Separate names with a comma.
Unlock hidden features. Sign Up for Free Today!
Discussion in '3D printers' started by Giuliano M, Oct 24, 2014.
Discussion in '3D printers' started by Giuliano M, Oct 24, 2014.
H-Bot with CoreXY movement, above-rail gantry and a large 400x400x400mm build volume.
Giuliano M published a new build:
Read more about this build...
Just now opened your SKP. Thanks for that.
Interesting how you plan the 4 corners of the z axis.
Regarding your use of the V Slot Gantry Plate running on a 40x20 x axis - I've used the Mini V Plate running on a 20x20 x axis on two designs with no problems. If you are going to use a pretty much standard extruder rather than something heavy, a 20x20 is plenty rigid and vibration free. Are you planning the V Slot Gantry Plate instead to provide mounting for the belts?
Yeah, E3D v6, possibly two via Bowden.
I may design another plate that uses the Mini Vs like the C-Bot does.
Also, one thing I dislike about the mini plate on the Z gantry ends os the plate itself... Its not a metric thickness!
Going to make cutting those bars to length a bit odd with a fraction of a mm.
You could take a look at how I made my Z axis. It could help.
I'm trying to avoid using rods as rails in mine.
I have a few ideas in the works..
Great thing about having the SKP model of the Mini-V plate is that it's pretty easy to resize by 20mm on each side.
New gantry plate is 70x70 mm, and double sided - two plates.
They'll be printed parts... but it sure would be nice if OpenBuilds made plates for 40 and 60mm...
Working on the layout of the CoreXY belt paths.
Continuing with the design, I've adjusted the thickness of the modified gantry plates so that the total front-to-back dimension is 40mm...
This makes the plates slightly thicker when printed, and intended to be used with the 40mm low-profile bolts.
The bolt heads are recessed on the front, and the nuts are partially recessed on the back.
With the spacing on the front/rear of the carriage an even 40mm, it should make it easier for referencing the Y-offset for dual extruders, one on the front and one on the back...
Boxed enclosure with 1/8" acrylic sheets, side panels are 550 mm wide and 610 mm tall.
I plan to add supporting pieces to the frame to hold it in place.
I've got the CoreXY belt paths and idlers pretty much set in place:
I'll probably optimize the mounting blocks with roundovers on some of the edges.
The updated Sketchup file has been uploaded if anybody wants to look.
20x40 beam color coding:
Purple: 500 mm Long
Orange: 430 mm Long
Light Green: 403.590 mm Long (approx)
Interesting belt layout. Any chance you could give us a better idea as to how the system works?
Well, it's the CoreXY system: http://corexy.com/
It works like an Etch-a-Sketch.
Both Motors clockwise = Gantry Plate moves left.
Both Motors counter-clockwise = Gantry plate moves right.
Left Clockwise, Right CCW = Gantry plate sits still, but the entire gantry moves forward.
Left CCW, Right Clockwise = Gantry plate sits still, but the entire gantry moves backwards.
That is CoreXY movement.
I'll upload the newest Sketchup file so you can see it from all angles.
Discovered that the 5-hole angle plates on the inside of the frame would restrict the maximum height of the bed, preventing it from reaching the tip of the nozzle.
Replaced them with the 40mm universal L brackets on the inside corners instead.
The bowden extruder that I'm going to hang off of the front bar is a RepRapDiscount Bulldog Lite, the one for 1.75mm filament, direct drive metal extruder. It's the one without the planetary gearbox.
Did a test print of the left side motor mount, and it came out pretty well.
I don't have any V-slot to test the fit yet, though.
A few structural changes:
Bottom rails have been increased to 20x80's to increase the height on the bottom to make room for the Z motors.
Designed Z motor mount blocks to take ball bearings at the top and bottom, taking the weight of the Z gantry off of the motors, rod captive at both ends.
Since the Idler pulleys can handle 9mm belt, I'm going to go with GT2 9mm wide for strength. Robotdigg supposedly has them, plus there's a guy on e-bay selling 9mm belt around $5 / meter - this will probably need about 4 meters.
Going to start populating the BOM soon, the model already has something like 85 T-nuts in it.
Sketchup file also updated.
Could all the orange/red pieces be replaced with 500mm beams to gain some volume? I imagine the gantry could be L-bracketed to the V-plates.
I am looking to design a printer that can be built without having to cut any beams, but it looks like the green beams will always need to be cut with the current design...
It looks like, in that design, that the green beams would also need to be cut exactly correct as the ends screw into the mount holes in the mini-plate. One way to avoid that would be to screw an L bracket vertical to the mini-plate mount holes and then bolt the green beam v-slots to the L bracket. That would offset the beams inwardly 3mm which looks okay and allow the beam ends to me somewhat imperfect. The green beams would be 5mm shorter on each end, and the red beams the build plate is attached to would need to be slightly shortened also.
