CoreXY belt path - new twist? (pun intended)

[makertr...@gmail.com]

Hi all,

Note:  This only applies if you are using CoreXY with the belts intersecting.

I've always had a challenge keeping the belts from rubbing eachother at the intersection in a corexy configuration.  Adding spacers and washers galore did the trick, but adding more of these seemed to complicate matters.  Then, I discovered recently that if you flip the gt2 belts at the intersection,  they avoid eachother altogether.   Specifically, at the intersection, the belts smooths sides are now facing eachother. I can use less washers and spacers now and the belt travel is quieter.

Thanks, Chris

[iquizzle]

That's good to know. I'd like to start experimenting with a corexy arrangement after I finish building my shapeoko 2.

[makertr...@gmail.com]

It does take a few more parts, but I find it just slightly more difficult to setup than the hbot config.

I highly recommend going CoreXY since it removes the racking problem that is common with non-crossed belt hbots.

Dealing with the belt rubbing was the part that I dreaded with CoreXY.  Now that this has been solved with the belt twist at the intersection, I find the setup simple.

[iquizzle]

Have you tried using spectra line? To me, that seems like another good potential candidate for corexy.

[makertr...@gmail.com]

I have tried spectra, but found that GT2 is more usable by customers and parts are readily available.  I, personally, like the idea and compactness of the spectra.  There is also the perceived value of reinforced belt vs. fishing line by some customers.

[Rupin Chheda]

Can you post a photo of how this flip is done? are you menaing to say that at the point of intersection, the two belts have been turned by 90 degrees?

[Chris Welch]

I'm building my setup with spectra line and 625vv bearings :)

Once I get a decent CAD made up, I'll post some pics.

[Ryan Carlyle]

I like those bearings. Where so you get your spectra line?

[Chris Welch]

I got 20 bearings (625vv) and 100m of 80lb spectra line off ebay...

Was about $26us.

It is cheaper for me to go this route over the belts and pulleys

(My printer will have a build volume of about 300mm3)

that would be allot of belt for a CoreXY!

[Ryan Carlyle]

Yeah, I got a couple goofy design ideas that will double or quadruple the amount of belting, and put in extra twists where belts would cause problems. Spectra line seems like the obvious solution.

What are you using for a motor capstan pulley?

[Chris Welch]

I'm not sure what I'm going to do yet...

I think I will be making my own pulley (I have access to a lathe)

- Ultimately, a blank pulley is what I have in mind.

I've seen a few others just use a GT2 belt pulley... not ideal, but I guess its working for them.

Are you going to use the line or the belts? If Line, what are you going to use?

[Ryan Carlyle]

Hadn't planned it out that far yet. I'll probably do the initial build with GT2 belts so I can share some spares with my R2x. Then maybe modify to a line system later when I do my special design that requires a lot of 90 degree belt twists. I'm worried about slipping on the line drive, but I know people make it work, so I guess I just need to do some research on how all that works.

[Chris Welch]

This is all I have so far, nothing extraordinary...

What do you have planned with all the twists?

[Ryan Carlyle]

One option is a combo capstan / tensioner system between two multi-wheel pullies, like shown here:
http://www.pcwirelines.co.uk/machine.php?id_pcl=37
These can develop unbelievable line pull but you'd have to lathe it up yourself.

Other option I was thinking is a standard curved shipboard capstan profile like this:
http://www.qrbiz.com/product/1157829/10-Ton-Electric-Capstan.html
Problem with these (and most designs) is that the line has to slip sideways when the wraps travel to the end of the pulley.

[Ryan Carlyle]

Main thing is a 3-axis gantry that I call CoreXYZ. It's pretty crazy impractical but it'll be fun to build.

Gotta get a straight CoreXY bot built first though.

[Ryan Carlyle]

I like this:
http://3dprinterhell.blogspot.com/2013/06/a-better-filament-drive.html?m=1

[Elliot Foster]

Apparently, we can't respond to posts on this group via email..

I replied off-list that I printed the pulleys involved with this and the system seems to work well.  I printed them in PLA, which is likely not a good idea for long-term usage, but was a good way to validate the concept.

If/when I get around to printing the spool of Talman nylon, I might try again using that for longer-term usage, or maybe even ABS.  Of course, if/when I get more time, I may even finish https://github.com/elliotf/corexy-printer.

Some pictures of what I came up with for the filament drive are attached:

[iquizzle]

Thanks for posting the pictures Elliot. I'm interested in testing a spectra line drive system for its low cost and because it simplifies the corexy crossover. What do you think about moving the idler out 90 degrees from to drive? Seems like that might help keep the spectra line from rubbing the part.

Also -- updated the group settings to allow email response. Sorry about that!

[Elliot]

"moving the idler out 90 degrees from to drive" -- I'm not sure if this is what you're talkig about, but the idler is actually on the far side of the drive pulley in my version.  This isn't ideal, since there's less filament contact on the drive pulley.  The idea I currently have is that the to/from filament path is a wide (~30deg) angle, rather than the one seen here which is 0deg.  The idler would sit inside of that 30deg angle with plenty of room to spare.  Of course, you're probably talking about something else.

