Why is a LOX/Kero SSTO not rather easy?

On Tue, 9 Sep 2003, Bob wrote:

Date: Tue, 9 Sep 2003 10:19:54 -0500
From: Bob
Newsgroups: sci.space.tech, sci.space.policy
Subject: Why is a LOX/Kero SSTO not rather easy?




Please review this thread for another, more detailed, SSTO concept using
NK-33s and a wet wing.

http://www.google.com/groups?hl=en&l...phi.com&rnum=1

I hope pasting this will work. If not, then I will paste in the entire
article. Thanks,

BobDL





-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----

=====================
Thanks a heap -- I remeber that article and have looked in vain for it for
a while. Much appreciated

-- Larry

Which has nothing to do with his assertion that simply scaling up and
throwing testing out would make it cheaper.

--
"Yea, all israel have transgressed thy law, even by departing, that they
might not obey thy voice; therefore the curse is poured upon us, and the
oath that is written in the law of Moses the servant of God, because we have
sinned against him." Daniel 9-11

Penguinista wrote in message m...
Len wrote:
Penguinista wrote in message m...

I was under the impression TPS scaled linear with the mass of the
vehical. For a larger vehical, unless surface is proportional to mass,
it will plunge deeper into the atmosphere before slowing and take a
greater intensity of skin heating.


Not necessarily so. If mass ratio gets better with
size, it is quite possible to end up with a planform
loading comparable, if not less, than for that of
a smaller vehicle.

Does not follow. Suppose you can make large structures with a wall
proportional to sqrt(scale). The resulting structure would have a
volume a*scale^3, mass b*scale^2.5, mass ratio c*scale^(1/2), and
surface d*scale^2, thus planform loading e*scale^(1/2).

To maintain planform loading with larger scale would require such things
as a constant wall thickness across scale. To reduce this loading would
require even thinner walls, not a realistic option. Above a mid sized
launcher tanks and engine comprise practically all non-recovery related
mass.

Please note that I said: "quite possible to end up with
a comparable planform loading.." I agree that--other
things being equal--smaller vehicles usually end up with
a lower planform loading. Nonetheless, the improvement
in mass ratio with size can, in some cases, compensate
for this main effect. And then, for reasons that follow,
heating may be less severe.

And then you can be ahead, since
temperature tends to drop with absolute distance, not
relative distance from the nose or leading edge.
Moreover, higher Reynolds number may get you more
removed from "slip flow" which tends to increase
heating.


Those aerodynamic factors might be enough to maintain heat loading with
scale. Anybody have solid figures on this?

My experience has been that--for a given design philosophy
--the smaller vehicle usually has lower heating, but may
have a higher TPS percentage of empty mass, since there is
proportionately more area to cover and there as certain TPS
components that do not scale with temperature. The main thing
for either size is to integrate propellants; external tanks
get you into a significantly higher heating realm for the
remaining spacecraft, since propellant tanks tend to have a
lower planform loading than other parts of an orbiter.

I have also found that TPS can vary wildly with particular
design philosphies. It's sort of an unstable situation:
when heating gets high, TPS weight goes up enough to increase
planform loading, thereby resulting in higher heating. If
you can manage to get the phenomenom going in the opposite
direction, good things can happen. Dynasoar TPS was
extremely heavy on a unit-weight basis--partly because of
exit heating, and partly because it was a research vehicle
designed to delve into problem areas.

Best regards,
Len (Cormier)
PanAero, Inc. and Third Millennium Aerospace, Inc.
( http://www.tour2space.com )

Am Sun, 14 Sep 2003 15:28:08 -0700 schrieb "Larry Gales":

[...]
Anywho, the problem with adding a small number of engines is
that because of that same limited throttle range you have to
rely on all of them. With three times as many engines and
all of them needed you've just upped the chance of *a* failure
involving an engine by a factor of 3. In order to keep the
failure rate the same you need to have enough redundancy so
that you have the same probability for equivalent
configurations. For example, if you have a 1 engine rocket
that uses a 1% failure rate engine then you have a 1% engine
related failure rate. But if you up that to 3 engines you
end up with a 3% failure rate vehicle. To get back below
1% you only need to add an extra engine though, you'll have a
4% failure rate for one engine out but only a .06% rate (I
think) of falling below the minimum number of engines needed
for flight. Though, of course, that brings symmetry issues
into play since you usually have to shutdown engine pairs
unless it's the center engine.

