I’m running a Shapeoko 3 that I upgraded several years ago so I can cut full sheets of plywood. I’m looking into upgrading the drive system, especially exploring rack-and-pinion or other alternatives to the current belt and pulley setup. I’m not a machinist, so options that aren’t super exotic are ideal.
If anyone’s done belt-upgrades, rack-and-pinion conversions, or hybrid systems, I’d love to see what your setups look like (pics highly appreciated!) and hear what worked for you — what parts you used, what issues you ran into, cost, accuracy, how you handled backlash, etc.
Very interesting! The belt method has been “okay” in the past for me. I have only had a belt snap once. I’ve just always heard that for more high accuracy jobs usually a rack and pinion or lead screw is the way to go.
We tried steel core belts, and while they are more rigid, when you route them through small pulleys, you fatigue the metal cores. They only last a few months before snapping. It’s really not recommended unless you plan on doing regular and proactive maintenance.
I ran steel core belts for several years in my 3xxl. With hdz snd 80 mm spindle. High feed rates and a 20 ish lbs z axis mandated them. They gave me lower deflection. But they would break , normally the x axis , every 6 weeks or so. ( machine ran 50 hours a week).
I ended buying 10 meter rolls and replacing the first Sunday of every Month. It was worthwhile for me. But it takes an hour to replace the belts and get the tension right.
Personally i wouldn’t consider a lead screw. It would be a ball screw or a rack and pinion. With ball screw being the more accurate more reliable longer lasting option. But you’d be better off. Buying a ball screw machine.
Ah, thank you for the update, Winston and Ryan and Rob. As Paul Harvey always said “and now for the rest of the story”. Unfortunately for many threads here, there is no closure. But you have provided that now!
An old thread that I can’t find here (but the beginning of it is still archived on the Wayback Machine) about retrofitting wider belts to Shapeokos inspired my own Nomad belt retrofit. There is an existing thread still here too.
Hey that is one impressive SO3. I have seen people upgrade regular so3 and XL to XXL but you really made an upgrade.
Although the SO3 is a great machine at some point you should consider simply upgrading to a newer machine without belts. Trying to re-engineer you machine with a trial and error method could be costly. Plus time moves on and maybe you should also. The SO5 4x4 can cut whole sheets but not in one cutting session. Depending on what you are making maybe you just need a bigger industrial machine.
@CCD with a machine that size, anything else than belts will likely turn out to be quite expensive. The key question is what parameter you want to improve, and to make really sure that any upgrade will actually address that (but then again if this is for the pleasure of upgrading because it’s fun, that is quite a valid reason too )
For a machine that size (I assume you joined up beams from two SO3 XXL ?) I would expect deflection to be a much more critical parameter than backlash.
Larger belt width would likely be the easiest upgrade (if you can find/order the right pulleys), I do love the 15mm belts on my Shapeoko Pro, though I could not say whether upgrading to larger belts would be sensible if reusing the original SO3 electronics/stepper motors.
Then you could replace the wheels and go for linear rails, but that is not an easy mod on a machine that size (you would need to order custom size linear rails, not such a big deal, but then you will need to design and mill multiple custom adapters parts to attach the beams)
And finally there is the ballscrew upgrade, as in latest Shapeokos. But ballscrews that long and the effort to fit them on a giant S03 (many custom parts to mill again) would likely be more expensive than buying a new machine as others have said, or at least very far from a quick & easy upgrade.
Rack & pinions came up a couple of times over the years on the forum, but as far as I can remember no one ever actually did it, probably because it is impractical.
