Go Back  PPRuNe Forums > Flight Deck Forums > Tech Log
Reload this Page >

A380 engine failure - Qantas

Tech Log The very best in practical technical discussion on the web

A380 engine failure - Qantas

Old 3rd Dec 2010, 04:14
  #81 (permalink)  
 
Join Date: Feb 2001
Location: Singapore
Posts: 89
From the ATSB report published today.

"A recent key finding from those examinations was the presence of an area of fatigue cracking within a stub pipe that feeds oil to the HP/IP bearing structure. That cracking was associated with a misaligned region of counter-boring within the stub pipe outlet."
ITman is offline  
Old 3rd Dec 2010, 11:16
  #82 (permalink)  
 
Join Date: Oct 2009
Location: Durham
Posts: 483
Bearfoil...

Good point but I don't agree in this case. In support of your view I would BAN all mobile telephone devices from the assembly areas of aircraft production.

A worrying note was sent out by GE/PandW this last week about oil line integrity in its products. This means that oil lines are being repaired. That means welded. Welding chrome alloy nickel steel formulated to high specs is notably difficult even for those who do it all day and everyday.

In this case the RR oil tube is assymetrical. Weak on one side. Given the vibrations and heat cycles this is subjected too it means that in this case the T972 as used in Qantas 388 it was literally vibrated until it broke.

The oil was probably leaking already due to micro fractures which would have altered the combustion times and set up a further harmonic that wore the spline shaft. Basically the engine was fighting agin itself.

Given all the fancy software design tools the kids use these days it is ironic that we have a failure here that is so familiar and has been with us since 1910.

Ban moble phones in work places a good start. If a company wants to subcontract out work then it should employ a full time qualified engineer on site...even if this is in some far flung part of the planet. This really is a very simple engineering failure.

Rolls Royce will get get stung mightily in litigation over this.
DERG is offline  
Old 3rd Dec 2010, 13:49
  #83 (permalink)  
 
Join Date: Dec 2010
Location: Middle America
Age: 79
Posts: 1,144
Lomopaseo

I would agree with your assessment. One additional point about these tubes and tube assemblies, chances are they are not produced by the engine manufacturer at all. Given today's world-wide supply chain, these parts are ideal candidates to be manufactured elsewhere, potentially part of offset agreements, e.g., you buy our engines and we will by a certain amount of components from your country.
Turbine D is offline  
Old 3rd Dec 2010, 15:48
  #84 (permalink)  
 
Join Date: Oct 2009
Location: Durham
Posts: 483
Manufactured where?

The fact is there is only a handful of places where this technology exists.

The one I am familiar with is the reserch dept. run by Dr. Magnus Hasselqvist of the company Siemens in the Swedish town of Finspong.

Of course GE, Siemens and RR are competitors so unless GE and PR will buy the patent and know how from Siemens this will get no where.

You can weld what you want to the alloys Magnus formulates...he is the leader in this field. The formal address is Siemens Industrial Turbomachinery AB in Finspong Sweden and the boss is called Vladimir Navrotsky.

Good outfit.
DERG is offline  
Old 3rd Dec 2010, 16:10
  #85 (permalink)  
bearfoil
Guest
 
Posts: n/a
DERG

I take it you are familiar with the technology surrounding turned tubes, or drilled, as it may.

In the rather sharp pics of the tube's end are two types of odd artifact. The Fractures are straightforward. What puzzles is the brownish off center coving at the lip's edge? It seems to be part of a patent "wear" signature, quite different than the irregular faces of the fractures. It has swirl marks as well. The area at the offset "drill" blunder also seems to have been "worn in" similar to the lip damage. Any thoughts?

I still am not clear on the need for such complex machining if this is but the terminus of a supply line. Everything I see suggests this is one end of a "coupling". If so, the wear evident in the image suggests a poor fit with the mate. Disclaimer: Now that we are all hyperfocused on this oil tube, what of some of the other telltales?

bear
 
Old 4th Dec 2010, 06:41
  #86 (permalink)  
 
Join Date: Oct 2009
Location: Durham
Posts: 483
The photograph

I am familiar with material failures as any trained engineer should be and moreover I am a free agent so I can comment without prejudice to my financial income.

The insitu postion of the pipe stub is unknown to me but there are several prima facie characteristics that should be noted. The ATSB has already noted some facts and should be read carefully.

