THE BEHAVIOUR OF SPACE TRUSSES INCORPORATING NOVEL COMPRESSION MEMBERS

by
Gerard Andrew Roger Parke

A thesis submitted in accordance with the Surrey University the of requirements of for the degree of Doctor of Philosophy

1>) 10 -5

Department of Civil Engineering University of Surrey Guildford, Surrey

May 1988

SUMMARY

Double-layer possess load. significant However,

space their

trusses

are

structural

systems load-carrying

which

reserves

of strength

beyond their

elastic

may limit

behaviour

and ultimate

capacity

depend on the topology of the structure, the support positions, and the load-displacement response of the individual members forming the If plastic buckling of the compression members is space truss. probable exhibiting then the collapse of these structures response. ways of improving the space has may be sudden, a brittle load-displacement investigation response of

The present load-displacement trusses.

has considered square-on-square

double-layer

investigation To undertake this an analysis is c.apable of tracing the full been written which This load-displacement behaviour of these structures. been used special to study soft to yield test the response of space trusses ductile members and tension

program

non-linear program has incorporating

compression

members which

were permitted

and deform plastically. program has been undertaken their to the of to develop the

An extensive novel soft

compression Also four been

members and refine fabricated and

load-displacement space truss The model chord create these

response.

model square-on-square obtained response in which tensile showed in the that from

double-layer

have structures load-displacement structures extensive structures. tested, ductility

tested

collapse. first two the lower to of

members was

permitteb,

yielding it is

post-elastic

possible behaviour

results presented the assumptions used in the analysis load-displacement response of unusual predict valid. collapse with the collapse of behaviour of This approach may be used with behaviour other soft the novel

The experimental

in this program the soft

study to

indicate both

that the to are the either

model members,

and

double-layer confidence

space trusses, to investigate types

double-layer compression

space truss members.

or without

ACKNOWLEDGMENTS

I wish to express my sincere gratitude to Professor Z. S. Makowski for all his continuous help, advice and compelling enthusiasm which I sustained my research.

ali

also

greatly during

indebted

to

Mr D. I. work.

Retief

for

his

unfailing

assistance

the experimental

fabricated the My thanks are also due to Mr A. Smith who carefully test structures, and also to Mr P. Disney for his valued advice. on the experimental investigations.

I would also I ike to express my thanks to both Or H. Nooshin and Or their P. J. Wicks for adv Ice and encouragement g iven freely throughout this work.

The Thesis was typed by Mrs E. Ryan, whom I would like her patience and care in preparing the manuscript.

to thank for

The

financially by the Department of Civil work was supported University Engineering, of Surrey; Constrado (the Steel Construction Institute); Research Council, to and the Science and Engineering

whom I am most grateful.

CONTENTS

Page

Summary

......................................................

Acknowledgements

.............................................

ii

Contents .....................................................

INTRODUCTION .................................................

1 CHAPTER

Material,

Member And Space Truss Behaviour

CHAPTER 2 Non-Linear Collapse Analysis

Of Space Trusses

.....

70

CHAPTER 3

Methods Of Improving

Space Truss Behaviour

........

97

4 CHAPTER

Novel Force-Limiting

Device

......................

142

CHAPTER 5

The Experimental Space Trusses

Behaviour

Of Double-Layer 211

....................................

CHAPTER 6

General

discussion

...............................

326

References ..................................................

344

INTRODUCTION

Double-layer frameworks, possessing to cover

space

trusses from an

are

reticulated of

three-dimensional discrete elements,

constructed high strength large,

assembly

not restricted Javits Convention trusses vertical At elastically design stress' have

used ratios, which are frequently (Makowski, 1987). Their use is areas column-free for in the recently to roof structures, completed to weight Centre used walls. space trusses 'working are stress' normally or 'limit In a designed state' 'working in to New York form City, double-layer floors, and space both intermediate

been

and sloping present,

double-layer

to comply with either (BS449,1969; requirements design the

BS5950,1985).

load carried by a elastic probable maximum is compared to an allowable stress which is member in the structure a fixed proportion of the critical This results or member buckling. rupture causing yielding, in the same factor of safety being stress

As the loads imposed dead to both structure. on acting and applied than imposed load in the greater much normally are variations is design load, in dead this the approach undesirable variations for double-layer where the space truss roof structures, especially load effect. In imposed load forms the major portion of the total 'limit state' design, separate partial safety both the dead and imposed loads. the truss that to ensure serviceability structure loading is sufficient and also to prevent A series performs that the collapse are applied to of checks are implemented satisfactorily ultimate occurring. strength under of the factors

The strength, or load-carrying capacity of a double-layer space by the Is truss considering elastic assessed normally load-displacement response of the structure up to the point where is taken falls. Normally, the first no account member compression double-layer the strength which reserves of of any post-buckling behaviour of the structure space truss may, possess. The inelastic is usually ignored in the design process because it is considered high degree of statical indeterminacy, due to their that, double-layer space trusses, fabricated from ductile materials, will possess reserves this However, of strength in excess of their is true. not necessarily elastic capacity. The complete

load-displacement of the structure, of the

behaviour

of a space truss conditions members

depends on the. topology load-displacement structure. in a double-layer the forming

the support individual

and the

response Post-elastic

reserves of strength will only exist space truss if after the failure of the first compression member the to carry the additional adjacent members have sufficient strength load transferred into the structure from the buckled member.

the compression members in a space truss are stocky then their post-buckling, load-displacement be ductile, response will load pI ateau as the member squashes. exhibiting a steady Alternatively, if the compression members are very slender, elastic buckling load of the struts carrying a gradual reduction in the occurs exhibiting However, practical capacity of the elements.

If

considerations usually dictate that the compression members used in have intermediate a double-layer space truss slenderness and inelastically. The inelastic buckle consequently or plastic in a buckling of a compresion member is a dynamic process resulting If sudden loss of both member stability and load-carrying capacity. the double-layer member can space truss encompassing the failed absorb this release of energy then the structure may also be able to However, if the force support an increase in the imposed loading. redistribution resulting then members to fail, buckling of the complete collapse from it is the likely failed member causes further that progressive inelastic occur,
the

compression members will of the structure.

in January of 1978 of

resulting

in

The sudden collapse space truss Connecticut, the forming

steel

double-layer in Hartford, of both ultimate

the roof

the Hartford the accurate

Coliseum

USA has necessitated

determination the

load-displacement response and complete load-carrying capacity of double-layer space trusses.

The collapse analysis of double-layer space trusses has been studied by several investigators and their work has indicated that double-layer trusses are prone to progressive certain collapse by the inelastic buckling of the compression members precipitated (Schmidt, Their et a], 1979; Collins, investigations have also 1981; Smith, shown that 1984b; Hanaor 1979). prone to space trusses

progressive fall

collapse,

like without

the

Hartford

Coliseum exhibiting

roof a

structure, brittle

suddenly

warning,

load-displacement

response. was to investigate into the load-displacement response of for design purposes, any and to quantify, of present study strength this of work it tracing which the may full exist necessary the elastic Using this in these was first through to write load-displacement range of program the

The major objective means of inducing ductility double-layer post-elastic structures. a computer response behaviour the space truss, reserves To facilitate program of

capable

of a double-layer up to the collapse behaviour characteristics to of

space truss of the space trusses were studied.

structure.

collapse

incorporating

members with

pre-defined

Two methods double-layer considered the

improve

the

load-displacement

response The first

of to

space trusses effects of

have been investigated. permitting extensive

method yield

tensile

occur in the bottom instability occurred a novel soft

truss before chord members of a double-layer in any of the top chord compression members.

The second method was more radical, elastic-plastic investigation heavily with

and involved the development of member which, when loaded in compression, exhibits an load-displacement This response. ' second has considered compression soft members. the effects of replacing the most members in a double-layer The design and development space truss of the soft

stressed

the novel

testing to determine extensive member has required behaviour of both full size and model soft members. In program order to to space check the assumptions the four incorporating carefully used in

and improve the

the

analytical of and model

and double-layer permitting structures

determine trusses yield,

load-displacement soft fabricated

response members, steel

tensile

to collapse. The behaviour of the four has been extensively test structures separate monitored and the load-displacement response of each space truss recorded on video tape for information for future investigators.

have been tested

1 CHAPTER

AND SPACETRUSS BEHAVIOUR MATERIAL, MEMBER

INTRODUCTION In inelastic order to determine the collapse load and assess the

response of double-layer grids, careful consideration must be given to the strength, stiffness and stab iI ity of the individual In double-layer grids the members forming the complete structure. structure as axial forces and, consequently, the collapse of these primarily flow of tension structures can involve both the yield and plastic the instability and buckling of members and, more importantly, the collapse behaviour instability by the in behaviour, this the and members of compression characteristics to the grid configuration can and support conditions, addition first determine whether the complete structural response after yield, will be ductile or brittle. compression members. Usually axially-loaded is dominated double-layer of grids Typical double-layer compression members used in the fabrication of and the imposed external loads are transmitted throughout the

very stocky nor very slender, to take place buckling members is likely inelastically, associated with a sudden and rapid loss of member load is these buckling The of columns critical strength. including initial imperfections, to column sensitive particularly during formed manufacturing. stresses and residual curvatures grids failure are neither of these load and collapse it is to behaviour grids necessary post-collapse trace, step by step, the behaviour of the entire structure under the This will require accurate action of an increasing app] led load. Consequently, to both evaluate of double-layer the numerical modelling of both the post-buckled response of of the the compression members and the tension members. Before this post-yield characteristics

is numerical modelling can be undertaken it necessary to consider the material properties of certain steels in behaviour of typical tension to the post-yield and addition space compression members used in the fabrication of double-layer

trusses.
4

STEEL MATERIAL PROPERTIES

The most important material properties which influence the behaviour of steel tension and compression members are the modulus E, and the yield stress, ay. Values for both the modulus of elasticity and the yield stress may be obtained from tensile tests on coupon specimens or from compression tests on stub columns. of elasticity, 1.1 and 1.2 show typical stress-strain relationships for steel tested in axial tension and compression. The curve shown in Figure 1.1, obtained from a tensile specimen, exhibits a linear the specimen yields elastic relationship until and deforms Figures plastically at a constant load. After plastic deformation has occurred the material strain hardens and is capable of supporting an increase in load before final rupture. The value of stress at yield is termed the upper yield stress, ayu, and the corresponding strain at yield for a structural steel specimen is in the order of As straining progresses, the mean stress in the specimen reduces to the lower yield stress value ayl and then increases to remain substantially slightly constant at a value termed the The value of the dynamic yield stress dynamic yield stress aydby the rate of strain applied to the specimen and as the strain rate reduces to zero the yield stress reduces to a It has been limiting value, termed the static yield stress, ays. found that an annealed cold-drawn steel tube the dynamic yield stress could be as much as 15% greater than the static yield stress (Smith et a], 1979).
As the structural elongation steel of plastic the yield static steel before cut stress yield the length of increases the for different grades of

0.12%.

is influenced

for

coupon,

plastic plateau, and overall both decrease. For a typical rupture grade 50c from a hot finished seamed steel tube, the amount after yield is in the order of ten times elongation at failure

strain strain

occurring

would be greater

and the percentage overall than 20% (BS4360,1986).

The stress-strain curve shown in Figure 1.2, obtained from a tubular dynamic and stub column compression test also exhibits static yield stress values, but does not show a well defined yield Consequently the static yield stress is usually defined by point. the 0.2% proof value which leaves the specimen with a permanent set of 0.2% when the specimen is unloaded. stress
5

which is the stress

b
In Ifl L 1.3

(J

Strain

I. I. Figure Typical Stress-Strain Relationship For A Carbon Structural In Tension. The Figure Steel Coupon Tested shows a typical stressin for tested steel structural coupon relationship a carbon strain The elastic tension. modulus E may be taken as 2.06 x 10s N/MM2 for all hardening the strain modulus EST is variable while steels structural history. depending strain on prior r0yu is the upper yield stress. 6yi is the lower yield stress. 6yd is the dynamic yield stress. 6ys is the static yield stress.

tn ul a)
U,

0-2%

0.5%

Strain For A Carbon Structural Relationship Stress-Strain 1.2. Typical Figure The Figure In Compression... Stub Column Tested Steel shows a typical for structural steel stub column a carbon relationship stress-strain defined The static in compression. 6ys is usually tested yield stress by 0.2% proof which is the stress value which leaves the specimen stress 0.2%. 0.5% Alternatively of proof stress, of set permanent a with be 0.5% the total to may value of of stress at a strain corresponding used (ECCS, 1984). E is the modulus of elasticity. 6yd is the dynamic yield stress. 6yz is the static yield stress. 61P is the limit of proportionality stress.

a 0.5% proof stress, corresponding to the value of The stress at a total strain of 0.5%, may be used (ECCS, 1984). lack of a clearly defined yield point is attributed to the presence of residual stresses occurring in the tubular stub column. Alternatively To predict the strength of compression members it is preferable to determine the material yield stress from stub column compression These tests tend to be vulnerable to locallsed bulging at tests. the ends of the specimens due to the end restraint the restricting Poisson expansion of the tube in the radial direction. As a result 6ulges the compressive yield stress of the material appears of the yield stress of the material. This apparent discrepancy between tensile and compressive yield Korol (1979), stress has been invesiigated by several researchers. has indirectly undertaken a comparison by providing test data for an experimental small-diameter, the mater 1a] investigation circular, stress-strain into hollow, the inelastic tubular bending of thirteen beams. Figure 1.3 shows in both tension and to be 5-10% lower than the tensile

relationship compression for specimens cut from the same cold-formed steel tube. It can be seen from the Figure that the tensile and compressive for the 0.2% stress indicating a very stress are almost identical between the tensile and compressive yield small discrepancy stresses. Wolford, et al, (1958), have comparison, forming part of an extensive Figure behaviour of small diameter welded and seamless steel tubes. 1.4 shows a comparison between the compressive and tensile yield For the high strength structural stresses obtained by Wolford.

also undertaken a investigation into the

yield stress was found to be generally higher steels the tensile This disparity tended to than the compressive yield stress. Similar results shown in decrease for the lower strength steels. Figure 1.5, were obtained by Yeomans (1976), who found that for 114 stress was steel tubes, the mean compressive yield 5% less than the mean tensile stress. yield approximately Conversely, from a limited series of tests on samples cut from one 8 mm thick steel plate it was found that the mean compressive static mm diameter yield stress was on average 5% greater stress (D. O.E., 1977). in Static Yield Stress
in yield stress can occur in nominally These discrepancies may be due to small, 7

than the mean tensile

yield

Variations

Significant identical steel

variations members.

04

E
31

2
b

L4

ul

0-2

0-5

1-0

1-5

2-0

Btrain

6[%]

1.3. Stress-Strain Figure Compression And Tension Relationships For ' The Figure Specimens Taken From The Same Cold Formed Seamed Steel Tube. between the compression and tension stress-strain shows a comparison from one coupon and one stub column test relationships obtained specimen The tube was cold formed and not stress tube. cut from the same steel before the samples were taken. relieved dimensions: Diameter Stub = 114.3mm, or oss-section column 1979). Korol, Thickness = 3.96mm, (from,

E JE 2 tr
U)

CL

E a u
Tensile Yield Btress cry [N/mmal

1.4. Comparison Between Compressive Figure Yield And Tensile Stress For Tube Specimens. The Figure Steel between the shows a comparison for test taken yield compressive and tensile stress obtained specimens from galvanised tubes. from 9-5mm to The tubes ranged in diameter steel 'Armco, Zincgrip' 75. Omm and were made by cold forming flat and welding In addition galvanised steel rolled sheet. standard steel pipe was also (Wolford, for comparison 1958). tested et al, Key: (0.02 to 0.06%) welded galvanised A Low carbon steel tubes - 0.2% proof stress. (0.21 to 0.29%) welded galvanised * Medium carbon steel tubes - 0.2% proof stress. * Standard galvanised point. pipe - yield steel (Figure from, Ellinas, 1985). 8

local

variations the

in the steel cold the work tensile working, member length. of variations of

composition Several in

in addition stresses investigators

to changes occurring

in

heat treatment, throughout the

and to residual tensile is

have assessed yield value.

magnitude and

and compressive of particular

stress

the

Yeomans (1976)

Yeomans obtained of specimens cut groups of tubes manufactured by

and compressive yield stresses for a series from 114 mm diameter steel tubes. Three different investigation, Figure each group of tubes 1.5 summarises the

were used in the a different

process.

in yield that variations results and indicates stress of up to t 13.5% from the mean value were obtained for the welded tubes and t 16.3% from the mean yield value for the seamless tubes.

in the tensile yield stress were obtained from Small variations This tests on 3 and 8 mm thick steel plates (D. O.E., 1977). in'yIeld investigation stress, although showed that local variations in in the thick thinner than higher the plate slightly were This was considered to be the combined result of plate. in chemical differential rates of cooling and local differences small, composition.
It is apparent yield that stress significant variations in both nominally tensile do occur in steels of for manufactured tubular and identical used less

compressive strength.

In addition, grids, yield

in double-layer than the tensile practice yield of

the compressive yield Consequently stress.

members, often stress is normally the generally using

accepted the tensile

stress

assessing compression member strength must be used with caution.

Strain

Aging Strain aging is an important physical the behaviour affect significantly property

of steel which tension of steel and can compression members. When an annealed or normalized mild steel is unloaded and then immediately strained beyond the yield strain, relationship of the material exhibits reloaded the stress-strain a to the relationship form very similar obtained for a continuously However, if after the material has been strained strained material. into the strain hardening region, unloaded and then allowed to remain unstressed at room temperature for a period of approximately three months before the stress-strain reloading, relationship

4
0 0 2 ul L CL

275

300

350

400

450

Tensile

Yield

Btress

cry [N/mm']

Figure 1.5. Comparison Between Compressive And Tensile Yield Stress For Steel Tube Specimens. The Figure between the shows a comparison compressive and tensile yield stress obtained for test specimens taken from continuously The welded, electric welded and seamless steel tubes. tests were carried out on tubes of the same size with a diameter of 114.3mm and a wall thickness of 3.6mm (Yeomans, 1976). Key: a Continuously welded tubes - 0.2% proof stress. A Electrically welded tubes - 0.2% proof stress. 0 Seamless tubes - 0.2% proof stress. (Figure from, Ellinas, 1984)

10

Btrain

Figure

1.6.

Stress-Strain

Relationship

Showing

The

Effects

Of

Strain

The Figure Aging. shows the stress-strain curve of an annealed or low carbon steel Al unloaded to point normalized strained and then (curve immediately restrained a) and after aging (curve b). AY is the change in yield stress due to strain aging. AU is the change in the ultimate tensile aging. strength due to strain is the change in elongation due to strain aging. &E , (from, Baird, 1963a and b).
10

Upon reloading, the the behaviour, shown in Figure 1.6. exhibits discontinuous yielding behaviour returns and the value of the yield In addition there may also be an increase in stress increases. ultimate tensile strength and a decrease in the overall elongation to fracture. (1963a and b) in an extensive review on strain notably are also affected, aging, has shown that other properties high-temperature ductile-brittle transition temperature, the and magnetic properties. strength, fatigue strength, and electrical Baird The phenomenon of strain aging was not adequately explained (Cottrell dislocation the theory and Bilby, of publication until 1949). It is now generally of strain accepted that the effects lock to by and atoms migrating carbon and nitrogen are caused aging This theory has been reinforced by the earl jer (1944), if both carbon and that Gensamer Low showed and who work of initial discontinuous from both the steel removed are nitrogen yielding behaviour and the strain aging phenomenonare eliminated. f ix dislocations. have in is specimens steel which pronounced most aging in tension and then reloaded again in tension been pre-strained is if However the after aging carried pre-straining after aging. out in compression then the return to a well defined tensile yield Strain point is suppressed. Other characteristics of strain aging are not (Tardif 1956). Ball, In by the and mode of pre-strain affected has only a small effect on addition the amount of tensile pre-strain the change in yield stress produced by subsequent aging. However, a and aging cycle produces a greater increase in repeated straining

both yield and ultimate tensile strength than that obtained If aging (Low and is carried out only after the completion of straining, Gensamer, 1944).
BauschingerEffect

the stress-strain When a metal has been plastically strained, is affected by both the behaviour during any subsequent straining direction If a of the initial prestrain. magnitude and relative in tension and then strained steel specimen is strained plastically in compression, a decrease in compressive yield stress is exhibited In comparison with the compressive yield stress obtained from a virgin specimen. This drop in yield stress, which also occurs if a in compression and subsequently strained in specimen is prestrained tension, was first observed by Bauschinger (1886), in wrought iron, and is referred to as the Bauschinger effect.
11

Several Bauschinger be due to

specific effect the

mechanisms have been proposed and it is now accepted that this

to

explain

the

phenomenon may

positioned

pile-up of dislocations moving against randomly (Orowan, 1966). These obstacles may be foreign obstacles existing oxygen or carbon between adjacent but are more likely to be crystals.

atoms such as nitrogen, grain boundaries

is most pronounced when the secondary The Bauschinger effect is directly direction straining opposed to the prestrain direction. is in a direction When subsequent prestrain to the perpendicular original the effect can be significantly prestrain, reduced. Rolfe, the effects et a], (1968), investigated which the state-of-stress have on the Bauschinger effect. In their criteria and yield investigation test specimens were cut in the longitudinal and directions to large steel each of three plates Both longitudinal radii. and transverse in tension and compression to assess the from

transverse cold-formed

small specimens were tested As a result magnitude of the Bauschinger effect. of extensive testing no Bauschinger effect in the transverse straining direction was observed only work hardening (Rolfe, et a], 1968). However, the effect straining was observed during transverse of simple tension specimens (Chajes, et al, 1963). The discrepancy was in prestraining considered to be due to the differences conditions (Rolfe, et al, 1968). Rolfe, et al, (1968), strained their material by the curvature of wide plates so that a plane-strain condition (1963), their thin applied, whereas Chajes, et a], strained These specimens in tension creating transverse compressive strains. to a marked reverse transverse compressive strains gave rise in the compressive yield strength being for the transverse strength yield Bauschinger

Bauschinger effect resulting than the tensile greater specimens.

The effect which the prestrain direction has on the Bauschinger investigated by Pascoe (1971). In this effect has been further a 915 mm x 150 mmx 12 m-nthick strip was cut from a larger plate of high-yield low-alloy steel. The strip was strained in tension and then specimens of square cross-section were cut from the plate in directions from the strain spaced by 150 Intervals direction Compression specimens 50 mm long as shown in Figure 1.7. The and tension specimens 150 mm long were subsequently tested. results 1.8. for both tension From these and compression tests are shown in Figure it is evident that the effect results of the
12

investigation

...................
c 00
TOO

Direction

of

tensile

pre-strain.

The Figure shows the Plan For Steel Specimens. Cutting Figure 1.7. 150MM by long 300mm from for test a steel plate samples cut plan cutting to the 11.4mm The pre-straining by thick. specimen each of angle wide low-alloy The plate was made from a high-yield is also given. direction individual before cut. 1.3% the were to specimens steel and was strained C= Compression specimen. T= Tensile specimen. (From, Pascoe, 1971).

500 z 400

go*

45*

30'

0.

300

M 200 IC, O
0 0-2 0-4

Valueson curvesare anglesof test specimens to pre-straindirection. Compression of unstrainedmaterial. --------Compression Tests After 1.3% Pre-tension

0 E3 Strain SM

0.8

1-0

1-2

E E
z '400

45.

Go*

75*

go*

300
200

11

Valueson curvesare anglesof test specimens to pre-straindirection. Tension of unstrainedmaterial. ---------Tension Tests After 1.3% Pre-tension

0 FijZure 1.8.

0.2

0-4 Strain

0-6 61%)

0.8

Compression

And Tension

Tests

On Specimens

After

1.3%

in The Figure Pre-Tension. shows how pre-straining a steel specimen the subsequent tensile influences tension stress yield and compressive in is If the a re-strained pre-strained specimen of material. in compressive then reductions stress are obtained. yield compression depends The magnitude of the reduction orientation of the on the (From, direction 1971). Pascoe, to the pre-strain original specimen

13

for produced a marked Bauschinger effect Also, as the subsequent compressive loading in the same direction. direction to the prestrain of loading changed from being parallel prestraining direction, to being perpendicular Bauschinger effect decreased. It to the prestrain direction, the

tensile

is apparent from the work of Rolfe et a], (1968), Chajes et a], (1963) and Pascoe (1971) that in addition to both the type and direction of prestrain, the chemical composition of the steel has a Killed, semi-killed significant effect on its behaviour. and rimmed carbon the same yield with approximately strength, behave in markedly different ultimate strength and ductility, ways is for the same amount of pre-strain. The Bauschinger effect steels all aging may

apparent in varying degrees in all steels, whereas strain killed steels. not be present in cold-finished All material characteristics mentioned will

occur to some steel members and consequently the mode of extent in structural manufacture of both the steel, and later the steel members, plays a significant role in determining overall memberbehaviour.
MANUFACTURE MEMBER In fabricated 1986). the United from Several Kingdom steel double-layer 50c structural are used the grids are usually (BS4360, the

grade cross

43c or grade sectional

shapes

steels but due to

by hollow offered sections, Round, square and rectangular predominates. advantages available are produced billet required tubes coiled formed with and their size mode and desired of

cross-sectional

manufacture Small diameter use.

these use of hollow sections are depends on their seamless tubes a steel the to obtain piercing

by hot working a mandrel

methods which involve it externally and rolling Alternatively, both small

cross-section.

ranging

from 40 mm to

sheet steel. into a circular heated tube to

600 mm in diameter is first The sheet steel tube shape. The tube's by electrical through

and medium size from are fabricated decolled longitudinal welding. and is then hollow plates and cold edges The hot

are then complete rolled Larger cold

and joined

resistance furnaces

is heated by passing

produce circular, square or rectangular diameter tubes are fabricated from single formed into a tubular shape and joined

sections. which are arc

using

submerged

weI ding.
14

The method of manufacture has a significant influence on the overall behaviour of the member. Large diameter cold formed tubes possess complex longitudinal and circumferential residual stresses which consideration when assessing column 1976). Smaller diameter cold formed tubes also possess residual stresses which have produced reductions of up to 40% in column strength (Sherman,1971). Fortunately, double-layer from cold-formed hot-finished, space trusses are usually fabricated seamed tubes which have very small residual stress levels, resulting in negligible reductions in column strength (Dwyer, et al, 1965). TENSION BEHAVIOUR MEMBER Tension members are efficient structural elements which are The behaviour of normally in a state of uniform axial stress. loaded steel tension members is very similar to the axially behaviour exhibited by a tensile test coupon cut from the member. Differences between coupon and member behaviour may be due to differences in the end fixing the methods, but it is generally in the manufactured or of residual result stresses present fabricated must be taken strength (Ross, et a], into

steel member. Residual stresses present in steel tension members do not affect the yield or ultimate strength of the member. However, they do induce early local yielding within the member which The causes a loss of stiffness as the yield load is approached. causes early local strain hardening, and premature local yielding consequently is shortened. A range of deformation crookedness occurring within similar effect is caused by an initial the member. The axial load applied to the member induces bending followed by local strain stresses which again cause early yielding the plastic hardening.
In coupon order to assess for the difference between member and test in double-layer sections hollow

behaviour

typical

grids,. the author tested two coupon specimens. and forty the hollow sections hollow before

members used steel in tension twenty one steel

Two coupons were cut they were tested in tension. from manufactured steels were tested. strip Figure for

from each of Both square

and circular (BS4360,1986) tubes

sections

structural

grade 43c and 50c The hot finished

were manufactured resistance axial stress

from steel welding. and strain

by electrical between the

with the tube seam formed 1.9 shows the relationship one of the 2.0 metre long,

15

500 Z
400

soo

tiri

(1

200

lzo

05 .

1-5 . STRAIN

Figure 1.9. Tensile Stress Vs. Strain RelationshiD For A Grade 50C Steel Square Hollow Section. 7be experimental relationship was obtained for a 2. Om long 76.1mm by 76.1mm grade 50C square hollow tube with a in tension thickness of wall of 3. Omm tested at a straining rate 9.25 x 10- 7 IS. Cy = 451 N/mm2 ;E=2.06 x 105 N/mm2 .

Soo

400

300

200

loo

) . lu.;,

.1

1.5 . STRAIN

.0

25 .

.3

3s

TEST
Figure 1.10.

SPECIMEN STRIP 9A
For A Grade 50C Steel

LONGITUDINAL
Tensile Stress-V-Strain

RelationshiD

Coupon. The steel coupon was cut from a 2. Om long 76.1mm by 76.1mm tube with a wall thicImess grade 50C square hollow of 3. Omm. The coupon in tension was tested at a straining rate of 2.22 x 10-5/S. cy = 464 N/mrn2 ;E=2.06 x 105 N/mm2 .

16

76.1 mm x 76.1 mm grade 50c steel square hollow tubes tested in Figure 1.10 shows the axial tension. stress versus strain for one-of-the-coupons--cut-from-the-corresponding relationship -steelit is evident that the behaviour of member. From the test results the tube and the corresponding coupons is very similar although the In tube members did experience a reduction in overall elongation. for both grades of steel, the rectangular hollow sections addition, deformation than the and ductility plastic exhibited greater circular tube members.
BEHAVIOUR COMPRESSION MEMBER

Column Strength have been studies experimental loaded undertaken to determine and explain the behaviour of axially be to first The relationship empirical compression members. published was derived from experimental work on timber columns and to the square of showed column strength to be inversely proportional Numerous theoretical and This early work was the column length (van Musschenbroek, 1729). developed by Euler (1757) who produced a quantitative expression for derived by van Musschenbroek. the experimental relationship Euler's expression is given as:
P2 Ek2 L7

named the term Ek2 the qual ity factor and the term that column Euler realised fundamental his and stability Euler the use of Experimental

"moment du ressort", where E is the k2 represents a dimensional factor. strength is primarily a problem of and extended with (Lagrange, 1770).

expression showed loads were over estimated for both short and that column critical Euler had neglected the medium length columns. In his calculations At axial compression and shearing forces. effect of both direct these two omissions were considered to be the cause of the However, discrepancy between experimental and theoretical results. it was not until 1845 that it was observed that the major reason for limit of the column material this discrepancy was that the elastic by crushing, followed by the was exceeded and that column failure first possibility elastic buckling, of inelastic buckling (Lamarle, 1846).
17

work was confirmed analysis an improved theoretical Euler's work undertaken to verify

occurred

before

failure

due to

Due to

the

limitations

imposed by Euler's

attempts have been made to modify his expression theoretical strength of perfect columns which buckle or plastically.

equation several to determine the inelastically

A tangent-modulus theory was proposed in which it was assumed that the perfect column would remain straight until failure the modulus remained constant and that across the limit. When the elastic limit cross-section even beyond the elastic of the material is exceeded the material modulus decreases, and the elastic modulus E, used in the Euler equation was replaced by the This approach was rejected tangent modulus ET (Engesser, 1889). by Consid6re who in 1889 suggested that as an ideal column stressed limit begins to bend, the stress on the concave beyond the elastic relationship, side increases according to the material stress-strain decreases elastically. on the convex side and the stress ' Consequently the may be strength a perfect column of determined by using an average modulus E in the Euler equation. This is based on the assumption that the value of E lies between the modulus of elasticity E and the tangent modulus ET-

theory and Engesser acknowledged the error in his original subsequently developed an improved average modulus theory called the based on the or double-modulus theory, reduced modulus theory, work was verified by undertaking a series of tests who in addition on rectangular columns. Several expressions for the reduced modulus ER, each depending on the column cross section, were evaluated and load obtained by substituting ERs, instead of the column critical the elastic
The describing However, investigations consistently buckling the

elastic modulus, (Engesser, 1893).

tangent

modulus

and the

column

This analytical checked his results

cross-section by von Karman

modulus E, in the Euler equilibrium

theory reduced-modulus the behaviour of perfect the results indicated from that became columns several the

equation.
the accepted inelastic theory range. theory column than and to the to

in the careful

experimental

reduced-modulus with the actual values led

over-estimated loads being closer

Column strength, to the tangent this Both

modulus theory, the

modulus values. reduced in use of the reduced difficulty adopt the tangent-modulus loads of perfect columns. The difference critical loads theory

discrepancy

modulus

engineers

to calculate

plastic

buckling

between the tangent understood

18

modulus

and reduced

modulus used

was not

until

the basic

assumptions

of the reduced modulus theory were re-examined in (Shanley, 1947). inelastic In important theory column paper on an the derivation of the reduced modulus theory it was implied that the while the axial load is increased to the column remains straight in the derivation calculated To obtain critical value, after this predicted critical which the column begins to bend. load, which is greater than the is needed in order to some strain reversal increase in stiffness required beyond the

tangent modulus load, provide the additional

The strain reversal required for the column tangent modulus load. to obtain the reduced modulus load cannot occur in a straight column. Shanley reallsed that there is nothing preventing a column from bending simultaneously with increasing axial load, and reasoned that column bending will begin as soon as the tangent modulus load is exceeded. Consequently the maximumcolumn load in the inelastic range will be greater than the tangent modulus load but less than the reduced modulus load.
Perfect Compression Member Behaviour 1.11 shows the theoretical relationship Petween the

Figure

for a series of steel columns Of and end shortening stress axial Curve A describes the loaded in compression. different slenderness, for a very short steel typical column with a low characteristics L 40. These in the order of stubby columns ratio slenderness -c r has been deform the stress yield material elastically until will initiating the column cross section plastic reached throughout deformation slenderness without ratio in buckling. Columns which have a high L> 100 are long and slender the order of r before the material compression yield stress cross section. The typical B in Figure behaviour 1.11. of The is shown by curve

and buckle elastically is reached in themember these slender columns

is load by characterised a plateau range followed path post-buckling For steel by a soft and ductile unloading columns with an path. L (40 100) buckling intermediate ratio slenderness plastic -c -c r behaviour shown by curve C in figure occurs and the post-buckling 1.11 is characterised by a sharp unloading path. is This behaviour typical and types.
19

for the

compression term

members commonly used in double-layer grids, "brittle-type" is describe the used to rapid path characteristic of these member

load-shedding,

post-buckling

cry

c) L
43 ul

*i x 4
CrF

End

Shortening

Axial Stress-End Shortening Relationshii) For A Rectangular Figure 1.11. The Figure shows the axial Bar In Compression. shortening stress-end bar different for three rectangular a with slenderness relationship in low loaded Columns ratio compression. with a slenderness ratios, the material (L/R 4 40; curve A), deform elastically stress yield until buckling. Columns deformation is reached initiating plastic without (L/R buckle 100; high B), ratio elastically curve slenderness with a is reached. Columns with before the material an yield stress buckle (40 100; C), L/R intermediate ratio curve slenderness \< load shedding path. a rapid post-buckling and exhibit plastically (from, Wolf 1973).

