Bjorn’s Corner: Efficient systems


By Bjorn Fehrm

By Bjorn Fehrm

05 February 2016, © Leeham Co: In recent Corners, we looked into technologies which have made the new breed of airliners more efficient.

We’ve talked about how new engines can raise efficiency by about 15% and how aerodynamic improvements, like more efficient split winglets, can add another 1%-2% over single blade winglets. We have also looked into modern ways to manufacture the more resilient and lighter composites structures that designers want to use to increase aircraft efficiency.

There is one area which we have not covered: the aircraft’s systems and how these can be made more efficient. An improved system architecture can add the efficiency improvement of a split winglet. So let’s have a look at the trends in aircraft systems.

We start this week with power distribution.

Power to those that have a need

A civil airliner has three ways to deliver the power that the aircraft’s different systems need.  It can be hydraulic, electrical or bleed air based systems that take power from the engines and route it around the aircraft. It is important how the power gets to its users. Why is easy to understand when one observes what happens to the aircraft’s engines efficiency when power gets zapped to feed the aircraft’s system.

At normal cruise conditions, the engine fuel consumption is raised by around 5% because of the power needed to drive things like movables, lightning and air condition. A 5% efficiency loss is major and any potential for improvements in how power is distributed and consumed will be welcome.

Let’s go through the different ways and see what gets done.


Each engine has one or two hydraulic pumps which supply power to the aircraft’s hydraulic systems (an aircraft normally has two or three independent hydraulic systems, and sometimes four, for safety reasons). The hydraulic systems have recently raised pressures from 3000 PSI to 5000 PSI to reduce pumping losses (less fluid needs to be moved, so less frictional loss) and increase power density.  The high power density makes hydraulics suitable to power the aircraft’s movable control surfaces (stabiliser, rudder, ailerons, spoilers), landing gear and brakes with comparatively small and light servos.

Further improvements in efficiency is more to do with simplifying the way system redundancy is achieved than fundamental changes in how the hydraulics system works on aircraft. More and more electrically driven extra pumps or servos remove the need for redundant extra hydraulic systems for the aircraft (on top of the normal two). It thereby lowers the aircraft’s weight.

Bleed air

Another system with high power density is bleed air from the engine’s compressors.  This bleed air gets cooled and regulated to 200-250°C and 275kPa/40 PSI in the engine pylon’s pre-cooler. It is then routed to users around the aircraft, Figure 1.

Bleed architecture

Figure 1. Classical mixed bleed and electrical architecture with pre-cooler in pylon. Source: Boeing

While bleed air is a simple way to get compressed air to the aircraft’s air conditioning system or heat to de-ice the wings, it requires cooling away energy in the pylon pre-cooler and it means routing 250°C air around the aircraft. Any leaks can destroy electrical cabling or even weaken aircraft structures. Therefore leak detection systems are required and different shut-off or rerouting valves need to be operated by the pilots should a leak occur.

More electrical system

Boeing’s 787 was the first aircraft project to say “there is a better way of doing this; we replace the bleed power with electrical power.” It is simpler and more efficient, Figure 2.

Figure 2. More electric architecture with high powered electrical distribution of energy. Source: Boeing

Boeing claims that the more electrical distribution of power is up to 3% more efficient than a mixed bleed and electrical system. Aircraft designers who have stayed with the classical mixed system say the gains are less but don’t dispute there are gains.

The reasons they haven’t jumped on the more electrical bandwagon is because there are problems before there are gains. Every major change in an architecture has a learning curve and the reliability problems seen during initial operation of the 787 has in no small part been caused by the system changes caused by a more electrical architecture.

Electrically driven air conditioning compressors have failed and the massive power electronics required have given trouble. The power electronics is needed because classical aircraft electrical systems had the main power, 115V 400Hz AC, created by constant speed Integrated Drive Generators (IDG). These delivered 400Hz power regardless of engine RPM.

Such IDGs are expensive, heavy and require maintenance. Therefore, modern aircraft, including the 787, bolt AC generators straight to the engines gearbox and out comes varying frequency 235V power.

Systems that don’t care about frequency consume this power directly, like de-ice mats or heaters in galleys. Others that need the stable 400Hz get that supplied via solid state power converters that make both 115V 400Hz and 28V DC from the AC power.  For classical mixed architectures with AC generators, these converters handle reasonable power levels, with each engine generator delivering around 150 KVA and part of this is converted.

The generators of a 787 delivers up to seven times that, or 1,000KVA, and the conversion electronics are now housed in water-cooled large racks. These new conversion racks have created problems and many user systems that were air powered but now changed to electrical power have added their initial problems.


Evolution in aircraft systems goes in steps. There is always an advantage to stay with a known system and be second with a new system; one does not have to go through the learnings the first mover endures. The more electrical architecture of the 787 has created a major part of the problems the aircraft has endured.  A well-known example is the Lithium Ion batteries.

Where other programs could change to the older NiCad batteries, Boeing was forced to make the Lithium Ion batteries work. The more electrical architecture changed the aircraft’s brakes from hydraulic brakes to electrically operated ones. The safe stopping of an aircraft with no engines running then requires the type of power curve that only a Lithium Ion battery could deliver (within a reasonable size).

