Hacker News Reader

"There's an inherent physics limitation when trying to strap a whole human body to the side of something small (and generally depicted as being strapped on your back, Bobba Fett style). The center of mass is off."

While obviously wasteful and goofy, there's always the option of putting an upwards-pointing (i.e. pushing down) thruster at the rear of a jetpack like a lever arm.

I was thinking that eventually they could design some kind of powered attachment between the arm pods and the backpack. It would accentuate your desired movement while relieving stress. Maybe they could control it with a Neuralink implant so that there won't be accidental movements due to buffeting. Then if you had that you could build an AI to make it mostly user-error-proof.

To be clear, this is all pie in the sky nonsense, but it's fun to think about. It would still be terribly impractical, but maybe it could be made reliable enough that using it over land would not be immediately fatal to most ordinary people who would want to try it.

“ this is all pie in the sky nonsense”

Nah, only the Neuralink part.

I imagine there will be a whole batch of grifters coming post Neurolink…

Similar to just add it to blockchain.

How about a neurolink to the blockchain? Well I guess we have Doge.

I don't work for neuralink nor do I have any economic stake in their success. Please don't make unjustified accusations of grifting.

> Please don't post insinuations about astroturfing, shilling, bots, brigading, foreign agents and the like.

From the guidelines.

I didn’t accuse you of grifting.

I’m making a prediction of what future grifts will involve.

The stability issues could easily be fixed with more closed-loop control. The much more serious issues are heat and the limited fuel supply.

Agree. This is the same reason why I'm skeptical on flying cars. There's so many fender benders. I don't want a Jetta crashing through my roof.

Sure, general aviation is safer than driving (e.g. less accidents per vehicle, and less accidents per hour/mile), but general aviation is more dangerous than commercial aviation. I suspect there's something going on with maintenance, training, and available safety features. Seems like lowering the barrier of entry to flying is simply going to increase the accident rate because there's just more less skilled people flying.

Of course most flights crash on takeoff and landing, and conceivably that would be automated, so maybe I'm just being paranoid. Still seems impractical, if for no other reason space restrictions on the ground.

Commercial aviation gets to pick from the cream of the crop and fully professionalizes it’s pilots - it’s all they do, all the time. That means staying current (keeping habits fresh, remembering key details, etc) happen easier.

Considering how low the frequency is for your your typical GA pilot, it’s honestly a wonder it works as well as it does.

Flying is not simple and for everyone - it is complex, but many things are.. and when mastered it is actually not that hard - so disagree, no wonder here.

Flying is easy, flying safely is very difficult. GA inherently allows a great many things that significantly increase risks. Going solo inherently makes you a single point of failure so for example many significant but survivable health issue basically guarantee your death.

And that’s just the start. Pilots run out of fuel, fall asleep, they land on the wrong runway/field, and so forth across lots of very small risks that end up killing people.

I am not saying GA should be more regulated, it’s just important to understand the trade offs in terms of freedom vs safety.

We can also get away with lower GA regulation because there’s relatively few aircraft. If you scaled 1346x to equal the number of automobiles, we’d probably have even more than 1346x accidents, simply because of a self selected pilot skill bias. We’d have to have more regulation then.

Even just 1346x as many accidents is ~1/2 million deaths per year which is unlikely to be considered even remotely acceptable.

My understanding was that most GA accidents occur in the 50 to 350 flight hours of pilots. By this, perhaps the issue isn't of handpicking the best of the best, but more that they, by definition, survived the early filter.

Getting a basic (entry level) commercial pilot’s license requires 1500 hrs.

What I mean by the ‘cream of the crop’ is you have to be more experienced and pass far greater hurdles (including a far more demanding medical exam) than any GA pilots need to, just to be licensed to try to get hired as a commercial pilot.

It doesn’t mean there aren’t great GA pilots out there - there are! It just means that only the top x percent of GA pilots could pass the requirements to be a commercial pilot.

> Getting a basic (entry level) commercial pilot’s license requires 1500 hrs.

That's not correct. For a US commercial pilot's licence, the hours requirement is 250. It's 1500 hours for an airline transport pilot's licence.

Thank you much! Got my wires crossed apparently.

Applicable FAA Regs btw [https://www.ecfr.gov/current/title-14/chapter-I/subchapter-D...]

