Welcome! Log In Create A New Profile

# CoreXZ

Posted by nicholas.seward
 Re: CoreXZ July 25, 2014 09:40AM Registered: 7 years ago Posts: 8
If I may contribute to this discussion,

I will be building a CoreXYZ soon, based off the CoreXZ, however I'll be adding the Y axis to the end of the XZ mechanism, I see no issues, do you?
What steppers are you using for yours? Microstepping ratio?
Have you seen this design:

Either way, your CoreXZ has more pulleys to increase torque yes? Using the above design, and my own Y addition, there are a total of 3 spools and 15 pulleys.

I'm aiming for a build volume of 200 * 300 * 300 mm
I will be making my version with Nema 17's at 48 Ncm torque , 0.9 degrees per step, 1/2 microstepping and anywhere from a 1/4 speed reduction pulley. (5 mm spool radius, 20 mm pulley radius).
This should in theory yield 40 um resolution (39.26990816 um), based on the equation:

Step Angle * Microstepping * Speed Reduction * Pulley Radius * Pi / 180 = Minimum Movement Value
Stepper Torque * Inverse Speed Reduction / Pulley Radius * 1000 = Applied Force

Is a speed reduction pulley a good idea or should I just go with 1/4 microstepping? The force in any direction will be no less than 96 N based on the above.

Edited 1 time(s). Last edit at 07/25/2014 10:15AM by BentCookie.
 Re: CoreXZ July 27, 2014 02:09PM Registered: 8 years ago Posts: 7
@nicholas and @bentcookie What are the equations of motion for your machines? I am looking a developing a platform that utilizes this concept but in a polar coordinate system. So all of my z movement would be from one side (with respect to your designs) and my x movement would be in my "r" direction.
 Re: CoreXZ July 27, 2014 03:00PM Registered: 9 years ago Posts: 979
X=x-3z
Z=x+3z
Y=y

The capitals are steppers and the lowet case letters are coordinates.

I would be interested in a sketch of what you have planned. I can't visualize it yet.

ConceptFORGE
Wally, GUS Simpson, LISA Simpson, THOR Simpson, Sextupteron, CoreXZ
 Re: CoreXZ July 27, 2014 05:26PM Registered: 8 years ago Posts: 7
What program do you use to sketch out your drawings in this post? I can use that to draw up my current concept idea. Thanks for the movement info!
 Re: CoreXZ July 27, 2014 05:44PM Registered: 9 years ago Posts: 979
I use AutoCAD and the post process in inkscape. There are many free 2D cad programs if you don't have AutoCAD. Hand sketches are also welcome.

ConceptFORGE
Wally, GUS Simpson, LISA Simpson, THOR Simpson, Sextupteron, CoreXZ
 Re: CoreXZ July 28, 2014 02:22PM Registered: 8 years ago Posts: 7
I'm almost done with the model for the concept, once it is completed I will post a picture or two showing what I'm talking about. I'm currently looking at roughly 8 pulleys. I did have a question regarding the spools for the wiring, what spools did you utilize on your machine? And would you by chance have dimensions or a CAD model I could use? I want to make sure I have my design modeled correctly and I'm having trouble finding a good example to use.
 Re: CoreXZ July 28, 2014 02:30PM Registered: 9 years ago Posts: 979
[www.tridprinting.com]

18mm diameter body. 25.5mm long

However, I would prefer 14mm or 16mm for better resolution.

ConceptFORGE
Wally, GUS Simpson, LISA Simpson, THOR Simpson, Sextupteron, CoreXZ
 Re: CoreXZ July 31, 2014 12:39PM Registered: 8 years ago Posts: 790
I know it's post hijack, I'll remove this after if you'd like.
I need some design critique as I have finally put this together after months of research and still another month to go waiting on parts and building it. For the framing and linear motion is what I'm mostly asking about, but if you see anything else worth noting, please note it.

Realizer- One who realizes dreams by making them a reality either by possibility or by completion. Also creating or renewing hopes of dreams.
"keep in mind, even the best printer can not print with the best filament if the user is the problem." -Ohmarinus
 Re: CoreXZ August 06, 2014 12:52PM Registered: 8 years ago Posts: 790
Question to the master designer: how would you implement this as a single axis for Z bed movement?
My feeble attempt was adding another layer of stringing [or in my case belting] so that it would be 4X the resolution to make a flat 10micron 1/4 step.

Used on one side and evened out using the QuadRap's bed leveling.

If you could somehow shove it down to one motor so that it could be used on both sides, that would be wonderful. Don't worry about specific parts, I can adapt a basic concept to my printer and parts.

Teach me your ways

Realizer- One who realizes dreams by making them a reality either by possibility or by completion. Also creating or renewing hopes of dreams.
"keep in mind, even the best printer can not print with the best filament if the user is the problem." -Ohmarinus
 Re: CoreXZ August 06, 2014 07:31PM Registered: 9 years ago Posts: 979
@MrDoctorDIV:

Here is a rough idea of how I would go about it. There are obviously many variations. As shown it is 3:1. It would be easy to make it 5:1 or 7:1. (any odd number) This is essentially the QuadRap solution made to be easier to use with a laser cut machine.

