N A SA TE C H N I CA L N O T E (^^B^ NASA TN D.

51 45

S
;
UQ
0:
^ ’"

^ ga |
c
AFW.
^^^ j
^^x^a
fi/VI H ’-)> 5

^^ S

MECHANICAL PROPERTIES
AND COLUMN BEHAVIOR
OF THIN-WALL Be-38Al
ALLOY TUBING

by Donald R. Rummler
Langley Research Center
Langley Station, Hampton, Va.
NATIONAL AERONAUTICS AND SPACE ADMINISTRATION WASHINGTON, D. C. APRIL 1969
 TECH LIBRARY KAFB, NM

0131133
NASA TN D-5145

MECHANICAL PROPERTIES AND COLUMN BEHAVIOR

OF THIN-WALL Be-38Al ALLOY TUBING

By Donald R. Rummler

Langley Research Center

Langley Station, Hampton, Va.

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION

For sale by the Clearinghouse for Federal Scientific and Technical Information
Springfield, Virginia 22151 CFSTI price $3.00

I.-.
 MECHANICAL PROPERTIES AND COLUMN BEHAVIOR
OF THIN-WALL Be-38Al ALLOY TUBING

By Donald R. Rummler
Langley Research Center

SUMMARY

The results of an experimental investigation to determine the room-temperature

mechanical properties and column behavior of commercially produced Be-38Al alloy
tubing are presented. The investigation included tubing in both the as-extruded and
annealed conditions. The diameter of the tubes ranged from 0.25 to 0.69 inch (6.4 to
17.5 mm). Wall thickness was 0.020 inch (0.51 mm). Microhardness measurements and
metallurgical studies were also performed. The mechanical-property measurenients
indicated that the tubing was comparable to other forms of the Be-38Al alloy. Column-
buckling loads could be satisfactorily predicted by using the tangent modulus, derived
from compressive stress-strain curves, in the inelastic column-buckling equation.
Column mass-strength comparisons showed that in the plastic range, the as-extruded
Be-38Al tubing was more efficient than beryllium tubing. Tne results also indicated that
the Be-38Al tubing could be commercially produced to dimensional tolerances comparable
to those of aluminum tubing.

INTRODUCTION

A continuing need exists in aerospace structures for materials which coinbine high
stiffness with low density. Although beryllium has outstanding specific stiffness, its
structural application has been limited. This limited application is, in part, due to the
brittle and anisotropic behavior of beryllium under biaxial stresses which, among other
things, makes it difficult to fabricate and join. A beryllium-aluminum alloy (Be-38Al),
which has recently become commercially available, has a specific stiffness intermediate
between beryllium and other structural metals. In addition, this new alloy in sheet form
is more ductile, is easier to fabricate, and has better impact resistance than beryllium
sheet.

Although the mechanical properties of Be-38Al alloy in sheet form have been well
documented (see, for example, refs. 1 to 3), only a limited amount of information on
extruded tubing is available in the literature. Because of the interest in lightweight
tubular structures, an investigation of thin-wall beryllium and beryllium-alloy tubing
for lightly loaded truss-type structures has been initiated. Results obtained on the beryl-
lium tubing are reported in reference 4.

The purpose of the present paper is to report the room-temperature mechanical

properties and column behavior of commercially produced Be-38Al extruded tubing. The
investigation included tubing in both the as-extruded and annealed conditions. The nomi-
nal diameter of the tubing ranged from 0.25 to 0.69 inch (6.4 to 17.5 mm). The nominal
-wall thickness of all tubing was 0.020 inch (0.51 mm).

SYMBOLS

The units used for physical quantities defined in this paper are given both in the
U.S. Customary Units and in the International System of Units (SI). Conversion factors
relating the two systems are given in reference 5, and those pertinent to the present
investigation are presented in appendix A.

A area, inches^ (meters2)

c column-end-fixity coefficient

D outside diameter of tube, inches (meters)

E Young’s modulus in tension, pounds force/inch2 (newtons/meter2)

E Young’s modulus in compression, pounds force/inch2 (newtons/meter2)

E+ tangent modulus, pounds force/inch2 (newtons/meter2)

e total elongation in 2 inches (5 centimeters), percent

e^ uniform elongation, percent

1 moment of inertia, inches4 (meters4)

k local-buckling constant

L column length, inches (meters)

P applied load, pounds force (newtons)

2
S-o standard deviation of outside diameter of tube

Sf standard deviation of wall thickness

t wall thickness, inches (meters)

W mass, pounds mass (kilograms)

^1 transverse sensitivity coefficient for longitudinal strain gage

(32 transverse sensitivity coefficient for transverse strain gage

e strain

e actual longitudinal strain

69 actual transverse strain

60 apparent longitudinal strain

60 9 apparent transverse strain

li Poisson’s ratio

P.Q Poisson’s ratio of isotropic bar

P radius of gyration, inches (meters)

o stress, pounds force/inch^ (newtons/meter2)

max average stress at maximum compressive load, pounds force/inch^

(newtons/meter2)

w density, pounds mass/inch^ (kilograms/meter^)

Subscripts:

cc local buckling

3
cp compressive proportional limit

cr buckling

cy 0.2-percent offset compressive yield

e effective

m mean

max maximum

min minimum

n nominal

tp tensile proportional limit

tu tensile ultimate

ty 0.2-percent offset tensile yield

A bar over a symbol indicates average value.

EXPERIMENTAL PROCEDURES

Materials

The Be-38Al tubing (38-percent aluminum by weight) used in this investigation was
furnished by the supplier in both the as-extruded and annealed conditions. Delivery of
the tubing was completed in the fall of 1966. The tubing diameter ranged from 0.2500 to
0.6875 inch (6,35 to 17.46 mm), and the nominal wall thickness was 0.020 inch (0.508 mm).
The nominal outside diameters, the three heat and lot numbers, and chemical composi-
tions of the tubing are presented in table I.

All tubing was extruded from hollow-core "canned billets." A steel mandrel was
used to fill the cored billet before extrusion. The canned billet and the extrusion tooling
were heated to 850 F (730 K) prior to extrusion. A reduction ratio of 25:1 was used
for all of the tubing. The extrusion ram speed was 11 in./min (4.6 x 10~3 m/s). After
extrusion, the tubing was dejacketed in nitric acid. The tubing was then resistance heated
and stretch straightened. Tubing supplied in the annealed condition was heat treated after
straightening for 24 hours at 1100 F (865 K) in a vacuum of less than 0.1 torr (13 N/m2).
4
 The specified dimensional tolerances on outside diameter, wall thickness, and
straightness for both the as-extruded and annealed tubing are presented in table II.
These tolerances are comparable to those for the 5000-series aluminum alloys presented
in ASTM designation B 221-67 (ref. 6).

Metallurgical Examinations
Metallurgical examinations included macroscopic and microscopic observations
and microhardness measurements. The tubing was examined macroscopically both in
the as-received condition and after etching in a chromic acid, hydrofluoric acid, and
water solution (HF-CraOj-HgO) to remove approximately 0.010 inch (0.25 mm) of the
outside diameter. Standard metallographic procedures were used to prepare specimens
for microstructure examinations. Polished sections were prepared in both the longitu-
dinal and transverse directions of the tubing. Knoop microhardness determinations were
made on polished sections with a standard microhardness machine. The penetrator load
was 100 grams, and the dwell time was 15 seconds. Each reported average microhard-
ness number was based on 10 microhardness determinations.

Dimensional Measurements
The outside diameter, wall thickness, and straightness of the Be-38Al tubing were
measured to verify conformance to the specified tolerances (table n) and to provide an
estimate of the variability of diameter, wall thickness, area, and moment of inertia. At
least six measurements of outside diameter and wall thickness were used to calculate
average values for each nominal diameter of tubing. Typically, measurements were
made at three equally spaced sections on the tubes which were from 1 to 10 inches (25 to
250 mm) long and at five equally spaced sections on tubes which were longer than
10 inches (250 mm). The variability in diameter was calculated by using the average
diameter determined at each measured section as a single observation. Wall-thickness
variability was calculated in a similar manner. Assuming a circular cross section and
a constant wall thickness, the area and moment of inertia of each measured section were
calculated and used to determine the variability of these two section properties. Straight-
ness measurements were made on tubes greater than 3 inches (76 mm) in length. Addi-
tional details of the equipment and procedures used for the dimensional measurements
are presented in reference 4.

