In order for single turbines like the PC12 to be used in commercial operations in some countries like Canada and Australia, they had to provide data on all loss of thrust situations, so they have pretty good data and good enough to satisfy the authorities. I think it required no more than 1 IFSD per 100,000 hours, which they easily had data for.

As far as that Meridian accident, the cause was never known. The FAA did sound analysis and did not detect a problem with engine pitch. If it were an FCU rollback, the MOR should have been used, was not. The plane did not land off field, they stalled it. If they wanted to return to field, without an engine, they were supposed to feather the prop. So a lot of ways this could have ended. The pilot did not do a single one of the things that is called for in standard training. The NTSB did not attribute this accident to an engine issue. There were a lot of possibilities discussed, but there was no evidence the engine failed, and in fact the plane hit with a lot of excess speed that could have been turned into climb, glide, whatever. In fact they said they did NOT believe that it was a rollback after completing the investigation. Here is the final:

The National Transportation Safety Board determines the probable cause(s) of this accident to be: The pilot's excessive bank angle and his failure to maintain airspeed while returning to the airport after takeoff due to an unspecified problem resulting in the airplane stalling and colliding with trees during the resultant uncontrolled descent.

If someone wants to read the whole thing:

https://app.ntsb.gov/pdfgenerator/Repor ... l&IType=FA

Seemed low on power.
Used lots of runway.
Climbed very weakly.
Low airspeed.
Cracked P3 line.

Seems like there is evidence of an engine problem. Note that evidence doesn't mean proof.

A plane flying slowly into trees is a sign of lacking power. It is hard to imagine this accident occurring with an engine making full power. Maybe that was pilot induced, maybe it wasn't, but it is hard to say this crash was not engine related. The NTSB certainly concentrated their efforts on the engine.

Mike C.

I feel your pain, I've been there...actually, I'm still there.
She really doesn't care about the parachute or the number of engines, it's just an excuse, what she really likes is the Cirrus.
Sorry man.

You just made that up. "They tend to not include".... Yeah. There are multiple redundant systems for all of them, Heck, the PC12 has 3 prop governors. Only single point failure on the items you listed would be the high pressure fuel pump. Not sure if there's a single failure ever recorded of that one.

Just as a data point, our company has operated PT6's for close to 500k hours, and we've had one uncommanded Py failure in cruise, easily recovered with MOR operation. I know of other major PT6 players who run around same numbers. MTBF in the 500-600k hour range.
I think the 300k+ MTBF number is accurate, unless you have actual data to show?

I think you underestimate how many PT6's are operating at any given moment, and think the 350k MTBF means you should be able to operate a single specimen for 350k hours without a failure.

Not original with me.

Like John Oaks earlier in this thread:

"That's great for the engine core. I'm also concerned about mundane things like oil lines, FCUs, oil pump, fuel pump. When those pesky things fail, I have been told that it doesn't count as an engine fail. Anybody else hear that?"

Yes, John, I've heard that from many people.

Here's Canadian assessment of PT6 failures (PWC home country, BTW):

"The PT6 engine has maintained an excellent record and reputation for reliability. Pratt & Whitney data for the PT6A-67B/D recorded 1 275 600 hr in fleet operations in 2005. There were 10 in-flight shutdowns for a rate of 0.008per 1 000 hr."

Works out to 127,560 hours MTBF.

There are about 1500 PC-12s in existence. At the above rate, and assuming 300 hours flown each year by each PC-12, there are about 3 PC-12 engine failures/year. Seems plausible.

Mike C.

Ours was a -21 P&W (right engine) on a C90A. Catastrophic failure in core. No loss of containment. Happened about 7 or 8 minutes after takeoff during climbout. Daytime in good weather. A SE TB might have been able to turn around and return to the departure field. Obviously the twin made it back no problem.

No. Especially on a twin (the -67D is used on the Beech 1900 et al) an in-flight shutdown is not an engine failure. There are lots of circumstances (eg., gauge problem) where on a twin it would be shut down now, sort it out later, on a single it'd be kept running, divert and land. So this metric overstates the engine failure probability of the PC-12 while P&W's preferred "just the engine" metric understates it. The number we care about is somewhere in the middle.

There's a great many semantic arguments to be made here. What is a failure? What parts are considered part of the engine? Do you include non engine reasons for engine shutdown? And so on.

But in the end, if you don't have thrust, that is a failure in so far as what the pilot has to deal with, regardless of cause. The pilot has to handle it no different if the failure is from a turbine wheel failing or an airframe fuel valve. They won't be thinking about whether this failure "counts" or not in the stats, but instead how they have to deal with the loss of thrust.

Mike C.

intentional inflight shut downs can turn into failures pretty quick.
I always thought of inflight shut downs as polite speak for the engine packing it in, and shutting itself down. But yes, I guess that includes intentional shut downs for say a chip light maybe and nowhere near to land, or fuel pump failing so you feather it rather than wait for the surprise.

A shutdown or a failure yield the same result. It doesn't matter what you call it. The point is once it stops producing thrust you can call it whatever you want. Perhaps you'd be reluctant to do the "precautionary shutdown" in a SETP, but if it's gone, it's gone.

There are a lot of things worth losing sleep over.... The reliability of the PT6 is definitely not one of them.