Discussion on RAAus Accident & Incident Summaries

[Admin]

Sorry Guys but I "stuffed up" with trying to split the thread to create a separate threads that just lists the entries and one for discussion...somehow all the discussion posts merged together...sorry.

 

Anyway, the list is at:

 

http://www.recreationalflying.com/threads/raaus-accident-incident-summaries.65804/

 

and I will try and keep it updated

 

It is great that we are finally getting access to some accident statistical data, may it continue....Thanks Ian..........Maj...

This is great!!!! I second what Maj has said. Should have been done years ago.

 

If I can be so bold - who is collecting these numbers, and are they collecting anything else? Weather conditions, fuel type, registration, recency, engine type etc....the more data the better. If this can be collected for every incident, there can be some real progress (in a couple of years!).

 

No complaints about the posting of information, but where is the investigation? All we have in most instances is a list of accidents with no details on what caused the failure (as distinct from the sequelae of the failure ie., engine stops or loss of full power, crosswind at the time vs designed crosswind capability, etc, noted in the summary).

 

What is needed is a published "root cause analysis" around the whole incident, similar to investigations done on major airlines if we are to learn anything and a relevant recommendation out of the incident report. Only then, with the cause of the incident known, can you do anything about it with any degree of confidence.

 

Simplest of investigation techniques is the "5 whys". Simply asking "Why?" until you cannot revel any more information.

 

---------------

 

Example

 

Catastrophic engine failure in flight.

 

Why? Engine valve jammed in cylinder and caused crankshaft to stop

 

Why? Exhaust valve broke off from stem near collet (Metal fatigue)

 

Why? Engine running hot for lengthy periods causing metal fatigue in hot elements of engine.

 

Why? Cooling shroud found to be partly collapsed impeding cooling airflow to engine.

 

Why? Cooling shroud fouled by upper part of fuselage engine cowl last time cover was placed over engine.

 

OUTCOME: Care is to be taken in replacing the engine cowl to avoid bending cooling shroud. Visual check of shroud position to be made in preflight checks.

 

--------------

 

Doing something like this can reveal other issues that are at play in an incident as in this hypothetical (but entirely plausible) example

 

Damkia, Some incidences are somewhat self explanatory. IE: engine failure on a certain type leading to emerg landing and damage. Or student pilot solo, having to land in an unexpected cross wind. As always of course, more info is best, but even a little is better than none at all !...................Cheers Maj...

 

Damkia, Some incidences are somewhat self explanatory. IE: engine failure on a certain type leading to emerg landing and damage. Or student pilot solo, having to land in an unexpected cross wind. As always of course, more info is best, but even a little is better than none at all !...................Cheers Maj...

They don't explain the cause of the engine failure, therefore do not explain the incident. see my amended post above... No engine failure should be treated as "normal".

Well with a certain engine you could take your pick of several causes that are common !......................Maj...

 

Well with a certain engine you could take your pick of several causes that are common !......................Maj...

........and collecting reliable data of what the incidence of each type of failure is may lead to an investigation of ways to mitigate the risk of that mode of failure.

Too may people are dropping out of the sky. Too little information is made available about WHY.

 

These look like a compilation of the Pilot Notes from the magazine, plus news reports, and I agree this is very helpful, particularly since the incident details can be quickly found again, and referred to, or repeated issues become undeniable, and "bashing" claims become irrelevant.

 

Where any information comes from Pilot Notes, I agree with Bandit and Damkia that the information often is not very useful, and it seems a waste to me that there's definitely a reporting system in place, but it doesn't gather just a few more pieces of information which would lead to useful training lessons.

 

For example when I was researching engines, going through hundreds of reports, one repeating incident jumped out from the pages, and that was a loss of directional control after touchdown, with the aircraft usually finishing up in a ditch, damaging one wing, or flipping head over heels, none of the incidents producing a fatality, but costing a lot in repairs. For the most part the report simply said "a FlyingFox lost control after touchdown and flipped after hitting a ditch" - no investigation or querying of the pilot, no advice on how to prevent this etc.

 

That could be improved enormously with just a little more effort.

 

Fatalities and serious injuries will continue to be a problem for reasons we've discussed several times, but ATSB have hinted they want to investigate "all" accidents, they have actually started on RA engines, and it seems the Coalition will give them extra budget (while not promising the same to CASA), so that level of crash too may provide better lessons for us in the future.

 

........ but ATSB have hinted they want to investigate "all" accidents, they have actually started on RA engines, and it seems the Coalition will give them extra budget (while not promising the same to CASA), so that level of crash too may provide better lessons for us in the future.

I really hope that the ATSB is geven the money to do this, it would benefit RA Aus a great deal.

 

Wether it be Labor, coalition or Bob Katter, who cares, as long as we get the money !...........nobody including CASA, ATSB or whoever can knock us for having a bad record without having supplied us with reliable and meaningfull accident reporting.

