Hydrogen-powered train makes UK maiden journey | Hacker News Reader

Hydrogen-powered train makes UK maiden journey

bbc.co.uk

76 pointsaries198067 comments

I'm very skeptical whether it makes any sense to use hydrogen on a train.

We know since many decades how to run trains on electricity. It's an established technology. It seems hydrogen trains are a cheap excuse for places where infrastructure is lacking and train tracks haven't been electrified. Except it's not cheap, because physics dictates using hydrogen will always be much less efficient compared to direct electricity use, as it includes an additional conversion step.

I think developing hydrogen infrastructure is important for sectors where no other alternatives are available. For sectors were established technologies for electrifiation already exist - not so much.

There's plenty of train lines in the world that see very few trains, maybe only one an hour, where the business case for electrification is much weaker. Many of these lines are already questionable from a business case point of view which can deter anything that would increase infrastructure costs, and if hydrogen trains can approach the operational cost of electric trains then there's a clear business case for them.

> if hydrogen trains can approach the operational cost of electric trains

That's exceptionally unlikely. You go from "use electricity to run a train" to "use electricity to run an electrolyseur, fill hydrogen into hydrogen tanks on the train, use hydrogen to power a fuel cell to generate electricity, use that electricity to run a train".

You need to add "build electric lines over entire length railway" to the left side of the equation and "build a hydrogen locomotive" to the right side.

Hydrogen is easier to store than electricity, and you can produce it at times where electricity price is zero or negative (very sunny days at photovoltaic plants, very windy days at wind turbine plants, 3am at nuclear plants). Processes more efficient than electrolysis do exist.

I suppose safe storage and transportation of hydrogen is a major cost, though. Hydrogen gets absorbed into metals, because their crystalline lattices act like sponges for molecules this small. Worse yet, hydrogen can seep through things that work as impenetrable solids for other gases. And hydrogen flame is all but invisible (most radiation is infrared, nearly no visible light).

Yep. The trains in the article (Cuxhaven-Bremerhaven) go once per hour. That entire region of Germany, and the area of the Netherlands that borders it, has an enormous amount of trains running on a 1x/hourly schedule. Groningen-Leeuwarden, Groningen-Delfzijl etc etc. All running (diesel-electric) Stadler FLIRT trains, which have taken the world by storm.

The political decisions in these less populated regions are similar to small town politics in two ways: the budgets are tiny (aka no money for electrification) and a fairly small group can push things through if it fits their agenda. For those reasons I think these hydrogen trains could totally become popular for less densely populated areas like this.

Just like just about every line in the Netherlands, there’s a train from Groningen to Leeuwarden and Delfzijl every 30 minutes.

The fixed cost and time consumption of electrification is quite high.

I'm not sure about hydrogen specifically, but just as steam trains had a coal+water "tender" behind the engine, it seems not unreasonable to have a "battery" car on a train. Whether that's a tank of H2 or just a boxcar full of 18650 cells (anyone care to do the maths on that?)

I tried doing the math a few years ago. I did another search and up popped this.

http://www.brooklynrail.net/science_of_railway_locomotion.ht...

Which has the comment: At the same speed, same load (GCW) and on level ground, any steel wheeled railway vehicle is 24.6 times more energy efficient than any large rubber tire road vehicle, regardless of the type of power source.

My hand wavy argument would be, to go 300 miles a passenger car needs 33% (guess) of it's weight to be batteries. But a truck is probably 2-2.5 times more efficient. So would only need 33% / 2.25 (split the middle between 2 and 2.5) or 15%.

If a 4000 ton train is 25 times more effecient than a truck as started above then. 15%/25 is 0.6%.

So the battery for a 4000 ton train would weigh 4000 * 0.6% -> 24 tons.

Seems to pencil out to me.

Note that the majority of hydrogen trains currently under development are all multiple units, and don't have dedicated locomotives. In the UK at least, track access charges push things towards distributed traction in part because of the axle loadings of any dedicated locomotive (or battery car). You see this to varying degrees across countries where speed is relevant.

To take as an example: each British Class 802 motor car has a 1550 litre fuel tank. If we (slightly optimistically) assume the diesel engine is 50% efficient, and diesel fuel to have 10.8kWh/L energy density, then that's 8370kWh of energy produced. The Tesla 3 battery pack manages 160Wh/kg, so to carry 8370kWh the battery pack would be 52 tonnes. That's more than each car currently weighs in total. And a nine-car train has five such fuel tanks and engines.

