CNC Router for Hardwoods: Evaluation and Questions


Page 1 of 12 123411 ... LastLast
Results 1 to 12 of 142

Thread: CNC Router for Hardwoods: Evaluation and Questions

  1. #1
    Registered
    Join Date
    Aug 2013
    Posts
    50
    Downloads
    0
    Uploads
    0

    Default CNC Router for Hardwoods: Evaluation and Questions

    Hello everyone,

    I am a furniture designer/maker who makes sculptural pieces of furniture out of solid wood. I have no experience with CNC, but I am sure that I will be able to incorporate one into my one-of-a-kind work flow. I am very excited about the whole process!

    I have been developing a design for a machine that synthesizes my needs with what I have learned from other posts. I know enough now to know that I know very little. Any comments or guidance would be greatly appreciated.

    The overview of the machine is: welded steel base, 8020 extrusion, plate aluminum cars, Hiwin linear rail, R&P, 4KW Chinese spindle with VFD and Nema 34 steppers…(all subject to your review). I am approaching the build in a few stages that will be spread out over the coming year.

    Clear space in my studio and evaluate size
    Weld the base
    x & y rails and cars
    z car
    steppers & spindle
    Electronics
    Learn how to use it!

    I am at the point where the various components are starting to work as a system in my CAD drawing. Thats the good part…the bad part is I am now seeing flaws in my plan and other areas of my inexperience are totally exposed! I have some specific questions in regard to sizing Hiwin blocks and in regard to setting up the Z axis. I will do my best to walk you through my design which will explain the thought behind some of my decisions. I am open to any and all suggestions that help me reach my goal of a capable CNC router for hardwoods. If the machine is rigid enough I could easily find some aluminum projects for the machine.

    CNC Router for Hardwoods: Evaluation and Questions-base-jpg

    Base: The size of the base is the big issue for me. My studio is a fixed size with no hope of expansion. It would be ideal to have a machine that could accommodate sheet goods, but it is far from mandatory. In reality 30” of cutting width would be fabulous for my workflow. The long axis is more important to me and cutting 8 plus feet will be very useful. I mention the size because the design of the machine tries to maximize the cutting with of a narrow machine. I should point out that my machine will keep the top of the 80/20 rails below 34”. I realize that this is quite short short, but it affords me the freedom of allowing many of my other machines to pass material over the cnc.

    The base will be welded out of 14g steel with levelers. I can’t expect the base to be either flat or square enough for my taste. I plan on using the aluminum plates as a transition to the 80/20 so that I can either flatten or shim with the help of my 6’ Starrett straight edge and machinists level.

    CNC Router for Hardwoods: Evaluation and Questions-spoil-board-jpg

    Spoil Board: I plan on having three layers of 3/4” material. The cut out in the front of the machine is for vertical work. I plan on hinging the plate so that it can angle to 45°. This will mimic some of the functionality of my multirouter but add CNC awesomeness! I should probably remove the bar in front of that part of the table but I am protecting my dream of supporting a full 4 x 8 sheet.

    CNC Router for Hardwoods: Evaluation and Questions-y-axis-detail1-jpg

    CNC Router for Hardwoods: Evaluation and Questions-y-axis-detail2-jpg

    Y-Axis:

    The rails will be made from 2040 series 10 80/20 extrusion attached from below and attached to the short sections of angle iron (to help reduce the tendency to rack). The Y cars are made of 3/4” aluminum with some of the vertical components made from 1/2”. I have designed the cars to allow for maximum movement in the x direction, a concession to the limited space in my shop. I have read enough about the Hiwin rails to know that I will be limited in terms of what legacy blocks I can actually find. I am curious if these blocks are adequate? LGW20HBZ1H blocks and rail. If I cant find them can I use the standard precision? What about the preload? For some reason I decided that Z1 appropriate?

    CNC Router for Hardwoods: Evaluation and Questions-x-axis-detail1-jpg

    CNC Router for Hardwoods: Evaluation and Questions-x-axis-detail2-jpg

    X-Axis
    3060 series 15 extrusion is at the core of the x axis. The same hiwin block and rail as the y axis are mounted on the top and bottom of the extrusion. Same questions about preload, precision and size?The spacing of the blocks also tries to maximize the reach of the router in the x direction.

