Menu Sign In Contact FAQ
Banner
Welcome to our forums

Glass cockpit aircraft more likely to have accidents which are fatal?

When I was in flight school during our ifr training we would start with a lot of classroom theory. They then gave us a stack of paper roughly 2cm high and featuring nothing but tables/numbers and had us calculate qdr/qdm interceptions etc… If you did all it was several thousand. “See you in a week” they said. After that we spent a few months in a screen-less plotting printer equipped FNPT based on a C310 where we did the most crazy interceptions and holdings one could imagine. The symmetrical cloverleaf was one of the simpler ones. We were so drilled that navigating in the real plane (calculating entries, interceptions and wind corrections) and flying it was not a problem on nothig but round dials!

Fast forward 10 years I now have a huge map in front of me and don’t know any of that stuff as I used too.

I think safety is about effort. Back then I was super safe, because I put in so much effort to know where I was without a map. Flight preparation was very detailed. I knew all the grid MORAs, all the MEAs, I knew all the step down altitudes etc…

If you put in the effort that is required for a non glass cockpit before a flight in a plane with modern avionics you have a safety bonus.

If you compensate effort with all the information provided by those screens I can see where that study makes sense.

always learning
LO__, Austria

The NTSB database can be searched for year, make…I just did one for 2015…

It shows 18 Cirrus crashes, 5 fatal. Only 2 were SR22T, both non fatal. http://www.ntsb.gov/_layouts/ntsb.aviation/Results.aspx?queryId=608effb6-bd78-4738-bcbb-ece96fdf5b6a

By contrast there were 206 Piper crashes, 52 fatal.

Not clear what a ratio of crashes to fatalities proves, but it does make sense to compare the number of active aircraft flying in a given brand to how many accidents happen in a given year. There are about 4,500 Pipers flying in the USA, 6,000 Cirrus. Yet 10x more Piper crashes in 2015.

Last Edited by USFlyer at 23 Dec 22:50

USFlyer wrote:

Cirrus fixed the problem by establishing strict training regimen for anyone getting into one.

That is the whole point of the studies: Only training can reduce fatality rates, be it due to avionics or aerodynamics. (I don’t know why many people instantly jump into apologetics-mode, once an objective view of their toys gets published… whatever).

I think the sense in the use of glass cockpit systems depends substantially on the task set to do. In a basic Day-VFR LSA, in my view any PFD is just a toy and the tape-ASI is inferior to a round gauge (digital or not doesn’t matter) because of the aircrafts low inertia. During flight testing of an LSA I refered much more to the conventional ASI than to the tape od the Dynon SkyView, especially in gusty winds. This changes, of course, the more stable an aircraft is designed. The MFD instead is very good in VFR and IFR use, modern engine monitoring systems can provide more awareness of engine parameter deviation and general operation.

But other than the standard steam gauges, they all do require more time to get used to the specific device and more training to know the units, because of some unconventional (or creative) menus and specialities. This makes glass cockpits unattractive for renters of standard bread and butter spam cans. Every pilot is pretty much proficient on a basic six pack and I know plenty people who would rather take the standard 6-Pack instead of a G1000. Not that they won’t be able to fly or learn, but they don’t see the added vylue to their kind of flying, and I get that. Our Aero Club did go the intermediate step and install an “MFD-light” for navigation: An iPad Mini with SkyDemon.

Last Edited by mh at 23 Dec 23:48
mh
Aufwind GmbH
EKPB, Germany

USFlyer wrote:

There are about 4,500 Pipers flying in the USA, 6,000 Cirrus. Yet 10x more Piper crashes in 2015.

That can’t be right.

mh
Aufwind GmbH
EKPB, Germany

USFlyer wrote:

By contrast there were 206 Piper crashes, 52 fatal.

The “death rate” in a Cirrus is 28% and it’s 25% in a Piper. Statistically speaking the chute doesn’t help much.

