Bjorn’s Corner: The challenges of hydrogen. Part 30. Integrated nacelles.

By Bjorn Fehrm

March 26, 2021, ©. Leeham News: This week, we look at combining the propulsion and hydrogen tank in an integrated nacelle as Airbus proposes in Figure 1.

Airbus calls it its “pod” solution. What are the advantages, and what challenges does it present?

Figure 1. Airbus concept for a turboprop with integrated nacelles. Source: Airbus.

Combine hydrogen tank and propulsion in a pod

Airbus describes its pod solution as a serial fuel cell solution, where the pod contains:

  • A propeller
  • Electric motors
  • Fuel cells
  • Power electronics
  • LH2 tank
  • A cooling system
  • A set of auxiliary equipment

Does it work? Can a typical turboprop use the pods as a propulsion solution?

Let’s do a reality check of the type we did before. The turboprop is described as a sub-100 seater with six pods.

We use the propulsion values from an ATR 72-600 80 seater to get our base data for the check. The gas turbine engines develop 4MW for takeoff, and the fuel needed on a maximum range route of 800nm is 2,000kg with 500kg reserves, occupying a volume of 3.2m3.

Tanks in pods

If we convert this to hydrogen, we need 850kg LH2 that occupies 12.6m3. The tanks to store the LH2 weigh 2,400kg (assuming a 35% gravimetric efficiency).

For a carbon fuel to LH2 transfer, we have 2,500kg fuel with an assumed 500kg weight for tank sealing, pumps, valves, and tubes versus 850kg LH2 and 2,400kg tanks. As these are in the pods, their pumps/valves/tubes can be set at 250kg. This includes a filling tube to all the pods.

The carbon fuel total is 3t versus an LH2 total of 3.5t. Weight-wise we are in an acceptable range.

Volume-wise we have 3.2m3 versus 12.6m3 plus the tank walls space, say 15m3 in total. It’s a five times increase, but it can be handled. The nacelles of a DH 8-400 are around 8m3 each, and while large, they don’t cause problems for the aircraft (they house the main landing gear, thus the size).

A six pod, four pod, or even two pod solution is doable from an LH2 tank aspect.

Podded propulsion

It’s more problematic when we check the proposed fuel cell propulsion. Let’s first conclude, if we convert the Pratt&Whitney PW127 turboprop engines to hydrogen, it works. The PW127s weigh 500kg each and occupy the forward part of a gas turbine + LH2 tank pod.

The fuel cell solution runs into the same problem we saw last time we calculated a turboprop using a serial fuel cell concept. The fuel cells for the 4MW propulsion weigh 2,100kg, the motors 300kg, and the inverters 200kg. We have a total of 2.6t instead of 1t.

An added weight of 1.6t for the propulsion adds to the 500kg for the LH2 tankage. A weight increase of 2.1t instead of 0.5t is problematic for an aircraft type with an empty weight in the 13 tonnes bracket. From a volume perspective, it works. The fuel cells with motors, inverters, etc., occupy around 1m3.


The Airbus pod concept has several attractions (Airbus has applied for a patent for the pod). It stores the LH2 close to the propulsive motors, and it offers the LH2 tanks a round shape.

The tanks can also be placed close to the center of gravity of the aircraft. This eliminates a center of gravity shift as the LH2 is consumed.

Placing the weight of the LH2 with tanks on the wings where the aerodynamic lift attacks is advantageous; it lowers the wing’s bending moment at the wing roots.

The aerodynamic drag increase of an enlarged engine nacelle is not dramatic. As long as the speeds are far from compressive flow (and they are for a turboprop), the air curves around objects with ease (the idea that frontal area is a significant drag cause is false, it’s not as long as the speed is below M0.6). The dominant drag increase comes from the increase in wetted area from the pods, not their shape (air friction drag is the dominant drag of an airliner).

20 Comments on “Bjorn’s Corner: The challenges of hydrogen. Part 30. Integrated nacelles.

  1. Placing the LH2 tank pods c.g. at center of lift at 1/4 wing chord will effect the c.g. as fuel is consumed moving c.g. forward increasing trim drag. Hence placing the LH2 pods at the aircraft c.g. will not shift c.g. as fuel is consumed. Designing a new turboprop aircraft with todays materials opens the possibility for slender wings with better glide ratio +30, using this you can reduce power requirements and still have the same take off distance, it might be slow but the ATR’s are slow and sell well.

    • It should have been CG and not aero center, I realized this, changed and then saw your answer. Good catch.

      Re wider wing, yes but everything else being equal it will increase the weight. The present turboprop wings have a high aspect ratio already, 12 for the ATR 72 and 12.8 for DH 8-400. You have limited margin before you run into roll clearance problems etc.

  2. Has the effect of going from 2 engines in total to 2 or 3 each side. Clearly the power required for a twin is enough for one engine to continue takeoff with one engine out. That all changes with 6 engines as one engine out means the power needed is , at a first estimate, spread over 5 other engines. I imagine Airbus has enough information to only need diminutive engines in the LH2 version, say 300kW when converting a 72 seater. Maybe 400kW for the 100 seater?
    We are talking some like the tiny RR500 TP

    • Look at MagniX Magni500, 560 kW / 750 shp each. Having LH2 available you have coolant for superconducting engines in the same pod for even more light weight powerful engines. Having it all in the same pod makes for easy systems testing and pretty easy upgrades of aircrafts bolting a new certified Mk (n+1) version under the wing.