Mechanical-Property Tests
Tensile and compressive mechanical-property tests were performed at room tem-
perature in a 120 000-Ibf-capacity (535-kN) universal hydraulic testing machine.

5
 The tensile specimens were prepared by adhesively bonding end fittings to the
tubing. To preclude specimen failure at the end fittings, the bonded tensile specimens
were etched in a HF-CroOj-HoO solution to remove a minimum of 0.010 inch (0.25 mm)
from the outside diameter of the specimen. The length of the reduced section was at
least 6 times the nominal diameter of the specimen. Carefully alined precision-machined
grip assemblies were used for tensile tests to minimize loading eccentricity.
Compressive specimens were tested in the as-received condition. The ends of
these specimens had been ground plane, smooth, and perpendicular to within 0.25 of an
axial line passing through the centroids of the specimen ends. The length-to-diameter
ratio of the compression specimens was approximately 4. The compression specimens
were supported on hardened steel disks. Annealed aluminum washers were inserted
between the specimen ends and the steel disks. The plastic deformation of the aluminum
washers helped to distribute the load uniformly on the specimen.
To prevent possible contamination of the laboratory air with toxic Be-38Al dust,
both the tensile and compressive specimens were enclosed in a protective cylindrical
sleeve during testing. This sleeve was made from multiple layers of 0.008-inch
(0.2-mm) latex sheet, wound around the steel support blocks of the compressive speci-
mens and the aluminum end fittings of the tensile specimens. The sleeves were sealed
with masking tape. After testing, the sleeves were removed while the assembly was
immersed in acetone.
In both the tensile and compressive tests, strain was measured with foil-type strain
gages adhesively bonded to the specimens. Each strain-gage assembly included strain-
sensing elements in both the longitudinal and transverse directions. The gage assem-
blies were equally spaced about the circumference of each specimen. Two such assem-
blies were bonded to each 0.250-inch-diameter (6.35-mm) specimen. Four gage
assemblies were bonded to each specimen over 0.250 inch (6.35 mm) in diameter. The
electrical outputs of the strain gages and of a load-indicating deflectometer attached to
the load dial of the testing machine were recorded on magnetic tape. Each electrical
signal was recorded at approximately 0.5-second intervals up to the yield strain and at
approximately 2.0-second intervals thereafter.
The nominal strain rate for the mechanical-property tests was 0.0015 per minute
up to the 0.2-percent offset yield strain and 0.015 per minute thereafter. The strain
rate was manually controlled throughout each test. The instantaneous strain rate was
monitored by comparing the output of a longitudinal strain gage on a strip-chart recorder
with lines drawn on the recorder paper at the desired strain rate.

Finely scribed pencil lines at 0.40-inch (10-mm) intervals along the specimen
were used to make elongation measurements on the tensile specimen. Both elongation
in 2 inches (5 cm) and uniform elongation were measured. Uniform elongation is the

6
amount of residual plastic strain in the unfractured portions of a tensile specimen. Uni-
form elongation does not include the region of the specimen near the fracture and was
determined by averaging the residual plastic strain in each portion of a fractured tensile
specimen.
Data from the mechanical-property tests were reduced by means of a digital com-
puter and associated automatic plotting equipment. To determine the tangent-modulus
curves, short segments (seven consecutive data points) of the stress-strain curves were
successively fitted to a second-order polynominal equation by the method of least squares.
The first derivative of the fitted equation was used to calculate tangent modulus at the
center of the fitted segment. To determine Poisson’s ratio, correction of the indicated
strains for transverse strain was necessary. By using the procedures presented in ref-
erence 7 for this type of correction, the following equations were developed to account
for the different transverse sensitivity coefficients in the longitudinal and transverse
gages of the strain-gage assemblies:

, 1 ^a.l^ ^l) ^^1 ^)

1 ^2
ea.2(l ^a.l^ ^l)
where e and e
e2
^ 1 ^2
are the apparent strains in the longitudinal and transverse gages,
respectively; p.o is the Poisson’s ratio of the isotropic bar on which the gages were
calibrated; and j3-^ and f3^ are the transverse sensitivity coefficients of the longitu-
dinal and transverse gages, respectively. The reported Poisson’s ratio for each speci-
men represents an average of the Poisson’s ratios determined from each gage assembly
on a specimen. Additional details of the equipment and procedures used for the
mechanical-property tests are presented in reference 4.

Column Tests
Column tests were also made at room temperature in a 120 000-Ibf-capacity
(535-kN) universal hydraulic testing machine. Column specimens ranged in length from
3 to 30 inches (76 to 760 mm) and were tested in the as-received condition. The ends of
the columns were ground to the same tolerances as the ends of the compression speci-
mens. Four foil-type strain gages were bonded to the center of each column. These
gages were equally spaced about the circumference of the column and were used to
establish an initial elastic modulus. The columns were enclosed in a protective latex-
sheet sleeve and were tested with fixed ends.

7
 Steel tubes with stiffness comparable to that of the Be-38Al columns were used to
establish the column-end-fixity coefficient c by the method outlined in reference 8.
The experimental buckling stress was taken as the average stress at the maximum com-
pressive load o^^x. Additional details of the column testing equipment and procedures
are presented in reference 4.
A graphical procedure which employed the tangent-modulus inelastic column
equation,

aer TT2^
where

- ^
/T
1-=-\
\^e/
\2
\P }

c
f
^ 94
d.\)t

was used to calculate the theoretical column-buckling stress over a range of effective
slenderness ratios Le/P for each type of Be-38Al alloy tubing. The theoretical column-
buckling stress o y at a given slenderness ratio was calculated by using the average
value of compressive tangent modulus for all of the tubing in a given heat-treatment
condition.

RESULTS AND DISCUSSION

Metallurgical Examinations

Macrostructure.- Both the as-extruded and the annealed tubes typically exhibited a
dull matte appearance which is characteristic of an etched surface (fig. 1). Some of the
tubes in the as-received condition contained minor surface defects such as small pits and
very shallow longitudinal grooves. Often these grooves were formed by a series of
closely spaced small pits. When tensile specimens were etched, these surface defects
were easily detected (figs. 2(a) and (b)). One of the as-extruded 0.250-inch-diameter
(6.35-mm) tensile specimens exhibited a "rattlesnake" surface after etching (fig. 2(c))
which was not evident prior to etching.
Microstructure.- Typical samples of longitudinal and transverse microstructure
of the as-extruded and the annealed tubing are presented in figures 3 and 4, respectively.
The gray areas are beryllium and the white areas are aluminum. In both the as-extruded
and the annealed conditions, the aluminum is somewhat elongated in the longitudinal
direction and essentially randomly dispersed in the transverse direction. The major
difference between the microstructures of the two types of tubing is an apparent coars-
ening of the aluminum (i.e., greater separation of the aluminum "islands") in the annealed
material. The annealed microstructures are similar to those of the annealed extruded

8
bar described in reference 1. Small angular inclusions (figs. 3(a) and 4(b)) were occa-
sionally noted. Similar inclusions were tentatively identified in reference 9 as beryl-
lium carbide.

Microhardness. The results of the microhardness measurements are presented

in table III. As expected, the as-extruded tubing was somewhat harder than the annealed
tubing. On the basis of the confidence intervals listed in table ffl, there do not appear to
be any significant differences in microhardness as a function of direction. This isotropic
behavior is in contrast to the anisotropic microhardness of extruded thin-wall beryllium
tubing (ref. 4). Except for the 0.250-inch-diameter (6.35-mm) tubing in the annealed
condition, the microhardness values of all tubing in a given heat treatment were essen-
tially the same. The microhardness values of two different 0.250-inch-diameter
(6.35-mm) annealed tubes were significantly different from each other and from the
microhardness values of the other annealed tubes. No completely satisfactory explana-
tion for this difference in microhardness could be found. The somewhat higher oxide
content of these tubes (table I) may have been responsible for their increased hardness.
This explanation, however, is not consistent with the fact that the microhardness of the
as-extruded tubes made with this lot of material was not significantly higher than the
microhardness of the as-extruded tubes made from material with a lower oxide content.

