis I am working on modeling. (Work in progress)




    Future considerations

    Given that I have a transformer running at half of it what it is capable (48VDC,) I want to upgrade the control board to some sort of Gecko board. I recall they can take up to 80VDC. Perhaps with easy connect/disconnects. Money is no longer as much an issue for the build. This seems like a reasonable upgrade, and i'll have an extra control board to build a smaller machine.

    I feel the calling to change to some shielded wire, and with higher gage. I'm not sure what is optimal yet, haven't dug into the equation for resistance based on diameter. Any vendor or product suggestions?

    I also need to find a better lubricant - preferably one that is dry. I've heard rumors of mysterious substances that don't are dry, work well, and are clean. Everything sat for so long it started to rust a tiny bit. I've been using some spray on white lithium grease so far in an attempt to protect all the linear motion systems. Unfortunately, it collects a lot of dust (better than rust :P).


    Major Design Decision
    After successfully modeling all the components I have currently, I will explore the various machine configurations. The big decision is how to proceed with the frame design/build from here. I have near complete wood frames built with 2x4's, but I have the resources(bridgeport mills) to finish this machine off right with aluminum extruded frame. Hopefully I get around to calculating the moment arm, and deflection based on extrusion section modulus while considering frame configurations.

    Here are some sketches of the configurations I'd like to consider. Please forgive my HORRIBLE handwriting. I didn't intend sharing this document when i drew it up.



    I think I'll try to quickly finish the machine with the wood frame, and then move to convert the machine to a sturdy 80-20 frame. Yet, due to my huge work range, I question the machine's ability to even be rigid enough to cut anything reliably with a wood frame.

    Either way i'm going to completely model everything, before I build it from here on out (lesson learned.)


    I welcome any advice, suggestions, recommendations, or questions with open arms and ears!
    Would the stars shine if nobody were there to observe them?

  2. #2
    Join Date
    Jul 2009
    Posts
    60

    Further modeling

    I've been working on this model between classes. Hoping to get a solid plan in order before moving any further with the physical parts. The price of 2x4s is so low, and since i'm so far along with the wood frame concept, I'll do that to keep moving.

    Oops! I just realized i need to go back and correct the Z-axis. It is upside down. I didn't notice because i did not attach the Nema-23 motor to the z-axis assembly. I've got all 3 axis moving in solidworks in relation to the ballscrew. I may upload animation video of this, as I make more progress.

    Solidworks says this frame weighs approximately 137 pounds. I imagine this is a decent calculation if but a little low. Unfortunately, solidworks does not have an option for douglas fir by default, so i went with pine. I may make this material correction later. It indicated each axis weighs roughly 40 pounds, which I believe is a fairly decent approximation since I've had to lift these before in the shop.

    Things to do:
    - Reorient Z-axis in model
    - Add holes in various plates in model
    - Model motor mounts
    - Start modeling 80-20 frame
    - Select spindle or router


    Spindle Selection Suggestions?
    I am concerned with the noise that the spindle will make and would like to keep about the sound of a normal bridgeport mill. I would like to get a spindle that can cut aluminum/steel metal, wood, and plastic, and accept cutters as large as 1/2". Preferably greater than 1HP. I do not have access to 3 Phase. I'd be happy to learn how to use a VFD to convert to 3phase and use something beefy.

    I've been eyeing the chinese spindles at Kelinginc I'd likely get a high quality hitachi VFD. Then there is the super-pid chip thing that can make cutting at lower RPM's possible on a router...

    Any recommendations?
    Attached Thumbnails Attached Thumbnails Wood Frame Revision1.jpg   Wood Frame Revision 1 - picture 2.jpg   Wood Frame Revision 1 - picture 3.jpg   Wood Frame Revision 1 - picture 4.jpg  

    Would the stars shine if nobody were there to observe them?

  3. #3
    Join Date
    Apr 2009
    Posts
    165
    Please I beg you :nono:,,,,Do not build that out of 8020!!!!!!!!!!

    Those NICE linear bearings and ballscrews are begging for some real metal.

    Trust me 8020 will do nothng but flex and cause you pain. You will end up doing a rebuild. If it were me I would look at steel and or epoxy granite. I actually built my router out of 8020 and wish I had used something even much more rigid than that! I am currently filling all of my extrusions with sand and epoxy to help with some issues I have now.

