Open-source Stationary Stage or z-direction-movement Stage Printers?

Hi all,

The Cartesian open-source repraps seem to all use a print head that travels in the X and Z directions, with the stage travelling in the Y direction. Are there any Cartesian printers that have a stationary bed, with a design that allows the head to travel in 3 directions, or an open-source project like the Makerbot printer with a print head that travels in the X-Y plane, and a stage that travels in the Z direction? I'm aware of the Delta printers, but I'm looking for designs that have 3-axis motion with precisely defined steps/resolution, that the deltas can't offer. Thanks!

I was going to ask this exact same question, I'm curious as to what others have designed with a movable gantry. I'm designing a movable XYZ gantry with stationary bed.

Here is a note that I wrote to myself earlier today, part of my design definition:

The moving gantry, and stationary build plate is foremost for DLP/SLA printing. The mass of a movable gantry is a disadvantage for filament printing but the movable gantry doubles the print envelope for filament deposition, so a small trade off for increased build volume for less print speed, and the ability to use two different additive technologies with one platform.

Quote
Kyle R
open-source project like the Makerbot printer

Makerbot is not open-source any more, they are patenting every thing that any one could think of, and actively suing competitors. It appears to be an attempt to close the door of competition, development, and sharing of ideas in open forums.

There is Tantillus (see also Ingentis), Ultimaker and Ultimaker 2. Two of Makerbot's early designs are still open source and still available on Thingiverse: Thing-O-Matic and CupCake CNC.

@MattMoses:

Quote
Kyle R
Are there any Cartesian printers that have a stationary bed, with a design that allows the head to travel in 3 directions

OP, and I are looking for examples of a stationary build platform, one that the bed does not move up/dn in the zed axis, all Cartesian movement is on the gantry. I've found lots of CNC routers like this, but not one dedicated to a 3d filament printer, let alone a RepRap 3d gantry printer.

Tks for pointing out potentially the last of the open-source printers from Stratasys-Makerbot. It will be interesting to see what happens with the Autodesk open, but not open-source software and DLP/SLA printer, and Autodesk future product offerings does to Stratasys market share, and stock valuation.

Autodesk has not yet determined which of several standard open-source licenses it will use, if any.
[bits.blogs.nytimes.com]

Autodesk Announces Open DLP 3D Printer and Spark API
[makezine.com]

I'm not addressing your comments, but I want to point out the following as it relates.
Bre Pettis said in an interview that he was working hard not to violate Stratasys patents. From what I have read in the news, and in forums it appeared all along that Bre Pettis was violating one or more of Stratasys patents with his open-source RepRap derived printers. If I recall correctly if this violation exceeds I think ~7 years without Stratasys bringing suite it effectively allows Stratasys patent to be freely copied by any one? I haven't read this argument being brought up in the forums, so it's probably not in play. I think one of the violations was a heated bed? The link is in the thread Stratasys sues Afinia thread.

Stratasys sues Afinia
[forums.reprap.org]

The result of this lawsuit will have a major impact on the future development of desktop 3D printing.
[makezine.com]

In The Fold, Autodesk News and Opinions
[inthefold.autodesk.com]

.... and this thread has been high jacked If any one want's to take this conversation further, please post your comments in one of my other threads, and I'll try to stay on topic in this thread.

Some insightful information about gantry 3d printers in this thread:

Newbie with ambitious build
[forums.reprap.org]

Gantry plans
[www.ebay.com]

Basically it's the mass of moving the gantry and trying to accelerate and decelerate. There's a lot of people out there that have put extruders on CNC routers and not very happy with it because of the acceleration rates. The first printer I built I designed my hot end and extruder first and tested it on a large gantry router that I had built which was great for testing the extruder but not optimal for a printer. The forces of cutting material even if it was just wood at a decent rate is quite high so things have to be massive the router that I had that I tested it on used to cut Baltic Birch at 90 in per min. so by splitting it up moving the axis separately you reduce the moving mass some of the other things to think about is where the mass is moving the lower on the machine more stable.