Dimensional Measurements
The measurements of diameter and wall thickness, normalized with respect to their
specified nominal values, are summarized in table IV. These normalized data and simi-
lar data on the calculated area and moment of inertia are presented in figure 5. As
expected, on a percentage basis wall-thickness variations were much greater than diam-
eter variations. The standard deviation of diameter was less than 2 percent of the speci-
fied diameter for all of the tubes, and the standard deviation of wall thickness ranged
from 1.5 to 7.5 percent of the mean wall thickness. Of more importance is the fact that
the average wall thickness was as much as 12 percent below the specified value. The
average values of the computed section properties A and I reflect the variation in
wall thickness exhibited by the tubing. It is interesting to note that extruded beryllium
tubing (ref. 4) exhibited similar variations in wall thickness, although in all cases the
average wall thickness of the beryllium tubing was in excess of that specified. It should
also be noted that the dimensional tolerances for both the Be-38Al tubes and the beryl-
lium tubes were comparable to those used for 5000-series aluminum alloys. Conse-
quently, as was the case for the beryllium tubes, the maximum deviation from any speci-
fied dimension in the Be-38Al tubing is comparable to that which could be expected in
5000-series aluminum tubing.

9
 The annealing heat treatment did not appear to alter significantly the dimensional
characteristics of the tubing. Both types of tubing exhibited similar increases in the
normalized values of average diameter and average wall thickness as a function of
increasing diameter.

Mechanical-Property Tests
The results of the mechanical-property tests are summarized in table V. The
results for individual specimens are presented in table VI for the tensile tests and in
table VII for the compressive tests. The tensile strengths of the as-extruded tubing
showed considerable variation over the range of diameters tested. All other measured
mechanical properties for the tubing of a given heat treatment showed little variation as
a function of diameter.
Strengths and proportional limit.- As expected, the annealed tubing had lower
values of strength than the as-extruded tubing (table V). Typical stress-strain curves
for these two types of tubing are shown in figure 6. Both the compressive proportional
limits and the compressive yield strengths of the tubing were usually somewhat lower
than the comparable tensile values. The lower compressive proportional limits and
yield strengths are probably due to the procedures used to straighten these tubes after
extrusion.
The data for the as-extruded tubing (table V) indicated a general trend of increasing
proportional limits and yield strengths as a function of increasing diameter. This trend
is shown graphically in figure 7. This trend, however, was reversed in the annealed
tubing; that is, the proportional limits and yield strengths decreased slightly as a func-
tion of increasing diameter (fig. 8). The greater susceptibility of the stronger, larger
diameter tubing to the annealing heat treatment, in terms of the reduction in yield strength,
is consistent with a similar observation reported in reference 2 for as-rolled Be-38Al
sheet exposed to heat treatment at 1100 F (865 K).
The data on tensile yield strength, tensile ultimate strength, and elongation pre-
sented in table V are in general agreement with similar data presented in reference 2
for Be-38Al extruded mill forms. Both extruded bar and sheet in the annealed condition
have a lower yield strength in compression than in tension (ref. 1).
Elastic modulus and Poisson’s ratio.- The average values of elastic modulus for
the Be-38Al tubing ranged from 28.1 x 106 to 30.4 x 106 psi (194 to 210 GN/m2). (See
table V.) These values are in excellent agreement with the values reported in refer-
ence 2 for extruded bar and sheet. No consistent differences in elastic modulus, either
as a function of diameter or as a function of heat treatment, were apparent. This result
is in contrast to the reduction in elastic modulus which was observed in reference 3 after
heat treatment of as-extruded Be-38Al sheet material. The stress dependence of the

10
tangent modulus is presented in figure 9. These typical curves illustrate the lower pro-
portional limits of the annealed tubing and its rapid decrease in tangent modulus for
stresses above the proportional limit. This figure also reflects the minor differences
in the shapes of the tensile and compressive stress-strain curves in each heat-treatment
condition.
The stress dependence of Poisson’s ratio for the Be-38Al tubing is shown in fig-
ure 10. Above the proportional limit, both types of tubing exhibited the expected increase
in Poisson’s ratio. This increase occurred more rapidly in the annealed tubing as a
function of stress than in the as-extruded tubing. In addition, the annealed tubing typically
exhibited a higher maximum value of Poisson’s ratio than the as-extruded tubing. The
differences in the shapes of the curves in figure 10 for the tensile and compressive spec-
imens were reproducible and are considered characteristic.

Failure modes.- In both the as-extruded and annealed tubing, tensile fractures
typically occurred at right angles to the longitudinal axis of the tubing (fig. 11). The
annealing heat treatment did not affect the shape of the fracture. However, the annealing
treatment increased the average elongation at failure (table V). The small longitudinal
grooves or pits noted in the etched tensile specimens did not seem to affect the strength
or the elongation at failure of either the as-extruded or the annealed tubes. The single
0.2 50-inch-diameter (6.3 5-mm) tensile specimen with the rattlesnake surface had a very
low strength (30.1 ksi or 208 MN/m2) and almost no elongation. This specimen is not
reported in table VI.
The failure modes of the compression specimens were different from those of the
tensile specimens and suggested that the annealed tubing was less anisotropic than the
as-extruded tubing. The as-extruded compression specimens developed longitudinal
cracks when the maximum compressive stress was attained (fig. 12(a)). These longi-
tudinal cracks were induced by the circumferential tensile stresses which were developed
near the restrained ends of the short compression specimens. The annealed compression
specimens (fig. 12(b)) continued to deform after the maximum compressive stress was
attained and typically sustained a considerable amount of circumferential plastic strain
without developing longitudinal cracks. Even after an annealing heat treatment, Be-38Al
extrusions have a crystallographic texture similar to that of beryllium extrusion (ref. 9).
Consequently, it is reasonable to assume that the behavior of the annealed tubing in com-
pression was less anisotropic than that of the as-extruded tubing because of the decrease
in preferred orientation which occurred as a result of the annealing heat treatment.

Column Tests
The results of the column tests are presented in table VIII. This table also pre-
sents the experimentally determined elastic modulus and the calculated buckling stress

11
a^v for the columns. No corrections were made for the lack of column straightness.
C.
The average value of the initial elastic modulus of the columns was 28.5 x 10 psi
(197 GN/m2). This compares closely with the average of 28.8 x 106 psi (198 GN/m2)
from the mechanical-property tests.
In figures 13 and 14, the results of the column tests are compared with the calcu-
lated buckling stresses. The elastic Euler equation and the Engesser tangent-modulus
equation for both the average value of tangent modulus and the range in tangent modulus
obtained from the compression tests were used in this calculation. For both the as-
extruded and annealed tube columns, the agreement between the predicted and the experi-
mental column behavior is satisfactory. It should be noted that the prediction band for
the as-extruded columns (fig. 13) includes the variation in compressive properties which
was observed for the four different diameters of tubing. Although the results of the
annealed tube columns (fig. 14) are well within the predicted range of column failure
stress, both the data scatter and the range of the prediction band reflect the lower repro-
ducibility of the compressive mechanical properties of the annealed tubing.
The effects of the low proportional limit, as compared with the yield strength,
exhibited by the Be-38Al tubing are well illustrated by comparing the Engesser tangent-
modulus prediction band with the elastic Euler curve. In the case of the annealed tube
columns, for instance, the use of the tangent-modulus inelastic column equation for
stresses as low as 15 ksi (103 MN/m2) is necessary to predict the column-buckling
stress.
Column-Efficiency Comparisons

In this section, the structural efficiency of the Be-38Al columns tested in the pres-
ent study is compared with the efficiency of minimum-mass Be-38Al columns. The
structural efficiency of the Be-38Al tubing is also compared with the structural efficien-
cies of aluminum and beryllium for use in minimum-mass thin-wall tube columns of cir-
cular cross section. The mass-strength equations utilized in these comparisons are
presented in appendix B. The material properties and compressive tangent-modulus
curves used for the comparisons are shown in table IX and figure 15, respectively. The
material properties of the aluminum (7075-T6) and beryllium tubing were taken from
references 10 and 4, respectively. The selected beryllium tubing is referred to in ref-
erence 4 as type BL.
Mass-strength data for the as-extruded Be-38Al columns utilized in the present
study are compared with the mass-strength data for minimum-mass (o^y Ogc) as-
extruded Be-38Al columns in figure 16. The symbols show values obtained from the
column tests reported herein, and the curves show calculated results. In addition to the
minimum-mass (o-gy a^c) curve, calculated curves for the maximum and minimum
DmA values of the columns that were tested are also shown in figure 16. These two

12
curves provide a frame of reference for the column results and show Dm/t "cutoffs"
for columns which are restricted by minimum-gage considerations to values of D^A
which are less than those of a minimum-mass column. (See appendix B.)

