3 ms·
Absolutely. Net output per $ of opex (+ amortization of capex, decommission costs, etc.) is the only relevant metric. Parasitic power loss is categorically not
by nkoren 10y ago
Absolutely. Net output per $ of opex (+ amortization of capex, decommission costs, etc.) is the only relevant metric. Parasitic power loss is categorically not an issue.
This is a mistake that engineers frequently make when they are too narrowly focused. You see it all the time in rocket engineering. The rocket equation dictates that performance drops off rapidly as the mass of the vehicle becomes heavier, or the efficiency of the engines worsens. So rocket engineers are obsessed with saving weight and increasing engine performance.
However: performance, in absolute terms, doesn't actually matter. What matters is that you get your stuff in orbit, whether that's done efficiently or not.
Up to a point, you of course do need to worry about vehicle weight and engine performance -- with too much of the former or too little of the latter, you won't be able to launch any payload. This is analogous to the "break-even point" for fusion power.
Beyond that point, however, there's a tradeoff to make: if you need to launch a larger payload, you can either improve the weight or the engines -- or you can just throw more propellant at the problem.
Many engineers scoff at the latter approach, because it is utterly inelegant (and doesn't require as many engineers to accomplish). But rocket propellant is cheap. Really cheap. A cost-driven analysis that compares improving efficiency vs. throwing more propellant at the problem will often favour the latter.
The reason SpaceX succeeded in reducing launch costs -- where NASA's engineers failed to do so for 50 years -- is because they were willing to do this analysis and go to market with a lower-performance rocket. They preferred to spend an extra $100k on kerosene than an extra $10M milling engine parts out of unobtainium. Having done so, they then iteratively figured out how to make a cheap rocket high-performance -- their rockets are now very high-performance -- which turns out to be much easier than figuring out how to make a high-performance rocket cheap.
Anyhow, this article gives me a strong whiff of that kind of engineer's bias, where the good is the enemy of the perfect.
[Edit: typos.]
- ckozlowski 10y agoI'm not entirely sure that what you listed is the reason SpaceX has been making such progress; NASA seems perfectly willing and capable of making cutting edge engines (or rather, their contractors do). Whether or not their organization has the risk-taking and agility to change approaches like SpaceX does, that's another matter. However, I don't think in rocketry that "throwing propellant at the problem" is ever a solution, as propellant has weight. Efficiency is indeed incredibly important, because adding more fuel means having to lift more weight, which means you need a bigger engine, which needs more fuel to run, etc in a vicious cycle. One of the solutions of course, is to burn more of it at once (ie. multiple engines), but there's limits to what's feasible. (speed/atmospheric considerations.) In other words, it's not a linear matter of throwing more propellant. Costs can go up, quick. Coming around to the issue at hand, I can absolutely see having a high minimum threshold for a fusion power plant, where the plant needs to be large enough to generate a sizable return (excess power) for it's initial outlay, but also be able to make the maximum return (high consumption) of that power generated. You can build a massive plant in the midwest, but if the demands on it are quite low, then the economics of it doesn't work out. That means that the economically feasible scenarios for a fusion plant at this time are probably quite limited. I'm sure that will improve greatly over time. But it seems to be that the author's point is that right now, even assuming the technological issues are solved, it is probably isn't as cost-effective as some make it out to be.