Hacker News Reader

Comment by walrus01 | Hacker News Reader

For reference from the UAV industry. The very best lithium ion batteries are about 250Wh/kg but limited in the instantaneous amperage they can put out. High amperage capable batteries are more like 170Wh/kg.

UAV compressed hydrogen tank systems with fuel cells in the 800-1600W range, the full system with tank, fuel cell and a small buffer battery works out to between 800-1200Wh/kg depending on size. Some guys in south Korea recently hovered a large hexacopter for 10.5 hours. The systems I've seen are typically configured for an output voltage equivalent to the float voltage of a 12S lipo.

Diesel fuel, jet a, avgas and other fuels are a whole magnitude more dense in Wh/kg stored than the above-mentioned hydrogen system.

View on HN

Aside from the (obvious) energy density downside of hydrogen as a fuel, there is another substantial problem.

Hydrogen is challenging/expensive to store/contain. Kind of the direct result of it being the smallest atom we have. Even more important: when it reacts uncontrolled, in particular with oxygen, it does so at a truly frightening speed. Once it goes, good luck doing anything about it.

The fact that this is almost never mentioned in every yet-another-hydrogen-is-going-to-be-our-future-next-year piece, always makes me wonder if that's the result of not-so-honest entrepreneurs trying to lure investors, or that none of these jokers ever talked to any experienced chemist.

This is no attempt to spread FUD regarding hydrogen. It sincerely worries (if not frightens) me, how so many initiatives appear to completely ignore that any volume of hydrogen that is sufficient to power something that can transport humans or cargo, is pretty much inherently a catastrophic disaster in the making. That is, if such a system would ever fail in a way that makes the stored hydrogen react uncontrolled.

Every widely utilized technology is bound to fails, even if only in edge cases, and of course: extra safety margins/features can be added to limit the effects of failures. But with hydrogen that might be a hell of a lot more expensive and/or difficult than most would realize.

So, the way I see this is that we either end up very cost ineffective solutions, or ones that will be far more dangerous than many would ever realize. That is, until something goes terribly wrong, and everyone realizes they have been ignorant and maybe even intentionally fooled by people trying to make money. Reminds me a bit of self-driving cars.

If your fuel tank catches on fire you have similar problems, so I don't think these problems with hydrogen are insurmountable.

Gasoline is pretty well-behaved until it's vaporized.

That goes for lots of things including powdered sugar and flour. But that vaporization step usually happens in a relatively controlled environment which makes it manageable.

Isn't that their point? That gasoline is a pretty well-behaved and easy to work with liquid while hydrogen isn't?

Kerosene burns, hydrogen explodes. And kerosene doesn't actively damage the tank it's stored in, like hydrogen does.[1] Also, while its energy density per kilogram compares favourably to kerosene, its energy density per litre does not.

Hydrogen is a great rocket fuel because its specific impulse is second to none, so why isn't it universally used? Because even by rocket fuel standards it's a bugger to work with.

There are lots of things that react poorly to being set on fire. That doesn't make them all equally dangerous.

[1] https://en.wikipedia.org/wiki/Hydrogen_embrittlement

Hydrogen embrittlement depends on the type of tank. It does not occur with every material. Look, I'm not saying hydrogen is great and has no problems. I'm saying the problems are not in theory insurmountable. I have elsewhere made the same point as you with regards to mass vs volume. I'm aware of the drawbacks of hydrogen and am actually pessimistic on its use in aviation, but I think it is worth trying out.

Hydrogen is not only in contact with the tank but with other materials en route to the fuel cell as well. And the problems don't need to be insurmountable in theory, they need to actually be solvable and that is a different thing entirely. I'm fine with people trying it out but I'm very skeptical about this ever making it to production. I think this is more of a PR piece than anything substantial.

> Hydrogen is not only in contact with the tank but with other materials en route to the fuel cell as well.

Isn't it also in contact with everything around the tank on account on there being basically nothing which is completely impermeable to hydrogen?

I don't think hydrogen powered planes will substantially replace normal planes for reasons having to do partially with safety and partially with performance, but I think the engineering problems are surmountable. There are existing prototypes that have flown using hydrogen fuel cells, and the Tu-155 which ran on liquid hydrogen (used as fuel for its jet engines though, not to power fuel cells). So it's not like this stuff doesn't work.

I think you are overestimating the difficulty of dealing with embrittlement. Many materials don't suffer from embrittlement at all, and others (like aluminum) only suffer at high temperatures.

