TRANSLATION CGD-32

A ROCKET DRIVE
TOR
LONG RANGE BOMBERS
(Tiber elnen Raketenantrleb for Fernboober)
by
E. Sanger and J. Bredt
Ai nr i ng, Aug "la t 1944
Deutsche Luftfahrtforschung
UM 3538
Translated by M. HAMERMESH
RADIO RESEARCH LABORATORY
Reproduced by
TECHNICAL INFORMATION BRANCH
BUAER NAVY DEPARTMENT c- 84296
A Note by the Publisher
Toward the end of the last century a. few farsighted individuals became
t thoroughly convinced that man could f l y. Today their names are al l but
forgotten bat their technical achievements wi l l endure for centuries.
Today we are on the threshold of Banned flight between the planets*
Drs. Eugen Sanger and Irene Bredt are prominent among the handful of pioneers
whose dedicated efforts have made possible thi s vi sta.
"Uber einen Baketenantrieb fur Fernbomber" i s based on more than a decade
of effort by the authors* The material i s condensed. This report contains
, only about one-third of ths Information which the authors had available at the
tine of writing j al l of the mathematical derivations and much of the support-
ing and supplementary information were omitted.
In spite of t hi s fact, the report i s , in effect, s definitive treatise.
It catalogs new problems and outlines solutions t o the more important ones.
For years to come i t wi l l serve as a storehouse of vi tal concepts for the
serious student of rocket science. For these reasons, i t s publication at
thi s time seems warranted.
Since 19U5, Dr. Irene Bredt (now sSngsr-Bredt) and Dr. Eugen Sanger
have ^ived in Paris, where they are employed by the Arsenal de l'Aeronautique.
Dr. Sanger i s also president of the International Astronautieal Federation.

sidle the Technical Information Branch, BUAER, Navy Department, has very
generously furnished copies of their translation of "Uber einen Raketenantrieb
t fur Fernbomber" t o many public libraries and research institutions, thi s i s
the fi rst time the report has been available for public sal s. The publisher
would like to thank the u\ S. Havy, without whose permission thi s publication
would not have been possible.
Robert Cornog
Santa Barbara, California
16 November 1952
FOREWORD
The application of pure rocket propulsion t o aeronautics suffers at present froa, l i mi t a-
tions imposed on exhaust -speed and fl i ght velocity by constructional di f f i cul t i es .
Because of the thermal st resses on the engines, the exhaust speed i s not raised to the
physically possible l i mi t s.
Because of the mechanical st resses on the airframe, the velocity of fl i ght has not yet
gone beyond the velocity of sound.
Cki the basis of extensive physical and physico-chemical st udi es, we shal l discuss some
possi bi l i t i es which are opened for the rocket propulsion of long-range mi l i t ary ai r cr af t when
these two l i mi t s are surpassed.
In addition several suggestions as t o construction are made, which should f aci l i t at e over-
coming the present l i mi t at i ons.
These investigations on thi> problem of long-range mi l i t ary rocket ai rcraft originated as
a j oi nt work of the two authors during the years 1937-1941 and were intended, together with the
material of report LM-3509, t o be a second volume of "Hocket Flight Technique", by the senior
author.
As a r esul t of circumstances caused by the war, publication was postponed and the r esul t s
of the work, issued in abst ract form in the present report .
Sanger
Uig.)
Bredt
Ainring (Upper Bavaria), August 1944
C-8 u6
-
TABLE OF CONTENTS
I , Fundamentals
I I , The Aircraft
1. Characteristics of the Motor
2. Effective Exhaust Speed of the liotor
3. Properties of the Air-Frame
U. The Glide-number of the Air-frame
I I I , Launching and Climb
1* Acceleration of the Aircraft
2 Catapult take-off
3. Climb Path
IV* Oil ding Flight and Landing
1. Supersonic Path of Gliding Flight
2. Path of Subsonic Gliding Flight and Landing
7 Projection of Bombs
1 Types Of Projection
2. Flight Path of the Boob
3* Bal l i s t i cs of Impacts
71* Types of Attack
1. Basic Types of Attack
2* Point Attack with Tiro Propulsion Periods and
Reversal of Path
3* Point Attack with Tiro Propulsion Periods, Partial
Turn and Auxiliary Point
U Point Attack with Sacrifice of the Bomber
5. Area Attack with Full Turn
6. Area Attack with Partial Turn and Auxiliary Point.
7* Area Attack with Antipodal Auxiliary Point
8
S
Area Attack with Circumnavigation
9> Evaluation of Procedures for Attack
VH The Line of Development of the Rocket Bomber
1, Development of the Combustion Chamber and Jet
of the Motor
2* Development of Special Fuels for Rocket Motors
3* Development of the Auxiliary Engines of the Rocket
Bomber
li. Development of Test Model of Complete Rocket Motor
5 Wind-Tunnel and Tow-Tests on Models of the Air-frame
6* Constructional Development of the Vehicle
7* Bench Tests on Interaction of Motor and Air-frame
9, Development and Teat of the Take-off Arrangement
9* Takeoff and Landing Tests on the Bomber
10, Fl i ght Tests of the Bomber
11* Navigation Tests on the Rocket Bomber
12. Bomb Release Trials
Bibliography
Table of Most Important Symbols
I. Fundamentals
The range of flight-speeds several times the vel oci t y of sound is the exclusive province
of the pure rocket, which develops the propulsive j e t ent i rel y froir. the fuel carried on board
the ai r cr af t . The pure rocket can also compete in cost at lower speeds, if propulsive forces of
great magnitude or short duration are required, or if no surrounding ai r is avai l abl e, e. g. under
water or outside the percept i bl e atmosphere of the eart h. These special charact eri st i cs aive
rocket propulsion a broad domain of application to mi l i t ary techniques, which can be outlined as
follows:
Propulsion of proj ect i l es or bombs, in which the r el at i vel y strong," short duration pro-
pulsive forces can be achieved in most cases by powder-rockets.
Auxiliary drive for propel l er-, or j et - ai r cr af t , with operating periods generally under a
minute, for which liquid rockets with compression-drive can be used.
Auxiliary or pri nci pal propulsion for vessels with period of operation of several minutes,
so t hat rocket motors having fuel pumps, but without high exhaust speed, aie required.
Main drive of aer i al torpedoes against land, sea, or ai r t ar get s, with moderate to long
times of operation, in which high exhaust speeds are important only for quite larse ranges,
Main drive of fighter or bomber ai r cr af t , e. g. for fighter defense at very hieh al t i t udes
or for military ai r cr af t operating over very great di st ances. Both propulsion-time and exhaust
speed set extreme requirements for the rocket motor. The last-mentioned application, the rocket
bomber, i s treated in more det ai l in the present report .
Pure-rocket engines make only very incompleLi; use of the energy made available by the fuel.
However since the craft is not loaded down by the energy carried on board but rather by the
weight of the fuel, t hi s disadvantage can be counteracted by use of fuels with the maximum pos-
si bl e energy content per uni t weight. Thus rocket fuels represent, on the one hand, car r i er s of
energy with maximum concentration of energy per unit mass and per unit tank space; on the other
hand, they are the car r i er s nf those masses from which the j e t pi" the engine i s developed.
According t o the method of feeding the fuel (which,- in the tank, i s liquid or sol i d) into
the combustion chamber of the rocket, we can distinguish between various modes of operation of
the rocket motor; e.gj- rockets with periodic propulsion, which are charact er!l ei by moderate
Values of the work for feeding the fuel, th temperature st resses in the walls in contact with
the flame, the exhaust speed and the t hrust ; and rockets with continuous propulsion, with arbi -
t r ar i l y high constant flame-pressures, high constant exhaust speed, maximum t hrust for given
dimensions and maximum thermal st resses of the furnace walls.
The type of construction of the walls in contact with the flame is determined by these
st resses,
The type of construction using the heat capacity of the wall-material gives very simple
solutions, which are however applicable only to periodic propulsion, or to continuous propulsion
over short periods at moderate furnace temperatures. For example, the 20 mm. thick metal wall
of the j et - t hr oat of a powder-rocket at 280$K and having thermal conductivity 4000 k cal / m*h
0
,
begins to melt on the side in contact with the flame aft er 2, 4, B, 10, 14 or 90 sec, i f i t is
made of Al, Ag, Cu, Fe or Pfi, Pt , or I r , reap; t hi s can be shown by calculation and can be qual-
itatively checked by t est s on welding torches,
Designs of combustion chamber walls using the best refractory materials gi v
e
somewhat more
complicated arrangements and longer propulsion times, which are in general limited mainly by
cberaical changes in the wall mat eri al . The best heat r esi st ant materials, (melting points given
in C) which would be of i nt er est in t hi s connection are for example: beryllium oxide (2500),
molybdenum (2600), zirconium oxide (2700), magnesium oxide (2800), thorium oxide (3050), titanium
carbide (3140), rhenium (3170), tungsten (3380), zirconium carbide (3500), tantalum and hafnium
carbide (3700) and graphite (4000). with these mat eri al s, using non-statiotiary thermal condi-
t i ons, the driving times can be extended further than the Values givei
1
previously.
Design using condenser j acket s around the combustion-chamber walls is similar to t hat
used in i nt ernal combustion engines for cont rol l i ng the hot, strongly superheated combustion
gases; i t i s however limited t o moderate combustio* temperatures and pressures for which the
heat flow through the wall i s everywhere less than 1 h. p
/ t D|
2 so t hat the velocity of the coolant
need not be raised above about 10 m/sec. J

~ 4.5 &
r
is
k
C-B4Z96
Fire-wall construction using forced ci rpul at i on of the refri gerant in channels, of pref-
erably one-dimensional ext ent , which cover pract i cal l y without gaps a l l the wall surfaces
couched by the flame, gives the possi bi l i t y of cont rol l i ng also those high heat t ransfers through
the chamber wall3 which occur unavoidably in using high-grade rocket fuels in uniform-pressure
rockets, and which go far above 1 h. p/
c m
2, and can be even 10 h . p / 2 or more in the j e t t hroat .
* This type of firewall for rocket motor construction i s used in the designs of the present work.
Aside from design of fi rewal l s, the supply of fuel t o the combustion chamber i s , for
uniform-pressure rockets, a special problem for the solution of which Tarioua methods have been
used.
Placing the whole fuel supply in the combustion chamber has proved sui t abl e in short -
period powder rockets. Pressure tank feed of liquid fuel, because of the considerable weights
of the tanks and compressed ai r , i s possible only for moderate driving periods and f i r e gas
pressures. Fuel supply through gas-pressure pumps limits the tank si ze and gives longer driving
periods at moderate flame pressures. Fuel supply with ordinary pumps and turbine drive r e qui r e
special propel1ants or exhaust gas removal from the combustion chamber and r esul t s in increased
fuel requirement per uni t momentum; nevertheless, i t does give high driving periods and flame
pressures. Fuel supply with ordinary pumps using a turbine driven by steam from the refri gerant ,
where the steam for the turbine i s developed by vaporizing the coolant in the canals of the
* chamber walls and f i r e- j et , limits nei t her driving period, driving pressure, or flame temperature,
and permits the use of the great est exhaust Bpeeds. This method i s the basis of the rocket Rotors
described here.
Fi nal l y, one of th most essent i al construction problems for uniform pressure rocket motors
i s the choice of furnace pressure. The high-pressure rockets with furnace pressures above 50 atm,
(which are necessary because of the high exhaust speeds requi red], are in pract i se driven up t o
iOO atm. They have small dimensions per unit t hrust and are especi al l y valuable combined with
highest grade fuels, where the already high exhaust vel oci t y can he increased by 2'2% through a
furnace pressure increase from 10 to 100 atm., and by 6% through a change from 50 t o 100 atm.
I t s domain of application i s therefore especially t hat of rocket fl i ght , e. g. for rocket bombers,
where the requirement of high exhaust speed i s most st r i ngent . The high requirements on the fuel
feed system are no t roubl e when they are taken care of by the coolant-steam turbine mentioned
above, which uses the heat from the forced cooling of the furnace. As a resul t of increased gas-
densi t y, -vel oci t y, -t emperat ure, and - radi at i on, the specific heatflow from the flame through the
furnace walls ri ses proportionally with the furnace pressure. This has as consequence the decisive
* di f f i cul t y t hat the protection of the walls in contact with the flame becomes more cr i t i cal as the
furnace pressure i ncreases, since the heat transfer from furnace wall to coolant only increases
a the 0.4 power of the coolant pressure, so t hat a pr act i cal l i mi t of furnace pressure i s reached
+
at about 100 atm,
Similar general considerations apply to the air-frame. To the fundamental question,
whether explosive propulsion by rockets over large distances shall be used with wingless, un-
manned rocket-torpedos or with winged and man carrying rocket ai r cr af t , i t may be said t hat for
the "r et ur ni ng" ai r cr af t , the range of use and the t ot al dest ruct i ve energy brought to the
t arget (weight of bomb % energy of explosive) i s as large as for the rocket torpedo for equal
i ni t i a l fl i ght speed, so that the conserving of the empty craft for repeated use and the probably
great er bombing accuracy speak in favor of the ai r cr af t . Since the i ni t i a l cost of the empty
craft is far greater than t hat for the bomb and fuel load, t hi s i s the basis for the choice. If
the rocket bomber doesn' t return t o i t s place of takeoff, i t s range for equal v
0
will be much
greater than that of the rocket torpedo, though, of course, the % weight of dest ruct i ve energy
brought to the t arget decreases. The extreme ranges possible with the rocket bomber are com-
pl et el y forbidden to the rocket torpedo.
Hie rocket bomber wi l l di ffer from the present-day propeller-driven bombing ai r cr af t in
the following essent i al poi nt s: in place of the propel l er propulsion from the fuselage front i t
has the rocket propulsion in the fuselage stern; the fuselage i s in the shape of a bul l et with
* tapered hind part , the wings have a thin wedge-shaped pr of i l e with sharp leading and t r ai l i ng
edges and high wing loading at the s t ar t of the fl i ght ; the cabin i s constructed as an ai r t i ght
stratosphere chamber.

For st ar t i ng, the use of i t s own fuel as in the usual propeller-driven ai r cr af t was con-
sidered, Because of the great difference i " s t ar t - and landing-weight t hi s leads to large wing
surfaces and too high fuel consumption in the range of speeds below the vel oci t y of sound. Ver-
t i cal s t ar t under i t s own power has only the l ast disadvantage, but even in a great er degree.
El i ng-st art i ng on a horizontal take-off path unt i l the sound vel oci t y i s reached appears most
favorable arid i s assumed here. In t hi s type of s t ar t by means of external forces, an especi al l y
5
C-8K296
energy-consuming part of the ai r cr af t acceleration is not obtained at the expense of the fuel
carried on board, so that the range of the ai rcraft catapulted in t hi s fashion r i ses markedly,
Mle at 5 * same time t he f l i ght - char act er i st i cs can be matched more easifry t o the st eadi l y
decreasing si ng loading during f l i ght .
As methods of fl i ght were considered; acceleration t o the point where fl i ght speed equals
exhaust speed, and then fl i ght at constant speed; also acceleration to such a speed t hat the
subsequent unpowered gl i de extends over the ent i re remaining fl i ght path. For equal fuel cost ,
the l ast method gives great er ranges and simpbv* power pl ant , and i s therefore assumed from now
on.
The whole procedure for use takes place, somewhat as follows: the rocket bomber at the
surface of- the eart h is brought t o a speed of about 500 m/sec. by a ground-fixed rocket drive in
a period of 11 seconds over a 3 km. st ar t i ng path; then climbs at full motor drive t o a height
of 50-150 km along a path which is inclined at 30 t o the horizon at f i r s t , but l at er becomes
f l at t er ; thus i t reaches final vel oci t i es up tjo more" than twiqe the exhaust speed. The duration
of the climb is 4-8 minutes; usually during t hi s t i ne a l l the fuel supply on board wi l l be con-
sumed. At the end of the climb the rocket motor is turned off, and the ai r cr af t , because of i t s
kinetic and potential energy, continues on i t s path in a sor t of osci l l at i ng gliding fl i ght with
st eadi l y decreasing amplitude of osci l l at i on. This type of motion i s similar t o the path of a
long-range proj ect i l e which from similar heights follows a descending gl i de-pat h. Because of
i t s wings the ai rcraft descending i t s bal l i s t i c curve bounces on the lower layers of the atmos-
phere and is again kicked upwards, like a fl at stone ricocheting on a water surface, though
during the entrance i nt o the dense ai r each time a fraction of the ki net i c energy is consumed,
so that the i ni t i al l y big jumps st eadi l y become smaller and fi nal l y go over i nt o a steady gl i di ng
f l i ght . At the same time t he-fl i ght speed, along the glide path of several thousand kilometers,
decreases from i t s high i n i t i a l value t o normal landing speed. If the descending path (which is
within certain limits cont rol l abl e by the pi l ot ) l i es in the di rect i on of the t ar get , the bombs
are released at a predetermined moment, and the craft returns t o i t s st ar t i ng place (or some other
landing fi el d) in a wide arc, while the bombs go toward the t arget along the ori gi nal di rect i on
of f l i ght . Even i f the t arget i s very di st ant from the take off point, the bombs are only dropped
near i t , so that the scat t er of bombs can be compensated for by a large number of rel eases on the
t ar get , which will in t hi s way be covered by a Gaussian di st r i but i on of hi t s . This mi l i t ary use
i s completely independent of weather and time of day at the t ar get , and of enemy counteraction,
because of the possi bi l i t y of using astronomical navigation in the st rat osphere and because of
the height and speed of f l i ght .
From the char act er i st i cs given for the rocket bomber i t follows t hat t hi s is not the de-
' , velopment of an improved mi l i t ary cr af t , which wi l l gradually replace present types, but rather
that a problem has been solved for which no solution existed up to now, namely, bombardment and
> bombing over distances of 1,000 t o 20,000 km. toith a single rocket bomber point attacks can be
made, e.R. from Central Europe, on di st ant point t arget s l i ke a warship on the high seas, a canal
lock; even a single man in the other hemisphere can be fired upon.
ni t h a proup of 100 rocket bombers, surfaces of the size of a large ci t y at ar bi t r ar y
places on the ear t h' s surface can be completely destroyed in a few days.
6
C~S*2*
II. The Aircraft
1 . fhnwi PtMH. aHj. ff pf fop Ro c kBt Mo t o r
The main parts of the basic construction of fche rocket motor considered here are shown in
f i g. 1. The fuel goes from the fuel tank t o the fuel pump, where i t ia compressed t o ISO at n. ,
then fed continuously through valve 5 t o the i nj ect i on head of the combustion chamber. The oxygei
goes from the thin-walled uninsulated oxygen tank i nt o the oxygen tank, i s compressed t o 150 atm.
t here, then forced through valve 6 and the pipe system of the condensers i nt o the i nj ect i on head
and the combustion chamber, aft er being warmed t o 0" C. There the fuels come together for the
f i r s t time, mix and burn producing flue gases at a constant pressure of lOO atm. and at 4000 C. >
In the head of the burner, the flue gas expands t o a very low pressure and forma the driving masaj-
j et with exhaust vel oci t y of 3-4000 m/sec, whose react i on for a fuel consumption of 245-327 kg/sale
produces a t hrust of 100 t ons. With a 90 ton fuel supply, the ai r cr af t can be driven with the
above t hrust for 367-275 seconds-
Aside from t hi s main process with energy conversion of about a million Kcal/sec the
econdary process shown schematically in Fig. 2 gives sn energy conversion of 20,000 Jfcftl/aec for
driving the feed-pumps. I t s parts-* vi si bl e in Fig 1, can be followed from the water pump which
puts about 28 kg/sec of water under 250 atm. press ut e, dri ves t hi s water, at the j et t hr oat , intc
the cooling pipes of the j et wal l s, where i t flows toward the mouth of the j e t and ia heated t o
bout 300 C. After coming out of t here, while s t i l l above the c r i t i c a l pressure, i t i s (again at
the j et throat) driven i nt o the cooling pipes of the combustion chamber wal l , where i t i s again
heated and vaporizes in the neighborhood of the c r i t i c a l pressure; f i nal l y i t is removed at the
injection head in the form of highly-compress ad superheated steam, led t o the steam t urbi ne; t here
i t expands t o about 6 atm. and goes to the liquid-oxygen-cooled condensers where i t is reconverten
t o water and gives Up considerable energy to the oxygen; then i t repeats i t s cycle going through
the water pump. The steam turbine drives a l l three pumps from the same shaft . During the process
valves 3, 4, 5, and 6 are open, and 1, 2 are closed, while 7 serves as a safety valve against t ec
high rot at i on of the t urbi ne.
The process described can be begun with the aid of the st eam-st art er, which produces the
small amounts of steam required by chemical means! in t hi s process the valves 3 and 4 are closed,
L, 2, 5, 6, and 7 are opened.
Aside from the det ai l s given in the l i t er at ur e, (16-30) the following things are important
for understanding the proposed construction:
The rel at i ve value of different fife-*a11 materials is determined by the available heat
flow <J*%(tf~tA4i through the walls for a given heat capacity of the wal l s. If the wall t hi ck-
ness a i s proportional t o the reciprocal of the breaking st rengt h <T- ( t ensi l e s t r e s s ) , or t o i t s
square root (torsion s t r e s s ) , then the possible heat flow (and al so therefore the worth of the
material) is proportional t o the product ^fCf-C/fJO", or &{tf$fj0
:r
reap. Here \ , tf and <f
are pure material constants, while the cool-surface temperature E/r (and so. the whole evaluation)
depends on the part i cul ar arrangement, coolant temperature, e t c . In the combustion chamber cooled
by live steam, the cool-surface temperatures are 500-600 C. From Fig. 3 one sees t hat the usual
heat -resi st ant metals chrome-nickel st eel , nickel, " vent i l "- st eel , e t c . , are favorable (as con-
firmed by construction experience) while platinum is even more sui t abl e. The t heor et i cal l y most
favorable materials like tantalum, tungsten, and molybdenum ar e, because of t hei r chemical act i v-
i t y and the di f f i cul t y in working them, act ual l y not a t a l l promising. In the hot-water-cooled
fi re .jet the cool-surface temperatures are at 400*500 C, because of the high heat flow; Fig. 4
shows (in accord with pract i cal experience) that copper i s unsurpassed as j et - wal l mat eri al .
The cooling system for t he walls in contact with the flame (21) i s required because of the
high heat flow from the combustion gases t o the fi re-wal l s; t hi s is presented in Fig. 5 for a
gas-oil-oxygen charge at 100 atm. combustion pressure, on the basis of cal cul at i ons and pract i cal
experience. For example, in order t o conduct 5 bp./
c| p
2 from the fire-surface t o the cool surface
through a 1 mm thick copper wal l , a temperature gradient fjt--6
f f
^ between the two surfaces of
100 C is necessary. In order for such thin walls t o withstand the mechanical st r esses due t o
flame-and coolant-pressures, they must be reinforced at very shor t - i nt er val s. At the same time the
heat flow through the fire-walls must be assured by a preci sel y prescribed and carefully maintained
high streaming velocity of the coolant behind the f i r e- wal l . Both requirements can be met by the
cooling-pipe system shown in Fig. 1, with forced ci rcul at i on of the coolant in st ruct ures of one-
dimensional ext ent . The necessary compromise between the r i si ng forced motion and increased
pressures required t o drive the coolant when the number of channels i s decreased has been so made
t hat the cooling system of the j et consists of several hundred par al l el pipes each only a few
7
C-BU296
e-
r* e
a>
(r*
T ,t~>
rr

* *
t
3
09
*
O
U "J
c*
a
QR
X
^ o
a
u
(B
Cxygftn
S5S Fuel
Fi g. Is Systematic represent at i on of the main par t s of the rocket mot'
of tho rocket boaber.
.."* Wa t e r
Fi rs
Steam
or, drawn i n the i nt er i or
Heat added to the fuels by
pumpwork f>2)
Energy cont ai ned i n t h e
st eam a f t e r t h e t u r b i n e
and added t o t h e f u e l
by condenser (~18?5)
Energy contained in the steam
before the turbine t~20)
/
I
WAWM90 Keat conduction through walls
^ ^ V
r
in contact with the flame
(-2056)
Fig, 2i Energy flow i n the working process of the rockflt motor.
sq. mm. in cross-sect i on, which j oi n together -part i al l y at the j et throat and again branch off
in the far parts of the j et mouth so that the small individual cross-section of the pipes is
retained; each pipe runs meridionally, and the whole surface of the j et is-completely covereci
with pipes, of which the side toward the flame consi st s of the smooth and t orsi on-st ressed f i r e-
wall. The requirement of a plane surface does not apply t o the furnace surface in contact with
the flame, so t hat there the s t at i cal l y more favorable ci rcul ar cross-section can be used. This
circumstance and the smaller heat-flow through' the furnace walls require greater wall-strengths
of the fire-wall and thus greater cross-sect i on for the individual cooling pipes, so that the
then necessarily, few pipes in par al l el can assure high ci rcul at i on of the coolant in the neigh-
borhood of the steam chamber which has varied and unstable flow conditions. These few cooling
pipes are wound on the furnace surface in the form of an evaporating coi l .
We see, from the above discussion, the requirements on the coolant i t sel f : large heat
capacity, heat conductivity and density! for t hi s reason mercury has advantages over water.
The reason,why the furnace-jet in Fi gs. 1, and 5 i s shown with the unusually large open-
ing angle of 60 is the following: (22) Aside from the fact t hat a speci al coolant, which best
sui t s the requirements, is ci rcul at ed around the f i r e- wal l s, the actual coolant is r eal l y the
fuel i t sel f , t o which the intermediate coolant t ransfers i t s heat in the pumps and condenser.
The heat -absorpt i vi t y of the fuel before i t is brought i nt o the combustion chamber i s limited,
and is only a few percent of the heat wiiieh is liberated when i t i s burned in the combustion
chamber. One must therefore take care, t hat the t ot al heat t ransfer per sec, from the flue-gas
t o the coolant through the walls of the furnace and j e t , which is given as 20% in the example of
Fig. 2, remains less than or equal ,to the aforementioned absorptivity of the fuel consumed per
sac. This t ot al heat t ransfer which must be regulated is proportional t o the t ot al inner surface
of the furnace and j e t . It can be decreased by diminishing t hi s t ot al surface. Concerning the
contribution of furnace and j et t o the t ot al surface in contact with the flame, the following i s
t rue: a l l experience shows t hat over a wide range of values, combustion in a furnace is more com-
pl et e, and efficiency and exhaust speed are correspondingly l arger, the great er the furnace
volume Vo as compared t o the smallest cross-sectiofi r of the furnace j e t . Because the t ot al wall-
surface is limited by the heat absorpt i vi t y oi the fuel used as a coolant, the furnace surface
can be increased provided that the j et surface is decreased keeping the sum of the two below the
permissible l i mi t . From Fi gs. 2 and 5, the t ot al heat-flow through the 154,000 cm of fi re-wal l
surface represents 2% of the energy developed, which corresponds t o the permissible heat absorp-
tion of the fuel, so t hat the heatflow per unit area of the furnace and j et walls is about 0.8
np/
C[n
2. If in place of the short 60 throat with 60,000 2 surface we used the customary Laval-
throat with a 1C opening angle, i t s furface (345000 ,.,,,2) could not be completely cooled i f the
same heatflow and absorpt i vi t y of the coolant were maintained; for the furnace there would be no
surface cooling avai l abl e at a l l . At the same time the length of the Laval-throat could not be
decreased below 9720 mm. By using large throat-angles we are enaDled t o f ul f i l the requirements
of the fuel-cooled rocket motor, and moreover the quantity Vo/f' which determines the completeness
of combustion can be increased, so long as the increased efficiency of the furnace J/" is not
overcome
by
t
j
i e
decrease of j et -effi ci ency fAtf&gyfc with increasing opening angle.
The pump system of the rocket motor consi st s of three pumps for fuel, oxygen and coolant
and the driving turbine, for these pumps. The vaporiied coolant of the rocket motor is used for'
feeding the t urbi ne, where the forced ci rcul at i on used for the combustion chamber permits the
chamoer t o be used as a high-pressure-radiating-steam-boiler with forced ci rcul at i on in the
manner of the Itenson - , Uhmt - , Vslox -, Sulzer - steam boi l er s . (27) The use of the vapor-
izes coolant for driving the auxi l i ary turbine has the advantage over the use of a separate energy
source that the t ot al fuel consumption per impulse by the rocket motor is not increased by the
auxi l i ary turbine dri ve; the advantages over feeding the auxi l i ary turbine from the flue-gas of
the rocket motor are t hat : use of the cooling capacity of the fuel for cooling the combustion
gases brings i t t o temperatures permissible for the turbine drive-rod; the di f f i cul t i es associ -
ated with condensation of, for example, metallic-oxide fumes in the flue gas disappear; the heat
absorptivity of the fuel as a coolant is increased by the work done in pumping; the important
decrease in momentum of the j et in the emission of a part of the flue-gas and transfer of i t s heat
content t o the remaining flue-gas is avoided: fi nal l y the construction of a hifch pressure steam
turbine is incomparably simpler than that of a high pressure flue-gas t urbi ne. Since according t o
Fig. 2, the efficiency of the 12,000 up, drive of the steam t urbi ne, which uses waste energy, r
unimportant, while we do demand very snai l weights of the i nst al l at i on, the simple Curtis-shaft
gives a sui t abl e sol ut i on. The 3 pumps can in view of the high t ot al fuel supply of over 1000
rrr*/hr., be designed as cine-stage turbines (despite the high intake pressures), so t hat the whole
pump assembly including the turbines consists of 4 running from the same shaft , at anout 12,0l>0
PMi. Thus the outer dimensions and weight of the whole i nst al l at i on can be kept below 600 x 1200
mm and 500 kg.
10
C-8U296
100 tOO JOO 400 500 000 700
1 e.' i-issiblc colu ~iae -tecfc-rat-re in '"
BOO
Fig. 3: Relative evaluation nuEfcr of combustion chamber wall materials
subject t o t ai si cn.
Permissible cold side temperature in
U
C
Fig. 4: Relative evaluation numbers of firewall construction materials
under bending (exhaust wall materials).
12
C-B4296
Heat ab-
sorbed by-
cooling
agent
*8
_Q^.Temperature
of cooling
agent
?Sm
Fig. 5: Conditions in a 100 ton rocket motor operating with
oaygan and oi l .
-ftux
Aside from the requirements of extremely l i ght construction the turbine, fuel pump and
water pump present no speci al constructional di f f i cul t i es, whereas in the construction of the
oxygen pump the choice of the construction mat eri al , the arrangement of the moving part s and the
feeding of the boiling liquid to the pump must be speci al l y considered. As a construction ma-
t er i al (for the oxygen pump) which at -180 C will he suffi ci ent l y strong, el ast i c and r esi st ant
to impact, suffi ci ent l y resi st ant to corrosion and non-inflammable in liquid oxygen, the ni ckel -,
A1-, and Mn- bronzes, as well as Monel-type al l oys and pure nickel have proven sat i sfact ory. In
view of the inflammability of- a]] lubricants in liquid oxygen, the problem of arrangement of
moving part s was solved by using a floating support for the pump shaft away from the oxygen. In
order to drive the boiling liquid Oj from the tank st eadi l y to the pump, an arrangement was used
whereby the oxygen Hows toward the pump over a long route in the di rect i on of an acceleration
field; n. g. , in the t es t i nst al l at i on, from a higher level; or in the ai r cr af t , from tanks lying
far toward the front. Because of the gradual pressure increase in the feed l i nes, accompanied by
only a sl i ght temperature increase, supercooling of the O2 occurs at the pump intake, so that no
more gas i s l i berat ed.
Fig. (J is a photo qf the experimental model .of a high-pressure liquid- O5 pump, which as
a rotary 6-stage pump with external bearings supplies 5 kg/sec of liquid O9
al
- 1*>0 a tin, pressure,
when running at 15,000 RPM; i t has proven i t s sui t abi l i t y and r el i abi l i t y m hundreds of experi-
ments.
The ignition of *he rocket motor i s not shown in Fig. 1, because ignition i s limited onlv
to st ar t i ng; once the combustion chamber gets going i t operates l i ke a welding-burner. The basic
ignition procedure chosen was the injection into the combustion chamber of materials which ignite
on contact with 0) or a i r . From the pyrophors to be considered, like the phosphorus hydrides,
"silanen", halogen-acetylenes, rare earth amalgams, metal al kyl s, e t c . , zinc-diethyl Zn(Cj %)2
was ohoser at the suggestion of H. Troitzsch; F. Zohrer developed a sui t abl e ignition fluid by
di l ut i on of t hi s with heayy hydrocarbons ( e. g. , machine oi l ) , and also an ignition apparatus in
the form of a small pressure bomb using compressed nitrogen and a remote-controlled valve; by a
simple movement of the valve and consequent_injection of the ignition fluid i nt o the combustion
chamber, arbi t rarv ignition time and arbi t rary repet i t i on of the i gni t i on is possi bl e. This
ignition procedure is notable for i t s sure performance and the very smooth st ar t i ng of combustion.
The pract i cal work on the development of the rocket motor described in t hi s section was
taken up by the senior author in 1933-34 at the Teehnisehe Hochschule in Vienna and gave in the
f i r s t experiments, on small models with 30 kg. t hrust s, controllable flame-pressures of 50 atm.
and high exhaust speeds; the fuel was tig (at up to 150 atm, i nj ect i on pressure), and gas-oi l (at
up to 500 atm. i nj ect i on pressure), and a Laval-throat of small opening angle was used, (19)
After a delay of several years, which were spent in constructing larger experimental i nst al l a-
t i ons, the t.ests were recommenced at the Trau Aeronautical Testing Station in 1939. The con-
st ruct i on of the experimental i nst al l at i ons was under the di rect i on of H. Zborowaki; the con-
st ruct i on of the components was directed by H. Ziebland; K. Hedfeld directed the experimental
work. Fig. 7 shows the testing. shed during an experiment with 1 ton t hrust and 5 minutes dura-
tion. Among the important. parts, one can see at the l eft on the embankment a cyl i ndri cal tank
of capacity 2.5 ro^ for the liquid &>, and j ust to the ri ght of i t the tap for the underground
tank of liquid Oj (see al so I I , 2). The drive tank, from which the apparatus i s di r ect l y fed, is
an open uninsulated thin-walled metal tank which (out in the open) vaporizes oxygen at the rat e
of 15 kg/hr. per sq. meter of tank wall, and whose varying weight during the t est is shown by an.
automatically-recording spring balance. From t hi s tank the liquid O2 flows at slow speed under
i t s own weight to the high-pressure liquid oxygen pump 8 m. below (see f i g, ) . Beyond the feed
pump the liquid oxygen, now at 150 atm, pressure, runs through a heat exchanger, in which i t i s
heated by. the warm cooling-water coming from the furnace, then goes into the combustion chamber
through a large number of i nj ect or nozzles. Following the corresponding path of the fuel, we see
in the l eft foreground a lm* fuel tank, from which the fuel flows under i t s own weight to the
high-pressure fuel pump. For t hi s purpose a cog-wheel pump i s used, which compresses the gas-oi l
t o 150 atm. at 3000 RTAi. In the experiment shown here, the fuel and oxygen pumps were driven t o-
gether by a D.C. Motor standing between the pumps; l at er the coolant-steam turbine was used i n-
stead. Beyond the fuel pump the fuel also i s forced i nt o the combustion chamber through a large
number of nozzles. The fuel - and 0^- streams are directed at 90 to each other and have i ni t i a l
entrance-velocities of about 100 m/sec BO that rapid spraying and mixing i s forced. In the fur-
nace, the three fluids - oxygen, fuel and ignitor - meet and form the furnace gas. The furnace-
gas pressure during the ent i r e run i s up t o 100 atm, with Vo/|"B0Q and about 30 opening angle
of the provisional expansion-Rozzle, The next photos 8 and 9 show a 1-ton t r i a l from the j e t
si de, 10 shows a small model using coolant vaporization; Fig, 11 shows a 1-ton t r i al in which a
high-percent Al-gas-oil suspension was used as fuel. The flame glows bri ght er in t hi s case, and
the resul t i ng aluminium oxide begins t o condense t o white corundum dust at a few meters from the
j e t opening, and then thickens i nt o a heavy white cloud. Fi nal l y, Fi g. 12 shows a short-exposure
Ik
Figure 6} Experimental high pressure l i qui d O2 pump which
has a 6 stage rotary pump. At 15000 H.P.M. i t
i s pumping 5 Ic^/sec of l i qui d 02 a t 150 a-tooa-
pherea,
15
r* d H " Ac
Fig. 7: Overall view of a 1-fcon, high pressure combustion
chamber experiment using cooling by evaporation.
Propellent tanks are above roof to the left. The
fuel pumps directly underneath. Combustion chamber
ia in operation in center* Note the cloud of con-
densed cooling agent. The observation stand is
above on the right.
Figure 8j Overall view of a rocket motor t eat atand. This
motor produbed 1 ton of t hr ust for a duration of
5 minutes.
17
rvsiMfi
K:,
Figure 9l View of ins^ruwentB and propel l ent l i ne s . This
t eat was run on t he 20th of Karon I9kl. The
chaialser pressure 100 atmosphereaf the t hr ust 1.1
tonsj the duration 3
#
5 minutes.
18
i
Figure 10; Small water cooled combustion chamber and teat
instrument in duration teat, 'feter, heated at
400 centigrade at 100 atmospheres pressure in
the cooling system.
19
C-849*
Figure 11; Rocket motor t e s t stand experiment using Aluminum
i n oi l di spersi on as f uel . Supersonic exhaust gases
from the nozzle of 1-ton experimental rocket motor.
Note compression l i n e s .
20
C-8429*
M
Fig. 17: Supersonic exhaust gasns from the nozzle
during an axperimantal run. Thrust was
compression lines.

