January 28, 2022, ©. Leeham News: Having discussed where investments would be the most efficient in alleviating our Greenhouse gas problems and identified the low-hanging fruit, we now look at new technology airplanes that can improve the situation.
We start with classical airliners, working our way from small types to the largest, then we discuss the impact of new transport forms like VTOLs for short-haul transportation.
As we will use the Leeham Aircraft Performance Model in some of the work, there will be extra articles (for this one, a Part 4P) which are Paywall, where we use the model to generate deeper data and understanding.
A lot is going on in the world of Sustainable Aviation, with an abundance of upstarts offering one project after the other. New ways of doing aircraft using new Greener technologies are proposed, like battery or hybrid-electric designs competing with hydrogen-fueled proposals.
As we wrote about in Part 2, the electric technology used for cars only has to beat energy hogs that are below 10% in efficiency, our everyday car. But to qualify for aircraft use, the technology has to equal or beat what is used today, which is at the 30% to 40% efficiency level using modern turboprop or turbofan engines.
The difference is a whopping 300% to 400%. That this is ill-understood, we will see in this and following articles.
We can find this in energy-spent analysis for a typical flight, using the performance model. But before we get there, let’s do the simplest form of analysis possible, a weight check (really a mass check) of some designs.
We start with two high-profile projects where airlines (DHL, United, Finnair..) have put in provisional orders for hundreds of planes. When airlines order aircraft, they do thorough checks, so the analysis should show everything is fine, shouldn’t it?
Well, you would think so. Let’s start with the simplest of checks, which you only need elementary school math to conduct, the weight check (in high school, you learn it’s a mass check).
In Figure 2, we have gathered the weight data from Eviation and Heart Aerospace for Alice and ES-19.
The top line is the Maximum Take-Off Weight (MTOW) as communicated by Eviation for Alice and the maximum possible for ES-19 as it’s an EASA CS23 project, following rules similar to FAA 14 CFR Part 23 (CS23 and Part 23 has a Commuter MTOW limit of 19,000lb/8,618kg).
Then the line with the claim payload capacity. For the Cabin version, Evation states 2,400lb payload; Heart Aerospace says 19 passengers with bags, which we put at 220lb/99.8kg each.
The following line is the stated battery mass, 8,200lb for Alice and 3,000kg for ES-19. This gives 820kWh of energy for the Alice (at 0.220kWh/kg). Assuming the same energy density for the ES-19 battery, it has 660kWh. How far this energy will bring the two aircraft compared with their claimed range is for a later article.
When we deduce the Payload and Energy mass from the MTOW, we end up with the Empty weight, called Operational Empty Weight or OEW in airline speak.
The question now is; how realistic are the OEWs in the table? Anyone who has worked with aircraft will tell you: “these OEWs are totally unrealistic.” If you have learned the most basic stuff about airplanes you know the rule of thumb: the empty weight is at least half the gross weight.
In the accompanying Part 4P Paywall article, we use our performance model to show why. The model output, which uses data from 74 presently flying airliner variants, from short-haul Commuters to long haul Widebodies, made with aluminum or composites, shows that the OEW fraction (OEW divided by MTOW) never go below 47%. The short-haul Commuter class has fractions beyond 60%. We also discuss why there is no plausible reason why the above aircraft should have a lower fraction.
The paywall article discusses what can be done. Here is a summary:
These projects have been around for years, yet no one has done this simple math. Why?
Can our abundance of industry “experts” not conduct the simplest weight calculations that you learn during the first theory lessons for a private pilot license, or are they just too lazy to put pen to paper for such checks? The amount of writing about these projects without any critical analysis is amazing.
You’re to conservative Bjorn, a battery break through is around the corner. It’s a question of commitment & good looks really.
All joking aside, I can see & understand your frustrations Bjorn. Unwanted realities are easily put aside these days & peolle are getting away with it.
I’m always surprized with the thousands of engineers working on these projects, who know realities damn well..
… using external oxygen for discharge …
How do you bring enough oxygen into the cell for discharge? ( O2 partial pressure reduced in typical cruise levels.)
One company build small rotary screw compressors to pump oxygen into them. Today you have nitrogen/oxygen separators in most commercial aircraft deivered.
I think electrical superchargers are already in use on some commercial vehicles and certainly Formula 1 sports. I know PEM fuel cells need compressors. I would imagine these Japanese developed batteries would have very low power output unless aspirated by some kind of pressurisation.
