Bjorn’s Corner: What did we learn in 2015?

By Bjorn Fehrm

By Bjorn Fehrm

08 January 2016, ©. Leeham Co: It’s the first Corner for the year and a look at 2015 as a year of technology advancements is due. 2015 will be remembered as the year when three clean sheet airliners passed important milestones. This will not happen for many years to come, so it will be worth to look at what they brought to world of aviation.

I’m thinking of Bombardier’s (BBD) CSeries getting certification for its first variant; the Mitsubishi MRJ doing its first flight’ and COMAC’s C919 being rolled out. Going forward, we will only have derivatives progressing through such milestones for years except for the roll-out of United Aircraft’s MS-21 single aisle airliner in 2016.

The Airbus A320neo was certified in 2015 and Boeing’s 737 MAX rolled out, but these are derivatives of in-service aircraft.

Embraer’s E-Jet E2 will roll out in February but this is a further development of today’s E-jet and Airbus A350-1000 is a variant of the in-service A350-900.

It will be a long time before we see so much new in a year, so it can be instructive to look at to what extent did these new aircraft bring the state of the art of airliners forward.

Airframe technology

The CSeries, MRJ and C919 are all similar in the overall airframe architecture. They are all classical tube with wings with the engines slung under the wings for highest aerodynamic and structural efficiency. The counterweight effect of the engines on the wings and the focusing of the propulsive and landing gear forces close to the wing’s center wingbox is an efficient way of building an airliner.

The fuselage is clean of forces except those from pressurization and the control forces from the empennage and nose gear. It was not long ago that we had engines in different locations, including at the root of the horizontal tailplane.

While the overall architecture of the aircraft is similar, the detail technology is not. None of the projects followed the lead from Boeing’s 787 and Airbus A350 of using Carbon Fibre Reinforced Plastic (CFRP) for the fuselage. Ten years ago this would have been unthinkable. 2005 was the year of the drug-like rush of the virtues of CFRP construction. It made things lighter, easier and therefore in the end cheaper to build and virtually maintenance free.

Since then a lot of practical experience has been gathered and the picture is more nuanced. While the initial promise was for a 20% lighter structure (that is the ideal gain for CFRP versus aluminium construction), it is today debated if there was any weight gain in making the core fuselage tube in CFRP.

While there might be gains for larger aircraft, they are harder to achieve for smaller types. The reason is that the CFRP fuselage skin of a smaller airliner gets so thin that is can’t stand up to everyday handling hazards. CFRP fibres are so strong in tension that the pressurization and bending forces for the fuselage can be countered by so few CFRP tape layers that the skin gets sensitive to forces from the side, such as hail, birds or sideways bumps from dropped tools, loose containers or catering trucks that don’t brake in time.

As a consequence, so much of the fuselage will be dimensioned from damage tolerance that a lot of the gains from CFRP’s strength gets lost. While there might be some weight gain for a larger dual aisle aircraft, the gain for smaller single aisles are dubious. As for ease of manufacturing, this seems today to be a pipe dream; Boeing is still wrestling with high manufacturing costs for the all-CFRP Dreamliner.

Airbus is keeping the cost of building the A350 to itself (the program accounting makes the Dreamliner build costs public whereas the standard accounting practice of Airbus does not reveal costs for building the A350). There has been no boasting about its all CFRP construction making it easier or cheaper to build.

This leaves maintenance as the final advantage for CFRP fuselages. CFRP allows fuselage tube inspections up to the 12 year heavy check to be made as external ultrasound de-lamination checks, instead of the internal inspections that require that seats and interior is taken out. So there are real benefits from CFRP design from a maintenance perspective.

The only problem is that BBD CSeries achieves the same 12 years heavy check schedule with its AlLi- based fuselage. BBD has furthered AlLi’s corrosion and fatigue resistance to the level where it has the same maintenance intervals as CFRP.

When one looks carefully at the CSeries, its AlLi fuselage seems to have gained the advantages of CFRP in most areas, yet stayed away from its unknowns in damage tolerance, electrical structural network (CFRP can’t work as a lightning protective cage nor as a return path for electrics without metal meshes added to the structure) and production costs. Its nominal 5% weight gain from making the aircraft’s fuselage tube in AlLi is what a CFRP based fuselage could have brought.

BBD chose CFRP for the empennage and wings, where CFRP offer real benefits. The flight loads demand skin thicknesses that make the surfaces stand wear and tear without extra weight for damage tolerance. Mitsubishi and COMAC stayed with classical aluminium wings but this was more to not wandering into the unknown with their clean sheet designs. It had more to do with prudence (it was a first ever for COMAC and Mitsubishi hasn’t made an aircraft for decades) than aluminium being a better choice for a wing for a 100 to 150 seater in 2015.

Both also stayed classical aluminium alloys for the fuselage rather than AlLi for the same reasons. Mitsubishi is venturing new territory only in the vertical tail wingbox; here it employs a novel “out of auto-clave” CFRP production method, something United Aircraft has announced as well for their up-and-coming MS-21 wing.


We have looked at the airframe technologies used in the clean sheet projects that passed major milestones during 2015. The common thread is: use CFRP where it has undoubted benefits (empennage, wings) if you know how to do it. Otherwise stay away and take the weight and maintenance hit. For the fuselage tube, make it with aluminium. If you have the time and knowledge, make it an AlLi design.

The reason is that CFRP does not offer any real benefits for the tube; there are simply too many other dimensioning requirements which are not playing to CFRP’s strengths. There are huge needs of electrical conductivity, yet CFRP cannot conduct. CFRP skins are not strong in the traverse direction, yet this is the attack angle of the catering truck.