The more efficient more electrical architecture has had its gremlins. But there is no doubt it is the way to go. An efficiency gain of just 1% makes it interesting and Boeing claims more. It is just a matter for other aircraft projects to pick the time when they think the technology is ripe and then get off the fence.

33 Comments on “Bjorn’s Corner: Efficient systems

  1. today progress is quite often driven by mass market products.
    Converter efficiency gains are today driven by photovoltaics
    and they make continuous progress having passed the 95% mark significantly. Obviously anything towards 100% gets increasingly difficult.
    Airliners are in this context a niche application and (should be) conservative in choice.

    IMHO Boeing entered this field one generation too early
    when converters had substantial cooling requirements requiring liquid assisted heat transfer and strong cabinet cooling. ( This aggravated electrical problems from condensation )

    Battery / Systems choices:
    There is a “good design” principle around:
    In interfacing systems the source should overprovide and the sink should underdemand as layed out.
    A “perfect match” is the foundation for endless problems.

  2. Some systems that moved to electrical on the 787 is hard to understand, like the Air conditioning. Having cabin compressors with an efficincy less than the Engine World class compressors and still needing an air cysle machine to get the right temperature and pressure seems cumbersome especially since you still have hot air offtakes from the engines for anti-ice. The hot air systems with High stage valve and PRSOV’s has been maintenence intensive, still it is no reason for giving them up, just do better engineering. But other selections might be brilliant as brakes with built in anti-skid logic.

  3. Thanks Bjorn.

    Just to clarify – when you say that electric systems can save 1-3%, does that mean fuel, COC, DOC, or something else?

    • Also – given the risk/reward profile of electric systems, would you expect Boeing’s MOM/NMA to follow the 787? You seem to imply that Boeing may be far enough down the learning curve now to use electric on its next clean-sheet project, even if it perhaps wasn’t worth it on the 787.

  4. Nice summary, Bjorn
    As mentioned above, the current killer for “more electric instead of pneumatic” is the power electronics, but I’d say it’s the power controllers as well as converters. The problem being that they are expensive, relatively heavy and generate concentrated heat deep inside the pressurised fuselage. That heat needs to be extracted and dumped overboard by a cooling system. The SFC gains due to power transmission efficiency are lost through these other cost drivers. Might change when thermal losses go down

    More generally, one has to consider that systems are involved in a variety of aircraft lifecycle costs : SFC as you mention, but also
    – weight (20 to 25% of totsl weight IIRC)
    – development costs (complexity driving testing needs)
    – procurement costs of sophisticated parts,
    – assembly costs for the myriad of equipment/ducts/pipes/harnesses,
    – and maintenance costs of these tricky systems

    Gaining on one aspect usually means losing something elsewhere

    • Thanks Airmagnac for chipping in with the details re the other side of the coin. It will be interesting to see who is next in going more electrical and what detail solutions they choose.

      • Is any of that generated heat turned into useful?

        I would think at altitude rather than need air conditioning you need to add heat to the system.

        In my field we have made the conversion form all pneumatic to all electoral. The electrics are very costly and far harder to trouble shoot.

        I liked our hybrid system where we used electronics retrofit to control the existing pneumatics. That worked very well.

        I have read the power distribution breakdown on the various 787s system, pretty amazing but also horribly convoluted.

        The one advantage would seem to be easier routing of the power lines vs the various pneumatic tubes.

        • The heat might be used as you say. Re the electronic control of a classical system, the A330neo and 737 MAX does just that, it converts a pneumatically controlled system to an electronically controlled one. This increases the efficiency marginally but it shall save on maintenance actions and therefore costs.

          • Bjorn:

            Is there any approach that assesses the whole of each system?

            Back to the pneumatics pipes and required routing and spaces vs electrical wires and all the other pluses and minus (weight of the components?)

            I never gave it any thought but do they use a pneumatic starter on a jet engine or is it airflow through the engine that spins it for firing up?

            I also scratch my head at 3% possible savings vs all the cost of the new components including the starter/generators (I have worked with those on small generators, they work well but only have seen it used on very small gen sets)

            The biggest reason they dropped off use of the hybrid system in my work is pretty easy to tap a power sources in a building and they prefer that to running a pneumatic supply system (and compressor for power) through the building.

            On the o the other hand that compressed power is already on the aircraft.

            I do continue to be stunned that they made it work and generally it seems to work well and pretty reliably.

          • This what Boeing says about 787 jet engine starting:
            “The 787’s engine-start and APU-start functions are performed by extensions of the method that has been successfully used for the APU in the Next-Generation 737 airplane family. In this method, the generators are run as synchronous starting motors with the starting process being controlled by start converters. ”

            “Unlike the air turbine engine starters in the traditional architecture that are not used while the respective engines are not running, the start converters will be used after the respective engine is started. The engine- and APU-start converters will function as the motor controller for cabin pressurization compressor motors.”