22
' v a VZ
photo of the j e t i t s e l f , i n which one can see with the naked eye the supersonic conpreaaion l i nef
which gi ve the ex^- i st j e t the appearance of a l arge, blue crys t al .
The B i n part of the pract i cal work consi st ed i n the constructi on of combustion chambers
for developing and wi thstandi ng of combustion gases wi th the high energy concentrati on mentioned
No time could as yet be devoted t o the conversion of the heat i nt o ki net i c energy of a j e t , t hat
i s t o constructi on of a j e t noxsl e. Ordinary amount* of fuel consumption i n the course of long
ind s t eadi l y run t e at s , gave ef f ect i ve exhaust speeds of up t o 2400 m/sec at about 35 atn flame
pressure, with gas- oi l - oxygen f uel and for k ^ f - 1 . 4 3 ( i . e . , s t i l l i ns uf f i ci ent performance of ?--^
the extended throat s e c t i on) . *> *~--
e
-..'.- ; _,." - - ". - ~ - V ' '-
f
With the r e l at i ve l y low %- val ues of t i e combustion gas {about 1. 25) caref ul l y constructed '
f i r e - j e t s should, according t o Fi g. 17, gi ve at l eas t P/
p
*=1.6 i n the t e s t set - up, and about
P/p j = 1.75 i n the ai r c r af t , so t hat the comvustiato gas developed corresponds t o ef f ect i ve ex-
haust speeds of 2700 to/sec. on the ground, an* ness 3000 tu/aec. i n the ai r c r af t .
Fi nal l y, Fi g. 12a shows a t r i al constructi on of the carburetor of a 100-ton rocket hi gh
pressure combustion chamber, which was not hofever used i n the experiments.
2. Effective Exhaust-speed of the Rocset Motor
Statements concerning the form and nature of the physico-chemical processes i n the j e t of
the rocket-furnace, which could gi ve a determination of the ef f ect i ve exhaust speed, assume a
knowledge of the processes and of the f i nal s t at e of the combustion gas i n the furnace.
di e may assume t hat most of the avai l abl e t*me i n the furnace apace (about 75 ndl l i s ee)
i s used up i n processes of sprayi ng, heati ng, vapoei i at i on, di s s oc i at i on, turbulence and di f -
fusion of the i nj ect ed streams of fuel and oxygen; and onl y a small part of the time i s used for
the actual combustion and coming to equi l i bri nm. The f nl l y prepared and mixed f ue l - , and oxygen-
molecules (or - atoms) c o l l i de , and react with eacu ot her, but wi l l immediately di s s oci at e again
i f there i s no means of t ransf erri ng the l i berated heat of react i on t o i nt ernal degrees of f ree-
dom, to other bodi es, or converti ng i t t o t ransl at i ons1 energy. The l as t pos s i bi l i t y e x i s t s for
atomic c ol l i s i ons (according t o the pri nci pl e of the conservati on of the center mass) onl y i f a
thi rd atom or molecule takes part i n the c ol l i s i on, so that the par t i c l e s present af t er the Tr-
act i on can repel each ot her. (Tri pl e c o l l i s i o n, wal l - c at al ys i s , exchange react i on) . A measur
for the probabi l i t y of occurrence of any react i on j s the ef f ect i ve number of c ol l i s i ons , which
s t at es how many c o l l i s i o ns of other part i cl es with the molecule under consi derati on are required
t o produce the desi red e f f e c t . According to en empi ri cal formula of Gwb (5)2jjji ntri pl e c ol -
l i s i ons occur i n every Z ordinary c o l l i s i o ns After each formation of a new molecule, l at er
c ol l i s i ons wi l l supply energy f i r s t t o i t s ' rot at i onal and then to i t s vi brat i onal degrees of
freedom at the expense of i t s t rans l at i onal energy, t i l l f i nal l y i n some cases di s s oci at i on oc-
curs. In the course of a s uf f i c i e nt l y long time, which i s dependent on the number of ef f ect i ve
col l i s i ons ' whi ch a molecule must undergo for each 4f these changes, and on the time i nt erval
between two such c o l l i s i o ns , an equilibrium skate (dependent on the pressure, temperature and
proposi ti ons of fuel and C^) develops i n the furnace, which can be exact l y descri bed in terms of
the kind, number and energy content of the molecules or atoms present . This s t at e i s assumed i n
the l at er cal cul at i ons of the e f f e c t i ve exhaust speed.
Si nce the great est pos s i bl e energy content af the t ransl at i onal degrees of freedom of the
combustion gas determines the maximum possi bl e valtle of the e f f e c t i ve exhaust speed, one Would
prefer for the rocket motor a more favorable f i nal s t at e than t hat of st at i onary equi l i bri um,
t hi s, seems t o be at t ai nabl e, si nce t ransl at i on, rot at i on, vi brat i on, di s s oci at i on and recombina-
t i on take successi vel y longer times t o at t ai n equi l i bri um, and the time during which the fuel
remains i n the furnace may l i e anywhere between these ti mes. According to Jost ( 8 , page 141) i t
i s conceivable that the newly formed molecules may, because of t hei r process of formation, not
have t hei r vi brati onal degrees of freedom completely exci ted by the end of the combustion pronet>e,
so that a greater f ract i on of the energy remains for the K.E. of the center of mass than corre-
sponds t o equilibrium; thus temperature and pressure at the end may be higher than that corre-
sponding t o equi l i bri um. <see K. Wohl and M. Magat, Z. Pbvs. Chem. Vol. 19, p. 117, 1932; al s o
( 10) p. $05, f i g. 6, "The Approach t o Thermal Equilibrium'').
While, i n the furnace of the rocket at say 100 atm. flame pressure and 3700 K, a molecule
experi ences an average of 10" c ol l i s i ons per sec, the number of c o l l i s i o ns drops af t er expansion
i n the j e t , with a corresponding decrease i n the i s t e of exci t at i on of the degrees of freedom,
the rat es of further chemical react i ons, such as reburning, and of physi cal reacti ons such as
condensation or s ol i di f i c at i on of the combustion products (provided the temperature drops below
t hei r s t at i c sublimation temperature during the expansi on).
23
C-B1M6
The processes in the j et are then treated exactly as those in the furnace, neglecting a l l
wal l -effect s; the expanding combustion gas i s assumed t o be an adi abat i cal l y closed system with
a t ot al energy E kcel / , ^ corresponding t o the beat content of the mixture; i . e. Af co
Accordingly as the time the streaming gas spends in the j e t i s long or short compared to
the times of dcrclopment of the various i nt ernal energetic and chemical equi l i br i a, three assump-
tions are possible concerning the expansion:
1, The time spent, or the number of Collisions which a molecule undergoes, on i t s path
through the j et , i s so small that no eneTgy exchange or changes of vi brat i onal and di ssoci at i on
energy can occur. The char act er i st i c conditions for t hi s case are dD = 0 ande
yas
^tJTmO'
i
when
put i nt o the generally valid energy rel at i ons between t ot al energy, E, heat supplied, (J, i nt ernal
energy, U, heat of di ssoci at i on, D, heat of vaporization at QK, R- , heat content, J , work done
in expansion, ApV, and ki net i c energyAe^; t o the equations:
a "smal l er" adiabat Exponent which r esul t s from only t ransl at i onal and rot at i onal speci fi c
heat s; (X ordi nari l y would be (C^'"*
+
ARl^* / (note by Transl ).
gives the vari at i on of the flow-velocity as a function of the pressure drop in the j e t .
2, If the time spent by the combustion gas in the j e t i s such t hat vi brat i onal st at es are
instantaneously in equilibrium, whereas no chemical processes can reach equi l i br i a corresponding
to the changed condi t i ons, then the flow i s characterized by dD = 0, C, osc. dT - f*(T). From
t hi s and the fundamental equations we get the rel at i on;
For a gas mixture with n vi brat i on frequencies and Dj different gases,
par t i al pressure of the gas as compared t o p and 6. the char act er i st i c temperature. The i nt egral
for t hi s type of flow i s ;
i^jfe+gfe-uh^*
r
J~ fp ""
f
' -** '*
3. The t hi rd possible type of flow in the j e t occurs when the time Spent i s the j e t bv
the combustion gas i s so long t hat al l energetic equi l i bri a, including chemical equilibrium, can
be attained for the instantaneous Values of pressure^and temperature. This type of flow in
awact er i i ed by tb* conditions: c*etc JT"~f'o% <&
m
4p'faT)
A
c-e(
Fro* t hi s and the fundamental equati ons, we obtai n the di f f e r e nt i al equati on:
Tbe function c&*p'<kU which now contai ns Jjfc* canrijt be represented i n e x pl i c i t form, so t hat na
general anal yt i c sol ut i on of the di f f e r e nt i al equation can be gi ven. (fee must t heref ore, uaing
the di f f erence equation
aa cl os e a c ol l e c t i on of Tallies of J, D, and M as ftmctiona of p and T as pos s i bl e, obtai n the
desi red connections poi nt for poi nt . Procedures far cal cul at i ng such t abl es for vari ous fuela
have been suggested by M. V. St e i n.
The behavior of temperature T, heat coatent J, Di s s oci at i on energy D, and the l i berat ed
ki net i c e n e r g y w^ f al ong. * pressure drop frorn^A / ( ! * to>4<r-0far a l l three types ef flow i s
shown for the combustion of octane i n l i qui d oxygen i n Fi g. 13.
The deci si on as to which of the three po s s i bi l i t i e s (or an i nt ernedi s t e one) i s most
probable for the flow of the combustion gaaes from the rocket furnace i s i nfl uenced by the f ol -
lowing consi derati on: i f the expansion takes pl ace i n an experimental short j e t 1300 mm. long,
then the mean time spent by the combustion gases can be computed to be about 2 x 10" sec-,
corresponding t o an average no. of c o l l i s i o ns of the molecule on i t s way through the j e t of about
2 x 10. If one consi ders that 96% of these c o l l i s i o ns already occur' i n the apace between the
furnace and the smal l est cros a- s ect i on, then one sees by comparison wi t h the ef f ect i ve number of
c o l l i s i o ns {/O />) for the Various energy i nterchanges, that we may expect equilibrium of rota4
t i ona, vi brat i ons and possi bl y even di s s oci at i ons l or the processes occurri ng i n the i ni t i a l part
of the j e t , up to i t s smal l est cros s s ect i on. Flow-type 1 becomes the more improbable the higher
the furnace pressure and temperature for given ve l oc i t y, the longer the j e t , and the smaller the
e f f e c t i ve number of c o l l i s i o ns of the combustion gas mixture. Therefore fl ow-type 2 i s general l y
assumed i n cal cul at i ons on rockets; i t gi ves cl osed i ntegrabl c formulae. The act ual l y rapi dl y
varyi ng9c can, for rough est i mat es, usi ng the usual adi abati c flow formulae, be replaced i n f i r s t
approximation by a mean valuea/m -Z/X whiah for stoichiometric burning of octane at 100 atm.
gi ves ar*= 1.248 * ^ - > * e
This mean value i s s t r i c t l y val i d onl y for an expansion to T 0K and b_ 0 atm. If for ex-
ample the mouth val ues are t o be cal cul at ed fur an expansion t o J 1 atm, then two equations
are required for a determination of/ Jl f*/ s i nce bot h/ % %* and T_ are unknown for a given
mouth pressure p
n
^ 0 For the example of octane oombustion given above, we obtai n i n t hi s man-
ner from the equati ons:
for an expansion t o / - 1 .atm, the v a l u e / j K ^ = 1.222 at T
m
- 1586 K. This method gi ves
f ai rl y accurate val ues of T
n
and c_ whereas a l l intermediate val ues between T and T , as wel l
aa the corresponding c^,, are more i naccurate. Flow-type" 3 , i n case i t s consi derati on csn not
be avoided, can be represented by the method of Mol l i er with the ai d of entropy chart s, which
show J + D + R as ordi nate and enable Cg, and temperature val ues t o be read off. Figure 14 shows
such a Hoilier-diagram for a combustion gas of gas- oi l - oxygen and gas - oi l ozone respect i vel y.
The .following t abl e shows i n summary form for the ease of stochi ometri c combustion of
octane i n On at 100 atm. pressure, how the important f i nal charact eri s t i cs of the streaming:
mouth vel oci t y C_, e f f e c t i ve exhaust speed C, j e t nouth temperature T, and the rat i o of ki net i c
energy to t ot al energy supplied (E), di f f er for the three types of flow and for di f f erent j e t
end-pressures.
25
C- MM
J
i 1
i 1
| , _ ,
i
1 1 1 1 1
; 3+D+A&2g*1t*2587Jm$
2sm
i
, _ ,
i
!
i i i
1
te
\jfD+Aci'2<3'Z357ffca//frq j
i ' ,
, -, i
!
|
1
i
V*i>
I 1
i
i
1
\
1
i
i
i
1
i
i
a
3 V
7"
f
1 '
!
5
a
3 V
l > ^
I I .
\\
i
i
: C
!
II
w
r
a
t

l
e
r
g
i

V.
L I
i p
5
1 1
t \
1 1
\
\
30w -ISOO

*
^
V?
i
m-*-
*
*
\
\ Y
V?
^ ^ ^ ^ ^
r '

* i
i i
i
% 1 V
^
1
t | , !
1
i
i
% L
*
^^^^
i 1
|
1
1
1
AM* VHb
W
^*
V 1 1
1 1
i
^
^*"
fe
, i
|
i
*
< ^
\ i
" 1
i
4 '
J-
1 1
w
1
^
V*

i
i
p '
i
*
V*
^
1 ^ p ^
"*'?
**>^
* ' - / ' l ^^
i
^ ^ j ,
1 "* *
1
z
*
1
9 f 2 i
-<
f
-s
i & - 7
Logarithm of tha chamber Dreasure in ctmosphere
Figure 1?; Temperature T, Heat content J, Dissociation loss Dpt, and
gain in kinetic energy plotted against the logarithm of the pressure
p according to the three possible streaming gradients in the nozzle,
(See text p.24 & FF). The gasses entering the nozzle are assumed
to be a completely burned stoichiometric mixture of the octane and
02 at a pressure of 100 atmospheres*
? * * *
i * V i l A7 2,1 If 3.0 3.1
Entropj i kcat/kg *
Figure 1*; Entropy diagram after M* von Stein for the
Hoxzle stream gradient and oil combustion.
S t r e a m i n g Pr OCBSf i Art ( 1.380) 2. Ar t 3. Art
Test stand conditioner
p
m
a 1 a t 1 at 1 a t
T
B
103DK 1577K 250aK
c
m
= 2518 m/sec 2759 m/sec 3O20 tn/sec
c = 2654 m/sec 2949 m/sec 3260 a/ >:
Gained energy i n t he
mouth of t he nozzle
i h % O f 29. 3 35. 1 42
Fl i ght conditions i n d/dn =0.14:
p ^ s 0.0729 a t 0. 1305 a t 0,250 a t
T,,, = 500K 1041K 2225K
c,,, = 2760 m/sfcc 3066 m/sec 3330 m/sec
c = 2820 m/sec 3178 m/sec 3520 m/sec
Gained energy i n the
mouth of the nozzle
in % Of E. 35.1 43.4 51
Optimum conditions i n case of complete relaxation*
r i u >
0 a t 0 a t 0 a t
! , _# 0K 0** 0K
c
m
n
l
c , ^ = [ 2970 m/sec 3496 m/ sec 4438 m/sec
Gained energy i n t he
mouth of t he nozzl e
i n % of Be
40i
7
56#5
9o.v
Completely l o s s l e s s conversi on of t he t o t a l i nput ene-rgy E i nt o k i n e t i c energy would gi ve '
a t he or e t i c a l exhaust speed _
The concept s of mouth ve l oc i t y C^,, maximal ve l oc i t y c max, and e f f e c t i ve exhaust speed
c
( 1 9 ) i r equi r e s or e d e t a i l e d expl anat i on. I n gener al , for rough c a l c ul a t i ons , s t andar dadi abat i c-
flow formul as, us i ng a f i c t i t i o u s aver age*; , whose magnitude and eval uat i on has been di scussed
al r eady i n pr es ent i ng t he t hr ee t ypes, of flow. Thus f or expansi on t o a mouth pr es s ur e *
m
, t he
mouth vel oci t y i s m = / ^ f f i - j V ra^mor expansi on t o t he t h e o r e t i c a l l y pos s i bl e l i mi t i ng pr e s -
s u r e
/
A, ^ , < ^
w
v | j S j & and from iese two r e l a t i o n s :
I f we mul t i pl y t he v e l o c i t y tt^ b t he mass bl-iwn out per second * % we obt ai n t he mouth-
monentum J , of t he j e t .
In experi ment s wi t h rocke> s , i f t he nobt h pSessureyb i n t he j e t i s equal t o t he pr essur e
of t he sur r oundi ng s t i l l a i r i
a
t h i s mouth momentum i s d i r e c t l y recorded by a dynamometer as t he
force P*; i t depends on t he ext nal a i r pr es s ur e, i . e . on t he barometer r eadi ng, a l t i t u d e of t he
t e s t i ng- pl a c e , e t c . and must no!, be confused wi t h t he t h r u s t P, as can be seen from a consi der a-
t i on, of Fi g. 15. There t he conv t i ons between r e s i s t a nc e and t hr us t for a f l yi ng appar at us and
t he i r val ues measured on he gr> ind ar e schemat i sed.
By r e s i s t a nc e t o e dr i ven appar at us I n f l i g h t we mean t he vect or sum of a l l t he aer o-
dynamic forces on t he a e a t e d s ur f aces . Let If be t h i s sum of a l l t he pr es s ur es and f r a c t i ona l
f or ces . I n a wind t unnel or a taw t e s t on a non-dri ven devi ce one always measures a s mal l er
r es i s t ance W"*W-pxf
n
i ' h e r e p' i * t he abs ol ut e a i r pr es s ur e beyond t he month sur f ace f
m
of t he
j e t . I t follows t her ef or e t ha t N=W'-fp'fm-
r
*
OT t n
* moving body, t he a i r pr essur e p' behind t he
38
C-Ma
- *i ** t( J
Pa m: Unavoidable
. addi ti onal t hrust
at the t e s t stand
^=Tr
Relationanlp between thrust and drag in flight
compared to sea level test values*
rear nr f ce ia always m l l i than the pressure ^ of the ai r at t eat ; for %>2/fyt-/)-klzjl'mo
Tie value p' must always be included in resi st ance meatureikents on undriven devices, in order t o
obtain the t r ue resi st ance W of *ke body.
By t hroat on an apparatus in fl i ght we mean the vector sum of a l l the flame pressures on
the surface of the rocket i n contact-with flame (inner wal l ?). In a t es t one always measures a
smaller t hrust P' P -frfi, where *
a
i s again the pressure of s t i l l ai r . The t hrust measured in
a t eat depends on the pressure of the surrounding a i r , and for the effect i ve t hrust we obtain
P'P' + pJw'
The effect of these two unavoidable addition* 1 effects in a wind-tunnel or a t es t can be
seen most cl earl y by cal cul at i ng the resul t ant force (t hrust -resi st ance) which accelerates
a mass nfefy
&<*& =p-W= P- iVV f
m
fa-/>'J
For measurements on the ground, an unavoidable additional force proportional to ; i s thus
included, which i s different in measurements of resistance and t hr ust , and for which speci al
correction must he made. The correction i s small at moderate vel oci t i es ( e. g. in the use of rock-
et s for takeoff) and becomes l argest for very high supersonic speeds ( e. g. rocket pr oj ect i l e,
long distance mi l i t ary rocket ai r cr af t ) .
In order to cal cul at e the effect i ve t hrust P of a rocket motor, one must add to the mouth
impulse J*G~ 4? the t ot al pressure of the combustion gas on the j e t mouth, ( / ^/ j ) ; or one must
add t o the free dynamometer t hrust in the t es t
the t ot al t hrust of the s t i l l ai r on a surfac* the size of the j e t mouth:
In the i ni t i al l y mentioned, most frequent c a s e ^ p *
3
^ , mouth impulse J
m
and dynamometer reading
P' are i dent i cal .
Thus the effective velocity C, which i s independent of external ai r pressure, and which
when multiplied by da gives the effective throat i s :
The determination of the effective exhaust sneed c has the following pecul i ar, technically
i nt erest i ng consequence; in the usual engine construction the thermal efficiency i s fAmrm
= c
"feJL
s: fT^/fc In the rocket motor the quantity which corresponds t o t hi s i s the j et efficiency: *
(see also (19) p. 6) . This expression implies that the higher temperatures and the quant i t i es
of heat at those temperatures are more " effect i ve" than lower temperatures and quant i t i es of
heat at lower temperatures. This represents no contradiction to the energy theorem, since the
effective exhaust speed i s not i dent i cal with the actual velocity of flow of the combustion gas,
but i s larger than i t . This rel at i on has nevertheless a technical value, because the part of the
i ni t i a l heat content at lower (temp.) ranges can be made available only by special technical
procedures (for gases, which soon tend to develop degeneracies, l i ke condensation, i t can' t be
done at al l ) and non-avai l abi l i t y therefore leads to rel at i vel y smaller losses than one would
expect according to the second law, for the non-available heat content.
Aside from the rel at i ons already given, we can, from the well known equation for the rat e
of gas flow through the j et throat _^^ .
and the equation for the effect i ve exhaust speed, obtain a frequently useful r el at i on between P,
V andA
0
in the form P = H A / " where ^_/
By means of t hi s factor k, by which the effect i ve thrust i s great er than the product oi furnace
pressure and j et - t hr oat surface, one can reduce the otherwise tedious determination of effective
t hrust of a rocket motor, for suffi ci ent l y lengthened j et , to a measurement of the furnace pres-
sure and the outside pressure, which can be easi l y done with ordinary manometers. This type of
thrust measurement i s very convenient in t r i al setups as well as in measurements on ai r cr af t in
30
C-B1296
V V f.s /,
Figure 16; Elregas flow with Xal.
QC/nax, CmlCmmt, dtyn,
, M , tt
25 K-Pbof tiCm-PIF
31
P-AKMA
Figure 17; Analysis of effective thrust of rocket motor with, x-1,
f l i ght . Conversely, i n a t e s t , i f furnace pressore and ef f ect i ve thrust (or dynamometer t hrust )
are measured, the act ual l y ef f ect i ve X can be determined.
Al l these rel at i ons : the factor k, the rat i os c/Camx., Gn/Smox, c/Cm, Tm/To and d'/djn **
shown i n Fi g. 16 as functi ons of the rat i o of furnace pressure to mouth pressure, for a frequently
used value of * * 1 . 2 S .
As i s shown i n Fi g. 17 for the s peci al ease 3f *1. 25, the i ndi vi dual f act ors i n the expres-,
si on for k ctn be i nterpreted very i nt ui t i ve l y; the t ot al thrust P of the rocket motor i s made Hp
of the part i al t hrust s: ft^/i,f
i
, ari s i ng from the pressure A of the combustion gas on t hat
spot, f , of the rear wal l of the' furnace which one s ees when looking down the axi s of the j e t ,
throat. A cylindrical tube, closed at one end', would show this sane effect;/ ^ =' / J. /
r
- ^^
?
J^7
h
the i ncrease rel at i ve t o Pj , ari s es from the lion-uniformity of the pressure di s t ri but i on on the
remaining furnace-wall surfaces due t o the pressure gradient i n the di rect i on of the Jet opening?-
This i ncrease r e l at i ve t o 2 i s caused by the pressure of the combustion gases agai nst the ex-
tended j e t .
The free thrust P* of the rocket motor i s
and i s shown for comparison i n Fi g. i t by the dotted curve.
Before these physi cal consi derati ons as chemi co-energeti c problems are treated i n rel at i on
to the e f f e c t i ve exhaust speed, we want t o di scuss bri ef l y t hei r s i gni f i cance i n terms of the
long-range rocket-bomber proj ect . The fami l i ar rel at i on, that j e t and rocket motors produce
t hei r energy conversi on most e f f e c t i ve l y when the vel oci t y of f l i ght i s (as c l os e l y as pos s i bl e)
equal and opposi te to the vel oci t y of the j e t , might l ead us to the f al s e concl usi on that the
exhaust speed should be adjusted t o .the instantaneous vel oci t y of f l i ght . Since i n the ai r c r af t
what matters i s not i deal energy conversi on per sea but rather the smal l est Consumption of f uel
i n termi of wei ght, f uel s with higher energy Content, which gi ve l arger C-val ues, are to be
preferred, even when t hi s greater energy content per uni t weight i s not used so e f f e c t i ve l y. Fot
a given consumption (of f uel ) by weight, the energy gi ven y> ^
tts
. ai rcraf t becomes great er, the
greater the value of c , because the usef ul part f%* *? Mfi^^tK/ *
t n e
dri vi ng energy c
2
/
2
from a uni t mass of fuel (*.#. *
5
> ^ c
J
) i i ncreases rapi dl y with exhaust speed. Despi t e t hi s , , the
great importance of the exhaust speed may be l i mi ted i f the fuel having .higher exhaust speed has
other disadvantage*. For example, a 50% aluminium-gasoil suspensi on and oxygen, having a tank
volume per kg. fuel requirement of 0. 64 dm, i s preferabl e despi t e the somewhat lower exhaust
speed, t o l i qui d H- and 0 , , which requi res 2.4dm
3
tank space per kg. of f uel , and for which more.
over s peci al heavy devi ces are needed i n the tank space because of the low temperature of the
l i qui d H,. The fol l owi ng general statement can be made concerning the ef f ect of densi t y on the
choi ce of f uel : - i f t he - f l i ght path s i s approximately proportional to the cube of the maximum
f l i ght ve l oc i t y V, and V i s rel ated t o the exhaust speed O and weight rat i o G/G
Q
by the equation
of page 207, V - 0. 443c + ft In G
e
/2G
t
then s - cons t . x C
3
X(In -ffe - 0. 25)
3
. One can therefore
obtai n equal ranges for di f f erent val ues of c i f 0/Go vari es properl y. By di f f erent i at i ng
(keeping i , V, and G
0
constant) we obtai n
A
%.
m
$^8 A dependence of the expressi on ^9/G on the
small change ^ ^ . i n the energy densi ty E^ can be estimated as, for example, ^ ^ a - A f & f "
Thus 4 c ^ = -aS-S- -Ag* . "o that i n the regi on of i nt ereat , % a 5 j i g- ^- "% vari a-
t i ons i n exhaust ve l oc i t y are four times as e f f e c t i ve as improvements i n densi t y. Furthermore
these consi derat i ons permit a comparative eval uat i on of di f f erent f uel s , by s t art i ng from a
standard f uel , say a hydrocarbon wi th l i qui d oxygen, having val ues C and Ey. From these and the
corresponding C, and Eyj of the comparison f uel , an eval uat i on number K can be gi ven i n the form;

