Go Back  PPRuNe Forums > Flight Deck Forums > Rumours & News
Reload this Page >

Qantas A380 uncontained #2 engine failure

Rumours & News Reporting Points that may affect our jobs or lives as professional pilots. Also, items that may be of interest to professional pilots.

Qantas A380 uncontained #2 engine failure

Old 15th Dec 2010, 14:44
  #1881 (permalink)  
 
Join Date: Mar 2009
Location: Inside
Posts: 285
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by SKS777FLYER
...the interview didn't involve rumours or hypothetical engineering discussions; thus offering the reader nothing to learn.
I have learned a great many things following this thread. In particular I have learned who to avoid at parties, the true meaning of 'Ad nauseum', the many meanings of 'obtuse' and that speaking in incomplete sentences does not make you sound clever, only pretentious.

May the learning long continue.
One Outsider is offline  
Old 15th Dec 2010, 18:51
  #1882 (permalink)  
 
Join Date: Aug 2008
Location: London, UK
Age: 68
Posts: 37
Likes: 0
Received 0 Likes on 0 Posts
777Flyer: Res ipsa loquiter
TiiberiusKirk is offline  
Old 15th Dec 2010, 18:54
  #1883 (permalink)  
 
Join Date: Aug 2009
Location: Texas
Age: 64
Posts: 7,179
Received 377 Likes on 231 Posts
If I may pile onto VFD's skepticism ...
Qantas pays for power by the hour, as far as cycles are concerned then the loss of revenue would be for the time for a 4 engine change, the turbine cost would have been on RR.
75 cycles, that is looking at turning the clock back to recip reliabilaty days.
I find it hard to believe that Qantas or anyone else would have bought into that.
Something in the last two pages doesn't make sense to me. I suspect that the info is incomplete.

75 cycles (launch to land) to mandatory replacement (take off wing) four engines from an A 380. If I may spell out why it puzzle me. Please explain where in my understanding there are gaps, holes, or poor assumptions.

A 380 is a long haul aircraft. I have a very modest understanding of long haul optimization decisions, which inform a company which airframe to buy and fly for the expected/forecast route structure.

I'll estimate the bounds of profitable legs/sectors to be between 5-12 hours long. (If this is way off, please advise). That would put a given A380 in need of new engines (four) in somewhere between 375 hours and 900 hours. At something like $12 million per copy, for ease of illustration, it looks like they cost more than that, that's $48,000,000 expense on engines alone in 75 sectors ...

Makes no fiscal sense to me.

For X amount of servicing, versus replacement, it would make more sense. That said, you still eat X' amount of time (which is money and days of no revenue) getting four new engines on and checked out before you release the airframe to the line for operations - the next 75 anyway. Is this a cycle that most airline companies would accept, in terms of standard/mean time on wing for an engine? I suspect not, but would be pleased to be educated if I am wrong in my guess.

RR has been building reilable and durable engines for airliners for some decades.
Qantas wishes to operate at a profit.
A few back of the napkin cost curves leave me skeptical of the 75 cycles at face value.

Some information is missing, surely, or some context. (Is it 75 cycles for a major inspection cycle? )

Reference point:

Twenty five years ago, I operated GE T58-8F turboshaft engines that had a "high time" replacement interval of 600 hours (+/- 10%) on sorties that averaged from 2.5-6.0 hours. You'd get from 100 - 240 sorties if two matched engines worked as advertised out of the can. Many did. Most came "off wing" within the first 60-100 hours if they weren't going to make the interval.

The state of the art of engine design and manufacture has advanced a couple of generations since those T-58's were put together. The engines I used later were "modular" in design, so that overhauls were both simpler and some sections not needing them in as many hours as others ... I don't understand the Trent well enough to grasp if it has that same sort of modularity designed into it, but think it does.

I have difficulty believing that a purpose built turbofan engine intended for a particular operating environment would have such a short replacement ( or overhaul) interval. Remove and replace, on a power by hour basis, leaves an awful lot of engines in rework/refit for the engine provider (RR) when you start looking at multiple airlines and dozens or hundreds of aircraft, each with four engines on wing ... my brain comes up with the term "churn" when I consider this prospect.

