Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Germany launches world's first hydrogen-powered train (theguardian.com)
207 points by ksourav on Sept 18, 2018 | hide | past | favorite | 115 comments


In related news, Bombardier is deploying a battery powered train in Berlin: https://montrealgazette.com/business/local-business/bombardi...

Most of the rail system in Germany is electrified anyway. With batteries to power the non electrified bits, why bother with hydrogen?


Overhead power lines require maintenance too, and getting 100% coverage often involves disproportionate spend on short sections that are expensive to cover and maintain. Getting rid of masts on those sections - or even whole lines - could be a massive cost saving.

Additionally, Germany is a big train tech exporter, and there's lots of unelectrified route miles in Europe and the world, many of which would be totally uneconomical to electrify.


The batteries are supposed to eliminate the need for 100% coverage, though, isn't it?


you don't want to remove 100% since you want to charge the batteries somewhere. Also it maybe more economical to keep 1km of power lines after the station when the train is accelerating - main power consumption.


That's a great idea. You could just put the overhead charging infrastructure at a few stations on the line. Just enough to top the batteries up. Most of the energy could probably be recovered through regenerative braking, but of course some will be lost through mechanical and air resistance.

This would be so much simpler than electrifying the whole line and much easier to maintain.

This infrastructure is often already at key stations, and so it might require no extra work to implement them. Just more money for the battery powered trains.


Gotta put little spinny generator fans and solar panels on the train too. Perpetual motion!

edit: also I don't think your idea will 100% work because it would take some time to charge the trains at the stations.


> Additionally, Germany is a big train tech exporter

They are, but the article mentions that these particular ones are built by the French TGV-maker Alstom.


They are built by their german division Alstom Transport Deutschland (former Linke-Hofmann-Busch) in Salzgitter and Alstom is about to merge with Siemens Mobility.


Could they not convert to third rail?


Because fuel cells provide range equivalent to diesel (600km, says the article), while these battery-powered trains can only go 40km or so between charges.


How much better is the energy density for fuel cells?


There's 2 measures of energy density. Per kg, and per liter (weight and volume).

Hydrogen scores really well on the per kg measure, and pretty bad on the per liter one.

The values are: Hydrogen 142 MJ/kg and 9.17 MJ/l.

For comparison: Fuel 46.4 MJ/kg and 34.2 MJ/l, Lithium polymer batteries 1.8 MJ/kg and 4.32 MJ/l, which is why Tesla batteries are a LOT heavier than they look.

So to answer your question: a hydrogen fuel tank (pressurized at 700 bar, also known as "you need a 10cm thick steel plate to contain this, and may God have mercy if you so much as dent it") is about 3x lighter but 3x bigger than an equivalent fuel tank, and 70 times lighter and half as big as an equivalent battery.

In case you're wondering, hydrogen in fuel is "compressed" to about 3000 bar, but in a way that's stable, not prone to sudden temperature drops and doesn't blow up if the octane number is high enough.


This is why gas/oil fuels most cars and will continue to do so unless there is a significant technological breakthrough in hydrogen fuel storage safety or battery efficiency/size.

People don't realize that a tesla sedan weighs more than an average sedan. And I laugh when people say you can keep extra tesla batteries and swap them to extend range. They think it's like changing smartphone or regular car batteries. They don't realize that tesla batteries weigh 1300 lbs. Good luck with that.

Once there is a real breakthrough by either the hydrogen fuel side or the battery side, society will steadily and gradually switch to one or the other and leave gas fuel behind. But who knows if/when such a breakthrough will happen. Either way, we are going to be using gas fuel for a few decades at the very least.


In a train, you could just swap the battery car. Much like swapping a coal car in the past. Not convenient but not terrible either.

Unlike swapping batteries in cars, the infrastructure exists already.


You could even do it with a slip coach while moving.


I don't think there's any need for a breakthrough in order to have world wide massively adopted electric cars.

The only bottleneck right now is battery production, not even demand.

The technology is here, the demand is here, the supply chain is being adapted so that price (not weight) drops.