There are a world of cooling and other subsystems around high performance battery packs. Custom Cells, the German battery maker being used by Lilium jet is claiming 300WHr/Kg energy density and the ability to operate at 150C which would mean little cooling systems would be required saving much weight. Lilium needs 300-330 Watt hours/kg.
Back to air batteries: Aluminium Air batteries already achieved 700WHr/kg in 25 years ago and seem to have a potential for 2000WHr/Kg. They need to be recharged by replacing the Al Electrode as a cartridge and would need pressurisation as well. The recent development of inert electrodes in Aluminium smelting means they can be recycled CO2 free. Not sure how energy efficient the ‘recharge’ is but probably over 60%. Those sorts of densities do make 1000nmi flights possible,
You shall always be cautious with the energy density claims of batteries for aircraft and VTOLS. To get to a certifiable battery SYSTEM you need to add around 50% of system components on top of the cell mass. Always ask yourself, what is the source stating? Cell values, or battery values (includes everything in the battery containers) or the system mass which includes external components including management electronics and cooling system. Lillium expressively states that the battery structure is not part of their density values.
“’m always surprised with the thousands of engineers working on these projects, who know realities damn well..”
They tend to sit in their labs under gag order. .. if the company actually has some eng-talent around.
What you read in the public domain tends to be fluffed up PR. ( if not just scraped from thin air.)
Below a much better link than the independent;
Below a link to the paper the Japanese team published.
Should produce about a 300-400nmi range electrical prop airliner if they get that certified at that density. About 1/6th of that possible with jet fuel.
Good reporting, Bjorn!
Personally, I don’t approve of upstart companies overpromising in terms of payload/range performance. It would be better to underpromise and then overdeliver, in my opinion.
The Pipistrel Velis electro has payload 172 kg and endurance 0.8 hours, while the Rotax-powered version Velis club has payload 260 kg and endurance 5.5 hours. So clearly the battery-electric version is severely limited in terms of payload/range.
so the pipistrel has a payload of one average american man + his lunch box….. lol.
I resemble that remark.
The Velis Electro has very low cost per hour.
It is a fantastic airplane for training schools.
“Can our abundance of industry “experts” not conduct the simplest weight calculations that you learn during the first theory lessons for a private pilot license, or are they just too lazy to put pen to paper for such checks?”
I’d say it’s more a matter of (deliberately) keeping things fluffy so that the funding will keep pouring in; after all, venture capitalists generally don’t have engineering degrees, and nowadays they tend to support whatever hype will get them positive exposure on social media.
For the public funding scene, things are changing . Fiscal tightening by the Fed means investors are being more wary of hot air stories: half the companies in the NASDAQ composite index have lost at least 50% of their value in the past few weeks. When asked about new projects in Wednesday’s Tesla earnings call, Elon Musk started “gassing” about a humanoid robot that he wants to develop to work in his factories. Those on the call weren’t impressed –Tesla stock went down 12%.
On this subject this morning:
“The tech sell-off has some venture capitalists worried the good times may be coming to an end”
Less than six months ago the Alice website had a different number for their MTOW: 6686 kg. They still listed the same battery weight (3720kg) and useful load (1134kg), which left an even more preposterous 1814kg OEW. And they published these numbers!
Dear Bjorn, Just some number form the Wikipedia Fuel Fraction article:
-The Rutan Voyager took off on its 1986 around-the-world flight at 72 %
-Steve Fossett’s Virgin Atlantic GlobalFlyer could attain a fuel fraction of 85%
-For comparison an A380-800 has a fuel fraction of 45% (with a small cargo).
These record numbers suggest to me that battery mass fractions of somewhat less, say 60%, are practical which should leave some room for payload. I’m assuming the higher volumetric density of battery packs saves both space and weight in the airframe. Your previous articles were eye opening in showing that the superior efficiency of electrical to power conversion is negated by the enormous penalty of carrying the extra weight and larger aircraft (making say electrofuel SAF more efficient).
However at a certain point, say at ‘normal’ mass fractions below 30% when operating at ranges of below say 100 nmi with current battery technology the penalties seem minimal and electrical aircraft would seem to have significant advantages.
There must be some cross over point where batteries are better?
