Pwdr - an open source powder 3d printer | Hacker News Reader

Pwdr - an open source powder 3d printer

pwdr.github.com

54 pointskephra35 comments

This is awesome, I had forgotten the kickstarter to make an inkshield, apparently the guy pulled it off and made 'em [1]. Way to go. I apologize in advance for killing his web site.

[1] http://nicholasclewis.com/projects/inkshield/

Indeed, I've used an earlier version of Nicolas' inkshield for the Pwdr machine, it works great!

My initial reaction was this might be a hoax, or at least vapourware, on account of the lack of any print pictures or videos, just the very nice glossy shots of the machine sitting idle.

However I did find one video: http://www.youtube.com/watch?v=bzX5x1A1AMg

And one photo: http://www.thingiverse.com/image:161441

So, seems the results may lag a bit behind the presentation, but this is still really exciting to see. Fair enough for something they're calling v0.1, as well.

The focus was mainly on the machine and not the results, as they can still be improved. But i've added a section on the Github website with pictures of print results. Hopefully this proves that it isn't a hoax ;)

Very nice, thank you!

Didn't mean to be too cruel about it, it's just that with a new 3d printer announced almost every day I always look to see what the achievable print quality is like.

It's a great project you're working on, I'm sure many people (me included) will be watching closely.

Oh no, I totally understand your first reaction. Hopefully, others will pick up the project too and start experimenting and improving the machine. And one of the reasons to get people's interest is to make it all nice and shiny. As you've mentioned, a new 3D printer is announced almost every day, but 90% just looks crap

It's very superficial of me, but I wish more open source hardware projects went to this much effort to "sell" their product.

Beautiful page design, well written copy, nice photography and a great looking device.

I completely agree, this kind of presentation is so great to see!

This is one of the hard problems open source hardware has at the moment. The community is just now realising its existence, so hopefully there will be more so us working on solving it soon.

Without going into too much depth (which probably warrants a whole article of it's own), the problem is that once you have designed something electronic and built a few, actually manufacturing more than a handful of PCBs (>10) not only becomes a whole different ball game, but requires a completely different skill-set to actually pull off well.

You'll have to remember that most hardware people aren't web devs or designers. Just as most web devs aren't ops people. Sometimes, you just need to get something out to show people. And often times, the people that hardware people show it to is other hardware people, and they don't mind it as long as there's good content.

While I do agree that the page design is nice I don't consider this a good presentation of the end product or an open source project. The first thing that you should show if you are selling the product—doesn't matter if it's to fellow open source developers or customers—is that it works. And the best way to show that is with a simple and short video.

It could be of course that the presentation is not yet finished completely. Nevertheless, the idea is really nice and I would love to see the project take off.

Thanks for your feedback. Meanwhile, I've added some pictures and a video of the printer. A more extensive video should follow soon.

Am I the only one waiting for a 3D printer which parts can be created with a 3D printer?

Isn't that the whole point of the reprap projects?

You're going to be waiting a while if you're hoping for a 3D printer that you can buil at home and can churn out steel rods, bolts and belts.

Pwdr can do aluminum oxide and likely also other sinter metals.

A step in the right direction. I've always felt that home 3D printers will truly become a game changer only when they can make metal items, like DMLS[1] does.

[1]: http://en.wikipedia.org/wiki/Direct_metal_laser_sintering

Not all valuable parts are made of metal. You can do plenty with plastic already. Just look around you.

A machine that can make metal parts from a blueprint can also make all kinds of weapon. Do you think such a 3d printer would be legally available to the general public?

They already are. They're not cheap though.

That's interesting. I would think that a machine capable of making arbitrary metal parts would very quickly be subjected to heavy government regulations?

The parent poster is probably referring to CNC milling machines. AFAIK, gun parts made on them are subject to the same regulation/registration as if they were purchased in a store or made in a machine shop. Also, using CNC machines takes skill -- they're not "set and forget" -- and I would imagine that the kind of people with said skill are smart enough not to tempt fate and make illegal gun parts.

