Aluminum (aluminium) printing

Hi,
I've seen your toolhead-idea "Laser etch tool"
especially the link mentioned:
[www.felesmagus.com]

So, what to do with my "I don't like to read CDs anymore" DVD-Burner laying around?
Exactly!
Now I do have a 200mW red Laserpointer, that's capable of burning plastics.

Exciting, isn't it?
So I was looking for the Tim Taylor version... "more power"
And: there are lots of people selling Laserdiodes at ebay... 2 and 3 Watts.
for example [cgi.ebay.de]
Remember: with any 3W laser you're capable of melting aluminumpowder!
the one above p.e. is not as expensive and fibrecoupled
Some aluminumpowder on the desk, the reprap for moving the fibre above the dust,
a controller for de/activating the diode, another layer of powder... and so on...
And voila, there you have an aluminumpart printed.

Oh and it's etching, too if you want;
this one has a 3W Laser too
[www.photon-energy.de]

'sid




No.. true... that's not "two material meltdowns", but the fibre doesn't get hot, so it won't hurt the reprap

'sid

warnings about thermite but not about how powerful a 3watt laser is? j/k... its pretty obvious that a laser that can melt aluminum is dangerous.

anyway, i DEFINITELY want to attach a laser print head to a reprap. once you get the cartesian bot down... theres SO much you could do with it. its a very crucial first step.

I could be warning on that laser, but i decided that i preached enough for tonight

Instead i will say that using aluminum powder as a base could yield bad results as it tends to oxidize in air when molten. heated aluminum powder may easily auto combust (ouchhh, that's hot!)

Why not try other metal powders like Zink or Bronze? We may have to look into sintering powders or additives that keep the powders from burning up in the air. How do they do that in the industry? Do they use N2 or noble gas atmospheres? Who's our sintering expert?

personally, i'd be just fine with a machine that can do laser etching / cutting of small material. that would be very cool.

Uuups, your right Zach!

Lasers are dangerous, even the 200mW DVD-Burner thingy I'm currently playing around with is hot enough do sting your skin, no need for talking about eyesight
after looking into the beam (there will be none left)
And keep away from the infrared lasers until you're absolutely sure that it doesn't emitting light. If you don't see it, it doesn't mean that it isn't there!

I did some "free hand etching" on my black plastics toolbox, visible but not beautiful
So etching some platics with a reprap and an abandoned dvd-burner would be very easy.

I've ordered some atmega8 to play around with those tiny little steppers I kept from the drive just to see if I can build a microetchingtool with it (36x36mm)
to find out if I can draw a nicer line that way,
or something useful


Fernando,
never stop warning.... you'll never know who's reading something
Well oxidisation, you're right, didn't thought about that.
lit up and burn?? I'm not sure, I don't know... mabe?!
But the laser is melting (the bad and hot 3W one) a lot of things
zink or lead (do not swallow!) or or or
Maybe I get hold of one of those 3W Laser do see what it can do.

'sid

I work with powdered aluminums all the time. A large part of my job is formulation of thermite derivatives. You want to be careful to use a coarser grade of aluminum (which is cheaper anyway ) but the surrounding metal is going to chill the molten material very quickly, likly far too quickly for their to be any real issue with oxidation. My concern would be that you wont get any interlayer adhesion because the metal will cool far too quickly. That of course can be solved by just using very thin layers with each cycle so that the laser melts at least two layers.

Of course then you have a tradeoff, fine enough powder to allow a single pass to melt two layers, but coarse enough to avoid oxidation issues.

Mike

The thoughts and ideas expressed in this post do not reflect those of my employer and are intended only as communications between individuals. Any attempts at implement are at your own risk

Edited 1 time(s). Last edit at 09/12/2007 09:39PM by ohiomike.

looks like we've found our sintering expert =)

Cool!
Anyway, for sintering I'll leave it to the people that know how to do this. I'll be avidly reading on your results, this would be such a great addition to reprap, both in sintering and laser-etching!

Concerning sintering.

My concern on sintering is it seems that the output would always be the equivalent of cast metal. I think maximum utility won't be reached until you can fabricate metal exibiting the properties of other forming processes. Drawn, for instance, or rolled.

Would it be possible, and feasable, to force a different crystalline structure by using an induced current during the sintering? What about standing waves from radiation, (radio, not isotopes?)

Excellent suggestions

My thinking is that it will be easier to extrude drawn wire onto a bed and weld the surrounding powder to it rather than try and change the crystal structure of the metal using radiation. Speaking of induction however, what about the possibility of using high current to melt a small amount of metal for an extruder rather than a high tempature coil?

Although I could see using the laser to heat and then timed application of water to temper the material. When the temps get hot enough for iron injection of carbon to allow for differental hardening in addition to the temper is also a possibility.