    CNC Router for Hardwoods: Evaluation and Questions-z-axis-detail2-jpg

    CNC Router for Hardwoods: Evaluation and Questions-z-axis-detail1-jpg

    Z-Axis
    This is where my plan kind of unravels. All of my efforts to maximize the x axis movement crash up against the reality of clearing the spindle and my lack of experience. I have many questions: What is the relationship of the bottom of the spindle to the bottom of the Z plate? My design has the blocks permanently mounted and the stepper, ballscrew rails and plate moving. How high up should the spindle be able to go? I would like to get the bottom of the collet in line with the bottom of the x car. The answer to the first question effects how high I need to move the bottom hiwin block. If I try to achieve this I end up with the spindle hanging way below the bottom Hiwin block. Am I going for too much movement? Am I going about this in the wrong way? Perhaps I should explore pre-made Z options?

    Here are the images that illustrate some of my many Z axis questions:


    CNC Router for Hardwoods: Evaluation and Questions-question1-jpg

    If I line up the mounting holes on the spindle with the bottom of the z plate I will have to move the bottom hiwin block up much higher.
    CNC Router for Hardwoods: Evaluation and Questions-question4-jpg

    moving the blocks up will result in quite the overhang of the spindle. I have been looking at other machines and it seems that this sort of overhand is not unheard of.
    CNC Router for Hardwoods: Evaluation and Questions-question5-jpg

    Many thanks for having a look at my plan.

    Similar Threads:
    Attached Thumbnails Attached Thumbnails CNC Router for Hardwoods: Evaluation and Questions-y-axis-jpg  


  2. #2
    Gold Member
    Join Date
    Apr 2004
    Location
    Oakland CA USA
    Posts
    4189
    Downloads
    0
    Uploads
    0

    Default

    [QUOTE=Sapele;1405722]\

    The overview of the machine is: welded steel base, 8020 extrusion, plate aluminum cars, Hiwin linear rail, R&P, 4KW Chinese spindle with VFD and Nema 34 steppers…(all subject to your review).

    [The welded steel base won't be flat, but you knew that, right? To run those linear rails, though, they need to be mounted on something very flat, so you can either attach something flat to it or cast some epoxy to make a flat surface. Some people seem pleased with their Chinese spindles, fewer of them with the VFDs they come with. I'd suggest a Hitachi, but make sure it can make enough 3-phase power for the spindle from whatever power you're giving it. If you're feeding it single-phase current, it will need more going in than it will put out. The steppers will work, but wood wants to be cut fast, so don't rule out servos at this stage.]

    Clear space in my studio and evaluate size
    Weld the base
    x & y rails and cars
    z car
    steppers & spindle
    Electronics
    Learn how to use it!

    [I'd say put that last item at the top of your list. As you get into learning 3D CAD modeling, you'll get a better idea of what you want the machine to do. For instance, this machine is entirely dedicated to 3-axis carving at this point, but you might decide you need a 4th axis to do the designs you find yourself making. If so, thinking about that at the outset would be good. You might want to leave a space open to mount a 4th axis centered at table height, since that will double the diameter that can be accommodated under the spindle.]



    Base: The size of the base is the big issue for me. My studio is a fixed size with no hope of expansion. It would be ideal to have a machine that could accommodate sheet goods, but it is far from mandatory. In reality 30” of cutting width would be fabulous for my workflow. The long axis is more important to me and cutting 8 plus feet will be very useful. I mention the size because the design of the machine tries to maximize the cutting with of a narrow machine. I should point out that my machine will keep the top of the 80/20 rails below 34”. I realize that this is quite short short, but it affords me the freedom of allowing many of my other machines to pass material over the cnc.

    [Consider making it even shorter. If you run your gantry on raised "walls" (like the Mecmate design) it will be stiffer than if it's raised up on wobbly verticals. The walls can be as heavy as they want, since they aren't a moving part. There's no particular need for the router to be at table height, but you can put a table top over the walls (when it's not being used) if you need it to function as an outfeed table.]


