Critique of Sarigul-Klijn paper
Mitchell Burnside Clapp
Launch Vehicles with Recommendations for Launch and Recovery"
(AIAA-2003-0909), available from
http://mae.ucdavis.edu/faculty/sarigul/AIAA_2003_0909.pdf,
the authors, Marti and Nesrin Sarigul-Klijn, make a number of
inaccurate assumptions and errors that do not support the conclusions
they have drawn.
The specific impulse figure asserted on page 3 and applied on the
example on page 4 for a pressure fed lox-kerosene engine is not
accurate. The reported figure, 225 seconds, is the sea-level specific
impulse for an engine designed to expand from 250 psi to 14.7 psi with
an engine efficiency of 90% of the one dimensional equilibium value.
The authors only report the chamber pressure and the specific impulse,
but the design exit pressure of the nozzle has an effect on the
performance. A 90% efficiency is probably realistic, for a low
development time and cost engine although personally I think you might
be able to do a percent or two better without too many cycles. The
exit plane pressure is, however, totally unrealistic. No one designing
an engine to proceed from sea level to space would expand to only 14.7
psi. A designer would, barring some specific operational reason to the
contrary, put on as much nozzle as possible to the threshold of flow
separation, which at sea level is about 5 psi. Finally, sea level Isp
is not the figure to watch, it's vacuum Isp. One properly uses vacuum
Isp, double checks the exit pressure at the ignition altitude to make
sure that flow separation is avoided, and assesses the losses due to
back pressure in the delta-V part of the calculation. POST, OTIS, and
similar codes do this, and it's really the standard for keeping the
bookkeeping straight. A 250 psi engine, properly sized for ignition at
sea level, would have a vacuum Isp of 285 seconds, at an expansion
ratio of 8, with an exit plane pressure of 5.111 psi, and the same
efficiency of 90%.
The delta-v figures asserted on page 4 of the paper have some
difficulties as well. The losses due to drag are understood by
aerodynamicists to depend on drag coefficient and reference area,
rather than a cube-square law. It's admittedly difficult to calculate
such losses from first principles, so vehicles of similar
configuration and resources such as DATCOM and other aerodynamic
databases are used. Statements such as "3/4 of drag losses are caused
by supersonic drag" and "gravity losses are on the order of 2000 to
4000 ft/s" are very trajectory dependent. The discussion of steering
loss neglects to mention aerodynamic steering and appears to assert
that a vehicle needs to turn its direction of flight to vertical if it
is horizontally launched. Back pressure losses are ignored here and
overaccounted in the discussion of Isp. Finally, vehicle thrust to
weight ratio is a design optimization depending on the engine
technology, the airframe technology, and many other factors, and can
be less than unity for horizontal takeoff vehicles (although in
general the performance is disappointing when it is).
The propellant mass fraction discussion asserts that historical trends
are valid because "Vehicle designers were highly motivated to increase
propellant mass fraction". Vehicle designers were in fact highly
motivated to meet requirements, such as combat survivability, cost,
risk, safety, performance, normal load factor, and many other things,
only one of which was propellant mass fraction. Requirements drive
design. If there's not a requirement for a 9 g sustained turn, the
wing can be lighter. If there's not a requirement for a 100 passenger
cabin, the fuselage can contain more fuel. Interestingly, the authors
omit mention of vehicles that are designed to be flying fuel tanks
much as rocket powered vehicles are: tankers. A KC-135 has a
propellant mass fraction of about 65% and the KC-10 of over 60%.
It would be possible to perform a similar critique of this paper for
the trade studies for each takeoff mode. Because there are problems
with the analysis of the ascent flight mechanics, however, these
issues should be considered and the paper revised in its entirety
before a review of the trade studies would have any constructive
validity.
In conclusion, the authors of the paper have made errors that
potentially undermine their conclusions in their consideration of
specific impulse, delta-V, and propellant mass fraction. The other
elements of the rocket equation are just fine.
Mitchell Burnside Clapp
John Schilling
>In their recent paper "Flight Mechanics of Manned Sub-Orbital Reusable
>Launch Vehicles with Recommendations for Launch and Recovery"
>(AIAA-2003-0909), available from
>http://mae.ucdavis.edu/faculty/sarigul/AIAA_2003_0909.pdf,
>the authors, Marti and Nesrin Sarigul-Klijn, make a number of
>inaccurate assumptions and errors that do not support the conclusions
>they have drawn.
[...]
>In conclusion, the authors of the paper have made errors that
>potentially undermine their conclusions in their consideration of
>specific impulse, delta-V, and propellant mass fraction. The other
>elements of the rocket equation are just fine.
Are you sure about that? Given the quality of the rest of their work,
I wouldn't put it past them to get the values of 'g' and 'e' wrong...
But I didn't bother to actually check the math. The claim that the
Concorde represented the Absolute Best Possible Mass Fraction for a
supersonic HTHL vehicle pegged my bogometer, and from that point I
was reading just for entertainment value.
--
*John Schilling * "Anything worth doing, *
*Member:AIAA,NRA,ACLU,SAS,LP * is worth doing for money" *
*Chief Scientist & General Partner * -13th Rule of Acquisition *
*White Elephant Research, LLC * "There is no substitute *
*schi...@spock.usc.edu * for success" *
*661-951-9107 or 661-275-6795 * -58th Rule of Acquisition *
Nesrin & Marti Sarigul-Klijn
commenting on it. We also thank one of our associates who pointed out
his review to us.
We note that a 250 psia LOX-kerosine engine, properly sized for
ignition at sea level, would have, as Mr. Clapp points out, a vacuum
Isp of 285 seconds, assuming a nozzle expansion ratio of 8, an exit
plane pressure of 5 psi, and an efficiency of 90%. This same 250 psia
engine has a sea level Isp of 225 seconds, which is what we state on
page 3 of our paper.