Some method to take drawn wire and partially oxidize a portion of the surface prior to inclusion in the final product would also make possible differental hardnesses that are impossible in current methods. Basically feed the wire though a electro-reactive bath and pulse the current based on your needs. You just have to leave sections unoxidized in order to allow good welds to be formed.

Of course its all version 2.0 and higher speculation.


Mike

The thoughts and ideas expressed in this post do not reflect those of my employer and are intended only as communications between individuals. Any attempts at implement are at your own risk

Edited 1 time(s). Last edit at 09/12/2007 09:38PM by ohiomike.

From the Stratasys website (proving you can FDM with metal):

"Metal parts are build up layer-by-layer from metal powder melted by a powerful electron Beam to exactly the geometry desired for each layer. The Electron Beam Melting Technology (EBM) allows for high energy to be used in the gun which lead to high melting capacity and a high production rate. Parts are built in a vacuum at high temperature and the resulting material characteristics are better than cast and comparable to wrought material. The EBM technology provides 100% dense material without any layering effect or weld lines in the microstructure."

hi
vacuum environment challenge is demanding
high voltage Electron beam generation is demanding
EBM diagram
[www.automotivedesignline.com]

tan

how about a hacked (for a wayyyy more powerful beam) old tv without a screen? then you could instead just make the whole chamber vacuum and put your metal powder where the screen should have been.

Is it just me or are these suggestions becoming progressively more dangerous!

I was hoping to make my fortune by coming up with a new rapid prototype method to replace conventional manufacturing, but I have now abandoned that idea. I'm going to put all my money in hostpitals specialising in eye surgery, cancer or the treatment of burns.

(I'm only joking....these ideas are really great....)

Hi

Just wanted to add my idea to this.

What if you exstruded aluminium in much the same way as the plastic but heated the build chamber so it was near the melting point of the metal so it bonded easily.

What I imagine is this. The build chamber would basically be an oven. The oven would be a box with metal sides and be well thermally insulated on the inside. Ordinary oven heater elements could be used to heat the enclosure and because of the insulation it will be able to achieve a very high temperature.

The build platform which would need to be all stainless steal to prevent corrosion and have carbon bearings on the sliding surfaces. The platform pulled back and forth by rods which exit through clearance holes in the sides of the oven chamber. These rods could be split and have a thermal barrier in so as to prevent any of the extreme heat reaching the control motors and mechanisms.

The metal extrusion nozzle would be long and move up and down in the chamber. On the end would be the hater element (Not sure how you would make that.) and it would be electrically attached by flexible copper cables going to ceramic fittings in the walls of the oven. These fittings would be far apart negating the need for electrical insulation on the wires. Where the long nozzle enters the chamber it passes through a close fitting cooling heat sink to prevent the heat reaching the sensitive wire feed mechanism.

The atmosphere inside the oven could be made pure Argon to prevent oxidisation be lightly pressurizing the chamber to 1psi. So the chamber would not have to be massively strong.

So after you have all that then you just have to find the optimal temperatures for extruding the aluminium and the chamber itself and hopefully you could bond it just like plastic.

Of course the flow properties of aluminium may make it all splodge up and go wrong so ignore me if I'm thinking in the wrong direction here.

... aluminium is melting at above 650°C, so you need around 600°C preheat in the chamber - and all parts in the enclosure have to withstand the ambient temperature!

---

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

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

That's why you would need carbon bearings.

[www.graphalloy.com]

They can be used up to a very high temperatures. And stainless steel will remain easily strong enough for the application.

Although I am sure there would be many many teething troubles. Dealing with expansion of parts and getting the metal to lay in a neat smooth line if that is at all possible.

But it seams a far simpler idea technically to build at home than trying to rig up lasers or build vacuum chambers for electron guns to do the work. All the parts are simple and almost everything is available.

The idea of using aluminum extruded in a high temperature chamber is interesting, but how would you manage the flow of liquid metal? I was thinking of something along the lines of surface tension printing, where a drop of liquid metal (small diameter) would be dragged over the surface to draw the structure. The main issue would seem to be that the first layer wont have much to stick to. However, assuming that you draw the first layer and also assuming that you can keep the drop formed well enough that it is constantly available for adding more metal to the local area, then it seems like we could build this way.

The idea is that as the liquid metal drop is dragged over the preheated metal surface (to just below the melting point) it adds enough heat to allow metal in the drop to bond with the metal below, but only local to the drop. How you do that without causing the whole section to liquify is beyond me.

It seems to me that this could be tested using something like wax... heat the wax until it is liquid and forms the first drop.. then drag it around a cold surface to draw your pattern.. keep drawing until the wax object is formed... if you can do it in wax, it might be possible to do with metal. And wax has a much lower cost for entry than attempting the control software and dynamics on a metal printer directly.

