Looking For Examples of High Lead Ballscrews With Gear Reduction in Z

I'm looking for some examples of high lead ballscrews with gear reduction in the Z axis.

When I say high lead, I mean anything over 13mm / rev.

Gear reduction could be planetary or belt driven.

Could be a commercial machine, a DIY machine, doesn't matter. Any kind of computer controlled machine with a high lead ballscrew and gear reduction in Z.

I'm just looking for any examples I can find please. Not sure if any exist.

Not exactly what you are looking for, but my router had 40mm lead on the X and 25mm lead on the Y The X was belt driven @ 1.3:1, and the Y was 1:1. The Z was 25mm lead also with about 2:1 belt drive.

Quote:

---

Originally Posted by Jim Dawson

Not exactly what you are looking for, but my router had 40mm lead on the X and 25mm lead on the Y The X was belt driven @ 1.3:1, and the Y was 1:1. The Z was 25mm lead also with about 2:1 belt drive.

---

Hi Jim!

It's been a long time. I have been away from the forums for a couple of years. Working alot of overtime, sometimes 80 hour weeks. Recently, like so many others, I've lost my job. So now I have time to post.

I almost sent you a PM a while ago about a Galil controller I was thinking of buying. It was sold in the meantime. Interesting that Mach 4 has a plugin specifically for Galil.

Yes, that is exactly the kind of info I am looking for.

I am contemplating using some precision ground 1620 ballscrews for a dual column, rising gantry, build. With either 3:1 or 5:1 low backlash planetary gear reduction. I'm concerned that this might result in possible damage to the ballscrews if accidentally run into the stops at full torque. I don't really know how the 20mm lead ballscrew compares with lets say a 5mm lead when it comes to accidentally damaging things.

I'm just so indecisive about this that I'm trying to find some examples.

As far as the math goes for inertia, and torque, it would give very decent performance....

So what was your experience like using a 25mm lead on the Z? What diameter? Was your Z axis counterbalanced? How much did it weigh? Any thoughts you have are appreciated.

Sorry to hear you lost your job.

I think my ball screw was 25mm dia, at least 20mm, not exactly sure. The weight was around 75 to 100 Lbs, not counter balanced. Driven by a 1280 Oz/in NEMA 34 stepper.

I really don't think you'll have a problem unless you rapid into a hard stop with a big motor. That's what they make travel limits for :) A low lead (5mm?) ballscrew generates a lot more thrust (4X) than a 20mm lead at the same reduction. Or to say it another way, the thrust produced by a 20mm at a 4:1 reduction is the same as 5mm lead at 1:1, all else being equal.

The mechanics of the system should be built heavy enough to resist the full torque of the motor. The motor should slip (decouple) in the case of a stepper or trip out on over torque or encoder error in the case of a servo. In either case the travel limit should trip out the motor before anything hits a hard stop.

Just to really put this in perspective, due to a stupid programming error on my part, I ran a tool into my lathe chuck (rapid, 400 IPM)). 2505 ballscrew, 1.8KW servo, direct drive. Wasn't close to the travel limits. The motor tripped out instantly, and no mechanical damage done except the motor coupling may have slipped on the shaft, I'll figure that out in a couple of days when I tear into it. The carriage in this case weighs around 1000 lbs.

Quote:

---

Originally Posted by Jim Dawson

I think my ball screw was 25mm dia, at least 20mm, not exactly sure. The weight was around 75 to 100 Lbs, not counter balanced. Driven by a 1280 Oz/in NEMA 34 stepper.

---

That's decent

Quote:

---

Originally Posted by Jim Dawson

I really don't think you'll have a problem unless you rapid into a hard stop with a big motor.

---

That's what I'm concerned about. I wouldn't be using big motors, but driving into the hard stops by accident, or accidentally having one Z axis go the opposite way during setup (dual column, rising gantry, so 2 Z axis lifting the gantry). The normal kind of mistakes you might make when setting up a machine for the first time.

Plus the precision ground ballscrews just don't "look" like they can take as much punishment, for whatever reason, lol. Perhaps I'm just not used to them.

Also, I'm looking at this from a perspective of "best practices and experience", so finding examples of other machines with higher lead ballscrews and reduction, like yours, that work well, makes me think it's not a bad idea.