Figure 16 shows the wide range of the structural index P/L2 that was covered in
the column tests. In addition, this figure clearly shows that many of the tested columns
were inefficient when compared with a minimum-mass Be-38Al column, particularly at
the lower values of the structural index. Even though the wall thickness was 0.020-inch
(0.51 mm), the columns were severely restricted by minimum gage limitations at the
lower values of the structural index (P/L2 < 10 Ibf/in2 or 69 kN/m2). For example, the
mass corresponding to the columns tested at a structural index of 0.1 Ibf/in2
(0.69 kN/m2) was 5 times as great as the mass of comparable minimum-mass (cr^r o^c)
columns.
In the rest of this section, the calculated
column efficiency of Be-38Al tubing is
compared with that of aluminum and beryllium tubing. The compressive tangent-modulus
curves shown in figure 15 clearly indicate thatthe four materials change in relative stiff-
ness as the stress level increases. Figure 15 does not, however, provide an adequate
materials comparison for tube columns. This figure does not reflect the density of the
materials and does not allow a direct comparison of the materials for a given set of
column design conditions P and L. A more useful comparison of the relative efficien-
cies of these materials for tube-column applications is shown in figure 17. In this fig-
ure, the column efficiency of the Be-38Al tubing is compared with the efficiencies of
aluminum and beryllium tubing on a column mass-strength basis. The curves are for
minimum-mass (o-gy Ogg) pinned-end columns and were obtained from equations (B6)
and (B7) (appendix B). As would be expected from the curves in figure 15, the beryllium
tubing is most efficient in the elastic range (low
values of P/L2). Because of the rapid
decrease in the tangent modulus with increasing stress in the beryllium (fig. 15), the as-
extruded Be-38Al alloy is the most efficient material at values of P/L2 greater than
0.1 Ibf/in2 (0.69 kN/m2). For values of P/L2 greater than 1 Ibf/in2 (6.9 kN/m2), the
annealed Be-38Al alloy is as efficient as the beryllium. However, both beryllium and
annealed Be-38Al alloy become less efficient than aluminum for values of P/L2 greater
than 10 Ibf/in2 (69 kN/m2).
In summary, of the materials considered in this comparison, the as-extruded
Be-38Al tubing is the most efficient on a mass-strength basis over a wide range of the
column structural index P/L2.
CONCLUDING REMARKS

The room-temperature mechanical properties and column behavior of commercially

produced thin-wall Be-38Al tubing in both the as-extruded and annealed heat-treatment
13
conditions have been investigated. The results indicated that the Be-38Al alloy tubing
could be commercially produced to dimensional tolerances comparable to those of
aluminum tubing. The results of the mechanical-property tests, which include three lots
and three heats of material, indicated that the properties of the tubing were comparable
to those reported in the literature for extruded bar and sheet products of the Be-38Al
alloy. The column behavior of the tubing could be satisfactorily predicted by using the
Euler-Engesser relationship. The column tests also demonstrated that the use of the
tangent-modulus inelastic column equation was necessary at stresses as low as 15 ksi
(103 MN/m2) for the annealed tubing because of its low proportional limit. Column
mass-strength comparisons showed that in the plastic range, as-extruded Be-38Al alloy
is more efficient than beryllium.

Langley Research Center,

National Aeronautics and Space Administration,
Langley Station, Hampton, Va., January 10, 1969,
124-08-01-05-23.

14
 APPENDIX A

CONVERSION OF U.S. CUSTOMARY UNITS TO SI UNITS

The International System of Units (SI) was adopted by the Eleventh General
Conference on Weights and Measures, Paris, 1960. (See ref. 5.) Factors required for
converting the U.S. Customary Units used herein to the International System of Units (SI)
are given in the following table:

TT c; pnctnmaT^
u’0- customary Conversion
Physical quantity factor SI Unit
unit
/^
Mass Ibm 0.4536 kilograms (kg)
Length in. 0.0254 meters (m)
Velocity in./min 4.23 x 10-4 meters per second
Temperature (F + 460) 5/9 degrees Kelvin (OK)
Density lbm/in3 27.68 x 103 kilograms per cubic meter (kg/m3)
Load Ibf 4.448 newtons (N)
Pressure torr 1.333 X 102 newtons per square meter (N/m2)
fpsi Ibf/in2 6895 newtons per square meter (N/m2)
Iksi kips/in2 6895 x 103 newtons per square meter (N/m2)
*Multiply value given in U.S. Customary Unit by conversion factor to obtain equiv-
alent value in SI Unit.
Prefixes to indicate multiple of units are as follows:

Prefix Multiple
milli (m) 10-3
kilo (k) 103
mega (M) 106
giga (G) 109

15

J
 APPENDIX B

MASS-STRENGTH EQUATIONS FOR THIN-WALL TUBE COLUMNS

OF CIRCULAR CROSS SECTION

The efficiency of a minimum-weight column can be expressed in terms of a mass

index W/L3 and a structural or loading index P/L2 (ref. 11). The relationship
between these two indices as determined by equating the column-buckling stress cr^p to
the local-buckling stress o^. For the purposes of the present discussion of thin-wall
tube columns, the following expressions from reference 12 were utilized:

^cr
2
CTT^D
----m- (Bl)
SL"

o,, ^EeEt)172 "m (B2)

By utilizing the relationship for the area of a thin-wall tube of circular cross section,

A 7rDmt
equation (Bl) can be solved in terms of the structural index:

S^cr2 Dm
^
^
1,2 CTrEf
(B3)

Setting o o^y a^ and eliminating t/Dm between equations (B2) and (B3) yields

P, 8a3 (B4)
L2 CTrkE^Ec1/2
The mass of a column with uniform cross section is

Solving equation (B4) for a

W O)AL a?

^ L

and substituting the results into equation

W_ 2 oj /P \2/3
(B5) yields
(B5)

/gg^
L3 (c.k)1/3 Ee1/6^172^
This equation is the structural-efficiency relationship of a minimum-mass column for
a < Op For o Ogy, the column-efficiency relationship becomes

16
 APPENDIX B

^ ^-^
L2
L3
For a minimum-mass tube column (o^r cr^^
cy

/ 2
reference 11 shows that

\V3
(B7)

{)/OCr=aCC
v t
2

mass tube column are a function of the structural index

^^
\C7T

Equation (B8) shows that for a given material, the proportions

P/L2.
Dm/t
^
of a minimum-

Minimum wall thickness (minimum gage) is often a controlling factor in the design
of a tube column, particularly for low values of the structural index P/L2.
Therefore,
it is appropriate to consider the efficiency of a tube column for which minimum-gage
restrictions preclude the design of a minimum-mass column. Solving equation (B3) for
Ogp and substituting the results into equation (B5) yields

W /8 \l/2^m\-l/2 . ^)\V2
/P

^ [-T-] (B9)
/^
L3 E772
Equation (B9) is the structural-efficiency relationship for tube columns which, because
of minimum-gage restrictions, do not have minimum-mass proportions; that is,

D^D^
t v ^cr^cc
t

On a column mass-strength plot, equation (B9) yields Dm/t cutoff curves which merge
with curves obtained from equation (B6).

17
 REFERENCES

1. Fenn, R. W., Jr.; Crooks, D. D.; Watts, G. H.; and Neiman, A. S.: A Mechanical
Property Evaluation of Be-38% Al Alloy From -320 to 800 F. Metals Eng. Quart.,
vol. 5, no. 4, Nov. 1965, pp. 37-47.
2. Fenn, Raymond W., Jr.; Crooks, Donald D.; Brodie, Roy W.; and Chinowsky, Stanley:
Comparison of Lightweight Structural Materials: Be, and Alloys of Be, Mg, Al,
and Ti. (Preprint) 660652, Soc. Automat. Eng., Oct. 1966.
3. Fenn, R. W., Jr.; Crooks, D. D.; Coons, W. C.; and Underwood, E. E.: Properties
and Behavior of Beryllium-Aluminum Alloys. Beryllium Technology, Vol. 1.
Vol. 33 of Metallurgical Society Conferences, L. McDonald Schetky and Henry A.
Johnson, eds., Gordon and Breach, Sci. PubL, Inc., c.1966, pp. 103-143.
4. Rummler, Donald R.; Dexter, H. Benson; Harth, George H., in; and Buchanan,
Raymond A.: Mechanical Properties and Column Behavior of Thin-Wall Beryllium
Tubing. NASA TN D-4833, 1968.