Another thing that I had in mind was to use a larger diameter drive pulley and use a 0.9deg (400steps/rotation) stepper motor, like ultimaker is using on their direct-drive 3mm extruders:

https://www.sparkfun.com/products/10846

This would allow a slightly larger drive pulley diameter, which would reduce the number of loops needed to ensure good traction, while still having high enough steps/mm.  Hopefully the motor would still be strong enough given the higher torque requirement due to the larger diameter pulley.

Spectra seems like it has real promise, especially with less conventional designs.  If you look at my corexy design, the motors are actually on the back of the printer rather than the top, allowing for more compactness.  So far, anchoring and tensioning the darn stuff is the hardest part I've found.  I bought some guitar machine heads to do the tensioning, but those are (relatively) heavy.

--
You received this message because you are subscribed to the Google Groups "H-bot and CoreXY 3d printers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to h-bot-and-corexy-3d-...@googlegroups.com.
Visit this group at http://groups.google.com/group/h-bot-and-corexy-3d-printers.

[Ryan Carlyle]

Rule of thumb for reel design is that you need 5 wraps on a spool for friction to prevent any line tension from reaching the termination. So I think 5x360=1800 degrees of line contact would be a safe slip-free upper bound. For a non-crossed arrangement, that's 10 half-wraps on the drive pulley.

[Elliot]

I would hope/imagine that the number of wraps would be dependent upon
the diameter, since really you're looking for a surface area (contact
patch). Five wraps around a pin/needle wouldn't give you much grip,
but one wrap around a wine barrel would.

I can't seem to find it, but someone experimented with a
large-diameter single wrap pulley drive. I can't remember what the
results were, but the steps/mm would be terrible, and the torque
required would also be substantial (I would assume.)

On Sun, May 4, 2014 at 8:48 PM, Ryan Carlyle <temp...@gmail.com> wrote:
> Rule of thumb for reel design is that you need 5 wraps on a spool for friction to prevent any line tension from reaching the termination. So I think 5x360=1800 degrees of line contact would be a safe slip-free upper bound. For a non-crossed arrangement, that's 10 half-wraps on the drive pulley.
>

[Elliot]

Ah, I found it. By the same author/designer as the figure 8:

https://www.youtube.com/watch?v=_e7yy9u6QmA

Not condoning this design, but I think it's interesting.

[Ryan Carlyle]

It's been a long time since I've looked at the math on this, but my understanding is that the drum wraps case is analogous to the way friction force only depends on normal force and coefficient of friction, not on contact area. A larger contact area with lower pressure creates the same friction as a smaller contact area with a higher pressure. As long as force is constant, friction is constant. 

To deflect the path of a tensioned wire/rope with negligible stiffness by 0-180 degrees, the total force required is a function of the total angle of deflection and the line tension. (You can easily see this in a free body diagram.) The size of the drum/pulley doesn't change the free-body force required to change the direction of the line. The tension vectors are identical, irrespective of the pulley size. So likewise, the size of the drive pulley should not affect the friction force to the line. A bigger pulley will result in less contact pressure and a larger contact area, but the same total normal force, and thus the same friction force. 

So all that should matter to line grip is the coefficient of friction between the line and pulley, times the total deflection angle (= wraps * arc length per wrap). This would not be particularly hard to calculate after doing some simple tests to determine coefficient of friction, but trial and error works fine too. Just run the motor pulling against a fixed object and make sure the stepper stalls before the line slips. 

Check this:

http://en.wikipedia.org/wiki/Capstan_equation

> To unsubscribe from this group and stop receiving emails from it, send an email to h-bot-and-corexy-3d-printers+unsub...@googlegroups.com.

[Chris Welch]

Would something like this work? (I only drew one side)

The idea is that the large wheel will only rotate about 350deg, pulling the cable the distance required.

This way, you do not need to worry about the cable walking around the pulley and there is no slippage, as the cable is secured to the pulley.

No mechanical advantage is lost and no backlash is introduced. 

The downside being that the wheel takes up more space and adds a little more rotational mass.

Thoughts? (before I go any further with this!)

  

[iquizzle]

I wouldn't be concerned with the rotational mass nearly as much as the size of the pulley. The circumference of the pulley will have to be at least as big as the corner-to-corner distance of your work area. It might get awkwardly large.

I think I agree with Ryan's explanation. Number of wraps and normal force determine the friction, not diameter (assuming all other things are equal).

[Ryan Carlyle]

Securing the cable to the pulley is an interesting idea, but yeah, the pulley would have to be huge. 

On the "giant pulley" train of thought, why not just give it three wraps and let it walk a little bit? The line will walk ~1 line width per pulley rev, so a giant pulley that makes fewer revs will walk a lot less than a small pulley.

[Chris Welch]

I'm going to scrap the large pulley.

And play around with this idea for a bit...

On Saturday, May 3, 2014 5:18:17 PM UTC-7, Ryan Carlyle wrote:

I like this:
http://3dprinterhell.blogspot.com/2013/06/a-better-filament-drive.html?m=1