I don't see how the weight of a car has any meaningful impact on adoption. Speaking of which, adoption is usually an S curve and I fail to see how electric car's adoption would be any different. That makes it everything but "steady" and it will take pretty much everyone by surprise, like any other technology before (computers, internet, laptops, smartphones... and petrol cars). That raises many interesting questions about the readiness of each country's infrastructure but that's another debate.


> I don't think there's any need for a breakthrough in order to have world wide massively adopted electric cars.

It's not a matter of opinion. candiodari posted the metrics just a few comments up. Gas/Hydrogen has 4 to 8X better power to weight ratios.

> The only bottleneck right now is battery production, not even demand.

No. It's also demand. Most people want big gas guzzling trucks and SUVs.

> That makes it everything but "steady" and it will take pretty much everyone by surprise, like any other technology before (computers, internet, laptops, smartphones... and petrol cars)

S curve applies to something "revolutionary" or better. Electric cars aren't better than gas cars. Regular cars are cheaper, get more mileage and are much more efficient (and powerful) than electric cars. The only thing going for electric cars is they pollute less.

Smartphones did something that telephones couldn't. Computers did things typewriters couldn't. Cars did things that horse carriages couldn't. That's not the case between gas powered cars and electric cars.


About that infrastructure: it isn’t only about charging infrastructure. If most cars become electric cars that weigh twice as much as existing cars and accelerate much faster, wear and tear on road surfaces and bridges will go up (or won’t it? Will accelerating more smoothly offset that?)

Certainly, in the USA, a country that already has problems maintaining its road infrastructure, that could become problematic.


I think hydrogen is here to stay. Japan is going full throttle on hydrogen, phasing out gas to a great extent in the next generation or so. The technology is viable and working already, just needs state sponsoring and determination to make it happen.

https://www.japanfs.org/sp/en/news/archives/news_id036048.ht...



100km with the updated batteries they are planning apparently. I imagine, you could stuff quite a bit more battery in a train at the cost of passenger space.


And you could add overhead pantographs or third rails at key stations to charge the battery whilst it is stopped, and for a mile or so after if it is deemed necessary.

A battery means that you don't need to build miles of infrastructure. You can just add a small amount.


Until you show up to the next station and its charging infrastructure is broken.

A train traveling more than 100km in a single trip is extremely common. 1000km, the range of the current hydrogen prototypes, is much more feasible.


The grandparent said that most of the rail network in Germany is already electrified anyway. So no need for diesel or hydrogen trains. He/She states that at most a battery would be needed to cover the non-electrified parts.


The intercity network is fully electrified, but there are plenty of long regional stretches without overhead power. The need exists.

These trains are being considered for use on similar regional stretches currently served by diesel-electric trains in the Netherlands as well.


And it's not only the poor and forgotten areas that aren't electrified.

- The link between Hamburg and the island of Sylt isn't (one of the most profitable ones)

- The ICE-link between Hamburg and Puttgarden (to Copenhagen) isn't (one of the most busiest around 110%++ capacity)

- Large parts of Swabia, one of the most industrious areas of Germany are completely un-electrified apart from the ICE-link between Munich and Stuttgart. Getting from Aalen to Lake Constance or Kempten via Ulm you will see a lot of "Mittelstand"-Companies but not a single electrified track.


Also the shunting tracks at the train stations aren't electrified most of the time.


According to [1] less than 60% of the rail system is electrified in Germany.

[1] https://de.m.wikipedia.org/wiki/Chronik_der_Streckenelektrif...


By rail kilometres, yes, but not by usage. Virtually every line experience heavy traffic (cargo, RE) is electrified, but there are many lines covering branches of the network where no heavy traffic occurs, which is not electrified.

That's what this hydrogen train is for; it's base (LINT) is one of the most common trains used on those "backroad lines".

(Coincidentally, the lint is also one of the most uncomfortable trains to ride if you are larger than a twelve year old boy)


All main lines are electrified, many small towns are connected by diesel currently.


