Sender: <marv@lancaironline.net> (Marvin Kaye)
To: lml@lancaironline.net
Date: Sat, 05 Jun 2004 16:17:33 -0400
Message-ID: <redirect-140921@logan.com>
From: "Gary Casey" <glcasey@adelphia.net>
Subject: Safety
MIME-Version: 1.0
Content-Type: text/plain;
	charset="Windows-1252"
Content-Transfer-Encoding: 7bit
Importance: Normal

I read with great interest the "unsafe at any speed" and the "unsafe in any
plane" threads.  While the argument over fault can go on forever, there are
always things that can be improved.  Pilot training and attitude adjustment
can compensate for aircraft performance characteristics and conversely
design changes can compensate for pilot weaknesses.  Neither is wrong and
both should be used to enhance overall safety.  Many years ago an automotive
safety study concluded that the lowest cost way to keep drivers from killing
themselves was to remove the things they crash into (move light poles
further from the road).  On the other hand MU-2's and Lear 24's are safely
flying today, mostly because of pilot training in spite of a terrible
initial safety record .

The incentive I had was to uncover any design weakness in the planes,
specifically the ES out in my garage.  My conclusion is that there seem to
be no glaring structural defects - all the incidents were single-crater
crashes.  Having said that I hear about cracks in the fuselage just ahead of
the empennage and by looking that does seem to be the weak link in the
structure, especially considering the location of the large inspection
panel.  What to do?  Since the plane is supposed nose heavy anyway, I'm
tempted to add unidirectional Kevlar diagonal reinforcements.

Another detail that bothers me:  The only retention of the rudder pedals are
the cables to the rudder.  Break one of those and you can probably fly, but
how do you keep it on the runway upon landing?  Some have added gears to
connect the two rudder pedals and this would solve that problem.  The
maximum stress on the cables is certainly on rollout trying to get it
stopped.  If a rudder cable let go then it would be a guaranteed accident.

Another thing:  I installed an AOA indicator with aural warning, but now I
find out that if the pitot tube is plugged BOTH the airspeed indicator and
the AOA are rendered useless and there is accident data to prove it.  In
fact, the failure mode is the worst - the airspeed reads zero and the AOA
says "stall," each reinforcing the false information from the other.  Should
there be a separate pitot system for the AOA?

Aerodynamically, the concern seems to be centered on stall behavior and spin
recovery.  It seems that to reduce wing drag the max-thickness point must be
moved back and that necessitates the leading edge radius be made smaller.
That small leading edge radius almost forces the stall to be more sudden.  A
more sudden stall encourages spin entry as the entire wing can be stalled
while the other wing is not stalled at all.  How would you fix that?  The
stall strip addition is certainly a step in that direction but it mostly
just increases the stall warning and doesn't necessarily affect the stall
itself.  Adding more twist would certainly improve things as shown by the
Columbia design.  Even on a finished wing a cuff could be added.  Maybe even
just rigging the ailerons to be slightly high would improve things as
someone suggested.  vortex generators are a way to do the same thing, but I
wonder what good they do to prevent a leading-edge stall as they are usually
located some distance back.  It seems as though no stall tests have been
done in the ES at full gross and at the aft CG limit.  If not, how did the
aft limit get established?  Just a guess?  I'm surprised that in spite of
large horizontal and elevator area some say that there is insufficient
elevator authority at forward CG and other say that any stall aft of mid-CG
is courting disaster.  Seems like the actual useful range is very narrow.?

Regarding spin recovery I wonder why the IV and ES have the same vertical
tail as I would think a longer wingspan would force a larger vertical
surface for the ES(the asymmetric drag in a spin has a longer lever arm).  I
have also read that vertical area below the horizontal is most important for
spin recovery as the area above can be rendered ineffective by the flow
across the horizontal.  A rumor had it that the Columbia 400 has a ventral
fin, but the photos on the website don't show that.  Should one be added to
the design?

In summary, what I read shows that while there is nothing truly "unsafe" in
the design, there seems to be room for improvement.  In Detroit there was
once a real sentiment against air bags - "if they don't wear their seat
belts they deserve what they get."  Experience shows that air bags save
lives of the deserving and undeserving alike.  Let's improve BOTH the pilots
and the aircraft in any way that helps.

Gary Casey
not a fighter jock or airline pilot
1.0 hours in type
ES #157, max speed to date 60 mph in a truck, no stalls, no spins.  Stall
strips already installed