Some thoughts on eliminating Leadscrew Whip

[Foxman]

I noticed that there is a trend to build bigger and bigger CNC's and that sometimes in machines with leadscrews at around 36" and longer the leadscrew sometimes whips when spun fast.

1. My first thought was, build it backwards. That's to say, fix the leadscrew and spin the nut. No whip!. The following drawing shows a hollow shaft supported by two bearings, spaced apart and locked in place with a nut. At the other end is a large diameter section that is machined as a timing belt pulley. Attached to the face of this is a CNC Dumpster anti backlash nut. The whole unit is driven by the stepper mounded above with a 1:1 ratio belt drive.



2. Second thought. If you take a section of leadscrew and cut lengthwise grooves in the thread and then give it some heat treatment you basically get a tap. Mount this in a lathe between centers and then feed the edge of a freewheeling disk into the tap and the disk will start turning, taking on the reverse shape of the tap, like an inside out nut. If you then mount some of the same threaded rod in a C channel and clamp onto the channel/thread combination with your new disk and a bearing under the channel you have a rack and pinion drive. No whip!

[FandZ]

I've seen a few builds utilizing idea number one successfully. I hope to go with a similar setup in my next build or the one after.

Idea number 2 looks interesting but I'm wondering how you'll be able to limit back lash. Do you have any ideas on that?

[jalessi]

The first idea is good and has been implemented many times in the past.

The second idea is weak because only one thread is fully engaged at a time.

Jeff...

[DareBee]

Simple fix is larger diameter screw.
Typically we don't go to a rotating nut til over 12'

[jsheerin]

My ~60" screws whip on my machine at speeds above ~250ipm. I thought about going to a rotating nut setup and actually designed something in CAD very similar to yours, although with thrust bearings incorporated as well. However, the more I thought about it, the more I realized that it would not give me much advantage for the complexity required. It would get me faster rapids, but my current stepper motors do not have enough torque at high speeds to actually be able to cut much faster than I currently do. It seems it would be worth it if you went with servo motors that could supply more torque at high RPM, but that would require replacing a large portion of my electronics. Alternatively, I could try to use steppers with more torque at high speeds, but I'd still have to replace all my electronics to drive the larger motors. If you were starting from scratch it could be a nice solution, but it's much more complicated to build, so it would be hard to do if this was someone's first machine.

[Foxman]

Originally Posted by FandZ I've seen a few builds utilizing idea number one successfully. I hope to go with a similar setup in my next build or the one after. Idea number 2 looks interesting but I'm wondering how you'll be able to limit back lash. Do you have any ideas on that?

I would be interested to read about them. As to the second idea there should be little or no backlash as the wheel was made to fit the thread by using the thread to make the wheel, and the thread is held against the wheel by the secondary bearing.

[Foxman]

Originally Posted by DareBee Simple fix is larger diameter screw. Typically we don't go to a rotating nut til over 12'

I was thinking more DIY, most of the leadscrews I have seen people use are 1/2", which means the core of the thread is somewhere between 1/4" and 5/16" in diameter. What diameter screws do you use?

[Foxman]

Originally Posted by jsheerin My ~60" screws whip on my machine at speeds above ~250ipm. I thought about going to a rotating nut setup and actually designed something in CAD very similar to yours, although with thrust bearings incorporated as well. However, the more I thought about it, the more I realized that it would not give me much advantage for the complexity required. It would get me faster rapids, but my current stepper motors do not have enough torque at high speeds to actually be able to cut much faster than I currently do. It seems it would be worth it if you went with servo motors that could supply more torque at high RPM, but that would require replacing a large portion of my electronics. Alternatively, I could try to use steppers with more torque at high speeds, but I'd still have to replace all my electronics to drive the larger motors. If you were starting from scratch it could be a nice solution, but it's much more complicated to build, so it would be hard to do if this was someone's first machine.

What diameter and number of thread starts is your 60" leadscrew?

[FandZ]

There defiantly could be some merit in idea 2. It could free up a lot of lost torque when you compare it to useing a derlin anti backlash nut. Certainly would simplify mounting the motor to it. It's chancy, since I haven't seen it done before, but I'm kinda hoping you try to pull it off.

What kind of material were you thinking the disk/gear should be made out of?

Originally Posted by Foxman I would be interested to read about them. As to the second idea there should be little or no backlash as the wheel was made to fit the thread by using the thread to make the wheel, and the thread is held against the wheel by the secondary bearing.

[jsheerin]

1/2-10 5 start. It's on a 4'x4' cutting area router.

[jsheerin]

Idea 2 is basically a rack and pinion. You would need to support your screw from the opposite direction of the pinion gear. It would be easier to support a rectangular 'screw'. Gosh, that sounds like a rack... ;> Seriously, why reinvent the wheel (rack and pinion)? I mean if you want to make your own, that's cool, but if the purpose is to make a machine that is reliable and as cheap as possible (including the cost of your time) in order to make other things, then I'd just buy a rack and a pinion gear and worry about removing the backlash. And if you do build your idea, you will still get backlash over time as the parts wear unless you build in a spring or something to compensate for it (like in a typical rack and pinion). That's half of the purpose of the anti-backlash nuts - compensating for wear. Also, delrin anti-backlash nuts are not that horrible in terms of efficiency as far as I know, especially with higher lead screws such as the multi starts. I don't know how rack and pinions compare, but I don't think there would be a large difference.

[LazyMan]

Just some things to think about rotating nut designs. If you are using acme rod then no problems but if you are using a regular ballnut not design for rotating you could potentially have problems. Most ballnuts have a preferred mounting orientations to ensure the proper recirculation of the balls. For example Nook Industry ball nuts are not supposed to be mounted with the ball recirculation track facing down. This is something you could get away with doing especially on a light machine. Rotating ball nuts are especially designed for this task but are unfortunately more expensive.