Yeah, the orange beams could be all 500mm... they're currently length-restricted based on the gantry beam being 500mm - so the one gantry beam would have to be longer than 500mm, or mounted in some other fashion.
I'll play with the idea a bit.
Yeah, that's the awkward part being that the mini-plate is 1/4", and not a specific metric thickness.
I'll try the L-bracket method and see how it looks.
Here's how it currently looks.
I've decided to go with RobotDigg and using custom 500mm-long threaded steppers for the Z-axis:
60mm Nema 17 motors, 92 oz/in torque, 8mm TR8*1 or TR8*2 (1mm or 2mm lead, single start) trapezoid/ACME leadscrews.
They're $55 each, with a 15-20 day lead time. Probably go with a TR8*1 lead screw, considering that a standard M5 threaded rod is 0.8 mm lead.
Anti-backlash nut is also probably not required, so it'll most likely be a POM / Delrin threaded nut block to match.
Some further revisions...
The chassis is now wider, with all the purple beams being 500mm long.
As a result, the main X/Y gantry bar has to be longer - around 570 mm long, but it doesn't have to be exact.
The light green bars still have to be cut, but they are to be cut to 400mm lengths - there's about a 1.5mm gap between the end of the bars and the mini-V plate, so it doesn't hurt if you're off by a small bit.
The brown bars on the bottom are 20 x 60, 500 mm long. They're taller because the Z motors are 60mm long.
The bed area I've managed to increase to 400 x 400mm, and I've managed to find a borosilicate glass supplier that may be able to get me 400mm cut sheets for not terribly much, 5mm thick. The glass bed will be on top of a ~3-5mm aluminum sheet, with a silicone heater underneath that.
Usable space on the X (left-right) axis is just about 400mm, but the Y axis is reduced by about 30mm due to the 60mm offset of the nozzles.
I've updated the Sketchup file if you want to pan it around.
Here's another idea that I'm toying with - a MIC6 cast aluminum heat bed, 18" x 18" (457 x 457mm), usable would obviously be less.
I can get a MIC6 cast aluminum plate, ground flat to 0.015" (0.38 mm) from McMaster-Carr for about $97-107 in 1/4" or 5/16" thicknesses.
I can pick it up from the local McMaster depot to save on shipping cost...
0.38 mm flatness tolerance isn't great, but I bet I could get a local shop to grind it flat to within 0.002 inches or 0.05 mm.
And it would appear to fit within the frame, though I'd probably have to work on the Z motors a bit.
So the cost of getting an aluminum plate surface ground is likely to be unreasonable at around $1 per square inch, and it's likely I'd have to get a much thicker plate to maintain any reasonable level of flatness tolerance.
So, back to borosilicate glass as the print bed.
The good news is that I got quotes from Howard Glass for a cut piece of SB-9 borosilicate, cut at 400 x 400 mm, 5mm thick:
$50 each, 1 week lead time.
Realistically I could have them cut pieces at 400x400, 450x450, etc.. somewhere in that range, for the same price or thereabouts.
Combine the glass top with an aluminum heat spreader and a silicone heater, should do just fine.
Playing with the geometry some more, and coming up with something workable for a dual-extruder setup.
Assuming a build plate of 400 x 400 mm, anything much larger than that and you're limited by the 500mm length of the rails.
So, placing two E3d v6's on the front of the gantry with their nozzles 30mm apart on the X axis, you get the following for usable build area addressable by both nozzles:
370 mm on the X axis (addressable by both nozzles) - it can reach the full 400 mm if you do one nozzle in the center.
352 mm on the Y axis - I'm going to increase this a bit by moving the back rail down by 20mm so the gantry can move a bit farther back.
Redesigned the XY Idlers on the far end of the frame to allow the nozzles to reach +400 on the Y axis (at the back).
The rear rail is now lowered so that the gantry can travel all the way back, and even slightly further off the rail.
You can see the XY Extended Idler part STLs in the build files.
Just ordered the V-Slot rails.
The OpenBuilds Part Store team was kind enough to quote me for custom-cut lengths for the following:
2 x 20x60, 500mm long, Black
10 x 20x40, 500mm long, Black
2 x 20x40, 460mm long, Black
2 x 20x40, 400mm long, Black
1 x 20x40, 570mm long, Black
Cost was about $124 for the V-Slot rails, should be here soon!
I have a BOM in progress for the rest of the parts.
500 on the Z axis by 6 and 700 seems more than do-able.
I've managed to fit in 400mm on the Z axis with 500mm verticals.
Could have gone taller, but that would have required even more custom cut lengths.
the weight on the bed seems to be a worry for me, in the move the bed up or move the XY axis up and down, thats a lotta plastick going down, do you think it could actualy stop the acme rod from lowering using lower end nema 17's?