[HN Gopher] Show HN: Luna Rail - treating night trains as a spat...
___________________________________________________________________
Show HN: Luna Rail - treating night trains as a spatial
optimization problem
Hi HN, I'm Anton, founder of Luna Rail. I've always thought night
trains are a fantastic, sustainable alternative to short-haul
flights, but they're often held back by a lack of privacy, comfort,
and poor economics due to low passenger capacity. I became overly
fascinated with this puzzle. I view it as a kind of night train
Tetris (my wife less charitably calls it "sardinology"). I spent
way too much time learning about and sketching various layouts,
trying to figure out how to fit the maximum number of private
cabins into a standard railcar, while making them attractive for
both day and night travel. This eventually led to a physical
workshop (in Berlin) and a hands-on rapid prototyping process.
We've built a series of full-scale mockups, starting with wood and
cardboard and progressing to high-fidelity versions with 3D-printed
and CNC-milled parts, with various functional elements. Hundreds
of people have come in to test our various iterations, because you
can't test ergonomics or comfort by looking at renderings (although
we did create a bunch of nice ones). The link goes to our home
page showing our approach and some of the thinking behind them.
It's been a lot of fun working on this puzzle, and we're excited to
share what we've come up with. We hope you think it's cool too and
would love to hear your thoughts.
Author : ant6n
Score : 105 points
Date : 2025-06-18 08:50 UTC (4 days ago)
(HTM) web link (luna-rail.com)
(TXT) w3m dump (luna-rail.com)
| rmccue wrote:
| What do you see as the benefits over Nightjet's Mini Cabins?
|
| Also, interesting to see just after the launch of
| https://noxmobility.com/ which is targeting the same market.
| ant6n wrote:
| OBB Mini cabins * night only * 40 pods
| per car * implemented
|
| Luna Rail * day and night * 65 pods or 40
| single rooms * in physical since 2023 * team
| focused on technical development and testing
|
| Nox * day and night * virtual concept
| * team very strong on marketing
| ant6n wrote:
| (Messed up my edit - I mean in physical development since
| 2023, not just renderings)
| graphememes wrote:
| the bed looks way more comfortable in this than luna rail
| fyrn_ wrote:
| Really cool concept. I wish we could have something like this in
| the US one day.
|
| Wanted to report a small typo, In the 3D model index menu,
| "Uppder" can be found. I assume this was supposed to read
| "Upper", as in "above".
|
| I hope to one day ride such a system when I visit europe, best of
| luck with your project.
| ant6n wrote:
| Thanks!
| ViscountPenguin wrote:
| What's up with the super high carbon intensity estimate of your
| trains? Is it because they're running at night and can't use
| solar? Does it include embedded carbon in the train? Or is
| Germany's grid just _that_ dirty?
| solardev wrote:
| Where do you see the carbon intensity? I just see a plane above
| a train with a single "250g" figure.
| dcan wrote:
| I see 15g CO2(eq)/km in the bottom left-hand corner
| ant6n wrote:
| 15g/km is very low. The number is dominated by energy mix and
| infrastructure. Deutsche Bahn publishes lower number, but they
| only look at operation (not infrastructure), and assume ,,green
| energy" rather than the energy mix of the country.
|
| The 15g is a good estimate, see discussion here: https://back-
| on-track.eu/de/klimawirkung-von-nachtzuegen-neu...
| solardev wrote:
| I love the concept and the renders, but I wonder... is the
| spatial optimization really what's holding train travel back? If
| you decrease passenger density in order to increase privacy and
| comfort, do you then have a corresponding increase in ticket
| prices?
|
| In places with good train travel, it seems like they already have
| several cabin classes, from sardine seats (still luxurious
| compared to air travel) to private cabins (at several multipliers
| of price). Pod style rooms would presumably be cheaper than that,
| but still a lot more expensive than a seat?
|
| Then in places without high speed passenger rail, like the US,
| this wouldn't really be able to address the major problems with
| train travel (slowness, lower priority than freight, low
| reliability, etc.).
|
| Under what scenarios would using pods instead of cabins be more
| economically viable? And could these be retrofitted into existing
| sleeper cars, or would they have to build entirely new trains?
| ant6n wrote:
| You yourself mention the trade off -- the ,,sardine class" is a
| six-people per cabin couchette (60 ppl per railcar). The
| private class is luxury (10-17 cabins per railcar).
|
| We got 65 private pods or close to 40 little single cabins - in
| a refurbished railcar. In a new car it would be more.
