On Thu, 18 Aug 2005 14:19:04 +0000, Mortimer Schnerd, RN wrote:

WE had one of our Lances crash at RDU one foggy morning that I believe
was due to the pilot running one tank dry while on the ILS inbound from
the outer marker. He hit the tops of the trees, spun around and down
one and landed flat in a densely wooded area to the north of the field.
The resulting fire destroyed the aircraft pretty completely.
Fortunately, the pilot walked away from it with just a superficial cut
on his forehead.



Let me be clear here, I am not talking about accidentally running a tank
dry. In fact, if done as Deakin and Cook prescribe, it probably would of
prevented the accident you mention.

He may have tried to restart but just didn't have enough time before he
sank into the trees. Clean, a Lance will come down at 1100 FPM; it's
gonna be considerably faster with the gear hanging out like it would
after intercepting the glideslope.

http://www8.landings.com/cgi-bin/nph...at&pos =71003

Again, this sounds like an accident caused by poor planning. This is not
what I'm talking about. Poor fuel management is poor fuel management.
Let's not confuse the two.

Now, to get back to the first question: how often does one run a tank
dry intentionally? If I was in a Cherokee Six with four tanks and had
passengers, hopefully never. The downside of this is that if you left
just a couple of estimated gallons in each tank, you have lost a
significant amount of useful fuel.

Deakin specifically addresses the issue with passengers on board. In his
opinion, proper passenger briefing, a timer, and calm behavior on part of
the pilot is key.


If I were trying to stretch a flight and even then only if I were alone,
I might consider running one dry. But I have to tell you: running a
tank dry in a Cherokee results in te longest 30 seconds of your life. I
ran a C-210 tank dry once and almost the second I hit the boost after
switching the fuel selector I got a restart. Not so in the Cherokee...
it takes a while. A loooong while. Kind of scarey.

If you have not done so, please read his article and see what you think. I
must say, my impression of your position is one of fear not one of reason.
Am I wrong?

I must admit, I certainly would not expect something like a 30-second
delay. I would expect something more in line of a couple of seconds at
most. Perhaps it was a could of seconds but felt like 30?

Greg