Sponsor Message:
Aviation Technical / Operations Forum
My Starred Topics | Profile | New Topic | Forum Index | Help | Search 
Boeing Vs. Airbus Wing Design Philosophies  
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Posted (2 years 7 months 1 week 6 days 14 hours ago) and read 29648 times:

In the A350 prototype thread no 1 post 236:

A350 Prototypes Production Thread Part 1 (by ferpe Oct 26 2011 in Civil Aviation)

I compared the A350 wing with the 787. They have many similarities and smaller differences but one difference that seemed large was the wingloading, Airbus going for a larger wing with more wetted area and Boeing preferring a smaller wing with lower weight and accepting a higher wingloading    .


I then decided I ultimately wanted to understand the consequences of it all. As only the 787-8 and not the 350 was available as PianoX model (excellent preliminary design tool with a free of charge demo copy here www.lissys.demon.co.uk/PianoX.html ) I started to implement a simple model my-selves.

Reading the Piano manual before was of great value and I have continuously been able to verify my results with the PianoX 788, 767 and 346 models that are provided for free (using the Output : Detailed flight profile and Point performance ), a big thanks to Dimitri Simos for making this really well designed tool available to us enthusiasts    and for having the complete Piano 5 guide on the web.


Dimitri also helped me to solve the biggest problem I had, things just did not jive even though I had implemented everything that made sense from http://adg.stanford.edu/aa241/AircraftDesign.html , the Aircraft design course written by R Shevell of Douglas fame.


   This part of the Piano manual is more true then one thinks:

http://www.lissys.demon.co.uk/pug/c03.html#s05


  ..........................WE ARE COMPARING APPLES AND ORANGES.............................  


The wingareas we are throwing around do not have the same definition, the one for 788 (325m2) is Trapezoidal, the 443m2 for the A350 is Airbus and even PianoX mixes definitions e.g. using Boeing Wimpress for the 767, Airbus for 346 and Trapezodial for 788    . Here the definition of the areas (from the Piano 5 guide):



And here how they pan out for some of the types we discuss all the time (I now measure them   Sad

m2............................................Trapezoidal.......Airbus........Boeing/Wimpress........Exposed
787-8, 9, 10.....................................325..............370...................360......................301
A350-8, 9, 10...................................400..............443..............................................362
777-200--300ER...............................428..............454..............................................372
767............................................................................................283...........................
330....................................................................362...................................................


Here the wingloadings (MTOW/Wingarea) I skipped Wimpress as I only have one value and it is more complicated to measure. Haven't measured the ACAP drawings of the 330 and 767 yet:

m2............................................Trapezoidal.......Airbus.........Exposed
787-8..............................................701..............616...............757
787-9..............................................772..............678...............833
350-8..............................................648..............585...............715
350-9..............................................670..............605...............740
350-10............................................745..............673...............819
777-300ER......................................821..............774...............945
777-8X............................................682..............630...............784
777-9X............................................731..............689...............840
330....................................................................649


So if we take e.g. the Airbus column (might be practical as Airbus publicizes these values, IIRC Boeing does not) we can see that the 788 and 359 as the two new main models have similar wingloadings!!!    .

And the next bigger models using the same wing (787-9 vs 350-10, (35J has a 4% larger wing but the reference area stays the same) has the same wingloading as well    , so all this talk about differences in their wing design philosophies re wingloading is    .


This conclusion came when I already had my model working and tracking PianoX within a couple of % so I am happy anyway because I now have a tool to check a lot of things, but more of that in another post    .

[Edited 2012-04-16 13:59:26]


Non French in France
81 replies: All unread, showing first 25:
 
User currently onlinePihero From France, joined Jan 2005, 4630 posts, RR: 77
Reply 1, posted (2 years 7 months 1 week 5 days 13 hours ago) and read 29386 times:
Support Airliners.net - become a First Class Member!

A very interesting subject.
IMO, difficult to differentiate between the twp designs in clean conditions. They are both high-aspect ratio with variable -geometry camber, high aeroelasticity wings.
The big difference is - and has always been - in the low speed, wing high lift devices :
- the Airbus uninterrupted trailing edge flaps generates more lift and less drag than the separated inner / outerr flaps plus an engine gate, high and low speed ailerons (the aspect ration of the uninterrupted flap configutration is a lot higher ).
-On the other hand, Boeing's leading edge device configuration seems to bez able to generate more lift...
So where is the comparison, I'm curious.
I personally don't have any info on the complexity of the mechanical designs.



Contrail designer
User currently offlinedlednicer From United States of America, joined May 2005, 547 posts, RR: 7
Reply 2, posted (2 years 7 months 1 week 5 days 10 hours ago) and read 29321 times:
AIRLINERS.NET CREW
DATABASE EDITOR

Quoting Pihero (Reply 1):
I personally don't have any info on the complexity of the mechanical designs.

You might be interested in these documents:
High-Lift Systems on Commercial Subsonic Airliners
Mechanical Design of High Lift Systems for High Aspect Ratio Swept Wings

Quoting ferpe (Thread starter):
using Boeing Wimpress for the 767

Funny enough, a couple of years ago I mentioned to Jack Wimpress (the guy its named after) that Boeing uses the "Wimpress area" to define the area of wing planforms and he expressed surprise - he wasn't aware that there was any such thing.


User currently onlinemandala499 From Indonesia, joined Aug 2001, 6956 posts, RR: 76
Reply 3, posted (2 years 7 months 1 week 5 days 5 hours ago) and read 29232 times:

Ferpe,
Fascinating ! And we haven't gone into the Krueger vs Slats and VGs vs clean wing debate which would explain:

Quoting Pihero (Reply 1):
The big difference is - and has always been - in the low speed, wing high lift devices :
- the Airbus uninterrupted trailing edge flaps generates more lift and less drag than the separated inner / outerr flaps plus an engine gate, high and low speed ailerons (the aspect ration of the uninterrupted flap configutration is a lot higher ).
-On the other hand, Boeing's leading edge device configuration seems to bez able to generate more lift...

  

Now, are patents driving these choices?

Mandala499



When losing situational awareness, pray Cumulus Granitus isn't nearby !
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 4, posted (2 years 7 months 1 week 4 days 22 hours ago) and read 29126 times:

Quoting dlednicer (Reply 2):
Funny enough, a couple of years ago I mentioned to Jack Wimpress (the guy its named after) that Boeing uses the "Wimpress area" to define the area of wing planforms and he expressed surprise

Before we dig into the differences can anyone (OldAeroboy?) explain why Boeing uses such a intricate scheme to define a reference area, it kind of factors down the body covered area of the Yehudi (why) and not the wing proper (why not if the if you factor the yehudi?)



Non French in France
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 5, posted (2 years 7 months 1 week 2 days 6 hours ago) and read 28750 times:

The mystery Wimpress seems to stay with us for a while, then we can cut into the meatier stuff  Wow! . To start things right of the bat I have copied a region of the little excel model I have fiddled with the last month or so after this B vs A apples-oranges wing mystery triggered me  :



What we see here is a snapshot of a spec range flight with the "maxPaxnoCargo 3" class load. The fuel is burned to Cl 0.5 for the 788 (83% which is about average fuel load for the spec mission for these frames which are close to 50% OEW/MTOW) and then all the rest of the frames are compared at this average fuel load. I have only included the FL where Cl is 0.5 for a 83% load (350), the other FL shows the same picture at the Cl 0.5 weights. Cl 0.5 is a typical design point for this class of frames.

Our figure of merit is as before L/D (yellow) and we can see that the new wing for the 777X is the class leader (based on the little data we have, span and 10% larger wingarea then 77W) and that the 350 comes next closely followed by the 788 and jumbo is 77W as is should be (this is clearly not the FL for the 77W, it is at Cl 0.6 and clearly close to buffeting, the 77W should be at FL 330 or even 310 but I kept it here for comparison. I have not added compressibility drag accordingly, to make sense of the 77W range vs model range I had to add some 10 promille comp. drag (compared to below 5 for the rest) which seems to about the design point for the 77W wing). Total drag is highlighted in lbf and the lift = weight (kg) at FL350.

As can be seen the 99% Breguet prediction is pretty close (I have adjusted the climb and reserve fuels to closer to actual then Shevells proposed values) so the model seems to work  . Here we can see that Albaugs statement of 8000nm design point for 777X is sandbagging   , seems to be good for 8500nm  .

So given that all the data is there for the clean wings lets now speculate why this is so (and what are the culprits of my model as per Shevell in the OP)  .

[Edited 2012-04-20 21:55:32]


Non French in France
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 6, posted (2 years 7 months 1 week 2 days 6 hours ago) and read 28720 times:

Just to avoid confusion, the 777-8XL is not the XL, it is the X, the normal -1000N competitor which I took the 65m wing for so they are in the same gate class, the predicted OEWs for the 777X does not come into play other then in the range prediction, not in the L/D. Those can seem high but are realistic, the new engines + wing will come in at about the same as 77W (10% more area and heavier engines due to higher PR) and the fuselage will add length which negates the 5% gain Al-Li parts will give (there other weights like MLW, MZFW is just put there and not very realistic, they don't influence the model, e.g. the 8X MZFW should be higher and so on, I will clean this up ).

The aspect ratio does include the effect of RWT (85% of physical length included in the AR) and winglets (45%) as per the other thread here at Tech/Ops.

[Edited 2012-04-20 23:07:17]


Non French in France
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 7, posted (2 years 7 months 1 week 1 day 13 hours ago) and read 28491 times:

I guess it is not so easy to start commenting on some calculations I have made when you only see a lot of results and have no idea what is behind them. As I had a lot of fun developing the model (very gradually I might say   ) here a little how to and some tips for those who want to have a go for themselves:


- the Shevell course is very good but his notation in the formulas can be a bit cryptic, I found that looking up the same thing in Wikipedia e.g for Induced drag helped a lot. Toorenbeek is a bit much to read for what it gives but it can give some good rule of thumb values for e.g. wing load distribution factor in landing config (0.7).


- One only needs to implement a model for drag really, as cruise lift = weight so once you have drag your have L/D, add TSFC to that and you have the Shevell range formula which is Breguet. Breguet works fine if one is careful with the inputs, the rule of thumb weights he proposes are surprsingly usefull, I got better reserve values from my payload-range chart exercises (where Zeke gave me some tips) but there is less then 5% difference.


- TSFC is tricky though as the engine guys don't like to give it out. PianoX is very good here as Simos knows TSFC in and out (his first Piano was made for RR Derby to model their engines flight behavior). TSFC varies with altitude more then the drag below 36000 feet after which the temp remains constant and TSFC calms down. Therefore using the right average TSFC is important in the Breguet equation. Use PianoX in Point performance mode (with weight to give same thrust for different altitudes) to get a table like this:

..................................788...........767
Cruise TSFC FL390...0,5304.....0,5918
Cruise TSFC FL370...0,5320.....0,5905
Cruise TSFC FL350...0,5375.....0,5977
Cruise TSFC FL330...0,5472.....0,6122
Cruise TSFC FL310...0,5581,,,,,0,6253

Other engines behave similarly e.g. I have reduced these values with 2% to get the 350 table and with 4% to get the 777X table (RR say you gain 1% TSFC per year in engine development typically).


- Once one have the Cl for e.g. the 77W one sees that it has to start it's cruise at FL310 vs 350 for e.g. the 788 or 350, therefore one needs to take a lower average TSFC for 77W, this is more important then the diff in drag or L/D for FL 330-350.


- Tip re Excel, use named cells and areas all the time, it makes things much easier, here the Breguet formula in this way:

=0,99*(Cruise_speed_TAS_kt/Cruise_TSFC_FL370)*L_D_cruise_370*LN(Wi_spec_range/Wf_spec_range)

This is the formula in each cell for the range rown with the typical frame data in the colums. If all the values are in the column with these labels to the left excel will automatically pick the right values. Constants like density for FL370 I have in named cells. The ,99 is for the inefficiency of the step cruise climb.

- some of the parasitic drag stuff one can short-cut in the first go, control surface and nacelle base drag is negligible (nacelle drag not), for the upsweep drag I have measured 0,0006 on the 350 and use that until further for all frames. The nacelles have the same typical length as the wings MAC and about 10% of the area so I just added this to the wing parasitic drag calculation. (4 holers need double dose of course  )

- for induced drag I use 0.9 for the span efficiency factor, it is a good starting point.

- I have put compressibility drag at 0.0005 for all frames except those with very high wingloading (77L, 77W) where I have it at around 0.0010

Much more to say but later... 

[Edited 2012-04-21 15:30:53]


Non French in France
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 8, posted (2 years 7 months 3 days 2 hours ago) and read 28006 times:

So later was quite a bit later    .

Based on initial results I have expanded the model in several areas. I expected to find that frame efficiency improved with increasing FL to an optimum where also L/D was highest and then tapered of. I found the aerodynamic optimum to be rather weak and that the true reason for frames seeking highest possible FL is a mix of aero and engine TSFC reasons. So I modeled in TSFC variantion as well and seeked a better figure of merit then just L/D. Breguet theory sais I can pick the range factor/constant (speed/TSFC * L/D) but even more meaningful is Specific Air Range (SAR) i.e. the Range factor divided by the cruise weight. This gives the nm covered per kg fuel burned at that FL which is a more interesting number then Range factor.

As can be seen from the improved model the optimum FL based on SAR differs from best L/D (all at average cruise weight of 83% of MTOW) for e.g. 788, 358, 359, 777-8X, 332 and 333:

http://i298.photobucket.com/albums/mm262/ferpe_bucket/Aircraftrangemodel788-330.jpg


So based on this model lets focus first of finishing the analysis/discussion of the B vs. A wings in cruise condition (easiest), then go to landing and finally to the most difficult, the start (at least I find it so  ). I will refrain from how the numbers are calculated here and rather make a "DIY Frame analysis tool" thread otherwise I wander of into "amateur analyst wonderland" and we never come to the differences  .

