Space Forum / Shuttle / May 2007

Here is a true story on me training RTLS's as NASA.  I hope you like it.

His name is Curt Brown.  He is one of the most arrogant and difficult to get
along astronauts in the program.  Unfortunately, I didn't know this during
my first training session with him as a student.  The training team planed
for a Return To Launch Site, RTLS, abort by killing an engine shortly after
lift off.  As expected, there would not be enough power left in the
remaining two engines to make it to an abort orbit, or even to get across
the Atlantic and land in Europe or Africa.

I have decided to attempt to train a Flight Rule I had recently read.  That
is if control is shortly after the main engines are turned off on an RTLS
abort, the crew will command a fast separation.  I was new to the field of
training astronauts, so I thought they would know to do this task.  I could
not have been more wrong.

The rule was written because on an RTLS abort, the shuttle is in significant
atmosphere at the time the engines are turned off.  It turns out the shuttle
with the tank on is unstable, but in trim if the attitude is kept close to
the desired zero sideslip angle and -4 degrees angle of attack.  The only
devices maintaining control are the small Reaction Control System, RCS,
jets.  If the shuttle moves very far from its trim attitude, the air loads
will overwhelm the control available and the stack will tumble.  If this
process starts, the best avenue is to get of the external tank ASAP.  After
separation, the orbiter by itself is stable and the flight control surfaces
become active.

The problem I had was to do something to cause the combined orbiter and tank
to go out of control right after main engine cut off.  I used some time in
the simulator by myself to come up with a way to do this.  I came up with
the amount of force to apply, the direction to apply it, and the duration to
apply it.  It worked in the simulator by myself.  If I applied the force as
soon as the engines turned off, the stack would go out of control, but after
the separation, the shuttle would regain control.

I briefed this to my team lead and got permission to try it in the training
session.  I kill the engine shortly after lift off and the crew goes RTLS at
the correct time.  The powered flight portion of the RTLS is flown without
incident.  As soon as the main engines are turned off, I put the force in.
It did not work as planned.  Control was lost as I expected, but rather that
perform the fast separation as called in the flight rules, the crew engaged
the Backup Flight System, BFS.  To make matters worse, there is known bug in
the BFS that if it is engaged in this time frame, it gets lost on what mode
it is in and it will not separate from the tank.  Needless to say, entering
the atmosphere while still on the tank is a disaster.  The rest of the
training simulation was a disaster.  The shuttle tumbled and tumbled.  We
finally declared the crew dead and stopped the simulation.

Curt Brown was furious.  In the debrief I showed him the flight rule that
says the crew should perform a fast separation in that phase if control is
lost.  Curt had never heard of this.   I thought it strange that a flown
crew member would never have been trained on such an important rule.  The
next day I asked Andy Foster, the lead instructor on ascent abort, about
this rule.  Andy informed me this was a bad rule and it should not be
trained.  I was stunned.  How can we have an official policy to not train a
flight rule?  The flight rules are the bible of operating the shuttle - for
both the crew and the flight controllers.  I also argued that BFS should
certainly not be an option in this phase if it is known to be catastrophic
to engage.

My complaints fell on deaf ears.  Everyone more senior than me was
comfortable having a rule on the books that "everyone" knows we should not
follow.  Everyone was also comfortable not training the crew engaging the
backup flight system in this phase - even though it is know to cause certain
loss of orbiter and crew if they do engage.