triangulation-free Winch system for 3D printer Z-axis movement

Hi guys, I am pretty new to reprap forums so please bear with me. I can't apparently delete a post I've made in another thread. I think this post fits better in the mechanics category so I will also post it here. If i am breaking any laws on the forum please let me know. I hope I am not

I've been posting for a while (and still do) on Nic Seward's ConceptForge about a Wally without the radial or linear Z and without a threaded rod or leadscrew. I know little is left but in many parts of the world these parts are difficult to find and there are other reasons as well. Over several posts I've come to an idea of a winch that would spool Spectra Line while lifting or descending the SCARA arm on the Z axis. You can read the why's and how's over in Ideas for the Next Wally thread.

The problem with winches is that they have triangulation errors that don't count for a normal winching job, but for 3D printing any triangulation would mean a cvasi-impredictible Z resolution as the spooled line drifts away from the vertical, creating an angle. This could potentially be calibrated in software with some trig functions but as long as I don't know how to modify the software, I tried to think of a mechanical solution.

Because this could eventually be 3D printed, a solution I came up with is a spur gear bar with a threaded profile embedded (or a threaded bolt with spur gear teeth cut along its axis), so that the whole thing can be rotated by a wormgear while also advancing through some nuts. I could afford all this complication in a 3D print and the end result is the spectra line will remain fixed on the vertical (to the anchoring point), while the line is spooled on the threads by the moving "spur-bolt".

To better exemplify i will attach a printscreen I have also posted on ConceptForge, but here I hope more people could pitch-in if interested, over there I mostly pester Nicholas, and I know he has a busy life.
Nicholas said he would be interested in prototyping it and I couldn't be more hapier as I have no real maker's experience (CAD or other engineering exp for that matter; I am also very new to 3d Printing). I can barely conceptualize things.

In the attached sketch, i've taken the stepper motor away from its position so that the spooled line can be better seen. I have the spooled line at the back, while the wormgear is at the front, to prevent interfering with it. Nic posted a CAD render with the wormgear at an angle with the bolt so that the contact is optimal. This could clear up some space at the back so that a position of the stepper with the wormgear pointing to the wall could be possible.

Until Nic gets some time for this I would be interested in getting some feedback on possible issues that this approach would have and what could be some potential solutions given the constrains of the design.

Edited 3 time(s). Last edit at 05/12/2015 08:20AM by realthor.

Looks like you'd have to cut the teeth on the spur bolt pretty deep to avoid the line fouling on the nuts that guide the spur bolt.

---

[3DKarma.com] - suppliers of quality, affordable 3D printer kits and filament for the UK market.

100lb Spectra line is .56mm in diameter (Model: 21101500300Y), so for a 25-30mm 3D printed bolt it wouldn't make much of a difference. I am more concerned about wormgear to straight spur contact, as it's not the most optimal. The best contact is a double throated worm gearing, as exemplified in the following images:


For my design the drum is a straight teeth spur gear so the best I can hope for is a single-throated worm that follows the curvature of the bolt. Also, because of the helical threads, the wormgear will have to be angled a bit to follow the best contact with the threads. Nic Seward has modeled a sketch of such a thing in CAD, you can see it in this thread: RepRap Lander on ConceptForge.

I can actually think that the backlash can be further reduced by designing the threads so that the tilt of the wormgear with the spur bolt would be even steeper so that the profile of the worm can almost fill the gap between teeth. I am sure an experienced engineer would express this better, I will come up with a sketch if I will be able to translate it.

If you're worried about triangulation errors why not just taper the drum to compensate?

I think that it would be similar to do the compensation in software, which would be in fact easier as long as the same trigonometric functions would have to be used. But I tried to recycle the existing code in the firmware and my design only needs some parameters adjusted.
Second reason was that my solution above is a challenge for my engineer within , i haven't seen it used elsewhere and wanted to explore the possibilities of 3D printing.

To be frank, i've seen tapered drums while doing the research about winches but was never drawn to this particular solution. Would be interesting though to think about it but would have to learn trigonometry again.

Edited 2 time(s). Last edit at 05/16/2015 07:43PM by realthor.

I can't seem to be able to edit my previous post anymore. I wanted to add that a tapered drum might not be scalable should i decide to have a longer Z. The Z depends on accuracy of the printer and if accuracy is achieved and the non-Z accuracy is good enough to prevent Pisa-tower prints then the potential of Z is limited only by the length of the "threaded spur gear" spool.

Here's a recent screenshot of how this will work:



Basically a motor with a worm will actuate this spur gear rod (3D printed) which will spool the fishing line that goes upwards, thus rising on the Z. To avoid backlash, although not particularly visible due to gravity, the rest of the spool is already coiled with another line that is anchored somewhere below so that the system is at all times tensioned between two anchors.

This is similar to a Delta-printer Z carriage that is fixed on the timing belt. The timing belt is at all times tensioned so the carriage goes up and down. With my design the Z motor is onboard lifting itself up via spooling the line above while the fishing line below needs to get un-spooled to keep tension. I hope this is clear enough.

I think I will create a thread in the Polar robots section to actually present the whole idea.