The ATSB state this is a feed line. In my view it could be the single scavenge line particular to the bearing case on the Trent. We will see in due course. RR has a patent on this design.

The most significant fact, in my view, is the fractured brown discoloured upper edge closest to the camera. This is an old fracture. Just how old I don't know. But certainly old enough to corrode.

Now this means that oil was leaking through this fracture and hot air was getting in. This would have added oil to the combustion process and cause a very slight delay to the energy transfer cycle viz. a vibration...micro second harmonic. This could be the cause of the spline wear.

What I do know for certain is the Rolls Royce are PARANOID about sharing technology. It is no surprise to me that they have been caught out with this. To be fair they try hard to get staff educated but obviously this simple component caught them out. Sad for UK engineering.
DERG is offline  
Old 4th Dec 2010, 08:37
  #87 (permalink)  
bearfoil
Guest
 
Posts: n/a
DERG

It's a bit clearer to me. You state that "Oil was getting in, and hot air." Are you saying this opening is part of a closed system? I repeat my question that the end of this pipe appears to be half of a line coupling, and losing the integrity of the line as the coupling failed may have caused the oil to drown the bearing case? Why is a counter bore necessary at all if this is a terminus to a (unpressurized, and un-checked 'feed')?

bear
 
Old 4th Dec 2010, 09:58
  #88 (permalink)  
 
Join Date: Oct 2009
Location: Durham
Posts: 483
Bearfoil..

Yes, excuse me..got my Toyota Corolla stuck in the snow..four dachshunds..irate wife..and a couple of pet huskies causing distress to all..finally got the thing out and all occupants and home.

If I am not mistaken the central disc that was exploded out revolves in the opposite direction to the other rotors. I believe this is a feature of RR aerospace turbines if not all in the RB-211 pattern. If someone can confirm this?

So WTF am I taliking about..right? Well we know that this engine is a bypass design and a lot of air does not enter the combustion process. If I am not mistaken I believe this is about 10/15% that gets inducted in the burn section to be mixed with kero.

We know that the oil feeds are subject to hot gas and that the oil should be kept to below 150C. We know that the oil line in question was fractured and I know personally that it was fractured BEFORE this final failure...final failure..shiny crystaline surface on the metal. Old fracture is seen brown oil/corrosion/colour.

Cause of fracture: stress due to harmonic vibration. Only possible cause. Heat cycles added..severe for T-972 'cause of higher energy transfers at max TO weights AND high ambient temps as in S Cal and Australia.

From first principals the engine SUCKS SQUUEEZES BURNS then THRUSTS. The thrust part depends upon the kero/air ratio. The thrust turns the sucker fan at the front. In these new engines all of this is data mapped so the engine KNOWS WTF is going on and adjusts the fuel flow to match.

If you have an oil leak..the turbine oil gets mixed with the kero and changes the burn characteristic..in particular the TIME taken to release X amout of energy. RR are very proud of the fact that their products get more thrust out per total unit mass than the rest. As they boast "High in power, yet low in mass" with " low emissions".

All very well if the kero/air mix is right and the software is not confused. If you have turbine oil mixing with the air then with the kero the whole model is sent to hell. So the thrust turbine thinks it should have X amount of air with a weight of X amount of kero where as the reality is different.
So the different parts of the engine are out of sync.

This sets up a harmonic and the situations gets progressively worse. The vibrations get worse, the temps rise and eventually a part fails due to stress fracture. Of course if the central disc is turning the other way this can make the situation worse.
DERG is offline  
Old 4th Dec 2010, 11:48
  #89 (permalink)  
 
Join Date: Jul 2007
Location: West Sussex
Age: 77
Posts: 56
DERG - A very entertaining scenario. But by what means does the fuel find itself with the oil. Is it a Swedish thing? Is Dr Hasselquvist involved?
CAAAD is offline  
Old 4th Dec 2010, 12:00
  #90 (permalink)  
 
Join Date: Oct 2009
Location: Durham
Posts: 483
Bear and Caad

Caad...amzn how simple the faults turn out to be. Just poor part manuafacture,

Bear
As far as I understand the guy who designed this bearing was sure they could not mess up the manufacture.

It is a continuous flow with pressure differentials in play. The system is vented to the atmosphere but only at a start up. In service the way the oil is contained is very clever using centrigul forces. But it DOES require the oil to say in grade..in that sense the oil has to be ON SPEC. Ie right visco and temp and pressure.

