As some of you may know, I like to tinker with airliner performance models based on the Breguet range equation. It's a neat little tool that allows you to explore the entire performance envelope of an airliner based on basic performance specs like max takeoff weight, range, payload, etc. The models are fairly basic and crude, but they do give answers good to a handful of percent.

Without further ado... here is a picture of how fuel burn is evolving over time:

The evolution of fuel burn

Boeing is in red, Airbus is in blue, and the marker for each aircraft is sized according to seating capacity (a bigger dot means a bigger airplane).

There are a few assumptions behind this chart, some of which were necessary to make the comparison as fair as possible.

(1) all airliners were 'flown' over the same mission length of 6000 nm. This removes the differences in design range, although it effectively 'abuses' ultra long-range airliners like the A345. Some will say I am comparing apples and oranges, and to that I would reply that I am comparing the entire fruit cart!  

(2) all airliners were 'flown' with a payload of 1 passenger per square meter of cabin floor. This metric gets us around the seat-count arithmetic that can so easily distort performance figures, and is so often the subject of heated debates here. (8 abreast! 9 abreast!! Unfair!!) The mission payload is therefore simply 95 kg times the number of passengers. This similarly 'abuses' some airframes that have higher seating density in their 'spec' configuration, like for example the A346.

(3) the 777-300ER is arbitrarily taken as the "baseline" (100%) aircraft to which all others are compared.

(4) I took the liberty of adding some models that do not exist (yet). These include the 787-10, the A350-900R, and an A380-900 with the same 'XWB' engines as the A350. The performance figures for these airliners are straightforward to extrapolate from the known performance specs of the basic variants, but should naturally be taken with a small grain of salt.

A few observations...

(a) we often have discussions about cost per seat-mile (CASM). Fuel burn, shown in this picture, is only one component of CASM, albeit an important one. This is not intended to be a chart comparing CASM, although fuel burn is often taken (rightly or wrongly) as a proxy metric for CASM.

(b) Despite the 'abuses' described above, the sales relationship of the 777 and A340 can be read right off the chart, if you squint hard enough  

(c) the A350 and 787 are likely to be a wash, at least in terms of fuel burn

(d) the A380-800 is not as 'green' as we are sometimes led to believe, and even the smallest 787 outdoes it

(e) if the market ever demands it, an A380-900 with A350 engines would put the A380 fuel burn back in the running, going head-to-head with the composite airliners... albeit in a much, much larger format

(f) revolution, or evolution? Fuel burn seems to follow a pretty steady trend line over time.

(note to mods: while this post is somewhat technical, I think it is of sufficient general interest that a move to Tech/Ops is not warranted)

Great. Thanyou for sharing that effort with us. Comparing fuel burn is difficult because of all the variables you mentioned.

I note that in the Boeing Press release regarding the 773ER order of Air NZ, they state that the current 773ER is 3.6% more fuel efficient than when it forst entered service.

Ruscoe

very interesting analysis. Its easy to see why the 332 is still selling so well given other options in its size and range.

So it appears that LH was correct in posting the relative fuel burn figures awhile back that indicated the A388 had better fuel burn than the 748? However, when compared with the new mid-size twins, it looks like trouble is a commin' on routes where "hub constraints" are not an overwhelming consideration.

Quoting Scorpy (Reply 2): very interesting analysis. Its easy to see why the 332 is still selling so well given other options in its size and range.

If you mean missions less that 6000nm, then I wholeheartedly agree. Otherwise, why wouldn't one opt for a 777 series?

This is a very nice effort. I hope you have your flame suit on!  

Just a question ,but is the statement that EK made about the A380 false? "A380 uses 18-22% less fuel per seat than a 777-300ER in comparable configuration"

If EK would be right the A380 would have to be around the 80% marker and effectively be the most fuel efficient per seat for all airplanes on your list.

Why is it that ailine companies all over the world still buy Airbus instead of Boeing planes because according to your numbers almost every type that Airbus Currently makes is outdone by a competing Boeing product ?

Sorry to sound sceptical but somehow these numbers don't line up with other sources (usually airlines that should have a good idea), or maybe I'm looking at it the wrong way which, granted, is also very well possible.

Quoting ArniePie (Reply 4): Why is it that ailine companies all over the world still buy Airbus instead of Boeing planes because according to your numbers almost every type that Airbus Currently makes is outdone by a competing Boeing product ?

...it only appears that the A345 and A346 are outdone. the A332 far outdoes the 763ER. And the 787 and A350 appear to be about on par.

As for EK's 18-22% figures, what are the conditions they are based on? Probably a completely different set of operational specs. It is quite possible both sets of figures are true, they are just measured under different conditions and metrics.

Many thanks for sharing this with us WM.   

Quoting ArniePie (Reply 4): Just a question ,but is the statement that EK made about the A380 false? "A380 uses 18-22% less fuel per seat than a 777-300ER in comparable configuration"

To be fair to WM, he's been quite clear about the "levelling" assumptions he was required to make in order to create this chart. He'll have had to make some other assumptions too, otherwise each blob would become a "circle of possibiliy" dependent upon range/configuration/mission etc.
See my next comment

Quoting WingedMigrator (Thread starter): all airliners were 'flown' with a payload of 1 passenger per square meter of cabin floor

Per our previous conversations, I understand why you've done this, but it may be potentially the biggest "distorter" relative to real life.