|
| We put together some explanation of the economics and the
| difference between old and new cars: https://luna-
| rail.com/approach
| bruce511 wrote:
| Clearly different markets have quite different requirements and
| comparisons to air travel.
|
| For example a "night train" maxes out around 12 hours. A train
| from 6pm to 6am is functionally equivalent to a 8pm flight,
| arriving at midnight, checking into a hotel, getting some sleep
| etc.
|
| How far you can go in that 12 hours (give or take) depends on
| the speed of the train etc. In Europe you can go to a lot of
| places in 12 hours. In the US not so much.
|
| Much longer and other factors come into play. You have to
| balance the time cost of "getting there" to the time benefit of
| "being there".
|
| But thats OK. This solution doesn't have to work _everywhere_.
| It can start where it works well and grows from there.
| clan wrote:
| That is like comparing range in gas cars and EVs. Some do
| that but there are other major benefits.
|
| The lengths I will go through to avoid air travel is much
| higher than a 1:1 ratio in comparable time. When I have to
| get the cattle treatment I prefer cattle cars over cattle
| cans.
|
| And even with 1:1 remember that layovers are a completely
| different beast. If Munich was a hub between Northern an
| southern Europe I would be happy to spend a well rested day
| before continuing on. Especially in spargel season!
|
| ...but only a fool does not fear German railroads. They could
| really learn from the Austrians.
|
| The reason night trains are not a thing is because there is
| no real network. Looking for tickets in Europe it is often
| once or twice a week departures on specific routes. No real
| good north south interconnected corridor from Scandinavia.
|
| And as a proper geek I have even sought them out but often
| found them sold out.
|
| They cost optimized themselves to obliteration.
| rocqua wrote:
| My experience on the Caledonian sleeper, in a 'room' was quite
| cramped in all senses (I am over 6 feet tall), and quite
| expensive aswell. If that had been optimized better, I would
| have enjoyed it more.
| Freak_NL wrote:
| These renders do not make me feel as if I am even considered
| as a desirable passenger at 200 cm.
|
| The new OBB wagons seem much more practical (and currently
| exist). A normal sleeper train wagon with stacked beds in
| compartments is fine for me. This origami concept looks
| claustrophobic, and the sleeping positions seem to allow for
| no room for the normal movements you make in your sleep, let
| alone getting out to take a piss or something.
| ant6n wrote:
| Well, you cannot assume ergonomics from renderings. That's
| why we run extensive testing. We tested on a large range of
| people. In the pods there's a fall off in perceived comfort
| around the 95th percentile. Even then, the feedback is
| overall quite good.
|
| In the larger pods, there's actually an uptick in
| evaluation for taller people. Testers were often surprised
| how well it works.
|
| All beds have at lest 2m, although there are different
| degrees of becoming smaller at the foot end -- just like in
| aviation business class (with ticket prices 1.5 orders of
| magnitudes higher).
| pimlottc wrote:
| As another tall person, that's good to hear, but the
| pictures still make me skeptical. It would help if there
| were some actual dimensions listed.
| mathis-l wrote:
| I'm 190cm and tested luna rail's prototypes. I was amazed
| how much space I had, even in the smallest cabin.
| Definitely much better compared to any night train
| experience I've ever had
| ant6n wrote:
| Hi Mathis-l, thanks for the shout out!
| matt-p wrote:
| Honestly they need far greater economies of scale to drop the
| price to where it needs to be. They should be competing with
| a daytime fare and giving you a bed in exchange for having
| the ability to run the train slow.
| ant6n wrote:
| > They should be competing with a daytime fare and giving
| you a bed...
|
| That's basically what we're doing, since the capacity
| approaches that of day trains, the ticket costs should be
| similar.
|
| > ...in exchange for having the ability to run the train
| slow.
|
| The railcars can go 200km/h. It's not super high speed, but
| pretty competitive in Germany at least.
| econ wrote:
| I've long joked that modules should be made easy to swap with a
| forklift. Trains are usually full of small defects that aren't
| serious enough to take them out of service.
|
| If they are comfortable you could rent out the cabins when not in
| use either fitted on the train or not. You could also retire the
| units there.
|
| You could make a platform only and make it easy for others to
| design modules in a broad price range. Maybe most modules should
| be in storage until booked.
|
| You could park the "hotel" module on the destination and put it
| back on the train for the return trip.
| ant6n wrote:
| I sometimes her proposals like that, and it sounds kind of
| attractive - you get into your pod and forget everything until
| you arrive.
|
| But in a sense, night trains are already like that. Since they
| can stop at multiple places, you can depart and arrive
| downtown. In the meantime you're in your cabin and forget
| everything.