20

it

To explain further the behaviour of perfect compression members is beneficial to consider the theoretical in stress distribution

behaviour. struts exhibiting perfect elastic-plastic 1.12 shows the distribution the stress middle at for a column which buckles elastically cross-section at a uniform stress equal to the Euler stress OE, which is less than the perfect Figure material The buckling of the column compressive yield stress ay. to occur in the column, which in turn causes a lateral deflection creates a bending moment on the section, which changes the stress distribution as shown in Figure 1.12. Equilibrium can be maintained throughout early stages of this process while the stress increases on the concave side of the buckled column. When the stress in this region reaches the material compressive yield stress the material behaviour becomes ductile and, in order to maintain between the external moment and the internal moment of resistance of the column the external axial load must decrease as the column deformation increases. equilibrium The changes in stress distribution which occur during inelastic buckling are shown sequentially in Figures 1.13, a, b and c. Prior is at yield as indicated to buckling the entire column cross-section causes the member to displace laterally and again causes a bending moment to act on the Similar to the elastic buckling procedure, column cross-section. the formation of the bending moment will try to increase the stress on the concave side of the buckled column, but as the stress is already at the yield throughout the cross-section value and must change rapidly to cannot be exceeded, the stress distribution maintain equilibrium to the level shown in Figure 1.13c. Associated to the equilibrium with this dynamic jump from the pre-buckling is a sudden decrease in the column loadequilibrium post-buckling This can be seen by considering the stress carrying capacity. distribution shown in Figure 1.13c where the compressive yield is no longer attained stress over the entire column middle These characteristic differences between elastic and cross-section. plastic contribute post-buckled buckling, which depend on the column slenderness ratio, large towards the differences in the experienced behaviour of steel
slenderness buckling

the

in

Figure

1.13a.

Any slight

disturbance

columns.
which determines initiated
21

The critical squashing

ratio is

whether

plastic

or elastic

is obtained

by equating

compression

111, E

tension

Figure 1.12. Sequential Changes In Stress Distribution During Elastic Buckling Of Columns The diagrams show the sequential changes ;., -hich in a perfect distribution stress column crossoccur in the theoretical buckling. during When the column section elastic at mid-height At the A is greater than the transition slenderness ratio slenderness deflection Lateral CEof perfect stress at the Euler column buckles the column increases the compressive on the concave side of the stress This the stress column and reduces on the convex side of the column. load. distribution is in axial with a constant change associated stress

compression

tension

[a]

Figure

1.13.

Sequential

Changes

In

Stress

Distribution

During

Plastic

The diagrams show the sequential Columns. Buckling-Of changes which in a perfect stress distribution occur in the theoretical column crossduring buckling. If the column at mid-height section plastic X is less than the transition It plastic slenderness slenderness ratio buckling in axial load. occurs with an associated reduction

22

the

Euler

buckling

stress

OE to

the material

compressive yield

stress ay. from a material column fabricated behaviour the Euler buckling perfect elastic-plastic exhibiting stress GE can be shown to be: For a perfect pin-ended
E T12 ( -V)

where L is the column length r is the radius of gyration E is the material elastic modulus. L by x gives: Denoting the slenderness ratio. r
112E Irx

equating the Euler stress

to the compressive yield

n2E x=

stress,
(3)

gives:

ay

so that

the

transition

slenderness

ratio

of

xt

is

given

by:

xt

IJIL ................ . ayy

Consequently for perfect pin-ended struts having a slenderness for is initiated by failure X<xt squashing plastic while of ratio failure is The buckling. by theoretical Wt elastic with struts 1.14. 43C is in Figure For a curve shown grade strength column 275N/mm2 transition the having of a minimum yield stress steel ay is 86. the Xt of order of ratio slenderness
Imperfect Compression Member Behaviour section considered the behaviour of perfect loaded

The preceeding

straight, axially columns which were assumed to be perfectly from a perfectly through the centrold and fabricated elastic-plastic It is beneficial to look at some of the imperfections material. present In real compression members and to consider are never how these straight the column strength. Real struts

affect and

perfectly

23

Plastic
cry

squashing

b
C, C.. 43

Elastic EULER

buckling curve

U,
0

Slenderness
Figure The 1.14. Theoretical the from X is Strenath theoretical perfect less

Ratio
Curve strength elastic-plastic 'At the than For Perfect curve material

L/r

Pin-Ended for a perfect When stress

ColumncL. pin-ended the column is reached is

Figure shows fabricated column

material. yield

slenderness throughout initiated

ratio the column by followed

mid-height the possibility

cross-section plastic of

and plastic buckling.

squashing

cry

b
U, 43

(I,
At
Slenderness Ratio A= L/r

1.15. Reduced Strength Column With Curve For A Pin-Ended Figure due to The Figure the Imperfections curve reduced shows strength for a pin-ended imperfections of column showing the combined effect initial material of loading, column out of straightness, eccentricity in hardening The greatest and residual stresses. reduction strain A is equal to the occurs when the column slenderness ratio strength 7t. transition slenderness ratio

24

loaded. they are always eccentrically In addition the steel from deviating which they are fabricated will have physical properties from the ideal elastic-plastic behaviour. The steel may not show a definite yield point and at large strains the material will strain harden. Also the physical properties exhibited may not be Residual consistent throughout the column length and cross-section. stresses extent, be present, their will magnitude depending to a large on the column cross section and fabrication process. was the f irst to investigate have on the strength of columns. loaded columns of eccentrically the effect which He considered the rectangular cross

Young (1807), imperfections behaviour of

theoretical treatise which considered section and produced the first in column strength the reduction due to eccentric loading and initial Young concluded from his investigation column curvature. column curvature decrease and initial This reduction is most pronounced at intermediate Column strength. values of column slenderness, with the greatest loss of strength The occurring at the transition slenderness ratio, when X=Xt. both eccentric presence of residual stresses also decreases column strength except for very stocky or very slender struts. Conversely, if the column material experiences strain hardening after initial yield this will enhance column strength especially at very low slenderness values deformation predominates. The combined effect where plastic of these imperfections is to decrease the column strength for most of the slenderness range as shown in Figure 1.15. The greatest overall reduction in column strength occurring at the transition slenderness ratio.
Post-buckling In order is it grids behaviour of indeterminate member, adjacent of both causes Behaviour to predict Of Compression Members the to post-elastic consider in response detail the of double-layer post-buckling

that

loading

necessary the structures a transfer the failed

compression

In members used in the structure. failure of an individual compression from the failed of forces member to If the load adjacent the members are capable any release of will remain in (1981), and Collins curve of and absorbing structure Supple the

compression carrying from the

members. increased

energy

member, then

loading. equilibrium with the external have shown that the characteristics of

post-buckling

25

compression member has a commanding effect behaviour of the structure. the Paris (1954), proposed a means of following

on the

overall

the post-buckl ing behaviour of rectangular compression members, and this successful approach has been extended by Supple and Collins (1980), to predict behaviour both the thinaccurately post-buckling of and tubes, typical thick-walled of members used in double-layer grid structures. method the axial deformation of a column during plastic buck] Ing can be considered as the sum of the deformat Ions due to direct stress and column flexure. Using the nomenclature of Figure 1.16:
LL

In this

+ Tr
Using ds = dx as:

PL

2 dy + dxi)

17 0

(ds - dx) ............

integral in equation

(5)

the

5 may be expressed

LL ! (cls - dx) = /2 2
0

NX-) dV\2 dx

on substituting

into

equation
LL2

011(
5 this

gives:

PL +2 T 0

dy dx EFX ............ x

(6)

it is necessary to assume a In order to evaluate the integral It has been shown that a good deflected shape for the column. to be approximation is achieved by assuming the column deflection sinusoldal (Lin, 1950). Then: y=a differentiating sin (7) TIX .............. (7)

equation (7) with respect to x gives: dy d-x IIIXN ]I a cosr-) L

26

ds

Column Parameters. Figure 1.16. Column parameters and dimensions used load-displacement in the derivation relationship, of the post-buckling 1954). (from, Paris,

tension

(h-g)

Stress Distribution Assumed Plastic At The Central CrossFigure 1.17. The assumed rectangular Section Of Column. plastic stress distribution for pin-ended column at the mid height a perfect cross-section, 1948). (Van den Broek, This simplified has been stress distribution (1954) in derivation the the used by Paris of post-buckled load-displacement for relationships with columns rectangular cross-sections.

27

Substituting gives:

equation

8 back

into

equation

(6)

and integrating

PL

7T

(an )2

-E -4

(9)

is it In order to el iminate the unknown displacement 'a' Then: at mid height. necessary to consider the strut equilibrium
(10)

Pa

....................

moment of resistance of the column at mid height IMI can be evaluated by considering an assumed plastic stress distribution at the central cross section of the column as shown in Figure 1.17, (Paris, 1954). From equilibrium requirements: P Al and MCZ Af
Substituting equation 6 PL

The internal

ay A

.............

(11)

ry

dA
into

.............
equation

(12)
(9) gives:

(10)

+ WE -T

2 2 (m 11

7)............

(13) of area A, depth h and width

For a rectangular b where b>h:

P=

column cross section

2g) ayb(h ..............

(14)

and

M= cyybg(h - g) .............. the

(15)

Eliminating substituting

(14) (15) from and and unknown g equations into equation (13) gives the required P-S relationship

LL

+p AE 4L

[h 112

A p CTYA

16

28

Using the approach outlined have extended thick-walled this work to obtain between the axial

above, Supple and Collins the post-buckling for thinand fixed

(1981)

relationship and The for a thin

load and end displacement both pinned

tubes with

ends.

between the axial relationship tubular strut walled fixed-end

load and end displacement is given as:

-2

A4 PL .R. +IWe L L[p Providing valid

cos (11 p w)....... ay

the tubes are not prone to local buckling Equation (17) is for tubular members with 1> 25, where R is the radius of t

the tube and t is the tube wall thickness. Paris (1954), and Collins (1981) have shown that there is good agreement between experimental in the advanced predictions especially results and theoretical both Figure 1.18 the of early portion regions. shows post-buckling the experimental and theoretical post-buckled load-versus-end From experimental investigations shortening curves for a 12.7mm square annealed, pin-ended column (Paris, 1954). The discrepancy between the theoretical and is in immediate the post-buckled region shown values experimental primarily due to the Hnear response reaction from the structure or test machine loading the column.
in a displacement test controlled follow the line AD, the slope values will machine the experimental If the test to the stiffness of the machine. of which relates If the column rigid machine was Infinitely line AD would be followed. redundant Paris, structure to the stiffness line AB instead of then the vertical For a column forming part of a larger the slope of the line AD would be proportional without the column present. by ABCD in Figure 1.18 is the area represented is tested

of the structure

has shown that

'plastic during dissipated the of magnitude of energy representative dynamic buckling' as the column undergoes a rapid change in Furthermore, he has shown that 'plastic dynamic states. equilibrium buckling' is most pronounced slenderness for columns with slenderness equal to the transition ratio.

29

40

A
X=76-3

30

Dynamic
r--l

Instability

20 m to

Data r.3 , --Test %eZ *-% c -% -. \: -*"Theory

10

Equation

16]

0 End

0-25 Shortening

0-50 [mm]

0-75

Figure

1.18.

Theoretical

And

ExDerimental

Load

Vs.

Deformation

For A Pin-Ended Square Relationship Column. This Figure shows the dynamic buckling. The experimental phenomenon of plastic valves were (1954). during by tests Paris a series of column obtained undertaken The column material hot-rolled was an annealed steel with an average The axial deflection yield stress of Cy = 239 WmM2. was measured in the direction and at the point of application of the load.

30

NON-LINEAR BEHAVIOUROF SPACETRUSSES

The complete range of behaviour of a space truss can be With a suitable to a large extent by the designer. controlled chords, compressive choice of member type and size, the tensile be allowed to yield or buckle as chords or web members may . is as the imposed load acting upon the structures appropriate, increased. Space trusses are usually designed to remain elastic when supporting an imposed load equal to the design load multiplied It may then be by a suitable minimum safety factor or load factor. because these highly are usually structures assumed that, indeterminant and will contain several under-stressed members in increases in stressed members, even further addition to the fully loading can be supported before collapse is imminent. However this have reported the Several investigators may not always be true. space trusses collapse of model square-on-square progressive the failure compression member after of the first occurring (Schmidt, et a], 1976; Collins, 1981). In addition it has also been suggested that square-on-square space trusses even may collapse (Hanaor, 1979; designed maximum capacity before reaching their This reduction in safety factor was thought to be Howles, 1985). length due to the commonpractice of using the concept of effective to design the truss compression members. This approach is only loaded if the axially and straight, are perfectly struts valid Lateral torsional buckling is excluded and buckle by bending only. is completely void if the compressive stress in the strut varies along the length and exceeds the yield stress at any This situation sway frames may arise in rigidly-jointed position. but it has been shown that in double-layer space trusses, in which length in forces the of effective used predominate, concept axial the design of the -compression members is acceptable (Massonnet, et a], 1986). the concept
Tensile Chord Yield

By allowing yield to occur in the lower chord tensile members it is truss possible to introduce a large amount of space of a into the structure, ductility post-yield provided the compression Several members can remain stable throughout the yielding process. and experimental studies into the behaviour of space analytical trusses in which tensile yield predominates have been reported.

31

Two model square-on-square double-layer space trusses have been (Schmidt, et al, 1978). fabricated The and tested to collapse nominally identical structures shown in Figure 1.19 were 1.83 metres square, 0.22 metres deep and were simply supported at each of the lower chord perimeter nodes. The upper chord, perimeter lower chord, and the web diagonal members were manufactured from 12.7 rWn outside aluminium tube which had the ends coined to fit aluminium TrIodetic nodes. All of the tensile chord members, apart from the perimeter members, were formed from 1.6 mm diameter soft diameter

iron wire. Figure 1.20 shows both the theoretical and experimental load-deflection It can relationships obtained for the structures. be seen from the figure that a large reserve of strength exists exhibited by both of the structures was terminated by buckling of the top chord compression Figure members which led to a sharp drop in load carrying capacity. 1.21 shows the theoretical and experimental compressive chord forces limit and also for one half of one truss occurring at the elastic just prior to collapse. The Figure shows that as tensile yield of occurs in the bottom chord members a significant redistribution force occurs throughout During tensile the truss. yield a is established compressive membrane force which increases the compressive in a non-linear chord forces manner as yielding increase in the external load resulted in a A four-fold progresses. increase in force in the most heavily stressed compression nine-fold In beyond the-elastic limit. This ductility

member.

the peripheral compressive chord members addition force tensile underwent a force reversal and carried a substantial just prior to collapse of the structure. The non-linear development in the of the compressive membrane force associated with yield tension chord prevents the use of the yield line method of analysis The yield I ine to determine the collapse load of the structure. approach requires that the limit moment be maintained at any point within the structure behaviour non-linear and this exhibited cannot be obtained with the marked by the compression members. The

behaviour of the two square-on-square double-layer structures shows clearly that it is possible to create a large amount of post-yield ductility However, and strength reserve within these structures. for these both of limit the elastic load was structures less than the collapse load. This difference could be significantly reduced to provide a suitable safety factor by increasing of the tensile members in addition to using a material tension exhibits a definite yield plateau and a smaller ultimate load ratio.
32

the area which In yield to

aT nil

1.19. Square-On-Square Figure Double-Layer Space Truss. The double-layer to favour space truss shown in the Figure was designed tensile chord yield with a ratio chord cross-sectional of tensile area to compressive The structure chord cross-sectional area of 1: 28. was joint. Upper-chord, at each lower-chord perimeter supported simply lower diameter chord and web members were 12.7mm outside perimeter tube and all tensile the perimeter alloy chord members, except aluminium 1.60mm diameter iron were solid members, soft wire (From, Schmidt, 1978).

Lli

10 _AMT Al

is

20

25

1.20. Theoretical And Experimental Figure Load-Deflection Behaviour. theoretical lection The Figure load-def shows the and experimental behaviour for two identical nominally trusses the space with 1.19. Pi is the test load at which truss configuration shown in Figure the first tensile Ai is the corresponding member yielded truss and deflection. The diagram shows a substantial reserve of strength after first (From, Schmidt, 1978). yield et al,

33

0'

.. I. Experimental iTheoretical 123456 [A

8-

6 L tL 4-

2.

E
0-

.0 9

I2

56

Member

Number

Chord Forces For Experimental Compressive And Theoretical Fb-, ure 1.21. Space Truss. The experimental Pin-Jointed chord forces and theoretical half for one of the structure only. are shown limit the member forces the Represent of the truss. elastic at Represent the member forces of the truss. at imminent collapse (From, Schmidt, 1978). et al,

1 970 mm t

9600

mm

Upper Web

chord members

and

Lower members

chord

Square-On-Diagonal Truss Layout. The structure was -from tubular mild steel members and simply supported at all The structure nodes. chord upper perimeter was loaded equally at each (Fromy Schmidt, nodes of the lower chord 1980a). 2_t al, Figure fabricated

1.22.

34

Another tensile (Schmidt, fabricated at all

double-layer

space truss

designed

to

chord

et from mild

before is shown yield collapse This square-on-diagonal a], 1980a). steel tubular

have substantial in Figure 1.22 space truss was

members and was simply

supported

The structure of the upper chord perimeter nodes. was loaded Figure 1.23 shows a equally at each of the in ner lower chord nodes. load-deformation typical relationship obtained for both tensile and compressive experimental As the chord to of members. Figure 1.24 shows both the theoretical and load-deflection obtained for the structure. increased, lower on the structure several However before buckling of any substantial ductility some upper chord members led of the structure and a loss was not et by as predicted They al (1980a). buckle behaviour

imposed load acting

members yielded. in the structure, occurred

in the stiffness reduction a further load carrying This behaviour capacity. analysis although that undertaken the first to the

the theoretical observed collapsed individual

by Schmidt compression

member to

at a load close

element compression

average maximum load obtained from tests, subsequent buckled compression

members failed This reduction rotation adjacent that tolerances,

test results. at loads well below 70% of the control in member capacity amount of was due to a large occurring at the end nodes caused by the collapse of members. joints Schmidt et a] (1980a) also reported to normal inserting both and members were fabricated in difficulty was encountered

compression increasing

although

Initial progressed. as construction members into the structure The maximum member force measurements were made for the structure. force was 9% of the yield lower chord tensile force and the initial maximum initial buck] ing load. forces caused compressive These by in the correlation results effect the initial chord forces force was 8% of the member were of the same sign as the loading into taken and were This slightly improved and a theoretical other factors analysis

imposed

consideration the poor

theoretical between

analysis. the that

experimental predominant

and suggested on the truss

may have

behaviour. used in the construction of this

The Bamford jointing structure, will allow

system

a certain amount of joint slip, and this joint together with excessive rotation may account for the large discrepancy between the theoretical truss and experimental behaviour. structure benefits Nevertheless possesses of allowing the experimental ductility to dominate results do show that highlights the the some post-yield tensile yield 35

which

the truss

behaviour.

r. -"

z
.D 0 -I m co

Displacement

Lmm]

Disp

lacement

[nnm]

TENSION

COMPRESSION

1.23. Load-Displacement For Tension And Relationships Figure load-displacement Members. The Figure typical Compression shows for mild steel behaviour tubular tensile members usedin and compressive 1.22. The compression the square -on-diagonal grid shown in Figure 87 had tested ratio of with slenderness end nodes a were and members (From, Schmidt, 1980a). et al,

04
r-

misplacoment

[mm]

1.24. Theoretical Load-Displacement And Experimental Figure For Square-On-Diagonal Space Truss. Curves A, B and C Relationships theoretical three different analyses of the square-on-diagonal represent in Figure 1.22. derived full Curve truss A the shown was using space behaviour 1.23. Curve B represents the as shown in Figure member load-displacement for theoretical the structure relationship allowing due to lack of fit. for initial Curve C also member forces represents behaviour for lack of fit forces but the theoretical structure allowing in compression Curves A, with an assumed 20% reduction member capacity. B and C were terminated when the first compressive chord member was (From, Schmidt$ 1980a). about to buckle et al,

36

One of the leading, commercially available systems, Space by Space Decks Ltd. of Chard, Somerset, is also Deck, fabricated designed to allow a limited amount of tensile yield to occur within becomes unstable. A the bottom chord members before the structure large square-on-square double-layer from space truss constructed Space Decks 900 zone roof units, was tested to assess the amount of before within the structure 1973). Figure 1.25 shows complete collapse details of a typical pyramid unit and a plan of the complete The structure structure. was erected in three sections and when into the space truss by complete, a predetermined camber was built the tie bars forming the lower chord tightening successively members. The structure was symmetrically loaded at the top nodes by using a system of spreader beams and cables pulling from beneath the As the imposed load was increased several bottom chord structure. f ailed in tension. members in the central region of the structure In addition some bottom chord members adjacent to columns and Figure 1.26 shows the several diagonal web members buckled. It is relationship obtained for the structure. load distribution and support evident that for this particular exhibits both a large reserve of strength condition the structure load. This will ensure an beyond the elastic limit and ductility if the large visible deflections acceptable collapse mode exhibiting load-displacement structure is grossly overloaded.
Theoretical Studies

load redistribution

that takes place (Space Decks Ltd,

CompressionChord

Collapse

depends, to a response of a space truss extent on the behaviour of the compression members considerable have compression Most space trusses will within the structure. L/r, L/r the 40 within range c members with slenderness ratios, The post-yield 4140 and consequently post-buckling
Several

these members will exhibit response when axially loaded.

investigations

a "brittle

type"

analytical

have considered

the

effect

which the individual has on the collapse dimensional the effect of

compression member post-buckled characteristics behaviour The two of the complete structure. fan" the shown in Figure post-critical 1.27 was used to study slope of the Individual (Supple, et al, 1981). the fan structure to node 1 and incrementally increased to in the individual The members. members shown in Figure and the resulting

"compression changing

compression A vertical produce a

members forming load was applied series of

failures

slope a, of the compression post-critical 1.28 was varied from 101 to 801 in 100 increments 37

18-000
Top Chord Members All Members

1200

m sqale PERIMETER ANGLE 50x 38.1 x38 1 v

TOP CHORD 50 38 1.6 35 L, Light Llrd (Standard 50 50 x6 35 L, Shear Und (Q 01,r-) V

DIA GON ALS 0 D I ubt 27-00_ , OD be b L, g ht Lg. j h 28-5o ohd bar.. Shear Und

BOTTOM CHORD MEMBERS 20 57mm 9 so lid bar

0 racinu

Iv, = mLrnr

Bottom

Chord

Members

Space 1.25. Plan Of Square-On-Square Double-Layer Figure double-layer truss was supported square - on- square space loaded at each top boundary and was symmetrically nodes behaviour is shown in load-displacement for the structure (Space Decks, 1973).

The Truss. at twelve The node. 1.26 Figure

3-0

T2
1.0
0

100

200

300 [mrn I

400

Displacement

Figure 1.26. Load-Displacement Layer SDace Truss.

Behaviour 38

For

Square-On-Square

Double-

Fi

ion

F:

48

12

20

24 -

verticEj defln. r,o.:!e 1 trnm)

1.29. Curves Collapse Figure For Compression Fan For Various Values The Figure Of(x. theoretical load-displacement shows the relationships fan structure for the compression 1.27. The curves shown in Figure and to different correspond numbers values of the compression member postbuckling 1.28 (From, 1981). Supple, slope, m, shown in Figure et al, 39

relationships are shown in Figure 1.29. From their analysis Supple, et a] (1981), showed that for low values is not abrupt, failure of a where the strut failure characteristic of the individual members did not produce a sudden collapse of the However, when the strut failure structure. characteristics sudden the structure exhibited several unstable modes. Earl ier were

theoretical

load-deflection

in detail the variation in truss work considered response due to changes in the post-buckled characteristics of the The double-layer compression members (Schmidt, et a], 1980b). square-on-square grid shown in Figure 1.30 was analysed assuming six post-buckling paths for the compression members. Figures 1.31 and 1.32 show the assumed post-buckled column behaviour and the As the corresponding response of the double-layer structure. post-buckling response of the compression members becomes more behaviour of the structure becomes critical. abrupt the post-yield However the results given in Figure 1.32 show that for this individual compression member capacities can be reduced by structure initial load before the complete almost 30% below their ultimate truss post-yield capacity falls below its initial yield capacity.
A more general theoretical model of column behaviour, -shown in Figure 1.33 has been used to assess the effect which changes in both length a and in the magnitude of the residual force plateau 0, will have on the post-yield ratio response of space trusses (Schmidt et a], 1976). The double-layer square-on-square grid shown the in Figure 1.34 1.30 was studied in this analytical investigation. shows the response of the truss due to changes in behaves symmetrically by the Figure a and 0, the of a

different

the structure assuming that loading sequence and finally symmetrical theoretical evident theoretical that

throughout formation

collapses

From consideration pattern of buckled members. load-deflection in Figure 1.34, curves given the larger the ductile region column behaviour,

of the is it the

the greater in the post-yiel d response of the truss. For values of a less than 0.5 and 0 less than 0.6, the initial Yield load for the truss represents

assumed in the ductility exhibited

for the structure. When the maximum load-carrying capacity than 0.5 and 0 is greater plastic plateau length a is greater than 0.44 the maximum load-carrying is greater capacity of the structure load capacity than the yield is obtained. and a reserve of strength the

40

121S

-..

-_. S"-S---.

"" ".

"

I
"

".. ". "" ."

"

"".

" ". ""

I
"

I
""

".. """ ". ""

""""""

I
"" .

"""_! S"S

I.
" -:

-:

;
I I I

""
"

"" .. -.; -

"

--

-:

-.

::
" ..
S.. ""S

: :i
"". ". "" "..
-

I I
--:
"".
b

..

I :
"

"". I "1 I
S..

".. ". "..

"

"

""" "".

j"".
--

I I I
--

"". ". "". "" .

-

"..

" ".

I "".. I
P. -.

E -:
--

"1
I"
-

i
" ---

I
--

E E 0 m co rl

".

:I
-. ""I !

ISS... ""S"
..

""

".. """ " .

"".. I
-

."

.I.
---

"" --:

I I.
--

.I.

.I

.. .I. -: --

.I.

.J I I

-..

---

p.

I
"". .

I
"""

I..
-. -"-"

III
"__ .

I
"
-S

. """.

._i
1B30 Top Bracing Bottom Loading members members members points

___

_. _.

__

mm

Spac Double-Layer 1.30. Plan And Elevation Of S uare-On-Square Figure The structure Truss. at all boundary nodes and was supported was simply lower loaded Member properties: at four chords nodes. symmetrically 2.28MN; lower AE AE = 2.52MN; web members, chord, = chord, upper 1980b; Schmidt, 1976). AE = 2.76MN (From, Schmidt, et al, et al,

41

1.0

0.5

0123456

8 8y
1.31. Theoretical ComDression Figure Member Load-Dis-Dlacement The idealised Relationships. compression member load-displacement by in form, is the are shown non-dimensionalised relationships where at buckling member end displacement and Py is the maximum member load. have been used to investigate These six relationships the theoretical double-layer the in Figure 11.30 of grid shown response 1980b). (From, Schmidt, et al,
15

1 Q QY

05

A Ay
1.32. Theoretical Load-Displacement Figure Relationships For A Double Layer Grid. Square-On-Square The Figure shows the theoretical load-displacement for the structure 1.30. relationships shown in Figure A to The lines E correspond to the theoretical marked assumed 1.31. load has compression The truss member behaviour shown in Figure by dividing been non-dimensionalised by the theoretical truss peak load, from compression Qy, derived D of Figure 1.31. The member behaviour deflection A has also been non-dimensionalised central by vertical dividing by Ay, is the displacement which member of compression behaviour (From, Schmidt, D at buckling 1980b). et al, 42

1.0
p py

0.5

01111
0246

10

%Y

Behaviour. Member Of Compression Model Theoretical General 1.33. Figure the theoretical member the compression of general model The Figure shows the the to square-on-square of behaviour response study used load in 1.30. The in Figure of changes effect double-layer grid shown 0 Py been have level load and assessed. length, residual and (5 plateau buckling, displacement load at 8y are the maximum column and column 1976). (From, Schmidt, et al, respectively

CLY

1.0

0.5 1.5 Q O. y 1.0 0 1

'c

0.5

1
'A

0 QY I. 0

load-deflection I 1.5

curves

load-deflection

curves

Q3 = 0.5)

0-5
0 1
'A

load-def
Figure

lection

curves

(P = 2.0)

Load-Displacement Relationship. The Figure 1.34. Space Truss load-displacement behaviour truss for the theoretical the space shows for 0 (0,0.44,0.88) 1.30 three in Figure valves of each shown (0,0.5,2.0). 3 Point A three (, separate valves of with associated initial load B the the yield of structure and point represents development the of lines of collapsed members occurring represents (From, Schmidt, 1976). the structure et al, across

43

the effect which changes' in the work investigating length of the plastic plateau has on the behaviour of trusses has The investigation been reported by Stevens (1968). was limited to the study of plane trusses and concluded that in order to obtain load capacity of the structure, full the length of the potential Additional plastic plateau required in the compression member behaviour must be Figure 1.35 shows details at least several times the yield strain. used by Stevens (1968) in his analytical of the plane truss mechanism shown in the Figure the load plateau length of 5.8 times member DE would require a plastic to occur, to the yield strain to permit sufficient redistribution in For truss trusses, the capacity. plane maximum which onlyobtain in-plane behaviour is considered it is important to consider lateral be the buckling of the compression members which may, in practice, investigation. most critical design requirement. For the collapse

The investigations trusses pattern

behaviour of space related to the post-yield have all assumed the formation of a symmetrical collapse is most However, it of buckled compression members. for

in to pattern occur symmetrical collapse completely a unlikely identical, in One nominally of member a group practice. symmetrically loaded mqMbers, will always buckle before the others in a completely imperfections. This may r'esult due to larger different collapse pattern of buckled members compared with that imposed if the full be was on symmetry obtained would which structure investigations throughout the behaviour of space post-yield undertaken trusses would be erThancedby considering how sensitive the structure In is to the premature buckling of one compression member. level it to assess the of would be beneficial addition, imperfections structure, sequence.
Compression Chord Collapse One of the first Experimental grids interconnected structure solid, Studies to be tested collapse plane trusses square and joined control a to

collapse into the

sequence.