Quoting WingedMigrator (Thread starter): Despite the 'abuses' described above, the sales relationship of the 777 and A340 can be read right off the chart

As far as I am aware, the 773ER has c. 10% more floor space than the A346. However, previous threads have generated data from every user airline of each type which shows that the average 773ER flies with 3% more seats than the average A346.
So caveat a potential 7% mismatch from averagereality.
Doesn't change the outcome  

Quoting WingedMigrator (Thread starter): the A380-800 is not as 'green' as we are sometimes led to believe

It's also not as much of a gas-guzzler as we've almost always been led to believe (on A-net).
It appears to be a WHOPPING 27% more fuel efficient than the Boeing 744   - that's gas-guzzling for you.
In truth, I suspect this corrobortates my earlier comment regarding REAL passenger loads vs. cabin floor space, but this time to Airbus advantage  


Couple of questions/points from me..
1) I'm astonished that the 772ER is damn near as effecient per seat as the much larger (and much vaunted) 773ER of very similar range. It doesn't appear to support the "reality" that the 772LR is more efficient than the 772ER, even at shorter ranges.
2) The 787-10 would in all probability lose its "pole position" if it was required to match the larger A350's and not constrained at 245t/c. 7200Nm.
3) Like you, I find the A3510 has surprisingly little advantage over the A359, most of which I put down to the 4 or 5 extra tonnes of landing gear required (see also my point re- the 787-10 above.)
4) Roll on the A389   

Lots to think about there........
Thanks once again.

Regards

Quoting Lumberton (Reply 3): However, when compared with the new mid-size twins, it looks like trouble is a commin' on routes where "hub constraints" are not an overwhelming consideration.

Yes indeed. A380 sales will in my opinion hinge almost entirely on this consideration.

Quoting Astuteman (Reply 6): To be fair to WM, he's been quite clear about the "levelling" assumptions he was required to make in order to create this chart. He'll have had to make some other assumptions too, otherwise each blob would become a "circle of possibiliy" dependent upon range/configuration/mission etc.

Indeed. Perhaps a way to satisfy these concerns is to show the same chart without the 1 pax / m2 assumption, where they are flown at their design payload (from the Airbus or Boeing spec sheet). The 787's are flown at a load midway between 8 and 9 abreast. Of course, the seating capacities of hypothetical models are just educated guesses based on known dimensions. One caveat: this chart is subject to the effects of Seat Magic (TM).

Evolution of fuel burn, using design (spec) payload

Quoting Astuteman (Reply 6): I'm astonished that the 772ER is damn near as effecient per seat as the much larger (and much vaunted) 773ER of very similar range. It doesn't appear to support the "reality" that the 772LR is more efficient than the 772ER, even at shorter ranges.

I only had a 1% SFC improvement from the 772ER to the 773ER... perhaps that's on the low side? Also, it is not surprising that the 772LR would burn more fuel than the 772ER... the 772LR is essentially getting 'abused' at such a short range. It is a heavier aircraft because the MTOW is much higher, but it isn't larger. I think the sales pitch also goes that it can fly maximum payload (with lots of cargo) quite far... but that's another analysis.

Quoting Astuteman (Reply 6): The 787-10 would in all probability lose its "pole position" if it was required to match the larger A350's and not constrained at 245t/c. 7200Nm.

Probably true, yes. You'd see the same relationship between the 789 and the 7810 as you see between the A359 and A3510.

Quoting Astuteman (Reply 6): Roll on the A389

You bet! I would be surprised if they didn't build it by about 8 years from now. If size really matters, nothing else will even come close. And a new Y3 (not shown on the chart) is unlikely to top it by much at all, unless there is a significant breakthrough like a switch to BWB configuration.

I should make another chart like this for freighters... an A380F with 'XWB' power would raise an eyebrow or two, as would the future A359F.



  

Quoting WingedMigrator (Reply 7): You'd see the same relationship between the 789 and the 7810 as you see between the A359 and A3510.

That would be my guess too...

Quoting WingedMigrator (Reply 7): I would be surprised if they didn't build it by about 8 years from now. If size really matters, nothing else will even come close. And a new Y3 (not shown on the chart) is unlikely to top it by much at all

Not bad for an "old clunker" with "no innovation" to recommend it, carrying "60t of extra dead weight" around.........  

Regards

Quoting WingedMigrator (Reply 7): I only had a 1% SFC improvement from the 772ER to the 773ER... perhaps that's on the low side?

I believe the difference is between 2-3% between the older GE90s on the 772ER and the newer versions on the 77W/L.

Quoting WingedMigrator (Reply 7): I should make another chart like this for freighters... an A380F with 'XWB' power would raise an eyebrow or two, as would the future A359F.

Except that the hassle of loading a double deck freighter will make it extremely undesirable for most operators. I think the 748 will be far more attractive to most operators, even if not quite as efficient.

Quoting Lumberton (Reply 3): So it appears that LH was correct in posting the relative fuel burn figures awhile back that indicated the A388 had better fuel burn than the 748?

The gap between the two giants (or semi-giants if an A389 appears) is even more surprising given the later engines on the 748. Looking at it the other way, it would give Airbus considerable room to improve even the A388 let alone the 389.

It does appear as if that thread on the aerodynamics of the 388 and 748 got close to the same answer as WM gives us here.

I like the entry of XWB power. It threw me for a moment, I was wondering what an A389-A350 hybrid was going to look like until I found it was only the 350 engines. It is going to be interesting to see just how good RR chose to make the XWB Trents (or bad I guess noting that the sun has just gone down!).

Great work WM, roll out the cargo chart, I know you are dying to do it!    

Is there some simple way of making it a 3-D chart with range as the third axis?