G
S.S0 C *
A t ypi cal hydrocarbon react i on i s the burning of octan i n oxygen:
<i4 + /*#<%= aCOe +9#
M
0+2S8r**tf^
i
<****&<. &S/
The value for the upper l i mi t of the heat of mixing i s referred t o 0K. From Fi g. 18, i t
i s cl ear t hat we must subtract from t hi s value the 9-1% l oss due t o heat required for physi cal
separati on ( mel t i ng, vapori zati on) and that the pressure dependent l osses due t o chemical aepara*
t i on ( di s s oci at i on) amount t o 34.4% at 100 atm. furnace pressure, s o that under these dri vi ng
condi ti ons the furnace e f f i c i e nc y i s ^ = *&=.&**? This graph of octane combustion at
33
C-M2M
1
P
o
1
Upper heating value of mixture
,
0
2587kcal/kg ,
C8Hie+f2%0z=8C02 +9HzO
Lower heat i ng val ue
of mixture
.Unavailable separa-
t i o n heat (for example
"heat of evaporation")
Available
chemical heat
of separation
Available heat
* content.
of fire pressure
Figure 18; Heat of evaporation, dissociation, enthalpy, firegas
temperature, theoretical and maximum exhaust velocity and adiabatic
gas exponent of the combustion products in the case of oxygen and
octane with static equilibrium.
34
3 f~
var yi ng furnace pr e s s ur e s , as wel l as t he l a t e r graphs and t abl es for ot her f ue l s , were c a l c ul a t e d
by M. V. St e i n, t aki ng account of a l l t he di s s oc i a t i on pr oduct s, and assuming t hat physi cal and
chemical equi l i br i um ar e r r a^. ^d in t he f ur nace, and t ha t expansi on can be fol l owed, as a flow of
t he second ki nd, t o an e xt e r na l pr essur e zer o. In a ddi t i on t he lower l i mi t of t he beat developed
by t he mixture ( E- Hi
0
) , t he r ^*t cont ent J
0
, furnace t emperat ure T
QI
t he or e t i c a l and maximum ex-
haust speeds ar e pl ot t e d, as a f unct i on of t he log of t he furnace pr es s ur e, for t he proper aver -
age X correspondi ng t o t h i s expansi on; a l s o t o give some i dea of t he range of va r i a t i on of Vt
for ot her t ypes of expansi on and ot her j et - mout h pr e s s ur e s , t he adi abat exponent cor r espondi ng t o
t he i nst ant aneous' f ur nace- s t at e i s shown as aa upper val ue, and t he val ue of 9C for t he c a l c ul a -
t i ons of fl ow-t ype 1 i s shown as an upper l i mi t . In t hi s and t he l a t e r gr aphs , which enabl e onl y
a r e l a t i v e comparison of di f f er ent f ue l s , t he e f f e c t i ve exhaust speed C, which i s t he det er mi ni ng
f act or f or f l i g h t , and whose val ue f or octane combustion has been pr evi ousl y gi ven, i s not es pe-
c i a l l y not ed. I t can be det ermi ned from t he maximal ve l oc i t y usi ng t he val ue of j e t e f f i c i e nc y:
The expr essi on f or t he t o t a l l oss of t he r ocket motor i s
numer i cal l y, for oct ane combustion a t 100 atm. furnace pr essur e* T *
The f l i ght i s fixed in t h i s and a l l t he l a t e r c a l c ul a t i ons by d / J b. 14, on t he bas i s of t he
fol l owi ng cons i der at i on: in f l i g h t , t he pr essur e behind t he s t e r n of t he roc"ket bomber drops
below 1 atm, so t ha t t he pr essur e gr adi ent p,Jv
m
a t s
*
y
1*^
B t m
furnace pr essur e must go f ar above
100. The combustion gas wi l l a c t ua l l y spread over t he whole 2. 50 m* sur f ace of t he s t e r n of t he
a i r c r a f t ; i . e . , t he r a t i o of j e t mouth sur f ace f
m
t o t he j e t t hr oa t sur f ace f wi l l be about 50;
throughout t he working peri od of t he r ocket motor, cor r espondi ng t o d / J =i / i r r =0 . 1 4 .
In addi t i on t o t he i mport ant and c a r e f ul l y s t udi ed hydrocarbon-oxygen mi xt ur e, t her e i s a
second group of r ocket f ue l s , having t he common pr oper t y t ha t t hey ar e el ement s i n t he f i r s t
columns of t he per i odi c t a b l e , and which when bur nt i n Qj gi ve much hi gher energy c o n c e n t r a t i o n
per uni t mass, and a l s o gener al l y per uni t volume, t han t he hydrocarbons.
The fol l owi ng r e a c t i ons were consi der ed:
Unt ersucht wurden folgende Beakt i onen:
Be + 0. 5 Oj - BeO + 5930 kcal / kg
2 B + 1,5 O^ = B2O3
2 Li + 0. 5 O2 = Li
2
0
2 Al + 1.5 Oj = A1
2
0
3
^ + 0.5 Oj = HgO
M; + 0. 5 0j> = MgO
be i 0K
+ 4930 kcal / kg
+ 4750 kcal / kg
+ 3920 kcal / kg
+ 3900 kcal / kg
+ 3330 ' kcsl / kg
( 6, 1 3 . 1 4 )
Combustion of At and IU are shown i n Fi g. IV and 20 in t he same form as for octane com-
bus t i on. The fol l owi ng t abl e gi ves t he most i mport ant c h a r a c t e r i s t i c s for t hese l i ght - me t a l -
fuel s wi t h s t ochi omet r i c l i qui d Oj a t 1, 10, and 100 at m. furnace pr es s ur e:
V . TA
Ca
//*l,7 , ,
BeO
&A
Li
2
0 Al
2
% %o MBO
\ \ * ^ ^
5930 4930 4750 3920 3900 3330
5040 2300 2780 1290 750 3750
0/ /a/in
13,8
18,0
23.2
-
4 -
-
-
43.0
48.0
54.5
af /da/**
20,6
23,8
28,4
40,5 .
46,7
51.4
21.9
24.1
27.0
21.5
24.4
28.1
45.8
51.7
59,7
28,7
35,0
44.1
&oifaf n*rf?fc*
3400 1990 1100 3250 373 3120
35
c-em
BeO
%<*
u
2
o
A l
2
^
11)0
UtO-
at /00#r?
af /00 a fin
3400
3920
4550
621
0
7010
7680
2350
2730
3200
3700
4070
4700
2950
3200
3560
3350
3970
IPSO
#"? /0O*B
1.24
1.22
1,20
1,125
1,110
1,100
1.320
1. 315
1,310
1,2 M>
1,260
1,240
1.300
1,250
1,220
1,11
L.10
I , OR
C
& *Ae<:J
7050 6420 6310 5730 5720 5280
^ / /tali*
* / /00erfn
3200
3440
3760
4090
4390
4610
2950
3100
32 GO
2660
2830
3030
3B70
4110
4420
2fi30
3130
3510
In t hi s cal cul at i on a l l heats of fusion were netciected r el at i ve t o heats of vaporisation; a l l
possible di ssoci at i ons of the end products of combustion were taken i nt o acccunt.
B> 1-i, Al and Hg do not di ffer essent i al l y from hydrocarbons in t hei r behavior during com-
bustion, but di ffer only in the numerical r esul t s; for example, Li and Al, despite t hei r higher
heat production, reach lower maximal exhaust speeds than for hydrocarbons, because of the vapor-
ization and di ssoci at i on l osses. Be and Mg are basi cal l y di fferent . In the case of Be, the
large heat of vaporisation of BeO permits only a part of the burning mass t o vaporize, while a
large part remains in the liquid st at e {fog} or solid st at e (dust ). These parts can then make
use of the upper l i mi t value of the heating value of the mixture, so t hat despite the large heat
of vaporization, high temperatures, and High heat content of the t ot al mass occur, with consequent
large values of C_
x
- These liquid and sol i d masses in the combustion gas have a great effect on
the expansion of trie t ot al mass, which expresses i t sel f in the form of a very small adiabat-
exponent: This $C of the t ot al mass is calculated from the weight fraction k_ of the gas phase,
the weight fraction kj and specific heat cj of the l i qui d-sol i d phase by the equation:
are obtaxnta \T!^M uYual way for a flow-type 2, t o mouth pressure p
m
a: o. The T
2
curve at the
same time represents the boiling point of BeO, since the combustion temperature is determined by
the boiling point of the Bed. For Mg, whose lower heat value is act ual l y negative, the large
fraction of l i qui d-sol i d phase r esul t s in very high temperatures and heat content of the t ot al
DBSS and especi al l y low A? values, so that the burning imss rapidly loses the char act er i st i cs of
gas or vapor and approaches, the behavior of a hot lava.
F.om the given values of maximum exhaust speed, the resul t ant effective exhaust speeds for
the individtial light-metal fuels at LOO atm. furnace pressure can be computed, if we take account
of the j et efficiency corresponding t o the & values for d/j =0.14, using the r el at i on
, 3 &
'
/ c,
iag
wrr#tF/rft-{*) **/
X
[n f i r s t approximation,** was obtained from the somewhat too large mean value between To and o,
instead of between To and T^; t hi s leads t o less favorable values.
/ " - e / c - - 0. 844;
0.690;
0,925;
BeO * - 1.20
fe^>
1.10
1.31
Al2J
1.24
1^0 1.22
MgO 1,08
' max
o.eso
0,865
0,630
3170 m/sec;
31 SO ;
3030 ;
2670 ;
3P20 ;
2210
(The approximate mean val ue for 9t i s -1.255 for octane and gi ves C 3120 m/ sec. For purposes of
uni formi t y, t hese numbers were used i n l a t e r comparison c a l c u l a t i o n s ) .
For a ppl i c a t i on t o r oc ke t s , i n addi t i on t o t he e f f e c t i ve energy concent r at i on i n t he mass,
the ef f ect i ve concent r at i on E
v
i n t he t ank volume i s i mpor t ant , s i nce i t det ermi nes t he s i z e of
the fuel t anks , r a t e of feed of t he i nj ect i on pumps and t he eval uat i on number n of t he f uel ,
3&
fOOOO
fff-
U
P
peg
heating value of mixture 5920 kcal/kg
Unuseable phys-
ical heat of separa-
tion (heat of fusion
and vaporization).
Conditionally
usable chem-
ical heat of
separation.
(Dissociation)
~
e
*Po
*ut
Available
Enthalpy
Common logarithm of fire pressure
Figure 19; Heat of vaporization, dissociation, heat content,
firegas temperature, theoretical and maximum exhaust
velocity and adiabatic exponent of flregases for the
burning of aluminum in oxygen with static equilibrium*
37
C-84296
100QL
K Upper
heating value of mixture 3900 fccal/kg
Unusable
physical heat
of separation
(heats of fusion )
and vaporization)
Conditionally
usable chem-
ical heat of
separation
(Dissociation)
, Available heat
content
0 1 2
Common logarithm of"fire pressure*
Pigure 20; Heat of condensation, dissociation, enthalpy, fire-
gas temperature, theoretical and maximum exhaust
velocity and adiabatic exponent of firegasea in the
burning of H
2
in 0? with static equilibrium.
Xo ui ejn^fijedaa) Wfd
A
Ber yl l i i i a a i t Fl ues i gaauer s t of f E, = 1S80 kcal / dm
3
; K - 1, 23;
Bor ni t Fl us s i gs auer s t of f 1640 ; 1, 27;
Li t hi um n i t Fl us s i gs auer at of f 820 ; 9. 64;
Aluminium mi t Fl us s i gs auer s t of f 1390 ; 0, 68;
Fl usai gwasser st of n i t Fl i i s s i g-
s auer s t of f 780 ; 1, 22;
Magnesium mit Fl i i s s i gs auer s t of f 6*0 ; 0, 25.
Die St andardwert e fur Oktan (Gas51)
mit Fl i i s s i gs auer s t of f bet r ugen , = 1240 kcal / dm
3
: K * 1, 00.
Thus we Arri ve a t t he r e s u l t t h a t , among t he f uel s i n t he second group onl y Be, B and
l i qui d Ho ar e s uper i or t o hydrocarbons. Be and B ar e immediately el i mi nat ed s i nc e , under t he
condi t i ons pos s i bl e i n t he t ank, t hey ar e i n t he s ol i d s t a t e , s o t ha t feedi ng them i nt o t he hi gh
pr essur e furnace of t he r oc ke t motor i s i mpossi bl e. If one shoul d t r y t o overcome t h i s d i f f i c u l t y
by usi ng t he mat er i al in t he form of wi r e or powder, t hen t he concent r at i Oh i n t he t ank would be
decreased so much t ha t t he cal cul at ed smal l s u p e r i o r i t y would be l o s t . Use of f uel i n t he l i q -
ui d s t a t e i s el i mi nat ed because of t he hi gh m l t i n g poi nt . So appar ent l y l i qui d hj. i s t he onl y
mat er i al i n t he second group of f uel s which can compete wi t h hydrocarbons. I t s 22% s u p e r i o r i t y
i s endangered by t he fact* t h a t a t t he t emperat ure of l i qui d Hg, condensat i on of t he sur r oundi ng
a i r wi l l s t a r t on t he met al t anks i n t he a i r c l a f t , so t ha t t he r a t e of evapor at i on of t he Hj
wi l l be i ncr eased and t he aerodynamic forces wi l l be af f ect ed, unl ess s peci al pr ecaut i ons a r e
t aken. Never t hel ess, l i qui d Hi , because of i t s easy procurement and a l s o for r easons t o be d i s -
cussed l a t e r , may be consi der ed ss a most promi si ng r ocket f ue l .
Si nce, -with t h i s except i on, t he l i ght met al s ar e onl y s l i g h t l y ( i f a t a l l ) s uper i or t o
hydrocarbons as f uel , a second p o s s i b i l i t y shoul d be t e s t e d , - a combination of two e . g. i n t he
form of me t a l l i c suspensi ons i n mi neral o i ' a . Suspensi ons not onl y combine good feed c ha r a c t e r -
i s t i c s (t hey ar e e a s i l y pumped) wi t h hi gh energy dens i t y (wi t h consequent low r equi r ement s on
t ank volume and feedi ng speed) ; i n many cases t hey have t he amazing pr oper t y t ha t t h e i r heat
cont ent i s gr e a t e r t han t ha t of t h e i r i ndi vi dual components such as hydrocarbons and l i ght met al s .
(30) Fi g. 21 shows cycl es f or Al - Og and oct ane - ftj a t 100 atm. flame pr e s s ur e , A comparison of
t hese shows t he i n t e r e s t i n g f act t h a t t he end t emperat ure i n t he combustion of oct ane (3700 K)
i s about 830 lower t han t he boi l i ng poi nt of AI2Q3. I f one assumes t hat equal s t ochi omet r i c
mi xt ures of oct ane -Og and Al -Oj a r e burned t oget her , t h a t r el axi on- f r ee heat exchange t akes
pl ace between a l l t he mol ecul es of bot h bur ni ng gas es , and t ha t f i na l l y t he two burni ng masses
do not i nt e r a c t appr eci abl y chemi cal l y, t hen one s ees from t he two cycl es t hat energy wi l l flow
from t he hi gher t emper at ur e l evel of t he Al^Orj t o t he col der oct ane. Thi s wi l l be heat ed at t he
expense of t he heat s of di s s oc i a t i on and vapor i zat i on of t he aluminium combustion gas u n t i l t he
boi l i ng poi nt of AI9O3 (4530 K) i s reached. Aft er t h i s pr ocess of t emperat ure equal i zat i on has
occurred, t he usef ul heat cont ent of t he oct ane wi l l have r i s e n consi der abl y, whi l e t he heat con-
t ent of t he Al gas wi l l not have changed appr eci abl y; i . e . , t he heat cont ent and maximum exhaust
speed of t h i s 70% Al - oct ane suspensi on ar e l ar ger t han for Al or oct ane al one. The advant ages of
t he combination ar e t he fol l owi ng:
1. The t emper at ur e of t he bur ni ng gases f or t he suspensi on has not t aken on a val ue mid-
way between t hose of t he i ndi vi dual components, but r a t he r t he whole combusting mass has reached
t he boi l i ng poi nt of ALnOa. The r e q u i s i t e energy has been obt ai ned a t t he expense of t he ot her -
wi se unavai l abl e energy of di s s oc i a t i on and vapor i zat i on of t he AI9O3
2. As a r e s u l t of t he t emper at ur e e qua l i s a t i on between t he two f uel s i n t he suspensi on,
t he r e l a t i ve l y s l i g h t di s s oc i a t i on of AI9O3 has been decr eased, whi l e t hat of t he oct ane has i n-
cr eased markedly. Thus t he average s pe c i f i c heat c^ J / T
0
of t he gas has i ncr eased, a s one can
see from t he smal l er s l ope of t he expansi on curve for t he oct ane. Both t hes e ci r cumst ances r e -
s u l t i n i ncr eased heat cont ent J cJT,
One r e a l i z e s , moreover, t ha t t he random example of a 70% suspensi on chosen her e, which has
equal pa r t s by wei ght of Al - and oct ane - combustion gas es , may not gi ve t he be s t r e s u l t s for
C^gv- The bes t Al - oct ane suspensi on . wi l l be one havi ng enough Al t o heat t he e nt i r e mass t o t he
boi l i ng poi nt of AI2O3. The upper heat val ues of t he Al wi l l be f ul l y used i n t h i s cas e, s i nce
no met al l i c oxi de vapor i zes . The ext r a heat of vapor i zat i on and di s s oc i a t i on of AI1JO3 ar e used
t o i ncr ease t he heat cont ent and di s s oc i a t i on of t he oct ane. The best r e s ul t s ar e obt ai ned for
a 60. SS Al - oct ane suspensi on.
The cons i der at i ons concerni ng t he Al - oct ane suspensi on can be ext ended t o t he ot her l i g h t
met al s . Fi g. 22 shows for Be, B, Li , Al and Mg suspensi ons wi t h var yi ng met al cont ent , t he chi ef
1*0
c-Bsm
CmmB
CmaxBe
cB-%U
* %W\
iO Of 0,2 Of 0,5 Oj6 0,7 0,8 0,9 iO
Metal content in the metal dispersion in parts by weight.
Figure 22; Properties of the metal oil dispersions against metal .,
content at 100 atmos chamber pressure.
C-84
char act er i st i cs of the combustion nt 100 at a furnace pressure; aaxiimn exhaust speed, average
adiabat exponent, effect i ve exhaust speed i n f l i ght , and evaluation number K. From the graph we
see t hat those fael r, l i ke B and Li, th* -boiling poi nt vt whose oxide i s below the end temperature
of combustion of octane, do not show the char act er i st i c effect of the di spersi on. For them the
curves of c , , , , * ! , and c, show no maximum; i . e . , the suspension i s no more favorable than the
bet t er-of i t s two components. For the remaining mat eri al s, Be, Al, and Mg, the char act er i st i cs
s t optimum composition s r e:
for Be - hydrocarbon fool with 39% by weight of metal c ^ - 4100 n / . .
e
X- 1.20; C 3400 m/
s e c
; K- 1.51; E= 1557 kfi| i ^ ^
for Al - hydrocarbon fuel with 60.5% by weight of net al C^
x
m
3760 m/
c
.
W= 1.189; C = 3140 m/
s e c
; K = 1.14; E = 1453 i c a l
for Mg - hydrocarbon fuel with 60S by weight of mehal C_ * 3725 */
act
'
t
tf-l.l?l; C= 3080 m/
s e c
; K = 1.00; E = 1330 kj al
dm
3
Summarizing we can say: Mg dispersed in hydrocarbon has no advantages over pure hydro-
carbon. Al and Be show a wide range of suspensions in which they are superior t o the pure hydro*
carbon by up to 14% and 51% resp. Because they are easi l y obtained, Al - suspensions have speci al
importance for mi l i t ar y rocket -fl i ght technique, while Be - suspensions come i nt o consideration
for special uses. All the st udi es were limited t o stochiometric proportions, so the possi bi l i t y
s t i l l exi st s t hat other mixture proportions may reach bet t er values of K ox O-
The preparation of 60% Al - gas oi l suspensions, which are s t i l l usable aft er many weeks
i f l ef t untouched, and which are easi l y fed through centrifugal pumps, was done in two ways on
the basis of suggestions by H. Troitzsch and E. Russer:
1. Increasing the Viscosity of the gasoi l by dissolving various materials such as met al l i c
sal t s of fat t y acids, wax^s, f at s, rubber or various synt het i cs. Good r esul t s were obtained in
t est s with natural and synt het i c rubber, and si mi l ar high-polymer hydrocarbons, the oppsnols.
The oppsnols have the further advantage t hat , being pure .hydrocarbons, they require no bal l ast
mat eri al s, but burn completely with large heat output.
2. Decrease of par t i cl e si ze of the Al dust while hindering surface oxidation as much as
possible, since with decreasing par t i cl e si ze, the sedimentation speed and, in most cases, the
vi scosi t y decreases. (?) For these large quant i t i es of metal, pulverizing by using supersonic*
seemed impractical. So the powder was ground in bal l - mi l l s in a nitrogen atmosphere.
In connection with the use of liquid On as i component of al l the rocket fuels discussed
to far, the problem of st ori ng very large quant i t i es of t hi s material i s important. Because i t s
boiling point i s - 163 C, i t wi l l be continuously boi l i ng as a r esul t of the steady flow of heat
from i t s warmer surroundings, and wi l l l i ber at e t he energy absorbed by vaporizing with a heat of
vaporisation of 51 kcal/kg, so t hat the resi dual mat eri al can maintain i t s el f at t hi s low temper-
at ure. This undesirable vaporization can be decreased by lowering the heat t ransfer from the
surroundings, which occurs mainly through the tank walls which are wet by the l i qui d. A f i r s t
method i s the decrease of the wetted surface by put t i ng a l l the material t o be stored i nt o a
single tank of spherical shape. The heat flow through t hi s smallest surface can be further de-
creased by the use of various standard heat -i nsul at i on procedures, of which heat-stopping ma-
t er i al s like loose powder of magnesium carbonate, with a heat conductivity of A = 0-027 kcal/mh,
have shown themselves effect i ve, in the form ol thick l ayers. In the exi st i ng temperature range
of + 20C to -183C, 131 kcal per square meter of tank surface will go through a 1 meter thick
i nsul at i ng layer during a 24-hour period; t hi s corresponds to Ojg-vaporization at the rat e of 2.57
kg per day per sq. meter of surface. With these fi gures, the dai l y loss by vaporization i s shown
in Fig. 23, for various containers up t o a million ton capacity, for three thicknesses of the in-
sul at i ng layer lm, 5m, and 10m, and taking account of the spat i al heat flow through the t hi ck
walls.
The resul t s of t hi s calculation were confirmed in a t r i a l i nst al l at i on of a liquid On tank
with 56 ton capacity and a magnesium carbonate i nsul at i ng layer of average thickness 2.6m. This
tank has been running at the ai r cr af t - t est i ng st at i on at Trau since 1938; the manufacturer i s the
" Aktiengesellschaft fiir Industriegasverwertung Berlin - Bri t z". Although even t hi s tank, (though
small as measured by rocket -t echni cal requirements) represents a brand-new development as compared
t o al l previously constructed liquid Oj containers, and no experience with tanks of such size
existed, the tank worked sat i sf act or i l y from the f i r s t day i t was used. I t s vaporization i s 140
kg/day, which i s exactly the t heoret i cal l y expected value, as can be seen by subst i t ut i ng the
1.2
i c-e*t
looooa
$50000
3
a ZOOM
o
#40000
(D
3
2000
10DO
SOO
WO
100
SO
20
to
s
^*TTT|J|
i l l
Tank contents in tons
Figure 23; Daily evaporation loss from cylindrical large scale con-
tainers for liq. Qg for various capacities and various thiciaiesses of
inorganic insulating material*
43
tank-volume, i nsul at i on thickness, dai l y vaporization in Fig. 23. In the photograph. Fig. 24,
can be seen the 8 m. high, 8m. diameter outside cover of the tank. Inside t hi s, st eel cover hangs
the actual container for the fl ui d; i t i s made of brass and has a capacity of 50 m or 56 t ons.
The space between the two metal shel l s averages 2.6 m in width, and i s loosely f i l l ed with finely
powdered magnesium carbonate. The whole tank stands freely in a subterranean space, where i t can
be approached and examined from al l si des. This room has a completely normal cel l ar climate; no
noticeable drop in temperature can be observed. The outer st eel cover of the tank i s also at
normal temperature; moisture in the ai r does not condense on i t . The 140 kg. of Og vaporized
each day are collected in flasks and are used for operation of the t est i ng st at i on and for welding.
Fig. 25 shows a schematic diagram of a large container for a million tons of l i qui d, which
has a dai l y evaporation of 13,000 kg. for a 10 m. i nsul at i on thickness. This container, when
once f i l l ed and l ef t alone, will become empty only aft er 200 years. The amount evaporated dai l y
can be used by simply f i l l i ng st eel flasks and t ransferri ng them t o the consumer. The cyl i ndri cal
inner container of Cu-alloy has a diameter of 103.S meters, and, including the arched base, a
height of 119 m. The bottom end i s conceived of as a hanging floor which r est s on a ri ng running
along the boundary of the outer cylinder. This ring i s supported on the ground by poorly con-
ducting pi l l ar s placed at i nt er val s. The top cover i s hung on the outer s t eel cover at various
points. The i nsul at i ng layer of Loose magnesium carbonate powder i s around the inner container,
and i s 10 ra. thick on the sides and cover, IS m. thick below the floor. This i nsul at i ng layer
is bounded and supported hy the st eel cover which i s ri gi d and strong. The whole thick-walled
container stands on a cyl i ndri cal subterranean bunker of reinforced concrete.
The liquid Og comes in through the intake at the l ef t and i s led through the intake pipe
to the floor of the container, in order t o exci t e as l i t t l e motion of the oxygen during f i l l i ng
as possi bl e. At the top of the tank i s the drain pipe for the vaporized gaseous CJ2, by means of
which normal atmospheric pressure i s maintained above the l i qui d surface. The liquid i s removed
at the lowest point of the container. At the same level aa t hi s point, a pump system stands on
the ri ght at the edge of the bunker, and feeds the l i qui d to the out l et at the top edge of the
bunker.
Further det ai l s and auxi l i ary equipment of the tank will not be mentioned here, hut the
behavior of the liquid 0>j in the large tank requires further consideration. In the liquid pool,
which has a depth up to 117 meters, the hydrost at i c pressure, as shown in Fig. 25, increases with
the depth up to 13.1 atm. The boiling point of O2 depends on the pressure, and increases with
depth in the liquid from -183 C at 1 atm t o -144 C at 14.1 atm. In spi t e of t hi s , the ent i r e
contents of the tank will stay at -183 C, the temperature of the surface of the l i qui d. For i f
the masses of l i qui d down below should warm to higher temperatures than the upper l ayers, t hei r
density would decrease and a convection would occur which brings the warmer masses upward to the
region of lower hydrostatic pressures, where as a r esul t of the decreased pressure they wi l l
begin t o boi l , give off the beat of vaporization of the l i berat ed gas, and cool to the temperature
of t hei r surroundings. Since t hi s process holds for layers at any depth, the lowest temperature
at the liquid surface wi l l bring al l the lower layers to the same temperature. Thus in pract i ce
one wi l l act ual l y find the liquid at the bottom of the tank to be at -183 C. Since the heating
occurs mainly through the side walls, the movement of the fluid wi l l be such that the heated boi l -
ing masses of liquid r i s e along the walls, while the cold masses sink in the middle of the tank
so that the t oroi dal streaming shown in Fig. 25 develops.
Liquid On, whose combustion propert i es and storage have been discussed in det ai l here, need
not give, in combination with the previously described fuels, the great est exhaust speed. There-
fore, two further notable candidates wi l l be discussed bri efl y - fluorine and ozone (or ozone-
enriched liquid &>).
Fluorine would be used only for met al l i c fuel s. The corresponding metal fluorides in some
cases have higher heats of formation and lower heats of vaporization than the oxides, give the
prospect of higher exhaust speeds, but are not considered further because of the technical di f f i -
cul t i es in using liquid fluorine in place of liquid On. The use of pure ozone can be easi l y
checked for the case of octane. Since ozone has avai l abl e a di si nt egrat i on energy of 710 " H i
the heat-output of the mixture increases from 2587 ^M* for O2 to 3140 j|l- for ozone, and the
t heoret i cal exhaust speed e
t n
= 4655 n/
8ec
increases By 10K to 5120 m/
s e c >
The use of pure ozone
increases the effect i ve exhaust speed by t hi s same order of magnitude.
To t est the appl i cabi l i t y of ozone t o rocket s, H. Schumacher in Frankfurt made t,ests on a
small scal e, with the following main r esul t s;
Both gaseous and liquid pure ozone are explosive under the working conditions in fl i ght ,
and are thus not sui t abl e. Gaseous ozone - Qj mixtures at N.T.P. will react , in pipes or spheri -
cal chambers, st ar t i ng at a 10% weight fraction of ozone, i f the reaction i s i ni t i at ed by an
10t
c-ai
Figure 24; External view of l i q . 0^ tank of 56 ton
capacity) 2.6 m i nsul at i on thickness and
HO leg daily evaporation.
45
n-BU*
u Figure 25;
4^3 Sketch of an enormous tank for 1 million tons of liq.
<J 3, with a diagram of the pressure and boiling point ratio
a-riin the liquid content.
incandescent body, but the reaction wi l l not spread to al l of the mass unt i l the ozone concentra-
tion reaches 17% by weight. Even then the pressures developed in the reaction are sc moderate
t hat they might be expected to require motors and armatures.(?) The pressure increases are
of the order of 2 atm. A luminosity s t a r t s only at ozone concentrations of 40-50% by weight, and
the pressure increases reach 6-10 atm.
Liquid ozone-oxygen mixtures at atmospheric pressure and the corresponding boi l i ng point
seem to have, for weight % of ozone above 25%, a tendency to explode with great dest ruct i ve ef-
fect, i f gaseous ozone above them explodes with emission of l i ght . The mi sei bi l i t y of l i qui d
ozone with liquid On and the boiling point of the mixture are suffi ci ent l y understood from the
sol ubi l i t y diagram and vaporization curve at atmospheric pressure, that we assume the following
regarding the storage of large quant i t i es of the mixture in open containers:
Mixtures up to 25% by weight of ozone are st abl e, do not separate, and have a temperature
of -183
s
C. Since such mixtures seem to he safe from explosion, they are of importance for tech-
nical use. Mixtures between 25 and 55 weight % of ozone are not st abl e; they s pl i t i nt o a heav-
i er , deep vi ol et , ozone-enriched, phase (>55% 3) , which sinks to the bottom, and a l i ght er , l i ght
blue, Og -enriched phase 075% "2) , which fl oat s above. The phase which sinks i s probably ex-
plosive. Mixtures with more than 55% ozone by weight are again st abl e but in danger of exploding,
so they are without immediate technical value.
For increased pressures the cr i t i cal sol ubi l i t y temperature of -179.5 is quickly passed,
and a mi sci bi l i t y gap no longer occurs. Liquid ozone-liquid oxygen mixtures which are stored
for long periods become ozone-enriched because of the more rapid boiling off of On, so t hat the
resul t ant explosive tendency of the tank contents must be counteracted by adding On. The gas
phase above the l i qui d surface reaches the cr i t i cal ozone concentration of 40-50 weight -% only
for ozone concent rat i ons^ 90%in the liquid phase, so> t hat the products of vaporization of liquid
ozone - On mixtures of up to 25% ozone content are scarcely dangerous even for rapid vaporization
on uncooled machine par t s . Fig. 26 shows the behavior of liquid ozone-liquid Og sol ut i ons.
A few other oxidizing agent's are mentioned in the f i r s t part of t hi s book; two of them,
HgOg and HNO3 have at t ai ned some pract i cal value for cert ai n rocket uses.
Aside from the three groups of non-self-acting rocket fuels : combustion of hydrocarbons
with Oj or ozone, burning of l i ght metals in Ck) or fluorine, and the combination of the two
groups i n the form of l i ght metal-hydrocarbon suspensions, a fourth group of sel f-act i ng fuels
i s possible, which use the heat l i berat ed in formation of molecules from the substance in atomic
form: atomic nitrogen (9) ( 11)
2N =. N
2
+ 6050 A-s' /kt . G
th
- 7120 m/
s e c
and atomic hydrogen
2H= H2+ 5l4O0*"7kg ( l l , 32) , c^, = 20,800 m/
s e c
Since the l i f e times of these unstable materials i s very short {the l i f e time of active hydrogen
i s given as at most 10 sec [?, page 253j, research in t hi s branch of rocket fuel development
must f i r s t take the following l i nes: 1. Finding a basic method for prolonging the l i f e of active
nitrogen or hydrogen. 2. Determination of the dependence of lifetime on temperature and pressure,
especially in the di rect i on of very low temperature. 3. If necessary, development of a method
for enriching liquid or solid material with the mOnatomic modification.
Though the di f f i cul t i es of .such research may be great , and the prospects of technically
valuable resul t s are small, i t should be noted t hat , because of the 10-20 times greater energy
Concentration of H as compared to present l y available fuels, the more favorable specific heat,
the higher St values - i . e . , greater j e t efficiency, the higher reaction vel oci t y - i . e . , great er
furnace efficiency for smaller furnace volume, and the highly diathermic behavior - i . e . , sl i ght
thermal st r ess on the furnace walls; even par t i al successes, say a 10% enrichment of H in Ho by
perhaps dissolving gaseous H in liquid H9, would be of extraordinary technical importance. For
the limiting case of 100% N or H concentration as st ar t i ng fuel for combustion, the following
charact eri st i c quant i t i es were calculated for a furnace pressure of 100 atm:
No: T = 8260 ;9fe 1.49; {% = 0.97; C= 4690 m/
8ec
H
2
: T. - 5500 ;9U 1.49; ^ - 0.97; c = 14100 m/
s e c
For atomic hydrogen these values are plotted in Fig. 27 for various flame pressures. This
diagram shows cl earl y the extraordinary properties of active hydrogen; i . e . , that the probabi l i t y
of regaining large portions of the di ssoci at i on energy by after-burning in tha j e t i s very l are.
-100|
Region of mutually
non misciMlity %wft
(mixture separates u
into ? phases of S.
non-equal density) 8
Practical,
immediately
usable non-
dangerous
region.
Oaone boiling point
Explosive region
above a. licuid
surface.
Region i n which an
explosion in the
gaseous phase chan
to the licuid phaa
ao that also the
liquid phase can
explode*
Cri t i cal temperata.:
of solution
Ozone boiling poln
Pig. 26*
Weight % of ozone
Behavior of mixture of liqald ozone and liquid oxygen at 76O nm
pressure according to Schumacher-Frankfurt.
48
10000 iou
9000 90
Heating value 5I4G0 kcal Ag
Conditionally
avai l abl e
chemical heat
of separation
(Dissociation)
*W00 g W
o
o
p
S
H
<300O*40
^t ft
j&MMB 70
I, Usable heat
^content
(tfcm
7 . J T 3
Common logarithm of fire pressure
Figure 27; Dissociation, heat content, combustion temperature,
theoretical and maximum exhaust velocity and adia-
batic exponents for fire gases in the case of assoc-
iation of active H2 to molecular H2 with static
equilibrium.
49
C-8H296
Finally another group of fuels i s worthy of note: These fuels baaed on nuclear reactions
can r esul t in exhaust speeds of 10
6
-10
8
/
Be e
i snd have recent l y, because of the reactions in
*a*iuA fi ssi on, aeved iirte the deaw # technical ite>eefc. ( 3 ) .
Summarizing the numerical r esul t s of t hi s section concerning the problem of exhaust speed*,
we nay s>y t hat by means of stochiometric combustion of hydrocarbons in 0, in rocket motors a t
100 atm. furnace pressure, exhaust speeds aboye C 3100 /
s e c
are possi bl e. For excess fuel,
5% higher values are obtained. By enriching (he 0g with ozone, a further increase of exhaust
speed t o C= 3400 m/
c
may be possible with stochiometric mixtures. The use of Al hydrocarbot
suspensions with liquid Oj gives si mi l ar exhaust Bpeeds, but more favorable proportions by weight
on the ai r cr af t , because of the higher fuel density. We nay expect exhaust speeds, of rocket
motors in f l i ght , over 3S00 W.^c *
o r
^ l " ^ % witlh l i qui d Gg, and over 4000 m/gee *' ozone i s
included, while the addi t i on of atomic hydrogen would give even higher values. For cal cul at i ons
of fl i ght and mi l i t ar y performance of the. rocket bomber we shal l use the values c = 3000 m/
s e c
t
and C= 4000 n /
s e c
- To see the effect s of higher exhaust speeds we shal l cal cul at e with c =
5000 m/
s e c
for comparison.
3. Properties of the Mr-Vraine
The external appearance of the rocket l^oiiber i s shown in Fi gs. 28-31 and discussed theo-
r et i cal l y in the next sect i on. The bow of the ai r cr af t ' s fuselage consi st s of an "ogi val " with
9.6 cal i ber radius of curvature, which i s cut by a plane through i t s long axis so t hat a f l at
underside r esul t s for the fuselage. Between the wings the semi-ogive goes over i nt o a roomy
chamber with perpendicular side wal l s, while the fuselage gradually tapers toward the st ern\ wi t h
a steady decrease in cross-sect i on. The large blunt end surface at the st ern of the fuselage i s
necessi t at ed by the si ze of the mouth of the j e t of the rocket motor. The r el at i vel y small wing
stumps serve mainly for st abi l i zat i on in f l i ght , and for landing; the wing cross-sect i on i s the
well-known t ri angul ar wedge profi l e with a maximum thickness of 1/20 of the depth at 2/3 of the
wing depth. (16, p. 170). To t hi s pecul i ar ai r cr af t shape there correspond the laws of flow for
very high Mach numbers. An angle of incidence between fuselage and wings i s unnecessary, so t hat
for the low-wing arrangement chosen, the l i f t i ng f l at surfaces of the fuselage and wings go over
into each other without a break, as can be seen most cl earl y in Fig. 31. For the t a i l surfaces,
a symmetrical quadrangular cross-section was chosen, which also has a great est thickness of 1/20
of i t s depth in the l as t t hi rd of i t s depth. The whole arrangement of the t a i l surfaces i s i n-
dependent of the streaming from the rocket, j e t , since use of the rocket motor and fl i ght below
sound vel oci t y never occur together.
The s i t e of the rocket bomber was chosen as a compromise between a ser i es of contradictory
requirements. The idea of making the ai r cr af t as large as possible i s suggested by the fact t hat
then the r at i o of additional load t o weight when empty i s generally more favorable, that the con-
st ruct i on of l arger rocket motors i s simpler, t hat with increased si ze of ai r cr af t the mi l i t ary
strength of a rocket bomber group increases while the number of capable pi l ot s required per uni t
of load transported decreases. If one computes a few comparison designs in the range of 10-100
tons starting-weight, one finds that with increasing weight of the ai r cr af t , the aerodynamical
l i f t i ng power contributed by the fuselage represents (for geometrical reasons) a smaller part of
the t ot al weight, so t hat the wings have t o be r el at i vel y l arger; fi nal l y the weight of the wings
predominates, without giving any noticeable improvement in gliding angle in the region of high
Mach numbers. Such considerations lead t o an apparently favorable take&ff weight of 100 tons, t o
which corresponds an empty-weight on landing of 10 tons. Thus a limit of 90 tons of fuel with
about 76 m
3
tank space must be included, which leads to the fuselage dimensions shown in Fig. 28.
The wing dimensions are determined by the permissible wing loading of the bomber. The
st ar t i ng procedure by rocket cat apul t , which has already been described bri efl y, permits pr act i cal -
ly high wing-loading; thus even though before landing the consumption of al l fuel and removal of
al l bal l ast reduces the weight to 1/10 the takeoff weight, the landing speed determines the wing
si *e. Though landing speeds of over 200 km/hr can be used in special cases, a permissible l i mi t
of 150 km/hr was f i r s t chosen because the landing of the rocket bomber i s to be considered a
glide-landing, and because one cannot count on the avai l abi l i t y of experienced t est pi l ot s . The
corresponding stagnation pressure i s q = 110 kg/m^, From Fig. 34, the very thin and sl i ght l y
curved wing profi l e leads one to expect a maximum l i f t coeffi ci ent of onlv C
a ma x
= 1.25 even
with landing ai ds, so t hat the wing-loading before landing i s limited t o <( C
a ma s
=- 137.5 kg/m^.
Aside from t hi s fi gure, the wing size i s determined by the fact t hat the fuselage ~plays an impor-
t ant role in the l i f t i ng power of the whole ai r cr af t . According t o the i nvest i gat i ons of the
next auction, for large Mach numbers 2/3 of the t ot al weight i s carried by the fuselage, and 1/3
by the wings. At landing speed, the l i f t coeffi ci ent of the fuselage, at the angle of attack
for maximum l i f t , i s C
B
= 0.45; with the already fixed fuselage supporting surface F
t
= 80.8 nr
the t ot al l i f t i ng power i s F_. C
a
. q = 4000 kg, while the residual landing weight of 6000 kg goes
50
C~B296
Suppo
Surfa
I
Figure 26; Outline sketch
of rocket bomber
of 10 T empty weight.
ZB.
6 2
t>~-
in
10
Fig. 29: Top view of rocket bomber of 10 ton empty weigit.
i
? 2
Fig. 30: Outer fotm of Socket Boniber of 10 tons empty
perspective view from front-above.
Fig. 31s Perspective view from "backwards below" of 10 ton emp
Rocket bomber.
t o t he wi ngs. Thus t he r equi r ed wing s i z e i s about 44 m*; accor di ng t o Fi g. 28, t he t o t a l sup-
por t i ng s ur f ace i s 125. 5 m
2
; t he mean wi ng-l oadi ng i s 10, 000/ 125. 5 a 79. 7 kg/m
2
and t he l i f t co-
e f f i c i e n t f or l andi ng i s C j S A/J.F = 0.74-i i n agreement wi t h Fi g. 34.
At t ake off, t he mean wi ng-l oadi ng i s t en t i mes as hi gh, i . e . , 797 kg/ or ; for t he assumed
cl i mbi ng speed of 500 m/
B e c
t he s t agnat i on pr es s ur e at t akeof f i s 15930 kg/m
2
and t he t o t a l l i f t
c oe f f i c i e nt C_ i s 0. 05, cor r espondi ng t o an angl e of at t ack ^= 3 , whi l e t he l i f t c oe f f i c i e nt
a t optimum gl i di ng angl e i s C_ = 0.173 for v/ . = 1. 5. At a 3 angl e of a t t a c k, 38% of t he t o t a l
wei ght of 10
s
kg f a l l s on t he f usel age, and 62% on t he wi ngs; t he wing l oadi ng i s t hus 1390 kg/ m
After t he a i r c r a f t r i s e s t o t he optimum angl e of a t t a c k of 8 , C^ becomes 0. 173; t hus t he l i f t
for haul i ng t he c r a f t a t V/_ = 1.5 i s 346000 kg, o l which 48% r e s t s on t he wings wi t h C
a
p = 0. 08S,
cor r espondi ng t o a wing l oadi ng of 3720 kg/nr* i n t h i s s t a t e of f l i ght which det er mi nes t he wing
s t r e s s .
Ot her cases of l oadi ng ar e i mport ant for var i ous pa r t s of t he r ocket bomber; e . g . , for t he
fuel t anks i t i s t he a c c e l e r a t i on at t akeoff; for f usel age and p i l o t i t i s t he a c c e l e r a t i on at
t he end of t he cl i mb; f or fusel age and l andi ng gear i t i s t he l andi ng which i s i mpor t ant .
I n es t i mat i ng t he r a t i o of empty-weight G, a l t e r consumption of a l l fuel and droppi ng of
t he usef ul l oad, t o t he f l i ght weight G, immediately a f t e r r i s i n g from t he ground, we s t a r t e d
from t he known weight d i s t r i b u t i o n f or overl oaded long range a i r c r a f t : ai r f r ame 18%, power pl ar-t
13%, a u x i l i a r i e s 3%, a ddi t i ona l load 66%, so G/QQ 0. 34. The main pa r t s of t he r ocket bomber
were est i mat ed BB: cabi n - 500 kg, r ocket motor - 2500 kg, wings - 2500 kg, a t o t a l of 56
kg/m
2
; f usel age 3250 kg, t a i l , l andi ng gear , bomb-hay, e t c . , al t oget her 1250 kg; t hus t he t o t a l
weight of t he a i r c r a f t i s 7000 kg, whereas f l i ght wei ght s for or di nar y bombers of s i mi l a r s i z e
ar e about 3000 kg. The reason t ha t t he r ocket bomber wi t h i t s 10 t i mes gr e a t e r t akeof f weight
i s onl y 2. 3 t i mes as heavy as an or di nar y bomber Oi t he same s i z e i s mai nl y because t he s uppor t -
i ng s ur f a c e s , e s pe c i a l l y t he wi ngs, car r y not 10 t unes but onl y 3-4 t i n e s t he wei ght , whi l e t he
remainder i s d i r e c t l y support ed by t he a i r wi t hout any i nt er medi ar y s t r u c t u r e ; moreover because
of t he de f i ni t e way of s t a r t i n g and cl i mbi ng t he f act or of s af et y f or t he r ocket bomber need be
onl y a smal l f r act i on of t ha t for an or di nar y bomber. Thus t he t o t a l wei ght d i s t r i b u t i o n f t he
r ocket bomber i s ai r f r ame 7%, power pl ant 2.5%, a u x i l i a r i e s 0.5%, a ddi t i ona l l oad 90%, so
<VGo
S
0 4 , All t hes e cons i der at i ons ar e val i d f or s cal ed- up wei ght s . One t hus obt ai ns d i r e c t l y
t he cor r ect performance f i gur es , i f t he bomb l oad i s di mi ni shed by t he excess wei ght , above 10
t ons, of t he a i r c r a f t .
Fi gs . 32 and 33 show an over al l schemat i c of t he r ocket bomber.
The f r ont view of t he c r a f t does not show t he r e t r a c t a bl e f r ont wheel, which oper at es i n
conj unct i on wi t h a r e t r a c t a bl e t a i l ski d and fche l andi ng gear which i s r e t r a c t a bl e i nt o t he
fusel age betweeh t he wi ngs. The front wheel s er ves t o pr event dangerous cont act wi t h t he ground
of t he bow end dur i ng t he bouncing motion of t he a i r c r a f t dur i ng l andi ng, and t o slow down (wi t h
t he ai d of t he l andi ng gear ) as qui ckl y as pos s i bl e t he a i r c r a f t which comes on t o t he ground a t
150 km/hr and has p r a c t i c a l l y no wind r e s i s t a nc e t hen because of t he smal l wi ngs. Behind t he bow
i s t he pr e s s ur e - t i ght cabi n, i n which t he s i ngl e p i l o t s i t s . I t i s t i g h t for i ns i de pr e s s ur e s
of 0. 4- 0, 5 atm. wi t h vacuum out s i de, and should per mi t r api d e xi t of t he p i l o t i n case of danger
( e . g. a f t e r t akeof f ) . Because of t he smooth ext er nal shape, v i s i b i l i t y from t he cabi n i s very
poor . In f r ee f l i ght a t high ve l oc i t y, s i de view s l i t s and opt i c a l ai ds ar e s uf f i c i e nt . For
l andi ng a ki nd of det achabl e wi ndshi el d can be used, s i nce t han t he p r e s s u r i i a t i o n of t he cabi n
nd mai nt enance of t he bul l et - s hape ar e uni mpor t ant . A f ur t her e s s e n t i a l arrangement f or t he
cabi n i s t ha t t he p i l o t ' s seat be so arranged t ha t t he p i l o t can t ake up t he hi gh a c c e l e r a t i ons
al ong t he a i r c r a f t axi s in t he bes t pos s i bl e pos i t i on, so t h a t not onl y body and head, but a l s o
f eet and arms have good suppor t i ng s ur f aces , and a t t he same pos i t i on can be s hi f t e d. The r e -
mai ni ng equipment of t he p i l o t ' s cabi n - i ns t r ument s , D/F and r adi o equipment, v e n t i l a t i o n , e t c .
i s not consi der ed f ur t he r . At t he back of t he pr es s ur i zed cabi n ar e t he t ank i n s t a l l a t i o n s , which
c ons i s t of two l ar ge t ubes 20, 5 m. long and wi t h maximum di amet er 1,9 m; t hes e c ons t i t ut e t he
main pa r t of t he f usel age. The upper f our t h of t he t ubes ' ci rcumference forms t he ski n of t he
a i r c r a f t , whi l e t he lower hal f and t he space between t he t ubes i s covered so t ha t t he r equi r ed
shape i s obt ai ned. For cons t r uct i onal r easons t he t ank t ubes ar e subdi vi ded i nt o t he act ual con-
t a i ne r s by means of c r os s - wa l l s . The purpose of t he cr os s - wal l s i s f i r s t t o have s epar at e con-
t a i ne r s of c or r e c t capaci t y one behi nd t he ot her f or fuel and l i qui d O2 and t h a t each f uel s ha l l
l i e symmet r i cal l y wi t h r es pect t o t he axi s of t he a i r c r a f t , so t ha t t hermal s t r e s s e s and t wi s t s
a r e not developed due t o asymmet ri es; second, as a r e s u l t of t he s ubdi vi s i on, t o lower t he l i qui d
pr es s ur e on t he r e a r end of t he t ank dur i ng a c c e l e r a t i on and t o pr event t he a i r c r a f t from becoming
t ai l - heavy as t he t anks ar e emptied; f i na l l y t he cr os s wal l s gi ve t he t hi n- wal l ed f usel age t he
s t i f f ne s s necessar y f or t aki ng up t he t or ques a t t he r oot s of t he wing*-spars. I t i s advi s abl e
t o put t he oxygen t anks i n t he f r ont end of t he f usel age so t h a t t he force dr i vi ng t he On t o the
55
e-s*
Figure 32; Scheme of Space Distribu-
tion of rocket 'bomber of
10T empty weight
Fuel tanks
^
s
^ Hoc
p | aux
Fuel tanks Bomb Retracted l
Figure 55j Total view of 10 ton Rocker Bomber
?
CD
pumps wi l l be as large as possi bl e. Between the wings and the tanks i s the bomb bay, sui t abl e
for proj ect i l es up t o 30 tons in weight. When the bombs are released, the floor of the bomb bay
must be opened completely for a short time so that the axis of the bomb i s par al l el t o t hat of
the ai r cr af t , since in the perpendicular posi t i on premature detonation may occur. Just before
the main spar meets the fuselage there are chambers for the extendable landing wheels; these
chambers are par t l y on the ver t i cal sidewalls of the fuselage and par t l y recessed i nt o the tank
cylinders.
Finally, Fig. 33 shows the rocket motor at the end of the fuselage;
4. The Glide-Number of the Ai r-Frame
During i t s fl i ght the rocket bomber goes through velocity ranges with ent i r el y di fferent
flow char act er i st i cs, e. g. , the ordinary Subsonic range, the supersonic range up to three times
the velocity of sound, the domain of large Mach numbers V/a, in which Newton's law of ai r r es i s t
ance i s valid - i . e . , the aerodynamic forces vary with the square of the velocity and the angle
of attack; and fi nal l y the range of gas-kinetic streaming with very long free paths of the ai r
molecules and thus speci al laws of ai r resistance which are s t i l l similar to Newton's.
In al l these regions the ai r cr af t n.ust have suffi ci ent s t abi l i t y char act er i st i cs, whereas
the glide-angle i s important chiefly in the Newtonian region for both density types, for i f the
rocket bomber at the beginning of i t s glide had Mach number 30, then 99% of i t s ki net i c energy,
would be consumed in the Newtonian region,, and only \% below a Mach number of 3.
In the region of higher densi t i es the gl i de angle can be estimated at only a few speci al
points on the Mach scal e, say V/
a
=
0. 1; 1.5; 3 and o , whereas for rarefied ai r closed gas-
ki net i c formulae can be given.
For a landing speed of 150 km/hr, i . e . , V/
a
=0. 12, polars of the bomber and the wings
were determined by wind tunnel t es t s on a 1:20 scale model; the r esul t s are shown in Fig. 34.
The best reciprocal glide number i s / ? = 7.75 for angle of a t t a c k- ^ 5; the best l i f t coef-
fi ci ent i s C
a ma x
- 0.575 f or =16. 5
0
,
For V/- between 1 and 3 the difference in pressure compared to the pressure of s t i l l ai r
i s , for two-dimensional flow at small angles of attack,
according to Ackeret and Busemann; for spat i al flow around the rotation-symmetric end of a cone
according to Busemann and V, Ka~rman, and the tangential fri ct i on st resses are T/ = 0.072 (QT)
according to V. KarmSn- With the aid of these rel at i ons the values given in Fig, 35 and 35, for
polars of the wings, fuselage and complete frame of the rocket bomber at V/
a
* 1.5 and 3, can be
computed. From the Figs, we read off the best reciprocal glide number Vf = 3.94 for V/
fl
= 1.5
and*-*8 and Vf
:
3.83 for V/ = 3 andoC-*?
0
. Kind tunnel measurements of the pol ars, which
are feasible in t hi s domain, could not be carried out. In the calculation a t ot al vacuum was
assumed behind the stubby st ern-surfaces. Actually- these surfaces,' in t hi s Velocity range where
the motor i s off and they are not in contact with the flame, can be acted on by oticeable pr es-
sures which make the wind resistance appear to be l ess than when the motor i s #oing; t hi s must
also be noted for wind-tunnel measurements.
For V/ ^o^ the thermal vel oci t i es of the ai r molecules and the pressure of the undisturbed
ai r are respect i vel y negl i gi bl e compared t o the fl i ght speed and the dynamic pressure on the sur-
faces struck by the ai r stream* The part of t hi s uormal pressure ^>j which ari ses from molec-
ules bouncing i nt o the wall can be taken di r ect l y from Newton's law for i nel ast i c col l i si ons:
/tff/er~gs//7
a
*< Whether there i s a further contribution to the pressure depends on how
these molecules leave t he surface. For dense ai r t hi s must take place along the surface of the
pl at e. If the wall were f l at and the density of the gas layer which streams away were i nf i ni t e,
then the ai r molecules coming off would receive no acceleration perpendicular t o the pl at e and
there would be no further pressure contribution. According t o V. Karman the r at i o of density
i ncr ease^JHo de ns i t y^ before col l i si on i s 4 ^ - - ? ^ - / wh e n the corresponding ai r pressure
r at i o is AfAf
00
, so t hi s case can occur only for 9t - \. For 9i - lA,*pfr
z
5, the ai r den-
s i t y at the plate i s 6 times as large as for s t i l l ai r , the layer has.a fi ni t e thickness, the
angle of impact of the ai r i s greater than the angle of at t ack, and the ai r pressure yd i s
greater t ha n^ ' ; according to Busemann/^
:
(^- rJJSMeC. As far as the value of X i s concerned,
the molecular col l i si ons in the flow of the condensations are suffi ci ent in number t o fully exci t e
al l molecular rot at i ons; i.e.,3t - 1.4. If the high stagnation temperatures necessary for
c-ase
4
o
A
O
o
V
* *
4 4
r
v

"F
-?
^ ^
#
r
m,
^ B o mb e r wi t h l andi ng
_ ai ds (aatiBiD^o^^
<?6
7^
1 M v v u y
Bomber wi thout l and-
i ng ai ds (measured)
<?6
4
/
. *
, -
s
1 M v v u y
Bomber wi thout l and-
i ng ai ds (measured)
A f
^
/
#s ^
f
/Z"
A i .
4+
y*
J ) A
* w
$3
Off*
A A
5/
' \ Win
ai d
gs 11
a (1
irith
lean
out l andi ng
tired)
$2
A d
m
i3
at
n
g * ft
W
w # fc
n ft V A 4 20 0
a? a & Q* V Q, to
Coefficient of DragCw
figure J4; Wind tunnel measurements at Reynolds
1
Number 4x10 and
Hach number corresponding to the landing speed. (The aero-
dynamic coefficients are based on the area of the support-
ing surfaces.)
59
r-sM6
QM 0003 0,0ft 0,016 0,020 0,02k 0,028 0,032 0,036 0,M Oflkk WW 0,052 0,056 0,060
Coefficient of Drag Cw
Figure 35; Calculated polars of wings, fuselage, and of the entire
f rafts of the-Rocket Bomber in the gas dynamic region and
U s 1.5. These aerodynamic values are based on the area
of the supporting surfaces.
60
C-84!
QQ02 qotX QQM 0,006
opn. ojm o/m qo& O,QZQ o,m qo o,o ym o,ex
Coefficient of DragCir
Figure 36; Computed polars of the wing, the fuselage, and of the
entire frame of the Rocket Bomber at M 3 (The aero-
dynamic coefficients are based on the area of the lift-
ing surfaces*)
C-8
exci t at i on of molecular vibrations are reached, then considerable osci l l at i on can occur in tKi
available streaming time, i . e . , 9C approaches 1.29. For?* = 1.4 the pressure on a f l at surface
sd in
curvature
Ind undergoes acceleration perpendicular to the pl at e, so that for convex curvature a negative
additional pressure p
9
r esul t s. For *
;
1, these were calculated by Busemann for general curva-
of
the object, and vanishes, in the case of the lens pr of i l e, at the paint for which the angle of
attack isCts//3\ for the bul l et shape i t vanishes somewhat l at er , isXfy/g, so that the average
pressure on the curved surfaces i s far l ess than for plane surfaces, For3*?/ the centrifugal
effect s, according to Busemann and Guderley, are somewhat larger, so that the pressures for
<K ' 1.4 are 1% smaller than forrff
:
1, in the case of the lens pr r f j l e.
As regards the fractional st resses paral l el to the wall between the fixed surface and the
ai r , for dense ai r , momentum paral l el to the wall is transferred to the wall only by a thin
boundary-layer of molecules near the wall, so that the usual fri ct i on laws are val i d. One i s
led, in the case of high supersonic vel oci t i es, to give main importance to laminar fri et i onal
effect s, so that the fri ct i on is determined by V. Karmans formula *2/W = /fyflfr In
fc
be Valleys
of the fl i ght path of the ai r cr af t , which determine the energy consumption, we may use a value
of Re = 10^, which gives "?/ = 0.00013 for fl at , unfilled surfaces. At f i ni t e angles' of at t ack,
the density, fri ct i on and temperature change on the leeward side (negative angle of at t ack) be*
come zero, while for the windward side they are 6 times as great as for free ai r ; at the same
time the viscosity of the ai r increases with the stagnation temperature V
e
sift%a?gQaaccording
t o the rel at i on
The fri et i onal forces on the surfaces to windward are approximately
lyfy - ef x /.3/fc*
ej
fe?7J*r&J?4ffee<M^fe73}*
76
With the aid of these equations we can cal cul at e the aerodynamic forces on an arbi t rary
body iarfy-*
00
. Fig. 37 shows the polars f i r s t for the i nfi ni t el y thin f l at pl at e which i s
known to be t heor et i cal l y the best wing for fl i ght above the velocity of sound, second for the
wedge profi l e with f l at sides and a thickness 1/2 of the wing depth in the second t hi rd of the
profi l e (8, p. 170) and finally for the symmetric double-convex lens profi l e composed of two
equal ci rcul ar arcs, also with thickness 1/20 of the depth. In the region of V/
a
- 1 to 3, where
the l i near dependence of the aerodynamic forces on the angle of attack is valid, and where the
excess pressures on the windward surfaces and the subnormal pressures on the leeward surfaces are
of the same order of magnitude, the biconvex lens profi l e gives the best glide-numbers; in the
Newtonian region, where the ai r pressure vari es quadratically with the angle of at t ack, and where
the ai r pressure vanishes in the shadow region, the flat-surfaced wedge profi l e i s defi ni t el y
superior. In the region of angles of attack' which are greater than the front bevel angle of the
wedge,- i t i s as good as the i nfi ni t el y thin fl at pl aw. In addition i t has the remarkable prop-
ert y, in t hi s velocity range, t hat the ordi nari l y s t r i c t requirement of minimum profi l e thickness
i s weakened, in the sense that even the limiting profi l e can be so thick that the whole wing sur-
face i s in the shadow. The i nferi ori t y of the lens profi l e to the wedge profi l e in t hi s velocity
region ari ses from the fact that larger angles of attack give poorer glide-numbers; since the
pressure depends quadratically on the angle of at t ack, the large angles for the front part s of
the surface outweigh the effect of the small angles at the back part s so t hat altogether a poorer
glide-number r esul t s than for the f l at underside of the wedge profi l e. A further reason for the
i nf er i or i t y of the lens profi l e i s that the curved surface, because of the centrifugal action, i s
acted upon by smaller normal pressures and about the same fri et i onal st resses as for the wedge
pr of i l e; thus to obtain the same l i f t , larger surfaces must be used, which involves .not only
greater weights but also greater fri et i onal forces. Since the rocket bomber needs the best pos-
si bl e glide number at high vel oci t i es, i t should be designed with wings having a wedge pr of i l e.
These considerations can in pri nci pl e be extended to spat i al flow. Fig. 38 shows the
polars for 3-dimensional flow around three surfaces with equal projected area and the same height
which might be considered for the bow shape - a ci rcul ar cone with height four times the diameter
of the base, a bul l et with the same height and base diameter (16.25 cal i ber radi us), and fi nal l y
hal f-bul l et with the same height and a semicircular base of the same size as t hat of the two
surfaces of revolution, i . e, with 8.25 cal i ber radi us. The aerodynamic coefficients are al l for
62
0,020
1% ft MS
0,020
1% ft MS
V
"OfiW
0
0,016
o
4-1
"OfiW
0
0,016
o
4-1
"OfiW
0
0,016
o
4-1
Pl at Pl at e
"OfiW
0
0,016
o
4-1
fa/Cw">0,t beia-3,7 ^
"OfiW
0
0,016
o
4-1 Wedfl-e Prof i l *
""7
0,010
0,009
ABOM
L(at/Cw-B76 beiec*#*) ""
""7
0,010
0,009
ABOM
*
#
0,010
0,009
ABOM
V
*y
0,010
0,009
ABOM
0,010
0,009
ABOM
4 ' /
/ ^i j ent i cul ar
Pr of i l e
0,010
0,009
ABOM
fcyfc-$ */*;
OftOf
QMS
OJ005
0flO4
0,003
0,002
0,001
OftOf
QMS
OJ005
0flO4
0,003
0,002
0,001
y
Jjl
OftOf
QMS
OJ005
0flO4
0,003
0,002
0,001
OftOf
QMS
OJ005
0flO4
0,003
0,002
0,001
lv
OftOf
QMS
OJ005
0flO4
0,003
0,002
0,001
OftOf
QMS
OJ005
0flO4
0,003
0,002
0,001
A
OftOf
QMS
OJ005
0flO4
0,003
0,002
0,001
OftOf
QMS
OJ005
0flO4
0,003
0,002
0,001
r
l
Tz
>
*
0" OflW
toe qms
oaf ficient of Drag fir
Figure 37; Folars of a plane curface* a wedge profile and a
lenticular profile at H-3 (The aerodynamic co-
efficients are based on the area of the lifting
surfaces)
the sane projected area. While the cone and bul l et differ l i t t l e in glide-number, the half-
bul l et with f l at surface t o the front, i s far superior, having a glidenumber with /'- 4.12.
l i t i s i s of decisive importance far the shape of the fuselage of the rocket bomber-* WhH* in *he
region V/a = 1 t o 3, the bul l et has the optimum glide-number, the laws determining the shape of
a body under 3-dimensional flow change for high Mach numbers in the same l ay as for the wing
sections; on the under side of the surface, which mainly determines the magnitude and di rect i on
of the net aerodynamic force, the ai r pressure should be as l arge, and the resistance components
as small, as possi bl e. The underside should consist solely of surfaces t i l t e d toward the course
wind. For a given average angle of attack of the undersurface, and for quadratic dependence of
pressure on angle of at t ack, one obtains the best r at i o of drag to l i f t on the underside i f i t
i s curved as l i t t l e as possi bl e in the di rect i on of flow. On the upper side of the body, which
with proper shape i s unimportant for determining the t ot al aerodynamic force, the pressures
should be as small as possi bl e. Therefore, the upper side should preferably consist only of
surfaces t i l t ed t o leeward. If t hi s i s not possible, the part s of the upper surface which are
t o windward should have the smallest possible angle of at t ack, and be curved convex t o the
di rect i on of flow, t o keep the streaming pressure low by taking advantage of the centrifugal
effect of the airmass streaming off the curved surface. Summarizing, for large Mach numbers,
bodies should be so streamlined with a point or dihedral at the front, t hat the l i f t i ng under-
surface is nol curved in the di rect i on of steaming, and t hat the upper surface should consist
as far as possible of surfaces to leeward; the unavoidable windward sections of the upper sur-
face should have a convex curvature in the direction of flow. (29)
The fuselege of the rocket bomber was shaped in accordance with t hi s recipe; i t s polars
forj&-#o are shown in Fig. 39. From the figure we see t hat the optimum angle.of attack for ^
wings and fuselage ar eot =f VO* andO-5*.5W' resp. Since the wing surface i s smaller than the
fuselage surface, the more favorable glide-number for the wings has l i t t l e effect compared t o
t hat of the fuselage, and one finds a t heoret i cal l y most favorable angle between wings and fu-
selage of -2 for which the optimum reciprocal glide number/ - 6.51 i s obtained for angle of
a t t a c k s S'SO'iot the wings and ct =*Z$o' *f<'
r
the fuselage. For constructional reasons, and
also t o have simple streaming conditions at the wing root s, the angle of incidence was chosen as
0; the polar for t hi s case i s shown in Fig. 36, which shows a best ^ = 6.4 f o r ^ ? " - This
value of glide-number, which can be at t ai ned by proper shaping of the fuselage and wings, i s
surpri si ngl y favorable for high supersonic speeds, and i s scarcely different from t hat of present
ai r cr af t below the velocity of sound. We may safely assume, for the velocity range --
:
30 to 10,
that the resul t s derived for j &- **
s
t i l l apply, and that below t hi s range the char act er i st i cs for
i
:
1 t o 3 gradually appear.
On the basi s of the previous investigation the vari at i on of glide number of the rocket
bomber for various angles of attack can be represented as in Fig. 40 for the whole range of veloc-
i t i es in dense ai r . The very favorable glide-number below sound-velocity, = a^,.. at &C = 3'
drops very rapidly as we approach the velocity of sound, then takes on the value s -^g^atoc=8"
for vel oci t i es sl i ght l y above the velocity of sound. Above -i. : 3, the glides-number improves and
then rapidly approaches i t s favorable behavior for high Mach numbers, with
= <&- *f-7
m
In al l the previous considerations on the aerodynamic forces, the ai r was considered as a
continuous medium. As the l at er sections show, the path of the rocket bomber reaches heights of
over 1000 km. There the densi t y of the ai r becomes so very small t hat on the one hand the st ag-
nation pressures for even extreme vel oci t i es no longer r esul t in aerodynamic forces at a l l com-
parable with the other external forces on the ai r cr af t , and also the laws for cal cul at i ng the
aerodynamic forces are no longer those derived for a continuous medium. Fig. 41 i s a plot of the
ai r density / i s a function of the heighfH for values up to 20,000 m from the well known for-
mula for a homogeneous isothermal atmosphere; for heights between 11 and 22 thousand m, the r el a-
tion i s <f/fa~ 4683?e'****
4
' (see also "Dinorm 5450" or (33)). This expression for density at
great heights i s used in the l at er (note: ori gi nal has error; uses J* instead of J^CT:M ) calcula-
t i ons. I t i s frequently assumed that the composition of the atmosphere at high al t i t udes i s the
same as on the ground - i . e . , mostly N2 and l ^, but t hat these materials are more or less di s-
sociated into atoms. Therefore two dotted curves in Fig. 41 are used to show the density i f al l
molecules are dissociated i nt o atoms ( l ef t curve), and i f only a par t i al di ssoci at i on occurred,
as assumed by Godfrey (middle curve). The molecular free path for constant composition vari es,
with the density according t o the rel at i on ^ *
0
~ /$' ^h*
5
rel at i on also holds approx-
imately for the dotted curves for the dissociated atmosphere, since the effect of di ssoci at i on
on the free path i s of the same order as the error in determination of the free path. At heights
of 40 km, the free path i s already greater than the thickness of the laminar boundary layers
(ZO~ /0r ) , reaches the size of the ai r cr af t dimensions at a height of 120 km,, and r i ses to
over 1000 km. Bt a height of 200 km. The stagnation pressure corresponding to the "vel oci t y of
escape" V : 12000 m/sec i s fess than 1 ^
E
/
m
2 at 100 km. al t i t ude; i . e . , at 100 km. al t i t ude the
aerodynamic forces have pr act i cal l y disappeared, and the motion of the rocket bomber i s along
pract i cal l y a pure i ner t i al path. The expressions derived for the aerodynamic forces on the as-
sumption of a continuous medium are already i nval i d at heights over 40 km, since the di st ances
4* 0,12 OH
Drag coefficient Cm
Fig. 33: Polars for three bodies having equal height and equal base cross-
sectional area.
1. A right circular cone
?. An ogival shape
3. A half ogival shape
Aerodynamic coefficients are based on the base area. Hach number
infinite. Gas dynamic region,
65
QOB
g ^
m
0,0*
0,03
0,02
Qfi1
0.00
-0,01
1 ' I
4- bt i - > **+ 4>**4 #*^4 n v t n a s i a+?*%m*+a
r
1
i
4. 1
i
4. 1
fl
witu x nc t i o n resi stance
fl
10
1
t /
/
* >*
^Complete
fXP**
*
f X
W
w'tyi ibeia~-7)
.
*~
j
ff
f f w
f
> >
* *
*Sl is el age
n
*Sl
t
t
$10*
I..W Wi ngs
s
j $
V
e
t9
ft
//
it
11
3L
^
0,OQZ Q,m 1^006 qOQ8 0,0f0 %01Z 0
t
01 0,(M
Coefficient of Drag Cw
Figure 39; Computed polar a of the wings, the fuselage, and the
entire frame of the Socket Bomber at H=3 (The aero-
dynamic coefficients are based on the area of the
lifting surfaces).
66
\
S
"
5
7
>>
6*
2
72
3
7
*
O 4
- 3"
7*
H
* if
-*
Jt
J
-
H
* if
0
t
i
o