The

theoretical

alter assuming symmetry is required

to

the not

collapse enforced

mechanism of the on the collapse

double-layer

of two sets of consisted (Canon, 1969). The complete fabricated together system water from by bolts. rectangular,

orthogonal was 3.8

metres

mild

steel

sections, a load

The test

was undertaken

using

loaded by the use of was uniformly where the structure filled The load-displacement membrane. plastic relationship for the simply supported 44 structure is shown in

obtained

Figure

IIII

PC

PC

PC c

full the develop To Truss. Plane Collapse Mechanism Of Figure 1.35. DE the member truss, the compression load plane of capacity potential total load a deform exhibit the and to squash be at required would 1968). (From, Stevens, the 5.8 times yield strain of strain

Be
LJ -0 Co 0 -J

E L2
0 C

10

20

30

40

50

60

Centre

Deflection

Fmm

Load-Displacement 1.36. Experimental Figure lattice-grid The square Lattice-Grid. boundary nodes and was at all supported section. solid, mild steel 4.76mm by 19. Omm Top chords 1.6mm by 25.4mm Bottom chords (From, Cannon, 1969).

Relationship test structure fabricated from

For A Square was simply rectangular,

45

procedure which assumed perfect Extensive moment redistribution elastic-plastic member behaviour. in a failure mode was assumed to occur throughout the grid resulting similar to that expected from an orthotropically reinforced concrete The actual collapse mode involved the plastic buckling of the compression members and did not result in the formation of yield or However, the analysis. collapse lines as assumed in the theoretical slab. load was within 5% of collapse proposed theoretical Unfortunately, collapse load as shown in Figure 1.36. method of relationship load application for the structure the full collapse was unobtainable. actual due to the the

A theoretical 'yield a presented

1.36.

analysis line'

was undertaken by Canon (1969) who

load-displacement

A large model tubular steel square-on-square double-layer space truss was fabricated and tested to study both its static and dynamic The structure was a scaled behaviour (Van Laethem, et a], 1975). for the double-layer truss the symbol zone of used space model of The space truss was 0.2 in Osaka, Japan (Tsuboi, 1971). Expo 117011 four deep, corners, creating a at supported simply was and metres tested The 4.0 was grid side. metres span along each clear span of the load to distributed load applied control with a uniformly under The top the to resulting by nodes. plates attached steel structure 1.37. The Figure in is load-displacement shown relationship line buckled of one only of structure collapsed Van the through bisecting the centre. structure compression members Laethem, 2t a], (1975) undertook only a linear analysis of the . the they However, of estimate realistic a obtained structure. the to bending load by equal moment a collapse calculating collapse load for ing buck] the one pin-ended member critical product of depth the the bars buckled by the of and number of multiplied load 15% This collapse a simple procedure predicted structure. by the formation greater than the actual collapse load of the structure. behaviour of another steel square-onThe elastic-plastic (Mezzina, been truss has double-layer et a], space reported square 1975). The structure shown in Figure 1.38 was simply supported at All the corners and had a clear span of 6.0 metres. from fabricated cold-formed welded steel tubes with an were members de di ameter of 60.3 mm and a wal 1 th 1 ckness of 2.9 mm. Each outs 1 the four member end was slotted to accept welded end plates which were in turn welded together to form the nodes. To determine the collapse behaviour of the structure a single concentrated load was applied
46

r--l cl,
Z _E

tu

10

Central

Displacement

IMM

1.37. Load-Displacement Behaviour Of A Double-Layer Fioure The Figure shows both the theoretical Square-On-Square Space Tnisst and for load-displacement relationship a square-on-square space experimental The steel structure truss. Was simply supported at the four comers and loaded by attaching steel plates to the top nodes (From, Van Laethem, et al, 1975). load-displacement Key: -----theoretical response.

47

In

161

-4

-1.1

-A

750mm

T
6000mm 11500

7he Truss. Space Steel structure Square-On-Square steel Figure to four point the a subjected was and comers at supported was simply The test a using top the was undertaken load applied node. central at tubes All steel welded cold-formed load control were members system. 2.9mm. thickness 60.3mm. diameter of wall a and of an outside with 1975). (From, Mezzina, et al, 1.38.

300

200

13

100

10

15

20

25

Displacement

IMMI

1.39. Load-Displacement Figure Relationship For A Steel Square-OnY-er Space Truss* Square Double-L; The Figure load-displacement the -shows for the space truss 1.38. relationship The structure shown in Figure load of 300XN with under a central collapsed point the simultaneous tensile (16-16; chord 1.38) yield of the lower Figure the members buckling of both the upper chord compression 1.38) Figure member (9-10; (From, Mezzina, and web member (10-16), Et 1q, 1975).

48

Under to the top central, node by a series of interconnected jacks. increasing load the top central node was effectively pulled through in a locallsed failure. Yield occurred in the structure resulting adjacent lower chord members shortly followed by buckling of four upper chord and the four web members joined together at loaded node. The load- displacement relationship obtained for is shown in Figure 1.39. Unfortunately, because structure experiment the the the the

the full was performed under load control, collapse showing the post-buckling unloading behaviour of the relationship To correlate the experimental was unobtainable. and structure behaviour of the space truss an extensive series of theoretical Themembers. complementary tests were undertaken on -individual in tensile stresses were obtained and compressive static yield to the load-deflection of the full size relationship addition members, tested in tension and compression. Mezzina et a] (1975) presented two methods of analysis used to In the first behaviour of the space truss. predict the theoretical approach was used assuming perfect elasticmethod a step-by-step for the tension members and an elastic response up behaviour plastic buckling load followed by a zero residual load for to the critical the compression members. The second method used a more approximate Both theoretical linear programming yield-line approach. analyses in loads but each agreement with other, very close collapse produced they were both approximately value.
Further the experimental behaviour work of undertaken to assess in particular has space trusses square-on-square 1976). Three nominally identical, were designed any tensile tubes so that chord which compressive The yield. had an outside Figure

13% less than the experimental

collapse

post-collapse

(Schmidt, been reported et a], six by six bay, model structures chord models diameter tubes collapse were would fabricated occur from

before

aluminium thickness Triodetic

of 12.0 on and a wall to fit into

of 1.5 mm. The ends of the aluminium nodes.

were coined

1.30 shows a plan and elevation of one of the model structure giving The structures dimensions. the principal at were simply supported each'of were by the lower chord perimeter nodes and four through were then equal point load loads sharing to each structure applied deflections Controlled system. the use of for all a turnbuckle. a four-point

imposed on the

structure

The load-displacement

relationships

obtained trusses

each structure behaved similarly,

The three are shown in Figure 1.40. however, the test on the first model of one link support. The

was halted

prematurely

due to the collapse

49

40

r---l

30
20

10

10

20

30

-1

40

r--l z

30

!e1 t

20

10

10 Diplacement

20 [MM]

30

40

r--n

30

20

10

10 C31placement

20 [mm]

30

For Three Load-Displacement Behaviour 1.40. Experimental Figure Square-On-Square Double-Layer Identical The Space Trusses. Nominally behaviour obtained from displacement Figure shows the load-displacement identical The layout tests on three nominally space trusses. controlled tests grid is shown in Figure 1.30. The and dimensions of a typical were simply supported at each of the lower chord perimeter structures loads to was loaded by applying nodes and each structure equal point four lower chord nodes forming a 600mm square, synnetrically about the (From, Schmidt, et al, 1976). truss the centre of
50

the support reinstated was unloaded, and the test resumed. As the load was increased several top chord compression members buckled which resulted in a drop in load carrying capacity. As the test progressed, moderate increases in capacity were obtained structure The tests were stopped when two complete as shown in Figure 1.40. lines of collapsed compression members had formed perpendicular to each other in the top chord. Schmidt, et al, structure (1976) undertook a theoretical analysis of the ]! near

a step-by-step procedure which used using approximations to represent the real compression members behaviour The analysis predicted a theoretical as shown in Figure 1.41.

load. load 27% greater than the actual collapse This collapse discrepancy between the theoretical and actual collapse load was (1976) Schmidt, by to be due to several factos. et al, explained They noted that during assembly of the models there was a variation into the nodes. This in the ease in which the members fitted into in a nonintroduced slip some joints which resulted during distribution the even strain structures within symmetrical the shows both the theoretical in forces the a section elastic range member within and experimental taken through the top chord members across a centre I ine of the (1976) difference that the Schmidt a] et suggested also grid. elastic range. and theoretical experimental inherent from imperfections loads may have collapse in the compression members. Figure 1.42

between

resulted The compression members used in the trusses had a slenderness ratio is length of the strut 67 close to the transition and which of imperfections. they to particularly were sensitive consequently Apart in the collapse loads, the experimental behaviour is the trusses theoretical post-yield space of very and A large amount of ductility was exhibited by the trusses similar. brittle compression members exhibited even though the individual from the difference However there was no reserve of the initial collapse of the first

type load-shedding characteristics. after strength in the structure compression members. A series capacity of tests

undertaken

reported space trusses were investigated and three nominally identical models for each type. The nine models were constructed were fabricated from brass tubes and brass balls of the boundary nodes in the
51

of square-on-diagonal (Saka, et a], 1986).

assess the load-carrying been space trusses has recently Three different square-on-diagonal to

and were simply supported at each bottom layer. The models were

10 , Experimental
r-n 8

2 J
C) .J4 6

------

Theoretical

approximation

L ---------------------7

------------

2468 End Displacement

10

12 [mm]

14

Experimental And Theoretical Compression Member Behaviour. Fizure 1.41. to the Both the experimental and the assumed linear approximation behaviour in The aluminium the Figure. are member shown compression had diameter an outside members of 12.0mmv with a wall compression 67 L/r thickneSB Of 1.5mm. and had a slenderness ratio, of (From$ Schmidt, et al, 1976).

Experimental Theoretical 81

0)

LL

(0 21
ol IIIII
12

Member

Number

The And ExT)erimental Top Chord Member Forces. Theoretical Figure 1.42. forces in the Figure shows both the theoretical elastic and experimental top chord members across a centre line of the double-layer space truss lack The in 1.30. the experimental in Figure results of symmetry shown due be to slip occurring between the members and nodes to was considered force distribution in addition to the presence of an initial caused by the lack of fit of members (From, Schmidt, et al, 1976).

distributed load, to a uniformly applied subjected under to the nodes in the top layer. displacement control, Figure 1.43 The shows a plan and elevation of the three model types tested. for two model types are relationships experimental load-deflection limit load and collapse load given in Figure 1.44. Both the elastic for the trusses have been estimated using the continuum analogy Their theoretical method (Saka, et al, 1968). values for the bay model were within 15% of the experimental values four-by-four for the elastic limit load and within 23% of the experimental values load. However these theoretical for the collapse values were bay models which had a denser mesh of The experimental members in comparison with the other model types. load-displacement relationships obtained for each of the model improved for the six-by-six behaviour with a load shows a large amount of ductile (Figure 1.44). load-shedding by followed a phase plateau -gradual behaviour of the model trusses is a direct This ductile post-yield exhibited by the brass result of the load-displacement relationships compression members used in the structures. Typical load-displacement trusses lengths two the of compression members are shown relationships From Figure 1.45 it can be in Figure 1.45, (Saka, et al 1984). ', long 330 the that mm members show an extensive load plateau seen led in the in turn to the load plateau exhibited has which for load-deflection behaviour. truss The model experimental 1.44 is in Figure of the post-yield uncharacteristic shown exhibited by steel space trusses.
In occurred which the brass models of Saka, et al, (1986) elastic buckling in the compression are compression in steel

behaviour behaviour

buckl ing

members, whereas critical,

space trusses

chord

usually compression members occurs, Nevertheless, it is of the structure.. capacity of load-carrying ductile from to a post-yield obtain response a compression possible This has been achieved to a large space truss. chord critical extent, by introducing into double-layer space trusses continuous connected at the joints chord members which are eccentrically investigation (Schmidt, et a], 1977). In this bay a seven-by-seven square-on-square Figure collapse. steel 1.46 double-layer shows a plan space truss was of chords chord the tested structure to and elevation section of the Figure

plastic in resulting

of the an immediate loss

with square hollow which was fabricated Continuity hollow section web members. achieved resulting in joint eccentrities. detail of the joint

and circular members was The space at

1.47 shows a plan

and elevation truss

used in the structure.

was 8.5 metres square by 0.86 metres 53

deep and was supported

Lr) %D

Cl)

c: b

-.0 -1

-1320

Type

Lr) to

CD,

01
CD cli cli

"I

A,

. 01

. 1-1

A,

fl) (11

CD cli C%j C) C\i cli ID cli Cj CD cli cli CD

Nor 1w, I.,

> > > > >

(D

CDI m CI)

%0

1320

-1320 A Type

Type

Fioure

1.43.

Plan

And Elevation

Of Three

Model

Diagonal-On-Square

Space

The double-layer Trusses. space trusses were simply supported at the boundary to a uniformly layer bottom nodes and were subjected in load at the top nodes. The structures distributed were constructed brass with tubular members with an outside diameter of 5. Ommand a wall 15.9mm to brass spherical thickness of 0.8mm bolted nodes of diameter (From, Saka et al, 1986).

54

13 J

2468 Displacement

10 at.

12 node

14 A

16 [MMI

Load-Displacement Truss Type B.

Relationship

For The Central

Wes

A and A' Of Space

75 60 45

to 30 15

2468 Displar-E3ment

10 at

12 nacie

14 A

16 [MM

18 1

20

Nodes For The Central Relationships 1.44. Load-Dis-Placement Figure Type C. The Figure shows typical A And A' Of Space Truss for the two diagonal-on-square load-displacement space relationships The average values were obtained trusses Types B and C (Figure 1.43). identical for each configuration from three nominally models tested (From, Sakai et al, 1986).

55

r--n

30 25

Je

to 0

20

A oi X 15 lo 5

1-0

2-0

3-0

4-0

5-0

Enci

Misplaaernent

[mm]

Figure

For Brass Load-Displacement Relationships 1.45. Average The load-displacement Members. compression relationships ession lengths test from Two tube of an average seven members. obtained were were tested for the pin-ended brass tubular members. Curve A represents the average behaviour of a 220mm long member while Curve B corresponds to a member length of 330mm. The brass tubular members had an outside diameter of 5mmand a wall thickness of 0.8mm (From, Saka, et 2.1,1984).

56

60mm T-8

'

: r. 4
-

E E

Co

-I-

8540mm

Loaded

Nodes

1.46. Plan And Elevation Of Square -Orn-Square Double-Layer Figure Space The square-on-square double-layer Truss. steel space truss was simply The top and bottom of the lower chord boundary at all supported nodes. hollow members were 25mm by 25mm by 3.2mm steel rectangular chord The web members were a cross-sectional with area of 277mM2. sections diameter hollow 26.9mm outside thickness circular with sections a wall Schmidt, area of 242mM2 (From, et al, of 3.2mm and a cross-sectional 1977).
hi S in terlle bolt.

Black

bolt

SZC'rIO-4

A-A

Figure 1.47. Plan And Elevations Of Space Truss Joint. The Figure for detail the square-on- square shows the joint space truss shown in 1.46. Figure The chord members in the structure through are continuous the panel The web members do not points and pass over each other. intersect the panel large at points and joint consequently (From, Schmidt, 1977). eccentricities exist et al, 57

loaded, The boundary lower I structure nodes. was under chord a] displacement control, at twelve lower chord nodes as shown in Figure joint details, 1.46. To assess the importance of member continuity, members on the strut behaviour, three different A typical sub-structure and were tested to collapse. sub-structures load-deflection the resulting are shown in Figure relationship joint 1.48. It can be seen from Figure 1.48 that member continuity, dramatically the overall details affect members adjoining and Figure 1.49 shows both the theoretical behaviour. and experimental and the adjoining The obtained from the full size truss. theoretical relationships were obtained from a simplified non-Hnear Hnear idealisations the representing analysis assuming different relationship as shown in Figure 1.50. Curve A strut load-deflection shown in Figure 1.49 represents the upper bound truss capacity load-deflection relationship assuming a concentrical ly-loaded, theoretical the represents fixed-ended behaviour strut. Curve P, assuming a Curve S shown in Figure

concentr ical ly- loaded, pin-ended strut. behaviour the theoretical the 1.49 assuming represents behaviour and obtained from the sub-assembly tests. capacity strut For this last

analysis two different sets of strut parameters were strength of the truss resulting used to represent the orthotropic from the joint characteristics. arge The and experimental results. behaviour analysis obtained using the strut closest theoretical derived from the sub-assembly tests, considerably underestimates the, ductility To improve of the truss. maximumcapacity and post-yield the theoretical estimate a rigorous non-linear analysis is required behaviour in the the actual reflect accurately element will which between member continuity truss by modelling the complex interaction The experimental behaviour exhibited by the and joint eccentricity. the advantages of using continuous truss does, however, highlight jointing A large amount of eccentric with system. members is ductility increase in created and a substantial post-yield strength pin-jointed Further exists beyond the ultimate load predicted by a non-linear analysis. From Figure 1.49 1t is discrepancy between theoretical ev Went that there is aI

the collapse behaviour work undertaken to investigate has been reported (Collins, 1981). This of space trusses fundamental work is of value because four models were accurately tested under manufactured to minimise Imperfections and carefully full displacement control. Figure 1.51 shows the model geometry,
58

Critical

1E

strut

to

Id -Sub-Structure Layout And Dimensions.

point 2 and 3 only

Tests

Test

No. 1

50

40
"'

30

20

10

10

20

30

40

50

60

Central

Def

lect

ion

Im

M]

Ix)ad Vs. Central Plots Deflection For Sub-Structure. Total Fij7.ure 1.48. deflection load vs. the top Total central relationships are shown for tested member in the sub-structure with three different end compression influence The Figure the joint shows of member continuity, conditions. details Schmidt, of adjacent members (From, and the presence al, -et 1977). 59

400

X 300 to

200 to

ental
100

10

20

30

40

50

60

Central

Oeflection

IM

mI

Theoretical 1.49. And Experimental Truss Behaviour. The Figure behaviour is shown for truss the space theoretical and experimental 1.46. Curve A represents the theoretical truss upper shown in Figure fixed-ended ical ly- loaded, bound truss assuming capacity a concentr behaviour P Curve the theoretical represents assuming a strut. ly- loaded, the Curve S represents pin-ended strut. concentrical behaviour theoretical assuming the strut capacity and behaviour obtained test. Curve E represents from the the theoretical sub-assembly from excentr ical ly- loaded behaviour obtained continuous chord members 1979). (From, Schmidt, et al,

70

r---l

60

2
50 13
J 40

30

20

10

10
Axial Deformation

20

30

IM

M]

1.50. Figure Compression Member Load-Deformation Behaviour. Both behaviour experimental and assumed member idealisation are shown in the Figure. Curve F shows the experimental behaviour for a fixed-ended FT represent the assumed linearisation member and lines of the member behaviour. Curve P and lines PT correspond to the pin-ended member and E and ET correspond to excentrically curve loaded continuous members (From, Schmidt, 1977). et al, 60

1800 360 mmj

6 1 -, 'jl /* I 5 t `15 /* '19 * 1 14 9 l -, /* s 1 ./5 \ 6 /* 1 -2 17 "2 2 7 28

mm

1 1 b
i

39 -/33

50 55

61

-9

60 /* 54 J -7 /* 1800 * \ -- -36 .:. /30 l /* 47 /* . 152 6 9 34 51 56 -----1 254.56 mm 57 53 mm 59

- -24 -18 -12 --------------

56 LO 0

57 LO

58
C14

59 0 tr) cli

60

"t

CO

51

52

53 C)

54

55

(0 73-

(D -9

CD

46
Cl) co

47
Cl)

48
co

49
Cl) cli

50
Co

11111 iM

ir, +

, V)

i c

N 45

o-

j2CD

41
r-

42

43

44
C14 cli

++

36

37

38

39
CN

40
L

(D

81
15-ID

82

83

84

31

32

jj

154

Node And Member Numbers For Model Sauare-On1.51. Geometry, Figure Space Trusses. The diagram Square Double-Layer shows the geometry, node by Collins (1981). and member niutoers for the model designed and tested from bright This model was fabricated tubes seamless mild steel with A= 22mm2; r=3.09mm; typical E= 203000 N/MM2; of: member properties
cTy = 285 N/MM2

61

node and member numbers. due to the sudden brittle was fabricated 9.52 diameter members structure hydraulic analysis f1 rst actual Figurel. structure a load of for were was this from bright,

The first collapse

test

is of particular exhibited. steel 0.81 the tube mm.

interest The model of nominal The steel

sequence mild thickness nodes matched and

seamless, a wall steel two at

mm and with into welded loaded by located grid,

complete model displacement-controlled

actuators

The theoretical nodes 20 and 42. indicated that the undertaken by Collins, be members 62,63,78 of of the test model displacement and 79. is shown The in

members to displacement 52. As

fa 11 wouId behaviour increments

the model behav ! our was both 7800 Newtons Beyond this larger

to the were applied syrrnetr ical and I inear unt iI

nodes. induced

at each of the two loaded was obtained load successive increments of equal displacement of

load on node 20 than on node 42. As the loads increased, member 63 buckled at a load of 8864 Newtons at initial This 7956 buckling 42. 20 Newtons of and at node node as of members across the grid member 63 led to a rapid buckling in quick succession. and 79 failed members 67,71 the structure failures suffered a sudden loss of as it developed into a mechanism. capacity Due to these load-carrying

increments

losses In load-carrying rapid capacity were not in the remaining three model tests. Model four was experienced similar to model one except that the tubes used in model four were had higher a and yield stress than those used in model annealed Such one. The model was loaded at nodes 9 and 29 and under test produced behaviour 1.53. load-displacement The the shown in Figure in Figure 1.53 show that the structure results given experimental limit has a reserve of strength beyond the elastic and that the load In the grid had buckled. attained after was members ultimate The theoretical analysis undertaken by Collins (1981) also predicted a reserve structure. The diverse behaviour of these models tested by Collins (1981) shows the importance which compression member behaviour and loading conditions have on the overall behaviour of the structure. Hartford coliseum roof collapse
in January roof of 62 1978, of the steel the Hartford double-layer Hartford,

of

strength

for

this

particular

loading

and model

The sudden collapse space truss forming the

Coliseum,

EI

T6
C 0 5

C C 0

Actuator .......... -----Actuator Theoretical -

at at

Nods Nods Result

No 20 No 42

"0 0 -I

10

20

Vertidel

Deflection

trnrn)'

Load Vs. Displacement Curves For 7he LceAed Nodes Of The FiMare 1.52. Space Truss. The Figure shows both the Square-On-Square Double-Layer between the load and and theoretical relationship experimental for the loaded displacement nodes of the space truss shown in The test structure loaded at nodes 20 Figure 1.51. was symmetrically displacement 42 two hydraulic using controlled and actuators 1981). (From, Collins,

. . .......

..

A C 1. 0 C 2 ly
2 0 1 .......... Actuator Actuator Theoretical at at Node Node Result No No 9 29

V 3

.....................

12 Vereical Cleflecticm

I [rnm)

24

Figure Model

in The members used typical tube with 203000 Wmm2; cy E=

1.53. Load Vs. Displacement Curves For 7be Loaded Nodes Of 7he Double-Layer Square-Q---Square -Space Truss. The Figure shows the between the applied and theoretical relationships experimental external load and the deflections at the loaded nodes for the model square-onThe structure loaded at nodes 9 square space truss. was symmetrically hydraulic and 29 using two displacement controlled actuators.
this model were annealed properties A= of: Collins, = 375 N/mm2 (Fran, cold-drawn 22MM2 1981). seamless r=3.09mm; steel

63

USA, led to the speculation that all space trusses are by the failure of prone to failure by progressive collapse initiated into the collapse Several investigations one member. of the findings have been published and their show how the structure between failed interaction compression members and compression Connecticut, member bracing structure. The primary structure of the roof was -a square-on-square double-layer grid with plan dimensions of 92.4 by 109.7 metres and a depth of 6.46 metres. The structure was supported at four interior lower chord nodes which provided a clear column-free rectangular area of 64.0 by 82.3 metres and an overhang on each side of 13.7 The chord members which were 9.2 metres past the column supports. formed from four long steel equal-leg angles bolted were metres The top chord members were together to form a cruciform section. centre by two steel angles forming part of an bracing 1.54. The system as shown in Figure additional bracing was to halve the effective primary aim of the additional length of the compression chord members and consequently they were length of 4.6 metres (Thornton, et designed with an assurned effective braced in their al, 1984). The roof structure collapsed co-. -npletely in the early morning during a freezing rain and snow storm (Figure 1.55). Two major lines fold formed from the across the structure resulting orthogonal buckling of upper chord compression members. It has progressive been estimated that the snow and rain supported by the roof just imposed load to to equivalent was collapse an of 623N/M2 prior 9 approximately half the imposed load which should have initiated (Lev Zetlin Associates, Inc., 1978). in the structure, yield into the collapse of the of the published investigations roof structure have concentrated on the behaviour and design of the unusual cruciform compression chord members. It has been suggested flexural buckling that these members had been designed to resist buckling, instead to be of torsional which was considered (Loomis, However, a more critical et al, 1980). significantly All into both torsional. and flexural buckling of cruciform members has highlighted several discrepancies In the work of Loomis et a] (Smith, 1983). detailed theoretical investigation influence load shedding during the collapse of the

64

-uciform chord
members

memoera

1.54. Bracing Figure Additional System To The Top Chord Compression The top chord compression Member. by members were braced at mid-span The single (125 by 125 by 7.9mm) were attached two steel angles. angles to the chord members with bent plates in large resulting eccentricities between the angle bracing Thornton, and cruciform chord members (From, 1984; Smith, 1980). et al, et al,

65

in.

ei l-:::, -

"4_"

J?-

4p'.

d. ,00:

RQQLHartford. CQ115QUITI Hartford ThQ Qt FiclurQ 1 5 QOiiaj25Q on square U, S A The coliseum root was a square Connegliclil 7 by 109 92.4 dimension of a plan truss with double layer space The structure was supported at 6.46 depth metres. of a metres and tree area of column a lower providing nodes chord tour interior 13 7 metres of side each on overhang 3 an 82 and by metres 0 64 Thornton !J_al. 1984) (Smith 11,1980. pL,, supports the column past

66

Smith (1983) noted that for a given boundary condition, the load for flexural length used to determine the critical effective buckling length used to would not be the same as the effective load in torsional buckling. He extended determine the critical (Blelch, buckling of cruciform sections existing work on elastic 1952) and presented relationships and column slenderness cruciform ratios between critical buck] ing stress buckling of covering the Inelastic residual stresses.

members with and without

In order to assess the ultimate load capacity and collapse mode several independent collapse analyses of of the Hartford structure Loomis et a], (1980) undertook an the space truss were undertaken. loads dead the and of structure only supporting analysis elastic loading, this that seventy four compression members under estimated by The buckled. have effectively analysis continued was should failed from by the the members structure applying the removing buckled loads the buckling to the end of at each nodes critical four the The then seventy structure with was reanalysed members. it sixteen members and appeared that an additional removed members load. in increase buckled have external also without any should This indicates a breakdown in both compatibility and equilibrium If have been a check was should apparent which requirements, to correlate node deflections with Loomis et al the assumed compression member shortening behaviour. (1980) assumed that compression member failure would be by torsional By coincidence, buckling. their buckling and not by flexural buckling of the torsional capacity of the members calculation incorporated into the analysis of the effects consequently they. obtained an estimate of the collapse the. actual collapse load. An improved analysis of Smith and Epstein (1980). the Hartford roof Their approximated the including to the flexural buckling capacity of secondary bracing the members, system, and load close to

was presented

by

analysis assumed non-linear flexibly-braced pin-ended members with the load-shortening behaviour 1.56. This load-shortening Figure In relationship was shown which were used to obtained by analysing a series of sub-structures influence the the middle of of the angle bracing restraining assess iterative The non-linear the cruciform compression members. of the structure up to collapse by replacing buckled compression members by pairs of equivalent nodal Equilibrium forces. and compatibility requirements were satisfied analysis
67

followed

the behaviour

to

x 4

buckling oading

Chord

Shortening

Fo Vs. Relationship Load Chord Shortening Axial 1.56. Figure by assuming the Curve A is generated that Members. Compression it is B for remains member while curve elastic compression pin-jointed have fibres that yielded cross-section. all at the central assumed Point Y on curve A indicates approximately yielding where extreme fibre the begins and the intersection of curves A and B at point M indicates (Smith, load et al, 1980). attainable maximum

68

before

The each new load increment was app] ied to the structure. (1980) Epstein Smith by and predicted a collapse results obtained load close to the actual collapse load and in addition, produced collapsed lines of failed truss. the collapsed compression members in the same bays as in investigation The collapse showed the behaviour of the compression members load from rapid effects resulting

importance of the post-buckled disastrous and the possible

buckles. compression member a when shedding The premature collapse of the Hartford Coliseum space truss has to the underdesign of the principal been attributed compression incorrect from an assessment of their effective members resulting The structure did not collapse. from an overload, but from length. However, it is not unreasonable to assume that an under capacity. designed to comply if the compression members had been correctly had been the the codes of structure practice, and present with the to of overload, collapse progressive a gross rapid subjected in have a manner occured any warning, without would structure This highlights similar to the actual collapse of the roof truss. the importance of assessing both the ultimate capacity and collapse integrity the to trusses the and also reapralse need space of mode the the small the a of after one, or group removal of structure of key members.

69

CHAPTER 2 ANALYSISOF SPACE COLLAPSE NON-LINEAR TRUSSES INTRODUCTION In order to trace both the I ! near and non-] Inear behav ! our of it is necessary to adopt double-layer space trusses up to collapse, method of analysis which is capable of dealing with a versatile plastic yielding of and strain hardening as well as the with highly the non-Hnear post-buckling unloading characteristics associated These

compression memb er s and other can be met by using the standard stiffness method of prerequisites To deal with the non-Hnear behaviour the usual approach analysis. by a series of is to represent the behaviour of the structure linear approximations which can be solved in the normal way. The loads acting on the structure in small are applied external increments and within each of these increments of loading, the equilibrium,
There process balance

compatibility a process of iteration.

both

and yield

requirements

are satisfied

by

are two main formulations of which nodal approach changes. imposed calculate forces. where In stress

associated and eventually The imposed first strain

with

the

iteration Out-Ofis a

minimise formulation Phanges of uniquely

residual

strain-to-stress determine stress stress-to-strain, changes. strain is necessary to strain plastic Increments of plastic strain.

the

second changes

type

uniquely formulation, determine it of

In problems associated with ideal plasticity, formulation because Increments adopt the first uniquely stress determine cannot the

uniquely

stress system, whereas Increments determine of

ANALYSIS OF COLLAPSE METHODS

Initial Stress Method

A procedure for handling elasto-plastic problems was proposed by (1968). This non-] inear method of analysis Zienkiewicz, et a]., has been extended. to include the post-bucki ing behaviour of the (Wolf, 1973). Wolf describes the analysis of a large space members hangar in which the non-Hnear truss aircraft stress-strain equations, resulting from the non-] Inear
70

post-bucki ing behaviour

of

the

compression The 'Initial is

members are stress best illustrated

solved, in 2.1.

using terms

the

'initial

stress 'residual degree of

method'. force

methodl,, Iater

renamed the of a one

method'

freedom problem as shown in Figure

line The assumed Hnear response is represented by the straight OA whiIe the true non- 11near response 1s represented by the curve OB. Using the stiffness method of analysis the displacement is calculated by solving:
JA6 'I [K]-IIPII = ................. 2.1

1P. [K] is initial is the load vector the stiffness matrix, where 11 JA6 01 is the nodal displacement vector. and Using the strain calculated from: jAe 'I to stress approach, the member strains are

[S]JAS 'I =

.................. matrix.
from:

2.2

[B] is the strain-displacement where

In turn the member stresses

can be obtained

fal = [D]JAc 'I

....................

2.3

[D) ic itY ix In Ing is def Itut Ive the el ast matr an member const where linear for the range of the material. relationship
Another member set of member stresses can be calculated representing using the the true

relationships, constitutive Hence: behaviour of the material.

non-]! near

fal = [Dep]IAE 01 .................. Referring to Figure 2.1,

2.4

Equation 2.3 evaluates the stress at point A and Equation 2.4 determines the stress at point C. The between these two stress RI is a residual difference values, as stress, which can be treated as a residual force and visualized additional equilibrium The fictitious nodal forces, required to fulfil nodal and compatibility iteration requirements. is implemented and by solving treating again the for

procedure

nodal forces, as external residuals displacements using equation 2.1, ie: jall
The iteration until the calculated

[K]-IIRII =
with

...................
the Increments

2.5
of strain added, A new

Is continued residual

Is considered

to be negligible.