What information is missing from that 75 cycles data point?

Last edited by Lonewolf_50; 15th Dec 2010 at 19:53.
Lonewolf_50 is offline  
Old 15th Dec 2010, 19:34
  #1884 (permalink)  
 
Join Date: Sep 2010
Location: Double Oak, Texas
Age: 71
Posts: 180
Likes: 0
Received 0 Likes on 0 Posts
OO
I have learned a great many things following this thread.
I have also learned much reading thru the thread, from multi-national contract law to recip longevity and theory of turbine engine oil fire genesis and turbine self destruct.


LoneWolf. I don't suspect that when pulling the engines off the wing at 75 cycles that they are just going to trash the engines. No doubt expensive procedure to follow, but most likely far less than replacement value of a new engine.
SKS777FLYER is offline  
Old 15th Dec 2010, 19:57
  #1885 (permalink)  
 
Join Date: Aug 2009
Location: Texas
Age: 64
Posts: 7,179
Received 377 Likes on 231 Posts
Thta was my guess, but that requires (see my edited post, sorry, I had not fleshed out one of the later thoughts) a significant churn of a very expensive engine, times four.

I admit to being in the dark. I'd have expected a new/recent generation engine to have a thousands (1,2,3, more?) of hours on wing, not hundreds of hours on wing, requirement to meet a long haul carrier's requirements when engines are proposed as the powerplant for a long haul airframe.

I should just have said that, perhaps.

If that idea is half cocked, or just plain wrong, please advise.

Since I have far more "hands on" with turboshaft engines than with turbofan engines, I may be missing a fundamental point.
Lonewolf_50 is offline  
Old 15th Dec 2010, 20:39
  #1886 (permalink)  
 
Join Date: Dec 2007
Location: Germany
Posts: 3
Likes: 0
Received 0 Likes on 0 Posts
LoneWolf:
What information is missing from that 75 cycles data point?
The claim quoted in the article is that parts "need replacement after operating 75 take-offs at the 72,000lb maximum thrust level", so I understand that the engine can sustain more take-offs at lower thrust settings.

Lets make the simplified assumption that maximum thrust is always needed when flying from LAX back to Australia, but for all other sectors the thrust can be reduced so far that they don't count for the engine life (reality will be in between). Then it might be possible to fly a single aircraft to LAX and back up to 75 times (150 flights) before its engines have to be changed. If you use the aircraft on multiple routes - most of which don't need maximum thrust - you will get even more flights between replacements.

Probably still not good enough, but less dramatic than replacing engines every 75 flights.
A664 is offline  
Old 15th Dec 2010, 21:18
  #1887 (permalink)  
 
Join Date: Jan 2009
Location: On a survey line somewhere...
Age: 42
Posts: 54
Likes: 0
Received 0 Likes on 0 Posts
On the 75 cycles bit - I'm certainly not an expert, but have read this thread from the beginning;

My understanding is that the 75 cycle limitation is as a result of spline wear problems documented in a few ADs, and is certainly not the original spec of the engine.
sgs233a is offline  
Old 15th Dec 2010, 21:19
  #1888 (permalink)  
 
Join Date: Dec 2010
Location: Middle America
Age: 84
Posts: 1,167
Likes: 0
Received 0 Likes on 0 Posts
barit1
The electronic engine controller (FADEC) and data recorders are an integral part of the engine and are mounted usually on the outside the forward part of the engine so as not to suffer from high temperature exposure. This enables shorter cable runs, weight savings and less complicated removal of the engine from the wing for off-wing servicing among other advantages.
Turbine D is offline  
Old 15th Dec 2010, 21:55
  #1889 (permalink)  
 
Join Date: Dec 2010
Location: Middle America
Age: 84
Posts: 1,167
Likes: 0
Received 0 Likes on 0 Posts
A664 and colpaz
I read and re-read this article and this part is still baffling to me. First it states the Trent 900's HP/IP support structure would be subject to "high severity" if the engines were to be operated at 540 psi at P30 on more than 75 flight cycles.
I assume 540 psi at P30 is a 72,000 pound non-derated take-off.
So then they say on November 12, 2010, Qantas, you must apply a minimum 8% derated take-off thrust to your 72,000 lb. engines.
So max derated thrust is now 66240 lbs.

