A330 Oil pump failure/Engine shutdown 2

[gearcutter]

Interesting summary of a yet-to-be-released report into the cause of an engine shut-down on an A330 while flying over Australia.

http://www.atsb.gov.au/publications/investigation_reports/2016/aair/ao-2016-101/

So far it appears like there's been an oil pump failure due to damage sustained to the drive shaft.
Looking at the picture of the failed shaft; I'm somewhat surprised at how it has been designed. Perhaps the undercut between the spline and bearing diameter is some kind of inherently designed shear point.

 

[MikeHalloran]

I'm guessing the gearbox that drives the pump is much, much more expensive than the pump itself, hence the frangible shaft.



Mike Halloran
Pembroke Pines, FL, USA

 

[SwinnyGG]

I would think Mike is right.

Doesn't look like a fatigue failure to me, although the fracture surfaces aren't really visible. Wonder if the failed shaft is the result of a gearbox lockup or other problem further down the chain.

 

[mfgenggear]

I agree with the previous post, I fabricate many shafts similar that have a built in shear neck. made to shear at a exact load in LBs or Inch LBs. the spline it self appears to be undamaged.
the wall thickness of the shaft appears to be very thick. hence tremendous amount of over load. but just in case , have a failure analysis of the shaft. correct material, heat treat , hardness of the core.
make sure the spline was correctly fabricated. with go and no go gages. and it was not binding up. even thou does not appear to be.

 

[mfgenggear]

hell gear cutter, it went right over my head, that this was a more show tell. I am bad.

 

[tbuelna]

The report linked in the OP notes the engine involved was a Rolls-Royce Trent 700 series. This engine has its lube oil pressure/scavenge pumps located on the EMAD. Most of the accessories driven by the EMAD (lube oil pumps, hydraulic pumps, IDGs, etc) typically have something like this shear neck to protect the EMAD geartrain if the accessory experiences overload/seizure.

This design approach actually has nothing to do with cost considerations. It is done entirely for safety and reliability reasons. If you look at the image below of a typical turbofan EMAD geartrain, you'll note that the gears are arranged in series. Failure of a gear in the middle of the geartrain will disable all the accessory drives past that point. Turbofan engines are designed to operate for a limited period of time with loss of lube oil circulation. The EMAD gears and bearings are also designed to operate under loss of lube conditions. The most important considerations in the event any driven accessory experiences a seizure are protecting the entire EMAD drivetrain all the way back to the PTO gear mesh at the compressor shaft, and ensuring continued operation of the other accessories located on the affected EMAD.

 

[MikeHalloran]

Thank you, tbuelna.

It smells like they are going to yell at the flight crew for the repeated attempted restarts,
and perhaps modify associated rules/procedures.

I hope there will be a followup report, outlining what caused the oil pump to lock up in the first place.




Mike Halloran
Pembroke Pines, FL, USA

 

[gearcutter]

Thanks for the explanation tbuelna.

 

[tbuelna]

MikeHalloran-

You'd be surprised at how much power an EMAD like this is designed to handle. The typical EMAD used on current commercial turbofan engines is also designed to last around 30K hours, so they are very durable.

Failure of a lube oil pump drive shaft like the example shown is unusual. The lube oil pressure/mass flow rates required by the engine are rather modest, so this should not have caused a fatigue failure. There are filter screens at each pressure/scavenge pump stage inlet that should have prevented any debris from causing a seizure. The most likely thing that might have caused an oil pump seizure would be a structural failure of a pump element.

 

[3DDave]

Mike - yeah, the restarts will attract attention. It's unclear if the pilots understood that operating an engine without oil is a useless exercise. What did they expect the warning was for? The worst is that it distracted them piloting the plane. Chasing that sort of distraction has killed a lot of people.

 

[SwinnyGG]

The lube oil pressure/mass flow rates required by the engine are rather modest

In comparison to what? You have sparked my curiosity.

With my very limited understanding of jet engines I would have assumed the pressure and flow rate requirements were rather high compared to say, the automotive world.

My understanding is that typical jet fuel lubricants are similar to slightly higher than automotive oils in viscosity- as a result I always assumed that actual bearing pressures and thus supply pressures at the bearing would be similar, with much higher volumes than a typical automotive engine due to the much larger number of components being lubricated and higher pump output pressures because of the distance of those components from the pumps.

 

[tbuelna]

In comparison to what? You have sparked my curiosity.

There is not much in the engine core itself that requires cooling by the lube oil. A few rolling element bearings, carbon face seals, shaft splines, and a single PTO gear mesh for the EMAD. The oil tank volume for this engine is only around 6 gallons. Compare that to the main rotor transmission of the CH-53K helicopter with far less installed power. It requires an oil volume of around 50 gallons, primarily for cooling purposes.

 

[SwinnyGG]

Makes sense.

What are the pump design pressures like?

I didn't think the oil was designed to provide much cooling- just figured that the rotor assembly would have enough bearings that the supplied quantity of oil would be high.

6 gallons is definitely less than I would have assumed for oil volume.