That is why oil oxidation due to volcanic dust is a factor in this design. It is in any turbine but esp in this one with this bearing lube design.

Basically the Trent is like a high strung race horse. Wunnerful when all the variables are on the mark.
DERG is offline  
Old 4th Dec 2010, 12:02
  #91 (permalink)  
 
Join Date: Oct 2009
Location: Durham
Posts: 483
How does oil mix with kero?

Squirts out the fractured oil pipe. Under pressure.
DERG is offline  
Old 4th Dec 2010, 12:14
  #92 (permalink)  
 
Join Date: Jul 2007
Location: West Sussex
Age: 77
Posts: 56
Yes, I can visualise the oil squirting out of the pipe fitting, but how does the fuel arrive in the bearing housing?

Ah, Sorry, I think I have misunderstood again. Are you suggesting that the oil mist from the bearing housing finds its way into the combustion chamber and thus upsets the engine control system? To such an extent that rotor integrity is compromised?

No oil fire in bearing housing, then?
CAAAD is offline  
Old 4th Dec 2010, 16:32
  #93 (permalink)  
 
Join Date: Feb 2005
Location: flyover country USA
Age: 77
Posts: 4,580
DERG - I'm having a hell of a time following post #88.

Firstly - Aircraft engines must burn a variety of fuels. Within the Jet A spec there is a range of specific gravity, a range of Hydrogen content, thus a range of LHV (BTU/lb) that must be accommodated. Add or subtract a bit of turbine oil, and the fuel/oil properties change very little, compared to the variation that already exists in the Jet A spec. In addition, engines are often approved to run on other fuels - Jet B (higher volatility for arctic use), etc.

Now then - at TO the engine is gulping a few dozen gallons of fuel per minute. The total oil tankage is only a few gallons. IF THERE WERE ENOUGH OIL ENTERING THE COMBUSTION CYCLE to make a measurable difference, the engine would start to overboost. Now what will happen?

The governing function of the fuel control or FADEC or EEC will trim back fuel delivery to bring rpm back to target!

So you see my dilemma: First - how will engine oil at a few dozen psi find its way into the main gas path (several hundred psi), and Second, how could this create overspeed?
barit1 is offline  
Old 5th Dec 2010, 04:54
  #94 (permalink)  
 
Join Date: Oct 2009
Location: Durham
Posts: 483
CAAD

"Yes, I can visualise the oil squirting out of the pipe fitting, but how does the fuel arrive in the bearing housing?"

Because RR use a patented oil bearing case with only one scavenge line. Lube system works by pump, air pressure differentials and centrifugal forces. Totally different to GE/PE. Very precise and neat design.
DERG is offline  
Old 5th Dec 2010, 05:10
  #95 (permalink)  
 
Join Date: Oct 2009
Location: Durham
Posts: 483
BARIT..glad you enjoyed my post!

Look up "JP8" and you will realise that commercial engines and military engines are two diff animals. If you think the Trent can handle ANY fuel you are mistaken. The way this Trent is designed it gets every last bit of energy from the kerro. Yes I do believe an oil mist would disturb the cycle. We are talking micro/nano seconds.

The days of running jet engines on paraffin are gone with these new units.
They are really efficient.

Remember only about 10/15% of air coming in is used for combustion.

At TO the 972 would gulp a few gallons per SECOND in the Qantas 388 at full load. Makes ya just slaver your chops don't it!

Not sure about overboost cause the software would start to mess with the kero supply. Hence the bad vibes and spline wear. The software will try to correct yes. But all the other parameters don't match the model..hence the stuttering and spline wear.

Did I mention overspeed? I don't think turbine oil was mixing ALL the time but I do think that at the 72K pounds the Qantas A-388 was putting out at TO it was. It would be interesting to know just how much oil had to be added after each flight.
DERG is offline  
Old 5th Dec 2010, 09:35
  #96 (permalink)  
 
Join Date: Jul 2007
Location: West Sussex
Age: 77
Posts: 56
DERG - Modern engines are cleared to use a wide variety of fuels and additives. We say 'fuel'. Not sure what kerro is, or paraffin in this context.

But I am very impressed by the power of your imagination in putting together a scenario whereby oil manages to make its way to the combustion chamber and upset the combustion process to such an extent that hazardous vibrations leading to shaft disconnect result.

I'm with Barit on this and consider it a very unlikely explanation.

