Bjorn’s Corner: Engine efficiency

By Bjorn Fehrm18 September 2015, ©. Leeham Co: The debate around the market’s two single aisle combatants is quite heated, with fans of the one side saying “the limited space for a high bypass engine on the 737 MAX will cripple it forever” and the other side saying “the tighter design of the 737 will make it highly competitive against the A320neo, it is the A320 which has a weight and size problem”.

One of the arguments is that each inch of engine fan diameter brings 0.5% in increased propulsive efficiency. Therefore the A320 with up to 81 inches fans will win against the 737 MAX, which has a 69 inch fan. Having all the tools to check out if this is really the truth, I fed our airplane model with all the facts and looked at the result. It’s not so easy, guys…

Engine efficiency

In our series Fundamentals of aircraft performance, we described the principal physics behind engine thrust in Part 2 of the series. It was described as:

Engine thrust = air mass moved * air mass over-speed

The turbofan engine is an air pump and it gets its thrust by pushing air out the back faster than it enters at the front. Air is quite heavy (1.2kg/m3 at sea level, one-third to one-fourth of that at altitude), and when 200kg of air gets accelerated to 100m/s over-speed by our high bypass engine, we get our cruise thrust of 20,000N or 4500lbf per engine for an A320/737 aircraft.

The engines propulsive efficiency is made up of:

2 * aircraft speed / aircraft speed + engine jet speed

The aircraft travels at M 0.78, which is around 250m/s. It means we have 2*250/250+(250+100), which is 500/600 = 0.833 or 83.3% propulsive efficiency. If we want to increase this efficiency by 0.5%, we need to decrease the engine jet speed with 4m/s to 250+96 m/s.

Note that we have defined how we get increased propulsive efficiency and By Pass Ratio (BPR) is nowhere in sight! Propulsive efficiency is 100% coupled to the over-speed of the engine’s air-mass flowing out the back (which is called its specific thrust by the experts). The BPR comes in way later and as a by-product of regaining thrust when the over-speed of the air goes down.

Now we have an engine with 83.8% propulsive efficiency but it only generates 200*96=19,200N or 4,317lbf of thrust. To get back to 4,500lbf, we have to increase the air mass with 8.5kg/s. The way to do that is to increase the fan area so more air passes the fan.

If our mantra of 0.5% efficiency increase per inch of fan should work, we should only need to increase the fan area with one inch diameter to get this increase in air mass flow (everything else being constant). In practice, we need a bit more. The one inch buys something like 0.4% propulsive efficiency increase. Finally it is this increase of air mass through the fan to regain our thrust that increases our fan size and mass flow and therefore BPR.

Drawbacks of increased BPR

Engine manufacturers have known the drawbacks of increased propulsive efficiency since the invention of the jet engine. As shown, it demands more air through the fan and therefore a larger fan and fan case. This increases engine weight and the engine’s and nacelle’s diameter. So while we have an increase in the engine’s efficiency, we have a decrease in the aircraft’s efficiency. What counts in the end is the total airplane efficiency increase.

Let’s use our example to understand these effects. We have an A320/737 style aircraft and equip it with two different engines, one an 81 inch engine and one with 91 inches. The latter has 5% better Specific Fuel Consumption (SFC) as per our mantra.

The first thing that happens in that the 5% lower SFC buys us an increase in max range from 3500nm to 3730nm. Then we start to include the drag of two nacelles, which are 10 inches larger. This cuts our range gain from 230nm to 160nm. Then we increase the aircrafts weight with 2*200kg, a low weight increase for engines and nacelles which are 10 inches larger. Now we are down to 110nm range increase. This means our engine gain of 0.5% gets more than halved on the aircraft level.


One shall always be suspicious of “simple marketing truths,” especially when they are formulated by the gaining party in a competitive situation. By a check with the rather simple fundamentals of aircraft turbofan engines, we can understand what brings an increase in engine propulsive efficiency and what affect it has on aircraft level efficiency.

In our example, we see that an increase of fan diameter enables a higher propulsive efficiency (by virtue of lower specific thrust that we can compensate with more air mass). But we also see that the engine efficiency increase gets halved by aircraft level effects.

This does not mean that lower specific thrust with increases in fan flow and thereby BPR is a negative. If done correctly, it has a positive effect. It is just not the Holy Grail that marketing mantras portrays it to be.

103 Comments on “Bjorn’s Corner: Engine efficiency

  1. One has to note that the net gain is still positive ( and significant ). And this projects into new/improved designs that have no/low secondary limitations. Thus Boeing arguing for “smaller is better” is obviously the tail wagging the dog.
    In this context: why do the current CFM56 engines for A320 and 737, though differing in BPR(and sfc), show about the same weight?

    • IF bigger is better was the way to go, then engine manufactures will always put the biggest fan size that the aircraft can accommodate. Since that is not the case, therefore finding the sweet spot that Boeing has found for the 737 is a more sensible approach.
      It is also common sense that an 81 inch engine will burn less fuel than one with 91 inches on the same trip if they were both mounted on the same aircraft type.

      • CFMI/GE put more technology into the -7B compared to the -5B helping Boeing. The -5B got some -7B technology with time just too keep up with the V2500-A5. Now it is a different game as the PW1100G with its slow turning big fan has a SFC advantage and CFMI has to pull all the stops to keep up, a bit depending on the big PW1100G nacelle induced interference drag as installed and PWA traditional engine problems they solve by time(black oil, HPC blade tip curs and surges, VSV hysteresis, Hot structures cracking). Boeing really has to convince GE to release all the latest GE9X stuff into only the LEAP-1B to keep up. It is a dangerous game and Boeing need some DoD support slowing down the latest engine technology into the -1A.

  2. Nice article,Bjorn.I would be interested to know why the b737 has been so competitive,at least until now.Acording to Leeham ,it’s lease rates are significantly more than the a320.Is the weight advantage due to the smaller fuselage diameter or regulation changes?If it’s regulations,has the extra weight saved any lives?I haven’t ever really noticed the difference in seat width when traveling shorthaul.Also, I am a bit confused by the evacuation regulations.I read in the aftermath of the B777 fire ,that planes are only tested at certification ,but not when reconfigured .I am a bit sceptical about the B777 high density narrow isle arrangement .

    • Grubbie, I fully agree with you on high density planes and their potential to cost lives. When you add extra rows and push seat pitches closer, it stands to reason that evacuation will be compromised, especially with a 777, given the number of seats.
      It is another example of airline greed and what they see as “managed risks”.

    • Emergency evacuation tests will indicate what the maximum number of passengers a plane may carry. If an airline’s configuration goes above the number indicated in Type Certificate, a new emergency evacuation test must be carried out.

      I do not know if that has ever been done before.

      One question I have about high density, low pitch configurations.
      If the exit door rows need to have a minimum pitch to allow egress out of the aircraft, why do not all the rows have to have the same minimum pitch?
      After all, they are also egress routes, just to a smaller degree.

      • If the exit door rows need to have a minimum pitch to allow egress out of the aircraft, why do not all the rows have to have the same minimum pitch?”

        Egress out of a door for all in contrast
        to egress into the aisles for max 3 persons.

        The area in front of the door is needed as a staging area
        to “flush^Hslide out” i.e. you have an FA or an assigned helper standing there ( and needing room )

        The formal evacuation test provides a common metric to measure arrangements ( for comparison and for go/nogo decision ). With prepared testers it can only test the topological arrangement in the plane. obviously. But that is what gets certified. How an airline handles the finer aspects of human horde behavior …
        Reality obviously will be different. But reality never provides for normalized repeatable testing 😉

        • Offer prizes for testers who can get their trolley suitcases out of the overhead lockers and down the slides…..

  3. I have always thought Boeing should pitch/optimise the 737 as a short haul plane for 500-1000 mile, max 2000, using lower thrust and MTOWs. Turn the aircraft´s weakness, that it was originally designed as a smaller aircraft, into a strength, I am sure it could have been made into an A320 beater on a 500 mile sector if Boeing had aimed it there.

    • I think the 737MAX is a A320 beater on a 500 mile sector by a fair margin already.

    • Are you thinking of something like a B737 400? Was nt that’s what it was? So the A320 had transcon range and Boeing had to compete Either that or buy a B757 which did both very successfully…….! AJK

  4. It might be interesting to inject a bit of gas turbine engine history here. The idea of increasing propulsive efficiency by using a fan on a pure turbojet and converting it to a turbofan was suggested (patented?) by none other than Whittle, one of the two gas turbine inventors. The first implementation of this was by P&W JT3D (based on its first 2-spool military turbojet J57), with a more radical GE rear-fan engine failing to get any market traction. This is what led to market domination by P&W until the 1980’s. But the interesting aspect is that aircraft engineers overestimated the installation penalties such as higher drag and weight, and for a long time, BPR was kept down to a value of around 1.0. Only when the mistake was realized, BPR started to increase to around 4 to 5. Then the C-5A program forced engine makers to consider even higher BPR and here we are with BPRs of around 10 and likely to go higher. How much higher, we don’t know and it all depends on the net decrease in the fuel burn, accounting for the increased inlet and nacelle drags and the increased weight of the engines.

    When Airbus built the A320, it made a deliberate decision to use a wider fuselage, quite aware of the fact that that decision would increase the aircraft empty weight and also the fuselage drag. But they felt it was worth it. Irrespective of the debate that goes on about the seat width etc., it is quite clear that the market has endorsed the Airbus approach, since the Airbus family has managed to intrude into this lucrative sector and now has begun to overtake Boeing with more orders for the A320-neo family as a whole compared to the B737-MAX family. Even with its weight penalty! The difference in block fuel burn between A320 and B737-800 is perhaps 1-2%. But the B737-800 has an 8% advantage simply because of more seats (162 vs 150) and that is what they have been emphasizing.

    No matter what the spin is, it is clear that Boeing would have loved it, if it could hang engines with fans bigger than the 69+ inch ones that it can, because of the 1950’s B707 legacy fuselage and the short landing gear. The time to change that was when the NG series was introduced. Now it is too late and the only way out is a clean sheet replacement.

    Look at the engine installation. B737 is one of the very few aircraft with the top of the nacelles very close to the wing itself, whereas almost every other aircraft has engines hung significantly below. It is amazing that Boeing engineers have squeezed out so much out of the 1950’s legacy 737, in spite of the severe limitations imposed by the short landing gear.

    Ultimately, it does not matter whether the rule of thumb is 0.5% per inch or 0.2% per inch. As long as there is a net, meaningful decrease in overall fuel burn, larger and larger fans (higher and higher BPRs) are going to be employed, if need be (as RR is planning to do also) by going to the GTF architecture. A basic design that permits that is likely to succeed than one that does not.

    • IMU core efficiency ( or better lack thereof ) limits useful increases of BPR. ( There is an RR pdf around that covers early FAN designs )

    • Would you happen to know what the advantage is to hanging engines lower from the wings?

      • Less wing interference.

        You can fit gapless “continuous” flaps that come with significant aero advantages. ( no thrust gate necessary )
        and no broken line along the leading edge either.

        The pylon is structurally “easier”. No forces carried all over the place.
        The moment arm of the engines that introduces torque into the wingspar(s) can be much shorter.

        Moving the engines forward places them more in harms way when landing with no/colapsed nosegear. Rare event. probably the least important variable.

  5. I think it is irrelevant to bring into the discussion the market share between a320 and the 737.
    Stating that ab has been successful because the market has endorsed its choice of engine size is way off the mark.
    Airlines decisions to purchase ab or boeing planes is not solely based on the engine’s propulsive efficiency.
    I will bet that the bean counters who decide whether to buy boeing or ab have no clue about the engine propulsive efficiency and care less about which engine has a bigger fan size.
    There are ab operators who have switched to boeing and also boeing operators who have switched to ab.
    That is to say that engine propulsive efficiency is the last thing that is in the mind of decision makers.
    Both aircraft types are very well engineered but the 737 has the upper hand because it can carry more Pax and thus bring in more revenue per flight.
    But again that is not the only factor considered by the airlines.
    It is no secrete that the main determinant during acquisition of aircraft is the monetary value of the deal that you get from the EOM.
    The same goes for engine manufacturers. When a 787 operator has to choose between RR or GE engine, the last thing they think about engine propulsive efficiency or engine fan size. It’s the value for money that you get out of the deal is the biggest determinant.
    My next flight will be on a B737. Have a good day gentlemen.