1
1

'

i

\
/ <
/ / /
t
,
-4
1
1
_1
-H-
f * *
' / /
r~
w
"I
11
11
l l
1
/ /
/
r
4
>
/
*}*
U
*
*
0

P^
1
1
i30=*
4 *
y
*r
3,6
P^
1
1
<***
*>~~
. -
42
1
X
*
Ifi
&
10
%6

0,B
An
|
2 * 6 B 10 12 1 16 1B 20 22 2 26 iS 30
Mach Ho, v/ a
figure 40; Probable behavi or of l i f t - t o - d r a g r a t i o and best angle
of at t ack of fiocket Bomber at 0. 1 U ^ 30.
Log of the free path length in meters
-7 -t > -* - f -i
! "
' 6
t i
1 3 4
L 5 6 7
1



*
/

i





1
'W










i

1
I



m
m
/




i
-:
i




S
/



*
i
!
A
v
/7
1
\
^



*

N
v , J
f

*


v
\Composition and temperature of the
atmosphere at al l heights equal to
1


>
V\ ^ those of the standard 5450 st rat os-
\ phere.
r


\
The same assumption} but al l ai r




After investigations by Godfrey


9
1 /
'ProcJtySocAWfimjMl




V
J
M

- w


1
0 -1 f 'I i -i -4 -4 r -t - ;
r -i
i -i
9 -i V 1 V -ft -1
a -/
bgttfi
Pig, 41; Variation of the specific weight of the atmosphere, and
the free path length with the flight height K.
between ai r molecules are already larger than the dimensions of the streams assumed in he cal -
cul at i on, e. g. , the thickness of the boundary l ayer. For such rarefied ai r , no procedure for
cal cul at i ng aerodynamic forces i s known. Only when, l or al t i t udes aboe 100 am, the free puma
approach the dimensions of the moving body, do the r el at i ons simplify suffi ci ent l y to enable a
f ai r l y exact estimate of the aerodynamic forces. ( 28)
In t hi s region of gaskinetic streaming the following equations are valid for the aero-
dynamic forces on a body moving at ar bi t r ar y yeloc-ty V at al t i t udes above 100 Ion, where the up-
per (lower) sign refers t o negative (posi t i ve) angles of at t ack:
l*/
9
- + cos*, f.. ^^^s^ccsa(f^^G-^)
where V i s the speed of f l i ght , Cfl
:
ifc?
(
3'?' i s
, n e
**>* probable velocity of the gas raoleculet
before col l i si on with the wall, c ^ - 1 ^ ^ ^
-
i a the most probable diffuse recoi l vel oci t y of the
molecules aft er contact with a wal l TcT
w
D
K. For v.ery large V, ^-*^ approaches$M^<fc*/*ce
JG~*o{}C*--]/3r) Bb^LoP/fc approaches zero, i e. we approach the Newtonian f o r mu l a ^ &
=
* ? *' ' ' v*.'
Apresent at i on of these aerodynamic forces in closed form as a function of Mach number i s not
possible; though a simple rel at i on between Cy and g i s known C^j
:
V^wt '
t
'
w r e
*
s n o
correspond-
ing r el at i on between Cp and .
All the equations refer to diatomic gates or gas mixtures. Also they require the know-
ledge of the char act er i st i c temperature 9, and thud involve more special assumptions than do the
gas-dynamic equations which require only a knowledge of the no. of atoms per nojecule and the
sound-velocity, since they involve only 9*and a.
The equations for the momentum t-jtrf of the molecules col l i di ng with the wall, and for the
tangential momentum <V given to the wall on r ecoi l are taken from (28). The recoi l momentum murt
be considered in more det ai l .
Before an expression for the recoi l mnmentuia can be set up, one must -jve a cl ear pi ct ure
of the r ecoi l of the molecules, which have a random thermal velocity c and a r el at i ve vel oci t y V
making an angle Ot with the pl at e. One must decide whether a bundle of molecules, st r i ki ng the
pl at e with a perpendicular velocity component V s i n + C
x
CO&fp leave the plate by specular
refl ect i on or under some other defi ni t e or random angles; further, whether the col l i si on veloc-
i t i e s are conserved for each individual molecule or have a defi ni t e r el at i on to the recoi l veloc-
i t i e s , or whether they are completely random so t hat we can derive r el at i ons only for the average
or most probable speeds
t
on the basis of energy considerations. One should i nvest i gat e whether,
during the col l i si on, the *velocitiea Cg as well as V are conserved, or whether they interchange
t hei r ki net i c energy in whole or in par t , or whether the i nt ernal degrees of freedom of the ai r
molecules or wall molecules take part in t hi s exchange. These questions were answered on the
basi s of a paper of Frenkel on "Theory of Adsorption and Related Phenomena" ( 4) , whose opinion
we give verbatim: "No matter how small the time of adherence, gas molecules col l i di ng with a
wall' do not simply bounce off, but are emitted again with vel oci t i es which need not have any
defi ni t e rel at i on t o the ori gi nal vel oci t y, ei t her in magnitude or di rect i on. The usual pi ct ure
of el as t i c refl ect i ons of molecules, from which the ki net i c theory of gases s t ar t s in deriving
the pressure on t'le walls of the container, i s false in pr i nci pl e. Tf one arrives at correct
r esul t s in t hi s case, i t i s because the sol i d body and the gas are considered to be at the same
temperature and at r est r el at i ve to each ot her. Under these conditions, the velocity di st r i but i on
of the emitted molecules i s the same as i f they were act ual l y reflected from the wall surface.
Actually a quite di fferent pi ct ure would r esul t i f une bombarded the surface in a def i ni t e direc*
tion with a molecular beam." Summarizing these considerations we find t hat , in set t i ng up a for-
mula for the recoi l momentum, we mst consider a gas whose probable recoil velocity Co i s deri v-
able by purely energetic considerations from the probable vel oci t y Cfl before col l i si on, from the
systematic velocity V and the wall temperature T^. The mass of gas st r i ki ng the wall per second
i s :
In the normal gas-ki net i c case (V - o and Co
:
CJJ) the process of emission of the molecules can
be considered as quasi-specular, with recoi l momentum Z+gas^Cq^'CtioT a mass col l i di ng per
"~~ ~ ~ {*&?), K times as many molecules col l i de eack second nfjo"- J^u Gs For
t n e
moving pl at e C? *^ ) , K times as many molecules col l i de each
second as for the pl at e at r est , and j ust as many Oust come off per second i f none Sticks per-
manently t o the wall. Since according t o the research ci t ed, the process of recoi l i s com-
pl et el y independent of the process of col l i si on, and i s of the same type as for r ecoi l at V - o,
the only possible difference can consist in a change of the factor K for the col l i di ng molecules
i\g=.j}'a&H osfajas C*=C*f For the sane reasons, the difference between recoi l and col -
l i si on temperatures can produce no change in the st ruct ure of the formula for the r ecoi l momen-
tum; one gets for Ca^Cu, / p = fittoAr.ty . The probable t r ansl at i onal recoi l speed Cp must be
considered further. According to the theorem of Conservation of Energy, for the assumed closed
system the sun of the energy Eg taken away by the recoi l i ng molecules plus the energy f% remain-
ing at the wall must equal the energy E, of the impinging molecules: # = ir-f-/#
where the values C>^and Case have only a computational significance and are therefore not
used in what follows.
Usually the r at i o of the energy loss of the impinging molecules to the excess of t hei r
i ni t i a l energy over the temperature level of the wall aft er col l i si on is called the accommoda-
t i on coefficient GC ; e. g. the average accommodation coeffi ci ent for the whole process has the
value Qt = {tEA'^eJ/tffjr-^,/. The accommodation coefficients for the individual degrees of free-
dom can be written as:
z
The average accommodation coefficient, ril i s -built up from the coefficients for the separate
degrees of freedom as follows:
Considerations on accommodation coefficients in the physical l i t er at ur e st at e in essence t hat
the external degrees of freedom take part immediately and completely in the energy exchange
during molecular processes, whereas the exchange of i nt ernal molecular energy, especi al l y t hat
of vibration, i s slowed down considerably and i s effective only after long relaxation times, so
that the accommodation coeffi ci ent s for t ransl at i on are I and also o(
ra
may be assumed to be
en/, while the deviations of the t ot al accomracdation coefficient Si from 1, observed in pract i ce,
are to be ascribed to the slowness of changes in vi brat i on. romO(t
no
,=/ i t follows' t hat
Cje*ran* = Cu,t
r
*s =/2g#7Z, where Cp** is thus the desired most probable recoi l velocity of
the molecules leaving the wall diffusely; T^ is the wall temperature in K,
Before cal cul at i ng the. aerodynamic forces one must estimate the wall temperature Tu. This
i s done by equating the energy radiated by the wall per cm
2
per sec. "
s
~i ^ S ('^t)
t o t n e
energy Eu transferred by the ai r molecules to the same surface in the same time. If one i nt r o-
duces the previously est abl i shed assumptions ttr-atrs /
3
<rot =/j=*'o&c = O . one obtains
#= 4p/iL~j-<?#( TG'TMTJJ where 3" is the mass of the molecules st ri ki ng unit area of the
pl at e per uniF time. These expressions were used in constructing Fig. 42, where the following
numerical values were assumed; temperature of s t i l l ai r TQ - 320 K, density of s t i l l ai r
f
jO>s/j~%y<tec%,*, (corresponding to H 9 0 km,) composition of the ai r 1455 O2 + 86% N
f
. Ioniza-
tion and di ssoci at i on were not considered; also the atmosphere was assumed to be at rest on the
ear t h' s surface. The optical absorption of the wall was assumed to be E - 0.80. Radiation of
heat to the wall from the outside (sun), from the ai r space or the earth 5 surface was neglected,
as well as from the i nt er i or of the ai r cr af t to the wall. Fig. 42 shows that the equilibrium
temperature of the pl at e at , r es t is 13$ K and t hat t hi s equilibrium temperature of the surface
of the ai rcraft increases with increasing velocity and angle of attack, as long as the l at t er i s
posi t i ve so that the surface i s t i l t ed toward the course wind. For surfaces to leeward the
equilibrium temperature drops to absolute zero, since the heat supply vanishes with the number
of impinging molecules. Altogether the temperatures remain-within reasonable limits for al l
angles of attack and vel oci t i es occurring in pr act i ce. At greater heights even lower temperatures
may be expected; at lower heights a t ransi t i on occurs to temperatures calculable by ki net i c
theory.
70
C-8SM6
4
1500
Plight velocity v in m/s
1000 2000 3000 4000 5000 6000 7000 6000 9000 10000
4 8 10 12 14 16 18 20 22 24 26 28 30
Mach number v/a
, A?: Surface temperature of a fl at plate in an atniosphere of 36 N
2
and V& O2
and with densitjr/' s 10~
7
kg setr/afr ( In molecular flow region) for various
flight vel oci t i es v/a and various angles of attack t
71
0-8*296
too
FUght velocity ymfflfS
1000 ZQOO 3000 4000 5000 6000 7000 8000 9000 40000
40 12 It 16 18 20 22 2 Z6 28 30
Uach nuirfber V/O
Fig. 4?: Coefficients iph/q of a shock impulse perpendicular to a fl at plate
for various fl i ght velocities v and angles of attack O-in an at -
mosphere of 86% N
2
and 14$ 0
2
and density f - 10
-
? kg secvuA
(molecular fLcm region).
ARft
W 2000 3000 mo 5Z8W 6000
Flight velocity
7000 8000 9000
vmm/s
10000
%ffO
0,95
Sk A/ 1
I
'i I
i I
i
I I
i V %ffO
0,95
Sk A/ 1
^485
V
V #1 an
^485
V
V #1 an
H 0,80
00
1 (27<?
I*
5
H 0,80
00
1 (27<?
I*
5
H 0,80
00
1 (27<?
I*
5
H 0,80
00
1 (27<?
I*
5
H 0,80
00
1 (27<?
I*
5
0,55
0,50
0
t
hS
+2f\
0,50
0
t
hS
^J5
**\tt\
3w
flflf
TOT
#Z5
TOT
0,20
TOT
0,15
TOT
0,10
0,05
^^r^VKnVv^^ 0,10
0,05
0,00
? 4 > 6 8 1l 7 / 2 * fr 1 $ / 8 2 0 2 2 ; * <
H i >8 3
Mach number l^(7
&i: Coefficient ipR/q of rest i t ut i on (rebounding impulse) on a fl at plate
for various flight velocities v and angles of attack OCin an at -
mosphere of 86S? N and U& 0g and density f- 10"
7
(molecular flow region),
73
G-SU296
iOO
0.95
t- e\9o
c
0
- qso
0,70
1000 2000 3O00 kooo 5000 6000
Flight velocity VinmjS
7000 8000 9000 10O0O
iOO
0.95
t- e\9o
c
0
- qso
0,70
I
I
! i
'"1
l "T
iOO
0.95
t- e\9o
c
0
- qso
0,70
|
iOO
0.95
t- e\9o
c
0
- qso
0,70
iOO
0.95
t- e\9o
c
0
- qso
0,70
+21\
i
iOO
0.95
t- e\9o
c
0
- qso
0,70 4
iOO
0.95
t- e\9o
c
0
- qso
0,70
n
V
iOO
0.95
t- e\9o
c
0
- qso
0,70
ml
\ tlf
fl/CC
VfOff
Q.60
*y
t
\|\
VfOff
Q.60
+7
*i l
+17
0,5$
Q50
0,45
A 9 C
0,5$
Q50
0,45
A 9 C
o'W
-*13
0,5$
Q50
0,45
A 9 C
0,5$
Q50
0,45
A 9 C
-3i\l\\
0,35
+9
0.30
Q2S
"'
J
Sl\\
+?:
0.30
Q2S
-7?'M\
0,20
*5*
fkA
U/lD
V +3
0,10
^ s
Q,Q&
O
a
0,00
9 i
t- i S i J 1 O i '2 1 * 1 6 1 8 V k V 2 '* i V 2 '8 30
Mach number V/Q
Fig. 45: Coefficient7/fy of the thrust tension bstwesn ai r and plane plates for
various flight velocities v and various angles of attack (. in an at -
mosphere of 86$ N- and U# 0, and density t - 10~" kg secVm^
(molecular flow region).
With the am ot the equations the aerodynamic forces were calculated for an assumed at -
mosphere with temperature 320 K. d e n s i t y X ? * ^ ^ , ^ and composition 14% O2 + 86X ftfe. The
momentum t
fi
ff of the ai r impinging on the pl at e i s pl ot t ed in Fig. 43 for various angles of
attack andvel oci ' t i es of f l i ght ; the corresponding recoi l momentum t > ^
i s
shown in Fig. 44, and
the tangential momentum in Fig. 45. Since *>=j r , Fig. 45 also represents the coeffi ci ent of
fri ct i on t/a between the ai r and the wall, for al l angles of attack and for the same range of
fl i ght vel oci t i es. In t hi s figure, the ext raordi nari l y large coefficient of fri ct i on in the
domain of gas-ki net i c streaming is noteworthy. While values of " ^ = 0.001 to 0.003 are custom-
ary in aerodynamics, we have in t hi s case values of the fri ct i on 300 to 1000 times as l arge, and
these are not accompanied by corresponding increases in the values of the perpendicular components
of the aerodynamic forces. The reason for t hi s astonishingly high fri ct i on i s the fact t hat in
the domain of gas-ki net i c streaming the protective boundary layer at the surface of the moving
body no longer exi s t s , so t hat al l the molecules which have the chance to t ransfer perpendicular
momentum t o the wal l , at the same time t ransfer t hei r tangential momentum completely on the aver-
age, whereas t hi s l at t er momentum transfer i s limited to the few molecules inside the boundary
layer, in the case of aerodynamic streaming. This very unfavorable behavior of the aerodynamic
forces in the rare upper atmosphere would completely prevent fl i ght at these hei ght s, i f the
stagnation pressures and consequently the forces did not decrease rapidly and fi nal l y vanish.
Obviously the shaping of the ai r cr af t to f i t the streaming conditions loses some of i t s signif-
icance of these hei ght s. Nevertheless, the Newtonian character of the ai r pressures ( i . e . t hei r
quadratic dependence on fl i ght velocity and angle of at t ack), and the rapid disappearance of
forces on leeward surfaces, i s very apparent at higher vel oci t i es of f l i ght . Thus in the gas-
kinetic domain al l the requirements are ful fi l l ed for the use of streamlined bodies, whose under-
side i s curved as l i t t l e as possible in the di rect i on of streaming, and whose upper side con-
s i s t s , as far as possi bl e, of leeward or convex surfaces, while the thickness of the body i s of
no importance. Reciprocal streamline-numbers and polars were derived for the f l at pl at e, from
Figs. 43-45, and pl ot t ed in Figs. 46 and 47. These graphs of the behavior of the aerodynamic for-
ces on a f l at , i nf i ni t el y thin pl at e, have di r ect pract i cal importance, because any wing profi l e
of fi ni t e thickness, with a f l at under-side and upper-side to leeward, undergoes exact l y these
same aerodynamic forces at high fl i ght vel oci t i es, regardless of the shape of the upper side and
the thickness of the pr of i l e.
Figs. 48 and 49, which show the aerodyi amic-forces on the rocket bomber in the range of
gas-kinetic streaming, were also derived from Figs. 43-45. In the same way as for the gas-dynamic
cal cul at i ons, the act ual l y curved surface of the ai rcraft was s pl i t up into a number of small
fl at surfaces, and the ai r forces calculated for each of these component surfaces.
Finally Fig. 50 shows a plot of the dependence of the ai r forces on height and velocity
of fl i ght for the domains of gas-dynamic and gas-ki net i c streaming. This pl ot of */ against
height, for the fl at pl at e, for various Mach numbers and with the fixed angle of attack of 7
(which i s the optimum for the gas-kinetic case), shows that the ai r-force-coeffi ci ent s decrease
with height in t hi s region. This phenomenon i s caused by the decrease of temperature of the pl at e
with height, so t hat cet er i s paribus the recoi l velocity and recoil momentum also decrease. The
forces in the range of heights from 40-90 km (which cannot be calculated exactly) were i nt er -
polated approximately in the form of the dotted curve.
Fig. 51 shows the final resul t of t hi s whole section, i . e. the dependence of the rocket
bomber's glide number (or i t s reciprocal) on the velocity and al t i t ude of fl i ght . With t hi s we
close the i nvest i gat i ons on glide-number of "the rocket bomber.
75
0,55
3 0,*5
2
w
id
o 0,M
9 0,3 J
Flight velocity V If) ffl/S
1000 2000 3000 4000 $000 6000 1000 8000 9000 70000
0,30
0,25
0,20
0,15
0,10
0,05
000
1
f T
i
- ^
1
v
%
\zt
V
a
9-
m*
"
7
o
" * * .
A
V
N
9
S*"
O*
3 ' "
**~
5 5
2^1 2^1
2 6 8 10 12 n 76 7fl 20 22 24 26 28 30
Uach number fyi?
46: Lift to drag rat i o of a fl at i nfi ni t el y t l dn plate in moleailar
flow region.
25^
F
7=P
/&'
F
7=P


J*
-
rty
Q-3X
J*
-
/w
fe
PJ*
fr

t7
T
j %
HrtQ
r
9>
13*4,
t-"''' >

/f
^r
^T
^
4
v q
2 a 3 a * < \5 q 6 q 7 q 8 0,
? if ? f ? * 2 1. 3 1fi it
Drag coefficient Cw
Fig, 47: Polars of a flat, infinitely thin plate in molecular flow region
for various flight speeds.
Flight velocity Vin m/S
1000 2000 3000 4000 5000 6000 7000 8000 9000 10000
0
f
dA " I
-
r
I
T
I 1
r
T ~
r
0
f
dA
<3 i
su
o
5
o n tilt
5
0
M
t/,03
**%
A VA
v
t
30
.'
/I 9C
CJZS
TjT
0,20 0,20
/ )*r
n
t/,/3
i ^ s
JI -ffl
, . /
\
V
v
t
lv
/) /)C
5-/
VfP3
/ "
0,00
, /
0,00
> < f < i y 7 ff 7 2 1
4 7 6 1 0 2 >0 2
>
2
2
* 2 6 2 8 3L
Mach number Zahl v/a
48: Lift to drag rat i o of the Rocket Bomber in molecular flow region.
Vfr4,
?
y
<&
?
/is
1
*w
30*
5
^
3*
?
w
2tf"
/
^
6*
VjW-ffl
-off"
I
5*
^ t r ft>*
A, ?'
f
" Oft QZ Q$ O
t
6 0,8 1,0 12 1fi 1,6 1,8 2,0 2
t
Z 2fr 2,6 2,8 3,0
Drag coefficient C^
Fig. 49
Polara of the Rocket Bomber in the molecular flow region for various
flight velocities.
.
. - '
r--*
.-
.-*
J
y
V
X
, ^ ^
^
y S
^ . - . .
. . . .
^
f-
^ ^ - ^
* *
^
f-
%

,
_ J
^ ^ ^
^^^^
M ^ ^
^^^
^ Q> <&
^ ^ ^
I I 1 a
&
51
% s* a s
at
8
I
I
I
I
I
1-
I
I
ma or vfiw *una
Fig. 50; Behavior of the coefficient p/q of the air pressure perpendi-
cular to a flat plate whan angle of attackd= 7 as a function
of flight velocity and flight altitude (conventional fluid
mechanics, molecular flow, and transition region).
200
t
H
V
ft
# 0
460
4W
420
400
80
60
HO
20
1
w
* * -

-
^ c
- - i - _ ___
* ^ * I
*^. ~*
^ >
\
\0fr" 5 V/ff- <5 Wfr*
oo
0 35 ^ ^5" 2,0 2J5 J
Ratio of l i f t to drag
Fig. 51;
Behavior of tha best lift to drag ratio of the Rocket Bomber plotted against
tha flight altitude and flight velocity (conventional fluid mechanics, mole-
cular flow, and transition region).
81
C-8 429 6
IK. LAUNCHING AND CLIMB
1. Acceleration of the Aircraft
The launching and climb of the rocket bomber have the purpose of giving i t , with a r-ininun
of fuel consumption, the high velocity necessary t o carry i t through i t s long gl i de path; they
are in the nature of an impulse which l ast s only for a few minutes. Despite t hi s, there are var i -
ous pos s i bi l i t i es for the variation, of t hi s external force during t hi s short time; we wish t o
find the most favorable, i , e-, the one with the l east possible fuel consumption for a given
final velocity.
In Fig. 52, the effect i ve exhaust speed c i s again taken as a measure of the useful energy
content per uni t mass of fuel and the attainable velocity increase V i s measured in uni t s of c;
the fuel consumption G - G
i s
measured in uni t s of the st ar t i ng weight G
or the weight of the accelerated mass GJ If one could completely transfer to the acceler-
at i ng body the energy avai l abl e from the combustion pases . - Gj^/cg, which i s t heoret i cal l y
possible in fi ri ng from a tube, then ?c = / ^ - / This rel at i on for ideal fi re i s shown
as curve 1 in Fig. 52. In the case of rocket drive, the acceleration process, i f i t was opposed
only by the i ner t i a! forces of the bomber and not by ai r resi st ance, gravity, et c. , would aft er
integration of the equations expressing the.Conservation
0
i[the Center of Mass (cdm+mctv=0 )
t
be represented by the fundamental rocket equation %=& ^ & - Curve 2 for ideal rocket accel-
eration thus gives much smaller final speeds for the same fuel consumption. The energy l ost t o
the acceleration process i s that of the motion of the combustion gases rel at i ve to the place of
launching; t hi s i s often called the external efficiency of the racket drive. This fundamental
curve for ideal rocket drive gives no indication of the actual accel erat i ons. Actually the
acceleration of the bomber occurs against ai r resi st ance and at l east a component of the weight
and these resi st ances become more important for the smaller accel erat i ons. For example, i f one
assumes t hat the ai r resi st ance and the component of weight along the path together represent a
fixed fraction, say 1/5, of the instaneous weight and that the acceleration i s constant during
the climb, then one can denote by k the r at i o of the rocket t hrust reduced by the r esi st i ng forces
to the effective t hr ust ; from the equation Ac*/m+mdv*=0 . <"
t ft
en derives a modified .
rocket equation %: = /f&l ^yfc If one takes the rocket t hrust to be equal to the instaneous
weight, then A"=ff
ae
Jf/ =
a s
and the curve for t hi s type of drive has the equation 5 = e J
8 c
or % = <3L8&IT- , This curve for K= 0,8 i s shown as number 3 in Fig. 52. In pract i ce one does
not obtain t hi s carve i f the ai r cr af t s t ar t s with customary small speed at the ground, because
for the assumed accel erat i ons the aerodynamic forces would increase much too quickly for subsonic
vel oci t i es. Carve 4 shows the climbing process for a commercial rocket plane with high requi re-
ments of safety and convenience, i . e . , small takeoff and landing speeds, and small, pr act i cal l y
constant accelerations along the path of climb; these are taken front a previous det ai l ed calcu-
l at i on of the path. (18) The long, tedious, uneconomical climb unt i l the velocity of sound i s
reached i s shown cl earl y; we see how costly low vel oci t i es at the ground and the accompanying
small accel erat i ons are for rocket f l i ght . Curve 5 shows a possible acceleration process under
the following assumptions; the rocket t hrust i s a. times the i ni t i a l weight G
p
and stays con-
st ant over the whole path of climb; thus the acceleration increases with the decreasing mass of
the rocket up to a Yalue determined by the weight of the crew and the ai r cr af t , and the rocket
motor i s used fully. Thus the weight l oss per second i s constant and equal to O-Qc'S/C
so t hat the weight at time i s : G=G
r
(/
a
&fy;) - By equating the i ner t i al and fri ct i on forces
(the l at t er with the same assumptions as for curve 3) to the rocket t hrust , i t follows that
Zb=Jir^ Ct2/?~^
r
Mt This calculation i s shown in Fig. 52 for three values of a. . P/G,, .
0.25, l.ff and 10. In the curve for P/G&* 0.25 the accel erat i on increases from 0.5 m/sec*. to
25 m/sec^. Because of the small acceleration the curve deviates noticeably from that for ideal
rocket dri ve. For the case P/G.= 1.0, in which t hrust and i ni t i a l weight are equal, the accel er-
at i ons vary between 10 m/see^ and 100 m/see^, and the curve approaches most closely t hat for
ideal rocket drive. Fi nal l y for P/Go = 10, the t hrust i s ten times the i ni t i a l weight, the accel -
erat i ons l i e between 100 m/sec
2
and 1000 in/see
2
, which i s cert ai nl y way beyond humanly bearable
l i mi t s. Thus the very great effect of the acceleration process on the final velocity at t ai nabl e
with a given G/(v i s cl ear,
The curves for P/Go - 1 and 10 are based on such reasonable assumptions t hat noticeably
more favorable types of launching could hardly be found; thus more det ai l ed study of the pro-
cesses of takeoff and climb under these assumptions i s advisable. The two curves di ffer only in
the accelerations during the climb. The usable accel erat i ons are limited only by the Strength of
the ai rcraft and by human r esi st ance. From our present knowledge of the physiological behavior
of the man under high accel erat i ons one must admit t hat an unmanned rocket ai r cr af t can be driven
at somewhat higher accel erat i ons than a manned craft ; nevert hel ess, the unmanned craft al so soon
reaches the point where great er accelerations are compensated for by the greater construction
weight necessary for craft capable of undergoing large accel erat i ons. In addition the r
Q
cket
4 **
u
v/c
1*
Q)fcfea/er Raftefenantrieb
ftakeienantriebmif massfoer
(D Beschfcuniguny gegenmassige
Bahnwefer&t&ntfe
(^Kakttenantrkb ernes langsam
startcnalen Verkchrs- IRaHetenffugzeugei
OfiaJfeJenantrieb e/Mi schnell
startenden Sate'en-Bombe/ifiugzeuges
ait verschiedwefl ertid//m'ssert von
StJiub zu Anfangs-fluggewkht, and
war P/%-%25
p 0,9 qe o,7 Qfi qs oj o
t
s o
t
z o/i o
t
o
weight of Wi be r / i hi t i dl weight Q/Go
?ig 52; Reference curves for the Rocket take-off.
(1) Ideal t r aj ect or y "shot"
(2) Ideal rocket take-off
(3) Rocket take-off with moderate accel erat i on agai nst
moderate path r esi st ance
(4) Rocket take-off of a slowly s t ar t i ng commercial rocket plane
(5) Rocket take-off of a rapi d s t ar t i ng rocket bomber with
various r at i os of t hr ust t o i n i t i a l weight
m
s
2
D
(J
O
4
4 A
i
i
1
1
1

