Bjorn’s Corner: Sustainable Air Transport. Part 23. Fuel Cell-based 70 seat airliner

June 10, 2022, ©. Leeham News: Last week, we looked at the different fuel cell systems that can go into a 70-seat airliner like the ATR 72. In this week’s Corners, we implement these in the aircraft and look at installation effects and efficiencies.

The deeper discussion is in the sister article, Part 23P. Fuel Cell-based 70-seat airliner.

Figure 1. The ATR 72-600 70-seater turboprop. Source: ATR.

A Fuel Cell-based 70 seat airliner

As discussed in our market articles, the market for airliners below the 70 to 100 seat segment is small. We, therefore, give the example of a fuel cell-based propulsion airliner as a 70-seat propeller airliner, very similar to the ATR 72-600 in Figure 1.

We use the base data from the ATR 72 to examine how our fuel cell systems fit such an aircraft. Here are the conclusions from the Part 23P article;

The challenge of fuel cell-based airliners

The fuel cell-based propulsion system weighs between five to seven times more than the gas turbines and fuel system it replaces for an ATR 72. That is before we have counted the tank mass.

Using liquid hydrogen, we can limit the tank mass to around seven times the tank and fuel system of an ATR 72. If we use gaseous hydrogen and have the ambition to keep the range in the ballpark of the original aircraft, we talk 40 times! If we cut the range from 800nm to a practical minimum of 250nm, we have a tank that is 15 times heavier.

Retrofitting airliners

The installation effects of gaseous hydrogen systems are unacceptable both for a clean sheet aircraft and in a retrofit case. There is no payload capability left in a retrofit case.

Using liquid hydrogen, we can produce a retrofit kit for an ATR 72, but the aircraft is reduced to a zero-emission functional demonstrator.

With a superconducting motor and cabling system and where the LH2 cool sink is used efficiently to help the thermal management system keep the Balance of Plant, we have a 250nm range and can take 36 passengers, a 49% payload reduction.

If we instead use a non-superconducting motor and cabling system, capacity reduces to a 17 seater with a 250nm range, a useless aircraft.

Depending on the exact implementation techniques, these figures can vary a bit. Still, these figures are before we have looked at problem areas such as the cooling drag from the thermal management system and the problem with a limited Maximum Zero Fuel Weight, MZFW, for the retrofit case. The actual retrofit cases will probably be worse.

The clean sheet Fuel Cell airliner

A clean sheet aircraft can be designed around liquid hydrogen and a superconducting motor/cabling system. This variant is 20% more efficient than today’s turboprop engines, rendering it an operationally acceptable 70 seater with functional payload-range characteristics and zero emissions.

The amount of development to get there is substantial, but it is the only route to a true zero-emission airliner. Long term, the fuel cell airliner will be there as the most environmentally friendly alternative.

40 Comments on “Bjorn’s Corner: Sustainable Air Transport. Part 23. Fuel Cell-based 70 seat airliner

  1. Will the Embraer new turboprops TP70 and TP90 be feasible with LH2 power and gas turbines? Now it is a 800nm JET-A1 powered aircraft project, a 400nm LH2 version would be useful on many turboprop routes but can you fit the LH2 fuel without building a new aircraft?

    • Yes, but Embraer would have to place a lot of the components in the main wing to avoid severe MZFW problems (I assume the LH2 tank and fuel cells are in the rear fuselage).

      One could foresee a wing that is at the same time the surface-based heat exchangers for the thermal management system. Its mass is in the order of the tank for such an aircraft.

      The probable solution is a LH2 burning turbine solution IMO.

      • My idea was LH2 burning modified gas turbines and maybe a fuel cell APU as a weight saving compared to fuel cell propulsion. You could add LH2 “drop tanks” as it is old technology and fairly low drag and easy to replace fully loaded with LH2, maybe not so easy to EASA/FAA certify?

        • Agree. The only problem area with H2 burning gas turbines is the water vapor emissions, but if the turboprop stays below say FL300 I think it improves the situation. We also have the WET condenser ideas for catching the water vapor on the way out of the nozzle. Let’s see where this takes us.