    • Me thinks you are very very wrong.

      Engines are not the ONLY aspect, but they are an important one.

      • Your comment makes me very very right as you have not demonstrated anything to the contrary of what I said.

        • You said the bean counters do not know fan size or fuel efficiency.

          I will bet my bottom dollar that the people that decide know exactly who has what fan size, how much it weighs, what the fuel efficiency is and considering its the single biggest cost to run thing they better.

          Yes you factor in the aircraft cost, maintenance but the engine is the single biggest issues.

    • Well, well. I did not suggest propulsive efficiency was the only consideration. I made it clear that it is the block fuel burn. Also the fuel cost is just one factor in a sale, since other operational costs come into play. Capital costs are also important and that is why the discount rates the airframer offers can tip the scale. Classic example is the A330-300, on which Airbus can offer larger discounts, whereas Boeing cannot match those discounts on the B787 and this has tipped the scale in favor of Airbus on occasion.

      But Bjorn’s point was: It is not just the BPR and/or the fan size. It is what happens to the aircraft fuel burn, when that engine is installed on the aircraft. I concur.

      • Well, it seems to have worked for Boeing and the 787 : Selling a wonderplane,
        developed in no time at no cost at a price
        well below the level that the established competition product could match.
        ( I am regularly flabbergasted when some posters spout a simplistic worldview were the US wins on technical merits ( and if not it was a pinch of bad luck ) and everybody else around the world works on undercutting prices and pure luck ). Manifest Destiny Semantics.

        • Both Boeing and Airbus make excellent aircraft! 🙂

          So the technical difference is often small. Nevertheless, it is those small differences that tip the scales often!

          • That is quite a noteworthy comment you made:

            Design differences aren’t binary 0 or 1.
            The delta quite often is in the lower single digit range.

      • I think the discussion here is about whether the fan size is all that matters in engine efficiency but your comments are drifting toward why one EOM sells more planes than the other.
        Whether AB can offer more discount where boeing cannot is definitely not related in anyway to their engine sizes.
        You are comparing 2 projects at different life cycle.
        Boeing could also offer very big discounts on a 767 Freighter if it was pitched against a a330 neo freighter.

        • The difference between the two freighters is Boeing has to sell 767 aircraft or freighters in order to keep the 767 line partly alive for the tanker. Airbus on the other side has no need to compete against the few 767 freighters Boeing sold because the A330 is still an attractive aircraft for many airlines. This is due to an attractive price…

      • True. This point is often lost. Airlines often compensate for a low load factor by carrying revenue loads in the cargo bay (in addition to the baggage of the passengers). However, that is only possible if the belly cargo space is enough and designed to accommodate modern containers. So aircraft designed with these points in mind fare better than those that were not! Thanks for pointing that out. For a really good description of how it all works, see Bjorn’s posting on the matter.

  6. F is the force required to push the aircraft through the air.

    Parasite drag + lift induced drag = -F

    Newton 2nd:
    The vector sum of the external forces F on an object is equal to the mass m of that object multiplied by the acceleration vector a of the object.

    F=m*a (mass times acceleration, a= dv/dt). Take dt=1 (or any other number).

    If a small fan moves 1 unit air and a bigger fan 1.2 units of air to produce the same F, the a required (acceleration) of the bigger fan can be less if F remains equal for both :1 than a=F/m = 1/1.2= .83.

    Kinetic energy (1/2 mv^2) required to speed up the air for the small fan = 1/2 * 1 * 1^2 =.50.

    Kinetic energy (.5 mv^2) required to speed up the air for the bigger fan = 1/2 * 1.2 * .83^2 =.41

    The bigger fan needs (1/2-.41)/ 1/2 * 100 = 18% less kinetic energy.

    If we assume the rest of the engine is the same, which is relevant in the LEAP-A vs -B case, the engine that takes significant more air through it’s bigger fan and accelerates it less to come to the same propulsive force is significantly more efficient (and silent).

    That is why Boeing has been investing so much to gain tenths of inches for the LEAP-B to start with.

    “Both aircraft types are very well engineered but the 737 has the upper hand because it can carry more Pax and thus bring in more revenue per flight.”

    A lot of assumptions must ne in there there 😉

    • I don’t see where you mentioned that if you suck more air you will also need a little more fuel to keep the balance of air-fuel mixture.
      Your analysis is not quite balanced.

  7. The 80 inch Pratt is an entirely different animal not comparable. The Leap-A has an 8 inch bigger fan then the Leap-B.

    “The first thing that happens in that the 5% lower SFC buys us an increase in max range from 3500nm to 3730nm. Then we start to include the drag of two nacelles, which are 10 inches larger. This cuts our range gain from 230nm to 160nm. Then we increase the aircrafts weight with 2*200kg, a low weight increase for engines and nacelles which are 10 inches larger. Now we are down to 110nm range increase. This means our engine gain of 0.5% gets more than halved on the aircraft level.”

    Some people could think unsubstantiated assumptions and handpick numbers and short cuts are selected to reach a predefined conclusion.

    It would be interesting to take the well known CFM56-5B and CFM56-7B as reference. Different dimensions, weights -> sfc are documented. (3 sources to make sure). The bigger fanned 5B ‘s always had a better sfc.

    • Keesje,

      LEAP-1B is not the same core as the LEAP-1A. The LEAP-1B core and engine is redesigned for the smaller thrust and fan.

      “Some people could think unsubstantiated assumptions and handpick numbers and short cuts are selected to reach a predefined conclusion.”