    Spoil Board: I plan on having three layers of 3/4” material. The cut out in the front of the machine is for vertical work. I plan on hinging the plate so that it can angle to 45°. This will mimic some of the functionality of my multirouter but add CNC awesomeness! I should probably remove the bar in front of that part of the table but I am protecting my dream of supporting a full 4 x 8 sheet.

    [You can always attach extensions to the table to support extra material. I like the idea of the cut-out for vertical work, but that will only work for short pieces (or very short pieces, if you take my suggestion above.)]



    Z-Axis
    This is where my plan kind of unravels. All of my efforts to maximize the x axis movement crash up against the reality of clearing the spindle and my lack of experience. I have many questions: What is the relationship of the bottom of the spindle to the bottom of the Z plate?

    [Your shortest tool still has to be the lowest point, or the Z-plate won't clear the material.]

    My design has the blocks permanently mounted and the stepper, ballscrew rails and plate moving.

    [That's pretty standard, although you can reverse that and have the blocks move while the rails and screw are stationary.]

    o How high up should the spindle be able to go? I would like to get the bottom of the collet in line with the bottom of the x car. The answer to the first question effects how high I need to move the bottom hiwin block. If I try to achieve this I end up with the spindle hanging way below the bottom Hiwin block. Am I going for too much movement? Am I going about this in the wrong way?

    [You just have to make sure that the collet is the lowest point, but can retreat so that it clears the work, or you're wasting headroom.]

    Perhaps I should explore pre-made Z options?

    [I've used automation actuators for this; they've worked pretty well. If the spindle is heavy, you might find it drops when you turn off the power. A gas spring or two can help with that. Since you're making a rack & pinion design (at least in part), take a look at the kits offered by CNC Router Parts CNCRouterParts They've figured out a lot of the things you're still wondering about, or haven't got around to wondering about yet.]

    Andrew Werby
    www.computersculpture.com



  3. #3
    Registered
    Join Date
    Aug 2013
    Posts
    50
    Downloads
    0
    Uploads
    0

    Default

    thanks for the review awerby

    I dont want to neglect you point about learning to use the machine so I will address it first. You are absolutely correct in your suggestion- a normal person would start by learning how to use a machine and then go down the rabbit hole of making one. Unfortunately I know that I have to jump in in order to learn. This is far from ideal and I am sure that I will have regrets that will manifest in future revisions of the machine. The fourth axis is a good example of a feature that is not that relevant to my work, but I might want to implement at some point so I should keep the option open. Great point!

    Thanks for the suggestions of servos and the hitachi VFD. I am open to both and will do some further research.

    I see what you mean about the mechmate design. It does look quite robust. I am going to have to think about that design in terms of my needs. It would further reduce the limited vertical capability of the machine. I am looking forward to this ability so I might be stuck trying to make the wobbly parts as rigid as possible.

    I was working on a new version of the machine before your response. This represents a more robust version of what I had with the added benefit of being easier to epoxy level. I will dam up and bridge the two 4" tubes together so tha I can level the top. Then I will mount the 2040 extrusions with the hiwin rail.

    CNC Router for Hardwoods: Evaluation and Questions-base2-jpg

    CNC Router for Hardwoods: Evaluation and Questions-base2-left-jpg

    CNC Router for Hardwoods: Evaluation and Questions-base2-front-jpg

    CNC Router for Hardwoods: Evaluation and Questions-base4-jpg

    lastly a quick Hiwin question: I think I have seen some people use 4 of the AGH15SAZ1H for the Z axis. Are these adequate? I plan on using long block LGW20HBZ1H for the x and y. I am having trouble finding square block legacy hiwins with a Z1 preload so my options might be limited.

    thanks



  4. #4
    Community Moderator ger21's Avatar
    Join Date
    Mar 2003
    Location
    United States
    Posts
    30725
    Downloads
    0
    Uploads
    0

    Default

    My design has the blocks permanently mounted and the stepper, ballscrew rails and plate moving.
    I don't like this, and don't think I've seen any machines with the steppers moving. Lifting the screw and stepper means you're moving a lot more mass, which means you'll need a lot more power to lift it rapidly. On a lot of home built machines, the Z axis is the limiting factor in the machine's performance.