Just an idea, though.

Hi everyone,

I'm new to the whole reprap thing, but the ideas coming out here are just awesome. I've only come across reprap online yesterday, and as I'll be busy with work and a holiday for the next few weeks, it's nice to be excited about something once it all dies down.

It seems to me from all the above comments that what we are trying to acheive is something that can extrude aluminium with the following properties:
- good wetting to allow it to bond to previous layers
- just the right temp / feed flowrates
- something that can withstand the heat of the forming process without incurring damage
- all in an atmosphere that will prevent corrosion / oxidisation during forming
- and just a little bit melting of the previous layer, enough to bond with it but not enough to go blowing holes or melting too much.


enter the MIG welder.


No need to re-invent the wheel here. I'm not sure if a MIG welder uses pure aluminium; it's probably a mixture of aluminium and some other metal and probably some flux aswell. I used to work at austal ships where they turn a pile of aluminium into a big-ass ship. I'm a sparky and not a welder / fabricator / boilermaker or anything like that, but I have seen guys building some aluminium up by hand using a mig, where the ally was cut too small / short etc. Surely all we'd need to do is automate the motion + operation of it, instead of coming up with something completely new.

Also some other things come to mind with this:

- Not sure what the smallest MIG welder you can get is, but it may be a bit too big for making small things with precision. if so then either it can be modded using it as a starting point and trying to keep this time-proven method working the same way just on a smaller scale, OR, why not just make a super-sized reprap, and use it to extrude large components instead of small ones, kinda like a cnc welder but used in a different way.

- May require some software that will make it move in "zig-zag" or "circles" pattern, kinda like real welding (hmm... if it's circles but each one is building another layer should I be calling it "helix" instead?.. lol )

- Also MIG's can weld a wide variety of materials (some rather exotic types aswell I believe), by utilising different "filling material" and setting up for different current, different technique etc.... so I suppose if we cnc it, maybe we could extrude components from STEEL???, stainless, or maybe even nickel and cobalt superalloys? - I've seen these welded via tig not mig, but I can't see any reason why this wouldn't work...


If anyone else is keen on working together to come up with some ideas, I'd be most pleased, as I'm hoping to purchase some components to try something reprap related in about 1 month. Once it's all running and I've made some stuff out of plastic and had some fun with it, I'd like to try some mods, and hopefully one day something like this where maybe eventually reprap could make some proper industrial equipment like pumps or valves, maybe even small rc engines? or microturbines??? (okay maybe that's a bit ambitious). Oh while I'm on the topic of making stuff, if anyone's got a link on some info about putting a milling head onto a reprap to make it into a 3d cnc milling machine for aluminium I'd be most interested.


Cheers guys,

Keen on hearing replys

... milling aluminium with a reprap is hard to achieve - the reprap isn't sturdy enough

Better go the other way around - repstrap a CNC-router ...

---

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

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

The reprap machine and its derivatives was really developed for lightweight applications such as Fuse Deposition Modelling (FDM) or paste extrusion, so is probably the wrong starting point for a CNC milling machine which benefits from a more rigid frame. Mig welding also gives off molten weld spatter which might melt a reprap machine!

You should also consider the cost and accuracy of producing a part using mig welding. Wire is not cheap, but also the electricity and gas usage would be significant. This is probably the reason why metal prototypes in industry are produced differently using methods such as sandcasting using rapid models, or sintering.

I would however encourage you to go ahead if you think there is merit in the design as there are too many "can't do that" postings on forums as it is and I dont want to add to them. The benefits of a direct to metal rapid prototyper are clear to everyone here.

There are a few alternatives that you might consider as a basis. They are not self replicating (They could be considered as Repstraps). But it might be worth a look.

HydraMMM

Mantis CNC (very cheap to make)

Also Google "diy cnc plans" and look in the images tab

I'm sure the guys here could also suggest other development streams for you to look at.

Marcin Wrote:
-------------------------------------------------------

>
>... but I have seen guys building some aluminium up by
> hand using a mig, where the ally was cut too small
> / short etc. Surely all we'd need to do is
> automate the motion + operation of it, instead of
> coming up with something completely new.
>,,,

I've done this, with steel MIG. (built up layers of weld to fill a gap)

this idea could be done, but there are some problems to overcome, here's the few I can think of off the top of my head.

-how are you going to seperate the created part from the bed?
I really don't know how you'd do this. you're going to need disposible beds for sure.

-you're going to need a way of gaugeing the heat of the part you're building up - half the skill in doing it by hand is knowing how far you could go before you had to stop and let it cool for a bit. (otherwise your hard work just slags away)
Maybe spray water after every layer?

-you're going to need a way to keep the splatter from gumming up the rest of the machine. never done aluminium but steel splatter can stick to steel pretty well - you don't really need little steel balls jamming up your axis bearings.