      Are you suggesting I made these numbers up? I have no agenda in this game, just want to put a little more perspective into the debate when people pedale half substantiated truths.

      Re the CFM56, you are missing the point of the whole article. It is not if a bigger fan/higher BPR gives better SFC or not, everything else being equal (it does). It is to show there is a trade vs installation effects.

      • I take it they are not two clean sheet cores. Is the core optimally sized for 1A, 1B, or did they split the difference exactly in the middle when choosing a core diameter size?

          • That is good to know. If all sizes weren’t chosen independently, it would be hard to choose various internal dimensions without favoring one engine or the other.

          • GE could not afford any slack in the designs.

            P&W offering a better (probably) longer term solution and Boeing counting on a bit lighter and smaller engine to be efficient enough to compete with the A320 series.

  8. Bjorn, it would be interesting to see your analysis on Trent XWB-97 vs GE9x as it’s a similar argument there of SFC vs Aircraft fuel burn.

    It would be an interesting discussion around the weight increase of fan diameter on the larger fans on these aircraft, as they are now pushing the boundaries of how big a fan can go whilst still maintaining competitive weight, aero performance and certification requirements of blade containment and bird strikes.

    • The TXWB 97 and GE9X are different generations of turbofans and they have very different fan designs. The TXWB has a ultimate version of Rolls Ti fans before they go CFRP, the GE9X GE’s third generation of CFRP fan and fan case.

      To compare these engines would be a bit apples and oranges as one is an OPR 50 design with BPR 9 with EIS 2017 and the other OPR 60 design with BPR 12 and EIS 2020.

      • Bjorn,

        Regardless, it would be nice to know the TSFC of the two engines, both installed and uninstalled. Uninstalled values around 0.51 and 0.49 for RR Trent-XWB97 and GE9X? You should be able to say what they are using GasTurb12! Forget the installed values for the time being. Or they proprietary to Leeham?

      • Appreciate your position but the apples and oranges comparison is what a lot of airlines will need to do. While the EIS dates are different, it’s a similar comparison between technology levels vs engine architecture as on the NEOs, albeit 3 shaft vs 2 shaft and high use of composites and ceramics vs more traditional materials. Can they achieve a similar level of fuel burn at an aircraft level with different SFCs? I read this as the main discussion point for your interesting article.

    • Looking at the ratio of MTOW to fan size, if it relates to thrust to fan size, the A359 seems to have a better fan ratio than the 789 or A3510.

      • MTOW does not directly relate to engine thrust and therefore fan size.

        The ultimate thrust of a twin airline is to a large degree dictated by the one engine out thrust required at the safety speed, V2. This is in turn to a large extent dictated by induced drag ie MTOW vs span. In fact the aircraft OEMs stipulate the thrust they require from the engine OEMs at V2 (close to M 0.25) and not at static sea level.

        This is why span and aspect ratio is so important, it helps at take-off, climb, cruise and descent. Of these the highest induced drag proportion as a part of total drag is take-off.

  9. Bjorn, excuse me for suggesting you made up numbers, I just thought they popped up without sources, matching conclusions I saw in the “New single aisles on home stretch” Topic.

    Regarding the LEAP-A and LEAP-B installation effects, I heard rumors about the weight and drag figures on the NEO and MAX.

    It would be interesting to know how much weight the MAX and NEO gained over their NG and CEO predecessors. That must be clear within a few percent at this stage.

      • Well said Bjorn,
        It seems some readers are dead set on the preposition that a bigger fan engine can always have a better SFC on an aircraft. Somehow, the guys at International Aero Engine pull the hell of a trick when their engine (V2500) with a smaller fan (65″) will give a better SFC than the CFM56-5B (fan with 68″) on the A320 family of aircraft.

        • Yep…I’ve avoided this conversation because there are people here who will disregard the analysis Bjorn put up and say “Whatever! The bigger fan always wins! Always!”

          Nice info, Bjorn. Thanks for the education

        • ALL ELSE being the same. Let us see why is the 65″ fan V2500 better than 68″ fan CFM-56? Can you tell me? Of course, I do know the answer.

      • I did read his Bio when he came on the site (as a regular I think some reports before) . Its probably the most impressive one I have ever read. Amazing breath in technical background and personal accomplishments as a flier and inventor as well.

        I would never disagree on his facts. There is an occasional typo that he correct immediately when he finds or pointed out (and says so and why)

        We are phenomenally fortunate to have someone with his capability even talking to us, let alone polite and patient responses with questions that are pretty basic and obvious if you know the field.

  10. Bjorn-

    Wouldn’t propulsive efficiency trade off against core efficiency – or require a bigger, heavier core? At constant thrust, a bigger fan means some compression stages – at least the fan itself and the first-spool compressor – are rotating more slowly. Thus wouldn’t the core need more stages to achieve equal OPR?

    As an aside – does a bigger heavier core lose more energy to friction? Or does its slower rotation speed mean less friction loss? Assume equal OPR…

    • Yes and no. Today’s higher BPR is to a large part the result of smaller cores delivering more shaft hp to the fan. This is directly proportional to the Turbine Entry Temperatures (TET), the higher the TET, the more hp from the turbines. A smaller core means higher BPR for the same size engine which enable lower fan pressure ratio which gives lower specific thrust = higher propulsive efficiency.

      When you model engines you typically assume mechanical losses from bearings etc of less than 1%. Losses follow RPM and residual disk forces in the engine.

  11. Another weight increase mandated by a larger engine diameter is due to the longer landing gear legs required. Why isn’t this factored into your analysis?

    • Probably because here it works in reverse.
      the MAX with the smaller Fan has significant work done on the gear. The NEO is unchanged.

      Any it was reported some time ago that the increase in deadweight is higher for the MAX. ( and runway performance further degraded by this)

  12. The fact that 320’s have had taller gear logically results in a weight penalty, though the gear didn’t need to be further extended for this engine change. The fact that Boeing increased nose gear length and weight doesn’t change the fact that its main gear is shorter, and thus lighter, than the Airbus.

