Moderators: jsumali2, richierich, ua900, PanAm_DC10, hOMSaR

 
User avatar
Matt6461
Posts: 3032
Joined: Wed Oct 16, 2013 9:36 pm

Re: Analyzing The Airbus A380’s Premature End

Mon Mar 28, 2022 8:53 pm

flipdewaf wrote:
Matt6461 wrote:
Checking in on the A380 postmortems...

Folks have made the most important points:

  • 1. A388 wing/empennage/engines/MLG were sized for growth with disastrous results
  • 2. A380 fuselage is not very efficient for a double-decker (though still more efficient than any single decker)
  • 3. Leahy's excuse about A380's engines won't do: GEnx/T1000 are only ~3% better and ~3 years later: a predictable improvement.
  • 4. P2P vs. H2H is a stupid debate for non-specialist consumption.
  • 5. A380's efficiency edge over smaller planes insufficient to compensate for yield dilution.

Re #2, this factor may be highly underrated. Here, for example, is a 24.5ft circular fuselage:

Image

If we restrict MD height to 84in (vs. 100in on A380 and MD12), we can fit a 12-10 Y layout in a cross section with smaller diameter (therefore drag and weight) than the A380's (~25.6ft circle-equivalent).

This fuselage is so efficient that we can fit ~A388 capacity into a ~200ft fuselage that has about as much surface area as the 777-300ER's fuse: 1300m2 (vs. 1564m2 for A388).

Attach that fuselage to an 80m wing with effective span of 90m (winglets or folding mechanism) and it's hard not to get an L/D of >25.

With Ultrafan-generation engines, this plane would have ~half the per-seat fuel burn of 777-9 and feasibly lower trip COC.


If you make the changes you suggest to the fuselage (if you can fit the PAX in to the 61m fuselage) then you get a reduction in weight of about 8t. Adding the extra length to the wing loses that plus an extra ~20t in extra empty weight.

The cruise L/D changes from about 17.5 and goes to about 19.4, nowhere near 25...

Fred


A388 cruise L/D is around 19.5 from what I've seen.

Anyway, it's difficult to respond without any insight into the model you're employing. I assume you're using A388 as a base? IIRC from a couple years ago it wasn't clear if/how your model incorporates fuselage bending stress. Probably you've refined it since. What base parameter for A388 fuselage weight?

Using 777-9 as a base, we have about the same fuselage Swet, slightly more empennage area (depending on ultimate parameters), same with engine area. But a 90m wing reduces Di by ~40% per unit of lift (vs. ~230ft effective span). What base parameter for B779 fuselage and wing weight?

Using B779 at 21 L/D and cutting its Di by 40% (applied to 45% of total cruise drag) gives 22% L/D delta or 25.6 L/D for a very rough first cut.

In either base extrapolation, final L/D depends on wing area and thus cruise altitude, which depends on weight and other parameters. For the A388 base case, Ultrafan engines and optimized dimensions for wing/empennage/engines size impact L/D as much as span.
 
User avatar
Matt6461
Posts: 3032
Joined: Wed Oct 16, 2013 9:36 pm

Re: Analyzing The Airbus A380’s Premature End

Mon Mar 28, 2022 10:44 pm

Matt6461 wrote:
A388 cruise L/D is around 19.5 from what I've seen.


Referencing, e.g., this old thread from Ferpe, aka Bjorn Fehrme from Leeham News.

Image

As I note towards the bottom of that thread, there's some calculation errors in the spreadsheet. That said, it accords with analysis Bjorn has since posted on Leeham, which presumably is checked against proprietary data from Leeham's clients and contacts.

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

One other quick way to check the L/D calc is using wetted aspect ratio (WAR). Calculus 201 will show that for max L/D (setting transonic effects aside), WAR is a good first-order approximation (breaks down as we diverge from the condition of Di=Dp).

For a notional 24.5ft fuselage with overall Swet ~equal to B779's, WAR predicts rough linearity with span and L/D. Thus a 90m wing on a plane of B779's Swet (fuse Swet equal but reconfigured to VLA dimensions*) would give L/D delta of +~30%.

*The VLA would suffer a penalty for, inter alia, higher Cdp at constant Swet, so L/D delta would be <30%.
 
flipdewaf
Posts: 4521
Joined: Thu Jul 20, 2006 6:28 am

Re: Analyzing The Airbus A380’s Premature End

Tue Mar 29, 2022 5:13 pm

[quote="Matt6461"][/quote]
I see ferpes models and they don't seem right. If you take those L/D numbers and apply then to breguet they all come out with higher ranges than the aircraft in question.

In terms of the models it doesn't directly apply bending stress, the method (Stanford university method) applies fundamental principles and then uses correlative methods to scale correctly.

In terms of the design Idea I drew it up on CAD and it doesn't really look feasible, even at 9 on the upper deck. I think a 7ft ceiling is too low as there is such a lot of stuff to put in there. At 9 abreast the upper deck looks a bit too tight for my liking on the window seats. (these are 17" in the mockups)

Image

I don't think a 200ft fuselage is reasonable at this scale but I'm still drawing it up to see what I can learn.

in terms of reducing Di by 40% per unit of lift then that is right but we have to be very careful because wing weight scales very non linearly. A wing with the same area but scaled from 71->90m gains about 81% additional weight.

I think overall I would say that there is too much emphasis on Aero savings an not much on weight. A 1.2X increase in fuselage diameter would have an increase of 1.2^2 weight for pressure carrying loads which, at these short lengths is the dominating force.

Fred
 
sgbroimp
Posts: 279
Joined: Thu Sep 01, 2016 6:35 pm

Re: Analyzing The Airbus A380’s Premature End

Tue Mar 29, 2022 6:04 pm

Someone forgot to tell SIA about the 380's demise. They are re-starting JFK-FRA-SIN with brand new interiors! Business Class is a real knock out based on the photos.
 
flipdewaf
Posts: 4521
Joined: Thu Jul 20, 2006 6:28 am

Re: Analyzing The Airbus A380’s Premature End

Tue Mar 29, 2022 7:29 pm

Matt, Not sure if you were looking to match the A380 cabin area but I calculated the usable width top and bottom and then assumed a nose taper of ~1.5 and a tail taper of 2. Then I assumed that the top deck could only be of use in the constant section and the lower deck would make use of about 1/2 the length of the rear tapered section (modeled as a traezoid for simplicity) from this I figured you'd need about a 67m (228ft) fuselage to get the same useful floor area as the A380.

Not really sure what you expected from a wing area perspective? At the moment my estimate puts the fuselage at heavier than an A380s...

Fred
 
User avatar
Matt6461
Posts: 3032
Joined: Wed Oct 16, 2013 9:36 pm

Re: Analyzing The Airbus A380’s Premature End

Tue Mar 29, 2022 10:33 pm

flipdewaf wrote:
I see ferpes models and they don't seem right. If you take those L/D numbers and apply then to breguet they all come out with higher ranges than the aircraft in question.


The discrepancy may trace to whether you're using OEM/airline rules for reserves and diversions to safety airports. I also find that Breguet over-predicts range based on published L/D and SFC values, but not when you use those rules.

flipdewaf wrote:
In terms of the models it doesn't directly apply bending stress, the method (Stanford university method) applies fundamental principles and then uses correlative methods to scale correctly.


This is basically what I've used as well, if you're going off the Ilan Kroo presentation that used to be available online. I lost my old spreadsheet in a computer crash and the Stanford stuff isn't up anymore...

Flipdewaf wrote:
A wing with the same area but scaled from 71->90m gains about 81% additional weight.


That seems about right from the fundamentals.