    In fact if you insist just please send all of those linear bearings and I will put them togood use

    Sorry for the rant!! Just my opinion, Im sure other will say Im crazy.

  4. #4
    Join Date
    Jan 2005
    Posts
    15362
    nickswimsfast

    Your machine is a different build to mine, but take a look at another concept of fabrication, You might change your way of thinking, of how you could put your machine together
    http://www.cnczone.com/forums/cnc_wo...ld_friend.html
    Mactec54

  5. #5
    Join Date
    Apr 2007
    Posts
    1955
    Hi, one of the challenges of working with so many material systems is that some really need to be cut dry, and others really benefit from lubricants. Of course, the cutting speed aspect you understand.

    I like that you are modeling and building up with wood - I am doing a wood mock up as well.

    I personally like fixed gantry designs, due to the rigidity. If you have the space, it might be an option to consider.

  6. #6
    Join Date
    Aug 2008
    Posts
    1166
    I'd agree with harryn on the fixed gantry suggestion if it matches up with the space you have. A lot of the moving gantry machines work well with smaller z axis travels, but it looks like you want to have a z travel about half what your x and y are. This is a work envelope that's scaled more like a milling machine, and that's what your sketches look like as well. If that's the direction you're going, I'd definitely look at the fixed gantry. You might look at the Mori Seiki NV6000 or similar mills for inspiration on frame shape (search on youtube - they show the structure of the machine).

    If you're going to be cutting steel and expect any kind of decent cut or material removal rate, you will want a much stiffer machine. Required stiffness to cut material with good performance corresponds roughly to the modulus of elasticity of the material, so while my router has a stiffness between the bed and spindle nose of about 2000 lbf/in, you'd really like more like 3k-12k, for aluminum you'd like something in the 20k-80k lbf/in range and for steel more like 60k-230k lbf/in (depending on exactly what your goals are). That's not to say you can't cut metals at all with more flexible machines. I've cut aluminum and steel on my router, and it works. It's just slow and I get a bad surface finish.

    On the noise, if you have a high speed spindle, the cutter will still create some significant noise. When I ran my router in a detached garage, the router running by itself was not really that audible but as soon as it started cutting you could hear it. If you want it to be as quiet as a Bridgeport, you might look into a lower speed spindle although this will limit how well it would work with wood. Small scale versions of this could be using a mini mill head. You can buy these from littlemachineshop.com among other places. I have one with R8 collets that I'm working on putting a DC motor on for doing drilling and milling on my router. Otherwise you could look at enclosing the machine in a noise isolation box.

    One more thought - usually you don't want to drill with a high speed spindle meant for wood. You want to limit your plunging to about a 10 degree angle iirc. You can look up data sheets from Columbo to see the kind of usage they recommend for their spindles, for example. So while this can work - you can spiral down into holes, etc., it might not be the best idea if you want to do a lot of milling work where you might need to plunge straight into metal. The milling head, on the other hand, can do this kind of cutting and is why I'm planning on using one as a drilling head.
    CNC mill build thread: http://www.cnczone.com/forums/vertical_mill_lathe_project_log/110305-gantry_mill.html

  7. #7
    Join Date
    Jul 2009
    Posts
    60

    Thank you for your insight

    From tskguy and mactec54, I've begun a new machine design using 1/2" mic 6 plate, 45x90Heavy and 45x45Heavy aluminum bosch extrusions. (attached) As apparent from the design i've added several internal members to stiffen the base. If I do go with this design, I will be milling the extrusions flat on a bridgeport mill.

    The 5x mic6 plates alone cost about 300$ shipped. Eeep :P It was originally 3/4" but to cut cost i've reduced that. This doesn't even include any of the material for the Fixed Gantry design. Once I complete the design, i'll cost it all out to get a rough idea, and see if I should change the design for cost. Under a $1000 USD for the frame would be nice seeing as this will be my first machine, and I do expect to make some sort of mistake. It's the $$$ vs. stiffness battle!

    -------
    Harryn,

    Cutting fluids is a great consideration, I will keep in mind the option of an enclosure for both particulate containment, and fluids. I'm thinking an enclosed table (walled) table would make for a simple solution. I'll try to incorporate it into the design. I'll have to ensure it doesn't bump into the fixed gantry.