28
26
i

28
26
2* \
22
| V
^ 0 i ^ V
lying
on back

i
lying
on back
i
\
12
10
\
^
sitting crouched
i
12
10
\
V
" -
-
1 1 1
sit*-* -
1'
o
\
V
" -
-
1 1 1
sit*-* -
t~
n
i
i
0
6
9
I
0
t~
i i _
0
6
9
I
0
0
6
9
I
0
.,
i . .
1
~T *
i
0
6
9
I
0
i i
! r
! I
:
9 5 12 16 20 2* 28 32 36 90 99 9B 52 56 60
Duration in seconds
Fig. 53;
Limits of accelerations withstood by trained pi l ot s i n terms of
"G's" as a function of time of duration
84
C-B4296
bomber, because of i t s long range and the accurate navigation necessary for the bomb rel ease.
cannot do without a pi l ot on board. Thus the *rmissible accelerations are limited t o value* t hat
can be withstood by t rai ned fl yers.
In Fig. 53 are shown the r esul t s of roost recent st udi es (2, 12, IS, 31)on the highest
' accel erat i ons t hat are bearable by a man in different body posi t i ons, as a function of the length
of the acceleration time. Vlher*as in the s i t t i ng position the l i mi t i s determined by diaturbanc
af the ci rcul at ory system especi al l y caused by loss of blood from the brain or heart due to
differences in hydrost at i c level i nsi de the ci rcul at ory system, t hi s danger decreases i n the
lying posi t i on, and the l i mi t then seems to be set by di f f i cul t i es in breathing as a r esul t of
greatly increased weight of the thorax. This more favorable lying position i s di r ect l y achieved
in the seating arrangement sketched in Fig. 3i , since the accelerations due t o the engine are
taken up by the pi l ot in directions perpendicular to the axis fit hi s body. In the lying position
in caroussel t est s accelerations of 20 g for mure than a minute Have been withstood by dragged
apes, and accelerations of 17 g for more than 1B0 s . by men.
During the takeoff and climb of a rockrt bomber there are two phases during which the
accelerations can reach c r i t i c a l values; the catapult takeoff and the end of the climbing path.
lha catapult process at takeoff l ast s several seconds; the accel erat i ons can be chosen freely,
the acceleration s t ar t s rat her suddenly, keeps i t s top value for about 11 sec. and then stops
again suddenly. Adjustment of the ci rcul at ory aysten i s thus scarcely possi bl e. Because of t hi s
impulse and rat her long action, one cannot go beyond an acceleration at takeoff of more than 5 g,
even for the favorable perpendicular posi t i on. Also higher takeoff accel erat i ons, though they
resul t in economizing on st art i ng-rocket s by decreasing the run on the ground, are unfavorable as
regards fl i ght performance, since they necessi t at e stronger construction of the tank assembly.
Conditions are quite di fferent at the "end of the climb. During the climb the accelerations
increase in the r at i o G /G as the resul t of decreasing mass of the ai r cr af t for constant motor
t hr ust . Even i f the ai r cr af t f l i es without payload, t hi s r at i o will end up at 10. This climbing
process l ast s several minutes, with the great est acceleration occuring in the l as t second. The
body has plenty of time to adjust i t s el f to the high accel erat i ons, and the highest accelerations
l ast for only a short t i ne. The conditions are thus very favorable, and the effect of the high
acceleration on fl i ght performance i s also favorsMe. One can, therefore, go cl oser to the l i mi t s
of resistance and, in view of the perpendicular position of the pi l ot , permit accel erat i ons of
10 g.
For the assumed G/Q_ = 1, t hi s means t hat the factor a equals 1, i . e . , the motor t hrust
equals the i ni t i a l weight. The limiting acceleration of 10 g i s of course not reached i f the
bomber varri es cargo. E. g. , i f i t has only a tons of bombs on board, then G/ ^ 0.-15 and.
the limiting acceleration i s 6.67 g. For P = G , the acceleration at the s t ar t of the climb i s
1 g, and increases in say t = e/gX CI . G/Go J = 340 or 360 sec. to 6.67 g or 10 g respectively
i f we use the rel at i on b = g/ (l -gt , '
c
) which neglects fractional forces. The l as t increase from
6.67 g to 10 g thus occurs in the very short time of 20 seconds. Only during these l ast 20 sec.
i s the pi l ot subjected t o the cr i t i cal accelerations, after he has become accustomed to the high
acceleration gradually and uniformly during the previous 340 s e c ; furthermore, t hi s holds only
for the pract i cal l y unimportant case where he takes off without bombs, with t hi s the assumptions
concerning permissible accel erat i ons during climb seem j ust i f i ed.
2. Catapult Takeoff
The foaket bomber i s to be brought t o a takeoff speed of 500 m/sec by means of equipment on
tfte ground, in order to increase as much as possible the final velocity at t ai nabl e with the fuel
on board, to minimize the di f f i cul t i es associated with the i ni t i a l wing loading of 800 kg/nr, and
to eliminate the uncertain fl i ght charact eri st i cs of the airframe in passing the velocity of
sound. For t hi s purpose a cat apul t -l i ke, perfectly st rai ght , horizontal st art i ng-t rack several
kilometers in length is needed; on t hi s s i t s a sled, which carri es the ai r cr af t , and which i s
driven by rocket apparatus having a large t hrust but moderate exhaust speed. Fig. 54 shows a
schematic of the sling arrangement. The downward forces acting along the takeoff path are the
weight of the bomber with sled and st ar t i ng rockets, about 150 tons; in addition there are large
forces acting vert i cal l y upwards or in the direction of the path due to driving and retarding
actions respectively. These forces and the necessity for very preci se i ns t al l at i on of the sl i de-
r ai l s suggest the construction of the roadbed out of reinforced concrete, whose cross-section fpr
the single-track arrang*-tm>nt chnsen has the isosceles t ri angl e shape shown in Fig. 54; the base of
the triangle serves to fix the st ruct ure in the ground, while the sled carrying the ai r er af t rides
on the apex of the t ri angl e. The upper construction of the r a i l arrangement consists of the main
r a i l lying or top of the wall, which has t o take up the perpendicular forces and also transmits
the very large horizontal retarding forces, and of two l ead- r ai l s halfway up the wall, which
support only occasional gin al l forces, and which prevent rot at i ons of the whole system of ai r cr af t
plus sled around i t s longitudinal axi s. Lubricants, of the proper vi scosi t y are used on the main-
85
C-B2Be
s
/,
e
>
Fig. 54; Starting sled of the Rocket Bombers
86
c-
and lead- rails, so tnat the sliding faces of the sled can glide on them with l i t t l e friction.
The sled itself is mounted on the sliding faces at A and B and also has a pair of fittings to
present overturning. The aircraft to be catapulted si t s up front on the main body of the sled,
the sled's power plant is at the bacl(, behind the aircraft. The high takeoff thrusts are trans-
mitted to the aircraft through the main body of the sled. Very high thrusts of 610 tons-for 11
seconds and the utmost safety in operation are required of the starting rockets, while their
exhaust speed can be moderate since they do not affect the flight performance of the bomber. In
the use of 40 tons of fuel for starting, one "i l l not, for reasons of safety and cost, use the
selfacting fuels like powder or HIJC^I which are customary for moderate exhaust speeds, but
rather the fuels consisting of two materials, say Hquid On and gasoil-water emulsion, are pref-
erable. For the calculation of the catapult process, we shall take, as exhaust speed of the
ground-based starting rockets, C - 1500 w/ s. so that the fuel consumption per second is Pg/c =
WOO kg/sec. The weight of the whole apparatus, for constant starting thrust Fj '& S
s
=^u~^fpi
at any instant. The air resistance of the whole catapult system is estimated to be 75,000 kg for
Y = 1.5, and increases during the takeoff according to the equation i = 0 3 V*. The sliding
friction of the sled on the takeoff rail is completely negligible compared to the two forces
mentioned. Using the second law of motion we can write for the acceleration at any moment during
takeoff: ^
=
gfP-O.3^^-Pgt/
C
)
If we require that the acceleration at the end of the takeoff shall not exceed 50 m/sec
Z
, then
#e obtain for the constant starting thrtistn P value of 610,000 kg if the final weight of the
take-off mass is 105,000 kg. By integrating unce, we obtain a relation between time elapsed and
velocity attained t-S*/399a/f-(j&e} ***/ , , _ zs/^'^s
For V = 500 m/sec we obtain for the time of takeoffJthe law of consump-
tion of the takeoff rocket vields a second relation: ff
r
/OJo*^
s
-^-ft
fc
/
from which t, a 10.96 sec and G 148700 kg. The total path length can be obtained most simply
by numerical integration. Summarizing, these relations concerning the takeoff process give the
results that: the takeoff rockets develop a thrust. P - 610 tons for 11 sec, use up 43.7 tons
of fuel with C = 1500 m/sec if the system to be catapulted Weighs 105 tons, the length of the
tow path is 2750 m; and that the accelerations during towing increase from an i ni t i al value of
40.2 m/s.-c
2
to the permissible limit of 50 m/sac^ at the end of the takeoff; thus for approximate
estimates one may use an average acceleration b ='45 m/sec and calcinate all the other
quantities; e.g.
=
^ = &>
/ / M
^f ^^tf*
7SO
t.
Fig. 55 shows the dependence on length of path of the velocity V attained and the resid-
ual weight G; also given are some of the data concerning the sliding jaws of the takeoff-sled.
The two sliding jaws at A and B can be assumed to be of bronze or hard steel and have a square
jaw-surface of 0.25 x 0.25 m; they are a sort of self-adjusting bearing-piate in Mi'chell bearings,
and are movable about a hinged edge; the flat undersurface of the jaw is bent up at' the front end
to assure the flow of lubricant to the edge. From Gumbel - Everling ( 7, P. 150/151) one- obtains,
for a square sliding jaw of side length t , if the point of application of the force corresponds to
the safest thickness of lubricating layer ( 7L* = 0.8), some fundamental equations; e.g., the
coefficient of friction i s ^ = 2.37/^^&z*, where p is the average pressure P/t? on the jaw; the
minimum thickness of lubricating layer is ^ = O-Sot~/-p>$(> and the angle of t i l t i s=t f ! 3^Sb^.
From these, Jt^Zf/Ad i . e. , the minimum gap h
o
between the sliding jaw and the rail ' " '
surface, which make an angle oCwith each other, should be kept as small as possible to permit
large loads P to slide with l i t t l e fractional resistance. In the desired state of pure liquid
friction, h
o
is limited by the roughness of the sliding surfaces; for the long sliding surfaces of
the takepff track which is subject to rough handling tnd all kinds of weather, we may expect val-
ues of" SO~ -~/Om. Thus the minimum gap-height h to assure dynamical buoyancy of the jaws is
h = 10 ro, with ^/i. = 0.00316. The largest sectior.-loading ever experienced by the jaws is
/>= 75000/625 = 120 bg/cm^; during takeoff this drops to 4 kg/cm^. From the previous equations
we see that minimum thickness h
Q
(and the minimum^44 ) can always be maintained, if the viscosit,
*f of the lubricant, decreases in the same manner as V/p increases; i . e. , if ^ ^ is constant.
At the start of the motion of the sled, the sliding speeds are so small that there is no ' f
sufficiently large to keep the product constant. In this case, the flat surface of the jaw will
l i e flat on the surface of the rail and will ris>e to the angle on'y as the velocity increases.
Fig. 55 shows the residual part G-A of the weight G which rests on the jaws"at any moment; i t is
assumed that the buoyance A varies quadratically with V from A = 0 for V = 0 to A = 100 tons for
V = 500 m/sec. Thus P = "y*- and the required lubricant viscosity at any position on the track
the viscosity of the hot oil in the film for minimum thick-
ness h
b
. The rise in temperature A T of the oil film can be estimated by assuming that all the
work against friction goes to heat the oil. Because of the very brief duration of this heating
process, the heat conduction from the oil will be extremely small. From rt/yUVt^, vjfc
&
A7'
;ity; taking values of i t follows that the heating is independent of the velocit
87
C-8 (296
so t hat as P decreases i t drops from 60 t o about 2 G ar the end of the catapult process, i s
shown in Fig. 55. If we assume an external temperature of + 15 C, then at the end of the
tow path a lubricant ia required which has a viscosity of^?=0. 00002 at +17 C; e. g. , benzine,
water, petroleum; three meters from the beginning of the takeoff we need a lubricant wi t h^ = 0 . 1
at 75 C; t hi s requirement i s sat i sf i ed by pi t ch. In betweenthere i s a whole range of
possible l ubri cant s. In the f i r s t three meters of the sl i de, pure liquid lubrication wi l l not
be possible; one wi l l have to use graphi t e-pi t ch mixtures. The equations given are valid only
for moderate sl i di ng speeds; act ual l y the velocity of motion reaches a final value of 500 m/aec
which i s 1/3 the velocity of sound i n the lubricant; because of t hi s we may expect a great in-
crease in the coefficient of f r i ct i on, but no physically different fri ct i on phenomena, especi al l y
since the high vel oci t i es are accompanied by small pressure on the sl i di ng jaws. The proper l u-
bri cat i on of the 3 km long surface should be maintained and protected by a special l u b r i c a n t
carriage which runs the f ul l length of the track, ilong the r a i l s . The brake arrangement at
point B can be modelled on t hat of a rai l road; i . e . , cast -i ron jaws are pressed onto the braking
surfaces of the r a i l ; aft er the ai r cr af t r i ses the forces are released and the-empty sled i s
brought back t o r est as quickly as possi bl e.
To t est the basic f easi bi l i t y of sl i di ng at high speeds, IL&chmid has made t est s on the
sl i di ng of pr oj ect i l es along curved metal walls lubricated with vaseline. The experimental
arrangement consisted of a drawn-steel tube of 10 mm inner diameter, which was s l i t along i t s
axis to produce a U-shaped groove. This channel was bent i nt o a closed ci r cl e 8 n. in radius,
so t hat the surface of the channel was toward the outside; then the proj ect i l e sl i di ng in the
channel i s kept there by centrifugal force. Through a branch tube which was tangent to t hi s
ci r cl e, an sS proj ect i l e was fired from a German 98 K, and into the ci rcul ar path. The bul l et
ran through the ci rcul ar path several times t i l l i t was brought t o r est ; after the t es t i t was
found lying completely unharmed in the channel (Fi gs. 56-, 57, 58)- The thin copper pl at i ng
of the steel was, in most t es t s , rubbed off at the posi t i on of the maximum diameter; the st eel
surface showed i sol at ed scratches apparently caused by fi l i ngs and burrs along the sl i de-pat h.
Immediately aft er finding i t , the bul l et was s t i l l warm to the touch. Also i t was compressed in
cross-section; t hi s i s probably caused by the fact that the spinning of the bul l et stops aft er
a short time and then the centrifugal force, which i s i ni t i a l l y 100 kg, causes the lead core t o
flow. After cessation of rot at i on and sl i ght deformation, the i ni t i al point contact with the
wall i s increased to a surface of about 1 enr, so that the 100 kg/cur pressure between the sur -
faces, which i s caused by the centrifugal force, corresponds approximately to t hat of the
sl i di ng jaws of the takeoff sled. Since the aS -bul l et is made of lead with a thin soft st eel
cover, and in the course of i t s 150 m. sl i di ng path goes through al l vel oci t i es from 800 m/sec
down to rero without being harmed, t hi s experiment may be considered as a proof of the f easi -
bi l i t y of construction of sl i di ng jaws for vel oci t i es of 500 m/sec; provided the s l i de- r ai l s are
properly constructed and lubricated.
If the accel erat i ons of the ai r cr af t during i t s launching are t o remain within permiss-
ible limits for both ai r cr af t and pi l ot , a launching path of about 3 km. i s required, so t hat
i t i s not possible to use a movable construction which would point the craft toward i t s t arget
at takeoff, (toe will therefore i ns t al l a takeoff track, usable in both di rect i ons, so t hat i t
points in tbe most probable di rect i ons for attack, say east-to-west; the more precise correc-
t i on of the di rect i on of fl i ght can be l eft to the pi l ot during the very f i r s t part of the
climb path, where the velocity and acceleration are s t i l l suffi ci ent l y small to permit such
changes of path.
If the bomber r i ses off the takeoff-sled at 500 m/sec, then i t s l i f t coefficient is c
a
*
0.05, whereas the best glide-number occurs at much larger angles of attack and a l i f t coeffi-
ci ent c
a
= 0.173. If one l i mi t s the normal acceleration during the f i r s t driveless part of the
climb-path to 20 m/aec
2
, then the i ni t i al radius of the turning-path, which goes from the hor i -
Eontal takeoff di rect i on t o the di rect i on of climb at $< = 30, i s 12500 m. This curved path
of turn, which begins at the end of the takeoff i s 5540 m. long and ends at an al t i t ude of
1700 m., with the required path i ncl i nat i on of 30 As a r esul t of ai r resistance and gain in
al t i t ude, the velocity at t hi s point has dropped to 370 m/sec, the l i f t coefficient i s c
a
=
0<093 - s t i l l much smaller than the one for optimum glide-number. The bomber then climbs fur-
ther at an angle of 30, at the expense of i d flying speed unt i l , aft er a l i near climb of
4000 m. to a height of 3675 m. and a vel oci t y V * 284 m/sec, the angle of attack for optimum
glide-number i s reached; at t hi s moment the rocket drive for the actual climb can begin. During"
t hi s 25-second-long, undriven motion aft er takeoff, the path of tbe ai r cr af t can be turned in the
di rect i on of the t arget . At the same time, t hi s procedure helps increase the range of the
bomber over i t s value i f one took off at V 250 m/sec with optimum glide-number and st art ed the
engine immediately' after takeoff, because we have gained a sl i ght excess velocity and a few
kilometers for the climb.
88
'i
C-8*
tOO 1*0 OpCOZ Mr
WD g BO * flOOSO* SO
I I t ;*
.5700 q f40^tqooz<^
K * , no - * qoa3M
t P cs S
o00 . 0 g 3D*-3
1 1 * *
M
sop ffojqoanjg
3 o no e qooac^Vo
3 S
MOH #O-C0O0^M
10 _
r
*a
JO*
9
qODttx**
e
JBD O 70 p^qOO* 21
1 J?
JDO,5 *oopoa 24
n
o 0
-SO 00 nOflOfO 20
JOB -o qoocv
o
130 *~* 30 qOOM fl
100 o ao opoo*
JO 10 00002 4
0 opooo 01
1
UM
6
JOO 400 900 800 WOO POO * > 4800 OQ JOQO JMO JfOO Jft MOO
Starting path In meters
O, p-t
O T 4
o o
H - D C
*
3
o
d
a
c
e
CQ
55j Gliding velocity v, moved load G loading on the runners G - A, heating of the
lubricating film AT, viscosity^ of the heated lubricating agent, and practical
manner of lubricating for various points along the starting path.
89
C-84296
Abb. 56; Tersuobanlage far Gleitrelbungsversuohe bel sehr
hohen (Heltgeschvlndigkeiten, beetehend aua elnen
MUltarkarablner 98E und elner nit S n Radius ge-
krumnten, gesohmierten und geschloflBen-kreisfor-
migen Gleltbahn for dae eS-Qeaehoaa.
90
c-i
3. Climb Path
The climb path of the rocket bomber i s determined by the forces act i ng on i t s center
mass. I f one momentarily neglects the rotation of the eart h, these are;
The weight of the a i r e raft-magnitude = {%. ~ ^*Aj/fy(tf*-/(/
di rect i on toward the center of the earth.
The aerodynamic lift-magnitude A = Cc ryV/3 , di rect i on perpendicular to
the tangent
1
to the path and in a plane through the center of the earth;
The ai r resistance-magnitude V^SA in the di rect i on of the tangent to the path;
The t hrust of the motor-magnitude P = 100 tons, t i l t ed from the tangent to the path by
angle of attack, lying in the plane through the center of the earth and the tangent t o the path-
The d'Alembertian i ner t i al force T, equal and opposite to the resul t ant of the other
forces, having a tangential component m-g and a normal component m Va%o .
These five forces al l l i e in a plane, so t hat the path i s also in a plane.
The rot at i on of the eart h makes the si t uat i on more complicated. If the ear t h' s atmosphere
were fixed in space, so that i t did not rot at e along with the eart h, then the path rel at i ve to the
ear t h' s surfaee could be calculated as i f a si xt h force, the Coriolis force, having ar bi t r ar y
direction and magnitude <?=2.u*'4tJ(Gy -?$) acted on the center of mass of the ai r cr af t .
Because of i t s arbi t rary direction in Space, t hi s force makes the orbi t , rel at i ve t o an observer
on the earth, appear to be no longer in a plane. The atmosphere, which actually r ot at es with the
eart h, produces dragging forces on the bomber which cause i t s absolute path al so t o be twisted,
so t hat the bomber i s pulled l i ke a weather vane by the apparent wind developed by the rot at i on
of the atmosphere, and will tend to turn except when i t i s flying in the west-east di rect i on, as
can be seen by considering a fl i ght st ar t i ng from the pole. This i nt erferi ng weathervane action
will have to be countered by the pi l ot ' s steering; i . e. , by a seventh force perpendicular to the
tangent, i f the absolute path prescribed by cel est i al navigation i s to be maintained. In the cal-
culation, i t i s assumed t hat the transverse aerodynamic forces excited by the elimination of the
weather-vane action do not noticeably al t er the glide-number. Since these transverse forces are
always perpendicular to the plane of the orbi t and only prevent a shi ft of the ai r cr af t out of
the plane, they need no longer be considered in the calculation of the or bi t . T
0
i l l us t r a t e the
the rel at i on between the forces act i ng on the center of mass, Fig. 59 shows two views of the ai r -
craft during the climb; the l i ne of si ght i s along the horizon at the level of the ai r cr af t . I f
al l the dependences were available in analytic form, one could introduce sui t abl e space coordi-
nat es, resolve the forces along these di rect i ons, apply the dynamical equation to each direction
and thus obtain three di f f er ent i al equations for the three coordinates of the climb path; the
integration would almost cert ai nl y be impossible, since the far simpler equations of ordinary
ext eri or bal l i s t i cs are not. The determination of the rel at i ve orbi t i s much simpler; one proceeda
in two steps; f i r s t the absolute path i s ealeulatea neglecting the Cori ol i s farce ( i . e. , using the
five forces previously l i s t ed) . Then one calculates separately th'e turning of the ear t h' s surface
below the ai r cr af t . Fi nal l y, the two components are combined on the sphere to give the twisted
rel at i ve path in space. The plane absolute orbi t i s determined by a step-by-step method which i s
familiar in Bal l i s t i cs . { 1, Vol. 1, p. 2Q7) As in Fig. 60, the continuously curved path i s
broken up i nt o a polygon, whose sides are suffi ci ent l y short that they can replace the arc of
which they are the chord. All the forces acting on the ai r cr af t are combined at the vert i ces, A,
B, C, et c. , and along with the i ncl i nat i on f* of-t he orbi t are assumed constant along
As ~#3 &C,*&- By set t i ng the resul t ant force in the tangential di rect i on equal to zero, we
***** 4 = S,f
cos
*< -

4
4
-
4
**J*fa
So t hat the velocity increases in f i r s t approximation from 1% & * / ~y* f-2^d$
in die time A = I/& - Mf/i, ^ l et t i ng the resul t ant normal force equal to Hera we
abtminjQj/ fy VOJSG
A
cosf%-/'s//?<K-'t#J&f The absolute vel oci t y Va
A
at the point A
can be-determined with suffi ci ent accuracy from the vel oci t y of fl i ght V^ and the angular velocity
of the point of takeoff. Thus the arc AHcan be drawn, and B, i s determined in f i r s t approxi-
mation. Now that the average value of the i ncl i nat i on and the average force between A and B, are
known, one can locate B in second approximation in the same manner. One repeats the procedure at
B, often enough so t hat the orbi t i s given in analytical or graphical form. The cal cul at i ons of
the orbi t by t hi s method were done numerically by A Moyczechowaki; the i ni t i a l conditions were
height 3675 m. above sea level, rel at i ve velocity of fl i ght 284 m/sec, and i ncl i nat i on of orbit
'i0, and the path was studied out t o G/r-
a
0-1- The orbi t was al so calculated for C_ 3000,4000
and 5000 m/sec and finally neglecting the earth' s rotation, and then with maximum effect of the
93
oa26
&.-
2
t ax* '
<
8
e u r l ^
Horizontal
Fig, 59; External forces on the Rockat Bonbar in cllabing.
?
o
J3 I
a. i
H '
1
A
&s
^ T ^ - -
.""*
5
3.
3
u
Q
s
I
I
s
faO
H
95
C-8ft6
H
O
.-1
(P
O
Is
43
f . .
o x
f i *
r-i
ft.
*
m
6
V
A3
a
>
*-

H
H
C
M
B
o
m
b

l
o
a
d

i
n

t
o
n
s

A
l
t
i
t
u
d
e

i
n

K
m

H
o
r
i
z
o
n
t
a
l

p
r
o
-
j
e
c
t
i
o
n

o
f

t
h
e

c
l
i
m
b

p
a
t
h

i
n

K
m

A
n
g
l
e

i
n

d
e
g
r
e
e
s

b
e
t
w
e
e
n

h
o
r
i
z
o
n

a
n
d

t
h
e

t
e
r
m
i
n
a
l

t
a
n
g
e
n
t

S
s
si
3s
g
in
E o-a
0 0
1*
-dp
m
P. a
As
1
fl?0O o,oo
50,0
37,0 55 +15,5 12,8 -6,8 12*<
1
2000 0,42 31,8 46,5 135 + 0,4 ZZ.0 -5.7 178
1
30W 0,30 20,0
41,5
230 -4,1
30/*
-4,8 214
1 4000 0,22 11,5 36 320 - 3, 7 41J0 +3,7 240
1
5000 0,*5
W
3**p 405
+ -W
55jB +37,0 264
1
0OW 0,10 0,3 MO 485 + 7,3 WA +39,1 174
I
a
/0OO 0,69 58,7 33,5 60 + 8,2 12,1 -7,1 124
I
a
2000 0,53 *| 3 31 150 -8,1 19,1 -5,5 180
I
a
3000 0,40 30,5 44,5 280 + 1,6 18,3 + 26,8 2U3
I
a
MOO 0,30 20,0 m
425 +9,5 28,3 -2,4 283
I
a
5000 0,23
13,3 66,5 560 +8,1 37,7 - 4 , 0 313
I
a
6000 0,18 8,0 84 675 +7,4 47,5 -3,0 333
I
a
7000 0,14 3,8 98,2 753 +7,3 64,3 -0,8 350
I
a
8000 0,11 1,0 109 860 +7,5 81,6 + 1,6 302
i
1
1000 0,75 65,0 31 60
*3,1
41,5 -7.3 128
i
1
2000 0,62 $1,7 18,5 155 -10,6 16,3 *5,0 196
i
1
3000 0/f8 37,5 43,5
335 +11,9 16,7
-w
267
i
1
WO 0J8 28,1 71
505 +7,3
22fi -6,3 315
i
1
5000 0,30 20,1 87,5 665 +4,4 28,8 -5,8 351
i
1
6000 0,25 15,0 98,5 830 +2,9 3* -5.1 380
i
1
7000 0,20 10,5 108 975 +2,6 4lfi -4,1 404
i
1
8000 0,17 6,5 112,3 1135 +1$ W,S
-2,0
424
i
1
9000 0,1k 45
117 1275 +2,0 60,3 '1,3 HO
i
1
10000 0,11 | 1,0
W,5 1400 +2,3 7^3 0,4 452
Numerical charact eri s t i cs f or three iraricus climbing paths of the
Rocket Bomber.
96
J

^
*
1 ^
I*
Si 0
25
a
fc #

8
f l
u
Q
-8
S-

1
a
& "1
u
Q
-8
S-

1
p
n
5
n
i
% E
n
5
n
i
fi B B
3 5 5
1
-
.
a
r
t
i
n
g

f
r
o
m

a

p
o
l
e

o
f

.
i
n
g

t
o

t
h
e

W
e
s
t

f
r
o
m

,
i
n
g

t
o

t
h
e

E
a
s
t

f
r
o
m

&
X
-
.
a
r
t
i
n
g

f
r
o
m

a

p
o
l
e

o
f

.
i
n
g

t
o

t
h
e

W
e
s
t

f
r
o
m

,
i
n
g

t
o

t
h
e

E
a
s
t

f
r
o
m

^
-
a
r
t
i
n
g

f
r
o
m

a

p
o
l
e

o
f

.
i
n
g

t
o

t
h
e

W
e
s
t

f
r
o
m

,
i
n
g

t
o

t
h
e

E
a
s
t

f
r
o
m


a
r
t
i
n
g

f
r
o
m

a

p
o
l
e

o
f

.
i
n
g

t
o

t
h
e

W
e
s
t

f
r
o
m

,
i
n
g

t
o

t
h
e

E
a
s
t

f
r
o
m

\m M
1
1
-

S
t
a
r
)

-

S
t
a
r
l

1
1
L
1
1
1 1
r - i
1 ; i I
5 5 J 1 1
1 s
i
& 5 5
I I
\ 5
5 ?
i
WS ir[ H qniTT JO iniPTH Pig. 61;
Absolute climbing path of the Rocket Bombsr with c = 3000 m/sac, without
consideration of the Earth's rotation, but with consideration of rotation
velocity of a point on Earth's equator, launch to East and launch to West.

I
| S i % f c $ S s S
62: Absolute clijnbing path of the Bockat Bomber with c = 4000 m/aec, with-
out consideration of the Earth's rotation, but with consideration of
rotation velocity, of a point on Earth's equator, launch to East and
launch to lfeat.
1 * *
na uj H qnr^TO jo VHpTH
Pig. 63s Absolute climbing path of the Rocket Bomber with c = 5000 m/eec, with-
out consideration of the Earth's rotation, but with consideration of
rotation velocity, of a point on Earth's equator, launch to East and
launch to West.
99
C-84296
1 . . , ,
adSfut Ut ^po[wA V&HTf ^xqpaq-B -[Birpoasi
6V: The absolute flight velocity of the Rocket Bomber attained during climb
with e = 3000, 4000, and 5000 m/aec as a function of G/G
0
, and without
consideration of Earth' s rot at i on.
rotation - i . e. , takeoff from the equator in an east erl y or westerly di rect i on. The following
figures show some examples of the r esul t s; Fig. 61-63 show the shape of the absolute orbi t , Fig.
64 shows the vel oci t i es reached. Table I I I gives a auntnarv of the moat important resul t s for al l
the or bi t s, neglecting the ear t h' s rot at i on.
A fai rl y accurate estimate of the final vel oci t y, which i s the most important datum for
the climb path, can be obtained .without doing the exact calculation of the orbi t , by using the
equation v/
c =
Jn
0/
/fc f(
/-
~ ^
0>
}This i s especi al l y so i f one f i t s the actual conditions by
set t i ng q ( r at i o of t ot al ai r resistance plus weight component along the path t o instantaneous
weight) equal t o 0.5 in the ranpe G,'
Go
= 1.0 to 0. 5, and q = 0 for G/r^ - 0.5 t o 0. 1. I f one
carries out t hi s rough calculation for the bomber with 100 ton takeoff-weight and 10 tons empt,
weight, for various exhaust speeds C, assuming i ni t i al velocity 284 m/sec, one finds t hat for C
- 4000 m/sec the bonier reaches the velocity of a long-range proj ect i l e even with a 50-ton bomb
load, so that i t car rnifpclr with lnng-range ar t i l l er y; without payload i t exceeds the vel oci t y
V = 7900 m'.sec. Between these to extremes l i e al l sort s of vel oci t i es and bomb loads. An
i nt erest i ng result which is easi l y obtained, i s t hat in the range of bond) loads sui t abl e for
mi l i t ary use, i . e. 5-4(1 tons, the exhaust speed has l i t t l e effect on the final speed. This means
t hat the development of a rocket bomber does not have t o wait for avai l abi l i t y of rocket motors
with C = 4000 m/sec; with C = 3000 m/sec one can already construct a very dangerous weapon.
Figs. 65 and 66 compare t hi s rough calculation to exact orbi t calculation and to the curve
for ideal rocket propulsion; the i ni t i al velocity r f t er takeoff i s included. One sees t hat the
rough estimate gives the actual behavior with suffi ci ent accuracy, especially for large values of
C; also the final vel oci t i es actually attained, especially with large bomb loads^ are far below
those for ideal rocket propulsion, because the weight component in the direction of propulsion i s
comparable to the thrust for the i ni t i al steep climb, and because of the rat her high ai r r esi st -
ance during the climb. The propulsion curve could thus be improved by lessening the resistance pr
increasing the propulsive forces. A lowering of the i ni t i a l nngle of climb $e would decrease the
harmful weight component, but would r esul t in great increase xn ai r resistance in regions of
dense ai r - we can expect no success from t hi s procedure. Another possi bi l i t y might be a great
increase in the t hrust during the i ni t i a l part of the path, so t hat , say, the tangential accelera-
tion i s constant in the range of the permissible l i mi t 10 g, and the resi st i ng forces are half
the instantaneous weight i ni t i a l l y and then drop to zero below WQ
0
= 0. 5. The rocket motor would
have to s t ar t with a t hrust of 1000 tons, and then be t hr ot t l ed gradually t o 100 tons at the end
of the climb. Taking \/=28U-+ 0.9Se.i
G
'/e for /. O > <%js. > O. S and V=S4-- aeSfet
c/f?
G
y&6 for O.S> ^ > , 0 . / we get the r esul t s shown in Fig. 67. *hile the 100-ton
rocket motor weighs 25O0 kg, one must assume a weight five times as great ( i . e. , 12500 kg) for a
well-regulated motor with 10 times greater maximum t hrust . If, despite the more powerful motor,
al l other weights of the bomber remain the same, then the empty weight would be 20000 kg. instead
of 10000 kg; i . e. , the most favorable G/Q would be 0. 2, and the bomb load would already have
dropped to zero at GQ/Q = 5. The propulsion curve shown in Fig. 67 for 10-ton empty weight would,
for the now necessary ' 20-tons, move considerably t o the ri ght ; despi t e the increased di f f i cul t i es
with the regulable 1000-ton motor one would achieve only a sl i ght improvement for short distances
of attack, while the 100-ton motor would s t i l l be superior for long range at t acks. Thus t he-sec-
ond method for improving the propulsion curve also f ai l s in pract i ce, and the curves of climb
shown in Figs. 65 and 66 represent the best solution of the problem.
The rocket bomber, at each moment of i t s f l i ght , ' t ends toward that part of the atmosphere
where the buoyant forces are equal to i t s weight. If t hi s equilibrium position i s such that the
al t i t ude stays constant for a few seconds, i t wjll he said to be st at i onary. Such a st at i onary
equilibrium occurs, for example, i f the aerodynamic driving force A of the bomber, increased by
the centrifugal force F, which for fl i ght at constant al t i t ude i s a consequence of the earth
1
1.
curvature, i s equal' to the instantaneous weigh"; G of the bomber.
During the climb the t hrust component Ps//r"C also act s as a buoyant force. For example,
st art i ng with a weight in fl i ght , Go 100 tons, at al t i t ude H = 35G0 m, and C__ = 0. 2, by
equating the weight to the propulsive force, we can obtain for the horizontal velocity a value
V f/lif Go/c^e i
1
^ = 300Sty&ec At any other al t i t ude, i f we set the weight equal t o the
buoyant force c
a
#
L
>

'fy
f
2g-*
<
3<sy^*Qvt* get a one-to-one rel at i on between al t i t ude and vel oci t y.
This relation i s shown in Fig. 68, for C - 4000 m/sec. tte see t hat these st at i onary al t i t udes are
low, below 60 km for vel oci t i es up to 7000 m/sec. If we include the vert i cal component,/Rr/>7
of the motor t hrust of 13920 kg, we get the upper curve. I t i s i nt erest i ng to note that for V =
4700 m/ sec, the curve becomes ver t i cal ; i . e . , no st at i onary equilibrium i s possible at f i ni t e
al t i t ude as soon as V = 4700 m/sec. In the f i r s t curve the st at i onary equilibrium i s no longer
at t ai nabl e above V=7900 m/sec. These assumed st at i onary st at es probably do not exi st in f l i ght .
In part i cul ar, the center of curvature of the path i s seldom at the center of the' eart h so that
the al t i t ude would remain constant - most of the time i t i s near the ear t h' s surface, and at
times i t i s even above the ai r cr af t ; also the radius of curvature i s usually much l ess than the
radius of the earth. As a resul t strong transverse dynamical forces are developed, which permit
" dynamical" equilibrium of the bomber at heights far above, the stationary ones.
101
C-B6
HQM
Bomb load in tone
75 70 65 60 55 50 *5 hO 35 30 25 20 iS iQ 5 0
'ffi 0,80 0,75
Fig, 65J
m QiO Q35 Q?Q Op 030 Of 5 QJ0
Remaining flight weight G/G
0
,
102
More exact determination of the relationship between bomb load, terminal
velocity of the Rocket Bomber attained for v - 3000 m/sec and v
0
- 284 m/aec
by means of the graphical determination of the course. For comparison are
plotted the approximate calculations and the carve for ideal Rocket accelera-
tion.
3
C-84
,4(000
Bomb load In tons
75 70 65 60 55 50 *t5 W 35 30 25 20 15 10 5 0
Q&5 0,80 0,75 0,70 0^5 0.60
055 CfiO ffi OfiQ 0,35 0,30 0,25 0,20 0^5 0^0
Remaining flight weight G/G_
Fig, 66:
More exact detennlnation of the relationship between bomb load, terminal
velocity or the Rocket Boriwr attained for T = iiOOO m/asc and v
0
= 2% m/sec
by means of the graphical detennlnation of the cnirse. For comparison are
plotted the approximate calculations and the curve for idea*! Rocket accelera-
tion.
to
O
I
43
id
3
!
si

4HW
9000
7000
kOOO
500
0
Ideal rocket acceleration
Acceleration of Rocket Bomber with
variable thmat up to 1000 tons max.,
in case 10 tons empty weight and
20 tons empty Height. -
Acceleration of the Rocket Bomber-
using a constant thrust of 100 tons
:
and an empty weight of 10 tons.
i i i r i i i
idsaler Ttaketenantrieb-
Antrkb des Bombers durch
rege/baren ftakefenmotor von
iQQOTonnen Grdtfrschub
bet iOTormai Leergewicftt
bei% 2QTonnenLeergewicht~
Anirieb des Bombers durch
HaHetenmotor voniOOTormen
konstanlem Schub and
lOTonnen LsergewichU
75 70 65 60 $5 50 45 W 35 30 25 20 15 40 5 O
Boob load In tons
Pig. 67:
Climb of a Rocket Bomber whose thrust is adjustable between 100 and
1000 tons, i.e, high almost constant acceleration with c = 4000 a/sec
exhaust Telocity.
i 104
0-8
The actual heights osci l l at e about the st at i onary ones, with very large amplitude, while the ad-
di t i onal dynamical forces tend to bring the ai r cr af t back to the posi t i on of st at i onary equi l i b-
rium; i . e . , they are restoring forces pointing toward the stationary position and increasing with
distance of the ai r cr af t from i t , l i ke spring-forces. The number of osci l l at i ons; for the ex-
ample shown in Fig. 68, i s not qui t e 2 for the whole climb; t hi s i s shown by the following con-
si derat i on.
At f i r s t , the st at i onary al t i t ude i s greatly exceeded. This r esul t s from the fact t hat
i ni t i al l y the air density decreases more slowly than the increase in velocity, so that the
ai r cr af t takes on a very steep position at angles = 45" or more. Already at fai rl y high ve-
l oci t i es, tli** ai r cr af t must now be turned through more than 45 from t hi s position to t hat of
horizontal fl i ght . Because of the small radius of curvature available, t hi s resul t s in powerful
centrifugal forces, which drive the ai r cr af t up above the stationary equilibrium heights of 40-60
km, t p heights over iQO km. One might now expect t hat the ai r cr af t would i[mediately drop down
again from these extreme al t i t udes, since the path i s approaching parallelism with the ear t h' s
surface, and the radius of curvature i s approaching the value of the ear t h' s radius. This drop
from the heights i ni t i al l y attained occurs only in sl i ght degree, i f at al l , because i n the mean-
time the velocity has increased very rapidly, and i s already near the value V= 4700 m/sec for
which the al t i t ude of st at i onary equilibrium of the bomber becomes i nf i ni t e. At t hi s speed, the
bomber i s in st at i onary equilibrium at every al t i t ude, so t hat i t does not have to f al l . If i t
exceeds V = 4700 m/aec, then i t suddenly finds i t s el f below the al t i t ude of stationary equi l i b-
rium, BO that i t has an excess of buoyant force, and begins to climb again so long as the
propulsion continues. This actual curve of climb i s obtained from the exact calculation of the
orbi t , and i s shown in Fig. 68. The whole climb may be considered as dynamic flight only in a
limited sense. For the most part i t follows an i ner t i al path, l i ke that of a proj ect i l e or a
heavenly body, at heights which are pr act i cal l y in empty space, since one can no longer speak of
an atmosphere in the sense of aerodynamics when the molecular free paths are greater than several
hundred meters. Fig. 69, obtained from exact orbi t cal cul at i ons, shows the actual dynamical
al t i t udes attained during the climb of the rocket bomber.
105
200
490
430
a
m
8
5 450
3 m
130
120
440
400
90
80
70
60
50
HO
30
20
10
0

J -
Equilibrium f l i ght a l t t *
? "tude from equilibrium of
l i f t , cent ri fugal force,
and Weight.
/
*q
j H * | W T
ullibrlum f l i ght a l t l -
de - considering l i f t ,
i ght , cent ri fugal force,
d the ver t i cal component
t hr us t .
tu
we
j H * | W T
ullibrlum f l i ght a l t l -
de - considering l i f t ,
i ght , cent ri fugal force,
d the ver t i cal component
t hr us t .
an
of
j H * | W T
ullibrlum f l i ght a l t l -
de - considering l i f t ,
i ght , cent ri fugal force,
d the ver t i cal component
t hr us t .
^ - Actual dynamic al t i t ude
11 W 21 100 31 W U W 51 100 6L 100 % TOO
m
Flight velocity in m/sec
Fig. 68i
Comparison of stationary (i.e. Instantaneous equilibrium) and actual,
dynamic flight altitude when the Rocket Bomber Is climbing with
c = 4000 m/sec.
SI&00GS-1
3
w u
o
. 43
n
4 Q 43
<s -So)
1-
+ 43
s
t
a
r