          • Is catching the vapor actually a good idea at a system level? Adding the O atoms to the H2 makes the resulting H2O molecule heavy, so adding structure to capture it and then carrying it around would result in more weight and fuel consumption. Would any gain in reduced radiative forcing of the vapor outweigh the added weight and additional required hydrogen? Keep in mind that there’s an opportunity cost component: any wasted energy making green hydrogen that we didn’t need is energy that could have gone to decarbonizing other sectors.

  2. Very nice to see that the final paragraph confirms that a (clean-sheet) fuel cell aircraft is feasible, thanks to exploitation of superconductivity.

    From the link:
    “While hydrogen fuel cell technology to power alternative-propulsion systems is still new to aviation, cross-industry collaboration, like the strategic partnership between Airbus and ElringKlinger, will be essential to maturing the technology’s potential in the years to come.”

    • You might get close to 20% efficiency increase in commercial TP engines if they used much more modern designs and materials.
      But no one wants to pay for it outside military

      So I dont see how Bjorns claim (‘ This variant is 20% more efficient than today’s turboprop engines…”) will make commercial sense when its allied to a whole new airframe as well. And not mentioned is the in flight gains as compared to bench testing
      This is the 70 seater TP market, which is more like a graveyard of ambitions.

      • Luckily for the rest of the world, Airbus *does* see potential merit in the idea.

        • No it doesn’t. They are talking hybrid which means fuel cell APU and a modified gas turbine with probably LH2
          It’s says so in your link. No one is talking about fuel cell main propulsion power.

          • Yes, the first step is fuel cell APU and LH2 buring gas turbines, they can be of high pressure ratio >60:1, 8 bladed Ham Sundstrand prop and high efficiency if using the LH2 cooling capacity smartly compared with the JET-A1 burning PW127 of approx 16:1 in pressure ratio and 6 bladed props.

          • “No one is talking about fuel cell main propulsion power.”

            Apart from the fact that this recent Bjorn series has talked about it (you apparently missed that), Airbus is also talking about it:

            “To further explore the possibilities of fuel cell propulsion systems for aviation, Airbus has entered into a strategic partnership* with ElringKlinger, a company with over 20 years of experience as both a fuel cell system and component supplier.

            “This partnership will contribute to growing our in-house expertise in alternative-propulsion systems,” says Glenn Llewellyn, Airbus VP of Zero-Emission Aircraft. “Today, Airbus has significant know-how in electric propulsion and fuel cells thanks to work carried out at our E-Aircraft System House and currently taking place at the ZAL in Hamburg. This partnership will be a phenomenal acceleration in bringing hydrogen fuel cells to future aircraft.”

            The agreement involves co-developing and co-validating aviation-compatible fuel cell stacks. In particular, testing the power density of fuel cell stacks for aircraft will be a key focus area.

            “Battery-powered propulsion to fuel larger aircraft over longer distances is not possible with today’s technology,” explains Matthieu Thomas, ZEROe Aircraft Lead Architect. “Hydrogen fuel cells could be a great alternative because they can generate—with zero emissions—significantly more power and energy for a given weight. This makes fuel cells an extremely interesting technology to achieve our ambitions.”


          • @Bryce, Gas turbines buring LH2 is the first step. Of cause they are researching fuel cells with 0 emissions, but it can be hard, heavy, hard to certuify and expensive to get certified. The big avtantage of LH2/H2 in gas turbines is the fast flame velocity making for a very short time at high temperatures and hence very low NOx production besides the cooling capacity of LH2..Todays JET-A1 burners of the latest design aslo have a huge margin for NOX emissions, so it becomes easier to lower it further with H2 combustion.

          • @ claes
            I’m quite aware of the advantages of gas turbines.
            I’m also aware of the fact that there will be a cascade of steps in the development and adoption of hydrogen propulsion.
            But I’m equally aware that fuel cell electric propulsion has the advantage of producing *zero* NOx and expelled water vapor.
            Bjorn says it himself in his final paragraph:
            “The amount of development to get there is substantial, but it is the only route to a true zero-emission airliner. Long term, the fuel cell airliner will be there as the most environmentally friendly alternative.”

            Some people just can’t or won’t accept that that is the direction in which development is pointing…regardless of how bumpy the development route is.