The figure on page 4 was prepared using a sea level Isp of 225 seconds
(not an average Isp). We thank Mr. Clapp for indirectly pointing out
this typo to us. The trajectory program used to prepare the figure on
page 4 properly adjusts Isp for altitude.
The atmosphere is not a friend to small launch vehicles. Drag depends
on reference area and drag coefficients, however, reference area
increases with the square of the vehicle's external dimensions whereas
internal volume (and hence mass) increases with the cube. For
example, if a vehicle's external dimensions are doubled, then
reference area is 4 times larger, but volume and mass are 8 times
greater. This means assuming the same drag coefficient, then the
amount of drag relative to the vehicle's mass is cut in half when, as
in this example, the external dimensions are doubled.
The KC-135 is subsonic. Historically for horizontal takeoff aircraft,
as maximum speed goes up, propellant mass fraction goes down. The
Concorde, at 51% propellant mass fraction, was the highest we could
find for an horizontal takeoff aircraft configured for supersonic
flight (not carrying external drop tanks).
Nesrin and Marti Sarigul-Klijn
UC Davis
mitc...@rocketplane.com (Mitchell Burnside Clapp) wrote in message news:<7a393d23.03060...@posting.google.com>...
Rand Simberg
Mar...@aol.com (Nesrin & Marti Sarigul-Klijn) made the phosphor on my
monitor glow in such a way as to indicate that:
>The
>Concorde, at 51% propellant mass fraction, was the highest we could
>find for an horizontal takeoff aircraft configured for supersonic
>flight (not carrying external drop tanks).
Highest that has been built != Highest that could be built.
--
simberg.interglobal.org * 310 372-7963 (CA) 307 739-1296 (Jackson Hole)
interglobal space lines * 307 733-1715 (Fax) http://www.interglobal.org
"Extraordinary launch vehicles require extraordinary markets..."
Swap the first . and @ and throw out the ".trash" to email me.
Here's my email address for autospammers: postm...@fbi.gov
Michael Walsh
Rand Simberg wrote:
> On 3 Jun 2003 15:30:52 -0700, in a place far, far away,
> Mar...@aol.com (Nesrin & Marti Sarigul-Klijn) made the phosphor on my
> monitor glow in such a way as to indicate that:
>
> >The
> >Concorde, at 51% propellant mass fraction, was the highest we could
> >find for an horizontal takeoff aircraft configured for supersonic
> >flight (not carrying external drop tanks).
>
> Highest that has been built != Highest that could be built.
True, but these data points are interesting. They are more important for
determining what can almost certainly be done because it has already
been demonstrated, but doesn't mean that it can't be bettered.
OK, I agree with you, in case my comment might be taken otherwise.
Mike Walsh
Jake McGuire
> The KC-135 is subsonic. Historically for horizontal takeoff aircraft,
> as maximum speed goes up, propellant mass fraction goes down. The
> Concorde, at 51% propellant mass fraction, was the highest we could
> find for an horizontal takeoff aircraft configured for supersonic
> flight (not carrying external drop tanks).
How hard did you look?
The B-1A is about 0.60, the XB-70 is higher still.
-jake
Ian Stirling
> Mar...@aol.com (Nesrin & Marti Sarigul-Klijn) wrote in message news:<cb83a01c.03060...@posting.google.com>...
>> The KC-135 is subsonic. Historically for horizontal takeoff aircraft,
>> as maximum speed goes up, propellant mass fraction goes down. The
>> Concorde, at 51% propellant mass fraction, was the highest we could
>> find for an horizontal takeoff aircraft configured for supersonic
>> flight (not carrying external drop tanks).
>
> How hard did you look?
SR-71, around 55%...
>
> The B-1A is about 0.60, the XB-70 is higher still.
>
> -jake
--
http://inquisitor.i.am/ | mailto:inqui...@i.am | Ian Stirling.
---------------------------+-------------------------+--------------------------
Windows 2000, software for next millenia. <latin pun alert> - Ian Stirling.
Edward Wright
>>>The
>>>Concorde, at 51% propellant mass fraction, was the highest we could
>>>find for an horizontal takeoff aircraft configured for supersonic
>>>flight (not carrying external drop tanks).
>>
>> Highest that has been built != Highest that could be built.
> True, but these data points are interesting. They are more important for
> determining what can almost certainly be done because it has already
> been demonstrated, but doesn't mean that it can't be bettered.
Although, if you want to use the Concorde as an example of "what can
almost certainly be done," you should consider the possibility of
removing some of the seats and replacing them with fuel tanks, which
would increase the propellant mass fraction.
I find it curious that the authors say they confined themselves to
looking at the mass fraction of supersonic aircraft without external
tanks. External tanks for suborbital vehicles are certainly not out of
the question, as the X-15A1 demonstrates.
Mitchell Burnside Clapp
> Jake McGuire <jamc...@yahoo.com> wrote:
> > Mar...@aol.com (Nesrin & Marti Sarigul-Klijn) wrote in message news:<cb83a01c.03060...@posting.google.com>...
>
> >
> > The B-1A is about 0.60, the XB-70 is higher still.
> >
> > -jake
Tu-160, also 60%
Nesrin & Marti Sarigul-Klijn
>
> SR-71, around 55%...
> >
> > The B-1A is about 0.60, the XB-70 is higher still.
Thank you for your continuing interest in our work.
XB-70: Takeoff weight of 542,000 lbs and empty weight of 300,000 lbs.
If we assume all of the difference is fuel, the propellant mass
fraction is 44.6%. Source:
http://www.aerospaceweb.org/aircraft/research/xb70/index.shtml
SR-71: Takeoff weight of 170,000 lb and fuel weight of 80,280 lbs
gives a propellant mass fraction of 47.2%. The SR-71 usual takeoff
fuel load was 45,000 to 65,000 lbs due to fuel leakage, tire and brake
heating, abort criteria, and single engine performance. It then
aerial refueled. Source: Jenkins, Dennis, "SR-71/YF-12 Blackbirds,"
Wartech Series, Volume 10.
