Bearfoil
old engineer
Re: Spline Wear AD. The document references accelerated wear at the "Abutment" Face. From the drawing, the "Trough" cut for the Spline shows a lazy exit for tooling at the aft, sharp end of the feature. Is this then the "Abutment" (Terminus) for Spline Travel? The "Crest" of the Spline is to be measured as often as every third Flight. Is the Spline subject to "sharpening" against the stop? Does this allow the Shaft to wander aft? It is of course far more difficult to make a "Pocket" of the Trough end, and expensive, yet a squared stop would absorb more energy, longer? What do you think?
What I'm thinking right now is that perhaps the source of the "oil fire" can be approached by considering the energy that the splined/geared coupling might have absorbed, and when and why it might have done so. (I'm deliberately using "splined" and "geared" interchangeably. It would take a full post to say why; but briefly a patent for AC application of such a coupling makes the same comment-- the GE patent?)
I'm now looking at there being a distinct possibility that the energy dissipated at the point of spline damage could well have caused the coupling to have ignited its own lubrication. For example, a spur gear running in a case containing an oil bath will often require an arrangement to cool the oil. This is true even though involute spur gears are 98% efficient. The other 2% can be a real problem. Our coupling has involute gear teeth/splines (again it is correct to equate the terms); the helical form of about 30 degree lead is just a twisted form of a spur gear.
If we lose the spline/gear teeth (say the oil fire they lit starts to soften them), and suppose we take at face value the AD comment that the IPT may then move rearwards (ruckwarts gegen, perhaps-- German is the language to discribe such events... fewer softening words), then the sequence of events is: Spline/gear overheats; its lubrication ignites, weakening splines-- mild steel would lose 1/4 its strength in a hot kitchen oven; splines are sheared completely off; IPT moves rearwards into LPT inlet vanes; IPT section explodes, oops bursts; in the burst the stub oil line to the IPT aft bearing is dislodged, breaking the already fatigue- cracked oil-inlet boss.
Post hoc, ergo propter hoc, eh? I think I've got all the events of the official communiques, just in a different order. Alright, I did add loss of the splines/gears. But the official preliminary ATSB report does show them missing, doesn't it? At least, posters are saying they can't see the splines. So the question remains, what event is post hoc, and what is propter hoc? (For those without a misspent youth, "came after, therefore was because of".) Not to mention, are the splines there, or not?
I did discover a few things about the Trent 1000 as used on the B787, in 6 US-certified versions, presumably those used on the test airframe. Some of this may shed some light on the 972 engine which failed:
0. L & dia of engine 15'-6" & just under 12'-6" dia.
1. The wt of all versions is 12,710.
2. The thrust of all versions is different, range about 20%.
3. The original version thrust is near but not at the top of that range.
4. Max thrust is permitted for 5 min; 10 if one engine out. Data ff w/aux's:
5. Max thrust TO is 74511; max cont 69523 lbs-- most powerful version.
6. Max thrust TO is 69194; max cont 64722 lbs-- original version engine.
7. Max temp TO, at entrance vanes to LPT, is 900-C (5/10 min as above).
8. Max continuous temp is 850-C.
9. Max overtemp 920-C, limit 20 seconds. Ascertain why and correct.
(Suggests to me 20 sec to see trend, decide, & shut down).
10. LPT shaft 100% rpm's = 2683. Max permissable rpm's follow:
11. LPT max setting N1 = 98% (or 2629 rpm), rpm's not reported to pilot.
12. LPT max setting N1 = 96.5% (or 2589 rpm) if fwd speed below 60 kts.
13. LP stages: compressor, 1; turbine, 6.
14. Fan diameter: 112 inches.
15. Shafts: LP and IP rotate CCW; HP rotates CW; looking from rear.
16. Engine is fitted with vibration sensor; sends to AC (then pilot, or HQ?)
17. Oil pressure 30-75 psi.
18. There is a list of lifetime-limited parts.
19. Engine does not supply compressed air for airframe ventilation.
20. Engine not rated for thrust reverser operation.
21. Engine not eligible for ETOPS.
22. Engine not rated for operation with faults present in the control sys.
23. Fuel inlet (@engine) min temp: minus 25-C. (I infer FOHE now OFHE
)
24. TC holder is RR plc, Derby.
25. Use is: OI-Trent 1000-B787.
I'm going to try to answer your specific questions ASAP. I don't know exactly where and in what form the stops are, and have been trying to find that out. I do think that on disassembly the LPT section is simply unbolted, and the shaft just pulls rearward with it, without any restraint except drag or slight friction. This would not be true of the splined/geared collar (of the coupling) into which its spline/gears fit; the collar must be restrained in some way, either to the forward shaft, or locally to some other element or structure. The collar being restrained by the forward shaft would however re-raise the same question you asked.
That is why I introduced the conventional geared coupling idea. These are so ancient and so not under a lid of trade-secrecy that the answer to the stops might be found there. The problem is the same; the solution might be similar.
I do now know (I didn't at my last post) that this shaft coupling is not a "curvic" coupling. Curvic couplings do occur in this engine (I assume, because it is so common in the ff use), but are used to join the centers of the disks (that carry the vanes) within a stage. They are like cups with gear teeth on their rim. Yes, gear teeth, although of a special form that would (possibly) permit the whole stack to curve without losing concentricity (ie, if my idea of the whole engine casing, LPT shaft and all, flexing into an arc under a max thrust TO, is at all true). They are a kind of bevel gear laid out flat as a pancake-- invented by Gleason Gear 60 years ago for keeping the tool posts on big lathes lined up true when taking a heavy cut, or somesuch.
Joined together, these "curvix" widgets also form a tunnel of sorts over the shafting. Then, perhaps, the hollow shafts can be more effectively be air-cooled. But I'm sure the shaft coupling is more on the idea of the GE-patent coupling with the parabolic gears/splines. That'll take a whole post, too. Well, a long post but I wanted to throw out some grist for ideas.
OE