Hacker News Reader

Comment by idlewords | Hacker News Reader

No one seems to mention that this whole architecture requires a specific technology, in-orbit refueling, that does not exist.

It would be cool to have it! But it's one of those technical problems that grows in difficulty and complexity the more you look at it in detail. It is very hard to move tons of propellant around in zero-G (or fractional G) and the learning curve for it has to be done in space. Multiply the difficulty by a factor of 10 or so if the propellant is also a cryogen.

View on HN

We already do in orbit refueling for the ISS. Those are different propellants and smaller volumes, but we know how to pump liquids in zero-G. Cryogenics are no harder to pump, the only difficulty is storage.

No one knows how to do fuel transfer in zero g, but there must be ways. Here's one: lash the orbiters and make them spin as a centrifuge and, of course, drain from the donor where the fuel piles up.

The problem is not a lack of ideas, it's in the engineering details of how to do it. "Lash together spacecraft" and "spin x as a centrifuge" are also hard technical problems.

What makes this particularly difficult is it can't really be tested and prototyped on Earth, you have to make all your mistakes the expensive way.

It doesn't have to be that expensive. They could start with a small scale demonstrator that could fly on Falcon 9 just to explore feasibility and make the mistakes at less cost. Or once Starship is operational they could fly several different small scale prototypes just to test them in one go, and they could even do this on an early flight that they wouldn't use to launch commercial payloads. Still expensive, because it might need two flights for them to get the confidence they need to go with one design. But not as expensive as "the expensive way".

I'm curious why a crygenic fuel is harder to transfer in space than something like kerosene? Space is cold, so my naive idea is that a crygenic fuel is easier to keep liquid (non-frozen).

Space isn't "cold", it's a thermal insulator. Objects in space are still illuminated by sunlight whenever they're not in shadow, which heats them up, and the vacuum makes it difficult for them to get rid of that heat.

That's why (for instance) the JWST has a sunshade that is considerably larger than the telescope itself, just to keep it cooled to 50K -- which is still well above the boiling point of liquid hydrogen. It's also why the ISS has large radiators so that it doesn't get too hot for human habitation.

A large sun shade for an object in low earth orbit shouldn't be that much of a problem. Transferring a liquid in microgravity should be equally difficult regardless if it's cryogenic or not.

That sunshade will turn into a big sail in LEO, annoyingly.

The issues are: a) boil-off (others have addressed that elsewhere in this these threads), and b) transfer in zero G (something close to an inertial frame of reference). (b) is hard because bubbles of gas form randomly as you drain one tank, so the pump can't work very well. Bubbles of gas are not a problem for rockets normally because they either are always under acceleration (so the gas floats to the top of the tank, while the outputs are at the bottom) or because they have small and full tanks that SpaceX calls "header tanks" to make engine start possible which then resolves the gas bubbles problem in the larger non-full tanks.

It's possible to avoid creation of the bubbles in the first place though. Some tanks could use pistons, and some - flexible membranes between displacing gas and the liquid fuel. Or tanks could be flexible themselves, squeezing the fuel out when their volume decreases and when being fueled, increasing their volume.

If bubbles appear, it's possible to liquify them back. Increase of pressure with possible decrease in temperature (active cooling) could help.

This is an old problem, with quite a few ideas how to solve, some of those ideas even tried in practice.

Retrofitting such things into Starship sounds difficult.

The environment in low earth orbit isn't cold, beyond the obvious direct sunlight the earth is warm and fills a lot of the sky. Shielding stuff from it is a challenge. Even near Mars orbit isn't cold enough to store liquid methane, let alone liquid hydrogen.

Reply on news.ycombinator.com