Building Another Printer - Which route should I take

Hello fellow reprappers!

I have been successfully printing on my Prusa i2 for a while now, which I built with a friend of mine. I'm looking to build another, and he would keep the Prusa.

So for my question, if you were to recommend a printer, which would it be?

Here's what I'm looking to accomplish
- Precision equal to, or better than the Prusa i2
- Speed equal to, or better than the (at least 60 mm/s reliable prints)
- Build volume of at least a cubic foot (304.8mm)
- Bowden extruder setup to remove weight from carriage area, allowing more speed, less moving mass, and possibly better accuracy
- Ability to run dual extruders if I want to in the future.
- Heated bed for ABS (may have to custom make a bed)

So far I'm leaning towards making some sort of printer based on the DeltaBot/Kossel/Rostock design, because (correct me if I'm wrong), I should be able to make one somewhat to my own specifications to accomplish a larger build volume than my current 185mm x 185mm x 85mm Prusa. The reason why I'm swaying away from the prusa is because I feel with a larger build volume, I won't want to have the build platform moving.

When I try and brainstorm of ways to have all x,y, and z moving, it seems like a deltabot is the way to go. All input/comments/questions/concerns are welcomed!

Thank you!
-Pat

---

- "Project Locus" - Mostly all printed parts, Corexy, Bowden Setup, 12x12 heatbed, ~1 cubic foot build volume
- SeeMeCNC Rostock Max V2

I am currently building a printer like that, just with a smaller build area, but the design can be scaled
[www.hackerspace-ffm.de]
It does have a moving build platform, but that is less of a problem than you might think.
Also for such a big print volume bowden setups will not add any precision, on the contrary. I am currently modifying the printer towork with this
[www.thingiverse.com] extruder one of our hackerspace members designed. It weighs <200g and allows for fast extrusion of flexible material, something you will run into problems with if you want to use bowden.

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

I have identical purpose for my next printer. Delta printers are fun to build, to have and to see, no doubt about that. I won't go for a delta printer for many reasons. First, there's a huge difference betwen building a delta kit of follow a BOM at dimensions, and dealing with your own delta design. Configuring a delta printer is realy difficult especialy if you explores new ways. Second, delta technology is very good for high speed moves. With FDM technology, you will not go that fast anyways because fusing plastic needs time. Third, even if it uses 4 motors, delta is not cheaper because it needs a stiff structure (preferably aluminium at last for the 3 rails) and linear precision guides (ex. MGM) to get a high print quality. At the same print volume, a delta needs more overall volume and longer guides. The heated bed cost more too, and you need a much larger build plate than that rouded triangle build area. Through, I still don't understand rounded beds for delta printers : 15% heats nothing usefull. Fourth, because of its shape, and except if you print diminushing vases only, at same volume size, only smaller parts can fit into a delta's. Especialy if you try to put several parts into a build plate. The maximum size for a cube or a sphere is smaller on a delta, at equals print volume. Fith, the resolution is not constant on delta printers, you have less resolution on the centre. Sixth, because of its delta headplate, it's not easy to deal with multiple extruders and as many fans as needed. Remember each extra extruder diminues the printable area too. Seventh, I think about enclosures to maintain heat and collect fumes. Making an enclosure for deltas is not that easy and requires much more materials.

The delta definition :


I'm realy excited by CoreXY printers. First, like the delta, it can works with 4 stepper motors only. The overall volume is smaller, and the ratio betwen print volume and overall volume is reduced. The 4 push-pull points of the CoreXY head carriage reduces vibrations and twist issues. It's easy to setup, even with custom dimensions. Imagine a 400x200mm printing area with two heating beds side by side. You only have to enlarge X dimensions. CoreXY can run as fast as FDM technology can go. It can be builded into a simple box shape, made ou whatever material you want, even cheapest ones. Enclosures are also cheap and easy to make. Maybe you can stack several core XY printers on shelves for production purposes. Last but not least, the better print quality I've seen comes from CoreXY printers. That's why I will probably build a CoreXY printer. I hope this can help you to make up your mind.

Edited 1 time(s). Last edit at 01/23/2015 05:43AM by Zavashier.

---

Collective intelligence emerges when a group of people work together effectively. Prusa i3 Folger (A lot of the parts are wrong, boring !)