You will achieve a stable flight, as long as the thrust vectors of all
engines will go through te center of gravity (CG) of the rocket
complex. So, if you provide gimballing enough to move the thrust
vectors according to the moving CG with emptying of the tanks, any
engine can cut out without losing stability. I know, that this is
problematic while in atmospheric flight because of aerodynamic loads,
but in principle this method works.

And this "trick" everytimes works then, when there are engines enough
to acieve the goal of not losing the payload, even when one engine
fails. And you don't NOT need to switch off engines symmetrically. The
disadvantage of this is, that you lose a bit performance, if the
thrust vector does not direct exactly to the acceleration vector. So
that is used (normally) only, if there is no possibility to switch off
the engine - e.g. solid boosters like found on many launchers.

Well known examples are the Delta-II/III, or the "old" Ariane-4, where
the booster engines have thrust vectors going through CG and not
parallel to acceleration vector. And there are even asymmetrically
mounted booster arrangements, like in Delta-74xx models...

In other cases, when all or main thrust is generated by engines with
throttling/switch off capability (usually liquids) it is mostly easier
to invest in gimballing capacity of (part of) the engines to reduce
the risk of necessity of off-switching or throttling of healthy
engines - as long as there IS enoungh thrust anymore. And that depends
on the total number of engines and the thrust reserves of them.


cu, ZiLi aka HKZL (Heinrich Zinndorf-Linker)
--
/"\ ASCII Ribbon Campaign
\ /
http://zili.de X No HTML in
/ \ email & news

Am Sun, 14 Sep 2003 15:28:08 -0700 schrieb "Larry Gales":

[...]
Anywho, the problem with adding a small number of engines is
that because of that same limited throttle range you have to
rely on all of them. With three times as many engines and
all of them needed you've just upped the chance of *a* failure
involving an engine by a factor of 3. In order to keep the
failure rate the same you need to have enough redundancy so
that you have the same probability for equivalent
configurations. For example, if you have a 1 engine rocket
that uses a 1% failure rate engine then you have a 1% engine
related failure rate. But if you up that to 3 engines you
end up with a 3% failure rate vehicle. To get back below
1% you only need to add an extra engine though, you'll have a
4% failure rate for one engine out but only a .06% rate (I
think) of falling below the minimum number of engines needed
for flight. Though, of course, that brings symmetry issues
into play since you usually have to shutdown engine pairs
unless it's the center engine.

You will achieve a stable flight, as long as the thrust vectors of all
engines will go through te center of gravity (CG) of the rocket
complex. So, if you provide gimballing enough to move the thrust
vectors according to the moving CG with emptying of the tanks, any
engine can cut out without losing stability. I know, that this is
problematic while in atmospheric flight because of aerodynamic loads,
but in principle this method works.

And this "trick" everytimes works then, when there are engines enough
to acieve the goal of not losing the payload, even when one engine
fails. And you don't NOT need to switch off engines symmetrically. The
disadvantage of this is, that you lose a bit performance, if the
thrust vector does not direct exactly to the acceleration vector. So
that is used (normally) only, if there is no possibility to switch off
the engine - e.g. solid boosters like found on many launchers.

Well known examples are the Delta-II/III, or the "old" Ariane-4, where
the booster engines have thrust vectors going through CG and not
parallel to acceleration vector. And there are even asymmetrically
mounted booster arrangements, like in Delta-74xx models...

In other cases, when all or main thrust is generated by engines with
throttling/switch off capability (usually liquids) it is mostly easier
to invest in gimballing capacity of (part of) the engines to reduce
the risk of necessity of off-switching or throttling of healthy
engines - as long as there IS enoungh thrust anymore. And that depends
on the total number of engines and the thrust reserves of them.


cu, ZiLi aka HKZL (Heinrich Zinndorf-Linker)
--
/"\ ASCII Ribbon Campaign
\ /
http://zili.de X No HTML in
/ \ email & news