Printrbot: Your Very Own 3D Printer | Hacker News Reader

Printrbot: Your Very Own 3D Printer

kickstarter.com

123 pointssidwyn36 comments

There have been a lot of crowd-funded and startup attempts to commercialize RepRap designs lately - here's a sample of a few:

eMaker Huxley: http://www.indiegogo.com/eMAKER-Huxley-3D-printer-kits

SeeMeCNC - uses software to drive stepper drivers via a parallel port rather than onboard control hardware running GCode or the like, saving hardware cost: http://www.indiegogo.com/SeeMeCNC-H-1

SUMPOD - http://sumpod.com/ - more appliance-y.

Good list - I think there's a definite hype bubble in cheap 3D printers. All the copy sounds very naive in relation with the reality of what these printers can actually do. Ex: "printer in every home" - that's like saying a CAD designer in every home - not going to happen. Even d/ling from Thingiverse isn't going to be that popular as most of the parts aren't super useful / attractive.

I think you're missing the beauty of having a a 3d printer at home and needing an enclosure to do an electronics project and just printing one out, or deciding to print out an ipad stand, fork, screwdriver, guitar, etc.

If 3d printers became widespread, we would have databases full of plans made by people and every 3d print wouldn't be a custom one off

I have the ee enclosure problem now. It's getting pretty bad with some of my deployed projects just kind of spread all over the place.

As far as the database full of plans: http://www.thingiverse.com/

Would you have imagined that one of these would be in every home?

http://www.ttsw.com/SuesClasses/OldComputers/ENIAC.gif

To be fair, there never were nor should have been ENIACs in homes. I don't think he's saying a future 3d printer shouldn't be a real consumer model.

A better analogy might be the early PCs sold as kits and targeted at hobbyists, which is basically what this Kickstarter is funding.

He said it was a hype bubble. Were computers a hype bubble when they were the size of entire buildings? Or later when you had to assemble them yourself and program them with toggle switches? Fortunately, a few people had the vision and patience to keep pushing the technology forward.

Hype means excessive claims - 3D printing guys right now are claiming the RepRap can do useful things for average people, and that's just not the case. Even the commercial machines, like on Shapeways, only produce tchotchkes.

DEC wasn't wrong about the "cloud" - just way too early and in the wrong environment. RepRap clones are at this point now.

It's not that I think the utility of the general public making things at home isn't attractive. It's that 3D printing, in current forms, will be obsolete by the time it's able to do this. People have been trained, and with Apple even more, to like beautiful things that combine multiple materials, the latest electronic hardware, and great packaging.

To believe this, you'd have to this we've got at least: - plastic/metal in the same machine - most likely simultaneously unless people are going to assemble it. - home printable silicon, maybe they could ship it in, but I feel like that defeats the point - they could just ship the device then.

Both of these are really hard material science problems that require tons of research.

The process/materials that 3D printing provide can do none of those things right now - probably not for 10+ years. So, I think all these RepRap clone guys who think, like GNU/Linux, they'll be able to go RepRap -> ? -> Replicator, are a bit naive / hyping it.

They should be creating things that compliment the 3D Printer - like a laser cutter like this:

http://www.kickstarter.com/projects/nortd/lasersaur-open-sou...

"SeeMeCNC - uses software to drive stepper drivers via a parallel port"

Parallel port? Isn't that an anachronism, or are these still popular in the 'hobby' world?

They're popular in the CNC world, yes. For instance, the recently Kickstarter-funded MyDIYCNC project also natively interfaces through the parallel port; Stephen McGloughlin, who's behind the project, is building an USB interface for it from scratch.

The main issue seems to be the timing with controlling... parallel i/o lines. From http://www.cnccookbook.com/CCBreakoutBoards.htm :

While there are boards the purport to simulate a parallel port with a USB connection, most of them don't work for CNC applications. The reason is that they are not high enough performance to maintain the exact timing relationships needed to produce a clean pulse train to control multiple servos or steppers.

They're nice for hobbyists because they bridge the digital<->voltage domains really easily, so you can just hook up wires to the pins and feed them into whatever you're building, rather than having to use a fancier microcontroller interface. But yeah it's getting to be more of a hassle that e.g. a typical laptop won't have one anymore.

In the past I've also used audio line-in/out as a simple two-channel (each way) I/O interface, but that runs into more problems with drivers getting in your way (buffering things and whatnot) than parallel ports do.

Does anyone happen to know where one gets the plastic for the printing? The price of $500 for this right now looks a lot more affordable than it used to be to me. Being able to source the printing material is the only thing stopping me at the moment.

You haven't said where you are based, but for starters look for either PLA or ABS on this list: http://reprap.org/wiki/Printing_Material_Suppliers (The Reprap wiki also has pages explaining the pros & cons of each.)