Edit: I also wanted to add to the person who made the claim "kerosene burns, hydrogen explodes" that kerosone can explode too, and it obviously is possible to handle hydrogen in a way that explosion is unlikely or it wouldn't be used as rocket fuel (where it is literally combusted with oxygen... combusted, not exploded). I agree that hydrogen is more dangerous than kerosene, but it's not like kerosene is that safe either and the question is whether you can create good enough processes around the storage and handling of hydrogen to offset the risk.

It's the huge difference in reaction speed that makes all the difference. Even in their gaseous/vaporized forms, conventional (fossil) fuels react comparatively slowly compared to hydrogen, like a sweet retriever puppy compared to a trained dog of war.

It's exactly this difference that makes them relatively safe to use and even still relatively controllable once something goes wrong. With hydrogen, you simply won't get that change. It will be all over in a blink. A very hot and bright one, I might add.

Feel free to believe whatever you want, you have every right. But I'm rather sure that your optimism regarding hydrogen is more rooted in ignorance than in knowledge.

Yeah, I'm actually more familiar with hydrogen than you think. I'm aware of the problems with reaction speed. One advantage hydrogen has is that it escapes very rapidly from its container and goes pretty much straight up due to how light it is.

I'm actually not as much of an optimist on hydrogen as you seem to think. I think it is unlikely to be used in aviation in any real sense, for a variety of reasons, some having to do with safety, others with performance. However, I think it is worth investigating, and I do not think the problems around it are insurmountable from an engineering perspective.

Fair enough.

Thought then I'm sincerely even more surprised you would say that any fuel tank on fire is a (relatively) equal problem.

As for hydrogen rapidly escaping and going straight up because of its low density, those properties I would only consider positive in scenarios where the hydrogen isn't reacting (being on fire, to be more precise). When on fire, quite the opposite. Buildup of heat on the ground might still be less of a problem than with heavier fuels (which indeed is a very real problem with those), but not much advantage of that once everything within a certain radius is reduced to the size of matchsticks by a powerful explosion.

But if you claim to be (some sort of) an expert, I'll believe you. I certainly am not, that much I admit.

I only based my opinion on what I've been told by chemists that I personally trust. They told me rather invariably that the engineering challenges for storing/operating hydrogen are such a bitch, that they are ultimately only a good for when you want something closely resembling a bomb, without actually being one (or just hoping it won't explode before it served its purpose). Like e.g. rocket engines.

I didn't say it would be an equal problem, I said that you will have similar problems if an airplane fuel tank catches fire. The problems you're raising are definitely real problems, I just think they can be designed around.

The engineering problems obviously aren't large enough to prevent something like the Tu-155 from using liquid hydrogen. Whether it can ultimately made as safe as conventional aircraft, I don't know. Maybe not. Either way, I don't think the economics are favorable for hydrogen.

> Feel free to believe whatever you want, you have every right. But I'm rather sure that your optimism regarding hydrogen is more rooted in ignorance than in knowledge.

You have made some great points about hydrogen. But personal insults like that don't help you get your message across.

It seems like you are not aware of existing production of hydrogen cars:

https://en.m.wikipedia.org/wiki/Toyota_Mirai

That particular model? No, I was not aware of it. But that production hydrogen cars exist, yes I did.

However, proof of their existence isn't proof of their safety. I have no doubt that they are (mostly) safe, even in almost every imaginable accident. However, it's the (unforeseen) edge cases that worry me. Not because of how often they might happen, but because of how bad the consequences could be when things go wrong.

It's a real bummer when whole a housing block gets wiped of the map, because one eccentric/rich guy just had to do something "for the environment" (#sarcasm).

From what I read, this car has storage capacity of 5 kg of pressurized hydrogen. I've seen similar storage tanks, and I have been amazed about how resistant they are to puncturing and tearing forces. Unlikely that any direct impact collusion would destroy one of these. But then there is always some corner case that it wasn't designed for. I certainly would not like to be even remotely around it, when 5 kg of pressurized hydrogen goes up in flames.

As hopefully every engineer knows, catastrophic failure is almost always a cascade of smaller compounding failures. Disrupt that chain-reaction and a catastrophe is usually averted.

However, with hydrogen the opportunity/window for that is often extremely short, with extreme consequences as a result. Not quite what you might call an optimal combination of factors.

Technically we could drive cars on nuclear reactors.

Substantially more than 5kg of hydrogen went up in flames with the Hindenburg, but I don’t think anything was wiped off the map there. Do you have any reference to the actual danger of stored hydrogen?

It's the tank pressures that make stored hydrogen almost indistinguishable from a bomb. 350 to 700 bar is not uncommon.

For comparison 7 bar is about the same as 100 psi for Americans.

One of the things that really worries me about the concept of mass market hydrogen cars. I'm talking in the millions of units. Ordinary consumers do all sorts of slightly dangerous, ignorant things with their vehicles. Including neglecting maintenance.

