Powder Printer?

How many folk are interested in building a powder printer?

As a sculptor, I'm interesting in making one that will make big plaster models cheap. And big plaster models are useful to make molds to make casts for wax.

Or we can go directly to glass or ceramics:
[open3dp.me.washington.edu]

We'd probably want to incorporate candyfab's powder mechanism:
[wiki.candyfab.org]
but I'd go with inkjetting down binder. Possibly laser sintering, although that's a long way off, and would seem more realistic after doing inkjet.

Anyone interested? I think it's a very useful basic technology to incorporate under the RepRap umbrella.

Also, we should start a list of various hobby-made powder printers.

1) [www.indoor.flyer.co.uk]

Edited 1 time(s). Last edit at 12/06/2009 06:11AM by SebastienBailard.

ive been looking into powder printing a bit myself. for those things that fdm style printing doesn't work as well for. also after seeing the NASA tests of the edm machine i was thinking of something that was using powdered metal and a low power electron beam. still in the conceptualization stages but putting a bit of serious thought into it right now. The reasons im thinking this might work in a home brew system is after seeing a site and video of a very simple crt/electron gun set up and the guy doing the tests stating that dry erase marker on a piece of paper florescences but after a little bit it burns a hole in the paper from the beam it got me thinking

it might require several beams focused on a single point to achieve micro melting of metal powder and it would require a shield gas at least but possibly a soft vacuum with minimal shield gas having been put in before the voiding of atmosphere.

A good place to discuss powder/binder based 3D printing is:

[tech.groups.yahoo.com]

Thanks John!

(via that group), Alvaro M. Fogassa seems to have it all figured out:
"Homemade 3d printer HM3DP;
a homemade 3d printer I've built from a Lexmark Z12 desktop printer."
[homemade3dprinter.blogspot.com]

I think logistics-wise the best way is to clone his setup, while also researching direct-drive bitbanging the inkjet head. (It's not clear to me that they're still making Lexmark Z12's.)

Direct-drive bitbanging the inkjet head means that we've got ownership of the entire toolchain, and aren't just dependent on scavenging a from a set of particular known-good printers. It also means we can go to a 16"x24" powder bed, and so on.

Fogassa and others have done a great job with the hardware and software, also materials, I have been following the yahoo group for some time and hope they would not mind me saying this but I feel the Reprap approach would be a big benefit to the development of a powder bed printer, in no way steal his project but partner its development.

Sebastian I have been following the HM3DP blog by Alvaro M. Fogassa.

I bought 3 Lexmark end of line Printers in ASDA = (UK Wallmart) 3 years ago as a cheap source of stepper motors. Also having several other old HP ink jet printers.

All of these printers in my opinion could be used instead of using the specific Lexmark Z12 used in the HM3DP project. In fact there is a You Tube link to a very similar print method using a HP500 printer in our forums here.

In the You Tube example the layers of plaster are being made manually between each print layer the result is shown of some finished printed parts.

---

Bodge It [reprap.org]
=======================================

---

BIQ Sanguinololu SD LCD board	BIQ Stepcon	BIQ Opto Endstop
BIQ Heater Block PCB	BIQ Extruder Peek clamp replacement	BIQ Huxley Seedling
BIQ Sanguinololu mounting	BIQ standalone Sanguinololu or Ramps mounting	Print It Stick It Cut it

My rep strap: [repstrapbertha.blogspot.com]

Buy the bits from B&Q pipestrap [diyrepstrap.blogspot.com]
How to Build a Darwin without any Rep Rap Parts [repstrapdarwin.blogspot.com]
Web Site [www.takeaway3dtech.com]

I haven't ever used a lexmark for powder printing, I've always used HP because you can order empty cartridges and fill them with whatever you want and HP offers tech papers that explain how to fire the jet(s).

Anyway, the hardest parts of building a 3d powder printer are the 3 axis platform, controller hardware, and controller software. Reprap has pretty much knocked our 90% of the work.

I'll get the HP cartridge numbers and post them up

Problem with Afgasso's blog is a lack of documentation for the novice using bought and scrapped parts. He was looking at doing up plans and parts and the lot similiar to what the Reprappers have done but seems to got fed up. I'd like to see a inkjet 3d printer go as far as Reprap has. Except with pinpoint precision on the goal of building the machine so it can print what ever. Not to replicate itself. If others want that, they can persue that through another group or with the Reprappers.