    -----------

    jsheerin,
    I'm convinced. Fixed gantry is the way to go for a machine such as this. [nomedia="http://www.youtube.com/watch?v=5IvFBngafFs"]YouTube - Mori Seiki NV6000DCG Vertical Machining Center[/nomedia] I looked it up as you suggested. It will be a fixed overhead gantry design similar in concept to the Mori Seiki.

    Machine stiffness - can I calculate this with FEA? I know how to roughly calculate the displacement, and Stress with von mises (example attached). Can either of these values be equivocated to the stiffness? Where displacement represented in inches, and Stress in lbf/in^2. Once I complete the mic6/8020 frame I'll compare it to the wood frame I've modeled. If I can compare it to your ballpark stiffness figures that'd be even better!

    Spindle noise - Still on the fence on this one. Seems such a shame to spend so much effort into designing a rigid machine then, putting a Taig or X3 spindle on it.

    A quick look at the KL-2200 says "Speed: 8000-24000 R/min" I'm not sure I understand it fully, but I guess that means it would take it out of the power curve likely optimized for those RPMs... Can you not run one of the high speed chinese spindles at lower RPM's to cut steel with a VFD?


    --------
    Finally, thank you for your input!
    Attached Thumbnails Attached Thumbnails mic6 and 8020.jpg   Sample FEA.jpg  
    Would the stars shine if nobody were there to observe them?

  8. #8
    Join Date
    Aug 2008
    Posts
    1166
    Quote Originally Posted by nickswimsfast View Post
    Machine stiffness - can I calculate this with FEA? I know how to roughly
    ...
    Can you not run one of the high speed chinese spindles at lower RPM's to cut steel with a VFD?
    Yes - divide the load applied by the displacement at the point you're interested in. I apply a load on the spindle, constrain the vice jaw, and look at the spindle nose displacement when I calculate k. If you actually model the stiffness of the bearings, this gets much more complicated but if you can figure out how to do it, it definitely helps design a better machine. My milling machine design has progressed quite a bit from where I started. In addition to stiffness, you can look at stress to see where you need more material or where you can remove material.

    On the spindles, I'd be worried that higher cutting forces of steel would lead to faster destruction of the spindle's bearings. Also if you have a fan cooled spindle, running it slower can overheat it. If you have a water cooled spindle, in theory you could control the temperature, but I'd still be a bit leary of doing it.
    CNC mill build thread: http://www.cnczone.com/forums/vertical_mill_lathe_project_log/110305-gantry_mill.html

  9. #9
    Join Date
    Jul 2009
    Posts
    60
    FEA Conclusions - Wood Design
    According to Jsheerin's numbers, it's not very rigid! Not bad despite the fact it's made of wood! The numbers should be taken at weak face value, because of the extravagant number of assumptions. It will more be used as a basis for comparison against future alternative designs. It is useful to see that it is nearly 3 times as rigid in the x-direction vs. the y. I need to strengthen the Y-direction (what is weakest in head based design.) It is useful to see the stress concentrations on the design so i know where to strengthen too!

    Quote Originally Posted by jsheerin
    Required stiffness to cut material with good performance corresponds roughly to the modulus of elasticity of the material, so while my router has a stiffness between the bed and spindle nose of about 2000 lbf/in, you'd really like more like 3k-12k, for aluminum you'd like something in the 20k-80k lbf/in range and for steel more like 60k-230k lbf/in (depending on exactly what your goals are). That's not to say you can't cut metals at all with more flexible machines. I've cut aluminum and steel on my router, and it works. It's just slow and I get a bad surface finish.
    --------------------------
    FEA Results - Pictures attached.
    Calculation 1 - (based on 200 pound load in y-direction) shows stiffness of 2777 lbf/in with the wood frame in the y-direction.

    Calculation 2 - (based on 200 pound load in x-direction) shows stiffness of 892 lbf/in.

    Calculation 3 - (based on 750 pound load in x-direction) shows stiffness of 892 lbf/in.

    ------------------------
    Assumptions
    I had to make significant assumptions to even produce results:

    • Wood is not precisely defined as metals in solidworks so i've attached a picture with the parameters i used to make this approximation. Most from google lookups for poisson ratio, and modulus of elasticity
    • I had to remove several machine elements from the solid model in order to combine it into one contiguous model. I can only FEA analyze "1 Part" based on the version of solidworks i have. The frame is assumed to be one big piece, so the fasteners, hole stress concentrations were not evaluated. I had to remove 2 wood cross members from the design in order to combine it into one part.
    • Spindle element is removed, all linear motion components have been removed.
    • etc. etc.. lots of assumptions weak analysis.