In looking at the Lilium Jet eVTOL testbed and reading about the effectiveness of the boundary layer control of the tilting EDF on its ‘flaps’ I wondered whether these could be used for eSTOL. Indeed upon researching there are now proponents of eSTOL that are promoting and developing the blown wing and active laminar flow control wings that take advantage of the multiple propulsors possible with electric flight to blow most of the wing (not just roots). This easily achieves coefficients of lift of over 4 not including the direct lift.
There are now two companies arguing that with a precision landing system an eSTOL can utilise the same landing facilities as a typical eVTOL port. This would mean runways of less than 100m.
I suspect electrical flight will look quite different from jet or turbo props in which the fuel and turbine engine is replaced with batteries and electric fans.
They will take advantage of low noise footprint, eVTOL and ultra eSTOL to allow tiny airports to be built in many more places probably no bigger than a train station. Autonomous pilot free flight is necessary to make this affordable. It’s already possible, we see drones doing it all the time. It’s a matter of increasing safety.
That is a massive increase in tail size. Did no one do an aero sim on the previous model?
Indeed it is. I have been watching the taxi trials (including the excursion) and the fin and tailplane just look ridiculously small to my eye. The fact they are showing a bigger version doesn’t surprise me. We haven’t seen them lift the nosewheel yet so will have to wait and see how much authority that tail provides for this first airframe.
A small — but interesting — example of the fact that the present-day aviation industry keeps incrementally achieving better environmental performance:
“A newly designed C295 testbed has made its maiden flight.”
“The aircraft, based on the Airbus C295 tactical airlifter, has been designed as part of the European Clean Sky 2 (CS2) and EU Horizon 2020 research and innovation programme, that allows technologies related to CS2’s future regional multimission aircraft to be tested.
“‘The [aircraft] modifications include new materials and technologies designed to achieve noise, CO2 and NOx emissions reduction,’ Airbus noted.
“‘With these technologies applied in a future regional multimission configuration, up to 43% CO2 and 70% NOx reductions can be achieved in a typical Search and Rescue mission of 400 nautical miles, as well as 45% less noise during take-off.'”
And this is much more realistic:
Thanks for the link. The P-Volt Electrical is shown as having a Battery Mass to Max Ramp weight ratio of 809kg/4086kg = 0.197 i.e. 20%. An A350 or B787-10 operates at over 40% mass fraction and many aircraft typically operate at 30% fuel fraction.
Tecnam replaced the 750L/600kg of fuel with 1100kg of battery. They kept maximum fuel payload about the same (800kg) by increasing MTOW and compensated with an increase in engine power from 2 x 275kW to 2 x 320kw.
The Tecnam Traveller’s airframe is built to a price of simple construction and ease of maintenance and acquisition.
I suspect a lighter fully composite airframe built around the batteries would probably be able to achieve more battery, say its 50% taking it to 30% mass fraction. Say the aircraft MTOW goes up a little as well to retain same cargo but say we reduce parasitic drag by 50% through use of moulded 3D structure to give clean aerodynamics. This will allow about a 41% better cruise speed that increase range.
So I think a purpose built Electric aircraft would get twice the range and 40% more speed. Say 170nm with current battery tech and 280nmi with future instead of 85/135.
Note this is still 60% of the 440nmi that Eviation is claiming/targeting for the Alice but its a far more useful range than 85/135nmi of the P-Volt.
The Tecnam approach is realistic.
And a short hopper with very low cost per hour of flight could be really useful for inter-islands or rough terrain in places with no roads (Tecnam is partnering with Rolls and Widerøe).
I guess that an ICE as range extender will be part of the package later.
The market for flights under 100nmi with the following constraints:
1 About a 1000m runway at both ends.
2 The aircrafts inability to be deployed on a longer flight, say 150nmi and above to earn revenue elsewhere and pay for itself must be extremely limiting. Who will buy that? Island hopping works by fast boat as well.
So imagine a purpose built aircraft. Keep the cargo the same but increase the MTOW and size by 30%. Use composites and a designed optimised for batteries to increase battery mass fraction from 20% to 30%. Improved aerodynamics will yield another 41% in range.
An first generation eVTOL with a range of even less, such as Airbus’s City’s conservative (80km or 43.2nmi) will take a lot of island hopping and float plane market all without the need of expensive runway. They can operate right of the shoreline.
