CNC Upgrades

Discussion in 'CNC machining projects' started by Al2O3, Feb 28, 2023.

  1. Al2O3

    Al2O3 Administrator Staff Member

    The time has come for me to make some upgrades to my CNC router. As far as the crowded market space for such hobby CNC machines go, I suppose the MillRight MegaVee was probably a decent value. Here’s a link to the specs.

    The primary interest in this machine for me was the large work envelope and (relatively high) speeds since my interest was largely machining lost foam patterns. It has ~4” of Z travel, but you’ll find out pretty quickly you can only use <2” of it with a 2” CL bit installed. When I bought, I knew under gantry height and Z travel upgrades were in the cards. Having used it for a year, I’ve also machined some hard plastics and aluminum. I find it to be wanting a bit in that use. There are some structural deficiencies that cause some deflection under heavier tools loads. I’ll get to that, but to cut to the chase, in addition to increasing height under gantry and the Z travel to 10”+, I also decided to upgrade linear motion guides on all axis from wheels to 20mm linear rail, and move to a 1.75HP router with more speed range and ½” collet. I plan to stick with the existing controller and NEMA 23 steppers at least for now, if they can hack the extra mass. I also kept the rack and pinion drive on X&Y but went with a 16mm ball screw on the Z.

    I went back and forth with myself about the Z-axis design as was whether to use a stationary or traveling linear rail/linear bearings. Most designs have the rails stationary, but to have more height under gantry and travel, that means (really) long cutting bits and/or having to move the table closer to the gantry or installing an elevated cutting platform to use shorter bits, and I have a lot of them. The trouble with the traveling rail Z is stiffness while extended. A comment that BattyZ made to me was almost no matter what I do my Z structure would be stiffer than a long (say 4-6” CL) ½” bit. It doesn’t package up quite as easily, but since I mostly cut soft materials (and like a challenge), this adaptability was very appealing to me since stiffness at full extension cutting foam (which is the only time I’ll ever use the Z travel) will never be a challenge and I could install a platform to use the Z where it is stiffest for harder materials, and meanwhile, focus on making the Z as light and stiff as it can be, yet still use the existing NEMA 23 steppers and my controller.

    1 Z Side View.JPG 2 Z-Top View.JPG

    Since the Z axis is all new, my plan is work that to completion, make the parts to adapt the X and Y axis to linear rail guides, to the point it can hopefully all be installed in a day, because until then, it’s tools making tools…..and you guessed it, there will be castings! Let’s start out with this one.

    3 Z Casting and Router.JPG

    More on that Z casting here:

    CNC Router Upgrade Parts | The Home Foundry

    -To be continued.

    Tobho Mott likes this.
  2. Smoking Shoe

    Smoking Shoe Silver

    A work around for limited Z travel is to 'change' the Z axis. A pic of a machine I was contemplating a few years ago that was going to make parts that needed both good Z travel and rigidity about the X/Y. Project was still born - machine never built.
    Side Z.JPG
    Rocketman likes this.
  3. Al2O3

    Al2O3 Administrator Staff Member

    That is clever. It would be a bit clunky as far as software, hard/soft limits, homing, etc. Won't fix height under gantry problems though for total cutting envelop.

  4. Smoking Shoe

    Smoking Shoe Silver

    No, more like desperation. "Clever' only happens every couple of years. ;)
  5. Al2O3

    Al2O3 Administrator Staff Member

    New 12" travel Z Axis is nearly a reality.

    7 Z Backside.JPG 8 Z Axis.JPG

    I said I was going to retain the rack and pinion drive on the X & Y axis but I can replace them with 20mm ball screws for $150. At that point, the only thing left of the original machine are the steppers, controller, and extruded aluminum rails. I'm having evils thoughts of just selling my existing machine to fund upgrading those too. I'd have a much more capable machine and money in my pocket from the proceeds of the sale.

    BattyZ and Tops like this.
  6. Tops

    Tops Silver

    Wow, if you can finish the new one and cash out of the old one, breaking even or coming out ahead, that sounds really good.
  7. Al2O3

    Al2O3 Administrator Staff Member

    The Z axis is done. I made the stepper mounting plates big enough to accommodate NEMA34 steppers. The CAM drawing still shows use of rack and pinion drive. I added a tail shaft support for the ball screw that tucks up inside the tunnel on the backside of the Z casting. It also serves as a mechanical stop to prevent the traveling Z from being run off the rails although I hope never to put that to the test. Fully extended it’s 37” from stepper motor to the base of the router clamp! I’ve drilled and tapped enough M5 threaded holes to last a lifetime.

    9 Ball Screw and Z Plate.JPG 10 Mounted Z Axis.jpg
    13 Z extended.JPG 11 Z and X Axis CAD Assy.jpg
    Tops likes this.
  8. ddmckee54

    ddmckee54 Silver


    What kind of speeds you hoping to get out of this machine? Just wondering, because it seems like that much mass is going to have a lot of inertia, changing directions quickly in X or Y isn't going to be easy is it? On the other hand the mass should be great for dampening vibrations.

