Fishing line traction drive

Sorry for the blurry image.

This is a design that is inspired by/based off the ideas discussed in this old thread.

The drive shaft (center) is a 13mm od aluminum m5 spacer, and the idlers (corners) are stacks of 608 ball bearings. I am using 1mm dyneema line.

(top-down: red = top loop, blue = middle, green = bottom)

My prototype has five loops around the drive shaft, for the equivalent of about three complete windings. I was not able to make the line slip when applying moderate hand tension. My only alteration to the off-the-shelf hardware was to roughen the surface of the spacer using a coarse file.

Pros (compared to spools):
-cheap, no specialized vitamins
-unlimited travel length
-small diameter (higher resolution)

Cons:
-bulky
-could possibly slip, in theory

I plan to use this design on an actual printer, and will update this thread with the results.

I like it - clever design. Idlers on the corners of the motor and you have a compact setup. This concept will be handy with a Cablebot design ( [astro.smugmug.com] )

---

You only learn when you change your mind.
[www.deltaprinter.co.za]

I wish I could visualize this a little bit better. I see in the photo that one side of the assembly has 4 bearings and the other 3. I guess this is to allow for a downward spiraling. On the side we cannot see, what is the arrangement?

You say your prototype has 5 loops around the center, but in the diagram I only see three. I assume that the diagram is simplified, and that you actually have an additional layer in your implementation.

I also assume that you have those bearings so tightly bolted together that the groove between them will not let the string slip through

I guess if I try to cobble something together that some of these questions will be answered.

Quote
jbernardis
I wish I could visualize this a little bit better. I see in the photo that one side of the assembly has 4 bearings and the other 3. I guess this is to allow for a downward spiraling. On the side we cannot see, what is the arrangement?

You say your prototype has 5 loops around the center, but in the diagram I only see three. I assume that the diagram is simplified, and that you actually have an additional layer in your implementation.

I also assume that you have those bearings so tightly bolted together that the groove between them will not let the string slip through

I guess if I try to cobble something together that some of these questions will be answered.

You are exactly right on all accounts. The bearings all spin at the same rate, so they can be clamped together tightly; the top and bottom plates only contact the non-rotating part of the bearing to provide clamping force. In addition, the line I am using is relatively thick (1mm), so the bearings would have to be very loose for the string to wedge between bearings anyways.

Here are some quick-and-dirty diagrams I threw together that should hopefully clarify things:



Some of the windings drawn in:


Forced-perspective head-on view from one end:

I ordered some bearings to give this a try.

I guess you're using an m5 spacer to fit over the 5mm stepper shaft. Did you insert a grub screw to anchor it?

One thing that concerns me about a spacer is that they are not manufactured to have a snug fit - they actually have a bit of clearance for the intended screw size. At McMaster, for example, the M5 spacers are 5.3mm ID. A grub screw would force this off center which would introduce a bit of wobble. I'm half thinking of trying the spools I had intended to use with the spool drive as the drive shaft.

Edited 1 time(s). Last edit at 10/14/2014 10:51PM by jbernardis.

Wow, I should probably update the first post with all the things I've forgotten to note

The spacers I'm using don't have holes for set screws, and I don't have the tooling to drill/tap holes, so I came up with a printed solution:


The slots are for m3 nuts, and the motor shaft and the spacer are each secured into the coupler with two screws. Though this should theoretically center the spacer over the motor shaft, I wrap the portion of the motor shaft that rests inside the spacer with Kapton tape to the diameter of 5.3mm to make sure it is properly aligned.

Can't wait to see your take on the design! I wish I had more time to play with hardware...

Your illustrations have been extremely helpful.

My bearings are due to arrive on thursday, but in the meantime, I've designed the mounting plates and printed one off. Once I have it in hand, I can play around with how it's going to fit into the printer. I haven't totally abandoned the spool approach yet, but given the stage I'm in with building the printer, now is the time to experiment with alternatives.

The traction drive works really well. I just printed a couple of test pieces and installed them on my printer - on one side only - and it was flawless. I was worried that the string would jump out of the groove on the sides where it changes level, but I couldn't get it to. I am using the original spool as my drive shaft and think I might stay with it - it too works well, and it is well engineered so I shouldn't see any wobble.

very clever thank you for sharing

Hi all,
can someone explain me what is the advantage in using such a system instead of the single spool?
I've read both this and the original thread but I haven't understood the reason this idea should be better.

Thanks,
Paolo

I'm actually in the process of moving back to the spool approach - for 3 reasons. 1) This approach is a nightmare to string, 2) unless I kept the system under extremely tight tension, I found that the string would pop out of the grooves too often, and 3) When I hit the end of travel, the string would slip. I know that I should never reach the end of travel, but if it would slip under these circumstances, I don't have confidence it wouldn't slip under others.

I was very close to getting the spool approach to work. I wish I could find larger spools, but for now, I am going with printed spools. I am a mentor on my local high school FIRST robotics team. The teacher has 2 lathes and a milling machine. I might ask him if he could make me a few aluminum spools.

@jbernardis
I wizzed a couple of spools out of 20mm Aluminium round stock, up on a lathe at work in just over an hour to make my experimental CoreXZ. I made mine 16mm dia which is not going to help you unless you make them even longer. Hardest part was drilling the offset hole. So the school teacher should be able to make some for you in no time.

Quote
jbernardis
I'm actually in the process of moving back to the spool approach - for 3 reasons. 1) This approach is a nightmare to string, 2) unless I kept the system under extremely tight tension, I found that the string would pop out of the grooves too often, and 3) When I hit the end of travel, the string would slip. I know that I should never reach the end of travel, but if it would slip under these circumstances, I don't have confidence it wouldn't slip under others.

I was very close to getting the spool approach to work. I wish I could find larger spools, but for now, I am going with printed spools. I am a mentor on my local high school FIRST robotics team. The teacher has 2 lathes and a milling machine. I might ask him if he could make me a few aluminum spools.

Hey sorry to ressurect an apparently dead thread but I am interested in a spectra drive solution that works and was wondering if there is any development in this area from you guys. I still find the spectra a good solution for applications when routing the cable would render using a timing belt impossible or a stretch at most. But I need to resolve the walking coil via some sort of mechanism and this continuous resetting looks the most simple.

Thanks.

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff