Diary of trip to California March 8-31, 1947

T]

A/^a- ^"id

p^/srL

NEW YORK UNIVERSITY

INSTITUTE OF MATHEMATICAL SOfNCES

LIBRARY

4 Waihington Place, New York 3, N, Y

Diary of Trip to r.'sllforrla > Farch 8 - 31, 1947

Harold Grad

Aerojet Corporation, Aguaa

The nroblorns of Irref^ular burning and genoral hydro-
dynawlc flow inside iiiultlplTf connected hollow tube rockets
were r^lscussed with T*. ^. D, Geckler and A, L. Antonio,

Secondary peak end Irregular burning phenoirena similar
to those observed at California Institute of Technology
durlnf? the war have been observed In both the hollow grain
and the rod end tube dealpns used by Aerojet.

T' ' "•

The Interrupted burning techniques used are still beln/a?
developed, but from a discussion of results already available
it appears probable that the resonance burning hypothesis
set forth by the New York TTniverslty group Is correct. The
physical basis of the phenonenon together with the qualitative
effects of various rocket design features were discussed.

-2-

Ir ordei* to verify certain aspects of the theory which h«a
been develot)ed. It waa decided to Derforro exioerlTnenta on
rockets of varying georretry and type of powder. Experiments
involving direct observation of powder reaction rates and
flow patterns Inalde the rocket are not feasible at present.
According to "Dr. Geekler, it would be v/orthwhlle to reduce
the irathematlcal solution to a forrri whereby the incidence of
pressure peaks and irregular burning could be predicted in
tenns of one or more parameters Involving the rocket eeometry
and powder grain constants.

In addition to the resonance burning oroblem, work done
at New York TTnlverslty on flow In multiple channel rockets
and on the validity of one dinenslonal treatments wts dis-
cussed.

T>r. Oeekler was also interested In heat +-ran8fer with
variable coefficients, and requested inforrration on the work
done by and under t^. ivacnonald on this subject.

Some problerrs in nozzle flow were discussed with
R. Gordon, in particular the calculation of thrust for an
inexact exnanslon nozzle. A Tremorenduin on this subject will
be sent to hire by Hew York TTniveraity.

■pjxperlTnents are being perforrred on rockets which have
purely diverging noszlea (i.e. no converging entry section),
ra^oek reflection natterns in the exhavist can be observed
directly by eye, since the various regions in the exhaust are
distinctively colored.

Several probleins involving straipjit pipe ^et wotors were
discussed with T5r. F, ?^Brlcl<T, who was interested in a Tnemo-
randuK being written on this subject, and who suggested so^e
possible lines of further development. He also expressed
interest in a renort on the Pulse Jet by Wae Donald and sehaaf
and requested a cony.

-3-

Dp. Tlwlcky, as well sa nun^erous other poraona encountered
In California, la e8«j;erly airaltlng the oppearanoe of the revised
Shook ?'eve I'nnual.

JPL-APL Sym-posluw

JPL - Jet Propulsion Laboratory - California Institute of

Technolopy

APL - Applied Physics Laboratory - Johns Hopkins University

The sympositur consisted mainly of a discussion of free
fli^t techniques and comparison with wind tunnel results.
It appears that there ere no results at present which allow
of direct coTTparlaon between the two rrethods.

Certain anor.alouc droft effects are observed at transsonlc
speeds, but different experiments vary widely in results.

k novel method of measuring drag of a projectile in free
fll^t using a so-called belllstlc pendulum was described.

California Institute of Technology

Dr. Hans LleTrann was consulted on his work on interactions
between shock and viscous boundary layers. It appears that
the very large velocity -radients in the neighborhood of a
viscous boundary layer at a well produce shock reflection
patterns which are entirely different froni what la predicted
by simple ehock reflection theory, but it is expected that
at sufficiently large distances from the wall the pattern may
apTwoxlciate that of the slinple theory. Slmiler curved shock
effects are found in I^ach reflectiona because of the large
velocity gradients In the sllpstresn. These effects, together
with Kiore complicated shock and boundary layer phenomena are
found to be essential in describing transsonlc flow around an
airfoil (which may be somewhat connected v'lth recent investigations
In New York University).

-4-

Some proT?«rtle8 of llrit lines were discussed by
Drs. Liepmerm and Cleuser at s lecture by the latter*

Several Interesting probler.s eoncernlnf^ characteristics
of hydrodynarnlc equations (with and without either heat
conduction or viscosity); characteristics of lower order
terms in parabolic equations; a comDarlaon of the equations
of transient nozzle flow and waves on sloping beaches;
some features of the Karwan-Moore perturbation method, end
«n equation of rlxed elllr^tlc-hyperbolic character were
dlacussed with Dr» D© Prima,

Two water tunnels (one for rreasurenent of drag, lift,
moment and one with a free surface at arbitrary air pressure),
a centrifugal tornedo impact tank, and the tranasonic wind
tunnel were visited.

Other presons contacted on various matters are Professors

Epstein and Bohnenblust,

Autoratlc Computing? I'achines

At California Institute of Technology Professor MeCann
of the Electrical Engineering Department was consulted about
the status of various computation developments on the coast,
TTe is at present constructing an analogue differential analyzer,
aimed principally at elastic beam and electrical system
stability nroblerrs, which treats on the order of 100 linear
ordinary differential eouations and some 10-20 nonlinear
equations. There is a nosslbillty of a digital computer being
constructed at California Institute of Tech-nology some time
in the future. Various electromechanical simulators such as
for torpedoes, guided missiles, etc. exist, and UCLA has a
GE differential analyzer.

Howard Hughes Aircraft is eonstrwetlni? various small
specialized machines, and one larp;e digital machine (details
secret) which Is nresvroably to solve !"'axwell'8 Equations.
They have obtained some computing machine men from Bell Labs.

Prom a consensus It appears thst the Bureau of Standards
Institute of Numerical Analysis i-vlll probably be located at
Stanford or Berkeley. According to the outspoken S, ^aymo.
Director of Research at Howard Hucjies, the Bureau of standards*
methods do not offer much chance for any development in
larf^ scale computlnp; nachlnea other then a rehash of already
exist Ini^ machines.

Professors Spencer and Polya who were cortacted at
Stanford are anxious to have the Bureau of Standards' Computing
Center either there or at Berkeley.

!5r. spencer has some problems on schllcht functions
which involve large scale computations and for which he la
thinking of usin^ the ?^iac.

The nroblem of Ir.pact of a cone in water v;hlch is being
put Into final form by Drs. Spencer and Shlffman was discussed,
end some necessary liaison between the two was arranged.

NEW YORK UNIVERSITY

INSTITUTE OF MATHEMATICAL SaENCES

LIBRARY

4 W»Ai.g;an Place,'i^^ York 3, N, Y

,'>«>

a'rso'ii