Quoting Baroque (Reply 11): Is there some simple way of making it a 3-D chart with range as the third axis?

Anything else you require sir?
Water into wine?
(might suit the "Hunters" just now).......   

Regards

Quoting Baroque (Reply 11): Is there some simple way of making it a 3-D chart with range as the third axis?

Quoting Astuteman (Reply 12): Water into wine?

Nowhere near that complicated; with 3D CAD it is quite simple. Putting it into a conveniently viewable format is a little more difficult....

Quoting SEPilot (Reply 13): Nowhere near that complicated; with 3D CAD it is quite simple. Putting it into a conveniently viewable format is a little more difficult....

What about with an Excel spreadsheet?  

Regards

Quoting WingedMigrator (Reply 7): I only had a 1% SFC improvement from the 772ER to the 773ER... perhaps that's on the low side?

How much credit did you give the 772LR wing tip extensions relative to the 772ER?

Quoting ArniePie (Reply 4): Why is it that ailine companies all over the world still buy Airbus instead of Boeing planes because according to your numbers almost every type that Airbus Currently makes is outdone by a competing Boeing product ?

1) Airlines don't purchase based on fuel efficiency alone (it's important but it's not the only thing)
2) A & B don't sell their aircraft at the same prices.
3) A & B aircraft have different lifespans.
4) A & B aircraft have different maintenance costs.
5) A & B offer different discounts and services to different operators at different times.

2-5 put together (plus a bunch of other stuff) all figure into what the "best deal" is for an airline.

Tom.

Quoting OldAeroGuy (Reply 15): How much credit did you give the 772LR wing tip extensions relative to the 772ER?

None at all... (i.e. I left the L/D the same)

I believe the LR's weight goes up enough that the change in induced drag is minimal, and very likely below the noise level of a first-order model based simply on the Breguet range equation. My numbers could definitely be a few percent off.

These major parameters determine the range: MTOW & OEW, SFC and L/D. I just listed them in the order of difficulty of looking them up... MTOW is always stated on the manufacturers' websites, along with OEW most of the time. Engine SFC can be found with some digging on the web, and L/D, well... that's the proverbial black sheep. I end up "backing it out" by constraining the range given by the Breguet equation to the published range, and then cross-checking for relative consistency e.g. between stretches. Refined aerodynamics this is not.   

I do think it's worth performing a reality check on the L/D numbers... am I totally off the mark here? Can you suggest a simple way to estimate it independently, in order to cross-check and improve my Breguet model? I am certainly not expecting the model to "close", because it's only first order.

Quoting Baroque (Reply 11): roll out the cargo chart

Here is a chart with the freighters.

Freighter fuel burn

As before, Boeing is in red, Airbus is in blue, and the marker for each aircraft is sized according to maximum payload.

Again there are a few assumptions behind this chart.

(1) all freighters were 'flown' at maximum payload over the same mission length of 4000 nm, excepting the A332F and 767F which do not have the capability to fly that far at max payload. For those two, the fuel burn is computed for a 3000 nm mission.

(2) the 748F is arbitrarily taken as the "baseline" (100%) aircraft to which all others are compared.

(3) I once again took the liberty of adding some models that do not exist (yet). These include the 789F (MTOW 250000 kg, range 4400 nm, max payload 78000 kg), the A359F, and an A380F with the same 'XWB' engines as the A350.

Observations...

(a) the A388F vs. 748F story is readily apparent, at least for the assumptions of this analysis (4000 nm range, for which the 748F was optimized and for which the A388F is disadvantaged, being well under MTOW)

(b) It's pretty clear the A388F will never see the light of day in its current configuration; in 2014 the threat will be more from the future midsize twins than the 748F. At the very least I would expect Airbus to strap under the wings a set of new generation 'XWB' engines. Even with those engines, the 748F will still do better at 4000 nm. What the chart doesn't show is that such an airplane will fly 6000 nm at max payload. At 6000 nm, the tables are turned: the A388F 'XWB' would burn 20% less fuel per tonne-mile than the 748F, which would find itself in a payload-limited regime.

(c) The 777F is nearly as efficient as the 748F, something I would not have expected

(d) The A332F is a good deal more economical than the 744F or anything else flying today, which perhaps explains the strong order book

(f) The new generation CFRP freighters (A359F and 789F) will outperform anything built until then. Whether or not they eat into 748F sales will depend largely on costs other than fuel... if there are economies of scale in flying a VLA with slightly worse fuel burn, the 748F will be fine for a long time. If not, I would expect Airbus to abandon the A380F altogether in favor of the quickest possible entry into service for the A359F.   

Quoting SEPilot (Reply 10): Except that the hassle of loading a double deck freighter will make it extremely undesirable for most operators. I think the 748 will be far more attractive to most operators, even if not quite as efficient.

Perhaps. I think these 'hassle' considerations (upper deck loading, no nose door, no 10 ft tall pallets) are largely second order effects compared to the basic operating costs of hauling x tonnes over y miles day in and day out. I don't have enough information to judge one way or the other.

Very interesting, thanks for posting!

Would be interested to see how you use the Breguet equation and where you got the specs from - would like to replicate the study.

Also, how does the A340-300 stack up?

Quoting WingedMigrator (Reply 18): It's pretty clear the A388F will never see the light of day in its current configuration

Quoting WingedMigrator (Reply 18): What the chart doesn't show is that such an airplane will fly 6000 nm at max payload.