71

U) to ei
to

cr

Strain

illustrates the diagram Stress Method. The Tnitial 2.1The rigure in force terms a one of or residual method initial method stress line (from, 1,1968). The freedom Zienkiewiczrrt a. problem degree of linear true the the assumed while response OA represents is by OB. The the represented curve response non-linear by the true between the stressrgiven of point ordinate difference by the linear stress, point of given ordinate assumed C and the Ri. to a residual This stress may residual stress A, is equivalent force and treated a residual visualized as an additional be as force fulfil to required fictitious nodal nodal and equilibrium requirements. compatibility

72

load is applied and the iteration increment of external process The major advantage of this method is that the original repeated. is used throughout the analysis, stiffness elastic matrix and consequently it is not necessary to reform the stiffness matrix, or solve a new set of linear equations or at each new load increment. Unfortunately for highly at each stage of the iteration,

non-] ! near problems the procedure has Consequently, difficulties can be expected a slow convergence rate. in trusses space which the majority of members have analysing when either yielded in tension or buckled in compression. Dual Load Method To overcome this proposed difficulty Schmidt and Gregg (1980b) have the highly capable of following of typical compression members

a method which is buckling behaviour post non-linear found in double-layer space trusses. In this method called member behaviour the

'Dual

Load Method'

the

known

Is modelled using a series of linear non-linear approximations as shown in Figures 2.2 A and B. Using a plecewise is only valid technique, linearization each linear . approximation df displacements. range member specified end a within limit of the pin-jointed The elastic Is first space structure determined using the direct stiffness Assuming method of analysis. that a group of tensile members are the first members to become the yielded and that all remaining members are elastic, non-linear following load-displacement the are which characteristics members of Figure 2.2A may be effectively removed from the structure, and replaced by two nodal forces of value R,, as shown In Figure 2.3. The external imposed load may be Increased by Incrementing the load factor X until the extension of the yielded members, undergoing pure (Figure flow, have the 2.2A). To 62 reached value plastic continue members must now be given a represented by the slope of the line BC and the value of stiffness forces must change from R, to R2 (Figure the nodal residual 2.2A). analysis yielded A similar the the

approach is adopted for modelling the behaviour of the compression members. However, It should be noted that the postbuckled stiffness of the pin-ended members is negative, and
73

E R4r-13RjR2 M .j B I POSSIBLE I Ic 1--ELASTIC RECOVERY

End rTr. uRF, 2.2A

Displacement Member

Tension

RI

13 0 0 J
r-12

143
r-I

End

Displacement Member

Compression

Figure 2.2A & B.

Plecewlse Linearization

Of Tension And Compression

Members usea in ine uual Loaa metnoa. ine memuer benaviour ot botn Real is ised by a series of I inear tension and compression members Each I inear approximation is only valid over a approximations. During each linear range of end displacements. specific range the member behaviour may be modelled by applying a fictitious non-linear to using nodal force R, at each of the member end nodes in addition to the line equal a member stiffness slope of the relevant the member behaviour throughout the range. approximating

Double-Layer Space Truss With, Yielded Pin-Jointed Figure 2.3. -Tension Members. WItR the dual load method ot analysis yielded tension members are initially removed from the structure and (RI) forces by two and opposite equivalent to equal nodal replaced the yield load of the members removed.

75

by the slope of the I ines EF, FG, GH and HJ, where each is only valid for the specified value of the stiffness range of end (Figure displacements 61-62,62-63P 63-64 and 64-65 respectively represented 2.213). Using this approach, addition the of actual the value nodal of the force force in acting a member is given by the

residual

at the member ends, and the value of the member force determined loaded simultaneously from the analysis by both the of the structure imposed loads and the nodal residual forces. external

each analysis and increment of the load factor, every structure must be checked to make sure that it is member in-the equilibrium and compatibility satisfying requirements dictated by both the structure and the linearization approximation of the member After may be undertaken by checking that the member displacements are within their current limits, and also that the If a with member end displacements. member forces correlate discrepancy has arisen, one or a group of members will-require a behaviour. linear from one range to the next. This required phase change phase from one region to the a straightforward shift change may entail from for range FG to range GH in Figure 2.2B, or the example next, involve may an elastic change unloading sequence or elastic phase line in dashed by the Figure. the During sequence as shown recovery the elastic unloading phase the member stiffness reverts or recovers linear pre-buckling. value. back to its initial
Whichever . the of the phase changes are occurring, changes in both forces to and nodal residual are required Successive linear then equilibrium. analyses maintain are the deflection has until either of the structure undertaken, limits, has degenerated into a or the structure exceeded specified member stiffness The maximum load carried by the mechanism. by the maximum value of the load factor X. Supple and Collins Algorithm structure is determined

This

A modification to simplify the 'Dual Load Method' has been made In this method the non-linear range by Supple and Collins (1981). represented by a series of linear of the member behaviour is still approximations, but the need for the nodal residual forces required in the 'Dual Load Method' to maintain equilibrium and compatibility between the member and structure, adding and updating is removed. displacement and force
76

is achieved by increments at each

This

This analytical change in the load factor. method has been used to accurately predict the initial collapse behaviour of 1981). four steel double-layer In his space trusses (Collins, incremental Collins used the mathematical model shown in analysis, non-linear Figure 2.4, which consisted of nine linear phases to represent the entire tension and compression member loading and unloading paths. Smith's Algorithm Another non-Hnear does Force Method', inversion method of analysis the not require which, like the 'Residual

stiffness force system of fictitious nodal forces is created, which, when applied to the original structure along with loads, produces a set of internal forces and the real external displacements that correspond to the required non-linear behaviour In this the members. method the member non-linear of are modelled characteristics technique shown in Figure 2.5. The full four distinct using the stepwise linearization

of the structure A residual Smith (1984a).

repeated formation and matrix has been published by

member response shown in the Figure is represented by linear elastic range A phases which cover the initial linear elastic

to B, a constant force plateaux B to C, a reversing down J. E F to D to step sequence and a phase

increasing the external The analysis is implemented by first load, until the first member, or group of members, reaches the end linear initial The external load is then the elastic range. of Increased by incrementing the load factor X by a pre-determined to cause any of the members within the amount just sufficient structure to change regime. The required increase in X, represented by &X, is determined by a series of iterations. to that adopted for the 'Dual Load Using a procedure similar Method', the internal force in each of the members in the structure can be represented by the expression: {P) = {Poj + {b); k + [A]{Rl internal is (P) of a vector where
{Poj is a vector of initial coefficient

{Rj ................... member forces.

member forces. vector

2.6

is an influence

of member forces.

[Aj

is an influence

coefficient

matrix

containing

values of

77

Member Idealisation. The Figure shows the nine linear Figure 2.4. approximations used to repF-e-sent member load-displacement behaviour (from, Collins 1981).

78

Used To Model Tension And Stepwise Linearization ; Ijurre 2 5. full Smith tfie Behaviour. Member member modelled on r ompress; linear B the A to by four distinct phases. represents response linear elastic range, B to Ca constant force plateaux with initial Ea linear D to hardening, reversing elastic or elastic no strain Ja down (from, F to Smith, 1984). step and phase sequence recovery ); numerical model ( ); step down Physical model (-

79

the member forces due to f ict it ious un it res ! dual forces appl ied at the ends and directed along the axis of each of the members in turn. {R) is a vector of fictitious Equation 2.6 may be rewritten as: .................... 2.7 residual forces.

{Pj = {Poj + {b), X + [A*]{Rl [A*] = where Change in the member forces

{AP) are given by the expression: 0* ... e.

into

{AP} = {b){A, \} + [A*J{ARI ...............

In order the matrices set The containing three to determine in equation all the value 2.8 are partitioned have

2.8

of AX required

to change phase, three sets, each

the members which considered linear are:

groups

elastic

members; group

group 2, the constant force elastic

behaviour. an identical 1, the initial linear plateaux members; and

group 3, the reversing

members.

Hence, partitioning Pj)

P2 AAA

equation 2.8 gives:

A12 22 32 AO

b
b2

it Ail &x +
A21 31

&R i &R2 ******* AR 3 2.9

A23
33

&P3

bAAA 3)

For the I ! near elastic members in groups 1 and 3, an increase in AX will be associated with an increase in the internal member in the forces JAP11 and II&P31, but no increases are required 101 1 1, J, forces, hence &R. = and &R31 0 01 fictitious nodal residual For the non-] ! near members in group 2 an increase in AX must be forces in fictitious the nodal residual change a with associated JAR21 but must not produce an increase in the internal forces in 1011 1AP21 hence these members, '4 Substituting
b! -pi 6p . 31 bA 3

these requirements
A12 --32 'j-1 A22

into equation 2.9 yields:

J-b2l

Ax

2.10

which may in turn be written

as:

80

P,

AP3

C3

..................

2.11

where

C,
C3

bi
b3

A12

--A32

[A

1-b2l

Using equation member from either

region. particular 2 must be checked to determine

2.11 a value for Ax can be chosen, so that a group 1 or 3 will reach the limit of their In addition, the constant force members in group if

a strain reversal occurs, and if the members are elastically unloading during a positive increment in The change in strain can be assessed by using the the load factor. 7 in the member, given by: apparent force F = K6 ....................... 2.12 of the member and 6 is the

initial is the K axial where member axial displacement.

The internal forces
IP21 0

stiffness

for
f721

the members in group 2 are given

JR21

by:

.............

2.13

and hence:
I AXIIT 1AP21 + 21

fAR21 7' 1, &R21 .......

2.14

substituting

for: JAR21 = [A *J-'If-b2l 22 A'

gives:
JEP-21

[A211-1

1-b21

AX

...........

2.15

Any stress reversal occurring in any of the members In group 2, increment AX, can be determined by comparing the for a positive 7 TPfor If a stress reversal Is each member in turn. and of signs member or group of members, then they are sensed in a particular 2 to from set and added set 3, so that the actual increment removed AX can be calculated.
the required increment in the load factor [A121s, [A21J [A32J {b), the of coefficients and only are required. These coefficients are determined and stored when a member leaves When a member leaves set 2 and enters set set 1 and enters set 2. In order to evaluate

81

is retained member coefficients and matrix At the end of every load increment the load vector repartitioned. forces for the start of the {Poj is updated providing the initial next load increment. are Members in group 2 may also shed load by plastic buck] ing, and in the stepwise linearization approximation such behaviour would cause the member to 'step down' to the next lower load plateau. load is held sequence the external member is reduced constant i. e. AX = 0. The force in the particular Hence substituting for AX = 0, by changing all JR21 values only. j,dRjj = 101 and fAR31 = 101 in equation 2.9 gives: Throughout
1AP21 and API AP31 A, 12 A32 JAR21 .............. [A'*] 22 JAR21 ................. 2.16

3,

the

[A*]

the

'step

down'

2.17

and from equation 2.11

JAP*21

[A221-1

1'P21

.............
it is evident that the state members to change

2.18
incremental change in turn

From equation 1, &R21 may cause

2.17 other

which will [A* J. in 2 The the of members set change number and change alter 22 JAR21 is appl led in Increments and set 2 is modified as required. Other members may also reach the 'step down' point and they may be effectively initiated After removed by setting the 'step their down' is group until &P21 'P21 =0 for these particular members. of a 'step down' all the members are from the

completion

if determine to any members are at the end of a load plateau checked If 'step downs' have been completed, top the step. of a all or at and all member behaviour then the is compatible with load may be external repeated for the new load increment. assumed linear incremented and the the

approximation, iteration

procedure

an algorithm implementing the analysis procedure previously described and has to this accurately model the test results reported by method used Schmidt, et a], (1976), and Mezzina, et a], (1975). The main advantage of Smith's algorithm Is that the procedure is based on the initial stiffness of the structure,
82

Smith

(1984a)

has

successfully

developed

and as such does not require

any

modification

eliminates linear equilibrium

or updating the need for

to the structure stiffness the repeated solution of

This matrix. large sets of

equations, and will undoubtedly provide valuable savings in processor time when undertaking the collapse analysis of large space trusses.

ANDGEOMETRIC NON-LINEAR ANALYSIS MATERIAL the methods of analysis, previously outlined, are capable of modelling the non-linear behaviour exhibited by a double-layer limit. Inherent in a] I of these grid loaded beyond its elastic methods, is the assumption that node displacements are small in comparison with member lengths, occurred, members remain parallel and that to their after displacements directions. linear elastic have original for the All

These assumptions are fully justified analysis of space trusses, and also for the analysis provided space trusses,

the collapse is stopped

analysis of before the

However, for certain space trusses displacements become excessive. to continue is beneficial, it the analysis, past the small limit, by taking into consideration displacement changes in 141th this modification geometry occurring with the structure. implemented it is possible to trace the full behaviour of double layer space trusses, which exhibit only tensile yield throughout it is also possible to assess the collapse sequence. In addition, the load carrying capabilities of a grossly deformed space truss deteriorated into have behaving as a catenary. may which by geometry can be taken into consideration iteration theprocedure adopted to model the member modifying behaviour For the 'Dual Load Method', characteristics. non-linear the Newton-Raphson or this can be achieved by adding either Newton-Raphson iteration iterative Modified to the existing procedure, and this modified sequence is then undertaken during load. Figure 2.6 shows the each increment of the external to a single degree of freedom Newton-Raphson method applied softening system. The load increment 101 is applied to the structure 16-LI displacements are calculated using equation 2.19 node f6il [K]-l = JAPI ................ 2.19 and the Changes in

83

0 -j

AP

Displacement

Newton-Raphson Method. The figure illustrates Figure 2.6. the degree of freedom Newton-Raphson method applied To-a single (From, Zienklewicz The true system et a]. 1968). softening _t7e _Furve is by OE. For a load response represented non-linear increment of AP the initial is estimate of displacement f6jj [K] calculated using a stiffness equal to the slope of the line OD. The displacement fSj is then used to update the stiffness to [K, j which when post-multiplied by the displacement 161 yields a new nodal force equal to CB. The residual force R, is found by the ordinate of point C from AP and Is then used with subtracting [K, j to calculate the updated stiffness the next increment of displacement 1621- The displacements are updated by summing the displacement increments and the iteration is repeated until acceptably small residuals are obtained.

84

[K] f611 is The displacements the stiffness structure matrix. where the member stiffness are used to update the node positions, the structure The stiffness submatrices and finally matrix. [K, ] is post multiplied by the updated structure stiffness' matrix displacements 1611 to obtain a new force vector fFjj. For the single degree of freedom system shown in Figure 2.6, the force JFIJ is represented by the distance BC. The out of balance vector nodal residual forces are then calculated IR, 1 = JAPI IF, 1 These residual 2.21: forces are then and the corresponding displacements

from equation 2.20. 2-.20

...........
as the

treated

load, new external 1621 are obtained from equation

1621 = [K IRIJ 2.21 ............. 11-1 The node displacements are updated by add! ng 1621 to f6j, and the by is again updating the member stiffness repeated procedure The stiffness matrix. submatrices and hence, the structure IRI be terminated forces the iteration may when vector of residual has become acceptably small. by using the The Newton-Raphson method can be modified, inverse of the initial structure stiffness matrix [K]-l in equation 2.21 instead of the inverse of the updated stiffness matrix [K This important difference between the Newton-Raphson and II". Newton-Raphson methods will increase the number of Modified iterations required for convergence but, because only one inverse is it will significantly the matrix required, stiffness reduce of the overall computation process and hence the cost. FORNON-LINEAR COLLAPSE ALGORITHM ANALYSIS PRESENT Introduction
one implement Inelastic of the principal behaviour alms of the present that could study, was to the This

a non-linear structural

computer of

program

determine

a pin-jointed in Fortran

space truss.

by been the author has written program implemented to run on Prime computers. using the direct stiffness approach, Method with load steps.

77, and has been is formulated

The algorithm

an optional

and is based on the Dual Load Newton Raphson iteration between available

85

Any computer program is inevitably a compromise between the degree of accuracy of the mathematical model, and the amount of The validity of the mathematical numerical computation required. idealisation, together with the subsequent analysis, checked by physical measurements on actual structures. is described in Chapter check involving four model space structures five. Other fundamental checks have been undertaken to ensure that the results of the analysis satisfy the conditions of equilibrium. The reactions acting at the points of constraint have been checked, can only be This final

to ensure that they are in equilibrium with the externally applied Also, the internal loads. has been member force distribution in internal forces were the by a structure analysing which checked, In addition, deduced from the conditions of static equilibrium. the finite element stress analysis prograrn LUSAS (1984), has been used to provide space trusses.
Member Idealisation Several pin-jointed transmission 1965). The types structural towers, initial of structures may be accurately modelled as

check results

for

the linear

analysis

of several

These include systems. plane trusses, (Makowski, and double and triple-layer grids

is almost structures response of these independent of the flexural and torsional member properties, in addition to joint however, these parameters, rigidities, may influence the behaviour have on the post-collapse an of structures. constructed Pin-jointed from individual space trusses may be considered elements, consisting to be of structural

by hinges to end nodes. The connected perfect members single is assumed to be completely independent individual member behaviour of adjacent is members, and progressive collapse of the structure from the sequential to result failure of individual

considered members.

The non-linear members has been approximation.

behaviour idealised,

between load and for an annealed steel small diameter tube tested obtained strain, The behaviour shown, is typical of that exhibited by in tension. the tension members used in the first test model, physical in Chapter five. The non-linear described response has been represented by six linear phases shown In Figure 2.7B, which cover
86

using a Figure 2.7A shows the relationship

of by

both

tension

and compression linear sequence of

12.5

10

7.5
't: Lij

is

PERCENTAGE

STRAIN

SMALL HOLL0W TUBE (BL UE)

2.5

12

1.75

5.0

B. 5

11.6

Figure 2.7B Linear Ideal Isation Of The Load-Strain Behaviour For The Member. Tensile Figure shows both the Steej Tffe-Annealed linear T1 loatT-strain the relaElon p and six phases, expFr-imental The phase T6 allows for to T6, used to model the member behaviour. the from the of member any recovery point on or unloading elastic load-strain curve.

87

the elastic, behaviour.

load plateau, strain hardening, and unloading member Elastic unloading is possible from any point on the loading path, and this implies that the member modulus will revert Table 2.1 gives the to the original value upon reversal of strain. modulus, nodal residual forces and their values of the elastic for each of the six linear phases.

corresponding range of validity

Figure 2.8A shows the load-strain relationship obtained from a of the members used in the test compression member, typical The non- I ! near behaviour of the compression member has models. been represented by seven I ! near phases shown in Figure 2.8B, six Table 2.2 gives the of which cover the post-buck] ing response. modulus, range and values of the nodal residual forces, elastic phase number for each of the seven regions representing the full compression member behaviour. Figure 2.9A shows the load-displacement relationship obtained This for a typical model soft member loaded in compression. behaviour has been represented by ten linear ranges as shown in Figure 2.9B. Two different values of the elastic modulus have been loading and unloading of the member given to represent the elastic Table 2.3 occurring before or after the end of the load plateau. values of elastic modulus, nodal residual gives the relevant forces, member.
the models, representing post-buckling do not allow for perfect compression member behaviour, rigid strut behaviour, both an infinite as this would require post-buckling infinite force. and an stiffness, nodal residual positive negative numerical However, the tension member model does allow for plastic behaviour force zero stiffness, and a constant which requires load of the member. equal to the yield Algorithm For Non-Linear Analysis nodal The

strain

range and phase numbers for

a typical

model soft

residual

2.10 shows a simplified flow chart, outlining the key The majority of the data required steps in the present program. for the analysis of the model square-on-square double-layer space Figure was generated using the principles of formex algebra 1984). The data consisted of member topology, member In addition, member properties, type and node co-ordinates. member model parameters, joint constraints and the imposed unit loads are structures (Nooshin, read into the program from data files.
88

U 4S- 0 0

'4"3 -0

4-

f>

2 Z. 01(0
c c:

tA t, U. (A M -1. -1 EE c2. S.. ro c :3 -0 r= c) 0 (L, fu V) u CZ U 4t.. -0 r ro C) CD Ln cM c cli

CD.

M
ll Cif
eri 0

t.0

C%i C\i

8.0

LO c7)

la

ro

du . LA c2.

19 -J

IV 4J
Z fli

= 10 S- 0E m

r=
L)

00

(A m G) 40

.*0

00 -

4-

00 0) #A CM Li c tA. C re :3 ro =

X. 0
>

1SC- CA 0 OL fo 4(1) 04L) ro 4-3

11

(D0 u

C:)
X C: ) 4 x

0 Z L)

cm CD c

CD CD c C: )

CD u;

CD (l --4

C) CD

CD c

(V m >

>.. s-

c:

LU

CL

c -0

F-

0MME

Ln rl
c

0 Zxo 10 LA in

4-

40

CD CD

CD LO c9

to

3- Z
>

c: (L) . rcs . -

cm r_ (Ki CD C:) Ln

CD

(0
4-3 vi

s-

i (D

pl : - P-4

C: ! Ln

C: )

lx CD

C ! CD

m
r Zm 0

tj

00 eKj

(A

c)
c: 0
c: cm

ci >>
cu

(D -0 .C r= CL a A vi C) . 10

G) c 5- E- c: F `o c) e) 0 S- 0u r= ro o

0)

.0 E Ln

0 m-

(M u

CL
du a (L) 0c 1-- ozj .6. J rd

11

1 -1 -1

w1

89

10.0

02 :
7.5

5.0

2.5

15.0
PERcENTAGE sTRAIN

30.0

LARGE HOLL OW YELL OW A

iental -Compression Member. Load-strain Relationshi
For A Typical

10.0

7.5

c 2.5

15.0 PERCENTAGE

30.0 STRAIN

Of The Load-Strain Behaviour For A Figure 2.8B Linear Idealisation I Compres-sion Member. The Figure shows both the experim and the seven I inear phases, CI to C7, load-strain relationships behaviour. The phase C7 allows for the to member model used unloading or recovery of the member from any point on the elastic load-strain curve.
90

 "I

4-)

4-)

= 10 .(A (Ij S-

0) E (1) u C,

r': U, SC-) C:11) a 0 Ln

Q) S.. cn. (U &ko U ci 4IA "I -a CL (1)

C: (1) E-= (D
CZ 00 ci

(U

-0 'D 0 -a 1-01 a) 40 C'o S. 0 Al 4-) CY) .0 C\1 CIJ r CA >

4-)

u t0 UC:) du m CD CD CD CD
LO m

C:) CD CD C:)
LO

CD
C) C) CM

CD CD
-4 m -4 .. cu

C)

de 0.
C Z

wo(1)
E
ZE
AI

0 oil

(n Ln

a2

--4

-0

4-3

fli
fli

fli
Z

C: -a
(D .-0 40 'AJ (A

de

'A .,ts x 0

4-

0 oc$ rfo C: (:) a) E> 10 Ln 0 0U

CDL CL O'D

=3 Zu

r 4_3 , 00 ra 0 cl) > 'm

F-

%I_ (1)

4)

CD W FF x CD
C: + c

(D

CD x Co
%2 4.0
gt CM

: X:

P
Q) CD

t A:
03 Lr)
cn 9

C>
Pli

C: )

CL .EC:

rl:

m (2, b
9-4 1

c;

Ln

-4

cz +

S-

to

LU

C:

C: (2) C: >> 0 Q) V) 'o 4(1) tA -0 W=C:

L- 4(1) 0 O u (U CD > S- CU
Lr) r. -4

(L) 4-3

4-)

r--i c;

cm c;

-e

CD c; le

C:) c; le

;z0E
to

WE -0 E=(1)

C: u (A (1) S- 'a 0-4 eo 0 C= .- -C= 0 CDtn (2) b al (1) (V 4r -0 0. SE 4--o CD.

c C: ) vi

tn "i CD

r. C*,! CD

Kt

M Lr)

C:) c;

C) 4A
(2.

E CM'o cu m CU 00 ol U. 9 "J C: C: -a
"0 410 0 u a ca. :3
>-) 4-) -0 eO

CL

1 c7) t. (D c2 .
Z= * fA s=3 U

U.
-I-) tA

4-J

;;

E o

(1) 10 = 4-)

40=

(ij -he 3: u
X)
(D 4_3 Al tA 0 (A .CL 1: u 00 0) V) ,1o 0 4-

ro 0 F

S-

Ca

0 0 -(U > U fil

.0L - t; -- L

LLJ

4-1 '.0

91

Ltj
-j

6
1

-j

5.0

10.0
PERCENTAGE STRAIN

15.0

Flaure R-emB er

2.9A.

Experimental

Load-Strain

Relationship

For A Soft

I
Figure 2.9B.

5.0
PERCENTAGE

10.0
STRAIN

15.0

L inear Ideal isat ! on Of The Load-Strain

uf--A---soft member. ine tigure the Shows both experimentaff I ! near phases, S1 to S10, To-aT--Tfrajn and the ten ships relatio behaviour. the The phase S9 allows to soft member for model used during to S1 S3 to occur phases unloading and phase S10 elastic to during for S4 to S8. unloading occur elastic phases allows 92

Rel at lonsh lp

0> u
4>o

(D

ro 0 -ll
qe vi

ni ci.

tn

c -4--3 c)(L) (MM -0 ni 0 c: fli

4-3 s- (n > z3 jm (1) (A (1) m -4j OKJ c2 --0 -C-

rd &4-3 4.3

.4-3 4-3 c0 0

nj S-

c0 0m LL-

D.

,cs E
c:

.- -c m0r

s--

C)

CD

co to
wil

Cl%j

to

LO
LO

C) c) c;
r-.

fA Cc3 C , 4.j -0 ,m-a C) -a 4-3 M0 00 .-0U m 00 X-- 4Of

(3)

CY)

to

C\j C%j co C; C6 -4 0 C; r,: (n tn 1-4

LO

10 a ' (L)
CL

4^ (M
CO

MO C: Zl) cu 0

Ln En

Ca. q15

:G
C:

C\j

to

en

r-4 +

C\j

Wo
L-3 I :3 .. W 4-J rto -0

r-4

C:

S0 (A 40

:3 .. (U r- 4-3 in did

0 St"
(A

-M wX ci, .2 tn 4J CJ.
t-

4-3 (L) c:

(L)- m"Z
(U

IA ,

Sro

C) (1) (A 0 0-

CD G) j _. M v)

tn

C14

Ln

4- E- (n O_ V) 0m (A (1) S- fg Z = 0fZ5 tu

E
a ni 0o cz c:
Z$ (D
tu 4-)

CD

LA (L) (U
0

r%. C\i Lf)

4m to

C> 1.4

X LO CD
C

C) C ILO
to

co r U-)
C)

C) C8 C\j
m

pl, C4 C\i
M

LO L m
r-4

r.. C LO
CYN

Ln C) x U') C)
cl

> (D _c 41 tn (1) M00 (1) (A 0 > 0

-4.1 L) (A

V)
0(1)

40
Z

C\j

cn

C\i

t.0

4-

4eij 0 CM c0
10

u;;

-0 fe >

cn cz
I

to

c: 4-J .vi m0
r-0 aj m 41 LA c:

w' -U m

_C
cn r c:

(A .

S(U Jol

4.3

_C c). ,0

r= M 00 CD M G) l) -- - M f-. c: (1) r= 4-3.> aj ) CD-.1.. o -&J V)

C\i ko ko (71 0)

(Y)

U-)

to

C"i

-4

C)

*10

C)

CD

C7%

CD

Lf)

C)

C C; C
C: ro

c
ko M cn m

c;

2X CZ

ni r=
X0 M S12.

-Z

M C: n L)
0 4-b

S- . b-) C

tA

#0

4-3

.,

c:
CO

c: v) ,

40

r= .(1) c7) t c: =; r=
.cz G) m
C)

tn c:

fei -a

Oj f13

ly! ll (71 C C C 9 . c) C: ) a o cj --d- (::, c)

ILO

CD

-a

(L) t

u tA

-0 (1) -0 eij
U

Jz

E- eu

2:

M eij (1) (L) e- _C c2.m j2

CZOO :3 (n . J .6-3 . -

c:

w10C

41 404-

t-

4-b.

E-= CM4c:

V)

-00

E=

0.

V) 4.. ) V) (V4-

Q)

C:
2= --4 C\i m LIO P W. V) LO V) to V) r-. VI co V) CY) u
V) .-

C:
C) -I
V)

cm (0 , -j
CL 4-

C: 4(1)

C]. (Lj >

CL G)

U
4-3

41

s-

J 0.1. 4(1)

(1) s- 0 . -

r LLJ

I LU

0 -1-10 r: C s- fo -ci 41

93

Read Data

Form external load vector JXP) Calculate memberstiffness sub-matrices Form primary stiffness matrix [X) Constrain [IC) and form structure stiffness matrix Solve for elastic displacements

[K]

JAI a [K]-1 fXP) Calculate memberforces i Increase ). to determine first critical member or next I

Modify critical submatrices member stiffness to represent required change in member stiffness I

Assemble modified structure

[Knl

stiffness

matrix

Modify external load vector to include new nodal residual forces JXP+ R)
Yes STO structure a is is # [KnI singular

mechanism w total [KnI-I No

I No displacem6nt fXP + R) predefined

# Are deflections within limits? I Yes

---!

No LF Newton Raphson iteration Acceptable Residuals? -TY es

I Calculate current member strains and forces

Ves Check if all critical memberbeha is consistent witft assumedlinear memberbehaviour Yes Reduce X to previous value elastically unloading

Fig 2.10

Flow Chart For Non-Linear Analysis Algorithm. Flqure 2.10. The 'F`ig-ureshows the main Reps in the present non-linear analysis is based The algorithm on the 'Dual Load Method' with programme. iteration Raphson Newton available between load steps. an optional
in the flow chart, A critical is a member member, mentioned force displacement corresponding to the intersection and a with of assumed in the representation ranges any two of the linear of the member behaviour.

94

the member stiffness construct in form turn to the primary stiffness are used which sub-matrices, The boundary constraints are app] ied to the primary matrix. This used matrix which is stored in a matrix, to form the structure stiffness banded form, to minimise storage requirements and processor time. The stiffness matrix and load vector form a linear systern of equilibrium equations, which are solved to obtain the components of joint are then used to determine the member forces and the minimum load factor required to is found. cause the first members to become critical displacements. The critical The joint displacements

data

is

to

phase change, members then undergo their first of the member stiffness, and the which requires the modification formation of residual forces applied at the nodes of the critical in forces These stored a vector nodal residual are members. latter load from because the the external only vector, separate load by factor. Once the stiffness the of multiplication requires the failed member has been modified, the structure stiffness matrix is updated, and the equilibrium equations solved to yield the new The Newton Raphson iteration is applied to total displacements. in geometry, and when acceptably small out-ofchanges update residual forces have been obtained, the current member forces and strains are calculated.

balance

The checks in the next section of the program, ensure that all is behaviour consistent with the assumed linearised member critical Any failed behaviour. has a member within the structure member The member may either follow choice of two options at any time. the pre-set means that This sequence of Hnear phases or unload elastically. a yielded tension member must be extending, and a buckled compression member must be shortening unless they are If failed there N members then there elastically. are unloading 2N-1 total possible alternative paths only one of which of a are Each the fulfil of equilibrium requirement and compatibility. will to the choice of member. path requires modification possible force vector, before and the nodal residual matrix stiffness solution correct Consequently, equations, of the updated equilibrium combination of member loading paths this assess if the has been chosen. to

procedure can become time consuming, resulting before compatibility is achieved. To reduce in numerous Iterations the amount of computation required, this section of the program was to enable the analyst to pre-select made interactive possible
95

into loading to take paths, and consideration member from symmetry. member behaviour resulting

any identical

Once all of the failed members are conforming with their linearised behaviour, then the load factor can be incremented to determine the next member, or members, to fail. These members are then set to follow their post-critical paths. The analysis and is repeated, until the structure degenerates into a in singularity mechanism, resulting of the stiffness matrix or, until nodal displacements exceed a predetermined limit. checking cycle
At each member phase change occurring analysis, displacements, failed critical for the current value of the throughout load the factor, ratios, collapse joint and The stress for a

member stresses,

member critical

stress

member phase stress ratio tension

numbers are read into an output file. is a ratio of member stress to yield and member stress and provides Detailed an to buckling of indication of

members,

stress how close

members, compression member is to failure. in the output which provided file

examination

the results a plotting

stored program

was simplified

by the use of

a graphical

representation

of the data.

96

]is.