          • The fuel cells in Airbus spiel is for APU only and thats where the named other fuel cell developer comes in.

            They have no mention of electric main propulsion of
            of former jets based on fuel cells.
            None what so ever
            Every one has ruled that out ( not forever but its not a pathway they can develop in foreseeable future )
            And for those who say look at how far gas turbines have come since 1940s to 1960s, technology like that has plateaued for the last 30-40 years with only slow movements. The real jumps have come with new areas of circuits/electronics and software.

            Surely theres one country that has the resources of people and money that can ‘make great leap forward’ in fuel cells and leave those ancient Leap-C engines behind.
            How’s their projects and progress, enquiring minds are asking.

          • Somebody needs to brush up on reading skills!

            The Airbus articles make extended reference to fuel-cell-driven *propulsion*.
            An APU is not used for propulsion.

          • You mean this
            ‘Today, Airbus has significant know-how in electric propulsion and fuel cells thanks to work carried out at our E-Aircraft System House and currently taking place at the ZAL in Hamburg.’

            The first link ( E aircraft) is about a now discontinued electric fan project using Bae 146 and Rolls Royce
            ‘ 2020, Rolls-Royce will deliver the electric motor that will replace one of the four gas turbines installed on the test aircraft—a BAe 146. The electric motor will be equipped with 2 MW of power. ”
            A dead end already by end of 2020 !

            The other link ‘ZAL’ is a ‘cluster’ with 60 people doing who knows what . But many projects unrelated to electric propulsion beyond a toy quadcopter with a fuel cell!

            happy to provide guidance for you firstly in ready the links you didnt and updating the latest news, but doubtfull you have the core understanding to keep up.

          • Even when it’s explicitly quoted for him above, he still can’t read it.

          • A quad copter thats the size of say a table top isnt a commercial passenger electric propulsion project

            Do tell us what’s Airbus ‘ electric propulsion project’ if you seem to see what no one else can

            Even the discontinued RR- Bae 146 was a turbine driven motor generator in the rear providing power for the main electric fan . Not a fuel cell either .
            Marketing people love those readers that accept cleverly worded spiels that hint at the opposite to the actual reality.
            Used to be called – one born every minute

  3. Anyone care to comment on Connect Airlines announced order for 75 hydrogen fueled ATR 72-600s flying by 2025. How does something like this benefit a company? Isn’t the potential damage to their reputation not worth the risk?

    • The planes in that order will use gas turbines rather than electric motors / fuel cells.
      The converted planes will lose a chunk of fuselage volume for LH2 storage, but the concept is still relatively viable.

    • “Connect Airlines and hydrogen propulsion solutions company Universal Hydrogen…..”
      Its just puffery from paper companies

      Connect Airlines seems to be a startup , ‘hoping’ to fly Q400 from Toronto to Chicago and Philly. yeah right, theres unmet demand there for a TP

      It always pays to go beyond the headlines

      • Breeze was once a paper company (called Moxy), with an out-of-the-ordinary plan to directly connect small city pairs using small NBs.

        Look at it now…it’s thriving!

        • It also has people of substance behind it.
          Who’s running Universal Hydrogen and how much money does it have
          But they have bunnies who lap all the puffery

          • Ryanair had nobody “of substance” behind it when it started with a single Embraer EMB-110P1, and then moved on to a mini-fleet of four second-hand BAC-111s.

            Look at it now.

          • Look at it ‘now’ 45 years later !
            The LCC niche was new to Europe but not invented by Ryanair
            And what about the part that matters -Universal Hydrogen. Whats their people, finance, development path ?
            No wonder you dont mention them

          • @Bryce, The staff of Ryanair were no rookies but some GPA staff and others that had experience and could calculate pretty accurate. They were good at negotiating with Boeing and secondary airports around Europe and got paid to fly there besides allowing their pilots to be paid to a consulting company in Ireland and pay arbritrary taxes. It is strange how Ireland and to some extent Netherlands/Luxenburg can be within EU and have such open holes to avoid paying normal EU income and corporate taxes. A bit like having Turkey in NATO.

          • @ claes
            The EU doesn’t regulate tax levels: they’re determined at a national level.
            A similar situation exists in the US.

