Trent 800 woes
H Andrew Chuang
An earlier rumor of a catastrophic failure of a Rolls-Royce Trent 800
engine during an Emirates' B777 take-off was confirmed by this week's
Flight International (9/24-9/30). Both R-R and Emirates have been very
quiet about this incidence. Earlier report indicated the compressor
failure was due to a foreign object damage (possibly a birdstrike).
However, Flight International alleged metal-chip detection from the
front bearing cage. (Does Cathay's grounding of the Trent 700-powered
A330 ring a bell?)
In the same issue of the magazine, R-R ran an ad boasting Trent 800's
consumption of less than one quart of engine oil (for lubrication) on
the B777 record-setting round-the-world flight. I was told by a very
reliable source that the engine from one of R-R's American competitors
"consumes" engine oil at least ten times more than the other two B777
powerplants. However, the Trent 800 seems to have experienced many
more woes than either the PW4000 or GE90. (Were I a Delta executive, I
certainly would seriously reconsider the B777 engine selection before
excercising Delta's B777 options.)
Malcolm Weir
On 28 Sep 97 00:53:39 , Chu...@cris.com (H Andrew Chuang) caused to appear
as if it was written:
>An earlier rumor of a catastrophic failure of a Rolls-Royce Trent 800
>engine during an Emirates' B777 take-off was confirmed by this week's
>Flight International (9/24-9/30).
[ Snip ]
> However, the Trent 800 seems to have experienced many
>more woes than either the PW4000 or GE90. (Were I a Delta executive, I
>certainly would seriously reconsider the B777 engine selection before
>excercising Delta's B777 options.)
Flight Internation 10 - 16 September states:
"Pratt & Whitney says that an upgrade effort to counter reliability problems
on more than 1,600 PW4000 engines is showing results, with a 'dramatic
reduction' to in-flight shutdown rates.
The upgrade effort, known as the Number 1 reliability programme, involves
around 100 service bulletins. More than 70 PW4000 customer airlines have
incorporated around 80% of the improvements into their engines, says P&W.
It adds that the 12-month rolling average in-flight shutdown rate for the
fleet has dropped from 0.014% in January 1997 to 0.008% in August as a
results".
Specific to the 777, "P&W confirms that it is replacing the low-pressure
turbine (LPT) stage-five vanes from PW4077 engines in service on Boeing 777.
'We saw this in validation tests for ETOPS, and we saw some areas in the LPT
vane that needed improvement."
So perhaps Delta's selection of RR is as good as P&W? Particularly since
the Trent's problems seem to be self-identifying (i.e. oil temperature
climbs through the roof), while some of the PW4000 problems involved
modifications to the leading edges of the fans due to cracking after foreign
object impacts...
Malc.
H Andrew Chuang
In article <airliners...@ohare.Chicago.COM>,
Malcolm Weir <ma...@mci2000.com> wrote:
>
>Flight Internation 10 - 16 September states:
>
>"Pratt & Whitney says that an upgrade effort to counter reliability problems
>on more than 1,600 PW4000 engines is showing results, with a 'dramatic
>reduction' to in-flight shutdown rates.
> The upgrade effort, known as the Number 1 reliability programme, involves
>around 100 service bulletins. More than 70 PW4000 customer airlines have
>incorporated around 80% of the improvements into their engines, says P&W.
>It adds that the 12-month rolling average in-flight shutdown rate for the
>fleet has dropped from 0.014% in January 1997 to 0.008% in August as a
>results".
The 94-inch PW4000 (which powers the A300/310/B767/B747/MD-11) has had a
poor reliability record relative to the RB.211 and CF6 engines,
especially the latter. Thus, the Reliability 1 program was established
to address the issue. However, both the 100-inch PW4000 (for the A330)
and the 112-inch PW4000 (for the B777) have very respectable record,
definitely much better than the Trent 700 and 800.
>Specific to the 777, "P&W confirms that it is replacing the low-pressure
>turbine (LPT) stage-five vanes from PW4077 engines in service on Boeing 777.
>'We saw this in validation tests for ETOPS, and we saw some areas in the LPT
>vane that needed improvement."