The article I linked uses the same statistic. Basically this train is intended for local trains on such tracks. I imagine that with some power just at the stations, it could recharge there as well.


Because the hydrogen trains are planned to run in regions with lots of wind energy — whenever there is too much electricity on the grid, just produce hydrogen from the excess energy, and use it later in trains.

It's a much better solution for handling the spikes in wind than most of the alternatives.


Large scale hydrogen electrolysers are capital expensive, so there is no such thing as to only use them when electricity is cheap.


Is that an option for offshore wind farms? Or would seawater be too expensive to "purify" before splitting into hydrogen/oxygen?


Split the water onshore, building those things offshore wouldn't make much sense when all you need is a cable to bring the electricity on shore to larger conversion installations


I'd be interested in seeing an analysis of batteries vs hydrogen when it comes to environmental impact. Does anyone know which one wins? I suspect hydrogen due to the environmental impact of mining / producing the battery materials, then disposing / recycling them later.


Almost 100% of commercially sold hydrogen is made as a byproduct of producing fossil fuels though. Yes, batteries take a lot of materials, but you can recycle them over and over again.


They can store excess wind energy as hydrogen and transport and store it with existing natural gas infra.


What happens with all the water that evaporates? This process will bring massive problems.


This is obviously great new tech, though one shouldn't see this as a solution to global warming. Until we find a cheaper scale-able way of producing hydrogen that doesn't come from hydrocarbons, this isn't really a solution. Might as well burn the the hydrocarbons directly...


Producing hydrogen when there is excess power from renewable source could be a great addition to the mix of "how to store energy when we produce more than we need, and use it when we need it".


One way it's used in France is pumping water from below hydroelectric dams to above, so the water can be reused by the dam during the day to produce electricity.


That's great, where it's feasible. Which isn't enough places, as you need abundant water, a higher reservoir, and a lower reservoir.


I don’t understand why this isn’t the answer to the problem of excess renewable energy at the times we don’t need it?


For simple grid storage the efficiency is just bad. The electricity-hydrogen-electricity round-trip efficiency is about 60%, pumped hydro or batteries can do better.

So it's only worth it if you have other uses for the hydrogen or need longer-term storage, e.g. gas reservoirs.

Plus any infrastructure you build will have to be overprovisioned because by definition it'll be designed to absorb peaks and run idle the rest of the time, which makes it even less profitable.


Mostly because of the cost associated with turning electricity into hydrogen. Which I hope will be coming down in future.


Because of security. Hydrogen is pretty dangerous, much more dangerous than anything else.


Is security the reason that hydrogen vehicles haven’t taken off yet, or is it a question of infrastructure?


Well, security is fixable. It's hydrogen after all and not a heavy gas; in case of leaks it goes up not down. leaks are still dangerous as it explodes when being mixed with oxygen. the flames are invisible, you only feel being burned but you don't see it.

but the biggest problem is psychological: The Hindenburg, still


Just compare hydrogen cars with electric cars and you'll see why people prefer electric. CA has 33 hydrogen refueling stations, so many people in CA could get by with one.


Is it that much more dangerous than say LPG though?


You can say the exact same thing about electric cars.

Hydrogen is not a primary energy source, it is an energy storage and distribution tech, just like batteries.


Not if the hydrogen comes from hydrocarbons, which right now is all hydrogen. In this case, it's just an inefficient proxy for fossil fuels...


hydrogen comes from hydrocarbons, which right now is all hydrogen

It's not! the Scandinavians are investing in "green" hydrogen, you should look at the photos of the giant electrolysers in this PDF: http://hydrogenvalley.dk/wp-content/uploads/2017/09/FCB-CPH1...


Ok, replace the energy source with a solar panel. The only thing this is doing is acting like a battery.


Exactly! If we can improve the methods of generating hydrogen, fuel cells have the potential to act as versatile, phenomenally efficient, dirt-cheap batteries, without all the issues of limited metals and dirty production - Being a better battery is a huge deal.