5. Comm. on Metric Pract.: ASTM Metric Practice Guide. NBS Handbook 102, U.S.
Dep. Corn., Mar. 10, 1967.
6. Anon.: Standard Specification for Aluminum-Alloy Extruded Bars, Rods, Shapes,
and Tubes. ASTM Designation: B 221-67. Pt. 6 of 1967 Book of ASTM Standards
With Related Material. Amer. Soc. Testing Mater., c.1967, pp. 172-196.

7. Baumberger, R.; and Hines, F.: Practical Reduction Formulas for Use on Bonded
Wire Strain Gages in Two-Dimensional Stress Fields. Proc. Soc. Exp. Stress
Anal., vol. H, no. 1, May 1944, pp. 113-127.
8. Lundquist, Eugene E.; Rossman, Carl A.; and Houbolt, John C.: A Method for
Determining the Column Curve From Tests of Columns With Equal Restraints
Against Rotation on the Ends. NACA TN 903, 1943.
9. Fenn, Raymond W., Jr.; Coons, W. C.; Crooks, Donald D.; Neiman, Alfred S.;
Robinson, John; and Watts, Glenn H.: Evaluation of Be-38% Al Alloy. NASA
CR-64343, 1965.
10. Anon.: Metallic Materials and Elements for Aerospace Vehicle Structures.
MIL-HDBK-5A, U.S. Dep. Def., Feb. 8, 1966. (Supersedes MIL-HDBK-5.)
11. Shanley, F. R.: Weight-Strength Analysis of Aircraft Structures. McGraw-Hill
Book Co., Inc., 1952.
12. Shanley, F. R.: Strength of Materials. McGraw-Hill Book Co., Inc., 1957.

18
 TABLE I.- CHEMICAL COMPOSITION OF EXTRUDED Be-38Al ALLOY TUBING^1
[0.020-inch (0.508-mm) nominal wall thickness]
Nominal
outside Hpat TLot Composition, percent byy weight
Condition diameter Heat r^ f f &
.____number number
in. mm Be BeO A1203 Fe Si Mg C Al
0.2500 6.35 gg_g gg g
33^^
As-extruded
.6875 17.46 ^ ^ ^ ^ ^^^ ^^g
^ ^ ^^^
0 4^7^
438-H 65-10 63.5 0.39 1.02 0.062 0.063 0.007 0.043 Bal.
.5625 14.29
0.2500 6.35 331-H 65-9 62.6 0.67 2.42 0.070 0.049 0.012 0.049 Bal.

Annealed -4375 ^l1

.5625 14.29 255-H 65-2 64.0 0.50 1.00 0.075 0.050 0.0035 0.049 Bal.
.6875 17.46
information furnished by supplier.

19
g TABLE II.- DIMENSIONAL TOLERANCES FOR EXTRUDED Be-38Al ALLOY TUBING
[0.020-inch (0.508-mm) nominal wall thickness]
Variation Variation definition Tolerance
Outside Size A Maximum deviation of mean diameter +/-0.010 in.
diameter /^^N a^ ^V ^Gtion along tube from speci- (+/-0.254 mm)
-*[[ )}" tied diameter: deviation is difference

Y^^ between i(AA + BB) and specified

A diameter
^--3-*^

Ovalness -^( r- Maximum deviation of diameter at any +/-0.020 in.

point along tube from specified (+/-0.508 mm)
/^N diameter: deviation is difference
-[[ jr~ between AA and specified diameter

’Wall Thickness Maximum deviation of mean wall +/-0.006 in.

thickness thickness at any section along tube (0.152 mm)
A/yA~B\\ B from specified thickness: devia-
\,\" }) ^lon ls d:ifference between
’- ^(AA + BB) and specified wall
thickness
1-----------------------------------------------------------
’Eccentricity ’Maximum deviation of wall thickness +/-10 percent of mean
/"^^
A\\ A
at any section from specified wall wall thickness
~^i thickness: deviation is difference
Y^r^ between AA and mean wall
thickness

Straightness Maximum deviation from straight at any point along +/-0.83 percent
length of tube: deviation is measured after rotating of length
finished tube through 360 and with ends resting on a
plane surface
 TABLE III.- MICROHARDNESS OF EXTRUDED Be-38Al ALLOY TUBING
[0.020-inch (0.508-mm) nominal wall thickness]

^^) <> hl
^_ )
Dn hi h^ hg h^
Condition
^ ^
KHN^oo CI ^NIQO CI KHN^ ci KHN^ ci
(a) (b) (a) (b) Ja) (b) (a) (b)
As-extruded 0.2500 6.35 204.4 4.8 188.9 7.7 197.6 4.3 212.9 5.0
.4375 11.11 193.4 5.4 185.8 7.2 193.7 4.7 221.3 8.7
.5625 14.29 207.8 6.1 207.0 5.9 194.5 7.8 213.7 8.4
.5625 14.29 203.9 9.9 208.3 4.2
.6875 17.46 199.8 11.4 200.6 10.0 196.8 3.9 213.6 12.2
Annealed 0.2500 6.35 187.7 4.2 186.8 5.8 184.1 2.9 196.5 4.5
.2500 6.35 166.5 4.1 172.4 3.5 157.9 1.8 164.6 1.8
.4375 11.11 149.6 6.8 142.6 6.1 138.8 3.7 154.4 4.6
.5625 14.29 141.5 6.2 147.6 4.8 137.1 6.4 144.6 7.4
.6875 17.46 145.0 2.2 139.8 4.0 134.1 4.4 146.1 7.6
aKnoop microhardness number determined by using a 100-gram mass.
^Half-width of 95-percent confidence interval based on 10 determinations of
microhardness.

21
M
(S3

TABLE IV.- SUMMARY OF DIMENSIONAL MEASUREMENTS ON EXTRUDED Be-38Al ALLOY TUBING

[0.020-inch (0.508-mm) nominal wall thickness]
Dn D^ D^ SD Number
Condition
in. mm
D_
Dn Pn Dn Pn
1
In
t^ Wn
tn tn
S^
tn observations
Number
tubes
As-extruded 0.2 500 6.35 0.995 1.012 0.970 0.008 0.880 1.00 0.770 0.060 103 32
.4375 11.1l 4.001 1.015 .993 .005 .930 1.06 .84 .050 166 43
.5625, 14.29 1.006 1.011 1.001 .002 1.020 1.10 .96 .035 128 32
.6875 17.46 1.008 1.016 1.001 .004 1.055 1.18 .98 .045 128 33
Annealed 0.2500 6.35 0.997 1.003 0.975 0.007 0.910 0.97 0.75 0.075 28 11
.4375 11.11 .999 1.000 .996 .001 .915 .95 .88 .015 33 11
.56251 14.29 1.002 1.004 .997 .002 .965 1.02 .92 .030 30 10
_________i .6875 17.46 1.005 1.009 1.002 .002 1.010; 1.06 .94 .025 51 17
’......JIBa-

TABLE V.- SUMMARY OF ROOM-TEMPERATURE MECHANICAL PROPERTIES OF

EXTRUDED Be-38Al ALLOY TUBING
p). 0200-inch (0.508-mm) nominal wall thickness]

in.
D"
mm ksi

0.2500 6.35 21
^P
MN/m2

140
ksi

22
"tp
MN/m2

150
ksi
%
MN/m2 ksi

66.2
%
MN/m2 ksi

As-extruded
456 71.2
cy
MN/m2

491 28.9X
psi
E

106
GN/m2

199
psi

28.5x
EC

106
GN/m2

197
^ ^^
0.122 0.134 <! <!
^
.4375 11.11 23 160 25 170 65.0 448 74.7 515 ,28.7 198 28.4 196 .106 .114 <1 <1
.5625 14.29 24 170 28 200 78.8 543 81.3 561 76.9 530 28.6 197 28.4 196 .102 .111 <1 <1
.6875 17.46 24 170 26 180 81.3 561 85.6 590 78.0 538 29.0 200 28.8 199 .107 .115 <1 <1
Annealed

0.2500 6.35 19 130 17 120 50.1 346 56.8 392 55.7 384 28.7 X 106 198 28.5x 106 197 0.120 0.107 2 2
.4375 11.11 17 120 15 100 50.3 347 53.4 368 48.6 335 29.1 201 28.6 197 .112 .114 2 2
.5625 14.29 16 110 15 100 48.8i 336 60.3 416 43.8 302 30.4 210 28.2 194 .110 .120 4 5
.6875 17.46 15 100 16 110 48.6 335 54.5 376 43.4 299 29.3 202 28.1 194 .120 .117 2 3

to
00
>tl- TABLE VI.- RESULTS OF ROOM-TEMPERATURE TENSILE TESTS ON
EXTRUDED Be-38Al ALLOY TUBING
0.020-inch (0.508-mm) nominal wall thickness]

------
(a) As-extruded

Dn tv tv E
Specimen ^------ ^-------, V
m_____________ 4tp ey e
in. mm ksi MN/m2 ksi MN/m2 ksi MN/m2 psi GN/m2
1 0.2500 6.35 20 140 61.8 426 28.0 x 106 193 0.120 <1 <1
2 26 180 60.8 419 29.0 200 .120 <1 <1
3 22 150 77.4 534 80.5 555 29.9 206 .116 <1 <1
4 20 140 61.5 424 28.6 197 .132 <1 <1
Average 22 150 66.2 456 28.9 X 1Q6 199 0.122 <1 <1
5
6
0.4375 11.11 24 170 ^~- 75.0 517 28.6 X 106 \ 197 0.102 <1 <1
26 180 62.7 432 28.8 199 .104 <1 <1
7 24 170 60.8 419 28.6 197 .108 <1 <1
8 26 180 58.4 403 29.0 200 .105 <1 <1
v
9 24 170 68.0 469 28.4 196 .109 <! 1
Average 25 170 65.0 448 28.7 X 106 198 0.106 <1 <1
10
11
0.5625 14.29 25
29
170
200 77.6
~^~-
535
73.7
81.8
508
564
27.7 X 1Q6
28.0
191
193
0.103
.102
<1
<1
~^
<1
12 26 180 78.3 540 82.9 572 28.7 198 .102 <1 <1
13 32 220 82.9 572 29.3 202 .100 <1 <1
14 ’’30 210 80.4 554 85.0 586 29.2 201 .105 <1 <1
Average 28 200 78.8 543 81.3 561 28.6 X 106 197 0.102 <1 <1
15 0.6875 17.46 23 160 81.9 565 85.8 592 28.5 X 106 197 0.102 <1 1
16 29 200 82.6 570 87.9 606 28.8 199 .110 <1 <1
17 26 180 80.8 557 85.7 591 29.2 201 .106 <! 1
18 24 170 79.9 551 82.9 572 29.4 203 .110 <1 <1
Average 26 180 81.3 561 85.6 590 29.0 X 1Q6 200 0.107 <1 <1
__-w-

TABLE VI.- RESULTS OF ROOM-TEMPERATURE TENSILE TESTS ON

EXTRUDED Be-38Al ALLOY TUBING Concluded
0.020-inch (0.508-mm) nominal wall thickness]

-------------------------
(b) Annealed

Dn ty
Specimen
in. mm ksi
^p
MN/m2 ksi MN/m2 ksi
^MN/m2 psi
E
GN/m^
^tp ^u e

1 0.2500 6.35 15 100 46.7 322 50.5 348 27.6 X 106 190 0.122 1 1
2 16 110 47.7 329 58.3 402 29.0 200 .119 2 2
3 20 140 56.0 386 61.6 425 29.5 203 .120 4 4
Average 17 120 50.1 346 56.8 392 28.7 x 1Q6 198 0.120 2__2_
4 0.4375 11.11 16 110 47.5 328 51.6 356 29.0 x 106 200 0.116 2 2
5 12 83 45.4 313 47.5 328 28.8 199 .114 2 2
6 18 120 58.1 400 61.0 420 29.6 204 .107 2
Average 15 100 50.3 347 53.4 368 29.1 x lp6 201 0.112 2 2

8
7 0.5625 14.29

T
18
16
120
110 52.8 ^
364
57.7
64.1
’398
442
29.8 X
31.3
106 205
216
0.112
.107
<!
5
1
8
9 12 83 44.7 308 59.2 408 30.2 208 .112 5 6
Average 15 100 48.8 336 60.3 416 30.4 x 1Q6 210 0.110 4 5
10 0.6875 17.46 12 83 ^49.4 341 51.0 352 28.2 x 106 194 0.114 <! 1
11 18 120 49.0 338 55.6 383 29.2 201 .124 2 2
12 14 97 41.6 287 53.9 372 29.4 203 .118 3 5
13 20 140 57.8 398 57.8 398 30.4 210 .115 1
14 17 120 45.3 312 54.2 374 29.4 203 .129 4
Average 16 110 48.6 335 54.5 376 29.3 X 106
202 0.120 2 3
---------------------------------l---------------l----------’------’---------’-------------’---------’--------’----’-----’
CO
01
M
0>

TABLE VII.- RESULTS OF ROOM-TEMPERATURE COMPRESSIVE TESTS ON

EXTRUDED Be-38Al ALLOY TUBING
[0.020-inch (0.508-mm) nominal wall thickness]
(a) As-extruded

cp cy max
Specimen
in.
^
-----i-----------------,----------,-------------------
mm ksi MN/m2 ksi MN/m2 ksi MN/m2 psi
^ GN/m2
n

1 0.2500 6.35 21 140 70.3 485 72.9 503 28.6 x 106 197 0.126
2 23 160 71.6 494 83.5 576 28.2 194 .089
3_____’ 18 120 71.6 494 86.4 596 28.8 199 .188
Average 21 140 71.2 491 80.9 558 28.5 X 106 197 0.134
4 0.4375 11.11 24 170 74.2 512 79.3 547 27.9 x 106 192 0.114
5 20 140 74.6 514 88.8 612 28.6 197 .114
6 T 26 180 75.4 520 80.4 554 28.6 197 .113
Average 23 160 74.7 515 82.8 571 28.4 X K)6 196 0.114
7 0.5625 14.29 20 140 72.5 500 85.8 592 28.4 X 106 196 0.112
8 26 180 77.9 537 90.1 621 28.0 193 .112
9 27 190 80.2 553 94.3 650 28.9 199 .110
Average 24 170 76.9 530 90.1 621 28.4 x 106 196 0.111
10 0.6875 17.46 26 180 79.6 549 92.6 638 28.8 x 10^ 199 0.116
11 22 150 78.6 542 92.4 637 29.0 200 .114
12 " v 25 170 75.8 523 87.7 605 28.5 197 .116
Average 24 170 78.0 538 90.9 627 28.8 x 1Q6 199 0.115

’.
!,-.<"-’

TABLE VII.- RESULTS OF ROOM-TEMPERATURE COMPRESSIVE TESTS ON

EXTRUDED Be-38Al ALLOY TUBING Concluded
[0.020-inch (0.508-mm) nominal wall thickness]
(b) Annealed

Specimen

1
2
3
Average
-------------------------
in.