 

Still a thorn in my side that is top of my list if I end up on the board..................Maj...

 

There is no column for "powerplant". Some aircraft could have about 5 possibilities. Engine failure tells little. Many engine failures on TO are caused by fuel selector foul up, carb heat left on etc.

 

Having ATSB investigate all. (They don't investigate fatal aerobatic ) will cost a bit. I have long given up wanting things for no cost as there is always a cost.

 

All you blokes out there who hate a certain engine, why don't you just be content to not fly one and give it a break and let others make their own choices? It would be a bit ordinary if they were grounded. And May I say If that sort of thing happens it will be THEM this week and perhaps YOU next week.

 

You WON'T get a full factual CAUSE for all failures or crashes. A fatal Jabiru (RARE, in the USA)done by the FAA got facts about the Carburetter totally wrong. The Carburetter WAS the cause but their analysis of it was incorrect. You just don't have the expertise in some of these areas any more. IF you were in a department and had to be involved with "those little Planes" (or a worse description) it would be taken as a "punishment". Hardly something job wise you could brag about at the Café Latte shoppe to your twenty somethings. Nev

 

Go and start the Indian, have a sniff of the excitement and you won't get so depressed about all this.

 

I forget where I read it now, but there has been a definite shift on ATSB behalf where they are talking about investigating all aircraft, and I took that to mean aerobatic and RA.

 

Facto, I don't think any of us 'hate' a certain engine. We may choose not to fly with them because we are used to more reliability, or we may simply be really over the fact that after 20 odd years, the factory has not taken the trouble, or spent the time or money, or given a damn about making them a safer engine.

 

We may also be very over at this point that the constant unnecessary mechanical failures, have, and continue to bring the overall RAA safety picture into disrepute...........................Maj......

 

All I'm saying is don' t bring on something that won't necessarily fix the problem. When you invite more inspection and overview don't be surprised if it bites you somewhere else. I believe in doing real thing s to get a FIX. I don't own a Jabiru ,but Plenty do here, and I don't think the constant harping and a lot of non constructive commentary is fair to the owners and operators who have a good run with these engines. The two main incidents that brought this on were not related to Jabiru at all. I am aware of a few others that predated them and were probably relevant too and the same applies. Get it into perspective. There are plenty of engines flying that are more likely to fail than a Jabiru. Beware of what you may find and the restrictions that may be enforced as a result, particularly on the qualifications to do this or that. We are not running an airline here and the less red tape the better. for the enjoyment of it . A bit more BS and it isn't worth doing..Nev

 

If you get the statistics correct, then the results are the results FH, emotion stays out of it, and issues can be addressed. The transport industry does this very well, with very little fallout.

 

You don't fly a U/L so you won't be affected. The same thinking brought us the ASIC. beaurocrats given statistics ( and you can interpret them anyway you like) will act on them. The safest U/L movement is a grounded one. You can go and buy two bottles of scotch and die by drinking them Anyone can do it.? No laws broken. Nev

 

What is needed is better investigations and complete data collection, not necessarily more data... ( = "Data validation")

 

Pointless data ends up having as much meaning as all these footy stats that infiltrate the TV broadcasts these days. Choose what you want to collect, how you want to collect it, and what you do with it after you have collected it. If you can't justify a basis for a point of data collection, then it is of little use.

 

I don't think it would take much to design an App to standardise the collection of data so that at least all people involved in making an incident report (even a self report) can contribute data in a common format. This would go some way to formalising a standard set of data that could then be turned into relevant information to draw conclusions and recommendations from. (Google the difference between "data" and "information" if you need to)

 

There is no such thing as bad data. One should collect as much data as one can so that in the future with a refinement of perspective one can do some interesting and quite valuable analysis.

 

If the data is of little interest or confusing at a given time then filter the raw data to provide a manageable perspective.

 

Some data might not be of interest, value or comprehensible to the viewer at that instance but will be a real gem to another person at a different time.

 

No ... but data can be manipulated and then it is called 'statistics'.

 

Winson Churchill once said ..."... there are lies, there are damned lies ... and there are statistics".

 

IMHO, statistics are data manipulated to prove what ever side of the argument you support. LOL.

 

No ... but data can be manipulated and then it is called 'statistics'.Winson Churchill once said ..."... there are lies, there are damned lies ... and there are statistics".

 

IMHO, statistics are data manipulated to prove what ever side of the argument you support. LOL.

and Josef Stalin allegedly said "I know about these statisticians, they can melt sh1t into bullets"

Why oh why is the world such a flat earth - because the naysayers like it that way.

 

Like it or not statistics governs our lives from the quality control of light bulbs to predicting the next earthquake - crap lightbulbs do get through but the majority do work, they miss out on some big earthquakes but if they followed statistical advise they wouldn't have built in Christchurch or Fucushima.

