Which is better?

[bobsch]

Is a cheap ball screw better than an expensive Acme?

[ger21]

Cost doesn't really tell you anything. You really need to see the specs. I guess it may also depend on the application.

[bobsch]

Originally Posted by ger21 Cost doesn't really tell you anything. You really need to see the specs. I guess it may also depend on the application.

I suppose more info would be helpful. I'm going to build a 36x48 CNC router for wood and possibly light aluminum and am trying to determine what to use for the x-axis.

[mighty3]

Go for ball screws as acme screws are inefficient and combined with stepper motors can compound the problems.

I learned this from a company site which builds CNC Routers.

[ahren]

Lots of people have made extremely impressive machines with ACME screws and stepper motors. These components are perfectly suited for hobby and light industrial machines, and in general offer a much better value than ballscrews for people interested in doing cutting tasks that don't require +/-0.001" accuracies or better.

While ballscrews are more efficient, high pitch ACME also runs smoothly and does a great job of converting rotational motion to linear travel. Furthermore, unless you get a double ball nut or a ball nut with oversized balls loaded in it for your ballscrew, as well as angular contact end bearings for them ($$$), it's possible to get worse backlash in a ballscrew than in a low cost anti-backlash ACME setup.

As to stepper motors, these are proven technology that are easy to setup, and can offer truly impressive performance. If one has unlimited funds, then by all means, go with servo drives and planetary gearheads, and have fun tuning PID loops. It is fun -- I used to do this for high speed pick and place robots. But for a beginner or anyone on a budget, steppers can offer an excellent performance/$ ratio, especially when coupled with a high end drive like the G540, and are quite easy to set up. Not sure what "problems they compound" -- they work well for a lot of people on this forum.

Best regards,

Ahren
www.cncrouterparts.com

[Kipper]

Originally Posted by ahren Lots of people have made extremely impressive machines with ACME screws and stepper motors. These components are perfectly suited for hobby and light industrial machines, and in general offer a much better value than ballscrews for people interested in doing cutting tasks that don't require +/-0.001" accuracies or better. While ballscrews are more efficient, high pitch ACME also runs smoothly and does a great job of converting rotational motion to linear travel. Furthermore, unless you get a double ball nut or a ball nut with oversized balls loaded in it for your ballscrew, as well as angular contact end bearings for them ($$$), it's possible to get worse backlash in a ballscrew than in a low cost anti-backlash ACME setup. As to stepper motors, these are proven technology that are easy to setup, and can offer truly impressive performance. If one has unlimited funds, then by all means, go with servo drives and planetary gearheads, and have fun tuning PID loops. It is fun -- I used to do this for high speed pick and place robots. But for a beginner or anyone on a budget, steppers can offer an excellent performance/$ ratio, especially when coupled with a high end drive like the G540, and are quite easy to set up. Not sure what "problems they compound" -- they work well for a lot of people on this forum. Best regards, Ahren www.cncrouterparts.com

I'm thinking in regards of Acme leadscrews that they are inefficient and a wearing part that isn't needed.....I'd go belt before Acme/Crapeziodal anyday....at night I'd go Rack.

[ahren]

While I do like the term "Crapezoidal" , my experience with ACME has been quite good in terms of wear, especially with nuts that have a backlash compensating mechanism -- one of my customers has run the equivalent of 5 miles of travel on the same delrin ACME nut, and when it does wear out, he'll just put in another $19 nut. Would a ballscrew save you this change out? Maybe, but at over double the cost, and in a less compact design. I don't disagree with you that ballscrews are a fundamentally more elegant technology, but the open-ended "which is better" question deserves consideration of cost, as well as the application the screw is going into.

As to belts, the downside of these is that unless you're using a clever design like Mike Everman's, they can stretch significantly, especially under cutting loads, which can lead to inaccuracies. They do have an advantage over rack and pinion in that one can run the belts with the motors fixed, rather than moving, but the added stiffness of R&P in my mind makes it a better technology for DIY CNC. My $0.02.


Ahren
www.cncrouterparts.com

[Kipper]

Originally Posted by ahren While I do like the term "Crapezoidal" , my experience with ACME has been quite good in terms of wear, especially with nuts that have a backlash compensating mechanism -- one of my customers has run the equivalent of 5 miles of travel on the same delrin ACME nut, and when it does wear out, he'll just put in another $19 nut. Would a ballscrew save you this change out? Maybe, but at over double the cost, and in a less compact design. I don't disagree with you that ballscrews are a fundamentally more elegant technology, but the open-ended "which is better" question deserves consideration of cost, as well as the application the screw is going into. As to belts, the downside of these is that unless you're using a clever design like Mike Everman's, they can stretch significantly, especially under cutting loads, which can lead to inaccuracies. They do have an advantage over rack and pinion in that one can run the belts with the motors fixed, rather than moving, but the added stiffness of R&P in my mind makes it a better technology for DIY CNC. My $0.02. Ahren www.cncrouterparts.com

I asked in his "original" thread how the patent application was coming along....no reply

My 2p worth is that the initial stretch of a belt drive outweighs the perceived superiority of the R&P.....plus the belt drive reduction for some kind of resolution.

Proper specification is the key to belt drives....5mm width belt and 1.2M travel is going to be bad news for resisting cutting forces....I predict 25mm width might just do the job for a router width moderate DOC....Everything is a balance (including cost)..

Open ended is a correct assumption of the OP's question....very ambiguous I'd prefer "I want to cut (insert material) at such and such DOC and X feedrate If I haven't done the same I don't reply.

There's not much that's new on Gods earth.....Mostly adjustments and reincarnations of ideas that have gone before.