As can be seen the new generation of high aspect ratio CFRP wings creates the basis for efficient frames (777-8X vs 77W for instance, starts at FL 350 instead of 310 (line 106 states initial and mid cruise FL, initial cruise is best SAR at 97% weight with Cl below 0.55), cruises at FL390 instead of 350 at mid cruise weight and there covers 0,075 nm/kg fuel instead of 0,062 ).

Also look at the 358 vs 789, they haul 270 vs. 280 Pax but the 358 who has a shorter fuselage with a lower wetted area (line 34) has the same SAR as the 789 despite 2% more efficient engines. Clearly the wing is a bit to large for this frame which can also be seen that the Dp is clearly higher the Di (line 127, 128) whereas the 789 has them close together (theoretical frame optimum is when they are equal but that is when one assumes constant TSFC. At FL370 and over TSFC is almost constant and there this rule applies).

One can also see that the 350-1000 wing is a bit stretched, initial cruise lands at FL330 (SAR 0.075) for MTOW flights which is below the more SAR optimal 350-370 FL. So with a clean sheet design the 350-1000 would probably have increased span or larger winglets in addition to it's 4% TE extension (primarily serving Landing and TO performance).

So what is the differences in the cruise wings between A and B? I would say smaller then within their own ranges i.e. 787, 350 and 777, in cruise config the 787 and 350 generation wings are pretty similar at least at this analysis depth (non CFD preliminary sizing theory), next level would involve supercritical wing design knowledge and I don't think the framers let us in on that  Wow! .

But there is a surprising level of behind the scenes things one can reveal with this simple model when trying to fit the model range with the OEM spec range, there one sees for instance that A counts on the 350 having a rather smooth surface whereas B has learned that the 787 really is not that major improvement  Wow! (or I have the OEW or TSFCs a bit wrong  Wow!  ) .

On a more general sense a good first analysis of a wing/frame combo is checking the wingload and spanload (MTOW/effective span, I will add it to my table). They give a good hint on initial and subsequent cruise FLs in order to get those touchy engines to breathe their preferred species of air  Wow!   .

[Edited 2012-04-27 03:09:39]


Non French in France
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 9, posted (2 years 7 months 1 day 22 hours ago) and read 27701 times:

Anyone in the know who can comment on the method I have used to compare the frames and their wings? Rather straight forward or?


Non French in France
User currently offlineCM From , joined Dec 1969, posts, RR:
Reply 10, posted (2 years 7 months 1 day 9 hours ago) and read 27561 times:

Hi Ferpe. I'm just dropping you a note to encourage you to keep adding to this thread. I know it seems like you are talking to yourself, but I (and I am sure others) am finding the topic really interesting; I just haven't hand anything of value to contribute... yet.

User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 11, posted (2 years 7 months 1 day 3 hours ago) and read 27489 times:

Thanks CM    . I could see that there where readers of the thread but not sure if it was of value to anyone, so feel motivated to take it to the meaty bits (after finishing of the cruise part below )   .


As Pihero so rightly stated:

"IMO, difficult to differentiate between the twp designs in clean conditions. They are both high-aspect ratio with variable -geometry camber, high aeroelasticity wings.
The big difference is - and has always been - in the low speed, wing high lift devices : "

we can see just that, it seems the wings are pretty similar in clean wing conditions:


SURPRISED
This is not quite what I expected when I started building the tool but the wingaera discovery kind of gave a hunch they could be more similar then thought and the deeper poking with the excel seems to confirm that. I was pretty sure the drag and L/D values would give some clear insights into e.g. the somewhat large size of the 358 wing and I expected that to be contrasted with a bit to small 789 wing. Both things were confirmed, the 358 has more parasitic drag then induced drag at optimum cruise level and the 789 has these at the theoretical optimum (see graph) but the wingloading causes it to start the cruise at the less then optimal FL330, something it shares with it's stretch cousin, the 350-1000.

http://i298.photobucket.com/albums/mm262/ferpe_bucket/Totaldragcurve.jpg



THEORETICAL OPTIMAL FL
Classical theory, which does not cater for transonic effects, says you can increase the lift until optimal Cl (given in Roskam as an optimal seeking L/D formula) which is somewhere around 0.65 or even more and thus a wing with smallest possible wing area and broadest span is the best choice for cruise. But that is ignoring transonic effects, a smaller wingarea forces a larger pressure differential and the ensuing lower pressure on the wing overside means higher local air speed = more supersonic area = compressive drag rises fast. PianoX seems to stay away from that by the rule of keeping the Cl below 0.55 and I have adopted his hint  being pretty blind when it comes to transonic effects.



BREGUET RANGE
One thing that has really surprised me was the power of the range calculation. I have learned that the Breguet range equation was kind of the "cheaty" way of calculating range (assuming all things constant except fuel=weight burn of) however it seems even Piano uses it but in a more clever way. As Shevell, Roskam etc states, if things do not remain constant (TSFC, TAS as a function of M below 36kft ) or you are forced to fly at constant FL (Breguet assumes you can always slowly cruise climb to keep Cl and therefore L/D constant) you just hack the flight into many Breguets each having it's new set of constant values and its start and stop weights. Hack it fine enough and you come close to reality, IMO this is what Piano does. I use the excel to find the average flight values instead and does only one Breguet with these, seems to be good enough to see what is interesting, the frames that fall out of the raster    .



RASTER FALLING   
This is a signal to try and understand what is not plain vanilla with the frame. Can be wrong assumptions on OEW (=how much fuel to burn) or TSFC (wrong MPG) but can also mean the framer assumes a very good roughness constant (350?) i.e. smooth skin due to e.g. CFRP construction. The 788 lost some 500nm between first ACAP and flight tests, 300 of those one can see came from weight but 200nm come from something else. The model suggest a parasitic drag increase. Now why wouldn't it be faulty TSFC or OEW?

TSFC
All 787 use the same engines so if I have the TSFC to low all should be better in range then the model, they are not, the 789 and 7810 are under. But the 789 and 7810 have laminar flow tails so I allocated a 30% friction drag reduction on the laminar flow tails and that seemed to be somewhere right, it made all the 787s fit the range model with std roughness factor. Given that 787 have two more air condition inlets I then reasoned; all in all they probably are comparable in the roughness factor to say the 777 as this factor includes all effects of skin roughness, intakes,outlets etc and intakes/outlets are draggy items and before things like frames etc are cocured that skin is only marginally saved from fasteners/rivets. The 787 range seem to fit the model with std roughness (which is probably a bit worse then B though 2005) and a 30% improvement of friction factor for the 789, 7810 VTP, HTP.

OEW
The latest ACAP OEW was 111.5t for a 224 seat 788, I compare with 250 seats as there is an official range figure there (7650nm). Add the 1.5t for the seat diff and we have 113t official OEW. Until there is more plausible evidence I can't place this figure and the model seems to support that, first at my 116t the range fits. If I put in the 113t we see 8000nm    something B would love to have seen in the NAMS testing   .


  
So range fitting gives good food for though and signals one to go hunting for reasons, perhaps one finds something    .



LANDING CASE   
After much ado on to the landing case. I am not on top what requirements the final approach has, kind of heard below 150kt but what other criteria would there be?

[Edited 2012-04-29 00:47:18]


Non French in France
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 12, posted (2 years 7 months ago) and read 27250 times:

The textbooks covers landing distance etc (here the Shevell course http://adg.stanford.edu/aa241/AircraftDesign.html ) but I will not venture into that, it is to complicated and beyond the scope of this thread, I will only look at the approach speed and what lift coefficient the wings would have to produce to fly these approach speeds safely.

According to the Schevell course FAR 25.125. states:

(2) A stabilized approach, with a calibrated airspeed of not less than
1.3 VS, must be maintained down to the 50 foot height.


So given that I have read that A and B wanted the 350-1000 and 77W to have an approach speed lower then 150kt I will use 150/1.33 = 113kt as the dimensioning speed for the landing case. At the MLW the discussed wings then need to produce the Cl in the table:



I have not included the drag data as this is only available with any accuracy via wind tunnel test even today according to the textbooks and A and B. Further, drag (as far as I know) is not necessarily avoided in landing config, you want a rather draggy config to have reasonable power on so that your power reduction at flare creates a positive deceleration and defined touchdown. Also having the engines at higher revs is desirable for the go around case, the time to TOGA config will be reduced. Once you go to that TOGA config (which should be similar to start config) you are interested in good L/D again and therefore low drag as margin for the engine out case, more on that in the start post.

Given the lower interest for maximum L/D the drawback of the B principle with a flaperon between the inboard and outboard flaps will be less critical, it will affect the spanload diagram somewhat and hence agument the induced drag but not on a scale where it will be a problem IMO. The start case is touchier as will be seen.


The required Vstall lift coefficients allow the modern wings to use simpler drooped hing flap mechanisms as discussed in the A350 thread, the highest Cl is for the 789/7810 to produce at 2.60. We know from Dynamicsguy that requires higher deployment angels and therefore reinforced flaps but the simple drooped principle is still OK.

The 77W need more Cl and did not have the advantage of spoilers that could droop and control the flap-wing slot, therefore a complexer Fowler arrangement was needed. One can also see that the -8X and -9X could live with drooped flaps for landing, it might be another discussion for the start engine out case given the rather low TOpower/MTOW ratio   for those frames.

[Edited 2012-04-30 04:29:03]


Non French in France
User currently onlinePihero From France, joined Jan 2005, 4630 posts, RR: 77
Reply 13, posted (2 years 6 months 4 weeks 1 day 23 hours ago) and read 27227 times:
Support Airliners.net - become a First Class Member!

Hello, Ferpe !
Fascinating study and thanks for putting so much time into it.
A few remarks, though :
1/- The basis for Airbus performance speeds is the Vs1g for all FBW airplanes. The concept is being taken slowly by Boeing into consideration, but AFAIK, the FAR 1.3 Vs is still current

2/6 The values of Clapp - don't think of anything but the lift coeff in approach (!!!) seem awfully high, especially for the 777 for which we have the data from the NASA study we've referred to already and which stands at Clapp = 1.5.
I do not believe one second that, considering the less than optimum solution (exhaust gate and mutiple ailerons), modern airplanes haven't improved on that.
I would admit, though to that value for a clean config at opt AoA.

I buy the champ if wrong.

Cheers !



Contrail designer
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 14, posted (2 years 6 months 4 weeks 1 day 20 hours ago) and read 27157 times:

Quoting Pihero (Reply 13):
The values of Clapp - don't think of anything but the lift coeff in approach (!!!) seem awfully high, especially for the 777 for which we have the data from the NASA study we've referred to already and which stands at Clapp = 1.5.

2) You were right, what I had in the table wrongly labeled was Clmax @ Vs where the frames have to show in flight tests (I presume) that they don't stall until 115kt in landing config. Here a better table with both Cl @ 150kt and @ 115kt. Airbus or Trapezodial at the end of the line stands for Airbus or Trapezodial measurement rules for the wingarea, I use this as PianoX use Airbus for Airbus planes and Wimpress for Boeing planes and that any other info most likely can be in Trapezodial like the NASA one you mentioned. Wimpress is only some 3% less then Airbus so I use the one where we have more data which is Airbus. By using the right reference area one can compare the Cl values (with Piano and other sources):



I have not seen the NASA study you referred to, can you give the link?


1) Could you explain the Airbus performance speeds a bit more? What does Vs 1g mean?

[Edited 2012-04-30 08:32:40]


Non French in France
User currently onlinePihero From France, joined Jan 2005, 4630 posts, RR: 77
Reply 15, posted (2 years 6 months 4 weeks 1 day 19 hours ago) and read 27142 times:
Support Airliners.net - become a First Class Member!

Quoting ferpe (Reply 14):

I have not seen the NASA study you referred to, can you give the link?

here it is :Rudolph study on hi lift dev

Quoting ferpe (Reply 14):
Could you explain the Airbus performance speeds a bit more? What does Vs 1g mean?

try this link :
getting to grips with aircraft performance
click on the screen icon to open the document.

on page 35, 36,; you have the definition and the application to approach speeds on page 113.



Contrail designer
User currently offlineOldAeroGuy From United States of America, joined Dec 2004, 3574 posts, RR: 67
Reply 16, posted (2 years 6 months 4 weeks 1 day 19 hours ago) and read 27136 times:

Quoting Pihero (Reply 13):
A few remarks, though :
1/- The basis for Airbus performance speeds is the Vs1g for all FBW airplanes. The concept is being taken slowly by Boeing into consideration, but AFAIK, the FAR 1.3 Vs is still current

The last Boeing airplane to use 1.3 Vs for approach speed was the 737CFM Classic in the mid-1980's.

All subsequent models have used Vs1g and the 1.23 factor for Vapp.



Airplane design is easy, the difficulty is getting them to fly - Barnes Wallis
User currently onlinePihero From France, joined Jan 2005, 4630 posts, RR: 77
Reply 17, posted (2 years 6 months 4 weeks 1 day 18 hours ago) and read 27095 times:
Support Airliners.net - become a First Class Member!

Quoting OldAeroGuy (Reply 16):

The last Boeing airplane to use 1.3 Vs for approach speed was the 737CFM Classic in the mid-1980's.
All subsequent models have used Vs1g and the 1.23 factor for Vapp.

I'd really like to see the proof of your statement as in 2000 I taught a course on aircraft performance in which FARs were still related to the Vsr = Vclmax / sqrt N where N is always and thedefinition was left to the manufacturer.
Reading the FARs is a bit confusing as the old notions do not seem forgotten as notations as Vs still abound.

