Coil-Threaded Rods

I was just perusing McMaster-Carr like you do and I came across this ridiculously cheap threaded rod and nuts that would be just about perfect for a z axis screw on a production Wally. (The RepRap Wally has a string drive that is amazingly fast on the z axis.)

The obvious benefit is the cost. The obvious draw back is they are plain steel so they can rust. However, with a thin oil coat they will stay rust free for a very long time. Almost forever. I have used plain steel acme screws in my CNC mill without oil and they show no sign of rust.

For Wally, I don't need to worry about backlash because gravity will keep the nut pressed to one side and the z motion is very slow.

What do you guys think? Has anyone had any experience with them?

Edited 1 time(s). Last edit at 10/03/2013 02:01AM by nicholas.seward.

I like it, and I almost used something similar if only i could get it in a smaller size than 12 mm

All the official RepRaps seem to have threaded rod for the Z stage, so this is well known and well accepted. What's the tolerance on the threads?

If the electronics are replaced with PID controlled servos or gearmotors, threaded rod has lower power consumption because it requires lower holding torque. So, the Z stage would only consume power during the z-adjustments. I'm not sure you could power down a stepper motor between Z moves, since the stepper might be at a microstep, not a full step, which is unstable when powered off.

My main concern with normal threaded rod is that the way it is made makes it unsuitable for high quality movement. I could see a marked quality difference when i switched my machines to the experimental drill bit z drive. It also sped up the drive, removing quite a bit of waiting, given Morgan's 200+ build hight, and max homing.

@qharley: Would this be in the normal threaded rod category?

@annirak: I am sure that the manufacturer would give you a horrible tolerance number but I suspect that the thread would be very self-consistent even if the pitch deviates from the manufacturer spec. I think it would be essential to actually measure. On a side note, we can also tune graceful variations in the pitch away with software. If only we had a distance transducer that could automate that. Maybe something with a laser beam. (For the record, if the thread needs that kind of tuning I am not going to use it. However, when I build my junk bot it will probably be very important to do a lot of clever tuning.)

It would be nice to have a delran / nylon nut for these
Or better an anti-backlash nut.

The Solidoodle uses a lift bed with leadscrew and early customers added anti-backlash fixes.

[www.soliforum.com]

@cozmicray: I think an anti backlash nut would be the wrong decision. They are more expensive. They have more friction, They have a shorter duty cycle. They also don't reduce the backlash in a system were gravity keeps the load on the nut pointing in the same direction.

At first glance it sounds like a good idea but this is one of those weird situations where spending more money decreases performance.

Coil thread rod is nice because it doesn't jam, rarely clogs, and nuts move across it quickly. They've been used for years in screw jacks in the automotive and construction industry. Rust doesn't really affect the nut travel because the nuts fit loosely with a mm or so of slop. The nuts are designed to wobble on the rod. Postion repeatability of a coil thread rod and nut combination is around 3 mm.

So yes, the coil thread rod is great for axis motion, but the coil thread nuts fit really loosely and are horrible to control.

You can print/cast your own coil nuts pretty easily to have tighter tolerances and remove the slop. I have done that before and gotten the position repeatability down to 1 mm. This was fine for that application, but 1 mm off in printing causes issues.

@criswilson10: I am so glad someone here has had some experience. I think I can do better than 1mm because of the 1 directional loading. I would never think about using this for an x or a y but I think a gravity dropped z would be perfect for this.

I also am planning to print out my own nuts.

Seems like these are concrete anchor product with NO Spec for thread quality control.

"They can be used in any fastener situation and excell in use as concrete forming systems."

I would think Z movement may need a bit more precise action?

Z-axis leadscrews on my Ordbot have probably got most of my frustration

But I am sure you will make up a nut and write software to make them work!!!

nicholas.seward Wrote:
-------------------------------------------------------
Maybe something with a laser beam.

Yes, this is called screw mapping, a laser measures the travel at many positions along the axis, and this
"mapping" or error list is refered to when moves are made.

---

Random Precision

@johnrpm: Yes, this is exactly what I was referring to. We could theoretically use junk thread that no one else wants. We just have to get the correction values once. Not that the laser system will be really expensive but we only need one system that can be used for all the machines.

However, when it only costs <$5 for the 225mm ACME thread it is probably foolhardy to go for a cheaper screw that needs an error table.

On the one hand, I would like everything as cheap as possible. On the other hand, I don't mind spending a little to make my life easier.

I am definitely going to get a coil-threaded rod to see just how good/bad they are.