So much for what happened on the airframe side; next week we will have a look at 2015 engine technology progress.

93 Comments on “Bjorn’s Corner: What did we learn in 2015?

  1. Hi Bjorn

    Interesting article (as always), one pedantic point, you continuously discuss weight ‘gain’ when it would make it easier to read if you said weight ‘saving’. Gain suggests an increase in weight

      • Bjorn:

        When I took German in high school we had a lot of interesting discussion about idiomatic expression and how they don’t translate.

        Not quite that but same thing, hard not to express in your roots.

        You do an incredible job and its a very miner item that is readily understood with a tad of thought.

  2. Hello Bjorn,

    Good old metal is not dead !

    Looks like Airbus is already on a 6/ 12 years cycle for 4C (intermediate D) / 8C (D check) checks on A330. Might still have some difference on the detailed tasks for the 4C. Boeing touted mainly “wide external structural inspection” pour the 4C at 6 years. Not sure for the A330.
    Intervals are quite similar, for sure, detailed content need a detailed dig in the MRBR…

    Bonne journée

    • I think we also have to realize the A330 MPD is on its n:th revision when they got up to these limits. The 787 and A350 has not even passed their first 4C/D yet, so there is no experience to say if it can be prolonged further. But be sure it will.

  3. Any idea Bjorn if Boeing’s switch to AlLi for the 777X fuselage will allow that tube to go for a 12 year check, a la C Series?

    • Boeing will not change their present range of Aluminium alloys for the 777 fuselage, AlLi will not be used.

  4. For some time now, I have been reading about dimples and possible a roughened skin to either reduce drag on the skin or to induce a fuselage friction that stops air from clinging so tightly that it increases drag. For these technologies to develop (dimples are concave circle in a skin like a golf ball) they would probably only be achieved on a metal skin. Roughened skins would probably need CFRP for maximum adhesion.
    It would be good to receive some thoughts on these advances on future designs.

    • As I recall the US did that to their boat on the Americas cup off Perth, won them the series.

      And hats off to the Aussies, that was the best darned Americas Cup ever, real racing in real seas.

      • More likely that fine grooves would be used on aircraft, more akin to that on shark skins. It would be done with adhesive film in selected areas. has been tested and does work at A320 scale. Not sure why its not more widely used.

        • Dirt in the things. Water easier as its cleaner (but you need to keep the hull clean but the green stuff take a lot longer to grow than a race does and they take the boats out of the water)

          I think Boeing is ditching the low drag tail because of dirt and bugs mucking it up.

  5. It seems the 787-8 and 787-9 are not lighter (OEW 120, 129t) then the metal A332 and A333 (OEW 119, 124t). Same engines (NEO), same dimensions. (page 2-2)

    Maybe Airbus will consider a simple A330 “-10” too, if CFRP isn’t all for everyone.

    I wonder if Boeing had known all they know today about the Dreamliner project in 2004, they would follow the same carbon strategy.

    Same for Airbus, if the had done the GenX A350 mk1 / NEO, rolling off the line from 2008, 10 per month..

    • The engines of the 787 are significant heavier than the A332/3 engines. A better comparison is to the A338/9 (about 5t gain). In that case, the 787 is about equal in weight or a little lighter but with much higher cababilities (range and/or payload). To subtract this would make the 787 lighter. But the difference is definitely not extreme for the 1st generation, yes. But looking at the A351 vs. the 77W (which are nearly identiacal in size AND cababilities) you get a better pictures which shows there are advantages. Just the weight loos is much less than the once thought xx% but more in the 4 … 8% range.

      • Dry weight Trent700: 4,785 kg
        Dry weight Trent1000: 5,765 kg

        I think for the A330NEO Airbus is shaving off further MTOW A343 dimensioned structure and looking for further composite quick winds. Not 5t gain by a long shot. maybe equal/or even less the CEO.

        • Airbus themself expects a SIGNIFICANT gain for the engines and the winglets as well as the reinforcements in the wingbox for the “bigger” wing. It may be slighly less than 5t, but at least it will bring the A339 to 789 levels.

          I’m to lazy to lock for it, but there was a power point from Airbus where they showes the gain and how it effects efficiency.

          [Edit:] I found something:

          It’s also the reason the A339 won’t be better than the A333 on short tripps (see Leehams analyses).

          • Little nit:
            they expect a +2% fuel impact from weight increases.
            ( + 1% for drag -4% for the changed wingtip devices )

          • @Keesje

            Straight from the horse’s mouth, I’ ve got the dry weights of the Trent 700 and Trent 1000 engines. They are 10,500 lbs (i.e. 4763 kg) and 11,924 lbs (i.e. 5409 kg), respectively.

            NB: Please do note that the number for the Trent-700 is without the RR provided nacelle. It may seem as if the extra weight of the nacelle is included in too many analyses that includes the Trent-700, further distorting the overall picture

          • @Uwe

            Yes, you’re of course right.

            FWIW, during the design phase of an airliner it’s known that an initial 1 kg increase in structural weight can increase the gross aircraft weight from 2 to 10 kg, and vice versa. Also, an increase in fuel consumption of 1 percent is typically the equivalent of an increase in the gross weight of between o.25 and o.75 percent, and vice versa. For wide-bodies, the fuel weight fraction is higher than on single aisles, while the OEW/MTOW fraction is typically slightly above 0.5.