But maybe I need to update my library.



Contrail designer
User currently offlineCM From , joined Dec 1969, posts, RR:
Reply 18, posted (2 years 6 months 4 weeks 1 day 17 hours ago) and read 27087 times:

Quoting Pihero (Reply 17):
I'd really like to see the proof of your statement

Just one example: Page 7 of the 777-200/300 TCDS under the Cert Basis - Equivalent Safety Findings section

Quote:"14 CFR 25. (several sections) Use of 1g Speed Instead of Minimum Speed in the Stall as a Basis for Compliance. (All 14 CFR 25 Sections, except structural, dealing with stall speeds/related factors for turbojet airplanes)."

You can find the TCDS here:
http://rgl.faa.gov/Regulatory_and_Gu...02e6/$FILE/T00001SE%20Rev%2029.pdf


User currently onlinePihero From France, joined Jan 2005, 4630 posts, RR: 77
Reply 19, posted (2 years 6 months 4 weeks 1 day 17 hours ago) and read 27081 times:
Support Airliners.net - become a First Class Member!

The date is Sep 2005


Contrail designer
User currently offlineCM From , joined Dec 1969, posts, RR:
Reply 20, posted (2 years 6 months 4 weeks 1 day 17 hours ago) and read 27074 times:

Quoting Pihero (Reply 19):
The date is Sep 2005

Not sure I am following you, Pihero. Are you saying you believe the equivalent safety finding for 777 stall speed was granted in September 2005?


User currently onlinePihero From France, joined Jan 2005, 4630 posts, RR: 77
Reply 21, posted (2 years 6 months 4 weeks 1 day 15 hours ago) and read 27019 times:
Support Airliners.net - become a First Class Member!

Don't know. Is it the original TCDS ?
On the same subject, comes to my mind the fact that , for Airbus and JAR, the Vs1g concept was for FBW aircraft only... When submitted to the FAA, it was accepted in the US with a proviso that Vs1g = ;94 Vs in order to keep some sort of continuity in perf determination ( so that 1.23 Vs1g = 1.3 Vs).
Therefore, when I did my course - and I clearly remember referencing on FAR and JAR books - the certification on US planes was on old Vs, although I confess the T7 did not entert the discussion. The 737, yes, but to which point , or type rather.
Tom, please enlighten me !



Contrail designer
User currently offlineCM From , joined Dec 1969, posts, RR:
Reply 22, posted (2 years 6 months 4 weeks 1 day 15 hours ago) and read 27007 times:

Quoting Pihero (Reply 21):
Don't know. Is it the original TCDS ?

Unless otherwise noted in the TCDS, items under the Cert Basis section are deviations from the Cert Basis Amendment Level used for certification of the type. When an Equivalent Safety finding or Special Condition has been granted subsequent to Type Cert, it will always note the date. In the case of the Equivalent Safety Finding for the 777 to use 1g speed instead of minimum speed in the stall as a basis for compliance, no date is noted. This means the Equivalent Safety Finding was granted with the original TCDS, dated April 19, 1995.


User currently onlinePihero From France, joined Jan 2005, 4630 posts, RR: 77
Reply 23, posted (2 years 6 months 4 weeks 1 day 14 hours ago) and read 26983 times:
Support Airliners.net - become a First Class Member!

Quoting CM (Reply 22):
In the case of the Equivalent Safety Finding for the 777 to use 1g speed instead of minimum speed in the stall as a basis for compliance, no date is noted. This means the Equivalent Safety Finding was granted with the original TCDS, dated April 19, 1995.

OK. But for some reason, in the yeart 2000 and for AI in 2002 (date of their "get to grip with Aircraft Performance ", the FAR document was still : FAR 25.103 Stalling speed :
(a) VS is the calibrated stalling speed, or the minimum steady flight speed, in knots, at
which the airplane is controllable, with Zero thrust at the stalling speed, or […] with
engines idling”.
FAR 25 doesn’t make any reference to the 1-g stall speed requirement.
Nevertheless, Airbus fly-by-wire aircraft have been approved by the FAA, under
special conditions and similarly to JAA approval, with VS1g as the reference stall
speed.

On the basis of that FAR in 2000, I have assuimed that Vs1g is something very new ( thought about the 787 cert, and not before.
I guess I may owe you guys a beer.



Contrail designer
User currently offlineCM From , joined Dec 1969, posts, RR:
Reply 24, posted (2 years 6 months 4 weeks 1 day 13 hours ago) and read 26960 times:

Using 1g speed instead of minimum speed in the stall as a basis for compliance is just one example of a larger trend by both OEMs; they are increasingly willing to deviate from the published regulations (whether it be via special conditions, equivalent safety findings, or other mechanisms) if it will help them to eliminate unnecessary margin in their designs.

Something which most people don't think about is the fact that airplane designs and OEM's understanding of performance are way out ahead of the regulators. The FARs lag the state-of-the-art in the industry by a considerable measure and an aircraft designed precisely to the letter of the FARs would accept many needless limitations. The OEMs make an effort to educate the regulators and to help them on a path for meaningful reform of the regulations, but it is an unduly bureaucratic process, meaning it is tedious, deliberate and very slow.

The lack of technical depth in the regulators is a risk to the OEMs in the sense there is always a chance it will end up limiting the product. It is not uncommon for a lack of understanding on the part of the cert agency to result in artificial limitations or other kinds of unwarranted conservatisms being imposed an airplane. This lack of expertise within the regulators combined with increasing aircraft complexity is the reason delegated authority is so important to the OEMs. Often the OEMs are the only ones who truly understand their design.


User currently offlineOldAeroGuy From United States of America, joined Dec 2004, 3574 posts, RR: 67
Reply 25, posted (2 years 6 months 4 weeks 1 day 13 hours ago) and read 27246 times:

Quoting Pihero (Reply 21):
The 737, yes, but to which point , or type rather.

Don't have access to the 737NG TCDS where I am right now, but I did play a key role in the 737NG Part 25 Aerodynamic Certification.

Remember that it certified after the 777 discussed above. I can assure you that it also used "1g" stall speeds. The last 737 to use Vmin stall speeds was the 737CFM Classic that certified in the mid-1980's.



Airplane design is easy, the difficulty is getting them to fly - Barnes Wallis
User currently onlinePihero From France, joined Jan 2005, 4630 posts, RR: 77
Reply 26, posted (2 years 6 months 4 weeks 1 day 13 hours ago) and read 27210 times:
Support Airliners.net - become a First Class Member!

Quoting OldAeroGuy (Reply 25):
I did play a key role in the 737NG Part 25 Aerodynamic Certification.

Respect where it's due. And thanks for the infos this site badly needs

Quoting CM (Reply 24):
This lack of expertise within the regulators combined with increasing aircraft complexity is the reason delegated authority is so important to the OEMs. Often the OEMs are the only ones who truly understand their design.

So I understand better the liberty the FAA gives the "applicant" for - in this case - the definition of stall speed.
CM, how quickly you've become a mine of explanations on perf and cert !



Contrail designer
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 27, posted (2 years 6 months 4 weeks 1 day 4 hours ago) and read 27282 times:

If I take 150kt approach speed and divide with 1.3 or 1.23/0.94 I get the same result, 115kt.

I understand that Vs1g is a better definition as Vs can be fuzzy (re if the inboard wing starts buffeting before the outerwing, is this stall or not) but apart from that, what is the big gain with the more modern definition, precision?

Seems not to lower the margins significantly or am I missing something?



Non French in France
User currently onlinePihero From France, joined Jan 2005, 4630 posts, RR: 77
Reply 28, posted (2 years 6 months 4 weeks 1 day ago) and read 27205 times:
Support Airliners.net - become a First Class Member!

Quoting ferpe (Reply 27):
f I take 150kt approach speed and divide with 1.3 or 1.23/0.94 I get the same result, 115kt.

That's what they were aiming for, for "continuity"'s sake.

Quoting ferpe (Reply 27):
what is the big gain with the more modern definition, precision?

Yep ! All protections come from that - for once - accurate determination of "stall speed".



Contrail designer
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 29, posted (2 years 6 months 4 weeks 2 hours ago) and read 27029 times:

The NASA Rudolp study of high lift devices is quite interesting, here the essential diagram where he compares the then prevailing aircraft in approach configuration and their approach speeds (I have added MLW/Wingare in kg/m2 in the diagram):

http://i298.photobucket.com/albums/mm262/ferpe_bucket/NASAhighliftdiagram.jpg

For the Boeing and Airbus models I have verified that he uses Wimpress and Airbus wingareas, thus the Clapp values are comparable (3% diff only), the approach speeds at MLW seems to center around 137kt for the first member of a new family thus I repost my table with this as the desirable approach speeds, I also added the MLW/Wingarea as lb/ft2:

http://i298.photobucket.com/albums/mm262/ferpe_bucket/Analysisof787350777Approach1.jpg

I now have Vs1g as ultimate test what the wings in max high-lift config has to produce.

The final parts of the Rudolp report makes it clear he addresses the US Airline industry and it is interesting that his recommendations for a simpler, less aggressive slat to go with a refined single slotted flap using the spoiler as a slot control element was the way that the 787 and 350 went. Boeing did not follow the advice of dropping the high-speed aileron to get an un-interrupted trailing edge however.

[Edited 2012-05-02 02:18:54]


Non French in France
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 30, posted (2 years 6 months 3 weeks 5 days 12 hours ago) and read 26643 times:

And so to the trickiest part of the analysis (framers! we need your windtunnel data   ) but also perhaps the most interesting, cause here the differences between the wings will start to show. For the start phase we will put all aircraft at climbout at speed V2, directly after liftoff, at 35ft and gear in. I have chosen 160kt as V2, it might be a bit tough for some as we will see but seems to be the most common one.

For the start phase drag counts, big time. Highest possible L/D will increase the possible payload, here a statement from B. Neild (Boeing) presenting the 777 high-lift solution in Aeronautical Journal (also referenced in the NASA Rudolph study):

“A 1% increase in take-off L/D is equivalent to a 2800 lb increase in payload or a 150 nm increase in range.”

For the start I have tried to model the drag as best can, we need it. As lift I have aircraft weight + tail trim load, I have not modeled the extra power needed for the climb angle ( later if need be ). The tricky part is to estimate the drag of the high lift configs used by the different aircraft, I have deliberately stayed away from trying to model As continuous TE vs Bs free flowing inboard aileron/thrust gate. It would only put in a range of arbritrary speculations as to what drag values it creates and how it affects the span efficiency. Instead I present different scenarios which we can discuss round and everyone can interpret the consequences of the different solutions chosen by the OEMs based on the data I provide.

To get a baseline drag increase for the high lift configs I had to go to "Aircraft Design" by Kundu, it is a well written and practical aircraft design book, a bit like a modern Torenbeek. He covers drag from high-lift devices on pages 282-286. For our purposes we pick 3 informations:

- slats add 20 counts parasitic drag (drag is discussed in counts which is % of % ie 0.0001 is a count). I would say this is slats in take-off position, ie non slotted.

- Dropped hinge flaps (787, 350) add 10 counts per degree flap (flap drag is directly proportional to how much they protrude into the wind, dropped hinge has a constant arc deflection).

- Well made fowler flaps drops less into the wind for take-off positions (mostly fowler motion up to 20° flap), therefore they add 5 counts per degree deployment.

I include Kundus interference drag in these figures, the increased induced drag from the high lift devices we cover in the span efficiency factor ( e, wrongly called span load factor in my tables, changed now  Wow! ). I have pasted together 3 variant of the tables, first with nominal values, then with 777-8 and 9X using dropped hinge flaps instead of my assumed fowler and finally a table with a degraded span efficiency factor of 0.85. The interesting values and changes are in red rectangles and in drag is in yellow. So after much preparatory aero speak here the start tables (click on the table to read it magnified):

http://i298.photobucket.com/albums/mm262/ferpe_bucket/StartV21engineanalysis787-330_.jpg

The lines in the large red rectangles are the excess thrust which is left at V2 should one engine fall out. This excess thrust shall cover stopped engine drag, drag from yaw rudder compensation and climb angle after engine failure.

As can be seen the -8X with it's 88klbf engines falls out of the picture, it has almost no excess thrust especially with dropped hinge flaps or a degraded effiecincy factor (inboard aileron). No wonder it has the same span as the -1000, 7 ton more weight and 9klbf less thrust. This frame will have to use the 68m wing but me thinks the derate will be less as well. The 9X is pretty OK despite 244t and 99klbf but that is assuming the wing will have fowler flaps and I would not be surpirsed to see the inboard ailerons go to outboard like Airbus to get max e factor. Let me just note that I don't know if 0.9 and 0.85 are the right factors or even the difference between A and B, it might be more like a couple of percent. 0.9 is a standard textbook factor and then I provided 0.85 to show a changed value.

If the Airbus droop nose inboard of the engines is worth 5 counts drag it will produce that 1% extra L/D as described by Nield .

So why have B stayed with the inboard high speed aileron for the 787? Here my shot, in the winglet vs raked tip (RWT) thread CM described how the RWT produces less wing torsion load and therefore less heavy wing for same induced drag reduction. An outboard high-speed aileron forces a torsional stiffer wing, so the 350 winglet + outboard aileron kind of goes together as does the RWT and inboard highspeed aileron.

But now for CM and others to comment  Wow!

[Edited 2012-05-03 16:11:05]


Non French in France
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 31, posted (2 years 6 months 3 weeks 5 days 3 hours ago) and read 26489 times:

For clarity the start flap setting in the tables are at equivalent Cl increase of about 1, this is a main flap element deployment angle for a hinged flap of approx 20° ( see Kundu p 284 ). The fowler motion vs flap angle diagram on page 115 in Rudolph shows that at a simple hinge flap angle of 20° the fowler motion is some 7% vs 10% for the 777 and 12,5% for the 330/340:


http://i298.photobucket.com/albums/mm262/ferpe_bucket/FowlermotionfromNASAstudy.jpg



It is pretty clear what those massive G90-115 has cut our for them when the 77W is at MTOW on a short runway   .