But it goes on to say (gives some Tflex temperatures for both 72K and 70K engine ratings) or a minimum take-off fixed derate of 8% (72K) or 4% (70K) should be used.
Since Qantas doesn't have any 70K thrust engines, what is this saying? Does this apply to other operators who's fleets are comprised of 70K Trent 900 engines? The 70K derate of 4% = 67.2K

It does say "the thrust derate was made in order to reduce the engine pressure ratio in the P30 area of the engine and therefore increase the life of the oil transfer tubes within the HP/IP support structure"
It does not mention spline wear.

Turbine D
Turbine D is offline  
Old 15th Dec 2010, 22:26
  #1890 (permalink)  
bearfoil
Guest
 
Posts: n/a
colpaz Turbine D

There is no track record for the "Stub Pipe" issue, and since RR "Sprung" the Stub Pipe issue on everyone, in hopes of diminishing the severity of the Powerplant issue, The 75 cycles is related to the Splines, the AD's, and the 972. 75 cycles is about ninety days for Qantas, as I reckon it, and the AD has not been satisfied. The Oiling issue is new, at least as far as RR would have it.

The AD is demanding of STRIP, not a simple Mx, and as I see it, not an on wing process. I stand to be corrected. The Spline Wear has to do with the HP/IP Thrust bearing end of the Shaft (forward), there are no Splines at the Aft Roller Bearings, the "Site" of the Oiling issue related to Fire, related to uncontained Burst, etc.

The fog is deliberate. The 972 should have been given a miss, relative to Qantas and their long haul aspirations. Eight thousand pounds of Thrust left in the Hangar in favor of 970's must seem like to drool over about now, for Qantas. What can 970's cost in MTOW and Pax/Freight? Some, I suppose, but again, the 972 is a mission killer as it stands now.

The IP Shaft/Spline Shaft couple had to do with the original report of lost IPT, QF32. "Loss of Rigid Coupling". It is this area that is causing the grief, the oil issue is a handy scapegoat, IMO. Worn Thrust Pocket "Abutments" that nest the Splines in their Troughs are the issue (according to the AD's). Though a "Module", if there are design issues here, the implications for permanent fix are problematic, and involve the 970 on up to the 980. RR can't possibly be entertaining a change out of The entire IP Module every ninety days, and the 970 is an identical engine; who will believe that a mere 2000 pounds of thrust separates identical twins from the same mechanical pitfalls??

Curiouser, Curiouser.

Alternate, and better prognosis. The AD is related to Oiling insufficiencies in the 972 only (By "happenstance"). If so, the entire Oiling system could be upgraded, changed out, and there you go. Peas under the mattress as it were. Let's hope it is that simple.

bear
 
Old 15th Dec 2010, 22:48
  #1891 (permalink)  
 
Join Date: Jan 2009
Location: On a survey line somewhere...
Age: 42
Posts: 54
Likes: 0
Received 0 Likes on 0 Posts
colpaz Turbine D

There is no track record for the "Stub Pipe" issue, and since RR "Sprung" the Stub Pipe issue on everyone, in hopes of diminishing the severity of the Powerplant issue, The 75 cycles is related to the Splines, the AD's, and the 972. 75 cycles is about ninety days for Qantas, as I reckon it, and the AD has not been satisfied. The Oiling issue is new, at least as far as RR would have it.
That's certainly my interpretation based on what little I know of such matters.... Back to lurking now!

Cheers,
Colin
sgs233a is offline  
Old 16th Dec 2010, 01:00
  #1892 (permalink)  
 
Join Date: Dec 2010
Location: Middle America
Age: 84
Posts: 1,167
Likes: 0
Received 0 Likes on 0 Posts
Bearfoil
I don't know if Qantas' entire Trent 900 fleet has been inspected for spline wear, but the engine that failed was: See the ASTB report. It must be something that can be accomplished on wing.