But only for constant area. I'd escalate wing area to ~630m2 (13 AR) for a plane launched ~today. That means our average chord is ~equal to 777X's, and so is our moment of inertia in the torsion/torque box. Thus wing bending weight should be roughly proportional to the root bending moment times the span: it should be ~ (span delta)^2 or 60% higher than 777X.

But that's just a first cut because the wing provides its own bending relief. All in, I'd see ~50% delta to wing bending weight.

If we assume 777X's wing is ~80k lbs and that 50% of that is bending-dependent, the VLA's delta to wing bending weight would be ~20k lbs.

Then we'd have another ~8k lbs weight delta for the area-dependent weight of the larger wing.

If we round up and call it a 30k lbs delta, that's ~7.5% of 777X's OEW. Well worth paying for a massive L/D and capacity delta.

flipdewaf wrote:
In terms of the design Idea I drew it up on CAD and it doesn't really look feasible, even at 9 on the upper deck. I think a 7ft ceiling is too low as there is such a lot of stuff to put in there. At 9 abreast the upper deck looks a bit too tight for my liking on the window seats. (these are 17" in the mockups)


Thanks for the graphic.

A lot depends on the specific numbers you've use for the following parameters - I'll give my values, please give yours:

  • Height above centerline of UD: My value is 44in.
  • Height below centerline of MD: my value -52in.
  • Sidewall thickness. At critical points that define available seating width, I assume sidewall sculpting a la 777X. On the UD I use 4in, on MD 6in (thicker elsewhere).

My visual sense of your diagram is your UD (and MD) are higher than mine.

Even on my parameters, though, you're right that 10ab Y doesn't actually work on UD due to sidewall angle. This doesn't matter to the floor area calculation, however, as on the UD we'd have 8ab PY and reverse-herringbone J. The latter especially can use the full floor width for the feet of recumbent J pax. This is how the A380Plus achieved a big escalation of J capacity of the UD.

With UD sidewall sculpted to 4in at the critical point, it's 240in wide at 33in above the UD floor - ample room for J-class feet. I used 235in effective cabin width as a conservative estimate because 8ab PY can't extend the full 240in.

flipdewaf wrote:
I think overall I would say that there is too much emphasis on Aero savings an not much on weight. A 1.2X increase in fuselage diameter would have an increase of 1.2^2 weight for pressure carrying loads which, at these short lengths is the dominating force.


Well I haven't said anything about weight yet, only L/D.

Comparing again to B779:

  • 44% delta to sectional pressurization.
  • -19% delta to fuselage length (200ft vs. 246.75ft)
  • ~17% delta to total pressurization stress.
  • BUT CFRP delta of around -15% vs. B779's metal so total pressurization weight about equal.

Then let's look at bending stress:

  • Moment of inertia would escalate ~cubically with diameter so is ~75% greater.
  • Bending arm of cantilevered fuselage length (total length minus ~30ft wing box) decreases by ~20% for VLA.
  • Fuselage loads:
    • Pressure-dependent weight of tube is ~same as 777X
    • Payload and floor beams and seats/lavs is ~60% greater
    • Let's say net load at each integrated bending arm station is 50% greater for VLA than B779
  • Net of moment of inertia (1.75x) and integrated moment arm times loads (1.5x * .8x), VLA bending weight should be less than B779's

So with (VERY CONSERVATIVE ASSUMPTION) equal fuselage tube weight, the VLA's OEW over the B779 should include the following:

  • Wing - ~30k lbs
  • Floor beams: ~11k lbs (+1800ft2 @ 6lbs/ft2)
  • BFE: ~8k lbs (50lbs/pax for +160 3-class pax)

Call it 50k lbs so far. Adding that delta to B779 gives ~456k lbs OEW, which implies <850k MTOW given Ultrafan engines (-7% SFC vs. Ge9X) and 25 L/D.

VLA would need a bigger empennage than B779 but it probably also uses smaller engines with lower total thrust: With lower wingloading and spanloading, and as a quad, its OEI V2 thrust requirements are much lower than B779.

MLG would end up being ~same as B779 as well because even with slightly higher MTOW, VLA's MLG will be significantly shorter than B779's:

  • Shorter fuselage so rotation angle achievable with less MLG elevation
  • Quad engines need less ground clearance.

flipdewaf wrote:
Not sure if you were looking to match the A380 cabin area but I calculated the usable width top and bottom and then assumed a nose taper of ~1.5 and a tail taper of 2. Then I assumed that the top deck could only be of use in the constant section and the lower deck would make use of about 1/2 the length of the rear tapered section (modeled as a traezoid for simplicity) from this I figured you'd need about a 67m (228ft) fuselage to get the same useful floor area as the A380.


I see two very large underestimates here:

1. By using trapezoidal taper sections, you've massively underestimate floor area in those sections. Look at A380, for instance:

Image

Everything fore of Door 2 and aft of Door 4 is in the tapered sections. You'd say effective width is 50% of max width; it's pretty clear that 95% is closer to the real figure.

2. The assumption regarding the top section is extremely ungenerous. Again look at A380, where the UD is ~twice as long as the constant sections. It may be optimal to use double-bubble UD geometry in the tapering UD sections, which would imply a manageable pressurization penalty (UD floor takes the accelerated tension loads in those sections).

My VLA would also use a 747-style UD cockpit. I would cram the pilots into the smallest feasible space to maximize the UD and MD space.

On the MD, this allows extending the cabin to ~4ft from the nose, as in 747. Below the cockpit, MD height will be adequate only for galleys/lavs - say 6.3ft.

I'd also do as the MD12 did: extend the aft bulkhead to ~21ft short of total fuse length. I'd use the last ~12ft still-wide tapering fuselage as a single-level bags compartment for ~600 bags, which ensures that all LD3 slots (~30) are free for cargo.
Last edited by Matt6461 on Tue Mar 29, 2022 10:42 pm, edited 1 time in total.
 
User avatar
Matt6461
Posts: 3032
Joined: Wed Oct 16, 2013 9:36 pm

Re: Analyzing The Airbus A380’s Premature End

Tue Mar 29, 2022 10:37 pm

flipdewaf wrote:
At the moment my estimate puts the fuselage at heavier than an A380s...


Forgot to respond to this...

There's obviously something wrong with your model then.

Even at 228ft, a 24.5ft D fuselage is narrower and shorter than A388 and it's CFRP rather than (mostly) metal.

A380 is taller but both fuses should be pressure-dominated and therefore the higher moment of inertia shouldn't be that relevant. Difficult to respond as I don't see how you can get to where you got.
 
flipdewaf
Posts: 4521
Joined: Thu Jul 20, 2006 6:28 am

Re: Analyzing The Airbus A380’s Premature End

Wed Mar 30, 2022 5:18 pm

Matt6461 wrote:
flipdewaf wrote:
I see ferpes models and they don't seem right. If you take those L/D numbers and apply then to breguet they all come out with higher ranges than the aircraft in question.


The discrepancy may trace to whether you're using OEM/airline rules for reserves and diversions to safety airports. I also find that Breguet over-predicts range based on published L/D and SFC values, but not when you use those rules.

When I apply a 45minute hold at 1500ft level my numbers are generally within 1-2% of spec.
Matt6461 wrote:
flipdewaf wrote:
In terms of the models it doesn't directly apply bending stress, the method (Stanford university method) applies fundamental principles and then uses correlative methods to scale correctly.


This is basically what I've used as well, if you're going off the Ilan Kroo presentation that used to be available online. I lost my old spreadsheet in a computer crash and the Stanford stuff isn't up anymore...

Flipdewaf wrote:
A wing with the same area but scaled from 71->90m gains about 81% additional weight.