CHAPTER 3 METHODS OF IMPROVING SPACE TRUSSBEHAVIOUR INTRODUCTION A review of the theoretical and experimental investigations, behaviour of double-layer undertaken to determine the post-yield grids, has shown their behaviour to be highly dependent on the individual characteristics of the members forming the structure. The results of physical tests on several different space trusses, all of which have failed by compression chord buckling, have shown load capacities have overestimated the that theoretical ultimate

actual truss capacities by some 20 to 25%. This has been apparent load capacities have been ultimate even when the theoretical determined, using precise strut behaviour obtained from physical tests on individual compression members. This discrepancy between theoretical imperfection and actual truss sensitivity of initial to force distribution an and occurring within members, lack of fit of members. structure resulting from any initial
Both of the uncertainty in assessing space has led the ical compress ! on-chord-crit behaviour, brittle post-yield improving by to space truss tensile extent, be created members or, top with the chord the allowing a lesser

is due to both the high capacity, length compression the transition the

trusses

load capacity ultimate in addition to their to investigate ductility can Post-yield

engineers

means of

behaviour.

compression members. aim of improving space truss

in bottom chord to occur yield by using eccentrically connected Alternative methods, undertaken behaviour, have considered introduction of an certain members and

of selected web members, the removal force system obtained by pre-stressing internal incorporation of force-limiting devices.

the

IN DOUBLE-LAYER GRIDS REMOVAL MEMBER only a small amount of work has been undertaken to In space assess possible improvements in the force distribution due to the removal of selected trusses, The web members. behaviour of a double-layer theoretical elastic square-on-square distributed a uniformly supporting load has been space truss, (Marsh, 1986). support conditions studied assuming three different is considered to be supported only The first structure at the centre of the grid and has thirty-six 97 web members removed in an At present,

attempt

to

even out

the

chord

member forces.

Figure

3.1

shows a

diagram of the structure and the distribution It can be seen from the the centre line. the thirty-six web members improves the the chords, an increase reaches are its

of chord forces along Figure that removal of distribution within

force

the structure is capable of supporting and as a result, in load of 24% before the most highly stressed member full capacity. truss Two additional truss spanning of supported continuously the corner trusses at the over supported also considered and columns. and the a corners, several truss

a square-on-square 3.2

square-on-square Figure force distribution

shows a diagram

in the chord members taken along a centre line, when f if ty-s ix web members have been removed from the pr imary The corresponding diagrams for the structure. continuous which has forty web members removed from each bay, is structure, shown in Figure 3.3. Marsh (1986) estimates that due to the are capable of removal of these web members, both of the structures in load of 20% before failure an increase of the most supporting heavily The removed to prevents chord optimum stressed internal chord member occurs. web members omitted of from the structures are

the transfer prevent an increase in tensile The object

members.

in turn, which, in adjacent and compressive forces the of the study was to determine shear

forces

of retnoved members, but no recommendat ions or pattern by Marsh (1986). However, Marsh has guidance has been presented increase the ultimate suggested that this procedure will capacity of only space trusses necessary to the with uniform the chords consider the that it is and in addition, linear behaviour of the trusses. response web to with of the structures has If the not both been

Unfortunately,

post-critical optional is not

and without with investigated, so it capacity member.

members determine the failure

possible is reached

of each structure

ultimate of the first

GRIDS IN DOUBLE-LAYER PRESTRESSING

Most attempts at improving the load carrying behaviour of compress ! on-chord-critical post-buckling

capacity, and double-layer

space trusses have concentrated on methods which introduce a small into ductility compression member behaviour. However, if amount of force initial distribution favourable a can be created in these structures which will pre-stress
98

the

critical

top

chord

17
Diagonals Present 36 Diagonals Removed

-Top
Bottom

Chord
Chord Web Members Diagonal Removed

Members

Load

1.24

1.52

Chord [Uniform

Forces Line Load,

Along Uniform

Centre Chords

Figure 3.1 Chord Forces In A Square-On-Square Space Truss Supported 7F-Th-e-Centre. top and bottom 1he Figure snows the theoretical line forces the centre a along of square-on-square chord load. The double-layer space truss carrying a uniformly distributed bracing forces all when members are present and when shown are chord thirty six bracing members have been removed from the complete the bracing Removal of members has improved the structure. force distribution theoretical within the chords and as a result the is capable of supporting an increase in load of 24% before structure the most highly stressed member reaches its full capacity (from, Marsh, 1986).
99

Oiagonals Present

56 Diagonals Removed

-Top Bottom

Chord Chord Web Members

Oiagonal --Removed

Members

Load Chord Uniform

1.0 Along Uniform Centre Chords

Forces Line Load,

In A Square-On-Square Chord__Forces Space Truss 3.2. Figure The diagr&fl shows the theoretical top and Supported At The Corners. I ine of a square-on-square aTong a centre bottom chord forces The structure is double-layer space truss supported at the corners. distributed load forces the a and chord uniformly are supporting the fifty before removal and after of six web bracing shown these The of memebers has improved the force removal members. the chords is capable distribution and the structure within of load in before 20% increase failure a of the most heavily supporting (from, Marsh, 1986). chord member stressed
100

Diagonals Present

40 Diagonals Removed

-Top --Bottom

Chord Chord Web Members

Diagonal Removed

Members

Load Chord Uniform

1.0 Forces Line Load, Along Uniform Centre Chords

Forces inA Chord Continuous Square-on-Square Fi Qure 3.3 -D-ou-ble-Layer Space Iruss. ine diagram shows the theoretical top and line bottom torces along a centre chord of a continuous The double-layer is truss. space structure square-on-square distributed load and the chord forces a uniformly are supporting web bracing lilembers per shown before and after the removal of forty By the removal of these membersthe structure is capable of bay. load in 200/10 increase before failure a supporting of the most heavily (from, Marsh, 1986). chord member stressed
101

compression members in tension, then the possibility also exists to The capacity of the structure. enhance the load carrying force distribution favourable initial can be created by modifying member lengths and has been used as a means of improving truss ' 1979; Spillers, design in a number of studies (Holnicki-Szulc, et An initial state of pre-stress a], 1984). can be created in any indeterminate the state of structure, statically and optimising for a particular pre-stress been achieved using linear 1985). has and loading condition programming techniques (Hanaor, et, a], structure

AI im1ted amount of experimental

work has been undertaken to

theoretical studies on undertaken val ! date pre-stressed However, two square-on-square double-layer space trusses. double-layer space trusses, each with the configuration shown in Figure 3.4, have been tested to compare the effects of providing an (Hanaor, Both of the truss initial et a], 1986). pre-stress structures exhibited exhibited constructed using compression members which a small load plateau together with tensile members which One structure was tested without a long plastic plateau. were

but with an unknown internal force pre-stress any prescribed due to the random lack of fit of members. The other distribution was tested after an initial pre-stress structure was imposed by Figure 3.5 shows the experimental shortening twe.lve members. load-displacement From relationships obtained for both trusses. the structure the Figure it is evident that pre-stressing has improved its ductility. load carrying The physical capacity testing but reduced the post-buckling of these double-layer grids

the benefits which can be achieved by pre-stressing, highlights but increases in load carrying it is apparent that to achieve similar in full the pre-stressing size structures, capacity must be undertaken with care to ensure that actual initial theoretical are in close agreement with their initial
FORCE-LIMITING DEVICES

member forces values.

improve the to load carrying order capacity and post-buckl ing behaviour of compress ion-chord-cr it ical double-layer several engineers have I nvest ig ated the poss 1biIi ty of grids, ductility into compression member behaviour introducing artificial In (Schmidt, et al, 1979; Hanaor, et a],, 1980).
102

This ductility

will,

4116 mm

Mwe

E E
qq

'\ x
p

\ Z

Figure

3.4(B)

RR
Figure 3.4(A)

Figure 3.4(A) Experimental Space Truss Dimensions. Fiqure 3.4. to truss typical assess used of a space elevation sKoWsa plan and The of members. the effects of a known imposed lack of fit and nodes were at all perimeter supported were structures jointing MERO tubular the system uniform using with constructed thickness diameter 41.2 of a wall of mm and members with an outside 1.6 mm. The first structure was tested without any prescribed internal distribution due to force but an unknown with prestress, random lack of fit of members. The second structure was prestressed by shortening twelve members as shown in Figure 3.4(B) (from, Hanaor, et al, 1986).
103

r. -. % 2

100 ,
13 (9

'i
50 (U

10

20

30

Central

Deflectian

Imm]

Figure

3.5.

Experimental

Load-Displacement

Relationships

For A

Prestressea Ana Non-prestressea Uoub I e-Layer Space Truss. The _ F-igure shows the experimental load-displacement for relationships both the prestressed double-layer and non-prestressed space trusses It can be seen from the relationships shown in Figure 3.4(A). shown in Figure 3.5 that prestressing the structure has improved the load but reduced the post-buckling carrying capacity of the structure (from, Hanaor, et a], 1986). ductility

104

in turn,

introduce

ductility

into

the structure

of progressive collapse occurring possibility ductility most heavily stressed compression member. Artificial be created in compression member behaviour, by incorporating the member a force-limiting limit the compression force device.

and decrease the after failure of the can into

This device would, ideally, in the member to a pre-determined level

increasing deflection. constant remain under would which Consequently a compression member, protected by a force-limiting load deflection the device, elastic-plastic would exhibit instead of the highly shown in Figure 3.6, characteristics, brittle, typical post-buckling characteristics of unstable, length compression members. The value of the load transition plateau set by the force-] imiting-dev ice must be lower than the average compression member buckling load to ensure that the device However, it should be becomes operative before the member buckles. of the force-limiting-device appreciated that the introduction itself, may alter the buckling load of the protected
from the

member.
devices

The characteristics to the ideal in addition and repeatibility. force limit constant allow redistribution with of

required

force-limiting

elastic-plastic must a load

behaviour be capable of to

are reliability of providing length a to

The device

plateau member forces

sufficient occur

In addition, the encompassing space structure. the function m1nimum of with maintenance and loading be independent should of characteristics time. To a large by energy absorbers force-limiting energy length critical types

the within device should the behaviour and sequence

these operating extent, requirements are shared and the design concepts used in energy absorbing devices, be for in can modified necessary where use devices. Energy absorbers are rated on both their per unit length. and their weight, ratio of stroke These parameters are generally not force-limiting in which acceptable devices predominate. Several different

absorptions to device in

assessing

load-displacement of

requirements

are available, energy absorbers of which may be all for use as force-I imit ing-dev Ices. To assist In their considered they have been classified to their comparison, according mode of deformation, material operation which may involve extrusion or friction (Ezra, et a], 1972).

105

Ideal Typical

Behaviour Behaviour

of of

Limiting Transition

Device Length Member

Ideal Behaviour Of A Force Llmltln Device. The Figure figure 3.6. limiting force loaded In device the behaviour of a when ideal shows The Ideal characteristics tension and compression. are a perfectly load displacement relationship. elastic-plastic

106

Material

Deformation

A wide force-limiting

range of energy absorbers, offering potential as devices, rely on the deformation of material for the absorption of energy. Energy is absorbed by rods, wires, cables or bending or tubes extended plastically or deformed by buckling, Thin metal tubes are particularly in absorbing efficient Tubes can be flattened, made to turn inside out, made to energy. as shown in Figure 3.7 expand or contract, or made to fracture (Ezra, et a], 1972). Axially loaded tubes which absorb energy by shearing. load-displacement a fluctuating behaviour, which has a mean value of load smaller than the initial laterally loaded tubes provide a smooth peak load. Alternatively, load-displacement response which is not affected by the direction buckling or exhibit The lateral compression of a single of the applied load. between rigid plates has been studied by several investigators Runtz, et al, 1963; Reddy, et al, 1979). De Runtz, deformation four quadrants, in Figure 3.8. et aI, of a laterally (1963) assumed the tube (De fracture,

separated Their analysis

collapse mode of loaded unrestrained tube to consist of by concentrated plastic hinges as shown of the deforming tube accounted for

perfectly plastic changes in geometry and assumed rigid, led behaviour. However, to these assumptions an material of the system when compared with underestimate of the stiffness The theoretical model of De Runtz, et al, experimental results. large (1963) was later improved by allowing for the effects of strain Even further hardening in the hinge regions (Redwood, 1964). improvements between theoretical and experimental results were obtained by using a more complex model which assumed regions of (Reid, localised instead 1978; hinges Reddy, of plastic plasticity et a], 1980). Figure 3.9 shows the non-dimensional load-deflection obtained by crushing laterally a welded mild steel relationship flat Figure 3.10 shows the tube between two rigid plates. load-displacement characteristics obtained by laterally compressing By constraining the tube and preventing the a constrained tube. horizontal are required diameter from increasing, to form a collapse in the collapse required hinges as many plastic mechanism In comparison with the twice

tube. number of an unrestrained Crossed-layers of tubes compressed laterally can also provide (Reid, 1983). suitable load-displacement behaviour characteristics

107

rgl

///////1//

//

Flattening

Tube

Inverted

Tube

Expanding Tube

Contracting

Tube

Tube and Mandrel

The plastic Metal Tubes Used As Energy Absorbers. Figure 3.7. is an efficient d-eformation of thin metal t7es means of absorbing flattened, be Metal tubes can made to turn inside out, made energy. to expand or contract or made to fracture as shown in the Figure (from, Ezra, et a], 1972).

108

P<

Po

P/2

P/2

-L 5/2
t

T
H

Po <

P/Z
o Plastic hinge

P/2

Mode Of A Tube Compressed Between Rigid Collapse deformation of mode of the tube is assume e collapse a es. by hinges. four separated concentrated plastic quadrants of consist tube accounted for large changes in The analysis of the deforming behaviour perfectly material plastic geometry and assumed rigid, (from, De Runtz, et a], 1963).

109

CL

0.1

0-2

0-3

0-4

0.5

0-6

0-7

0-8

6/0

Load-Deflection Experimental Relationship Fiqure 3.9 Obtained From Crushing A Mild Steel Tube. The Figure shows a typical experimenta F'onship load-deflection laterally obtained by crushing rela: a welded tube between two rigid flat P is the app] ied mild steel plates. def lectionL load, 6 is the lateral and D are the length and mean diameter of the tube respectively, Po is the initial load collapse per unit length of tube (from, Reddy, et a], 1979).
110

120

100
80

J 11\ CL
60

40

20

0.1

0-2

0-3

0-4

31 r.3

Obtained Load-Displacement Relationship Experimental Figure 3.10. Aluminjum Tube. 7-rom-77u-shing Laterally The Fig-ure A Constrained by load-displacement tFeobtained experimental relatiTnF's-hip shows aluminium alloy tube with an outside diameter crushing a constrained t of 0.9 mm,. P is the applied load, 6 D of 25 mm and wall thickness deflection By is the lateral and L is the length of the tube. horizontal tube diameter the and preventing constr aini ng the hinges in the twice increasing, as many plastic are produced co II apse mechan i sm compared wi th the co II apse of af ree tube (f rom, Reid, 1983).

ill

Figure 3.11 shows the experimental load-displacement relationships tube systems. obtained from both 'open' and 'closed' crossed-layer In an open crossed-layer tube system, the lateral separation large so that no contact between adjacent tubes is sufficiently In this system each occurs between them as the system deforms. tube deforms in a similar way to a single tube crushed between flat In a closed crossed-layer tube system, adjacent tubes are plates. kept in contact with each other, before and during deformation of the system (Reddy, et_al, 1979). From Figure 3.11 it is evident that the load-displacement

obtained from a closed crossed-tube systems exhibits relationships in load, and would be difficult large fluctuations to use in a load However, the load-displacement limiting behaviour capacity. obtained tube shown In Figure 3.9, a single unrestrained suited to the compression characteristics required particularly device. a force-limiting for is in

Tube Inversion Tubes which are made to turn inside out and absorb energy by bending and stretching, are also capable for use as plastic devices. This process consists of pushing a thin force-limiting internal tube on to a radiused die, which will cause either or (M-Hassani, inversion the tube of et a], 1972). Figures external load-displacement 3.12A and B show the process and the resulting Figure 3.13 shows how this mechanism could be relationship. forceI imit ing-dev ice to limit the load In a to provide a adapted Inversion tubes tubular compression member (Ezra, et a], 1972). must be ensure stability of the inversion could be a major disadvantage process, and consequently this in double-layer grids. their use prohibiting axially Hinge Systems loaded to

Plastic

device which may be carefully A versatile to proportioned load-deflection is shown in obtain the required characteristics The W-frame shown in the Figure Figure 3.14 (Johnson, 1972). forces by acting in flexure resulting in compressive axial resists hinges forming at the elbows. Figure 3.15 shows the plastic load-displacement relationship obtained for the W-frame. The

112

150

Closed Crossed System

Tube

E 100J CL 50-

P
Open Crossed Tube System

0.1

0-2

0-3

0-4 B/D

0.5

0-6

07

0.8

7r-ossec-F-Tube Systems. The experime a oa -e ec ion from crushing both closed and open were obtained FeT-ationships The aluminium tubes had an outside systems. alloy crossed-tube t of 0.9 w. P is the D of 25 mm and a wall thickness diameter is deflection the load, 6 and L is the length of the tube. applied tube system the lateral between In an open crossed-layer separation large is sufficiently tubes so that no contact adjacent occurs In a closed crossed-layer between the tubes as the system deforms. tube system adjacent tubes are held in contact with each other as the system deforms (from, Reddy, et a], 1979). 113

Figure

3.11.

Load-Deflection

Characteristics

of

Closed And Open

I"
m to 0 J

(D

\ ", \ m(D PC
CL
L

> C

(D > G) c) c2c -0

V) CD ro - =

0) 41 C

C)

>

73 CU 4'-

'Cl _

4-)

l< 4-J Q) cu 4-1 ci

(2) >Z Z 110 (1)

C)

-C3
0 C) CO -

c7) _c

'4CD

-0 CD

4-> > 7f5

.D ) 4.

Q)

CD

<Z cm

tn

(3) c: _Z

vl c (M CY)
---1

m tu

13 M 0 CL J

rn

(A cu

'0 -0 CD c: C-l --

.LL-

-0 :3 4-3

4-J u :3 Z

4--, , C)

x w

114

20-0
r--n

13 co 0

2 .X
J 10.0

10.0

20-0

30-0

40-0

50-0

Displacement

[mm]

Load-Displacement Relationship Obtained F igUre 3 12B From The . --d i r`xtTr-naT-_T_nversion Of An Aluminium Tube. The Io ad nt from the external shown i 11 the Figure relationship was obtained inversion The tube of an alu, -ninium alloy cold drawn seamless tube. diameter of 50 mm, a length of 76.2 rnm and a wall had an outside in the 'half-hard' thickness of 1.625 mm. The tube was supplied The external inversion annealed condition. was undertaken under die conditions and the tube and mild steel quasi-static were lubricated throughout the test (from, with polytetrafluorethylene M-Hassani, et a], 1972).

Invertube Figure 3.13. Internal Force Limiting Device. The Figure shows a cross-sect ion throujF-a--J`evjcwhich couTT-'Te used as a load- I imiter' for tubular However, to ensure stability members. during operation, the device must be loaded axially (from, Ezra, et al, 1972).
115

s Typical 3.14. Figure -CETEFr-ess i ve Loads . Geometry Of A W-Frame

E3 Under 0 osed

750

500

(a

250

50 Oeflection

100

150 [mm]

200

Load-Displacement Relationship For The W-Frame Loaded Figure 3.15 The load-displacement relationship In Compression Figure shown in . testing 3.15 was obtained from the quasi-static of a W-frame When loaded in compression fabricated from 5 m, Ti square mild steel. the spechilen deforms elastically hinge at first a plastic until forms at C. As the joint deforms plastically and closes, the ]ever decreases. 'When the arrn increases and the load correspondingly joint at C has fully closed the load increases rapidly until plastic hinges form at joints B and D (from, Johnson, 1972; Rawlings, 1967). 116

complete requirement carrying should material

load-displacement of a force-limiting capacity be possible occurring to achieve

behaviour device after if

is

well

suited drop

to

the

the gradual

in load This from a

yield can be eliminated. by fabricating the W frame strain hardening.

which exhibits

moderate

Cyclic Bending Device An unusual device which uses cycl ic bending to absorb energy for use as a force-] imiting device. has a high potential Figure 3.16 shows a cross-section though the, Idad-1 imiting device which has been termed the "rol I ing torus load-] ! miter" (U. S. Army, 1971; Johnson 1973). gap fit interference in the The device uses a continuous helix of wire gripped between two axial telescoping An cylinders.

between the wire and tubes ensures that the wire is rotated when the two cylinders telescope under the action of an (1973) load. It Johnson has been that the by shown applied axial mode of analysis, deformation of the wire is essentially bending assuming rigid-perfectly plastic force operating per ring of: an gives
P=I 3 Tiaot (Johnson,

material

and an behaviour,

1973)

where a is the radius stress of the wire.

An analysis plastic However, shown material that the of the

of

the wire

and at

is the tensile

yield

rolling of of the et a], the

torus

behaviour testing

has also

elastic-perfectly (Ezra, 1968). been undertaken

assuming

experimental

estimate

underestimate considerably Johnson, et al, 1975). friction considerable devices. Figure 3.17 shows force produced influence

a wide range of torus devices has (1973) (1968) Johnson and Ezra limit load of the devices (Johnson, (1975) the found that the tubes load 'rolling' had of a the

between

on

and wire limiting operational

both

the for

shortening a torus

and

lengthening from copper Figure it is in

load-displacement wire gripped

relationships mild

constructed From the force reduced

between

steel

tubes.

large fluctuations that although evident of they are greatly the first cycle, shortening and subsequent cycles, resulting

are

present limit load

in the second

in an almost constant

117

12 12 tu
9A 40 C CZ 4Q) 00
(1) .3: 4J

a
in
,u0-

:3 4-3 .-0 (1) (n 4-3 V)

) ci cA 1- m0c

00 u a) r- CEM 4-3 (Ufo f_U)

CD

r-

m=-

lu

1--

c3 .-m 0 %0) (1) -4-3 c .-. U 4-3 =O 0) fL) (1) c)- -

ro

(L) >

-13

4 10

5 4)
CLCL
4-3 fo

S-

(1) %A CD -cy -j 00.

v G) (L) 4-3

S-

(A

(1)

to

c:

tA

ul >
4
(A (L) 4-3 e0

in
=

32

Z \, F-

Int

93

to 41 ID

s- -;

4-3 #0

41 *

CC3: -i-i

13

(A c7

0M ,' 0 4. -3

-T

L ID 4.3 Z 0
4VIO
z3

Ei c: m0 IV 4-3 -tn L)

3: u
4-

L)

(A

0 4(L) 4-1 (L) m Ni r, fe 4-3 4.1 r-M r41 -40

(A

.-0

10 u0

c 4-3 fo

(L) M 4) 0 -4..

- (n "f

118

0 6 co

kv0

CM Ln (L) 0 4(A >, S- WC =m (1). 0 4-1 r_ (A m 41 0 10- Z .-0 (1) 0 CL -4-3 3: -c T3

' -0 mu -, U L- -EUcZ 45 2 -V rr. vIE(L) CL od

c: 4-3 .
0 vi

CZ

CD 41 w

-G LY

(A

to CD (1) CD U)

- -f .0 to =
40

3:
0 ta

to
LLtn

s(1) c (Dc2.

0 0 I0

m (A to CL CD M :2 c; 4-3

4..,

Lr)

c3t:

10

3:

3::

0)

of

t. _C (1) (1) E 4-) JL2 t= a) :3

40
tu r--l

r= -1-3 ro e
c) cn vu

c CD zi G) 0)

6M a 14-3 . 4- S- (L) :j

-,

- 4-3

to

, (1) (L) 0

(L) s-

E (?L_E
Q)
J

4(1)4-

c C) 43
CI)

'72. VI -0 u -c
tu
CZ mE0

-4 CM

c s-

(V

mw c 4j rZ
10 (L) 4-3 M

ro

r-

r.A

r=

0 0

0) .2GE p--4
3: , _) c

s: 4..)

E Lt) du --4 r1(L) cm a 4--1 xt m LO 9-4 ; ; V Lr) " m ro C 4- 4-J :0

-i
0 to 0 (A

r= (2) (L) ' - tu m CD M 0%$ CO r= h4-J c;

tn

0 4-

(c

0 0
I-

0 6 [NX] Gojo=j

119

response.

(1975) report considerable difficulty in Several materials a working assembly of the torus. constructing depths. For a smal I were used over a range of interference interference sliding occurred between the wire and tubes, while for Johnson, et a], large interferences off the contact the outer tube grooved and material faces of the tubes. relationships was sheared

shown in Figure 3.17, show torus is capable of producing a constant limit that the rolling force in both tension and compression. However, it is not clear if the experienced in the first shortening of the torus, but absent from subsequent cycles, would return after a Both the possibility of time dependent load period of recovery. fluctuations and the accurate fabrication required to ensure the large displacement characteristics torus as a force-limiting severely limit the use of the device in double-layer space fluctuation

The load-displacement

required rolling trusses.

Shearing

and Machining

The shearing and machining of metal requires energy, and this process has been used in both energy absorbing mechanisms and in 1977; Hanaor, device (Kirk, the construction of a force-limiting 1979). through The force-limiting device consisted of a metal rod pushed A cross-section too]. the centre of a four bladed cutting into several through the device, which has been incorporated space is shown in Figure 3.18 (Hanaor 1979). In order to obtain trusses load-displacement three the required different characteristics, were tested in the device. Brass, aluminium and mild steel The brass rod

metals

cut depths. rods were used over a range of different load-displacement chips and an undulating produced discontinuous depths. Both aluminium and mild at all cutting characteristics steel of test proved more ductile, less than 0.32 m. m. producing Figure continuous by chips Hanaor 3.19 obtained shows typical

at cut depths experimental (1979). His

load-displacement results

relationships it is show that from this

characteristics formed during is apparent

ideal to obtain possible almost device, provided continuous chips are However, operation from Figure of this 3.18 it is device

the cutting that the level bolts.

mechanism. successful of resistance

dependent

on the

by the restraining at

torque a predetermined from corrosion. must be protected

tools provided to the cutting Consequently these bolts should be set the complete assembly and in addition 120

3.18. Metal Cutting Force-Limiting Figure Device. The device Tp-er,it-es by pushing-or pulling a metal Fo__t'Frou`J-FF_thecentre of a To obtain the required too]. four bladed cutting load-displacement depth the of cut must be accurately characteristics set by adjusting bolts (from, Hanaor, 1979). too] restraining the cutting

121

m tu

Oisplacement Rod Brass

("0 0 -J Displacement
AJUMiniUM & Mild StE3el ROCI

From The Obtained Relationships 3.19. Load-Displacement Fiqure Metal cutt i ng i-orce-L IrnILi ng yev ice. I ne io aa-a i sp i acement Brass, three different materials. are shown for relationships aI umini um and miId stee I rods were each used in the dev i ce Ai ch was depths. The brass rod produced tested over a range of cutting discontinuous and undulating chips characteristics at all cut Both the aluminjurn depths. and steel rods produced a duct iIe reponse for all cut depths below 2.0 an (from, Hanaor, 1979).

122

Crushinq A honeycomb core is a useful structure which can be used in a device to provide a constant limit force-limiting force over a is essentially A honeycomb structure large strain. a group of hexagonal cells, which under axial load has a collapse behaviour Figure 3.20 shows a to that of a cylindrical shell. similar load-displacement behaviour obtained from crushing typical an (Coppa, 1968). honeycomb As the compression structure aluminium the response is Hnear elastic until the honeycomb buckles, producing six circumferential waves in each structure At this point the load rapidly decreases to a hexagonal cell. lower level, which is maintained at an almost constant value for load increases, the remainder of the displacement. Phenol ic-g I ass re inforced honeycomb structures a] so exh ib It for use in characteristics suitable acceptable load-displacement Figure 3.21 shows the stress-strain devices. force-limiting honeycomb structure, for a phenolic-glass and gives an relationship can be even when of how ductile a honeycomb structure (Coppa, 1968). brittle from Under a a material constructed honeycomb fails by progressive compressive load the phenolic forming small particles fracture, brittle adjacent to the loaded indication ends. load, The process continues under an almost constant external is crushed. the material Although the all until

for the phenolic-glass reinforced than the In1 um acceptable al um Is more adaptable for use as a An aluminium honeycomb force-I imit Ing dev ice in space trusses. forces, in addition tensile to would be capable of resisting in load limit compression, although the device would a providing characteristics is more honeycornb structure honeycomb, the I atter structure load-displacement have to a post-buckled load peak which is a typical initial load-displacement behaviour. be deployed in condition to avoid of the the characteristic

Extrusion
A large extrusion An extrusion suppress of number of devices a material damper designed earthquake have been constructed which use the

as an efficient

means of absorbing energy. to be incorporated into a structure to exhibits to 1977). 123 the load-displacement requirements A cross-section of a

oscillations

ideally suited character ist ics (Robinson, device force-] Imiting

drawn

80-0

60-0 2

13 (0 40-0

2 0'0

0
50-0 100.0

Displace

me

nt

[rnrn]

3.20. T Ical Load-Displacement Figure Relationship For An Migure Structure. The Aiuminjum Honeycomb sho -ws---ETFe Joad-displacement from crushing a low density relationship o5tained (128.1 kg/M3) aluminium honeycomb structure. Initially alloy a is obtained Hnear response with the load, increasing the until buckles producing six circumferential structure waves in each After this initial hexagonal -cell. buckling the load rapidly level to a lower decreases is almost constant which for the (from, displacement the Coppa, 1968). remainder of

124

r--l CV

E15-0 E z
U, U, C) C4.3

U)
10.0

E 0 u 5-0

L 12.

25-0 Compressive

50-0

75-0 Strain[9/-]

100.0

for A Phenolic-Glass Relationship Stress-Strain 3.21. Fiqure Under a compressive load the phenolic glass Honeycomb tructure. behaves the plastically, although actual honeycomb structure is brittle. The honeycomb behaviour of the phenol ic glass material locally brittle fracture by fails occurring progressive at structure This failure in process is continuous, the loaded ends. resulting load the majority plateau until of the honeycomb steady almost an (from, Coppa, 1968). is crushed structure

125

The device through the extrusion damper is shown in Figure 3.22. consists of a thick walled tube encasing a piston surrounded by Tensile lead. or compression loading on the system forces orifice, producing the extrusion of the lead through a restricted Figure shown in Figure 3.23. characteristics for a similar device, 3.24 shows the load-displacement relationship in which a bulge on a shaft is moved through the lead. During the of the process, changes occur in the microstructure extrusion load-displacement and of the grains occurs adjacent to the orifice, has been the the occurs extruded recovery with material after The the material. rate of of and grain growth recrystallisation depends deformed on the type of material, material a of recovery (Rollason, 1982). time and degree of deformation temperature, lead. Elongation its properties Deformed lead will completely at room recover temperature in under ten seconds, whereas copper under the same to takes one hundred years completely about conditions recrystallise. characteristics for is device the that 3.24 suited particularly extrusion show and in identical behaviour force-limiting The in capacity. a use is a valuable characteristic of the tension and compression, device, space trusses, would permit when used in which different loading forces to several occur under of redistribution conditions. Friction A var 1ety of dev 1ces have been conce I ved AI ch mobI 11se forces to absorb energy. The load control device shown in friction Figure 3.25, has been used in the foundations of a large power to minimise the effects of differential settlement on the station, (Clark, 1973). The device framed jointed structure et al, rigid friction force be developed that designed a constant so would was together by high prestressed on each of two faying surfaces, bolts. friction Considerable difficulty grip was strength load-displacement desired the obtaining encountered load from the device. Several different control characteristics faying surfaces were tested and the best characteristics were Figure obtained with sliding surfaces of firm unrusted mill-scale. load-displacement relationship 3.26 shows a typical obtained from Under a compression the load control device shown in Figure 3.25. load a peak value was reached initiating Intermittent sliding,
126

The load-displacement

shown in Figures

3.23

in

)R IGI NAL : RAINS

SEALS

c AD ELONGATED GRAINS

Ex -r;tv s CRIFICi

RECRYSTALIMATION

GRAIN G m. ow-r H

Section Through An Extrusion Energy Longitudinal Figure 3.22. Wsorber. The diagram Shows a longitudinal section tHrough an Fxtrusion energy aborber and the changes in m1crostructure of the device the During the operation of working working material. through is a restricted orifice and elongation of extruded material After extrusion the material recovers the material grains occurs. Robinson, 1977). by recrystallisation and grain growth (from,

127

20-0 15.0
10.0

L1

5-0

0-0 0 IL 5-C
10-C

15-( 20-(

Load-Displacement Figure 3.23. T-ube-Energy Absorber..