|
| Entering the train ,,with" the pod instead of just yourself is
| gonna make boarding and alighting take forever, and the
| logistics of storing and moving the pods are a nightmare. It's
| going to reduce capacity by a lot because you cannot optimize
| the layout and every pod needs to be ,,insertable" as a whole.
| (3x reduction in capacity means 3x increase in ticket cost).
|
| Homologation is going to be a nightmare - in Europe,
| realistically, it's gonna take more than 10 years or develop
| something like that. You need a new infrastructure because
| right now stations are for people - that'll probably take 20
| years (in Europe).
| Freak_NL wrote:
| Besides, sleeper trains really are modular already, using
| existing standard rail infrastructure. Wagons are the natural
| unit of a train consist.
|
| Ideally, we end up with a big enough network of sleeper lines
| that it makes sense for OBB, European Sleeper, or whichever
| operator to have the luxury of taking some wagons out for
| maintenance and standby, and even scale up and down depending
| on the season.
| ant6n wrote:
| Yes exactly!
| bruce511 wrote:
| When traveling I often consider trains, and especially overnight
| trains. It's by far the most comfortable way to travel.
| Innovation in this space is a good thing.
|
| While I'm aware that feature creepy is the enemy here, I would
| suggest a way to "combine " two pods for those traveling as a
| couple. If I'm traveling with my wife we don't want to be in
| "separate pods".
|
| A retractable "wall" between 2 pods would be fine. It doesn't
| have to be elaborate, but you wanna point to something outside
| and say 'look at that' etc.
| ant6n wrote:
| We considered connecting pods. The orientation ,,behind" each
| other makes interaction difficult.
|
| They way this issue was ,,resolved" more less naturally during
| testing is that the pods all have The same orientation, so pods
| across the aisle approximately face each other. In our lab we
| had two iterations of the pods set up to face each other, and
| tester and testee interacted quite naturally --- once we set up
| our test rig like that, the questions ,,what about couples"
| reduced a lot, most understood the vis-a-vis intuitively.
|
| Our bigger cabins have two ppl versions, but a lot (if not
| most) travel is individua anyway, especially if night trains
| will be used for work travel.
| raybb wrote:
| I gotta say I love this? Are you in contact at all with the folks
| advocating for more night trains across the EU https://back-on-
| track.eu/ ? Seems like it could be pretty mission aligned.
|
| Btw I run a weekly newsletter about urbanism and while your
| trains may not be exactly related I think it's cool enough that
| we'll feature it in the upcoming week!
| https://urbanismnow.substack.com/
| ant6n wrote:
| Yes, we're connected with back on track! I'm a day-one member
| of the German branch. A bunch of their members came and
| participated in our tests.
|
| We're participating in their night train conference in
| September in Berlin, probably bringing a physical prototype.
| PaulRobinson wrote:
| Great work so far, but I note you are looking at a custom double-
| decker railcar with what looks like a very large loading gauge.
| My understanding is that for most of Europe, double deckers are
| not used due to loading gauge limits imposed by tunnels, bridges
| and so on. I presume your planned routes take that into account?
| Is that one of the reasons your mocked journey planner app
| doesn't include the UK?
| ant6n wrote:
| The ,,current" concept is for refurbished standard passenger
| coaches, which is compatible everywhere in Europe except UK.
|
| The ,,long term" concept is built around to be compatible with
| profiles UIC GB and G2, which work in most of Europe.
| matt-p wrote:
| We have sleeper trains in the UK, it would obviously be great
| if they were cheaper but honestly there's not _that_ much
| demand for travel from London - Scotland or Penzance in
| general. This is exacerbated by the fact a daytime train is
| easily doable so it 's a bit of weird, niche product.
|
| However the amount of demand there'd be for an overnight
| train from (let's just say) London to Barcelona at PS99 each
| way would be immense.
| ant6n wrote:
| You can run London to Europe via HS1 and tunnel, but the
| operating costs are insane. Anybody who propose that would
| get shut down by the ,,rail bubble" extremely hard (like
| midnight trains did a couple years back). I think it would
| be nice, but it's far off.
|
| As for intra-UK sleeper: yes there are some possible
| routes, but the market is small. Developing custom trains
| for that would be very expensive. And the concepts we
| propose don't work well cut the profile is so small.
| matt-p wrote:
| This was true a few years ago, but today the regulatory
| environment is very positive for these kind of services.
| You will see many non-eurostar open access operators on
| those tracks within the next year or two.