Each back dot could be a plastic block that is clamped to the string. It would be easy to have this block loosely screw into the top of the bed with thumb screws for quick bed leveling. One of those blocks could have a guitar tuner or other tensioning device.

If this is done right, you only need two rails for the Z direction.

Edited 1 time(s). Last edit at 08/06/2014 07:34PM by nicholas.seward.

ConceptFORGE
Wally, GUS Simpson, LISA Simpson, THOR Simpson, Sextupteron, CoreXZ
 Re: CoreXZ August 06, 2014 09:24PM Registered: 9 years ago Posts: 979

This may be a more sane arrangement. I made it a 7:1 for fun but I probably would stop at 5:1. The green pulley is the drive pulley and it can be located at the top or the bottom depending on your goal.

I got rid of the crossing at the bottom to make it easier to envision electronics and such fitting under the printer.

Again, all the holes for pulley mounting could be on just 3 pieces of flatstock making this a great approach for laser cutting. If you are going to hand drill, all the holes from the right board would line up with the holes on the left after you flip the board over.

Blah, blah, blah. Thanks for asking. I have been meaning to solidify how I want to do this for another project.

ConceptFORGE
Wally, GUS Simpson, LISA Simpson, THOR Simpson, Sextupteron, CoreXZ
 Re: CoreXZ August 06, 2014 10:20PM Registered: 8 years ago Posts: 31
Seems like the left side of the reduction pattern pulls up hard, and the right side pulls down hard.
The (logically separate) leveling system cables will oppose this, ideally, but it still seems to me it would be better to align the upward and downward regions so their forces approximately cancel out. That reduces the tendency to twist the build plate/box, and reduces the tensions in the leveling cables.
Dave
 Re: CoreXZ August 06, 2014 10:26PM Registered: 9 years ago Posts: 979
@DaveGadgeteer: Good catch. As drawn this will cause an S curve on that edge of the build platform. It would be a simple matter to balance this out.

ConceptFORGE
Wally, GUS Simpson, LISA Simpson, THOR Simpson, Sextupteron, CoreXZ
 Re: CoreXZ August 07, 2014 12:25AM Registered: 8 years ago Posts: 790
What I meant by one motor was then using one on both sides. Using belts would make using both sides with one motor undesirable.
Here's my next adaptation, this time taking the time to show my full idea. Green points are motors. Any ends of the belts could be used to tighten/align.

Edited 1 time(s). Last edit at 08/07/2014 12:28AM by MrDoctorDIV.

Realizer- One who realizes dreams by making them a reality either by possibility or by completion. Also creating or renewing hopes of dreams.
"keep in mind, even the best printer can not print with the best filament if the user is the problem." -Ohmarinus
 Re: CoreXZ August 07, 2014 01:05AM Registered: 9 years ago Posts: 979
Using two steppers adds cost and doesn't really solve anything. As drawn the bed can get out of level.

ConceptFORGE
Wally, GUS Simpson, LISA Simpson, THOR Simpson, Sextupteron, CoreXZ
 Re: CoreXZ August 18, 2014 08:02AM Registered: 7 years ago Posts: 26
I did not want to create a separate post at it is related, although it is more for regular coreXY system, so I will post it here.

Obviously it is best to route lines in one plane, but it also reduces print aria, so I though routing line in two planes. This way line system should have minimum impact on printing aria. What do you think about it?

Also I have read that some people are using 50kg/100lbs rated fishing line for their systems, so it get me wondering just how much tension is needed for line based systems, because you also need strong enough pulleys/bearings.
 Re: CoreXZ August 18, 2014 09:09AM Registered: 9 years ago Posts: 979
@Endlane: It will be fun to see what people work up for variant machines.

I have to note that I contemplated off plane solutions (for months) but kept the 1 plane for flexibility. Within that framework I have gotten the machine to a very comfortable width. 475mm frame for a 225mm range. That is comparable my Printrbot Plus. (425mm frame for a 200mm range.) I am sure I could do better but simplicity dictated that I keep one plane.

One thing that can be done is to use a 2 parallel planes solution. One plane for each motor. That way bearings can share bolts.

I usually have about 5lbf of tension in my strings. Even that is more than necessary. I built the prototype and did the initial tensioning and tuning. I haven't adjusted it since. I have transported it in the car many times and have printed 5kg+ of PLA. (The tension has gone down but is still more than enough.) I store it with the tension applied. No ugly suprises yet.

ConceptFORGE
Wally, GUS Simpson, LISA Simpson, THOR Simpson, Sextupteron, CoreXZ
 Re: CoreXZ August 21, 2014 06:26AM Registered: 9 years ago Posts: 248
any firmware available?
 Re: CoreXZ August 21, 2014 10:09AM Registered: 9 years ago Posts: 979
Yeah. PM me your email and I can send you my hackish version to get you started. I (or someone else) needs to make a more general solution.