Edited 1 time(s). Last edit at 06/05/2014 08:46AM by cnc dick.

@A2: Here is what OP said:

Quote
Kyle R
Are there any Cartesian printers that have a stationary bed, with a design that allows the head to travel in 3 directions, or an open-source project like the Makerbot printer with a print head that travels in the X-Y plane, and a stage that travels in the Z direction?

Tantillus, Ingentis, Ultimaker and Ultimaker 2 all have "a print head that travels in the X-Y plane, and a stage that travels in the Z direction". So does the original Darwin.

I toyed with the idea of making an XYZ gantry, but later moved on for the sake of simplicity. I can go back to that if you'd like, even if just to make the design/concept.
Are you looking for belt driven on all axis? My idea used leadscrews, four of them. Not a cheap idea, but it wasn't meant to be.

---

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

Thanks for all the helpful suggestions. Price isn't a concern, I'm trying to take the open-source printers to the next level and drive down the price point for 3D bioprinters. I'm a PhD student in Biomedical engineering and want to bring the contributions the world has made to filament printers to the world of bioprinters, so any lab can build and begin experimenting with cell deposition.

Does anyone out there have an Ultimaker 2? We always hear about Makerbot, I'm sure a lot of the hype is in marketing, but I've used the Replicator plenty of times with great results. If the price point for building an Ultimaker 2 gets high, I may do the unthinkable and buy a Replicator 2 and try to reprogram it. I may start a new thread to get ideas on hacking a Replicator 2.

@MattMoses: OK, strange what the eyes see. So the OP is open to a couple of configurations....my apologies.

@cnc dick:

I've read a lot of posts about the CNC router gantry configuration not being ideal for filament printing due to the mass, and the use of lead screws. I'm not going to copy it exactly, it will be a modification of the router gantry configuration that will account for its shortcomings.

Knowing the weakness of the router based gantry for filament deposition I've added more constraints, and eliminated the lead screw for a belt. But the expense of the modifications is making it hard (impossible) to keep the cost under $2000.00, it easily could go as high as $3000.00. But it's 2 printers in one, (DLP/SLA, & filament), plus it can be converted to a Laser cutter, engraver etc. A gantry 3d printer is a very reconfigurable (alternative technology) friendly design. I look forward to maximizing it's utility with other additive, and subtractive technologies.

Note: the build envelope is big, 12 in wide by 12 in tall by 24 in long. I was originally aiming for 12x12x12 build envelope, but a gantry with a fixed bed makes it easy to double your Y-axis without increasing the footprint by much, I also like the object to remain stationary during printing, and that is very appealing to me for some of the larger projects that I have in mind. It's possible I could revert back to the 12x12x12 build envelope to save money, when the BOM is done I'll know where I stand.

To move the mass I've located a NEMA 23 stepper motor with 2.8Nm (400 oz.in), and I'm wondering if I should use 2 of them on the Y-axis? I'm also trying to keep the gantry weight to a minimum, regardless it's going to be heavier than a typical printer, and I have to account for it accordingly.

I'm in the iterate, iterate, iterate stage of the design process, commonly refereed to as R&D, and I'm in no rush to complete it.

Tripteron : [robot.gmc.ulaval.ca]

In the Reprap world, based on the tripteron, there is the Sexupteron : [forums.reprap.org] which I presume can be simplified to 3 axis.

Edited 1 time(s). Last edit at 06/05/2014 12:10PM by Mikadoh.

Quote
Mikadoh
Tripteron : [robot.gmc.ulaval.ca]

In the Reprap world, based on the tripteron, there is the Sexupteron : [forums.reprap.org] which I presume can be simplified to 3 axis.

That would require some tight, tight tolerances to be suitable for 3D printing.

---

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

Quote
A2
OK, strange what the eyes see.