Between a longer range purpose built electric aircraft and eVTOL what will remain for the narrow range of Tecnam P-Volt?
Ultimately there are two parts of the arguments: Part 1 of the argument is about emissions per passenger per km travelled. Part 2 of the argument is can the electric aircraft perform the mission in terms of range.
Below are the End-to-end CO₂ footprint including emissions from operations,
production and infrastructure assuming use of renewable energy. Taken from the Lilium NV website but from independent sources. (pkm passenger grams per km)
Passenger Jets CO₂/pkm 189g
Gasoline Cars CO₂/pkm 142g
Electric Cars CO₂/pkm 31g
Trains CO₂/pkm 18g
Lilium CO₂/pkm 13g
Lilium NV eVTOL is claiming to handsomely outperform even electric cars and trains. These are end to end emission calculations including manufacturing costs, airfield costs as well as direct energy costs. They need to be taken with reserve. A Wizz Air A321neo will achieve direct CO2/pkm 57 which is 30% of the end to end calculations above. (probably based on a B727 or B747 in intercontinental). If we fuel the conventional Airliner with 100% electrofuel SAF of about 60% efficiency and use an advanced model I assume energy costs come down by a factor of 5-10.
Even if we quibble over the numbers being +/-50% or so I’m somewhat pleased the eVTOL aircraft appears to outperform rail.
The problem with all electric flight is range. Even if we supersize the aircraft to increase range with the same cargo the weight rapidly reduces efficiency.
Lilium is targeting a range of 250km. They are relying on some high performing 300W.Hr/kg cells but also the advantages of eVTOL: operating without a heavy undercarriage for high sink rates, 2 high wing loading and 3 no empennage.
With a operating range of 250km it can’t fly the 600km from Dublin Airport to London Gatwick but it could do the journey via a hub in Liverpool England. Furthermore the low takeoff/landing and environmental footprint would allow multiple minor ports to be established in Dublin and besides John Lennon Airport in Liverpool. This saves an enormous amount of time in travel to and from large airports.
I suggest that electric flight if in the form of eVTOL can compete on the sub 500km flight market via a single hub and that they will have a speed advantage by having ports closer to the destination. They will also work well with high speed rail and conventional airports.
The numbers are from the Lilium NV Fact Sheet in their news section.
International Energy Agency. Fraunhofer Institute. The International Council on Clean Transportation. Umweltbundesamt. Öko-Institut. Lilium engineering estimates.
Note: Analysis assumes that electric cars, trains and Lilium Jet run with renewable energy and that the batteries of electric cars and Lilium are produced with renewable energy.
“Analysis assumes that…are produced with renewable energy.”
Which, of course, isn’t the case: apart from Iceland (geothermal), Norway (hydroelectric) and France (nuclear), no country has a sizable portion of renewable electricity generation in its mix. See link: Germany comes in at the top of the list with 12.74% renewable.
Electric vehicles don’t reduce emissions: they merely hand the emissions problem to the people responsible for generating/distributing electricity…and to the people responsible for processing mountains of used batteries.
And the present-day CO2 emissions from aircraft are, indeed, a LOT lower than the figure quoted in your table. Here’s data for 6 different European carriers in 2021 — for airlines that hadn’t yet switched over to significant numbers of neos:
I’m not sure exactly how the calculations for emissions were done using renewable energy I presume they were theoretical calculations around wind energy in Germany and are fairly trustworthy since they were from a Fraunhofer Institute, which are reputable. Even if not accurate they are approximate and very indicative of relative values.
Generally a wind turbine is counted as having only 20% of the emissions of a gas fired power station having a 5:1 payback. Again I understand there are costs in network, network stabilisation and energy storage quite possibly not realistically accounted for.
I’m on the record as saying that I believe that nuclear will need to be part of the mix but some countries may be so rick in renewables it will work for them.
“Electric vehicles don’t reduce emissions: they merely hand the emissions problem to the people responsible for generating/distributing electricity”
This was was probably once true, I believed it, but it does not seem to be the situation anymore.
I found it surprising that BEV are more efficient by a factor of around 2.5 over ICE Vehicles in the US mix of fossil, nuclear (about 20%), renewable(10%).
The above calculator/chart is from an MIT study. I can’t find the direct link but there is an app. Someone posted the direct link here once.
The problem with BEV is their extraordinary up front costs.