The reason that fuel cells are so expensive is because of limited metals. The best fuel cells are made with a lot of platinum. If fuel cells were cheap and did not need the use of limited metals, then they would be a huge deal. Until then, how you produce the hydrogen is a minor detail. (Maybe there have been amazing advancements in fuel cell tech since I last looked into it. Any links?)


Batteries don't do that great with the limited metals problem either.

The bigger issue for railroads is range; current battery operated trains have a range of about 40km, whereas this hydrogen train has a range of 1000km. You could have charging stations at every station, but then you run into issues if the distance between stations is more than 40km, or if the charging equipment at a station becomes broken.


The problem is that any non-fossil fuel derived hydrogen made from solar/wind loses more than 60% during conversion. Compare that to lithium with around 10% loss that includes outputting the energy (hydrogen would lose even more).

Hydrogen is a terrible battery with very little chance to catch up to current battery tech.


I looked for a "source of hydrogen" diagram/paragraph in the linked pdf but it's nowhere mentioned. They just call it "green" hydrogen, so I am left with the belief that it's sourced from water electrolysis, is that so?


... and for those wondering why: it's so inefficient to produce hydrogen with electricity that, even if you have to pay a carbon tax, it's currently almost always produced using natural gas.

That's why it's stuck in the demonstration phase, essentially.


Yes, except you need to produce about 3x more energy for the hydrogen powered car to get the same range as the battery powered ev.


If the price of grid energy is negative, it pays to produce hyrdogen from it.


Hydrocarbons too: longterm storage, with the sun as primary energy source.

Not a moot point, since depleting them has consequences.


Could you please expand on this? My understanding was that hydrogen was produced through electrolosys. What process does one use to go from a hydrocarbon to H2? And does that process produce CO2?


Unfortunately, the electrolysis of water is not efficient enough to produce enough H2 on an industrial scale at the moment. Most H2 is produced by steam reforming of gas: https://en.wikipedia.org/wiki/Hydrogen_production https://en.wikipedia.org/wiki/Steam_reforming


The first link says that the efficiency is 80%. Seems like that’s enough to scale, especially if electricy prices go negative during peak hours of renewable energy production. So I’m still trying to understand what the roadblocks to this are.


Historically it was the most expensive approach thanks to the high cost of electricity and higher initial capital. Here is a pretty easy to skim paper on the cost breakdown from 1980: https://pubs.acs.org/doi/pdfplus/10.1021/bk-1980-0116.ch001

The approach is quite competitive technologically and the efficiency is there but the problem is electric cost. Thankfully renewables are bringing that cost down so some areas with very cheap electric could support such manufacturing.


It's still a lot cleaner than diesel!


Why? That hydrogen most likely is created with fossil fuels, you don't avoid emissions. Actually, you will create more because the energy density of hydrogen is lower than the density of diesel, mostly due to the heavy storage tank. In other words, you need more energy per mileage to move the trains because the weight of the trains is heavier which leads to more emissions.


I don't like these in-the-grand-scheme-of-things-everything-is-the-same. There is a difference between emissions from a power plant and emission from diesel cars in densly populated cities. Just step outside.

In the grand-scheme-of-things the universe will cool out either way.


I think a lot of people are missing the point. A higher demand for hydrogen will incentivise companies to improve the production methods for hydrogen, hopefully with a more conscious effort to keep it green, i.e. Improving electrolysis methods.

Nobody is claiming that all hydrogen is greener than diesel, the point is hydrogen has the potential to be vastly greener than diesel, and the best way to explore that in a capitalist society is to increase the demand for it.

A small disclaimer, I know nothing about hydrogen.


With the steady uptake of solar panels on private homes and wind energy, it is not too far of a stretch to envision hydrogen production whenever there is more energy returned to the grid than needed.


Upvoted for honesty


There are locomotives in the US powered by liquefied natural gas. The Florida East Coast railroad's 24 locomotives are all LNG. BNSF is looking into it. It's cost driven; natural gas is cheap right now.


LNG was also considered for the French TGVs and the American rail system in the '60s, but then the oil crisis happened.

