Homebuilt safety and NTSB reports

[danielfindling]

While browsing the NTSB site for January 2013 incidents, I came a cross this preliminary investigation regarding an RV 10

http://www.ntsb.gov/aviationquery/br...05X35922&key=1

- - basically an engine failure at 6 hours during flight testing resulting in substantial damage to the airplane.

If the NTSB could provide more information regarding engine type: e.g. Certified, experimental or auto conversion. . . It would appear to me to be relevant to safety. Anyone have more information on this incident?

does EAA follow up on these incidents to improve safety, compile data etc.? What about Vans?

If a goal of EAA, FAA etc. is to improve safety, it would seem relevant to understand the data.

Daniel Findling

[danielfindling]

Ok, my goal is to start a dialogue, not to bash experimental aviation. It appears we are under a microscope for saftey concerns compared to certified aircraft. The Seawind had a history of engine trouble and recently replaced fuel injector and overhauled prop.

http://www.ntsb.gov/aviationquery/br...12X64931&key=1

What is missing is information on the engine. Certified, experimental or conversion . . .

Daniel Findling

[WLIU]

The problem with better data in the accident report is that generally the FAA has an Aviation Safety Inspector from the FSDO look at the accident and those individuals know very very little about the universe of homebuilt aircraft out there. Or any aircraft that carries some kind of Experimental airworthiness certificate. And they are not tasked with learning about that topic. You will note that the vast majority of homebuilts get their certification through Designated Airworthiness Representatives, not FAA inspectors. With the FAA budget priorities focused on the air carriers and sexy programs like NextGen, it is hard today to get the ASI's out of the office to look at anything that is recip powered general aviation. When was the last time that you met an FAA inspector on the ramp at your home airport on a Saturday?

The other problem is that the information is often no longer available. When a Beechcraft goes down, the FAA can call Beechcraft to confirm the details of the aircraft. Engine, propeller, number of seats, etc. The builder is the repository of all info about an individual experimental amateur built aircraft. If the builder perishes in the crash, the accident investigator has no source for the information that you suggest would be useful in an accident report.

So all we are left with is gross statistics. Maybe number of hours on the aircraft, the type of airworthiness certificate, number of engines, general configuration.

So when the airplane says "Experimental", there are great aspects of what you do, and potentially some not so great aspects.

Fly safe,

Wes
N78PS

[Kyle Boatright]

That airplane had a lycoming. There has been some discussion on the RV boards. It may be that the registration is so recent the FAA's databases aren't current enough to show the information you're interested in.

[rwanttaja]

There are four basic accident reports.

There is the FAA preliminary on the FAA web page, that identifies the type of aircraft and the basic circumstances in tabular format. This usually appears the day after the accident, or the Monday after if it happened on a weekend.

The NTSB preliminary comes out about two weeks later. The preliminary is basically a narrative version of the FAA prelim. Usually, it just describes the circumstances of the accident ("Aircraft lost power and landed in a field") without any greater depth of description.

The next report is the NTSB Factual Report, which comes out quite a few months later (8-10). The Factual Report is basically a set of filled-out forms, and will normally include details such as the engine type, pilot total time, pilot time in type, etc. It includes a narrative (sometimes several pages long) that summarizes all the results of the investigation, including any lab work. It does not identify the cause of the accident, but usually it can be inferred from the facts provided.

Finally, about a year after the accident, the Final Report comes out. This is basically the Factual, with addendum that provides the NTSB Investigator's Probable Cause of the accident.

The NTSB accident report databases are available online for public download and analysis. They are in Microsoft Access format. You can download a year's worth of accidents, and build your own summaries based on aircraft type, engine type, etc. These are located at:

http://www.ntsb.gov/avdata/Access/

I have been downloading and analyzing this data for about ten years, now, mostly dealing with Experimental aircraft accidents. When I transfer the NTSB data to my personal database, I add columns to make it easier for me to break out accidents by other factors, such as Engine Type (Certified, auto, two-stroke, etc.), homebuilt type (the NTSB designators are all over the map, but I standardize them), and cause (I read the narrative for each accident, and assign the cause to one or more of 51 categories).

