Hacker News Reader

Comment by ortusdux | Hacker News Reader

Exactly. Many reusable starships flights would be needed to ferry up all the trappings of a working station, while returning reusable elements like the engines from the now mothballed fuel tanker ships.

I just meant to point out that a stripped down starship would consist of 30+ tons of easily weldable steel pre-fabricated into a reinforced pressure vessel. Spacex's interplanetary goals would benefit from treating the tanker starships as expendable, and if someone was inclined to start building habitations similar to those depicted in the link, they could buy up the building-blocks for a song.

View on HN

Nothing is easily weldable in space, and empty hulls aren't habitrail elements that you can connect into something useful (unless a lot of design goes into it up front). Even if the hulls were free, the cost of creating such a space station would be prohibitive.

Don’t metals spontaneously weld in vacuum?

It's not perfect and you wouldn't want to trust it for a habitat. It requires the surfaces to be pretty meticulously clean and it's not as strong as traditional welds.

Yes, under certain circumstances. Two pieces of identical metal, with no surface coatings or oxides, will directly bond if placed in intimate contact. Done properly, the weld would be as strong as the metal. As Feynman put it, the atoms have no way of knowing they are in different pieces.

Similar results can be reliably recreated on earth. Ultrasonic welding rubs two pieces of metal together until the oxide breaks apart leaving pure metal to fuse. Explosive welding creates a plasma that strips off the oxide layer, and then propels the metals into each other. This method has the benefit of bonding dissimilar metals, and usually produces bonds that are as strong as the weakest metal.

They also become brittle due to cosmic particles, if one e.g. wants to think about interstellar travel.

Nothing is easy in space, but 304L SS is easier to weld than 2219 Al or Ti, the main structural metals on the ISS. Welding in space has been well studied by both NASA and Roscosmos. Most testing was done in the 60s and 70s in preparation for possible repairs due to high-velocity impacts. More reciently, testing has been focused on 3d printing via additive welding.

The main issue with welding in space is the lack of convection based cooling, which means the welds take longer to cool through conduction, which can result in a larger HAZ. Increasing the mass and heat capacity of the adjacent material greatly reduces this.

The lack of an atmosphere and contaminants makes space a near ideal welding environment.

The lack of welders who can breathe vacuum makes space a challenging welding environment.

They are already using robots to weld starships. Doing it in space removes variables like air, humidity, and shielding gasses.

The difficulty is on Earth you have a convenient stable mass to anchor your robot to while in space everything is unmoored and unstable if you start swinging around a robot arm.

I assume you would build in a rail that the welding robot would clamp onto.

exactly. there are no large masses in space let alone ones we can anchor things to. the country of canada tried to build a robot arm for space and they ran into this exact problem. seriously, if they would just consult smart people like us then maybe they could have saved themselves the trouble of launching silly robotic arms into literal orbit.

Oh no I totally didn't think of the most famous robot arm in space, one that moves slowly and carefully and was planned and designed to work with the single largest and heaviest thing we've ever assembled in orbit. My point was you can't just grab a generic industrial arm and sling it around it needs a lot of consideration on how to mount and move it, Canada arm moves quite slowly and carefully.

no, i think "welding is hard in space" is much more insightful. it is much better to make a broad and sweeping statement and then discard a bunch of important possibilities.

Reply on news.ycombinator.com