Bowed rotor

[AeroTech]

Hi,

I have few questions regarding bowed rotor on PW 1100 engine. If I am not mistaken this problem was discovered after the engine was certified.

a) Why PW didn't discover this problem earlier?

b) Did Airbus discover this problem before the certification of A320 NEO?

It seems this problem is related only to PW 1100. Assuming this is true:

c) Why other PW geared turbofan engines (PW1200G, PW1400G, PW1500G) don't have this problem?

I understand that aircraft and engines manufacturers usually issue service letters or bulletins after the aircraft is certified.

d) Can you mention examples where aircraft manufacturers (Boeing, Airbus, ...) or engine manufacturers (PW, GE, RR)
missed to discover major problem(s) on time (after certification). Please be specific by mentioning the system/subsystem... on aircraft or engine/APU.

Thank you.

[tdracer]

Bowed rotor is a common problem, affecting most new technology turbine engines. It's exactly what it the name describes - differential cooling of the HP rotor causes it to bend or bow. As a result, there is period after shutdown - anywhere from a few minutes to several hours after shutdown depending on the engine - where restarting the engine will result in issues - anything from high vibrations to blade rubs and -worst case - blade damage. To achieve max efficiency, designers are minimizing blade to case clearances - good for efficiency, but makes bowed rotor start worse.

I have no first hand knowledge of the NEO issue, but I'd bet pretty good money that it affects all the PW geared fan engines to a similar extent.
To be fair, it also affects the Trent, GE90, GEnx-1B and -2B, LEAP, and to a lesser extent, the CF6...

[pattern_is_full]

Not just new-technology engines. Concorde's Olympuses (Olympii?) required a "de-bow" procedure for re-starting (using very low rpms) between 15 min. and several hours after shutdown (time period when the engine was cooler than operational, but not yet cool enough). The debow start allowed the temperature to increase slowly to operational, while minimizing vibrations while it was still bowed.

That was a bowing of the shaft(s), however, not the rotor (turbine/compressor disk).

The PW1100G engine with problems on the NEO also appears to be an issue with differential or asymmetrical shaft cooling and bowing, that in turn "tilts" the rotor slightly out of alignment.

The bowing is due to differential expansion and contraction forces on the relatively long skinny shaft. Like pushing on both ends of a drinking straw - it will bend in the middle.

EDIT - PW CEO mentions that the bowing is less pronounced when the engine is "fan mounted" rather than "core mounted" on the pylon. The engine was tested on a 747, with room for the whole engine UNDER the wing - the smaller A320 mounts the engine higher and cantilevered out ahead of the wing for ground clearance. I'm guessing that is what he meant.

Their temporary fix is a long, slow spool-up to warm up the engine parts - like the Concorde de-bow procedure.

http://aviationweek.com/commercial-aviation/new-pw-president-has-nothing-hide-gtf-starting-issue

[vapilot2004]

To be fair, it also affects the Trent, GE90, GEnx-1B and -2B, LEAP, and to a lesser extent, the CF6...

I would assume all of the above handle this condition automatically by lengthening motoring times during start.

[JammedStab]

To be fair, it also affects the Trent, GE90, GEnx-1B and -2B, LEAP, and to a lesser extent, the CF6...

I did a couple of flights on the same aircraft last year with CF-6 engines that had encountered some sort of bowed rotor start. I suppose they have sensors that send the info to maintenance. Anyways, both days during the turnaround at the outstation, maintenance came on board and did a motoring procedure for about a minute or so at the gate.

[tdracer]

I would assume all of the above handle this condition automatically by lengthening motoring times during start.
It's not quite that simple. The GE90 does simply motor longer before fuel on, but on the GEnx, max motoring can result in resonance of the shaft which actually makes things worse (and if I wasn't clear earlier, bowed rotor is a bowing of the shaft, not of the discs). On the -1B (787), since it's electric start it's a simple matter to reduce the starter torque to keep the motoring speed down, but on the -2B (747-8) the pneumatic starter air valve is either open/closed, so to keep the motoring speed down they reschedule the stator vanes during the bowed rotor motoring to keep the N2 speed down.

[FullWings]

The GE90 displays the worst symptoms that I’ve come across, probably because the turn-round time of a LH aircraft puts it in the period where bowing is most noticeable. It doesn’t take that long to even out but the wobble is quite pronounced sometimes.

If it’s actually causing damage, I’d have thought one solution would be to have a small electric motor that slowly rotates the N2/3 section when you think it’s not long before starting (or use the motor that’s already there, like the 787 above).

[vapilot2004]

It's not quite that simple. The GE90 does simply motor longer before fuel on, but on the GEnx, max motoring can result in resonance of the shaft which actually makes things worse (and if I wasn't clear earlier, bowed rotor is a bowing of the shaft, not of the discs). On the -1B (787), since it's electric start it's a simple matter to reduce the starter torque to keep the motoring speed down, but on the -2B (747-8) the pneumatic starter air valve is either open/closed, so to keep the motoring speed down they reschedule the stator vanes during the bowed rotor motoring to keep the N2 speed down.