We did not previously look at the B-1B because of a conversation that
Marti had USAF Major Baltrusaitus, B-1 Test Pilot at Edwards AFB in
1995. He stated that the B-1B was not capable of supersonic speeds
nor could it climb above 20,000 feet with full fuel and internal
weapons.
B-1B: Maximum takeoff weight of 477,000 lbs, Empty Weight 190,000
lbs, Weapons load 75,000 lbs (internal only) to 134,000 lbs (plus
external). Fuel load estimated at 212,000 lbs. Propellant mass
fraction of 44.4% at maximum takeoff weight or 52.7% without weapons.
Source: http://www.combataircraft.com/aircraft/bb1.asp
Note that many supersonic jet aircraft can have propellant mass
fractions that exceed 50% if they carry external ferry tanks, but then
they can't fly supersonic in this configuration.
Nesrin and Marti Sairgul-Klijn
Henry Spencer
Nesrin & Marti Sarigul-Klijn <Mar...@aol.com> wrote:
>> > The B-1A is about 0.60, the XB-70 is higher still.
>
>Marti had USAF Major Baltrusaitus, B-1 Test Pilot at Edwards AFB in
The B-1A, however, was supersonic. The B-1A and B-1B are different
aircraft.
--
"Bin Laden must be laughing his beard off." | Henry Spencer
-- Wes Oleszewski | he...@spsystems.net
Mary Shafer
wrote:
> In article <cb83a01c.03060...@posting.google.com>,
> Nesrin & Marti Sarigul-Klijn <Mar...@aol.com> wrote:
> >> > The B-1A is about 0.60, the XB-70 is higher still.
> >
> >...We did not previously look at the B-1B because of a conversation that
> >Marti had USAF Major Baltrusaitus, B-1 Test Pilot at Edwards AFB in
> >1995. He stated that the B-1B was not capable of supersonic speeds...
>
> The B-1A, however, was supersonic. The B-1A and B-1B are different
> aircraft.
The B-1A was supersonic. The B-1B can exceed Mach 1, but it's still
transonic, and it only does it going downhill with the wind at its
back. Maybe 1.1 at best, but not on the deck. That's not supersonic.
The B-1A and B-1B are very much the same aircraft, actually. The
airframes are virtually indistinguishable. However, the inlets are
quite different. The A had variable inlet ramps and the B has fixed
ramps.
Mary
--
Mary Shafer mil...@qnet.com
Retired aerospace research engineer
"The guy you don't see will kill you." BGEN Robin Olds, USAF
Rand Simberg
Shafer <mil...@qnet.nospam.com> made the phosphor on my monitor glow
in such a way as to indicate that:
>> >...We did not previously look at the B-1B because of a conversation that
>> >Marti had USAF Major Baltrusaitus, B-1 Test Pilot at Edwards AFB in
>> >1995. He stated that the B-1B was not capable of supersonic speeds...
>>
>> The B-1A, however, was supersonic. The B-1A and B-1B are different
>> aircraft.
>
>The B-1A was supersonic. The B-1B can exceed Mach 1, but it's still
>transonic, and it only does it going downhill with the wind at its
>back. Maybe 1.1 at best, but not on the deck. That's not supersonic.
>
>The B-1A and B-1B are very much the same aircraft, actually. The
>airframes are virtually indistinguishable. However, the inlets are
>quite different. The A had variable inlet ramps and the B has fixed
>ramps.
So, it sounds like a propulsion problem, which is only weakly related
to mass fraction.
Scott Lowther
>
> On Wed, 04 Jun 2003 18:58:44 -0700, in a place far, far away, Mary
> Shafer <mil...@qnet.nospam.com> made the phosphor on my monitor glow
> in such a way as to indicate that:
> >The B-1A and B-1B are very much the same aircraft, actually. The
> >airframes are virtually indistinguishable. However, the inlets are
> >quite different. The A had variable inlet ramps and the B has fixed
> >ramps.
>
> So, it sounds like a propulsion problem, which is only weakly related
> to mass fraction.
It's not a problem, it's a feature. The B-1A inlet was optimized to go
like a bat outta hell, but the B-1B was optimized for stealth. Radar
absorbing vanes were installed within the inlets that blocked direct
view of the turbine blades; RCS dropped through the floor, but so did
speed. It was considered a worthwhile tradoff.
--
Scott Lowther, Engineer
"Any statement by Edward Wright that starts with 'You seem to think
that...' is wrong. Always. It's a law of Usenet, like Godwin's."
- Jorge R. Frank, 11 Nov 2002
Rand Simberg
Lowther <lex...@ix.netcom.com> made the phosphor on my monitor glow
in such a way as to indicate that:
>> >The B-1A and B-1B are very much the same aircraft, actually. The
>> >airframes are virtually indistinguishable. However, the inlets are
>> >quite different. The A had variable inlet ramps and the B has fixed
>> >ramps.
>>
>> So, it sounds like a propulsion problem, which is only weakly related
>> to mass fraction.
>
>It's not a problem, it's a feature. The B-1A inlet was optimized to go
>like a bat outta hell, but the B-1B was optimized for stealth. Radar
>absorbing vanes were installed within the inlets that blocked direct
>view of the turbine blades; RCS dropped through the floor, but so did
>speed. It was considered a worthwhile tradoff.
The point is, it wasn't a mass-fraction issue.