Characteristics

Of A Constricted

40-C 30-(
r--l

Z 20-C L--j 0 10.1 u L 0 0.1 LL

10.1

2030, 40-

Of A Bulged Shaft Load-Displacement Characteristics Fiqure 3.24. The diagrams show the load-displacement -Energy Absorber. lead filled Figure for two extrusion energy absorbers. relationships 3.23 relates to the device shown in Figure 3.22 while Figure 3.24 is for a device in which a bulge on a shaft is moved through the lead. The devices were tested at a rate of 10 mrn of movement per minute (from, Robinson, 1977).
128

-0
cr 0CDCD k-

C) 4-J ro cl

-ci ei -a

4--) 4-) 0 cn 1

cli

co M
m LLJ C) z LU 0-1 LLJ

Z,,
E E co (D co

0 -C

_0 (3) 00
(A (1) S- (n Ln CT)
c4-)

Wo U 4>C

r
Caro CD

1) CE: D
I

CL

CD r",

4-

uc

(1)

Lo

Li

C: )

c: c 01404. ) Ln 4-J C) -) -0

S0

110 4-)

WWOBS

10 -1 0 0

C: 40 _0 0-

C-

C) ;m CD
LL-

C)
Ln (3) Ts

c
C) 4-, ) L) . S-4-

lu

Co

im

hZ ::
LIJ L'i

LO ('

0 10

k-Cq

-A __j
CD LLJ

0)

C) cn C= S- S- LA

= cn (1) (L) _0

UV)

>1 to LL-

129

80

Load
60,
r---l

40

20

20-0

40-0

60-0

Displacement

Imm]

Relationship Obtained From Load-Displacement Typical Fiqure 3.26. The Device. load-displacement Load-Control 7FeFriction diagram loading by the the obtained in was shown relationship device shown in Figure 3.25 at a rate of load-control friction
behaves elastically The device the first 3.3 kN per second. until load the decreases to a residual which after value about occurs slip takes large fluctuations Very in place. slipping subsequent which depending have been reported the behaviour on the faying surfaces, (from, loading Clark, test and the test machine characteristics rate et al, 1973).

130

load. Clark, et al, (1973) also residual producing a fluctuating investigated the short-term behaviour of the device. They found that if the device was re-tested after a period of several hours had elapsed then the first slip in the re-test occurred at a peak load approximately 14% higher than the original No residual load. explanation has been given by Clark, et al, for this time dependent discrepancy or residual between the residual occurring and peak loads. However, they have shown that the magnitude of the long term slip load is unaffected by time.

A similar load control device to that tested by Clark, et a], (1973) has been used in the construction of a long span composite forces resulting steel and concrete bridge to absorb horizontal Figure 3.27 shows details of the device which from earthquakes. resistance of polished stainless on the friction steel relies (Loo, ferrobestos to against absorb energy pads et plates rubbing materials for the a], 1977). Loo, et a], tested several different faying surfaces over a small range of bolt tensions in order to optimise the load-displacement characteristics. and Clark, et al, (1973) found these et devices to be extremely sensitive to both the type of material used for the faying surfaces and the magnitude of tensile force in the Similar sensitivity bolts. was encountered in load prestressing Loo, a] , limiting friction framed structures et a], 1983). devices to changes in faying surfaces and bolt tension is friction devices in space use as load limiting a major drawback to their Although acceptable load-displacement characteristics trusses. can the repeatability be obtained, of the behaviour cannot be guaranteed. WITH FORCE-LIMITING DEVICES TRUSSES SPACE the concept of improving space truss behaviour by force-limiting devices has been the subject incorporating of theoretical only a very limit ed amount of. experimental studies, ' to has been assess the physical performance of undertaken work Although these modified structures. have been tested In order both with structures, (Hanaor, et a], 1980). sets of space trusses to compare the behaviour of similar the force-limiting and without, devices Three different devices (Baktash, incorporated into reinforced The sensitivity concrete of the (1977)

131

: errobestos 3ix High

Pads Tensile

Slots

Bolts

Load-Limiting Friction Device. The diagram shows Figure 3.27. f_Fiction load- I imit i-n-g-revice used to protect a through a sections f rom horizontal displacement from resulting composite bridge The faying surfaces are provided by polished stainless earthquakes. ferrobestos During the against rubbing pads. steel plates it was found that testing of the device proper experimental in preventing both alignment of the two sliding components was vital premature damage to the ferrobestos surfaces and ensuring a steady load response during operation (from, Loo, et a], 1977).

132

were all statically the effect determinate of an from the lack of fit force-distribution initial of resulting layout the trusses, the 3.28 Figure of shows general members. tubes aluminium alloy and nodes. using constructed which were determinate Three statically one with structures were tested, The tests were devices and two without the devices. the displacement simply structures control with under out carried bottom loaded the boundary the at centre and nodes supported at theoretical 3.29 both the Figure and experimental shows node. for deflection load the between and central external relationship force-limiting the trusses 3.30 Figure shows the without for deflection load the between central external and relationship four has of the top chord compression members which structure Figure from be devices. It force-limiting seen can equipped with devices show a 3.29 that-the two structures without force-limiting the behaviour, structure brittle whereas post-buckling force-limiting devices. incorporating behaviour, four force-limiting devices exhibits an additional a ductile load in and is capable of supporting limit load. the elastic excess of

The first

group of space trusses tested chosen to eliminate structures,

The undulating response of the load-displacement relationship is direct the behaviour 3.30, Figure in of poor a result shown devices by Hanaor force-limiting the used of characteristics The devices Used a hydraulic cylinder and ram, fitted with valve which could be set to open at a predetermined a relief for load-displacement A typical relationship obtained pressure. Hanaor, et a], device is shown In Figure 3.31. the force-limiting (1980). in the the that response, resulting nature of oscillating reported due to in load carrying capacity, was primarily large fluctations In the open the between difference area effective valve relief the and closed positions. The second group of space trusses tested by Hanaor (1980) were double-layer The steel structures grids. square-on-square small loaded the boundary at and supported under simply nodes, again were displacement of the three control from the central lower chord node. Only one incorporated force-limiting devices In the structures

compression chord. For this structure the hydraulic force-limiting Instead, devices operating on the metal devices were not used. (Hanaor, 1979). Figure 3.18 shows a principle used were cutting
133

RR

RR
t Plan t R S13.5 mm

2-15-3

mm

p Elevation

Fiqure
7-r -S. US

3.28.

Plan

And Elevation

Of Statically

Determinate

Space

-The trusses were Simply supported at the boundary nodes loaded drawing, bottom the R the by and at centre node. on marzlE identical All members in the structure were nominally and fabricated tubes with an outside diameter from aluminium of 12.7 mm and a alloy 55.4 All mm2. of area of the members were 305 mm cross-sectional 2 had 41.1 GN/m in long equivalent elastic an of modulus and 2 in tension (from, GN/m 49.8 Hanaor, and et a], 1980). compression

134

2 20 13 0 ..........

Theoretical Test joints Test joints with with stiffened unstiffened

L 10
x ------------........................

10

20

30

Central

Deflection

[Mml

Relationship For The Statically 3.29. Load-Deflection Fiqure Te-termina-te Space Trusses tho t orce-Limiting Devices. he had in tested truss members which were effec ive y pin-jointed space fixed in the vertical In the horizontal planes and partially plane. influence the instability, to the nodes of joint assess an attempt top chord compression members were stiffened by of the four central the structure (from, epoxy cement before applying was re-tested Hanaor, et al, 1980).

135

2
20

Theoretici

41 x

L1 0

Top chord ...................................... Test

[stiffened

ul

without

F. L. Ds.
joints]

10 Central

20 Def lection

30 [ mm

For The Statical] Relationship Load-Deflection 3.30. Fiqure Devices. Ihe rigure -D-eter-m in AeSp ace Truss With Force-Limiting introduced has force-lim-iting devices of shows how tMe incorporation the devices the Without into ductility the structure. post-yield dotted by the in brittle curve behaves shown as manner a structure in the Figure (from, Hanaor, et a], 1930).

Q u L 0 LL

Oef lection

Force-Deflection 3.31. Relationship For The Hydraulic Figure The device consists Device. Force-Limiting of a hydraul ic cyl in-Ue-r with a rel ief valve which can be set to open at a and ram f itted During operation the device exhibited large pressure. predetermined in load which was reported fluctuations to be due to the difference between the effective rel ief valve area in the open and closed (from, Hanaor, et al, 1930). positions

force-displacement typical and a cross-section in Figure 3.19. is The shown gently undulating relationship behaviour exhibited by the metal cutting device is an improvement hydraulic device. the However, behaviour the of characteristics on this improved behaviour was obtained at the expense of some loss in through the device load. Figure in setting the required operating limit 3.32 shows the experimental load displacement relationship obtained behaviour for for the three trusses, in addition to the theoretical flexibility the that truss the with large force-limiting differences Figure Hanaor, et al, reported between theoretical and experimental force 3.32, was due to an initial devices.

behaviour distribution

shown in occurring

throughout

lack of fit of members. truss with force limiting load carrying a exhibits limit load.

from a the structures resulting Nevertheless, the response of the space devices is ductile, well and the structure in excess of its elastic

capacity

In order to observe the effect of incorporating force-limiting devices in full size trusses, Hanaor, et al, (1980), undertook four A truss. tests space plan square-on-diagonal steel on one separate in Figure 3.33. The the truss space shown of are elevation and for designed tensile tested a and yield originally was structure a compression chord critical four the the and central system altered was structure, In addition, bottom chord members were stiffened. several top chord compression members were welded to their end nodes as shown order loading achieve The structure in Figure 3.33. was simply supported at the edge four displacement the loaded, at central control, under and nodes For the first test on the structure, the bottom chord nodes. tensile bottom chord members were unstiffened and eight devices were attached to the critical top chord force-limiting hydraulic The members. compression test, investigation and for the first the level to structure make as a such devices the were used In this limit force was set at failure, but in to

compression chord critical. In the second test, the limit force of the devices were raised to in to lower before the tensile occur chord members any yield enable For the third test, top chord compression members became critical. four members were stiffened and structure reverted to being compression chord In the fourth test, the force-limiting devices were Figure 3.34 bottom chord

central the consequently, critical.

the

removed and the compression members allowed to buckle.

137

No.
200-0

Theoretical

180.0

1&2
Theoretical

160-0

140-0

m tu 120-0

100-0 c L 80-0 w 40-0

20-0

4-0

8-0

12-0

16 O

20-0

24-0

28-0

Central

Deflection

Imm]

Identical For Nominally Relationship Load-Displacement 3.32. Figure Devices. Without The Ana Force With Both Space Trusses To_wsboth the experimental load-3isplacement 3Ta-gram77h, relationship Theoretical behaviour truss test. from space a steel obtained (-). The structure behaviour Experimental was from steel tubular diameter fabricated members which had an outside 1c kness II 1372 3.2 The th 33.7 of members were rnii. mm and wa qrn of p 15 0 q long and had an equivalent elastic of modulus U^! n2 both in For test number 1 the structure tension was and compression. devices Hanaor, im iti I f ng and et al, reported orceany without lack of member fit during occurring asser-nbly of the significant fit lack I. This of members was reduced for the second test of mode devices present. which also had no force-limiting on the structure, test force- I imit ing devices were fitted For the third to the four top chord compression members (from, Hanaor, et a], 1980). central
138

9601

mm
X

XIK 4.. /\

I
E E

0
OStiffened joints

' 1 X , II IVF )< X \\Q /N, \\ T \\ / X ,

I

\-Z \, a . A
Plan

C0 AV r, Cj)E

yy -y-

VN

Elevation

And Elevation Showing Of A 3.33. Details Figure -Plan 5'p-ace-T-rus -are---Un--Z -aqon-a-I The square-on-diagonal 7qTu s. i spac--e-l-r-uss I loaded al at edge nodes and supported at four simply central was The structure from steel tube lower chord was fabricated nodes. diameter members having an outside of with the top chord compression thickness 48.3 mm, wall of 4.0 ;nrn and an equivalent elastic modulus 2. le tensi The tubular diameter 120 GN/m me-mbers had an outside of 3.2 thickness 33.7 of a wall ffn and an equivalent mm, elastic of 2The 14-8 GN/rn tubular web members had an outside of modulus thickness diameter of 33.7 mm, a wall of 3.2 mm and an equivalent (from, 150 GN/M2 Hanaor, of modulus et a], 1980). elastic

139

Theoretical
200-0

No.

r
r--l 2

Experimental

No.

M150-0

(a 0 J

c L G)IOO-o ) 4. x ui

Experimental Theoretical No.

No. 4

20-0

40-0

60-0

80-0

Central

Deflection

[mm]

Relationships For Square-On-Diagonal Load-Deflection 3.34. Devices. Tfie di agr am T_rus -Wiitf_And Without Force-Limiting ion relationship' Th-ows the load-defle for the steel space truss force-] imiting dev ices Eight hydraulic both with and without . to top chord compression members for test number devices were fitted These devices were removed for test number four and the three. (from, to buckle Hanaor, et a], allowed were members compression 1980). Fiqure

140

shows the experimental load displacement relationship obtalned for These relationships tests three and four. shown in Figure 3.34, including the of the influence also give an indication of is apparent that the inclusion has had a slight of the detrimental With the effect on the behaviour of the structure. devices present, the experimental load displacement behaviour has become erratic and deviates widely from the theoretical a], behaviour. force-limiting devices in the structure. It force-limiting devices

The investigation the

of Hanaor, et

(1980) have highlighted

importance of obtaining devices. the force-limiting and last

group of tests, dominated and adversely affected the experimental which capacity Hanaor, et al, used three different In their investigations tests. for It is assumed trusses a total of ten non-]! near tests. space a] I the tensile members which have yielded, and compression However, members which have buckled are replaced after each test. this can only be achieved for all the failed members if the forces in the The majority of elements in the structure. monitored are large discrepancy between theoretical and experimental behaviour that obtained in the investigation, may be a direct result for several tests, structure involves a completely different where each non-linear collapse sequence. of using one investigation

a steady and consistent response from The hydraulic system used in the first large fluctuations in load exhibited

141

CHAPTER 4 FORCE-LIMITINGDEVICE NOVEL

INTRODUCTION

from the use of recurring problems arising devices, is the general unsuitability and lack of Several of the control of their load-displacement characteristics. devices considered in the preceeding chapter, showed either large in their working limit loads, or a high initial fluctuations load occurring before the load decreased and stabilised at the working In an attempt to overcome these problems, a novel I imit load. compression member has been designed to act both as a 'soft' stiffness, and as a load-limiting member, with reduced axial The initial device. concept of a 'soft' member, with reduced axial (The by Constrado Steel Construction proposed was stiffness The aim of the present work was to ascertain if improvement could be made in the force distribution occurring double-layer by replacing space truss, a selected within Institute). member. These initial compression members with a 'soft' concepts, have been investigated and the role of the 'soft' member extended to incorporate a load-limiting potential.
Mode Of Operation Details of Of Soft a full Members size soft member are shown in Figure of two square The smaller two tube hollow hollow plus section section strips just 4.1. tubes tubes fit

One of force-limiting

the

The compression and four are proportioned inside the large

member consists strips. the so that tube.

rectangular

soft

Figure 4.2 shows the load-strain member loaded in compression.

relationship obtained for the Under a compression load the but At with a reduced

a pre-determined compression load acting on the soft member, the four middle strips load for in tension the complete soft and provide a plateau yield (B 4.2). At Figure C, this to stage, the member is only member Under this capable of supporting a constant compression force. of the middle strip steady compression, yielding will continue inner tubes mate simultaneously at both ends of the and outer until the soft member, represented by point C in Figure 4.2. When this

behaves initially elastically, member soft (A B, to Figure 4.2). stiffness elastic

142

40

INNER

TUBE

0 04
0 Cl) 0 Cl)
-1

THROUGH PLUG WELDED OUTER TUBE TO STRIP MIDDLE

OUTER

TUBE

0 C11

MIDDLE

STRIPS

25

TOP

OETAIL

0 N-

LO 04 r-

INNER 5mm

TUBE FILLET WELD

Lv
0I '-I

MIDDLE

STRIP

i
0i

40

BOTTOM
60

DETAIL

OF TOP OF LOWER POSITION END BLOCK SEE FIGURE 4.9

6Ox6Ox4 S.H.S.
25x5 STRIP S. H. S.

Notes
1/ ALL DIMENSIONS 43c IN MILLIMETRES. STEEL. 2/ GRADE STRUCTURAL

40X40X4

SECTION

A-A

SOFT

MEMBER

SM5

The Figure Soft Member SM5. details 4.1. Figure shows fabrication of End blocks 4.9 were the full member SM5. soft size shoun in Figure into the member at both f itted ends and allowed a total vertical inner between the tubes The pin-ended and outer of 40. Omm. movement in tested was compression member under displacement soft control at an This rate was continued initial rate of 0.006% per minute. strain until had 4% in the middle total of occurred strain the a strips whereupon 0.06% increased to was per minute. rate strain 143

500

400
0 z

Ld

300

200
PI ED -i

100

10 PERCENTAGE STRAIN

SOFT MEMBER No, SM5

Load-Strain Compr ssive Relationship For Soft Member SM5. load-strain by shows the experimental relationship exhibited in SM5 tested The member beha,,, -es initially compression. member soft but (points A to B). At a a reduced with elastic stiffness elastically load four the acting on the compression member, middle strips pre-set load for in the complete tension a plateau provide and soft member yield Yielding B to C). (points the of the middle strips continues until inner both tubes the mate simultaneously at ends of and soft outer has When (point C). this the the occurred, stiffness of member soft increases the and compression member acts as a pre-stressed member is increase in load (points capable of supporting which a further column buckles At point D the C to D). column exhibiting type' a 'brittle buckling path. strut 4.2. Figure The Figure

144

member increases and the compression member acts as a pre-stressed column which is capable increase (C in further loading to 0, a compression of supporting At point D, the column buckles exhibiting Figure 4.2). a 'brittle type' strut buckling path. has occurred, the stiffness The characteristics 4.2, can be significantly cross-sectional area of of the soft member shown in Figure by varying both the length and

of the soft

The initial steel strips. increased be by shortening the length the can systei of stiffness and the magnitude of the load plateau can be of the four strips, In area of the strips. enhanced by increasing the cross-sectional load the by of plateau can be controlled addition, varying the end distance occurring between the inner and outer load-strain Figure 4.3 shows the tensile behaviour square tubes. the length in Figure 4.6. from the If the soft shown soft member obtained in tension, four loaded then the is steel strips are member and both the inner and outer subjected to compression forces, If the gap between the square tubes are stressed in tension. strips and tubes is very small, then the steel strips will yield in If is this compression. not the case, the strips and squash deflect they by the inner and buckle are until restrained and will outer tubes, whereupon the member will carry additonal load, until When the the strips yield in compression and deform plastically. the soft member will carry an almost constant the strip hardens, and the material strain load. The tensile tensile load additional an supports member increase, the by will member until either the end welds carried fall, or the inner or outer square tubes yield, and finally rupture in tension. strips have yielded, load until tensile member shown in Figure 4.1 have been made. The relative maximum load capacity of the device was enhanced, by changing the two square tubes to round by replacing the four separate steel tubes, and more significantly, tube. The three by tubes were fitted steel a round also strips Several of soft inside the the other, and assembly was fabricated using one closely to that for the square soft members. adopted procedure similar a These triple-tube soft members have a simple mode of operation, but the ultimate capacity of the novel triple-tube compression member to assess, due to the inderdependence of the three is difficult
tubes. 145

altered the four

improvements to

the

type

500

400

300

z 1--f

200

-J
100

PERCENTAGE STRAIN

SOFT MEMBER No, SM3

Tensile Load-Strain Relationship For Soft Member SM3. The Figure 4.3. load-strain by soft Figure shows the experimental relationship exhibited in tension. When is loaded in the tested tension SM3 soft member member in compression. The middle strips the four middle strips are strained buckling from their by both the outer and about weak axis restrained are load applied tensile inner tubes and an increasing to the soft member in deform to This the yield compression and strips plastically. causes in the short load plateau shown in the figure. behaviour results The buckled load plateau when the four strips was curtailed about their displacement to the tube walls. This small with a parallel strong axis in the concave edges of the SUPS compressive strain caused additional hardening in the material. the The soft member of strain onset and load 462KN tensile before total the test was halted of to a supported to the test machine. prevent damage occurring

146

Theoretical Members

Behaviour

And Ultimate

Capacity

Of Triple-Tube

Soft

response of a perfect soft member supporting a investigated load, be by considering the stiffness can compressive interaction. Figures 4.4 of the individual tubes and their joint A to F show the changes in stress occurring within the three tubes of the soft member as the external compressive load is increased. Before the soft member has closed, each of the three tubes will in load. to force the This will equal magnitude external a carry result in a. compressive stress in the inner and outer tubes, and a As the external tensile stress in the middle tube (Figure MA). load is increased, the force carried by each of the three tubes Provided by the that the increase same amount. compression will to cause force in the inner and outer tubes is always insufficient buckling both these the tensile of of members, elastic premature When force in the middle tube will continue to increase ]! nearly. to cause on the middle tube is sufficient this tube to yield, plastic flow occurs within this member and the (Figure increase in load support not any external will member soft The middle tube continues to yield until the three tubes 4.4B). form tube triple compression member. pre-stressed a and up close load is increased, the force within each of the As the external the tensile force acting The middle tube unloads elastically, three tubes now alters. while in the inner and outer tubes increases the compressive stress increase in the external A further loading will (Figure 4.40. increase in the in the compressive stress corresponding a cause inner and outer tubes, and consequently increase their tendency to buckle. are in very close buckling to premature each other, elastic of either the proximity inner or outer tubes will not occur due- to the restraint provided in tension. As the load is further by the middle tube still increased, the compressive stress in the inner and outer tubes will approach elastically. yield If and the middle tube will continue elastic-plastic material perfect inner the tubes in then and outer when yield compression assumed, increase in external load they are unable to support any further In addition, (Figure 4.4D). when the inner and outer tubes have unload behaviour is to However, because the three tubes

The theoretical

bending the stiffness of the triple have column will yielded, decreased to a value equal to the bending stiffness of the middle increase in the external load must now be Any further tube only.
147

LOAD Outer tube riddle tube lInner tube

(A)
Tension Compression

INCREASINGLCAD 13NSCFT MEMBER

Elastic with reduced stiffness -city

INNER MIDDLE .OmR

Tust TUB& TUB&

(B)

C13NSTANT LOAD ON SOFT MEMBER

Tension Compression

Yield of middle tube Zrro stiffness

(C)
Tension compression

INCREASlNG L13AD

IDEAUSED SOFr MEMBER (OPEM

LOAD

I-edy

Soft member- closed Middle tube unloading Normal stiffness.

imuta

TUE TUBC

(B)
Tension Compression

INCREASINGL13AD13NS13FT MEMBER I CFV (f II

Inner and outer at yletcL 'tubes

MIDDLB Oma

""Z

(E)
railure d'y member elastic of the of the soft with the buckling mldcftt tube

Comprevislon

IDEAUSED SOFT MEMBER (CLOSED)

(F)
1 CompresgillOn Fagurt member CIV ptastic 1, of the of the soft with the buckling midero tube

jOuter

tube

riddle

tube

Inner

tub

Diagrams Showing Stress The Theoretical Response Of-A An Axial Soft Member Supporting Compression Load. 'I'he diagram Perfect the three tubes of the soft occurring within shows the changes in stress load is increased. Initially the compression member as the external is elastic but with a reduced elastic (Figure member behaviour stiffness in compression load on the soft An increase 4.4A). the member causes in tension. Ubile tube to yield the middle tube Is yielding the middle increase in compression loading support any further member cannot soft Yielding 4.4n). tube stops when the soft (Figure of the middle member tube compression increase A further up to form a triple member. closes load in the compression on the soft the compressive member increase tubes and causes the middle in both the inner tube to and outer stress (Figure 4.4C). load causes both the inner Additional unload elastically in compression tubes to yield decrease in the and a further and outer in the middle (Figure tube 4.4D). If tensile the slenderness stress large tube is sufficiently to allow this of the middle ratio member to buckling, by elastic fail collapse of the complete soft member will in the middle tube reaches when the stress the elastic occur critical buckling for this flexural value component member (Figure 4.4E). If the Is less tube its of the middle ratio than transition slenderness tube the middle does not buckle ratio# slenderness elastically and in the middle member occurs when the stress of the soft tube collapse (Figure of the tube material 4.4F). stress reaches the yield

148

Consequently, tube. by the the stress in the middle only supported from tensile to compressive, and will tube change will middle increase until the tube becomes unstable, and the middle portion displace horizontally. begins to This horizontal tube the of displacement will occur with buckling at the Euler buckling load, by elastic buckling, or will occur middle tube falls load if is the by tube tangent-modulus the prone to failure at buckling If buckling. elastic of the middle tube occurs, plastic the bending stiffness of the member has been decreased to zero, and if the provided to both the inner and outer tubes by all lateral restraint the middle tube vanishes, causing the complete soft member to If the middle member falls by plastic buckle (Figure ME). will support a load greater than the tangent modulus However, buckling of load but less than the double-modulus load. the middle tube will still cause buckling of the complete soft (Figure Consequently, buckling 4AF). load the the of member buckling it buckling load be by the of will controlled member soft perfect If fails by the tube buckling, tube. middle elastic middle fall load, theoretical will at equal to a member soft perfect inner loads tubes, the the and outer squash of plus of s um buckling load of the middle member. However, if Euler critical the a the the the

buckling by falls tube a perfect soft member will plastic middle load equal to the sum of the plastic buckling fall at a theoretical load of all three tubes. elastic-plastic material behaviour is not assumed, then material strain hardening will allow the inner and outer tubes to sustain a compressive stress greater than the yield stress, in the below Its tube the middle stress remains critical provided If perfect In addition, because the middle tube has been buck] Ing value. in before tension being in strained stressed extensively stress and the member compression, both the compressive yield be decreased direct in will as a compression result of stiffness These small decreases in compression decrease both the elastic stress, will and yield and stiffness loads the tube, in turn buckling of middle which will plastic load the of the complete soft member. collapse reduce the Bauschinger effect. The presence of imperfections will also decrease the collapse load of the soft member. If both the inner and outer tubes yield in compression before the middle member fails, It will be imperfections occurring within the middle member, which will have
149

the dominating member. EXPERIMENTAL

influence

on the collapse

load of the entire

soft

investigation has been undertaken to assess An experimental the behaviour of soft members loaded in both tension and Accurate and reliable values are required for the compression. initial stiffness, yield load, collapse load and the post-buckled response of the soft member. This data was used to compare the behaviour of the soft member in theoretical and experimental the theoretical response of space trusses addition to obtaining incorporating Several soft members. The members have been tested. groups of soft load-displacement behaviour has been obtained for. both model soft In number of full size soft members. members and a limited several small diameter tubes have been tested along with addition, test coupons cut from each steel tube and strip.
Full Size Soft Members

Six ful I size soft members have been fabricated and tested. Five have been loaded in compression and only one member tested in investigation During this tension. modifications were made to improve the behaviour both the connection components.
of members

of the soft member. Changes were made to details and also to the lengths of the

individual
Fabrication

from grade 43C of the soft members were fabricated (BS4360,1986). This grade of steel was chosen steel structural The f irst and good weldability. because of its high ductility from one six metre three members SM1, SM2 and SM3 were fabricated length of 100 x 100 x4 mm thick square hollow sections, one six All six 80 80 thick hollow x5 qrn x square of section and metre four, eight metre lengths of 50 x5 mm thick flat steel strips. The-second group of soft members SM4, SM5, and SM6 were fabricated from one six metre length of 60 x 60 x4 mm thick square hollow length of 40 x 40 x4 mmthick square hollow six metre one section, lengths four, 25 eight metre and of x5 an thick flat steel section length strips. All of the square hollow sections
150

were hot finished

seamed

tubes supplied to meet the dimensional Part 2, (1975).

tolerances

given

in BS4848

details of soft member SM1. Figure 4.5 shows the fabrication Fabrication details of members SM2 and SM5 are shown in Figures 4.6 The fabrication of each of the soft members and 4.1 respectively. to the correct and then milling cutting was undertaken by first The four the required steel tube and strip components. (metal Inert gas) welded, one each side, to then mig strips were This assembly was the bottom of the smaller square hollow section. slid inside the larger square hollow section and the four carefully four then the sides, to the attached, one on each of were strips length For the first top of the outside tube. soft member, SM1, the top was formed using a connection, made between the tube and strips, (Figure 4.7). high bolts strength of and plug welds combination This joint proved to be unnecessarily complex so the top joints on the remaining strip. During fabrication of each of the strips In addition these measurements were taken at both cross-sections. dimensions but the the tubes the outside and only not wall of ends Also the thickness could be measured along the tube length. curvature of the soft members the width and thickness was carefully measured at three different members were formed using two plug welds for each

and twist of the completed soft members were carefully depth The a straight edge and gauge. straight using edge measured the from to the tube soft and member raised adjacent placed was blocks by two positioned at each end of the member. gauge side Measurements were made along the tube from the top of the straight depth tube the to side using a vernier gauge. edge coupons were taken from each of These hollow sections and the eight flat strips. tensile the determine to material properties used machined to comply with the requirements given in 4, (1971). Two test Equipment A SATEC screw-type universal testing Loading machine: machine Control Unit SATEC for by a was used each of the six controlled tests - The machine has a capacity of 500 kN and will allow the Before cross-head to be moved at a constant rate of displacement. the tests accuracy of the load cell undertaken were was any
151

the

four

square

test coupons were and were cut and BS18, Parts 2 and

!4 80
FILLET ER TUBE
0 0

80

vi
50 x5 mm

WELDII

L 15 j- 20 r- -- -T--r---l

1 151

91

qo. 11mm DIA ' LES EACH FACE, I 10mm DIA C/S JS GRADE 8.8-SEE DETAIL 1

-111 iji -1

Ii

'1 Ii
[I

-41

I I
rER TUBE

ii I ii
I'I 1I

I
I,
I

r1

j.J'
A

ill

ILI

\
45 50 OETAIL EE DETAIL 2 I

PLUG WELD THROUGH OUTER TUBE TO MIDDLE STRIP

INNER 5mm

TUBE FILLET WELD

cli

MIDDLE
1 100 1/1 1111 77J

STRIP

[1-

,oo(-

100 x 100 x4 mm S. H. S. 4 No. 50x5mm STRIPS

80

"i

80 x 80 x 5mm S.H.S.

DETAIL

SECTION

A-A

The Figure 4.5. Soft Member SM1. details Figure shows the fabrication This SM1. member was the first member soft member to be tested of soft designed to support loading the additional not compression was after and had in tension. The yielded strips pin-ended member was tested middle at a cross-head control movement of O. 1mm per minute. under displacement to 1.0mrn per minute This rate was increased after a total of 4% strain in the middle had occurred strips.

152

80

INNER

TUBE

0
0

THROUGH PLUG WELDED OUTER TUBE TO STR; P MIDDLE

1: jI II II II --I

cli '4 0
50

OUTER

TUBE

MIDDLE

STRIPS

1 A
(0

i TOP A OETAIL

INNER 5mm

TUBE FILLET WELD

T
Lf) 0

MIDDLE

STRIP

80

"1

C0
00 Z, 100x100x4 50X5 STRIPS 80x8Ox5

BOTTOM

DETAIL

OF TOP OF LOWER POSITION END BLOCK SEE FIGURE 4.9 Notes 1/ ALL DIMENSIONS 43c IN MILLIMETRES. STEEL.

S H. S.

S. H. S.

2/ GRADE

STRUCTURAL

SECTION

A-A

SOFT

MEMBER

SNA2

details The Figure Soft Member SM2. 4.6. shows the fabrication Figure SM2. The to full proportioned the member allow member was soft size of between the inner movement to occur 150.0mm of relative and vertical in The tested tubes. soft pin-ended member was compression under outer displacement of 0.006% per minute. at an initial strain control rate in the a total This rate wa-s continued until strain of 4% had occurred 0.06% increased the to strain rate per whereupon was strips middle minute. 153

Figure

4.7.

