Rehashing ACME screw drives

I know its discussed fairly regularly... Can we spin it a bit? Yes acme lead screws are expensive (I have a freebie source) and slow...

It's that slow part that has me pondering... Perhaps you mechanical wizards can help me hash it out? It's the slow speed thats usually referenced as to why acme leadscrews aren't used - great accuracy, slower than sin... I'm wondering - what if instead of direct driving the acme - if I ran it through a gearset (or even a small belt drive)? I was actually eyeballing the herringbone wades extruder gears when I was coming up with this...

Yes it'd be another bit of complexity, and another possible problem spot...But would it not get some of the speed back + have a higher accuracy?

Anyone willing to help me out - and point me into some gearing ranges I would want to be looking at?

Or talk me out of my insanity?

Thanks in advance!

The problem with gearing up for speed is you reduce the torque and we already do not have enough torque with a Nema17. You could go with a much larger motor but then you could go with a steeper lead angle and use direct drive and no gearing. You could also go with a Ballscrew which has very little friction and can have high lead angles but you are then talking about hundreds of dollars for the ballnuts.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

With a given motor, you have a given tourque and a given maximum RPM, so you can trade speed for accuracy or vici versa. Higher thread pitch = more speed, less accuracy (and less tourque). Screw drives are limited in speed, too. Because on high RPM they start to vibrate.

Given belt drives are accurate and stiff enough for printing, there is simply no need to go with multi-hundred-dollar drives.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

In addition to adding inertia and friction to overcome, ACME screws (and any gearing to get the speed up) will add backlash. Unless you use anti-backlash components, which add yet more mass and friction. And then you have to maintain it all. Free isn't cheap enough, you'd need to get paid to use ACME screws.

Z axis is another story, of course.

The biggest issue, asside from speed limitations, is acceleration limitations. The maximum acceleration of an axis will be dramatically reduced with a screw drive, even if the steps/mm is made to be the same as a belt drive.

Well here's the outright just of it... eventorbot -enlarged and stretched a little (should be good for 18"/450mm travel on xyz). That's +/-1000mm of belt for x/y belts, I don't think that's so workable even some rough tests with some 12mm belt I have lying around - that's a lot of play...

Was leaning towards NEMA 23 steppers to avoid the not enough torque issue, yeah means more headaches in electronics...

Trying to find a nice middle ground...