Hydrogen powered multiple units seems like an operational pain in the butt. If you have to fuel each and every one of them on a regular basis.

Seems simpler to put a 120kwh battery on each one and use a common charging buss.

Overhead lines are an eyesore and third rails are dangerous. I wouldn't consider electrification the be all and all for railways. It comes with tradeoffs like anything else.

I’ve never understood the aesthetic argument given the railways themselves — and indeed major roads — are something I find extremely ugly.

You never understood why people think they’re ugly given that you think they’re ugly as well? What?

I’ve never understood why electrification makes it worse when it’s ugly without, and when the alternatives are also ugly.

The point is electrification via gantries makes it much more ugly.

A railway line disappears into the ground somewhat, and to some people is actually attractive, even with emissions from a combustion engine.

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

Electric gantries are seen to many people are far more intrusive to the landscape, and even destructive to landmarks, since they stick up into the air.

By way of comparison, the Royal Border Bridge, with relatively lightweight and specifically designed gantries: https://upload.wikimedia.org/wikipedia/commons/9/93/60163_To...

I've literally never heard this argument before. It's baffling.

I have heard "wind turbines are ugly", which I extremely disagree with (they're huge impressive things), but at least it's sort of justifiable because you might see them. Who the hell is staring at railway tracks?

And who could possibly care about aesthetics when we're talking about something so important? Clean energy is not about which color you paint the bikeshed, it's not about your personal preferences.

When sacrifices have to be made, aesthetics probably is going to be one of the first things to go. But in this case we're talking about a technology that is an improvement over diesel while still maintaining the aesthetic and safety of the countryside. Just because you consider it a lower priority doesn't mean it shouldn't factor into the equation at all.

I certainly agree that asthetics should come after function here, particularly because we need to try to slow environmental degradation. On the subject of roads and rail being ugly, I think it may be a degree of normalisation - if you looked at the general opinion of rail and road at the time of invention, you might well find the same reservation about their asthetics as for wind turbines. (see https://www.mtholyoke.edu/courses/rschwart/ind_rev/rs/denaul... and also probably for windmills).

More recently the opposition to HS2 in the UK has produced many arguments about asthetics, mainly due to route location.

TBH I'd prefer to see smartrail systems rather than a rehash of an existing transportation system. Something like: http://openprtspecs.blogspot.com/2016/

I never found railways or roads ugly. But find them extremely polluting for the environment in loud sound(trains are terrible at this), toxic chemicals (volatile asphalt compounds, wheel dust, fumes and specially petrol oils in the case of trains), lights and so on.

It is also very bad for the environment because it divides the territory in two and animals and humans could not cross easily like they did in the past for millions of years. You can not cross or it is extremely inconvenient crossing whole cities because of trains.

On the other hand, trains make transportation way cheaper and territories to prosper.

I also find electrical posts ugly and dangerous stuff.

Electric trains are pretty quiet. Even Japan's bullet trains can sneak up on you.

Well... that makes sense given that you already find them ugly. But then why would you like to make them even more ugly?

Passenger service is usually high-usage enough that the cost of electrification is worthwhile. But many freight-only lines still use diesel, even in areas where a majority of the rail net has been electrified, because little-used supply tracks lack an incentive to electrify them. If we can replace those diesel lines with hydrogen, that's a win.

Also note that the motors themselves are electric, so when this train is running on an electrified section, it could still use the overhead lines. Not sure if modern diesel engines offer that same option (i.e. if the diesel is purely used as a generator for an otherwise electric engine).

Yes, all the diesel train engines are diesel-electric, and on some routes they use bi-mode (eg electric London to near Bristol, diesel-electric to Cardiff) trains.

Something like 60% of the UK rail network is unelectrified and recent electrification projects have been ruinously expensive, we need something to fill in the gaps and hydrogen definitely has potential.

> Something like 60% of the UK rail network is unelectrified and recent electrification projects have been ruinously expensive, we need something to fill in the gaps and hydrogen definitely has potential.

It's not as bad as is often made out; with relatively few exceptions (GWML most prominently) most recent electrification programmes in the UK haven't been too much above European norms.