    ---------------------
    Special thanks to Jsheerin for the information and advice!
    Attached Thumbnails Attached Thumbnails deflection 200 lbf in x-direction.jpg   stress 200 lbf in x-direction.jpg   deflection 200 lbf in y-direction.jpg   stress 200 lbf in y-direction.jpg  

    deflection 750 lbf in y-direction.jpg   stress 750 lbf in y-direction.jpg   fine mesh density.jpg   removed members zero-thickness geometry.jpg  

    wood parameters.png  
    Would the stars shine if nobody were there to observe them?

  10. #10
    Join Date
    Aug 2008
    Posts
    1166
    Nice work. Yeah, sometimes the hardest part is figuring out how to simplify things and still get usable results. For your purposes, removing the linear bearings is probably not that bad as they're probably a lot stiffer than the rest of the frame. On my mill, it's the other way around - the stiffness of the bearings is a big contributor to the overall stiffness of the machine. But you do know that adding all the linear motion parts will make the stiffness go down, so if you have a target and you're below it without all the linear motion parts, you know you need to make changes.
    CNC mill build thread: http://www.cnczone.com/forums/vertical_mill_lathe_project_log/110305-gantry_mill.html

  11. #11
    Join Date
    May 2010
    Posts
    307
    Glad to see some FEA work done. I would be interested in seeing how a more accurate loading would effect the results i.e. a load at the end of a mock spindle.

    Ideally you would make the spindle much stiffer than the frame but I realize you are limited to one material.

  12. #12
    Join Date
    Apr 2009
    Posts
    5516
    nickswimfast, your mill looks like a cool design. I don't have a clue how to model things in solidworks, let alone do those cool stress analysis things!

    I will say, however, that you could probably make your z 'column' stonger just by adding plwood ribs inside the column and skinning the outside with plywood... unless there's a reason you;re keeping them open...

  13. #13
    Join Date
    Jul 2009
    Posts
    60
    jsheerin, you make an important point with the stiffness of the linear motion components. I am assuming the same bearing elements for any frame design, as such i think it is fair to completely remove them from this analysis. This might not give an accurate representation of what the actual stiffness value is, but it will allow me to consistently compare designs.

    ---------------------------------------
    New Frame Design!

    Well I've been reading up on giz's design, as well as adapting basic concepts from the mori seiki machine. I'm sure this design will be more rigid, I've utilized only mic6 aluminum, and 2.5"x2.5"x0.125" steel members for this frame design. As such, this frame is likely very expensive, I'll post the cost when i price it out. I suspect it is somewhere in the 500-1000$ range for frame materials alone.

    I'm more interested in the rigidity of this design, so there are a few details I'd need to clean up if this was selected to be the final design. I was thinking of two sayings when making this, "when in doubt, make it stout" and "Go big or go home."

    This frame weighs 426 lbs according to solidworks, and this is not including welds or fasteners. I'm attempting to make the upper portion of the frame detachable with bolts so I can take the machine apart into two sections. (attempt to make it easier to move.)

    Design Concern
    My only concern with this design is mounting the rails to the nonuniform cold rolled steel. (Just as is heavily discussed in giz's, and many other similar designs.) My idea is to utilize 1/2" thick mic 6, and bolt it directly to the steel frame in 3 locations. Thoughts on doing this?

    I'll attempt to do some FEA on this sometime in the near future, and post the results. I expect this frame will be much more rigid than the wood frame. I'm hoping to hit higher rigidity numbers closer to what jsheerin posted before. I'm getting the itch to get my hands dirty with cutting oil or welding sparks so I hope to settle on a design soon.
    Attached Thumbnails Attached Thumbnails Complete frame.jpg   front view.jpg   mic6 parts for X-axis.jpg   Removable Z-axis frame.jpg  

    X-Y table.jpg   Z-axis.jpg  
    Would the stars shine if nobody were there to observe them?

  14. #14
    Join Date
    Aug 2008
    Posts
    1166
    I'd suggest rethinking your bearing mounting on the lower frame. You've thrown away a lot of the stiffness of the steel beams by raising the mounting surface of the rails up off the steel and only attaching the two in 3 places per side. The mounting surfaces should be attached directly to the steel at many points and supported over the entire surface. Same for the bearing block attachment for the screw - that is where all the reaction forces get channeled through from pushing your carriage around. You want it mounted solidly to a big structure.