Scott Lowther
>
> On Wed, 04 Jun 2003 20:02:17 -0700, in a place far, far away, Scott
> Lowther <lex...@ix.netcom.com> made the phosphor on my monitor glow
> in such a way as to indicate that:
>
> >> >The B-1A and B-1B are very much the same aircraft, actually. The
> >> >airframes are virtually indistinguishable. However, the inlets are
> >> >quite different. The A had variable inlet ramps and the B has fixed
> >> >ramps.
> >>
> >> So, it sounds like a propulsion problem, which is only weakly related
> >> to mass fraction.
> >
> >It's not a problem, it's a feature. The B-1A inlet was optimized to go
> >like a bat outta hell, but the B-1B was optimized for stealth. Radar
> >absorbing vanes were installed within the inlets that blocked direct
> >view of the turbine blades; RCS dropped through the floor, but so did
> >speed. It was considered a worthwhile tradoff.
>
> The point is, it wasn't a mass-fraction issue.
Well, yes. Ditch the ECM systems, the ejector seats, the other
military-specific items, and mass fraction improves again.
Rand Simberg
Lowther <lex...@ix.netcom.com> made the phosphor on my monitor glow
in such a way as to indicate that:
>> The point is, it wasn't a mass-fraction issue.
>
>Well, yes. Ditch the ECM systems, the ejector seats, the other
>military-specific items, and mass fraction improves again.
Those have what to do with inlets, again?
Jake McGuire
> If we assume all of the difference is fuel, the propellant mass
> fraction is 44.6%. Source:
> http://www.aerospaceweb.org/aircraft/research/xb70/index.shtml
Every other source I have found in a brief online seach shows the
XB-70 empty weight at around 200,000 pounds, not 300,000.
http://www.google.com/search?q=xb-70+valkyrie+empty+weight
> B-1B: Maximum takeoff weight of 477,000 lbs, Empty Weight 190,000
> lbs, Weapons load 75,000 lbs (internal only) to 134,000 lbs (plus
> external). Fuel load estimated at 212,000 lbs. Propellant mass
> fraction of 44.4% at maximum takeoff weight or 52.7% without weapons.
> Source: http://www.combataircraft.com/aircraft/bb1.asp
If you are trying to determine the maximum empty weight fraction
possible, why on earth would you include externally mounted weapons?
I further note that you are not considering mounting the weapon-bay
fuel tanks, which would push the propellant mass fraction up above
55%.
Or more to the point, in 1943 (!) there was a production (!) wooden
(!) aircraft with a MTOW of less than 10,000 pounds (!) with a 50%
propellant weight fraction and the power to go supersonic [1]. The
knowledge of aerodynamics wasn't up to par so it didn't quite make it,
but it's still an example of what can be done under hugely adverse
conditions. I'd hope that we can do much better now.
The issues with your analysis of horizontal takeoff vehicles have been
pointed out in other messages; but I'll just reiterate that you really
need to do an ascent analysis that takes into account thrust, drag,
lift, and gravity. Such an analysis doesn't have to be complex; there
is a great section in the book "Aircraft Design: A Conceptual
Approach" that has all of the math that you need. Keep in mind that
rockets will give you a much higher T/W than any aircraft, and that
optimizing for an acceleration rather than a cruise mission will lead
the design in a different direction than most aircraft.
Finally, there are at least two companies that disagree with your
conclusion that horizontal takeoff aircraft cannot meet the X-Prize
requirements:
http://www.rocketplane.com/tech1.html
http://www.xcor.com/suborbital.html
It could be instructive to ask why they disagree with you. The
founder of Pioneer Rocketplane started this thread and the founder of
XCOR (Jeff Greason) is a frequent contributor to sci.space.policy.
[1] The Me-163B Komet, of course. You even reference it later in your
paper.
Ian Stirling
>>
>> SR-71, around 55%...
>> >
>> > The B-1A is about 0.60, the XB-70 is higher still.
>
> Thank you for your continuing interest in our work.
> SR-71: Takeoff weight of 170,000 lb and fuel weight of 80,280 lbs
> gives a propellant mass fraction of 47.2%. The SR-71 usual takeoff
> fuel load was 45,000 to 65,000 lbs due to fuel leakage, tire and brake
> heating, abort criteria, and single engine performance. It then
> aerial refueled. Source: Jenkins, Dennis, "SR-71/YF-12 Blackbirds,"
> Wartech Series, Volume 10.
"as maximum speed goes up, mass fraction goes down".
Which was my point, as the SR71 has pretty close to the same
mass fraction as concorde, but goes half again as fast.
--
http://inquisitor.i.am/ | mailto:inqui...@i.am | Ian Stirling.
---------------------------+-------------------------+--------------------------
Acting is merely the art of stopping a large number of people from coughing
- Sir Ralph Richardson
George William Herbert
>> >1995. He stated that the B-1B was not capable of supersonic speeds...
>>
>> The B-1A, however, was supersonic. The B-1A and B-1B are different
>> aircraft.
>
>The B-1A was supersonic. The B-1B can exceed Mach 1, but it's still
>transonic, and it only does it going downhill with the wind at its
>back. Maybe 1.1 at best, but not on the deck. That's not supersonic.
>
>The B-1A and B-1B are very much the same aircraft, actually. The
>airframes are virtually indistinguishable.
There's a lot of reinforcement in the Wing Carrythrough Box and
more fuel tanks in the B. But I am reasonably sure the mold lines
are all the same.
> However, the inlets are
>quite different. The A had variable inlet ramps and the B has fixed
>ramps.
Yep. B's also use ejection seats instead of the module.
I was looking at a B-1B with one of the B-1A intake sets
grafted back on as a RASCAL carrier vehicle. Sick and wrong.
-george william herbert
gher...@retro.com
Scott Lowther
>
> On Wed, 04 Jun 2003 21:33:22 -0700, in a place far, far away, Scott
> Lowther <lex...@ix.netcom.com> made the phosphor on my monitor glow
> in such a way as to indicate that:
>
> >> The point is, it wasn't a mass-fraction issue.