|
| I wouldn't massively bother with intra-uk yet, I agree
| wholeheartedly.
| herbst wrote:
| We have them in Switzerland. Often had to change train at the
| border and never thought it could be a technical limitation.
| matt-p wrote:
| The UK has adequate loading gauge for double deckers on
| HS1/Chunnel and HS2 (though I'm not sure if the Euston
| extension will support it or not).
|
| We can run London - Europe double decker today and if we
| connected HS2 to HS1 it would probably be possible to run from
| Birmingham, Crewe and Manchester to continental Europe too.
| eqvinox wrote:
| Hi! What's your perspective on the shortage of manufacturing
| capacity for night train rolling stock? Last I heard OBB can't
| build them fast enough for demand, and few other companies are
| able to actually produce these? Are you planning to build your
| own industrial manufacturing capability for this? And what about
| 2nd level suppliers?
| ant6n wrote:
| Yes, the bottle Beck for night trains is rolling stock.
|
| The plan is to initially target refurbished standard passenger
| couches. We have a large research and development consortium to
| develop the technology and adaptation of coaches.
|
| The refurb itself can be done by many companies, we don't need
| a large supplier like Siemens.
|
| As for the OBB nightjet, there's talk about shifting the last
| ordered nightjet to rail jets (for day travel). Speculation is
| that the low capacity of the Trainsets result in poor unit
| economics.
| eqvinox wrote:
| Interesting, Thanks! (and sad to hear about that shifting to
| railjets :/)
| Freak_NL wrote:
| Sleeper trains are held back by flying getting subsidised heavily
| by not having kerosene taxed, and national governments giving
| airports effectively unlimited room to grow; happily
| externalising the environmental cost. Why take a train if you can
| fly for a fraction of the cost?
|
| Trains in general are held back by governments not investing in
| rail infrastructure, because the pork barrel of another motorway
| link is so hard to resist (and we're not properly maintaining
| these either).
|
| Sleeper trains are held back, because cross-boundary
| collaboration between the various semi-national rail companies is
| tough (for Europe).
|
| Sleeper trains are held back, because there is a lack of modern
| rolling stock. Not completely new concepts; just up-to-date
| sleeper wagons (the OBB has the leading edge here now with their
| new wagons).
|
| There is room for improvement in the wagon designs, but it is
| almost irrelevant in the face of the other challenges.
| ant6n wrote:
| You start off by essentially claiming the unit economics of
| night trains being too poor compared to aviation is the largest
| hurdle, then finish off by claiming that unit economics are not
| that major issue.
|
| Our perspective is that with much improved unit economics, the
| problem overall becomes much more easily solvable. You can
| compete with aviation on price. You can pay for prioritized
| track access. You can operate trains privately without direct
| involvement of national operators.
|
| Finally, the refurb approach skirts the rolling stock bottle
| neck.
| Freak_NL wrote:
| I've listed four major causes. These are cumulative, not
| mutually exclusive.
|
| You _can 't_ compete with aviation on price currently. Not
| while its environmental costs are graciously overlooked and
| left out of the ticket price. This is a political problem.
|
| Comfort, reliability, sustainability, elegance, ease-of-use:
| those are the major selling points for night rail travel done
| right. This includes legroom, luggage space, and the ability
| to move around outside of your seat and to toss and turn in
| bed1 -- something current generation sleeper trains provide.
| If your goal is to cram as many people as possible in every
| metre of train length, you are optimising the wrong parameter
| at the wrong time.
|
| 1: The coffin beds in these hotel pods are clever from a
| space filling perspective, but I fear I would be nastily
| banging my knees several times in the course of the night,
| and that's ignoring the fact that sleeping that low close to
| the rails is just not pleasant.
| bluGill wrote:
| You missed a couple other points against night trains.
|
| first trains work best if they stop many times - how will you
| wake people up at 3am for their stop? For that matter who would
| agree to that? Without that churn many destinations are not in
| range.
|
| second, track needs maintenance. If the track is running at
| night as well when will you repair it? I makes sense to just
| close nost track every night for repairs. For busy two rail
| sections you can close on track and run very reduced service on
| the other - but this reduction means you don't want people
| sleeping as you can fill your trains just on people working
| night shifts.
|
| all of the above are challenges. They can be worked around in
| various ways however they to be considered to see if it is
| worth it-
| matt-p wrote:
| No they don't. You use the same hub to hub model as airlines.
| E.g London to Edinburgh, not London to Edinburgh stopping in
| 4 places.
|
| One of the nice things about traveling at night is you have
| less time pressure and congestion so when doing track repairs
| it's usually fine to divert the train.