Turbine D
Turbine D is offline  
Old 16th Dec 2010, 01:59
  #1893 (permalink)  
 
Join Date: Feb 2005
Location: flyover country USA
Age: 82
Posts: 4,579
Likes: 0
Received 0 Likes on 0 Posts
Turbine D:
The electronic engine controller (FADEC) and data recorders are an integral part of the engine and are mounted usually on the outside the forward part of the engine...
So true, not only for Rolls but others as well.

But it's the thermocouple lead from the LPT to the electronics, undercowl but outside the core casings, that I was referring to.

I suspect that the cockpit indications were also lost when the cables in the wing LE were severed.
barit1 is offline  
Old 16th Dec 2010, 02:08
  #1894 (permalink)  
 
Join Date: Feb 2005
Location: flyover country USA
Age: 82
Posts: 4,579
Likes: 0
Received 0 Likes on 0 Posts
AD applies to - - T972 only?

I am really puzzled by the corrective actions in this situation. Rolls Royce must have some really solid evidence to support the application to the T972 only, and not to the T970. Will they come to rue this decision?

And if the distinction is genuine, should not Qantas elect to derate their fleet to T970 thrust as an interim measure, until they are confident of the final fix?
barit1 is offline  
Old 16th Dec 2010, 02:14
  #1895 (permalink)  
 
Join Date: Dec 2010
Location: Middle America
Age: 84
Posts: 1,167
Likes: 0
Received 0 Likes on 0 Posts
Bearfoil
Let me propose a scenario that might be appropriate for the Trent 900 situation today.

Generally, engine manufacturers that develop, test, gain type certification and go through a multi-aircraft certification program do not stop at the conclusion and move on to the next engine. In fact, there is much more to be learned to be passed back to design engineers, performance engineers and the service business arm such as Rolls Royce's "Total Care". Usually an engine is reserved in a test cell where a cyclic program is run to simulate "the life of the engine". One or two other engines are available initially to run special cycles to examine component improvements or even new components. Why? Because the money to be made comes from the engine services, not the original build and sales of the engine itself. I don't think this is what you would call a "long runway engine program" where thousands of engines will be built so services is where the money will be made. Data from the "life of the engine" and planned periodic teardowns of this engine becomes the basis for refinement of design practices, comparison of performance over time real verses design theoretical, and most important, the criteria for establishing long term service contract costs. It behooves the engine manufacturer to improve all the components or systems in the engine to be capable on wing for 5 or 6 years enabling a profit to be made as part of the "Total Care" program while more accurately setting the per hour fee to future engine users. That all said, here is what I think happened to Rolls Royce on the Trent 900 Qantas engine.

One way to look at it is this way: (A) leads to (B), (B) leads to (C), (C) leads to (D) and (E) leads to failure. This is the classical failure methodology of a complex system. Looking at engine #2's failure,

(A) = Unknown
(B) = Unknown
(C) = Unknown
(D) = Stub pipe
(E) = IPT Disc burst/major engine failure/damage to aircraft

The trick is to identify (C), (B) and (A), (A) being the root cause.

From the aircraft flight certification program and the "cyclic Life of the engine" testing, I think Rolls Royce found some results (B) & (C) that didn't match original design life analysis theories and were in the process of making corrections. But the stub pipe failure, either Mfg. defect or wear, came much sooner than anticipated, if anticipated at all. If (B) & (C) could have been incorporated in engine #2, perhaps the failure could have been prevented. I don't know. Has the true root cause (A) been identified? I don't know, perhaps this is in the newest "C" mod. all the operators are wanting ASAP.

Turbine D
Turbine D is offline  
Old 16th Dec 2010, 02:20
  #1896 (permalink)  
 
Join Date: Dec 2010
Location: Middle America
Age: 84
Posts: 1,167
Likes: 0
Received 0 Likes on 0 Posts
barit1
Yes I see what you mean relative to the thermocouple leads.

Turbine D
Turbine D is offline  
Old 16th Dec 2010, 03:04
  #1897 (permalink)  
 
Join Date: Jan 2008
Location: Denver, CO
Posts: 265
Likes: 0
Received 0 Likes on 0 Posts
Okay, Gurus...