I built my delta around the rostock mentality. For experince I can say that it can be done. My printer has a built volume of roughly 320mm x 350mm..... You also have to take into account that the printer will be tall, wide and heavy. You need to make certain your frame is built tough.... I can sit on top of my printer while printing and it wont effect the print. 190lb soaking wet here.... I went cheap and used thick wood for top and bottom. I used recycled aluminum with threaded rods in compression for the frame and threaded the smooth rods at the end to place them under stretch. If my budget was larger I would use two slabs of aluminum plate at least 3/4inch thick.

So far my printer does the following reliably and I keep pushing it harder and harder with every print.

110mm/s outer shell to keep it looking recognizable
140mm/s infill
130mm/s inner shell

2:1 extruder is what I am using and hoping to retail soon enough (waiting on parts at this point)


Overall printer dimensions are ~4foot tall and ~ 16inch round (and growing as I add more bells and whistles) My build blog is below. I warn you a large delta takes up a huge chunk of room. As for the bed, if you get a heat spreader between the glass and a regular 250mmx250mm bed it will work just fine for ABS. I have heated mine up to 100c. It takes a while but eventually gets there. You better stock up on thermal paste.

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11

I have both a Cartesian printer (Ormerod) and a delta printer (Mini Kossel). The moving bed of the Ormerod increases considerably the amount of desk space it occupies, and this will be true for any Cartesian printer with a bed that moves in the X or Y direction. Delta printers have a more compact footprint because of the fixed bed. The extra height of a delta printer is of little consequence unless you want to fit it under a cupboard. Also, the rigid frame of a delta printer makes it easy to put filament spool holders on the top, saving even more desk space.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

I'll echo the comments made by Zavashier. It's hard to use a Delta printer as a production printer, IMO, though I'm sure it's being done. Deltas are fun to watch but there's constant wear on all linear motion components. That was the first type of printer I built, and that was a bad idea. Insanely hard to calibrate without autoleveling for a beginner. The stepper noise I had with my Delta was also intense. I'm a big fan of the CoreXY and I'm currently designing/building a large volume one. I've found that CoreXY is much quieter and it's great to capture Delta type speeds with cartesian precision.

Edited 1 time(s). Last edit at 01/23/2015 01:10PM by runninfarmer.

I have a Prusa i3 and im grabbing bits and pieces I like from others Core XY printers and im going to make a nice package when im all done and have built one.
Seems a lot of Reprap people have great ideas but have poor information when they are done so others can try to make the same unit.
Im also developing in Solidworks so that gives me MUCH more freedom over just a pile of STLs people post.

Quote
runninfarmer
I'll echo the comments made by Zavashier. It's hard to use a Delta printer as a production printer, IMO, though I'm sure it's being done. Deltas are fun to watch but there's constant wear on all linear motion components. That was the first type of printer I built, and that was a bad idea. Insanely hard to calibrate without autoleveling for a beginner. The stepper noise I had with my Delta was also intense. I'm a big fan of the CoreXY and I'm currently designing/building a large volume one. I've found that CoreXY is much quieter and it's great to capture Delta type speeds with cartesian precision.

I will be honest, my printer was hard to get going. Took me a month to get my first print. However I can say that a delta printer is not necessarily more wear prone than a cartesian printer. They both use linear motions. I used chinese lm8uu bearings and tool steel rod, 6 months worth of use and abuse as I keep finding the capabilities are greater and greater with each and every push and each and every improvement. Neither is close to dead just yet. With open beam v-wheels the wear will be lower. Having seen "Comercial Units" eg makerbots MSRP $2500+ being outclassed by my home built and home designed printer its frankly awesome.

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11

I saw your delta Jaguarking11, you did a nice work. Threaded rods in compression works pretty well. You might be an ingeneer or at last have some skills. Because linears are longer on a delta, with the same guides the wear must be more important on a delta than on a carthesian. Note on a CoreXY the 4 points push-pull reduces wear. That's why I think you must go directly on linear rails and carriage, at last MGM style or better; since the start. This is a wise choice that makes the delta more expensive. When I studied costs of both types, CoreXY was 20% under for comparable build area.

---

Collective intelligence emerges when a group of people work together effectively. Prusa i3 Folger (A lot of the parts are wrong, boring !)