            As regards Tony Ryan: his stint at Aer Lingus — which, at the time, was one of the smallest and worst-run “government” airlines in Europe — was hardly a natural stepping stone to setting up an LCC. Ryanair started as a regional carrier with a single route that was not served by anyone else (Waterford – Gatwick).

            We agree on Turkey/NATO.

          • @Bryce, Yes nation decides corporate taxes and income taxes for crews, if they allow a system with loop holes that differ alot from i.e. the German tax system they should be kicked out of EU and loose traffic rights to/within the EU. France have been fighting Ryanair on this and Ryanair does not fly German domestic rotes while Lufthansa Technic does some MRO for Ryanair.

  4. I have 37 years of experience as an Air Force Mechanic, FAA A&P and Aircraft Designer on McDonnell Douglas F-15s, General Dynamics F-16s, Boeing 707s, 720s, 727s, including the UPS 727QF Tay 651 Re-engine, 737s [all versions], 747s, including NASA 747SP Shuttle Carriers, 757s, Douglas DC-8s, DC-9s [all versions], MD80s [all versions], SR-71s, NASA Shuttles, T-38s, USAF “Groom Lake” Test Planes, Edwards AFB Test Planes, Boeing B-17s, NAA B-25s, P-51s, P-38s, P-4os, T-6/SNJ, ATR42/72, Swearingen Metro, Beech 18, Beech 1900, and several dozen Experimental Aircraft – I live in Alamogordo – Holloman – White Sands, NM and if you provide me with the ATR 72 Fuselage [airworthy or can easily be made airworthy, preferrably 2 or 3 for Flight Test repeatability], the engineering support and the LOX / LH2 / Fuel Cell Equipment, I will convert the ATR 72 into a Test Bed and test it over the White Sands Desert .. one LH2 Modified Twin-Turbine Powered ATR and One Twin [or Quad] Electric Motor/Fuel Cell powered ATR .. we can use this test vehicle to determine FAA airworthiness standards of the LH2 and Fuel Cell systems, the reliability / TBO of the Electric Motors and LH2 Turbine Engines, “safety of flight”, weight and balance considerations, whether the vehicle is going to be aerodynamically stable [unlike the 737 MAX], can the pilots mechanically/physically recover from a nose down / too far forward CG or a nose high / too far aft CG [ the 80 year ‘airworthiness’ safety of flight standard the 737 MAX FAILS at ], and exactly what kind of changes need to be made to the fuselage to accomodate passenger safety and load carrying [ 24, 36, 48, 60, 72 ] for 2 hours / 250 nm .. with an engineer designing these changes as we discover them, and the proper structural metal fabrication capability available, we can make these changes to the fuselage as needed, remain structurally safe and airworthy, and come up with a definitive “shape” or “prototype hull design” that will best fit the “Hydrogen Type Rating” needed for a commercially viable 70 to 150 passenger airplane ..

    • The modelling done Bjorn using his data and adatations of common software packages shows the unfeasibilty of most of what you suggest.

      And that’s the way any minor or major developments are done these days. It’s proven in the software first before even taking off like you suggest and there isn’t even a workable APU fuel cell to test. Hydrogen based turbines is being done via very large aircraft with a few LH2 tanks in the cabin.

      • You’d better go back and read my post – NOWHERE did I say “taking off before the software” .. nowhere did I say ignore the software .. nowhere did I say we don’t need the software .. you completely either did not read or did not comprehend exactly what I was saying .. I have worked with dozens of software designed and software “proven” aircraft and spacecraft – and every single “software proven” aircraft and spacecraft – AFTER it is finished being tweaked in software has to eventually be built ..

