|
My area of expertise is failure prevention. I hope
you guys will consider my analysis:
First, I don't want to detract from the excellent
analysis so far. Very open minded, willing to consider multiple causes. I
particularly like the diode concept, because it makes the plane insensitive to
the cause. You admit it will happen again, but change the design so the failure
can't affect the plane. Good stuff. You guys might be saving a life.
Consider this: Both Al and Eds failures have the
exact same design oversight. Your planes have only a handful of switches that
are flight critical. Both of you had switches in
the wrong position. Both of you were ignorant of the incorrect switching. This
is a system error. You fell into the trap of designing your plane in traditional
method.
These are not freak occurrences. Al's failure is
the leading cause for Cozy deaths. 7 or 8 so far. I know of at least two other
guys that fell into Ed's trap. Both of these could have been fatalities.
Solution: Flight critical switches need to be
designed in very thorough manner. You need everything headed in your favor. Ed,
move your emergency switch to the corner of your panel. You don't want any
routine switch activities to be anywhere near your emergency one. When you
operate your emergency switch, it needs to be an unfamiliar hand movement to
different area of your panel. Place a switch guard on it (NOT one of those red
aircraft latch switches, they cause more failures than they prevent). Place a
flashing LED above it that says "Alternator Disabled".
Al, checklists are very ineffective. They are
traditional, lot's of people believe in them. They are just highly unreliable.
You need method that is not sensitive to the distractions that periodically
occur during preflight and takeoff. My Plc method is the best approach. But any
method that includes visual, aural warning will work. As long as system doesn't
provide false warnings. Checklist should be used as backup method.
I found a microswitch location on my canopy system
that's very effective. Switch is only activated when canopy is not latched. No
false warnings. Switch is protected from pilot egress damage.
Glad no one got hurt.
----- Original Message -----
Sent: Wednesday, June 03, 2009 3:40
PM
Subject: [FlyRotary] Lessons
learned
Ed set the example
by pointing out contributing factors in his ‘incident’, and for the sake off
completeness I should do likewise.
We were planning on
being wheels up by 8:00
am; and making the
roughly 7 ½ hr flying time to Austin that day.
But the airport was socked in with unusually heavy marine layer – 400 ft
ceiling. This generally breaks up pretty early 20 miles inland at
F70. We waited, and waited, looking at the sky; no change. Finally, and
rather suddenly we saw breaks in the clouds. At 10:15 I decided we were
good to go; we strapped in, and headed out. Still some low clouds in the
direction of takeoff, but I saw we could be off, make a 180 and climb.
IF
the cause was that
the door latch was not fully engaged it was because I was impatient to get
going; not thorough on the pre-takeoff checklist. Hey, I had closed and
latched that door 100 times – of course it was fully
latched.
Don’t ever be in a
hurry when you are going flying. Don’t ever be in a hurry when you are going
flying.
Al
G
|