a

p
o

S
t
a
r

a
t

e

>
0

0
1
i
1
t
\
?
&>
1
\
*
A
rH
1
\
;
a
r
t

t
o

K
a
s
t

,

e
o
u
a
t
o
r
,

L
i

_
4
i
n
fc
\
;
a
r
t

t
o

K
a
s
t

,

e
o
u
a
t
o
r
,

* 1 4 3
i

_
4
i
HJ
_
B
e t-,
0
D 43

\
;
a
r
t

t
o

K
a
s
t

,

e
o
u
a
t
o
r
,

* 1 4 3
i

_
4
i
1
S>
Via
4 3 1
0 nj
1
^
\ \
A *S
\ \
I
k \
r~
4>
a
Hi
b
"6
<
k \ ^N b
o

E
J

a
t
o
r

'*
b
&
V
\ V
\ i 1
\
1
43 -*J
en a
c
5
f t
(0 o
|_ _ _ -
l
4 3 43
si
0

\
\
* *
X *
\
K K
"*1
g
V
* >
V
!
\
8
1
\ :
v
* 'T '
I f' /
t

^
3
V
i S
1
y
' N / & T
i
ff*
~s*
" ' J
\ \
1 i 1
i
J
j
v,
1

'
l f L
$
^
1
I
1
^ ^
5
1
1
_ J
1
U/if Ul pBuys^* epn^-nxB xwrp&iB-J
W. GLIDING FLIGHT AND LANDING
1. Supersonic Path of Gliding Flight
The supersonic path of gliding fl i ght of the rocket bomber i s determined by the forces
acting on i t s center of mass, in the same manner as for the climb path. These external forces are:
the weight of the ai r cr af t , G = rag
o
( r y
R +
u)
2
. the aerodynamic l i f t ^ = C^/yat^&
f
the
ai r resistance IA/= SA < and the d
1
Alerobertian i ner t i al force T. The same r esul t s apply
t o the rot at i on of the eart h and the atmosphere as in Sec. I l l 3; in par t i cul ar , the supersonic
gliding path i s to be considered as occurring in a plane, for an observer out in space; the
deviating effect due to atmospheric rot at i on, which i s especially effective in the lower layers
of the atmosphere, most be eliminated by steering. Oily in special cases will the Cori ol i s force
be eliminated by st eeri ng, so t hat the orbi t r el at i ve to the earth i s plane; t hi s wi l l occur i f
the bomber i s to go a l l around the eart h on one hop, and land at i t s st ar t i ng poi nt . To do t hi s
in a perfectly plane absolute orbi t i s possible only i f the takeoff field i s at the pole, or i f
i t i s at the equator and the plane of the orbi t coincides with the equatorial plane. Otherwise
the procedure of circumnavigation wi l l go as follows: unt i l the bomb rel ease an absolute plane
orbi t will be flown for purposes of preci se cel est i al navigation; only the weather-vane action
will be eliminated. After the bomb rel ease a sui t abl e r el at i vel y plane orbi t will be flown, i . e . ,
both weather-vane and Cori ol i s effect s will be eliminated by st eeri ng. The arrangement of forces
used in calculating the path are shown in Fig. 70 for gliding fl i ght in two aspects, both viewed
along the horizontal at the level of the ai r cr af t . The procedure of calculation corresponds
exactly to the stepwise method used in determining the climb path; f i r s t the absolute pr bj t i s
determined neglecting the Cori ol i s force and then the separately computed rotation of the eart h
i s combined with"the absolute orbi t t o give the desired r el at i ve or bi t . The i ni t i al conditions
for the decending path are given by the endpoint of the climb path. Since the climb path can be
broken off at any point, there i s a singly i nf i ni t e manifold of possible paths of descent for
each ascending path. Actually we st art ed from the previously calculated ascending paths with C
= 3, 4 and 5000 m/sec, and each of these was broken off at V
o
= 1,2,3 up t o 8000 m/see, provided
t hat al l these vel oci t i es were act ual l y reached. They represent the i ni t i al vel oci t i es for the
descending path. Each descending path i s followed unt i l the vel oci t y has dropped to 300 m/sec.
In Figs. 71 and 72 two paths are shown; one for C = 3000 m/sec, V = 4000 m/sec; the other for C
" 3000 m/sec, V =6000 m/sec at 6.36 times the al t i t ude. In the fi gures, the strong osci l l at i ons
of the paths, especi al l y in the f i r s t portion, are notable. Because of the considerable t i l t of
the or bi t resul t i ng from the climb, the bomber overshoots i t s st at i onary al t i t ude of fl i ght , then
approaches i t from above, passes through i t because of i ner t i a, then i s driven upward again by
the greatly increased aerodynamic' forces, unt i l after several such osci l l at i ons the amplitude.has
decreased so much t hat the ai rcraft levels off at the st at i onary al t i t ude and continues i t s fl i ght
i t that al t i t ude. This ricocheting generally has a favorable effect on the range of the bomber.
I t has the advantage t hat the thermal st resses of the external surfaces of the ai rcraft which
face the course wind vary in time at high vel oci t i es of f l i ght . The osci l l at i on of the path will
be hindered by st eeri ng, only i f fl i ght at the st at i onary al t i t ude i s needed for some special
reason, say aiming before the bomb release.
Figs. 73 and 74 show the main elements of the two orbi t s in great er det ai l . In Fig. 74,
one can easi l y see the following connections: the orbi t shown in Fig. 72 i s now represented by
pl ot t i ng the al t i t ude H agai nst the log of the distance S. One can easi l y locate the peaks and
valleys of the orbi t for both curves. The climb path ends after t = 300 sec. at an al t i t ude
H = il.'l km, with i ncl i nat i on * - 6.2. At t hi s moment the velocity V reaches 6000 m/sec. When
the rocket motor i s turned off the tangential acceleration drops from &=+ 7.5"%6*c.*
to -5.5 in/see^, so the ai r cr af t i s greatly retarded, considering the f l at ascending path. At the
same time the normal acceleration acting at the cockpit drops from S = -f-4~f- * / * ; aft er
the normal component of t he engine t hrust has ceased, to 4>
n
* >*-/c7 " ' / s e c * This posi t i ve
residue r esul t s from the instantaneous center of curvature being above the ai rcraft . In the
f i r s t part of i t s motorless supersonic gl i de path, the bomber climbs to H * 143.8 km, while the
inclination of the path goes through a maximum J**= i'
1
SWalso the tangential acceleration gradnally
increases t o 4 * "
a t
the crest of the wave, while the normal acceleration increases up t o
h= sotifaec*, so t hat over a long period of time objects in the ai r cr af t will appear to be
weightless. At t he same time the velocity drops t o a minimum of V = 5600 m/sec at the crest of
the wave. At the posi t i on, S = 1700 km, of the f i r s t crest in the orbi t , the curves for J# and
also pract i cal l y for & (because of the low ai r-densi t y) must go t o zero, while A
n
and V pass
through a minimum. Then follows the f i r s t fl i ght i nt o a valley, with a height loss of 108.8 km.
in 250 sec. During t hi s the velocity again r i ses to 6000 m/sec, the l argest i ncl i nat i on i s -8,
the l argest tangential acceleration i s & - +/'mj&ec* *
n t n
trough the height i s H = 35 km,
and the l argest normal acceleration acting at the cockpit i s ^> = -f sf3"*/scc
z
108
4
C-B2)6
Absolute path
Horizontal
tangent
Horizontal
Q
Surf**
0*
Fig. 70: External forces on the Rocket Boraber in the case of sup
?
01
e
3
O
o
ti
not UT qipf
B
M W*TU
Fig. 71: Flight path of the Rocket Bomber with c
and a bomb load of 11*5 tons.
= 3000 m/sac,
ofts or Mr-pT^iprU 9 aM/nTn: A^XIXOOTU. a n 7
* * 3L..5__5L ?_.*___-.-* ? ? ?
j JH ; re * n
* L J* * * * * * * * *
,^r^ ticx^vioxsobv 'f
s,lu0
*
u
5
H
113
c Vo
inm/sec
Bomb
load
In tons
Fl i gi t
path
length
in &s
Time of
f l i gi t
in
seconds
liaxinniffi Maximum pos-
f l i gt t itive normal
heijtot accelerator
in Km In m/sec.
f
1000 50,0 303 490 40 38,5
f
2000 31J8 1528 1300 46 21,7
f
3000 20p 3639 2180 45 11,4
f
4000 11,5
6692 2620 47 10,5
f
5000
w
12171 4330 76 34,9
f
6000 0,3 20371 5800 143 46,5
t
1000 58,7 295 530 34 28,7
t
2000 43,3 '1367 1160 37 15,2
t
3000 30,5 3477 1100 49 26,0
t
WOO 20,0 6950 3040 80 35,6
t
5000 13,3 12592 WOO 104 45,3
t
6000 8,0 21139 5320 160 48,8
t
7000 3,8 39363 8840 283 50,3
t
8000
1,0
91870 16015 1296 55,7
1
1000 65,0 291 455 30 22,0
1
2000 51,7 125$ 1120 31 19,2
1
3000 37,5 3847 2225 68 313
1
WOO 26,1 7454 3200 81 37,0
1
5000 20,1 12180 4290 102 46,5
1
0000 15,0 21531 5990 1H 35,9
1
1000 10,5 42091 9120 128 36,3
1
8000 6,5 293720 41600 I 775 2,5
Numerical characteristics of 22 various descent paths of the Rocket Boater,
. 1 l I 3 I I I I i $ i
* a ^ e & j s * * s B
B
1
orfoBp uj tKS^witauf
- ~ ? = 8 5
n
?
;t Deitftt in tot ni ght hl#il in v.-
vi -i te f c- *i . . t _/ j -
h
v- >' * -* _ _ .
Fig- 77* IalluftiK* of tlws i-rtsas* of a 3,3 ton boab fron * haigtit of W) km and t t
vfll^eity 31 6069 s^sss on Wi f l i ght path of th Rodent BombHr *lt!i c =
:, r
0
= 7000 mfiac.
1
a
*
nonul *acalsF>tlosk--<'
i i
axi al acceleration t,-V ,
Inelination in dora-wa,
fl i ght naight *-*-
a S S i 3 5 5 S 3 .S- B ^ 3
flight Telocity * In /ate t flight tine In c,
w
This procedure i s repeated at each jump which the rocket Dcmoc> ,Ke4 toward i t s t arget , while
the extreme values gradually die down and the fl i ght becomes more and more smooth i n the neigh-
borhood of the st at i onary al t i t ude. The t ot al length of the projection of the glide path on the
ear t h' s surface i s 20371 km, the duration of the fl i ght is 5830 sec. In t hi s manner various
orbits were calculated; Table II gives the most important data - length of the orbi t , glide.
number, maximum al t i t ude attained, and maximum posi t i ve normal acceleration.
All these or bi t i nvest i gat i ons were carried out neglecting the ear t h' s rot at i on. For
t hi s reason Figs. 75 and 76 show the rel at i ve and absolute orbi t s for supersonic fl i ght of the
rocket bomber for C = 3000 m/sec and with the velocity at the end of the climb path V = 6000
m/sec, i f the ai r cr af t takes off in an exact east erl y or westerly di rect i on from a point on the
quator. In the case of an eastward takeoff, the velocity of rot at i on of the takeoff point
rel at i ve to the ear t h' s center adds to the velocity of the ai rcraft r el at i ve to the takeoff
point, so the centrifugal force increases. Because of t hi s influence of the ear t h' s rot at i on,
the range S = 20370 km. in Fig. 72 i s increased to S = 23470 km, more than 13%. This supersonic
path rel at i ve to the ear t h' s surface is shown in Fig. 75. The most i nt erest i ng features are the
much greater heights of the orbi t peaks and the longer distance between peaks, compared to Fig.
72. The absolute path of the ai r cr af t , as seen by an observer out in space, i s shown dotted in
Fig. 75; nat ural l y the absolute path length i s much greater (S 26410 km.). In the case of
takeoff t o westward, the rotational velocity of the takeoff point i s subtracted from the velocity
of fl i ght , the centrifugal force decreases and the range drops from 20370 1cm to 18200 "km, i . e . ,
more than 10%. At the same time the heights of, and i nt erval s between, the f i r s t waves of the
orbi t decrease' These r esul t s are shown in Fig. 76; the absolute orbit i s shown dotted for
comparison. The effect of the ear t h' s rotation on the range and height of the supersonic orbi t
becomes even great er if the maximum velocity of fl i ght approaches the velocity due to the ear t h' s
rot at i on; on the other hand i t decreases rapidly i f the st ar t i s made 'from the equator i s an
intermediate, rather than in an easterly or westerly di rect i on, or i f the takeoff point moves
from the equator toward the poles. For example, the ranges for c=4000 m/sec and V = 7000
m/sec are 32430 km. for westward takeoff from the equator, 50440 km for eastward takeoff from
the equator, and 39363 km for takeoff from the pole; the corresponding ranges for C - 5000
m/sec, V
o
70QO m/sec are 32660 km, 58880 km and 42091 km.
nhen the bomb load i s dropped during the supersonic gl i di ng fl i ght , the weight suddenly
decreases by 10 tons from the value G, and the stationary al t i t ude increases b;
&tf=434h& tye. This decrease and the diminished bal l i s t i c loading of the ai r cr af t produce an
effect on the osci l l at i ng orbi t , as shown in Fig. 77 for a definite case. I t was assumed t hat ,
along the orbi t for C = 4000 m/sec, V = 7000 m/see, a 3.8 ton bomb was released horizontally at
40 km al t i t ude and 6060 m/sec velocity, so that i t struck the point on the earth opposite to the
takeoff point. For a bomb throw of 850 km, the rel ease must occur aft er 19150 km. At t hi s
point the plot of the orbi t s pl i t s into three curves; the path of the fal l i ng bomb, the dotted
path which the bomber would have followed i f no bombs were released, and the sol i d l i ne showing
the orbi t after the bomb release. In the l ast case the waves in the orbi t are higher and broader,
so that aft er several osci l l at i ons a definite phase shi ft i s observable, the stationary part of
the path l i es 1670 m. higher, and the final velocity of 300 m/sec is reached a few dozen ki l o-
meters sooner. The difference in range i s so sl i ght that one need not make a special i nvest i -
gation of the orbi t aft er the bomb release, but can use the approximate orbi t calculated for full
bomb load. Fig. 78 shows the elements of t hi s orbi t ; i t i s i nt erest i ng to note t hat at the point
of release the normal, acceleration jumps discontinuous!y from + 7 m/sec" to +19.5 m/sec
2
, because
the ai rcraft was in a trough and Before the bomb release (when i t weighed 13.8 tons) i t was in
dynamical equilibrium with the buoyant forces of the ai r .
The range of the rocltet bomber is largely determined by the length of the supersonic
glide path. This important quantity can be estimated t o a f i r st approximation, without doing
the exact" orbi t cal cul at i on, by set t i ng the i ner t i al force equal to the ai r r es i s t ance, ^^' " Ky
1 1
_- <j from which , , s a/-
c
This simple calculation i s sat i sfact ory up to V = 2000 m/sec. Above t hi s, the cent r i -
fugal force due to the curvature of the bomber a orbi t around the earth can no longer be
neglected. In second approximation we may set <&$$< = /^3-G^ y^V/ifi from which (18) :
us
C-BBZ96
o
8
<5
1
s 3
1
g
5
I
l
5?
OOOOOl
00006
00008
00001
00000
00005
0000<! A
OOOOf
HOOK
p
a
t
h

i
n

ooooz
OOOSI
oooet
1
i
OOOOi
0008
0008
OOOL
)

s
u
p
e
r
s
(

0009
$
0009
o
000* &
000
oose
om
OQSl
OQZt
0001
006
008
001
009
005
m
OOP
OQd
002
ost
Oil
001

. 9 o Q S 9
S 5 g (^ g
S/LU Ul peeds y$TV TWFl"
Fi g. 79: Coaparison of the ranges of (toe supersonic descent pat h for
equal i n i t i a l vel oci t i es eonputed by
a) Constant f l i ^ i t al t i t ude .
f k
_ .^w*
Innr
. .
b) Graphical determined f l i ght al t i t ude ^
c
" gj',^
1
*
c-
suoq irj g pw^ qmog
OOOOOi
00006
00009
00001
00006
00005
St
$
s
2 s
--
Cf C?
cr
Cf
<"
C3
o cs
st s s a
cf or cr cr
Cf
est
cr
8
Cl-
ip &
-CW*J y&wx
i 21
C- 8429*
These equations give good r esul t s for the st at i onary fl i ght paths unt i l we approach the velocity
of points on the ear t h' s surface. Kith the aid of these equations we can also include the effect
of the ear t h' s rot at i on, and obtain values of V
fl
, t , and S
g
for the absolute orbi t , i f in place
of the rel at i ve velocity V we use the absolute velocity V of the takeoff point, which i s
calculated from I t s velocity V
e
and the compass angle f of the i ni t i al velocity V
o
by the
equation tf - *?*&*+2\&\f
m
sri $ The effect of the ear t h' s rotation on the path length i s
greater than 1% in the most unfavorable ease for V - 2500 m/sec, but decreases considerably when
we change from absolute t o rel at i ve path length. The path lengths calculated in t hi s way can be
compared to the ranges of fi re for the sane munl e velocity V ; up to V
o
6000 m/sec and for the
glide number/ = 6.4 used here, they arc bet t er by a factor 3.k, and as we approach the velocity
of points on the ear t h' s surface the factor increases rapidly; i . e . , for the sane i ni t i al velocity
one can fly much farther than one can shoot.
Fig. 79 shows a comparison of the r esul t s of t hi s second approximation with the stepwise
calculation of the oscillating^dynamical path. The reason why the dynamical paths are consider-
ably longer than the stationary^for the same i ni t i a l speed i s mainly t hat whereas in the st at i on-
ary gliding fl i ght the energy consumption is di st ri but ed uniformly over the whole path, i t is
concentrated in the troughs of the dynamical path; the f i r st (and also the longest) jump, which
const i t ut es 15-30% of the t ot al range, has a trough only at i t s end, and only.loses there half
of the energy which would be l ost at the stationary al t i t ude ( t hi s energy represents 15-30% of
the t ot al energy st or e) . Other reasons for the great er range of the dynamical paths are that the
regions of great est energy consumption occur at low al t i t udes where favorable gas-dynamic gl i de-
numbers at high Reynolds numbers exi st , while the higher part s of the path, in which unfavorable
gas-kinetic glide-numbers and low Reynolds numbers exi st , occur in regions of rarefied ai r or
pr act i cal l y empty space; fi nal l y, the process of turning out of the climb path, in the case of the
st at i onary orbi t , has to occur under unfavorable angles of attack, resul t i ng in increased energy
consumption for the same length of path. Even more clear is the pl ot t i ng of the supersonic range
of fl i ght against bomb load or weight r at i o, as in Fig. SO. The last two figures do not include
the effect of the ear t h' s rot at i on.
Also important for the use of the rocket bomoer i s the si t uat i on where, aft er release of
Combs, the ai rcraft s t a r t s toward i t s home base, If the antipodes of the home base or other part s
of the opposite hemisphere are being attacked, the bomber will return simply by continuing along
i t s glide path after the bomb rel ease and flying al l around the eart h. For nearer t ar get s, there
i s the possi bi l i t y of return by shi ft i ng course after the bomb release. Three pos s i bi l i t i es are
Considered. Fi r st , instead of having the path be a great ci r cl e, after the i ni t i al propulsion,
l et i t be a ci r cl e, lying on the ear t h' s surface, with diameter Sf equal to the distance
between takeoff point and t arget . For fli ght along t hi s ci rcul ar cap, the ai r resistance for a
given velocity increases in the r at i o /&esjS&>^trfr
f;
fi.y V^fas/r/mft-Vfajf
because of the oblique centrifugal forces, while the ci rcul ar cap i s shorter than the great
ci rcl e only in the r at i o *"/$ Flight along the ci rcul ar cap will bp more favorable than flight
along the great ci rcl e only for vel oci t i es below 5600 m/sec, so t hi s method of turning seems sui t -
able for shorter distances of at t ack. However, in t hi s range there i s in most cases another
favorable turning procedure - to reach the t ar get along a great ci r cl e, to turn the ai rcraft at
the t arget as sharply as possible, and return home along another great ci r cl e. If the bomber
turns through a small angle **, then the r at i o of the work done, on the element of turn path of
l engt h/ """t o the ki net i c energy of the bomber at the beginning of t hi s element i s 2 f tfir^-/^/
which can drop to ^ = ^/ because of large centrifugal acceleration which can approach
the permissible l i mi t . The turning path then becomes a spi ral along which the tangential and
radi al accelerations are constant. I t i s not integrable and must be computed step by step. From
the equation, the r el at i ve loss of weight during the turn i s independent of the velocity at the
s t ar t of the turn, so i t can be computed once and for al l for al l the spi r al s. I t i s shown in
Fig. 81 for t ur n- spi r al s with tangential deceleration 1,75 m/sec
2
and cent ri pet al acceleration
50 m/sec
2
. After a 90 turn only 6QJS of the i ni t i al energy is l ef t , aft er a 180 turn only 37S
i s l eft ; i . e . , the sharp turn s t i l l costs a great deal of energy. There are intermediate pro-
cedures for turning, which are a combination of narrow turn with non-great ci r cl es; these are
characterized by l ess energy consumption than the l ast limiting case which only permits the use
of osci l l at i ng fl i ght paths.
A third turning method consi st s in the ai rcraft using only enough energy in the climb, to
enable i t t o reach the t ar get ; there i t turns at i t s small residual speed, and with the aid of a
fuel store on board, gets another push to give i t the energy for the return home. This method of
using two driving periods has the feature that the ai r cr af t t ravel s slowly over the t arget at low
al t i t ude. So on the one hand the bombs can be dropped with great accuracy on the other hand,
the fi re power i s less than fpr long range at t ack, and fi nal l y the bomber get" into the enemy
zone of defense at the t ar get . Fig. 82 shows the ranges of the rocket bomber wbpn i t i s turned
by the l ast two methods.
122
C-829(
30 *5 60 75 90 105 120 135 150 165 130 195*
Turn angla (5* in degrees
. 81; Loaa in kenatic energy during the turning through an angle
along the spiral turn, (see p. 122)
$
p a f l H O
o a> o o o
o o P- B
9 o P a.-a
f S e*- a P
4 * H* *( - -
P O 9
o n o
*-^ o _^
a
1
? ffuic
o Q =r
Bomb load B in tons
8 8; 3 a
O 8!

8"
-KZ
a
(0
g> 25000
<8 JOOOT
40000
B
50000
6Q00Q
70000
80000
90000
100000
I
400
95
90
t
85
H
S so
49
a
a
75
70
t 65
3 60
55
$Z>
&
e
o
&
g 4ff
H
w
35
30
25
20
15
10
0
' "
'
'
_
*
Bomb l oad
.OTonnen
i
flQTonmn
///WTormen
t
/ / ,
Y/SOTonmn^
/rWTonnen
.V
** *
*.
\
,
// On
/jr-5
H
OTonnen
0 Tormen

(
^
A
*
_ * *'
' - . ^ _ "
%
y .
v
-.
^ *
/ ^
*
>>

m
*^
*
m
1

^
-
^
8000 7000 6000 5000 " lOOQ 5000 2000 4000 Q
Flight velocity v in m/sec
Fig. 83: Stationary flight altitudes of the Rocket Bomber during the
supersonic glide nith various bomb loads.
125
C-84Z96
J
300
290
2B0
270
260
250
2W
230
220
210
%200
<
f
I
I *'
|
(dynamic ni ght height
- - - -
_
- - - -
_
I
[il
I f
. st at i onary f l i ght height
J
A^v/v'
*"*
r
. .
180
470
160
450
iQ
430
120
110
400
90
80
70
60
50
W
SO
20
40
0
8000 7000 6000 5000 WW 3000 2000 4000
fl i ght velocity v in m/sec
Fig. 84: Comparison of the actual and stationary fl i ght altitudes of
the Rocket Bomber during a supersonic glide path, (c = 4000
m/sec, v
Q
= 7000 m/sec.)
126
1
The same considerations apply to the al t i t udes of supersonic gl i di ng fl i ght as to the
climb especially as regards " st at i onary " and " dynamic " al t i t udes. The st at i onary al t i t ude- ,
determined by the equilibrium between the constant flight-weight and the propulsive plus cent ri f-
ugal forces (j ust as they were used for the second approximation to the length of the supersonic
path) are shown in Fig. 83, where the curves are drawn sol i d out to the point act ual l y at t ai nabl e
with the given bomb load and C= 4000 m/tec. The dynamical al t i t udes of fl i ght , which r esul t
from the varying i ni t i al conditions for the dynamical orbi t of supersonic glide flight, tend to
approach the st at i onary al t i t udes of fl i ght , as shown in Fig. 84 which shows the dynamical and
stationary paths for c s 4000 m/aec and V
Q
= 7000 ra/sec.
2. Path of Subsonic Gliding Flight and Landing
The subsonic gliding fl i ght , by i t s definition, begins at V =300 m/sec with G * 10 tons.
The l i f t coefficient for a favorable gl i de position i s then about C
a
= 0.2 and the corresponding
height i s H * 20 km. The supply of potential and ki net i c energy i s s t i l l 24580 kg m/^ , with
which a distance of 24580/tf = 98200 m. can be t ravel l ed for an average subsonic glide-number
f = ii. One can follow the descent in small vel oci t y-st eps V,, \
2
< V3 . . . by using the resul t
that the decrease in ki net i c ener gy^' *- I^J/Sf P*"* t hat ol the pot ent i al energy (Hj-H
2
)
must always equal the work against ai r resistance As.E - i n the stratosphere we get
ASmfrS^/kef+tS+pfeAfyiir
r
*#= 63?/J# Kytf " the
The subsonic gl i de path obtained from these equations i s shown in Fig. 85. We see that the sub-
sonic descent l as t s 11 minutes, and ends near .he surface of the earth at a velocity of 288 km/hr,
whereupon the landing can occur. The actual var i abi l i t y of the subsonic glide-number can affect
his path of descent.
The landing process begins at V = 288 km/h with C = 0. 2. The behavior of the aerodynamic
forces i s given by the upper polar of Fig. 34, so that the velocity of the bomber can be lowered,
by using landing-aids^ t o 288/^30^= 150 km/hr which i s required for mi l i t ary glide-landings.
With these polars we can determine the ai r resistance and the required angle of attack C for
al l vel oci t i es between 288 and IM^tm/hr, for G = 10 tons, so the landing procedure can be
followed by using the dynamical equations. I t i s shown in Fig. 86-.
tOO km
Fig. 85: Subsonic descent path of the empty Rocket Bomber assuming
a constant average L/D of 4
127
f
8L
9
O
r-
:x
a
o
n
a
c*
t d
f
N
Oi
Angle of at t ack #
<a oi S $ E
Resiaining f l i ght time fif\ sec
a ti ^
Gliding velocity v in km/hr

*
^
<&
1 1 8

a.
(?
P>
3
/ A
M
/
/
/

/
V V
k^
3
- W 1
v 1
1
/i
3. W 1
v 1
1
j*
* \
V
s
i
Knd of the descent
Beginning of terminal gl i de
\
i
i
Ending of the gliding
Beginning of roll
v. f-KOjEcnoN or THE BOMBS
1. Types of Projection
The rocket bomber has two di fferent methods of attack: point attack and areal at t ack.
The two procedures di ffer mainly in the mannei- and accuracy of the bomb rel ease.
In the procedure of attack on a pai nt , the bomb will be aimed preci sel y toward a point"-
shaped individual t arget , and released at moderate al t i t ude and velocity under the same condi-
tions as for ordinary bombers. In pract i ce the same sub-types of bomb release are available to
the rocket bomber as for other bombing ai r cr af t , e g. bomb release during horizontal f l i ght ,
dive bomb attack, bomb rel ease during climb, low-altitude bombing, et c. The well-known condi-
tions and di f f i cul t i es of these types of release apply pract i cal l y unaltered t o the rocket bom-
ber, especi al l y as regards the accuracy at t ai nabl e and the need for adequate vi s i bi l i t y at the
t arget , ao nothing new can be said about these t yp
v
s.
Things are quite different for attack on ai area. Here the bomb i s thrown from great
al t i t udes (50-150 km) and at very high vel oci t i es of fl i ght (up to 8000 m/sec), i . e . under con
ditio&s far beyond those of long-range ar t i l l er y f i r e. Since the t arget , for the distances i n-
volved, wi l l not be vi si bl e, the release on an area wi l l be aimed i ndi rect l y, e. g. by cel est i al
navigation. Thus i t is independent of weather and vi s i bi l i t y at the t arget . Because of t hi s ,
i t does not reach the accuracy of release on a point, and we must expect spreads of several
kilometers. So with areal bombing one cannot hi t part i cul ar poi nt s, but rather a correspondingly
large area, with suffi ci ent probabi l i t y. To achieve an anticipated effect on t hi s whole sur-
face, a single drop wi l l not suffice, rat her we wi l l have to project several bombs toward the
same t arget ; these will di st ri but e themselves over the surrounding surfaee according to the laws
of chance. The di st r i but i on of hi t s inside the area will not be uniform; the bombs will st r i ke
more frequently in the neighborhood of the t arget than far away; there will also be unavoidable
bomb-hits far outside the area being attacked. However, on the basis of laws of probabi l i t y, the
bomb di st r i but i on can be predicted well enough so that the goal of the attack can be achieved
with the same or even greater accuracy than for point at t ack.
2. Flight Pat h of the Bomb
In order to make calculations concerning the bombs thrown from rocket ai r cr af t , we must
make some assumptions about the external shape of the pr oj ect i l e. Best suited to the exi st i ng
conditions i s a bullet-shape, with f l at base, with i cylindrical tapered portion at the rear,
with l argest possible r at i o of height h to calibei i. With say -j}-- 8 we get, from the wel l -
known gas-dynamic laws for very high Mach numbers, i resistance coefficient of C,
;
0.014-*- 1.43
A /\2,~ where in the fri ct i on contribution C^,- 0.004 we neglect the st abi l i si ng surfaces, since
they probably can only manage to set the proj ect i l e spinning, about i t s axis in the i ni t i a l part
of the flighL, before t hei r thin walls are destroyed due t o the temperatures developed by Fric-
tion
If one assumes t hat the explosive const i t ut e; 50% of the bomb's weight, then bombs weigh-
ing 30 (or 5 or 1) tons have lengths of 11.20 (or 6.16 or 3.60) m,, and cross-sect i on loadings
of 19.5 (or 10.7, or 6.2) tons/nr. In a pract i cal si t uat i on any bomb load could be made up out
of these three si zes. The considerable space required for the pr oj ect i l es will require an ad-
justment of the bomb-load-and tank-space to the purpose at hand, in the sense t hat larger fuel
loads will be accompanied by decreased bomb loads and vice versa; i . e . , the ai r cr af t described
ear l i er i s best suited to the f i r st case.
In estimating the path of fl i ght of the bomb *e can proceed in the same way as in det er-
mination of the ascending or descending path of the ai r cr af t i t sel f . The force pi ct ure di ffers
from that for the climbing ai r cr af t , since we have assumed that no aerodynamic l i f t forces act
on the bomb and that the bomb can no longer be kept in a definite orbi t plane by a pi l ot , so
t hat i t follows the tendencies to sideways motion due to atmospheric rot at i on (weather-vane
action) and earth rot at i on (Cori ol i s-force), und describes a twisted orbi t in space. As shown
previously, the weather-vane action i s caused by the fact that the bomb, as i t f l i es over places
of different l at i t ude, continually moves through layers of ai r of different absolute speed
(depending on l at i t ude) , and thus is acted on by a cross-wind, which produces horizontal t r ans-
verse forces; these do not act at the center oi mass of the pr oj ect i l e, but behind i t , at l east
i f the proj ect i l e has fi ns. This force as well as the Coriolis force was assumed, in the case
of the ai r cr af t , to be eliminated by transverse steering which develops forces equal and opposite
to them. In the case of the bomb there are no controls; as a resul t of the weather-vane action
i t will not only dr i f t to the si de, but also as a consequence of the resul t ant torque, i t will
s t ar t to rot at e about a vert i cal axis through the center of mass. The fl i ght path of the bomb
i s t hus det ermi ned by f i ve ext er nal f or ces, of which four act at t he cent er of mass and t he
f i f t h a c t s behind i t : weight of t he bomb, a i r r e s i s t a nc e , t r ansver se a i r f or ce, Cor i ol i s f or ce,
and d' Al embert i an j n e r t i a l f or ce.
The t r ans ver s e force i s per pendi cul ar t o t he t anpent t o t he pat h, poi nt s t he way t he
t r ans ver s e wind i s bl owi ng, i s pr opor t i onal t o t he a i r r es i s t ance and t he square of t he angl e
Under which t he unpert urbed a i r st ream meets t he Symmetry pl ane of t he a i r c r a f t . Thi s angle in
t ur n i s det ermi ned by t he s t r e ngt h of t he cross-wi nd, i . e . by t he course of t he bomb and i t s
vel oci t y r e l a t i v e t o t he ground, by t he magnitude of the moment of t he crosswi nd about t he cen-
ter- of mass of t he bomb, and f i na l l y by t he slowness wi t h which t he bomb responds t o t he c r os s -
wind and i t s t or que, i n t ur ni ng i t s apex toward t he crosswind. Thi s i n e r t i a shoul d be as l ar ge
as pos s i bl e, s i nce t he tendency t o t ur n always e xi s t s except when t he bomb i s in t he pl ane of a
l a t i t ude c i r c l e . We can get an i dea of t he magnitude of t hi s ef f ect by consi der i ng t he 300 km.
long pat h of f al l for a poi nt of r el eas e over a pol e of t he ear t h. The di f f er ence i n cross-wi nd
between t he poi nt of r e l e a s e and t he poi nt of impact i s , in t h i s case, t he absol ut e ve l oc i t y of
' r ot at i on 22 m/sec of t he poi nt of i mpact . Thus t he angl e of t he cross-wi nd blowing a ga i ns t t he
p r o j e c t i l e i s 1/2 degree for a mean forward speed of 3000 m/ sec. For such angl es of i nci dence,
in t he Newtonian ve l oc i t y r egi on, t he t r ans ver s e force would be equal in magnitude t o i he r e s i s t '
ance of t he p r o j e c t i l e ; t he pr oj e c t i l e wi l l t hus car r y out t he def l ect i on r api dl y des pi t e i t s
l ar ge moment of i n e r t i a , so t hat t he t r ans ver s e force cannot i ncr ease t o such magni t ude.
The remai ni ng f or ces , t he Cor i ol i s force and t he d' Al emhert i an i n e r t i a l f or ce, can be
given from our pr evi ous anal ys i s . In doi ng t h i s we should note t hat t he system of forces act i ng
on t he bomb i s gener al l y a s pa t i a l syst em, i . e . i t cannot be balanced by a s i ngl e i n e r t i a l f or ce,
but r a t he r by a " s cr ew" .
I f f i na l l y t he bomb s t a r t s t o r ot at e about i t s axi s of symmetry, then gyr oscopi c, Magnus-,
and Pbi sson- ef f ect s s t a r t , which af f ect t he course of t he bomb.
Jn or der t o obt ai n t he pat h of t he bomb, t he force components al ong t he pr i nc i pa l di r e c -
t i ons of t he compass and al ong t he ve r t i c a l must be determined. and t he equat i ons of motion for
t he t hr ee di r e c t i ons i n space must be wr i t t e n down. I nt egr at i on of t hese would give an exact
des cr i pt i on of t he t wi st ed pat h in space. Thi s ext ensi ve gener al i zat i on of t he well-known
' Fundamental Problem of Ext er i or Ba l l i s t i c s
1
i s not di r e c t l y s ol ubl e.
To get a pr el i mi nar y pi ct ur e of t he pat h of pr oj ect i on, t he range, f i nal ve l oc i t y of t he
bomb, time of f a l l , angl e of impact, e t c . , we can use t he well-known si mpl e procedure of PonceleV
and Didion for st epwi se gr aphi cal cons t r uct i on of t he path of f a l l , Thi s procedure t akes ac-
count of t he e a r t h cur vat ur e, convergence of v e r t i c a l s , and var i at i on of a i r dens i t y and gr a vi t y
wi t h a l t i t u d e , and i s t her ef or e t he best s ui t ed of t he usual b a l l i s t i c pr ocedur es for t he rocket
bomb. In t he act ual cal cul at i on of the bomb pat h by t hi s procedure, i t t ur ns out t hat t he ef f ect
of a i r r e s i s t a nc e on t he pat h i s ext remel y smal l , f i r st , because of t he low a i r dens i t y along most
of t he pat h, second because of t he small c oe f f i c i e nt of ai r r es i s t ance for t he s l ender b u l l e t -
shaped body. The pat hs can t her ef or e be very accur at el y descr i bed as Kepl er e l l i p s e s , and then
t he range and angl e of impact can be det ermi ned from t hese. The act ual vel oci t y of impact of t he
bombs wi t h t he e a r t h wi l l be decreased by a few per cent because of t he a i r r e s i s t a nc e .
The range of the bomb thrown hor i z ont a l l y and f a l l i ng al ong a Kepler e l l i p s e i s given by
W=S?arKeosfU{t-{#**i>//e&~}$
t,
'
A
*
re
E =S-r.

{#+H)A
/
*
Z i s t h e
e c c e nt r i c i t y of t he e l l i p s e .
From t hese equat i ons , trie r e l a t i on between t he hei ght at r el ease H, t he vel oci t y at r el eas e V
Q
,
and t he range W, i s pl ot t e d in Fi g. 87. The dur at i on of f al l can be det ermi ned by i nt e gr a t i ng
t he pat h l engt hs or by a s t ep- by- s t ep c a l c ul a t i on of t he or bi t , and gi ves val ues correspondi ng t o
t hose shown i n Fi g. 88. For l ar ge v e l o c i t i e s of r el eas e, the ear t h r ot a t i on has a not i ceabl e
ef f ect on t he r ange. Since t he r el eas e poi nt i s out si de the e a r t h ' s s ur f ace, one can i n s e r t for
t he vel oci t y of r el eas e t he absol ut e vel oci t y of t he bomb at t he poi nt of r e l e a s e , and then ob-
t a i n t he absol ut e l engt h of t he range from Fi g. 88; the r e l a t i ve range on t he ear t h can then be
cal cul at ed in t he usual manner.
One t hi ng t o be checked i s how warm t he bomb becomes in f a l l i ng t hrough t he lower l ayer s
of a i r . In t he case of t he bomber, which has i t s high speeds i n r egi ons of r ar ef i ed a i r , so t hat
i t f l i e s at moderate s t agnat i on pr es s ur es , i t was assumed t ha t equi l i br i um between heat i nt ake
and r adi at i on can be mai nt ai ned by having a s t r ongl y r a di a t i ng ski n for s u f f i c i e n t l y low wal l
t emper at ur es, or t hat c r i t i c a l thermal s t r e s s e s can be wi t hst ood in t he t roughs of t he pat h by
havi ng a ski n wi t h s uf f i c i e nt heat capaci t v. During t he f al l of t he bomb t hrough more denst l ay-
e r s of a i r , t he heat t r ans f er per uni t ar ea of t he bomb sur f ace wi l l i ncr eas e gr e a t l y, but t he
bomb' s f al l l a s t s a much s hor t er t i me, so t h a t one may compute usi ng t he heat capaci t y of t he
s hel l of t he bomb. .The s t agnat i on- and al s o approxi mat el y t he f r i c t i o n - , t emper at ur e can be
130
C-S24
150
130
120
110
109
v
m
cd
IB

S 90
O
*>
6
H
V
X
80 \
7U
60
50
40
30
20
10
T-

n
n~
"
f
f
jr
#
# #

b:
H 4/
&
I /
i
f i
i

I
/
?
/ / /
/ / '


/
/
/
~~f '
1
: ,- ,*
: : "
4
!
:
*
'
.-'
i :
!
/
"
/ - y
'
i i f
i

.* ,*
' !
f
"
,
: : : - :
: :
s
t
*
'
: '

.
!
/
i
/
.

i
?
t
/

;
,'
.'"
:
/ /
.'
y
..
i
/
.*'
.' /
*
1
/
/ ,

_
s
. * '
4 *
/ / ."
* "
200 WO 600 860 1000
Throwing
(Path as
and also
1200
Throwing di
distance of surface bomb according to second approximate cal
a Keppler ellipse, taking into consideration the Earth's cur
the approximate air resistance) See p. 130.
Jt h.
u
?
150
%1>tO
iZO
o
to
d
9
H