Looking at the cert for the T900 http://rgl.faa.gov/Regulatory_and_Gu...E/E00075EN.pdf I'm reading that the basic continuous max thrust for everything in the family is 71,850 lbs.


The difference is in the T/O thrust for the different variants - from 71,152 to 84,098 lbs.

The engine itself is apparently identical, and these different ratings are "plug" controlled.

So what makes the operation different?


Thanks in advance!



RR
rottenray is offline  
Old 16th Dec 2010, 12:46
  #1898 (permalink)  
 
Join Date: Aug 2009
Location: Texas
Age: 64
Posts: 7,179
Received 377 Likes on 231 Posts
A664:
The claim quoted in the article is that parts "need replacement after operating 75 take-offs at the 72,000lb maximum thrust level", so I understand that the engine can sustain more take-offs at lower thrust settings.
Light went on, I hadn't digested all of it. Thanks! You helped me connect the max rated versus "max typically used" dots.
Then it might be possible to fly a single aircraft to LAX and back up to 75 times (150 flights) before its engines have to be changed. If you use the aircraft on multiple routes - most of which don't need maximum thrust - you will get even more flights between replacements.
Got it.
Colpaz

On the 75 cycles bit - I'm certainly not an expert, but have read this thread from the beginning; My understanding is that the 75 cycle limitation is as a result of spline wear problems documented in a few ADs, and is certainly not the original spec of the engine.
I missed that, thanks!

RR: max continuous registers. The others would be for me "contingency power."

bear:
The 972 should have been given a miss, relative to Qantas and their long haul aspirations. Eight thousand pounds of Thrust left in the Hangar in favor of 970's must seem like to drool over about now, for Qantas.
What can 970's cost in MTOW and Pax/Freight? Some, I suppose, but again, the 972 is a mission killer as it stands now.
Potential mission killer, bear? If RR have got an adaptation or mod that accounts for the promised verus currently realized gap, will they not return to all of Qantas 972's full capability (or agreed/contracted) in due course as each engine is modified/inspected/refitted on (or perhaps off?) wing? Who and how that is funded is of course something for Qantas and RR to sort out between them ... and I will guess make sure the regulating agencies are happy with the final resolution.

I may have misread your post, and you are only looking at the short term with that prognosis.

Turbine D: much appreciate your well crafted thoughts, insights and explanations.
Lonewolf_50 is offline  
Old 16th Dec 2010, 15:15
  #1899 (permalink)  
bearfoil
Guest
 
Posts: n/a
Lonewolf 50

In my #1883, and in barit1's #1887, mention is made hypothetically that in hindsight (!), Qantas and perhaps Rolls would have been better off giving the 972 a miss.

The specificity of the AD's and the focus on Spline wear underscore something I see too in Turbine D's post relative to failChain. Either Roll's emphasis on the singular source of the problem (Stub Pipe manufacture) as being exclusive to the 972 is accurate, or some sort of wishful thinking.

Stub Pipe 6/7 is vulnerable to fracture due poor machining or wear (caused by vibration?).

The engine's are "identical" in design and construction throughout the inventory (900), one assumes the Oiling is consistent?

Vibration, assumedly, is not present on Install or early service life, but develops.

The AD directs inspections of Spline Wear "At the Abutment Face".

Is Spline Wear due lack of Lubrication? Does it make a difference, at this point?

People do guess at the cause of important problems, especially when involved in the Industry displaying same. Rolls serves no one, least of all themselves, by being so coy.

Cross-Type Modularity is a boon, but carries some important pitfalls. One size fits all, but also screws up as a Unit. Splines carry the very reason for the powerplant's existence, a problem of any kind with these portends other than a changeout of a stub pipe.
E IPT disintegration
D Stub Pipe failure, Oil Fire
C Vibration
B Spline Slop/Imbalance issues
A Choose one or more: Poor Lubrication, causing friction and rapid wear. Poor Design, Splines don't "Pocket" in the Abutment Face. Poor Metallurgy, Insufficient Ductility. Stretching too far the limit of "Lightweight/Power". Insufficient attention to "Production Rate" verse "Quality".