That seems about right from the fundamentals.

But only for constant area. I'd escalate wing area to ~630m2 (13 AR) for a plane launched ~today. That means our average chord is ~equal to 777X's, and so is our moment of inertia in the torsion/torque box. Thus wing bending weight should be roughly proportional to the root bending moment times the span: it should be ~ (span delta)^2 or 60% higher than 777X.

But that's just a first cut because the wing provides its own bending relief. All in, I'd see ~50% delta to wing bending weight.

All wings provide their own bending relief.
Matt6461 wrote:

If we assume 777X's wing is ~80k lbs and that 50% of that is bending-dependent, the VLA's delta to wing bending weight would be ~20k lbs.

No way the 777X wing is 80klb, unless you mean each? The wing on the 77W is over 100klb.
Matt6461 wrote:


Then we'd have another ~8k lbs weight delta for the area-dependent weight of the larger wing.

If we round up and call it a 30k lbs delta, that's ~7.5% of 777X's OEW. Well worth paying for a massive L/D and capacity delta.

flipdewaf wrote:
In terms of the design Idea I drew it up on CAD and it doesn't really look feasible, even at 9 on the upper deck. I think a 7ft ceiling is too low as there is such a lot of stuff to put in there. At 9 abreast the upper deck looks a bit too tight for my liking on the window seats. (these are 17" in the mockups)


Thanks for the graphic.

A lot depends on the specific numbers you've use for the following parameters - I'll give my values, please give yours:

  • Height above centerline of UD: My value is 44in.
  • Height below centerline of MD: my value -52in.
  • Sidewall thickness. At critical points that define available seating width, I assume sidewall sculpting a la 777X. On the UD I use 4in, on MD 6in (thicker elsewhere).


My visual sense of your diagram is your UD (and MD) are higher than mine.

You are right, I had the center of the main deck at the center if the tube. I have redone it as you described. I reduced wall thickness down to 100mm for the sake of ease (rather than varying).
You can fit 10x 17" in there but I think its only reasonably 9x (for the head smashnig reason. I'm still not convinced that its reasonable to assume no overhead lockers.
Matt6461 wrote:

Even on my parameters, though, you're right that 10ab Y doesn't actually work on UD due to sidewall angle. This doesn't matter to the floor area calculation, however, as on the UD we'd have 8ab PY and reverse-herringbone J. The latter especially can use the full floor width for the feet of recumbent J pax. This is how the A380Plus achieved a big escalation of J capacity of the UD.

With UD sidewall sculpted to 4in at the critical point, it's 240in wide at 33in above the UD floor - ample room for J-class feet. I used 235in effective cabin width as a conservative estimate because 8ab PY can't extend the full 240in.

flipdewaf wrote:
I think overall I would say that there is too much emphasis on Aero savings an not much on weight. A 1.2X increase in fuselage diameter would have an increase of 1.2^2 weight for pressure carrying loads which, at these short lengths is the dominating force.


Well I haven't said anything about weight yet, only L/D.

Comparing again to B779:

  • 44% delta to sectional pressurization.
  • -19% delta to fuselage length (200ft vs. 246.75ft)
  • ~17% delta to total pressurization stress.
  • BUT CFRP delta of around -15% vs. B779's metal so total pressurization weight about equal.


CFRP isn't saving 15%. A330->787 CFRP weight savings...? 77E-> A359 CFRP weight savings?
Matt6461 wrote:

Then let's look at bending stress:

  • Moment of inertia would escalate ~cubically with diameter so is ~75% greater.
  • Bending arm of cantilevered fuselage length (total length minus ~30ft wing box) decreases by ~20% for VLA.
  • Fuselage loads:
    • Pressure-dependent weight of tube is ~same as 777X
    • Payload and floor beams and seats/lavs is ~60% greater
    • Let's say net load at each integrated bending arm station is 50% greater for VLA than B779
  • Net of moment of inertia (1.75x) and integrated moment arm times loads (1.5x * .8x), VLA bending weight should be less than B779's

So with (VERY CONSERVATIVE ASSUMPTION) equal fuselage tube weight, the VLA's OEW over the B779 should include the following:

  • Wing - ~30k lbs
  • Floor beams: ~11k lbs (+1800ft2 @ 6lbs/ft2)
  • BFE: ~8k lbs (50lbs/pax for +160 3-class pax)

Call it 50k lbs so far. Adding that delta to B779 gives ~456k lbs OEW, which implies <850k MTOW given Ultrafan engines (-7% SFC vs. Ge9X) and 25 L/D.

VLA would need a bigger empennage than B779 but it probably also uses smaller engines with lower total thrust: With lower wingloading and spanloading, and as a quad, its OEI V2 thrust requirements are much lower than B779.

MLG would end up being ~same as B779 as well because even with slightly higher MTOW, VLA's MLG will be significantly shorter than B779's:

  • Shorter fuselage so rotation angle achievable with less MLG elevation
  • Quad engines need less ground clearance.

flipdewaf wrote:
Not sure if you were looking to match the A380 cabin area but I calculated the usable width top and bottom and then assumed a nose taper of ~1.5 and a tail taper of 2. Then I assumed that the top deck could only be of use in the constant section and the lower deck would make use of about 1/2 the length of the rear tapered section (modeled as a traezoid for simplicity) from this I figured you'd need about a 67m (228ft) fuselage to get the same useful floor area as the A380.


I see two very large underestimates here:

1. By using trapezoidal taper sections, you've massively underestimate floor area in those sections. Look at A380, for instance:

Image

Everything fore of Door 2 and aft of Door 4 is in the tapered sections. You'd say effective width is 50% of max width; it's pretty clear that 95% is closer to the real figure.

2. The assumption regarding the top section is extremely ungenerous. Again look at A380, where the UD is ~twice as long as the constant sections. It may be optimal to use double-bubble UD geometry in the tapering UD sections, which would imply a manageable pressurization penalty (UD floor takes the accelerated tension loads in those sections).


The problem is this is really difficult to visualise and take account of in a 2D space, I would wager the best way to understand the length is to take the A380 as a starting point (for a double decker) and see the difference at 10*8 vs 12*9. At 21 abreast 880 seats takes ~42 rows and 880 seats at 18 abreast takes 49 rows. at high density (30" pitch?) that's a reduction on the A380 of ~5.3m giving a fuselage of ~65m.
Matt6461 wrote:

My VLA would also use a 747-style UD cockpit. I would cram the pilots into the smallest feasible space to maximize the UD and MD space.

It isn't great for transonic flow. There is a reason the modern jetliners have massive foreheads.
Matt6461 wrote:

On the MD, this allows extending the cabin to ~4ft from the nose, as in 747. Below the cockpit, MD height will be adequate only for galleys/lavs - say 6.3ft.

I'd also do as the MD12 did: extend the aft bulkhead to ~21ft short of total fuse length. I'd use the last ~12ft still-wide tapering fuselage as a single-level bags compartment for ~600 bags, which ensures that all LD3 slots (~30) are free for cargo.


Fred
 
dynamo12
Posts: 79
Joined: Mon Dec 29, 2008 6:07 am

Re: Analyzing The Airbus A380’s Premature End

Wed Mar 30, 2022 6:12 pm

lightsaber wrote:
Many issues:
1. "Cruise ship" stairways that took up too much space. That could have added "free seats"
2. Prior weight discussion
3. Prior generation engines
4. Weight added by Catia error

I believe there was a market for the A380. It just wasn't optimized. It became a horse by committee.

Lightsaber


Agreed with all of this. The advantage if anything has to be in CASM. The slot limits stuff is bogus. If you have a (massive) CASM advantage there is still a story there.