        Modeling is not always accurate or correct and I can absolutely guarantee you I can make my incorrectly named “unfeasible” test vehicle and FAA supplemental type certificate work – and I absolutely guarantee you that I can build a financially feasible Commercial air carrier powered by Hydrogen – whether it be Hydrogen powered gas turbines or fuel cell powered electric motors IT DOES NOT MATTER – someone, Somewhere, is eventually going to BUILD a workable fuel cell powered apu outside of a computer model .. someone, somewhere, is eventually going to build a workable, Hydrogen powered turbine engine installed in a 30 to 100 seat passenger airplane – and I am 100% capable of building that 30 to 100 seat passenger airplane USING BJORNS MODELLING – but NOT on the first day – it’s going take at minimum 12 months and probably 24 months and maybe 36 months – and I NEVER said “ignore Bjorn’s modeling or “ignore the computers” or “take off before the software is finished” – not once did I ever say that – YOU MISUNDERSTOOD MY WORDS .. – Bjorn is modeling [ according to what is publicly posted ] for a COMMERCIALLY VIABLE 36 – 72 passenger turboprop airliner – based on the ATR72 FUSELAGE – that can take off tomorrow flying paid customers .. modelling is only the FIRST STEP – it is not the only step and it is not the last step .. you still have to build the EXPERIMENTAL AIRCRAFT [ a test bed that CANNOT FLY PASSENGERS ] – let me repeat that [ because it probably went over your head ] – a test bed that cannot fly passengers – it is a violation of of FAA regulations – and you need to Test it for several thousand hours and usually 1 to 2 years AT MINIMUM .. and I absolutely guarantee you that CHANGES will need to be made to Bjorn’s model .. it will not be perfect right out of the gate.. mistakes in calculation will be discovered .. mistakes in weight and balance will be discovered .. computer programming won’t work as intended.. components won’t work the way they were intended .. and NEW components and NEW programming will have to be built .. Engineering Changes will have to be written and implemented.. new parts will have to be manufactured and installed.. new flight tests will have to be conducted.. new failures will be noted .. and then new Engineering Changes will have to be written and implemented – and the cycle repeats, over and over again, day after day, week after week, for 12 months or two years or ten years .. this isn’t my first airplane rodeo r my first day on the job .. I’m not an idiot .. I have an IQ higher then 98 percent of the entire world’s population over the last 12000 years .. I’ve been doing this for 37 years .. hundreds of experimental flight tests .. thousands of hours of flying .. tens of thousands of hours buried in a fuselage or an engine or a system [ including the apu ] doing repairs and modifications – and. Guess What? – working, every single day, with Aviation and Aerospace Engineers who design things in computers on software .. software proving their designs .. Ive been doing it for 37 years .. Pratt and Whitney. General Electric, Rolls-Royce, Boeing, McDonnell Douglas, Airbus, NASA, USAF, Garrett Airesesrch, UPS, Learjet, Bombardier, Lockheed, General Dynamics, and dozens of other Aerospace Companies – and I know a hell of a lot more about the process than you do .. if you don’t want to do it on the ATR THAN PICK ANOTHER FUSELAGE – the ATR is not the only fuselage on planet earth nor is it the roomiest, lightest or most aerodynamic .. I’ve worked on all of them .. pick a Bombardier CRJ .. pick a Bombardier Q400 .. design a brand new fuselage and I will build it .. pick an MD80 .. pick a Bombardier A220 .. pick a C130 Hercules .. once we have all the kinks worked THEN we can concentrate on Commercial financial viability .. a USED/already designed fuselage that is SIMILAR to what we want – it doesnt need to be exact – that has the internal capacity we need for the Hydrogen systems – is GOOD ENOUGH for flight testing .. it will either PROVE or DISPROVE Bjorn’s software modeling .. if it is too large or too small or too heavy or has too much drag or not enough lift or doesn’t have the heat vent in proper place we can deal with that in testing and MAKE THE NECESSARY CHANGES in the software modelling when we get ready to .. wait for it .. TRANSFER THE TECHNOLOGY .. I am NOT talking a production airplane .. I am talking a TEST MULE – every single airplane in production started out as an EXPERIMENTAL AIRCRAFT – and the FAA will REQUIRE IT before the first production airplane ever flies its first minute .. so will Canada, Europe, Asia, Africa, and the Pacific Nations .. NOTHING will come out of a computer model .. the “software” has to be converted to HARDWARE – and then TESTED IN FLIGHT .. and that is EXACTLY what I was talking about .. I NEVER once said “take off” and ignore the software .. Brad Hartliep. A&P IA Pilot Engineer ..