GA is actually less safe than driving. It's about as safe as riding a motorcycle: https://inspire.eaa.org/2017/05/11/how-safe-is-it/

GA aircraft crash every day in the US. They don't make the news outside of aviation circles because only 1-2 people are affected. Often the occupants survive.

I'm not a huge fan of those statistics because "deaths per 100k hours" isn't really an appropriate way to compare driving vs. motorcycles vs. GA. If you're comparing driving vs. GA, I think one possible appropriate measure is "deaths per mile"; let's say the average speed of a GA craft is 150mph and the average speed of a car is 37mph, that means that GA is 6 times more deadly than driving a similar distance.

Even that is a flawed measure, however. Few people use GA as a replacement for driving. I think in general, an appropriate measure to use is to estimate the (negative or positive) expected DALYs associated with an activity over a lifetime. That said, this measure does (in my opinion) penalize moderately-unhealthy things worse than very risky activities. For example, BASE jumping has a fatality rate of 0.04% per jump. If you do a handful of crazy BASE jumps, that's only a slight DALY difference - very roughly comparable to not brushing or flossing as far as I can tell from this study [0].

[0] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4484374/

I would expect drone like tech so a computer will keep whatever it is easy to steer, drive, and land, and/or just self drive.

That tech doesn't handle mechanical or electrical failures very well. The mishap rate for drones is far higher than for manned aircraft.

And drones don't scale linearly. Small unmanned "toy" drones can get away with RPM control of their props to maintain level flight. RPM control is easy with brushless DC motors and it's mechanically simple. But drones big enough to carry people can't do that (in general) because you cannot change RPM on a big prop nearly as quickly as you can on a small prop, because of inertia. So big props often need collective pitch just like helicopters. Collective pitch is far more mechanically complicated than RPM control and it thus presents more avenues for failure.

The good news is that multirotor drones (even big ones) usually don't need cyclic pitch which is even more complicated than collective pitch, so that makes them less complicated than helicopters, but big drones are still not as mechanically simple as toy drones.

Can't you build an array of smaller (say brushless DC motor) propellers to achieve the same thing? I could have sworn I saw a guy on Youtube do that and hover around his backyard in a home-made frankendrone comprised of a ton of props

You can, and hobbyists have done exactly that. But such aircraft are extremely inefficient and only have enough battery power for a few minutes.

It's kind of like that guy who tied a bunch of little helium balloons to his lawn chair and went flying. It sort of works but it's pointless and dangerous.

Yes, you can certainly do this. But it won't be as efficient because in general large propellers are more efficient than small ones. So your batteries won't last as long.

A closed-loop control system won't fix the stability issues if the pilot's arms are pointed in the wrong direction.

Nonsense, Galvani solved that problem in the 18th century.

Nonsense. It doesn't matter how good your control system is if you can't generate the necessary thrust vector.

It was a joke. Galvani's experiments were on contracting frog muscles with electric currents. The joke being that the jetpack will take control of your muscles.

I wouldn’t be so sure. Fundamentally this is a control input/feedback/positioning/power problem - one we don’t currently know how to model or control in an automated fashion.

With the right mental model and control/power systems, it wouldn’t have to be much different from riding a unicycle. Helicopters have similar issues (it’s a bit like balancing a spinning bowling ball on the end of a stick in practice - yet many people are licensed and safe helicopter pilots)

I’m guessing it’s something that not everyone can do, but something a large percentage of people (20%+?) can do if they spent the time learning and it was systematized usefully.

Minor nitpick. A fixed wing airplane is generally known to be stable. But many other aircraft like multirotor are inherently unstable and require a fast PID loop to stay in the air.

> An aircraft is stable. Let go of the controls and it will keep flying.

That's true for most fixed-wing aircraft, but it's certainly not true for most rotor-wing aircraft. I'd estimate you've got about a second, if you're lucky, between letting go of the cyclic in a Robinson R22 and unrecoverable loss-of-control.

Even with loss of power for rotary wing aircraft don't you have the option to auto rotate to the ground?

You have to have enough forward speed. Auto rotation is a joke in a real emergency most of the time.

The problem with Robinson helicopters specifically (older R-22 & R-44 more than most) is their very low mass rotor system, which results in very low inertia. The time you have to recover is significantly shorter than other high inertia systems on other helicopters.