            Now, the A330neo is already based on a mature airframe and it will have a wing that was originally designed to carry a maximum of 139000 litres of fuel for the A340-300 – an amount that will never be used by an A330neo. So, the wingbox is optimally too heavy for the A330 neo, and if was designed today it would most likely have a smaller wingbox (i.e. reduction in wingbox-size between the engines and smaller centre wing box)). The wing area, though, would, remain about the same

            So, an increase in the fuel consumption of one percent would, in a worst case scenario, increase the OEW by some 1.5 percent – or 3 percent increase in OEW if the fuel consumption increases by 2 percent. Therefore, a 3 percent increase in OEW of the A330-900 would result in a weight gain of about 3.75 tonnes – that is, if we are using the “worst case” scenario” as describe above.

          • The integral wing box fuel tank isnt a problem, as you still need long span wings for most efficient long range cruise. Normally the airlines, when they are going shorter range is utilise the ‘wing box’ fuel tanks but leave the centre fuel tank empty.
            The largest fuel tank capacity was for the 340-600 with 195,000L, The A330 neo models have about same capacity as the older 330-200 version, around 140,000L

          • A340NG has a different wing and wingbox.
            tankvolume thus does not compare.

            The A330/A340 family all have the same tankage provisions ( as fits long range requirements ). This IMU continues into the NEO upgrade ( See, it is a New Engines Only thing :-).
            Only difference until recently was that the A333 did not use its center wingbox as tankage. a dry space. With the recent MTOW markups to 242t it starts to make sense to use it.

          • Youre right, I forgot about the ‘wing insert’ they did for 345 and 346, a sort of long narrow triangular shape widening about mid chord.

          • @dukeofurl

            The centre wing box on the A330-300/A340-300 is the equivalent of 10 fuselage frames – or 0.533m* x 10 = 5.33m. The centre wing box on the A350 is the equivalent of 8 fuselage frames – or 0.635m** x 8 = 5.08m. Hence, the centre wing box of the former is slightly longer (in the x-axis) than than the latter. The centre wing box on the A333/A343 is also heavier than the centre wing box on the A350, partly due to the latter having a 50 percent CFRP content.

            Also, keep in mind that the wing of the A350 is about 20 percent larger in area than the A330 wing, and still the A330 centre wing box has a larger “footprint” in the centre fuselage than what’s the case for the A350 centre wing box. Ideally, therefore, the wing box could have been reduced by the equivalent of 2 frames in the centre wing box (x-axis) and with the forward spars moved further aft in the wing – i.e. in the inner wing area, between the centre wing box and the attachments points for the engines. The wing planform, though, would remain the same. Of course, this would be an expensive undertaking, but the volume of the wing would be more “right-sized” for the fuel volume requirements for A330neo, rather than the current one which was optimised for the A340-300. The weight would be significantly reduced for such an optimised A330neo wing box, especially if the smaller centre wing box would be re-designed and having 50 percent CFRP content. If you take a closer look at the wingboxes of the 777, 787, A350 and A380, they all have a pronounced kink in the forward spars inboard of the engines (i.e. inner engine on the A380), while the forward spar on the A330 wing is straight all the way to where it’s attached to the forward spar on the centre wing box. If the forward spar on the the inner wing of the A330neo wing would be re-designed with a similar kink, then you could reduce the “footprint” of the centre wing box by, say, 2 frames.

            The integral wing box fuel tank isnt a problem, as you still need long span wings for most efficient long range cruise. Normally the airlines, when they are going shorter range is utilise the ‘wing box’ fuel tanks but leave the centre fuel tank empty.

            *One fuselage frame on the A330 fuselage is 21″ long, or 0.533m.
            ** One fuselage frame on the A350 fuselage is 25″ long, or 0.635m.

          • Addendum

            That last paragraph in my comment above was apparently yours. 😉

    • I recall Airbus designers were reluctant to move to a CFRP fuselage for the A350 Mk I. Basically they did it for marketing reasons. Customer perception was that CFRP was better, so planes made in that material were easier to sell.

      A long time ago I read an article about some young engineers that visited Boeing, where Boeing admitted their spun barrel technique was an expensive way to create fuselage sections. They went for it because it allowed a relatively simple integration of sections made by inexperienced partners. And because they were using a technique that was already industrialized in producing racing yachts, so there would be a shorter route to market. Based on that, I suspect the A350 panels are cheaper to manufacture.

      • Correction:Airbus designers were reluctant to move to a CFRP fuselage for the A350 Mk II

      • Well, Boeing was driven by their own PR. Nobody would have taken their offer based on some well known technology.
        They had to use some new fangled material with “super whatnot” pedigree. And they had to “one up” every established technology to leverage all the announced superlatives in a marketable way.
        What remains is a lot smaller.

      • “They went for it because it allowed a relatively simple integration of sections made by inexperienced partners.”

        And it went so well that Boeing had to buy those partners to save the programme. 😉

        • They’ll never do it again either is my guess.
          ( Using barrel sections that is )

          My heating fuel tanks are done in barrel sections ( though not ‘C’FRP ).
          Well over 30 years old.
          Boeing spouted so much “tail wags dog” stuff to argue their case for those “innovations” … .

          • “They’ll never do it again either is my guess.
            ( Using barrel sections that is )”

            On paper the idea was good. But personally I always had problems with the industrial risks involved. More so than the performance advantages because of the weight savings, which I never really questioned, but which also remain to this day hard to evaluate.

          • I never understood why it was such a great idea to build the 787 fuselage sections the way you do composite mast manufacturing in the yachting industry, by first wrapping (laminating) the fabrics around a mandrel (male mold) and after curing, pulling out the mandrel to have the finished mast. The difference between a mast and a fuselage is that the former is not a semi-monocoque structure (i.e. including stringers and frames for stiffening purposes). A mast has a very thick skin, and one that’s been cured on a male mold (mandrel) is a barrel structure with no further need for strengthening and/or mechanical fasteners – unlike the 787 fuselage which still needs a lot of fasteners in order to attach the internal circumferential frames.