FWIW I don't expect the A380F, when re-launched, to have the same max payload capacity as it did previously....  
Regards

Quoting SEPilot (Reply 13): Quoting Baroque (Reply 11): Is there some simple way of making it a 3-D chart with range as the third axis? Quoting Astuteman (Reply 12): Water into wine? Nowhere near that complicated; with 3D CAD it is quite simple. Putting it into a conveniently viewable format is a little more difficult....

Viewing 4-D is where the difficulties really come in as you construct a hyperbox to plot the graphs. And yes, water into wine is a speciality of Australia these days, you should come and view the process Astuteman, although whether we have quite enough to run trials on nuke subs is not known. Why not bring one and check! If trying a simulation, do not consume too much of the wine before choosing which version of 3D CAD you will use.

Quoting WingedMigrator (Reply 18): Here is a chart with the freighters.

Thank you muchly kind sir.

Quoting WingedMigrator (Reply 18): (b) It's pretty clear the A388F will never see the light of day in its current configuration; in 2014 the threat will be more from the future midsize twins than the 748F. At the very least I would expect Airbus to strap under the wings a set of new generation 'XWB' engines. Even with those engines, the 748F will still do better at 4000 nm. What the chart doesn't show is that such an airplane will fly 6000 nm at max payload. At 6000 nm, the tables are turned: the A388F 'XWB' would burn 20% less fuel per tonne-mile than the 748F, which would find itself in a payload-limited regime.

Aha, so 3-D will be good!! Those are two really interesting observations, that the T900 might have to go to make the 388 competitive and that it really gets quite a bit better at >4000 nm. I guess once you get those huge wings up aloft, it is better to "flap" them for a longer time.

Apart from the desirability of later Trents, presumably all this shows that the thing that would help the A380 is longer sectors, presumably both for the passenger version and the F model.

WM, you did leave out the A343, A342 and A333. Is it because of lack of data? What is missing?

Your estimations about the A388 and the 748I seem to concur with LH's expectation that the 748I needs about 3-4% more per passenger.

SailorOrion

[Edited 2007-08-06 15:47:36]

Very interesting indeed...

...How about the A330-300?

Very interesting thread and very good work WM! I'd like to join if I may and add some relavent points to the discussion. I've done some scholarship on these issues and have some data to present and I'll get to that as well. One concept that I don't necessarily put much credence in is so called "seats per area" An area density seating model leaves out one very important aspect which is the shape of the cabin or its aspect ratio. For a given same area of cabin a higher aspect ratio fuselage i.e. longer and thinner, will be able to have more seats with the same pitch/size seats. A good example of this is the 777-300ER and A340-500. In operation the seating arrangements are quite close in number despite the fact that there is a notable disparity in cabin area. The A340-500 can hold more rows bringing parity to the cabin usefulness.

To really match cabin density you have factor in the cabin length and width as a determiner of seat rows and abreast seating, the next factor is distrubution of classes and class seat pitch. Throw in galley and crew spaces and then you have the deal just about done. However there is really no substitute for doing actual cabin plans as there are many surprises that abstract geometry won't account for... like doorways...

Take for example the metric of one seat per square meter of cabin area. This measure would put the A380 in at over 550 seats and a 747-8 at 420 or so. This is totally at variance with realistic considerations of seating. Using the same/similar seats, pitch, and class distribution this disparity would never materialize in the real world. While 420 seats could be a realistic consideration for a 747-8 in a premium international configuration with generous percentages of the total seats being buisiness/first class the same would not hold true for the A380. 550 seats on an A380 would require a larger proportion of economy seats to make up the total if the same larger pitch business/first class seating rules are used.

With that said let's use that metric as a standard for the case of simplicity and als use it to measure the differences for more practical arrangements...

Next thing is mission models and weights. This is connected to the first point in this way. lower cabin density, reflected by comparing a standard metric to operational data. In practice lower cabin density means heavier cabin interior items such as seats and galleys are used, albeit in different proportions, to make up the total. As an example on the 777-300ER NH uses a 247 seat four class arrangement while AC uses a 2 class 349 seat arrangement. I'm sure it would surprise most to know that the AC ships are lighter, lower DOW, than the NH planes. That is because the cabin interiors are heavier for the NH birds, especially the economy seats. DOW or Dry Operating Weight when we are talking about practical examples of aircraft can vary wildly with manufactures spec or even typical OEW. Those are the weights the OEM's use for quoting public performance figures on and they can be close to what the majority of operators will see or nowhere near them. This makes substantial differences in range performance. Performance quotes are a dubious thing to someone like me when it comes to comparing airplanes. My first question is what is the basis of the results, weights, seating, allowances, etc. Once that basis is established then we can adjust as needed to represent practical situations.

Manufactures basically use three kinds of airplane weights to quote performance. Spec weights, generic operating weight, or typical operating weights. The first two can be wildly off the mark as far as what the majority of operators use but the last, Typical OEW, is usually pretty close to the weights operators will see. Right now there are basically two aircraft that use the manufactures Typical OEW for public performance quotes and those are the A380-800 and the 777-200LR. The weights used for public performance quotes on those aircraft are very close to what operators will see across the board. This makes OEM performance numbers more applicable to reality, but if dramatically lower weights are used, i.e. A340-300, public information can be radically different what operators see in practice.

Quite a mouthful there but getting to my point. With respect to WM as the pioneering effort here I'll follow his prescribed methods tempered with a little more data and my own insight. The tables I've done are the background data set for the chart done in WM style. I did include the cast-down data from two mission models, the first being max payload for the 6,000nm segment, and the other being passenger only payload for the same distance. The point of that was to illustrate the cost of carrying additional payload on the passenger aircrafts which would normally be cargo and to put a cost measurement on it. Carrying cargo on a passenger aircraft is usually incidental to the main purpose of the operation but an important revenue generator. The cost of carrying additional payload in terms of aircraft operation is simply the cost of fuel which is detailed in the dataset.