If you did a highly optimized design focused on CASM (not glory), can you recover the 4 engine penalty? I think so based on all the things they didn't do here. No CATIA, do a combined engine development with a twin to leverage stuff the new gen is using, go light and optimized for existing size.

For someone more expert, are intra-deck stairways a must? The massive space use (some of these are inline with seating or curve down hugely) are crazy. Can these planes not just be loaded from upper desk, maybe one narrow set of stair somewhere down?
 
User avatar
william
Posts: 3975
Joined: Thu Jun 10, 1999 1:31 pm

Re: Analyzing The Airbus A380’s Premature End

Wed Mar 30, 2022 6:39 pm

Taxi645 wrote:
flipdewaf wrote:
I did some calculations on how the A380 would have performed with nominal 787 engines. For this I ran my model with the A380 with standard engines rated with cruise SFC of 0.525lb/hr/lb over a near spec range flight. Then the SFC was dropped to 0.5 to represent a 787 technology level engine. Finally the geometry and weights of the A380 were optimised so that the excess performance afforded by the lower SFC engines was not there.

The simple addition of the lower SFC engine reduced fuel burn on the spec mission by 5.8%. When the airframe was optimised around this new engine the fuel burn dropped 8% from original.

The MTOW required dropped from 575t down to 550t. The empty weight dropped from 277t down to 272t.

Fred


Sent from my iPad using Tapatalk


What would be interesting is to know the brochure range and fuelburn per seat mile delta if they had gone simply for an optimized 800 of say 505T MTOW.

- 15% less wing surface with both less induded and parasitic drag.
- lighter engines with less surface area
- lighter and smaller control surfaces
- weight savings for not having so much 900 capability built in.

Would it have had enough range and enough of an CASM advantage to sufficiently offset the yield and frequency advantage of the twins?

A lighter 800 could possibly have had a pathway to a 2nd generation that the actual unoptimized 800 did not.


What kind of weight savings are we stating here? As a percentage, 10 percent lighter?
 
User avatar
Matt6461
Posts: 3032
Joined: Wed Oct 16, 2013 9:36 pm

Re: Analyzing The Airbus A380’s Premature End

Wed Mar 30, 2022 7:01 pm

flipdewaf wrote:
All wings provide their own bending relief.


Magnitude and the relative values of wing (+engine) vs. fuselage (+contents) matter.

Here we're increasing wing weight by ~40% (first cut) while fuselage+contents weight increases by only ~20%.

What's more, the quad layout provides ~50% more engine bending relief in the 6g static maneuver condition, due to the longer moment arm of the outboard engine. In -2.5g taxi bump condition it will be worse but the relatively much smaller fuel load (relative to intrinsic wing strength) will obviate that concern.

The critical bending moment, net of wing/engine relief, will therefore be significantly smaller than what a span-only analysis would show.

flipdewaf wrote:
You can fit 10x 17" in there but I think its only reasonably 9x (for the head smashnig reason.


Re the UD I've already conceded that 10ab Y isn't practical. I only said "12-10" to give an approximate idea of the floor capacity. Once more, using the UD for reverse herringbone J allows using the full 240 inches of floor space, which makes the UD effectively wider than 777X and 747 for J class. A good VLA will shift proportionally away from Y, as PY will cost less on it than Y on other planes. So all basic Y seats go on the MD (this also restricts UD floor beam weight).

Your MD seat layout cheats the arrangement by several inches because you extend the armrest to the seat bottom. The outboard contour would be more like this.

Image

Let's analyze the requirements for 18in 12ab, 18in aisles, 1.5in armrest:

Seats: 12 * 18 = 216in
Armrests: (12 + 3) * 1.5 = 22.5in
Aisles: 2*18 = 36in

Sum: 274.5in

flipdewaf wrote:
I'm still not convinced that its reasonable to assume no overhead lockers.


I don't so assume.

Here's A220's cross section:

Image

As you can see, 7ft max headroom and plenty of space for overhead bins. Except on the VLA there's much more space because the bins aren't constricted by sloping sidewalls. They don't need to angle downwards (horizontal, rather) so actual headroom above seated pax will be greater on the VLA than on A220.

flipdewaf wrote:
CFRP isn't saving 15%. A330->787 CFRP weight savings...? 77E-> A359 CFRP weight savings?


That's at plane level, not component level. The plane-level savings I've modeled for CFRP fuse (~10k lbs) would be ~2.3% of OEW.

flipdewaf wrote:
No way the 777X wing is 80klb, unless you mean each? The wing on the 77W is over 100klb.


I've been surprised by how light the actual core wing components are when that data has been available - IIRC ~20% of OEW on A380. There's a pdf of the A380 transport regime somewhere online (maybe gone?), showing the weight of transported loads (wings, fuselage sections, empennage). Can anybody share that?

Even if the 777X wing is 100k lbs, that makes my wing weight delta only ~7,500lbs greater - now OEW is ~463,000lbs.

Even if the VLA's OEW is 500k lbs (can't see any case for that), with ~60% higher capacity and ~20% higher L/D, VLA is still burning ~half of 777X's fuel/pax.

flipdewaf wrote:
The problem is this is really difficult to visualise and take account of in a 2D space


I don't find it that difficult. Here's a rough schematic of the foreward tapering MD section:

Image

I calculate the area of that section at 85% of a constant section.

Here's the rear:

Image

I calculate that section at 90% of a constant section. Note this is more tapered than, say, a B777 aft:

Image

...which is probably ~93% of a constant section.

Then it's a matter of simple math for a 200ft fuselage:

LOA: 200ft
Fore taper: 1.6*D = 39.2ft
Aft taper: 3*D = 73.5ft
Constant section=87.3ft

Parallel sections (UD and MD)

Length: 87.3ft
Width: 42.58ft (276in MD, 235in UD)
Block coefficient: 100%
Area: 3717ft2, 345.5m2

Tapering MD
Distance from nose: 4ft
Distance from tail: 33ft
Cabin tapering length: 75.7ft
Block coefficient: 90% applied to max width of 276in
Area: 145.6m2

Tapering UD
Distance from nose: 17ft
Distance from tail: 33ft
Cabin tapering length: 62.7ft
Block coefficient: 85% applied to max width of 235in
Area: 114.1m2

TOTAL AREA: 605m2

...that's actually 80m2 more than A380's 525m2.

So if you want to apply a lower "block coefficient" to the tapering sections, that's fine. Even at 70% it's hard to get a smaller cabin than the A380.

Fact is the A380 was just a horribly inefficient design. It has a massive empty tailcone (~45ft) and wastes a ton of space in the forward UD between the grand stairway and the isolated space ahead of it (EY's super-F suite reclaims some of that). The tailcone is so big because the H-stab is the size of an A320 wing, which was only necessary because the wing/engines/MLG etc. were too big because Airbus wanted an infeasibly large -900 on an 80m wing. Plus the MD height is too great for little real economic benefit. You'll either stick price-sensitive Y pax on my VLA's MD or more-than-compensate them for headroom by providing 8/9/10ab PY/Y+ at less than the cost of a standard Y seat.

flipdewaf wrote:
It isn't great for transonic flow. There is a reason the modern jetliners have massive foreheads.


I'm aware but the effect can't be very big. 747 cruised at slightly higher speed (.855M) than other planes, which wouldn't be done if the transonic effect of the UD cockpit were great.

This arrangement give ~12m2 of space to concentrate galleys and a couple lavs. That's ~2.5% capacity enhancement. There's no way a 747-size transonic drag penalty equates to 2.5% L/D penalty.

