BACK TO THE FAMILY OF FOUR
NTSB and FAA
Lets revisit the family of four that was killed in the airplane that I had just torn apart, put back together, made a mistake and they all died on the first trip out of our hangar.  My name and signature was in both logs, the airframe and powerplant logs signing off the 100 hour inspection which included an oil change and full service.  I expected some intense interviews and testimony in front of the NTSB and FAA over the next whatever period of time.  It was likely both would be required since the NTSB is involved in fatal accident investigation and the FAA would be reviewing what I did during the maintenance visit. The NTSB is independent of the FAA to enable causal findings against the FAA in an accident if necessary.

A little later in the same day after I was told about the crash, the F.A.A. told us that the pilot of the plane had probably wandered into cloudy zero visibility conditions for which he was not certified to fly, and he flew it into the side of an 11,000 foot mountain east of San Bernardino.   But what about that mistake I made during the job?

The Piper Cherokee 180 came into our facility for a 100 hour inspection including servicing the engine with a leak down compression check, oil and filter change and dual magneto synchronization checks in addition to other visual inspections and aircraft documentation checks.  It was a Saturday, so I was working alone for the day.

Lycoming 0-360  180 hp

The engine was an Avco Lycoming 0-360 A4A, a 360 cubic inch displacement four cylinder opposed air cooled piston engine of 180 horsepower, naturally aspirated and carbureted. 
Nothing fancy, very simple with 8 spark plugs and two magneto ignition systems.  (Ironically, seven years later I would become an aircraft accident investigator employed by Lycoming.)

While changing the oil in the engine, I had just replaced the oil filter.  Then the phone rang in the hangar and I answered it.  That led to some running around and a little momentary confusion.  When I got back to the airplane, I finished some other tasks, then started up the engine for leak tests.  About 30 to 40 seconds later I noticed it!  Horror overcame me and I shut down the engine. 

No oil pressure.  I had forgotten to put oil in the engine. 

My first error was neglecting to put oil in the engine.  My second error was not fixing my eyes on the oil pressure gauge on engine start like I always do.  I thought about it for a minute, and then I decided that since I had just drained the oil about an hour prior and it had only run 30 seconds, it was probably not damaged.  "What a dummy!" I thought to myself.  No strange noises came out of it at the time.   So I finished the service, tied up some loose ends, took the airplane out to the runup area and tested the engine coming up with normal results in all parameters.   I signed off the airframe and engine logbooks and released the aircraft for service.  I went home, coming back to work to the bad news. 

Even though later in the day we'd gotten a hint of the possibility that the pilot had probably flown into the side of a mountain while in the clouds, I lived for 4 days in misery.  On the fourth day, the FAA informed our facility that the accident was a CFIT (controlled flight into terrain) accident with the engine operating at full 100 percent climb power setting when it impacted the side of the mountain and that no mechanical issues were involved. 

I had dodged a bullet. 

The pilot was probably climbing through a cloud layer in zero visibility conditions and never even saw the mountain coming at him and his family.  It's easy to do.  You can think that the cloud layer is thin and you'll only be in it for 10 or 20 seconds to pop out of the top of the layer.  This guy must have gone into the underside of the layer and that was it.  Unreal.

I was not at fault, the pilot was.  The FAA and NTSB never even interviewed me, not once.  I haven't thought about that in a long time.  In fact, up until I started looking into some of the subject matter for this writing, I can't think of the last time I had ever thought about it.  I must have wanted to forget it.
 
About three years later I made another mistake that cost 100 thousand dollars in a corporate aircraft turbine engine test fire that was very much my fault.  More about that later.

Now we'll look at some experiences with the MEDA investigation approach.

PAGE FIVE:  Experiences with the MEDA Process

BACK to TM page one