          • Normand: What’s funny is they nailed the risk, no real issues with the tech on the use of CFRP anywhere in the 787.

            Other issues yes, but that has given them no trouble (cost not a tech issue though totally relevant to the program return)

            What missing here is the modular nature of all of it and the so called snap together. Maybe has not worked the way they wanted but its a design factor.

            there is a lot less riveting and fastening involved as well. Some and some it seems to me.

            And yes Airbus did not have the tech ready so they tried to do what they knew how to do. Now Hazy is all excited about the plane old A330 with new engines (and while a nice setup fro the right range, its also not gone into orbit sales wise, niche player with Air Asia X a major supporter and they keep deferring A330s and then shifting to the A330NEO or as we say in the US, kicking the can down the road)

        • @TransWorld

          “Normand: What’s funny is they nailed the risk, no real issues with the tech on the use of CFRP anywhere in the 787.”

          When I said that “I always had problems with the industrial risks involved” I had in mind the problems they eventually faced trying to fit the first barrel sections together. It took a few ship sets before they could nail it down (the problem, not the barrel!). I also had genuine concerns with the sub-contracted work they had set up on the East Coast with inexperienced suppliers. The plane was going to be preassembled by these partners, taking lightly the fact that their inexperienced employees were not at the time at the same level Boeing’s one employees were. I take pride from the fact that I had anticipated all the trouble they faced down the road. But why did I see it all coming and not them? My answer was, and still is, that Boeing had been taken (run) over by a pack of computer hot shuts who thought they knew better than all the generations before them combined. They had entered the computer world at a young age and thought everything could be simulated numerically. Here is my message to them: We are not there yet boys!

          • Normand:

            I think you completely miss the upper management muck up, not young hot shuts.

            Boeing was in the throes of not even doing an aircraft, the new method was a ploy to get an aircraft (new) back into production and continue Boeing on a patch as an airplane manufacturer.

            Oversold, note how many managers quit right after the rollout of an empty aircraft , they saw it coming and did not want to get rolled over by the juggernaut.

            And again, the CRRP tech went extraordinarily well, what did not go well was the management end (production centers scattered all over the map but had nothing to do with the hot shuts, that was strictly and upper management decision to avoid investment cost 0and we have seen how well that did not work out.)

            As for assembly, there is always going to be problems and learning curve, nothing new, they seem to have it nailed down.

            And other than the battery, so far there are no show stoppers as noted by its now quietly going about its business with 300+ in service and you don’t hear anything about them these days.

          • “… its [787] now quietly going about its business with 300+ in service and you don’t hear anything about them these days.”

            The Dreamliner has something in common with CFRP: resilience.

          • If you go to avherald and look for B788/B789 there is this recent strange Jetstar incident and some engine shutdowns ( GenX predominantly ). No obvious sore thumbs around.

            “No icing” limitations on the GenX engines are still active.

      • @OV-099 I think Boeing adopted a technique for making yacht hulls, not masts.

        @Uwe. The article was interesting because I believe it got out before the CFRP story was taken over by Boeing PR. So it may be a pointer to the “real” thinking behind Boeing’s decisions.

        @Normand “They went for it because it allowed a relatively simple integration of sections made by inexperienced partners.” – And it went so well that Boeing had to buy those partners to save the programme.

        Ha Ha. True. But think of what would have happened if they had gone for a “complex” integration.

        • tickled pink but if your are referencing the article that was provided by keesje.

        • @FF

          No, the technique was adopted from composite mast making on a male mandrel. Making composite yacht hulls is still a labour intensive affair.

    • “Same for Airbus, if the had done the GenX A350 mk1 / NEO, rolling off the line from 2008, 10 per month..”

      No real gain.
      As happened they had 8..9/m A330 coming off the line
      and they now have a further type available that sits in the segment above the A330.

      • Airbus had a A333 sized aircraft too, the A350-800. But the airlines told them they liked a serious A330 upgrade even better. Lower range, lower payload but lower costs & complexity too.

        An A330-1000 wouldn’t be doing heavy Asia flights, but be a very good, affordable and profitable 5000NM aircraft. Keeping the payed for, popular A330 line going, freeing up A350 slots for airlines with bigger requirements. Just like the A330 NEO, that also shouldn’t exist 🙂

        • Hi Keesje

          That A330-1000X looks like it’s only been stretched aft of the wing. There are 18 windows between doors 1 and 2 on the A330 -300/-900 and so does your dash 1000. 😉

          BTW, I would assume that a stretched A330-1000X would have about 23/ 24 windows between doors 1 and 2.

          • It’s a 4.7m stretch, a little more in front then aft of the wing for cg reasons. Forget the windows 😉

            It would splash on top of the A350 family and “compete”. Reason to buy could be different range/capacity requirements, slot availability, purchase costs and A330 fleet commonality.

          • It still doesn’t look right, though, as it looks like a dash 900 ahead of the wing. 😉

            What you describe is a 9 frame stretch – 5 frames ahead of the wing and 4 frames aft of the wing.

    • A330-1000 at the same weights as the A330-900 seems like a good idea for Hawaii and Iceland flights.

  6. Bjorn, I would have liked to read more about the 777X and the choices that were made in regards to the empennage, wings and fuselage. I still don’t know if the 777X fuselage will be made of conventional aluminium or Al-Li. I do know that the wings will made of CFRP, but what about the empennage (vertical and horizontal)? And what should we expect from the 777X in terms of the material choices that were made versus a clean-sheet design?

    Thanks, Normand

      • Bjorn:

        do the CFRP single aisles fuselage sidewalls get thinner because it needs less structure to support less weight.