In any event, we'd design the cockpit in light of what's been learned over the last 60 years. Boeing's 1990's VLA studies, for example, included a UD cockpit faired into the fuselage contour:

Image
 
flipdewaf
Posts: 4521
Joined: Thu Jul 20, 2006 6:28 am

Re: Analyzing The Airbus A380’s Premature End

Wed Mar 30, 2022 7:38 pm

Matt6461 wrote:
flipdewaf wrote:
All wings provide their own bending relief.


Magnitude and the relative values of wing (+engine) vs. fuselage (+contents) matter.

Here we're increasing wing weight by ~40% (first cut) while fuselage+contents weight increases by only ~20%.

What's more, the quad layout provides ~50% more engine bending relief in the 6g static maneuver condition, due to the longer moment arm of the outboard engine. In -2.5g taxi bump condition it will be worse but the relatively much smaller fuel load (relative to intrinsic wing strength) will obviate that concern.

The critical bending moment, net of wing/engine relief, will therefore be significantly smaller than what a span-only analysis would show.

Except it isn’t a span only analysis… it’s span, thickness, MZFW, area MTOW. My estimate has it at over 105t right now.
The fuselage is 37t, slightly less than the A380.

Overall it’s 251t empty weight…it’s always taking things in isolation followed by special pleading round here.
Matt6461 wrote:

flipdewaf wrote:
You can fit 10x 17" in there but I think its only reasonably 9x (for the head smashnig reason.


Re the UD I've already conceded that 10ab Y isn't practical. I only said "12-10" to give an approximate idea of the floor capacity. Once more, using the UD for reverse herringbone J allows using the full 240 inches of floor space, which makes the UD effectively wider than 777X and 747 for J class. A good VLA will shift proportionally away from Y, as PY will cost less on it than Y on other planes. So all basic Y seats go on the MD (this also restricts UD floor beam weight).

Your MD seat layout cheats the arrangement by several inches because you extend the armrest to the seat bottom. The outboard contour would be more like this.

Image

Let's analyze the requirements for 18in 12ab, 18in aisles, 1.5in armrest:

Seats: 12 * 18 = 216in
Armrests: (12 + 3) * 1.5 = 22.5in
Aisles: 2*18 = 36in

Sum: 274.5in

flipdewaf wrote:
I'm still not convinced that its reasonable to assume no overhead lockers.


I don't so assume.

Here's A220's cross section:

Image

As you can see, 7ft max headroom and plenty of space for overhead bins. Except on the VLA there's much more space because the bins aren't constricted by sloping sidewalls. They don't need to angle downwards (horizontal, rather) so actual headroom above seated pax will be greater on the VLA than on A220.


Except that isn’t negotiating past a centre group of 6!! Seats.
Matt6461 wrote:

flipdewaf wrote:
CFRP isn't saving 15%. A330->787 CFRP weight savings...? 77E-> A359 CFRP weight savings?


That's at plane level, not component level. The plane-level savings I've modeled for CFRP fuse (~10k lbs) would be ~2.3% of OEW.

flipdewaf wrote:
No way the 777X wing is 80klb, unless you mean each? The wing on the 77W is over 100klb.


I've been surprised by how light the actual core wing components are when that data has been available - IIRC ~20% of OEW on A380. There's a pdf of the A380 transport regime somewhere online (maybe gone?), showing the weight of transported loads (wings, fuselage sections, empennage). Can anybody share that?


https://www.google.co.uk/url?sa=t&rct=j ... 3q6osINev7

Matt6461 wrote:

Even if the 777X wing is 100k lbs, that makes my wing weight delta only ~7,500lbs greater - now OEW is ~463,000lbs.

It’s the 77W wing that’s 100klb, the 779x is more like 122klb
Matt6461 wrote:

Even if the VLA's OEW is 500k lbs (can't see any case for that), with ~60% higher capacity and ~20% higher L/D, VLA is still burning ~half of 777X's fuel/pax.

It’ll be closer to 550klb, and with 555pax doing sin-lax nil wind use about 176t of fuel. Assuming SFC of 0.48 lb/hr/lb
Matt6461 wrote:

flipdewaf wrote:
The problem is this is really difficult to visualise and take account of in a 2D space


I don't find it that difficult. Here's a rough schematic of the foreward tapering MD section:

Image

I calculate the area of that section at 85% of a constant section.

Here's the rear:

Image

I calculate that section at 90% of a constant section. Note this is more tapered than, say, a B777 aft:


It’s difficult in 3d and so you show in 2d? Here’s the clue, the roof gets lower…
Matt6461 wrote:

Image

...which is probably ~93% of a constant section.

Then it's a matter of simple math for a 200ft fuselage:

LOA: 200ft
Fore taper: 1.6*D = 39.2ft
Aft taper: 3*D = 73.5ft
Constant section=87.3ft

Parallel sections (UD and MD)

Length: 87.3ft
Width: 42.58ft (276in MD, 235in UD)
Block coefficient: 100%
Area: 3717ft2, 345.5m2

Tapering MD
Distance from nose: 4ft
Distance from tail: 33ft
Cabin tapering length: 75.7ft
Block coefficient: 90% applied to max width of 276in
Area: 145.6m2

Tapering UD
Distance from nose: 17ft
Distance from tail: 33ft
Cabin tapering length: 62.7ft
Block coefficient: 85% applied to max width of 235in
Area: 114.1m2

TOTAL AREA: 605m2

...that's actually 80m2 more than A380's 525m2.

So if you want to apply a lower "block coefficient" to the tapering sections, that's fine. Even at 70% it's hard to get a smaller cabin than the A380.


Fact is the A380 was just a horribly inefficient design. It has a massive empty tailcone (~45ft) and wastes a ton of space in the forward UD between the grand stairway and the isolated space ahead of it (EY's super-F suite reclaims some of that). The tailcone is so big because the H-stab is the size of an A320 wing, which was only necessary because the wing/engines/MLG etc. were too big because Airbus wanted an infeasibly large -900 on an 80m wing. Plus the MD height is too great for little real economic benefit. You'll either stick price-sensitive Y pax on my VLA's MD or more-than-compensate them for headroom by providing 8/9/10ab PY/Y+ at less than the cost of a standard Y seat.


Take that same approach and apply to the 777 or 787 or A350 and see if your numbers are close to real world, if now why not.
Matt6461 wrote:

flipdewaf wrote:
It isn't great for transonic flow. There is a reason the modern jetliners have massive foreheads.


I'm aware but the effect can't be very big. 747 cruised at slightly higher speed (.855M) than other planes, which wouldn't be done if the transonic effect of the UD cockpit were great.

This arrangement give ~12m2 of space to concentrate galleys and a couple lavs. That's ~2.5% capacity enhancement. There's no way a 747-size transonic drag penalty equates to 2.5% L/D penalty.


And yet it isn’t applied to subsequent aircraft….
Matt6461 wrote:

In any event, we'd design the cockpit in light of what's been learned over the last 60 years. Boeing's 1990's VLA studies, for example, included a UD cockpit faired into the fuselage contour:

Image


And that was a concept and so didn’t have to be certificated against the standards for cockpit visibility….

Fun times.

Fred


Sent from my iPad using Tapatalk
 
User avatar
Taxi645
Posts: 606
Joined: Wed Feb 22, 2017 7:29 pm

Re: Analyzing The Airbus A380’s Premature End

Thu Mar 31, 2022 7:51 am

william wrote:
Taxi645 wrote:
flipdewaf wrote:
I did some calculations on how the A380 would have performed with nominal 787 engines. For this I ran my model with the A380 with standard engines rated with cruise SFC of 0.525lb/hr/lb over a near spec range flight. Then the SFC was dropped to 0.5 to represent a 787 technology level engine. Finally the geometry and weights of the A380 were optimised so that the excess performance afforded by the lower SFC engines was not there.