0.2500
^ mm

6.35
ksi

22
16
20
19
cp
MN/m2
150
110
140
130
ksi

66.0
49.6
51.6
55.7
"cy
MN/m2
455
342
356
384
ksi

76.6
58.9
65.8
67.1
max
MN/m2
528
406
454
463
28.0
28.2
psi

29.4 x

28.5 X 1Q6
EC

106
GN/m2
203
193
194
197
^cp

0.143
.098
.081
0.107
4 0.4375 11.11 17 120 44.2 305 56.9 392 28.4 x 106 196 0.121
5 22 150 60.2 414 69.7 481 28.6 197 .115
6 12 83 41.5 286 58.4 403 28.8 199 .106
Average 17 120 48.6 335 61.7 425 28.6 X 1Q6 197 0.114
-------------------------j-------------------------------------------------------------------------------------------------------

7 0.5625 14.29 17 120 43.6 301 55.5 383 28.8 x 106 199 0.126
8 14 97 43.6 301 52.1 359 28.0 193 .116
9 17 120 44.1 304 49.4 341 27.7 191 .119
Average 16 110 43.8 302 52.3 361 28.2 X 1Q6 194 0.120
10 0.6875 17.46 14 97 43.2 298 54.6 376 28.0 X 106 193 0.118
11 16 110 42.4 292 54.7 377 28.4 196 .110
12 T " 16 110 44.6 308 56.0 386 28.0 193 .124
Average 15 100 43.4 299 55.1 380 28.1 X 106 194 0.117
CO
-3
 ALLOY TUBING
TABLE VIIL- RESULTS OF ROOM-TEMPERATURE COLUMN TESTS ON EXTRUDED Be-38Al
["0.020-inch (0.508-mm) nominal wall thickness]
Specimen
L n Lg GC "max_ "cr^ ffmax
p
}_MN/m2 -^--
in. mm in. mm psi [GN/m2 ksi j MN/m2 ksi cr

As-extruded__
1 3.00 76.2 0.2500 6.35 18.5 27.9 X 106 192 78.3 540 82.6 570 0.95
3.00 76.2 18.5 27.7 191 80.2 553 82.6 570 .97
2
6.97 177.0 43.0 28.0 193 58.4 403 62.3 430 .94
3
6.97 177.0 42.8 29.5 203 58.4 403 62.0 427 .94
4
16.00 406.4 98.9 29.0 200 25.4 175 26.0 179 .98
5
16.00 406.4 98.2 28.8 198 23.2 160 26.4 182 .88
6
30.00 762.0 182.7 30.0 207 8.0 55 8.5 59 .94
7
30.00 762.0 184.4 29.0 200 7.8 54 8.3 57 .93
8
9 3.00 76.2 0.4375 11.11 10.2 28.2 194 90.8 626 89.0 614 1.02
3.00 76.2 10.2 27.1 187 83.7 577 89.0 614 .94
10
8.00 203.2 27.0 27.2 188 72.9 503 75.5 520 .97
11
8.00 203.2 27.0 26.2 181 72.1 497 75.5 520 .95
12
13 30.00 762.0 101.6 29.0 200 24.2 167 25.1 173 .97
30.00 762.0 102.0 29.4 203 25.4 175 25.0 172 1.02
14
8.00 203.2 0.5625 14.29 20.9 27.8 192 80.1 552 80.6 556 .99
15
8.00 203.2 20.9 27.0 186 78.3 540 80.5 555 .97
16
25.55 649.0 66.8 29.2 201 43.5 300 43.6 301 1.00
17
25.55 649.0 66.6 28.6 197 43.9 303 43.8 302 1.00
18
8.00 203.2 0.6875 17.46 17.0 28.3 195 87.0 600 84.0 579 1.04
19
20 8.00 203.2 17.0 28.0 193 86.0 593 84.0 579 1.02
27.83 706.9 58.8 29.9 206 48.7 336 49.1 338 .99
21
27.83 706.9 58.9 30.3 209 54.5 376 49.1 338 1.11
22

[
2
3
4
8’.00 20372
8.00
8.00
8.00
203.2
203.2
203.2
0.2500
.2500
.4375
.6875
6.35^ 49.1
6.35
11.11
17.46
49.1
27.2
17.0
~-----
Annealed
45.3
39.0
47.4
56.8
312
269
327
392
38.0
38.0
47.2
52.5
262
262
325
362
1.19
1.03
1.00
1.08
5 8.00 203.2 .6875 17.46 17.0 47.3 326 52.5 362_90

28
 TABLE DC.- PROPERTIES OF MATERIALS USED FOR
COLUMN-EFFICIENCY COMPARISONS

Material a>. Ec ^cy

Ibm/in3 Mg/m3 ksi GN/m2 ksi MN/m2
Aluminum (7075-T6) 0.101 2.80 10 500 72 71 493
Beryllium .067 1.85 40 000 276 42 290
Be-38Al (as-ext.) .075 2.08 28 500 197 75 517
Be-38Al (ami.) .075 2.08 28 500 197 45 310

29
 As-extruded

Annealed

1 in. (25 mm)

Figure 1.- Typical macrostructure of extruded Be-38A1 alloy tubing. l - 69- 1209

30
 1m. (25 mm)

.It • ) lot" l ...- .......,..,..

• I ~
" I .
<, -
"'1
~ ~

;~)."t ....... "," .

(a) As-extruded.

Strain gages
\

(b) Anneal ed.

~.!<i, _ ___ ..... "-'_ ... ,...--,...,.. ._

(c) As-extruded, O.250-in ch (6.35-mm) diameter.

Figure 2.- Surface defects on extruded Be-38AI alloy tubing after etching. L-69- 1210

31
 ..
Inclusion

(a) Longitudinal.

(b) Transverse.
L-69- 1211
Figure 3.- Typical microstructure of extruded Be-38AI alloy tubing in the as-extruded condition. Unetched; x 1000.

32
 , ,
. ,. 0: - ..
'Of ~

, .
. ~

(al Longitudinal.

Inclusion

(bl Transverse.
L- 69- 1212
Figure 4.- Typical microstructure of extruded Be-38AI alloy tubing in the annealed condition. Unetched; x 1000.

33
 As-extruded Annealed

Dn in. 0. 2500 04375 0.5625 06875 0. 2500 0.4375 0 5625 06875

(mm) (6.35) (11. 11) (14. 29) (17.46) (6.35) (11. 11) (14. 29) (17.46)

1. 02 j-

x
D/D L 00
99
1, T ^ ^
98

-
97

1. 2 [-

/t" 1 :’ : T I J T ^ z I
7 Twice ("
Standard^[ c><-Average
L2
r T
Range

^L- Deviation
+/-
1. 1
-^- +
A /A" L^ T
7
I T ^ ? ^
1. 2
F
^^
i /,
n "
6
T-)- l ^ T"p
s
^ s

Figure 5.- Normalized dimensional variation of extruded Be-38AI alloy tubing.

34
 ,^^3B-^
^Bff----’--’"’"---"-^

700
i

As-extruded

80 /
/?
^^^^^/ ,1
/’
^- ’"’
Tension
Compression
600

/
/ /
X Ofu
500

/ /

60 /
I 400
/ Annealed

;/ ’___ o, MN/m2
o, ksi
^-------
j / 1^ ^ 3m
40
/ /

/ / / / 200

/ 100

F 0.002 -\ E

CO
ul Figure 6.- Typical stress-strain curves for 0.6875-inch-diameter (17.46-mm) extruded Be-38AI alloy tubing.
W
os 100 -1 1m
Diameter, in. (mm)

06875(17.46) 0.5625(14.29)

X ^^ ^’ ^^’^ a 2500 (6.35)

600

^ // 0.4375(11.11)
^^^--^
/ / / i/ m

o, ksi // // / / 0. MN/m2

200

/ / 100

0
-0.002
-^
Figure 7.- Typical compression stress-strain curves for extruded Be-38AI alloy tubing in the as-extruded condition.
 100 7"

600

80

500
Diameter, in, (mm)

02500(6.35)

^____----------"---"
/
i^^’^ ^^-^^~~~^ ------’^"T’T’n^
^----_----
0.4375(11,11)
0,6875(17.46)
400

^ ^ ^^2
^ /
/
/’ /’ A^^ 3()0

---t---i---I----------------------______-_ o
o
^-
a 002 -i

03
"^ Figure 8,- Typical compression stress-strain curves for extruded Be-38AI alloy tubing in the annealed condition.
&0
oo

o MN/m2
50 x io3 100 200 300________W________500________600

-I 300
40 Tension
Compression

30 =______
-------^^g^-^---^^ 200

^^ ^^. ^^--^ ~^^^^^^ E, GN/m2

20 \ ^’\ ^ As-extruded

10 I- Annealed
\ ^K
\ ^^^ ^^ 1()0

0 20 40 60 80 100
o ksi

Figure 9.- Variation of tangent modulus as function of stress for typical extruded Be-38AI alloy tubing.
^wSSF--
’^^^S’.S^^a^^SS^s-

o MN/m2
100 200 300 400 500 600

Tension
0,4 Compression

03 ^
^
/ ^ Annealed
M /

a2
- ^ y^ ^--
-^’ As-extruded

0,1 -=~^---’--- "------

0 20 40 60 80 100
o ksi

Figure 10.- Variation of Poisson’s ratio as function of stress for typical extruded Be-38AI alloy tubing.

w
CD
 -I

,j,
c
:c
.2
>-
~
ro
«
00
'i'
<1>
CO
"0
<1>

-E
"0
::l
~

-0
Q) -0 ><<1>
-0 Q)
:::l
E '0
~
ro
Q) Ln
V>
<1>
....... C N
"0
0
E
X
Q) C E
I
V') « c ::l
t:)
« .......-I
E
~
"Vi
c
~

'"
u
"0.
>-
I-

;:::l
E
::l
en
w:

40
 en
c
:a
.3
>-
E
-;;;
<C
00
~
OJ
CO
"0
OJ

.s
"0

X
E OJ

'0
E -ci
OJ
"0
on
OJ

Lt'\ ;: "0
0
E
N ;(

. !
<C
~
:::l

C :§
~ OJ
>
.~

~
0-
E
0
u

'"
u
.5.
>-
......