    • “I’d say it’s the power controllers as well as converters.”

      There is no real difference between controllers and converters ( nowadays ). All are essentially VFD like devices.

      No rotating converters (Motor-Generator-Sets) around anymore.

      • Actually there are, still considered the best source of GPU power by one of the GPU mfgs.

        Its neat, clean and simple. Also close to zero maint (change the bearings every 100k)

        On the other hand the GPUs we got from Europe blows capacitor every 4 to 9k amongst an occasional other fault.

        Of course a MG set weighs a whole lot more so for an aircraft is logical, but they do sill live on and are admired by some.

        • “.. from Europe blows capacitor every 4 to 9k amongst ..”

          That tends to be due to grid quality being well below specced for the device.
          With the rise of indeterminately switching on/off wind and solar power generation I start to see similar probs in the vicinity here.

          • Kind of the same issue we have with the contactors.

            Built to a lower spec than NEMA and far more failures.

          • Nobody in Europe would build to NEMA spec.
            DIN and or EN is what you get here.
            Same with stuff required to be UL certified.

            you have different specs that overlap but are not identical. If you order the wrong parts you will have problems.
            ( looks like a bit of market protection to me )

      • It was just to clarify a point by Björn about heat generating power equipment. These are not only used as gateways between say, the 270VDC and 115VAC, but also to chop/modulate the power signal going into the electric motors and actuators.

        • What I wrote.
          converters between system busses aren’t really all that different from variable frequency controllers/converters for driving multiphase (a)synchronous electric motors.
          And the same group of tech advances determines their efficiency.

  5. This piece contains enough material that it needs to be read twice. In addition to efficient systems, it touches on the subject of incremental vs innovative product development. Often, despite rigorous methods to reduce it, strong personalities on the development team prevail, and introduce risk. I always thought it was a form of intellectual arrogance. Sounds like a lot of that may have been present in the 787 program.

    • Frankly I don’t think it was arrogance, given a clean sheet design they went with what looks to be the trend (rightly or wrongly)

      It does seem to be that they did not make the trade off studies between the two system.

      I do think one key item that is maybe impossible to assess but extremely relevant is the route savings.

      The pneumatic piping has a fixed sized per the pressure and length of service, it can’t bend too much and has to fit in between wing structure and the fuselage structure.

      Everything else has to route around it. I also suspect areas have to be designed for the pipe size and re-enforced to accommodate it.

      electrical route on the other hand is a lot more flexible, power density is much higher (smaller pipes as it were) and you can bend it as much (pretty well) as you need to.

      So it may have benefits in the other areas that make things better.

      Question I have is that ALL translated into the saving or is it really impossible to calculate?

    • Engine and nacelle deice is the same as on any other turbo fan engine.
      Just no bleed air into the pylon and beyond.

      • Dont forget the engine itself needs bleed air too for its internal processes, and the extra generators dont create power all on their own.
        One of the extra advantages with electrical powered cabin a/c units is the ability to use variable speed ( in the domestic space its marketed as inverter) or even variable heating for de-icing.
        Even the 787 APU is a ‘no bleed’, just generating power.

  6. Another ‘efficient’ system is E-taxiing.

    There was many talk about how efficient e-taxiing would be but I still miss an airline to introduce this system on large scales.

  7. 5% (different aircraft, I know) Airbus wouldn’t just leave that on the table would they? With electrical architecture they knew Boeing were committed and that if they didn’t follow and got it wrong they won’t be able to change their minds. Split wingtips would seem to be an even easier choice, too proud to use something that they didn’t think of?Aviation Partners asking too much to for a licence? What was the result of the litigation?

    • “Aviation Partners litigation”

      IMU the patent was voided. Too much prior art around.

      split wingtips do have higher cost.
      You have to make the gains work for the cost incurred and savings provided.
      Boeing at the time of introduction on the NG needed the winglets dearly to be competitive to Airbus.
      The sharklets for Airbus made sense after fuel prices had risen enough to make the extra cost acceptable in view of fuel saved. The MAX needs that little bit extra from the splits to close up to the A320NEO while being ahead all the time 🙂

      Then one has to keep in mind that A320 and 737NG wing designs are different enough to also influence usefulness of the various wingtip solutions.

      • Yep, they went with the crank wing end on the P-8 but the situation with the profile of flight on the 737MAX was different and they went with the split (though I still think the Scimitar winglet is a lot cooler). Seems odd it was not AP they went with but……

        All new wing per the 777, 747 and 787 they used the crank wing end.

  8. The older the aircraft, the more system upgrades it needs to keep up.

    If one of the OE’s sees it’s aircraft is good enough to dominate the segment, further expensive system upgrades can be avoided/ delayed.

    Commonality is important for airlines.

    • Yep, same old balancing act of what pays returns and what does not.

      Everyone takes their best guess, usually a lot more surprised than expected when you switch but management will never admit that (bonuses and careers involved)

    • Yes thats true, Southwest has been a brake on development on the 737 for many years as they highly prized commonality almost to a fetish, luckily many other airlines can make big orders as well.

Leave a Reply

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