> >
> >Well, yes. Ditch the ECM systems, the ejector seats, the other
> >military-specific items, and mass fraction improves again.
>
> Those have what to do with inlets, again?
Rand: I've said it before. You'll have to stop looking for excuses to
argue. It's undignified. Makes you look like a Nader supporter.
Mary Shafer
Herbert) wrote:
> Mary Shafer <mil...@qnet.com> wrote:
> >> >1995. He stated that the B-1B was not capable of supersonic speeds...
> >The B-1A and B-1B are very much the same aircraft, actually. The
> >airframes are virtually indistinguishable.
> There's a lot of reinforcement in the Wing Carrythrough Box and
> more fuel tanks in the B. But I am reasonably sure the mold lines
> are all the same.
It's an airplane. Those are loft lines, not mold lines. Aircraft
were originally drawn in lofts, not molded. You should see the
splines they used for the full-sized drawings.
Nesrin & Marti Sarigul-Klijn
> possible,
Lets try to put the propellant mass fraction argument to bed. There
is a great figure in Raymer's book "Aircraft Design, A Conceptual
Approach" on page 17 in the third edition titled "Fig. 3.1 Empty
Weight Fraction Trends." Also Dr. Jan Roskam 8 volume set "Airplane
Design" is great. Refer to pages 19 through 46 in Volume 1. He plots
the actual aircraft data points for his trend lines.
The takeoff weight of an aircraft is the sum of the crew weight,
payload weight, fuel weight, and empty aircraft weight. These can be
expressed as percentages or fractions of the takeoff weight. So as an
example, consider the Concorde again, the crew weight is much less
than 1%, the payload (passengers and bags) are about 7%, the fuel is
51%, and the empty weight is 42% of the takeoff weight. Can the
passengers, passengers seats, and bags be left on the ground, and fuel
tanks installed in the cabin to increase the fuel fraction? Sure.
Propellant mass would go up to about 58 to 59%. Great, we got a
400,000 lb plane that carries 3 people.
We can do the same thing to the Tu-160 Blackjack and the B-1 Bomber,
i.e., replace the weapons with fuel in located in tanks inside the
bomb bay and external tanks. Yep. Might get propellant mass fraction
up to 0.60.
Now take a look at Raymer's figure 3-1. He says as much since he show
an empty weight fraction of about 0.40 for military cargo and bombers.
Subtract 0.40 from 1 gives 0.60 for fuel, crew and payload.
However no one is proposing to build a Concorde or B-1 bomber sized
X-Prize vehicle that carries 3 people. Most currently proposed
X-prize vehicles have takeoff weights under 20,000 lbs with some above
that. According to Raymer's figure 3.1, at 20,000 lbs, subsonic jet
transports have empty weight fractions of 0.55 and jet fighters are at
0.65. That leaves 0.35 to 0.45 for fuel and crew. At 600 lbs for 3
people and assuming a 20,000 lb X-Prize aircraft the crew fraction is
0.03. That then that leaves 0.32 to 0.42 for the fuel fraction.
For example consider the T-38: 7,247 lbs empty (62%), 426 lbs crew
(3.6%), and 3,916 lbs fuel (33.6%), and flight design gross weight of
11,651 lbs. Source: Volume V, Airplane Design, Dr. Jan Roskam.
Note that propellant mass fraction decreases with decreasing takeoff
weight. Why? Because items like crew, seats, flight instruments,
windows, etc., don't scaled down as the aircraft is scaled down.
Conclusion: Lets go back to the statement in our paper that is the
focus of this thread: "Although not a combined powered vehicle, the
Concorde's propellant mass fraction of 0.51 probably represents the
maximum internal fuel load possible for a supersonic horizontal
takeoff concept." Perhaps if we had added the statement "for a 3
person aircraft sized for the X-Prize" we would not have had this
discussion. We think that 0.51 is an optimistic propellant mass
fraction for a horizontal takeoff X-prize vehicle when 0.32 to 0.42 is
norm for a 20,000 lb class vehicle. Incidentally the Me-163 had a
propellant mass fraction of 0.37, which could be improved by removing
its two 30 mm guns and amour plate.
> The issues with your analysis of horizontal takeoff vehicles have been
> pointed out in other messages; but I'll just reiterate that you really
> need to do an ascent analysis that takes into account thrust, drag,
> lift, and gravity. Such an analysis doesn't have to be complex; there
> is a great section in the book "Aircraft Design: A Conceptual
> Approach" that has all of the math that you need. Keep in mind that
> rockets will give you a much higher T/W than any aircraft, and that
> optimizing for an acceleration rather than a cruise mission will lead
> the design in a different direction than most aircraft.
We have an analysis in work that is close to what you describe and are
in the process of writing a paper for the AIAA. Obviously the
propellant mass fraction required to climb to a particular altitude
will be a function of drag (which varies with velocity, air density,
drag reference area, and drag coefficient which in turn varies with
Mach number, angle of attack, and Reynolds number), climb angle
profile (which will be continuously changing), velocity (which varies
continuously), engine specific impulse (which varies with attitude,
propellant combination, and chamber pressure), climb velocity (which
varies with continuously), and thrust to weight ratio (both
continuously changing).
However we think we have discovered a simple method that allows one to
directly compare a vertical trajectory that is fully borne by thrust
as compared to a climb trajectory that is borne by both wings and
thrust.
Thanks again for the interest in our work.
Nesrin & Marti Sarigul-Klijn
Michael Walsh
Edward Wright wrote:
Well, the paper was a rather broad-brush approach.
I don't fault them for not covering all examples, but their
brush off of horizontally launched vehicles, whether rocket
or airbreathing propelled for a relatively low performance
goal such as the X-Prize did not sound convincing at all,
especially since they relied on previous history rather than
presenting any design details showing why it was not a
viable approach.