  13. Probably more than offset by the less efficient mounting position of the 737MAX engine.

  14. A number of Operators have ordered both MAX and NEO (whereof eg Norwegian) and hopefully will bring their pinch of salt to this discussion by rendering public the results of both types on a given citypair operated at same operating rules … we may learn something ?

  15. “the other side saying “the tighter design of the 737 will make it highly competitive against the A320neo, it is the A320 which has a weight and size problem”.

    I didn’t run into this view very often.

    If public perception would be it’s a draw between the two aircraft, that would be a win for Boeing. Airlines don’t go for perceptions.

    • Are Boeing going to do the thin insulation/side wall thing with the 737?New quiet engines could help with this.

      • There is a reason for thick and thin isolation. The price for thinner material with the same values is higher. Especially not concerning heat rather noise.

  16. Kant is correct that the turbofan was patented by Whittle, but I believe the first turbofan/bypass engine in service was the Rolls-Royce Conway. The numbers for thrust increase per inch of fan diameter are clearly dependent on the size of the engine and are an over simplification. The 737 was designed for the JT8D with a low bypass ratio and frontal diameter, resulting in the engine being mounted below the wing, with a short undercarriage and acceptable ground clearance. Boeing had great difficulty installing a higher bypass engine on the -300, finishing up with the engine mounted much further forward. The A320 from the outset was designed for a bypass ratio around 5, resulting in a longer undercarriage. As one of the oldest inhabitants it is useful to point out that Douglas overbuilt the undercarriage on the DC8, which eventually permitted the major fuselage stretch on the Series 60 , which Boeing could not match on the 707. Fan diameter, while important, is only part of the equation. The PW GTF for neo has a much larger diameter than the LEAP1A but the performance for both appear similar. I have seen no credible numbers on either engine in the open literature , certainly not in the manufacturer’s web sites. If you track the split in orders for the neo engines you will find CFM slightly ahead of PW, with about a third of selections still to be made ( see ).

    • The engine selection waters are probably muddied by GECAS ( resp. engine _and_ airframe financing )

      For similar overall sfc the bLeap needs significantly higher thermal efficiency against the aLeap ( due to propulsion efficiency getting partly stunted by integration losses less than the difference in engine propulsive efficiency.

      With further improvements for the GTF already waiting in the aisles I expect the final numbers after NEO EIS to give some hint.
      finally: A gtf for the MAX really makes less sense than for the NEO. GTF advantages materialize with a (very) large fan. 737 constraints make that nigh impossible.

  17. As more cross section of air moved per amount of thrust required helps engine efficiency,
    CSeries 73″^2/21K = 253
    A320 81″^2/27K = 243
    have a high ratio of cross section to thrust. Although maybe the average thrust over a flight profile is a more important number.

  18. Thank you bjorn for not going for the simplification/ truism route. By illustrating the interplay between core/ fan etc etc you give a far more realistic understanding of the sophistication of the turbines as they currently stand.

    The linear relationships often touted re inches and efficiency always seemed a bit too good to be true. I am very interested in the current engine competition as at both the smaller (leap vs gif) and at the top (ge90 vs rr Trent) there seem a wide range of design solutions that threaten to become even wider going forward.

  19. I just happened to come across this article. I think it is a must-read for us all debating this BPR issue. The RR plot included there along with the equation for the overall efficiency of an engine is quite illuminating, as I will explain below.

    Ignoring the installation and integration penalties for the moment: The overall efficiency of an engine all by itself depends on two efficiencies: 1. Thermodynamic and 2. Propulsive. The product of the two is the overall efficiency and this is what is important for the thrust specific fuel consumption (TSFC, abbreviated as SFC). If you look at the equation Aspire Aviation article presents, it can be rewritten as follows: SFC – U0/(etao*hpr), where U0 is the flight speed (usually about 250 m/s at M 0.85 at FL 340 say), eta0 is the overall efficiency of the engine (thermal to propulsive power) and hpr is the fuel lower heating value (about 43 MJ/kg). If you look at the limit shown in the RR chart (with an open rotor, which has the potential for highest propulsive efficiency of 95% and stoichiometric combustion, which has the highest potential thermodynamic efficiency of 60%, to which both TET and OPR contribute), we get an eta0 of 0.57, 57%. This is the best we can do ever and is the HOLY GRAIL of engine makers. Substituting this into the above equation, we get a potentially attainable SFC of 0.36 in British units or 10.2 micrograms/s/N. This is what engine makers are shooting at. P&W is going the way of increasing propulsive efficiency with higher BPR and CFM is going the way of increasing thermodynamic (Brayton cycle) efficiency by the way of CMCs and less cooling air in their GTF and LEAP-1 engines, respectively. If we could ever build a 100% efficiency engine (a thought experiment, impossible in practice), we can realize an SFC of 0.205. This is the absolute theoretical limit (unattainable) for jet fuel for an airliner traveling at M 0.85 in the lower stratosphere! Right now, it is about 0.50 for engines on the B787 and A350 and could go down to 0.48-0.49 (?) for GE9X.

    P&W has experience in building high performance engines for F35 and F22. If it wants to, it can transition that high temperature technology (ceramics) to civilian engines. But what clinches its case is the GTF architecture, which lets the fan rotate at 1/3rd speed than the LPT, and this is a HUGE game changer. Why I won’t go into here.

    So far so good. We know what an engine by itself can do. But when we install it on an aircraft, we incur penalties. One is due to the unavoidable intake and nacelle drag. This is usually the responsibility of the nacelle/airframe maker and it is an engine/airframe integration issue. The rule of thumb, for whatever it is worth: SFCI = SFCU*[C1+C2*(BPR-1)], where SFCI is installed value, SFCU is uninstalled value, C1 and C2 are constants. SFCU is uninstalled. SFCI is installed and is obviously larger. C1 is typically 1.03-1.04, C2 is typically about 0.04- 0.05. The precise values are known only to the airframers. Thus a BPR of 11 would cause about 8% increase in SFC due to installation penalties.

