Rack & Pinion or Screws?

[Rob K]

I'm designing / building a 2x3 machine to route wood and occasionally mill pcb's.

At first I thought R&P would be the better route - no bent screws, no whipping, and possibly less expensive?

After thinking about the z-axis, I'm having second thoughts about R&P. Does anyone have any recommendations or experience here. I'm thinking screws will be more accurate for pcb milling, but will severely limit speed when routing wood. What if I venture into routing aluminum? Which drive mechanics will be better?

I thought I'd invest in linear bearing guides to improve rigidity (and accuracy?) for pcbs.

Is overall weight more important?

Thanks in advance for any comments or suggestions.

[Dorsal]

How about a compromise? R&P for the X and Y axes and ACME or ball screw drive for Z? The controlling software doesn't care and that will help reduce the weight of the Z-axis.

[HereinCS]

From what I recently learned, ground screws are the most accurate way to drive any axis. They are more maintenance though. Needs lube and are sensitive to debris compared with rack and pinion. But no back lash. I wanted to go fast on my machine as well, but fast is only as fast as your cutter can cut through the material. No sense going 800 ipm if you are using a small spindle through tough material.

[gio666]

Ballscrews are far better than R&P , especially at only 3 feet

[Rob K]

Does the lead of the screw affect accuracy? Obviously it affects precision, but what about accuracy (i.e. does higher lead mean more backlash)?

I understand keeping the weight of the gantry down for acceleration and reducing flex of the supporting axis members, but is it good or bad to add weight to the frame?

I got significant quality improvements when I upgraded from my "light weight" contractor's tablesaw to a full cabinet saw. The thing I noticed most was that the weight eliminated a lot of vibration. Will I see similar improvement with a heavier base?

I like the simplicity of aluminum extrusion, but I'm concerned that it doesn't offer enough weight.

[ahren]

Rob,

I think for a machine this size, ACME screws are definitely the way to go. While we offer both rack and pinion and ACME drive kits, I typically don't recommend rack and pinion until you get above 3 feet in axis length. Under this size, a multi-start ACME solution is plenty fast and is typically more economical. I always recommend screws for a z axis, simply so your axis maintains position if power is dropped.

Your recognition of the difference between accuracy and resolution is a good one, and is often lost on people. With a 10x microstep driver, and a 5 start 1/2-10 screw, you'll have an effective resolution of 0.00025". Will an ACME solution ever see this type of accuracy? Not likely, but the error is not due to the lead of the screws. A single start screw will have a theoretical resolution of 0.00005, but will face similar issues of general pitch error or backlash at the bearings. That being said, with backlash compensating nuts and a tight thrust bearing arrangement, it's not too hard to get a system with less than 0.001" of backlash, and more than reasonable accuracy for cutting wood. Hope that helps!

For a good example of a 2' x 3' machine using all ACME, check out one built from our plans here:

http://www.cncrouterparts.com/produc...products_id=47

Or to confuse things, see a similar sized machine using R&P on the X and Y

http://www.cncrouterparts.com/produc...products_id=72

Ahren
www.cncrouterparts.com

[Dean448]

I built a machine using the bed of a bumper pool table, which is 32" X 48", and have no issues with wipping of acme lead screws that are 1/2-10 5 start. I have the steppers geared so they wont exceed about 700 rpm at the highest feeds. I checked for wip using a formula before building all this.

My understanding of rack and pinion is that its used for both long travel and high speeds. I'm sure they can be accurate but you will pay more then for acme.