Metal Reprap Developer forum request from MetalicaRap

Metal Part 3D printer developer forum request

Criteria ; Design a Metal 3D Printer ,
that produces dimensionally finished parts +- 1µ m , largly self reproducing, electrical supply limited to single phase ( ie not 3 Phase,), minimum consumables beyond metal powder ( avoiding need for e.g.argon gas ... etc,),cost for parts less than a small car, buildrate can be slow i.e. .2KG per hour, no bigger than a wardrobe.

could you add the forum link in page
[reprap.org]


If you could remove the development stub references it would help me a lot
kind regards
thanks

Tansen

Rapatan,

I've created a MetalicaRap, and put it in Machines Variations.

I wasn't sure if should go there or in our "types of fabrication" folder.

However, you may want to use reprap-dev instead, since you're developers.

---

-Sebastien, RepRap.org library gnome.

---

Remember, you're all RepRap developers (once you've joined the super-secret developer mailing list), and the wiki, RepRap.org, [reprap.org] is for everyone and everything!

hi
If you have a use for 3d metal printing, do get in touch,
we need you, fund raising, spreading the word, backing us up.
kind regards
MetalicaRap team Denmark

Hi,

I've been looking into this for about 4 months now and here is a summary of what I think is the goings on.

1. x-y tables are not a problem
2. metal power supply is plentiful and cheap with a huge variety of different metals available
3. A blower is needed to blow powder to a surface - hardly a challenge
4. A roller needed to roll over the powder laid on the construction surface - hardly a challenge
5. A laser is needed to melt the powder rapidly and fuse it to the object that is being built

And point no.5 is where it all comes apart.

The traditional CO2 lasers and optics are all bulky so they can be ruled out.

There are semiconductor laser suppliers - but too low in power or dammed expensive if high power.

But a new generation of fibre optic lasers are being made and these may have the right high output power
and the compact power supplies needed to operate them available. Supposedly fibre optic lasers are just
a huge bundle of lasing fibre optic cables with the internals of fibre being made of materials
that allow it to lase such that if you put a coil of this fibre optic cable in a light box, it will pick up the surrouding
light and lase. Cheap. Prices I don't have for fibre optic lasers. Some of the latest semiconductor lasers may
also have gotten cheaper.

That is all about that I have on this.
Where to get them, and who would supply them to developers to make metal repraps I cannot
at this moment say.

If someone has contacts or further information, please do share.

---

--
Wiperman [www.enemygadgets.com]

Great work
The laser route needs to be worked on too,

Laser solution is tricky for the following reasons;
Price of a laser that is powerful enough is currently too high, as you pointed out ? From my research its currently 100,000 euro for nd Yag Q pulsed laser 400W ,
The metal reprap must produce dimensionally finished parts, so it can self replicate (i.e. little use if parts are not between IT grade 7 (+-20µ over 20mm) and IT grade 0)

To get a laser to focus to 5µ is not a problem, as long as the laser light is axial (ie parallel light ie not diverging beam , at the metal powder) this demands that you are over 0.7m away from work piece for high power Nd Yag laser optic's, given a reasonable build size of 250mmx250mm.
So this distance and the required finished part tolerance demands expensive laser aiming device ,

So this makes both the laser and the aiming device too expensive and unlikely to be reprapable for the foreseeable future.
So a few hurdles to be overcome for now,

keep up the research!

P.S. If any body has a used Xray switchmode power supply to donate do get in touch! We need more of them!

A largely non self reproducible machine producing lower dimensionally spec parts eg IT grade 15 (0.8mm over 20mm ) , with highly porous/weaker metal parts either by; printing with Mendel with wax coated metal particle filament and then kiln firing to remove the wax leaving a metal form, or by SLS powder process is possible.

( As Adrian Bowyer says best to stay away from Selective Laser Sintering (SLS) .)

Hi
Do get in touch if you want to back us up

kind regards
rapatan

Hi rapatan

On the MetalicaRap wiki page it says your on the look out for metallurgists to help with the project.

I just finished a science undergrad with a major in materials sciences which briefly covered microstructural development, microstructural design, modelling and manufacturing processes.

If I can be of any help let me know.