    Next comes the increased weight. This is also simple to account. If we assume a constant L/D, one percent increase in weight causes 1% increase in lift during cruise and hence 1% increase in drag. For the same distance traversed at the same speed, the fuel burn rate is 1% higher.

    Now we have all the elements that go into the fuel burn issue and hence the fuel cost considered. Next come other operating costs, crew, navigation/landing etc. On top of that we have capital costs. When oil prices are high, airlines tend to focus on fuel costs. However, lower acquisition/leasing costs can offset to some extent, higher fuel costs. This is where discounts matter. If development costs of your plane have already been reclaimed, the airframer can offer bigger discounts (A330-300-ceo). If you have overshot your budget and cannot really afford to give huge discounts (B787, $30 billion development cost). The discounts over list price are usually around 50%.

    Clearly, if we install more seats in the same aircraft, the fuel burn goes up slightly (because of increased weight similar to increased engine weight), but that is more than offset by the increased number of seats. For A320-ceo and B787-800, the fuel burn is supposedly similar (1-2% in favor of 737) but B787-800 has 8% more seats (162 vs 150). Therefore the specific fuel burn (SFB), which is computed in terms of liters/PAX/100 km is 8-10% lower for the Boeing aircraft! Boeing PR emphasizes that, whereas Airbus PR emphasizes fuel burn and comfort etc.

    • I forgot to add: The lure of GTF is that the fan tip Mach number can be brought down significantly, without compromising the LPC (booster) and LPT performances. Since the fan is the major contributor to overall noise during takeoff, this is a major advantage (e.g. Bombardier C Series vis-a-vis B737-MAX with LEAP-1B and A320-neo with LEAP-1A).

      Enough said. I will shut up and wait for any comments.

      • The “real” GTF lure is the unlinking of Fan tipspeed from turbine tipspeed.
        A large diameter fan requires low rpm which on the same rpm shaft requires a weighty large diameter turbine with high blade count.

      • Kant thank you very much for this excellent report technical ( like any other for that matter). For a management specialist like me, it is very informative . Also, I have a question for you : Can you repeat the same thing but playing the role of a buyer of aircraft ? Indeed, if you were a buyer of single-aisle aircraft , between a 737Max, a 320Neo and possible CS500 (with 165-180 seats) , given all the parameters covered, what would be your only choice ? And why ?

        • I could, CaptaineScarlet. But this is Bjorn’s column and hence he should be the one to answer you. I am quite sure he can do a better job than I can. Bjorn?

        • We do this kind of work for airlines and lessors as part of our consultancy practice. Here I can only say that the 737 MAX and A320neo are close enough in their key operational parameters that which aircraft is the best for an airline will depend a lot on the airline in question and its history (installed base) and priorities. Also the airframe and engine pricing will be important.

          Re a possible CS500, we would have to see what the spec looks like if and when it happens.

        • The PWA lower speed geared fan makes for reduce local fan supersonic losses and slow turning makes it possible to make it mainly out of cheaper and more precice gemotery aluminum.
          The A320 neo family now are over 63% market share, hence the shift from the A319/737-700 to A321/737-900 helps Airbus getting this market share as its A321 is more popular. Boeing must do its 727/757 replacement soon. Will the Open rotor Engines be ready soon for this or will Boeing remake the 767-2C and put GEnX-2B Engines and 787 systems, new APU and landing gears into it besides a new carbon wing? The USAF might pay for this for the later KC-45B?

    • “But what clinches its case is the GTF architecture, which lets the fan rotate at 1/3rd speed than the LPT, and this is a HUGE game changer. Why I won’t go into here.”

      I will, but only to cover the basic advantages for the benefit of the uninitiated:

      In the current GTF the LPT turns 2 1/2 times faster than a conventional engine. That allows the LPT to operate at a more optimal speed; which makes it lighter and more efficient, with less parts. In a conventional engine this is impossible because the fan is connected directly to the LPT; and it is physically not possible for the fan to turn faster, for it is already at the limit. But with the addition of a gearbox in between the two the fan speed can be reduced by a factor of three, which will make the fan actually turn slower than a conventional engine (in the same ratio of 2.5 to 3, or -17%). And the slower the fan turns the quieter it is. Also, because the fan turns slower its diameter can now be increased without exceeding its speed limit; which will yield a higher bypass ratio: 12/1 for the C Series and 11/1 for the neo.

      • Normand,

        Now that you have opened the proverbial box …

        Leonard Euler, a Swiss mathematician, who invented the symbols pi, e and i and formulated the most beautiful mathematical equation e^(i*pi) + 1 = 0, was a great scientist. In addition to formulating equations that modern high performance computers use to calculate the flow field around entire aircraft, called Euler’s equations, he also formulated the Euler Whirl equation that tells us how much power we can extract from a turbine rotor and how much power a compressor rotor can put into the fluid. Suffice it to say that the blade linear speed (equal to rotational speed and the radial distance) figures in prominently there. The higher the blade tip speed, the higher the rotor power. Initially, designers were vary of increasing blade tip speed beyond M of 1 because of losses caused by shock formation on blades. However, modern high pressure ratio compressors would not be possible without supersonic tip Mach numbers (around M 1.3?) and careful CFD analysis to MINIMIZE shock and other losses in blade passages.

        The moral of the story is that the blade tip speed must be as high as feasible to extract more power. So the rotational speed is important. GTF allows the LPT to rotate at 10,000 rpm (?) and so you can extract roughly 3 times the power than if you let it rotate at 3,000 rpm as dictated by the fan as in RR and GE designs. This means fewer stages for the same LPT power. The same holds for LPC, the booster on the low speed spool. It generates a stage pressure ratio of only 1.14 because it is rotating at 3,000 rpm dictated by the fan. Look at the GE90 series. They had to make LPC blade mean radius quite large even to achieve that (LPT is also large in diameter). Compare that to RR, which lets the booster (called IPC) rotate at a much higher rpm because of the 3-spool architecture. IPC is much more compact. Now, GTF allows high power input from LPC in a compact design, because it is rotating at 10,000 rpm also. All this means lesser number of LPC and LPT stages in a GTF and hence shorter/stiffer and lighter engine. Each 100 kg decrease in engine weight is equivalent to 2 additional revenue-paying PAX! More importantly, there are roughly 1,500 LESS blades (aerofoils) in GTF engines!

