Menu Sign In Contact FAQ
Banner
Welcome to our forums

Do we learn from GA accident reports?

I hit a fence post after aborting an up-hill take-off in long grass at near max AUW. That post broke, the wires pulled the nearest posts, and the Jodel stopped with slight wing damage and a prop tip broken. I flew it out downwind and downhill after a day’s work by an engineer. It would need a lot of safety design to give a pusher aircraft the protection of an engine in front.
Maoraigh
EGPE, United Kingdom

As I recall, the NTSB had about 100 reports of LongEZ or VariEZ accidents if you search by type. If you discounted the accidents where the aircraft landed in a car park or road the fatality rate was bad. Obviously some people will still make it. OK here goes:

from http://www.ntsb.gov/aviationquery/ :

LONGEZ 4/12
Long EZ 14/49
varieze 20/70
vari ez 2/8

40 fatal accidents of a total of 149. What I did in the past, but don’t have time to do now, is to read __all __ the summaries and look at the number of those accidents in which the aircraft happened to be in the circuit and land on a prepared runway, road etc… Or took off, had a power failure, landed then ran off the end of the available runway. Even the successful off-airport landings were generally on flat American cornfields and not on Welsh mountainsides. A lot of the nonfatal accidents were e.g. hand-starting aircraft running away, clipping wings on objects etc..

One thing that I may or may not have done (I don’t quite recall) was to count things like hitting trees after a power failure as an off-airport landing. Does it matter? Even here, you might expect to do better in a slower aircraft that lets you stall fully before hitting the canopy, or where you are more likely to find an usable landing area.

If you do an equivalent search for piper J3, about 54 of 449 of the accident reports record a fatality, and a higher proportion of these involved drink, drugs, suspected suicide, mountain flying etc. rather than engine-out scenarios. And I’d wager that the a lot of cub engine failures never get reported to the NTSB anyhow because no damage occurs and you can fly out with no fuss, after you’ve put some fuel in the tanks.

There are 4154 accidents for PA28s, of which 893 resulted in fatalities, so admittedly not a great deal worse than the EZs.

As ever, a high proportion of accidents in all types involved stupidity, low level aerobatics, tree-skimming, terrain in IMC etc. You can certainly improve your chances considerably simply by not being an idiot.

Last Edited by kwlf at 04 May 16:41

before coming to the conclusion that forced landings in them on rugged terrain (i.e. where I like to fly) should basically not be considered survivable.

There have been 4 reported forced landings with Rutans canards here in Norway. In 2 of those the aircraft were totally damaged, in one the aircraft turned upside down with some damage, and one landed on an airfield but run off the runway with some minor damage. In all accidents the pilot could walk away, except for one where he had some serious leg damage. The passenger in that one could walk away though.

The elephant is the circulation
ENVA ENOP ENMO, Norway

before coming to the conclusion that forced landings in them on rugged terrain (i.e. where I like to fly) should basically not be considered survivable

I personally know a real live pilot who climbed to 18000ft in a canard homebuilt, passed out through hypoxia, and woke up in a forest on the side of a mountain. This was in the Alpine region, but may not have made any accident reports. The cockpit stayed in one piece despite what must have been a full power (way past Vne) impact. He had his legs rebuilt.

So maybe it depends on how well you build them?

Administrator
Shoreham EGKA, United Kingdom

Those are slightly different: ‘That worst day’ is written in the third person; ‘ILAFFT’ is first person. Both are useful and in my view, very healthy. The ‘worst day’ articles are in my view very nicely written by someone who clearly knows what they’re talking about.

A while ago I was quite interested in road safety and design, and it’s odd to consider that it seems far easier to paint a picture of a ‘typical’ general aviation accident than a typical road traffic accident. There’s much more raw information publicly available. I did learn that a fairly common pedestrian-crossing accident involves a car stopping, then being rear-ended pushing it forward onto the crossing and hitting the pedestrian. Part of the reason behind advanced-stop lines which everyone seems to ignore.

I’m not sure where I put the raw figures, but a while ago I worked my way through lots of NTSB accident reports in order to decide what sort of aircraft I would like to fly (had an obsession with the Rutan canards) before coming to the conclusion that forced landings in them on rugged terrain (i.e. where I like to fly) should basically not be considered survivable. Of course, I ended up flying Cessnas and Pipers like everybody else but I ended up buying plans for a reasonably conventional non-canard homebuilt.

Last Edited by kwlf at 04 May 10:49

I wonder who writes these?

Maybe I will put my two or three of four (yet unwritten) ones to paper some day… Some magazines call them “I learnt about flying from that”.

EDDS - Stuttgart

Another form of accident report is the ‘That worst day’ format, where the idea is to write something pithy, readable and anonymised so that all the human factors can be discussed without any embarrassment.

I wonder who writes these? I like to think I’ve learned quite a lot from them.

The rumour in the UK is that the UK AAIB gets a great deal of “input” (not always input they have asked for) from the lawyers for the various parties involved, and the final report comes out only when all the lawyers for all the parties have agreed to not sue them.

I agree about “hearsay” but it could be to some extent verified. For example if you have a pilot whose attitude is “I always fly” (I have known such types, and not all of them are alive now) it would be fairly easy to look up his logbook entries versus archived wx records.

OTOH the AAIB might argue their job is not pilot training, but merely reporting the known facts.

It would sometimes be really handy to publish the name of the maintenance company that didn’t open the oil filter, but the AAIB would prob99 get sued by them because I don’t think checking for metal in the filter is a requirement under Part M.

Last Edited by Peter at 03 May 19:31
Administrator
Shoreham EGKA, United Kingdom

Would they be able to print it? Legally yes, morally it would be questionable.

The problem is: It is “hearsay” information about a deceased person who can not defend himself any more. Therefore it is better left out of the report, even if it would provide useful information. But I have read a lot of reports where previous accidents and incidents of the same pilot were cited (e.g. Crossair 3597, which is not a GA accident but can happen in the GA environment as well).

EDDS - Stuttgart

I have specific issues with some UK and some French accident reports.

One UK AAIB one, G-AVRP, had a fair chunk of its camshaft in the oil filter, which was obviously never opened by the “maintenance” company. Nobody noticed, probably for years. Action taken? You are kidding… I can see why now; if they were to stop this, they would shut down half the UK “maintenance” business.

The one from the BEA on N2195B was similarly useless and obviously inaccurate. No background on the pilot, his lack of carrying oxygen, his problems working out Eurocontrol routes. And some other issues which could have been established with some quick local enquiries. The attitude is… flying is dangerous, people will get killed. Learning value = 0 unless you already know how to do that flight properly, or stay on the ground.

I try to learn from these reports but most of the time I struggle, and whenever I get an opportunity to speak to somebody who knew the pilot personally I hear a whole new piece of the story which obviously was not in the report. One “highly experienced” pilot of a turboprop, fatal crash, was a well known cowboy who a friend of mine (well qualified on type) would not fly with anymore and his demise was widely predicted. I suppose one can’t expect the investigators to do research beyond a certain point but if they visited the pilot’s home base and asked around they would get useful info. Would they be able to print it? Legally yes, morally it would be questionable. Yet this sort of stuff is a crucial part of most crashes where human factors were involved.

Administrator
Shoreham EGKA, United Kingdom
22 Posts
Sign in to add your message

Back to Top