I don't know what filament diameter(s) the Printrbot will support.

I apologize for not saying where I'm based, (US by the way). But that link is absolutely perfect.

According to the comments on the kickstarter page it'll support 3mm to start with and they plan on having a 1.75mm hotend available for $59+shipping by the time that all the kits are shipping.

EDIT: corrected price of hotend

A home-made 3D laser scanner would complement this nicely.

As would a 3d CNC router: scan with the scanner, print it roughly, then the router cuts it to high quality perfection.

It seems that movement hasn't been as rapid in hobby CNC machines as in 3D printers, probably because CNC machines are a lot more expensive. Good CNC usually requires two tool-tilt axes in addition to the standard X, Y, and Z - for a total of 5 axes, and everything has to be a lot stronger to apply enough pressure to the tool without warping. Hopefully once everyone has a 3D printer they'll use it to print some CNC parts and kick that movement into gear...

With that said, a 3D printer can be repurposed into a very weak 3-axis router pretty easily. Provided you move the tool head slowly enough and are cutting something very weak, they work okay.

The fundamental concept (having a computer move a tool head) is the same - 3D printers are just built out of weaker/smaller (and hence cheaper) parts.

what I'm longing for is a good G-code generator that is affordable, runs on my Linux or OSX, can generate toolpaths for milling as well as 3D printing and can work with files from Sketchup.

(Most makers I know don't use Windows and I think there is a huge opportunity for a polished product that will run on either Linux or OSX)

FYI: sketchup doesn't support linux yet.

I know, but since I will be using either EMC2 on Linux or Grbl on an Arduino to run the G-code, I can live with CAM software that runs on Linux.

Slic3r is an interesting project that goes in that direction. Only for printing though. You can feed it an STL exported from sketchup and it will generate a toolpath. And of course it's all on github and completely free and opensource and multiplatform.

The major problem with it is that sketchup is a mesh editor, not a solid editor, so it sometimes exports geometrically impossible (non-manifold) objects.

> 78 backers for $499 or more

Wow! A true kick-start.

Is moving the print head in 3 axis the best approach? I'd imagine moving the print head in x and y and leaving z to the build surface easier.

You don't want to be moving the warm plastic..

Good point. You either have to move the warm plastic or the print head and its positioning mechanism... :-/

Engineering is full of unsatisfactory choices...

I'm confused by this Q&A - I can't think of a hobbyist 3d printer design that doesn't move the build platform:

In the Printrbot it appears the print head moves in X & Z, and the build platform moves in Y. Same as the RepRap Mendel.

In the MakerBot the print head moves in Z, and the build platform in X&Y. In the Up! 3d printer the print head moves in X, build platform Y&Z. In the Ultimaker, print head X&Y, build platform Z (this is the combination that rbanffy describes.)

So that's a good bunch of possible permutations, but the build platform always moves...

The original RepRap design (Darwin) moved the print head in X and Y, and moved the build surface in Z only (downward as the print progresses).

The Ultimaker machine is based off the Darwin design, and works similarly. Ultimaker has a very lightweight printhead, so they can print incredibly fast.

I haven't seen a moving-head deposition robot that moves only the head. I think this is because the simplest way to make a 3-axis robot involves two tools/surfaces, each moving in at least 1 axis. A tool/surface that moves in all 3 axes needs to be very stiff to avoid juddering around as the machine starts and stops line segments. It's much simpler to make a 2-axis robot and a 1-axis robot, than a 3-axis robot. :)

Can the 3D printer, print other RepRap 3D printers?

No, and unfortunately it's very very very very highly unlikely [functionally impossible] that this will ever be a reality.

Imagine printing a stepper motor, something you need several of for a 3D printer.

Now imagine printing an extruder head...or a PCB for driving the steppers.

Just being able to print PCBs would make tons of hackers glow. The closest thing we have to that right now is putting adhesive copper sheets into a vinyl cutter. That is to say, it is a very very long way off.

(After you print blank PCBs, then what...is the rep rap also a pick and place machine? Can it "print" the components to populate the board?)

Looks promising. Is it open hardware (i.e. anyone can make one without having to buy from the vendor) like the MakerBot?

He is promising to release the designs after they're finalized so it should be possible. The part that caught my eye is that it should be incredibly easy to expand the size of the device compared to many of the other reprap like printers out there. For height alone you just need to get some longer rods from a hardware store, doing the length and width looks like it would be slightly more difficult as you have to make a larger bed but it doesn't look that difficult.

Good point!

What are some uses cases for something like this? For what reasons would schools care?

Why do some schools have Computer Aided Design and Drafting classes? Or Wood Shop? Or an Auto Shop? Or Art?

To get kids excited about learning how to do things they haven't done before.

Reply on news.ycombinator.com