It needs to be simple. No machining required. Buy the parts. Lets not make it complicated...or try to pass it off as it's not complicated when it is if you don't have practice.

Build the machine, use the machine.

Hey criswilson10.....got a link to HP on how to fire the jets....gotta love control...especially if ya can use a heated type cartridge to fire wax or low melt metal/plastic. Ya I know, not the first to to think it.

Lets see the link!

The european electronics magazine Elektor did a collaborative deal a couple of years ago, running a series of articles with a kit available for a cnc pcb drill. The thing that was unusual about it was that it was a polar coordinate machine rather than cartesian coordinate. That might be a good scheme for a powder printer, with a continuously rotating climbing build rather than discrete layers. A continuous dump of powder and a blade planing excess over the side. Just a thought.

The clever and difficult to make parts of Afgasso's HM3DP printer are the two stainless steel chambers with their sliding bottoms/pistons. Im sure these parts are now possible to print using RepRap. That leaves the construction of the leveling mechanism and the compresion roller.

The other part a slide mechanism for the printer chasis is as simple as one axis of RepRap and is easy to construct using a belt drive with skate bearings on 8mm bar.

The firmware required is also very simple as it is only needed for two axies the Y axis and the Z axis the rest is performed as usual bt feeding an image/png/jpg of the layer to print to the un touched printer in the normal way.

Next is replicating the mixture to use as the print medium.

---

Bodge It [reprap.org]
=======================================

---

BIQ Sanguinololu SD LCD board	BIQ Stepcon	BIQ Opto Endstop
BIQ Heater Block PCB	BIQ Extruder Peek clamp replacement	BIQ Huxley Seedling
BIQ Sanguinololu mounting	BIQ standalone Sanguinololu or Ramps mounting	Print It Stick It Cut it

My rep strap: [repstrapbertha.blogspot.com]

Buy the bits from B&Q pipestrap [diyrepstrap.blogspot.com]
How to Build a Darwin without any Rep Rap Parts [repstrapdarwin.blogspot.com]
Web Site [www.takeaway3dtech.com]

Here is an old concept that uses nylon powder with an infrared lamp to fuse the powder, cheaper than lasers, sls


The rotational recoat is used by EOS i think on the commercial machines also on metal sintering machines, saves a lot of moving parts but limits the build envelope somewhat, but does give a speed advantage.

To Dman
Firing wax through an HP cartridge is difficult because it only does microheating at the nozzle and to get to that nozzle the fluid (ink, wax, glue, whatever) has to be able to flow through a narrow (10 mil I think) spiral channel. If you want to do wax, I would go with the reprap extruder at a lower temp.

To Johnrpm and others considering nylon powder
The old infrared lamp design does work well, even if it is slow, but using
a laser is faster and cheaper. If you use a laser then you do need to design it to have a removable lens because the fumes coming off of the nylon tend to cloud the lens which creates a hot spot and then the lens cracks. Cleaning the lens after every use is a necessity.

To CrisWilson10 & johnrpm

Thats a very interesting bit about nylon powder and infrared lamps from your notes above was it a nylon material that the Deskop factory were using in its sub $5000 3D printer?

"The Desktop Factory 3D printer, which has a list price of $4,995, uses an inexpensive halogen light source and drum printing technology to build robust parts layer by layer from composite plastic powder. "

---

Bodge It [reprap.org]
=======================================

---

BIQ Sanguinololu SD LCD board	BIQ Stepcon	BIQ Opto Endstop
BIQ Heater Block PCB	BIQ Extruder Peek clamp replacement	BIQ Huxley Seedling
BIQ Sanguinololu mounting	BIQ standalone Sanguinololu or Ramps mounting	Print It Stick It Cut it

My rep strap: [repstrapbertha.blogspot.com]

Buy the bits from B&Q pipestrap [diyrepstrap.blogspot.com]
How to Build a Darwin without any Rep Rap Parts [repstrapdarwin.blogspot.com]
Web Site [www.takeaway3dtech.com]

... for protecting the lens you should apply a stream or air (or you work with inert gas) and you have to exhaust the fumes or all your surrounding parts will start to oxidize, as you often have acids and other critical (and carcinogenic) stuff in the fumes ...