In general, the reasons why costs have been higher in the UK are:

* other countries having fairly constant ongoing electrification work leading to availability of skilled staff (versus massively spikes of electrification, then nothing for a decade or more),

* many more structures (e.g., overbridges, tunnels) close to the limits of the loading gauge and hence more structures work needed

* track access being more difficult in the UK (in general prolonged closures are harder to get approval for), and

* operators being compensated for service disruption.

At this point the UK really just needs an ongoing programme of electrification, as we now have high-quality modern off-the-shelf parts for most needs approved and in use on new projects, and we have designs that meet the requirements of almost all current routes.

Also worth bearing in mind is the recently announced Scottish Government plan to decarbonise rail in Scotland by 2030, which includes electrifying all the major lines in Scotland. (A few won't be, and hydrogen trains have been mooted for them, but these are mostly lines with a few trains per day where the business case for electrification is always going to be weakest, especially combined with them often being in the most challenging terrain.)

Have there been any experiements (on a long distance route) to have battery packs on trains that are fast-charged at stations? In the UK a lot of routes are at least partially electrified, and EV chargers have proved it is fairly easy to supply a few hundred kilowatt electric supply to urban areas.

I know there are trams that work like this, but I'm wondering if the tech could be scaled up to trains.

https://en.m.wikipedia.org/wiki/Acumulador_de_Carga_R%C3%A1p...

And it's a good way to justify ever increasing train prices in UK.

> because physics dictates using hydrogen will always be much less efficient compared to direct electricity use

You have to look at the whole system (and life-cycle) and not just compare single trains. I assume people with enough insight and expertise did the math and they don't just try this tech out for fun.

EDIT: Also looking at the video, the giant transformators you usually have in a Lok aren't there. That weight difference already could make hydrogen competitive.

How do they store the hydrogen? Just big pressurised tanks? Also be interesting to see how they refuel the train, because I imagine that small hydrogen molecules are difficult to confine, while the gas is so diffuse that it must take ages.

Edit: I found a better link which has more technical details:

https://www.railway-technology.com/projects/hydroflex-hydrog...

A total of 20kg of hydrogen is stored in four high-pressure hydrogen fuel tanks whose pressure is regulated and maintained at 8.5 bar by the pressure drop regulator.

The fuel cell unit is powered by hydrogen from the tanks, while oxygen is sourced from ambient air. The fuel cell converts the mixture and generates pure water and electricity up to 100kW. Electricity generated by the fuel cell will be sent to the lithium-ion battery pack.

20 kg of hydrogen is about 2400 Mj. The Class 319 has 1 MW of motors. So the hydrogen fuel will run the train for 40 minutes at max power. Maybe 3 or 4 hours in typical use?

20 kg may not seem like a lot, but hydrogen has three times the energy content of diesel.

It sounds as if the train has a battery pack as well, so it will recover breaking energy. Trains are uniquely suitable to energy recovery since they have mostly predictable acceleration and deceleration curves.

Keep in mind that trains expend a lot of their output power only during takeoff. Maintaining a constant velocity requires relatively little power, and with regenerative braking some energy may be recovered (the primary reason for the Li-Ion cells between the power source and the motor).

I don't have any numbers on this, high-power electronics was never my field. But I wouldn't at all be surprised if cruising power was less than 10% of max engine power.

Cruising power can be almost zero. Trains run on shallow gradients, and most of the power is used to climb them.

When the gradient is negative trains freewheel down the slope, and the power source is only used to keep the lights, heating, and electronics running.

>Trains run on shallow gradients

This is genius. But could anyone confirms this. I try to do some google search and nothing useful or related came up.

Does that mean Trains are constantly running downhill? Because that doesn't seems to be the case long distance rails.

It means an ascent or descent of less than 1.5% is normal for railways, and there's so little friction a train can coast downhill without power.

They obviously need power to maintain speed going uphill.

Not unusual for trains to be at 50-75% throttle even at cruise. No getting around v^2

A lot of this is just due them being relatively underpowered since acceleration isn't usually a priority. For instance the BR Class 319 mentioned up thread has 1300hp to move over 120 tons - and the aerodynamics of a barn.

Note that hydrogen-powered trains have been in regular passenger service in Germany for a couple of years now: https://www.theguardian.com/environment/2018/sep/17/germany-...