        A bit confused about that aspect.

        787 seems pretty impervious and having see the section in Everett its an amazingly thick wall.

        • The problem is another one. The dominant dimensioning stress in a fuselage tube is the cyclic pressurization (you dimension to the stress level that gets you fatigue life of 25-60k cycles, way below the materials mechanical capability).

          For a circular fuselage or a double bubble with the floor dividing the bubbles these forces are mainly tension in the skin. CFRP is directional in strength and very good in tension, not so good in compression or transverse loads.

          It is possible to get the weight saving for a CFRP skinned fuselage but the CFRP layers needed for that tension strenght will not create a thick enough skin to resist all the transverse loads that can occur (ramp rash, hail etc). When you increase the thickness to also resist these forces and you add the mesh for lightning/return path and you add the internal electrical return network and…. you have lost most of the point of a CFRP tube section, you don’t save any mass. The point is the tube section of an airliner has so many functions that are not playing to CFRPs strenght, it is full of stuff that need a conducting/shielding attachment and skin.

          The rest of the aircraft is well suited in most aspects for CFRP construction.

          For large aircraft a CFRP tube need enough skin thickness to resist the fatigue tension from the pressurization that it gets durable enough to be damage tolerant without extra layers and therefore mass.

          • I would have thought that the sidewalls would be thinner owing to the fact that the skin is sized for impact resistance and therefore structurally stronger than it would otherwise be resulting in the frames being slightly thinner, is this right?
            Another (fairly obvious when you think about it) reason wings work better is that you can run more of the fibres lengthways, while the fuselage gets its stiffness from its huge section. Cfrp works best for long thin things, wings, yacht Masts,etc.

  7. What is the weight savings from using AlLi vs conventional aluminum? I read somewhere the material costs more, but being able to use less of it keeps construction costs in line with conventional aluminum

    • IMU the per volume weight is slightly less.
      But the advertised gains are only available if you design for the advanced properties. Not by replacing Al sheet metal with Al-Li stuff 1:1.

      • So simply replacing metal on a hypothetical revamped (757, 767 that some argue for) wouldn’t yield any advantages because they aren’t designed for Al-Li?

  8. I always wonder what would have happened if Airbus persisted with the original concept for the A350. With 787 delays, I think they would have sold a BOATLOAD of them. So much for listening to your customers. I wonder if Lehey rubs that in to Udver Hazy

    • Mark, correct. But, they sold a boatload ofA330s too (800) since the launch of 787, without re-engining. Looking back, not upgrading the A330 then was ok. Did they foresee the 787 development would be a drama, or was it just plain luck?

      In 2004 Airbus engineers said Boeing had rushed through the technology, before it sufficiently matured.
      Everyone jumped on them, knowing they were just jealous and behind the curve.. They scheduled 8 yrs for the A350 development iso the 7e7’s 5 yrs.

      • Look at when the A380 bashing became really virulent.

        That should reference the growing realization the project was going pear shaped nicely. ~12..15 month ahead of the fake roll out ?

    • The A330 has done extraordinarily well since the launch of the 787. Now, Airbus also has a very good 9 abreast platform with the A350 and 8/10 abreast platform with the A380. In the future, I would not be surprised if Airbus launched an all new 10 abreast aircraft/platform and re-designed the main deck of a second generation A380 (i.e. re-profiled frames, slightly “higher” main deck floor etc.) in order to accommodate 11 abreast – with 18″ wide seats, 2″ armrests and 20″ aisles.

    • They had to do a bigger Mark II A350 because the 777-300ER slaughtered their A340-600 offering. If they had had a modern competitor to the 777, they would have stuck with the Mark I A350, I suspect.

      • The A350-900 is a modern competitor to the 777-200ER, while the A350-1000 is a modern competitor to the 777-300ER. The 777X/777-9 is a step-up in size, but will IMJ be vulnerable to either a super-stretched A350-1100X – featuring a 777X-sized wing, or an all new 10 abreast platform from Airbus.

        The Mk-I A350 was IMJ too much “new aircraft” and too little gains for the return of investment.

    • Mark,
      In my view (formed after flying a few journeys in the 787) Airbus has done very well to shelf the original A350 and increase the width. Airlines insist on putting 9 abreast seats into the B787, turning it into a horror after 3 hours in it. It just does not work. At 8 Abreast, the 787 will still be very competitive, but will be much more comfortable. A 9 abreast A350, with the additional 10 or so cm, just about is the minimum a self respective airline can offer customers for missions beyond 5 hours.

  9. I assume that Boeing will only move to Al-Li if required to meet specifications. the B777X is a tight design (not much room left for unexpected weight gains). The increase in window size and fuselage differential pressure will require a complete redesign anyways, so why not go Al-Li.

    • The original 777 came close to being made of Al-Li. But in the end Mullally decided against it because the material was not mature enough at the time. I think the 777 Classic is flying with only approximately 400 pounds of Al-Li in total. Today Al-Li is a much better known material, and one which has considerably evolved over the years. But to incorporate it to the existing 777 fuselage would requires extensive, and expensive, reengineering of a core element of the Classic. The new CFRP wing is already a major challenge and a very expensive proposition indeed. But Boeing has to keep the price of its main cash cow competitive. With a new Al-Li fuselage and CFRP wings we would be very close to a clean sheet design. And for an plane of this size it would perhaps make it more expensive than Boeing wants.