For 'low-energy' lasercutting or melting, where a focus diameter above 30 microns is enough, i use a single lens with 200mm focal length, so i have enough room, but here exhausting is enough.

For smaller diameters and higher energy-densities i use a beam-expander and a lens with 35mm focal length, here i have to be carefull with fumes, so i blow the fumes out of the optics path and exhaust it directly after ...

Viktor

I was a builder long ago. When we were pouring concrete floor slabs, we would have level edges on the side formwork and we'd pull a vibrating screed that spanned the edges along the slab as the concrete was poured to level it and give a preliminary finish. I wonder if a vibrating wiper would help consolidate the powder bed surface as the build progressed. It might reduce binder bleed through the powder and help with surface finish.

I run a ZCorp Z510 day to day at work. I can get as much pictures and info as someone who was keen to build one likes, including some technical wiring schematics.

proto, that's a good idea in theory but a bad idea in a world with lawyers. "Yap, that's a suin' alright."

------------------

More generally, there's some good work going on in this thread; I'll comment further when I have another break.

... vibrating with powders isn't so a good idea

You'll mix air in the powder, so it becomes 'fluid', but more voluminuous too.

You can apply vibration when filling, but before leveling and sintering you should let it settle, what can teake a while with fine powders or dust ...

Viktor

Lasers are way out of my budget, The commercial machines have a nitrogen atmosphere
so need the glass to maintain the chamber,which must be cleaned before every build, the video ( in the link)shows a machine that uses inhibitor but another patent uses the opposite effect, print onto powder with black ink, this absorbs the heat from the infrared more than the surounding white material, the nylon powder is a by product of some cleaning process, (don't know which) but as with all these things you have to buy in bulk.
One possible source of nylon powder could be the commercial fabbers, the powder can be re-used from each cake but degrades each time untill its not much use for high quality work, they then tend dispose of it.
I have an old Epson 2000 printer with bulk ink feed waiting to be made into a powder bed printer, the only thing holding me back is the mess that powders make, the dust gets everywhere, it realy is something to consider.

If any one can think of an easy economic way to construct the two chambers required to build a poweder printer that we be a good starting point.

The chambers would need to be 210mm / 8" wide by at least 5" or 8" long then the depth is how every high you want the print to be.

The build area would be 8" x 5" x 5" maybe the next problem having made sutible chambers is to find a very good piston seal material for the chambers to prevent powder escaping and filling the works of the piston drive system.

Solve those problems and the ability to re-use standard printer technology is relativly easy.

Printing the chambers with Reprap would mean they would be the biggest print ever made on RepRap so maybe that is not a possible solution.

Here is my sugestion for the chambers finding a rectangular or square cake baking tin with the flat slidey bottom building an MDF Frame for support and finding some way to stop too much powder egress from the sliding action.

May be we could examine the techniques used by comercial printers to get arround some of these problems when powder is the medium. I gess you would also need to have some heat or de-humidifing system in place as well.

---

Bodge It [reprap.org]
=======================================

---

BIQ Sanguinololu SD LCD board	BIQ Stepcon	BIQ Opto Endstop
BIQ Heater Block PCB	BIQ Extruder Peek clamp replacement	BIQ Huxley Seedling
BIQ Sanguinololu mounting	BIQ standalone Sanguinololu or Ramps mounting	Print It Stick It Cut it

My rep strap: [repstrapbertha.blogspot.com]

Buy the bits from B&Q pipestrap [diyrepstrap.blogspot.com]
How to Build a Darwin without any Rep Rap Parts [repstrapdarwin.blogspot.com]
Web Site [www.takeaway3dtech.com]

In Afgasso's blog, I think he uses MDF sheet for the bins, judging by the amount of powder that drops through the zcorp the seals do not need to be perfect as long as some sort of tray is fitted to catch the debris, I would give some thought to having a hopper type feed, this will give a smaller footprint for the build size and reduce the distance of travel for the recoater, although its not trivial keeping it replenished.
Last week I put new door seals on the back door, they have aluminium extrusions with C shaped rubber covered canvas type material, would make a very effective seal for powder???, (from B&Q) , an overflow bin will be needed to catch the surplus powder on each recoat.