As can be seen in the table the 330 is up there with the best in start L/D, a class act from a 1990ies Alu design. It achieves this table L/D without me giving it credit for it's superior fowler wing area increase (over the 787/350 dropped hinge) when calculating the start Cl and thus the 5% * 2/3 ( flap span fraction) less delta Cl needed from deployed slats/flaps.

As stated before I have consistently used the reference wingarea to show the Cl needed at different speeds, the measurement rule follows the Airbus method which is very close to the Boeing method (Bs method gives about 3% lower area). This method of using the same reference area for cruise, approach and start makes all Cl values comparable, increases in usefull wingarea then shows up as an increase in Cl capability for the wings that benefit from strong fowler motion (LE and TE). Alternatively one could use the fowler influenced wing area for start and landing to better compare between the models but the above seems to be the industry practice.

[Edited 2012-05-04 00:56:19]


Non French in France
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 32, posted (2 years 6 months 3 weeks 4 days 3 hours ago) and read 26359 times:

Some more tidbits on start and landing;


WHY NO KRÜGERS ANY MORE?   
Quoting mandala499 (Reply 3):
Ferpe,
Fascinating ! And we haven't gone into the Krueger vs Slats and VGs vs clean wing debate

In the Airbus vs Boeing thread Mandala499 and me tried to understand why A did not use Krueger flaps any more (used as gap fillers on A300 and 310), we speculated there might be some mechanism patent problem vs Boeing. The Rudolph paper gives a simpler answer, Krueger has the advantage of simplicity and non interference with the wingnose overside but then many disadvantages for a high performing wing. The deployment forces are high, the shape is dictated by the undersides shape and therefore generates a less then ideal start shape and flap-wing transfer of the airstream. The first can be fixed but at the cost of simplicity but the second is what virtually kills it once the elegant 757 rack and pinion slat drive was invented. A 3 position slat is just so much better at the LE, the Krüger is now relegated to shot-gun services like gapfilling around pylons and such    .



THE 777 FOWLER
The less then stellar fowler motion of the present 777 flaps makes makes the 0,0100 Cdp a bit flattery, if one put in something like 150 counts instead the L/D declines another 1.3%. I don't think Boeing will keep this mechanism for the 777X and would therefore leave the -X at 100 counts.



INBOARD HIGH-SPEED AILERONS LOSSES
I have found one indication of what one looses in Cl with an inboard aileron, in Van Dams grand summary of "The aerodynamic design of multi-element high-lift systems for transport airplanes"

https://docs.google.com/a/marketlogicsoftware.com/viewer?a=v&q=cache:UQ-AC63BE3oJ:cafefoundation.org/v2/pdf_tech/High.Lift/Elsevier.Prog.AS.HighLift.vDam.02.pdf+&hl=fr&gl=fr&pid=bl&srcid=ADGEESiapSjJIutb7jsVYbFZ1vpOMKr6EQT6cX81bcnMODNTy7-E9J_EbE5aAa7nfPV1WaPdMKRZNfMS8PowFk6sinPIhi-slomdYBlIhvq34i6aohM9EieVzenEiVrJP4l0vvRmazkK&sig=AHIEtbRQvJ_mI8iM30XxcqccQZQTHY_k3w&pli=1

On page 134 (he starts at 100   ) he has this diagram:



Not quite modern times    but we can deduce the scale of the Cl loss, for the drag I can find no source. In the case of introducing a inboard high speed aileron one lost 0.4 Clapp and by moving the high speed aileron from outside the flap to between the flaps one loose 0.2 Clapp. Thus for a moden FBW desing where one can droop an outside aileron lets say one loose something like 0.2-0.3 Clapp. For the landing case the loss in L/D is not important but the loss in ultimate Clapp might squeese things a bit. The more important loss is in attitude, the inboard aileron requires another 2-3° Alfa (the dropped hinge and the fowler flaps gains about 0.1 Cl per degree Alfa ).

For the max start case requiring Boeing flaps 20 I would halve this effect, ie we loose some 0.1-0.15 Cl and need another 1-1.5% Alfa at liftoff. The typical L/D of 10 would take a 6-9 % hit (Cl typically 1.6 as per table and we loose 0.1-0.15 ), a significant loss before counting any drag increase   .



Non French in France
User currently offlineOldAeroGuy From United States of America, joined Dec 2004, 3574 posts, RR: 67
Reply 33, posted (2 years 6 months 3 weeks 3 days 20 hours ago) and read 26310 times:

Quoting ferpe (Reply 30):
As can be seen the -8X with it's 88klbf engines falls out of the picture, it has almost no excess thrust especially with dropped hinge flaps or a degraded effiecincy factor (inboard aileron). No wonder it has the same span as the -1000, 7 ton more weight and 9klbf less thrust. This frame will have to use the 68m wing but me thinks the derate will be less as well.

I think you're using the wrong wing definition for the -8X. From what I've read, it will have the same wing as the -9X. This would make sense from a design/manufacturing standpoint. You'd be designing the same basic wing planform for the two new derivatives. If you use the larger wing, the lower thrust rating works.

With regard to the inboard aileron, lift carry over losses and the resulting induced drag increase can be reduced if you use a flaperon (ala 777/787) rather than a simple alieron.

At the same time, overall flaps down induced drag can be improved by extending the outboard flap as a flap rather than relying on a dropped outboard aileron. As always, wing design is a series of trade-offs and compromises.

You're doing a great job with this thread. It's nice to see data rather than the usual arm waving.



Airplane design is easy, the difficulty is getting them to fly - Barnes Wallis
User currently offlineOldAeroGuy From United States of America, joined Dec 2004, 3574 posts, RR: 67
Reply 34, posted (2 years 6 months 3 weeks 3 days 19 hours ago) and read 26275 times:

Quoting ferpe (Reply 4):
Before we dig into the differences can anyone (OldAeroboy?) explain why Boeing uses such a intricate scheme to define a reference area, it kind of factors down the body covered area of the Yehudi (why) and not the wing proper (why not if the if you factor the yehudi?)

The Wimpress area factors the non-trapezoidal parts of the wing area as projected into the body as a function of the exposed span of the non-trapezoidal region (e.g. a Yehudi).

It's an attempt to normalize the wing area effects of mulitple leading edge (gloves) and trailing edge (Yehudis) planform breaks.

All the exposed wing area is counted but only part of the area projected into the body is added to the total wing reference area.



Airplane design is easy, the difficulty is getting them to fly - Barnes Wallis
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 35, posted (2 years 6 months 3 weeks 3 days 15 hours ago) and read 26196 times:

Quoting OldAeroGuy (Reply 33):
I think you're using the wrong wing definition for the -8X. From what I've read, it will have the same wing as the -9X. This would make sense from a design/manufacturing standpoint. You'd be designing the same basic wing planform for the two new derivatives. If you use the larger wing, the lower thrust rating works.

This has been my conclusion as well, no point in doing 2 wings for 2 models, especially if the 71m wingspan gate problem is solved by folding the raked wingtips, might as well use this solution then for both frames. Nice that the above analysis could basically put to bed the Ostrower reporting of 3 different winspans that where on study, 65 and 68 with winglets and 71 with RWT. Here the table with the -8X using the 71m wing:

http://i298.photobucket.com/albums/mm262/ferpe_bucket/Analysisof787350777-8X71mwing.jpg

Now one can see that the start thrust excess is the same as -9X at 1.17 so this seems to be the wing for the quoted 88klbf thrust for the -8X. The funny thing is that the larger wetted area and lower wingloading (Dp some 20% larger then Di at cruise just like the 358) actually reduces the optimum midweight cruise level from 390 to 370. The specific air range (nm per ton, I have change to ton instead of kg, makes the nm covered easier to grasp) has gone up from 74.9 to 81 and the projected range goes from 8300nm to 8900nm, this is definitely the wing to choose for the -8X. I might now have underestimated the OEW a bit given the larger wing span, any increase would only shorten the ample range with some 100-200 nm however.

Quoting OldAeroGuy (Reply 33):
With regard to the inboard aileron, lift carry over losses and the resulting induced drag increase can be reduced if you use a flaperon (ala 777/787) rather than a simple alieron.

This is correct for landing, the 777 and 787 inboard aileron is a single slotted flaperon in such case and the span efficiency does not fall that much especially on the 787 who matches that with a single slotted dropped hinge flap each side. Dynamics guy have said however that the flaperon is left free flowing (no hydralic pressure) for start, ie it leaves as hole in the span load map and creates interference drag due to vortices and separations on both flap sides besides it. So for the L/D critical start case the flaperon does not help, that is why I only proportioned the 727 result for the start case, it would not apply for the approach case.

[Edited 2012-05-05 13:18:19]


Non French in France
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 36, posted (2 years 6 months 3 weeks 3 days 15 hours ago) and read 26184 times:

Quoting OldAeroGuy (Reply 34):
All the exposed wing area is counted but only part of the area projected into the body is added to the total wing reference area.

I have gradually come to understand that, initially I wondered why not use only he exposed area but then saw this interesting picture in a 787 presentation:



Clearly the wing induces a low pressure area onto the fuselage inside the wing halves, the Trapezodial, Airbus and Boeing Wimpress wingarea methods values this resulting lift force in somewhat different ways.

Re Wimpress I just tried to grasp the logic behind the ratio used, the exposed Yuhedi LE span divided by the exposed TE span, still can't figure that out logically. Perhaps there is no finer logic   , that ratio is just a good empirical fit to actual total lift measurements over the wing-mid body region  .

But then we have a lift area as well just behind the finishing of the nose curvature, does one try to include that in the wingarea reasoning or is this left to CFD and windtunnel runs to get the magnitude of?



Non French in France
User currently offlineOldAeroGuy From United States of America, joined Dec 2004, 3574 posts, RR: 67
Reply 37, posted (2 years 6 months 3 weeks 3 days 15 hours ago) and read 26207 times:

Quoting ferpe (Reply 35):
Dynamics guy have said however that the flaperon is left free flowing (no hydralic pressure) for start, ie it leaves as hole in the span load map and creates interference drag due to vortices and separations on both flap sides besides it.

This is only true for the initial part of the takeoff roll where the engine jet plume is free to expand as there is no freestream velocity to contain it. The result is flaperon buffet that would cause flaperon actuator fatigue if it were held rigidly in position.

Around 60 kts, the flaperon hydraulic pressure is restored and the flaperon deflects to fill the gap between inboard and outboard flaps. There is even less discontinuity for takeoff flaps than landing flaps. For takeoff CL and 2nd segment L/D, its very close to a continuous flap span.

This is very obvious if you observe the 777 flap system on a takeoff run. The video link shows the flaperon position during takeoff climb out.

http://www.youtube.com/watch?v=kQH8lTLqPCU&feature=related

It also shows the flaperon performing its aileron function.

[Edited 2012-05-05 13:19:17]


Airplane design is easy, the difficulty is getting them to fly - Barnes Wallis
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 38, posted (2 years 6 months 3 weeks 3 days 14 hours ago) and read 26185 times:

Quoting OldAeroGuy (Reply 37):
Around 60 kts, the flaperon hydraulic pressure is restored and the flaperon deflects to fill the gap between inboard and outboard flaps. There is even less discontinuity for takeoff flaps than landing flaps. For takeoff CL and 2nd segment L/D, its very close to a continuous flap span.

OK, this then explains a lot, thanks for pointing it out. It seemed rather wasteful of B as I had it described (it might be that Dynmicsguy gave this info as well, I missed it in this case). So there would then be rather low L/D loss both for the approach and start case dependent on how well this flap to flaperon transit on both sides would be made.



Non French in France
User currently offlineOldAeroGuy From United States of America, joined Dec 2004, 3574 posts, RR: 67
Reply 39, posted (2 years 6 months 3 weeks 3 days 14 hours ago) and read 26196 times:

Quoting ferpe (Reply 36):
Re Wimpress I just tried to grasp the logic behind the ratio used, the exposed Yuhedi LE span divided by the exposed TE span, still can't figure that out logically. Perhaps there is no finer logic , that ratio is just a good empirical fit to actual total lift measurements over the wing-mid body region .

Remember, the Wimpress area convention was developed during the 1950's when these beautiful CFD graphics weren't available. It did a reasonable job of correlating the differences in wing planforms with the observed wind tunnel and flight test results.

It worked well enough that it is still used today by Boeing to describe wing reference area.



Airplane design is easy, the difficulty is getting them to fly - Barnes Wallis
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 40, posted (2 years 6 months 3 weeks 3 days 14 hours ago) and read 26173 times:

Quoting OldAeroGuy (Reply 39):
It did a reasonable job of correlating the differences in wing planforms with the observed wind tunnel and flight test results.

Exactly what I thought, thanks this means I can stop my logical hunt  , makes sense.

Re the flaperon function, here a nice video from the 787:

http://www.youtube.com/watch?v=w2wz2l97KL4&feature=related

Great to see that the engine flux is not enough to hold the flaperon initally, it only raises as the freestream raises it, then at 60kt start to works as a start flap + aileron. Deflections are pretty large, much larger then the outboard aileron, surprising  Wow! (sure it has low area but I thought the outboard low speed aileron would be doing the predominant job at start, not to move the flaperon out of it's ideal position to much ).



