On Mon, 30 Aug 2010 03:48:11 -0400, JF Mezei
Post by JF MezeiPost by matt weberThere are in fact huge differences. At Sea level most of the thrust is
generated by the fan, at FL400, you have essentially a turbojet rather
than a tubo fan engine.
OK. Are computer modeling/simulations now sophisticated enough to be
able to project how the engine will behave at altitude based on ground
tests, or is this still constidered an "art" and only actual testing at
altitude will reveal true performance ?
The general answer is yes, but would I like to bet my life on it? NO!
As the PW4000 turbine case demonstrated, the analysis is only as good
as the things you remember to check for. Mother nature tends to be
pretty unforgiving about things you don't check for.
Post by JF MezeiAnd prior to RR delivering engines to the 787 programme, wouldn't they
have also used some other aircraft to test the engine's performance at
altitude ?
I don't doubt the engines were tested on an airframe, but I don't know
if RR owns one big enough, and frankly RR doesn't have the resources
that GE has.
Post by JF MezeiPost by matt weberThe RR engines currently in use for the test program are simply not
good enough (SFC) for commercial operations. Boeing is now saying the
issue really is engine availability from RR.
Considering that aircraft engines are extremely design heavy, with a lot
of computational fluid dynamics etc involved, what sort of tweaking can
they do to increase engine efficiency by 4% ?
GE worked wonders when they had a major re-think on the design of the
interstage seals. You can also tweak the FADEC. In fact just using one
is often good for gain several per cent in SFC. You can also tweak the
scheduling of the stator vanes, the EGT's etc. In short there is no
shortage of things you can try. The question is how much do you impact
the reliability and/or durability of the engine in doing so.
Post by JF MezeiIs this a case of the prototypes being different than what the plans
called for because of manufacturing limitations ?
Early examples tend to be hand build and individually machined. The
early examples are almost never within the intended design
specifications. It isn't talked about very much, but NW took many of
the early 747-400's. IIRC, they are all at least 10,000 pounds over
spec OEW.
Post by JF MezeiOr are the engines exactly what had been designed, and they find out
their fancy computer simulations had overestimated engine efficiency by 5% ?
You design to what you know you can build. Often as a result of test
and experience, you find better ways to do things, and discover you
can in fact build or run closer to the 'edge' because the margins are
better than expected.
Post by JF MezeiIn the later case, considering they spent years fine tuning the shape of
everything in the engines, can they really squeeze out an extra 4% of an
engines with just a few more weeks of thinkering with the design ?
Probably, but there are no guarantees in this business. The RB211-524G
never made guarantees for QF, and RR actually decided it was less
expensive to pay the penalties than to fix the problem.
The simulation is only as good as the data and the assumptions that
go into it. The result can often be unpleasant surprises when you miss
something.
Post by JF MezeiFor aircraft engines, where every part has to be precisely manufactured,
is there such a thing as a prototype engine ? Would they bother
assembling an engine with parts that do not perfectly match the design ?
Aren't GEnx and Trent 1000 state of the art today ? If so, is it really
realistic to expect the state of the art to magically increase by 4% ?
They are state of the art, but on the other hand there is no
experience with no bleed air engines. The RR engine is a lot different
than anything RR has ever designed and built previously. There is a
lot that is new in the GEnX, but a lot of GEnX is based upon the GE90
experience and GE90 technology.
Post by JF MezeiPost by matt weberIt is fairly rare for early examples of engines to meet SFC
guarantees, just as it is fairly rare for early examples of airframes
to make spec weight.
I can understand airframes getting overweight. Add an extra rib here and
there to strenghten weak areas. But the overall aerodynamic shape does
not change. And they find that some areas are too strong, so later
production can use less material in thsoe areas to reduce weight. They
change the construction based on experience.
But for engines, since tolerances are so small, how can they build an
engine that does not precisely match the engine design ?
