Drive Fundamentals : R&P vs belt for 120" router X-axis

[tlharris]

Companion post to my earlier one re: linear guides...

I'm about to just jump into build mode after a year of lurking, reading, learning, and overanalyzing. The new machine (my first CNC) will be a nominal 8'x4'x12" 3-axis w/906oz-in steppers, <125# gantry, and a welded steel frame (with a few epoxy-bedded bolted connections). 3kw Chinese water-cooled spindle.

The challenge is to build the most machine I can for the money. Target is acceptable quality for a small production shop working in wood, foam, acrylic and occasionally some aluminum sheet. I'm not trying to mill steel, nor make NASA-quality parts. At the end of the day, however, I do need to get usable cabinet parts off this thing.

I've read so much I'm bordering on overload. When it comes to drives, I'd really like to stick to ballscrews, but they start to become cost-prohibitive when they get big. And frankly, I'm afraid of screw whip. I tell myself to just simplify and go R&P, but don't know a lot about the whole split pinion thing. Belt seems easiest, mechanically, but after reading a great many posts, I'm convinced I'll build myself a giant slingshot, with so much spring it would be unusable. The belt would be my first choice, because it seems simple and elegant. R&P would seem to follow, if someone could make the split pinion thing simple for me, and the ballscrews would be ideal, if anyone has surplus that they'd like to donate to the cause

I know there are a lot of calculations to be done, and a lot of information not given. I'm looking for more seat-of-the-pants experienced advice. I don't so much need a best-in-class machine at the end of the day, but I do need to avoid spending a lot of time and money making a piece of scrap metal.
Thanks very much in advance.

[louieatienza]

You could spin the nut instead of the screw, that would eliminate whip and allow faster rapids. Check eBay for ballscrews, I'd come across many long ones for pretty cheap prices... Also have you looked at the CNC Router Parts r&p system? Lots of users here, and no need for split pinions. There are many ways to implement belts; check out the Solsylva design, Momus design (they have a forum here), Mike Everman's ServoBelt (thread here too); even the MechMate has a belt drive option.

In fact, it might be worth the $100 to get the MechMate plans, since all the figuring is done for you!

[ger21]

I've never seen a belt driven Mechmate. Almost all are regular rack and pinionnot much different at all from the CNC Router Parts setup.

On the subject of split pinions, I don't recall seeing any DIY machines using them. Most people these days use the CNC Router Parts R&P, or a similar DIY version.

As for ballscrews, everyone says "spin the nut". But, if it was that simple, you might actually see someone who's built a spinning ballnut.

All you really need is a screw with a 1" (or 25mm) lead. A 1" screw should work fine.

But for a 120" axis, you'll need an even longer screw, which really isn't a viable option, imo.

Belt seems easiest, mechanically, but after reading a great many posts, I'm convinced I'll build myself a giant slingshot

I agree 100% here. Belts are too springy on a larger machine. The exception, as Louie mentioned, is the ServoBelt. I'm going to attempt to use a stepper powered version on my next machine, but it's untested right now.

The simplest option (and most common) would be the simple rack and pinion. Rather than a split pinion, you just spring load a regular pinion into the rack.

My question is, where does the 906 oz stepper selection come from? What drives do you plan on using?
Ideally, motors and drives should be matched to the application. If you don't know how you're going to move the machine, you shouldn't be choosing motors yet.

[tlharris]

Thanks, both.

Louie - Right on... I agree with you, the spinning nut thing would seem the right way to go, but as Gerry says..... you don't see many of them around, which makes you wonder why. Maybe I'll play with it on my next machine. More likely, if I were going to invest more R&D / tinkering time, I'd head toward the ServoBelt setup that Gerry's been working on.
I've got the Mechmate plans, and there's no doubt it's all thought out, and a completely workable design. I'm aiming for stiffer & stronger, as we really need 12"-16" Z to be useful for us. While Mechmate offers the extended Z option, there's a big disclaimer there re: rigidity.

Gerry - You are quite right re: motor sizing. That's my seat-of-the-pants guess, nothing more. Simply based on eyeballed gantry weight & target speeds, and I agree, there's another round of calcs to be done when I get past drive system. I recognize intellectually that the loaded R&P is probably more than adequate for our needs... just fighting the tendancy to oversize/overbuild. My wallet will probably pull things back into perspective here, shortly Thanks.

[bearwen]

If your interested in belt drives they will work just fine as long as you use the right parts.

Most here seem to think that belts are rubber bands this is only true if you dont know what your doing and use the wrong parts.

You would be well off to call the tech guys at brecoflex they are really nice and informative.

I spoke with one of there guys for over an hour on the phone and he walked me through there online build system and I have to say that the machine you have discribed is not much different than my machine which is in final design check status.

Belts are fine you just need the right stuff.

Belt systems are not cheap
you should not need anything bigger than a 400 ozin motor for your machine

I used brecoflex online designer and an Excel spread sheet I have and was able to design a 4'X8' machine that could rapid at 1800 IPM and do it with a single 280 ozin motor of course cutting would be much lower the machine could go from dead stop to full speed in less than 1 second and brake in the same distance and could travel the whole table in 6 seconds. Belts and gear ratios you can due alot.

My machine is looking at 400 ozin motor single for X drive and rapids in the 1000IPM range and has enough torque to machine everything you listed as that is the same list of materials that I have.

I will post the spread sheet I did not make it I got it from the mechmate forum and I have made a few additions to it for metric conversion stuff for using brecoflex's online designer (its all metric) the spreadsheet was designed for rack and pinion systems so it asks for Diametral pitch in one of the boxes this is the only questionable thing that myself and every pully engineer I have spoke with are not 100% about as in pullies they calculate stuff differentlly than the rack and pinion guys but from searching the net and asking the engineer at brecoflex who said I am not sure but what you have seems to be logical. anyways you will need the pully pitch diameter and the number of teeth (if you use SDP-Si online stuff it gives you the info you need for the spreadsheet) you put that in the boxes and it spits out the diametral pitch for the spreadsheet calculations. if you use it and its not right I TAKE NO RESPONSIBILITY!!!!