Powder or Wire

There is a rule of expeditions that nobody argues with: you cannot carry everything. Every polar party, every ocean crossing, every mountain camp has run into the same wall. Somewhere out there, something breaks, and the spare is not in the bag.

Space is the extreme case. A resupply run to the Moon is days. To Mars, months. To the asteroid belt it is years, and it is not coming. A crew working the Belt has to make what it needs from what is around it, and what is around it is metal.

In the last article I argued that the steel mill never goes to space, and that what replaces it is a printer and a drum of powder. I stand by that. But I also admitted I had not seen a convincing design for making the powder, and that admission is where this article starts. Because there are two ways to print metal, they need different feedstock, and both of them assume you are standing on a planet.

Small pieces of the answer already exist. Plastic parts have been printed on the International Space Station since 2014.¹ A metal printer built for the European Space Agency has been running in orbit since 2024 and has made the first metal parts ever produced off Earth.² China demonstrated its own orbital metal printing this past May.³ Simulated regolith has been run through a printer on the station as feedstock.⁴ On the ground, a full-scale habitat has been printed out of concrete and crews are living in it a year at a time.⁵

What has not been done, anywhere, is the whole chain. Nobody has started with raw rock, ended with a finished machine part, and done it without a supply ship. That is the gap. I do not have the answers. I have some ideas, a few of them mine, and a set of questions I think are the right ones. I would like yours.

Powder: the bin

The first family is powder bed fusion. The build volume is a bin of fine metal powder, spheres a few times the width of a hair. A blade sweeps a layer across a plate, a laser or electron beam traces that layer's shape and welds it solid, the plate drops one layer, and it repeats thousands of times. The part grows buried in loose powder, and the powder around it holds it up, so overhangs come for free. When it is done you dig the part out, cut it off the plate, and heat treat it to relax the stresses from all that melting and freezing.

This is the precision option. It is also the one that makes the least sense off Earth, because everything about it depends on gravity keeping the powder in the bin and on a chamber full of argon so the hot metal does not burn.

And the powder itself is the hard part. You do not make it by grinding. Grinding gives you jagged flakes, and flakes do not flow, do not spread flat, and do not pack. The printer needs powder that behaves like fine dry sand, and the way industry gets that is gas atomization: melt the alloy, pour it in a thin stream, and hit the stream with high-pressure argon. The metal shatters into droplets, and the droplets freeze into spheres as they fall through the tower. Sieve out the size band the machine wants and melt the rest again.

Read that once more with space in mind. The droplets fall. The tower works because gravity pulls the spray down through cold gas and gives each drop time to freeze into a ball. Take away gravity and the spray does not fall anywhere. It drifts, collides, and clumps. The single most basic step in making printer powder assumes a planet under your feet.

Wire: the spool

The second family feeds material instead of spreading it. A nozzle moves, wire feeds into a laser or an electric arc, and the melt is laid down like a weld bead, pass after pass. The feedstock is a spool, it cannot float away, and it welds happily in vacuum. This is why the one metal printer flying today is a wire machine, not a powder bed.² The trade is that you get a rough, near-net shape that needs machining wherever a tolerance matters.

It is tempting to stop there and call wire the answer. It is not free either.

You still have to make the wire. On Earth, wire is drawn: a rod is pulled through a series of dies, each one smaller than the last, with lubricant between rod and die. Lubricants outgas in vacuum. The dies wear. The rod has to be cast first, which means a mold, which means controlling where molten metal goes without gravity to pour it. None of that is impossible, but none of it is solved, and a spool of wire is not something you find lying on an asteroid.

Then there is heat. A weld bead on Earth cools by conduction into the part and convection into the air around it. In vacuum there is no air. Heat leaves only by conduction and radiation, so the melt pool runs hotter and stays molten longer, and a part that would be fine in a shop can warp or slump in orbit. The station printer has already had to stop a build after an anomaly and prove it could resume where it left off,⁶ which is exactly the sort of unglamorous thing that matters when nobody is coming to fix it.