~
~
:::l
.!:!'
u..

41
 E -ci

E '"
-0
:::J
-ci u
Lt'\ ~ c
co 0
N u
'"cc
C
. <I:

8
~
~
:::J
........ '"
u:::

42
 70()
100

\ \ (’ Diameter, in. (mm)

\
D
V\
\ \
\\ max -\
0
D
0-2500
0.4375
(6.35>
(11.11)
600

; 80
\
f^7
Cy\
\ \ [_
0
V
0.5625
0.6875
(14.29)
(17.46)

^E^\\
\
\\ o based on compressive
500

^\\ \_____Euler mechanical tests

60 >-\\
H\ \ \\ Range-j ---Average
400
\\\0 \
o, ksi V \ o, MN/m2

^\ \ 300
40
\
^
Engesser^^ ^iY^ 2ao

20

^^ ’^:’^,^^
’^s^^
100

0 80 120 160 200

40

^P
’i Figure 13.- Column behavior of extruded Be-38AI alloy tube columns in the as-extruded condition.

,1
^ 80 r-

\ r Diameter, in. (mm) 500

\\ \\ max
O
D
0.2500 (6.35)
0.4375 (11.11)

60 x
\ \
\ ^Euler I 0 -6875 (17.46)

0 \
\ o based on compressive
mechanical tests ^
\ \ \

^s. ^k 0 \ \ Range -----Average

o, ksi 40
\
x ^YYn \
\ \\ ^_ -_^^-__
300 o, MN/m2

Y ^Y 200
^/^. -K
20
Engesser^
’^^ ’^
0 40 80 120 160 200

^
Figure 14.- Column behavior of extruded Be-38AI alloy tube columns in the annealed condition.
 ^^y^t^SK^^S^’-^’’^’"^ --^^-^- -^^^-. -""^^^.-^^--^SSi^--^. .-.^--^--.---.-. ---,^r_^------^-^,---EW.^-----"".,--~--- -----"-1-??aa=^=f^ae..S5Ss.a^^.,,_J!Blf

S
S

I
5
i| o MN/nT
| 50 x 103 1()() 200 300 400 500 600

300
40
\ .^- Beryllium (type BL, ref. 4)
\
30
-------^:--\,-^
\ -! 200

20
^^ ’’^^.^
’""’-^^
As-extruded Be-38AI (from fig. 9)
-;
Ep GN/m

____._\ \, \ \^
.._.._"^’--.
""""’"""’--^-’’^^^^^
’^,^ \. ^-Annealed Be-38AI (from fig. 9)
^Aluminum (7075-T6, ref. 10)

0
---------------------------I
20 40
^--+-------------\____________________
60
0
80 100
o, ksi

Figure 15.- Compressive tangent-modulus curves for selected materials.

^ui
iiiriiiiiii

P_. N_
L2 m2
102 ID3 10" 105 106
10-_ |------------------i------------------------------------------------------I

// ~1
1(11

10’4 /

^> 10
CD -^/

W, O^ ^/ L3’
? ^^^y "-1
Ibm
^-^m ^^^^^^ ^^ / /

lo6-^-^^-^,/
^^^’=33.4 ^^ Diameter, in. (mm)
.^ 0 0.2500 (6.351
-7 / 0.4375 (11.11)
10 O 0.5625114.29)

^’ ^ 0.6875 il7.461

10-3

lO-8------------------------------------------- ^----------------------------------------^ -^
1C-2 10-1 10 101 102 103
P_, IM
L2 in2

Figure 16.- Mass-strength comparison of experimental and calculated column results for extruded Be-38AI alloy tubing in the as-extruded condition.
c 4.0; k 0.6.

46
I ^^^.^s-a^gSaSS-^^^^’""^^-^85"’^’ ^a^~"e^.:-^ ^s-^->--..-" -=?^^^^-.-_-."
---^ "~_’"."^^_ -^"^^^S^S’^^^"’’"^ ^’. "^’"^’’^-’^’^^^"’-as-^a^^.^^^J^II

^1
?
.,,-3
L2 ;2
^_______________^________________^________________^________________l6
Eo’
y
/

//
/^ "
T 10^ // /
CT> /^//
*-
CO
///
.^//’
M
V/
/^// in0

10-5

.-y / /’/ ^/^

.-^’^^
w.
L2

W
r
ibm

-
10’’
y /
/’’’
/^

y^/
./
/^

/y
.^

//^
y^^
^’
’^^
x^^
.y
^’/^
/^
/y<^
/
,/
./.^

-----Al 17075-T61

Be-38AI lannealedl

-----Be-38AI las-extrudedl

-----Beryllium

tP
10
H-3

^
/ ./ 10-3

10-8 ------------------------------------------------------------------------------------------------|

10’3 10’2 ID’1 100 101 102 K|3

P_, lbf_
L2 in2

Figure 17.- Efficiencies of minimum-mass round-tube columns for selected materials, c 1.0; 0.6.
>^.
-3
 ^t
NATIONAL AERONAUTICSAND SPACE ADMINISTRATION POSTAGE AND FEES PAID
NATIONAL AERONAUTICS AND
WASHINGTON, D. C. 20546 SPACE ADMINISTRATION
OFFICIAL’BUSINESS FIRST CLASS MAIL

69086 00-50 ^
"-u

^ ^^^ ’^^
A I A P-’<-
K i-’.TL ’^’
-, ,,.-,,^ r^S.

.O.^ AC T
^ ^
.,.-
’^
r n l- F
s m-

T .Lri L J
,^ C LU

l-’UblMA&lhK.
If UndeliverableSection 158
postal Manual) Do Not Return
^
"The aeronautical and space activities of the United States shall be \
conducted so as to contribute to the expansion of human knowl-
edge of phenomena in the atmosphere and space. The Administration
shall provide for the widest practicable and appropriate dissemination \
of information concerning its activities and the results thereof."
-NATIONAL AERONAUTICS AND SPACE ACT OF 1958

NASA SCIENTIFIC AND TECHNICAL PUBLICATIONS

TECHNICAL REPORTS: Scientific and TECHNICAL TRANSLATIONS: Information
technical information considered important, published in a foreign language considered
complete, and a lasting contribution to existing to merit NASA distribution in English,
knowledge.
SPECIAL PUBLICATIONS: Information
TECHNICAL NOTES: Information less broad derived from or of value to NASA activities,
in scope but nevertheless of importance as a Publications include conference proceedings,
contribution to existing knowledge, monographs, data compilations, handbooks,
sourcebooks, and special bibliographies,
TECHNICAL MEMORANDUMS:
Information receiving limited distribution TECHNOLOGY UTILIZATION
because of preliminary data, security classifica- PUBLICATIONS: Information on technology
tion, or other reasons. used by NASA that may be of particular
interest in commercial and other non-aerospace \
CONTRACTOR REPORTS: Scientific and applications. Publications include Tech Briefs,
technical information generated under a NASA Technology Utilization Reports and Notes,
contract or grant and considered an important Technology Surveys.
contribution to existing knowledge. ^
Details on the availability of these publications may be obtained from:

SCIENTIFIC AND TECHNICAL INFORMATION DIVISION

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION
Washinslon, D.C. 20546 ^

c^i?"’