    Also you might want to triangulate something in the y direction (across the gantry). Right now your beams are all at right angles to each other. While welded joints are stiffer than theoretical pinned joints, the structure will still tend to fold up on a small scale.

    After that the next problem you might see is that your base is flexible compared to the gantry - if you push down in the middle it will bend down. But this may or may not be a problem given your stiffness goal.
    CNC mill build thread: http://www.cnczone.com/forums/vertical_mill_lathe_project_log/110305-gantry_mill.html

  15. #15
    Join Date
    Apr 2009
    Posts
    5516
    I can't find the link, but there was a guy who built a huge 4x8 and use steel epoxy to true the mating surfaces before bolting everything together... But maybe you can set your tooling plate surfaces over such epoxy, then mechanically set them in place with screws/bolts...

  16. #16
    Join Date
    Jul 2009
    Posts
    60
    Quote Originally Posted by jsheerin View Post
    I'd suggest rethinking your bearing mounting on the lower frame. You've thrown away a lot of the stiffness of the steel beams by raising the mounting surface of the rails up off the steel and only attaching the two in 3 places per side. The mounting surfaces should be attached directly to the steel at many points and supported over the entire surface. Same for the bearing block attachment for the screw - that is where all the reaction forces get channeled through from pushing your carriage around. You want it mounted solidly to a big structure.

    Also you might want to triangulate something in the y direction (across the gantry). Right now your beams are all at right angles to each other. While welded joints are stiffer than theoretical pinned joints, the structure will still tend to fold up on a small scale.

    After that the next problem you might see is that your base is flexible compared to the gantry - if you push down in the middle it will bend down. But this may or may not be a problem given your stiffness goal.
    Your comments concerning rigidity make complete sense. As far as the mic6/rails - I rationalized my way out of it due to the rail to frame mounting issue. I also was intending to add gussets to the vertical members, just haven't gotten around to it.

    jsheerin, Would you recommend bolting the mic6 directly to the steel and then the rails to the mic6? Just doesn't seem there is a clean solution on this one. Doing so could have dimensional issues as the steel won't be truly straight, but I guess some sort of shimming could be instituted. Such a conundrum, that has me going in circles.

    I'd like to draw my answer based on what other people have successfully done. What have you done in the past in this situation? What have you seen be successful?

    If this gets out of hand, I may resort to the KISS principle (Keep it simple stupid) and return to 8020, and simplify the design at the cost of rigidity. Thoughts?
    Would the stars shine if nobody were there to observe them?

  17. #17
    Join Date
    Aug 2008
    Posts
    1166
    I haven't done this yet. My router uses the CRS rail / cncrouterparts carriage setup. I'm working on converting it to THK rails, but I haven't figured out if I want to rebuild the entire frame or try to stick with 80/20. So far I've just modeled the current frame and confirmed it's not very stiff.

    I think in the build Louie is referring to, the guy had access to a large surface plate and he used that to set the frame upside down on with filled epoxy between the plate and the frame to create flat rail mounting surfaces. That would be a cool way to go but requires the large plate.

    If you want to go all out, use all steel, mount the rails to the steel, maybe bolt on some thicker plates on top of the tubes, and then mill / grind / scrape them flat. The problem is milling requires a large machine, grinding you could do yourself by pouring an epoxy surface plate (or surfaces next to your rail mounting spots and setting up a temporary carriage to grind your own mounting surfaces, and scraping you need a reference straight edge and / or surface plate. For my mill build I plan to scrape. I bought 4 precision straight edges a few weeks ago. The longest is 6' long. However, once again it depends on your goals - maybe all of the above are overkill. Maybe just pouring an epoxy surface plate and mounting your rails on the epoxy is good enough. I don't think I've seen anyone actually do this though and I have various objections to it. You'd still need a horizontal reference to align the side of your rail to as well. You could make that via an epoxy surface plate as well - pour it flat and then sit it at a 90 degree angle to the first plate. Or how about this - epoxy surface plates surrounding your rail mounting surfaces. Then set a piece of mic6 or better a small surface plate across the epoxy surface plates, covering your rail mounting surface but elevated slightly above it. Fill the area between the plate and the frame with a steel-filled epoxy. Once cured, remove the plate, move down and do it again until you have a bed for the entire rail. You could then do something similar to create the horizontal rail reference surface. So this would be similar to what Louie was suggesting but you would create your own surface plate on the machine instead of needing access to a really big one (and having to move the machine to it, etc.).
    CNC mill build thread: http://www.cnczone.com/forums/vertical_mill_lathe_project_log/110305-gantry_mill.html