Each printer configuration has advantages and disadvantages. Delta is nice because of the small footprint, simplicity of a fixed bed, and regularity of the build - once you have the design of a single tower right, you have them all right. Whereas with a conventional Cartesian printer, you need to get 3 different axis designs right. If that 20% extra parts cost is right (and I suspect it isn't if you include electronics, power supply etc.) then I think that is a small price to pay for the regularity, unless you intend to go into mass production. CoreXY looks nice, but moving the bed in the Z direction while keeping it level is surely challenging.

People complain about the time taken to calibrate a delta, but it's a one-off event unless you dismantle and rebuild your printer. Many firmwares can do auto calibration these days. People also complain about the reduction in precision of a delta near the centre of the bed; but all that matters is that you get at least the precision you need. Who needs 0.012mm precision, when the extrusion width is 0.3mm or more?

If I had time, I would try designing a 3D printer around a Stewart platform. The fixed head would simplify the build of everything apart from the platform, and the tiltable bed would offer interesting possibilities for printing overhangs!

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

All was included in the quotation until the last cable or last nut.

Edited 1 time(s). Last edit at 01/23/2015 05:22PM by Zavashier.

---

Collective intelligence emerges when a group of people work together effectively. Prusa i3 Folger (A lot of the parts are wrong, boring !)

HOLY REPLIES! Keep them coming, this is a lot of useful information. Thank you thank you thank you!

After reading through these replies I do like the core-xy design. I guess I'm not opposed to a moving platform, but I would prefer the bed to only move in the z direction if it's going to move. As for the Bowden thing I mentioned, I'm currently running a wades geared setup and have no problems with it. I'm more thinking bowden because it seems pretty common, although I've read retraction can be troublesome. The benefits of a cleaner look and less moving mass and a smaller size (in turn possibly allowing slightly larger build volume?) vs a geared setup just stood out to me. Once again, these posts have been very informative, thank you. You all brought bring up some very valid points and some nice useful information that I did not know about deltabots.

As of now, I've been starting to look more into the core-xy setups. My one concern regarding the budget is the use of extruded aluminum. Has anyone heard of, or tried making the extruded aluminum parts, but made of PLA or ABS? I understand if there are bearings rolling on them it would be a rough surface, but I know that stuff isn't cheap. I could always use smooth rod attached in some way to the main frame to allow the bearings to ride on.

I am proficient in creo, so I'll probably design the printer first and go from there. Here's what I'm talking about.

Option 1, print these in pieces, clamp together, glue, and make them to length. This option allows you to buy smaller (1/8"?) steel from the hardware store, and slip it in the corner holes for added rigidity. Maybe even a threaded rod/nuts on each side through the center hole.



Option 2, create the same thing but with lego-type connectors to help alignment when assembling. Again, a center rod could be ran through the middle if necessary.

---

- "Project Locus" - Mostly all printed parts, Corexy, Bowden Setup, 12x12 heatbed, ~1 cubic foot build volume
- SeeMeCNC Rostock Max V2

For fun, I've decided to try and test out some of those blocks. These are 35mm x 35mm, however I feel they may be too large so I've scaled it and will try some 26 x 26mm as well. I plan on trying to load these similar to a bridge to test the strength in bending, as well as physically twisting it to see how it holds up to torsional loads.

What do you all think about this? Is it a waste of time?

---

- "Project Locus" - Mostly all printed parts, Corexy, Bowden Setup, 12x12 heatbed, ~1 cubic foot build volume
- SeeMeCNC Rostock Max V2

It's kinda pointless. If you want reliability, you need a stiff frame. Printed extrusion parts will not get you near the stiff required. Plus, depending on the size of your printer, extrusions are not that expensive.

---

greghoge.com

HUGE 3D PRINTER PARTS SALE!!!

After some testing, I've decided it was indeed probably not worth my time to actually implement.

What are my options for heatbeds? Besides doing two standard sized (200 x 200) side by side, has anyone successfully made a single, larger one, say 300 x 300?