Interesting. Thanks TD.

I assume temperature is a factor in the digital decision making process? I wonder if time since shutdown also plays a role.

[JammedStab]

It's not quite that simple. The GE90 does simply motor longer before fuel on,
So the reason for the fairly long time to add fuel on the GE90 compared to the CF-6 is an intentional design to reduce rotor bowing?

[tdracer]

Interesting. Thanks TD.

I assume temperature is a factor in the digital decision making process? I wonder if time since shutdown also plays a role.


The logic looks at time since shutdown and residual EGT (for the GEnx bowed rotor it's between 30 minutes and 6 hours after shutdown) - if the time and/or EGT are outside the bowed rotor threat zone it just performs a normal start.


Fullwings, that's exactly what's being proposed for the GE9X on the 777X - a small electric motor on the gearbox will slowly rotate the N2 shaft for some time period after shutdown. We're calling it the rotisserie :E


JammedStab - yes, the GE90 will max-motor for several seconds (17 seconds IIRC) before turning fuel on to alleviate bowed rotor.

[Intruder]

a small electric motor on the gearbox will slowly rotate the N2 shaft for some time period after shutdown.
Large ships have had that for decades; keeps the shaft rotating even when at the pier. Called "jacking gear".

[lomapaseo]

Large ships have had that for decades; keeps the shaft rotating even when at the pier. Called "jacking gear".

and lots of other stationary installations as well.

I wonder where they will place the motor and get the power from in an engine pod and under what part of the FARs?

[JammedStab]

JammedStab - yes, the GE90 will max-motor for several seconds (17 seconds IIRC) before turning fuel on to alleviate bowed rotor.

But no big deal for a manual start to just select fuel on at initial max motoring(22% I think) without a delay.

[spannersatcx]

fuel flow should start after 3 secs on manual start or 18 for autostart on the GE90

[oldpax]

"Intruder"and "full wings",large steam turbines and Industrial gas turbines have this arrangementbut are called barring gear.Jacking gear usually refers to a small H.P. oil pump that puts a small wedge of oil under the shaft to lift it slightly before turning commences.

[FullWings]

Fascinating stuff. Having just looked it up, “barring" comes originally from the need to move large engines off top-dead-centre with a crowbar! Barring gear is a more sophisticated method of achieving this.

Sounds simple. For low-cost operations, a FO and a crowbar should do the trick...

[TURIN]

I wonder where they will place the motor and get the power from in an engine pod and under what part of the FARs?

I'm guessing a small motor on the engine gearbox would be capable of turning the HP shaft, problem is it would have to turn the fuel pump, oil pumps, hydraulic pump(s) etc. This small motor is suddenly getting bigger.

How about just reversing the hydraulic pump flow and using it as a motor?

[FE Hoppy]

PW1500G have the same problem but it's not uncommon as can bee seen in this thread. You just need procedures to manage it and not hold up operations which can be tricky at small airports with short taxi times.

[barit1]

I have proposed a simpler method to avoid bowing, although it requires some unusual timing for the crew.

The bow is the result of thermal convection within the case - warmer air rises to the top. If one were to "tickle" the starter for a few seconds, some xx minutes after shutdown, it would stir up the air and randomize the rotor positions, thus causing the bowing to be somewhat reversed. I haven't seen where anyone has tried this gimmick.

[flyhardmo]

I haven't seen where anyone has tried this gimmick.

Then you haven't seen the pilot of a Garret powered aircraft (eg: metro, Donier 228 etc) and hand spin the prop for a few mins after shutdown.

[msbbarratt]

The bow is the result of thermal convection within the case - warmer air rises to the top. If one were to "tickle" the starter for a few seconds, some xx minutes after shutdown, it would stir up the air and randomize the rotor positions, thus causing the bowing to be somewhat reversed.

You wouldn't even have to move much air. All that would be required is that all the shafts are kept turning, even very slowly, to allow heat dissipation to be uniform around the shafts. That way there'd be no bowing at all.

You often see engines windmilling whilst the aircraft is on the stand. That's probably not quite going to achieve the desired result; the auxiliaries must stop the one of the shafts turning, leaving it to soak in the heat.

[tdracer]

I'm guessing a small motor on the engine gearbox would be capable of turning the HP shaft, problem is it would have to turn the fuel pump, oil pumps, hydraulic pump(s) etc. This small motor is suddenly getting bigger.
I don't recall the size of the proposed rotisserie motor on the GE9X, but's quite small (yes, gearbox mounted, with a BIG gear reduction). It's only intended to turn the shaft at something less than 1 rpm so it doesn't take much torque.
More critical is the one way clutch - otherwise that little motor would be going a zillion rpm at redline N2..

[fruitloop]

IAE V2500 also suffers from "bowed rotor effects" Mainly the A1 variant..