No worries, man

Quote
Kyle R
want to bring the contributions the world has made to filament printers to the world of bioprinters, so any lab can build and begin experimenting with cell deposition

Can you tell us a bit more about the requirements for this?
What is the work volume form factor (i.e. Xmm x Ymm x Zmm)?
What is the range of desired printing speeds (mm/sec)?
and accelerations (mm/sec^2)?
What would the "tool" look like, is there an off-the-shelf micro-pipette gizmo or something?
Why is it important that the stage be stationary (is this so that the medium the cells deposit into doesn't slosh around)?

If you have a small work volume that requires high precision (it seems like most bioprinting experiments would be small, unless you are trying to print a cheeseburger or a new liver) then you could in principle design a very precise Delta manipulator that should be able to do everything that an XYZ table could do.

For the work volume, along the lines of 150-200mm in the X, Y, and Z directions, though initially I would be fine with short and wide, where the max height of a printer object is smaller than this. I am looking for consistent 50-100 micron resolution in all directions, where the printing speed and acceleration takes a backseat to accuracy in deposition. I have several "tools" in the works, but the first would be a microvalve which controls the flow of the living material to be deposited, the material being under light pneumatic pressure. So the Arduino controls the amount of time the material can flow from the valve, and thus the amount of material. You hit the nail on the head, it would be optimal of the stage didn't move as the material is similar to Jello after deposition, and shouldn't be subjected to the movement you get with a Prusa or similar.

So the gantry will not have to bear much weight, as only microvalve 'heads' will be attached to it.

It seems many of the reprap projects use the build platforms that move in the Y direction, and the ones that don't are much less developed than the Prusa projects. Do you think a delta would serve me better than an XYZ table? The change in resolution with distance from the center of the build area concerns me. Any other additional thoughts would be great, thanks!

Since resolution (50 micron), accuracy, and object stability (Jello like) are key for a successful build, I think the better option is to use a gantry style printer with a fixed stage. You plan to have your machine used in all corners of the world, so I think a recognizable machine configuration would help with the adoption of it's use.

Do you need a design of the pump to go with the microvalve, any special storage required for the cells, does it need to be in a clean room like enclosure, does the enclosure need to be heated, does it need to be transportable (easy to stow), is grease allowed to be near the cells, are certain materials not allowed near the cells, (lead barium, cadmium), does light have to be blocked out, do the materials to construct the machine need to be common off the shelf hardware, etc.?

Post a few pictures, and a video of the current machine, the object being printed, and the microvalve being employed, I think that will help with the brain storming.

In your case what you're talking about sounds like a gantry is a good idea I would use precision ballscrews no belts and you might want to try to run it on something other than the standard reprap electronics CNC controls. But I know nothing about the material you're talking about or the size of the orifice or nozzle you would be using. The Gantry is a fine design unless you try to go fast with high acceleration

A2, thank you for your input. I have a team working with me on all above-mentioned 'extras.' we have an enclosure capable of controlling the microenvironment around the printer and have a working deposition system. Since there are so many open-source designs for positioning systems out there I wanted to go with a reprap, and modify it over the course of my graduate work. So the question boils down to gantry style or delta style. And, if gantry style, which designs fall under my constraints (fixed platform or platform the moves only in the z). A few good suggestions were made, but As the price point gets higher, hacking a MakerBot replicator 2 is a possibility.

CNC Dick, thank you for your input too. Any suggestions on open-source designs?

You both mentioned leaning away from the deltas, what are the down sides?

Thanks!

With the Delta I don't think the possible resolution in the print is linear it changes from center out but again I may be wrong I am trying to work it out in my mind

Edited 2 time(s). Last edit at 06/05/2014 10:57PM by cnc dick.

I have read that too...The quoted resolutions for the deltas seem incredibly high also. The look is so elegant though. The Tantillus design might work, not sure about the Ultimaker. I'd like the system controlled by the MEGA/RAMPS combination, so some other systems I've seen are out. Any other ideas to achieve this end?

