Hacker News Reader

Comment by topaz0 | Hacker News Reader

"No one knows why this device works" --> "There are still some small gaps in our understanding of the details of this device's performance"

View on HN

In this case, it's different than lift. We can run CFD simulations of wings and get accurate estimations of lift/drag etc. We cannot do that for Hall thrusters without already having run the thruster, as we need to calibrate the anomalous transport parameters to make the simulation match experiment. This makes the simulations much less useful in design than they are in other fields, which is a real drawback.

Another point to make is that testing these devices is expensive. They are low thrust propulsion systems that are designed for thousands of hours of continuous operation. They need to be run in very high vacuum, and for higher power devices there are only a handful of vacuum chambers in the world sufficient for the task. Additionally, Xenon is expensive (dramatically moreso since the Russian invasion of Ukraine), so developing better predictive simulations is critical.

Thanks for the context. My comment was not intended as an indictment of the post. It's an interesting problem and the missing physics is clearly not trivial. All I mean to say here is that I think saying "no one knows" trivializes all the knowledge that people (like you!) do have.

It is, per the article, a large enough gap that designs from scratch can't predict performance and that simulation isn't always fully reliable. I don't know that the author is looking for a metaphysical answer to all of existence here. It would be nice to understand the functioning of something that moves billions of dollars of material around well enough that design is less dominated by trial and error.

To be clear, I'm not saying the gaps are not interesting or economically important, and I am not belittling the work that the tweeter does. I'm willing to entertain the possibility that my rephrase is too extreme. But I'd still argue "No one knows why X works" is an unhelpful frame that could give the casual reader the wrong idea.

The "small gap" here is an order of magnitude difference between theoretical and practical performance.

No it's not, the gap is an order of magnitude between a pen and paper calculation, or a fudge factor in a more sophisticated simulation.

I'd say we understand it about equally well as the weather.

I don't think the weather is a good analogy, at least if you're just referring to the unpredictability that comes from chaos. That the weather is unpredictable (when projecting out a few multiples of the relevant Lyapunov time) is an unavoidable consequence of chaos. Indeed, if you could hypothetically run a full atomic-level quantum simulations, the wavefunction would contain different branches where different weather events happened; there is literally quantum uncertainty in the outcome, even if we had a complete theoretical description and understanding of the different outcomes.

In contrast, the situation with a Hall effect thrust is that we cannot explain a repeatable effect. And that could change with better theoretical understanding. It's in that sense that Hall effect thrusters are "not understood", which I think is a reasonable thing to call it.

(And if you are talking about some aspect of the weather that, unlike the chaotic aspects, is empirically repeatable but not predictable from our models, then I think it would be fair to characterize that as not understood too!)

Pretty much this. Also, we still do not 100% understand aircraft lift and Bernoulli's principle.

Or the physics of lithium ions, which is why sometimes a new battery design will just spontaneously catch fire (because the time it takes it to denature into something dangerous exceeds the time the designers spent on practical testing).

What are you talking about?? We completely understand aircraft lift. Bernoulli's principal is a nearly irrelevant footnote that gives a small boost to efficiency.

Aircraft lift is caused by the upward angle of the wings forcing air downwards as the jets move the plane horizontally forwards. This "angle of attack" is usually very small so it's hard to perceive with the naked eye when looking at a commercial aircraft.

That's not totally correct. See here:

https://www.grc.nasa.gov/www/k-12/airplane/wrong2.html

It says it in the article. Lift is created by turning a fluid. Just as a spoon in a faucet, it’s all how you “throw” the fluid in the direction opposite the direction of lift. This is angle of attack.

The year is 2312, and we still have to combat the Bernoulli-airfoil misinformation. Even though all you have to do is stick your hand out of the window of a self-driving drone and twist it around a bit.

This sounds like an XKCD comic!

https://xkcd.com/803/

I find it hilarious that in the future everyone is more capable than the present at inventing things and also simultaneously these same people are incapable children who can’t use anything manual. It’s concerning that people dream that way.

>> do not 100% understand aircraft lift

Do you mean turbulence? There are some blank spots there but we have very good equations to approximate the behavior, equations good enough for practical aviation.

From what I understand, the challenge is the following: Even though we know precisely how to calculate the lift generated by a wing, all attempts to reduce those equations and their predictions to "ELI5" arguments leads to incorrect explanations.

Basically: We can explain aircraft lift to a college student, but not to a 5 year old. Or, we can explain aircraft lift using equations, but not using informal arguments.

Air pushed down. Plane go up. More air pushed down, more plane go up. Most everything else is about preventing and controling turbulence, keeping the air moving smoothly around stuff.

Exactly. I see more and more of this clickbait-style title applied to scientific articles. I instinctively no longer read them. The fact that this is ALSO a twitter thread is just icing on the cake.

It's about one step away from "We thought scientists knew how rockets worked. You won't believe what we uncovered next"

For what it's worth, the rest of the thread is a good read.

Reply on news.ycombinator.com