But do tell us how you think the oil arrives at entry to combustion chamber. I think I may have asked this before but may have overlooked the explanation.
CAAAD is offline  
Old 5th Dec 2010, 12:10
  #97 (permalink)  
 
Join Date: Oct 2009
Location: Durham
Posts: 483
CAAD

Well I call it kerosine mainly because of the smell...fuel..not definitive enough for me.

The disc thingy was exploded out because the oil/air was on fire inside the bearing case.

Now you want to know how the turbine oil got mixed in with the kero...

Well if you look at the photo you will see that the fracture on the part nearest the camera is old. So that pipe has been porous for enough time for the surface to be discoloured.

We know this bearing was in the mid section of the engine. The oil is sent around in pipes which we know are subject to fracture. Air and oil was mixing.

So we have fractured pipes feeding hot oil into the engine domain. The reason why I am convinced this happened is the premature wear on the spline shaft.

Thank you for your comment on my imagination. I can see why the disc exploded out,but I use my imagination to apply a reason for the spline wear. All comments welcome by return.
DERG is offline  
Old 5th Dec 2010, 13:41
  #98 (permalink)  
bearfoil
Guest
 
Posts: n/a
I am intrigued by the Splines wear. It seems reasonable to assume that the TRENT is a well balanced and smooth running machine. I also think it possible to entertain other than lack of lubrication as a partial factor in the loss of Spline metal.

Under the great stress of Climb thrust, the engine is reliant on its internal balance for long service life.

Splines, though accomplishing a "Rigid" coupling, are subject to vibration (chaotic) and Harmonic (synchronous). Each wave upsets the face of Spline, Groove with opposite and energetic input. Hence the loss of metal from the "Crest" of the Spline, and the faces as well.

Under its ultimate Power loading, any impact that impinges on the face/groove of the Rigid coupling produces wear. Almost all of the Thrust from nozzle and Fan are used to propel the a/c, some is lost as friction, it is friction that may have done in the Rigid coupling. The "Tug O' War" between LPT and Fan is "Refereed" at the Thrust Bearing, any load impinges on the Splines, if there is very rapid loading/unloading in this couple, the joint beats itself into the wear rate that caused the AD's IMO

I still think that while very important, the oiling is not the entire answer. It may be that the extra one tonne of max Thrust is a bit too much for the system. That is a very big Supersonic Fan, any vibration and harmonic whether mechanical or 'acoustic' is a very big load.

Just some thoughts

bearfoil
 
Old 5th Dec 2010, 13:51
  #99 (permalink)  
 
Join Date: Dec 2010
Location: Middle America
Age: 79
Posts: 1,144
Bearfoil and DERG

In the photograph of the fractured stub pipe in the area of the fracture, I think the surface having swirl marks on it is the result of subsequent forensic examination. That is to say, the swirls come from a radial cutting wheel under coolant spray (may account for the brownish color). The examiners were probably trying to get a better view of the fracture itself, and cut away an unaffected area to enable this. I am sure there are other unreleased photographs that show this component in its original state as found before analysis.

Turbine D
Turbine D is offline  
Old 5th Dec 2010, 14:07
  #100 (permalink)  
bearfoil
Guest
 
Posts: n/a
TurbineD

Point taken. However, the 'damage' is consonant with the offset of the 'miscreant drill'. I still think it possible that the 'smoothed' offset is wear induced, perhaps caused by an 'aspirator' affixed to the end of the Stub. An open and unadorned exit of an oil supply/scavenge doesn't fit with a supply that is highly engineered to 'mist' its lubricant.

I could see a rotating tip, an 'aspirator' to induce an aerosol mist into the bearing cavity.

The AD specifically mentions Vanes. Restrictor Vanes. The AD also addresses "clogging" vanes. If there was such a "fan" at the tip of this line, its loss would increase oil supply. If this increase in Oil Supply is too much for the scavenge system, the case would flood with oil. This unscavenged Oil would be noted at Flight's end by line people, or even spoken by a CEO: "There is Oil where it shouldn't be..." Alan Joyce.

If this "imagined" aspirator became "clogged" it would resist the Oil's flow, and perhaps, if out of balance, cause the wear that is evident on this terminus.

I think your point is well taken, but I also doubt any destructive testing would want to 'follow' an out of round (eccentric) fault.

conjecture.

bearfoil
 

Thread Tools
Search this Thread

Contact Us Archive Advertising Cookie Policy Privacy Statement Terms of Service

Copyright 2018 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.