Poor Man's Digital Dentist Core XY build

I am building a printer similar to Digital Dentist's build and will use many of his ideas.

I wil take some close up pictures soon, as i have made some changes (READ: Sacrifices) due to lack of a milling machine.


Anyway, here is a video of a potential problem:

[youtu.be]


In the video, I am holding one side of the y grantry still, but as you can see, i am still able to move the other side a little bit. Is that a problem? I assume once i get the belts on it probably won't be a problem, but i wanted to get some opinions on that first! Thanks.

So you are showing it can skew a bit, but the belts will precisely control the movement on both sides, and they in theory should prevent any slop like you are showing. I think it will be fine.

By the way, looks like awesome work, even without a milling machine!!!

Not a problem. Once the belts are in place it will behave itself. However, it will move as you adjust tension on the belts. You will have to check it to make sure the X axis is perpendicular to the Y rails when you're finished tensioning the belts.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Ok, thanks guys

More pictures/video please!

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Here are a few I took yesterday. I need to adjust/redesign the xy interface because the linear rail sticks out in front of the bearings. My plan is to add a second piece of 2020 extrusions between the bearing block and the x rail.






I'm sort of impressed with this. I can move the printed piece side to side to make adjustments that ensure the belts are parallel, then I can slide it back to tension the bets, it tightens with four m5 t nuts.



Edited 2 time(s). Last edit at 05/06/2017 11:44AM by Qdeathstar.

Your Y-rails are on top of the frame. That loads the carriers sideways from belt tension. I wonder, if that makes a difference long term?

There's a piece of t-slot on the X axis, and its ends are anchored on the Y axis bearing blocks providing lateral support. That will prevent belt tension induced side loading of the Y axis bearing blocks.

My printer OTOH has only one end of the X axis rail bolted to one Y axis bearing block. The other end has a second X axis bearing block. That allows the Y axis rails to separate when the machine's frame heats up and expands without adding any side loading to the Y axis blocks, but it also means the belt tension is side loading the Y axis blocks all the time. The Y axis blocks and rails are rated for much higher loads than the belt tension will ever apply to them, so I am not concerned about it. This arrangement is an experiment and I'll be examining prints under a microscope. If I see any quality issues that are attributable the X axis I'll add an aluminum tube to the X axis.

Edited 1 time(s). Last edit at 05/07/2017 07:45AM by the_digital_dentist.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

I was able to finish a version of the effector mount. I haven't fully decided on my hotend it yet so there is definitely going to be changes.... im thinking about either dual e3dV6 or i've already a cyclops i modified to have water cooling. I want to do a direct drive (no bowden) system, however. I have might have some Nimble extruders for the cyclops, and if they get here and work that might be OK. The only issue is now i have a big printer and i really need a .8 nozzle or larger to make practical use of it and they are only available with e3d v6.

-Not bowden
-Dual Extrusion
-different nozzle sizes available

Would be my main criteria in that order. If anyone has any ideas on the front i'd appreciate it!





The motor mount didn't have enough "play" to get it to where i need to be in order to make the belts parallel, so i've got another one printing down that shifts the upper square over. derp. You can also see where i've added some mechanical stops to prevent the rail from sliding off the guides. Had that happen to me yesterday. Not again! (not a big deal, the rails have ball bearing retainers).





Thanks to The Digital Dentist for the idea of the belt mount. However, it seems like on one side, the smooth side of the belt faces its self rather than the the teeth. Is that how you have it DD? Or did I interpret the design incorrectly.












Overall, the belts look parallel, which is good news. I'm waiting on the slowboat to get here for the larger pullys for the idler bearings... only option was to get them on aliexpress and wait forever. Well, unless you wanted to pay $15each for them. I've got an old RAMBO board hanging around so once i get the motor spacers reprinted i might fire it up and see how she chootches!. I've made some hole in the effector mount to attach a spare linear rail to so i can try the pencil thing DD did. I think that is a great way to check for backlash/optimum speeds, since you can easily see the ripple. Though, i expect with the hotend attached you might have to run a bit slower because of the added mass.

Inserting the belt into the clamp is done like this:



Sorry for the poor image quality- I had to shoot with my phone while the printer was running.

You have to keep the "tail" short so that it doesn't prevent the belt's teeth from touching the clamp as the belt enters the clamp. That would cause the belt to not be parallel to the X axis guide rail and would cause all sorts of bad behavior.

The design of the clamp could be modified to allow the tail to extend beyond the edge of the clamp- you'd have to bring the edge of the clamp in a little to ensure that the belt stays parallel to the X axis rail.

Physical stops are always a good idea. I used 5/16-18 carriage bolts to mount flat plates on the frame, and screwed the Y axis linear guide rails to the flat plates. I positioned the carriage bolts and cut the Y axis rails to just fit between them so the carriage bolts act as the physical stops at both ends of the Y axis.

Edited 1 time(s). Last edit at 05/07/2017 11:08PM by the_digital_dentist.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

ah, i see now. I have the belt entering from the front and the tell sticking out the back (closer to the aluminum plate in your picture)

I am starting to get the xy plane of my moving via some steppers. I have got it to move, however i am running into some annoying software problems.