All engine companies constantly provide performance imporvement packages
just like the airframers. Durability is important but reliability is
perhaps more important. Maintenance associated with durability issues
is usually predicatable and schedulable. However, the same cannot be
said about incidences associated with reliability issues.
>So perhaps Delta's selection of RR is as good as P&W?
It's only my opinion. You're certainly entitled to your opinion.
>Particularly since
>the Trent's problems seem to be self-identifying (i.e. oil temperature
>climbs through the roof), while some of the PW4000 problems involved
>modifications to the leading edges of the fans due to cracking after foreign
>object impacts...
The Trent 700/800 is supposedly a derivative engine, hence, it should
be a more mature engine in the field and a less risky choice for airline
customers. However, the Trent 700/800 seems to have encountered even
more field problems than the GE90, a brand new engine which was expected
to have more teething problems. Most airlines love the economics of
flying twins. However, Cathay Pacific seems to hate twins. Perhaps, the
fact that CX only has Trent-powered twins (the A330 and the B777) is not
a coincidnece.
Exiled Expat
Malcolm Weir wrote:
> >An earlier rumor of a catastrophic failure of a Rolls-Royce Trent 800
> >engine during an Emirates' B777 take-off was confirmed by this week's
> >Flight International (9/24-9/30).
> So perhaps Delta's selection of RR is as good as P&W? Particularly since
> the Trent's problems seem to be self-identifying (i.e. oil temperature
> climbs through the roof), while some of the PW4000 problems involved
> modifications to the leading edges of the fans due to cracking after foreign object impacts...
I work on the Trent 800 at Emirates and despite a few teething problems
as a whole I find the Trent Engine to be the best engine I have ever
attended to. It's built like a brick Shit House and as access to
components is far easier than on the GE90 and PW4000 it offers very quick
turn time. At least it hasn't needed 100 service buletins to bring it up
to speed ( but that is typical for PW, Just look at how many mods were
made to the JT9D to bring it up to it's reliability ). The recent failure
of our RR Trent is so far believed to have been induced by foreign object
damage wich is difficult to fault the engine for.
As far as oil temps going through the roof, I know of no such failure in
our fleet of 14 operating engines.
Incidently, a pilot told me that UAL wasn't happy with the way the PW4000
is guzzling far more fuel than it should be for their present service age.
Malcolm Weir
On 03 Oct 97 01:18:30 , Exiled Expat <lut...@emirates.net.ae> caused to
appear as if it was written:
>Malcolm Weir wrote:
>> So perhaps Delta's selection of RR is as good as P&W? Particularly since
>> the Trent's problems seem to be self-identifying (i.e. oil temperature
>> climbs through the roof), while some of the PW4000 problems involved
>> modifications to the leading edges of the fans due to cracking after
>> foreign object impacts...
>As far as oil temps going through the roof, I know of no such failure in
>our fleet of 14 operating engines.
Sorry, I should have been more clear: That was a reference to the A330
Dragonair failure of a Trent 700, blamed on faulty lubrication systems which
resulted in insufficient oil getting where it should be, hence the
temperature rise.
The point was, while all in-flight shutdowns are serious, possibly the most
serious type of event involves fan blade failures... It's bad enough to
lose an engine without bits of the engine whizzing around the rest of the
airframe (c.f. Sioux City...)
Malc.
Art Intemann
> Incidently, a pilot told me that UAL wasn't happy with the way the PW4000
> is guzzling far more fuel than it should be for their present service age.
Funny---I work for UAL and haven't heard anything like that.....
Regards,
Art
H Andrew Chuang
[snip]
>The point was, while all in-flight shutdowns are serious, possibly the most
>serious type of event involves fan blade failures... It's bad enough to
>lose an engine without bits of the engine whizzing around the rest of the
>airframe (c.f. Sioux City...)
Malcolm, I think you know that the Sioux City accident was not due to
a fan-blade failure. It was due to a rotor-disk failure. All engines
do have to go through a fan-blade-failure test just like they have to
go through birdstrike tests (but not a rotor-disk-failure test). In
most cases, a fan-blade failure should not cause a catastrophic failure
of the engine.