Quote:

---

Originally Posted by Jim Dawson

The mechanics of the system should be built heavy enough to resist the full torque of the motor. The motor should slip (decouple) in the case of a stepper or trip out on over torque or encoder error in the case of a servo.

---

The stages I have (yes, these are complete stages) come with 300W AC servo motors. If I use them on X and Y, I will use the servos. I've started a different thread about the servos. If I use them on Z, I will use planetary reduction and Nema 23 steppers.

Are you saying that the physical motor coupling should decouple before causing damage to the ballscrew?

Quote:

---

Originally Posted by Jim Dawson

Just to really put this in perspective, due to a stupid programming error on my part, I ran a tool into my lathe chuck (rapid, 400 IPM)). 2505 ballscrew, 1.8KW servo, direct drive. Wasn't close to the travel limits. The motor tripped out instantly, and no mechanical damage done except the motor coupling may have slipped on the shaft, I'll figure that out in a couple of days when I tear into it. The carriage in this case weighs around 1000 lbs.

---

Please let us know how it turns out. I am interested what the weak link was. Also, had you set torque limits on the servo during setup? How might the outcome have been different if it was a stepper with the same torque?

Here are a few things that got me started on this line of questioning.

If you look at the manual here:

https://www.nskeurope.com/content/da..._MR-Series.pdf

At the bottom of page 15 it says that the maximum input torque is 3.8 Nm. And that is for the heavy modules that I have, although mine came with servos installed.

The Nema 23 steppers I'm considering have a maximum torque at near zero RPM of 2.2 Nm or 3Nm, which would give me between approx 300 and 400 IPM at 0.5G acceleration with a 3:1 gear reduction. That is with an 80lb gantry and no counterbalance being driven by 2 Z axis.

But 3Nm x 3 = 9 Nm, so it's possible to overtorque it if you run into the end stops? Even though the actual torque needed to run the machine at those parameters is less than 1 Nm, or 3Nm after gearing.

But if you look at this manual

https://www.nsk.com/common/data/ctrg...e3162g_bss.pdf

On page B433

BSS1520-2E Is the ballscrew I'm guessing I have, or at least it's almost identical.

This lists a Dynamic load rating of 5070N and a static load rating of 8730N

Using the equation Torque(Nm) = [Force(N) x Lead(mm)] / [2000 x Pie x efficiency]

17 Nm = [5070 x 20] / [2000 x 3.14 x 0.95]

17Nm is a heck of alot more than 3.8Nm....so I'm not sure what's going on there or if I made some erroneous assumptions.

Now if I look at the manual that came with these, it says that the rated thrust force is 300N. I quickly realized that this is talking about the whole system, and that rating has nothing to do with what the ballscrew can handle, but is related to the rated output of the 300W servo motors.

Still not sure what I will do.

What would be the point of adding gear reduction to a ball screw? A preloaded ball screw will eliminate backlash, but the gears will add it back. If you've got a high-lead screw, you can drive it directly with a stepper motor; they like going slowly. If you feel it doesn't have enough torque that way, do what Jim did, and give it some mechanical advantage with timing belts, which don't add backlash.

Quote:

---

Originally Posted by awerby

What would be the point of adding gear reduction to a ball screw? A preloaded ball screw will eliminate backlash, but the gears will add it back. If you've got a high-lead screw, you can drive it directly with a stepper motor; they like going slowly. If you feel it doesn't have enough torque that way, do what Jim did, and give it some mechanical advantage with timing belts, which don't add backlash.

---

The planetary gears I'm considering are very low backlash. Less than 5 arcminutes, or 5/60 of one degree of backlash.

With a 20mm lead that equates to 0.0045mm or 0.00018" of possible backlash from the planetary. I can live with that. I don't think you will do better with belts. Plus, I don't have the space unless I hack up the ends of my nice stages.

I'm not going to drive a 20mm lead directly in the Z axis. That is out of the question for a couple of reasons. This project is mainly going to be an industrial quality 3d printer, but with the quality components I will be using, it will easily double as a decent router.