The elephant is the circulation
ENVA ENOP ENMO, Norway

mh, would have to trawl through all the Piper types, but the PA18 production run alone was 7,500, the PA28 was 32,000, the PA23 over 5,000, you have from the J2 through the PA46, so possibly around 100,000 produced in all the types? what the active fleet might be is difficult to estimate, 60,000? They are relatively un complicated with limited ADs, and for example around 2,000 PA18s active in Alaska where they don’t lead a sheltered existence with a relatively high accident rate reflecting the hostile weather and off airport operations.

I don’t know if the Piper fleet flies ten times the hours of the Cirrus fleet, but it might. To ‘control’ the statistics, you might compare the Cirrus to the Saratoga or the PA46, but they are still different vintages/types (Saratoga mainly produced in the 1980’s, the PA46 a high altitude, more complex aircraft), and I expect these Pipers to be average, or above average on the fatal rate, with the rate creeping up as the fleet gets older. Controlling for a specific year of production and type of operation would yield too small a set of data to be meaningful.

I would give the Cirrus the benefit of the doubt that it is probably one of the safer cross country high performance SEPs ever built, even though not so long ago (up to 2012) the report card was average to worse than average. The poissonic nature of the accident curve means the last three years may not yet be meaningful, statistically.

Which reminds me of Zhou-en-lai responding drily at the Korean peace talks in Paris, what did he think of the French Revolution: ‘too early to tell’. I like the quote as this period prompted the US to ship several thousand PA18s to Europe to help re build the Western European air forces from the bottom up, including little old L18C 51-1555 sent to the French ALAT.

Oxford (EGTK), United Kingdom

In the last ten years there have been 92 Cirrus fatal accidents in the NTSB database, out of a total of 215. A fatal rate of around 30% to 40% per accident does not seem to out of line given the type of operations and kinetic energy of this GA aircraft (it was 33% in 2015), and as LeSving points out the CAPS may have less statistical influence on the fatal accident rate, when compared to similar aircraft without CAPS, than we might think a priori. The 33% rate seems a fair estimate, at the lower end of the historical average.

A statistician then might calculate what is the Poisson probability that the Cirrus fleet accident rate stays at the 2015 rate of 15 accidents in the year, compared to a historical average of 20 plus, before concluding that it is the safest of its type.

In any event this is hair splitting. Compared to Part 121 operations, GA has such a higher accident probability (in the order of 50 to 100 times higher), that whether a GA type experiences 1 or 3 fatals per 100,000 hours is not that relevant. Survivability, and this is linked to type of operation and aircraft type, is a factor, and here I would champion the PA28 or DA40 or C172 as superior.

Oxford (EGTK), United Kingdom

USFlyer wrote:

There are about 4,500 Pipers flying in the USA

Here we go again… @USFlyer, could you please be more careful with your claims.

ESKC (Uppsala/Sundbro), Sweden

RobertL18C wrote:

mh, would have to trawl through all the Piper types, but the PA18 production run alone was 7,500, the PA28 was 32,000, the PA23 over 5,000, you have from the J2 through the PA46, so possibly around 100,000 produced in all the types? what the active fleet might be is difficult to estimate, 60,000?

Thanks, that’s what I expected and why I didn’t believe USFlyers “facts” for a moment.

@USFlyer: Why?

mh
Aufwind GmbH
EKPB, Germany

There’s more than 10.000 Super Cubs built, almost 5000 Warriors, more than 5000 Senecas … and that’s only 3 types Piper made/makes. . … other examples: 5000 Aztecs were built, 20.000 J3 Cubs… So “4.500” is of course completely wrong

The numbers I know from COPA are: Cirrus fatal accident rate for the past 36 months: 0.63 for 100.000 hours of flying. (In the past 36 months, there have been 17 fatal accidents and approximately 2,700,000 flying hours for a rate of 0.63 fatal accident per 100,000 hours of flying time).
Numbers for the GA fleet: 1.05 overall, and 2.38 for Personal & Business flying

Last Edited by Flyer59 at 24 Dec 11:28
Sign in to add your message

Back to Top