Bolted

Connection

Of Soft

Member SMI,

154

by R.D. P. - Electronics Ltd., to comply with the British certified (B. S. I) Grade 1.0 requirements given in BS Standards Institution 1610, (1985).
Displacement members six To obtain axial measurement: I inear variable differential the strain transducers in were attached to deformation transformer the outside three of the soft tranducers tube three sides

To determine were used. R. D. P. D5/2000 transducers of the member.

different

had a working range of 50 mm and were cert if1 ed by R. 0. P. -E I ectron 1cs Ltd. to h ave a , than 0.15% of their better linearity The working range. were R. D. P. D5/IOOOOC transducers remaining three transducers which range of 250 mm. These transducers were certified Ltd., to have a linearity better than 0.25% by R. D. P. -Electronics had a working their of working arranged Steel range. Two of the D5/10000C transducers of the middle sides to were tube of the carefully and strips. the soft bottom through (Figure to measure the displacement wires were threaded the steel on the through strips top

Each of these

opposite and attached wires

members adjacent to of the middle tube. the 4.8) soft the readings

Because the

two steel of the

passing tube were

member rested obtained

outside

from these

two transducers

displacement the tube to the to of outside obtain the correct added for displacements the middle tube and strips. The third values of displacement D5/10000C transducer was used to measure the overall by the transducer achieved attaching was of body to a collar at the bottom of the member and the associated ferromagnetic at the top of the member. Figure plunger to a collar the soft member. This 4.8 test. shows the transducers mounted -tested. The same transducers on the second soft were used for each of the soft member member

At

the

start over with its

of their

cal ibrated calibrated

each test each transducer was carefully full The transducers were working range.

respective amplifiers using accurately The complete procedure was monitored using a slip gauges. measured The transducer displacements program. computer were scanned simple Interval the best straight line was fitted and pre-set each at through the data points using the method of least squares to obtain the millimetre coefficient. - voltage calibration Temperature measurement: throughout each test The temperature was measured continuously thermometer. Al I of the using a Digitron
155

i,.

M,

iq ur e_4_.8

F(i Iiz,

2 So fL

le:nber

SM2

156

including load experimental work, reported in this investigation were undertaken at an ambient and transducer calibrations cell 'C. Before any experimentation temperature of 20 03 was undertaken, temperature This this equipment was kept within of the test range for at least one hour before the test commenced. to be minimum period allowed for near stable conditions all temperature induced drift of the amplifiers

achieved and mininised and power supply.

A Spectra-xb data-logger Computer and data-logger: made by Interco] Measurements and Control Systems was used to monitor and record data obtained from all of the experimental work reported in The load cell and transducers fed continuous this investigation. to the data logger via their amplifiers and these signals intervals time at pre-set under the control scanned of a were The microcomputer is configured around the 16-bit microcomputer. Software Equipment Corporation LSI-11/2 micro-processor. Digital signals language in BASIC the which enabled the operator to was written select the number and frequency with which the measuring devices This data was stored on floppy discs and a subroutine were read. into the software which enabled data files incorporated to be was This was opened and closed and the discs changed during a test. from necessary to prevent the possible loss of test data resulting frequently failure Interruption or more to a power surge. a power the power source erased all data stored in an open data file. The data stored on discs was re-read into a plotting program which from load the cell and transducers into load and voltages converted displacement These results were then plotted values respectively. T221A Graphics Packard Plotter driven by the Digital Hewlett a on

microcomputer. Test procedure An endeavour was made to test f1 ve of the fuIIs1 ze sof t

members as pinned-ended compression members. A perfect pinned-end both to any not give restraint should column-end rotation condition Any is influence the restraint which present warping. will and load ratio and slenderness ultimate of the column. effective of the end fixings shows details used to provide a for the The soft members. end blocks shown in Figure pinned-end into the tube ends and the tapered holo in the tightly 4.9 fitted fitted blocks over a rounded plinth projecting from each of the end Figure 4.9
157

Hallow iare , =tIon

E E LO VM!

E E 0 (D

loomml

E E
c4

4
1-52

CY

E 0

U)

(32)

mm.

-4

4.9. Soft Member End Fixing Details. Figure 7he diagram details gives used to provide for pinned-end of the end fixings the full conditions tests. The end blocks size soft member compression were H=hined out of in the test toughened and were placed steel machine on top of rounded from each of the cross-heads projecting plinths of the machine. The in brackets dimensions shown on the figure to the end block correspond into inner fits the the tube of the soft member. end of which

158

cross-heads of the testing

machine.

The soft member test specimens were positioned and a] igned in The centre line of the outside the test machine with great care. tube was accurately al igned with the centres of the top and bottom crossheads. The displacement transducers were fixed to the members by spec fal hangers and were al Igned using a spirit level. The of the sloping transducers, used to measure angle of inclination the displacement of the inner tube, were also measured. After the soft members were positioned in the test machine they were loaded up to approximately 10 kN to ensure bedding In of the end fixings operation of the data logging and test equipment. test was undertaken the load was removed from the readings of the measuring devices taken. specimen and the initial and correct Before the The specimens were tested under displacement control at a 4% 0.1 had minute until strain of mm per movement crosshead four At in this stage the cross-head the middle strips. occurred displacement was increased to 1 mn per minute for the remainder of These cross-head displacements correspond to a constant the test. 0.006% in the specimen of approximately per minute and strain rate This very low strain 0.06% per minute respectively. rate was test conditions which would chosen so as to approach quasi-static a basis for comparison investigation. in this reported form Model Soft Members soft members have been fabricated and Sixteen members have been tested in compression and eight tested. in To determine both tension. the tensile tested yield members buckling individual the the stress of critical component and stress been twenty-six have tubes tested in the soft members, of parts Twenty-four model When the model soft member Is and twelve in compression. loaded in compression, the middle component, which is sandwiched After between the outer and inner tubes,, is stressed in tension. the middle component unloads the member has closed, soft tension becomes in finally To assess stressed and compression. elastically the middle components in the influence straining of initially tension before loading the component in compression, tests on individual tubular twenty-seven members have been undertaken. of all of the experimental work

159

Additional

have also been undertaken, to assess the influence of strain aging on the soft member. In this investigation several in tension, tubes have been initially strained aged and then restrained in tension. tests Fabrication of members

Each of the model soft members were fabricated from three small diameter, cold-drawn, seamless, annealed steel tubes BS6323, The three tubes had different Part 1, (1982). diameters and wall thicknesses and were carefully proportioned so that they would just fit inside each other as shown in Figure 4.10. The members were by first fabricated cutting and accurately machining the three Four small slots were then milled tubes to the required length. into the middle and outer tubes close to one end. These slots enable the three tubes to be plug welded together using a sequence to that previously similar large soft members. The design details of described in the fabrication of the

evolved in an attempt to Figure 4.11 shows the three principal stages in the development of the model soft member. The members shown as type 1 in Figure 4.11 in the compression test behaved satisfactorily machine but Inner the top the tube occurred in some of at of rotation premature into the test model double-layer the soft members incorporated described in the next chapter. This failure truss structures space incorporating by a small collar was removed at the characteristic top of the soft member as shown in the type 2 soft member In some preliminary (Figure 4.11). tests undertaken on both the type 1 and type 2 members the weld at the bottom of the member, the movement of the connecting the member to the node, constricted This difficulty inner and middle tubes. was overcome by modifying the tube lengths and increasing the height of the bottom plug as (Figure 3 4.11). in type the members shown of the soft members both the external diameter and the wall thicknesses of all of the component parts In the initial bow of the measured. addition carefully were fabricated soft members was measured using a machined steel surface depth gauge. a vernier and plate During the fabrication

the model soft member have slowly improve the behaviour of the device.

160

Outer Tube

Middle Tube

Inner Tube

Section Soft

Through Member

Fabricated Soft

Model Member

Figure

4.10

Model

Soft

Member

ci a)
0
4. ) 4-)

4 LO 1-. .4-) -a :M 4-) 0

-0

a) -0
5

a)
-r. 4j

4 4-4 CL) 0

co

V) z V)

to

0 uj - cc uj Lu. E 3: - w 2 3: 0m w cc E 0m Z LU 00 MI LU UJ =) - -J Z N _j CC :E uj 00< CC LL LIJ '

N cli c-,

a -J

02

a) a) -, 4 C) 1-4
U) r _) 9'. 0 -,, co 0) (u 'r-,

,-4 4-4 0 o Ln

Cd

U) 0

4- -
LU CL

Lr) (0

Q) to
cl Lo

Ln (1)

"0 o

Q)

LO
'0 4 C0

) 4--

rA r-4 QH0

C) ao
4 0

4J
4-)

-a 4)

75

C 0)
4-)

Q)

; - 0

LU

Ln

I-

0 al 1 11 0 -P

0 ul

En

-, :3
LU

F,

CL

Ln En a r0 r. uo

a)

a)

0 U

cd

02
-4 -1 -P 4 C00 C -

bc Q) -H -a

r.

162

Experimental equipment equipment and microcomputer used to full the size soft members tests were also used in all of monitor the same capacity to monitor the tests undertaken on the model soft However, a different test component tubes. members and their machine and displacement measurement system was used for the model soft member and component tests. A Howden EU500 twin screw drNen testing machine was used for all of the tests on the model soft members and The test machine has a maximumcapacity of 50.0 component tubes. by a Howden E179A Control Unit. kN and was controlled Both before and after the tests were undertaken the linearity of the load cell Loading_machine: R. by D. P. Ltd. to certified comply with the British was -Electronics Grade I requirements given in BS1610 (1985). Standards Institution One R. D. P. D5/2000 linear measurement: variable transformer transducer was used to obtain the total differential axial deformation of the test samples. The same transducer, which the for tests, had all of a working range of t 50 mm and used was by R. D. P. -Electronics Ltd., to have a linearity was certified Displacement better To determine accurately the specimens a R. D. P. DHE 25/50 dual The was used. extensometer had a gauge length of extensometer 50 mm and incorporated two D5/4OG8 tranducers each with a working Before the test program commencedboth of the t1 mm. of range working range. transducers than t 0.15% of its in the test strain Both the data logging

in the extensometer were calibrated by R. D. P. Howden Ltd. and compiled with the Grade C requirements given In BS3846, (1970). Throughout the test program these transducers were over their full working range of lm., calibrated regularly n using the data logger and a specially
Test procedure

modified micrometer.

Four separate sets of steel tubes Soft member and component tubes: have been used in the fabrication of the soft members. Each of these groups termed white, yellow, red and green consisted of three diameter six metre long tubes. One length of tube from different for each of the outer, used was group each middle and Inner components of the soft member. Four, type 3 model soft members, frorn each of the white and yellow (Figure 4.11), were fabricated

163

tube groups.

compression. lengths metre six

in of these eight members were tested to failure In addition, four samples were taken from each of the

All

of tube and two of each of these four samples in in failure tension two tested failure to to tested and were were four from All the taken samples each of the six of compression. tubes were cut to the saine length as the corresponding component tubes used in fabrication A further of the soft members.

eight type 3 model soft members also were fabricated Two tensile specimens from each of the red and green tube groups. in the three tubes both The from taken each of used groups. were fabricated from the red tube group were tested members soft eight from in compression while the eight members fabricated to failure in tension. the green tube group were tested to failure of the model soft members and component tubes tested in investigation this carefully were first mig welded into recessed fillet four 5 blocks symmetrically positioned mrn welds. using end Every specimen was accurately aligned in the test machine using a jig which moved up and down the main frame of the made specially All lightly The extensometer clamped to the outside of was machine. the specimens at mid height and the body of the larger tranducer to the top the test fixed cross-head of machine. vertically was All of the specimens were tested under displacement control at an This initial movement of 0.2 mil per minute. cross-head 4% of tensile displacement was continued until or compressive in the had specimen whereupon the cross-head occurred strain These cross-head displacement was increased to 2.0 mmper minute. displacements correspond to a strain 0.063% and 0.63% per minute respectively. rate in the specimens of

In order to determine how the critical Pre-strained middle tubes: buckling load of the middle tube of the soft member is affected by initial tensile test of pre-strain, specimens were amounts varying identical, lengths two, from the nominally six metre of prepared Twelve tube 313 long, from tube. samples, each mm were cut middle length One sample was failed in first of tube. the six metre in compression. The tension and three samples were failed in tension before being were pre-strained samples eight remaining Six of these eight samples were in tension 2.0 before and the loading was extended mm yielded in failed the compression. This test sequence samples and reversed tested to failure
164

in compression.

models the behaviour of the middle member in the type 3 model soft (Figure The two remaining samples cut from the 4.11). members first tube, were extended in tension by 4 an dnd 6 mmrespectively before they were failed in compression. A similar procedure was adopted for the fifteen samples cut from the second six metre length of tube. Three of the fifteen in tension and another three samples samples were tested to failure in compression. Four samples were failed in were tested to failure initial tensile after an compression from a 2.0 mmextension in tension. pre-strain of 0.639% resulting The remaining five samples cut initial from the second tube were given tensile pre-strains from extensions of 4,6,8,10 and 12 mm, before the resulting loading was reversed and the samples failed in compression. of these samples were welded onto end blocks and tested procedure to that outlined above for the model using an identical soft member and component tests. Additional tests on samples cut from another six aging: tube have length been undertaken to assess changes middle of metre due to strain aging. Eleven samples in yield stress and ductility each 313 mm long were cut from the same tube and two of the samples Strain failure in to tension. The tested remaining nine samples were were in tension, each to a different initially strained predetermined level, removed from the test machine and iramersed in boiling water for one hour. After each sample was boiled it was allowed to cool for ten minutes before being returned to the temperature at room in tension. Each of these samples were test machine and failed also welded onto end blocks before they were tested In tension. However, these samples were tested at the lower strain rate of in the specimens had exceeded 0.063% per minute until the strain 13% whereupon the strain minute. RESULTS
Full Soft Size Soft Member Tests component measurements: Table 4.1 gives soft the

All

rate

was increased ten fold

to 0.63% per

member and

individual the of component parts of the measurements SM5 and Table 4.2 gives the initial out-of-straightness

member

and twist

165

44

S- :3 LA

G) c:

tm

c Lr) CDr- mr -t ri Ln CO CO(n CDD Co u-. Co (n ........

ri ei en C" (" cn mm

Lr) C--

c31 c2

4-3

c c
4-4 0 49
'44

CD CO "

ni- tr)

mt

to

cn

C-4-)

0,. (D% aN c% c%

CL
FE

rw4

cm
Z 41

CD
cr x CD :t

c: to
to
41 v

1 0
14

rA

0 "

() yj to a) P-4jQ
49 ',Jl

P.

4-)

w U- N bO

ei

%A

Z c> o. m to Co -, CO CO CO CD a, CZ (n vt

(7%

CD

r0 %. 0

C> 9

LA (0 r-

c;
$, 4
44 o

-0 FEci .

-.
Ln

-(n in

CO -4 %0 %Z r
Ci Ln C% (n tn Ln

c, ) Ln
c% tn

%LI
c

CD
c5

Q) (a U) Q) >
4-4 0 ) 4m,

(U

G)

to

&.

Q)

E: 0 u

CD CD .0

> ej VI c2

0 ;2 4 111

-ri

(1) 0)

Wj S on) r. . 4
4-)

"c3 (L) S- 0 :i (A
tu c:

C) C) CD fl) Mm
Ln U') V)

cn CD C) "m fe)
Lr) tn Ln

-e- Ln -4 m ri M

(n

""

Co Co

CD

ci

"

-4

e)

CD

Lr; t

u;
4) 0 r. 4-4 i.0

(L)

mtm

-4

-e

mr r

r,

c% c%

"

Co " Ch

l co W

.4j

Ln Ln CIJ

ci

i C,

CD
(D

r_ 0 c2.

vi Ln
CL
x 4.3 tr)

E Co Ln (A
c2.
. 4.-2 un

CD.tn
x A-) Ln

CL V')
x Ln

c> c: 'M 0
M
C.

44 0

k 'w

(0
-

vi

(14

ro (1)
M

vi

4.3 (L) vi c3

(0 >

166

Nd

4) 41

C:

4 0) , *9 'w : cd 4) r: cu
Cl 'a C, C', 5

c)
ILD

CD C)

C 0

lo:` cliU-) cli ul

-4 U)

ll`
LO Ln

En

E
Q

'2

0
Cl 9 Ln ll

8?
E C) C) E ,a Q "2 0 0 Q cc
0 N 41 .4

0
CD m

Lr) Ln

en " Ln LC)

g0
to Q >.

14 " 13

_4

Ln

cli

%0

0
0 ow C: a)

'2 8 cr-

en cli

kn C)

Lr) LO

Ln Ln

9 a) 0Q E IOU,
4)
44
8 44

Cc: ;3
'o S'.
8 ri

E
Cl) cil (1)

ce = V) C

cn

Ln cli r C4 cl; Ul) u')

cli %m C) -IT Lr) Ln

,--

2-5
co co 9 Ci fn cli Ln Lr) cli co ll cli Ul) Lr)
.8r
to

.w C"
0 U)
0

4-)
E

C) C) cli Ln Ln

CDC5 cli cli Lr) LC)

C. 4

'm

EE
CDC%j cl! C"! c) C) co Ci Clt CD CD

APIA.
o ul 44 0 P. 0 Is "9

.4h

EE
C; %A %A M a% r_ C cu saj 4L, 5X) %A -A qm Im CL, -@a) W

-@
40

(1) cu (U Gi 4.. sCD -4 W 0) V) ZZ *0 10

sw I

(U t

C> 40 Ln (1) w M"a a

VI V) .-- ; IA

V)

--

4- Lr) 0x V) V)

00

40 vi

A-j kM
X: V)

m .00

167

of fabrication specified smaller for their all

the completed of

soft the

members SM5 and SM6. members were all wider the

The strips and thicker tubes

used in the than the

nominal and thinner of the full

dimensions than size

whereas

were all

slightly were-made

specified. soft

These measurements

theoretical

members and have been used to compare behaviour. and experimental Table the in 4.3 gives test the experimental from size the full values every soft

Material obtained tube

coupon tests: for each of used of the

two tensile the

coupons cut of

and strip All plateau

manufacture

members. plastic the pairs

Large differences

tested exhibited long material a relatively before failure. and a large amount of ductility In upper yield stress were obtained from some of

of coupons cut from the sarne square hollow tubes, however did not occur in either the lower yield stress or these differences Both the upper and lower yield values. stress stresses ultimate higher than the sections are significantly 2 275 N/mm for grade 43C structural specified steel minimum value of (BS4360,1986). The upper yield from the stress values obtained obtained a mean value of 376.7 N/n,112 and a standard The mean value minus three times tKe deviation of 34 N/.mm2. gives the minimum specified value for the yield standard deviation 25 indicating 275 is a three In one N/mm that there of stress falling below the minimum specified thousand chance of a result tube samples value. stress results lower yield values for upper yield, and ultimate from the strips, lower than the obtained were generally from the square tubes. The mean value for the obtained stress was 309 N/MM2 with. a standard deviation yield of 34 The test that only sixty eight percent of the test yield results stress of to be above the minimum specified had from the tube

upper 2 indicating N/mm

can be expected 2. /Mm N 275

Soft member SM1: The load-strain relationship obtained from This member the f irst soft member SM1 is shown In Figure 4.12. in tested compression was designed only for the four was which in to tension. The yield strips member yielded at a load of middle 305 kN and 156 mm of movement of the Inner tube, relative to the before the test tube, The initial occurred was halted. outer of the soft member, obtained from the Hnear stiffness load-displacement portion of the experimental relationship, was 69.4 kNImm. This experiment value was less than the theoretical elastic value of 77.4 kN/mm, which was calculated
168

by

estimating

an

(D a) 14 4 r-q 0 4-) tC-4

0
4-)

44 0

0 0 4-)

wc mw
-o W 1. c Z2 W. 41Um C ow W- w
o x

) C.

9! .0

1 "! V! C '! '1 9 9 m (Y. cn Im a, t- m C-i fn M

'9 m

el m m

19 19 m m M m m

1 M m

C! 11: en cn en

4-) 0

tn

%n

kn %D V! rl:

19

c-

1 ll ! -

C! 9 Ci 1 9 r4 C. cm -

C? V 42

4-3 0 90 (d 1:4 Pa 0

.p

"

"

a-

co

C',
co

(D

co
co
C-

In

47%
10
-

--r
-

in
-

cn
C
ol

-a

(D
Cc
co

w
-

m
z C) -4

W
C;

C.

rz

4D

C6 9

4- r. a, W r-4 14

C C;

C9
0a -

r!

11

Il

I't

C9

"I CD w

C co p .
.

0: 'CO %o
. CIA

C CD w

u = 4)

': 9,

00 m

"

co

m co

cn

ON

co

00
r=
Ln

0 4)-R 4-) rA P4

(v

16

I
I

; co "; m

m "; _; 17 m

8 m

(n

C, M

cn M

ON CIJ

C> 'o eq r

6 .) C

co M

r
M

1 C
M
m Cm

"

Cl#

GO

C W!

CO

14-44-) --" 0 U) Cc 4)
9 co 2 CD

: ;aE
WX4 c

Li -: ll
C> . C) . (D . I

Ci <D . I 4D w (n (D COl CD C> CD 0 C) a C (D CD -

11: 8 C, m C3 C4 CD CD, co

1 CD

0) .H Q)

I
m

I 'a co co

I "0: Ci
10

I r cm m in

I a, r
C5

0c
Li -C

co Go
-

co C)

C6

C
`0

Cli

C)

Ci w %o I
o 10 1

0r

m Jl 0 ox

a, .

Cc!
(D C, co w . P .

%n %n . tn 4v

Ln m c.

" Ln I ! ol -W
10 . . co .

, -:

'9

-'r

C C'!

1 "!
In in 0; 44 cl;

e
W C4

C
r ri e4

El) (1)
LI! r4

C2, g V! j j

10 ri 4

LI! t-

41

0 w w0

4 C) C, C, x 9 (D CD C> C,
.9 .9
4, m

x C> (D CO co x x x
g
41 c C -

43
W r

K C
-

C,
S CL -

x
m -

CD

CD In 6

6m
L -

j CD CD CID, .9 w 'o w x CD x x 3 w w v 1 43 43 43 4p -9
I-I-" II41 = U w c Iw c c 46 s-

I
A
" cx

I
1 1

I
K

CIJ

K
C%l CL

Ln

fn S P rA 9z A) 0
C) CO

41

!L I-

u -

CD

vi

6111 Ln

LI)

V)

I %n I In I In

C)

10

1-

Z;

I 4n

4n

L t4
1-4
u

_j

II c o ! - - - - - - - - -- a !
.c _j -c _j ca j cc 4-) Lj cz j cz _j

i 49
c3

c vw

w v) 4A

4.4 L3 0 0 F4 w4 44 4-) -t-j 41

;4
(L) cd _4

P4

169

Soo

400
0 z M 300
Ld z

C3 -1 YZ Pl EI -i

200

100

10

STRAIN PERCENTAGE

SOFT MEMBER No, SM1

Load-Strain Relationshi Soft Member SH1. EigLw ft 4.12. The Figure load-strain from the relationship obtained soft member SM1 tested shows The member was tested in compression. under displacement control at a O. 1mm 4% of strain had occurred per minute until of movement cross-head At this in the four strips. middle stage the cross-head movement was The member yielded to 1.0mm per minute. increased under a compression force tuberelative of 305KN and 156mm of movement of the inner to the before the test tube, occurred was halted. outer

170

effective and did not take bolted connection.

length

for

each individual into consideration

component of the soft member the flexibility of the long

Soft member SM2: The second soft member SM2 was designed to close-up to form a compound compression member after 150 mm of movement had occurred between the inner and outer tubes. relative The experimental load-strain relationship obtained from testing the is in The soft shown in Figure 4.13. compression soft member load 308.4 kN initial of exhibited at a and an yielded member of 59.4 kN/mm compared with a theoretical stiffness elastic 150 had kN/mm. After total 63.4 a of an of extension of stiffness the soft member closed to form a in the middle strips This was accompanied by both an increase in compound column. load The the test carrying capacity of member. was and stiffness halted prematurely before the member failed at a compression load occurred just below the maximurncapacity
Soft

of the test machine.

The third soft member SM3 was nominally member SM3: Figure 4.3 shows the to the second soft member SM2. identical from testing load-percentage strain relationship obtained member SM3 in tension. The soft member had a physical elastic stiffness kN/mm. strips theoretical 63.4 kN/mm 63.0 stiffness of compared with a of the four steel When the soft member is loaded in tension inside these yield tensile exhibited load the soft

Due to the closefit member are in compression. of the strips sandwiched between the two tubes, strips should in compression and deform plastically when an increasing load is is applied the to the part soft of member. the test This behaviour was during early and a small relationship parallel strain tensile

plastic

in the load-strain evident plateau The load plateau Figure 4.3. was curtailed buckled laterally with a small displacement walls. edges material. before the test of This the caused strips additional and the compressive onset of a total

shown in when the four strips to in the hardening the in tube

concave the

strain

The soft the test

member supported was stopped in order

load of 462 kN to

to prevent

damage occurring

mach'ine.

member SM4: The last three soft members tested were three soft members. designed to be more slender than the first This was undertaken so that an assessment could be made of the load the buckling of soft members. The fourth soft member critical to be tested SM4, had an overall length of 2.0 metres and provision Soft
171

500

400
0 z C3 300 z C3 -j z S-4
p .CE 100

200

10

PERCENTAGE STRAIN

SOFT MEMBER No, SM2

Load-Strain Relationship For Soft Member SM2. The figure Figure 4.13. from load-strain testing the relationship obtained soft member SM2 shows The member was tested under displacement in compression. control at the SM1. The soft member member SM2 yielded as rate at a load of same 59.4KN/mm. 308AHN and exhibited an initial The elastic stiffness of compression load after soft member closed up and supported an additional movement had occurred between the inner and of 150mmof relative a total outer tubes.

172

was made for 150 mm of movement to load-strain The experimental closed. testing the member in compression by premature 50.8 mm of buckling is failed member after inner load inner only

occur

before

the

soft 4.14. inner

member from The tube

relationship the top

obtained the

shown in Figure of

of

movement had occurred between the The soft member failed under a compression and outer tubes. in the of 189.5 kN which caused an average compressive stress tube of 336 N/mm2-. This short portion of the inner tube relative 98.2

beyond have been the tube outer should capable mm projecting load have 220 kN just compression of a which would supporting of in compression. Due to the fabricating caused the tube to yield tolerances in this soft member it is possible for the inner tube to is line to the tube, 1.5 with outer equivalent of which out mm move in the top portion imperfection occurring of the inner a large displacement A small lateral tube. of the inner tube can occur if one of other force the three four strips stress inner strips has a higher yield cause inner In soft stress than the under load will of the inner tube. one of the four

which of

an unsymmetrical member SM4 the strips was 27.5%

to act on one side yield

average

the than the three average yield stress of remaining strips greater in have the buckling this resulted premature of the top of may and failure the inner tube and overall of the soft member.

Soft member SM5 and SM6: The soft members SM5 and SM6 were identical to allow 40 mm of and were proportioned nominally before form to to closing occur a compound compression up movement 4.15 4.16 the load-strain Figures and show respective member. from SM5 testing the SM6 in obtained members and relationship load 152.9 kN at a of compression. load 331.6 kN. buckled The member had an finally at a of and initial of 27.3 kN/mm and a theoretical physical elastic stiffness initial of 28.0 kN/mm. Member SM6 yielded at a elastic stiffness load of 147.4 kN and buckled at an ultimate load of 298.9 kN. The experimental kN/Mm, almost stiffness initial elastic identical to stiffness of the the experimental member was 27.5 initial elastic Member SM5 yielded initially

of member SM5.

collapse load of soft members SM5 and SM6: The collapse load of members SM5 and SM6 can be obtained by theoretical behaviour the Individual both the of components of the considering joint interaction their and member An with each other. soft Theoretical
173

500

400
0 z C3 FL'i 300

z C3 -j
z

200 1--f
C3 -i
100

STRAIN PERCENTAGE

SOFT MEMBER No, SM4

Relationship Load-Strain For Soft Member SM4. The Figure Figure 4.14. from SM load-strain testing the relationship obtained member soft shows 7be member was tested under displacement in compression. control at the Soft SM1. SM4 member member was more slender than the soft as rate same three soft members tested and was designed to assess the flexural first load the buckling The by of member. member failed prematurely critical tube after buckling of the top of the inner only 50.8mm of relative inner between the had occurred and outer tubes. movement

174

0 z 13 Soo Ld z D

I. -I
D -j

200

100

10

STRAIN PERCENTAGE

SOFT MEMBER No, SM5

Fimwe
0 z C3 Lij z C3 -i 300

4.15.

Load-Strain

Relationship-For

Soft Member SM5.

200 .4
C3 -j 100

10

PERCENTAGE STRAIN

Relationship For Soft member SM6. The Figures Inad-Strain FiLfUre 4.16. from SM5 testing load-strain the relationship obtained members soft show These two identical in SM6 members compression. were and nominally and 40mm before to to of allow movement occur closing up. were proportioned under displacement The members were tested control at a cross-head 1mm 4% had O. in the four until per minute of strain occurred of movement At this to stage the cross-head movement was increased strips. middle Member SM5 yielded I. Ommper minute. at a load of 152.91N and buckled had 331.6KN. The initial load member an of elastic stiffness of at a 27.3HN/mm. Soft Member SM6 yielded at a load of 147AHN and buckled at load 298.8KN. The member exhibited of an initial elastic an ultimate 27.5HN/mm. of stiffness 175

SOFT MEMBER No, SM6

ing load imperfect buckl tubular individual the of of estimate . by initial bow occurring be the representing obtained may members, in the unloaded pin-ended member by the relationship: x (Ayrton, (4.1), 1886), et a], Wo = aj sin-IL .... displacement the transverse Wo of the unloaded represents where X and a, represents the maximum central member at po int displacement of the unloaded member (Figure 4.17).
p

Initial Untoaded Position Loaded Position

L
wo 1

Pin-Ended Colurrin With Initial Deformation. 7be f igure Figure 4.17. Aows the parameters used in the derivation of the Perry-Robertson 7he dashed line indicates the position formula. of the unloaded column transverse displacement the WO represents of the unloaded member at and displacement The maximum central of the unloaded member is point Xgiven by ai.

Following transverse

the procedure outlined by Allen, I, (1980), .L displacement of the loaded member is given by:
31 lix (4.2)9

the

where

aL p .......... 'Fe
The maximumstress in the loaded compression member will be at the is the both bending the the sum of and strut axial of and centre stresses: "imax =P+ 7 PNI Z -1
176

is A the cross-sectional area of the compression member, I is where the second moment of area of the compression member and Z is the distance of the centrold of the section to the extreme fibre on the concave side.
Substituting I= Ar2 into equation 4.4 gives: (4.5), Oel

"max = cr 1+

where

a, Z
r2

....

(4.6)9

r= radius of gyration of the member, a is the mean stress is the Euler buckling stress of the member.

and 17e

load acting on the column is increased the As the external However, an Increase In mean stress in the column also increases. the mean stress is accompanied by a larger increase in the maximum When the maximum stress in the stress given by equation'4.5. stress of the column material a zone of occurs on the concave side plasticity at the mid-height This of the column and the member loses stiffness. cross-section indicates the onset of collapse and decrease in column stiffness consequently the value of load P at which the maximumstress equals the yield stress may be taken as a lower bound to the collapse load PC. Substituting equation in equation amax '2 ay in terms of a, (ae-a)(ay-a) = naea .......... 4.5 yields a quadratic column reaches the yield

The smaller root of equation 4.7 is: aecly a+ (a 2*... -aeay)"

(4.8)

where a=

(n ' + 1) ce ay 2 .0.