o
J5
bQ
H
X
90
80
70
60
50
30
20
10
~f-
4

A
~f-
4
f
A
7
-
^

~"
~Jr
- - - - - -
Fi g. 88:
50 100 150 200
Dura
Dependence of durat i on of throw as a function of i n i t i a l a l t
vel oci t y.
Flight plane
Depth oX natural
horizon
.Natural horizon
Earth sphere
Flight plane
Fig. 89: The third aiming phase of bomb release (see p. 134).
133
C-B4296
est i mat ed from the gas-dynamic energy equat i on t o be AT- V
2
/2600 i f one assumes t hat for hi gh
t emperat ures of t he boundary l ayer , t he vi br at i onal degrees of freedom of t he a i r molecules ar e
compl et el y exci t ed, but t ha t no di s s oc i a t i on occurs in t he boundary l ayer . The laws governing
t he t r ans f er of heat from t he heated r egi ons of t he boundary l ayer t o t he r i gi d wall ar e unknown,
but one can s et up a rough energy bal ance by usi ng the est i mat ed coef f i ci ent of laminar f r i c t i on
of t he boundary l ayer Cf
:
0.0003 t o cal cul at e t he work of f r i c t i on per uni t sur f ace per s e c . as
Trv = Cff>v* ASc/fv* for V - 6000 m/sec at the e a r t h ' s s ur f ace, t h i s gi ves 5.4 hp/cm
2
. I f
one assumes t hat a t hi r d of t hi s heat goes t o the wal l , then one obt ai ns t he condi t i ons of t he
j e t t hr oa t of a powder r ocket mentioned on page 8; i . e . an i r on j a c ke t ntily 2 cm. t hi ck would
onl y begin t o mel t af t er 10 s e e . as a r e s u l t of t h i s heat i ng. From mire r ecent measurements of
heat t r ans f er at high Mach numbers, i t appears t o i ncr ease more sl owl y than the r a t e of f r i c t i on'
a1 work, i . e . i t goes l i ke JO ' f t h i s i ndi cat es t hat t he bomb can go through t he c r i t i c a l
lower atmosphere even wi t hout t he use of st r ongl y r a di a t i ng r e f r a c t or y pr ot ect i ve l ayer s for t he
out er cover i ng.
Fi nal l y some gener al cons i der at i ons ar e necessary concerni ng t he accuracy of h i t t i n g for
bomb r el eas e from gr eat a l t i t u d e s at high speed. In gunnery one assumes a di amet er of t he SOX
c i r c l e equal t o t he range of f i r e . For t he rocket bomber, t he range of f i r e in t hi s sense i s
t he l engt h of t he pat h of f al l of the bomb. For a mean l engt h of say 600 km, t he probabl e
s c a t t e r would be 6 km, i f one could r el eas e from t he bomber wi t h t he same accuracy as one can
f i r e from a cannon, The act ual e r r or wi l l be the r e s ul t a nt of e r r or s in r e l e a s e , which ar e
determined by pos i t i on- , ve l oc i t y- , and di r e c t i on- e r r or s a t t he poi nt of r e l e a s e , and of devi a-
t i ons duri ng the f a l l , which ar e caused mainly by f l uct uat i ons i n t he dens i t y and flow of the
a i r . The navi gat i on of t he rocket bomber t o t he r el eas e poi nt i s di vi ded i nt o t hr ee aiming
pr ocedur es. The f i r s t phase c ons i s t s in having t he c a t a pul t appar at us l i e i n t he di r ect i on of
t he t a r ge t , i f pos s i bl e; because of i t s fixed i n s t a l l a t i o n devi at i ons up t o 90 may occur . The
cor r ect i on of t hi s e r r or occurs in t he second phase r i ght a f t e r t akeoff, dur i ng t he rootorless
f l i ght or r i ght at t he s t a r t of t he pr opul si on per i od, when t he di r ect i on t o t he t a r ge t can be
s et by means of t he compass t o wi t hi n a few minutes of a r c . During t he gl i de , t he a i r c r a f t must
be s t eer ed very a c c ur a t e l y, si nce syst emat i c ef f ect s - r ot at i on of t he e a r t h and atmosphere, and
acci dent al i nf l uences , such as small asymmetries, e r r or s i n s t e e r i ng, f l uct uat i ons of a i r dens i t y,
and movements of t he a i r , cont i nual l y tend t o br i ng t he a i r c r a f t out of i t s or bi t . During t he
gl i de above t he t r opospher e, t he t hi r d phase, on which t he accuracy of t he bomb h i t s depends, i s
completed. In t hi s one can t hi nk of usi ng ast ronomi cal methods for poi nt i ng; t hi s wi l l be i n-
dependent of t he i nf l uence of t he weather or t he enemy; in t he i nt er f er ence- f r ee s t a t i ona r y gl i de
pat h before r e l e a s e , i t per mi t s det er mi nat i on of pos i t i on t o an accuracy of a few seconds of a r c ,
correspondi ng t o pos i t i on e r r or s of l i t t l e over 100 m. To det ermi ne t he apparent horizon under
t he excel l ent v i s i b i l i t y condi t i ons a va i l a bl e , one can s i ght on t hr ee poi nt s of t he nat ur al hor i -
zon; from the depth of t he hori zon t he a l t i t u d e can be det ermi ned. When t he apparent hori zon has
been f i xed, t he maintenance of t he pr escr i bed or bi t in space can be assur ed by choosi ng one or
s ever al s t a r s i n t he pl ane per pendi cul ar t o t he hori zon through t he pos i t i on of t he t . t r get , and
fol l owi ng . t hei r apparent motion wi t h a t he odol i t e . I f the gui di ng s t a r , dur i ng t he dynamic
f l i ght , s t ays in t he pl ane of t he or bi t ( i . e . on t he ve r t i c a l c r os s - ha i r ) t he p i l o t knows t hat
he ia in t he pr escr i bed or bi t - pl a ne . On t he s t a t i ona r y f l i ght path he can det ermi nt hi s absol ut e
vel oci t y from t he appar ent motion of t he s t a r al ong t he ve r t i c a l c r os s - ha i r , and hi s pos i t i on and
t hat of the bomb r el eas e from t he a l t i t u d e of the s t a r . Whether t he t a r ge t i s vi s i bl e at t he
i ns t ant of bomb r el eas e i s uni mport ant . The det er mi nat i on of the poi nt of r el eas e i nvol ves only
t he small er r or i n angl e measurement, which corresponds t o a few hundred met er s. Fi g. 89 i s a
p i c t o r i a l pr es ent at i on of t he condi t i ons duri ng t hi s t hi r d aiming phase. The e r r or s a r i s i n g
duri ng the f al l of the bomb ar e more s i g n i f i r a n t . I f among t he f act or s af f ect i ng t he spread of
pr oj e c t i l e s : di f f er ences in pr opul si ve charge, vi br at i ons of t he b a r r e l , met eor ol ogi cal f l uc-
t uat i ons , di f f er ences in t he p r o j e c t i l e s and e r r or s in di r e c t i on, onl y t he l a s t two ar e consi dered
and each i s given equal wei ght , t hen t he spread would be about 2/ 5 of 1%, i . e . .6% ( ??) , which
for a range of pr oj ect i on of 600 km, gi ves a di amet er of 3. 6 km for t he 50% c i r c l e . To t h i s i s
added t he er r or in navi gat i on of t he a i r c r a f t i t s e l f , so t hat one obt ai ns for t he probabl e devi a-
t i on from t he t ar get of a s i ngl e r e l e a s e , H^, - 3 km, wi t h which we can make approximate cal cul a-
t i o n s .
3. Ballistics of Impacts
The process of impact for poi nt r el eas e and ar ea r el eas e di f f e r fundamentally in having
very di f f er ent ve l oc i t i e s and angl es of impact. The pr ocesses i n poi nt bombing ar e s i mi l ar t o
t hose for ordi nary bombing or heavy mort ar f i r e , so t hat t he necessar y r e s u l t s can be wr i t t en
down; e . g. in dive bomb a t t a c k by t he rocket bomber wi t h 30, 5, or 1 ton bombs, for a f i nal
di vi ng apeed of 500, 300, oj, 260 m/ see, t he penet r at i ng power of bombs through t he e a r t h ' s cr us t
i s 100, 30, or 12 m., for r ei nf or ced concr et e t he val ues ar e 10% of t he above; t he correspondi ng
penet r at i ons through armor pl at e ar e 200 cm. ( 1. 43 c a l i b e r ) , 60 cm (0. 86 c a l i be r ) or 25 cm.
; 0. r o c a l i be r ) , in otUer word
1
; I ' v a ^ r than t he s t r engt h of a l l known s h i p ' s armor.
Ent i r el y ttpw condi t i ons orcur for t he ar ea bomb, which has a ve l oc i t y of impact 10 times
as gr e a t . Trie en'-rgy of ini;-acl is much ; / r eat er than t he energy cont ent of t he expl osi ves in t he
boirb. Th^ s t r engt h of the n
;
at er i al of the bomb i Lsel f wi l l permi t i t t o penet r at e a s t r u c t u r e ,
or even t o pn through a c i t y wi t h numerous bui l di ngs , because of t he smal l angl e of impact; i t
wi l l not permit penet r at i on i nt o t he e a r t h. 'Hie e l feet wi l l thus be s i mi l ar t o t ha t of a mine.
The ratige of t he expl osi on of a de f i ni t e amoutit G Lkg) of expl osi ve can be est i mat ed as h = /4"fl
under t he assumption t hat the sur f ace dest royed i s pr opor t i onal t o t he expl osi ve char ge; here r
i s t he r adi us r.f des t r uct i on in met er s, and k i s a const ant which gi ves t he degree of de s t r uc t i on;
for a i r pr es s ur es of about 2lii;t)U kg/or which produce t he worst ef f ect s on bui l di ngs and smash anv
except s r c c i a l U r ei nf or ced one, k i s 3; for pr essur es of 5000 kg/ill , k i s 12, for which val ue
s l i ght damasv t o s t r uc t ur e s KCUFS, wal l s ar e over t ur ned, and gabl es ar e dest r oyed; for k
;
25
t he sai'i; di st ance of or di nar y hui l di nga from expl osi ve st or ehouses i s reached, and k
:
200 gi ves
the c i r c l e at which window panes and, pa r t l y, window frames ar e broken.
These e f f e c t s a r e di s t r i but e d uniformly in a l l di r e c t i ons from t he poi nt of impact of a
i t r y t hi n- wal l ed bomb, and t he r e s ul t of impact of t he r i gi d body of t he shock wave moving wi t h
gr eat er than sound ve l oc i t y through t he s t i l l a i r . The development of such shock waves in fiutit
of bl unt pr oj e c t i l e s f l yi ng at super soni c speeds i s well known. The.phenomenon of an expl osi on
wave i s qui t e s i mi l a r , except t ha t hern t he e xc i t a t i on comes only t o a smal l ext ent from the
bomb fragments thrown out of t he homh cover, and mostly from the combustion gases of t he expl o-
s i ve, which for udi abat i c expansion from t he pr essur e and t emperat ure of t he expl osi on t o a
normal pr essur e reach r adi al v e l o c i t i e s of f2f&>\ 3400 m/
s e c
for E
:
1400 k c a i / ^ whi l e i ndi -
vi dual por t i ons j f t he pas can reach even gr eat er ve l oc i t i e s at t he expense of ot her por t i ons .
I f t he expl osi ve energy i s a l s o shared wi t h t he cover of t he bomb ( r epr es ent i ng say 50% t he t o t a l
mass), then t he vci ori Ly of the r adi al expl osi on o.-ve drops t o 2400 m/ sec, a f i gur e which checks
well wi t h act ual Iv measured ve l oc i t i e s of fraptiients. The expl odi ng mat er i al of t he bortb c ol l i de s
wi t h the surroundi ng s t i l l a i r at t h i s vel oci t y, and s t a r t s t he powerful and f ar - r eachi ng ex-
pl osi on wave in i t .
From t he mechanics of t he expl osi on process we can get a c l e a r pi c t ur e of the ef f ect of
high impact vel oci t y of an ar eal bomb on t he expl osi ve e f f e c t . In t he fol l owi ng cons i der at i on
we s hal l assume an impact Vel oci t y of 8000 m/sec, *>hieh we would get i f t he a i r c r a f t descended
t o t he e a r t h ' s sur f ace at 8000 ms e c and r e l e a s e ! t he pr oj e c t i l e at s hor t r ange. Aft er det ona-
t i on of the ar eal bomh on or above t he ear t h, ' s sur f ace t he mass of t he r e s ul t a nt bal l of f i r e has
i-.:'l onlv i r adi al vr l oc i t y of ~ H)U in' ' sec, but al s o the rtmvected forward speed of 8000 m/ sec.
'!i'p i i . l i mi t n:s superpose as shown in Fi g. "0 t o produce a vel oci t y r e l a t i ve t o t he surround-
i'i." - t i l ' a i r . The i r oni face of t he expl osi on sphere c ol l i de s wi t h t he a i r a t a vel oci t y of
_' Hit) * fi'jllt) - IUIUU ni s ec, and e xc i t e s a shock wave as i l tilP evpl osi ve had (1O4OQ/240Q)
2
= i i ), 7
Limes a.i much energy cont ent . The i nt e ns i t y of t he expl osi on wave t her e i s 18 times as gr eat as
l or a bomb exploiting a t r e s t . The i nt e ns i t y drops r api dl y for the sideward di r e c t i ons , and d i s -
appeai s compl et el y at the r e a r .
Si ncr tlw area of des t r uct i on by an expl osi ve ch:iri_'e i s p r o p o r t i o n ! t o t he weight of t ht
i -.j'li.n i v, or morp pr eci s el y Ln the energy avai l abl e for t he expl osi on wave, t he area of dps t rue-
' ;i i' u( an ar eal bomb i s i ncreased in the r a t i o of t he sum of impact energy and expl osi ve e;irr?v
\<-. t tie. Ut t e r al one, i . e . in the r a t i o (l U00' + 80UU") MOU- - 12. 1. At t he same time t he ..(.
! roveo an? a l oses i t s c i r c ul a r shape, and becomes a dr op- l i ke area along' t he di r e c t i on of r el eas e
w
;
K-ii- out l i ne can he . i l rul . i t ed from the square of the r esnl t ai i l of i mpact - and expl osi ve shnck-
vi "l oci t i es . The r a t i o ot dr st r o>cd ar eas for the same bomb for |-*nnt r el eas e and area r el eas e
i.-, shown n l i p . <>U for an. irr.p.it t vel oci t y of fJOOO m/ sec. The de s t r uc t i ve act i on of a bomb land-
iup af t er ar eal r el eas e i s antch gr e a t e r tlian t hat of an e^ual - si zed normal l y-dropped bomb, i s
fan-shaped and poi nt s in the di r e c t i on of r el eas e. Thus t he ef f ect of t he bomb no longer depends
onlv on t he enerev cont ent of llit' expl osi ve; t he ki t i et i c cnersy of t he bomb a l s o produces i t s
full ef f ect . The ef f ect of t hr fragmentation of thp burst itifi bomb-shell i ncr eases and d i s t r i b u t e s
i t s e l f in t he same manner as t he i nt e ns i t y of the pxpl osi on wave. We pet t he i ns t r uc t i ve con-
cl usi on for the r ocket bomber t ha t , hrratiM
1
of tl . addi t i onal i:nerry of impact, the des i r ed
decree of des t r uct i on of a given sur f ace can he accomplished by ar ea bombing wi t h much smal l er
bnmh loads than for poi nt bowling.
135
C-B4J96
VI. Types of Attack
1. Basic Types of Attack
The type of at t ack procedure t o be used by t he r ocket bomber in any s pe c i f i c case i s
determined by t he nat ur e of t he t a r g e t and i t s di s t ance from t he home base.
The - Ur aor di nar y var i et y of t a r g e t s i s di scussed in Sect i on VI-9. There we di scuss i n
gr eat er c . ^ a i l t he bas i c di f f er ence between poi nt and area t a r g e t s , accordi ng t o which t he t ypes
of at t i ' . ' .in be subdi vi ded i nt o poi nt - a t t a c k and ar ea- at t ack pr ocedur es.
The i ndi . j - ' ual t ypes of poi nt at t ack follow from t he requi rement t hat t he bomber f l y as
slowly as possi bl e over t he t a r ge t , so t ha t i t may have r a t he r smal l r es i dual energy t he r e . I f
in s pi t e of t h i s , t he bomber i s t o r et ur n t o i t s t akeof f f i e l d wi t hout a st op- over , then af t er
dropping i t s bombs, over the target, i t rft-*t be pr opel l ed by i t s own r ocket motor u n t i l i t has
acqui red a s uf f i ci ent speed t o {fit l.c:f- on t he correspondi ng energy. Thus we ar r i ve at a pr o-
cedure for poi nt at t ack i nvol vi ng two pr opul si ons and r et ur n home, which cons i s t s e s s e nt i a l l y i n
having t he bomber, a f t e r bei ng cdtapt>it-;d a t t he home base, accel er at ed onl y t i l l i t acqui r es
enough energy t o bri ng i t over t he t a r ge t . There i t r el eas es and t ur ns at t he l owest pos s i bl e
speed, then s t a r t s i t s motor with t he r es i dual s t or e of fuel , t o get up enough energy for t he
home t r i p , and lands back at i t s home bas e. Very l arge qua nt i t i e s of fuel ar e r equi r ed for t h i s
double pr opul si on, so t ha t t h i s procedure can be used only for l i mi t ed ranges of at t ack (up t o
600ti km) and l i mi t ed bomb l oads. Poi nt a t t a c ks over gr e a t e r di s t ances or wi t h l a r ge r bomb l oads
than i n t h i s f i r s t procedure ar e pos s i bl e i f t he bomi-er can land not too f ar from the" t a r g e t ,
and t ake on new f uel .
For t he poi nt - a t t a c k procedure wi t h two dr i vi ng per i ods, p a r t i a l t ur ni ng and a uxi l i a r y
poi nt , t he bomber i s agai n accel er at ed a f t e r cat apul t from t he home base, u n t i l t he acqui r ed
energy c a r r i e s i t j u s t t o t he t a r ge t . Then i t r e l e a s e s , t ur ns through t he r equi r ed angl e at
l eas t pos s i bl e f l i ght speed, s t a r t s i t s motor with a small r esi due of fuel on board, t o get t he
small amount of energy which c a r r i e s i t t o t he auxi l i ar y f i e l d not f ar from t he t a r ge t ; i t l ands
t her e and t akes on new fuel , ni t h t h i s , i t t akes off agai n in normal fashion and r et ur ns t o t he
home base; i t has t he p o s s i b i l i t y of making f ur t her bombing at t acks on t he way home.
I f a ooi nt at t ack i s t o be car r i ed i u> over a l ar ger di s t ance or wi t h very gr eat bomb
load and t her e i s no pa s s i bl e a uxi l i a r y l anci ng pl ace f ai r l y near t he t a r ge t , then r ocket -
t echni que, as seen at t he pr esent t i me, gi ves no p o s s i b i l i t y of r e t r i e vi ng t he bomber and br i ng-
ing i t back t o i t s home f i el d. I f at t ack on t he t a r ge t seems more i mport ant than t he bomber
' , t s-i l f {which has only * r e l a t i ve l y smal l mat er i al val ue}, then t her e i s t he p o s s i b i l i t y of
s a c r i f i c i ng t he bomber a f t e r t he a t t a c k. This procedure of poi nt at t ack wi t h a s i ngl e pr opul si on
;ipriod and s a c r i f i c e of t he bomber i s , i n pr i nc i pl e , appl i cabl e t o a l l poi nt s on t he e a r t h ' s sur-
i.fp. [L i s , nat ur al l y. , t o be appl i ed t o at t acks and t a r ge t s of very s peci al s i gni f i cance, as
Idr cxumple the s ur pr i s e des t r uct i on of a government bui l di ng and t he governi ng group assembled
t her e, t o t he k i l l i n g of a s i ngl e, s pe c i a l l y i mport ant enemy person, t o s i nki ng l ar ge enemy
t r ans por t s or warshi ps, bl ocki ng of i mport ant avenues of commerce ( say canal s or s t r a i t s ) , and
t o s i mi l ar speci al cases; t h i s i s l es s because of t he l oss of t he a i r c r a f t than for t he more
valtiahln p i l o t .
For procedures of at t ack on an area t he need t o f l y slowly over t he t a r ge t di s appear s , so
t hat one has more freedom in car r yi ng out t he procedure. The most obvi ous procedure for area
at l . i ck. with s i ngl e pr opul si ve peri od and r et ur n home, cons i s t s in t he bomber bei ng cat apul t ed
liom i t s home base, and then dri ven u n t i l it, ge t s s uf f i c i e nt energy t o get t o t he v i c i n i t y of
the t a r ge t , t urn and get back home. The t ur n pat h uses up very l ar ge amounts of energy, so t hat
t hi s at t ack procedure remains l i mi t ed t o small di s t ances and bomb l oads.
Area at t ack over gr eat di s t ances i s very much s i mpl i f i ed, i f an auxi l i ar y f i e l d e x i s t s noc
t if- i"ar fron, t he t a r ge t , so t hat t he ban,her can lano and t ake on new fuel for t he r et ur n t r i p .
]:-. I I.is ens*- t he area at t ack goes as fol l ows; af t er r el eas e t he bomber makes a p a r t i a l t ur n
t hrr (itli .in .ingle l ess than 180 ( t h i s r equi r es smal l er energy consumption t han for a complete
t.ii'iiil. ri-;n ! l i e s t o t he auxi l i ar y f i e l d on i t s r esi dua] energy. Thi s area at t ack wi t h s i ngl e
pr-i, i-i ziii.rj, pa r t i a l t ur n and auxi l i ar y f i el d i s appl i cabl e t o a l l di s t ances on t he e a r t h; i t
aST,.- Li-never, t hat wi t hi n at most a few thousand km. from t he t a r ge t t her e i s a s ui t a bl e
iuxi (i.ir-. l i ul d, for l andi ng, and which has a t ukcof f appar at us. In view of t he l ar ge number of
r c s s i bl e t a r ge t s for ar ea at t ack, t h i s requi rement can be f ul f i l l e d onl y in except i onal cas es .
1%
rf0
jfetmifast ^
f
TV
\
12.1 times greater area destroyed
by tangential release
Relative velocity with
respect to air
Bomb velocity at
time of hitting
Explosion velocity
Destruction area of the bond)
for a point drop
o
c
o
O
e
h
D
Fig. 90:
Comparison of the destruction area for same bond}
a; point hit
b) tangential velocity of 8000 m/sec.
(see p. 135)
u
n
137
C-?4N6
The value of the auxi l i ary point can vary Considerably, not only according t o i t s distance
fTom the t arget , but also because of the size of the required angle of turn. Since large angles
of turn are much more harmful than great distances, an obvious idea i s to provide auxi l i ary
points beyond al l forseeablc t ar get s; e. g. beyond the two population concentrations outside of
Europe (North America and S. . Asia), say on the Marianaf in the Pacific Ocean or on the
islands of t hat ocean off the Mexican coast; or to secure a single auxiliary point at the ant i -
podes of the home base, say in New Zealand or on the islands east of i t . This auxi l i ary point
at the antipodes can always be reached by st r ai ght fl i ght without turning, no matter what point
on the eart h i s attacked. I t s distance from the t arget can be very large. On t hi s discussion
i s based the method of area attack with si ngl e propulsion and auxiliary point at the antipodes.
Such a single auxi l i ary point also has the advantage t hat i t can easi l y be fully equipped to
enable ai rcraft also t o make bombing attacks on t hei r journey back to the home base, and that i t s
island location can be easi l y protected against enemy at t acks; against the most dangerous attacks
by enemy fl eet uni t s, t hi s could be done by the rocket bombers.
If such an auxi l i ary point at the antipodes i s not available, area attacks over large
distances can be carried out by having the bomber, aft er rel ease, fly a st rai ght course al l the
way around t he ear t h t i l l i t r eaches t he home base. Thi s i s t he procedure for area at t ack wi t h
s i ngl e propul si on and Ci rcumnavi gat i on of t he gl obe.
Summarizing: a l l pos s i bl e procedures for area at t ack coi nci de in what happens up t o t he
bomb r e l e a s e ; t hey di f f e r onl y i n t he manner of br i ngi ng t he bomber home af t er i t r e l e a s e s i t s
bomb l oad.
2. Poi nt Attack with Two Propul si on Per i ods and Rever sal of Path
I f we denote t he weight of t he bomber in t he s epar at e phases of t h i s at t ack procedure by
G for t he f ul l y loaded bomber weighing 100 t ons , Gj af t er consumption of fuel for t he outward
t r i p , Gj af t er bomb r e l e a s e , and G, t he empty weight of 10 t ons af t er consumption of t he fuel for
the homeward fl i ght , then the i dent i cal distances of outward fl i ght and return are determined
by i dent i cal r at i os Gj/G and G
3
/G
2
according to l i p. 80, while the bomb load li ^dj - O, , Thus we
obtain for each bomb load
G
y^
B
=
:
^ ^ , = W %f * ^ ^ f 2 t f nd ih" rel at i on between bomb load and range
shown in Fig. 82, which also gives the range of t hi s attack procedure, Figs. 91, 92, 93 show
these ranges of attack, for three exhaust speeds G = 3, 4, and 5000 m/sec and with approximate
inclusion of the ear t h' s rot at i on, as contours of equal permissible bomb load, on a map of the
e a r t h ' s sur f ace. The pi c t ur e s show t he ef f ect i venes s nf t he rocket bomber in a very per suasi ve
fashi on. Despi t e t he unfavorabl e double pr opul si on "the bomber. for t he i nt er medi at e exhaust
speed, i s able t o car r y out poi nt a t t a c ks wi t hi n a r adi u* of 2000 km., which i ncl udes a l l t he
s t r ong poi nt s in Lurope between Moscow and Madrid, !\'ort hern bwden and Tr i pol i , I r el and and
Ankara; t hese at t acks can be car r i ed nut wi t h pvtrenie .i'-i-iir.irv r>n iiny nn mat t er how small obj ect
on land or sea, wi t h a bomb load of 3(J t ons, which bnmh triad wi l l .stove in a l l except s pe c i a l l y
rei nforced s t r uc t ur e s wi t hi n 30U m; penet r at e ear t h wort.-: lull m. t.liiek, and s t eel armor 1 meter
t hi ck; and t hen t he bomber Can r et ur n home without a sUipuveri ttith a smal l er bomb load t he Same
bomber can car r y i t s at t acks t o over 6000 km; from Gerrany tn Cent ral Af r i ca, Hi ndust an, East ern
Siberia, the North polar regions, to the east const nT North America and over the whole North
At l ant i c,
For C = 3000 ra/scv, t h i s range shr i nks t o Kurope ,,nd i hf immediately adj acent r egi ons; for
C= 5000 tn/spc i t expands beyond t he heni sphere ni t l i Kiiri.j"- .i* cent er . T li r- p[ .teediire of poi nt
at t ack with two propul si on per i ods t hus appears In havt- ext r aor di nar y praiLir. al importance, and
i s appl i cabl e t n a l l . i d iiin.s i nsi de kurape or in ne if;hlnir j np r ep i ons. In i i .* favor i s t he fj et
t ha t , though tlie bomber iliiriiin ,i poi nt at Lark comes inlit t he cni-mx defense mm- at t he Larget at
low vel oci t y and a l t i t u d e , t he at t ack wi l l n>npr;i1 |v V Mt.-h ., s ur pr i s e that, even a warship
on t he a l e r t wi l l s car cel y l.ave s uf f i ci ent Linn
1
l o l<ot her t he boiiher, much I PRS t o ward off t he
attack.
3. Point Attack with Two Propulsion Periods, Partial Turn and
Auxiliary Point
This nethod of poi nt at t ack di f f e r s from t.h.it of t he precedi ng s ect i on only because t he
r et ur n f l i ght can be s hor t e r than t he outward t r i p from t lie home base t o t he t ar get si nce t he
l andi ng i s t o be made at a s ui t abl e auxi l i ar y l ocat i on ot her than t he home base. Because of t he
smal l ki ne t i c energy over t he t a r ge t , t he angl e of i urn i s uni mport ant ; t he only i mport ant
quant i t y i s t he di st ance from t he t a r ge t t o t he a uxi l i a r y poi nt , measured as a f r act i on It of t he
di s t ance from t akeof f poi nt t o t ar Ret . Vlith t he not at i on of t he precedi ng s ect i on, one has
13S
i
c-e6
Cities of more than
one million.
x Horns base
T
St
Fig. 91r Bomb load in tons {i.e. percentage of the initial fl
Rocket Bomber in the case of point attack - double p
mediate turn around - and with exhaust velocity c =
ire 4a.
L.4S0
Cities of more than
one million
X Home base
Fig. 9?: Bomb load in tons (i.e. percent of the initial flight weight) of
Rocket Bomber in the case of point attack - double propulsion wit
intermediate turn around - and with exhaust velocity o ~ 4000 m/a
'
#
on^million
01
"
8
*****
F l g
"
9 3 : Bomb l o a d i n t o n a
*
1
'
e
* P
e r C8nf c
t
t h
i ni t i a l
Hocket Bomber in the case of point attack - do
X Home base intermediate turn around - and Tilth exhaust Te
* .
as
90
65
60
GO
l
5 70
m
65
60
55
SO
5
40
J5
J0
25
20
IS
10
\
y&\
z m >hlii
s.
a
<*. g I
s s
s e i
S3 3
8 SS
II
s
f s.
Fig. 94: Relationship between attack range a, and return range k. a and bomb load B,
that i s load rat i o, in case of a point attack and a second propulsion at
the knee of a dog l eg path to a secondary base, c = 4000 ra/sec. (see p. 138
sec. 3)
Cities of more than
one million
xHome base
^Secondary base in the
Marianas Islands,
n
^
95
.
B o m b l o a d o f a R o c k e t
Bombar in tons (percent of th
In the ease of a point attack, dog leg path, second
the knee, and a secondary base in the Marianas Isla
c = 4000 a/sec,
90
m
es
Sao
m
a 75

H 70
6S
60
55
50
1*5
*0
35
30
20
15
10
5
r
"
i
!
! !
i
\
^
i
\
\j
,
N
N
X
\
" ^
-*.
*
*
*
k

V
s
t
X
s,
*.
^ .
- . . .
"
V
- j
"
L
" ^ rx
s
x,
J
' -
V
*.
*.
^
S
' r
^
V
V
.
<= ^ m
/
5
o o
Q
l
v\ tt
- 1 <Si
")
1 1
g
^ R s

r>
1
3 R 8f

Attack range a
96: Relationship between attack range a and bomb load B, i . e . load
rati o , i n the case of point attack with double propulsion
and landing at the antipodal point for c = 3000, 4000 and 5000
m/sec.
two char act er i st i c weight-ratios Gwipo
fa
*^ fi*2 *
o r t n e
outward-bound fl i ght and the fl i ght to
the auxiliary point, respectively. Thus to each distance of attack a, and to the corresponding
return distance ka, the required weight-ratios and the possible bomb load B = GJ-GJ can be gotten
from Fig. 80. For k = o, we get the curves of Fig. 80; for k = 1, those of Fig. 82, and for al l
other values of k we get intermediate values, which for the case of c 4000 a/ ,
e c
, are shown in
Fig. 94. Compared t o the previous maximum possible range of attack of 5000 km, we now have
unlimited ranges of attack up to 20000 km. i f an auxiliary point is available suffi ci ent l y near
near the t ar get . Fig. 95 shows the contours of constant bomb-load in using an auxi l i ary point
in the Marianas - i . e. when a landing i s made at the auxi l i ary point aft er the at t ack. Since a
point attack with two propulsions and use of an auxiliary point i s sensible only for those parts
of the ear t h' s surface which are nearer t o the auxiliary point than to the home base, the l argest
ci rcl e, whose points are equidistant from hone base and auxi l i ary point has been marked on Fig.
95, The bomb l i nes are limited to the region of the ear t h' s surface beyond t hi s. The possible
range of attack in t hi s case includes al l of East Asia and a large part of the Western Paci fi c.
For c = 3000 m/sec, the bomb-load curves shrink to small ci rcl es around the auxiliary point; for
C= 5000 m/sec.they spread out over pr act i cal l y the whole hemisphere opposite Europe. Ahen using
an antipodal auxi l i ary poi nt , a + ka = 20000 km, and the possible bomb loads have the values
shown in Fig. 96. In t hi s case bomb loads are possible only for qui t e large ranges of attack -
for C= 4000 m/aec, only beyond 17,800 km, i . e . in a small ci rcl e around the antipodal point with
a 2200 km. radius. Whereas an antipodal auxi l i ary point can be important for area at t acks, i t i s
of value for point at t acks only when the point i t s el f i s t o be protected (say against attacks by
a fl eet ) by rocket bombers from the home fi el d.
As an example of point attack with two propulsions and auxiliary point, an attack on the
locks of the Panama Canal and landing at an auxi l i ary field on the American ftest coast will be
described bri efl y. For C= 4000 m/t, the bomb load i s 2 tons; the char act er i st i c numbers for
the attacking fl i ght are: Takeoff: time 0 sec, weight 100 tons, velocity Orn/sec, al t i t ude 0 km,
distance t ravel l ed 0 km; Climb from takeoff track t o northwest; 11 sec. after takeoff; weight
1Q0 tons, velocity 500 m/sec, al t i t ude 0 km,, distance 3 km. End of the motor less fl i ght : time
36 sec, weight 100 tons, velocity 284 m/sec, al t i t ude 3.7 km, distance t ravel l ed 12 km; End of
Climb Period: time 332 *C, weight 26; tons, velocity 4560 m/aec, al t i t ude 60 km, distance t ravel l ed
5/2 km; End of Supersonic descent: 3682 s e c , weight 26 tons, velocity 300 m/sec., al t i t ude
14 km., distance 9390 km. End of Subsonic Descent: the subsonic descent ends with the s t ar t of
the diving attack; the final al t i t ude i s thus determined by the succrding dive. Since for an
attack on the Canal-locks maximum accuracy of hi t s is more important than high impact velocity,
of the bomb, the end of the subsonic descent i s chosen as 2 km. al t i t ude. From t hi s we get the
other numbers: time 4162 sec, weight 26. tons, velocity 142 m/sec, distance travelled 9450 km;
bad of Dive-attack: t he dive-attack goes from 2*km. t o about 0. 5 km. al t i t ude; the final vel -
ocity of the dive i s about 200 m/aec; then the bombs are released and the ai rcraft goes off with
small loss in vel oci t y, approaching to within negl i gi bl e distances from the ear t h' s surface.
From t hi s we get the values: Time 4172 sec, weight 24 tons, velocity 200 m/sec, al t i t ude 0 km,
distance t ravel l ed 94S0 km;
End of the Second Climb Period: time 4405 sec, weight 10 tons, velocity 2800 m/sec, al t i t ude 22
km., distance t ravel l ed 9710 km;
End of the Second Supersonic Glide-Flight: time 6125 sec, weight 10 tons, velocity 300 m/sec,
al t i t ude 20 km., distance t ravel l ed 12550 km;
End of the Second Subsonic Glide-Flight; time 6785 sec, weight 10 tons, velocity 80" m/sec,
al t i t ude 0 km., distance travelled 12648 km;
Landing: time 6810 sec, weight 10 tons, velocity 0 m/sec, al t i t ude 0 km., distance t ravel l ed
12650 km.
4. Point Attack with Sacrifice of the Bomber
According to the previous considerations the rocket bomber with moderate exhaust speed crn
carry out attacks against small individual t ar get s up to 6000 km. distance from the nome base; i f
t here i s an antipodal auxi l i ary point, the radge extends to 2000 km. from t hi s point; for
arbi t rary auxiliary poi nt s di st ri but ed over the ear t h' s surface, the range i s the same distance
from each of the poi nt s. If however the point attack i s to be directed at a t arget in whose
t i ci ni t y there i s no auxi l i ary point, there i s the possi bi l i t y t hat the bomber proceeds exactly as
described in the previous section, and lands at a corresponding point near the t ar get even thougn
the technical i nst al l at i ons of an auxi l i ary point dp not exi st t here. In t hi s ease i t will no
longer be able to take oi l from t hi s point under i t s own power but will not in general be l ost if
the landing does not occur in enemy t er r i t or y If there i s no other possi bi l i t y than landing in
enemy t er r i t or y, then we are l eft with a l ast , not to be neglected, way out-point attack with
si ngl e propulsion and sacri fi ce of the bomber.
1*5
C-S1296
o*
ua < *IB.-X. U so
W *Q TO ^ o J - W 1 -HO
Cities of more than
one million
+ Antipodal bass point
X Home base
Fig. 97* Bomb load of a Rocket Bomber in tone, that i
in the case of point attack with a single ac
bomber and with exhaust velocity c = 3000 m/
I
c
4
o
u
* o
43 H

a
s
I
43
2
.3
43
*-
o
s
o
e
43
43
flj ff
ID Q
18
W M
(0
43
V
o
o
K
a! tM
O
^ -a
to n
3
X!
43
"8
a
u
J3
43
6 C O
m .H .o
DO
%
3
n
w
43
o
a
1
e
n
flj
I
5
8.
(0
1
147
C-8U296
Since the ai r cr af t gains al t i t ude rapidly aft er bomb rel ease in a point at t ack, the pi l ot
can, at the 'end of t hi s bri ef climb, parachute from the plane and destroy the empty ai r cr af t to
keep i t from get t i ng i nt o the hands of the enemy. He will land a few km. away from the point
of the impact of hi s bombs, and be captured. From the point of view of performance t hi s pro-
cedure represents the limiting css of the point attack procedure described in the pre*ions
section, for which k = o, i . e. the return fl i ght distance i s zero. Figs. 97 and 96 show the
possible bomb-loads for t hi s method of at t ack. Is al l cases, the attacking range covers the
ent i re surface of the eart h; at c c 3000 m/sae. 0.S tons of bombs can s t i l l be carri ed to points
most di st ant from the hone base; for c = 4000 m/aee. t hi s figure increases to 8 tons, and for
c = 5000 m/aec. to 17 tons, This procedure i* nat ural l y also sui t ed to unpiloted use of the
rocket bomber.
5. Area Attack with Full Turn
This f i r s t procedure for area attack corresponds to the f i r s t point attack procedure
described, with the difference t hat the energy supplied at the s t ar t of the fl i ght must suffice
for the outward fl i ght and the ent i re return t r i p to the home base; thus large ki net i c energy i s
present over the t ar get , and considerable fractions of t hi s energy are l ost in turning. An out-
l i ne of the ent i r e fl i ght path including the path at the bomb i s included as a sketch in Fig. 99.
In order t o cal cul at e the rel at i on shown between bomb-load B and range of attack a, one can
proceed as follows: -from IV 3, S-, the length of the subsonic gl i de i s known. "From Fig. 79,
we get the value of V . for (a - S, - *) ; from Fig. 99 we get V, j . Now the length of the climb,
S
3
**iust be estimated by reading off from Fig. 64 a f i r s t value of G/
G(J
for an assumed Vj, and
then get t i ng from t hi s a f i r s t estimate of S3. Then (a - S3 - *) i s given from S5+ (a - 85 -*>
- S
v
The i ni t i a l vel oci t y Vj on the supersonic descending path, in order t o have vel oci t y \'
w
j
aft er a length of gl i de of (a - S3 - *) , can be obtained as described from Fig. 79. This value
i s to be compared with the estimated value and improved, i f necessary. From Fig. 64 we now get
the desired G and B= G - 10. From Fig. 63, the st at i onary al t i t ude h\j for given V^j and G, i s
known, and fi nal l y the range of projection of the bomb, W, corresponding to V j and tl, i s read
from Fig. 67. The range of attack, a, i s thus a = Sg + (a - Sc - *) + *. This calculation con-
t ai ns a few assumptions which should be considered br i ef l y. Fi r st , the assumption i s made t hat
rel at i ons between vel oci t y and distance calculated for part i cul ar fl i ght paths Cab be transferred,
unaltered, to similar fl i ght paths. More important i s the assumption t hat the supersonic descend-
ing path, during the fl i ght before bomb release and during the turn, i s carried out at the st a-
tionary al t i t ude, r at her than in the strongly osci l l at i ng dynamical fl i ght path. This i s necess-
ary tor the decisive t hi r d phase of aiming, in order to at t ai n the necessary aiming accuracy for
the bomb release and in order to release the bombs during horizontal fl i ght . Stationary al t i t udes
of fl i ght are al so necessary for the period of turn in order t o set up the aerodynamic turning-
forces,
This l ast circumstance i s connected with the fact t hat turning i s possible only up t o
definite vel oci t i es of fl i ght below the velocity of points on the ear t h' s surface; for higher
vel oci t i es, other methods of attack must be .used. For t hi s reason the procedure of area attack
with single propulsion and full turn i s limited to the ranges (up to 12000 km.) marked in Fig. 99.
Inside t hi s space i t proves to be extremely effective despite the very cosl.'y turning process.
Figs. 100 and 101 show the l i nes of equal bomb-weight dropped at the t arget by t hi s area
attack procedure, for CB 3000 and 4000 m/ scc For the former value of exhaust speed, the domain
of attack i s bounded by a closed curve which deviates from a ci rcl e because of the ear t h' s rot at i on,
and whose periphery touches the North Pole, Newfoundland, Central Africa, and Central Asia. For
C = 4000 m/sec, the ring expands so that now only Australia, the South polar regions, the South
Pacific and the southern t i p of South America do not l i e within i t . For Cm 5000 m/sec. i t would
cover the whole surface of the eart h.
As an example of area attack with single propulsion and full turn, we use the attack on ftev.
iork at a range of 6500 km. For C 4000 si/sec, the bomb load i s 6 tons, and the detailed attar!.
runs as follows: the motor s t ar t s to work 36 seconds after the take-off at 12 km. distance from
the take-off poi nt , and consumes the t ot al fuel supply of 84 tons in the next 336 sec. At the end
of the climb process, the ai r cr af t reaches a velocity of 6370 01/sec, an al t i t ude of 91 km, a
distance of 736- km. from the pomt of take-off, and a weight of 16 tons. Using only i t s store of
potential and ki net i c energy, the bomber f l i es on t o the point of bomb release, 5550 km. from the
take-off point, and 950 km. in front of the t ar get . At t hi s point, which in reached 1150 sec.
after take-off, the velocity has decreased to 6000 m/sec, &t:A the stationary al t i t ude to 50 km.
After the bomb rel ease the weight i s 10 tons. Then the ai r cr af t goes i nt o a turn :m<i in 330 sec.
goes through a t urn-spi ral 1000 km. in diameter unt i l i t has reached the di rrei i on for the return
fl i ght to the home base. Luring turning, the al t i t ude i s great l y decreased in nrtier to develop
f-SJms
osyfe-tf.
0
**
1
" ti
O
8 J U
S R 8 S
-
x
cs 9 r
!
4
-11-11.
\ \ \
E
\ 1 \ *
\\\ 41
WW/
\ "1 I /
U l l
\ \ \
E
\ 1 \ *
\\\ 41
WW/
\ "1 I /
U l l
\ \ \
E
\ 1 \ *
\\\ 41
WW/
\ "1 I /
U l l
\ \ \
E
\ 1 \ *
\\\ 41
WW/
\ "1 I /
U l l
\ \ \
E
\ 1 \ *
\\\ 41
WW/
\ "1 I /
U l l
\ \ \
E
\ 1 \ *
\\\ 41
WW/
\ "1 I /
U l l
\ \ \
E
\ 1 \ *
\\\ 41
WW/
\ "1 I /
U l l
\ \ \
E
\ 1 \ *
\\\ 41
WW/
\ "1 I /
U l l
/
\ \ \
E
\ 1 \ *
\\\ 41
WW/
\ "1 I /
U l l
/
f
\ \ \
E
\ 1 \ *
\\\ 41
WW/
\ "1 I /
U l l
/
\ \ \
E
\ 1 \ *
\\\ 41
WW/
\ "1 I /
U l l
f
\ \ \
E
\ 1 \ *
\\\ 41
WW/
\ "1 I /
U l l
\ \ \
E
\ 1 \ *
\\\ 41
WW/
\ "1 I /
U l l
\ \ \
E
\ 1 \ *
\\\ 41
WW/
\ "1 I /
U l l
*" ^ r ^
0 S
1 J
a
/
!
Jf
^
ft/
d
^
/
NV

r^r*
y
. ^4.
"/
.A
S
* .
5/
y
>
) /
1
/
^ *
1
/
i
1
OOOOOl
0000s
0000s
OOOOl
00000
0000s
OOOOl E
oooor
00052
OOOOi
OOOQt
OOOZl
ooooi
0006
000s
0001
0009
000s
000*
000c
00a
tax
OOQl
OO&
oaoi
006
008
OOL
009
005
00*
00c
osz
002
oil
OOl
0
o
*rt O
<*>
O J1 C 1ft
U w
&
* <o
K
*-
W 0 *n n
* ^
3 8
^
& *>
suo*} orp p-BOT cpaog
5
n
IS
CO
o
gs
cfl X!
58
>
*5
N-i
03 Vt
at 5
o 9
H
2
,-g *
o c
> t+
_
3

'ai
g *
S
S (DO
P > O
49 C O

3
C C-H
D- H4>
5^g
8:
14-
C-6W296
*H 4
o a
i\
43
V
*
43 d
S3
o.g
is
H
a
fa _.
m o
n
4> O fll
Is**
id j n
H43 H
8
150
43 C
38
*
n t ivc -H0...1. agio to to -HI uo 10 - I M& L - W 160 1*0
#
Cities of more than
one million
X Home base
Fi g. 101: Bcanb l oad of a Rocket Bomber i n tons ( i . e . percent of t he
weight) i n t he case of area at t ack wi t h t ur n around, and a
Tel oci t y c = 4000 m/ sec.
the aerodynamic forces necessary for the turn. At the end of the turn path, the velocity i s
s t i l l 3700 m/sec. and the corresponding stationary al t i t ude i s 38 km. The supersonic glide-path
in the direction of the home base goes over 5450 km, in 2600 sec. and ends 100 km. before the
home base at an al t i t ude of 20 km. and velocity 300 m/scc. Subsonic gl i de and landing are com-
pleted in customary fashion. The whole fl i ght l ast s 4755 sec.
6. Area Attack with Partial Turn and Auxiliary Point
This type of attack corresponds to the point attack with two propulsion periods, par t i al
turn and auxi l i ary point. Like the l at t er , the area attack discussed here represents the most
general case in i t s cl ass, and includes a l l other procedures for area attack as special cases.
The course of an area attack with single propulsion, par t i al turn and auxiliary point
consists essent i al l y of f i r s t giving the bomber, during a single propulsion period, al l the
power which i t requires until landing at the predetermined auxiliary point, having the bomber
release i t s bombs in front of the t arget , at high al t i t ude and fl i ght velocity, as for area
bombing, then carry out a turn at the existing high speed immediately aft er the bomb rel ease,
which takes i t i nt o the direction of the auxi l i ary point at which a landing is contemplated, and
fi nal l y glides with i t s residual energy to t hi s auxiliary point and lands there.
The f i r s t thing t o determine i s the rel at i on between bomb load B and attack-range a. This
rel at i on i s affected by the distance ka between t arget , and auxi l i ary point, the angle of turn If
through which the bomber goes aft er rel ease, and the exhaust speed C. Because of the large num-
ber of independent variables, the r el at i ons are many-sided. For example, Fig. 1012 shows the
rel at i on between a and B for a large number of distances of return fl i ght , ka, and for a defi ni t e
angle of turn, 0~ 60. The procedure of computation which gave these curves was the following:
Assuming a defi ni t e C, a bomb-load B was chosen. To t hi s there corresponds a mass-ratio G/Q *
(10 +
B
J/1QQ, a length S, of climb path, and from Fig. 59 a final vel oci t y V,_ For t hi s maximum
velocity V,, one can now choose various ranges of attack a m S^ + S, + * in such a way t hat before
the bomb release, Ci e. at the end of the outward fl i ght over S^ -> Sn + S3 * S^), a suffi ci ent
supersonic speed s t i l l exi st s. For t hi s velocity and various turn-angles &" , we can, with the
aid of Fig. 81, calculate the loss of speed and the distance S
w
travelled during the turn, and
from the residual velocity finally cal cul at e the " distance of return " from release point to
landing-point. By several repet i t i ons of t hi s procedure, and interpolation between the rough Su
- values found, the ka -curres of Fig. 101! were obtained.
Fig. 103 again shows the l i nes of constant, bomb load for a rocket bomber using t hi s pro-
cedure of attack, i f an auxiliary point on the west coast of America i s used. I t should be
pointed out t hat for turn-angle ff~ = o, there i s a defi ni t e bomb-load with which the bomber
reaches the auxiliary point and can release i t s load at any point enroute without al t er i ng the
range, provided the rel ease occurs after the motor i s turned off. Thus for fl" ^ o the range of
i s anywhere between take-off point and auxiliary point. According t j Fig. 103, t hi s bomb load
i s 5 tons for C = 3000 m/sec. With smaller bomb loads the possible area of attack st ret ches
over al l of North America and considerable portions of South America and the Pacific Ocean.
7. Area Attack with Antipodal Auxiliary Point
A special case of the area attack described in the previous section occurs i f the aux-
i l i ar y point i s at the antipodes of the home base. In t hi s case the turn-angle i s iero for
al l t ar get s. Thus there are no energy losses due to turning at speeds above that of sound.
The r el at i on between bomb-load B and attack-range _a can be obtained from the equations of the
preceding section for 0" = o and a 4- ka * km. For a r ect i l i near fl i ght with t ot al distance
a (1 -f k) * 20000 km, with the present approximations i t does not matter at what place on the
gl i de-fl i ght the bombs are released. Thus, within the 20000 km. range of flight, the range of
attack i s arbi t rary and independent of bomb load. From Fig. 80., the possible bomb load i s 0.7
tons for C 3000 m/sec, B = 8 tons for c = 4000 m/sec, and B - 17 tons for c - 5000 m/sec, '
8. Area Attack with Circumnavigation
Another special case of the general method of area attack discussed in Section VI 6 - the
area attack with si ngl e propulsion period and circumnavigation - resul t s for <F = 0 and a + k
40000 kin. The rel at i on between bomb-load B.and attack-range a_ can be read off from Fig. 80
4
MB
-8*2*4
o$/g-v <vn*j
P
in
> **\ J2
** tir v c cr
a s
c? or
I
" o " ? i
1
S
* . * N* W* M W
QT Q* <tf C- Q- <^
ft
| "
1
i
1
i
ft ) ft ) ft )
i
b
"H
C
o
>
* : _
K5 /
i
b
"H
C
o
>
* : _
sX
i
b
"H
C
o
>
* : _
i
b
"H
C
o
>
* : _
if
i
b
"H
C
o
>
* : _
J
i
b
"H
C
o
>
* : _
J
i
b
"H
C
o
>
* : _
i
*
i
0
i
b
"H
C
o
>
* : _
i
*
i
0
i
b
"H
C
o
>
* : _
i
*
i
0
i
b
"H
C
o
>
* : _
i
*
i
0
i
b
"H
C
o
>
* : _
i
*
i
0
i
b
"H
C
o
>
* : _
i
*
i
0
i
b
"H
C
o
>
* : _
i
*
i
0

* &
i
b
"H
C
o
>
* : _
i
*
i
0

i
b
"H
C
o
>
* : _
i
*
i
0
i
b
"H
C
o
>
* : _
i
*
i
0
i
b
"H
C
o
>
* : _
i
*
i
0
i
b
"H
C
o
>
* : _
i
*
i
0
i
b
"H
C
o
>
* : _
i
*
i
0
6! 6!
/ / /
* * * *
^f "MI
"i^>y/
*
3
6UO) trf po-[ qnog
o >o o vi
$j tv: K o
o
3
s
13 O p < !o
*> I * " *M -
S
ooooot-
00006
QQOOi
000OL
00009
0O00S
MOO*
E
oooos
OOQSZ
oooaz
OOOSi
oooit
QQOOI
000b
0009
0001
0009
ooos
000b
000?
oosz
0002
oost
OOZl
0001
006
008
Oft
009
00$
00*
out
osz
002
051
Oil
001
3