I tried to be as specific and realistic as possible given the contraints of 1 seat per square meter. This necessitated changing cabin appointment levels based on what is used in practical operation so it's not necessarily and apples to apples comparison in that regard. Cabin appointment levels are a significant factor in determining the DOW of each airplane used for the mission models. As I said before The A380-800 vs 747-8 using this metric is a prime example of this. My DOW are operational weights at the cabin seating levels.

Future models:

I have to add the discalimer that the A350XWB models and 787-10 performance data are highly suspect in my opinion. As far as the Boeing goes the weights I am comfortable with but not the aerodynamics. I think that spec of aircraft without any growth version of the current engines will be hard pressed to make the range figures Boeing wants. On the other hand the Airbus aero looks to be there but the weights are... well very ambitious IMO given the size of the wing, materials and fuselage construction methods.

A330-300:

The A333 is not an airplane capable of the 6,000nm mission model. I have included it at the lower specified range of 5,000nm.

The last factor which is not take into account at the is aircraft price via debt service or lease, and maintenenace. I will update the data to reflect some information that includes other cost of ownership foactors.

Without further delay:

LONG HAUL AIRCRAFT COMPARITIVE FUEL BURN DATASET

LONG HAUL AIRCRAFT SEAT SPECIFIC FUEL BURN BUBBLE CHART

Please feel free to note any need for corrections of typos.



-widebodyphotog

So you are expecting the A350XWB to burn a good bit less fuel in absolute terms than the 787?

Actualy looking at it, might want to check all your A350XWB figures as the A350-1000 EIS date is off on the bubble chart, and the 3 models make no sense as far as fuel burn Vs size to me atleast.

Quoting XT6Wagon (Reply 25): Actualy looking at it, might want to check all your A350XWB figures as the A350-1000 EIS date is off on the bubble chart, and the 3 models make no sense as far as fuel burn Vs size to me atleast.

Bubble chart shows 2015 EIS for the A3510, as currently stated by Airbus - or were you trying to make a point?  

As I commented earlier, somewhat surprisingly, the A3510 doesn't improve on fuel burn per seat as much as you might expect, primarily because at its MTOW, the triple bogie landing gear is required, which ups the OEW proportionally more than a simple stretch.

If Airbus were to make the A3510 the same size, but limit it to the 565 000 lb (and reduce the wing area accordingly), the A3510 would appear remarkably like a slightly larger 787-10, i.e. 350 seats, c. 7200nm range, and virtually unmatched fuel burn per seat.
As it is, its fuel burn per seat is remarkably similar to the A359, which currently seems to be the pick of the A350's.

Regards

=

Quoting XT6Wagon (Reply 25): Actualy looking at it, might want to check all your A350XWB figures as the A350-1000 EIS date is off on the bubble chart, and the 3 models make no sense as far as fuel burn Vs size to me atleast.

Don't blame me for the burn, the Airbus figures come out just that low. As I said in my caveat Airbus range and payload figures based on the airplane are rather ambitious. However, relative to the the aircraft Airbus is targeting they are right on as far as the Airbus claims. Based on this metric the fuel burns make sense as relative to size the cabin density is basically reversed from a real world application. However we want to stick to the 1 pax per square meter and that's the way the seating come out. If we were looking at it in practical terms we would have numbers that would fall along these lines:

A350-800 - 238
A350-900 - 288
A350-1000 - 332

This would change the picture somewhat...


I will say this for the A350XWB models though. Airbus has taken the A380 route and gone with a massive lower aspect ratio wing optimized for the largest variant. That has the effect of creating an airplane the size of the 787-9 with greater wing area than a 777 in the A350-800! The airplane is going to be very "floaty" with very good takeof and climb performance combined with low cruise thrust fairly low cruise thrust requirements. Burning less fuel to get on cruise quicker on the way up makes a big difference even if the big wing creates higher induced drag along the way. The 787-9 on the other hand is is where the A350-1000 is in terms of an optimized arrangement with a "properly" sized wing and good overall performance. I don't have an opinion on which approach is the best way but just reporting the facts as I see them.


Corrections noted and updated

-widebodyphotog

[Edited 2007-08-06 23:19:36]

Quoting Astuteman (Reply 14): What about with an Excel spreadsheet?

Well, it might come out something like Rubik's Cube...

Quoting Widebodyphotog (Reply 24): I have to add the discalimer that the A350XWB models...performance data are highly suspect in my opinion...the Airbus aero looks to be there but the weights are... well very ambitious IMO given the size of the wing, materials and fuselage construction methods.

Another airliners.net member privately sent me what they believed to be the DOWs for the A350XWBs that were on average 50,000lbs lower per model then what you have.   

Quoting Widebodyphotog (Reply 24): Please feel free to note any need for corrections of typos.

763ER from the graph is 0.0145, your table has it at 0.0189, looks like you plotted the A380-800 next to it.

Quoting Stitch (Reply 29): Another airliners.net member privately sent me what they believed to be the DOWs for the A350XWBs that were on average 50,000lbs lower per model then what you have.