  10. When I was initially looking at the design of the CFRP body of the 787 and the A350 I was stunned: They had simply copied the metal design with skin, ribs and spars!
    It reminded me of the first aluminum bicycle frames back in the 70s (Alan, Vitus, Trek,…) that had also copied the tube and lugs construction from traditional steel frames and so wasted all the possible advantages that the material offered.
    I am so sorry that I have to say that, but the engineers that designed the 787 and A350 body apparently either did not have the time and recources to develop a new design that could use the material to its best qualities or were lacking the guts, genius or freedom.
    If you look a the potential of carbon fiber and the sorry planes they have designed from it – it is really a shame.
    Back to the comparison with bicycles: Todays carbon fiber racing bikes are down to 5 kg in weight from 10 kg when all was aluminum.
    If you use carbon fiber to its potential in a plane you should be able to make it at least 30% lighter, maybe 50%. And of course its design and manufacturing would be completely different.
    Maybe it will take a new generation of engineers – one that did not grow up with metal heads. 🙂

    • You need to keep in mind some things.

      Guts and freedom do not turn out successful aircraft. Its a commercial enterprise (or supposed to be, I think you could argue otherwise with Boeing these days)

      You also cannot economically make a flying structure out of CFRP that is self supported. It would never get off the ground. ergo, you put rings in it, take advantage of what CFRP does do and assist the areas it doesn’t

      And they are put together in sections so you need a flange to bolt the two together (to put it in somewhat crude technical terms)

      You save a lot of labor at the expense of a lot of cost. Where it all falls out maybe Bjorn could with in.

      Airbus did not have the tech in place to do what Boeing did as they planned on making aircraft the old way forever. their approach indeed was to exactly duplicate the old structure system using composites (though with large CFRO panels and not small aluminum ones, 9 I think)

      They still produced what is an economical aircraft so? I would have though not but the orders say otherwise. They forced Boeing up an aircraft size so that part worked.

      what you don’t see is smaller simpler things are far easier than large flying structures.

      Aircraft have to be conservative, a company is riding on its success and the Comet shows what can happen when you don’t understand fully what you are doing.

      Pretty much like all materials, there is a learning curve, you start heavy and work down.

      There have been some spectacular structural failures for those who did not follow that.

      Boeing has taken an amazing amount of weight out of the 787 (I think 5 tons over) and now they are under.

      but they also made a mistake on the wing joint and that’s the hazard of going too far.

      • Take a look at gliders. The old ones made from steel tubes and cloth, wooden wings with skin, ribs and spars. You might have copied that in fibre glass or carbon and what would you have achieved? 10%?
        Now take a look at modern gliders, their capabilities are so far away from anything you could build from metal, wood or any other “classic” materials (yes, including AlLi of course) that it really is a different world.
        I am not saying that I have the perfect design in my pockets, but I can clearly see that something had gone absolutely wrong both at Boeing and Airbus when those planes were thought out.
        For those who are not familiar with the figures: A CFRP-part features easily 4 times the specific strength of aluminum! Almost all of that advantage is lost due to the conservative and (sorry) boring design of the 787 and A350.
        I am absolutely sure a true CFRP plane can be a world better product. So why has it not been done? My best guess is that the lawyers and financial geniuses that rule A and B today have shied away from the R&D projects that would have been necessary to develop such a true CFRP plane.

        • I don’t think the major impediment is lack of inventiveness or inability to ‘blue sky’ a new and radically different plan form. To contrast with gliders is a tad simplistic in my opinion as they are designed with only a couple of parameters.. A commercial airliner has a vastly more complex set of considerations

          Consider Bjorn’s repeated consideration of day to day operation that limit the application of a truly optimised CFRP fuselage.

          We have reached a point where the aircraft form is determined, near circular fuselage, high aspect ratio wings, high bypass podded underwing (x2) engines etc etc. the whole infrastructure is based around this standard,note the outliers and how they have all succumbed over time.

          The benefits of a market disrupting technology would have to be massively compelling (-30% cost or more) to make it worthwhile and this appears to be limited to either engine development (gtf please work) or based around fbw systems (reduced empennage, btv, etc etc).

          We are at the very end of an evolutionary process in terms of commercial airline layouts and the OEMs are looking to eke out the last efficiencies they can from a tried and tested format.

          • The technological battle for enhanced cost-effectiveness is only sighting with the cross of the collimateur one half side of the yield equation, whilst limiting explorations to encompass one half side of the operational picture = flight … Remain the other three quarters of the picture, where there is room for disrupting (r)evolution(s) to come : in the revenue side of the yield equation, plus in the groundworthiness of modern aeroplanes … when you’ve done every bit of PIP for the engines, the aerodynamics, the structure, the flight controls, the avionics etc etc, you need to re-focus onto the revenue-generating features or onto ground operations, with quicker airport rotations in the mire.

            With a typical cost split for feeder operations of 45/20/35 between hourly/cyclic/fuel costs, if the average sector is 140′ trip time = 90′ flight time/50′ ground time, a saving of 16.5 minutes in the airport ground turn-around gives a trip cost decrement of 16.5/140 x 45 = 5.3 % or the same as fitting NEO vs CEO engines : 0.15 x 0.35 = 5.3 %

            So there is a lot of new ground left yet to be covered in aviation !

          • When you are thinking of something around the ‘scale of gliders’, the best example of different thinking to give an advantage is the Piaggio Avanti.
            They have optimised the forward fuselage shape- supposed to provide some lift-, and using rear mounted pusher engines allows the wing spar to pass through the center of the fuselage- the best location aerodynamically- rather than with a faired under fuselage location as is normal. This gives above average fuselage capacity. The use of a small forward wing wing with flaps allows the tail surfaces to be lifting and overall gives a smaller wing so for the same power provides higher speed.

            Its clear with smaller size you can look to design choices that are better than an entirely conventional design such as Kingair.
            Im assuming that these sort of changes dont scale up satisfactorily, even to a 20 seater passenger plane.