The simple addition of the lower SFC engine reduced fuel burn on the spec mission by 5.8%. When the airframe was optimised around this new engine the fuel burn dropped 8% from original.

The MTOW required dropped from 575t down to 550t. The empty weight dropped from 277t down to 272t.

Fred


Sent from my iPad using Tapatalk


What would be interesting is to know the brochure range and fuelburn per seat mile delta if they had gone simply for an optimized 800 of say 505T MTOW.

- 15% less wing surface with both less induded and parasitic drag.
- lighter engines with less surface area
- lighter and smaller control surfaces
- weight savings for not having so much 900 capability built in.

Would it have had enough range and enough of an CASM advantage to sufficiently offset the yield and frequency advantage of the twins?

A lighter 800 could possibly have had a pathway to a 2nd generation that the actual unoptimized 800 did not.


What kind of weight savings are we stating here? As a percentage, 10 percent lighter?


I get to about 12.5% lighter. I think Matt and Fred, who have a better technical understanding, consider lighter still possible.

Weight is everything in airplane design. Leaving so much on the table in a competitive market, really was a big and hard to understand error on Airbus part.
 
User avatar
Taxi645
Posts: 606
Joined: Wed Feb 22, 2017 7:29 pm

Re: Analyzing The Airbus A380’s Premature End

Thu Mar 31, 2022 7:59 am

I have a question regarding fuselage width' relation to induced drag. If you widen a fuselage within a fixed span as Matt has done in his proposal, one would expect a negative effect on induced drag as less span is left for the wing itself. So more lift is created from a shorter wing (the added lift from the fuselage is marginal I assume), which would increase induced drag even though the basic formulas about induced drag and span don't account for this. Would that be correct?
 
User avatar
Matt6461
Posts: 3032
Joined: Wed Oct 16, 2013 9:36 pm

Re: Analyzing The Airbus A380’s Premature End

Thu Mar 31, 2022 7:35 pm

Taxi645 wrote:
I have a question regarding fuselage width' relation to induced drag. If you widen a fuselage within a fixed span as Matt has done in his proposal, one would expect a negative effect on induced drag as less span is left for the wing itself. So more lift is created from a shorter wing (the added lift from the fuselage is marginal I assume), which would increase induced drag even though the basic formulas about induced drag and span don't account for this. Would that be correct?


The wing area definition includes the "virtual wing" - the area under/over the fuselage. Here's a good primer on some related issues.

The fuselage interferes with the wing to a certain extent, reducing lift over the "virtual wing." A double decker likely increases the degree of this interference actually, reducing the span efficiency ("e") and therefore somewhat increasing induced drag. The larger driver of fuselage interference is width, however, as a percentage of total span.

A couple thing to watch out for in relation to this:

1. The wetted area of the wing will overestimated unless subtracts the virtual wing (or subtracts virtual wing area from fuselage area but easier to subtract the virtual wing).

2. I've seen folks on a.net attribute the lift of the virtual wing - usually ~15% of lift - to "fuselage lift," which is a grave error. Significant fuselage lift is something strenuously to avoid in conventional airliners because the ridiculously low AR of a fuselage implies immense induced drag from fuselage lift. A BWB improves fuselage lift's AR significantly but IMO the BWB won't ever be the optimal solution in part because its AR limits it to L/D's in the mid-20's (at best) in cruise and much worse at takeoff (due to high Di but low Dp). A good 2-deck VLA with a 2030's wing the mid-teens AR should already beat the BWB on L/D (likely on structural efficiency as well unless CNTP solves the pressurization problem).
 
User avatar
Matt6461
Posts: 3032
Joined: Wed Oct 16, 2013 9:36 pm

Re: Analyzing The Airbus A380’s Premature End

Thu Mar 31, 2022 9:39 pm

A 2-deck VLA along these lines might be best if/when the OEM's figure out aeroelastic wings like Airbus's albatross project, which makes 18 AR feasible. A VLA with a 700m2, 18AR wing would be around 30 L/D.

Its future single-deck competition would also benefit but not nearly as much as the VLA. Why? Because high-AR wings further reduce Di, which is >50% of total drag on A380 but only ~30% on big single-aisles. The VLA Swet efficiency (thus Dp efficiency) would become only more relatively prominent as typical wing AR's increase.

The other idea for a VLA launched in the foreseeable future would be to set the program up as (somewhat) easily convertible to liquid hydrogen power. The primary drawback for airliners of LH2 is that fuel's massive volume requirements. A large, voluminous fuselage cross section containing LH2 tanks is probably needed even for 3hr range. A 24.5ft diameter VLA launched with a ~200ft fuselage could be stretched to 777-9 fineness ratio (~300ft fuselage). Using the added volume as LH2 tankage should confer ~TATL range. That's a less efficient plane but of course all planes will be less efficient on LH2, ceteris paribus. Alternatively, if the future path is towards drop-in sustainable aviation fuel rather than LH2, the VLA's fuel efficiency edge will shine brighter when everyone is paying >$6/gal for SAF. At 30 L/D and ~450k OEW, an A380-sized 2030's VLA could be getting ~350 seat-miles/gal (economy seats only), which makes aviation economically sustainable even with expensive SAF.
 
flipdewaf
Posts: 4521
Joined: Thu Jul 20, 2006 6:28 am

Re: Analyzing The Airbus A380’s Premature End

Fri Apr 01, 2022 11:55 am

Matt6461 wrote:
Do you have a cite for 77W figure? Not disagreeing, my 20% seems to be an underestimate from the last time I thought about this stuff much a few years ago.

https://courses.washington.edu/ie337/Bo ... 0Facts.pdf
ACC131 is the wing-body join station, They lift the wing structure.

Matt6461 wrote:
Even at 122k lbs 779x wing, on the above-discussed parameters wing weight delta still doesn't exceed 50k lbs (and before you say it, I previously used +20% delta to fuse+contents due to payload/MZFW, which is implicit in my calculation that the VLA's 200ft fuse tube alone is probably lighter than 779's).


The weight the wing has to carry:-
The fuselage itself:- its 20% higher diameter than the 779X and so that equates to 44% higher weight per unit length. You claim its 20% shorter so that then equates to a total gain of 20%. We have ignored the fact that there's an extra floor...
The payload reasonably goes up by 30% (the current 779X cannot even take the 880pax with its payload limits)
The Weight of the cabin furnishings goes up by 60% (550->880)
The same rule applies to the required environmental and entertainment system weights.
The wing also needs to carry the additional size and weight from the Vertical and horizontal stabilizers.

Unsurprisingly, if you put more people and things in a tube that's bigger it weighs more, a lot more.

Lets for the sake of argument say that the increase in weight needing to be carried by the wing was increased by 30%.
You say the wing has the same root chord and increases span from 71.8m to 90. this is then a case of simply stretching the wing over that length and assuming the same spanwise loading distribution makes the comparison quite simple.

The spanwise loading increases 30% per unit length based on additional loads needing to be carried and reduces by 20% based on span increase giving a total increase in spanwiswe loading of ~3.8%

the increase in bending moment for any given section of wing then is given as wl^2/2 and as such the bending moments to be countered at any given section scale as (w*1.038)*(90/71.8)^2

Gives a bending moment to be carried increase of 1.63* the requirement of the 779X. The wing then has to carry these increased bending moments over a distance of 25% greater than the 779X (90/71.8) so the material required to carry these increase loads is 2.04* the 779X. not 50klb but 116klb more.