MDF does seem to be the easiest to get and if one can buy with laminate coating(fake wood/color) at say Home Depot, Building Box...whatever, they will cut it for you. It's how one puts it together that will result in a square and straight box. If you use the uncut edge with the laminate to butt each together, one might get good results without shimming.
The seal for the building platform isn't critical like johnrpm suggested. The issue isn't some leakage but that the platform raises and descends parallel and accurately. I think if one uses a strip of felt glued or stapled along the edge of the moving platform, this should provide a easy to find and attach seal. Like johnrpm suggested, an overflow bin is needed...ofcourse there should already be one for the leveling.
Also, like johnrpm said, instead of two bins, use one building platform and a v-shaped hopper bin with an auger type screw driven by a rack and pinion gear along the length of travel so no need for a motor to drive auger. I believe with some proper thought into the design, the slot opening could be opened with the travel aswell. One direction it opens and then is closed at the end of travel(or very near it) and the other way when it's printing it isn't activated...slip gear or something to that effect.

I was just looking at something i had seen a while back and thought of a posible idea.

mount a hp cartridge next to a reprap extruder and a powder dispenser. then have the extruder build a outer wall, deposit the powder against one edge and have a fixed distance wiper on a actuator to drop it below the level of the extruder nozzle to level out the powder to the height of the formed wall, then scan the print cartridge over the powder as normal for a powder printer.

[spritesmods.com]

was what gave me the idea but attach the print control to the extruder controller and poof, quick powder printer. it would use a small ammount of plastic to build the wall but keep that about 2 layers thick and it shouldnt be to much of a burden.

Brilliant thread. The RepLab project considers a powder-bed printer one of the basic technologies, mostly for working with water-bound powders in ceramic, glass and metal with kiln sintering.

The best notion so far has been combining the HM3DP inkjet hack with a CandyFab gasketed powder dredge bed and using Open3DP formulas for ceramics as the test medium. This seems to be pretty much the consensus of this thread also.

This is a project I want to work on soon, as in over the coming year. Excited to see where this is going.

Oh yes I had miss rememberd the build bins they were lined with stainless steel not made of stainless steel. As i said earlier.


http://homemade3dprinter.blogspot.com/2009/08/build-bin-parts.html

---

Bodge It [reprap.org]
=======================================

---

BIQ Sanguinololu SD LCD board	BIQ Stepcon	BIQ Opto Endstop
BIQ Heater Block PCB	BIQ Extruder Peek clamp replacement	BIQ Huxley Seedling
BIQ Sanguinololu mounting	BIQ standalone Sanguinololu or Ramps mounting	Print It Stick It Cut it

My rep strap: [repstrapbertha.blogspot.com]

Buy the bits from B&Q pipestrap [diyrepstrap.blogspot.com]
How to Build a Darwin without any Rep Rap Parts [repstrapdarwin.blogspot.com]
Web Site [www.takeaway3dtech.com]

Too many so called ideas of different types of printers. I thought this thread was about building an inkjet powder printer. Not combining with Reprap. Reprap is not a god like machine. We know of the different types of printers and we have seen people homebrewing them, so we know what works and what doesn't at this point.

Focus on building an inkjet powder printer. Why do people talk so much? Move forward not backwards. Let some corporation invent a new 3d printer while we can build a cheap printer based on earlier designs that have been around for many years. Everyone seems to want to start something cool but never finish it or want to take it in a direction that is pointless. So quit taking the conversation in other directions.

I know I talk big and have nothing to show because college consumes me, but I've followed many of these conversations and they all led down a rabbit hole of nothing.

Like I've said...build the damn machine and use it.

Dman
Not wishing to offend anyone but I take the opposite view, there is no such thing as to many ideas, I understand people getting frustrated with to much talk and not enough action, but the exchange of ideas and information is the life blood of forums such as this one, RBisping has made a valid point in my opinion, very few if any 3d printers can handle multiple materials, So please please please let us encourage as many ideas as possible, maybe from many ideas a gem will appear.

... And as nice as it is to be able to copy expired patents, part of the reason that existing 3D printers are so expensive is because of the technology they are using. Seeing as our design requirements are quite different than 3D printers in terms of cost, precision, reliability, and construction materials, it shouldn't surprise anyone that we'll need to do our own innovation at least some (and probably most) of the time.