Fossil fuel prices are extremely volatile, so alternate power sources are actually more financially secure for railroads.


The interesting thing about electricity is that it's a great decoupling interface to the power source: you can change the power source and basically leave the remainder of the device/infrastructure as is (not that it's not challenging per se to move say from coal to nuclear to renewable, but hey, if TGV were LNG-based we'd be in a tough bind to move away from it)


The Soviet Tupolev Tu-155 from 1988 also ran on LNG: https://en.wikipedia.org/wiki/Tupolev_Tu-155


as a totally lambda citizen : natural gaz is a co2 emitter, so it's a no. The fact it's cheap is not enough anymore.


> "a totally lambda citizen"

I've never encountered this use of "lambda" as an adjective; care to explain?


I imagine she/he's is a native french speaker. In french a "lambda" object/person is a non-specific, average one.


ah, it's a bad influence from my mother tongue. Lambda means "anybody", but that's just, well, not english :-) I wonder how somebody says it in english.


You might want "layperson", meaning non-expert. Is that want you intended to convey?


I would say the common word is "average".


I've seen at least two comments so far saying that this is only a stopgap in the face of electrification. I do not think this is true, because electrification has very high upfront capital costs, which would not be justified on lesser-used lines.

To back up this statement, I can use some numbers from the Caltrain (California, US) electrification project, where the existing commuter-heavy Peninsula corridor is being converted from diesel-electric to 25 kV overhead electric.

Caltrain quotes the cost of electrifying the line at USD 697 million. See http://www.caltrain.com/Assets/Caltrain+Modernization+Progra...

Caltrain quotes the length of electrified track as "51+ miles", which I'm going to treat as "51 miles". See http://www.caltrain.com/Assets/Caltrain+Modernization+Progra... pages 4 and 7. The length runs from San Francisco's station at 4th & King streets, down to San Jose Diridon station.

Those two numbers above give us a cost of USD 13 2/3 million dollars per mile. But, that's not fully accurate.

Most of Caltrain's right-of-way is double-track. There are three areas with four tracks (for fast trains to pass slow trains), one area with three tracks (near the Millbrae station), and three areas with more than four tracks (one is Caltrain's yard and shops, the second is the many platforms at San Jose, and the third is the many platforms & yard at San Francisco). See an old Caltrain timetable at https://web.archive.org/web/20110513180702/http://acm.jhu.ed...

Let's say that San Francisco is about one mile of eleven tracks (so, 11 track-miles), each of the three passing points is two miles of four tracks (so, 24 track-miles), Millbrae has a mile of three tracks (3 track-miles), and San Jose has around one mile of twelve tracks (so, 12 track-miles). That leaves around 46 miles left of the 51 original, which would be double-track, giving us 92 track-miles. So, that's 142 track-miles, giving us a cost of USD 4.91 million per track-mile.

Of course, costs may certainly be higher in the US, where electrification (especially for mainline heavy rail) does not happen much (if at all), but still, electrification of an existing line is not cheap. For Caltrain, it makes sense, because this is a heavily-trafficed line in an area with many polluting vehicles (automobiles) that also wants to have a 'green' sensibility, as well as wanting the reduced noise profile.

When you consider freight, there are other issues. For example, in the US, heavy freight traffic lines use gondolas that can support two containers, stacked one on top of the other. That means wires would have to be higher than they would otherwise, and I wonder how how well those extra-extended pantographs would work at higher speed.


There are also "partly electrified" possibility e.g. only 10% of track electrified (stations for example). Trains can recharge batteries or supercapacitators there from wire. There already are such existing light rail lines.

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


Sydney is doing this for a new light rail line to eliminate overhead wires from the CBD.


Electric has some added advantages over fossil fuels; in general, it's cheaper to run off electricity than diesel. Electrics also have better maintenance requirements and better performance characteristics; if you can squeeze the margins and increase utilization of vehicles, you can actually wind up on top. https://www.greaterauckland.org.nz/2013/10/24/electric-train...