Their very low cost makes them pretty common for schools and instruction in general, but the low inertia (and resultant fatalities early in its operation) was significant enough that the FAA requires a Robinson specific endorsement.

> Ultimately, 'jetpacks' will probably never exist

So no Aeroball and Harlem Heroes any time soon :(

https://talkingcomicssite.wordpress.com/2018/04/27/the-compl...

> since you have jet turbines strapped to your arms

So soldiers that use them would need guns strapped to their feet?

> It's inherently unstable and tiring since you have jet turbines strapped to your arms.

If you look at the photos in the linked article, you'll see that the turbines are on the backpack. It still is tiring for your arms per the article, but for different reasons.

If battery tech continues to improve at current rates then certainly electric jetpacks will be possible and practical, if there's still demand (and we don't destroy ourselves one way or another). Without doing the math though it might be many decades.

Unlikely. The specific energy of jet fuel is more than 40 times that of batteries, and I don't see batteries improving by an order of magnitude. A jetpack would need 40 times more battery mass than fuel to produce the same thrust for the same duration. Of course, the additional weight of the batteries necessitate even more thrust...

Nailed it.

Even in cars where there they do a bunch of gliding (rolling) batteries are just becoming plausible (looking back over time).

Existing aluminum-air batteries have 1300 Wh/kg, or 4.7 MJ/kg in modern units, and it's plausible they could improve to over 20 MJ/kg. The more expensive lithium-air battery already exceeds 6 MJ/kg. Jet fuel is 43 MJ/kg. Pure aluminum is 31 MJ/kg, so to the extent that you can build "new Al-air batteries" during flight with the cathode and electrolyte from "old batteries", you can approach that limit. That sounds ridiculous but we've been doing it with hydrogen-oxygen fuel cells since 01838. Hydrogen has 118 MJ/kg, and there's no fundamental physical reason hydrogen-air fuel cells can't be epsilon above that, though current models are large and heavy.

It's true that all rechargeable batteries like Li-ion batteries are so far under 1 MJ/kg, which more than 40 times worse than jet fuel just as you say, but it's extremely plausible that Al-air could beat Li-ion by an order of magnitude, and Li-air already does, except for the best Li-ion cells. It's already most of the way there. And fuel cells can beat jet fuel by almost a factor of 3, and for low-power applications they do.

Betavoltaic cells and TEGs have even higher energy density but you can't turn them off, so you need to dissipate the power they produce when you're not using it to fly or whatever. You could build a hybrid system with a large glide ratio that uses chemical batteries for takeoff and a betavoltaic battery or TEG for long-distance cruising, but you might have to only land it on water to keep it from overheating while it was on the ground.

https://en.wikipedia.org/wiki/Aluminium%E2%80%93air_battery

https://en.wikipedia.org/wiki/Energy_density#In_chemical_rea...

https://en.wikipedia.org/wiki/Energy_density_Extended_Refere...

As you say, practicable batteries in production have specific energy under 300 Wh/kg (1 MJ/kg), 40x worse than fuel.

The other technologies you mention don't seem to be practical for now. I don't think it is reasonable to expect any breakthrough making it to market in the next decade. With specific energy doubling every 10+ yrs, electric flight might be sensible for niche applications (training flights, occasional urban transport for example to the airport) in a few years, but for anything else it'll be several decades realistically.

Hydrogen-air fuel cells and aluminum-air batteries don't need any breakthroughs to work; they're well-understood, proven technology. They're just more expensive than kerosene and, in the case of hydrogen, more voluminous.

To put it differently, the reason nobody is flying across the country today on aluminum-air battery power is not that the existing batteries don't have the range; it's that they can fly across the country more cheaply on a combustion-powered jet.

(I think that will continue to be the case even when the combustion fuel is synthesized with electrical energy.)

> If battery tech continues to improve at current rates

What do you consider current rates? They improve slowly, maybe 3-7% a year. Last I looked, batteries are so much worse in terms of specific energy than fossile fuels that we're looking at several decades before electric flight becomes practicable (except for very short distances and training flights).

7% a year means doubling every decade, 3% means doubling every two. Fossil fuel efficiency isn't moving at nearly that rate.

If current rates continue, battery powered jetpacks should be compelling within 60-120 years