For larger launch vehicles with higher delta V requirements
things like landing gears for high launch weights make things
difficult unless you go to things like launch sleds or
refueling approaches after takeoff.
Mike Walsh
Michael Walsh
Rand Simberg wrote:
> On Wed, 04 Jun 2003 18:58:44 -0700, in a place far, far away, Mary
> Shafer <mil...@qnet.nospam.com> made the phosphor on my monitor glow
> in such a way as to indicate that:
>
> >> >...We did not previously look at the B-1B because of a conversation that
> >> >Marti had USAF Major Baltrusaitus, B-1 Test Pilot at Edwards AFB in
> >> >1995. He stated that the B-1B was not capable of supersonic speeds...
> >>
> >> The B-1A, however, was supersonic. The B-1A and B-1B are different
> >> aircraft.
> >
> >The B-1A was supersonic. The B-1B can exceed Mach 1, but it's still
> >transonic, and it only does it going downhill with the wind at its
> >back. Maybe 1.1 at best, but not on the deck. That's not supersonic.
> >
> >The B-1A and B-1B are very much the same aircraft, actually. The
> >airframes are virtually indistinguishable. However, the inlets are
> >quite different. The A had variable inlet ramps and the B has fixed
> >ramps.
>
> So, it sounds like a propulsion problem, which is only weakly related
> to mass fraction.
Well, depending on the weight difference between the variable and the
fixed ramp inlets.
Although, come to think of it, that does sound as if it probably would
be weakly related to mass fraction.
Started to post before completing the thought.
Mike Walsh
Edward Wright
> Lets try to put the propellant mass fraction argument to bed. There
> is a great figure in Raymer's book "Aircraft Design, A Conceptual
> Approach" on page 17 in the third edition titled "Fig. 3.1 Empty
> Weight Fraction Trends." Also Dr. Jan Roskam 8 volume set "Airplane
> Design" is great. Refer to pages 19 through 46 in Volume 1. He plots
> the actual aircraft data points for his trend lines.
> The takeoff weight of an aircraft is the sum of the crew weight,
> payload weight, fuel weight, and empty aircraft weight. These can be
> expressed as percentages or fractions of the takeoff weight.
That, however, is a historical relationship. It shows correlation,
which does not imply causality. Two aircraft with identical takeoff
weights can have very different crew, payload, fuel, and empty
weights. Voyager, as an example, carried a small crew, no payload to
speak of, but lots of fuel, because that's what the mission required.
Another aircraft with the same takeoff weight would have much less
weight devoted to fuel and more to crew, fuel, or structure (depending
on mission).
Since larger aircraft have different missions, that in itself could
explain the statistical differences.
> So as an examle, consider the Concorde again, the crew weight is much less
> than 1%, the payload (passengers and bags) are about 7%, the fuel is
> 51%, and the empty weight is 42% of the takeoff weight. Can the
> passengers, passengers seats, and bags be left on the ground, and fuel
> tanks installed in the cabin to increase the fuel fraction? Sure.
> Propellant mass would go up to about 58 to 59%. Great, we got a
> 400,000 lb plane that carries 3 people.
Which might be perfectly acceptable, depending on the mission. One
shouldn't reject a design simply because it falls outside of
historical curves, especially if it's designed for a new mission not
represented by those curves.
> Note that propellant mass fraction decreases with decreasing takeoff
> weight. Why? Because items like crew, seats, flight instruments,
> windows, etc., don't scaled down as the aircraft is scaled down.
Huh? If you scaled down from a Concorde to a T-38, the weight of
windows and seats certainly do scale down. So does the crew weight,
because the Concorde requires a crew of three, while the T-38 has a
crew of one or two (depending on whether you consider a student pilot
crew or payload). I suspect the Concorde carries considerable more
instruments, as well. Since you're relying on Raymer for weights, you
should note Table 11.6 which gives avionics weights as a function of
aircraft weight and mission type.
johnhare
"Edward Wright" <edwrig...@hotmail.com> wrote in message
news:32b558f9.03060...@posting.google.com...
> Mar...@aol.com (Nesrin & Marti Sarigul-Klijn) wrote in message
news:<cb83a01c.03060...@posting.google.com>...
>
> > payload weight, fuel weight, and empty aircraft weight. These can be
> > expressed as percentages or fractions of the takeoff weight.
>
> That, however, is a historical relationship. It shows correlation,
> which does not imply causality. Two aircraft with identical takeoff
> weights can have very different crew, payload, fuel, and empty
> weights. Voyager, as an example, carried a small crew, no payload to
> speak of, but lots of fuel, because that's what the mission required.
> Another aircraft with the same takeoff weight would have much less
> weight devoted to fuel and more to crew, fuel, or structure (depending
> on mission).
>
normally have very heavy airbreathing engines and their auxilery
systems. Replace a 10,000 lb jet, with 10,000 more lbs of inlet
and afterburner, with a 1,000 lb rocket engine that produces the
same 100k of thrust, and you drastically change the dry fraction
of a >100,000 lb supersonic vehicle.
Additionally the drag producing inlet is forward of the mass of the
air breathing engine, requiring some attention to stability issues above
and beyond the rocket case, which will sometimes lead to more aft
aero surfaces to compensate. The extra bending moments of the
aero surfaces on the fusilage can generate more structure. Not
to mention the asimetrical loads of the airbreathing systems on the
vehicle. History is a start point, not an end restriction.
I'll see if I remembered how to post instead of mail this time.
Nesrin & Marti Sarigul-Klijn
historical curves, especially if it's designed for a new mission not
represented by those curves.>
Thanks for taking the time of comment on our paper. Since the 1960s
(after the introduction of composites and the finite element theory)
there has been very little change in empty weight fractions of
aerospace vehicles. Once categorized by type (i.e., bomber, jet
trainer, single engine general aviation, etc.) historical empty weight
fractions all fall along very narrow trend lines that shows decreasing
empty weight fractions with increasing takeoff gross weight.