The point I am making is that if one built an 29 seat electric passenger aircraft about the size of a DC-3 whose maximum mission was to fly 29 passengers 100nm-150nmi i then it would be an extraordinarily efficient aircraft from the point of view of using renewable electricity.
If required it to to fly the same cargo 500nmi it would be a massive aircraft with a tiny fraction of cargo so heavy it would be far less efficient than a conventional aircraft using electro fuel SAF.
However the electric aircraft can still fly the route in 1-2 hops and retain its efficiency. This would be impractical for a fixed wing aircraft because of the cost of the runways and the time wasted taxiing but it could work in an eVTOL
I do not have much hope for fixed wing electric aviation except maybe in the even of a 2000kW.Hr/Kg Alumium Air battery or similar.
“BEV are more efficient by a factor of around 2.5 over ICE Vehicles in the US”
That statement deserves a little more nuance — for example, as provided in this detailed link:
“The report — authored by Michael Sivak and Brandon Schoettle — notes that an electric car recharged by a coal-fired plant produces as much CO2 as a gasoline-powered car that gets 29 miles per gallon. (For context, the average mpg of all the cars, SUVs, vans and light trucks sold in the U.S. over the past year is 25.2 mpg.) A plug-in recharged by a natural gas-powered plant is like driving a car that gets 58 miles per gallon.
Solar, wind and geothermal do far better on this score, but they generate a small portion of the nation’s electricity. More than 64% of electricity is generated by coal, natural gas or other fossil fuels.
The U of M researchers calculate that, given the energy mix in the U.S., the average plug-in produces as much CO2 as a conventional car that gets 55.4 miles per gallon.”
As you correctly point out, this doesn’t include the emissions required to produce the car (about 17.5 tons of CO2 on average), and also the emissions associated with constructing/expanding and maintaining a power distribution grid. Importantly, it also doesn’t include the battery recycling costs at EOL.
But, on the basis of this data, I’ll amend my original statement to:
“Electric vehicles don’t *do away with emissions*: they merely hand the emissions problem to the people responsible for generating/distributing electricity…and to the people responsible for processing mountains of used batteries.”
The Fraunhoffer figures are an attempt to incorporate end to end emissions. For instance. Direct emissions from the engines of the vehicle, train or aircraft plus indirect emissions from construction and maintenance of the vehicle, train or aircraft and its associated infrastructure such as road, rail, airport or vertiport. All of those French rail workers and contractors do nothing but produce emissions maintaining and operating rail.
Obviously an airport consumes more than a vertiport and both are probably better than a rail link. Per passenger emissions of high speed VFT rail are excellent so long as the track is utilised with a high capacity factor. (France good, Spain bad)
It looks like indirect emissions may be greater than direct emissions. Obviously a Lilium veriport about the size of a dozen tennis courts in the middle of a field groing cabbages or hay to act as a 250m sound buffer has a lower foot print that a 1500m runway needed to handle a 19 seater and much lower footprint than a rail track.
However eVTOL, if it achieves the required range of about 100-150nmi will work very well with very fast trains and conventional airports.
Quite a lot of countries have 80% or more renewable electricity production
Brazil is biggest country in this range just over 80%. Albania is 100% ,NZ is 84% Kenya and NZ are quite high with geothermal as well.
But, apart from Brazil, how many of those countries are heavily industrialized? And/or heavily populated? Sure, Kenya has 53 million inhabitants, but its economy comes in at position 65 in world ranking.
Some of them rely — at least partially — on burning biomass, and that’s no longer categorized as being “sustainable”.
And new hydroelectric schemes are also increasingly considered to be a no-no, because of other types of environmental damage that they cause.
Sweden is pretty industrialized in many areas and have a mix of nuclear, hydro and wind power. The cost per MW wind power keep dropping as competition heats up and parts manufacturing is spread around the world to the lowest bidder. Even Germany’s Enercon high quality windmills has to adapt and join the race to lower cost/MW. We expect that the sea based +12MW units will follow the same cost curve down.
@cleaes, You have to watch the wind turbine manufactures and the wind farm operators. They will quote the load levelized cost of producing energy at their wind farm output in KW.Hr what they won’t mention is:
1 The significantly higher cost of transmitting and stabilising that power which is as much as the energy itself doubled in the case of offshore.