  18. #18
    Join Date
    Apr 2009
    Posts
    5516
    Quote Originally Posted by jsheerin View Post
    I haven't done this yet. My router uses the CRS rail / cncrouterparts carriage setup. I'm working on converting it to THK rails, but I haven't figured out if I want to rebuild the entire frame or try to stick with 80/20. So far I've just modeled the current frame and confirmed it's not very stiff.

    I think in the build Louie is referring to, the guy had access to a large surface plate and he used that to set the frame upside down on with filled epoxy between the plate and the frame to create flat rail mounting surfaces. That would be a cool way to go but requires the large plate.

    If you want to go all out, use all steel, mount the rails to the steel, maybe bolt on some thicker plates on top of the tubes, and then mill / grind / scrape them flat. The problem is milling requires a large machine, grinding you could do yourself by pouring an epoxy surface plate (or surfaces next to your rail mounting spots and setting up a temporary carriage to grind your own mounting surfaces, and scraping you need a reference straight edge and / or surface plate. For my mill build I plan to scrape. I bought 4 precision straight edges a few weeks ago. The longest is 6' long. However, once again it depends on your goals - maybe all of the above are overkill. Maybe just pouring an epoxy surface plate and mounting your rails on the epoxy is good enough. I don't think I've seen anyone actually do this though and I have various objections to it. You'd still need a horizontal reference to align the side of your rail to as well. You could make that via an epoxy surface plate as well - pour it flat and then sit it at a 90 degree angle to the first plate. Or how about this - epoxy surface plates surrounding your rail mounting surfaces. Then set a piece of mic6 or better a small surface plate across the epoxy surface plates, covering your rail mounting surface but elevated slightly above it. Fill the area between the plate and the frame with a steel-filled epoxy. Once cured, remove the plate, move down and do it again until you have a bed for the entire rail. You could then do something similar to create the horizontal rail reference surface. So this would be similar to what Louie was suggesting but you would create your own surface plate on the machine instead of needing access to a really big one (and having to move the machine to it, etc.).
    I still can't find the link! But I thought of another way to level the rails. Get a piece of Mic6 that will fit between them. Then make a base for the router out of Mic6 that extends about 12 inches on one side, and place a couple 5 pound dumbbell weights on it as a counerbalance. Put a 1/2" 4-flute bit in and set it to take about .005" at a time, set the router speed the slowest it will go , and slowly skim away. This way, since you're referencing off the plate, both rail surfaces should be normal to each other. You might even be able to Bondo the rails, and use the router to skim the Bondo down to where it just covers the steel. This will only work if your machine is in place and has no way to flex (maybe bolted down to a concrete pad?) Maybe even use rectangular tubing for the base...

    But as jsheerin has mentioned you will still have some frame flex, and at the very least you should add gussets to all corners. I would tack weld the frame, then weld the gussets, then complete the tube welds last, because welding all that mild steel will cause some warping.

  19. #19
    Join Date
    May 2010
    Posts
    307

  20. #20
    Join Date
    Apr 2009
    Posts
    5516
    That's the dude! Check out the surface finish he gets on 2024 aluminum... awesome build!

    As for Haydn, his is one of the first DIY CNC vids I saw on YouTube... another awesome build...

Page 1 of 4 123

Similar Threads

  1. Build Log: RMGVideos XZero Raptor Build
    By sagreen in forum DIY CNC Router Table Machines
    Replies: 150
    Last Post: 05-10-2011, 09:00 PM
  2. Newbie - To build or not to build Router/Plasma Table
    By dfranks in forum Waterjet General Topics
    Replies: 10
    Last Post: 04-08-2011, 05:16 AM
  3. Joe's 4x4 hybrid build/2006 build
    By krap101 in forum DIY CNC Router Table Machines
    Replies: 1
    Last Post: 08-09-2010, 11:30 AM
  4. New Large Table Build in Houston, TX (Build Log)
    By anitel in forum Plasma, EDM / Other similar machine Project Log
    Replies: 12
    Last Post: 12-30-2008, 09:45 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
  •