-costs could be high, you've got the steel wire, argon, and a fair amount of electricity.

-maintanance on the welder.
this could be just a matter of manualy cleaning the tip, shroud etc after every x min of use. welders can be fairly tempermental (at least, the ones I used were, though they were fairly old machines)


if you're welding auminum you need some slightly more fancy welding equiment than your run of the mill steel welder from what I understand (never did aluminum myself) the fancy stuff I think does some cool things with the power supply iirc. it's less forgiving heatwise than steel, which has a much larger temprature range from being solid, to compleatly liquid.

... think of feeding a wire to the surface and melt the connecting point with a laser - much easier than with a welder.

Synchronize the moving head with the wire-feeding and you can place traces ... even with turning the feeder for curves and sharp edges.

Separating the wire from the surface is made by moving the head up while heating.

For separating the object from the bed you have to set first some sort of support that's easier to brake - maybe a thin sheet of the same material or some brittle structures...

---

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

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

Quick thought along these lines, how about using a tig machine and some powdered material? Then you'd simply be relying on heat to fuse the powder. Then using a argon filled welding chamber would get around the need to use up a lot of gas.
It'd still need a fair bit of electrical power though. Although I've heard that inverter style machines are more efficient.

se5a Wrote:
-------------------------------------------------------
> Marcin Wrote:
> --------------------------------------------------
> -----
>
> >
> >... but I have seen guys building some aluminium
> up by
> > hand using a mig, where the ally was cut too
> small
> > / short etc. Surely all we'd need to do is
> > automate the motion + operation of it, instead
> of
> > coming up with something completely new.
> >,,,
>
> I've done this, with steel MIG. (built up layers
> of weld to fill a gap)
>
> this idea could be done, but there are some
> problems to overcome, here's the few I can think
> of off the top of my head.
>
> -how are you going to seperate the created part
> from the bed?
> I really don't know how you'd do this.
> you're going to need disposible beds for sure.
>
> -you're going to need a way of gaugeing the heat
> of the part you're building up - half the skill in
> doing it by hand is knowing how far you could go
> before you had to stop and let it cool for a bit.
> (otherwise your hard work just slags away)
> Maybe spray water after every layer?
>
> -you're going to need a way to keep the splatter
> from gumming up the rest of the machine. never
> done aluminium but steel splatter can stick to
> steel pretty well - you don't really need little
> steel balls jamming up your axis bearings.
>
> -costs could be high, you've got the steel wire,
> argon, and a fair amount of electricity.
>
> -maintanance on the welder.
> this could be just a matter of manualy
> cleaning the tip, shroud etc after every x min of
> use. welders can be fairly tempermental (at least,
> the ones I used were, though they were fairly old
> machines)
>
>
> if you're welding auminum you need some slightly
> more fancy welding equiment than your run of the
> mill steel welder from what I understand (never
> did aluminum myself) the fancy stuff I think does
> some cool things with the power supply iirc. it's
> less forgiving heatwise than steel, which has a
> much larger temprature range from being solid, to
> compleatly liquid.


Why does aluminum welding need special inert gasses, but dumping a ton of aluminum in cauldron/furnace doesn't have this problem and the pieces melt into a nice clean pool? does the oxide rush to the surface?

... when pure aluminium 'sees' oxygene, it's instantly reacting to aluminiumoxide, what's preventing bonding to other surfaces, already coated with aluminiumoxide.

For moulding big volumes this isn't a problem, as only the surfaces will react and when pouring the fluid al in the form, the impurities/oxides will swim on the topmost surface and removed with the overshoot later.

For welding or fusing dust you need an inert gas around, what's either present in the surrounding chamber or is created from the brazing paste while welding ...

---

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

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

Okay, thanks. This is informative. As I've heard similar statements before, what I need to get at is how important is it and why, because there are often conditions where it isn't necessary.

I think a reprap has two things in its favor: Fast time between passes (assuming oxide forms over seconds-hours), and high consistency between passes. My guess is that for low consistency (I.e. worked by hand) it is much more important to have the inert gas. It might be that the reprap doesn't use it and has slightly less ideal parts, but that the commercial version uses inert gas and a more fire-safe air-proof enclosure.

One more thing: We can print aluminum that is 5% magnesium. There are also other varieties. Surely something has to work.

Argons readily available and not particularly expensive, just call you're local welding supplier, you have to but the bottle the first time, but after that you're just paying for the gas.
I'm not an expert, but I can tell you the difference between welding with an inert gas and without one is night and day.

I should point out it isn't "welding" in the classic sense probably reaches much higher temperatures and sparks. Perhaps it generates far more oxide.
This is only going to melt alumn to 1300-1400 F. Does that help?

I would of thought using argon make it oxygen free would help stop oxidization