      • Yes, cryogenically cooled engines and cooled inverter and fuel cells are simpler to achieve. The pod installation has a lot going for it. The only problem is the high weight of the fuel cell. If you go hydrogen gas turbine until the fuel cells have improved it’s an elegant solution. Yes, you still have (reduced) NOx, but you improve the situation immensely. Sometimes you reach your ideal state in two steps.

  3. If three engines are placed close together I can see EASA demanding 2 engine out continuous take-off and initial climb

    Combined that with a reasonable MTOW and the typical thrust/shp required, for that MTOW. Divide by 4 and you have the required shp per engine.

    An engine significant less than 500kg PW127 I assume, making everything (wing, pods, etc) lighter.

    A rough estimation of 1000shp TP per engine seems feasible (4000shp left when 2 engines fail at V1) . A full ATR72 has 2500 shp asymetric to clear the fence at the end of the runway..

    Easy replacement of an engine, LH2 tank for inspection, maintenance seems attractive from an operation standpoint. .

    The LH2 tanks out and away off the passenger cabin can also be good thing.

    I think I agree with Bjorn this approach has better feasibility than large tanks in the fuselage assumed on earlier LH2 concepts.

    • There is also an advantage in flexibility if they built it in; need >550nm range, just remove 2 pods. Need >270nm range remove 4 pods. LHR-AMS is one of the busiest routes with just 231nm, 2 pods would be perfect for that route. This is of course assuming that 2 pods will deliver enough thrust.

      • The frequency of flights and variable pricing can fill the flights rather than ‘adding or removing pods’.
        Ask Ryanair how that works

    • One advantage with cylindrical slender tanks under the wings is that they can take internal pressure really well and are pretty streamlined allowing them to grow in length.

      Interesting comment above of just installing enough “propulsion tank/engine pods” for the mission and sold pax tickets. The airline would then have a “bomber crew” driving out and installing as many LH2 filled tanks of correct size needed from the “freezer” and hang another pair of pods with engines for higher MTOW requirement missions.

      Still a risk that is will be SAF fuel and everything will look and smell the same.

  4. Interesting corner. I bet using electric and spreading the required thrust over so many propellers will also create a pathway to reduced noice and vibration as compared to existing turboprops.

    • Wont be electric, as shown in previous stories. Batteries are too heavy and to create electricity on board still requires a turbine.
      The propellers are the ‘noisy/vibration’ part and having electric motors driving propellers wont change that

      • Uhhh, the AirBus proposal that is the focus of this corner is for hydrogen fuels cells driving electric motors. Bjorn’s analysis shows the fuel cell to be heavy but not “absolutely non-starter” heavy like batteries.

        As for my point; several smaller and more lightly loaded disks with lower tip speeds may well result in less noise and vibration.

        • Yes, if you increase prop size and reduce rpm you get something similar to Lockheed YO-3 “Quiet star”. With large sailplane wings like the NASA/Boeing Truss Braced Aircraft you are not as weight sensitive.
          Most gliders have the ability to carry water ballast. The sole reason for carrying water ballast is to increase the cross country speed on a task. Water ballast achieves this by increasing the wing loading of the glider. A glider with a higher wing loading has the same polar curve one with a lower wing loading but the polar curve moves sideways along the higher speed range. This means a high wing loading gives the glider the same sink rate but at a higher cruising speed. For example, a glider with no water ballast might have a sink rate 0f 3 knots at 60 knots cruising speed. That same glider with full water would achieve the same sink rate but now at 70 knots. This in turn translates to a higher cross country speed in cruise.

        • “AirBus proposal that is the focus of this corner is for hydrogen fuels cells driving electric motors”

          The focus of Bjorns story is LH2 for turbine/propeller combination.
          There is a paragraph only of fuel cells but seen as too heavy, The propellers remain for either.
          “It’s more problematic when we check the proposed fuel cell propulsion”

  5. > same sink rate but now at 70 knots
    I think you intended to say same glide ratio. It is the L/D diagram that is shifted sideways. The polar diagram actually shifts AND drops with heavier aircraft having a greater sink rate.

    Power on a glider comes from potential energy and if the sink rate were to remain the say the implications is it takes no more power to fly the heavier glider faster than it takes to fly the lighter one slower. Which would be great but unfortunately is not the case.

    • Yes, CL/CD
      The best L/D ratio is equal for different wing loadings, but is occuring at different speeds – the higher the load, the higher speed. The minimum/stall speed is also higher for higher loads.

  6. I think that if the integrated nacelles approach is pursued, that could lead to an entirely new look at the wing shape, construction en dimensions. It wouldn’t have to function as a fuel tank anymore, but loads, weights, flight requirements would be entirely different form current wing designs.

    • I was wondering about that. How much of the root thickness is to accommodate fuel and how much to handle the bending moment.

      I was also wondering if electric driven prop still need to be constant velocity. Electric motors have great tourque curve so they don’t need to be restricted to a narrow rpm range.

Leave a Reply

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