Non French in France
User currently offlineOldAeroGuy From United States of America, joined Dec 2004, 3574 posts, RR: 67
Reply 41, posted (2 years 6 months 3 weeks 3 days 14 hours ago) and read 26201 times:

http://www.youtube.com/watch?v=Q0_juw3lOo0&feature=related

This video is an even better illustration of 777 flaperon takeoff movement. At first the flaperon droops with no hydraulic power. As the airplane gathers speed, it takes a freestream orientation. At around 60 kt, the hydraulic power is restored and the flaperon assumes its correct takeoff postion.



Airplane design is easy, the difficulty is getting them to fly - Barnes Wallis
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 42, posted (2 years 6 months 3 weeks 1 day 13 hours ago) and read 25774 times:

After finishing the start phase I found this picture in the van Dam paper:



and also the formula for the climb gradient after clearing the 35ft obstacle:

climb angle in % = (Thrust / Weight) - (1 / (L/D) )

For the 2 holers losing one engine they should climb 2.4 ft for every 100 ft forward. As I put that in I could see our excess power ratios were way to low, you need about 1.3 times the power to climb with that shallow angle  Wow! . Further the engines looses 25% of their power at 160-180 kt and not 20% as in the previous table. It was clear we had way to much drag for our engine power, the model or assumptions were not OK as several of the frames have FAA/EASA certification and thus could pull the stunt    .

It turned out the model is OK but one assumption was to harsh, the tail load due to longitudinal static stability. Shevell give it as UP TO 5% of MTOW negative lift needed from the tailplane during cruise I had it as average and times 1.2 as start max. I now needed the AVERAGE cruise loading and the maximum engine out start loading. Luckily Shevell provides as simulator for tail and additional wing load, cruise varies from -5% to +0,5% by MAC 0.2-0.4 with an average of about 2% by 0,3 MAC static margin. For the MTOW start case the framers limits the CG range to something like 0.25 - 0.37 of MAC (Mean Aerodynamic Cord). 3% seems to represent the most unfavorable case with max nose heavy static margin of 0.25 MAC- I had 6% in the last table gradually diminishing as the tail arm got longer (for the stretch frames).

When putting in 2 and 3% things changed a bit and I could increase the misc term in the parasitic drag (bundled with tail upsweep drag to 6+7=13 counts). Further the 789 and 7810 laminar flow now fitted with 20% reduction of surface friction for VTP, HTP, a more reasonable figure given the implementation on the nose part only. When I checked the 789 and 7810 with the laminar flow modelling and without the diff was 1.8% of overall drag, darn close to the rumored "couple of percent drag reduction" from Boeing    .

Further I now put in a row with needed single engine thrust for 2.4% climbout and then adjusted the climb-out speed until I had a fit with the 75% remaining single engine thrust. I also added 45 counts engine out drag from Kundu (caters for windmilling engine and yaw rudder, a bit to simple but OK for now), reduced the start flap to position 15 as this is a more reasonable trade between start field length and climb out obstacle clearance. Flap drag was adjusted accordingly. Here the model with the changes:

http://i298.photobucket.com/albums/mm262/ferpe_bucket/Analysisof787350777Start2.jpg

In red rectangle the V2 climb out speed need for attaining 2.4% climb-out with OEI (One Engine Inoperative). The lower the speed needed the better. The 359 is the winner with 333, 332 and 788 following. Here I have put the span efficiency at 0.85 for all frames, changing it to 0.9 for eg the 788 lowers the speed with 7 kt   .

One can see from the powerloading and wingloading row that this gives a good hunch of start performance for a frame.

[Edited 2012-05-07 14:49:55]


Non French in France
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 43, posted (2 years 6 months 1 week 2 days 15 hours ago) and read 24960 times:

After the last post I wanted to summarize the thread and include also the typical climb performance in the final table. Before that I went through all the areas that seems a bit fishy   in my model and I found and corrected some things:

1. The lack of range for the A350 series came from a fault in treating their winglets  Wow! , I had the effective span as 64 meter something when it should be 65.8m. Once corrected the A350 fitted the model fine.

2. There were a tendency for the stretched frames to come out short on range, ie they had to much drag. Further the 4 engine frames had to little drag. I therefore revisited all shortcuts in drag modelling and found I needed to model even finer to catch these things.

Consequently I separated the surface roughness factor and the misc. items drag (air cond, flap fairings, ...) and I also modeled the pressure drag coefficients for fuselage, wings, engine+pylons and tail separately, they all have quite different thickness ratios and an average like before discredits long frames.

3. I also implemented a function for compressive drag for cruise around Cl 0.5. I made a curve-fitting function based on PianoXs results and wingloading. The 767 has to be jacked up separately as it does not have a transonic wing design.

4. For thrust I put in the TO values (5 min) of the most common engine and then put in max continuous as 92% of that.

5. I also found by comparison with PianoX that my 2.0% trim load was a tad low, 2.5% for shortest family member gave me the same trim drag as the Piano models. The A380 with it's short tail arm fitted with a 3.0% trim load.


DRAG FACTORS
As can be seen from the aero constants section it got a bit more complicated but was fun doing (click on the table to view it better) :

http://i298.photobucket.com/albums/mm262/ferpe_bucket/Analysismodelfactors20052012.jpg



OVERALL MODEL RESULTS
The model now works quite fine, here the interesting parts for the most common frames (interesting values in yellow):

http://i298.photobucket.com/albums/mm262/ferpe_bucket/Analysismodel20052012.jpg


A couple of observations:

START PERFORMANCE V2 OEI restriction
The 788 has good OEI field performance with the GEnx-1B of 72.3 klbf but the 789 and 7810 with the top of the line 74.1klbf version will have the worst field performance of the lot with a V2 OEI of a whopping 184kts  Wow! . They clearly need more span or more power. The original 358 at the same MTOW had 5m longer span and 5klbf more thrust, no wonder they are a bit short on both.

The Airbus range have good field performance as has 748i (4 holer). The 77W and 777X has the same 171kts V2 which is OK for an ultimate stretch like the 77W but is it good enough for a new design?

It shall be noted these comments and values are without me degrading the B wings L/D with the lower span efficiency factor from the active inboard flaperon at start.


CLIMB PERFORMANCE @1500ft and 250kts
The -8X and the 380 form the jumbo league as climbers at 2600fpm, once again the Airbus range are the leaders together with the brute force 77W.


CRUISE PERFORMANCE
The last line of the table shows the kg/nm/pax of the spec trip, one can see why the 7810, 359 and -9X are the talk of the town but the 35J and -8X are just a tad more thirsty. If one load the 359 to the same 320 pax as the 7810 and run the same 6700nm it comes out at 0.0355 kg/nm/pax, ie just it beats the 7810 on it's kind of legs. It will be interesting to see if the 350 can hit it's numbers  or there will be a sobering experience for that program management as well   .

Should the 35J engines indeed have lower TSFC as Zeke posted (and "not any worse" as A and RR have said) then it might be on the level of the leading bunch. The 748i has lower fuel burn then the A380 but if one set the pax count to the debated 407 range they have the same fuel burn per pax, so then we know what that debate is all about  . Anyway, the 748i is no slough when it comes to fuel burned per nm and pax, once those PIPs get rolled in and the OEW settles. Come 2020 it will be encroached by the -9X but that is 2020  .

[Edited 2012-05-20 14:06:32]


Non French in France
User currently offlineWingedMigrator From United States of America, joined Oct 2005, 2243 posts, RR: 56
Reply 44, posted (2 years 6 months 1 week 2 days 11 hours ago) and read 24855 times:

Any chance you could make an educated guess about the performance of a hypothetical A380-900 ?

Maybe give it a stretch to 85 m, MTOW around 600 t, same wing, a bit shorter tail, and some Trent XWB donks.

I'm curious to see if the A380 stretch's performance can stay out in front of the best members of the A350 and 777X families. Does size matter anymore?


User currently offlinemffoda From United States of America, joined Apr 2010, 1103 posts, RR: 0
Reply 45, posted (2 years 6 months 1 week 2 days 9 hours ago) and read 24808 times:

Quoting WingedMigrator (Reply 44):
Maybe give it a stretch to 85 m

And if you thought flying a 80m A/C was a problem... What do you adding 5 meters will do?  



harder than woodpecker lips...
User currently offlineastuteman From United Kingdom, joined Jan 2005, 10174 posts, RR: 97
Reply 46, posted (2 years 6 months 1 week 1 day 22 hours ago) and read 24703 times:
Support Airliners.net - become a First Class Member!

Quoting mffoda (Reply 45):
What do you adding 5 meters will do?

Make some notable customers very happy, and also ensure that we will continue see that when Boeing aircraft start to encroach on these sorts of dimensions, we won't hear a peep.
As usual..  

Rgds


User currently offlineOldAeroGuy From United States of America, joined Dec 2004, 3574 posts, RR: 67
Reply 47, posted (2 years 6 months 1 week 1 day 21 hours ago) and read 24666 times:

Great work and even if there are a few areas of uncertainty, it's interesting to make a few observations.

1) At 525 pax, the A388 has the worst fuel burn per pax on the page. This runs counter to the Airbus statement that it will be equal to or better than the A350's. It may also explain why A380 sales are slow, independent of the low production rate.

2) The A3510 may have difficulty bringing home the claim that it will be 25% better fuel burn per seat than the 773ER as these data put the increment at 20%. It's hard to see how the 97K, 118" fan RR engine could make up that 5% shortfall.

3) The 77X's appear to compete well with the A350's in terms fuel burn per pax.

Time will tell how accurate Ferpe's estimates are, but his work helps bring some of the marketing statements into perspective.



Airplane design is easy, the difficulty is getting them to fly - Barnes Wallis
User currently offlineOldAeroGuy From United States of America, joined Dec 2004, 3574 posts, RR: 67
Reply 48, posted (2 years 6 months 1 week 1 day 21 hours ago) and read 24668 times:

Quoting astuteman (Reply 46):
Quoting mffoda (Reply 45):What do you adding 5 meters will do?
Make some notable customers very happy, and also ensure that we will continue see that when Boeing aircraft start to encroach on these sorts of dimensions, we won't hear a peep.
As usual..

I suspect the greatest peeps will come from the airports where 85m long airplanes are expected to operate. Few gates at the world's major airports are capable of handling this size airplane. It won't be a minor question for them.

[Edited 2012-05-21 07:16:55]


Airplane design is easy, the difficulty is getting them to fly - Barnes Wallis
User currently online747classic From Netherlands, joined Aug 2009, 2196 posts, RR: 14
Reply 49, posted (2 years 6 months 1 week 1 day 20 hours ago) and read 24629 times:

Quoting ferpe (Reply 43):
The 788 has good OEI field performance with the GEnx-1B of 72.3 klbf but the 789 and 7810 with the top of the line 74.1klbf version will have the worst field performance of the lot with a V2 OEI of a whopping 184kts . They clearly need more span or more power.

GE is already prepared for this scenario and will certify the GEnx-1B PIP2 with a raised T/O rating of 78.000 lbs.

Although Boeing hasn’t asked for the extra 3,000 pounds of thrust, the increase will help prepare GE for potential applications on the 787-9 and, eventually, the 787-10. “Boeing has told us they don’t need it,” according GE, but "we have designed it just in case. All other programs have always wanted a little more thrust.” GE expects to gain certification for PIP2, incl. the increased thrust rating by June 2012.



Operating a twin over the ocean, you're always one engine failure from a total emergency.
User currently offlineastuteman From United Kingdom, joined Jan 2005, 10174 posts, RR: 97
Reply 50, posted (2 years 6 months 1 week 1 day 20 hours ago) and read 24647 times:
Support Airliners.net - become a First Class Member!

Quoting WingedMigrator (Reply 44):
Maybe give it a stretch to 85 m, MTOW around 600 t, same wing, a bit shorter tail, and some Trent XWB donks.

I'm curious to see if the A380 stretch's performance can stay out in front of the best members of the A350 and 777X families. Does size matter anymore?

On those criteria, I wonder if an alternative wingtip solution would also be considered.....

Quoting OldAeroGuy (Reply 47):
At 525 pax, the A388 has the worst fuel burn per pax on the page

couple of points...
Firstly, the A380 also has WAY the lowest passenger/m2 cabin ration on the page as its measured. It is being disadvantaged by 15% compared to the next nearest aircraft in terms of pax count vs cabin size, and 22% compared to the furthest away.
On an area basis, the model figures would show the A380-800 on a par with the 777-300ER, and measurably better than the 748i.
And I think that still underplays the A380's position, for the reasons below

Secondly, I am led to wonder if Ferpe's model overplays the parasitic drag for the A380 - Airbus quoted a 10% improvement in parasitic drag per m2 for the A380's wings (over any previous Airbus wing) based on the surface treatment and resulting smoothness (and having seen them it's hard to disagree)

Thirdly, I'm unsure how the model might deal with the huge amount of body lift that the A380's fuselage also develops. For that matter, I'm unsure how it could deal with it.

Fourthly, there have been a goodly number of videos on you-tube, shot from inside the cockpit, including of the early flight phase, and whilst I accept that this is hardly scientific data, I haven't seen any that show steady state fuel flow per engine anywhere other than in the 2 750 to 3 000 kg/hr range - i.e. 11 000 - 12 000 kg/hr total - a lot lower than the 13 800 kg/hour that the model shows.

I put these forward as discussion points (except for the first, which is a perennial fact in any conversation about the A380's efficiency, or lack of)

Quoting OldAeroGuy (Reply 47):
It may also explain why A380 sales are slow, independent of the low production rate.

There's no way yet of saying that the slow sales are independent of the low production rate.   

Quoting OldAeroGuy (Reply 47):
The A3510 may have difficulty bringing home the claim that it will be 25% better fuel burn per seat than the 773ER as these data put the increment at 20%

Interestingly, EK quoted a fuel burn delta of 21%, which kind of supports Ferpe's model, as opposed to the Airbus figures

Quoting OldAeroGuy (Reply 47):
his work helps bring some of the marketing statements into perspective.