I am using the latest marlin release on a rambo board... (will be getting a duet for the final)


Right now i don't have any endstops (except ones to mechanically prevent the rails from decoupling) so i moved the carriage to one corner of my machine and issued a g92 x0 y0 z0.

From there, i am able to move in the x and y plane forward, but only 90mm. After that, if i keep pressing the up button on the manual control the values of position on my computer screen will change, but the carriage doesn't move... but, if i press move backwards, it will move right away, so my zero is screwed up. Anyone have any ideas about that. I'm sure it's something mismatched in the firmware but i can't figure it out. I checked the bed area settings in the marlin, min_x_value and min_y_value are set to zero but they have no problem going negative....

Are there some config variables that specify the maximum movement in each axis? Maybe they are set to 90 mm...

Make sure the printer's origin is at the front-left or right rear of the machine, otherwise prints will come out mirrored.

Edited 1 time(s). Last edit at 05/20/2017 10:32PM by the_digital_dentist.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Got her running. A bit slow and loud with the stepper whine, but its moving

[www.youtube.com]



Got the bigger bearings for the idlers installed and remade the motor mount to allow all the belts to be paralell as well.... so its looking up

Tomorrow i am going to try to see if I can attach a pen to the machine and do some speed tests.

A few pics swell, got the bigger bearings on

It looks pretty good!

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

I got accomplished what i wanted to today.. I need to figure out the plotter software because it took about 4 hours to figure how to go from .png to .gcode in inkscape!

[www.youtube.com]

I just drew a couple shapes and twisted/tilted them in CAD, then exported the STL. I sliced as a spiral vase with no bottom and exported the gcode. The gcode doesn't know that the machine doesn't have a Z axis so it just moves the extruder around as if the bed plate were dropping with each layer. Tilting and twisting the objects in the vertical dimension translates to spaced lines on the paper.

[vimeo.com]

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

good idea, cause inkscape is a pain in the ass...


I took the speed up as fast as she would suffer without having to adjust jerk/acceleration settings in the firmware.

[m.youtube.com]

The real speed/acceleration/jerk test comes when you run 45 or 135 degree infill- only one motor is moving everything.

It looks like the belts are flopping around a little more than I'd expect. You might want to check the tension.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

i will try some 45/135 movements tomorrow.. The belts are about as tight as i can get them... but i haven't checked to see of the stepper teeth are level with the rest of the idlers beyond just visual inspection... I think i might need to make the belt claps a little tighter, as i have the belt bath as 3.5mm instead of 3mm to give myself a little room, but it's possible the belt will get loose with time.

45/135




Printed at 4000mm/min according to simplify3D.. looks good I think. Their is some slight movement of the frame when it changes directions repeatedly at the corners, but I'm not sure if it's that the frame needs stiffening or just because I have it siting on a carpet... it didn't seem to affect the print too much.

Edited 1 time(s). Last edit at 05/23/2017 12:26AM by Qdeathstar.

If you want to know how much your frame flexes during direction changes, just put some ( painters ) tape in diagonal directions on the extrusions like cross braces.
You'll see the tape rip off or get loose. There are special tapes for such purpose, but a nonstretching tape will do.

I am moving slow. I just got my endstops for x and y hooked up.


I've designed the printer so that i home y first, and then x. Is there a way to change the home all function to home in a specific order?

Also,

There is another annoying issue.

When i home, i have my max build size set to 360mm x and y in both simplify3D and the firmware.

However, when I jog the controls past 360 the machine will not move (expected) but simply3D shows the head position as being past my software endstop (400mm for example) When i go back the opposite direction, the machine moves backwards the commanded amount, except for instead of being at for instance 360-100 = 260mm, simplify3D things I'm at 400-100 = 300. I have to re-home the printer :-/

Is this expected behavior? I never noticed it with my delta, but i didn't really jog the controls that much.

I normally tell the slicer that the bed size is only the size that is accessible to the extruder nozzle. In my printers the bed accessible areas are a little smaller than the full motion that the machines are capable of making. By using just the accessible bed size the slicer will accurately drop prints at the center of the bed. Offsets are used in the slicer if the machine origin and the bed origin are not the same. If the machine origin is away from the bed origin (front left or right rear corner of the nozzle accessible space on the bed), use negative offset values to specify the distances in X and Y.

In smoothieware you can specify the homing order in the config file. If you don't specify anything all three axes home simultaneously. I don't know if you can specify the order in Marlin or other firmwares.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Just a small update today.

I got the endstops for x and y going. I am using hall effect sensors. I am planning to print a small cover to cover the sensors The wiring routes underneath the extrusion so that you don't have to see it


[www.youtube.com]






Edited 1 time(s). Last edit at 06/03/2017 08:52PM by Qdeathstar.

Nothing to write home about, but it should protect the Hall effect sensors... just clips onto the protoboard... I put little holes on the top so I can see the end stop status.... didn't really show up in the picture.

Edited 1 time(s). Last edit at 06/04/2017 09:08PM by Qdeathstar.

The drawing looks interesting..
Reminds me of some wall art in "Star Trek" original series.



Just a sample, I couldn't find a real one.

Why Hall-effect sensors instead of much simpler snap action limit switches (or even magnetic reed switches)?

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]