  9. Al2O3

    Al2O3 Administrator Staff Member

    200in/min for foam cutting. Much less for most other things.

    The max acceleration can be set to slow the rate of speed changes yet still run at the higher speed in between where speed is spooling up and down. Most determine this by running their machines up until they start to missing steps and then back the accleration down to about 75% of that.

    On the existing machine, the limiting factor is the holding torque of the steppers. As I mentioned, I'm seriously considering a new build from scratch, If so, I'll upgrade to NEMA 34 steppers. The economical ones have ~4x the torque of the present steppers. To keep the rpm in the more optimal torque/performance range and still use the economical ball screws, the most common approach is to use a timing belt between the stepper and ball screw to gain up the speed. The small amount of wind up/down in the belt translates to small positioning error but mostly at the highest acceleration regimes so as you back off accel values, it becomes less of an issue. At high speeds, the accuracy should be well within my needs for machining foam and quite good at the slower speeds for harder materials.

    This would probably only be the case on the X&Y axis. I dont think it would be necessary on the Z. As long as the stepper holding torque is sufficient so steps aren't missed, it's dialing back the acceleration so the machine doesn't shake and walk and to be easier on drive components.

    The Z does have some additional mass, but you don't get +10 inches of travel for nothing. I think my total moving mass will be considerably less than most commercial machines in a similar class and drive line.

  10. Al2O3

    Al2O3 Administrator Staff Member

    I did some testing on my new Z.

    I changed the resolution to suit the 10mm lead ball screw, and increased the max travel. I left the acceleration set at 700mm/sec2. The max rate has been set at 4570mm/min and I have been using it with these settings for the past year and a half. I arbitrarily doubled that to 9000mm/min. I put a dial on it and was jogging back and forth at those settings. It repeated without loss of steps to within .001". I applied some resistance with my hand, probably 30-40lbs, and still repeated as close as the dial could measure. The X&Y rapids at 16500mm/min. Here’s the link to a short video.

    The entire Z assembly weighs 38.8lbs. The router is 6lbs of that. The weight of all the moving parts of the Z axis including the router is 18lbs.

    Mark's castings, DaveZ and Tops like this.
  11. ddmckee54

    ddmckee54 Silver


    Have you tried bolting the equivalent mass to your existing router's Z axis and moving it in X&Y? That might tell you if the new axis will be a problem to move at speed. (Before you rip everything apart to install it.)

  12. Smoking Shoe

    Smoking Shoe Silver

    That is a hazard of this kind of upgrade project. By the time you are finished you have a pile of parts to build the original machine - just minus one.
    Kind of makes us wonder why we bought the machine in the first place rather than build from scratch to start.

    Then I remember all the stuff I've have already made and put the original back together. It isn't as fast or accurate as the 'modified' version so ........ rinse/repeat.

    Then you run out of workspace. o_O

    If you do upgrade the steppers have you considered hybrid stepper/servo?
  13. Al2O3

    Al2O3 Administrator Staff Member

    Still learning, but if I knew then what I know now, I would have just built my own. But if I hadn't bought the machine in the first place, I probably wouldn't know what I know now ;):). The old machine is probably decent value for money and not everyone is equipped or inclined to build their own machinery. My time was better spent learning CAD, CAM, & G-Code. I still think I'll be money in pocket after sale of the old machine, not counting my time of course, but I enjoy that part. I think it would be $10k-13k to buy the machine I'll build.

    If I pulled the trigger right now, it would be closed loop NEMA34 (hybrid) steppers. I'd probably gain them up 3:1 to with timing belts to stay with economical (5mm-10mm) lead screws and within the more optimal torque range of the stepper. Do I need 34s versus 23s? Probably not on the Z, but can't pass up 4x the torque on the heavier axis' for $100-$200 additional cost. I will have quite a bit more mass to push around. Certainly can't beat the servo performance, but for my purposes and typical use, I sort of doubt I could make good use of all their advantage. Closed loop NEMA34 steppers, though not quite servo performance, would be a quantum leap from where I'm at, at considerable cost savings over good servos.

    12Nm 4 axis stepper packages complete with stepper, drive, power supply, encoder, and cables, are about $160/axis. I need two for the Y and one for the X. Probably dont need that much for the Z, but may as well keep them all the same. I figure the only downside besides small overall difference in system cost is the damage that can be caused my the more powerful motors in a crash.

  14. Smoking Shoe

    Smoking Shoe Silver

    Sounds like a good plan.
    IMHO it's better to keep the steppers comfortable in the mid range of the planned cutting speeds rather than go all out for top speed.
    There are formulas to figure out the sweet spot if you have the stepper specs. One example:

    Found an OLD spread sheet that I have used. don't know where I got it and I've not double checked the math.
    Change the file extension to use.