| Freak_NL wrote:
| Yup. These two points are not an issue at all, and are in
| fact strengths for sleeper trains. Being able to redirect a
| sleeper train without any (or minimal) impact on its
| timetable is a big plus.
|
| Usually they depart from the last boarding station around
| midnight, and the first disembarking station won't be hit
| until six in the morning. Some outliers exist, but the
| number of people getting on or off there is negligible. For
| most travellers you are in the train before 22:00, and
| won't leave before 6:00.
|
| This OBB line is typical:
| https://www.nightjet.com/de/reiseziele/oesterreich/wien
|
| Yes, you can get off in Nurnberg at 4:08. But almost no one
| does that. The train just happens to have a halt there1,
| but 95% of the passengers get on in the Netherlands and the
| Ruhr Area, and won't get off until Austria (and vice
| versa).
|
| 1: I suspect mostly for rail topological reasons.
| bluGill wrote:
| Hub to hub - does that mean you wake everyone at 3am to
| change trains? Nobody switches cars with humans in them.
| Freak_NL wrote:
| > Nobody switches cars with humans in them.
|
| What do you mean? Changing the train consist during the
| trip? That is exactly what they can do.
|
| Hamburg/Amsterdam on one end, Innsbruck/Wien on the
| other. Those wagons get shuffled about like a deck of
| cards en route! So in Amsterdam there is one train with
| wagons marked Wien and wagons marked Innsbruck. You get
| in the right numbered wagon, and end up where you want.
| In the middle of the night, in the middle of nowhere
| (actually, probably Munchen) they turn two trains into
| two different trains.
|
| Most people never notice.
| gruez wrote:
| >Sleeper trains are held back by flying getting subsidised
| heavily by not having kerosene taxed
|
| Is whatever fuel trains use taxed? If not, I don't see how this
| is relevant.
|
| >and national governments giving airports effectively unlimited
| room to grow
|
| Which countries are those? For instance in UK they wanted to
| expand Heathrow since as early as 2006, yet due to various
| government shenanigans isn't due to complete until 2040,
| assuming it doesn't get backtracked again.
|
| https://en.wikipedia.org/wiki/Expansion_of_Heathrow_Airport
|
| Moreover since you're comparing against trains, don't trains
| need land as well, for the tracks and stations? Why do trains
| seemingly get a free pass from you on that?
|
| >Trains in general are held back by governments not investing
| in rail infrastructure, because the pork barrel of another
| motorway link is so hard to resist (and we're not properly
| maintaining these either).
|
| What makes motorways more of a "pork barrel" than train tracks?
| schmorptron wrote:
| Cool idea! Small typo on the landing page, in the media section
| it says "german nationa media" without the l
| ant6n wrote:
| Thanks!
| octo888 wrote:
| Looks like some heavy inspiration from modern airline business
| class cabins? Love it
| MrJohz wrote:
| How do you plan to handle groups or families with this system?
| Having larger compartments gives a lot of flexibility and allows
| families or groups to travel together easily and spend the time
| with each other. The hotel pods look like they only go up to two
| people in size, and the seats are very isolated from each other,
| not even allowing people to sit next to each other as far as I
| can see.
|
| I'm excited to see more work being done to improve night train
| travel though!
| ant6n wrote:
| We're working on standard railcars, which can be running
| together with traditional couchettes railcars. I would argue
| that group travel is largely a ,,solved" problem, using
| couchettes (railcars with cabins having 4-6 beds).
|
| For couples, the smaller pods actually kind of ,,face" each
| other across the aisles.
| torwag2 wrote:
| Whatever you do, add ventilation and if you think you have added
| enough, add some more. If you do not believe me, go and visit
| some of Tokyos capsule hotels close to the party districts.
| Nothing is more worse than the feeling to breath what has very
| recently left intoxicated people from one or the other side. Also
| for the sake of avoiding drama on those trains, add snoring
| tests.
| ant6n wrote:
| As a startup, there are two perspectives on these kinds of
| issues: just get something built and running as quickly as
| possible, or make sure you work out the kinks.
|
| So far we've been very much doing the second. Built it,
| experiment on it, test it, make sure it works. Only publish
| what's actually feasible. Set up more projects to make sure all
| the kinks will be worked out. Work together with experts to
| ensure proper ventilation, noise and vibration control, etc.
| etc.
|
| But as a startup, this process can be frustratingly slow. I am
| concerned investors may want to see quick results, not
| perfected solutions.
___________________________________________________________________
(page generated 2025-06-22 23:00 UTC)