ConceptFORGE
Wally, GUS Simpson, LISA Simpson, THOR Simpson, Sextupteron, CoreXZ
 Re: CoreXZ August 22, 2014 07:16AM Registered: 8 years ago Posts: 903
 Re: CoreXZ August 22, 2014 09:22AM Registered: 9 years ago Posts: 979
A quick skim of the code from my phone seems to indicate that it doesn't account for the 3:1 reduction in the Z. This is complicated because some people may want other ratios. 5:1 or 3:2.

ConceptFORGE
Wally, GUS Simpson, LISA Simpson, THOR Simpson, Sextupteron, CoreXZ
 Re: CoreXZ October 02, 2014 09:34PM Registered: 9 years ago Posts: 1,320
I was so inspired by this thread that I decided to abandon the design direction I had been going and try to implement this. I had a lot of aluminum extrusions laying around, plus some V wheels from openbuilds from an earlier abandoned project, and decided I would use those in the build.

Currently, I only have the X and Z axes built, but I have started experimenting with moving things around and I am very happy with the direction things are going. I still have a lot of work to do, but I thought I'd share some photos of where I'm at.

There are two things that bother me right now. 1) the strings are in contact with one another at the bottom center. I'm worried that the constant rubbing will weaken them. I had thought that I could just offset the pulleys different amounts, but the string walks along the length of the spool, so I don't think that would work. I was next thinking that I would just take a short length of any diameter smooth rod, and mount it so it is vertical and between the strings. It would keep them apart. They would rub on the bar of course, but I would hope that it wouldn't be as abrasive as the other string.
2) the string walks across the spool as the axes move. I've seen it bunch up against one side or the other. One thing I could do here is to wrap the string fewer times around the spool. I think right now it's wrapped 4 times, but perhaps I could get away with fewer.

Anyway, here are the pictures:

 Re: CoreXZ October 02, 2014 09:46PM Registered: 9 years ago Posts: 979
Sweet! I can help you with the spool. You aren't using it quite right. Basically, you put the string through the off axis hole and then wrap the two free ends in opposite directions until you meet in the middle. [trains.socha.com]

ConceptFORGE
Wally, GUS Simpson, LISA Simpson, THOR Simpson, Sextupteron, CoreXZ
 Re: CoreXZ October 02, 2014 11:20PM Registered: 9 years ago Posts: 1,320
Thanks for that. I was wondering what those extra holes were for. I would guess that I need enough string on each "end" to reach to the furthest point of the axis.

I have a question though. If I'm doing it like this, shouldn't I have a tensioning device (guitar tuning peg) at each end of the string? If I don't, the tensioner would need to be able to pull the string through the off axis hole to balance the tension between the two sides, and it doesn't seem to me that it could do that.
 Re: CoreXZ October 02, 2014 11:50PM Registered: 9 years ago Posts: 979
Fill the spool up with string. This will give you plenty of travel and the string will help track itself.

Additionally, when you tension it will just rotate the spool a bit. As you have it now, tensioning both tuners will tension the system. Tensioning one will tilt the gantry. No change in your system is required.

Also, I wouldn't worry about the string rubbing against itself. Low force, low friction. The pulleys probably will cause a failure first. (High force, tight bend)

ConceptFORGE
Wally, GUS Simpson, LISA Simpson, THOR Simpson, Sextupteron, CoreXZ
 Re: CoreXZ October 03, 2014 09:51AM Registered: 9 years ago Posts: 1,320
When the spool is filled like you say, as one side pays out, the other side takes in. This means, to me, that if I start with the spool half filled from each side, I should start with the gantry at about the middle of its travel and string from that position.
 Re: CoreXZ October 03, 2014 12:07PM Registered: 9 years ago Posts: 979
 Re: CoreXZ October 04, 2014 12:55AM Registered: 9 years ago Posts: 1,320
I had a lot of trouble wrapping the strings this way. Because of the size of the printer, I had to wrap more string than the spool had capacity for. The string that was paying out kept getting fouled by the string coming in. I am currently in the process of printing a set of self-designed spools. I probably won't actually get a chance to try them out until late on saturday, but either they need to work, or I need to find larger spools.
 Re: CoreXZ October 04, 2014 01:27AM Registered: 9 years ago Posts: 979
pi*d*wraps should be greater than x_travel + 3* z_travel. What is the length of travel?

There are three ways to fix this. Larger diameter but that hurts resolution. Longer pulley but that adds to the triangulation error. Thinner string. Depending on how close you are it might be worth it to try lighter test string. (40lb test with 5-10 lbs of tension should do the trick. The 60-100 lb test that we usually use is over kill.)

Printed spools that aren't finished will probably add 100-200 um of error to your positioning. It will be subtle over the 50mm period of the pulley so it will still let you make amazing prints.

ConceptFORGE
Wally, GUS Simpson, LISA Simpson, THOR Simpson, Sextupteron, CoreXZ
 Re: CoreXZ October 04, 2014 11:41AM Registered: 7 years ago Posts: 127
Why did you choose to use guitar keys instead of springs to maintain tension at the ends of the wire?
Sorry, only registered users may post in this forum.