Space Forum / Space Flight / January 2004

Consider vehicle major maintenance checks.

Major reliability drivers that we can predict ahead of time
are going to be main engines and TPS.  We can also predict
that a lower stage engine failure or a lower stage TPS problem
are less critical than that of the orbiter, coming back
from 2x the velocity and 4x the energy...

Consider for example a vehicle maintenance rotation where
vehicles spend a year doing orbital work, then get their
engines rotated for the short nozzle models and OMS pods
unbolted, and then are used for booster flights for 2 years,
and then undergo the equivalent of a D-check and are put
back into orbital flight for another year.

System upgrades and such get introduced along with the
checks and refurbishment, in the vehicles that need it
the most... the ones which will be flying orbital
missions for the next year or so.  But the
vehicles flying orbital missions don't have to undergo
D-checks every year.  As their systems and structures
age somewhat you just shift them into less demanding
booster work for a while, with the short nozzles,
and then push them back into orbiter service after
the next major rebuild and upgrade cycle.

-george william herbert
gherbert@retro.com

Bimese is one of those ideas that just doesn't work when you start looking
at the details.

Forcing the two stages to be identical isn't free. It isn't even cheap.
Among other things:

The fixed proportion of stage sizes constrains the staging point to a
specific value around Mach 3-4. This is a low staging velocity by normal
standards and is a very non-optimum split between the two stages, forcing
the overall system to be much larger then necessary, even before the
inefficency of the identical design is taken into account.

The low staging point also results in a relatively high dynamic pressure at
staging (well past max Q, bust still decidedly endoatmospheric. The staging
velocity is low enough that a glide return is possible, but high enough to
make it touchy - attention must be paid to the glide characteristics of the
booster and the wind profile.

Bimese requires crossfeed (transfer of propellant from the booster to the
orbiter in flight) because otherwise the orbiter runs out of propellant
around the same time as the booster, and you would effectively just have two
SSTOs bolted together. Crossfeed is not impossible, but it is quite complex.
In particular, a Bimese vehicle requires that the crossfeed flow be shut
down and the feed system switch to the internal tanks while the engines keep
running. It is therefore not especially analogous to the Space Shuttle
Orbiter/ET connection.

On top of that, the functional requirements for a first and second stage
really aren't all that similar. A true bimese configuration forces the
duplication of wholly unnecessary systems on the two stages. For example,
the booster has essentially no need for a TPS system because of the low
staging velocity. It also has no need for OMS, or a long duration power
supply such as fuel cells (which may or may not be required on the second
stage depending on the mission duration requirement). In turn, a booster
stage must, by definition, have a stage vacuum thrust/Weight ratio well in
excess of 1. An optimally designed orbiter can get by with a T/W of 1 or
less at staging. Therefore, forcing commonality puts more engines on the
orbiter than it really needs.

Obviously, carrying around the deadweight of these superfluous systems makes
the overall system substantially heavier than an optimized design. To which
one might argue that mass isn't what matters - cost is. So consider this,
does it really make sense, from an cost standpoint, to needlessly duplicate
the components of the system that are the most expensive to buy and
maintain - engines, TPS and power?

It can be relatively painless to share only the subsystems that are most
expensive to design - engines, avionics, software - without making the
airframes look the same. That kind of commonality has a much better payoff
than a pure Bimese system.

For more on this subject, see "Selection of Lockheed Martin's Preferred TSTO
Configurations for the Space Launch Initiative" paper number IAC-02-V.4.03
from the 2002 World Space Congress. It describes a trade study in which
bimese placed last out of twenty TSTO configurations. It is interesting to
observe that all three of the SLI/2GRLV contractor teams plus NASA studied
bimese concepts, and that bimese was the initial baseline for Boeing and
Orbital, yet by the end of the contract no one thought it was the preferred
choice.

Josh Hopkins

What if you have a hundred (or more?) small chambers instead of a
couple large ones?
In addition to being deeply throttle-able, you could also run just a
few on the return.

And as for L/D, well... there is a classical solution to that.

Aloha mai Nai`a!

Option 4; Design hybrid (combined cycle) systems with some of the advantages
of both, optimised for the specific mission type possibly at the expense of
engine
life.

You want the first stage to contribute a fair bit of horizontal velocity,
not just altitude, to the second stage.  The dominant problem of getting
into orbit is velocity.  A first stage that just goes straight up requires
the second stage to have near-SSTO performance, and the big reason for
using two stages in the first place is the belief that SSTO is Too Hard.

Also, a vertical reentry from high altitude is a lot harsher than one with
a horizontal component, because it takes you down into thick air before
you've had a chance to decelerate much.  From 100km, it's manageable; from
much higher than that, it gets pretty nasty unless you've got something
like a deployable drag brake.

Am 30 Nov 2003 01:42:55 -0800 schrieb "George William Herbert":<