It's always hard to get extreme high accuracy at a low cost it's always a balance again I am not familiar with bio printing although I would like to be I don't know how accurate you really have to be the print on the cellular level. I've been around large CNC machines and built a few myself that can be extremely accurate but again that's not something that you are talking about you want something small cheap and very accurate. Me personally I would stay away from anything makerbot just because of the parent company which of course invented FDM which is fine and I have a lot of respect for the original MIT students that started it but now it's all corporate bull

Edited 1 time(s). Last edit at 06/05/2014 11:09PM by cnc dick.

CNC Dick,

I'm sorry, price is not a concern for me. I just want to find something that suits my specs, I have a several thousand dollar budget for this project. Thanks again for your input.

If price is not a concern then I would stay away from belts precision ballscrews is the way to go you can have extreme accuracy with just stepper motors. As long as they are sized correctly. When you buy the screws the c grade rating the lower the number the more accurate they are over the span of the screw. In your case you don't need very large ones and you're not moving fast so you frame doesn't have to be that rigid. But you do have to factor in the SAG in any all your linear systems. The first 3-D printer I designed and built is a large floorstanding one with all ballscrews very accurate but uses CNC-based controls. The latest little printer uses the Arduino and the ramps of course it's only 16-bit system and I don't think the interpolation is as good as the large printer I made here is a picture of a bottle opener I printed on the large machine I made you can see the layer lines very accurate

That is one sexy print!

I will take your suggestions into consideration, I appreciate the insight from hackers who have been around the block. Thanks again!

Kyle,
The idea that the delta printers have reduced resolution at the edges is somewhat inaccurate. I have modeled a fairly large one in SolidWorks and I discovered that when the ratio of the pushrod lengths to axis locations are correct the steps per mm are almost linear over the whole work envelope. The machine I am working on with a friend has a work envelope of 22" round x 24" tall, and a full enclosure so it the chamber can be heated. We set the maximum dimensions as needing to pass through a 36" wide x 80" high door way. The frame is hexagonal not triangular so that the build envelope is as big as practical. The frame is also heavily triangulated for maximum stiffness. Also the slides are hardened steel rollers on hardened steel ways supported by the aluminum extrusions in the stiffest way. It is very modular and can be assembled in different sizes from the same basic parts. I have come up with a simple way to gauge all the parts so that the frame is very accurate too. We are using carbon fiber pushrods held magnetically against precision ball bearings. However there will be no sheetmetal in the bearing cars or head frame. The cars and head are CNC machined from one piece with pockets for the balls and holes to mount the wheels. The motors are Nema 23, the timing belts have steel cords, and we are driving the motors with Leadshine digital microstepping drives from a 36 volt power supply. The extruder is geared stepper mounted on the head. Control is via a smoothie board.

I have nearly all the materials, including motors,drives, power supply smoothie board linear rails and bearings and I am currently machining the pieces that join the frame members together. I still have a little design work to do on the cars, motor mounts and head. My friend is a real computer whiz and is working on the smoothie board and software side.

Contact me and I'd be willing to share what we are doing. It is going slow because we are both employed full time, have kids,and our workplaces are very busy. I've thought about selling kits of parts once we get the first one built, but that's maybe not realistic considering how long it to to build my CNC machine.

[s811.photobucket.com]

I miss the mention of the CoreXY system, which should get a bit more attention imo.

Garyhlucas--

That sounds like an incredible project, good luck to you! I've decided to build an Ultimaker 2, the plans and BOM are open-source. I will also build a delta-style printer, I think I'll go with something fairly stock at first, a Kossel or Rostock, and expand upon the designs once my knowledge grows. I'll be starting a new thread to gather others who have built an Ultimaker 2. Thank you everyone for your experience and input, I'm very glad this resource is available!

Kyle