And the bead itself changes shape. On Earth gravity flattens it. Without gravity, surface tension wins and the bead sits up rounder and taller. That changes how layers bond and how much machining you need afterward. Which brings you back to hauling a mill.

So the honest scorecard is this. Powder gives you the better part and the worse process. Wire gives you the better process and the rougher part. Neither one was designed for the Belt, and each fails in a different place.

What an asteroid actually gives you

This is where the Belt starts to look better than the Moon.

The metallic asteroids, the M-types, are not ore in the sense a miner on Earth would use the word. They are not iron oxide waiting to be reduced in a furnace. They are largely iron-nickel metal already, alloyed by the same processes that built the core of a planet, with cobalt and platinum-group metals mixed in. There is no smelter to build. The job is not refining so much as separating and shaping. And the surface is loose, so you are scooping, not blasting.

A rough chain looks like this. Scoop regolith. Pull the metal grains out of the silicate dust with a magnet, which costs nothing and needs no chemicals. Then either melt the metal and shape it into wire or powder, or take a route that is over a century old: the Mond process, in which carbon monoxide gas reacts with nickel at low temperature to form a vapor, nickel carbonyl, which decomposes on a hot surface to lay down pure nickel and give the carbon monoxide back to be used again. It refines and deposits in one step, and it has been studied for asteroid use for precisely that reason.

Every step of that chain has a problem in it. Those problems are the questions.

Question one: how do you hold powder down?

If you want the precision of a powder bed, you have to keep the powder in the bin with no gravity to do it.

The obvious answer is a vacuum, and the obvious answer does not work. A vacuum moves powder by moving air past it. In space there is no air. You would be, as I put it to myself while thinking this through, sucking against the powder.

But that points somewhere useful. Powder bed machines on Earth already hold the bed down with a pressure difference across a porous build plate. It is not airflow past the particles, it is a gentle clamp. That could work in a sealed chamber with a small charge of argon. The cost is that you are now hauling or manufacturing inert gas, and running a machine that must never leak.

Here is my idea, offered knowing it has holes in it. A magnetic build plate. Asteroid metal is iron-nickel. Iron-nickel is magnetic. An electromagnet under the plate could clamp each fresh layer of powder flat, and switch off to release the finished part. Magnetic force on a small particle is strong relative to anything trying to move it, so the field would not need to be enormous.

Two catches. The laser heats the metal past the point where it stops being magnetic, so the part loses its grip as it is built. That may not matter, since the fused part is anchored to the layers below. And a strong field tends to stand particles up in chains along the field lines instead of letting them lie flat, which fights the blade. Neither looks fatal. Both need somebody to test them.

It only works for magnetic metals. Titanium and aluminum are out. But titanium and aluminum are not what the Belt is made of.

Question two: how do you make feedstock without a fall?

Say you have clean iron-nickel. You want powder or wire from it. The atomization tower assumes gravity and the drawing bench assumes lubricant. What replaces them?

One idea for powder: spin. A rotating disk or a spinning electrode throws droplets outward by centrifugal force, and a large enough chamber gives them time to freeze before they hit the wall. This is a real process on Earth and it does not care which way is down.

One idea for wire: skip drawing. Cast rod directly to size and feed the rod. It will be rougher than drawn wire and the feed mechanism will have to tolerate that, but it removes the dies and the lubricant in one stroke.

A third route makes neither. The Mond process deposits metal atom by atom out of a gas. Whether that can be steered into a shape rather than a coating is an open problem, and if somebody has worked on it I would like to read it.

Question three: where does the heat go?

Both families melt metal, and both have to get rid of the heat afterward. On Earth the air does most of that for free. In vacuum, radiation is the only way out, and radiators are big, heavy, and easily damaged. A printer in the Belt is as much a radiator design problem as a printer design problem, and I have not seen that treated seriously anywhere.