For new designs in which there aren't any historical trend lines
established yet, engineering judgment can be used to interpolate from
other trend lines. For example, consider the X-15 and the X-34.
These two aircraft designs are separated by 40 years, one is mostly
metal and the other is mostly composite, and they have similar
missions. The X-15 empty weight fraction is 0.364 (11,374 lbs empty /
31,275 lb gross) and the X-34 empty weight fraction is 0.362 (17,000
lbs empty / 47,000 lb gross). Note that there is not much difference
in the empty weight fraction, even though the X-34 has a higher gross
weight.
Also note that these fractions are smaller than horizontal jet powered
takeoff aircraft, mostly due to their lighter landing gear (sized for
landing at empty weight only), smaller wings (don't have to be sized
for a runway takeoff), and rocket power (rocket engines are lighter
than jets for similar thrust). Hence based on engineering judgment,
horizontal takeoff rocket powered aircraft should have higher empty
weight fractions as compared to these air launched aircraft. Landing
gear and wings will have to be larger and heavier.
The empty weight fraction for a combined powered aircraft that uses
jets for takeoff and then switches to rockets can also be estimated as
higher than these air launched aircraft. Taking off a combined
powered aircraft empty and then aerial refueling is similar to air
launching in terms of aerodynamic loads, wing size, and landing gear,
however the combined powered aircraft carries the additional weight of
the jet engines, inlets, and fuel tanks, so its empty weight fraction
should be higher.
>Huh? If you scaled down from a Concorde to a T-38, the weight of
windows and seats certainly do scale down.<
They do, but not linearly. The T-38 is over 30 times smaller than the
Concorde. The pilots, pilot seats, windows, cabin, etc., are not 30
times smaller. T-38 pilots don't weigh 7 pounds :) There are also
other components, not related to the pilot, that do not scale down
linearly. That is why decreasing takeoff gross weight typically
increases empty weight fraction.
Marti & Nesrin Sarigul-Klijn
Edward Wright
> Note that there is not much difference
> in the empty weight fraction, even though the X-34 has a higher gross
> weight.
Why should there be? Mass fraction is determined by mission
requirements, not the desire to set records. The X-34 didn't require a
better mass fraction for its mission.
Note that you just contradicted your own statement about the
relationship between mass fraction and vehicle size.
> Hence based on engineering judgment,
> horizontal takeoff rocket powered aircraft should have higher empty
> weight fractions as compared to these air launched aircraft. Landing
> gear and wings will have to be larger and heavier.
Let's test that hypothesis.
The air-launched X-15 had an empty weight of 13,000 pounds, gross
weight of 34,000 pounds, and a wing area of 200 square feet.
The ground-launched F-104 had an empty weight of 14,082 pounds, gross
weight of 28,779 pounds, and a wing area of 196 square feet.
I also note that when Scott Crossfield was forced to land the X-15
with full tanks, he broke the spine of the airplane, but the landing
gear did just fine. So, it appears that neither the wing nor the
landing gear was sized by the empty weight.
>> Huh? If you scaled down from a Concorde to a T-38, the weight of
>> windows and seats certainly do scale down.
> They do, but not linearly. The T-38 is over 30 times smaller than the
> Concorde. The pilots, pilot seats, windows, cabin, etc., are not 30
> times smaller. T-38 pilots don't weigh 7 pounds :) There are also
> other components, not related to the pilot, that do not scale down
> linearly. That is why decreasing takeoff gross weight typically
> increases empty weight fraction.
You are making some very odd assumptions. Surely you know that the
Concorde carries many more people than the T-38 and that pilots are
not the only people who get seats and windows.
Rand Simberg
Mar...@aol.com (Nesrin & Marti Sarigul-Klijn) made the phosphor on my
monitor glow in such a way as to indicate that:
>> If you are trying to determine the maximum empty weight fraction
>> possible,
>
>Lets try to put the propellant mass fraction argument to bed. There
>is a great figure in Raymer's book "Aircraft Design, A Conceptual
>Approach" on page 17 in the third edition titled "Fig. 3.1 Empty
>Weight Fraction Trends." Also Dr. Jan Roskam 8 volume set "Airplane
>Design" is great. Refer to pages 19 through 46 in Volume 1. He plots
>the actual aircraft data points for his trend lines.
<snip>
I got this response from an emailer who wishes to remain anonymous,
but has some knowledge in this area.
<begin email response>
Yes, indeed, let's try to "put the propellant mass fraction argument
to bed".
Raymer and Roskam's books are excellent -- but the tables you refer to
are all statistically derived. In other words, they are a
distillation of what has been done in similar vehicles in the past.
By definition, if you change nothing, you can expect similar results.
Predictions like that are intended for rough cut sizing of a new
vehicle; they indicate areas where, for example, if you need a better
mass fraction, you know you have to change something compared to the
prior art. They are not intended to be laws of physics -- nor are
they represented as such. For example, immediately prior to the graph
in Raymer's book, he says:
"To get the 'right' answer takes several years, many people, and lots
of money. Actual design requirements must be evaluated against a
number of candidate designs, each of which must be designed, analyzed,
sized, optimized, and redesigned any number of times" ....
"The simplified sizing method presented in this chapter can only be
used for missions which do not include any combat or payload drops.
While admittedly crude, this method introduces all of the essential
features of the most sophisticated design..."
As one simple example, if I take the "composite: homebuilt" line in
that book and project it to 10,000 lb takeoff weight, this method
predicts a empty weight fraction of about 0.43. That is the type of
vehicle with the lowest weight projection of any of those presented in
that size range.
