Core XY

[Asa DeBuck]

Starting my Core XY builds soon. Is there a good Consolidated source of information for these bots? Lesting the deferent belt layouts, gantry configuration, size lemtation, lead screw placement exc. I find a lot of piecemeal info. I'm especially interested in the different belt configurations with advantages and disadvantages of each.

[Ryan Carlyle]

Peruse the old threads here, there's some decent info: https://groups.google.com/forum/#!forum/h-bot-and-corexy-3d-printers

[Brandon Pomeroy]

In terms of belt configuration, I'd suggest this one:

Belt 1

Belt 2

I haven't seen anyone else use it so far, but I'm employing it in my latest build, and it is so much easier to deal with than the "traditional" belt configuration (For reference, I have also built a bot that uses the belt configuration on the corexy.com site. If I had to do it over again, I would pick my config over that one). It takes a few extra idlers, but it makes clamping the belts a lot simpler. I used two pairs of flanged bearings as idlers, as seen in this picture (Though this was an earlier belt config that ended up not working out, so ignore the belt path!).

Past that, most of the normal rules for printers apply -- You can maybe get away with a little less structure on your cross beam since you're not moving motor mass, but it doesn't affect lead screw placement or things like that.

[Asa DeBuck]

For the Core XY is Sailfish 8 bit still the best way to go?

[Asa DeBuck]

Does the belts having different height off sets, effect the accuracy of printing?

[adam paul]

My coreXY has non intersecting belts connecting to a gantry running on 20x40. I don't think I experience any issue with that. The big weak spot in my printer is the hbp.

[Ryan Carlyle]

None of the CoreXY belt arrangement decisions seem to matter as much as people like to think they do. More/fewer pulleys, intersecting vs planar stacking, etc just doesn't seem to matter much.

Here's the key belt path requirements in my opinion:

1) All belt paths attached to the moving X carriage are parallel to the X motion hardware

2) Belt paths between the Y carriages and stationary points are parallel to the Y motion hardware 

3) The above two points imply this, but it's worth saying: 90 degree net belt deflection through the pulleys on the Y carriages

Belt paths between stationary points can be run at whatever angles you want. The belts can cross or not cross. Really doesn't affect performance. 

Helpful but non-mandatory recommendations:

1) The belt attachment points on each side of the X carriage should be reasonably close together (vertically and horizontally) to reduce skew torque loads on the gantry caused by differences in belt tension. They do not need to be super-close, because belt forces are balanced with CoreXY. Just in the same neighborhood.

2) Don't use a tiny idler radius. Idlers should be bigger than your drive pulleys, if possible. This will A) reduce noise and tension vibration where the toothed side is on the idler, and B) protect the minimum bend radius of the belt where the flat side is on the idler. You DO NOT need toothed idlers or to flip belts around to roll the flat side on the pulleys. (The flat side is actually more likely to damage the belt and needs a larger bend radius, believe it or not.)

[Bonekollector]

Ive been running the original crossing belt config on one coreXY for almost 2 years and Ive never had a problem & get great prints. My newly commissioned larger CoreXY bot is the same way with the exception that the belts are inside the frame rather than outside & so far the print tests have been very good.

[adam paul]

I really like the clean look of running my belt path inside the v slot. Apples and pluot's, but in the end you just end up with good pie.

[John Gelnaw]

On Thursday, January 28, 2016 at 9:36:37 PM UTC-5, adam paul wrote:

I really like the clean look of running my belt path inside the v slot.  Apples and pluot's, but in the end you just end up with good pie.

I'm building a triple C-Bot as well, and I suspect the use of stacked flanged F623zz bearings is too small a radius for the belts, but I'll revisit that bridge once I get the printer running.

I found a document on how to determine the correct radius for GT2 belts, but it was really late at night, and I haven't found it since.  :)

[Asa DeBuck]

Ryan thank for the input as a sidenote I can't wait for your book for your book to come out. Having a good consolidated source of information will be a godsend. Something personally have no problem paying over $200 for.

[Ryan Carlyle]

Well, if I price it at $200, I'll probably sell about ten copies  :-)

[Asa DeBuck]

That may be true but having a cuslodated reference with solid information would have saved me $1000 In time. And hundreds of dollars in materials. Sad thing is most people feel someone else's years of learning instead he should be free to them.

[Lukamfs]

I would most deff buy a book like that for 200$!

[Rich Webb]

I don't know anything about the economics of publishing but that makes me 100% qualified on the Internet.  ;-)

Given the caveat above, perhaps an introductory guide for the new/casual user, explaining such arcana as "What is gcode" and "Tram is not a baby carriage" that's priced to sell and then the $200 volume (yeah, I'd buy one too) with everything *and* sprinkles.

[Bonekollector]

$200 is a lot for a book (generally), but a bargain for training. So I guess it depends on how you look at it.

[Bonekollector]

Meant to say a bargain for training & good reference manuals

[Ryan Carlyle]

If I put everything in one book, yeah, it would realistically have to be somewhere in the $150+ range just to cover publishing costs. And you'd have to be super-serious about 3d printing to buy it. What I'm working towards right now is 4+ volumes in a series, alternating practical-application books with theory books. So if you just want to build a frame and drivetrain that gives good performance, all you need to do is buy volume 1. If you want to get into kinematics and motion control theory, you would also buy volume 2. Etc. 

[Ryan Carlyle]

Incidentally, if anyone is curious, I post book updates and occasional discussion items here: https://groups.google.com/forum/#!topic/3dp-ideas/ldouQeTXtXk

Yes, I've been working on it for a year and a half. I have >100,000 words written, which is a solid book, but I've been writing chapters out of order so it'll probably be another year before I have volume 1 ready to publish. (You know how it is -- day job, family, building printers, etc.)

[Steven Butterfield]

Excuses excuses :) I for one looking forward to getting a copy. Oh how do I post things to your group? There have been some things I would like to comment on from time to time. It's intersting stuff.

[Ryan Carlyle]

Request to join, we'll approve you.

[William Cook]

I had never seen your group before, and boy is it a treasure trove of info. I will request to join also if that's OK. I promise not to comment unless I have something valuable to add. I have designed and built my own large delta and gone through close to 100kg of filament in the last year (PLA, PETG, PolyCarb), and I'm working on a CoreXY design.

[Ryan Carlyle]

Anybody more or less familiar with 3D printers is welcome to join and post, we just don't intend it to be a newbie help group like most 3DP forums.

[William Cook]

Gotcha, that's perfect. There's a very high signal/noise ratio on there.

[Brandon A.]

We?  I always assumed it was your personal GG, are whosawhatsis and other common posters also mods?

@William  "There's a very high signal/noise ratio on there." <-- well said =]

[Ryan Carlyle]

I made it and obviously have posted the majority of the threads, but the ultimate goal is to have a focused community for people to talk shop on technical subjects and exchange ideas. I happen to have a lot of ideas that I don't have time to pursue, so I post a lot. Most of this sort of stuff has historically been put in blogs, but those aren't good for collaboration or discussion.

The RepRap forums have a couple nominally-similar areas, but the large amount of bad info posted there makes my eye twitch. I can't correct it all, so I mostly stay away for the sake of my sanity.

[Brandon A.]

That aligns with my observation, I was just curious.  Collective "we" as in all participants of the group.