Early on, I figured this problem would cost about a Billion dollars. Exclusive of Marque Tarnish. Rolls needs "Marqued" improvement in Disclosure.IMO IMO.

bear
 
Old 16th Dec 2010, 15:50
  #1900 (permalink)  
 
Join Date: Aug 2009
Location: Texas
Age: 64
Posts: 7,179
Received 377 Likes on 231 Posts
Bear: you probably know most of this, but I am in part thinking aloud in response to your thoughts.
The specificity of the AD's and the focus on Spline wear underscore something I see too in Turbine D's post relative to failChain.
Stub Pipe 6/7 is vulnerable to fracture due poor machining or wear (caused by vibration?).
bear, as I understand it, the proposed defect on that is off center manufacture of what ought to be a concentric counterbore, which means wall thicknes min dimensions are not met for that particular piece. If that is what you mean by poor machining, concur. I find it easy to support your estimation that vibration contributes to premature fatigue failure/ (strain/creep) for that element of the engine's problems. It fits with other acft type I've see have odd problems (nsv causing things elsewhere to break) that beigin with a vibe initiated weakening of a part. Its relationship to the spline's problems will or won't be confirmed, by the time this is all over, if I understand RR's reporting requirements to the FAA/CAA/EASA etcetera.
The engine's are "identical" in design and construction throughout the inventory (900), one assumes the Oiling is consistent?
I don't recall seeing non similar lube on those "family" engines I worked with in the fleet, but that doesn't mean that RR may not have a few tweaks here and therebetween various mods of a given base configuration..
Vibration, assumedly, is not present on Install or early service life, but develops.
Vibration is a part of the flight environment. In my very humble opinion, the state of the art currently allows some predictable or estimatable fatigue (and thus safety factor) calculations to be made by those taking advantage of the decades of work and improvement in vibe analysis and failure analysis. I have no idea what sort of HMS (with vibe sensors in critical points) is built into A380, or if the rotary wing world's lessons on that have fully crossed over into fixed wing. I'd guess that the data would go to something like the ACARS/telemetry system right to the maintenance crews at various bases, so that a change in trend, spike, would trigger at least an inspection, and possibly further maintenance action when the hull arrives on a given sector's terminus. If X line is slightly mismounted or slightly misrouted (used to happen quite a bit in helicopters, which led to some innovative and cheap mitigation) it may pick up a vibe at a freq it isn't designed for.

How RR might, in the wake of this incident, back track that particular vibe/mounting for the tube in question is unclear to me, but I suspect it's in the realm of the possible.
The AD directs inspections of Spline Wear "At the Abutment Face".
Is Spline Wear due lack of Lubrication? Does it make a difference, at this point?
Good question.
People do guess at the cause of important problems, especially when involved in the Industry displaying same. Rolls serves no one, least of all themselves, by being so coy.
They serve themselves in not speaking prematurely, since they cannot ignore the reality of the litigous nature of modern society (global sense) and have to use sound survival skills to remain RR, maker of fine jet engines.
Cross-Type Modularity is a boon ... . , but carries some important pitfalls. snip the rest of that thought.
Concur
Early on, I figured this problem would cost about a Billion dollars. Exclusive of Marque Tarnish. Rolls needs "Marqued" improvement in Disclosure.
bear, I'll take a bit of RR's side on this, and advocate patience for us on the receiving end. They need to be very careful that whatever they say publicly and officially is dead right. They cannot afford to be the least bit careless about official utterances on this very important matter: the health and soundness of their amazing products.

I flew a WW II era designed trainer (T-28) some years ago (Wright Cyclone engine) that had a man-satisfying amount of horsepower at the sea level stop. When you put that throttle full forward, the engine roared that deep throaty bellow of power, and you leapt forward ... you'd better not be miserly with the right rudder application when you did so!

I think of 72,000 lbs of thrust in a Trent 972 and it makes me feel all tingly inside. Four of them? Tingly squared.

Last edited by Lonewolf_50; 16th Dec 2010 at 20:16.
Lonewolf_50 is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service

Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.