Let’s take one example, a composite homebuilt line in that book and
assume a 10,000 lb take off weight, this method predicts a empty
weight fraction of about 0.43. That is the type of vehicle with the
lowest weight projection of any of those presented in that range of
sizes.
The best example I can think of right now hangs in the National Air
and Space Museum, right over the entrance. It’s called the Voyager and
it flew around the world without stopping and without refueling, and
carried two people. Going to the NASM web site we find from
http://www.nasm.si.edu/galleries/gal108/gal108.html
That the Voyager had a gross weight of 9695 lb and an empty weight of
2250 lb. That makes it an empty weight fraction of 0.23, or just over
half the dry weight projected from the statistical method.
Look at the photo. This is an aircraft carrying a lot of wingspan,
which was necessary for its mission. It had to achieve the highest
possible cruise lift/drag ratio. It should be clear that suborbital
flight does not require that much wingspan. Neither is it reasonable
to assume that a rocket powered system can’t have better thrust to
weight ratio than a piston engine driving a propeller.
While a suborbital vehicle is not a subsonic homebuilt, it isn’t a jet
fighter either. Jet fighters carry a bunch of subsystems not needed
for a suborbital flight. And jet fighter engines are usually *much*
heavier than a suborbital vehicle would need.
But if it is fair to assume from the one case that this is impossible,
it is just as fair to assume from the other that it is possible.
<end email response>
Nesrin & Marti Sarigul-Klijn
tanks, he broke the spine of the airplane, but the landing gear did
just fine. So, it appears that neither the wing nor the landing gear
was sized by the empty weight.<<
components that are sized by structural loads. However when Jack
McKay was forced to land the X-15 with excess fuel, the landing gear
failed first.
>> ... the "composite: homebuilt" line ... predicts a empty weight
fraction of about 0.43 ... the Voyager had ... an empty weight
fraction of 0.23, or just over half the dry weight projected from the
We are not ignorant that light weight aircraft are possible for
certain missions. For example, foot launched paragliders have empty
weight fractions below 0.2. Also UC Davis has been participating in
annual SAE sponsored university aircraft design, construction, and fly
off competitions for the past 9 years. The goal is to carry the
largest possible payload while powered by a single 0.61 model airplane
engine. UC Davis students have built 15 different aircraft and Nesrin
has been the faculty advisor. Their best aircraft carried 27.5 lbs of
payload with an empty weight of only 5.45 lbs, or an empty weight
fraction of 0.165 - better than Voyager - and had a 12 ft wing span.
To achieve such a low empty weight fraction, CFD, finite elements,
CAD, aeroelastic design methods, flight simulation, vacuum bagged
graphite composites, and payload placed span-loaded across the wing
were used.
These models share the flight characteristics of Voyager. Namely,
very low rate of climb, large deflections in the wings and fuselage
that have low frequency bending modes, long takeoff rolls, very small
difference between stall speed and aeroelastic flutter speed, and very
small structural safety margin beyond what is needed for straight and
level flight. When fully loaded, they can breakup in-flight if banked
too steeply, flown too fast, pulled-up too sharply, flown in turbulent
air, or if the pilot's control inputs get out of phase with the
deflections of the fuselage and wing.
>> ...While a suborbital vehicle is not a subsonic homebuilt ...<<
We agree. An X-prize vehicle will need a high rate of climb and
stiff structure in which the fuselage and wing bending modes don't
couple into the controls, can withstand a pull-up right after takeoff,
turbulent air, supersonic flight, relatively high aerodynamic
pressure, and have a reasonable flutter speed - the sort of
characteristics demonstrated by aircraft that have higher empty weight
fractions.
>> ...Jet fighters carry a bunch of subsystems not needed for a
suborbital flight ...<<
We agree. The supersonic T-38 does not carry extra subsystems
(except for ejection seats) and still has an empty mass fraction of
0.62.
To summarize, air launch requires the smallest delta V and
propellant mass fraction to reach 100 km, followed by vertical
takeoff, and then horizontal takeoff. Our analysis shows that a rapid
pitch up to a vertical trajectory is needed for either horizontal air
or ground launch to minimize both delta V and propellant mass
fraction. Vertical takeoff vehicles can carry the largest propellant
mass fraction, followed by air launch, and horizontal takeoff can
carry the least. Hence, horizontal ground takeoff is the most
difficult concept - it requires the largest propellant mass fraction
and has the least capability to carry it using current technology.
Nesrin & Marti Sarigul-Klijn
Dave Fowler
specialist astronaut interviewee in 1985.
Dave
Edward Wright
>>> ...Jet fighters carry a bunch of subsystems not needed for a
>>> suborbital flight ...
> We agree. The supersonic T-38 does not carry extra subsystems
> (except for ejection seats) and still has an empty mass fraction of
> 0.62.
The T-38 has quite a few subsystems not needed for a 20-minute
suborbital flight. Replacing the jet engines and inlets with
propellant tanks would significantly change the empty mass fraction.
By the way, the T-38 also has less wing area than the air-launched
X-15.
> To summarize, air launch requires the smallest delta V and
> propellant mass fraction to reach 100 km, followed by vertical
> takeoff, and then horizontal takeoff. Our analysis shows that a rapid
> pitch up to a vertical trajectory is needed for either horizontal air
> or ground launch to minimize both delta V and propellant mass
> fraction. Vertical takeoff vehicles can carry the largest propellant
> mass fraction, followed by air launch, and horizontal takeoff can
> carry the least. Hence, horizontal ground takeoff is the most
> difficult concept - it requires the largest propellant mass fraction
> and has the least capability to carry it using current technology.
Now, that might be a defensible position. It's unfortunate that you
did not take it originally. I generally find papers that explore the
pros and cons of various approaches more useful than those that make
categorical statements about things being physically impossible.