I am not sure I want to see the average untrained person handling a precision refueling apparatus that connects to a tank at 250 to 700 bar pressure. It's a lot more complicated and technical of an interconnect to refuel compared to just sticking a petrol station nozzle into a tank and pumping.

I don't think a system that requires a professional refueler staff person at every hydrogen fueling station is a good idea either.

That is what worries me too. Not so much the current state of affairs, in which maybe several thousand to tens of thousands units driving around. But as the production scale goes up, so do the chances of hitting those unforeseen (or willfully accepted as "cheaper" to mitigate with financial compensations, because that is how the industry functions) corner cases where things will go wrong.

Actually liquid H2 has energy density of 39 kWh/kg, where common CH fuels are around 12-13 kWh/kg. So about 3x more energy per kg then conventional fuels.

Note that H₂ boils around 20-30K (depending on pressure). Once you take into account the weight of the insulated tanks, liquid H₂ is going to have much worse energy density.

Liquid storage I believe typically gets up to about 20wt%, so you're talking around 7.8kWh/kg. Compressed storage is around 6–10wt%, so up to 3.9kWh/kg.

That's without the weight of the storage and potentially fuel cell surely.

Yes, but that's without the container, which is very heavy for the hydrogen, while a plastic bag is already good enough for fossil fuels.

Hydrogen becomes significantly more efficient as you scale the tank. Don’t forget many rockets use Hydrogen over Jet-A and they really care about weight.

Rockets use liquid hydrogen, which is a pain to work with, but not too bad if you can fuel up and launch relatively quickly before too much boils off. The real reason rockets use hydrogen is because Hydrogen and Oxygen gives the highest specific impulse of any chemical rocket. You really don't care about specific impulse in a jet aircraft, and you really don't want the fuel boiling off before the end of your 12-hour flight.

Also, if you're thinking of pressurized gaseous hydrogen instead of liquid hydrogen, then the tanks get really heavy because high pressure tanks don't scale well (unlike low pressure liquid tanks which do). Hoop stress is proportional to diameter, which means tank wall thickness needs to grow with diameter too. https://www.engineersedge.com/material_science/hoop-stress.h...

Boiling off is a function of surface area which means it scales really well with volume. Rockets have issues because they don’t want to bleed off hydrogen on a launch pad or use significant insulation, but aircraft can just feet that to the APU or engines in flight. You will want to drain hydrogen for maintenance etc, but that’s not a significant issue.

Some rockets, and generally on lower-thrust high efficiency upper stages. Hydrogen has a low energy density in terms of volume. It requires larger tanks and so is rarely used to get off the pad.

The rockets that use it in the first stage all get most of their initial thrust from solid strap on boosters too.

The Delta IV heavy uses hydrogen/oxygen exclusively, without any solid boosters:

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

It is a very expensive vehicle though.

RP1 is effectively a purer version of Jet-A, and is a very common propellant. I'd say it's more popular than hydrogen, but I'm not sure I can back that statement up. It's what fuels the Falcon 9 lower and upper stages.

> UAV compressed hydrogen tank systems with fuel cells in the 800-1600W range

They didn't say how they would be using the hydrogen. I haven't been able to find it again, I did read about an article about a long endurance (days), high flying plane. It was powered by hydrogen. It burnt it. As I recall it was in an ICE, but at 50% efficiency using it as jet engine fuel isn't such a bad idea. It has an energy density of 120 MJ/kg, 3 times as much as normal fuels so it makes sense either way, if you can contain it.

In other news, the country where I live (Australia, 6th largest exporter in 2019) is contemplating exporting hydrogen. Why? Because if renewable power continues on its current price trend it will be cheaper to produce hydrogen than it is to mine the natural gas we currently export. Again, I presume no fancy technology - it will be burnt just like the natural gas is.

What about liquid hydrogen, which is much denser than compressed one? Not an option for UAV?

If you had a hydrogen tank in a plane, would you be better off using it as a fuel in a jet engine or using it to power motors with a fuel cell?

“Hydrogen needs a bit more work” for use in gas turbines, but it is something Rolls-Royce is examining, along with electric.

Both are being researched. The answer to that isn’t clear-cut.

Could you kindly provide a link for the hovering hexacopter please? Thanks.

I assume they're referring to https://www.youtube.com/watch?v=_U_dHZZ3vxQ&feature=emb_titl...

That one and also Doosan.

https://www.youtube.com/c/DoosanMobilityInnovation

There is more than one south Korean company working on combined packages of lightweight hydrogen tanks and fuel cell power systems. And several American companies.

Reply on news.ycombinator.com