Honestly, if they are that much at variance I would have to say they are blowing smoke! I mean that would put for example, the A350-1000 more than 100,000lbs lighter than the 777-300ER and the A350-800 lighter than a 787-8. These weights are basically impossible. Only possibility is if Airbus was using a spec weight based performance model with an all economy cabin like they did for... ahem the A330's... My weights are very well informed and methodically produced reflecting operational realities. Another avenue is that they may estimate weights that low because if you ignore the Airbus claims you would have to assume higher mass fractions of fuel than Airbus is claiming in a normal analysis, thus lower structural weights. Given the proposed aerodynamic properties the fuel burns make sense however the weights do not. Based on a loose structural analysis the DOWs would be significantly higher and the ranges much less again Airbus fuel burn claims ignored. Additionally assumptions about the Trent XWB powerplant may be off the mark as well. At the low cruise thrust required the propulsive efficiency may be out of the optimum range for three shaft engines especially for the -800. This would result in a lower cruise speed or increased SFC unless BPR was around 13:1!

Quoting Zeke (Reply 30): 763ER from the graph is 0.0145, your table has it at 0.0189, looks like you plotted the A380-800 next to it

Noted and updated



-widebodyphotog

Quoting Widebodyphotog (Reply 27): Airbus has taken the A380 route and gone with a massive lower aspect ratio wing optimized for the largest variant

What do you mean by "lower"?
Lower than normal?   
The aspect ratio of the A350 wing is c. 9.2, which is commensurate with most other wings around today, and may be even a tad higher.
Lower than the 787?   
The 787 has an extraordinarily high aspect ratio of 10.4, combining the best attributes of low wetted area drag, and low lift drag due to the large span.

The A380, by contrast, has an aspect ratio around 7.5, which IS a "lower" aspect ratio wing  

Quoting Widebodyphotog (Reply 27): Burning less fuel to get on cruise quicker on the way up makes a big difference even if the big wing creates higher induced drag along the way.

Higher induced drag???
The big wing might create high induced wetted area drag, but the induced lift drag resulting from having a 64m wingspan on a 245t aircraft will be absolutely phenomenal (c. 7% better than the 787-9).
In that sense, relative to say, the 787-9, the two drags might well cancel each other out, and result in both aircraft having similar drag, but the 358 having substantially better climb.

Quoting Widebodyphotog (Reply 31): and the A350-800 lighter than a 787-8

What the A358's wing can't do is be lighter than the 787's   
The A358 should have an OEW c. 4t heavier than the 787-9 (it's direct comparator), and thus considerably heavier than the 788 (which is a much smaller plane..)

Regards

Quoting Widebodyphotog (Reply 24): Very interesting thread

Thanks for your kind words, and for joining the discussion.

I'm glad our numbers look so similar (at 1 pax / m2)... I plugged your results into my spreadsheet, and it turns out you and I have a systematic offset of 7% (in engineering units) and a standard deviation of 5%. Definitely good to see my simple Breguet model work out to better than 5% on a relative basis   

There is one discrepancy that stood out like a sore thumb, the A350-800. On that one we were something like 15% apart... and that drove the above-mentioned 5% standard deviation across the data set. Physically I would expect the A359 to burn less than the A358, and your numbers don't show that... so I wonder if another closer look at your computations (not the values, but the calculations behind them) is warranted just for that one aircraft type. I think XT6Wagon is potentially onto something.

One other tiny little nit... 1 nautical mile = 1.852 km exactly

Quoting Widebodyphotog (Reply 27): Burning less fuel to get on cruise quicker on the way up makes a big difference even if the big wing creates higher induced drag along the way

This is precisely the sort of subtlety that my model cannot capture. Incidentally, would it be professionally impolite to ask what sort of model you've got under the hood-- i.e. something as simple and crude as the Breguet equation (nothing wrong with that ) or, I surmise, fancier mission simulation software?

Quoting SailorOrion (Reply 22): WM, you did leave out the A343, A342 and A333. Is it because of lack of data? What is missing?

No, simple omission... here they are (A333 at 5000 nm):

The evolution of fuel burn, at 1 pax / m2, with more types added

I made a mistake in my second chart (the one showing 'spec' payloads)... I hadn't really removed the normalization by seat density. Here is the data for a 6000 nm sector at 'spec' passenger numbers, with no cargo:

The evolution of fuel burn, at design (spec) payload

It sure is different, isn't it?

Quoting Widebodyphotog (Reply 24): One concept that I don't necessarily put much credence in is so called "seats per area"

   point well taken, but I had to start somewhere... I very much see the importance of doing 'proper' seat counts. I think the comparison between the two charts really illustrates the power of seat counts and how strongly they drive the comparisons. That only reinforces the importance of getting the seat counts right, when you're doing real world calculations and not making theoretical leveling assumptions like I did with 1 pax / m2.

It just goes to show that there are lies, damn lies, and statistics.  

And oh how much hot air is vented on a.net about these statistics!

Quoting Stitch (Reply 29): Another airliners.net member privately sent me what they believed to be the DOWs for the A350XWBs that were on average 50,000lbs lower per model then what you have.

For what it's worth, I had the OEW's pegged at 122, 130 and 143 t (compared to WBP's 124, 132 and 150 t) for the A358, A359 and A3510 respectively. Airbus has yet to publish numbers, so one must resort to educated guesses.

Quoting Astuteman (Reply 26): Bubble chart shows 2015 EIS for the A3510, as currently stated by Airbus - or were you trying to make a point?

when I viewed it the A351 was EIS in 2005 or so.