A bunch of the questions I saw asked about hydrogen on the UK rail network was the comparatively large size of the hydrogen tanks needed, and whether packaging of tanks of sufficient size was going to practical.

Please note these trains have a dirty secret: They run with hydrogen made from fossil gas.

The hydrogen source might change pretty soon, so. There are a ton of initiatives right now aiming at green hydrogen, all over Europe, Northern Africa and the Middle East.

Can be made with excess wind energy.

Not really. Who will invest in a facility, that runs occasionally on excess wind energy and stands rest of the time still?.. I want my investment working 24/7 with max 5 minutes maintenance each week.

they aren't. They're investing in something that creates power whenever there is wind, and hydrogen whenever there's excess power.

How often is there excess power? To be economic there will likely be purpose built to run on wind whenever the wind blows, not just when the wind blows AND there is excess power. The capital cost of electrolyzers is substantial and will need to run a lot to pencil out.

It happens quite a lot, actually. Search the net for negative power prices and you'll understand the mechanics (for example, [1]).

So if a supplier can use that excess power to generate another product which can be sold, that's a win-win for them: they won't have to pay the penalty for oversupplying the grid, and they can sell the generated hydrogen for additional profit.

[1] https://www.cleanenergywire.org/factsheets/why-power-prices-...

"The number of hours with negative power prices in Germany increased by around 50 percent to 146 hours in 2017, which translates into 1.6 percent of total time"

So if you're only going to run it when power costs are negative, you're amortizing the capital costs over running the plant for the equivalent of six days a year in 2017 Germany.

This is changing rapidly in many countries. House and large farm PV is getting super popular, more wind generators are being built, etc. Wind grows 10+% every year in Germany and solar ~5%.

I agree it's going to change. Overprovisioning of renewables is going to be necessary if we're going to rely on them alone, and long term I am excited about hydrogen.

But parent's link didn't really back up their claims, and concerns about capital efficiency for infrequently used plants are valid.

This is in dutch, but it is a very clear exposition on why that doesn't really work out: https://www.wattisduurzaam.nl/16238/energie-opslaan/watersto...

I will bite. This doesn't rule out why it cannot work out.

There are some strong assumptions behind this article, it chooses the following scenario: Wind + Solar + H2 for seasonal storage. And then proceeds to dismantle it. But nobody said this has to be the scenario.

I think it's far more likely something like: Wind + Solar + hopefully clean base load (e.g. Nuclear) + Hydro + Hydro storage + Lithium storage + H2 storage for transport (planes, trains, ships) + H2 storage for seasonal purposes

Fact of the matter is, there is no clean alternative to H2 (at the moment) for large transport vehicles (i.e. cargo shipping, planes, trucks) so unless you are willing to keep using carbon based fuels for these purposes you are stuck with H2.

Biodiesel is a nearly drop-in replacement for No. 2 diesel. It's not carbon-free in current form, but neither is hydrogen, and there are viable carbon-neutral biofuel cycles, albeit ones that are substantially more land-intensive than modal contemporary practices.

Solar being preferred energy source from what I heard. All you need after that is some access to ports and the hydrogen can be shipped wherever you want. Nice idea actually.

That's all great, except that it simply doesn't exist at the moment. Nearly all of our hydrogen comes out of oil refineries as a byproduct. If you're going to use electricity to make hydrogen you might as well use it to make synthetic fuel and skip the whole hydrogen nonsense.

Even if everyone bought electric cars tomorrow refineries would still be there operating at about 50% of capacity. Isn't it better to capture waste hydrogen?

> it simply doesn't exist at the moment

Sure it does:

https://www.bbc.com/future/article/20190327-the-tiny-islands...

https://edition.cnn.com/2020/03/13/business/orkney-hydrogen-...

What straightforward process do we have for making synthetic fuels with electricity?

My understanding is that synthetic fuels tend to be some form of hydrocarbon, which means they need a source of carbon. Either you make the synthetic fuel at the exhaust stack of a fossil fuel plant (so you still have to burn fossil fuels) or you spend huge amounts of energy collecting CO2 from the air (is that even a proven process yet?)

Hydrogen has a straightforward and well understood process to make it from water. What's the hydrocarbon equivalent?

Is there a better way to make synthetic fuels?