 

Please have a read of http://en.wikipedia.org/wiki/Epidemiology and tell me if you still think statistics is just for old cobblers spreaders.

 

No ... but data can be manipulated and then it is called 'statistics'.Winson Churchill once said ..."... there are lies, there are damned lies ... and there are statistics".

 

IMHO, statistics are data manipulated to prove what ever side of the argument you support. LOL.

Churchill was a politician, after all. They are usually short on truth and long on rhetoric. You have to remember that Churchill didn't win the war but depended upon a whole stack of people who were, statistically, going to die, be maimed or whatever. Statistics for them was a fact of life as it was for those who designed all the aero engines - if we soup it up, statistically, will it make it across the channel and back without a major overhaul or failure.

 

and Josef Stalin allegedly said "I know about these statisticians, they can melt sh1t into bullets"

The only statistics Jo cared about were the number of enemies of the grand Soviet who could be murdered - and this included large numbers of dedicated soviet citizens. He was simply a murderer - perhaps on that day he decided to kill all statisticians because ... well he could.

I happen to agree with you Col, it was momentarily my cynical side coming out. Statistics used ethically are of great community value.

 

And the data needs to be ethically collected in as much depth as possible and certainly to collect the serial numbers. And maybe the L2/maintainer/LAME so that one can make statistical analysis including cluster analysis.

 

What matters are the statistics we can use to save lives and reduce injury.

 

If you collect data it's useless unless its set up so you can drill down and get meaningful results.

 

In this case there's absolutely no point in the academics running off on their usual tangents, because we simply have no data where we need it most, and that's on the reasons people we actually know by name haven't been coming home.

 

Well done for making a start by publishing this listing.

 

It may not be complete, it may not give all the facts but it is better the no data at all - it is a starting point and hopefully interested parties can contribute by adding what they know too.

 

Can any of our ASRA gyro colleagues make a contribution of a similar kind into the public domain - my own listing is far from complete.

 

Regards

 

Steve

 

http://gyrocopteraccidentsinaustralia.blogspot.co.uk/

 

What matters are the statistics we can use to save lives and reduce injury.If you collect data it's useless unless its set up so you can drill down and get meaningful results.

In this case there's absolutely no point in the academics running off on their usual tangents, because we simply have no data where we need it most, and that's on the reasons people we actually know by name haven't been coming home.

Academics can probably advise where the holes in the data is. To paraphrase a good friend of the deputy sheriff - you can do stuff with the known knowns, you might be able to second guess the known unknowns but the unknown unknowns will probably kill you. I can't remember what it was about the unknown knowns but that was probably where the data gatherer thought it was irrelevant or just forgot to mention it and that is what really kills you.

 

"academics running off on their usual tangents" - What gratuitous claptrap.

 

Some great thoughts Col.. You are all fired up. Facts are good. Targeted incomplete or selectively intepreted, they are an instrument of oppression .Nev

 

What is the best way for this site to use the data in a way that will assist site users...assist in promoting safety through awareness by having the information readily available? Each one of these no doubt gets discussed in threads just after they happen

 

For example:

 

- do we have a thread that contains each one as a post and is updated as each one gets added but locked to just the Admin to add extra records...each post would have a link to a thread on the site?

 

- any other thoughts?

 

I think the chart as you have it, with a reference number in column 1, and only accessed by Admin.

 

Then if there's some discussion on, for example, nose overs people can start a thread and refer back to the reference number or numbers, and that clean data will aways be there, and not buried by comments going all over the place.

 

There is a lot of nonsense that comes out of academia - the trick is to have a good research design, no political axe to grind, and a complete and open reporting of the process followed. Unfortunately good research is tarnished with the brush of bad research, and the selective use of and reporting of results out of context. If you want examples, just read the news. You might find a statement in the news like "Researchers at the University of Blah Blah Blah have found that drinking coffee is actually good for you" when the actual published research article will say that there was a statistically significant difference between coffee drinkers and non-coffee drinkers on a measure of health and quality of life outcomes, although the effect size was very small.

 

What I will say is this - it is of little use to be looking at 5 years of data and saying to yourself "Hmm.....I wished we had recorded XYZ from each incident, as that would have been really useful....."

 

What is the best way for this site to use the data in a way that will assist site users.....

Here is another thought Ian. I don't know if there is an incident reporting mechanism for RA-Aus that is non-punitive (ie people won't get punished for reporting) but that would be a really useful thing. For every fatality, there could be half a dozen more accidents which only result in injury or less. There is probably plenty more where there are engine failures resulting in no damage or precautionary forced landings. On top of this, how many incidents occur where people end up in risky situations, come out the other side and nothing ever gets heard about it.