-Capo

hi
Excellent that would be great ;
Can I suggest you choose which part of the electron gun you would like to research?

from: Rf input, Rf Ariel,cathode, anode, Lens coil, alignment coils, nozzles, L. alignment coils, nozzles (aperture ) , deflection coils.

see [www.twi.co.uk] schematic shows position of the above
Calculate for 6kw 150KV
unipolar or bipolar cathode to anode connection.
Note:

1. special awareness of print ability of design
2. is their a cheap bootstrap metalicarap option.

kind regards
Rapatan.

I will send direct contact details.

What type of material are you printing? How dense are the parts once built? What kind of surface finish do you get?

Jamie Gearig
GPI Prototype - DMLS

Laser sintering without a inert gas (Argon or at least Nitrogen) will be a problem. Maybe we should consider a small device to generate the gas? it will of course take a lot of work and some engineering, but also has the added value of making the most important technical gases (CO2, Argon, Nitrogen and Oxygen) readily available wherever you run that machine.
Most likely the Linde Process would be the way to go there.
[en.wikipedia.org]

hi
Fascinating idea, maybe you would like to add it to future development section of MetalicaRap?

kind regards
rapatan

I worked on a EOS metal sintering machine some time ago, it had a unit that extracted nitrogen from compressed air, but
do not have any details of how it worked, the unit was small but very efficient and filled the build chamber in a short time, the
3d systems machine had a massive unit with very large tank, the EOS system was much better.

---

Random Precision

rkoeppl Wrote:
-------------------------------------------------------
> Laser sintering without a inert gas (Argon or at
> least Nitrogen) will be a problem. Maybe we should
> consider a small device to generate the gas? it
> will of course take a lot of work and some
> engineering, but also has the added value of
> making the most important technical gases (CO2,
> Argon, Nitrogen and Oxygen) readily available
> wherever you run that machine.
> Most likely the Linde Process would be the way to
> go there.
> [en.wikipedia.org]

I used to do heliarc welding. The cheapest and easiest method of creating CO2 gas was to fill a sealed container like a modified welder's oxygen bottle with filled dry ice. The bottle was modified by cutting off the top about 1 1/2 inches above where the taper blends into the full diameter of the vessel. A closure was constructed similar to a deep diving bell or a submarine hatch and rubber gasket. The diameter of the hatch was the same as the outside diameter of the cut opening on the bottle. 3 or 4 spreader bars were attached underneath the hatch similar to any high pressure apparatus . These bars were mechanically leveraged outward against the remaining taper left on inside of the bottle. A provision was made for a gauge and regulator to be mounted on the hatch after it was secured. A couple of infra-red heat lamps would warm the container and began to melt the dry ice and CO2 would build within the container.

hi
I have been told that EOS 3D printer uses nitrogen it makes itself within a few minuets,
Would be great to know how to create the Nitrogen.

great work
Kind regards
MetalicaRap Team / Rapatan

I think its membrane separation, this site shows some info, here

---

Random Precision

Scientific American had a DIY UV laser article using N2. Reprinted in a SA DIY laser book ~1980 - MIA - Cleanup?

This used a burner followed by bubbling through CaOH solution then drying in ? (CaSO4?, CaCl2). No memory of purity.

Interesting method for N2 ,
If you feel like pricing out the different methods that would be great!
kind regards
MetalicaRap team

I can't price out the different methods for you, but here is some info you might find handy, as I once looked into extracting pure oxygen from the air, and many of the methods are the same, I won't go into detail but you can google them from these keywords:

Air liquefaction and fractionation is the obvious one which is used commercially to do the bulk of the separation. It is possible to greatly improve efficiency over a naive system through careful use of counterflow heat exchangers and regenerators, and understanding of thermodynamics.

Pressure swing adsorption is common and used in those portable oxygen units for medical use. The adsorbent in those is zeolite, but ther are a wide range of adsorbents including activated carbon and activated alumina and others. Different pressure swings are similarly used.

Membrane separation uses a membrane which is extremely thin, usually pure silicone made from dimethylsiloxane. It's permeability is different for different gasses, they diffuse across at different rates and so this can be used, with engineering, as the basis for a separation system. Depending on the gasses you want etc. and the purities it will probably have to be multi stage or combined with other methods because the permeability does not vary by all that much between gasses.

wiperman use a welding machine with tugsten tip have a enclusure that has a small argon purge on it. at least thats an option
thanks

so you guys are talking about what can be used to fuse layers of metal powder cheaply? what about HV arcing? very short arcs at very high frequency low amps, the arcing tip moves over the surfaces where metal dust is suppose to be heated to be fused? its like welding but at a very very high frequency pulse? does that work?