        • Don’t forget that aerospace is now an intrinsically super-conservative industry – and that’s partly down to certification.

          Every… single… tiny… change – to procedures or to materials or to design methods or to manufacturing – must be proven to be at least as “good” as before in engineering terms; then tested and certified to be at least as “safe” as before for the authorities.

          It’s these considerations that make everything move really slow in our world!

          • Thanks alot for this comment. Do I hear the voice of a somewhat frustrated engineer who knows perfectly well that CFRP has a lot more to offer when it is put in its best shape?

          • Ha ha ha! I’m genuinely amused that you should take that from my comment… In fact, while I have been working on this topic for many years now, I’m one of those who thinks CFRP isn’t the great leap forward other people were spinning it to be!

            It certainly has its place in some applications, but I’m not so sure it’s had any real advantage on the fuselage – and I have my doubts about the long term issues we haven’t had time to run into yet (maintenance, fatigue and damage tolerance (particularly at the joints), fire safety, etc.)

          • Imagine a 2-piece body (right and left side) in sandwich design. Could be honeycomb for example. Inside and outside made from various prepregs, each 2 or 3 m wide. 10 cm fiberglass honeycomb. No stringers, no spars. One fat, stiff, extremely strong shell. No corrosion, superior fatigue, 30-50 % reduced weight.
            Can’t be done?

          • Regarding sandwich monocoque fuselage – I’m sure it can be done… in fact I think it already *was* done back in the seventies by Rutan or someone like that for a small run of small aircraft.

            But it would cost billions in research and design work before I could even begin to speculate on any potential weight savings – let alone if it was economically viable. You are talking about turning the entire (glacially slow) aircraft development process on it’s head since literally everything would need to be researched, tested, designed, manufactured and certified almost from scratch.

            That was my point. The industry has settled (largely by regulation) into a pattern of long-term research looking for incremental changes which can be slotted into existing designs and processes once they’ve been tested to maturity. And even this costs billions.

            We already took a “moon-shot” with the jump to composite fuselage and wings – which cost everyone greatly, in return for modest gains. It simply isn’t worth betting the farm on radical changes unless it’s been totally proven to be worth it. The more radical the change; the larger the burden of proof required before making a switch.

            I’m sure there are people doing blue-sky research into such things though, especially since such sandwich construction is commonplace in flaps etc., so the fact we haven’t embraced it elsewhere presumably means the evidence of a large enough advantage just isn’t there at the moment.

          • The De Havilland Mosquito was made in two halves. You can find a very interetsing picture on this site, scrolling down almost to the bottom:
            And indeed there are companies working on such designs of two-piece bodies from CFRP:
            Oh, and the fuselage of the Honda Jet is made in two halves:
            And here I have found something that should really get you thinking: A Boeing patent where the fuselage is made from large pieces of sandwich. So there is a little research going exactly in that direction.
            It proposes to cut the body horizontal not vertical. Well, I guess I could agree with that. 😀

          • Just a couple of comments on Gundolf’s last post:

            The Mosquito is the well-known poster-child of all early composite aerostructures…

            I’m not really sure what your point is about the two halves… it’s basically the same as the four very large panels of the A350. The difference is scale – Mosquitos and HondaJets are a good deal smaller than A350s and 787s so while it’s useful for them to be in two halves, when building a widebody the benefits don’t stack up so well against manufacturing disadvantages.

            The Boeing patent (from 2006 you will note) doesn’t really mean all that much – just today I read about another Boeing patent for a freighter to engulf shipping containers like Thunderbird 2 (or the Eagle from Space 1999). Highly doubtful they are serious about that one! Not even sure what they are trying to patent in your link since it’s just a long, long list of prior art except for the integrated floor stanchions, as far as I can tell – given the timing I’m guessing it was probably just to make life difficult for Airbus if any A350 developments “infringed” on it.

            Something else I forgot to mention before – Rutan, Honda and glider manufacturers have much less to lose when trying novel manufacturing methods – less stringent certification, less costly aircraft, less economy-critical customers. Large aircraft designs at the big manufacturers are so close to each other in performance that no-one can afford to be “off” when trying a new technique. The small guys have much more room to play around.

          • @gundolf

            That LM experimental composite freighter still has large circumferential frames (NB: a spar is the terminology used for the main structural member of the wing and is not used for the structural members – frames – that maintain fuselage shape and transfer load).

            What you’re talking about – one fat, stiff, extremely strong shell – is more akin to the thick composite casings of a solid rocket motor having no cutouts for doors and windows. That won’t cut it for a fuselage “live specimens”.

            Here’s a link to two decade old concepts for CFRP fuselages, where the latter one – the Gondelkonzept option – can have a SOFI-type loadbearing design (Stringer Outside, Frame Inside). The stringers, btw, where are located in a sandwich with foam multifunctional filling.


          • @ OV-099 So for the body it is “ribs” then, right, not “spars”? Ok, got that.
            Two decades ago no doubt there were very clever engineers at work, but in the meantime technology has advanced and a lot of experience is gained. Only think of the many thousands of wind turbine blades built over the past 10 years. In my company we are making bows for string instruments from carbon fiber. The resin content in the finished sticks is as low as 25%. 20 years ago that would have been near impossible. They are 0.7mm of even wall thickness, curved and tapered, with a solid one piece head made in one piece. If you care to take a look:
            Cut-outs for windows and doors are no problem at all for a big sandwich tube. It is actually quite easy to incorporate reinforcements into it.
            Did you check out the Boeing patent? What do you think of that?