In short: Do you have a question about homebuilt accident statistics? Ask it here, and I'll try to answer it. If I can't or won't answer it, I'll explain what the difficulty is.

The one thing I'll caution you about is that my database (currently) only runs through 2011; I prefer to wait until the investigations are complete and a probable cause is assigned.

Ron Wanttaja

[rwanttaja]

Ok, my goal is to start a dialogue, not to bash experimental aviation. It appears we are under a microscope for saftey concerns compared to certified aircraft. The Seawind had a history of engine trouble and recently replaced fuel injector and overhauled prop.

http://www.ntsb.gov/aviationquery/br...12X64931&key=1

What is missing is information on the engine. Certified, experimental or conversion . . .

The FAA registration database list this as "Lycoming IO-540 series." If it's a commerically-modified version of the engine, the registration database sometimes indicates this, but this listing should not be assumed to mean it was stock.

The NTSB factual report usually gets specific as to the engine type.

Ron Wanttaja

[rwanttaja]

does EAA follow up on these incidents to improve safety, compile data etc.? What about Vans?

To steal a line from "The Right Stuff": "You're looking at him." I write an annual summary of homebuilt accident trends for EAA, published in Sport Aviation. Just submitted the 2012 report a couple of weeks back, think it'll be in the March issue. The 2011 report was published April 2012. I've also published numerous articles on the subject in Kitplanes magazine, including analyses of the statistics for specific homebuilt types.

BTW, just so it's clear: I'm a freelancer, I don't work for EAA. Nothing I say here should be construed as coming from the EAA.

Ron Wanttaja

[1600vw]

The FAA registration database list this as "Lycoming IO-540 series."
Ron Wanttaja

I believe I wread somewhere that this engines are having crank problems.

[WLIU]

What you likely read was that Lycoming had a problem with a vendor and recalled a number of crankshafts. Lycoming provided a discount on replacement cranks and some $$ towards the labor of replacement. I have a number of friends who bought new cranks.

But owners of experimental aircraft may not receive FAA Airworthiness Directives, may not look at service bulletins, and are not obligated to follow them. The experimental category has fewer constraints than Standard, Normal, Utility, Acrobatic etc certified aircraft.

So from Ron's posts above we now know who tracks Experimental - Amateur Built aircraft incidents and accidents. That kind of data base has value, and I suspect that if Ron wanted to be the steward of what in the high tech world we call an "open source project", that there might be a volunteer or two able to make some diplomatic calls to follow up on accidents so as to obtain more detailed information to include with the statistics. I suspect that not only EAA can use the data, but I will suggest that insurance companies, manufacturers like Vans' and others would provide funding to support a database hosted by a "neutral" party, if that database offered more insight on the types of issues that result in accidents than what we have today.

Best of luck,

Wes
N78PS

[rleffler]

Unfortunately, there have been two recent RV-10 incidents that preliminary information indict the root cause is builder error.

As Kyle mentioned, there has been significant discussion on these incidents in the RV community. On appears to be a fitting that came off in flight and the other appears to be a stop not that wasn't installed. Over the last couple years, there seems to be more of these types of accidents occuring. I suspect that may be due to the increased numbers of experimentals being completed.

In all cases, the problem could have easily been avoided. The opportunity for the community is how to educate builders to pay closer attention to detail when it comes to fittings and fasteners. Torque seal works well when used appropriately to validate inspection. I think the problem is that the builder is too close to the project and starts making too many assumptions as the build gets complete, or simply doesn't remember or overlook testing the torque on the various connections.

This is probably the one area that keeps me up at night, since I'm about a month away from getting a pink slip for my RV-10. I think my approach is to invite a group of 3 or 4 builders (not at the same time) over to inspect the RV-10 before the final FSDO visit. To motivate them a little, I'm going to offer them their favorite adult beverage for each building defect or issue they find. In my opinion, there is no room for ego here. Worse case is that I may owe some friends a few drinks, but the positive is that I hope they find anything that I may have overlooked or missed.

But back to reality, what keeps me up at night is what if there is something that we all missed in our inspections? I know the odds are minimal and the risk is low. But I'm sure these other builders thought they had all their bases covered too or they wouldn't have flown their aircraft.

bob