Designing a new orbiter

The existing shuttle is due to be retired before long, even if NASA
decides to fly it again at all, and will need to be replaced. Now, I
rather doubt that anyone from NASA will be inviting input from me any
time soon, but I thought it would be interesting to play the game of
designing an orbiter.

So, if anyone wants to flame me for not having aeronautical
qualifications, or for thinking I know anything about this, don't
bother. It's a game, OK?

Raising issues and questioning assumptions is another matter.

Way back when, before the shuttle design was decided, there was a
proposal for an orbiter that had straight short wings, and which start
flying until its speed dropped below the speed of sound. That is, it
would be in a deep stall, at 90 degrees angle of attack (AOT) for more
of its descent. When I original read about it I imagined that it would
decelerate until it was approaching terminal velocity, and then pitch
over pretty much through about 76 degrees to start flying, albeit in a
dive, and then pull several g to return to horizontal motion, before
flying to a landing.

I came across this page today:

http://www.aerospaceweb.org/question...s/q0150b.shtml

One thing that comes out of this is that vehicle in a deep (90 degree)
stall has about twice as much lift as it would it if were flying
normally at an angle of attack that gives maximum lift. This was
initially rather counter-intuitive for me, until I realised that the
special thing about wings is not that they give lift, but that they do
it at right angles (roughly) to the direction of airflow. Anyway, one
implication is that the terminal velocity for vertical descent at 90
degrees AOT is substantially below the speed for level flight, let alone
for pulling out of a dive. It would make no sense to approach terminal
velocity before the transition to flight.

It suggests an alternative approach to the transition from a vertical
descent to flight, which is for the craft to attain a 45 degree angle of
attack (AOT) and hold that until the craft is descending at about a 45
degree angle before changing to a 14 degree AOT.

What next? A design parameter: Touchdown speed - 100m/s (approx 225
mph). Taking the coefficient of lift to be 1.0 (bit lower than the
aerofoil in the URL above), and the density of air to be 1.225kg/m^3,
gives 6125 N/M^2, so the craft needs about 1 square metre of wing per
600Kg of mass, or 1.66 square metres per metric tonne.

I did some simulations using these figures. Starting at 10,000 metres
falling at 200 m/s (there is a comment on this starting point below) the
sequence of events is approximately:

T = 0 sec, 10,000 metres, descending at 200 m/s, 2.4g, start pitchover
at 5 degrees per second, until 45 degress AOT is reached.

T = 9 sec, 8,500 metres, descending at 141 m/s, 1.3g, reached 45 degree
AOT. Since there is now some horizontal motion, the nose is 29 degrees
below the horizontal.

T = 24 sec, 6,700 metres, descending at 96 m/s, 1.2g, Nose is 4 degrees
below horizontal. Start pitchover at 5 degrees per second, until 14
degrees AOT is reached.

T = 30, 6150 metres, descending at 102 m/s (yes, it speeds up again a
bit), 1.1g. AOT of 14 degrees attained. The nose is 31 degrees below the
horizontal.

T = 47, 5000 metres. No longer descending. Moving forwards at 169 metres
per second in normal flight.

The g forces are highest at the start of the procedure, when the craft
is falling vertically. The lowest g force, 0.6g, occurs at about T = 29,
and corresponding to local minimum in the lift coefficient at an AOT of
about 20 degrees, and this minimum is the reason that the rate of
descent increases again at about T=23.

The maximum pitch down relative to the horizontal is 31 degrees.

I've ignore the effects of compressibility of air for this purpose, but
taken into account the changes in air density with height. Also ignored
is the induced drag which will be a function of the aspect ratio.

By comparison, going immediately to 14 degree AOT (76 degrees below the
horizontal) gets the craft to level flight in 18 seconds, having lost
3000 metres. However the g forces reach 3.5g, and the craft is doing 250
m/s when it achieves level flight.

The starting point of 10,000 metres and 200 m/s is based on some other
simulations, which show that the speed at 10,000 metres is not very
sensitive to the initial conditions, whether just falling in from
orbital height, or falling in at 8000 m/s (orbital speed). Either way,
the speed at 10,000 metres is less than 200 m/s, so 200 m/s at 10,000
metres must be an achievable starting point.

*Stability*

The transition from 45 degree AOT to 14 degree AOT takes the craft
through an aerodyamically unstable part of the envelope. This defintely
requires artificial stability.

*Landing gear*

The underside of the craft is the heat shield. The existing shuttle has
landing gear doors in its heat shield, which seems questionable to me.
I'd prefer to have the heat shield uncompromised, and have the main
landing gear come from above the wing, and the nose gear from above the
nose. This may look absurd at first sight, but I'm thinking of a
structure that pivots up from the wing, about the longitudinal axis of
the craft, and then pivots forward around the wing about the axis of the
wing, placing the wheel under the wing, and directly beneath the pivot
point.Something similar can be constructed for the nose gear. One
concern is whether these structures can be made stiff enough.

Of course, having the gear deploy from the top of the wing would get any
first year student kicked out of aeronautics engineering school, because
of what it does to the nice clean top surface of the wing, but in this
case we're not looking to minimise drag.

See

http://members.optusnet.com.au/sylviae/gear1.jpg
http://members.optusnet.com.au/sylviae/gear2.jpg
http://members.optusnet.com.au/sylviae/gear3.jpg
http://members.optusnet.com.au/sylviae/gear4.jpg

*Flight controls*

Roll control during reentry could be achieved by means of heat protected
spoilers that project into the airflow (which is vertical relative to
the wing).

http://members.optusnet.com.au/sylviae/fc1

During transition to flight, the controls would be rotated further into
the airstream to function as more conventional ailerons. A similar
mechanism could be used as elevators on the tail plane to control pitch.
The latter controls need sufficient area to handle the movement forward
of the centre of lift that occurs as the AOT is reduced to 14 degrees.

During normal flight, a fin and rudder will provide yaw control. During
reentry coordinated use of the the roll and pitch controls together
provides yaw control.

*The wing*

The wing shape is not conventional, because we don't want it to have a
sharp trailing edge, to avoid heating problems.The characteristics of
the wing profile given at the aerospace.org URL show that it's even
possible to fly a conventional wing profile backwards. I infer from this
that a wing
with a blunt trailing edge is acceptable, if one isn't too concerned
about drag.

The main drawbacks that were identitied with the stubby wing concept
appear to have been concerns about the transition to flight, the lack of
cross range ability (the ability to land at a place that's offset to the
side of the orbital plane), and limits on its ability to land back at
the launch site after a single orbit. There were no doubt some politics
involved too.

It still seems a shame that the concept wasn't further explored. The
transition to flight doesn't appear to be that dramatic. It certainly
doesn't consist of going into a dive and then trying to pull out of it.
The reentry phase is a ballistic one that seems to have been well
mastered years back by both the Americans and the Russians. By going for
a hypersonic delta-winged glider, NASA gave its engineers some difficult
problems to solve. It's a credit to them that they managed it, but
sometimes the best way to deal with a problem is to choose a different one.

Sylvia.