The problem left is that it's likely still somewhat (or much?) inferior to a lithium battery EV - or if there a good comparison? Energy efficiency is definitely much reduced.


Energy efficiency is a red herring. Theey are using surplus hydrogen for this project, i.e. they are getting useful work out of a waste product.


Gasoline was originally a waste product.

If hydrogen takes a life of its own that may not be a great thing.


French Alstom rolled out a hydrogen-powerded train in Germany.


Considering how closely interconnected the EU tech industry is — half the train will be German parts anyway — I don't think it makes sense to try to attribute it to a specific EU country.


Counterpositioning Germany and France never lead to anything good, on a global scale.


The Coradia LINT platform this train is based on is built in Germany. Alstom has two locations within Lower Saxony (which is the state funding a lot of this).


French Alstom and German Siemens recently announced that they'll merge their rail units.


I hope the news get to other countries where fuel trains are numerous (I think India network relies on that)


how can it be zero emission?


If they use excess electricity during peak renewable energy hours to produce the hydrogen it would be zero emissions. That’s an exciting prospect.


From the article:

"Hydrogen trains are equipped with fuel cells that produce electricity through a combination of hydrogen and oxygen, a process that leaves steam and water as the only emissions."


But why tho?


I wondered why they didn't just electrify the line but then I saw the picture and it is so much more beautiful than overhead lines.

But I guess the question is, can you create a third rail that's ~100% safe? Walking along tracks is fun, and I think an important part of rural society.

Then I'd be interested in the cost benefit.


> But I guess the question is, can you create a third rail that's ~100% safe?

Regular trains in Germany use 25 kV. Third rail => third system, but what for?

> Walking along tracks is fun, and I think an important part of rural society.

Walking around in the trackbed is not a recommendable activity in Germany.


There are third rail systems that only switch the power on when a train is in range:

https://en.wikipedia.org/wiki/Ground-level_power_supply


Third rail is incompatible to the existing system in the area, not suitable for high voltages and creates issues at railroad crossings. Moreover, even if unused sections are switched off, there will be some minimum section size, which means there will still be a risk.


RIP Stan Meyer


I'm not sure you're meant to launch a train — launching is for rockets and cream pies.


This seems like a very niche solution, as surely in the longer term 100% of railways will be electrified, like in Switzerland.


Stringing wires does not make sense for 100% of rail lines due to low volumes of trains -- does anyone make batteries for such occasions, or do you expect some rural lines to be abandoned?


SEPTA in Philadelphia actually abandoned many miles of commuter rail because it built a tunnel that could not handle diesel trains.


Why not throw a few Tesla power walls on cargo carriage, call it a battery car and swap it for a charged one at stations along the way as necessary..

You just need rails to be able to quickly attach/detach a car, that's much less infrastructure than full electrification.


> You just need rails to be able to quickly attach/detach a car

A single attaching maneuver takes 5-10 minutes (from personal experience). Add to that the detaching and driving the carriage off the rail to free it up, and you are quickly at 15-20 minutes, which is unfeasible for most long-distance train rides. Might work with short distances where you already have such a long pause planed in at the ends, but otherwise not so much. It also puts more strain/traffic on the track switch infrastructure at stations, which is already a bottleneck AFAIK.


Actually, on many RE routes the trains are nowadays getting fully automatically coupled/decoupled at certain places.

e.g. the RE7 Kiel-Hamburg and the RE7 Flensburg-Hamburg both arrive in Neumünster at the same time and are coupled together before continuing to Hamburg.

It takes roughly a minute to do the coupling due to modern automated coupling solutions.

On certain lines the ICE1 and ICE2 also did the same, and the IC2 is running the same rolling stock as nah.sh is using on the RE7 line, so that could also use the same technique for coupling.

But this all is very expensive and complicated, and in no way worth it just for coupling a wagon with batteries to the train


Switching a standard container might be much faster, but you'd need more infrastructure than - crane could be on the rail car though.


This problem is solved.

https://www.cargobeamer.eu/




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: