It's weird to consider "parasitic" loss a problem. You just factor that into the total output. If it's high enough you're good.
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.]
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.
In theory, you are correct. However, you are making the assumption that efficiency is a static number. In reality, parts age, get dusty, out-of-alignment, etc. Let's say you are building something that makes 10 units of power, but 8 of those units are needed to keep it running (when working as designed). Say that the part that uses the 8 units of power is nominally 90% efficient, but after you've run it for a few months it drops to 80% efficiency. Now you need 9 units of power to keep running, cutting your output in half. This is not just a problem in terms of dropping output; it is also an issue from the standpoint of understanding your machine. It's hard to engineer something when small changes have large effects. Your output just dropped 30% - is something broken, or is the fan just dusty again? How long do those electrodes really last anyways? Did someone use the wrong cleaning fluid? Maybe the pressure gauge is just off by 1% again.
All those things add up, just like coding a big ball of mud can make debugging difficult.
Almost none of the problems listed there seem to be deal breakers. Most of them seem to be a matter of small improvements that probably aren't the primary focus of research right now.
For example I'd imagine researchers have a fairly steady supply of tritium, they don't need to focus fairly hard on recovering it all, just enough to save on costs since it's pretty expensive. Why bother trying to recover 100% of tritium for a reaction that you can't even get energy positive yet? Trying to make it more efficient when you can't even make it work seems to be putting the cart before the horse.
And that seems to be the case with the fuel, too (at least how described in the article). Once we get the process working with the easy fuels it seems that the next target for research would be using fuels that might be better but harder to work with.
>Corrosion in the heat exchange system, or a breach in the reactor vacuum ducts could result in the release of radioactive tritium into the atmosphere or local water resources. Tritium exchanges with hydrogen to produce tritiated water, which is biologically hazardous. > Most fission reactors contain trivial amounts of tritium (less than 1 gram) compared with the kilograms in putative fusion reactors. But the release of even tiny amounts of radioactive tritium from fission reactors into groundwater causes public consternation.
He seems to negate his own point by saying that the tritium output is less than a thousandth of what it is in fission. It might not be perfect but that sure sounds like an improvement to me. Even better once they start recovering larger amounts of it (he said himself that they need to recover at least 99% of it).
Also, from my (limited) knowledge of tritium it isn't that dangerous. It is in watches, weapon sights, emergency signs, and many other glow in the dark things. I've researched it before due to having several tritium containing products and it seems that in small amounts it simply vents into the atmosphere and becomes a non issue.
Anyway please correct me if I'm wrong on anything.
You'd need to consider gain, loss and on/off cycle.