If we do the same calculation based on your 20% estimate then the figure comes out as 99klb more. There's a reason why aspect ratios seem to all fall in the same bands for the same categories of aircraft...


Fred
 
User avatar
Matt6461
Posts: 3032
Joined: Wed Oct 16, 2013 9:36 pm

Re: Analyzing The Airbus A380’s Premature End

Fri Apr 01, 2022 8:37 pm

@flipdewaf

Things you're ignoring:

  • The ~75% higher moment of inertia in the 24.5ft D fuse.
  • The prominence of bending reinforcement in the 779's fuse, which is 20% longer, made of heavier stuff, and has ~40% less inherent sectional bending resistance.
  • The actual Payload of a good VLA. There's no compelling reason seat 880 PAX. To repeat once more, the primary strategy is to sell more space rather than simply more tickets. At ~700 max exit-limited capacity and because we have little belly space, payload need not exceed 777-9's much.
  • your pressure-dependent weight calculation stems form a refusal to believe one gets >A380 capacity at <200ft fuselage length. This stems from a simply indefensible conception of tapering fuselages as mostly-empty trapezoids.
  • The wing bending weight impact of quad layout.
  • The Stanford wing weight factor relating MTOW/MZFW. Due to L/D and SFC, VLA's is much better at equal standard range.
  • that I've already added 11k lbs for floor beams

Taking all the foregoing together, I'd estimate that net bending moment adjustment for weight/arm of (1) fuse+contents+empennage vs. (2) wing+engines+fuel (latter relating to -2.5g condition via MTOW/MZFW) - that this factor favors the VLA at some point north of 190ft fuse (still A388 floor area) but perhaps south of 200ft fuse.

Rather than continue the discussion in this somewhat disorganized fashion, and because the thread is approaching 6 months and will be locked soon, I'll shoot for a more structured discussion in TechOps when I have time to pull one together.

Final note re wing AR's consistency across designs. You're correct of course but ignore two important factors. First, AR is increasing with tech - we now have ~12AR wings on the recent small NB's. Airbus's near-term goal is 14 AR. Second, The VLA's relative efficiency on fuse weight implies a different distribution of OEW, favoring a higher % of wing weight.
 
User avatar
Matt6461
Posts: 3032
Joined: Wed Oct 16, 2013 9:36 pm

Re: Analyzing The Airbus A380’s Premature End

Fri Apr 01, 2022 9:12 pm

flipdewaf wrote:
https://courses.washington.edu/ie337/Bo ... 0Facts.pdf
ACC131 is the wing-body join station, They lift the wing structure.


Thanks.

Do you know, btw, whether the ACC131 lift includes things like the outer MLG and high-lift devices? Absent component values for items like spars/skins/stringers/ribs, it's sometimes hard to get a read on what's true wing structural weight and what's stuff appended to the wing.

Other factors would include engine pylons and the fuel plumbing system. Presumably at least the latter is always installed in a skinned wing set. Even if the MLG struts aren't attached, their attachment points and local reinforcement would be inherent to the wing structure.

Another small factor is that 777x's center wing box remains metal to avoid complexity in a metal-plastic wing-body join. Probably worth a ton or so, and it's wing weight that doesn't confer bending relief.
 
flipdewaf
Posts: 4521
Joined: Thu Jul 20, 2006 6:28 am

Analyzing The Airbus A380’s Premature End

Fri Apr 01, 2022 9:30 pm

Matt6461 wrote:
@flipdewaf

Things you're ignoring:

  • The ~75% higher moment of inertia in the 24.5ft D fuse.

Moment of inertia is of little to no relevance in pressure dominated regime.
Matt6461 wrote:
  • The prominence of bending reinforcement in the 779's fuse, which is 20% longer, made of heavier stuff, and has ~40% less inherent sectional bending resistance.


  • Made of heavier stuff? Can you show that? 787 and A350 can’t.
    Matt6461 wrote:
  • The actual Payload of a good VLA. There's no compelling reason seat 880 PAX. To repeat once more, the primary strategy is to sell more space rather than simply more tickets. At ~700 max exit-limited capacity and because we have little belly space, payload need not exceed 777-9's much.

  • But somehow the 779X needs 70+ t of payload? Suddenly this aircraft operates ina different paradigm?
    Matt6461 wrote:
  • your pressure-dependent weight calculation stems form a refusal to believe one gets >A380 capacity at <200ft fuselage length. This stems from a simply indefensible conception of tapering fuselages as mostly-empty trapezoids.

  • This was accepted in my calculation. I still don’t believe it’s reasonable.
    Matt6461 wrote:
  • The wing bending weight impact of quad layout.


  • Because of a smaller outboard engine saving 50klb of wing weight through wing bending relief? If that was reasonable then you’d see ballast added outboard to save weight inboard…. If you think it’s true by all means show some calcs.
    Matt6461 wrote:
  • The Stanford wing weight factor relating MTOW/MZFW. Due to L/D and SFC, VLA's is much better at equal standard range.

  • I don’t know what this means.
    Matt6461 wrote:
  • that I've already added 11k lbs for floor beams


  • Taking all the foregoing together, I'd estimate that net bending moment adjustment for weight/arm of (1) fuse+contents+empennage vs. (2) wing+engines+fuel (latter relating to -2.5g condition via MTOW/MZFW) - that this factor favors the VLA at some point north of 190ft fuse (still A388 floor area) but perhaps south of 200ft fuse.

    What are your estimates based on?
    Matt6461 wrote:

    Rather than continue the discussion in this somewhat disorganized fashion, and because the thread is approaching 6 months and will be locked soon, I'll shoot for a more structured discussion in TechOps when I have time to pull one together.

    Final note re wing AR's consistency across designs. You're correct of course but ignore two important factors. First, AR is increasing with tech - we now have ~12AR wings on the recent small NB's. Airbus's near-term goal is 14 AR. Second, The VLA's relative efficiency on fuse weight implies a different distribution of OEW, favoring a higher % of wing weight.


    Higher AR often less impactful on NBs due to slower speeds.

    Fred


    Sent from my iPad using Tapatalk
     
    frmrCapCadet
    Posts: 5465
    Joined: Thu May 29, 2008 8:24 pm

    Re: Analyzing The Airbus A380’s Premature End

    Fri Apr 01, 2022 11:20 pm

    I suspect, all of the techie discussions aside (which I have enjoyed reading), was there any possible 4 engine fuselage with available engines which Airbus could have designed? And how much bigger than a 747 should it have been? The 787/350s seem totally devastating competition to everything from the 767/330s on up to 340/777s and even the yet to be certified 779.
     
    User avatar
    Matt6461
    Posts: 3032
    Joined: Wed Oct 16, 2013 9:36 pm

    Re: Analyzing The Airbus A380’s Premature End

    Sat Apr 02, 2022 12:52 am

    frmrCapCadet wrote:
    I suspect, all of the techie discussions aside (which I have enjoyed reading), was there any possible 4 engine fuselage with available engines which Airbus could have designed? And how much bigger than a 747 should it have been? The 787/350s seem totally devastating competition to everything from the 767/330s on up to 340/777s and even the yet to be certified 779.


    Re size of a VLA: Whichever size points confers the optimal tradeoff between trip cost and per-seat efficiency. A rough metric of this is "marginal capacity cost" (MCC) = (delta trip cost / delta capacity). Where MCC < 50% keep increasing size.

    IMJ there are kinks in the capacity-efficiency curves for various fuselage forms.

    One kink occurs when fuselage diameter must be increased to fit a second aisle - thus NB success and why, IMO, the A350 (or C929) is the last new single-deck WB fuselage we'll see for a couple generations (absent a switch to LH2). This adds Swet per passenger and adds fuse pressurization weight (~proportional to cross section volume) quadratically with fuse diameter. Very inefficient relative to A320/B737.