The design is only half the battle. Engines have lots of variables in
how they run. The days of direct connection between the power levers
in the cockpit and hydromechanical engine controls are long past.
You can have the best engine design in the world, but in the end, the
engine is controlled by Software these days.
Need I say more?
Post by JF MezeiOr again, is this a case of their fancy computer simulations not
providing realistic numbers for that design, and it is only once built
that they realise that some parameters are off and they have to fine
tune the simulation parameters which results in fine tuning of the
design ?
Post by matt weberProbably later, but later means on the order of a few months. I
suspect that the handover of the first GE powered 787 will be impacted
a lot less than the RR handover.
But if the GEnx's performance is also sub par, wouldn't this equally
delay deliveries using whatever excuse Boeing will find to explain the
delay ?
Post by matt weberGEnX is essentially a no-bleed air engine (I say essentially, because
bleed air is in fact still used for some anti-ice functions, but not
for cabin air).
Is this a case of Boeing going back on its "no bleed air" PR promises
after it realised that for some purposes bleed air was better ?
Not really. There are some places where electric de-ice is simply
impractical. In normal operation, the engines are no-bleed. Even when
Bleed air is used, it is a tiny fraction of what was required to
provide cabin pressurization and ventillation.
Post by JF MezeiI seem to recall that electric de-icing had been cited as example of
"all electric" design being better because you only need de-icing for a
relatively short period of flight and don't want to waste bleed air for
the cruise protion where you don't need it.
De-ice on everything except the engine itself is electric on the 787.
Post by JF MezeiIt is interesting that we have not heard much about how the "all
electric" concept has panned out for the 787. I suspect they would now
have a pretty good feel for how the all electric systems are working ?
No news is probably good news. I suspect the electric parts work fine,
it is the software the controls them that has been a major problem.
Post by JF MezeiIs this the shape of things to come, or was this 787 experiement
interesting but not really worth it for subsequent aircraft ?
The demand is for fuel efficient aircraft.
Fuel consumption is (Drag/Lift) * Weight * Specific Fuel Consumption
(SFC).
Within the design constraints for commercial aircraft certification,
there isn't much room left for L/D improvement (If the unconditional
stability requirements are changed, all bets are off).
SFC improvement is occuring, but it has taken almost 30 years and tens
of billions of dollars to improve SFC 15%.
Consider this for a moment engine SFC at cruise
RB211-524D4 .63 (vintage ~1980)
RB211-524G/H-T .57 (Vintage ~2000)
Trent 970 .52 (Vintage ~2008)
The path of least resistance is weight.
Realistically the only way to achieve signifcant weight savings is
increasing the use of composite materials. So at least in that regard,
the 787 is indeed the shape of things to come.
Post by JF MezeiPost by matt weberThe assumption is your vendors are going to delivery the product
somewhere near on time, and on specification. Bad things happen when
either of both don't happen.
Considering that Boeing's last truly new aircraft was the 777 in the
early 1990s (first delivery 1995), and considering that the 787 was so
radically new, did anyone seriously expect it to be on time and with
target weight ?
The problem is the same people tend not to last the 20 years between
programs, and many of the lessons learned in prior programs are
forgotten in the intervening years.
Ask A340-200/300 operators about the 'surprises' they got with the
A340-500/600. Many of the issues with the early A340-200/300
reappeared with the -500/600 because EADS changed some key vendors in
between, and lost the experience associated with those issues on the
earlier aircraft.
Post by JF MezeiSeems to me that the technical problems of the 787 have been expected
and are accepted.
It was the project management nightmare and the PR pressure to assemble
787s before they were ready which caused so much of the delay. Instead
of blaming a shortage of rivets, Boeing would have been better off
stating that it has found a design problem and is ordering new rivets,
and this will cause some delay. That would have been better understood
and had more credibility.
Evidently, the south carolina facilities appeared to be truly under par
because Boeing had to buy them and put its own people to run them.