Question four: where does the power come from?

Melting metal takes a great deal of energy. The Belt gets a fraction of the sunlight Earth does. A solar array big enough to run a metal printer out there is a large structure, and it is a structure you would want to build in place rather than ship. Which means the printer has to build the array that powers the printer. Somebody has to think carefully about which comes first.

The alternative is a nuclear source. Small fission reactors for space are being designed now. They solve the power problem and create a different set of engineering questions, all of them solvable and none of them solved.

Question five: what do you build first?

This is the one that ties the others together. A machine shop in the Belt has to bootstrap itself. The first printer arrives on a ship. What is the first thing it makes? Probably parts for a second, larger printer. Then the separator. Then the radiators. Then the array or the reactor shielding. There is an order to this that nobody has written down, and getting it wrong means a crew sitting in the dark with a broken machine and no way to make the part that fixes it.

What I am asking for

I have laid out a chain from asteroid to part and marked the places where it breaks. I have thrown out ideas, including one of my own about magnets, and tried to be honest about which are borrowed and which are guesses.

Now I would like to hear from you. If you are a welder who has thought about what a bead does with nothing pulling it flat, I want that. If you are a metallurgist who knows why the Mond process would or would not scale, I want that. If you are a student with a sketch of a spinning atomizer on a napkin, I want that too. Send it in. This is what the magazine was built for.

The people who figure this out will not all be at a space agency. Some of them are reading this.

Ideas in Star Bound are published open — take them, build them better. Our open proposal.

Comments

Star Bound is not a forum, so comments run the old way: if you have comments, please write to us at hello@starboundmagazine.com and they will be published here in the next update, with your name on them. Say if you would rather stay anonymous.

No comments posted on this one yet. Yours could be the first.

Share this article

Copy link Facebook X Email

A note on where this comes from. I am not a metallurgist, a spacecraft engineer, or a degreed engineer of any kind. But I am a practical one, with more than forty years in heavy industrial construction: welding, millwright work, hydraulics, mining, and thirty-three years of electrical construction and commissioning on the North Slope of Alaska, in a place that punishes anything built wrong. Much of that time I have worked as a field engineer, which means I am the one standing in front of the equipment when the drawing turns out to be wrong, and the fix has to be made on the spot, with what is on hand, and has to work the first time. Anyone who has done that long enough learns that a lot of engineering gets corrected in the field by people with more experience than the people who drew it. That is the eye I bring to these subjects: not what should work on paper, but what actually gets built, and what breaks. The facts I take from other people are footnoted so you can check them and read further. The ideas built on top of those facts are my own, and I would rather see them knocked down by somebody who knows better than left standing because nobody looked.

Sources

1. AMFG, 3D Printing in Space: A Guide to Lunar and Orbital Additive Manufacturing — polymer printing on the ISS since 2014 and the Additive Manufacturing Facility. ↩
2. Airbus, The world's first metal 3D printer for space is on its way to the ISS (January 2024); NASA, 3D Printing: Saving Weight and Space at Launch — the ESA Metal 3D Printer producing stainless steel samples in microgravity. ↩
3. 3DPrint.com, China Becomes Latest Space Power to Demonstrate Metal 3D Printing in Orbit (May 15, 2026). ↩
4. Additive Manufacturing in Space: Technologies, Flight Heritage, and Materials, Technologies 14(3):165 — the Redwire Regolith Print experiment on the ISS. ↩
5. Max Polyakov, How 3D Printing in Space Has Transformed in the Last Decade — ICON's printed Mars Dune Alpha habitat and NASA's CHAPEA analog missions. ↩
6. Metal AM, ESA reaches fifth ISS metal Additive Manufacturing milestone — build interrupted by an anomaly and the recovery procedure to resume it. ↩

Gas atomization, wire drawing, the Curie point, and the Mond process are standard metallurgy and are described in any materials engineering text; they are not footnoted individually.