In more ways than one....  

Rgds


User currently offlineOldAeroGuy From United States of America, joined Dec 2004, 3574 posts, RR: 67
Reply 51, posted (2 years 6 months 1 week 1 day 16 hours ago) and read 24557 times:

Quoting astuteman (Reply 50):
Firstly, the A380 also has WAY the lowest passenger/m2 cabin ration on the page as its measured. It is being disadvantaged by 15% compared to the next nearest aircraft in terms of pax count vs cabin size, and 22% compared to the furthest away.

OK, let's try to adjust for these numbers.

Assume we increase the passenger count by 20%.

.20*525 = 105

But the additional passengers will increase zero fuel weight, increasing fuel burn.

Assume the increase per pax is: pax + seat + catering = 210 lb + 30 + 10 = 250 lb.

Total zero fuel weight increase = 105 * 250 = 26,250 lb.

For the A380, I'd expect the block fuel burn to increase be 1% for each 8000 lb of zero fuel weight.

26,250/8000 = 3.28 or 3.28%

From Ferpe's numbers, Block Fuel = Block Fuel per pax * pax = .0525 * 525 = 27.56

Block Fuel Adj: 27.56 * 1.0328 = 28.46

Pax adj = 28.46/630 = .0452

Using Ferpe's Fuel per Pax data:

773ER: .0452/.0470 = .961 A380 burns 3.8% less fuel per seat

A351: .0452/.0378 = 1.196 A380 burns 19.6% more fuel per seat

779X: .0452/.0361 = 1.252 A380 burns 25.2% more fuel per seat

Given the lower risk to fill 365, 350, or 407 seats vs 630 seats, the A388 would seen to be at a disadvantage on all but the most dense/slot restricted routes. It needs a new technology engine upgrade ASAP. Even this will leave it 10% to 15% behind the A351 or 779X.



Airplane design is easy, the difficulty is getting them to fly - Barnes Wallis
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 52, posted (2 years 6 months 1 week 1 day 16 hours ago) and read 24744 times:

Quoting astuteman (Reply 50):
I'm curious to see if the A380 stretch's performance can stay out in front of the best members of the A350 and 777X families. Does size matter anymore?

On those criteria, I wonder if an alternative wingtip solution would also be considered.....

Thanks for those comments, I am also puzzled by the results for the A380. I found a slight misfit, I had the OEW at the ACAP 270t, I now have put in 276t vs the 569t MTOW, I alos included a lot of additional ratios to try and understand where the 380 spends those fuel kgs. I have also included the 79,3 m proposed 380-900 with TXWB engines:

http://i298.photobucket.com/albums/mm262/ferpe_bucket/Analysismodelincl38921052012.jpg


As per the A380 fuel burn, it is in drag but it is both parasitic (where Astutemans comments are valid, all frames are measured by the same token in the model and if someone has a smoother surface I can't include this without knowing it) and induced drag. Those 80m limits make the wings have an effective aspect ratio of 7.8 when the others are above 9.5, only the 748i is at the same 7.8. Also look at the spanloading, 50% or more above the others.

Re the drag level, the A380 flies 8300nm in my model, just as per A figures. It can then mean:

- the TSFC (0.556) is to low to compensate the high drag (more mpg to attain 8300nm)

- the OEW (270t and now 276t) is to low to compensate the high drag (more fuel for the 8300nm range)

- the reserves are to small (18300kg) to leave to much fuel to burn in order to attain 8300nm

If nothing of the above applies it does not matter what drag or lift we have, the fuel consumption is then OK IMO. Of course one shall then pack the frames to the same level for an honest apples to apples comparison.

I think we will see some clever feather winglets on the 389  Yeah sure .

[Edited 2012-05-21 11:35:09]


Non French in France
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 53, posted (2 years 6 months 1 week 1 day 15 hours ago) and read 24695 times:

Some more comments:

- please read the average fuel flow at line 243, the one at FL370 is a momentary FF at 83% weight which is mid cruise weight. As the fuel burns of the induced drag reduces and the FF follows.


- I realized that some would like to see my assumptions re MTOW, MZFW and OEW for the 389, I added these in this table, I also added cabin area in line 43 and fuel burn per nm and m2 cabin area in line 246 as an additional efficiency measure:

http://i298.photobucket.com/albums/mm262/ferpe_bucket/Analysismodelincl389weightscorrected21052012.jpg


Edit: corrected the 748i cabin area, please chip in if still not correct.

[Edited 2012-05-21 13:17:52]


Non French in France
User currently offlineOldAeroGuy From United States of America, joined Dec 2004, 3574 posts, RR: 67
Reply 54, posted (2 years 6 months 1 week 1 day 15 hours ago) and read 24692 times:

Quoting ferpe (Reply 52):
I had the OEW at the ACAP 270t, I now have put in 276t vs the 569t MTOW

I don't understand how increasing OEW decreases fuel burn per pax.

Quoting astuteman (Reply 50):
I haven't seen any that show steady state fuel flow per engine anywhere other than in the 2 750 to 3 000 kg/hr range - i.e. 11 000 - 12 000 kg/hr total - a lot lower than the 13 800 kg/hour that the model shows.

Ferpe's model shows a fuel flow of 11,680 kg/hr, in line with your observations.



Airplane design is easy, the difficulty is getting them to fly - Barnes Wallis
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 55, posted (2 years 6 months 1 week 1 day 15 hours ago) and read 24687 times:

Quoting OldAeroGuy (Reply 54):
I don't understand how increasing OEW decreases fuel burn per pax.

My bad, it is the other way around as would be natural (270t gives 0.0515 and 276t gives 0.0521).

I have another model where I count backwards from the OEMs spec values and given I have estimated the OEW and reserves correctly the FF falls out as I know all other weights. Here a lower OEW gives a higher fuel burn, I got mixed up after fiddling with a 290 lines * 50 rows Excel for to long       .



Non French in France
User currently offlinesunrisevalley From Canada, joined Jul 2004, 5142 posts, RR: 5
Reply 56, posted (2 years 6 months 1 week 1 day 14 hours ago) and read 24671 times:

Quoting astuteman (Reply 50):
Interestingly, EK quoted a fuel burn delta of 21%, which kind of supports Ferpe's model, as opposed to the Airbus figures

I have been wondering about various fuel burn claims and took a look at a still air comparison between what QF might expect with one of the A380 variants loaded in PIANO-X and V Aus. with a 77W for the same payload on the same LAX-SYD sector distance.
I arranged for a flight plan to be done on Jeppersen FlightStar for a 77W 167.829t OEW with a load of 44.76t for a 6535nm sector ( zero winds) The fuel load was 130.410t and the burn 110.109t. with a sector time was 13hr 38m. Burn was 7.876t / cruise hr
I ran in PIANO-X a 569t MTOW / 286t OEW A380 with an identical payload over an identical distance again zero winds. The fuel load was 185.296t with a burn of 167.851t. Sector time was 13hr 51m. Burn was 11.925t / cruise hr.

Below is a summary of the two aircraft.

A380 RANGE REPORT {fixed block distance & payload}
____________

{TOW 515057.kg./ OEW 285000.kg./ Fuel 185296.kg./ Payload 44761.kg.}

Range mode: fixed mach, step-up cruise

Climb schedule: 250./ 318.kcas/ mach 0.844 above 30675.feet

Cruise at Mach = 0.850 {FL 360 380 400}

ICA 36000.feet, 488.ktas, 284.kcas, CL=0.51, 61017.newtons/eng=MCR-13%
FCA 40000.feet, 488.ktas, 259.kcas, CL=0.43, 43745.newtons/eng=MCR-23%

Distance Time Fuelburn
(n.miles) (min.) (kg.)
_________ ______ _______
Climb 178. 25. 10929. {S.L to ICA}
Cruise 6230. 767. 152449. {ICA to ICA}
Descent 127. 21. 957. {ICA to S.L}
_________ ______ _______
Trip total 6535. 813. 164335.
Block total ===== 831. 167851.

Emissions: taxi,t/o climb cruise descent app,taxi total
(kg.NOx) 51.3 362.0 2577.2 3.6 7.4 3001.5
(kg.HC) 0.00 0.00 0.00 0.26 0.11 0.37
(kg.CO) 2.2 4.7 204.8 19.2 9.2 240.0
(kg.CO2) 8074. 34536. 481737. 3026. 3037. 530411.

Manoeuvre allowances:
taxi-out 2000. kg. {extra to t/o mass} 7.0 min.
takeoff 555. kg. 1.0 min.
approach 462. kg. 3.0 min.
taxi-in 499. kg. {taken from reserves} 7.0 min.

Reserves {at landing mass 349705.kg.}:

Diversion distance 200. n.miles
Diversion mach 0.589
Diversion altitude 20439. feet
Diversion fuel 6449. kg.

Holding time 30. minutes
Holding mach 0.316
Holding altitude 1500. feet
Holding fuel 5228. kg.

Contingency fuel 8268. kg. {5.% of mission fuel}

Total Reserve fuel 19944. kg.

FUEL SUMMARY 77W
Fuel Calculation Time Fuel(Kg) Reserve Time Fuel(Kg)
Climb T/O 00:30 8318
Cruise 12:48 100813
Descent 00:20 476
Approach 00:00 1
Total Trip Fuel 13:38 109609
Taxi 500
Burn Off 13:38 110109
Min. Required Fuel 15:58 130410


User currently offlineastuteman From United Kingdom, joined Jan 2005, 10174 posts, RR: 97
Reply 57, posted (2 years 6 months 1 week 1 day 14 hours ago) and read 24693 times:
Support Airliners.net - become a First Class Member!

Quoting ferpe (Reply 52):
Thanks for those comments, I am also puzzled by the results for the A380. I found a slight misfit, I had the OEW at the ACAP 270t, I now have put in 276t vs the 569t MTOW

In reality, both of those figures are light.

The R/P chart Airbus use implies an OEW of 282 tonnes (assuming the 569t variant).
In real life it's dry operating weight will be nearer 290 tonnes.

That may account for this...

Quoting ferpe (Reply 52):
the OEW (270t and now 276t) is to low to compensate the high drag (more fuel for the 8300nm range)
Quoting OldAeroGuy (Reply 54):
Ferpe's model shows a fuel flow of 11,680 kg/hr, in line with your observations

Cruise fuel flow at line 143 = 13 607 kg/hr

Rgds


User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 58, posted (2 years 6 months 1 week 1 day 13 hours ago) and read 24674 times:

Quoting astuteman (Reply 57):
Cruise fuel flow at line 143 = 13 607 kg/hr

As I stated this is a momentary value at mid cruise weight, the average value is at the bottom. The FF varies from 15.5t at start of cruise down to 10.5 just before descent.


I am sorry to have to repost the table again but the cabin areas for the 380 and the 748 are not correct. For the others I have checked the values myselves but for these double deckers I have relied on others given correct values, they have been way of    . Here what I have now calculated after taking the dimensions of cabin length from Wikipedia for the 380 and multiplied with 6.6 and 5.6 (the upper usable width is only 5.6, check the ACAP). For the 748 I have taken the 744 lower cabin length from Flightglobal (usually correct) added 5.6m and then times 6.1 and the upper I have measured to 18m times 3.5m (you can't use any more, see the ACAP). Makes for 580m2 and 449m2 :

http://i298.photobucket.com/albums/mm262/ferpe_bucket/Analysismodelincl389weightscor21052012.jpg

[Edited 2012-05-21 14:52:29]


Non French in France
User currently offlineWingedMigrator From United States of America, joined Oct 2005, 2243 posts, RR: 56
Reply 59, posted (2 years 6 months 1 week 1 day 11 hours ago) and read 24624 times:

Hi Ferpe, thanks for all that additional data.

Having tinkered with my own Breguet spreadsheet over the years, I have a basic concern with your final Breguet specific fuel flow figures. I think you're comparing apples to oranges. For any aircraft, the fuel flow is going to depend on the mission length, since it takes fuel to carry fuel. That inherently penalizes aircraft with very long range.

The comparison should be for a fixed range, regardless of aircraft capability. I've found that 6000 nm works nicely.

The next issue is payload-- spec seat counts are often determined by marketing as much as by engineering, and there again I've tended to go to a standard 1 pax (95 kg) / m2 load to compare everything on an even basis. I realize that's a gross simplification, but I'm at a loss for how to come up with a simple and fair way to load aircraft in proportion to their size, without getting dragged into the minutiae of armrest widths and cabin monuments.

As to your figures-- why is the A350-1000 range off by 1368 nm?


User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 60, posted (2 years 6 months 1 week 1 day 4 hours ago) and read 24527 times:

Quoting WingedMigrator (Reply 59):
For any aircraft, the fuel flow is going to depend on the mission length, since it takes fuel to carry fuel. That inherently penalizes aircraft with very long range.

You are correct, at this stage I have verified the drag and TSFC variation model (the OEMs spec range at spec Pax+Bags is a good check) more then actually compared frames from different range brackets with each other, the comparison should only be made for frames with about the same shown range. I will now work to make an equal range comparison possible, it is more to it then just set MTOW and therefore fuel to the set range, I also have to find a model for the reserve calculations (the 5% trip fuel reserve).

Quoting WingedMigrator (Reply 59):
The next issue is payload-- spec seat counts are often determined by marketing as much as by engineering, and there again I've tended to go to a standard 1 pax (95 kg) / m2 load to compare everything on an even basis. I realize that's a gross simplification, but I'm at a loss for how to come up with a simple and fair way to load aircraft in proportion to their size, without getting dragged into the minutiae of armrest widths and cabin monuments.