    As far as Z travel goes. For me, I want to be able to reach the table with a 1" long tool, and if I have a 3" tool in the spindle, I want it to clear the bottom of the gantry when retracted.


    I think I have seen some people use 4 of the AGH15SAZ1H for the Z axis. Are these adequate?
    Yes, they should be more than adequate. However, careful Ebay watching can get you better quality pre-owned THK's for the same or less, if you're patient.

    I plan on using long block LGW20HBZ1H for the x and y. I am having trouble finding square block legacy hiwins with a Z1 preload so my options might be limited.
    Good luck. I have 8 LGW20's that I bought when the supply was very low, over a year ago. I still need to grab my 10' rails.

    Gerry

    UCCNC 2017 Screenset
    http://www.thecncwoodworker.com/2017.html

    Mach3 2010 Screenset
    http://www.thecncwoodworker.com/2010.html

    JointCAM - CNC Dovetails & Box Joints
    http://www.g-forcecnc.com/jointcam.html

    (Note: The opinions expressed in this post are my own and are not necessarily those of CNCzone and its management)


  5. #5
    Registered
    Join Date
    Aug 2013
    Posts
    50
    Downloads
    0
    Uploads
    0

    Default

    ger21 you have me wondering if the LGW20 are in stock or if they are just on the sellers website! Thanks for the confirmation that the AGH15SAZ1H would be adequate. I will continue the search. I have not been looking at the THK blocks so perhaps I will be able to find something since I have the time to search.

    I will re work the z axis with the blocks and rails flipped. I am curious to see which way results in better support for the spindle. I understand that there is more to the comparison when one takes into account the weight of the screw and stepper.

    thanks



  6. #6
    Gold Member
    Join Date
    Apr 2009
    Location
    USA
    Posts
    5261
    Downloads
    0
    Uploads
    0

    Default

    I normally don't see any big problems having the stepper on the carriage itself, provided it's sized right for the job. Usually the limiting factor with homemade machines is that finer pitch screws seem to be (I believe unnecessarily) used on the Z, along with wussing out on the power supply. The X and Y steppers move far more weight, and the stepper itself is pretty light in relation to the other componnents on the Z carriage.

    As to the bearing block arrangement on the Z, my belief is that if your overall Z travel is going to be small, say under 4", it would be best to mount the rails onto the Y saddle and the blocks on the carriage. If you need more Z travel, it would be best to mount the rails onto the carriage and have the bearing blocks fixed on the saddle. The latter would require more clearance below the gantry beam, but I believe it's a lot easier to beef up the gantry uprights than to add gussets and other supports to add to the weight of the carriage. There will always be a tradeoff, but one compromise is to mount your spindle on a plate with multiple index holes, and corresponding holes on the carriage plate. This way you can mount your spindle to suit the job at hand. I use a lot of stub endmills (<1/2" flute length) and my machine is set up where I can get the collet within 1/2" of the table. This is a more ridgid setup than having the collet 1" over the table and using a longer bit.



  7. #7
    Registered
    Join Date
    Aug 2013
    Posts
    50
    Downloads
    0
    Uploads
    0

    Default

    Awerby's point about the 4th axis encouraged me to think a bit more about the future use of the machine. The 4th axis is not that high on my list, but making a provision for it made the following clear to me. Having a section of the spoil board that can be removed to reveal a lower table would be very useful to me. This sunken pocket would be the place where I would mount the 4th axis. I will bolt the tube in light blue rather than welding it so that I can cut the spoil board out to reveal the pocket when it is needed.

    CNC Router for Hardwoods: Evaluation and Questions-sixorwindow-25-jpg

    Thanks to everyone for the information on the Z slide. I understand the following.

    If the blocks are mounted on the z carriage then the carriage plate must be beefed up to accommodate for the overhang. The overhang is needed because the bottom block can not extend below the lowest point of the yoke. The plus with this arrangement is that it is much easier to clear the bit over the bottom of the yoke. The downside is that the more Z travel then the greater the overhang which requires further stiffening.