H Andrew Chuang
In article <airliners...@ohare.Chicago.COM>,
Exiled Expat <lut...@emirates.net.ae> wrote:
[snip]
>( but that is typical for PW, Just look at how many mods were
>made to the JT9D to bring it up to it's reliability ).
Your example reinforces what I'm trying to say: reliability is very
important. The only application the JT9D did well was the early B747
market, in which the JT9D initially had the sole source. After Boeing
started to offer the CF6 and the RB.211, P&W share of the B747 market
drastically fell. P&W practically lost every major European B747
customer, such as, Lufthansa, KLM, Alitalia, BA, Air France, etc.
>The recent failure
>of our RR Trent is so far believed to have been induced by foreign object
>damage wich is difficult to fault the engine for.
The Trent 800, as well as its competitors, went through rigorous
birdstrike tests, including one with a 4-kg bird. The purpose of the
test is to make sure engines can maintain certain level of thrust even
in an event of birdstrike. (Furthermore, R-R has long claimed one of
the many advantages of wide-chord fans is foreign objects seldom get
into the core flow.) Thus, even if the damage was due to a birdstrike,
the engine should not have had a catastrophic failure. If there is any
truth to Flight's speculation (that metal chips from the bearing were
found), then I think R-R has a much more serious problem at hand. R-R
claimed to have solved the Trent 700 bearing problem by putting the
Trent 800 bearing in the Trent 700. Perhaps, R-R has not solved the
problem, yet.
Malcolm Weir
On 07 Oct 97 14:10:18 , Chu...@cris.com (H Andrew Chuang) caused to appear
as if it was written:
>In article <airliners...@ohare.Chicago.COM>,
>Malcolm Weir <ma...@mci2000.com> wrote:
Yes, I am aware that UA232's accident was caused by a rotor-disk failure.
I'm sorry if I didn't make it clear that the point was that having bits of
engine part company with the rest of it is more significantly more serious
than lubrication problems. An extreme example of that point was Sioux
City.... compared with the Eastern L1011 that lost all lubrication on all
engines...
It is obvious that uncontained failures are more serious than contained
failures, and that *neither* should happen. But both do.
Didn't P&W have "concerns" about fan containment with those engines, anyway?
Malc.
Iain Stuart
In article <airliners...@ohare.Chicago.COM>, H Andrew Chuang
<Chu...@cris.com> writes
>>The point was, while all in-flight shutdowns are serious, possibly the most
>>serious type of event involves fan blade failures...
> All engines
>do have to go through a fan-blade-failure test just like they have to
>go through birdstrike tests (but not a rotor-disk-failure test). In
>most cases, a fan-blade failure should not cause a catastrophic failure
>of the engine.
Let's get technical. FAN BLADE FAILURE
All engine marks need to show that a single blade release will be
contained. This needs to be shown for any stage of compressor / turbine,
and for failure at the worst possible position on the blade (ie not just
the tip being released). Damage to other blades must also not cause
hazardous effects.
This can be demonstrated by full running engine test, and suitable rig
test, or by analysis. The cost of the engine test is massive, and thus
rarely done. The analysis method is cheaper, but usually requires
reading across data from engine tests. (NB I believe that the GE90
engine test involved blade failure at the "most likely" failure position
(not the worst), and the FAA agreed !)
However, multiple blade release or disc burst are not required to be
contained. (Think about the energies involved...it's not practical.) The
risk of such failures must be shown to be acceptable low, and this is
why discs have declared service lives and inspection schedules. This
does rely on the professionalism of the engine operator, and several
failures I know of cannot be blamed on the engine design.
Now....BIRD INGESTION etc.
Similar rules cover bird ingestion, hail ingestion etc. There are bird
ingestion rules for small birds, medium birds and large birds, and the
numbers to be swallowed vary with engine size. Again, compliance can be
demonstrated by engine test and/or analysis. The current requirements
are that bird ingestion causes no hazardous effects, and that the engine
continues producing at least 75% thrust for 15 minutes without throttle
tweak.
And, yes... the widechord fan blades do CF out the debris / nuggets away
from the core, and the core suffers less damage. Ask the airlines,
they'll confirm this.
-----------
Iain Stuart
They're my opinions !
D Eunson
H Andrew Chuang <Chu...@cris.com> wrote in article <airliners...@ohare.Chicago.COM>...