  5. I need to understand why Bjorn sees superconductive motor and cabling as viable / big gain. I can not see it. If you optimise inverter/cabling/electric machine, you can achieve over 90% efficiency. The motor can run over 95% electrical efficiency if well designed and correct inverter control algorithms are applied (this is key). You run them at almost constant RPM and control power with pitch. The weight, complexity and maintenance costs of superconductive components are just not worth a 5% improvement. The key to these systems is to cool them down to superconductivity in the factory and keep them cooled enough during operation. That means at the gate, parked in the hot sun, etc, you need to keep the cooling running. I just can not see that for aircraft. Once you lose superconductivity, they will need to go to the hangar to be cooled down again (you can not do that while operating, because once superconductivity is lost, internal heating due to internal electrical resistance shoots up). And it did not work for wind turbines and they operate 24/7 and are always connected to the grid to draw auxiliary power when needed.

    AMSC (American Superconductor) has been trying since 1987. They make great products for electrical power systems and have a top notch engineering team, but no current commercial products (as far as I know) of them are in the area of superconductivity. They just could not make it work in a way which is commercially viable.

    • I think the cooling comes from the LH2 which is stored in tanks ( at say -250C) and then phase changed to a gas as a fuel for the fuel cell.
      ‘With a superconducting motor and cabling system and where the LH2 cool sink is used efficiently to help the thermal management system keep the Balance of Plant, we have a 250nm range and can take 36 passengers, a 49% payload reduction.”

    • @ NdB

      “The three-year demonstrator project aims to show that an electric- or hybrid-electric propulsion system complemented by cryogenic and superconducting technologies can be more than 2 to 3 times lighter than a conventional system—through a reduction in cable weight and a limit of 30kW/kg in power electronics—without compromising a 97% powertrain efficiency.”

      • I am still not convinced. The Airbus link is to a technology demonstrator. It is far from commercially ready.

        Some of the claims sound impressive but it is also depending on how you package it: “half of the losses” sounds far more impressive than “improvement from 96% efficiency to 98%”. Reducing voltage below 500V sounds impressive, but there are SiC devices rated at 1300V and you can run them at 800V and have significant reliability margin. That said, there are many other devices which can not run at 20K – driver ICs are rated most likely -30C to 125C and getting them close to cryogenic systems is a problem. Power electronics / motor drivers can be integrated with the electric machines and save significant weight.

        If you have a fuel cell, that one will run internally at 800C. So you have a significant heat source close to cryogenics. If you run a battery, they need to be kept at 20C-40C or they lose significant efficiency.

        I believed that in the end, a superconductive solution is just too complex and a maintenance nightmare.

  6. A broad Dutch consortium of 17 participating organizations — including Fokker Aerospace, TU Delft, the Dutch government and the Royal Dutch Aerospace Center — are starting a hydrogen conversion program to produce a turboprop with 40-80 passengers and a range of 750km, to be flying in 2028. The plane will use electric motors and fuel cell-based propulsion.

    The original link is in Dutch. The second link is an English version.

  7. @ Bjorn
    Can you perhaps provide some ballpark figures for the weight that can be taken OUT of an ATR72 as part of a conversion to hydrogen propulsion?
    – Obviously, a certain number of passenger seats can be removed, together with overhead consoles for ventilation/lighting, etc.
    – One can also remove all the cabling/ducting associated with those removed passenger seats. Moreover, climate control / power supply units can be smaller, to cope with the smaller number of passengers.
    – No need for decorative sidewalls or ceiling either for the removed seating portion. And the wall insulation can also probably go.
    – What about the passenger floor? Surely the tanks and other equipment can be directly mounted onto the fuselage beams and/or cargo floor?
    – Then the wings: everything associated with fuel storage in the wings can go, including insulation, piping, pumps, valves and cabling.

    Any indications as to what sort of weight that all adds up to?

  8. -There is another route to zero emissions that is the Aluminium Air battery. The battery may achieve up to 2000WHr/Kg.
    -Another may be not quite zero emissions but would be close the closed brayton cycle using supercritical CO2 cycle or sCO2.
    -sC02 can operate as both a liquid and a gas within the cycle greatly reducing turbo machinery size. They can recover low grade waste heat of a open brauton cycle or operate directly. They intrinsically have multi fuel capability.

Leave a Reply

Your email address will not be published. Required fields are marked *