You are 100% right, this was also my reaction after seeing the differences with eg how the 380 is loaded etc. With my m2 quick fix I came closer to the truth (if there is one  Wow! ) but I shall also load the frames to that passenger number (I concur that 1 pax per m2 seems the right compromise for a spec density). When doing that one also have to adapt the furnishing level to the corresponding seat number, ie the adjustment factor is 95+63 kg per pax for any diff from the spec values (or 95.3+63 for B models).

When I do that I will also adjust the cabin areas to also take into account taper as per CMs post in the Cabin area post, right now I have only cabin inner width * cabin length, it is however checked figures and should not be off by 20% or more as we have seen previously.

Quoting WingedMigrator (Reply 59):
As to your figures-- why is the A350-1000 range off by 1368 nm?

It was my quick simulation of fueling for a 6700nm mission to compare it to the 7810 (look at the MTOW), the correct fuel burns at 8100nm are 0.0359 (pax) and 0.0395 (m2).

[Edited 2012-05-22 00:08:14]


Non French in France
User currently offlineastuteman From United Kingdom, joined Jan 2005, 10174 posts, RR: 97
Reply 61, posted (2 years 6 months 1 week 1 day 2 hours ago) and read 24526 times:
Support Airliners.net - become a First Class Member!

Quoting ferpe (Reply 58):
The FF varies from 15.5t at start of cruise down to 10.5 just before descent.

That seems like a big spread.
It might be worth seeking some corroboration from some of the more informed members able to point to actual figures..

Rgds


User currently offlineparapente From United Kingdom, joined Mar 2006, 1642 posts, RR: 10
Reply 62, posted (2 years 6 months 1 week 22 hours ago) and read 24455 times:

I think we will see some clever feather winglets on the 389

I feel this is very likley.When the present wings were designed they were designed to the best of their ability at that time.But time has mooved on a decade or two.Materials have advanced but more importantly so has wing tip expertise (esp Airbus catching up). The aspect ratio will always be compromised, also itwill force majeure have larger votex generation (than smaller aircraft) wich compromises spereation distances.

However these new "feathers" as displayed by both Aviation partners and Boeing look live a very obvious way for them to go.It does (in effect) increase the aspect ratioo and will most certainly reduce drag by a very large amount as we have seen on the 320 NEO programme and indeed even more so on the next 737. Winglets have a particularly strong benefit in cruise and that is exactly what this aircraft does mostly.


User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 63, posted (2 years 6 months 1 week 20 hours ago) and read 24426 times:

Quoting astuteman (Reply 61):
That seems like a big spread.
It might be worth seeking some corroboration from some of the more informed members able to point to actual figures..

Corroboration with PianoX:



Loading plane: Airbus A380-800 (569t, 270t OEW, 525 Pax)....Done.

DETAILED FLIGHT PROFILE {design range & standard payload}

Seg. Speed mode Thrust mode Stop condition

1. v3 at-%mto to alt 1500.
2. specify-cas at-%mcl to alt 10000.
3. specify-cas at-%mcl to mach 0.84
4. same-mach at-%mcl to alt 33000.
5. specify-mach level-cruise to next stepup
6. same-mach at-%mcl to dalt 2000.
7. same-mach level-cruise to next stepup
8. same-mach at-%mcl to dalt 2000.
9. same-mach level-cruise to next stepup
10. same-mach at-%mcl to dalt 2000.
11. same-mach level-cruise to next stepup
12. same-mach at-%mcl to dalt 2000.
13. same-mach level-cruise to mass 343929.
14. specify-mach at-idle to cas 322.
15. specify-cas at-idle to alt 10000.
16. specify-cas at-idle to alt 1500.
17. vapp match-grad to alt 50.
-----------------------------------------------

Manoeuvre segment 5 starts at:

Initial Mass 552475. kg. (CL 0.48 airbus)
Initial Altitude 33000. feet (1.40g buffet)
Delta-ISA +0. degC
Airspeed (CAS) 304. kts (input Mach 0.850)
Flaps 0 deg. (hi-speed)
Undercarriage up
All eng.operative
Thrust per engine 66530. newtons (level cruise) 87.6%MCR

Climb/Descent rate 0. feet/min
Flightpath angle 0.0 deg. (grad.0.00%)
True airspeed 494. kts

   Fuel Flow rate 15586. kg/hr

L/D ratio 20.36
   Total aircraft drag 266121. newtons = 59 816 lbf

Segment 5 ends at total time = 93.7 mins, endmass = 535665.kg.(1.44g buff)

-----------------------------------------------------------------------------
Manoeuvre segment 13 starts at:

Initial Mass 394512. kg. (CL 0.51 airbus)
Initial Altitude 41000. feet (1.34g buffet)
Delta-ISA +0. degC
Airspeed (CAS) 253. kts (fixed Mach 0.850)
Flaps 0 deg. (hi-speed)
Undercarriage up
All eng.operative
Thrust per engine 48542. newtons (level cruise) 89.4%MCR

Climb/Descent rate 0. feet/min
Flightpath angle 0.0 deg. (grad.0.00%)
True airspeed 488. kts

   Fuel Flow rate 11170. kg/hr    my FF 11553. kg/hr

L/D ratio 19.93
   Total aircraft drag 194168. newtons = 43 643 lbf

Segment 13 ends at total time = 1033.5 mins,    endmass = 343929.kg.(1.53g buff) (this is what I have as 10.5t/hr, PianoX does not give it as it starts decent at next segment with engines at low thrust).

------------------------------------------------------------------------------------------------------

Block summary: Time Fuel Burn
-----------------------------------------------
Ramp,taxi-out 420. sec. 2000. kg.
T/O to screen height 52. sec. 180. kg.
In-flight sequence 63419. sec. 222645. kg.
Taxi-in 420. sec. 499. kg.

   Block total (8509.nm) 1072. min. 225324. kg. = Fuel Flow rate 12594 kg/hr
-----------------------------------------------
   Kg / nm / Pax 0.0504

   Conclusion: my fuel burn and fuel burn per nm is OK but I have to look into how I handle time flown.

[Edited 2012-05-22 08:23:09]


Non French in France
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 64, posted (2 years 6 months 1 week 19 hours ago) and read 24406 times:

Quoting parapente (Reply 62):
Winglets have a particularly strong benefit in cruise and that is exactly what this aircraft does mostly.

If Airbus can add 1m effective span each side it will increase the range with 200nm and put the fuel burned per nm and m2 at 0.0458



Non French in France
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 65, posted (2 years 5 months 6 days 15 hours ago) and read 22960 times:

Quoting ferpe (Reply 63):
Conclusion: my fuel burn and fuel burn per nm is OK but I have to look into how I handle time flown.

I have corrected this faulty formula, at the same time I took the time to look over and improve other items I need for a more general comparison over equal distances like proposed by Wingedmigrator. I will do this in a separate thread here at TechOps.

Things that needed fixing was the 5% reserve fuel calculation which brings a recursive problem (fixed with setting iterative solving in Excels options), more general climb fuel calculation and finally I adjusted all ranges to exactly the OEMs range by virtue of introducing DOW and adjusting the empty weight to fit (within a fraction of a metric ton, otherwise I hunt for reasons why things don't fit. For the fitting calculations I employ Excel solver which is more robust then the "What if" Excel function, given I have a recursive loop in the calculations). This fiddling with a DOW adder to get a 100% fit to the OEMs range might seem foul but the OEMs OEW figures are very doubtful, when a 738 has the same official OEW without winglets and with winglets one realizes these figures are to be taken with a grain of salt, Airbus has skipped OEW altogether (as has Boeing sometimes) in the new ACAPs they publicized for the 320 series 1 June.

Here the comparison chart with the correct FF kg/hour. The efficiency measures like burned kg fuel/nm and m2 passenger area is virtually the same as before:

http://i298.photobucket.com/albums/mm262/ferpe_bucket/Analysisof787350777-8X71mwingwithFFcorrection.jpg



Non French in France
User currently offlineWingedMigrator From United States of America, joined Oct 2005, 2243 posts, RR: 56
Reply 66, posted (2 years 5 months 2 days 16 hours ago) and read 22411 times:

Curious about the A389 numbers... should V2 not be higher than the A388, given the same wing?

User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 67, posted (2 years 5 months 1 day 13 hours ago) and read 22192 times:

Quoting WingedMigrator (Reply 66):
Curious about the A389 numbers... should V2 not be higher than the A388, given the same wing?

No, the formula is dependent on available thrust when one engine has fallen away, the TWB thrust of 3*88klbf vs 3*72klbf for the T900 gives the difference in the necessary V2 to obtain a climb angle of 2.4%.

Here the formula:

V2=ARCTAN((Engine_out_thrust_at_0_75_lapse_lbf*lb_kg/MTOW_kg)-(1/V2_L_D))

The 4 holers should in reality obtain 3.0% climb angle (reply 62), I made a solver makro to calculate these V2s and only had one with 2.4% as target value, should have had one with target 3.0% for the 4 holers but as a comparison V2 to others it is correct e.g. vs the normal A380. Those TXWB would be a real asset for the A380   .



Non French in France
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 68, posted (1 year 3 weeks 3 days 3 hours ago) and read 11457 times:

Over the time since I started this thread it has been referenced many times by me and others as it contains a good rundown on how one can analyse a civil airliner with the techniques used in aircraft preliminary design. I have plowed through many standard books and courses to verify the formulas and constants I apply, I consistently find the formulas in the Stanford course the most usefull. For the most cases one also finds the constants one need there:

http://adg.stanford.edu/aa241/AircraftDesign.html

Other works where I find info are:

- Synthesis-of-Subsonic-Airplane-Design-Torenbeek
- Airplane-Aerodynamics-and-Performance-Roskam
- Aircraft-design, Kundu
- Airbus Getting_to_Grips_With_Aircraft_Performance
- Aeronautical-Engineer-Data-Book

On winglets this is the best I found:
- Wingtips devices; What they do and how they do it, McLean

For the atmospheric constants there is only one place , superbly made:

http://www.aerospaceweb.org/design/scripts/atmosphere/

There was one aspect of aircraft performance that I avoided in the last stint, start and landing performance where I only did the phase after lift-off (to avoid dealing with ground friction) and before touchdown. Especially the start problematic is hotly debated in view of the 777X and any extension of the A350-1000, therefore I have attacked that as well. More in the next post   .



Non French in France
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 69, posted (1 year 3 weeks 3 days 2 hours ago) and read 11444 times:

On Leehman.net I posted this in a discussion around the possibility of a A350-1100. There was the usual dicussion where different values of wingloading etc was thrown around proving nothing as they used different measurement standards, therefore this post (is shows the results of my start acceleration distance calculations) :

There is a lot of Apples and Oranges comparisons going on in this debate, here’s a try to stem that. To make a -1100 out of the -1000 with minimal changes to counter the 777X one has to first understand where things stand, based on that one can discuss needed changes. Wingloading and Thrust/Weight will dictate the acceleration distance to lift off, Sloff, thereafter if will be lift to drag L/D to get past the 50ft obstacle at V2. Lets first compare the wingloadings using ONE consistent wingarea measurement method then the the acc distance to Vloff and V2 speed and L/D. For background read here Boeing Vs. Airbus Wing Design Philosophies (by ferpe Apr 16 2012 in Tech Ops):

Airplane…..MTOW….Wingl……T/W….Sloff…..V2 L/D…V2 kts
300ER……..351t….785kg/m2…0.30…2300m….11.4…….173
-9X………….351t….700kg/m2…0.27…2400m….13.2…….179
3510………..308t….665kg/m2…0.29…2100m….12.0…….166
3511………..320t….695kg/m2…0.28…2200m….12.2…….167

The table shows that a stretch .-1100 with a 100klbf Trent97k and a Scimitar wingtip stacks up well against the -9X in all parameters discussed here. In practice Boeing is using the high -9X L/D to increase the flap angle so that they get the Sloff to around 2200-2300m but the point is a -1100 is fully doable without large changes and there are margins. To keep the Sloff the same as the -1000 without wing changes you need 104klbf as Scott sais, with a scimitar wing and the better then expected low speed lift of the A350 wing (ref flight tests) you can stop at 100klbf. With a non PIPed Trent you fly 8000nm, with the expected 3% improvement by 2020 8300nm.

REPLY
ferpe
NOVEMBER 1, 2013 @ 8:34 PM
Just to complement, the -1100 burns 10% less trip fuel then the -9X and flies the 400 pax in 18” seats at that   so it is a no brainer. Airbus will skip the -800 and do the -1100, as outlined it is a larger effort but not by much.


I put in the correct V2 safe climb height here (50ft), I wrote 35ft in the post which was the wrong FAR me thinks. The acc distance formula is the Shevell one under 11.1. He does not give any values for ground friction, I got 0.025 for concrete from Roskam and 0.03 from Toorenbeek, I use 0.025.

Another critical value is the Cl at vmu which together with the safety margins in FAR25 gives the allowed Vloff and V2. We don't have the Clmu (sh-t that there was no ADB-S reciever close to Vatry   ) so I have used values from PianoX and also calculated backwards from Vr speed from FCOMs and test flights we could observe. Clmax of 2.0 seems to be a good first bet, then one can assume that Boeing knows what they are doing with the 777X and realize then can trade a bit more flap angle and thereby flap drag as their start L/D is in the 13+ region (due to the span, after Vloff the induced drag is 80% of total drag, this is where the 71m span is handy  ). So it seems Clmx of 2.2 is a good guess for the 777X, might be able to go a tick higher also for other frames. The 787 would be good candidates, they seem a bit short on start performance compared to the A350 (I had to put in 76klbf engines on the 789 to get these values, something to watch as we get closer to EIS  Wow! ).