          • De Havilland saying, “no glue without a screw “.There were 30,000 small brass screws helping out the glue joins.
            Whilst the Mosquito was quite a tough plane,it was vulnerable to moisture just like modern honeycombs.
            They weren’t built to last and several broke up in mid air when deployed to some of the more humid places in the world.
            Laminated wood is amazingly strong, a place where I used to work called it “young carbon ”
            I am not a design engineer but l really like the look of the Vickers Wellington structure, they used to come back from bombing raids with huge chunks missing.
            Strangely enough the latest and greatest racing yachts are moving to thin skins and lots of ribs,this hasn’t been completely successful.
            With a yacht, when the skins start to disbond you have the option of slowing down or stopping before the whole thing unzips it’s self, this doesn’t always work for a plane.

  11. @ transtech:

    “What missing here is the modular nature of all of it and the so called snap together. ”
    That is long established procedure elsewhere. 1970tyish.

    “there is a lot less riveting and fastening involved as well. Some and some it seems to me.”

    Have you ever looked at how many fasteners those super barrels actually sprout 🙂

    “And yes Airbus did not have the tech ready so they tried to do what they knew how to do.”

    Your fallacy assuming such.

  12. Filament winding around a tube is almost as old as polyester resin.The machine used for the 787 was invented by North sails for sail making,but the first one was nabbed by Boeing, so wasn’t really fully industrialised.While developing carbon sails they also had to overcome the fact that sails are very thin but you also need the fibres to be placed in several directions
    This resulted in very thin individual layers, they called this thin ply technology.Thinner individual layers mean that the stiffness stays the same (the same amount of fibre), but it makes the laminate stronger. Just like the difference between 3 ply plywood and 9ply plywood.This seems to be just what’s needed, but unfortunately it takes much longer to place all these extra layers and is a bit tricky because they are very fragile. The learning curve will be very different because carbon is so new for airliners, but I believe that there are still quite a lot of big gains to be made. Airbus one piece cfrp door frame’s seem like they are going to save money and weight.Maybe it was a bit early but someone had to jump in and it would have been very risky for Airbus to ignore what Boeing was doing.

    • It seems Airbus took a more gradual approach on composites during the eighties and nineties. Boeing was conservative on composites on the 777 and jumped on the 787.

      Much of the composites production technology and structure design and modelling was developed by R&D centers like NASA Langley in the nineties and handled over to Boeing. Same as in Europe.

      • Most of Airbus know how is indigenous I’d guess.
        For Germany there is a range of competency places down the river from XFW to Cuxhaven centered around Stade.
        Already well established when I went to Uni in the early 80ties. Only they are not so busy with press releases like NASA is. ( Same goes for ESA actually. Good work but low on projecting that.)

        NASA: I seem to remember some exposé from them that saw more advantages from panels than from one piece barrels.

    • Hello grubbie ,

      Do you know what happend to the North 3DR Sails. They were advertized with a big marketing campain for some moth (including 3D glasses in the major sailing magazines to vie the photos of this 3D rotating laminating machine). And only a few months later, the whole 3DR product line was no longer available. Was the manufacturing process not working, or did they sold everything to Boeing for the Dreamliner programme?

      Best regards,

      • North 3DR process ( for a time ) seems to have had problems with the tape laying machinery.

        After that? no idea but fashions move fast in the luxury racing market. some small changes and a newly named marketing ploy emerges.

        Over the summer i see enough $”tapelayed somehow sails” around. also shredded CFRP spars to boot 🙂

        Boeing’s problem at the time was that tape laying speeds and progress therein did not in any way match expectations.
        I didn’t follow this too closely later on so I can’t say if speeds have finally gone up or the jobs have been parallelized

      • 3Dr sails were supposed to be a more economical version of the incredibly expensive (but incredibly good) 3di sails. They were delayed by the first machine going to Boeing, who were apparently highly impressed. I don’t really know what happened after that. It’s said to be much cleverer than it looks at first glance. Perhaps it was all a bit harder than they thought, ring any bells? I don’t think any one uses TPT for mass produced stuff yet as it takes too long to get all those layers down. Cfrp yacht Masts are a real killer app,much lighter where it really counts and also much stronger.

      • OK, I ran into some guys from Norths and asked the question. Basically the answer seems to be they don’t want to talk about it.There was a serious fire which destroyed one of the three machines and then it was abandoned.There were big problems with delamination and durability also a consistent shape. The shape itself was regarded to be a bit flat.When you look at the machine in detail this isn’t completely surprising, a bit over ambitious trying to combine all the processes in one go.

  13. To be considered is that a glider has a payload of 2 at most (or vast majority) and its hull is not pressurized nor full of passengers and freight.

    As Bjorn noted, fuselage serves a lot of functions on a commercial airliner, and to add to heat heat runs, electrical runs and HVAC.

  14. Structural liposuction applied to A319 LWO (Light Weight Option) taking a no-compromise approach when revisiting RdM/DdS protocols (Résistence des Matériaux – Dynamique des Structures) based on 2015-level know-how for AlLi, AlLiMg and Carbon components, all the way to the minutest poutrelle en ‘I’, stringer or skin-plate etc would permit a weight-reduction of such magnitude that the OWE of this aircraft would end up close or equal to the present-level OWE of the C Series competitor, whereby based on the better revenue-generating features of the smallest Airbus feeder, the tale is told.

    Why on earth Airbus are not using their design and engineering skills to their full present-time capabilities to muster A319 resilience in the market is beyond my understanding. Latest new programme – A350 XWB – is upn’flying since long so the brain-guys must be spinning thumbs going idle around in Europe … or itching to leave for a job to do c/o Elon Musk or Sir Richard Branson where the beat is on ?

Leave a Reply

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