2 The accelerated depreciation rates they receive for their short lived wind turbine power plants.
3 Cost of converting their cheap energy into reliable dispatchable power. It requires peaking plant, batteries, storage. If hydro is around its easier.
No doubt things will get cheaper and cheaper but I doubt we’ll ever get cheap power again.
Most people and corporations mislead. Renewable energy sector is no better or worse.
In an attempt to get on topic. We do need to start thinking in terms of the system and life cycle emissions cost of delivering a service. Wind power needs transmission lines and backup, cars need roads, trains need rail and maintenance and aircraft need airports and they all need to be manufactured/recycled.
Interesting link Duke. Some 77% of Brazils electricity generation is from hydro. Little of the planet is in this position but certainly New Zealand, Iceland & Norway. I would hesitate to call hydro renewable or CO2 neutral. In many cases it is very CO2 negative such as the Aswan Dams.
Most countries can with modest increases in electricity prices (say 50%-60% above fossil fuel power) can achieve 15%-20% renewables. This is because traditional turbine plant was always built to have 20% spinning reserve of immediately dispatchable power. This was necessary to drive power into a fault (powerline short till circuit protection cleared, bypass a power line or faulty power station or substation or deal with a poor power factor or harmonics). Without it networks become rapidly unreliable and unstable.
There is a cost in running the base plant in an inefficient manner and of the cost in building the extra long poor utilisation factor transmission lines needed by renewables. Ireland actually recorded no reduction in emission despite heavy investment in wind turbines for many years.
Above the 20% renewables peaking plant and or energy storage is needed. Prices rapidly escalate. A country that can supply 50% of its energy from hydro can easily achieve 100 renewables simply by turning the hydro on an off as renewables such as wind and solar wax and wane. Few countries are in this position.
As we saw in Germany and the low countries all wind turbine energy has stopped for several months and most of the PV cells are iced up.
This was explained to them and statistical data provided but its hard to get sense into someone fixated on a fantasy or ideologically motivated into group think.
At this point 70% renewables looks like wind and solar providing energy with about 2-4 hours battery backup giving thermal power stations plenty of time to get started. Backup energy for night and long term (months) is provided by combined cycle power stations running on natural gas and/or hydrogen. The hydrogen would be made from excess electrical energy r or it would be blue hydrogen made from natural gas and coal with the CO2 sequestered under ground.
It will work but it will be expensive.
Building damns is one of the worst things we ever did as humans.
There are many complex factors to consider. For instance. Combined Cycle power stations can achieve nearly 60% efficiency and there were hopes of 70% when the industry died because of the subsidies going into renewables.
There is an very strong empirical argument that it was better to build combined cycle power stations than wind turbines backed by less efficient peaking plant. Such nuanced arguments are no longer possible.
It would be more honest to had the TGV and its 2g of CO2 per km (it’s even less with the Ouigo service). Here they act if they were the penacle of low carbon transit. I agree though that their commercial jet numbers are outdated.
That TGV figure applies only in France, where 70% of power is nuclear.
For travel outside France, the figure is much higher (see link).
Also, it doesn’t include the CO2 associated with providing, upgrading and maintaining the track infrastructure along the route.
But your point certainly demonstrates the advantages of using nuclear rather than fossil fuels. And, yes, for high-volume transport on *existing* infrastructure, electric trains are wonderfully efficient. Whether that performance merits the costs/CO2 associated with constructing *new* lines is very much open to debate.
And then there’s this!
“This flying car could be the future of travel after regulators officially cleared it for flight ”
(p.s.: it’s not electric)
I am not sure I missed it, and it is implicitly mentioned in the part 5, but I think the concepts of MTOW and OEW do not apply as such to the electric technology, as the energy for electric technologies does not have mass, and therefore the mass of the batteries does not vary based on their electric load, which is obviously the case for the hydrogen fuel. And, therefore, a) there is no trade-off between range and payload, as the ‘energy mass’ of a battery does not vary, and b) the diminishing cost of carrying fuel (mass) during a flight also does not apply to the battery. In performance terms, for electric energy you always carry the same ‘energy’ mass during the flight, independent of the phases of the flight and the flight mission (unless you remove batteries).
Hi Bernhard, it’s covered in Part 5 as you say. Look at Figure 3 and the text around it.
Heart seems to have revisited their design and agreed with your analysis and moved on to a 30-seat variant called the ES-30. Have you looked at the new design to see if it is any more realistic?