This gives the following table for our normally debated airframes (as always click on the image to see better):

http://i298.photobucket.com/albums/mm262/ferpe_bucket/Startperformancefor787350777_zpsafa8e39a.jpg

[Edited 2013-11-02 01:26:42]


Non French in France
User currently onlinePihero From France, joined Jan 2005, 4630 posts, RR: 77
Reply 70, posted (1 year 3 weeks 2 days 20 hours ago) and read 11237 times:
Support Airliners.net - become a First Class Member!

Quoting ferpe (Reply 42):

For the 2 holers losing one engine they should climb 2.4 ft for every 100 ft forward.

...expressed more often as 2.4% climb gradient for a twin, 2.7 % for a tri and 3.0 % for a quad.

Quoting ferpe (Reply 69):
I put in the correct V2 safe climb height here (50ft), I wrote 35ft in the post which was the wrong FAR me thinks

AFAIK, the *screen height* is still at 35 ft

Thanks for the pain of this research. An absolute gold mine, ferpe !!!



Contrail designer
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 71, posted (1 year 3 weeks 2 days 18 hours ago) and read 11166 times:

I did some further checking with some FCOMs, Clmu of 2.0 is to optimistic, we shall rather have 1.7 as a good start value. I have remade the table with 1.7 for all frames, here the data:

http://i298.photobucket.com/albums/mm262/ferpe_bucket/Startperformancefor787350777withClmu17_zps94789d71.jpg

Vr is now close to Vloff in my table and V2 is close to the values in my table as well. I have made the V2 stuff for Flaps 15 or config 3 for Airbiiees.

Edit:
Thanks Pihero, I did remeber right in the Leeham post, screen height is 35ft, all nicely summed up in post 42 of this thread.

[Edited 2013-11-02 10:51:00]


Non French in France
User currently online817Dreamliiner From Montserrat, joined Jul 2008, 2553 posts, RR: 2
Reply 72, posted (1 year 3 weeks 2 days 17 hours ago) and read 11161 times:

Quoting ferpe (Reply 69):
This gives the following table for our normally debated airframes

Two questions:

1) How is the wing area for the 787 370m2? A google search shows it at 325m2, which figure is correct? If using the 325m2 figure, I get an aspect ratio of 10.79, using the effective wingspan shown in the table.

2) How do you calculate the effective wingspan shown in this table?

Im curious because im designing an aircraft for a uni course im doing, hope you can answer my questions  



Reality be Rent. Synapse, break! Vanishment, This World!
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 73, posted (1 year 3 weeks 2 days 17 hours ago) and read 11160 times:

Quoting 817Dreamliiner (Reply 72):
1) How is the wing area for the 787 370m2? A google search shows it at 325m2, which figure is correct? If using the 325m2 figure, I get an aspect ratio of 10.79, using the effective wingspan shown in the table.

Look in the opening post (OP) of this thread, there you have it all. The 325m2 is the value that the original PianoX analysis put on the 787 wing. A reference area is a reference area, you can use any measurement standard you like as long as you use one and only one for all compared frames. The Airbus way of measuring is close to the Boeing way but much easier to actually measure on a drawing, the trapezodial is only used in some Piano and other cases. As Airbus publicizes the reference wingareas and it is easy to measure them for Boeing frames (and one can do a quick and dirty check with the Wimpress being 97.5% of an Airbus area (this is not always true)) I have come to use Airbus method consistently, I advise you to do that as well. You have all the values in the threads first posts.



Quoting 817Dreamliiner (Reply 72):
2) How do you calculate the effective wingspan shown in this table?

The effective wingspan in the wing without wintip device + the equivalent wing length for the wingtip device. There are separate threads for this but as a rule of thumb (see the McLean pdf referenced in post 68 ) :

- a blended winglet counts with 45% of it's physical length

- a raked tip counts with 80% of it's physical length

- an Airbus fence counts as wingspan increased with 2%.



Non French in France
User currently online817Dreamliiner From Montserrat, joined Jul 2008, 2553 posts, RR: 2
Reply 74, posted (1 year 3 weeks 1 day 15 hours ago) and read 10645 times:

Quoting ferpe (Reply 73):

Thanks for answering my questions. I understand it a lot better now. 



Reality be Rent. Synapse, break! Vanishment, This World!
User currently offlineferpe From France, joined Nov 2010, 2805 posts, RR: 59
Reply 75, posted (1 year 3 weeks 17 hours ago) and read 10270 times:

I gave some 777 values in a General Aviation forum which was debating the 777X and the existing models, after my posting -200Er and LR numbers Poncho came up with Boeing numbers that contradicted mine (which is not uncommon of course) but they were outside of normal margins. Here how I found the fault and how my model then tracked Boeings numbers:

Rumour BA Unimpressed With Emiratisation Of B777X (by Blueshamu330s Oct 31 2013 in Civil Aviation)

"
Thanks Poncho, I found the fault, you send me a link to the following excellent presentation (truly great work to find it http://www.boeing.com/assets/pdf/commercial/startup/pdf/777_perf.pdf ) .

I was really puzzled as I know I am normally within a % or two so I started to check and saw that all values jived with mine within the tonne or so for 6000nm, strange !.

Then I found it! , I remembered wrong when I posted the numbers, it shall be 82.1t (200ER) and 81.1t (200LR) (fule used for a 6000nm leg, my addition). The Boeing numbers where for best engine 82.6 (200ER) and 85.0t (200LR). So why this difference? Because they give it for airplanes with realistic OEW numbers, ACAP has 138.1t for the 200ER, I used 140.0 and they now give 145.3, ACAP and thus my model has 145.15t for 200LR, now they say 155.5 ! .

The same for all other numbers, the closest is 300ER where they give 168.8t and ACAP has 167.8t. When I changed the OEW to these values my model versus Boeing goes like this:


Block fuel for 6000nm leg:

777-........OEW...Boeing....Model...% miss
200ER.....145.5....82.6.......82.1.....0.6%
200LR.....155.5....85.0.......84.3.....0.8%
300ER.....168.7....94.4.......94.1.....0.3%


So my model tracks Boeing's figures within -1%, I am just below their figures but within the %.

Given all the assumptions I have to make regarding climb fuel adders, TSFC (which is never given correctly, I have mostly learned from PianoX how to value them and now make my own analysis in Tech/Ops with Gasturb) etc is is not to shabby! .

I shall point out that I have not needed to change any aero constants to get the model to fit, mostly it is about weight and TSFC as I have said to many times. The aero (which is about 10+ constants) is surprisingly stable, I normally don't touch it for a new airplane model unless it is motivated for technical reasons like 789 and 7810 laminar flow tail or such or the 777X getting a CFRP wing (which is smoother).

This gives me new confidence in the model, I must say I am a bit surprised ! .
"

So just to comment for the Tech/Ops reader, that I track so well is common but not always, I try to avoid chasing rumor numbers just for the sake of getting things close. A good example is A350-1000. At 153t OEW and 0.52 engines I am at 8382nm and Airbus says 8400. I also know that Zeke has said 153t was not correct ( or was it the Dubai 155t, can't remember but he is in the know and can't give out such data which one has to respect, good that he indicates it is not the real figure    ).

The value 153t was deduced from Airbus charts at the last Dubai airshow and before that at Farnbourough when the showed the Payload-Range charts for the -1000 vs the 77W and vs the original -1000 (MZFW is 220t ) :

http://i298.photobucket.com/albums/mm262/ferpe_bucket/A350-1000vs77WDubaislide_zpsd323fe10.jpg
Dubai airshow slide October 2011

http://i298.photobucket.com/albums/mm262/ferpe_bucket/AirbusParisslideA350-1000payload-range_zps9daf7758.jpg
Farnborough June 2011announcment of -1000 changes

My ratio analysis would say it should be closer to 150t, but then my model goes 270nm to far. When an OEM gives their data for a new model they keep a couple of % as margin, until I know more I will stay with 153t, to often the model has indicated something which is sandbagging from some end and it is not me  Wow! .



Non French in France
User currently offlineAstuteman From United Kingdom, joined Jan 2005, 10174 posts, RR: 97
Reply 76, posted (1 year 3 weeks 17 hours ago) and read 10253 times:
Support Airliners.net - become a First Class Member!

Quoting ferpe (Reply 75):
My ratio analysis would say it should be closer to 150t, but then my model goes 270nm to far.

What makes me laugh about your analysis here Ferpe, is the huge amount of scepticism that met the original A350-1000's performance specifications..

I for one, tend to trust both OEM's to have the engineering skill to determine their new design's capability with reasonable accuracy...

I accept that the 298t A350-1000 was out-hauled by the 773ER between 6000 and 7000Nm, but it seems reasonably clear that the 308t A350-1000 will do pretty much anything the 773ER will do, with a (small) bit in hand

Ah well.

Rgds


User currently offlineKarelXWB From Netherlands, joined Jul 2012, 12698 posts, RR: 35
Reply 77, posted (1 year 3 weeks 16 hours ago) and read 10240 times:

Quoting Astuteman (Reply 76):
I accept that the 298t A350-1000 was out-hauled by the 773ER between 6000 and 7000Nm, but it seems reasonably clear that the 308t A350-1000 will do pretty much anything the 773ER will do, with a (small) bit in hand

According to zeke (if I remember it correctly), the improved 308t A350-1000 will be able to do everything the CX 777-300ER can, but with a 20% lower fuel burn.



Only two things are infinite, the universe and human stupidity. And I'm not sure about the universe.
User currently offlineAviaponcho From France, joined Aug 2011, 641 posts, RR: 9
Reply 78, posted (1 year 3 weeks 13 hours ago) and read 10112 times:

My best guess for A350-1000 is 131t MWE ... put in a 350PAX Interior and I got 150t BOW and 155t DOW... (working on 2006 slides from airbus vs 777 and tracking MWE creep

User currently offlineStitch From United States of America, joined Jul 2005, 31243 posts, RR: 85
Reply 79, posted (1 year 3 weeks 8 hours ago) and read 10021 times:
Support Airliners.net - become a First Class Member!

Quoting KarelXWB (Reply 77):
According to zeke (if I remember it correctly), the improved 308t A350-1000 will be able to do everything the CX 777-300ER can, but with a 20% lower fuel burn.

  

He also noted that the A350-1000 is about 8 tons lighter than a 777-300ER in four classes (160t vs. 168t) with effectively identical passenger seating.


User currently offlinezeke From Hong Kong, joined Dec 2006, 9210 posts, RR: 76
Reply 80, posted (1 year 3 weeks 7 hours ago) and read 9981 times:

Quoting Stitch (Reply 79):

About 40t lighter at takeoff.



We are addicted to our thoughts. We cannot change anything if we cannot change our thinking – Santosh Kalwar
User currently offlineStitch From United States of America, joined Jul 2005, 31243 posts, RR: 85
Reply 81, posted (1 year 2 weeks 6 days 19 hours ago) and read 9791 times:
Support Airliners.net - become a First Class Member!

Quoting zeke (Reply 80):
About 40t lighter at takeoff.

With a 43t lower MTOW, that's not surprising.  

Seriously, it does show the benefits of the lower fuel burn - less fuel needing to be tanked.


Top Of Page
Forum Index

Reply To This Topic Boeing Vs. Airbus Wing Design Philosophies
Username:
No username? Sign up now!
Password: 


Forgot Password? Be reminded.
Remember me on this computer (uses cookies)
  • Tech/Ops related posts only!
  • Not Tech/Ops related? Use the other forums
  • No adverts of any kind. This includes web pages.
  • No hostile language or criticizing of others.
  • Do not post copyright protected material.
  • Use relevant and describing topics.
  • Check if your post already been discussed.
  • Check your spelling!
  • DETAILED RULES
Add Images Add SmiliesPosting Help

Please check your spelling (press "Check Spelling" above)


Similar topics:More similar topics...
Exit Signs On Boeing Vs. Airbus Aircraft posted Tue Feb 14 2012 14:57:49 by 1337Delta764
Differences Between The Boeing And Airbus Wing posted Fri Dec 31 2010 14:26:02 by Thrust
Aircraft Wall Insulation Boeing Vs Airbus posted Wed Jul 14 2010 11:19:42 by samair
Boeing Vs Airbus Climbout Rates posted Sun Aug 16 2009 14:48:36 by Chrisrad
Did Boeing Hide 707 Wing Design From Russia? posted Wed Feb 9 2005 20:46:51 by KAUST
Differences Between Boeing And Airbus Wing Designs posted Sat Feb 5 2005 23:53:59 by Thrust
DRY Air ... Boeing Vs Airbus posted Thu Feb 19 2004 01:26:25 by Airbear
Boeing Vs Airbus-A Business Perspective posted Fri Jan 10 2003 08:34:07 by Zachary92373
Normal Airbridges Vs. Over-wing Airbridges (AMS) posted Tue Oct 25 2011 09:01:36 by Pacific
Pressure Above Vs Below Wing posted Thu Sep 15 2011 04:44:29 by faro
Aircraft Wall Insulation Boeing Vs Airbus posted Wed Jul 14 2010 11:19:42 by samair
Boeing Vs Airbus Climbout Rates posted Sun Aug 16 2009 14:48:36 by Chrisrad
Did Boeing Hide 707 Wing Design From Russia? posted Wed Feb 9 2005 20:46:51 by KAUST
Differences Between Boeing And Airbus Wing Designs posted Sat Feb 5 2005 23:53:59 by Thrust
DRY Air ... Boeing Vs Airbus posted Thu Feb 19 2004 01:26:25 by Airbear
Boeing Vs Airbus-A Business Perspective posted Fri Jan 10 2003 08:34:07 by Zachary92373
Normal Airbridges Vs. Over-wing Airbridges (AMS) posted Tue Oct 25 2011 09:01:36 by Pacific
Pressure Above Vs Below Wing posted Thu Sep 15 2011 04:44:29 by faro

Sponsor Message:
Printer friendly format