 

There is so much that could get collected, if you could get across a culture of voluntary reporting. You just need the mechanism (online form) and the awareness that it is everyone's responsibility to be involved. This is probably something RA-Aus should be doing rather than just recflying.com.

 

Research has found that there is a correlation between the incidence of fires, and the presence of firemen in the vicinity.

 

80% of fires in the Sydney region in 2012, had an average of 6.5 firemen within 298 metres of the fire centre.......

 

Here is another thought Ian. I don't know if there is an incident reporting mechanism for RA-Aus that is non-punitive (ie people won't get punished for reporting) but that would be a really useful thing.

Watch this space Bandit as I believe it is coming...and as always, talk to your local rep

 

What is the best way for this site to use the data in a way that will assist site users...assist in promoting safety through awareness by having the information readily available? Each one of these no doubt gets discussed in threads just after they happenFor example:

 

- do we have a thread that contains each one as a post and is updated as each one gets added but locked to just the Admin to add extra records...each post would have a link to a thread on the site?

 

- any other thoughts?

Would it be possible to set up a read only spreadsheet (visible to all), with write only by an "authorised spreadsheet administrators" (not necessarily Admin here, but someone with the knowledge of Excel and technical issues involved with incidents from written incident reports).

Reason for this is when the spreadsheet is exported for use elsewhere, the Excel search capabilities are more powerful than what is available to Word. Excel also allows for input data validation to minimise transcription and human errors in the spreadsheet.

 

Watch this space Bandit as I believe it is coming...and as always, talk to your local rep

I can't - not a member so I have no rep One day.....in the mean time, I weigh in because I do care about what happens in recreational aviation, regardless of what the rego numbers/letters are.

 

The RAAus has always "leaked like a Sieve". CASA have a way of reporting anonymously. This only suits certain situations as there are situations where you would be entitled to question your accuser. There is also the situation where you might be ostracised or suffer discriminatory treatment for doing it. There is work needed Here.... Nev

 

No ... but data can be manipulated and then it is called 'statistics'.Winson Churchill once said ..."... there are lies, there are damned lies ... and there are statistics".

 

IMHO, statistics are data manipulated to prove what ever side of the argument you support. LOL.

Think you'll find that quote was from Benjamin Disraeli, another famous British PM (1804-1881) - long before Churchills' famous WW2 statements. It's also often attributed to Mark Twain,(1835-1910), by our US friends, but he certainly liked and often used the phrase.

 

happy days,

 

 

[Bandit12]

Oh man.....it reads like we are all talking at once in a really loud pub

 

 

[turboplanner]

The chart is the part which important so no problem.

 

 

[coljones]

Thanks Ian, much appreciated

 

 

[fly_tornado]

you can create a spreadsheet under google docs that allows you to do all sorts of neat stuff, people can copy it if they want to tweak it

 

 

[damkia]

Given the news today of the safe emergency landing by Doug and his student, my "hypothetical" seems eerily relevant........

 

Here's hoping the investigation follows suit...

 

 

[Dafydd Llewellyn]

I agree most emphatically that accurate and complete data on accident causes is needed. Unfortunately it is more difficult to achieve this than one might expect. Different people tend to analyse causes differently. So any form of tabulated analytical data MUST be done in accordance with a well-defined (and public) set of criteria.

 

I'll give you an example: A little old lady trips over a loose carpet strip, and falls, breaking her hip. She pick up a viral infection in hospital, is given an antibiotic to which she has a violent allergic reaction, and dies. What was the cause of death?

 

Answer (as from a Coroner): She tripped over a carpet strip. Because without that, none of the rest would have happened.

 

This sort of thinking was applied by the WA Coroner in relation to an accident involving a seized fuel injection pump shaft in a Cessna, which stopped an engine at low altitude after takeoff. The pilot chose to turn against the dead engine because of some powerlines that ran diagonally across his path a couple of Km from the airport, lost control, and crashed. He had not turned on the electric fuel pump. The Coroner's finding was that the cause of the crash was an incorrect choice of bushing material in a modification to the fuel pump, about 1100 hours TIS previously.

 

There was no consideration that a pilot is supposed to be able to fly a twin-engine aircraft after an engine failure (that's why people make twin engine aircraft); or of the fact that the electric fuel pump is there to deal with that situation, or the fact that the fuel system of the engine concerned is designed in such a way that one cannot take-off with the electric pump running, or of the fact that the pilot turned against the dead engine when perhaps he did not need to.

 

So unless the individual compiling the data has a considerable depth of expert knowledge, there is a good chance that the statistics may still be misleading. It's no easy matter to get this right, and almost no national air safety bodies in the world do so all the time. The American NTSB is probably as good as any of them, and the FAA often disagrees with its findings.

 

So yes, it's a good idea and very much needed; but how do you get it right?