IFSD Rate for current technology helicopters

[PT6ER]

I'm trying to get some reasonable ballpark numbers for expected IFSD rates for todays current crop of twin engine helicopters.

From a regulatory point of view, the only mention I can find regards engine reliability is for ETOPS rated planks - not really useful I 'm afraid.

An internet search has not yielded much useful data so input from you grand assembly of rotary wing people would be much appreciated.

Thanks in advance for any gems of information.

Regards

Sixer

[Ian Corrigible]

Honeywell was quoting 0.22/100,000 hours at the show for both the current LTS101 and the HTS900 (target IFSD rate at EIS).

I/C

[Shawn Coyle]

You need to be careful with this - for Part 27 helicopters in the US and Canada (can't speak for Europe) if you have an engine failure and land safely without extensive damage (or words to that effect), you don't need to report the failure.
Hard to believe, but it really does skew the situation badly.
The UK CAA may have some data on this - I seem to recall a quote that one French engine manufacturer had an inflight failure rate of 5% (yes five percent) of that of a popular US engine manufacturer (but then again, that took into account the total life of the US egine, and the French engine was introduced much later in the game).
Be careful with statistics....

[NickLappos]

Shawn,
The IFSD rate s a maintenance number, not an accident number, so your worry about whether it would be recorded is not a problem. The basic credibility of the manufacturer is at risk if they bury data.

For the GE CT7/T700 the IFSD as recorded by the US Army is 1:200,000 engine hours for all causes (engine, fuel, pilot, etc) and 1:500,000 hours for engine causes.

I don't have any other data, but will look.

As I have posted in the past, for twins these engine failure rates are so low that pilot and maintenance errors completely dominate accident causes, making it important that we work causes, not opinions, to make our accidents vanish.

[212man]

PT6ER,
are you talking about IFSD or flame out/seizure? The figures will be somewhat different. IFSD can result from, chips, low oil pressure, governor problems etc, that are pilot actioned. other causes (as Nick says, fuel starvation etc) are unexpected.

I doubt very much that you will find meaningful statistics of all in flight shutdowns, as the pilot actioned ones will simply result in internal company incident reports in most cases.

[SASless]

212man....

How does the vaunted EC-155 at Shell/Nigeria rate in this catagory?

[NickLappos]

212man's doubts aside, the data is there, just ask the engine manufacturers. If they don't provide the data, just don't examine their engines, and exclude them from consideration in the next purchase.

Funny how that works.

[212man]

SASless,
vaunted would be your expression. You would not expect me to discuss technical detail of our operations (or would assume you don't). What I can say is that if considering IFSD from all causes, there are events that are in line with both Nick and my comments i.e. there's more to ending up on one engine than a simple failure, whatever its cause.

Nick, on reflection, you are probably corrrect that engine manufacturers will have details of most events that result in OEI flight, however when considering the statistics that are used for exposure period analysis I think it is true to say that many of those events will not be included.

[NickLappos]

212man,

The authorities simply write the manufacturer and ask for the data. Then they calculate using it to determine the exposure risk, a very neat and scientific way to capture safety. For recalcitrant engine manufacturers, the threat of not getting approved is enough to move the mountains, I think.

JimL, can you help here?

[JimL]

PT6ER,

The reason that you are not receiving much feedback on this is because it is a contentious issue and manufacturers do not wish to exposure their numbers to the public gaze. At high level seminars, which are attended by executives of the operators and manufacturers, reliability figures are routinely presented. At one particular conference last year Pratt, GE and Turbomeca all presented their yearly reliability figures (at least for engine families) and just before Christmas there was an ICAO committee meeting during which the reliability figures for the Allison were discussed.

One of the problems you have already seen is a use of terminology; IFSD is more applicable to ETOPS than for helicopters. For helicopters (and SEIMC in fixed wing) we are more concerned with Powerloss (and for the calculation of exposure Sudden Powerloss) – that these definitions do not exist in Airworthiness is evidenced by their presence in various operational texts (ICAO Annex 6 for SEIMC, JAR-OPS 3 for exposure etc.)

Manufacturers tend to base their statistics on core engine failures. (Core engine failures do not take account of maintenance errors, operational related failures and airframe associated failures (fuel systems, particle separator failures etc.).) For the purpose of the calculation and approval for exposure or SEIMC all engine failures have to be considered.

(While we are on the subject of SEIMC, your namesake engine which powers the Cessna Caravan has a tremendous amount of collected data as it is the primary engine for SEIMC on fixed wing (where the ICAO required reliability rate is 1:100 000 per flight hour). Evidence appears to indicate that the PT6 can achieve reliability rates better than that figure - this engine in the Cessna Caravan is well documented as UPS have been collecting data and monitoring it (using UMS) since its introduction.)

Depending for what the data is being used, it is sometimes parsed into that which is of concern to the application and that which is not. Using the example of approval for exposure - where only the take-off and landing events are being considered - events like chip warnings, which lead to precautionary landings in singles and IFSD in twins might not be taken into consideration. The reason; exposure for the event rarely extends beyond 9 seconds for twins and 18 seconds for singles (which for Europe puts the event within their safety target of 5x10**-8 – when the reliability of 1:100 000 is achieved)

My experience (for what it is worth) indicates that a (ball park) range of 0.3 to 1.3 core engine failures per 100 000 flight hours is about right (with the mean being about 0.6). Engines rarely improve on the low figure and it is unlikely that further substantial improvements are possible. With regard to the actual engine failure rate (for all causes), that does rather depend on the application (aerial work being worse than CAT for a number of reasons).

If you were to search a good database which contains such events (like the UK CAA mandatory occurrence reports), you would find engine failure rates for all causes is in the range of 1 to 10 failures per 100 000 flight hours. The number of these could be reduced by the use of Usage Monitoring Systems (UMS), improved inspection, a reduction in maintenance errors and reduction in operational related failures (abuse of engines and FOD).

In the recent discussions in ICAO, we agreed that the reliability rate for engines to be used in SEIMC in helicopters should be the same as fixed wing - i.e. 1:100 000 per flight hour for Powerloss events (the manufacturers did not object to this).

Jim