Safety: EAA type accidents

[Bill Greenwood]

Last year the FAA held a meeting with some of the leading folks in the EAA field.

The FAA says that records show that the accident rate for amateur built type airplanes are 600 to 800 % higher than for all gen av accidents.

Now, some if not most of many on this forum may want to argue with these facts or even propose that the FAA should direct their attention elsewhere, and leave us alone.
I hope not,, I hope that our primary reaction is better than that.

I just wrote above, that the facts are that the accident rate for homebuilts is 6 to 8 times higher than all gen av, according to FAA.

That should be a red light to pilots and builders and make us determined to do better. The first step in
improving a problem is to discover and recognize the problem, and not ignore or wish it away.

At this point I don't have any more real info. For instance, I would really like to know fatal accident rates, as I think that is more significant. I'd like to know the details of the categories of causes. The Nall reports from AOPA give these for gen av, but not broken down for EAA types.

Whatever the details, lets all try to be safety concious, and to share with our other pilots, even if sometimes they don't want to hear it.

[steveinindy]

The major problem with determining cause in experimental crashes is that very few of the crashes are thoroughly investigated. A lot of them only get a paragraph or two of description of the wreckage and therefore mechanical issues are likely getting overlooked by the FAA local inspectors assigned to handle the on scene investigations of many of these. Honestly the EAA (and the AOPA for the rest of GA since those are also shoddily investigated) should be pushing the NTSB to have a designated representative involved in the investigation so that more can be learned from them.

[Bill Greenwood]

Steve, i had not read or heard that. Are you saying that the FAA sizes their accident investigation as to the type of plane involved?
That a fatal accident with 4 people in a homebuit would get less attention than if it was one person in a C 172?


Obviously the airlines and really large passenger planes are going to get a lot of attention, as they should.

So often, cause can be seen pretty readill, when the pilot fail in imc wether or low altirude acro.

[Kyle Boatright]

Bill, it is indeed unfortunate that the accident rate is as high as it is in GA - experimental or certified. However, consider this report, which contends that the difference in the accident rate between Homebuilts and certified aircraft is much lower than the figures you listed.

http://www.eaa.org/news/2010/homebui...t_wanttaja.pdf

[Kyle Boatright]

Steve, i had not read or heard that. Are you saying that the FAA sizes their accident investigation as to the type of plane involved?
That a fatal accident with 4 people in a homebuit would get less attention than if it was one person in a C 172?

Bill, the answer to this question is "yes".

Since no two homebuilts are alike, it is hard for the FAA/NTSB to get a lot of traction with findings from HB accidents. It isn't like they can issue an AD or anything, like they could on a C-172, where (at least theoretically) there is a huge amount of commonality across the fleet. If the seat rails on one C-172 turn loose unexpectedly, there are 30k other airplanes in the field which are subject to the same problem. Not so (in general) in the HB world, although with the proliferation of Van's designs, there are enough of those in the field where the results of an accident investigation might benefit a thousand or more owners of the same basic airframe.

[Bill Greenwood]

Kyle, The figures that I iisted came from the FAA at that meeting, I didn;'t compile or invent them.

I really hoped that other EAA folks seeing this high accident rate would want to try to find ways to fly safer, more so than trying to doubt the FAA findings. Even if their figures are off by a third, then the rate is 4 to 6 times as many accidents for homebuilts.

They do seem to agree with you that first flights carry a higher danger rate.

[martymayes]

Steve, i had not read or heard that. Are you saying that the FAA sizes their accident investigation as to the type of plane involved?
That a fatal accident with 4 people in a homebuit would get less attention than if it was one person in a C 172?

No, that is not true. The NTSB Aviation Investigation Manual does not have different procedures for certificated vs homebuilt aircraft.

[Kyle Boatright]

Kyle, The figures that I iisted came from the FAA at that meeting, I didn;'t compile or invent them.

I'm certainly not saying you invented those stats. However, I do believe that flawed statistics can misrepresent the facts, and the study I linked paints a substantially different picture than the FAA did in the meeting you attended. One thing you might consider is that whoever is presenting for the FAA is trying to grab your attention. Shocking statistics are a good way to do that, regardless of how well vetted the data is/was.

[steveinindy]

No, that is not true. The NTSB Aviation Investigation Manual does not have different procedures for certificated vs homebuilt aircraft.

Look at how many of the reports for GA and especially the homebuilt ones indicate that the on-scene investigation was handled by the FAA FSDO employees and were not investigated by an NTSB investigator except as a paperwork review. I've been on scene for several and have witnessed firsthand how they often approach this with a "Let's get this done and get home" attitude. One of the former FSDO employees up in Michigan once made the comment "We know it's pilot error. This is ****ing pointless" as he got out of a vehicle and walked towards the wreckage. It's that sort of thing which explains how we wind up with reports like:

http://www.ntsb.gov/aviationquery/brief2.aspx?ev_id=20091005X51915&ntsbno=ERA10LA007 &akey=1

http://www.ntsb.gov/aviationquery/br...09LA182&akey=1 And I quote: "A post crash assessment performed by members of the EAA familiar with the airplane found small contaminant particles within the fuel tank’s drained fitting, which supplied fuel to the engine. The fuel tank was cut opened, the interior was observed deteriorating and flaking apart." They didn't even bother to maintain a chain of custody on the wreckage. If the pilot's friends had wanted to save the pilot's family the knowledge that his negligence cost him his life, they very well could have buried the evidence.

It's not an official policy but we all know how what is supposed to happen and what actually happens are often very different things. If they were following the NTSB regulations (or even the FAA regulations), we wouldn't have a roughly 60% rate of reporting basic and vital information like restraint use. What's funny is that if you go back and look at the local police or coroner/ME reports for a lot of those crashes, you find such information was easily documented if someone was just willing to put the effort into filling in all of the information requested by the investigation documents used by the FAA and NTSB.

So often, cause can be seen pretty readily, when the pilot fail in imc wether or low altirude acro.

Even those crashes need a thorough and complete investigation (including autopsy of all fatally injured occupants because a medical problem with a passenger could explain the pilot's distraction, etc as happened in a couple of cases I know of over the years) especially if you want to identify things that can be readily fixed to improve outcomes in future crashes or prevent them outright such as restraint issues or failures with the seat attachments/adjustment systems. The latter is well documented in the case of the various Cessna aircraft including a crash that happened here locally at Eagle Creek.

[steveinindy]

the study I linked paints a substantially different picture than the FAA did in the meeting you attended

The issue with that, as much as I like and respect Ron Wanttaja (in fact, he's one of the few people in the homebuilt community who I hold in high enough regard to trust most things he tells me with minimal fact checking), he isn't exactly without his biases either. None of us are and if he's working from the same FAA/NTSB data (which he seems to be) then there is a question of who is skewing the data to suit their means. However, as Ron points out, the issue may be simply that the FAA and NTSB are underestimating the homebuilt fleet. It doesn't mean there's anything illicit or untoward afoot, but just that the data set is incomplete and that may be simply because people aren't doing their jobs in documenting things (see my earlier points).

To be honest, the truth is probably somewhere between the two interpretations and may be explained by the way the data sets were analyzed (such as Ron's choice to eliminate training crashes from the control group because of the idea that training isn't done in homebuilts when it actually is but just in a "self-teaching" or at least informal manner in most cases (see page 5 of the report)).