> Malcolm, I think you know that the Sioux City accident was not due to
> do have to go through a fan-blade-failure test just like they have to
> go through birdstrike tests (but not a rotor-disk-failure test). In
> most cases, a fan-blade failure should not cause a catastrophic failure
> of the engine.
You seem to use the word catastrophic somewhat loosely. You are quite
right that tests are carried out to demonstrate that a fan blade can be
released and contained without hazarding the aircraft by releasing high
energy debris. As you say, this is not catastrophic. Tests are also
carried out on both rigs and engines to demonstrate the engines ability to
withstand birdstrike of a number of 1.5 pound birds and (rig test, fan
only) one 8 pound bird (well a block of gelatin). after the engine test,
the engine must be able to continue running for a period of time with no
intervention from the driver (which of course the Trent 800 can do). This
then is not catastrophic. While it is true that the Wide chord Fan will
centrifuge a high proportion of debris down the bypass, anything impacting
near the spinner does stand a chance of being ingested by the core (bear in
mind that one of the 1.5 pound birds is fired at the spinner). If a bird /
foreign object of sufficient size were ingested by the core, then damage
could result to the core compressor blades and it would not be
unreasonable to assume that there is a risk of failing a damaged compressor
blade at some time in the future if the engine continues to run on in
service. As with the above situations, there is nothing catastrophic, nor
any risk of release of high energy debris associated with a core compressor
blade failure (the casings being designed for containment). Many engines
have run on with sections of compressor blade missing, the damage only
being found on borescope inspection.
Essentially, the T800 failure was definitely not catastrophic!
By the way, who says Cathay hate twins?
JWizardC
>And, yes... the widechord fan blades do CF out the debris / nuggets away
>from the core, and the core suffers less damage. Ask the airlines,
>they'll confirm this.
I think you'll find that the shape and size of the 'spinner' of the engine
also has a lot to do with it's ability/tendancy to keep rain, hail, etc.
out of the engine core.
TheFNG
Malcolm Weir
On 16 Oct 97 11:59:51 , Chu...@cris.com (H Andrew Chuang) caused to appear
as if it was written:
>> Essentially, the T800 failure was definitely not catastrophic!
>
>I don't know for a fact that it was catastrophic. I was merely repeating
>what I had heard from two different sources that it was a catastrophic
>failure. For the second time, Flight International reported the
>incidence as catastrophic and cited the failure was due to a high
>compressor blade fatigue. If you have evidence that the FI report was
>inaccurate, perhaps you should say so to FI.
Here's the relevant stuff from Flight International, 24-30 Sep (Pg. 7)
EMIRATE 777 Trent engine fails during take-off
A Trent 800 engine from an Emirates Airlines Boeing 777, which suffered a
catastrophic engine failure during take-off is being examined by
Rolls-Royce. The take-off was continued and the crew shut down the engine
and returned to Dubai, where the aircraft was landed safely.
Emirates declines to comment, beyond saying that there was an incident
involving the flight to Male in the Maldives and that the aircraft returned
to Dubai.
Unconfirmed reports from pilots in Dubai say that a failure and subsequent
engine fire occurred after V1 (take-off decision speed), but before rotation
speed. The shutdown on 16 September is reported to have resulted in a
failure in the high-pressure compressor.
It is possible that the incident was prompted by a birdstrike, although one
source says that metal debris in the mass chip-detector is believed to have
come from the front-bearing cage. R-R says only that "...there was an
in-flight shutdown on an Emirates aircraft. The engine is being stripped
down and we are investigating."
Preliminary indications were that the aircraft's thrust-asymmetry
compensation (TAC) function also failed to operate. The TAX automatically
applies up to 10degress rudder when one engine produces greater than 10%
more power than the other, but only under set conditions.
When the engine failed, the Emirates' 777 engine-indication and
crew-alerting system displayed the advice "THRUST ASYM COMP". The 777's
operations manual explains this as: "Thrust asymmetry compensation is
inoperative". Boeing says: "It looks as if the TAC operated as it should
have. Certain failures will not activate the TAC function, to ensure that
there is no [rudder] input in the wrong direction.".
Malc.