    The other kink occurs when capacity justifies a switchover to the double deck. You save 30-40% of fuselage Swet and ~20% of fuselage pressure weight (at a rational capacity level you also save bending weight). The virtues of this kink can be largely extinguished with a sufficiently stupid design such as A380, however.

    There'd be another kink when you can switch to a triple-decker but I don't foresee capacity demand rising to the level within my lifetime.

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

    What gets analysts and OEM's (more rarely) into trouble is by thinking as if size categories are things-in-themselves rather than things that fall out from the underlying aero-structural and economic fundamentals. I was at the Paris Air Show in 2019 when Airbus launched the A321XLR and talked to MANY supposed experts who doubted the business case. Long haul was for widebodies, these folks thought. They were stuck in the thinking I'm describing - size categories as inherent things rather than contingencies explained by deeper fundamentals.
     
    User avatar
    Matt6461
    Posts: 3032
    Joined: Wed Oct 16, 2013 9:36 pm

    Re: Analyzing The Airbus A380’s Premature End

    Sat Apr 02, 2022 6:43 am

    flipdewaf wrote:
    the increase in bending moment for any given section of wing then is given as wl^2/2 and as such the bending moments to be countered at any given section scale as (w*1.038)*(90/71.8)^2

    Gives a bending moment to be carried increase of 1.63* the requirement of the 779X. The wing then has to carry these increased bending moments over a distance of 25% greater than the 779X (90/71.8) so the material required to carry these increase loads is 2.04* the 779X. not 50klb but 116klb more.



    OK I spent a minute doing this a little more systematically.

    In the below spreadsheet I:

    (1) estimate the lift-induced bending moment at MZFW per (a) cantilevered span, (b) MZFW, (c) an guesstimated value for the average spanwise moment-arm of lift, (d) the percentage of lift that is cantilevered (i.e. not from the "virtual wing" over the fuselage).

    I then:

    (2) estimate the reverse-direction "relief" bending moment from wing mass and engines via (a) wing mass, (b) estimated center of wing mass spanwise, (c) engine mass - dry weight, nacelles, pylons, (d) spanwise moment arm of engines per ACAP - for VLA I used the A380's engine locations.

    I then:

    (3) Subtract the relief bending moments from the lift-induced moments to get "net bending moment." The ratio is what's really important here.

    I then multiply that net bending moment ratio by span ratio to get the expected total ratio of wing bending material for 779 and my VLA.

    BUT - and this is critical - matching actual and predicted wing weight ratios is an iterative process because as you add wing bending weight you create wing bending relief. Thus the RED CELLS have to be reconciled iteratively.

    Here's how it looks in Excel on one set of presumed parameters:

    Image

    Edited to change to half-spans.

    Anyone can play with document here: https://docs.google.com/spreadsheets/d/ ... ue&sd=true

    Note that I penalized the VLA for having more its mass be bending-dependent and thus having less relieving arm.

    I have also assumed what I think is a fairly high non-wing OEW, still quite a bit above the 777-9.

    I'm not sure this is correct because I expect the VLA's fuse tube to be >10k lbs lighter than 779's, mostly because of fuse bending weight differential and CFRP. As of now I've given VLA a 37k lbs non-wing weight delta.

    VLA will have more weight in floor beams, furnishings, and empennage. It's hard for that to add up to ~47k lbs, however. Engine weight should actually be lower (though that costs a bit in wing weight bending relief) too and MLG ~same (due to shorter gear). I'm still thinking 460-470k lbs is the OEW.
     
    User avatar
    Matt6461
    Posts: 3032
    Joined: Wed Oct 16, 2013 9:36 pm

    Re: Analyzing The Airbus A380’s Premature End

    Sat Apr 02, 2022 7:24 am

    Note that this is still an over-estimate of the VLA wing weight delta because:

    (1) The MTOW/MZFW factor favors the VLA markedly.

    Per Stanford (IIRC) wing weight is ~related to SQRT(MTOW/MZFW)

    Using a 7% SFC edge and 25 L/D (vs. 21 on 777-9), a basic Breguet equation gives the VLA a -5.5% wing weight delta or -8,800lbs.

    That's first cut, however. We'd then be lacking wing bending relief and would have to back and iterate until we reach equilibrium. Net delta should be ~5,000lbs lighter.

    To give a sense of the scale: VLA MTOW (for equal range) would be ~70,000lbs higher absent the L/D and SFC effects. And that's holding OEW constant. It should be easy to see that MTOW reduction at this scale will reduce wing weight and OEW in general.

    (2) I also haven't given VLA any credit for wing bending relief outboard of the inner engines but inboard of the outer engines (where B777 obviously has no engine bending relief).

    Again, 460k OEW seems increasingly likely to me, perhaps with the original 200ft fuselage and capacity ~11% greater than A388.
     
    User avatar
    SQ22
    Moderator
    Posts: 2769
    Joined: Sat Feb 25, 2012 9:29 am

    Re: Analyzing The Airbus A380’s Premature End

    Sat Apr 02, 2022 2:48 pm

    Please remember to provide a link to your source when posting any kind of image or chart.
     
    User avatar
    Matt6461
    Posts: 3032
    Joined: Wed Oct 16, 2013 9:36 pm

    Re: Analyzing The Airbus A380’s Premature End

    Sat Apr 02, 2022 9:22 pm

    SQ22 wrote:
    Please remember to provide a link to your source when posting any kind of image or chart.


    The image/chart I posted is sourced to me and Excel. Should I state that?

    Re some of the parameters in the chart, I reference OEM ACAP's for things like engine spanwise displacement from centerline. The operating assumptions are general aero-engineering principles, such as the lift-induced bending moment is proportional to integrated lift arm * integrated lift (I use average arm/lift as a proxy). What else should be cited?
     
    User avatar
    SQ22
    Moderator
    Posts: 2769
    Joined: Sat Feb 25, 2012 9:29 am

    Re: Analyzing The Airbus A380’s Premature End

    Sun Apr 03, 2022 6:35 am

    Matt6461 wrote:
    The image/chart I posted is sourced to me and Excel. Should I state that?

    Re some of the parameters in the chart, I reference OEM ACAP's for things like engine spanwise displacement from centerline. The operating assumptions are general aero-engineering principles, such as the lift-induced bending moment is proportional to integrated lift arm * integrated lift (I use average arm/lift as a proxy). What else should be cited?


    I was not referring to your own charts.

    Who is online

    Popular Searches On Airliners.net

    Top Photos of Last:   24 Hours  •  48 Hours  •  7 Days  •  30 Days  •  180 Days  •  365 Days  •  All Time

    Military Aircraft Every type from fighters to helicopters from air forces around the globe

    Classic Airliners Props and jets from the good old days

    Flight Decks Views from inside the cockpit

    Aircraft Cabins Passenger cabin shots showing seat arrangements as well as cargo aircraft interior

    Cargo Aircraft Pictures of great freighter aircraft

    Government Aircraft Aircraft flying government officials

    Helicopters Our large helicopter section. Both military and civil versions

    Blimps / Airships Everything from the Goodyear blimp to the Zeppelin

    Night Photos Beautiful shots taken while the sun is below the horizon

    Accidents Accident, incident and crash related photos

    Air to Air Photos taken by airborne photographers of airborne aircraft

    Special Paint Schemes Aircraft painted in beautiful and original liveries

    Airport Overviews Airport overviews from the air or ground

    Tails and Winglets Tail and Winglet closeups with beautiful airline logos