Widebodyphotog,

My problem right now is that The 787 family follows a fairly sane size -> fuel burn pattern, while the A350 makes no sense. The A359 and A351 burn equivalent fuel while the A358 is massively better? Dah hell? This would be alot like waking up to finding that Boeing didn't build the 771 because it was 10% more efficient than the 772... or 100% out of reality.

Now I asked for the double check not because I doubt your methods, but rather to confirm that the Airbus numbers currently supplied make no sense.

Personally I think we are atleast a year early on tossing around the numbers on the A350XWB, as Airbus apparently hasn't even completely locked down the very basic configuration, much less gotten into the detail engineering work to make the paper and reality match.

Quoting XT6Wagon (Reply 34): The A359 and A351 burn equivalent fuel while the A358 is massively better? Dah hell?

Did you see the explanation I posted earlier?

Quoting Astuteman (Reply 26): As I commented earlier, somewhat surprisingly, the A3510 doesn't improve on fuel burn per seat as much as you might expect, primarily because at its MTOW, the triple bogie landing gear is required, which ups the OEW proportionally more than a simple stretch

Look at the respective jumps in WM's and WBP's OEW estimates for the A3510 relative to the A358/A359.
I assume (in one case know) that this jump in OEW is caused by the assumptions these gentlemen have made regarding landing gear weight.

Quoting WingedMigrator (Reply 33): For what it's worth, I had the OEW's pegged at 122, 130 and 143 t (compared to WBP's 124, 132 and 150 t) for the A358, A359 and A3510 respectively

You can easily see why the linear relationship gets broken.
The break is at 565k lb MTOW (because of the bigger landing gear required).
Any 787 derivative over 565k lb would suffer the same fate.....
Hence Boeing's "dilemma"  

Regards

Quoting Stitch (Reply 29): Another airliners.net member privately sent me what they believed to be the DOWs for the A350XWBs that were on average 50,000lbs lower per model then what you have.

Quoting Stitch (Reply 29): Honestly, if they are that much at variance I would have to say they are blowing smoke!

Actually, it was a case of me being an idiot and comparing the wrong set of numbers. Their actual data tracks close to yours. I have apologized to them privately for my mistake.      

Quoting XT6Wagon (Reply 34): My problem right now is that The 787 family follows a fairly sane size -> fuel burn pattern, while the A350 makes no sense. The A359 and A351 burn equivalent fuel while the A358 is massively better? Dah hell? This would be alot like waking up to finding that Boeing didn't build the 771 because it was 10% more efficient than the 772... or 100% out of reality.

Quoting Astuteman (Reply 32): The big wing might create high induced wetted area drag, but the induced lift drag resulting from having a 64m wingspan on a 245t aircraft will be absolutely phenomenal (c. 7% better than the 787-9). In that sense, relative to say, the 787-9, the two drags might well cancel each other out, and result in both aircraft having similar drag, but the 358 having substantially better climb.

It is interesting that some folks I talked to felt the A350-800 would be the "777-100" of the family, but now it looks like it might just be the best seller...   

Quoting Astuteman (Reply 32): What do you mean by "lower"? Lower than normal? The aspect ratio of the A350 wing is c. 9.2, which is commensurate with most other wings around today, and may be even a tad higher.

"Lower" in terms of the route Boeing chose... I for one would have thought that Airbus would move more towards a higher a/r wing in addtion to the higher sweep. I also thought it would have made sense structurally again with sweep being the largest difference in the A350 wing from the A340, in order to save significant weight through the model range although this would have called for more powerful engines...

Quoting XT6Wagon (Reply 34): My problem right now is that The 787 family follows a fairly sane size -> fuel burn pattern, while the A350 makes no sense. The A359 and A351 burn equivalent fuel while the A358 is massively better? Dah hell? This would be alot like waking up to finding that Boeing didn't build the 771 because it was 10% more efficient than the 772... or 100% out of reality. Now I asked for the double check not because I doubt your methods, but rather to confirm that the Airbus numbers currently supplied make no sense. Personally I think we are atleast a year early on tossing around the numbers on the A350XWB, as Airbus apparently hasn't even completely locked down the very basic configuration, much less gotten into the detail engineering work to make the paper and reality match.

Checked the data/database as requested and I did find an error in the display... The A350-800 data cells had some "lookup" values from the database that were not in the right place. Not values with big enough differences to be noticed so readily but this made the sums in the table significantly wrong. Thanks for your persistence and the data has been updated...



-widebodyphotog

Quoting Widebodyphotog (Reply 37): "Lower" in terms of the route Boeing chose... I for one would have thought that Airbus would move more towards a higher a/r wing in addtion to the higher sweep. I also thought it would have made sense structurally again with sweep being the largest difference in the A350 wing from the A340, in order to save significant weight through the model range although this would have called for more powerful engines...

Agree completely, FWIW.

I guess I was just "checking for understanding" again, when the A350 wing's A/R was compared to the A380's.

I'd like to see the A358 get the wing you've just described, but still at 64m span.
Less OEW, Less area drag, same lift drag - it'd be a better plane... (IMO)

Regards

Quoting WingedMigrator (Reply 17): Quoting OldAeroGuy (Reply 15): How much credit did you give the 772LR wing tip extensions relative to the 772ER? None at all... (i.e. I left the L/D the same) I believe the LR's weight goes up enough that the change in induced drag is minimal, and very likely below the noise level of a first-order model based simply on the Breguet range equation. My numbers could definitely be a few percent off.

Since the increased span of the 772LR will produce about 12% less induced drag than the 772ER at the same weight, it would be safe to assume that the 772LR L/D was at least 4% better.