        That is because LPT has 4 less stages and LPC one less. Not only that. Since LPC performs as it should, HPC does not have to work as hard as GE HPTs do, generating pressure ratios of 23 in GEnx and 27 in GE9x. PW GTF has a 8 stage HPT compared to CFM’s 10 stage one.

        So it is not just BPR, fuel burn. It is also engine weight, stiffness and maintenance costs over the lifetime of the engine. Geared architecture is nothing new, but PW spent $1 billion and 20 years PERFECTING it. Once airlines see it in operation and realize how good it is …. Lighter, quieter, shorter, stiffer, less (?) maintenance! IMHO, it is the architecture of the future, especially if RR can figure out how to reverse the fan pitch (as in propellers) and eliminate heavy conventional turbofan thrust reversers.

        Mea culpa. In my earlier post I said
        SfcI = SFcU*(C1+C2*(BPR-1)), C1 ~ 1.04 but I had C2 wrong. It should be 0.004 to 0.005, not 0.04 to 0.05!

        Also Herb is right. Brits were the first ones to implement a turbofan on their Conway, which was a single spool engine with bleeding to avoid surge. But Conway did not become as popular as the two-spool P&W J57 (civilian version JT3C-7 and turbofan derivative JT3D-7) and single spool GE J79, each of which found novel ways to avoid dangerous surge in high pressure ratio (12 to 13) compressors of that time. It is JT3D that made P&W dominate civilian engine market till the 1980’s, when GE came back with CF-6 and eventually GE90 and took over the long-haul engine market along with RR, which went bankrupt developing their 3-spool architecture (RB211) in the 1970’s (for Lockheed-1011 Tristar) and had to be rescued by the British government. GE spent $1.5 billion dollars and 10 years developing GE90. Engine making is an expensive and arduous business.

        Enough said.

        • “IMHO, it is the architecture of the future.”

          I couldn’t agree more! Especially when it will have been successfully integrated with the RR three-spool architecture. I am glad to have opened the door for you because you have written a wonderful post.

        • I don’t follow my own dictum “Check before you hit return”. A few typos:

          1. should be “equal to product of the rotational speed and radial distance”
          2. GE HPTs should be GE HPCs!
          3. 8 stage HPC not HPT

  20. “This is usually the responsibility of the nacelle/airframe maker and it is an engine/airframe integration issue. The rule of thumb, for whatever it is worth: SFCI =..”

    I think to assume the MAX engine integration is similar to the the NEO’s, is a big assumption. They had little room to start with and pushed for tenths of inches.

    “Next comes the increased weight. ”
    Articles I saw state the MAX took a hit. They had to place the heavier engine further in front of the wing. Imagine a hard landing with that & you know where the weight is added.

    “, the fuel burn is supposedly similar (1-2% in favor of 737) but B737-800 has 8% more seats (162 vs 150).”

    A double count. In the articles that give the 737 a better fuel consumption, its per seat, including +12 seats for the 737-8. Which is only relevant if you can/ want to fill those extra seats. The smaller A320 sells beter than the 737-8.

    The sfc for the CFM56-5B has been better than the -7B in every overview I linked. Furthermore for some reason Boeing takes Sharklet-less A320’s but Wingletted NG (2-3% ?) as reference.

    For me, BPR & sfc doesn’t say everything, but enough in this case. Boeing has little other options than to say it’s far more complicated & throw in 100 more variables. I would too, in their situation. If you can’t convince them, ..

    Ideally Boeing should make the MAX’s cabin slightly wider, for 18 inch wide seats plus a wider aisle, make it pallet/ container capable, make cabin and cockpit more silent, standardize the cockpit with its bigger brothers, increase fan size up to 80 inch for better sfc/ lower noise, offer engine choice, introduce more composites and fly-by wire replacing mechanics, maybe do a -9 stretch MoM version with 4000NM range, go for smarter galley’s/lavs, move doors for more flexibility.. A 737NEO.

    I would not be surprised if Boeing comes up with a 3-3, 10% better ~A320 as NSA, like the Y2 became a 10% better ~A330.

    • Yes, and maybe every tenth of engine diam. and 7″ center of gravity shift was all necessary for the MAX9. Boeing should have optimized the engine around the MAX 7.3 and 7.7. Smaller engine and lighter frame to compete directly against the A320 with two sizes at 115′ and 125′ long.

    • That is a CLEAN SHEET design that is going to cost not less than $10 billion and I don’t know if Boeing is ready to spend that kind of money given that it is still in the hole to the tune of $30 billion (?) on the B787 and starting the expensive B777X as well. This much is clear. If it does not do a MOM, it is going to lose that segment to A321-neo-LR. IMHO for whetever it is worth.

      • afaics the MOM is Fud thrown against the A321(LR).
        As a project it will have a shorter lifetime than the NSA.
        Nobody found it worth waiting for another early promise late delivery offering against the NEO. NSA disappeared from public view, enter MAX from left.
        Problem in this new round is that Boeing has no real fall back path.
        MAXing the 767 into a competitive airframe probably is a bigger project than the 777x is. If it is possible at all.

        Finally a new airframe with a flat oval fuselage as floated will be at a structural disadvantage as that geometry introduces compression forces into the flooring. All the new materials don’t take kindly to compression forces.

        • THTA – two hundred twin aisle. 14′-2″ circular 2-2-2 cabin, 145′ long. Then 165′ stretch, next add long range wing, 185′ stretch.

      • “That is a CLEAN SHEET design that is going to cost not less than $10 billion”

        – I keep seeing such double digit billion dollar figures being tossed around as the cost for developing the clean sheet NSA but is it really that much?

        FWIW, Bombardier’s spending a about $5.4 billion according to 2015 figures including all the cost overruns, to develop the clean sheet CSeries.