    If the blocks are mounted on the yoke and the rails on the carriage then it becomes easier to reach the spoilboard but harder to clear the bottom of the yoke when the bit is raised all the way up. This situation forces the spindle to be placed on an adjustable mount and/or the y-beam and yoke to be raised. A bonus, as I see it, is that the overhang of the spindle is not fixed. The higher the spindle is raised the better it is supported (I am not sure this is actually true because it changes other forces?). When I think of the 1/2" bit scenario that louieatienza brought up isn't scenario 2 going to be better?

    Scenario 1- short bit with the work resting on the spoilboard

    CNC Router for Hardwoods: Evaluation and Questions-scenario1-jpg

    Scenario 2- short bit with the work supported on a fixture

    CNC Router for Hardwoods: Evaluation and Questions-scenario2-jpg

    I am in the process of thinking about me actual Z travel needs rather than my Z travel desire. The drawing below reflects me Z Desire

    I raised the y beam up almost 2 inches from where I had it and I

    CNC Router for Hardwoods: Evaluation and Questions-sixorwindow-20-jpg

    this view shows the spread of the THK bearings (HSR 20R) and the bearings height above the spoilboard.

    CNC Router for Hardwoods: Evaluation and Questions-sixorwindow-23-jpg

    I have no idea if I am being reasonable with my Z movement? Too much?

    The actual movement in this model would be 9" of travel from the spoilboard.

    Thanks and I hope everyone enjoys the last day of the year!



  8. #8
    Community Moderator ger21's Avatar
    Join Date
    Mar 2003
    Location
    United States
    Posts
    30725
    Downloads
    0
    Uploads
    0

    Default

    I don't know why your mounting your bearings on the gantry like that. There's no reason to have that box surrounding the gantry beam. It actually causes a few issues.
    1) It's adding 1-1/2" to the bottom that you need to account for when calculating clearances.
    2) Having that box sliding from one end to the other makes it difficult to attach anything to the gantry. You typicallt want to use a cable chain, and with that design, you'll need a separate gantry beam or tray to support it

    It's much simpler to mount the rails to the front of the beam. That let's you drop the beam 1-1/2" and still have the same clearance, and reduces your travel requirements.

    As for your Z travel.
    What if you need to cut a sheet of plywood? In your scenario 2, you're not going to build a fixture to support large sheets or large parts.

    Also consider that realistically, you can't really cut much deeper than 3-4" with a 3 axis router. The reason is that once you reach your maximum tool length, you start running into clearance issues. You have to make sure the collet clears, and then the spindle body and carriage itself. I would guess that the max you'd be able to use would be in the 5-6" range, but it could be more in specific situations.
    Having more travel does allow you to make 3" cuts on 10" thick parts, though, if you have that need.
    But I would expect that 6-7" clear is more than enough for almost anything you'd be doing.

    Also consider that you can mount the spindle lower on the plate, and use some of the spindle length for your travel distance.
    On the machine that I'm building, I have the bearing blocks mounted to the spindle plate, and IO have just over 5" of clearance from the bottom of the gantry to the bottom of the collet. This gives me about 6" clearance with a 1" bit reaching the table. With 15" rails, I can left the collet about 4" above the gantry bottom.
    Here are some pics. http://www.cnczone.com/forums/cnc_wo...ml#post1155834

    Gerry

    UCCNC 2017 Screenset
    http://www.thecncwoodworker.com/2017.html

    Mach3 2010 Screenset
    http://www.thecncwoodworker.com/2010.html

    JointCAM - CNC Dovetails & Box Joints
    http://www.g-forcecnc.com/jointcam.html

    (Note: The opinions expressed in this post are my own and are not necessarily those of CNCzone and its management)


  9. #9
    Gold Member
    Join Date
    Apr 2009
    Location
    USA
    Posts
    5261
    Downloads
    0
    Uploads
    0

    Default

    If the blocks are mounted on the z carriage then the carriage plate must be beefed up to accommodate for the overhang. The overhang is needed because the bottom block can not extend below the lowest point of the yoke. The plus with this arrangement is that it is much easier to clear the bit over the bottom of the yoke. The downside is that the more Z travel then the greater the overhang which requires further stiffening.
    The other thing is that you'll have this overhang regardless of Z position.

