upgrading shapeoko cnc router

  • We are thinking of upgrading shapeoko cnc router to nema 23 stepper motor.

    Any ideas on upgrading the stepper motors and set up?

  • You’ll need bigger motor drivers. Upgrading the stationary motors is no problem… but the more weight you put on the gantry, the more flex and vibration you’ll get. Need to compensate by stiffening it.

    What is the bottle-neck on that you are trying to address with bigger motors? Faster rapids?

    I’ve never seen it stall, and the only missed steps I recall seeing diagnosed were from sawdust in the wheels, which @tinfoilknight fixed admirably with toothbrushes.

    It could use stiffening on general principles. Could also use some serious looking into reflashing GRBL. It was doing some crazy things with the finger maze last I saw. We tore the G-Code apart and loaded it into linux-cnc and could find no reason in simulation for it to be wandering all over the table and randomly mirroring axis’

  • The spindle on it is pretty weak too. It also does not stay exactly vertical, which goes back to what @pierre suggested about stiffening it. In particular the moving gantry flexes under load. We should probably look at spindle upgrade and options for better securing the Z axis on the moving gantry.

    I have seen the motors skip (not the belt) in several occasions, but it has been a long time since i used it.

    Faster rapids may seem not important, but actually can make a huge difference in how satisfying the machine can be. One of the really cool thing about the laser is that it is fast. If we can make the CNC router faster in both rapid travel and cutting we can win big time.

  • There’s a mod that flips the motor on the Z axis, it buys about 1.5" of Z travel at no extra weight.
    The makerslide on that axis cannot currently use its full travel because the router hits the motor mount. I’ll see if I can dig up a reference.

    Beefing up the spindle motor would tend to allow faster cutting, because you can push the speed a bit more and not end up rubbing the bit against the wood… At that point bigger motors, power supplies, and drivers start to look really tempting. I think we could actually put a better spindle motor on there and reduce the weight on the Z-axis at the same time, using some sort of fabricated mount for a replacement motor with better collets. I am not convinced we need to shell out for a full water-cooled VFD rig like the froggy router, but replacing the current trim router with a different router motor (available without the lift and handle assembly as a replacement part) and a precise-bits type collet could move the cutting axis closer to the gantry (makes the cantilevered forces much smaller,) remove the moving weight of whatever extra plastic crap they put on the handle part, remove the flex of going from metal->plastic->metal, and make it possible to design the mount with a built-in dust-collecting feature.

    The most interesting z-axis mod I’ve seen documented is this one: http://timf.anansi-web.com/wp/possible-new-z-axis-design-for-the-shapeoko/ By moving the lifting screw off to the side Tim has managed to move the spindle axis considerably closer to the gantry. I’m not sure that the off-center lift is worth it though… might rack the makerslide too much.

  • So what can we do to make it cut through wood and run the way it is supposed to do?

  • @Ron_Ron my best suggestion is to make sure the cutting bits are sharp.

  • @Ron_Ron

    Faster cuts (as opposed to faster rapids) will need stiffening of the gantry, and a stronger spindle. If you can make the motors move faster, you’ll just overrun the spindle cutting speed, and make the gantry flex, which will give you chatter, inaccuracy, and a tendency to bind up the mechanism. If the motors are strong enough and you try to cut fast with the gantry flexed hard, it’ll start to strip gears and belts.

    Assuming that you address the flex and the horsepower, at some point, speeding up the cuts, you will need to switch from 1/4" to 1/2" shank router bits… so if you are planning to have cutting forces that high, you might as well plan to put a 1/2" collet in when you do the spindle upgrade.

    Without doing any upgrades, the current system is capable of good results (if somewhat inconsistently) by taking several passes, making light cuts at reasonable speeds. The semi-random behavior where it decides to fuck off across the table and randomly mirror your work is much more concerning to me than how fast it cuts wood. If anything the slow speeds give me more reaction time to punch the E-Stop.

Log in to reply