Need help with a toolhead for a working powder printer (focus)

8 months ago I decided to build a powder printer, at that time mostly centered toward SLS. This printer is called focus. a few weeks ago it finally printed with powder in 3D, with repeatable results.

pictures, video's and progress can be found at: http://focussls.blogspot.nl/
Buildfiles, firmware and schematics can be found at: http://www.thingiverse.com/thing:33697

Focus is a reprap based powder printer with a buildbox of roughly 115*120*80mm. The toolhead can support relatively large weights (because of the experimental nature of this machine) and has a top speed of around 30mm/s. It uses ramps as electronics and costs around €600 to build, excluding the toolhead. It can be built with printed parts, simple mdf parts and standard reprap bearings and rods.

Focus is currently fitted with 2 808nm 1,5W laser diodes. It currently prints in a mixture of PA2201 and 1% carbon powder. Even though this does manage to print things (with HUGE amounts of shrink) I want to be able to do more. All powder systems are fully functioning. The firmware has some (yeah... some) problems, but it does work. The lasers heat stuff up just fine. The problem is that I want more out of this printer.

I have tried building a piezo inkjet nozzle (a small attempt) but frankly, that did not work well. the nozzle just leaks, even when there is no presure. Even though there are some amazing people that have already managed to build one, no one has bothered yet to make a complete schematic for one, including all circuits and manuals. I don't feel like fully developing one myself, since there are already some that would work perfectly for me.

The biggest thing that can use change is the toolhead. Mine has 2 lasers, but I would love a drop on demand or stream inkjet nozzle. With inkjet, I can print in Zcorp materials (I might have acces to this material), Ceramics and basically anything that fits through or can be binded with a inkjet nozzle. A heated buildbox is also an option, but that will cost around €200 extra to build. I am however open for suggestions.

Is there anyone that is willing to help in any way form or matter, with either suggestions, investigation or even development of any parts.

Many thanks in advance

Well done dragonnator on this build, I suppose you have already considered stripping out the printhead and rail from a cheap printer, that would give the posibility for colours eventualy.

---

Random Precision

I have C6602 (black) cartridges on stock here cause I concidered putting on the cartridge on the printer. I do not feel much for putting the whole rail on there cause I would have to redo most of the control and let the computer take over, or completely redo all of the firmware and write a new slicer. Also those cartridges are bubblejets and can print far from any binder.

I have actually looked at your work and tried to replicate it, but that will take me more time, which I do not currently have. I was concidering putting a DOD inkjet on there and print with ordinary Gcode. All of you have done amazing work in single nozzle inkjet, but no one has yet made full build instructions to my knowledge.

You have probably seen this, but if not [open3dp.me.washington.edu]
its a bit old now but may be usefull, Adrian built a reprapable printhead with some instructions I think, but
I have no plans to revisit the printhead, I go where my interest takes me, and its elsewhere at the moment,
I did hope that others would follow up on it and improve it, but not so it seems.

Just a thought, but what if you fired just one nozzle of a commercial printhead, if you had the voltage and pulse period
(google) an arduino with firmware modifications may be the quickest way, one thing that I would advise is that the printhead
has a reqular cleaning cycle over a rubber squeegee, as powder collects and hardens on the surface and tends to block the nozzle.

---

Random Precision

What about [nicholasclewis.com] although 96dpi a bit low?

Ben

Hi dragonator,

I've posted your blog in the metal-bot forum too: [www.metalbot.org]

Maybe you can help Jethro with his powder deposition system?

If you need some fibercoupled IR-laserdiodes with 975nm @9Watts out of a 0.1mm-Fiber, I have some ten of them salvaged from pumping banks ... 9Watts on a spot of 0.1 to 0.2mm would be enough energy for ceramics and metal powder too

White powder isn't so good for the IR-lasers, as the absorbing properties changes dramatically when the powder darkens while melting.

Better use dark powders and reduce your energy, so the melting would be much 'softer', no evaporating and with lower shrinking after cooling down.

I'll test soon 'preheating' a zone of maybe 1mm diameter around the sintering spot with a defocussed stronger laser (or bundle of more of the 9W-diodes) and a single melting spot of 0.1mm in the center ... should give much better results than a single spot ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Hello VDX

I know about the white powder problem, but I refuse to accept that I can't print in powders that naturally come in white (such as Nylon powder). The adding of carbon gives some grip and evens out the differences between powder and molten, but there is still change. Darker powders are harder to come by.

Also I am always interested in adding ridiculous amounts of laser power to machines never meant to take that, is there more information about the lasers, cause if they wont cost the whole project budget, I might be interested.

Yvo

Hi Yvo,

look here for some informations regarding the diodes - the 9W-diodes are similar to the 5W ones, but can emit 9Watts of power @9Amps of current, where the 5W-diodes emits 5Watts @5Amps: [reprap.org]

I'm giving away the 5W-types for 50€, the 9W's for 80€ + shipping.

They are salvaged out from 'wrecked' fiberlasers, where some ten of them pumps the fiber resonator ... so the fiber length is shorted to maybe 15 to 30 centimeters (where the backreflection has burned the fiber inputs).

I'm breaking the ends of the fibers, so they will have a clean, perpendicular end-surface and can emit the light in a round spot (see the images) ... will try to polish the ends, but this can take a while ...

Comercial diodes with same powers but until 2 meter long fibers and polished ends will cost maybe 350 Euros in single counts ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]