---

- "Project Locus" - Mostly all printed parts, Corexy, Bowden Setup, 12x12 heatbed, ~1 cubic foot build volume
- SeeMeCNC Rostock Max V2

I've designed this heat bed for my large format CoreXY printer. The bed is 21" x 20" (508 mm x 533 mm) using a 120 VAC heat tape and 1/8" thick aluminum plate. The heat tape is 8' long and I coiled it onto the plate using Kapton to secure it. Should only draw around 2A and I won't need an extra power supply to run it since I'll be using AC mains for power. I'm using a solid state relay with the AC and using RAMPs heat bed output for the signal. The type of heat tape I'm using is designed so that it can be used on electrically conductive surfaces such as pipes, etc. Omega Engineering sells them, mine is a Conrad. It's rated to operate at 400C, but I definitely will be under that for my PLA printing. Would work great for filaments needing really high temps. Disclaimer, AC electricity is dangerous so if anyone's interested in it, build at your own risk!







Edited 3 time(s). Last edit at 01/28/2015 01:26AM by runninfarmer.

One thing to note. We love to use 3d printed parts to create our printers, however this leads into a problem we all must face. The printed parts make the structure more brittle and flexible. My advice is that any printer you build you should look to make the frame as stiff as possible, I know this goes against the reprap mission, however if you want a reliable machine with low wear then the frame has sto be stiff. If your frame can't support at least 100lb of weight ontop then its not stiff enough. I am generalising here, but being on the overkill side of the spectrum is allot better than breaking stuff or chasing red herring's....

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11

Quote
dc42
People also complain about the reduction in precision of a delta near the centre of the bed; but all that matters is that you get at least the precision you need. Who needs 0.012mm precision, when the extrusion width is 0.3mm or more?

The reduction of resolution is an easy problem to solve. Just use the lowest resolution as the baseline...away from the center lobes your always increasing resolution!

Quote
jaguarking11
I know this goes against the reprap mission

Are you sure ? Reprap is not fixed into 3D printing. A machine able to replicate itself could be a CNC mill. See first page of reprap.org

---

Collective intelligence emerges when a group of people work together effectively. Prusa i3 Folger (A lot of the parts are wrong, boring !)

Update, the build is underway. I will post back with a link to the build thread once started. I have designed every part of the printer at this point in cad and have most of the parts printed. I've decided to go with a corexy setup along with... (Gasp) primarily all printed parts, and yes that includes the frame. Details to follow, thanks again for everyone's input!

---

- "Project Locus" - Mostly all printed parts, Corexy, Bowden Setup, 12x12 heatbed, ~1 cubic foot build volume
- SeeMeCNC Rostock Max V2

I have the same interest as you Redhatman,

I am looking forward to see the evolution ^_^

Quote
J.A.C.
I have the same interest as you Redhatman,

I am looking forward to see the evolution ^_^

Great to hear! Yes, I'm also very anxious to see if this works out.

Here's some teaser photos for everyone until I can get around to starting an official build thread.

Front







Rear view/ X-Y Motor Mounting

---

- "Project Locus" - Mostly all printed parts, Corexy, Bowden Setup, 12x12 heatbed, ~1 cubic foot build volume
- SeeMeCNC Rostock Max V2

Are the "Cube" structure hold by printed plastic parts? if so, at lest add 8 metal brackets to hold them and remove any future stress for the structure.

Quote
J.A.C.
Are the "Cube" structure hold by printed plastic parts? if so, at lest add 8 metal brackets to hold them and remove any future stress for the structure.

Could you elaborate please? I'm not quite sure what you're referring to.

---

- "Project Locus" - Mostly all printed parts, Corexy, Bowden Setup, 12x12 heatbed, ~1 cubic foot build volume
- SeeMeCNC Rostock Max V2

Update: CAD work/parts printed and ready to assemble ~90% complete. Assembly of complete printer itself ~75%. I'll be starting a detailed build thread soon (Project name is still pending).

---

- "Project Locus" - Mostly all printed parts, Corexy, Bowden Setup, 12x12 heatbed, ~1 cubic foot build volume
- SeeMeCNC Rostock Max V2

...and here's a picture of my Kossel Mini rebuild, with printable area more than 300mm diameter (who said you couldn't get a large build area from a delta?):



Auto calibration (endstops, tower radius and positions, and diagonal rod length) takes all of 30 seconds. The power supply and the SSR for the 350W mains-powered heated bed fit neatly under the bed:

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
dc42
(who said you couldn't get a large build area from a delta?):

I think it's been proven that you CAN get a large print area from a delta.