Same thing with the A380. Initially, they just said "industrial problem"
(or whatever excuse). It was only once they fessed up to the fact that
the germans had incompatible software which generated wire bundles of
incorrect lengths that the real problem was understood.
Post by matt weberOne manufacturer at this point is well known for their
overweight aircraft....
At the end of the day, would the "overweight 787 with underperforming
engines" still be significantly better then the 330s ? Or would the 330
remain competitive against early production 787s ?
The A330 is a non-competitor even with an overweight, underperforming
787. Airbus tried to sell a warmed-over A330 design as the A350 MKI
and then again as A350MKII. I think it was Udvar-Hazy who flat out
told them to start over if they planned to sell them. It took Airbus a
very long time to realize that the 787 had fundementally changed the
landscape.
Post by JF MezeiPost by matt weberThere are enough issues with the test articles that Boeing concluded a
long time ago that they could not be sold to customers.
Yes, but a lot has to do with the structure, and how many times they
were disassembled/re-assembled, rivets changed etc etc. This is not
meant to happen to an aircraft structure.
The big question is what will happen to the batch of aircraft of the
same vintage as the first 6 test articles that has been in various
stages of assembly for a couple of years now. Evidently, this includes
the first commercial aircraft to ANA.
Post by matt weberYou'd be amazed at what sales will do to win orders. Take a good look
at what EADS did to win the initial A380 order from SQ (and my guess
is the T&C's added about 4 t to the weight of the A380 aircraft)...
But "sales" would still have to talk to "legal" when making performance
commitments, right ?
Legal is subject to intense pressure from the top.
Post by JF MezeiIn the case of the 787, did Boeing engineers really believe that they
would reach those performance numbers, or did sales decide that Boeing
needed some sensationalistic PR with a big improvement in efficiency to
drum up support ?
In most companies, the path to the top is via sales and marketing, not
engineering.
Post by JF MezeiRemember that at first, the 7E7 was to be a "fantastic" aircraft with
totally new everything, which would revolutionlise air travel and
dramatically reduce costs etc etc etc. And Airbus' A350 V1.0 was
ridiculed for being too little too late in part because of all of the
Boeing PR that had made such great promises.
In the end, Airbus' 350 may end up being more profitable if Airbus'
promises were more realistic and a more conventional design will allow
earlier entry in service (aka: less delay).
Don't bet on it.
Post by JF MezeiPost by matt weberThere is no clear answer at this time. Boeing originally published OEW
for the 787 products. That information was redacted about 2 years ago.
So at this point it isn't clear what OEW is supposed to be, or how far
overweight the aircraft is.
2 years ago would cover the first 12 or so hauls that were delivered to
Boeing until Boeing realise how much fo a fuck up iot was in and told
its partners to stop production.
The big question is how long Boeing waited to restart production, and
whether improvements were made before restarting production. For
instance, when you look at the wing/body joints. They had to retrofit
already assembled aircraft by sending people inside the wing. For parts
already made but not yet assembled, they had to retrofit those parts
more easily.
But at what point will the parts be manufacturerd/delivered to boeing
with the changes already embedded ?
Post by matt weberYou have to build some aircraft and fly them to verify the
calculations, and determine where you can remove weight.
Wouldn't Boeing had that information by now ?
It is in fact a continuous process, and very much like peeling an
onion. Each layer you tweak reveals something you can do in the next
layer.
Post by JF MezeiFor an aircraft like the A380 that are built in low numbers, it is easy
to incorporate changes. But Boeing is so late with the 787 with so many
unfilled orders that it would have to weight the cost/benefit of
delaying production ramp up to incorporate the changes, or allowing
production ramp up and then spending all the time and money to retrofit
aircraft already assembled.
You obviously haven't been reading Airbus literature. Airbus expects
to sell thousands of A380's in case you haven't read their
forecasts...
--
misc.travel.air-industry is a moderated newsgroup. Please mail messages to
***@airinfo.aero, and see http://mtai.airinfo.aero for the FAQ and policies.