Interesting. I have not used the dashboard itself, but I downloaded the xlsx database and started playing with the numbers to get some interesting data…
The database currently runs over 11 years, but I guess it is updated regularly from 2012 so it depends on when you actually download it.
I also downloaded the FAA registry database at the start of the period to have a reference for fleet size.
For example, since I fly a P210 I looked up the number of accidents in the database and I found the following numbers (I also downloaded the FAA registry database and made some utilization assumptions so I could gauge rates):
I then decided to run the numbers for the “competitor”, more modern airplane, the PA-46, and found the following
The fatal rate for pistons on both is similar (1.7 and 1.6) and slightly above the classical 1/100000FH. Of course the actual rate depends heavily on utilization assumptions and if you assume , for example, 150 hrs per year then the rate will be almost exactly 1/100000. I think utilization is closer to 100/year and since these are travelling machines maybe flying over 100FH/yr in typical private use but also typically a portion of the fleet will be grounded at any point in time.
The PA46 turboprops do significantly better.
However the non-fatal rate is much higher in the P210 with engine failures featuring prominently, so I decided to go through the 12 engine failure reports in the period and took the following notes:
I am still pondering what conclusions to draw…
[I corrected some errors in my numbers]
The causes are well distributed which one would expect for proven technology.
That turboprops do better is what one would expect with higher performance alone, and by extension that is a major factor in airline jet safety.
What happened to the PA46 engine failure rate, especially prior to the change from Conti to Lyco? We don’t seem to have a standalone thread here but in one simple survey 10% of owners reported an engine failure. Obviously a reporting bias there but still… And a very high rate of cracked cylinders, with one pilot telling me they changed theirs at ~700hrs or some such.
Peter wrote:
What happened to the PA46 engine failure rate, especially prior to the change from Conti to Lyco? We don’t seem to have a standalone thread here but in one simple survey 10% of owners reported an engine failure. Obviously a reporting bias there but still… And a very high rate of cracked cylinders, with one pilot telling me they changed theirs at ~700hrs or some such.
Not that fast! I don’t think we can fairly draw conclusions like what you are searching for with this data. At least not as straightforward as it seems.
You raise interesting questions and in trying to find the Conti&Lyco split that you mentioned I discovered some major flaws with my data above. As with most complex matters, there is devil in the details… resulting in the following corrected and amplified results:
Before you draw conclusions, let me explain what happened so you understand where I am coming from.
First of all, a big part of the PA46 piston fleet was converted to the Jetprop variant with PT-6 turboprop engines. I don’t know the exact number but it is at least around 300…if someone knows the number in 2013 and how many produced yearly in the observed period plse chime in.
Second, the FAA registry has 25 PA-46 type variants (clerical rather than technical, things like PA46 vs PA-46…) .
Third, none of those variants capture the Jetprop conversion, or sometimes even the correct type variant, making it difficult to know exactly how many and which aircraft were or were not converted.
Fourth, the NTSB database is inconsistent in this respect with both the registry and the actual type, with turboprops as well as pistons and Jetprops being registered, for example, as PA46 in the “aircraft model” field of the database.
Last, in my first analysis one of the 25 registry “variants” had got lost in the numbers…
If you want to make it even more interesting for the investigator all you need is the one or two owners who decided to buy a new M500 after the crash of his Mirage and decided he would use the same registration marks….
So I have spent some time re-running the numbers and trying to obtain something useful. First I assumed 300 of the pistons are Jetprops (again if someone knows better plse let us know).
Second, I verified each one of the accident aircraft to identify the exact engine type.
In this way I can obtain reasonable accident rates for each engine type.
Last there is the assumption that utilization is the same for all engine types. It may or may not be the case, as perhaps turboprops tend to fly more than pistosn in this particular fleet?
Last warning: do not let the PA-46 fatal rate of 2/100000FH lead you into thinking the Contis are disastrous vs the Lycos. Once you split hairs so thinly, the statistical significance gets low and one single fewer COnti fatal accidents and one single more Lycos would make the rates identical.
Perhaps more interesting than the “per-engine” split is what is causing the fatal accidents and no, it is not the engines in any single case: unsurprisingly it is LOC, CFIT and WS/TS.
Even more curious is that all of the three PA-46 engine failure events that made it into the DB were PA-46-500TP’s : not a single piston.