Since the Breguet equation includes Weight terms, not making the span correction results in the higher 200LR weights being applied to the lower 200ER wing span. I'd also credit the -115B with at least a 2% TSFC improvement relative to the -94B.

I appreciate the trouble that you're going to produce these data.

Great job WBP! The latest set of corrections brings everything into the correct relationship (within the odd 1% - 2% or so), at least from my undertanding.

Quoting Astuteman (Reply 20): I don't expect the A380F, when re-launched, to have the same max payload capacity as it did previously....

Missed that comment at first... neither do I. I think Airbus missed the mark with too much range and not enough structural payload. At the same MTOW (590 t) they can add 10 t of structural reinforcement in the fuselage and floors (but not in the wings or landing gear, since MTOW does not change) to support 25 t more payload (175 t). Range drops from 5600 nm to 4400 nm but fuel burn per tonne-mile drops by 6% and matches the 748F.

Now if they wanted to increase MTOW (and there's plenty of growth potential...) 

Quoting Stitch (Reply 36): It is interesting that some folks I talked to felt the A350-800 would be the "777-100" of the family, but now it looks like it might just be the best seller

FWIW the numbers appear to put the A350-900 at the "sweet spot".
The 358 has an enormous wing area for its weight.
The 3510 gets a much heavier landing gear....

Now if Airbus od a "787", and change the wing so that the A358 gets a smaller area, lighter wing (but same, or similar span).....  

Quoting WingedMigrator (Reply 40): Now if they wanted to increase MTOW (and there's plenty of growth potential...)

That was where my comment was going.

An A388F with the extra reinforcement you mentioned, AND an MTOW of 625t would be a beast indeed.

200t payload anyone?  

Regards

Quoting Astuteman (Reply 41): FWIW the numbers appear to put the A350-900 at the "sweet spot". The 358 has an enormous wing area for its weight.

Alas, my comments were based on the incorrect numbers for the A358 which showed it having the best fuel-burn per seat mile of the lot. Now that it has been corrected, it is now the worst member of the family, though overall it remains good.

Quoting WingedMigrator (Reply 18): (f) The new generation CFRP freighters (A359F and 789F) will outperform anything built until then. Whether or not they eat into 748F sales will depend largely on costs other than fuel... if there are economies of scale in flying a VLA with slightly worse fuel burn, the 748F will be fine for a long time. If not, I would expect Airbus to abandon the A380F altogether in favor of the quickest possible entry into service for the A359F

What accounts for the big improvement of the A359F versus the B789F when the passenger versions are just about equal?

Thanks!

Quoting Justloveplanes (Reply 43): What accounts for the big improvement of the A359F versus the B789F when the passenger versions are just about equal?

Better wings and engines.

Quoting Stitch (Reply 44): Better wings and engines.

The A359F will have different engines and wings than the A359?

Quoting Justloveplanes (Reply 45): The A359F will have different engines and wings than the A359?

The A359F will have the same 295 tonne MTOW and six-wheel MLG bogies as the A350-1000. So yes, it will have slightly more powerful engines than the A359. Some of the extra structural weight pays off in payload capacity, which in turn cuts down tonne-mile costs.

Keep in mind that these fuel burn numbers are only educated guesstimates; Airbus has yet to publish detailed information on the A359F-- although they did announce the maximum payload, 90 tonnes. For what it's worth, the 777F still has a better payload to OEW ratio than 'my' A359F, so there's certainly no 'funny' math going on here. The real thing should be pretty close, within a few percent.

You should consider my A359F and 789F numbers to be a wash, within the margin of error. What is more notable is their relationship to the 748F.

Quoting WingedMigrator (Reply 46): The A359F will have the same 295 tonne MTOW and six-wheel MLG bogies as the A350-1000. So yes, it will have slightly more powerful engines than the A359. Some of the extra structural weight pays off in payload capacity, which in turn cuts down tonne-mile costs.

Aha!

Thank you, and it will be VERY interesting to see how fast Airbus pushes the A359F as it does seem to have much more promise than the current A380F.

And thanks for this thread, my favorite A.net to date (I have it bookmarked)!

First,I've really enjoyed reading through this thread. I've read every post, graph, and table in detail.   

Its very rare I feel like I should just sit back and read a technical discussion on a.net and just 'suck it in' and not comment much. Thank you WM, you're now on my R/U list. Widebodyphotog is pegged there.  

Quoting Astuteman (Reply 41): The 3510 gets a much heavier landing gear....

Interesting design tradeoff. With all the material improvements in aircraft, it amazes me how nothing takes the shock load as well as steel for landing gear. I see nothing replacing steel there anytime soon either.

Quoting WingedMigrator (Reply 33): Here is the data for a 6000 nm sector at 'spec' passenger numbers, with no cargo:

I'm a bit confused as to why the A388 performance became so much worse compared to the 77W at 6000nm than at 5000nm. Is that a typo?

Also, does the A333 really belong on the 6000nm chart?  

Lightsaber

Quoting Lightsaber (Reply 48): Interesting design tradeoff. With all the material improvements in aircraft, it amazes me how nothing takes the shock load as well as steel for landing gear. I see nothing replacing steel there anytime soon either.

I would suppose titanium would work, but the cost would probably be prohibitive. Titanium may not have as good wear resistance, and it is springier than steel, which might actually be a disadvantage. In my current job (designing equipment for semiconductor manufacturing) we use titanium a lot, mostly for its thermal characteristics, plus its lack of interest in oxidizing. But the largest parts we use it for are less than a kilo, and most of them have the largest dimension a couple of centimeters.