\
5
n a
u h
* h
to
H
1MJS# 4N * 1W
+ Secondary base on American West Coast
onl^Silllon Fig* 103s Bomb load of a Rocket Bomber In tons (
weight) in the case of area attack usin
X Home base American nest coast and an exhaust vel
where the rule again holds that the place of bomb rel ease has no effect on the range, so t hat
the range of attack, a_, i s completely arbi t rary and independent of the bomb load. From t hi s
figure, the l argest bomb load with which circumnavigation is s t i l l possible i s B a 3 tons for
C= 4000 m/sec, B.a* 12 tons for c = 5000 m/sec, while for c - 3000 m/sec, circumnavigation
cannot be achieved even without a bomb load. I hi s method of attack shows most clearly the
extreme technical superi ori t y of the rocket bomber which with a size and empty weight equal
to t hat of a medium mi l i t ary craft can, at moderate exhaust speeds, reach every point on the
ear t h' s surface with a bomb load of 3 tons, and f l i es 40000 km. al l around r.he earth without
an intermediate landing.
As an example of an area attack with single propulsion period and circumnavigation, we
shal l use the attack on the ci t y with a million population most di st ant from Germany - Sidney
in Australia. In t hi s case the range of attack i s 16500 km,, the possible bomb load i s 3 tons.
The flight goes as follows; Take-off and motion aft er take-off do not differ from the same
phases of previous examples. 36 sec. aft er take-off the motor begins tp operate and consumes
the 87 tons of fuel on board in the next 348 sec. At the end of t hi s climbing process the vel-
oci t y i s 7200 m/ s ec, the al t i t ude 101 km., distance from takeoff-point 815 km. and weight 13
tons. This very high i ni t i a l speed drops to 300 m/sec. in the course of the supersonic descent
which i s 39185 km. long. After a 10000 km. journey, the strongly osci l l at i ng descent must be
damped suff i ci ent l y' so t hat at the release point, 15400 km. from take-off, i t runs smoothly enough
at the st at i onary al t i t ude to enable accurate aimiflg for the bomb rel ease. At the release point
the al t i t ude i s 49 km., the velocity i s 6400 m/ s ec, and the range of projection of the bombs i s
1100 km. After release the bomber s t ar t s i t s supersonic glide with only 10 tons weight, during
which the course which was previously in a plane has to he al t ered sl i ght l y in order to lead
back to the home base. There the bomber lands 13060 seconds aft er take-off, having t ravel l ed
40,000 km.
9. Evaluation of Procedures for Attack
Procedures of point attack are directed against individual houses railroad st at i ons and
tracks, tunnel entrances, st r eet s, bridges, dams, single ships, canals, dikes, breakwaters, gas-
water-and oi l -t anks, munitions depots, magazines, power st at i ons, transformer st at i ons, ai r -
dromes, harbors, fact ori es, troop concentrations, et c; they are limited t o a radius of several
thousand km. around the home base, except for special cases where the bomber i s sacrificed or
f l i es on to an auxi l i ary fi el d, when the range of attack can extend over the whole of the ear t h' s
surface.
Procedures of area attack can be directed against the ent i re ear t h' s surface. The prob-
able scat t er of bombs over several kilometers limits them t o t arget areas of t hi s magnitude, e.g.
ci t i es with over a million population, large i ndust ri es, fl eet s, et c. I f in an area at t ack, the
t ot al energy Z in kcal i s released against a single t arget with a probable scat t er *_ = 2 km.,
then half the hi t s l i e in a ci rcl e of 2 km, radius; the average density of hi t s on t hi s uni t
surface i s
z
~^/Zw
r
W the actual density follows a Gauss error curve
Z.f=/3$8z. &~
A
*W
tr
^' has the value 1.4 5 at the center, and is % of t hi s at
the boundary of the area 4 km. in diameter. Fig. 104 shows such a di st ri but i on curve of bomb
hi t s over a map of New York. If larger connected surfaces than the unit surface described are
to be attacked, several points of the t arget can be aimed at, so t hat the individual Gauss error
curves par t i al l y overlap somewhat l i ke those of Fig. 105, where the distance between aiming-points
was chosen as Mfc/ffW so that the average density over the whole surface i s X, while the
local densi t i es are shown as contour l i nes.
From the appropriate l i t er at ur e the following rel at i ons can be gotten between average
density of destructive energy z in kcal/^^2 and the resul t ant dest ruct i ve effect; 1 = 7j(10<
i
kcal.
pUts i ndust ri al i nst al l at i ons completely out of operation for several days. kHz
(Degree of destruction 1)
z = 1.4X10" "THJ^r destroys ci t i es so that al l except speci al l y reinforced buildings collapse,
and only cel l ar s and foundations are usable (Degree of destruction II)
I = 1.4X10
9
tyjj$ destroys ci t i es so that cel l ar s are also smashed in, al l people inside the
area are ki l l ed and only foundation wells remain standing (Degree of destruction III)
1 = 7X10
9
^f makes ci t i es so fl at that t hei r location i s no longer discernible against the
background. (Degree of destruction IV).
I f the energy content of the bomb at rest i s assumed to be fc^i- 700 ^f^-, then with E =
15b
Zone of the second
order destruction
Bounding of \
the unit area ,
-Probable dispersion 2
r
of the-j
surface attack j
Maximum i ntensi ty of .
^coverage z
0
" l ^xKrkcal / km
2
1
Intensity of coverage
' = l . /
(
xl 0
8
kcal / kg
Intensity
of/coveragf
7xl0kcal/km
Hlefagsdichte UWO%Jj.
Second order
destruction
Target
: order deatructi
156
Fig. 104i Hit diatribution in area attack against a target point in the center
of New Iork.
distances
probable dispersion
Target
point
v
v
I
(
Target ;
' point ;
V
4>
.-**
^Target i
point
_/
105: Target distribution in the case of attack on very large surfaces
and contour lines of equal bomb hitting density in percent of average
value
157
C-8 479 6
N 800
N
g 700
* 600
0 SBO 1000 1500 2000 2500 30003500 WOO 4507 5000 5500 6000 6500 7000 7500 8000
Striking velocity in J^sac*
Fi g. 106: Required bomb hi t densi t y t o produce f i r s t t o fourth order dest ruct i on
as a function of the st r i ki ng vel oci t y of the boob.
158
r-BJJw.
Attack rang
Fig, 107: Number of required sorties to cover the unit area and ita sur
using various methods of attack, with a total of 3.52 x 109 k
destructive energy, aa a function of the attack range a.
700-+Av
2
/2g, the density of bomb hi t s in f ons/ ^2, independent of t-he imf.aci velocity V nf the
bomb i s obtained from Fig. 10fr.
In order to destroy, to the 2
n d
degree, the surface of a ci t y having an area e<(ual to
the previous unit surface, one must release against the t arget Z = 2X12.56 X 1-4M.0
8
= 3.52X10^
kcal of dest ruct i ve energy, corresponding to 5000-420 tons of bombs depending on the velocity of
impact. Then the concentration, as shown in Fig. 104, ri ses to 1.96X10 "fcgj near the aiming-
poi nt s, i s 0.98**10
B
^fj^A at the edge of the unit ci r cl e, and 1.14X10* ^ j * " at 4 km. distance
from the aiming poi nt .
The number of f l i ght s necessary for dropping t hi s amount, Z, of dest ruct i ve energy on the
unit surface i s 84 i f the bomb load i s 60 tons and the impact velocity i s low, or 420 i f we
assume the smallest bomb load - 1 ton - , and an impact velocity of 8000 m/sec. Fig. 107 shows
plot of the number of attacks necessary against the uni t surface, for various procedures of
attack, against range of attack, for c
9
4000 m/sec. If Che unit surface i s to be attacked with
the minimum number of f l i ght s, then the procedure of point attack with double propulsion and
full turn i s best . If a single unit surface i s to be bombed, t hi s superi ori t y i s doubled,
becauae then al l the energy l i es inside the unit ci r cl e, and only half as many bombs need t o be
dropped. For great er ranges of attack up to 8000 km., the procedure of area attack with single
propulsion and full turn i s far superior to al l other procedures, especi al l y since i t does not
depend on the use of an auxi l i ary poi nt . A remarkable thing about the curve for t hi s procedure
i s t hat the required number of fl i ght s does not increase mono onically with the range of attack,
but rather that the decreasing bomb load i s completely compensated by the increasing impact
energy. Corresponding t o the ful l curve, the number of fl i ght s actually required when using
area attack with full turn or circumnavigation fluctuates between 64 at 1000 km. and 322 at
20000 km. range of at t ack. The evaluation of attack procedures shown in ig. 107 assumes that
the t ot al consumption of take-off fuel, fuel for the ai r cr af t , and of bombs, which together
represent a constant amount of 133.7 tons per fl i ght , shall be a minimum. Since the bombs are
much more valuable per uni t weight than the fuel, one can also set a requirement of minimum
consumption of bombs. In Fig. 108 several procedures of attack are plotted from t hi s viewpoint;
we see^nere the point attack procedures are very i nferi or while^attack procedures, which operate
with high impact vel oci t i es of the'bombs, are most favorable, especially, area attack with ci r -
cumnavigation and - at long ranges-, with full turn, which requires in t hi s region the l east
t ot al consumption as well as l east consumption of bombs.
Fig. 109 shows an idealized di st ri but i on of hi t s , according to the laws Af probabi l i t y,
over a ci t y map of Berl i n, where i t i s assumed that 84 bombs of 60 tons weight are drepped on
the aiming point with low impact velocity; the half shown l i es in the 50% ci rcl e; about each
point of impact a ci r cl e of destruction of diameter 618 m r esul t s in which the energy density
1.4K10
8
kcal / ^2 required for degree II exi st s,
Fig. 110 shows the corresponding di st ri but i on of hi t s for 140 releases with 8000 m/sec.
impact velocity and 3 tons weight per release; again the average energy in the unit ci r cl e i s
1.4X10" kcal/. 2 but the area of destruction for the same energy per release is. now drop-
shaped and includes 180000 n?, as previously derived.
lbo
-8*3
Vn. THE LINE OF DEVELOPMENT OF THE ROCKET BOMBER
The development of the rocket bomber project wi l l follow roughly the sequence of 12 stages
outlined below:
1. Development of the Combustion Chamber and Jet of the Motor
The main problems in t hi s stage concern ttit introduction of the sol i d, liquid, or already
vaporiied fuel and the combustion-maintaining; mate-rial, i nt o a combustion chamber through in-
j ect i on nozzles; then the rapid di st ri but i on, mixing, heating and i gni t i on of the fuel, i t s most
complete combustion at more or l ess constant high pressure to a combustion gas at very high tem-
perature; then the expansion of these gases in a j e t to convert them to a beam of ess with as high
streaming velocity and as low temperature as possible. The very high pressures and temperatures
in the combustion space have the consequence that not only the tubes for the streaming process in
the motor, but al so the construction of al l walls in contact with the flame, becomes a very s er i -
ous problem, whose solution as regards choice of mat eri al s, methods of cooling, and construction-
al arrangement, should be a main point of study. Also important are questions of shape and r e l -
ative size of combustion chamber and j e t , choice of most sui t abl e flame pressures, measurement
of stream temperature and velocity, arrangements for rapid heating and mixing of the fuel, optical
and acoustic phenomena, mixing of the j e t with the surrounding ai r behind the motor, di ssoci at i on-
and detonation-problems, and numberless others.
2. Development of Special Fuels for Rocket Motors
In many respects qtlite different requirements are to be set for rocket motor fuels then
for the fuels of ordinary ai rcraft motors. In the f i r s t place, what counts i s the available
energy content per unit mass of the combustion mixture of fuel and, say, oxygen, and not the
heating value of the fuel alone Thus a combustible material which has a lower heat of combustion
than the usual hydrocarbons, but consumes much less oxygen in burning, can develop a far superior
heat output of the mixture. In additiuu to the heat output of the mixture, otner combustion
charact eri st i cs such as ease of i gni t i on, rat e of combustion, tendency to detonate, degree of
dissociation, st at e of aggregation of the combustion products, reaction temperatures, e t c . , are
important. One must also consider properties not so di r ect l y connected with the combustion, such
as procurement and cost , ease of storage in tanks on the ai r cr af t , density, danger, ease of feed-
ing of the fuel, et c. If one enumerates the problems of atomic hydrogen end nitrogen, of nuclear
reactions and of takeoff fuels, then one has outlined in broad strokes the scope of fuel research
which should lead to the development of new and more sui t abl e rocket fuels.
3. Development of the Auxiliary Engines of the Rocket Bomber
Just as with ordinary ai rcraft motors, the rocket motor requires for i t s operation a few
auxiliary engines, of which the most important are those for feeding fuel and coolant and the
associated driving assemblies. These additional i nst al l at i ons present some not-too-simple prob-
lems, since the feeding rat es are very high, of order of magnitude 50-100 H.P. per ton of t hrust ,
and the material to be conveyed can be in altogether unusual s t at es , say a liquefied gas, a me-
t a l l i c suspension or even solid or liquid metal - and must be very accurately proportioned as
well as being fed against very high pressures. A complication ari ses because the feeding i nst al l a-
tions must be designed under extreme limitations on weight of construction. In addition to these
arrangements for feeding of fuel and coolant, ignition systems and in some cases intake and reg-
ulator systems require consideration.
4. Development of Test Model of Complete Rocket Motor
Even i f the development of the previously enumerated most important parts of the motor had
been achieved taking account of t hei r i nt eract i ons, putting them together into a ready-to-fly
rocket motor and examining t hei r interplay i s s t i l l a separate and important step. Only now, on
the apparatus ready for flight can a bench study be made of the mutual i nt eract i ons of the com-
bustion chamber, j ^ f
ue
i
ari(
j auxi l i ary engines, so t hst by sui t abl e adjustments the best r esul t s
are obtained for exhaust speed, r el i abi l i t y of performance and construction weight- These bench
t est s of the complete rocket motor are especi al l y important and thorough because they reproduce
the conditions during fl i ght very closely; t hi s is in contrast to the ordinary ai rcraft motor
where these conditions can be imitated only with di ffi cul t y and not even completely by al t i t ude
t es t s . This i s mainly connected with the fact that the rocket motor accomplishes the j e t forma-
tion, combustion and cooling only with i t s own fuels, and without use of the surrounding atmos-
phere, SO that the differences of velocity, temperature and pressure of surrounding ai r between
the bench t est and actual fl i ght can scarcely al t er the conclusions. One of the few places where
the rocket motor comes in contact with the surrounding ai r i s the mouth of the j e t . There the
161
tU.:-
Range a in
Fig. 103: Amount of required bombs in tons in order to cover the unit surface
and the environment by various methods of attack with a total of
3.5? x 10' kcal destructive energy as a function of attack distance a.
?
?
[inn
Fig- 109: Ideal Gaussian hit distribution of 42 hits, each of 60 tons of
bombs, with small striking velocity in the unit circle in case
of surface release against the target, and with area of destruction
for each bomb (for example, bombardment of Berlin).
163
C-8 tt 296
Fig. 110: Ideal Gaussian hit distribution of 70 hits, each of 3 tons of
bombs, with small striking Telocity In the unit circle in case of
surface release against the target, and with area of destruction
for each boob (example, bombardment of Berlin).
4 H.M
combustion gas expands t o the pressure of the ai r behind the ai r cr af t ; i n f l i ght t hi s pressure
i s pract i cal l y zero whereas in the open bench t est the j e t can expand only to the pressure of
the s t i l l ai r around i t . St i l l the differences ere not very great, are l argel y amenable t o cal
culation, and are moreover such that the bench t es t considers the more unfavorable case. A r e-
action of the slip-stream and the j et on the ai r cr af t (in par t i cul ar on the control surfaces)
need not be considered for the rocket bomber, because operation of the motor and subsonic veloc-
i t i es of fl i ght never occur t oget her. Aside from the purely developmental and t eat st udi es on
the model, a whole seri es of physical researches on the rocket motor can be carried out most
advantageously during t hi s phase of the development: more careful, investigation of atomizing oi
the fuel) the actual behavior of the pressure in the combustion chamber, the temperature di s-
t ri but i on in the combustion -gas, the actual streaming speeds, the conditions of heat transfer
from the combustion gas t o the walls and from them to the coolant, the j et formation and sound
emission in the furnace-jet, as well as numerous otner questions. Only when in the course of
these studies the motor has proven i t s complete r el i abi l i t y of performance, can one consider in-
s t al l i ng i t in an ai r cr af t . The development of t hi s ai rcraft will have to be done alongside t hat
of the motor, so t hat both may be ready for flight at the sane time.
5. Wind-Tunnel and Tow-Tests on Models oi tne Air-Jrrame
The external shape of the rocket bomber in determined by the requirements t hat exi st i ng
ixperience on screw-propelled airfareft shall be used as much as possible, that the special con-
di t i ons of supersonic fl i ght and rear i nst al l at i on of the motor must be considered, t hat the
glide-number of tbe ai r cr af t shsll be as good as possible at very high Mach numbers, where the
aerodynamic forces are proportional to .the square of the velocity and angle of at t ack. Under
these general requirements models of the rocket bomber can be designed; these cannot be tested
in the most i nt erest i ng region of very high Mach numbers, in the upper part of which actual* chem-
i cal changes of the streaming medium are possible, because as yet such ar t i f i ci al ai r - bl ast s are
not available; one can perform wind-tunnel t es t s in the range of Mach numbers from 0.08 to 4.
In the range of vel oci t i es below that of sound, which i s important for the landing process, using
models of reasonable sine one can carry out measurements on the aerodynamic forces, and especi al l y
on the maximum l i f t of the ai r cr af t , aad i t s improvement by the use of aerodynamic aids suited t o
the part i cul ar pr of i l e. In t hi s range t unnel -t est s on s t abi l i t y, vi brat i on conditions and manoeu-
vr abi l i t y of the vehicle are important. In the supersonic domain, aerodynamic forces, di st ri bu-
tion of "
r
"-assure, s t abi l i t y and manoeuvrability, vibration conditions, ai r temperatures at
the stagnation poi nt s, and heat transfer to the ai r cr af t are of i nt er est . A whole seri es of these
t est s on very small models can be carried out in a supersonic wind tunnel.
For t es t s on l arger models the takeoff track of the rocket bomber i s i t s el f a very conven-
ient towing-track; on i t even very large models can be towed up to any desired high speed with the
aid of rocket propulsion,, and then studied. Such tow t es t s can be carried out st reasonable cost
on a 15-30 km. long track, with models scaled 1:10 up t o 1:1 at the ori gi nal speeds of 800 up to
5000 km/hr, so that the Mach number i s duplicated exactly and the Beynolds number approximately,
Such t est s would especi al l y consider the general behavior of the ai rcraft in the neighborhood of
the velocity of sound, i . e . between 800 and 1800 lan/br, for various shapes, angles of attack,
arrangements and t i l t s of control surfaces, e t c . ; also the di st ri but i on of ai r pressure and tem-
perature on the surface of the ai rcraft by means of t aps, thermoelements, e t c . , for al l vel oci t i es;
measurement of the location and magnitude of the forces' act i ng un une ai r cr af t , di r ect t es t s of
St abi l i t y by means of a Cardan suspension of the moving model at i t s center of gravi t y, st udi es of
the reaction of the j et on these conditions especi al l y on the aerodynamic forces and the s t abi l i t y
investigation of the el as t i c properties on el ast i cal l y- si mi l ar models; e. g. as regards vi brat i on
of the wings and control surfaces, et c. ; takeoff and free-fl i ght char act er i st i cs of sel f - st abl e or
remotely controlled models especially during the passage of the vel oci t y from values below the
sound velocity to values above, and others. K special problem in such t ow-t est s, aside from the
track i t sel f , i s the r el at i ve sl i di ng behavior of lubricated surfaces at sl i di ng speeds up to
5000 km/hr. For fl i ght at arbi t rary speeds in regions of very rare atmosphere, in which the so-
called gas-kinetic laws are val i d, and where nei t her tow t est s or ordinary wind-tunnel t est s are
feasible, an extension of the theory i s desi rabl e, as well as i t s experimental j ust i f i cat i on, say
by extension of the well-known moleculer-beam methods to the present problem.
6, Constructional Development of the Vehicle
Once a preliminary decision as to the whole arrangement and shape of the rocket bomber has
been made on the basi s of wind-tunnel- and t ow-t est s, one can commence the construction of the
fuselage, wings, control surfaces, and important i nst al l at i ons such as the pi l ot ' s cabin, the
tank i nst al l at i ons and auxi l i ary equipment. In view of the fixed external shape, the decisive
factor foi the construction of the fuselage, wings ind control surfaces i s the fact the ai r pres-
sures which occur are far greater than those on ordinary ai r cr af t . The ai r pressures on the rock-
et bomber will cert ai nl y be in the neighborhood of 3000 kc/m , and will therefore lead t o types
1.&5
c-*e
of wing construction like those used i n buildings or ships. The problem of the pressurized cabin
has already been treated i n other connections. The r esul t s assembled there can be made use of.
There are no prototypes on which to base tank i nst al l at i ons for large quant i t i es of fuels l i ke
liquefied gases, metallic suspensions, e t c . , and these must be suited to the special conditions.
An especially extensive group of apparatus requiring development i s the auxiliary equipment of
the rocket bomber, including vel oci t y- , al t i t ude- , and acceleration-meters, instruments for st eer -
ing, navigation and aiming of bombs, opt i cal instruments and many other apparatuses and instruments
connected with the special char act er i st i cs of rocket fl i ght .
7. Bench Tests on Interaction of Motor and Air-Frame
These t est s represent the l as t stage before the f i r s t fl i ght t est s, and should check the
sat i sfact ory operation of t he motor (which has so far only been bench-testedj under the condi-
tions where all the par t s have been assembled i nt o a vehi cl e. The exact rel at i on between the
driving force and the cent ers of mess and ai r pressure of the ai rcraft should be checked to
avoid i nst abi l i t i es caused by i ncorrect placement of the engine; and finally during these t est s
the pi l ot can become accustomed to some of the pecul i ar i t i es of the new ai r cr af t .
8. Development and Test of the Takeoff Arrangement
The several kilometers long tBkeoff path with i t s arrangement of r ai l s represents a r e l -
at i vel y simple engineering construction which presents no special problems. On the other hand,
the takeoff sled, which under a load of several hundred tons must be accelerated in a very short
time to 1 1/2 times the vel oci t y of sound and then slowed down even more rapidly, requi res speci al
development. This wi l l be concerned, especially with the ext raordi nari l y powerful takeoff rocket,
the sl i di ng contacts of the sled and the method of braking. Completely rel i abl e operation of the
takeoff track will have to be carefully tested by catapult t r i a l s on dead weights comparable t o
the weight of the ai r cr af t .
9. Takeoff and Landing Tests on the Bomber
The takeoff t est s begin with small fuel load on a t axi - s t r i p which i s as long as possible
the pract i cal l y empty bomber, .after very brief operation of i t s . own rocket motor obtains the velo-
city required to float, aft er which i t r ol l s on i t s own landing gear. After a short hop i t drops
down again and carri es out i t s f i r s t landing, which does not di ffer from the later landings aft er
long f l i ght s. These takeoff t est s should be carried out in such a large space that the landing
can occur immediately aft er the takeoff without turning the ai r cr af t , in order not to endanger
the ai rcraft by manoeuvers near the ground when i t s fl i ght char act er i st i cs are s t i l l uncertain.
If after many such t es t s , the takeoff and landing char act er i st i cs are understood, then by using
more fuel the ai rcraft can be brought to somewhat higher al t i t udes, sey a few hundred meters, and
then glided back to eart h. Thus the essent i al fl i ght char act er i st i cs at low speed will be de-
termined, the arrangement of the control surfaces will be undertaken, the trimming, s t abi l i t y and
manoeuvrability will be checked, and thus the airworthiness of the craft in al l fl i ght at t i t udes
at not too high al t i t udes wi l l be determined and eventually improved. If, during these t est s the.
ai rcraft acts rel i abl y for the pi l ot , then the takeoff t est s on the track can be repeated, by
placing the empty ai rcraft on the takeoff sled, catapulting i t and landing i t . If these t est s are
satisfactory the t r i a l s can be extended to higher fl i ght speeds. This phase of the t r i a l s wi l l
best begin by ten; t est s on a very lonR tow path, so that the ai r cr af t doesn' t get to fly but i s
only accelerated on the towing path and then immediately slowed down again. These tow t est s on
the actual ai rcraft at gl i di ng speeds of 800 t o 1800 km/h pre t o be devoted especially to the
following individual problems: rechecking and refinement of the resul t s of the tow t es t s on mod-
el s concerning pressures, temperatures, aerodynamic forces, s t abi l i t y, vibrations, effect s of the
driving j e t , et c. ; studies of the behavior of instruments, apparatus snd j e t engines under the
influence of high accelerations and vel oci t i es, development of sui t abl e safety devices, e. g. spe-
ci al seat s, get t i ng the pi l ot accustomed to the new phenomena of high accelerations, high vel oci -
t i es , special engines and new arrangements of the ai r cr af t ,
10. Flight Tests of the Bomber
Finally one can go over t o fl i ght t est s in which one cat apul t s the ai rcraft with an i n i t i -
ally small but gradually increasing supply of fuel, and l et s the motor operate for longer and
longer periods. Thus longer and longer climb paths will be traversed; one will soon surpass the
vel oci t i es attained up tb now by the fast est ai r cr af t , and go up to al t i t udes never previously
reached. The takeoff speeds also increase at the same time. Now begins the most di f f i cul t part
of the fl i ght st udi es, since the fl i ght conditions depart fart her and farther from any known at
present and checked by experience; completely new aeronautical t er r i t or y must be conquered. As
far as can be visualised at present, the fl i ght t est s wi l l extend to maintenance of requirement!
of l i f e in the pi l ot ' s space taking account of the high al t i t udes, accelerations, and tempera"
t urcs, the effect of the aerodynamic forces on the wings and controls of the ai r cr af t , t est i ng of
the heating of the walls at many points of the ai r cr af t , maintenance of st abi l i t y and manoeuvrabil-
i t y beyond the velocity of sound, etc- This part of the fl i ght t est s will have to be done very
circumspectly and with stepwise increase in velocity, a nice at very high al t i t udes, , low ai r dens-
i t i es and high speeds, any sl i ght accident can lead to a catastrophe since i t is pract i cal l y im-
possible t o leeve the ai r cr af t by parachute. These t est s of fl i ght char act er i st i cs will be con-
tinued up to speeds not yet reached by pr oj ect i l es, i . e. about 6 times the velocity of sound.
Finally the actual t r i a l fl i ght s will be completed with performance t es t s . They will serve f i r s t
to determine simple performance data such as takeoff speeds, takeoff di st ances, landing di st ances,
landing speeds, climbing performance, consumption of fuel, et c. , and l at er to determine the maxi-
mum speed, cei l i ng and range of the rocket bomber. Since theae three limiting performance figures
are interdependent, they can be determined during the same fl i ght . As the velocity i s increased
from the values attained during the t est s of fl i ght char act er i st i cs up to the maximum value, the
temperature di st ri but i on on al l windward surfaces will have to be carefully checked by means of
a remote-reading thermometer system, in order to catch in time any dangerous heating as a r esul t
of ai r fri ct i on and stagnation. Too great heating, especially of proj ect i ng par t s, such as the
sharp end of the fuselage, the sharp wing edge, et c. , i s most dangerous because the sl i ght est
melting or other deformation of these carefully shaped cr i t i cal points and t hei r consequent bl unt -
ing leads to instantaneous enormous increase of the stagnation temperature at those places and
then in steadily- spreading molten regions, and would r esul t in immediate burning up of he whole
ai r cr af t . If the performance fl i ght s are carried out unt i l these expected limiting values of
7000 nt/sec velocity, 150*altitude and 40000 km. length of fl i ght are reached, then the fl i ght
studies can be considered completed.
11. Navigation Tests on the Rocket Bomber
The next important task of development concerns arrangements for navigation, which, aft er
propulsion has ceased, give the rocket craft exact knowledge of i t s path, enable corrections of
i t s course, and permit an exact calculation of the moment of bomb rel ease. This precision naviga-
tion will have to be checked in a very large number of fl i ght t es t s , since the success of any
attack depends on i t s accurate and rapid performance.
12. Bomb Release Trials
These const i t ut e the l as t phase of the research end development work on the rocket bomb, r
and should give a pract i cal veri fi cat i on of the preliminary t heoret i cal work on the processes
occurring during the fal l of the bombs and t hei r contact with the eart h. So far as the rel at i on
between point of impact and point of release from the ai rcraft for various heights and speeds of
fl i ght i s concerned, the oceans of the whole earch-.provide r sui t abl y l?rge t est i ng press. The
study of impact of the bomb on land will be somewhpt more troublesome, since very large uninhabited
areas are required. However a few t est s in the Arctic regions untrodden deserts or in our own
possessions will suffice for t hi s purpose.
167
c-e*nt
BI BLI OGRAPHY
(1) C r a n i , C. : Lehrbuch der Ba l l i s t i k . Spr i nger , Ber l i n.
(2) D i r i n g s h o f e n , H. T . : Bi s ?u wel cher St ar ke kann der Mensch i n F.'ugzeug Zen-
t r i f uga l kr a f t e ver t r egen und welchen Ei nf l us s hat hi er auf
di e Anderung der K&rperhaltung. Jahrbuch 1937 der Deut-
schen Luf t f ahr t f or achung, Aosgabe Ausrust ung, S. 128.
(3) F 1 u g g e, S. : Kann der Ener gi ei nhal t der Atomkerne t echni sch nut i bar
genacht werden. Nat ur wi ssenschaf t en 27, S. 402, 1939.
(4) F r e n k c 1, I . : Theor i e der Adsorption und verwandt er Erschei nongen, Zs.
f, P- si k 26, S. 117, 1924.
( 5) G e i b, K. H. : At emreakt i onen. Er gebni sse der exakt en Nat urwi ssenschaf-
t en Bd. 15, S. 44, 1936.
(6) G l i w i t i k y , \K,; Messung des Druckverl aufs bei -Al unt i ni umst subexpl osi enen.
Zs. d. VDI Bd. 60, S. 689. 1936.
(7) G ii n fa e 1, L. und E v e r - Reibung und Schmieruog ira Maschinenbau. Be r l i n 1925,
l i n g , E. : Krayu.
(8) J 0 a t : Expl osi ons- und VerbrennungsTorgSnge' i n Gasen. Ber l i n
1939, Spr i nger .
(9) K n e s e r , H. O. : Der ekt i ve St i cks t of f . Er gebni s s e' der exakt en Nat ur wi ss-
enschaft en. Bd. 8; S. 229, 1929.
(10) k n* e s e r , H. O. : Die skust i schen Rel axBt i onserschei nungen. Phys. Zs. 39,
S. 800, 1938.
(11) L m d o l t - B o i m t t i n ; Physi kal i sen-chemi set t e Tabel l en, Ber l i n 1923 - 1936,
Spr i nger .
(12) R a I k c, O. F. : Kr ei sl auf unt er Beschl euni gung; Versuche uber di e Wir-
kung der Beschleunigung auf der Zent r i f uge an Mensch und
Ti e r . Jabrbuch 1937 der Deutschen Luf t f shr t f or schung,
Ausgabe Ausrust ung, S. 123.
(13) R o t h, a. A. , l o 1 f, U. und: Die Bildungswanne von Aluminiumoxyd und LanthanoVyd. Zs.
F r i t i , 0. f. Elektrochem. 46. S. 42, 1940.
(14) R u f f, 0. und K o n s c h a k , VerdampfungswSrme des AI2O3. Zs. f. Elektroehem. 32,
M-: S. 32, S. 518, 1926.
(15) R u f f ) S. : Die physiolc-gische nirkung hoher Beschleunigungen. Luft-
wi ssen 7, 5. 24, 1940.
(16) S a n g e r , E. : Uber Bsu und Lei st ungen der Rsket enf l ugi euge. DeutschS-
s t e r r . Tageszei t ung, Folge 35, S. 11 und Fol ge 38, S.
9, 1933.
(17) S a n g e r , E. : Der Entwieklungsweg der Hsket enfl ugcechni k. Zs. Fl ug,
Heft 5 - 10, Wien 1933, Pi t t n e r .
(18) S a n g e r , E. ; Raket enf l ugt ecbni k. Miinchen 1933, Oldenbourg.
(19) S 8 n g e r , E. : Neuere Er gebni sse der Raket enf l ugt echni k. Zs. Fl ug,
Sonderheft 1, Wien 1934, Pi t t n e r .
(20) S a n g e r , E. : Der Rsket ensnt r i eb fur Fl ugzeuge. Der Pi l o t , Heft 1, S.
5, 1935.
(21) S a n g e r , E. : Raketenmotor und "Verf ahren i n sei nem Bet r i eb. Oe. P. 144
S09, ttieff 1935.
168
C-84
(22) S a n g e r , E. : Expansxonsdnse mit- geki l hl t er Wand fur di e Feoergase von Raketenan-
t r i ebsmaschi nen. Oe. P. 146 000, Wien 1935.
(23) S a n g e r , E. : Der Verbrennungs-Raketenmotor. Schweizer Bauzei t ung, Bd. 107.
Nr. 2, S. 1, 1936.
(21) S i n g e r , E. : Kurze Dar s t el l ung der Reket ent echni k. Deutsche Luf t f ahr t f or -
schung, Ber i cht 828, Ber l i n 1936.
(25) S a n g e r , E. : Hohe Auapuffgeschwindigkeiten beim Rnket enant t i eb, Deutsche
Luf t f ahr t f or schung, Ber i cht 829, Ber l i n 1936.
(26) S a n g e r , E. : Rake tenmo t or . D. P. 44/ 39, Ber l i n 1937.
(27) S i n g e r , E. : Verfahren zum Bet r i eb ei nes Raketenmotors mi t Dampfkraftmaschinen-
MHs e n t r i e b . - D.P. 380/40, Ber l i n 1938.
(28) S i n g e r , E. : Gflskinetik sehr hoher Fl uggeschwi ndi gkei t en. Deutsche Luft fahi v-
forschung, Ber i cht 972, Ber l i n 1938.
(29) S a n g e r , E. : Gl ei t kor pev fur aehr hohe Fl uggeaehwi ndi gkei t en. D.P. 411/ 42,
Ber l i n 1939.
(30) S a n g e r , E. : Verfahren zum Bet r i eb ei nes Raketenmotors mi t Sonder br ennst of f t n.
D. P. a. S. 143 284, Ber l i n 1940
(31) S c h u b e r t , G. : Physi ol ogi c des Menschen im Fl ' gzeug. Ber l i n 1935, Spr i nger .
(32) S e f e ' r i a n , M.D.: Flamme d' hydrfigen* at oai rui e; r echer ches sur hi dlr5^tJv.u^xt-ii Lies
gs i dans l ' s r c . Chal eur et I ndus t r i e 19, S. 80, 1938.
(33) Z e n n e c k, J . ; Physik aa hohen Atmosphere. Er gebni sse der kosmischen Physik '
Bd. 3, S. 1, 1938.
169
C-BU296
TABLE OF IDS! L*.TOETAHT SYMBOLS
a* ..attacking distance (m, Ion), sound velocity (m/sec)
a critical sound velocity (for example, in nozzle throat)
b acceleration (m/sec )
b normal acceleration (m/sec )
b^....., tangential acceleration (m/sec )
c .....effective exhaust velocity (m/sec)
c
jnax
..........raaximo flaw velocity of fire gases (m/sec)
Oj-jj........... theoretical maximum flow velocity of fire gases (m/sec)
(^..........fire gas (exhaust) velocity at the mouth (of nozzle) (m/sec)
eg .probable velocity of the air molecules before they strike
rail (m/sec) a
cp
t
. . . . probable diffuse rebound velocities of the ai r rxilecules
from a Trail (m/sec)
c
a
lift coefficient (0^)
Cao initial lift coefficient ( C ^ )
Camax maxiaum lift coefficient
Cgp lift coefficient of Tings
c^ drag coefficient (C)
c frictional component of CTJ, drag coeff.
Cj> surface drag coefficient
c_ .....specific heat at constant volume (kcal/kg)
c.,,,,,. specific heat at constant volume, portion due to molecular
vtrans translation (kcal/kg)
c.,+ specific heat at constant volume, portion due to molecular
107011
rotation (kcal/kg)
c__ ...specific heat at constant volume, portion due to molecular
TOSC
vibration (kcalAg
0
)
d Trail strength, caliber, diameter (m)
diameter of nozzle mouth (in)
nozzle t hr oat diameter (m)
cro: : s-sect i onal areas (in )
cr oss- sect i onal areas of t hr oat of f i r e nozzle (m
2
)
area of surface bounding mouth of nozzle (m }
accel erat i on due t o gravi t y (ra/sec
2
)
l engt h of pr oj ect i l e (m)
smal l est thickness of l ubr i cat i ng l ayer (m)
ver t i cal shock i r pi l s e which i n uni t t i ne i s t ransmi t t ed
surface of pl at e by impinging a i r molecules (kg/ m)
ver t i cal rebound impulse T.tich i n uni t time i s given
surface of pl at e t o rebounding a i r molecules (kg/nr)
by
.impulse par al l el t o trail) of the a i r molecules Tiiich
impinge i n uni t t i ne on a uni t area of the trail (kg/m
2
)
free pat h l engt h of molecules (m)
mass (kgsec^/m)
air pressure, gas pressure, pressure (kg/m
2
)
external pressure of static air [kg/to
2
)
initial pressure, gas pressure in furnace, stationary gas
(kg/m2)
mouth pressure (kg/m
2
)
dynamic pre-sure, heat florr (kcal/m
2
sec)
trajectory radius of horizontal turning curves (m)
flight distance (m,km) (pathlength)
starting distance (ra,km)
length of partial distance after starting (m,km)
length of accelerated climb (m,km)
length of unaccelerated supersonic glide path {m,ko)
subsonic Lide path length (m,km)
s length of turning distance (m,km)
t*............time (seconds), wing depth, sliding slipper depth (m)
t
f
hot aide temperature of fire wall(C)
t
k
cool side temperature of fire wall (C)
v flight velocity (m/sec)
j- ............velocity vector (m/sec)
v ..absolute velocity (m/sec)
v initial flight velocity (m/sec)
o
V vel oci t y of a poi nt on t he e a r t h' s surface (m/sec)
e
v - ...,,,flight velocity at beginning of turning arc (m/sec)
v^.. flight velocity at end of turning arc (m/sec)
w ..bomb range (m,km) (literally "throwing distance")
w probable scattering of bomb trajectories (km)
r
...,.average density of boob hits for surface attack (kcal/km )
A.............mechanical equivalent of heat (1/A27 kcal/kg), lift (kg)
B bomb load (kg)
C... coriolis force (kg)
D. , .dissociation energy (kcal/kg)
E ...reaction heat of fuels, upper mixture coefficient, total
energy content (kcal/kg)
K spacial energy concentration (kcal/Liter)
K
R
rebound energy of air molecules per unit time and unit wall
surface (kcal/m sec)
K .....energy of air molecules remaining in walls after lir.pact
w
(kcal/m sec)
T, energy carried by air molecules per unit time and unit surface
A
to wall (kcal/m sec)
F ..aerodynamic supporting surface (m )
F carrying wing surface (m )
*
2
Fg.. carrying fuselage surface (m )
G mi ght (kg)
G i ni t i al weight (kg)
o
G, weight of bomber after using up fuel required in approach of
target (kg)
G weight of bomber after release of bombs (kg)
G_... empty weight of bomber (kg)
G ..starting weight (kg)
s
G
OT
-... initial starting weight (kg)
H flight altitude (m,km)
J heat content (kcal/kg)> impulse (kgsec)
J heat content in state of rest, state of furnace, initial state
(kcal/kg)
J mouth impulse (kgsec)
m
K . . . . . eval uat i on number of rocket fuel s
If .molecular weight
P ef f ect i ve thrust, load on one of a sl i di ng slipper (kg)
P . . . . . . free thrust measurable by dynaffioaeter {'kg)
Q. . . . . . . . . . . . . qua nt i t y of heat (kcal/kg)
R individual gas constant (m/), radius of earth (m)
R Reynold's number
e
R,
Q
i nternal hal gi t of evaporation at 0K (kcal/kg)
T temperature (K), d'Alembert's i nert i al force (kg)
T . . . . i ni t i a l temperature, furnace temperature, st at i c temperature (TE)
T mouth temperature (K)
m
To*-. temperature of air molecules before impact with wall C&)
T^..... temperature of a wall surface struck by air molecules ( K)
T temperature of air molecules after rebounding from wall ( K)
R
U internal energy (kcal/kg)
7 . speci f i c gas volume (m/kg)
V ...furnace volume (m-')
o
W. resistance (kg)
W .resistance measured in win<l tunnel (kg)
Z ...total energy thrown at a target (kcal)
\ .....angle of attack ( ), coefficient of accommodation
^j. ..angle of attack of leading edge of a curved surface
*". specific Height (kg/m )
o specific gas weight at the critical Telocity (kg/nr)
u or
v
0
speci fi c gas weight at rest^in the neighborhood of Earth's surface
(kg/m-*)
*. . . . . gl i de number, numerical eccentri ci ty of a Keppler el l i pse
, optical absorption coefficient
5 compass rose angle ( i . e . "heading") ()
p, . vi scosi t y (kgsec/m
2
)
fa mmrliBmn effi ci ency, external effi ci ency
fo* .nozzle effi ci ency
pi Internal effi ci ency
p
0
furnace effi ci ency
f therm". thermal effi ci ency
Jl. adiabatlc exponents
?f. . . .average adiabatlc exponents
A thermal conductivity (kcal/mh
0
)
/* coeffi ci ent of f ri ct i on
P . densi ty, radius of path, (kgsec
2
/m^) (m)
f mass of gas stri ki ng unit surface area of a plate each second
(kgsec/m3)
P
m
..gas density at the mouth (kgsec /or)
/*t>... gas density in furnace, Initial state, rest state (kgsec /m )
~. ....strength, turning angle (kg/cm
2
) ()
X,. tangential stress due to air or gas drag (kg/m )
f. path inclination ()
H angular velocity (l/c)
^..... ...change in a quantity
.....characteristic temperature of vibrational excitation of a gas
molecule (K)
i
\ns