Dreamliner first flight delayed yet again Boeing has announced the first flight of its 787 Dreamliner will not now go ahead on 30 June as planned due to "a need to reinforce an area within the side-of-body section of the aircraft". The Boeing Dreamliner in All Nippon Airways livery. Source: Boeing The company's press release explains: "The need was identified during …
These types of delays are to be expected in the creation of such a novel aircraft. The fuselage is the first all-composite fuselage in an aircraft of this type; while simulations have no doubt been irreplaceable in the design process, the number of unknowns means a revision somewhere was likely. It's far better for Boeing to delay the first flight to make a better airframe: remember the Comet; a great plane that flew for a long time as the Nimrod, but the shape of the windows caused great metal fatigue in the airliners with catastrophic results.
Engineering is not as exact a science as laypeople expect. There are approximations and simplifications made, and real objects will always be different from a simulation. For this reason factors of safety are used, and the purpose of the static test aircraft is to check their assumptions.
The A380 had manufacturing problems too; due to software inconsistency the front and back fuselage sections didn't fit, and Airbus overhauled their IT policy and unified it across their organization. Live and learn!
As Henry Petrosky states, the only way engineering can progress is through failure. Boeing's correction of this problem is a failure that caused no damage, and that is the best time to find one.
I'd rather Airbus and Boeing delayed and got the plane just right instead of having the damn things fall out the sky because the sides fell off after takeoff.
Remember these are machines that will have lifespans of 20-30 years, a delay of two years because of getting these new composite materials and other technologies just so is really nothing.
Don't they have CAD systems to tell them about this sort of thing before they build the planes? This sounds alarmingly like the Microsft approach to software development, throw it together so it looks good on the outside, and patch it when bits break. Remind me to add the 787 to my list of "don't fly until it's been around for a while" planes...
No they shouldn't...the flight test data would've been invalid cos they would, effectively, be flying a different airframe. Which isn't helpful for the FAA/JAA certification. So they would've had to have repeated the tests...which would have been even more demoralising for said Boeing employees, pushing to do a pointless test knowing full well they'd have had to repeat the test anyway.
Sounds like Boeing management had their project-planning caps on for a change. Airtime in a flying test bed is veeeeeeeery expensive, hence why they stuff the thing full of instrumentation and follow the 'do it right first time' approach, and try, as close as possible, to match the production aircraft.
Incidentally, at a presentation given to us as final year engineering students, shortly before the launch of the A380, the Airbus wonk claimed that "we don't build prototypes anymore".
Me: "So who are you going to sell the first aircraft off the line to, the one that's been stuffed with the aforementioned instrumentation?"
AW: "Oh, that's got too many holes and extra wires in it, we keep that, as a flying test bed, for testing future upgrades and the like"
Me: "So the difference between that and prototype is?"
AW: "Er...."
@ SteveX
Yes they do, but structural analysis (FEA, etc) will only tell you so much, especially as far as composites are concerned...hence why they test it 1:1 scale before they let the wetware take it for a spin...
It's worth bearing in mind that CAD stands for Computer AIDED Design - doesn't remove the need for the good ol' Mk1 Greymatter. Nowt more dangerous than a CAD-jockey just unthinkingly throwing stuff together in the virtual environment and expecting it to work IRL...
The one with a copy of "Basic Concepts of Engineering" in the pocket and the flame-proof lining.
CAD systems are for designing the airframe - it won't tell you how it will behave when some rainstorm dumps a few thousand gallons of saltwater all over your nice new weetabix fuselage, or how well it will withstand crosswinds, windshear etc.
Computational Fluid Dynamics and materials modelling is another kettle of fish entirely.
That's why they use the old method of hooking up one of the early airframes to a load of strain gauges and then pulling it every which way they can. It's better to lose one of your 'iron bird' test pieces than a scheduled flight full of fare-paying self-loaded cargo...
That said, to get a week from first flight and then announce another f***-up like this is pushing the limits a bit. Nice one Boing, Airbus sends their thanks!
If memory serves me Boeing went on loud mouthed media wankfests saying how good they were and how crap Airbus was due to the delays in getting the A380 production ready.
I do not seem to hear Airbus going on a loud mouthed media wankfest now that Boeing have done a faceplant with their new product.
And yes, the Boeing fanbois are conspicuous by theior absence.
And being an IT geek, I do not use a product until it has been given a good shakedown. Now that the A380 has been in service for a while it seems to be a stable, reliable platform so I will fly in it. I'll leave the first flight crap to those out there that make up for their small wedding tackle by waving their ego's around instead.
The TeeCee Boeing PR translation service thunders into life and renders:
"Consideration was given to a temporary solution that would allow us to fly as scheduled, but we ultimately concluded that the right thing was to develop, design, test and incorporate a permanent modification to the localized area requiring reinforcement."
as:
"We were going to slap some gaffer tape over it and hope it held, but we couldn't get it to stick so we're having to pop-rivet on a patch and cover it with filler."
...someone might have mentioned the 30 minutes it takes to boot the integrated avionics system.
Seriously. And before anyone thinks that's normal for this kind of thing, it isn't. Could be something to do with the sheer number of systems in there (350+?).
This might make cycling things in-flight 'interesting' even with the redundant failover. Can't think of anyone yet who plans to go on one of these things, especially the people working on it...
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