    Attached Files:

    Last edited: Mar 24, 2023
  15. ESC

    ESC Silver

    That Z was impressive Kelly. Even though it is what you designed I thought it was going to run off the rail.
  16. Al2O3

    Al2O3 Administrator Staff Member

    Stepper torque starts to really suffer above 300rpm. By 900rpm they are usually 1/10 their rated torque. You just need to decide how much loss you can stand in between.

    It's good summary of some of the trades and considerations

    It's a text file.

    yee of little faith :)

    -Always compromises. I wrote this up for a post on another forum.

    First thing to consider is my intended use of the machine. It’s still a hobby machine and after all, still just a cnc router. I have no interest in production work, more so just maximum versatility at the risk of the old saying….jack of all trades, master of none.

    My CNC Router occupies 4ft x 4ft floor space. For that space claim in my shop, it had better be a darned useful machine. I have a knee mill with DRO and use that for heavy stock removal, accuracy, and of course harder metals. Still, I machine quite a bit of aluminum and aluminum castings. For high feature count and complex tool paths, sure would be nice to have CNC aluminum cutting capability.

    I machine Styrofoam patterns for lost foam casting. It’s not a very challenging material to machine. Cutting forces are essentially nil. I just want a big work envelope and high speed. You can rough foam faster than 200in/min but you can also use large diameter bits with large depth of cut for huge stock removal so there isn’t great need for much higher speed in roughing, just finishing 3D surfaces.

    When I get into very fine details with very small cutters (say .031”D), sometimes speeds are down in the 5-10in/min range. Of course wood, hard plastics, and aluminum are in between. When I’m cutting foam, machine rigidity or loss of torque from high (stepper) rpm operation isn’t much of an issue other than being able to withstand the inertial loads from high speed and rapid changes in direction which to some degree can still be mitigated with the max acceleration settings. For harder materials rigidity is much more important.

    So, my plan for the non-moving parts of the machine is very rigid and accurate (steel structure), and make the moving parts rigid yet light as possible, which basically means the Z axis, traveling gantry (X) beam, and gantry end plates. Young’s (elastic) modulus for all aluminum alloys is more or less the same, regardless of the alloy’s strength and is 1/3 that of steel. So even though they are stronger, using high strength alloys does not help flex and vibration. But, aluminum is also 1/3 the weight of steel and that is a big deal. I plan to make box section aluminum fabrications (high polar moment of inertia for stiffness) for those to get a good compromise of rigidity and low weight for low moving mass.

    To help with rigidity when/where needed, I plan to make the gantry height easily adjustable. For example, for machining aluminum plate I don’t need much Z stroke nor height under gantry, so move the gantry and Z as close to the stock as possible where the gantry and Z are most rigid. I’m sure I’ll occasionally try to machine an aluminum casting where I need height and Z travel, but in those case I’ll just have to be more conservative on stock removal rate.

    The motion control hardware is more compromises. My existing system is rack and pinion drive. It’s easy to keep the steppers in their happy place with R&P. They’re strong and durable, but can foul with chips and suffer a bit with accuracy compared to a ball screw. The ball screws are accurate but not as strong nor durable but can be addressed with size/diameter. The import ball screws are really the only ones that are within the budget of a hobby machine, but they are typically only available in 5mm or 10mm lead. 5mm puts directly driven steppers into untenable speeds over my cutting speed range (~1000rpm at 200in/min), thus my comment about gaining them up with a timing belt. I can afford the some loss of rigidity and torque at 200in/min because I’m just machining foam and soft materials. At aluminum cutting speeds if falls back into the stepper happy range.

    There’s a whole other discussion on compromises in the motors and drivers and there settings but I’ve distilled that down to closed loop NEMA34s steppers, microstepping.

    I haven’t talked about Rapids speeds. Sure, I’d like high speeds there too. My existing machine rapids at 650in/min in X & Y but it's not very important to me. I figure if the machine is moving it should be cutting not going to a new position. I only make one or small number of anything so having a highly optimized run time doesn’t mean much after you’ve spent a bunch of time designing and CAM programming.

    So as you see the thread progress, that’s where I’m coming from with the choices and compromises I’ve made.

  17. Smoking Shoe

    Smoking Shoe Silver

    Yes, this site won't accept but a few file extensions. Kind of common for Xenforo based sites.

    Work around is to rename the file with an acceptable extension and let the recipient change it back.

    Rename the file by deleting the ".txt" and replace the "_" with a dot.

    I suspect the forum allows the administrators to do this to help reduce the spread of viruses? Use due diligence when opening this kind of modified file.
  18. Tops

    Tops Silver

    Smoking Shoe, would it be possible to load up the spreadsheet on your end and save it to a *.csv format spreadsheet please? I cannot get the save and rename of the old *.xls to work on my end. LibreOffice is not recognizing the renamed *.xls as a spreadsheet, just a text file with lots of non-text characters and no delimiters.
  19. Smoking Shoe

    Smoking Shoe Silver

    Tops likes this.
  20. Tops

    Tops Silver

    Thanks for doing this. All three versions read fine into LibreOffice.

Share This Page