    The actual movement in this model would be 9" of travel from the spoilboard.
    I don't think that's an unreasonable amount of travel.

    For extrusions, look into Misumi GFS extrusions. Very heavy duty, and they can be purchased with wither short or wide faces milled flat.



  10. #10
    Registered
    Join Date
    Aug 2013
    Posts
    50
    Downloads
    0
    Uploads
    0

    Default

    great points. I had thought abut the e chain mounting issue and I thought it was an acceptable trade off for bringing the spindle 1 1/2" closer to the y beam. The way you suggest is much easier for sure. I will have to re draw it as you suggest and see how it all works together.

    the driving force of the larger Z travel has to do with my expectation of using fixtures under my 8/4 solid wood parts. I do cast a lot of concrete which could result in the need to carve out foam to make half of a casting form.

    Thanks for the link to your spindles. thats a nice setup! A great explanation of preload in the thread.

    louieatienza I forgot to mention that! I take that as the real advantage of placing the rails on the carriage.

    I am going to rework the whole system further exploring a number of different scenarios.



  11. #11
    Registered
    Join Date
    Aug 2013
    Posts
    50
    Downloads
    0
    Uploads
    0

    Default

    I just drew the z slide with the blocks mounted on the z carriage. The results surprised me. The spindle seems to be supported as well as the reverse system in most positions. Especially in the position that I plan on using it the most.

    shooting for the 9" of Z travel
    CNC Router for Hardwoods: Evaluation and Questions-sixorwindow-30-jpg

    all the way down
    CNC Router for Hardwoods: Evaluation and Questions-sixorwindow-29-jpg

    Overlaid with the previous system. The bright green blocks belong to the other arangement
    CNC Router for Hardwoods: Evaluation and Questions-sixorwindow-31-jpg

    If you all think that the overhang of the spindle in relation to the bearings is OK the blocks on the carriage seems to be an easier setup with the greatest usability (without re bolting the spindle)



  12. #12
    Gold Member
    Join Date
    Apr 2009
    Location
    USA
    Posts
    5261
    Downloads
    0
    Uploads
    0

    Default

    Quote Originally Posted by Sapele View Post
    I just drew the z slide with the blocks mounted on the z carriage. The results surprised me. The spindle seems to be supported as well as the reverse system in most positions. Especially in the position that I plan on using it the most.

    shooting for the 9" of Z travel
    CNC Router for Hardwoods: Evaluation and Questions-sixorwindow-30-jpg

    all the way down
    CNC Router for Hardwoods: Evaluation and Questions-sixorwindow-29-jpg

    Overlaid with the previous system. The bright green blocks belong to the other arangement
    CNC Router for Hardwoods: Evaluation and Questions-sixorwindow-31-jpg

    If you all think that the overhang of the spindle in relation to the bearings is OK the blocks on the carriage seems to be an easier setup with the greatest usability (without re bolting the spindle)
    In the way you have it drawn, you have to have the Z trucks pretty close together to get the movement you want; i.e. you negate the advantage of the lower bridge by having the Z trucks close together like that. With the trucks fixed on the Y saddle, you could simply mount your spindle such that the collet is just slightly below the Z carriage plate. In short, with the moving truck design, regardless of how beefy you make the carriage assembly and Y saddle, the system will only be as ridgid as the spacing of the Z trucks dictate. You can gusset and plate the carriage all you want that will just add unnecessary weight. It is however obviously easier to build this way, and again if your Z is only moving 2-5 inches, the truck spacing in relation to the overall rail length would be greater.



Page 1 of 12 123411 ... LastLast

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
  •  


About CNCzone.com

    We are the largest and most active discussion forum for manufacturing industry. The site is 100% free to join and use, so join today!

Follow us on


Our Brands

CNC Router for Hardwoods: Evaluation and Questions
CNC Router for Hardwoods: Evaluation and Questions