        Yes, the CSeries is a bit smaller than what the NSA would be, but even so, its development costs can’t be less than half the cost of doing an NSA.

  21. Seems like fan area is very closely proportional to thrust going from the Leap 1b to the GE9x.

    69^2/28K=170, 133.5^2/105K=170

  22. Another consideration is the cost of the engine. (This was alluded to in some commentators reference to “acquisition cost”.)
    In general, for a given thrust, the bigger the diameter the more expensive the engine will be.
    If the fuel burn is within 1-2% at moderate ranges (and less at short range), the lower cost of the engine may make up for the higher fuel burn.
    (Of course total airplane cost also includes the cost of the airplane itself. I am not sure who as the advantage here, but the Boeing has a slightly smaller fuselage diameter and shorter landing gear, etc. so it might be cheaper to make.)

    • I would not expect the BLeap to be cheaper to produce than the ALeap. Higher techlevel like CMC and stuff comes at significant cost ( and also risk ).

  23. I think a 4% sfc advantage of the Leap -A versus the Leap-B is realistic. It was the case for the CFM56-5 & -7 too.

    CFM knows, but wants two happy customers, so won’t publish apple to apple numbers.

    The PW1100 is another story. I think CFM (GE, Safran) is speeding up gear research.

  24. It would be nice to get the 767 MOM Leeham out from behind the pay wall.

    While Winglets (scimitar of course!) and a new engine would work wonders, is it enough along with some other cleanup if possible.

    Can’t say without numbers and them I don’t have.

  25. The Conway was a two spool low bypass engine, I believe the BPR was about 0.25 : at that time the British design philosophy was to bury the engines in the wing ( Comet, Valiant,Victor,Vulcan). Passing the engine through the main wing spar restricted the diameter and BPR. The advent of pods on the B47 lifted this diameter limit and when PW converted the JT3 to the JT3D they were able to go to a BPR of about 1. The big increase in BPR came with the C5A, followed by 747, DC10 and L -1011.
    The first significant use of geared fans was due to either Lycoming or AiResearch around 1970. Lycoming used the T55 turboshaft with a new gearbox driving a fan instead of the helicopter rotor, resulting in the ALF502 which was used in the BAe 146 and the Canadair Challenger 601. The 601 installation was very troublesome and was soon dropped in place of the GE CF34. AiResearch developed the TFE731, which dominated the mid range of bizjets for many years. Over 12.000 of these entered service but it is worth noting that this was followed by the Honeywell AS900 which did not use a gearbox. By this time there must be a wealth of information on gearbox maintenance and reliability, but how much is in the open literature?
    In the GTF there is a saving in LPT weight, which is offset by the gearbox weight, so it is not obvious that there will be a decrease in weight. I suspect many airlines are still not convinced that the GTF is the way to go.
    Regarding the choice of twin spool or three spool, if both high BPR and high pressure ratio are required , the three spool is the only way to decouple the fan system from the compression system. RR have been successful with this through the RB211 and Trent ( originally the Trent was designated RB211-524L for Large) but in their proposed future technology UltraFan they intend to use a geared variable pitch fan. It appears that for very large engines of very high BPR the LPT weight becomes too high and a higher speed turbine with less sat ages becomes necessary.
    There is lots of OPINION ABOUT, let’s see some FACTS!

    • Herb, yes. You are right about why Conway was a low BPR turbofan. Fatal crashes of Comet did not help, although the culprit was the hull, not the engine.

      As for GTF, true it has been around as I said. The gear box mass does offset TO SOME EXTENT the decreased mass of LPT etc. But it is a net positive. Many airlines are still hesitant about GTF because of the lack of maintenance and dispatch reliability history unlike CFM engines. But the fact it has been selected to propel A320-neo family, Bombardier C-series, Mitsubishi MRJ, Irkut MC-21, Embraer E2 series speaks for itself. Also fully 1/3rd of the A320-neo orders have not settled on the engines yet. Are they waiting to see some GTF-related data, perhaps? It is also noteworthy that once it withdrew from IAE, RR is essentially out of the narrow-body engine market.

      Sure there is a BPR value beyond which GTF becomes more attractive, but that does not mean it can only be used then. I have always liked the RR 3-spool design, for obvious reasons. But I am glad it is going to a 2-spool GTF architecture for its Ultrafan.

      • “You are right about why Conway was a low BPR turbofan. Fatal crashes of Comet did not help, although the culprit was the hull, not the engine.”

        The Comet flew with the Rolls-Royce Avon, a long time before the Conway ever found its way on an aircraft; i.e., the Boeing 707, Douglas DC-8 and Vickers VC-10.

        • It is difficult to keep track of these old engines, especially with memory like mine. Thanks for pointing out the error, Normand.

          I think we have gone way off on a tangent from what Bjorn’s main message was, namely, it is not just the BPR.

    • “The first significant use of geared fans was due to either Lycoming or AiResearch around 1970.”

      The Garrett Air Research TFE731 came a number of years before the Lycoming ALF 502, and was much more successful commercially with more than 11,000 units sold.

    • “Regarding the choice of twin spool or three spool, if both high BPR and high pressure ratio are required , the three spool is the only way to decouple the fan system from the compression system.”

      Can anyone here explain to me why the GEnx-1B is much more successful with the 787 than the Rolls-Royce Trent 1000? When the Dreamliner was introduced I thought the Trent would have the upper hand because it is a three-spool engine and the the IPC would be the ideal power source to drive the ultra powerful starter-generators. In other words it would be more efficient than the GEnx. At least on paper. In my mind the three-spool engine is ideal for bleedless aircraft.

      Perhaps Bjorn could answer this burning question (burning as in fuel burn, of course). 🙂

      • Suggest the commercial clout of GECAS may be more important than any technical differences. RR advertising currently claims sfc advantage and I think the Trent has a better IFSD rate

  26. Pingback: Bjorn's corner; Turbofan engine challenges; Part 2 - Leeham News and Comment

Leave a Reply

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