Isn't this the exact same argument people made against electric vehicles - for the majority of Tesla's existence the reality of owning an electric vehicle has meant burning fossil fuels to generate electricity to charge the car.

The point of running off hydrogen or electric is so that you can figure out the problem of generating the source at scale.

Apparantly, the biggest users of hydrogen is, among others, the chemical industry. It could also be a potential solution to power stuff like container ships. But we'll see what comes out of all these initiatives.

Seemingly on the same day as:

https://www.independent.co.uk/travel/news-and-advice/bus-ele...

1st October was a good leap forward for transport in the UK!

Unlike battery you carry only 1/2 the weight of the fuel as you burn them whilst battery is dead weight even drained. Hence airline considered that more.

Just wonder why hydrogen car is so under the radar. All E except a few advertisements or waiting.

Battery-powered trains are also being developed and likely a more cost effective solution. https://www.railtech.com/rolling-stock/2020/07/09/over-400-b...

>wire-free range of 50 kilometres

Yep, there's your problem...

Look, I love lithium batteries as much as anyone, but the range really is a huge problem. If you take a look at the real world places that the existing diesel locomotives serve (on non-electrified lines) you have a lot of places where you're basically using one or two trainsets to go back and forth a rural line. They usually don't sit in stations for more than a minute or two, maybe 5-10 mins at each end. If each one is out of service to charge for 30 minutes once an hour, you're either demolishing the schedule or adding significantly more trainsets ($$$$).

Many lines connect rural to less rural places. As the train will get closer to the less rural place, it will start running under the wire. At the rural end of the line there can be a dedicated charging station for layovers. This way, there are plenty of opportunities to charge the trains.

In Germany, a very high percentage of Diesel Services could be replaced by trains having about 50km of autonomy - and for the rest it would be possible to fill the gaps using only a relatively small amount of electrification.

That's simply not how the lines are usually built, unfortunately [0]. In general, electrification was done for entire rail lines. If you look at maps of non-electrified rail, it's a relatively small amount where you have a sufficiently long electrified section near a station to charge. And usually, at the end of the line, you don't have enough time to get a good charge in.

Certain regions of Germany are an exception, partly due to their early electrification of the rail, then subsequent damage in WW2, then strange rail layouts, but if you look at non-electrified rail maps of the Netherlands, France, the UK etc it will be extremely clear that battery trains wouldn't work well.

Part of the reasoning is that you want these massive batteries to last, so you're going to want to charge at the fastest around 1C, aka 1 hour for a full charge. Turnaround time at the end stations being usually around 5 minutes for these services, your choice is either to absolutely destroy your schedule, fork up far more $$$$ for more material or ignore battery rail.

Look, if they made economic sense, we'd already have electric locomotives - the technology is extremely simple, just toss a bunch of batteries into a box; the engines themselves are already electric. Similar batteries to electric buses, too. But while the economics of electric buses do make sense - and you see electric buses in most of the world's cities today - trains running on batteries really don't make sense in most cases. That's why you don't really see them. Out of the nearly 2000 Stadler FLIRTs that are running across Europe, zero of them run entirely on batteries today - and there will only be a projected 55 in 2023 (!) in one region of Germany.

[0] https://en.m.wikipedia.org/wiki/List_of_railway_lines_in_the...

Typo: meant to say battery-driven instead of electric there in that last paragraph.

I'm extremely skeptical of hydrogen trains. First, we all know that hydrogen cars are never going to take off because of the massive success of batteries. But as production of batteries ramps up, prices will fall and power densities will increase in a virtuous cycle. Adoption of battery cars (already widespread) drives the adoption of cargo vans, people movers and pickup trucks (a multitude of models shipping next year), but it doesn't stop there. BYD, Tesla, Daimler and others are working on battery-powered semis and buses, and once you're there, you're basically in train territory.

Think about a train - one expensive, custom, bespoke, finicky engine and 100 dumb cars. Why not make mass-produce 100 cars with batteries and motors built in, reusing the same cells or packs that you use in cars, van, picks, lorries, etc.

The fixed nature of train routes also means you can electrify parts of it strategically (ie, a few km around a station to help a train accelerate) and rely on a mix of battery and overhead lines.

At this point hydrogen still seems possibly usable for ships and long-distance planes - short-haul planes will be electrified in the coming decade due to the same mass production that will drive EV adoption.

Reply on news.ycombinator.com