To get decent mechanical resolution for printing (in the Z axis), with a stepper motor, gear reduction is needed. Also, for a machine that will be running for many hours or days on end, with the Z axis often being not very active, gear reduction will help save on the electricity bill.

I may also mention that if I use the two identical stages I have for the Z axis, that I may use this one for the gantry.....

I pulled this pic off EBay because it's taken apart and you can see it easier, but I have the exact same one.

Attachment 444876

Not sure yet if I want to do that or if I want to go with an all ballscrew machine (Thoughts?)....so this thread might simply be academic at the end of the day. I figure I might as well make something decent though, that could also be used as a general cnc.

The rising gantry moving table design means that the parts that actually need to be accelerated in X and Y can be fairly light in relation to an overall beefy machine.

I'm looking at a working envelope of around (in mm) 750 x 500 x 600 (Z) or 600 x 600 x (500-700) (Z) depending on how I finalize the design.

I plan on eventually having two full on pellet extruders with good acceleration and speed performance. Hence the need for this kind of quality.

Hi,
when I built my mini-mill I used Vexta 23sized 5 phase steppers with original equipment low lash (< 2 arc min) 10:1 planteraies.
They are coupled to 5mm ground (C5) ballscrews. The 10:1 reduction and the low lead (5mm) result in slow rapids but great acceleration and
thrust.

The steppers and gearboxes are superb, after my intinial tuning they have not lost a step in six years. I use genuine Vexta drivers and run them
quite comfortably at 2400 rpm, or 240 rpm at the gearbox output.

I've had two crashes, both operator induced, in six years. I used some cheap spiral cut couplers to attach the gearbox shaft bto the ballscrew. The
couplers exhibit a small amount of trosional flex, which is bad for accuracy, but when a crash happens they shear, a bit like a mechanical fuse.
My machine has a resolution of 1um. The lash of the gearboxes, smaller than I can measure, is 2 arc min which works out to 0.7um linear. The flex
of the couplers add about 4um of 'lost motion' by virtue of the torsional flex. So in practice my machine has a repeatable accuracy/resolution of 5um.

I'm not quite sure why you want to go to the expense of low lash planetaries and then use high lead ballscrews? Why not direct couple a low lead,
say 5mm pitch screw, direct to a low inductance (high-speed capable) stepper or better still a servo? The money you save on a low-lash
planetary will easily get you from a stepper and driver to a genuine AC servo.

Craig

''Are you saying that the physical motor coupling should decouple before causing damage to the ballscrew?''

No, a stepper motor will magnetically decouple if overloaded. That will happen long before the ballscrew is over loaded. You can use Oldham couplings and those will shear under excess load, there are some torque limiting couplings for higher load systems also. With 20 or 25mm ball screws, you could not overload them with a 400 Watt motor under any possible condition even with normal gear reduction ratios.

Frankly I would go with 2505 or 2005 ballscrews and direct couple them to small servos. Most small servos are rated around 5000 RPM. That way you preserve your rapids and thrust, and still have sub-micron resolution. There is really no reason to over complicate this one.

Hi,
according to the manufacturer the stepper/gearbox combination has a holding torque of 8Nm, and when coupled to a 2005 ballscrew its
capable of 1400kgf, so whopping great thrust. No doubt it reduces markedly when at operating speed but none the less very impressive for a small machine.
It has also a very sobering thought that this small machine could crush/sever hands or limbs without even noticing the increased load.

I found however when you crash the peak impulse torque is more like 20-25Nm and thrusts of 3000-4000kgf.

I did not intend that the couplers would shear, but they did when I crashed....and was glad that they did. For that reason I left them in place
rather than chasing that last few micron lost motion.

In more recent times I bought a 400W Delta B2 series AC servo with a view to experimenting and then perhaps replacing one of my steppers to see how
it works in practice. To say that I was impressed is an understatement.....modern AC servos truly rock and leave any and all steppers for dead.

I was so impressed that I bought three 750W Delta B2 series servos and drives, including a brake one one of them, for my new build mill. They will be direct
coupled to 3205 C5 ground double-nut ballscrews. At rated speed (3000rpm) that equates to 15m/min rapids and if I go into field weakening mode (5000rpm)
25m/min.

Craig