Yeah, that's what I thought. My surprise was at the notion of set and forget machines for arbitrary metal parts.

Why the downvote? I was just pointing out something that the gov't is sure to latch on to.

Actually, there was a big kerfuffle on Thingiverse when someone published, as proof of concept, a design for a plastic rifle magazine: http://www.thingiverse.com/thing:11636

People have also done the lower receivers for the same rifle (an AR-15): http://www.thingiverse.com/thing:11770

Indeed, I've tried alumina, but any other powder materials should work. Creating a solid part is done by sintering or fixating (e.g. with resin)

what type of laser do you use for the SLS process?

The SLS process hasn't been implemented for the Pwdr machine yet, but a small IR Laser has been used by others for SLS: http://builders.reprap.org/2010/02/first-tests-building-powd...

So how does this compare to the quality of the rep rap? Is this a better approach?

It will compare badly at first.

RepRap/Mendal/Prnrbot/Makerbot all use a plastic extrusion mechansim (think robo-spider) which can make a lot of cool things but cannot make a bunch of other things. In particular 'spinning' machines (as I tend to think of these) cannot make large overhangs because the material they are using has no structural stability while being extruded. [1] The "Dual 'struder" on the Makerbot Replicator opens the possibility of a 'support' material that can be removed, this requires filling in with the support material to support the non-support material being extruded over it. The support material is later removed with a solvent.

In powder based machines the 'unbound' powder it its own support. So overhangs are not an issue. The challenge with powder machines has always been that the materials have relatively poor structural stability (in the cornstarch variety) or require a precise (and) delicate sintering process, or a 'firing' type process. In all cases having both bound and unbound material present can challenge the finishing stages.

[1] These machines melt ABS plastic to the weak/sticky stage as they push it out (typically 220 - 240 degrees). At that temperature a strand cannot hold its own weight.

I'm not understanding how it works. Can someone explain?

1: A very thin layer of powder (plastic) is evenly spread. Then the print head prints a binder on the places where you want to create a slice of your 3D model. Then another layer of powder is layered on top of it and GOTO 1. When all slices are printed your 3D model will be covered in powder which is also the support material for overhanging slices.

It's very hard to make this up from the photos but I think you can see the container for the powder and model with a movable bottom.

-EDIT- This printer uses a binder, but other models melt the powder with a laser (Selective Laser Sintering).

Thanks. A couple follow up questions:

So does the user sprinkle the powder after each slice, or does the machine somehow do it? How would the machine make an even layer of powder?

Why doesn't the powder stick to the print head, or get pushed around unevenly by the print head?

Why doesn't the powder fall out during printing?

So does the user sprinkle the powder after each slice, or does the machine somehow do it? How would the machine make an even layer of powder?

I can't make it up from the pictures, but most printers just trow a pile of powder on top of the last layer and flatten this with an arm. But I think this video will show it much more clear: http://www.youtube.com/watch?v=wD9-QEo-qDk (take a look from 1:50).

Why doesn't the powder stick to the print head, or get pushed around unevenly by the print head?

Because the print head never touches the powder. Just like it doesn't touch the paper in your inkjet printer.

Why doesn't the powder fall out during printing?

Check the video ;)

Well, you've covered pretty much all questions, but here is a reply from the builder :)

So does the user sprinkle the powder after each slice, or does the machine somehow do it? How would the machine make an even layer of powder?

The storage bin moves up a little and the build bin down, and then a counter-rotating roller (mounted on the XY-carriage) deposites the powder from one to the other bin.

Why doesn't the powder stick to the print head, or get pushed around unevenly by the print head?

Indeed, the printer head is mounted ±4mm above the powder bed and can never touch the powder. See the printer head moving across the powder bed in this video: http://www.youtube.com/watch?v=bzX5x1A1AMg

Why doesn't the powder fall out during printing?

Because the pistons form a seal and don't let the powder fall through.

You add layer of powder, melt together the parts you want to print, repeat (add another layer of powder, ...).

Reply on news.ycombinator.com