(4.9)

177

can be used in conjunction with the initial measurements recording the out-of-straightness of the soft buck] ing loads of both the members, to estimate the critical Equations 4.1 to components of the member and the complete soft member itself. Estimate of the critical SM6. flexural buckling load of members SM5 and

4.9

Average values measured for the width and thickness of the component tubes have been used to calculate the section properties. For the outer tube of soft are as follows: member SM5 the section properties

average tube width = 59.67 mm average tube thickness 1 area of moment second cross-sectional radius of gyration slenderness ratio average yield average elastic x area A r 3.94 mm 457287.41 rrrn4 879.13 MM2

22.81 mrn L/r = 76.29 (from test coupons) = 370 N/MM2

stress

2 (from kN/mm 205 test modulus coupons) =

From table

of straightness lateral from the bottom of the member. Representing the Initial deflection measurements for the soft members by the sin curve given in equation 4.1, gives a value for a,, of 1.27 mm. From equation 4.6 1.27 x 59.67 0.0729 = 2x 22.812 The Euler buckling stress
7r2

4.2 the maximum measured value for the out-ofsoft member SM5 is 1.21 mm occurring 1044 rn-n up

ae is given by:
105 2.05 X x 79.2.79

7r 2E_

cro ==

347.6N/MM2

178

From equation 4.9 370 + (1 + 0.0729) 347.6 2 -,

1.48N/MM2

From equation 4.8

a=

347.6 x 370 371.48 + (371.482 - 347.6 x 370;'

274.60
Repeating stress 2 N/mm for the the inner steps

N/MM2

outlined

above,

the

critical

buckling to be 143.37

tube of member SM5 was calculated

soft member was . by both the strips and outer tube the tube would unrestrained buckle under an applied compressive force of 80.75 kN. However, by the other the middle tube is restrained against buckling components of the soft member and consequently the buckling of the occur when the outer tube becomes complete soft member will critical. D shows how the stresses In each of the component parts of soft member SM5 change under an increasing Before the soft member has closed a external compressive force. is force acting on the outer tube, Inner tube and equal constant The force In each of these components is equal the middle strips. When the strips to the externally applied force (Figure 4.18A). Figure have just yielded the average stress In each component Is shown in These stress levels are maintained through the Figure 4.18B. When the member has yielding process until the soft member closes. closed each component Is subjected to the same Imposed displacement in stress to occur in the inner tube, outer changes equal causing Figure 4.18C shows the stresses In the component tube and strips. parts of the soft member after the closed-up member has been subjected to a displacement of 0.5 mm. At this stage the soft load an carry additional can until the stress In the outer member buckling value of 274.0 N/MM2. When this tube reaches its critical level of stress has been attained in the outer tube, this tube 4.18 A to

Consequently

if

the

inner

tube of

the

179

Puter

tube_

Middle ITotal

strip

i1nner

tube

Area 879-13mm2

Areal 534-66MM2

Area 563-26MM2

i I

EXTERNAL FORCE ACTING 13N SOFT MEMBER

Tension IE37-03N/mm2

I All components elastic Compression 113-75N/nim2 Compression 177-64N/mm2

100-OkN

Tension 285-96N/MM2

152-89kN
I Cornpressior Sof t member just closed. tubes Inner and outer 1173. h SIN /,, St ielded i e y av ps r elastic. . Compressior 2 P 71-43N / mM2

FIGURE
Tension 227-C35N) Sof t member closed. A 0.5mm displacement imposed on alt component parts causing a stress change of 50.91N/mm'.

4-185

269-35kN ;; ompres 330-33

Tension IBB-27N,

CRITICAL BUCKLING L13AD

351-95kN

Soft
ultimate

member at
load

1274-SON/mM2 Cornpression

=ompressior 372*11N/mm2

Fioure

4.18.

Stress

Diagrams

For

Comment

Parts

Of

Soft

Member

SM5.

The Figure shows the changes in stress which occur in the component due increasing to SM5 When an compression load. parts of soft member the member is supporting a compression load of 100KN all the component behaving (Figure 4.18A). the The member are elastically soft parts of four middle strips and the soft member yield under a load of 152.89XN. the soft member closes This load is maintained at a constant level until 4.18B). Once the soft member has closed the device up (Figure can (Figure 4.18C). 7he soft member compression loading support additional fails tube looses stiffness at a load of 351.98KN when the outside and buckles (Figure 4.18D). 180

loses stiffness to the strips and no longer provides any restraint and inner tube and consequently the complete soft member buckles. The force in each component part of the soft member can be obtained the stresses given in Figure 4.18D by the relevant by multiplying The critical buckling load of the areas of the tubes and strips. soft member is then obtained by summing up the forces in each of From Figure 4.18D the the component parts of the member. buckling load of the soft member SM5 is 351.95 theoretical critical kN which compares favourablY obtained from the experimental Using the with a buckling investigation. load of 331.58 kN

the s ame procedure as previously outlined, flexural buckling load for the soft member SM6 theoretical critical to be 328.17 kN. This value is less than the was calculated of mernber SM5 due to the bow of member SM6compared with the initial larger initial slightly The flexural buckling load bow measured in member SM5 (Table 4.2). was of soft memberSM6 obtained from the experimental investigation estimated theoretical 298.89 kN. soft members can also be calculated using the column curves given in the for the structural Standard Code of Practice British use of These c6lum6 curves are also derived using the Perry-Robertson formula and are based on bow of L/1000, and the assumption that the column has an Initial (Young, distribution 1971). of stresses parabolic residual a and Four column curves are presented, each curve depending on the type and the axis about which buckling can of column cross-section steelwork flexural To estimate the critical buckling load for square occur. hollow sections it is permitted to use the most favourable of the This curve is based on a Perry factor four column strength curves. the by expression: given n in buildings BS5950, Part 1, (1985). The critical flexural buckling load of both of the buckling load

n=0.00la(x-xo)
where a is the Robertson constant equal to 2.0 x is the column slenderness ratio

and

xo is the limiting
(ay ]12E 2

slenderness which should be taken as:

(BS5950), Part 1, (1985)

Xo = 0.2

181

Using the tabulated values of the column strength curve given flexural in BS5950 (19853, the critical buckling stress for the outer tube of the soft member is given as 253.50 N/mm2. This will tensile stress of 206.37 N/mm2 in each of result in a theoretical the four strips and a theoretical compressive stress of 351.01 2 in the inner tube at failure of the soft member. This stress N/mm leads to a theoretical flexural buckling critical of 310.23 kN for both of the soft members SM5 and SM6. distribution Comparison between theoretical soft membersSM5 and SM6 summary of both the theoretical and flexural buckling experimental results obtained for the critical load of the soft members SM5 and SM6. The differences occurring buckling loads between the theoretical and experimental flexural Table 4.4 gives a members SM5 and SM6 can be due to several buckling load, allowed for The estimate of the flexural bow of the column but did not take into consideration the initial the twists of the columns or any residual stress distribution obtained factors. for the occurring the soft member. The method of fabrication of the large amount of welding to form soft members requires a relatively the internal connections and this may cause a significant residual to occur within the member. If the estimate of stress distribution buckling load of the soft member Is to be Improved an the flexural within assessment must be made and account stresses. The experimental data could also be enhanced by measuring, deflections during the loading process, the lateral occurring at the centre of the members. These displacements can also be used to buckling load of the soft members (Southwell, estimate the critical to these extra measurements an Increase in the scanning rate of the data logger as the member approached the load would have improved the accuracy of the measurement critical of the maximum load attained by the members. 1932). In addition Model Soft MemberTests A total of twenty-five member component tubes in tension: in tension. component tubes have been tested to failure Two Soft samples have been taken from each of the twelve tubes used in the
182

load

and experimental

buckling

loads for

taken

of

these

residual

> "o x 4fo

41, c

0)

P4 P LO ci 4 ., to

4
.4 LO r-j I ri 0 Co
Ln

4-3 0 rn

(n

Co

C%

c
4) , 4.4
0

EO Cd t

LO jz

rq

r-q

E-

Ea

ta CS (1) 0 P (1) DI

c7 V')-c ro -- CO
X 4ro 41 $CD

ID .

r
8 m0 F- (A 02 0

Q)

10 DI . 4 :

tr. 4c:

4-) 44 0

Q) W 0 0)

0
=O

Ei

4.4
0 =0. 3: X CL)a 0 (L) j2 4- j '0 0A: o T-4 14

C% Ln

rr-4

c ri

0
Q .'

P4 q
44

0 S. 0 EO .Ha 4

CL) .00 0) r,

0 1C3 w ro (U =O= 1-- W

4--) CD 4

: Ln
40 vi

Z %0
44 vi 40

i V) a.

0 a) : P-q w0 LO
a) 4-) r-4 CO 9 :3 r-i C) 0W r-q

ca

183

of the soft members and one additional sample has been Table prepared from extra material remaining in the red tube set. 4.5 gives the experimental values of upper yield stress, ultimate stress, elastic modulus, plastic and overall percentage elongation, tests. Figure 4.19 shows typical obtained from the tensile obtained from an experimental load-end displacement relationships Both the yield stress and ultimate inner, middle and outer tube. stress values obtained from all the samples exceeded the minimum requirements of 170N/mm2and 340N/mm2respectiv ely, specified for this grade and finish of steel tube, CFS3GBK, Table 4.5 shows close agreement between the two values of yield stress and ultimate stress obtained from each However, there is a pair of samples taken from the same tube. in each of these values obtained from different large difference but nominally identical tubes. All of the tube samples exhibited a type, particular BS6323, Part 1 (1982). large amount of ductility and load plateaux of the middle and inner tube specimens. were obtained from all

fabrication

Two test samples Soft member component tubes in compression: been taken from each of the inner, middle and outer tubes of one sample taken yellow and white tube groups. Unfortunately, the middle tube in the white group was damaged during welding insufficient material

have the from and The

was available to replace the specimen. buckling stress and elastic experimental values of the critical modulus obtained from the compression tests are given in Table The results presented in Table 4.6 show differences 4.6. varying buckling stress obtained from from 1.4% to 9.4% in the critical identical test Table 4.7 specimens. each pair of nominally

buckling stress obtained from compares the experimental critical values calculated each compression sample with the theoretical of the specimen and the using the measured out-of-straightness In tensile the the yield stress of material. addition, average flexural buckling stress has been calculated theoretical critical for the test specimens using two different values of the Robertson member bow of 1/1000 constant of 2.0 and 3.5 assuming an initial (BS 5950, Part 1,1985). Soft members in tension of eight nominally identical model soft members were All of the members were type 3 (Figure 4.11) tested in tension. from the three steel tubes allocated to the and were fabricated A total
184

2 "9 3:
r4 I

got 4,c CI C,
. =Q) u

Lo 4-
co 410 " C

cl m

0: U! at CD

9 1 P! CO at

't %o

1
qt

'1

1 1 In

"! CO

4:

1 'I cn CO

C1 'I 4 CV %* C,

C'! -1 ON C

4- U) W
4-). 0 0

.4 K

It

c wc

c;

c;

Ea 4-)

4-4 000 w cc
D o 'o co c5 -: r w 'o (n Lm 'n La

4-)

r. q r-4 4)

(D

4-)

41 r.: cn w cn co m co co '1 I" o m co m w m co -1 al co m a cq co co a% co It

0) to a) 4-) 4-) ., j p F=

4-)

NO= cla s. c co co In co co co m c; w (n 4 Ln c; lo r ; co c r ol c c cq w lo c> W co co m

0 4-) Cl)
ey

I CD V4 C.
0 4.4
0

4-)

tv

l .2 2el

-C =

4 CD
-

C; -

9
" -

r
-

c
CD -

g C.
-

C> -

c r
-

co

C>
-

CDO -

CD.

:2

c9

qr

. C w. %b CD

a!
-

In

w! 9
r

a!
C

W:

Lo a)

4-)

l c
rz -W

In

In

w cy

co

03 C4 -

to 14 39

16

-W

CIS A C. ej C-4 -

to ol -

U
47,

IC! cn

'D
(n

CO CD
Cl 0 C

C.

C -

3,

co

.0 w w w

.0

.0
w w w w

.0

.2

.0

0 1
U

W 4-)
U % . W U w U w

4! 4! !! T W W uw U . m , 0 -0
I "-

w w ID
I

% .

4-4 0
I I-

4-) ca

iI *

" 2 2 w C> 3c E E CD C)

re
C)

0: C, un

-4

'' 2

:w

I--

: 4 -

! E IF I ;: F:

4- 4 4-) w

0
f4 44

>b
Cd 4)

185

7-0 610 5.0 p 4-0 3-0 2-0 1.0

(mm) END DISPLACEMENT

SAMPLE No. GT02G

6.0
^ z 5.0

4-0 3-0 2-0

1-0

(mm) END DISPLACEMENT

SAMPLE No. GTMlG

e% z

6-

C3 I

4,1

2"

END DISPLACEMENT (mm)

SAMPLE No. GTIlGA

Behaviour Of Soft Member Component 4.19. Load-Displacement Figure load-displacement behaviour tensile The Figure Tubes . shows typical inner in fabrication from tubes the the and used middle outer, obtained All three members. specimens were tested under of the model soft displacement control strain at an initial rate of 0.063% per minute. to 0.63% per minute after This rate was increased a total of 4% strain had occurred in the members. The relationships shown in the Figure are for the samples cut from the three tubes allocated to the green group. 186

r.

ca r.

:3 @A a) x LU s- , L,- V) M . v 5-

%0 c;
r-

r_ 19
tn

4.1 c
%0

_A t
r,

m c;
CD

Co

-4 r
Co

m c9
cli

gr
-4

W
-4

CY%

cli

ci

cli

ci

cli

4-)

4-)

f-,

to 0 -i 4), 2

;jw
C% -e
C%i C) Ln

C
-4 _A

C:)
(D --:r tr)

m
fqm M

cli

r_

C-M

mt

-4
Co ei r,

0
P +-) b4 (1) ;4
A

e;
C% :r Ln

l
Ln

-0:
c% r m

M 0, % M

c CY%
Ln m

r
410 (P

cz Co
cli

00
r4

4-) U)

tA

cm

-0ocy Z .2

zc

E
CD

Co
-1

c;

r-

CD

c cli

ri
c -4 cli

CD r CD . Ln C10%

CD %0 L
CY% C%
1-4

c -4

C7N

Co _A

rl
CD

cy,

Lr! r_
4,4 4-) 0 -, 4 "I 'S P-i 0

L.i

1450
0) -9

CJD 4-;

LM

CY%

ri

V%

%Z

Ln

to

110

Co

ka

Co

Co

V) tu

0% CY%C% CN cn

c CY%C% 0% r4

c cli

c cli

44 0
Jmj ,

0 4m)

8 49

4-) a) Cd 4-)
CY% C% CI) M Cl% e) C% e) CA 41) C% fli

c-

ri

rl

fli

m
M

cJ Cj

r-4 0 Cd -A

C:

(Cuo
>, 1
tu 0 s(L) 0 J2 $-

c7) 1-0 ei:3 4)

m all t- 0 cu
Z (V
>b

CL (L) :3 00

Z t-

c2. CL) M 00 :3 s-

Z 4v 0 d2 _ Z)

:3 ajo

, im 9 c7.
1- 0

CL. C) :i 12 0

s.

Ci. cu :1 0 . 92

4) = 0 . 92 :3 S-

(L) :i 0 . 92 :3 Z-

c) 3

Ci. =j 0

ei s

CL = 0

s..

4 m -0

a -0 (L) (L)

u cu
V)

s- CU cu 41
Zm
0

CU
Zm
0 3.

c2

1-0 (U r- r-

1- CL)

--

c-

CU

(Li

o -

0-

-0 .-

-0 -

I5 C . u w -2

t
2A-4 4: 3

2:
-# CD u 0

: 3:
C: ) C-) -4 (-)

>u

x
-4 L)

2:
f-) 1-4 u

>, C3

Ir

>.

:c
-4 L)

2:
cli L)

Z c2. v)

CD L)

2:

t-)

2:

Z:

Z:

44

187

La
x m
w-

(L) W r" ru;

LU1c

to
c;

0
u

CD
c-;

-4
cz

Q
c; f c

-t CO --4 cy cli -4

Em

rz-

to "

tLI CY

rcJ

C -4

CO

C) cli

0
1-4

4-)
llo, s

w
'S,

92.. x LLJ (j

Co

sl

(U
:3 r_ it CD cn Co r" Co 0% cu (%i

4-4 C3
LA. 1
cm

2 1 -, ,
4,0 i. L. V)

-4, (L) w. m

M
u

cn cli

-0

41 CD V) r_ V)
m CY%

cli

%D

cli

to

"

l Ln

cli

c U'>

; 19 r,
-4

; r,

r tm

%D " r4
C%i

cu

rL)4 4 C44 o 0 ;.4 0 cc 0) C) to 0

u-n : 3=OOW

4A CVI cm

F- W to

ce U-

(2
,

IL, S4,

2-

.0 u- vi

.Z
tA

ci 11

c3

4-3 -H 4-3 tr)

t

P0

LO

Cl%
c

No
c

-4
; -

CD
Lr ;

Co
r. :

#A
L

t
c9

0%
c

-4
c

c2

Im
-;

CM -0 41 (D %A C: Ln 41 m wu d m CYN .5 c4

r,

cli

"

ci

"

r_

0w

1- Z5

E s- 4j V) u
-0 r-

-4

9m

-4

-1

Co

-4

Co

:r cli

ui

-rl "

le cli
C13 ::, -4 f+4 .0

cm
x: .0

b-- w, ra -W

00

U-,

Co

4-)
x (L) (L) =. O Cll! (, l00 to tm cm 41 -4 r, " CD m r, -4 CIN --t C> CYI

w -. LL v)

74
-4 44 m C%i M cli ei cli E 4J .H04. P 44
0

a) E(I
0 0

G)

0-

r_ cli

c cli

OD

CYN

se
ra-

C. 4
: 0

0 rj) ) ;4
9

-0cu E

-E :i
Vb

0% cli
-!

%0 m

cli

C>

(2 M

Ln C%i
CD

-e

Ln

tM m

%0 e)
C:)

Ln cm

4M ci

Co %D

D)

2 w0

3:

C"! Il 0 CD

-: CD

C:)

c
0

-! (D

-! (D

-! C:)
4.4 a) 0

c12

c0 t 2 - Z (L)

CD N -4

C: ) Il -4 " c"i .

fIl

rrl -4

to c -zr

to Ill -e

C%i c m

CM c9 m

: r l' -t

-& -&

88
4.) 00

Co

00

Co

Co

-4

AN Oo .HMM
4

Co

CD

Co

Co

Co

CD

CD

Co

13 -

-P4 w 4-)

fi
$--0 m Ul)

fi
U')

li
to

ll!
ul OD

9
Co

9 11:
Co CO

p fi
to to

fi
%0

ll!
to

CI-4

0 9t w
P-4

Q) U) C)

LO

. 0-

:s0 "I ': 2 4L

;Z
(L,(2 0 cm M. ; V)
1CL)

Co

CO

Co
Co

c Ch
t)

c C)
fl)

Ln r C: )
m

cm

"

f"

ch

dm

cli

"

Lin f (D
M

e) u CD
fe

M t (D
m

C: ) r Co
"

(D f Co
ci

rc;

(U c >

s-

CN "

r. c;

Ln
Co cli

Ln
Co ci

0% "

4-) 4-4 o P0 44

0) P-. 4
9 CD (-) C: ) (-) :c -4 CD X " (D :x -4 2:

(D

CO

ti (D

(..> (D

C-) CD

ri -

-4 -

u LD

ci D

cli -

cli >-

.00:
f-.) C-13 c) (D

u (D

L> f.M

#. -) CM

9z vs 2

188

Figure 4.20 shows the experimental load-displacement test specimen. All of the obtained from the fifth relationship remaining test samples, apart from the sixth sample which failed due to a poor weld, exhibited very similar prematurely green group. load-displacement relationships Table 4.8 gives the experimental load, extension, overall plastic and percentage percentage for each of the soft member tensile tests. elongation at failure loads have The values obtained for the member yield and ultimate in the the corresponding stress levels been used to calculate These member stresses are given in inner, middle and outer tubes. Table 4.8 in addition to both the experimental and theoretical member stiffnesses. When the triple-tube soft members are strained in tension the in compression and the outer and inner The cross-sectional strained in tension. shown in Figure 4.20. values of the yield load, ultimate to that

middle tube is strained tubes are correspondingly tube tubes that the three the middle are proportioned so of areas in before deform the outer plastically compression and yield will in tension. The individual tube stresses inner tubes yield and in 4.8 that tension in the Table show soft member yields when given the average compressive stress in the middle tube is 30% greater This than the average tensile yield stress of the tube material. in compression, indicates that when the middle tube reaches yield expansion of this tube is prevented squashing and lateral plastic by the outer tube. As the middle tube is restrained from deforming in load increase the tensile an applied to the soft plastically, in the the middle compressive cause stress will member exceed the yield stress of the tube material.

tube to

load required to For the soft members under test, the tensile 12% is less than the tensile load required to inner tube the yield Consequently the tensile load acting on the yield the outer tube. soft member can increase until the tensile stress in the Inner tube At this reaches yield. tension, and effectifly of the soft member.
Soft members in compression A total of sixteen type 3 (Figure 4.11) model Figure obtained soft members

In point the inner tube deforms plastically controls the yield and post-yield behaviour

have been tested to failure load-displacement compressive

in compression. relationship 189

4.21

shows the from the third

"4

z
C3 I Ld 1 z Ld f--

7-0 6-0 5-0 4-0 3-0 2-0 1.0

(mm) END DISPLACEMENT

SAMPLE No. SMGT5

Load-Displacement Tensile Behaviour Of Model Soft Member. Figure 4.20. load-end displacement The Figure shows a typical relationship obtained from testing The member yielded under a a model soft member in tension. force of 7388N. force of 440ON and supported In a maximum tensile initial the stiffness of the member was 2745 Nlmm. The soft addition in the middle tube reached a value 3W. when the stress member yielded The greater than the average tensile yield stress of the tube material. behaviour of the soft member is governed by the and post-yield yield load-displacement behaviour of the inner tube. tensile The soft member failed when the inner tube necked and ruptured in tension.
dol% Z

Y. .

7-0 6-0 5.0

13 >-4

4-0 3.0 2-0 1-o

END DISPLACEMENT (mm)

SOFT MEMBER SAMPLE No. GS3Y

Load-Displacement Figure 4.21. Relationship For AModel Soft Member In 7he Figure shows the experimented Compression. load-end displacement from testing obtained relationship model soft member GS3Y in The member yields under a force of 3916N and buckles under compression. a force of 7865N. The member has an initial elastic of 3258.8 stiffness N/mm. 190

, *I
0, w 'o C-i v

4-4 --r Ln P cn Ci "i

Ln w
WQ. w C4o

00

4-4

w t" Z Zu co

4-)
3

4-1
I N

.uc al a.

Lj

-j

4-)

rn

41 s

4, )
co en
(1)

4! 41zS,
0-= cm
41)

47,
m

f,
en m

Ln .2
m

fn
m +

-o
-0

pq 4-) 44

C, .5 11 0)
4-) -rj

r8
4-)

0
Www-a

71 .;
Z , 10 P. In cli

.0 'n 4: 0- c -9 -

'a

1% N
o wi 10 ul

CLwm
0-2 0

10 In

nL) r, p
4-) V2

14

41

E 0 LI) r
Cc

co

4m -

t,

%a

V) en m m

r m m

R.00 0 ca
4-
rA a)

C14

4- Ea :3
(1)
9 CH >

a)

19

4-)

po bO W 4-)

! a)

-tf)

2:

C4 co

t,

14

C 00

00

Ci 1
01
P ba w 4j

Vb=)-J-

O;m

7' : w.9 QP r c w

ul "

C. CIA .

en m co ; 4 f V
. .

cv + +

V! 't CD

4-) 4-3 0

DI 0 Cd . r-I w 0
L) 0. "', ;4 '8

4
4-) a)

8, () 0) 4J H . r..C)
4J 4-)

0 a)

LJ

w 31. C%-

at r, (n CO at r m " m rn m m en (n I I

M P m m en M

A CO Z0

g
; -4

* ro

'A

cs
'9 C5 rw ri + c 4 0c L! C> +

r.

0 m

a) .00': 4-)

4-) -rq
4-)

? !! ZL
.

'I
en

" '!
ol %0

ON (D

co "

In 8

.0 9 U-0

LI! C'i m

0
4 1* 4) 4-) 4)
R :

V)

$4 44 Id

c0 4,0 4 ! 0 1 . S. 4- A

-4

40

co

rn

4) . a 2

X.

-=

r co

Zo 03

"; in

-W Go %D r, cv ev (L

-; (),

a CO E

co
ri r

4,

r, i C-

z C,; Ln co

to cm

to (L)
0 ;4

a)

44 4-) -t: r
a) c 1 0 t:
41 J-

4J

C9
.1 Ln m

C9
W$ 19) Ln rn

9
V In cn

Ci
ul W) 4"

eq
Ul
fn

Ci
Ul
M

eq
%n
en

Ci
Ln
gn

w U R

to

io 2
v)

'n

'n

vb

;F ;2
vi

191

from steel tube allocated to the member which was fabricated The soft member exhibited yellow group. an initial sitffness of 3259 N/mm, yielded under a compressive load of 3916 N and buckled test The soft member characteristics shown in of the load-displacement behaviour obtained test members. The experimental values of load, buckling load and member initial yield stiffness obtained from each test member are given in Table 4.9. In addition Table 4.9 gives the average stress levels in the three'component tubes at the yield load of the soft members and the average stress in the outer tube at the buckling load of the soft members. The stresses in the outer tube have been obtained from strain measurement made by the dual extensometer while the stress in the middle and inner from the member loading and the tube tubes have been calculated Table 4.9 also gives the theoretical areas. cross-sectional of each of the test soft members. These member stiffness have been calculated the measured tube using stiffnesses areas, the average elastic cross-sectional modulus obtained from initial tube tests described previously and the lengths of the The tube lengths were taken as the distance component tubes. between the ends of the members to the centre of the plug welds for both the outer and inner tubes and the distance between the two end the tensile plug welds for the middle tube.
Theoretical critical flexural buckling loads of model soft members

at a load of 7864 N. Figure 4.21 are typical from the other fifteen

load of the model soft using the physical dimensions of each members can be calculated, component tube together with the measured bow of the complete soft values of the soft member buckling loads member. Two different Estimates of buckling have been calculated assuming two different stress conditions to be The first estimate of the valid when the soft members buckle. buckling load is calculated assuming that the soft member loses when the stiffness of the outside tube becomes zero. The stability second estimate is calculated assuming that the soft member buckles Both these of the middle tube becomes zero. when the stiffness buckling stress of one of values depend on assessing the flexural the component tubes of the soft members and this has been undertaken using equations 4.1 to 4.9 described previously. The buckling loads of the complete soft members have been flexural calculated using a procedure similar
192

the

flexural

to that

ut! I ised to estimate

4--l Cd
W t4 Cd 4J 5 (D p Cd 0) 4--) 4-) e-I

*4 r-4 5

Ld P4

44

M :3F!

la) 0 k [a +) :5 44

0 r-I rq

C)

a)

44 4-) 0

A A

4 A

14

14
10. $4 M Id

A [a 0
0
m A P ;4 A
co

ce)
.

C)
41
4

a) P,

0 (L)
R C! C!

... -9 V

I'!

li
,8

Cd Z
;4 0

10

0) -P
$4

"-, +) 4 . -s ', q :s 4-) 0) D) EO 0

5 -QO) (L)

(L)
'0,4-1

-r-, 4-) a) a) 4-) 4 4- JZ

r
rI P,

-H

rMc4J .

.2wF. 4-)
rj

ol

a) 4 4-)

to 0 $.4 C)

4-)

0) 4

4.

00 )

44 0 4

b-

p B r-4 Q) P :3 Q) Cd t 0) 0 f--4 "0 En; to

A 0-

.1

s:

W,

Iz

c4

to . 49 C) P

R., 9 04 C')
0 41 8 44 0 -8

lo

41 10C)a 0 Cd D) r. 4-3 1 4 x" ") 49 U) > Ld 0

EQ r . . e4

4 rz.

I all

14
g

1181
Rig

is

J
9

bT j3

bt

u. IS

b 1. IS

1.2 13

jH

b 15

44 r- 001 V) -H 4. ) ? It
x x y x x x x 0--

.)m00
0,

C3 0 0 4 DI 0 44 Q)

-j ,

Fl 1

ig

-0

05

-0

A qV f...4 .,

; - W 44 r-q P-4

r.

X
9 ": r 4-) Cd .I

T4-4 0 0

4-) La

Ea td

> cd

(a 0) 4

La ;: 4S

9U

00 4-)

,8 a) ;4 rj) 4-) $ U) F-

193

the buckling loads of the full size soft members SM5 and SM6. The both estimates of procedure described has been used to calculate load of the soft members, each value buckling the flexural depending on the The stress conditions assumed at failure. presented have been based on the average physical calculations properties obtained from the four soft members fabricated from the steel tubes allocated to the white group.

buckling of the flexural members when buckling is controlled outside tube. Estimate

load

of the model soft by the stability of the

The parameters given in Table 4.10 have been used to calculate buckling for the outside flexural tube. the critical stress in both position Assuming the outside tube to be restrained and in position only at the other at one end and restrained length Le Of the tubular strut may be end, then the effective taken as 0.7 times the actual length of the tube. direction The tube slenderness ratio x is given by X Le r

0.7 x 318 = 71.23 3.125 Representing lateral deflection average initial measurements for the four soft members by the sin curve given in equation 4.1, gives a value for a, equal to 0.14 mm. From equation 4.6; the

TI

a,
r2x3.125
2Z0.14

x 9.52
2=0.06824

The Euler buckling

stress

ae is given by;

ae _n2E_, X7

2x2.09

x 71.232

105 =

406.56

N/MM2

194

EQ 0)
LM

C)
=3 9(1) C: cy) m co ILD tD 1.4 q:r 4D co 1-4 x km CD C% CY) co cli

p
w

CH

4.0
all 1-4

LO CV) %0

0
En

0 CH be I
., 4-4 (a C (71 co
co

.0 =3
W ,a a co ci Csi
--I

C)

rl_ -4 %0

LO -4 C)

4-)

P%

Cl%i

cli

co C%i

4-4

94 a) Q) e. I

s41 :3

ON co
CI

LO
4 C: LC!

Ln cli
M

C!

CY)

Ul)
en c::,

Q) 4J

;4

C)

1-4

1-4

C73

(1)

$4

4J

.8
4 0

4-4

w 4) r-q

E C4

I =

-1 LLJ
u

fo

S-

in

4--) c+4 0 >1

t) %A

4o

0
413
a 0

41
4 cl) to

Ln
013 W

0 4 44

E
4cu

%A

u aj

40
0

(1)
S-

4-A V)
(L)

0 u 1w

%A 0 S-

0 . 6-1

"a #10

V)

ct:

>-

195

From equation 4.9; 303.53 + (1 + 0.06824) 406.56 2 368.91 N/mm Substituting Ocrit equation 4.8 gives the critical for the outside tube as; 406.56 x 303.53 = 368.91 + (368.91' - 406.56 x 303.53/'
"crit Considering yield load = 256.25 N/M, n2 . soft member, the from the average steel

into

buckling

stress

Ocrit

the behaviour for the four

of the complete soft is given by;

members fabricated

allocated

to the white

group

Average yield
where:

load = 286.02 x 19.98 = 3712.47 N

the average yield stress of the middle tube ay = 286.02 N/MM2 the average cross-sectional area of the middle tube A= 19.98 MM2 in compression the magnitude of When the soft member yields the force acting on each of the three tubes is identical and the be as shown 1n Figure 4.22 di agram at yield wi II stress
Yield Load 3712-47N

286-02N/m Tension

IBB-S3N/mM' : ompression

iaa. a3
Compressic

N/m

Outer

tube

Middle

tub?,

Inner

tube

Figure

4.22.

Stress

Diagram

For

Model

Soft

Member At Yield.

196

The stress diagram shown in Figure 4.22 will remain unchanged during the yielding After the sequence of the soft member. phase is completed, the soft member closes up and all yielding three component tubes are subjected to the same end displacements. Any small end displacement of the soft member will now cause equal strain and stress changes to occur in each of the three tubes.
Consider displacement the soft member to be subjected to a small end

in the outside tube to its which will raise the stress buckling This will critical value. cause a stress change in all three tubes of:

256.25 - 188.93 = 67.32 N/MM2 is proposed that when the outside tube of the soft member becomes unstable the complete soft member buckles. Consequently, 2 to the is given adding 66.23 N/mm the stress diagram at failure in both the outer and inner tubes and compressive stresses 66.23 N/MM2 from the subtracting tube. The stress diagrarn at failure Figure 4.23. tensile in the middle of the soft member is given in stress It

Load Failure 7232-5N 2IE317ON/md Tension 256-25N/mM2 -ompression . 256-15N/mm Compressio 11

Outer

tube

Middle

tube

Inner

tube

Stress Diagram For Model Soft Member At Failure, FigUre 4.23. Collapse Is Govermed By Buckling Of Outside Tube.

TAliere

the component tube stresses shown in Figure 4.23 tube areas gives an average theoretical by the corresponding load for the four soft members of 7232.5 N. failure Multiplying (11) Estimate of the flexural buckling load of the model soft by the stability of the

members when buckling middle tube.

is controlled

197

The parameters given in Table 4.10 for the critical flexural been used to calculate this

Before the middle tube component. compression both the outer and inner tubes will have yielded. in compression. This will reduce the restraint offered to the middle tube by both the outer and inner tubes and as a result the middle tube may be considered to be heldin in position but not restrained direction at both ends. x-1.0 x 318 = 121.53

the middle tube have buckling stress for becomes stressed in

The tube slenderness ratio

Using an average member bow of 0.14 mm in equation 4.6, gives:

2j=

0.14 x 7.94 2x2.6177 ae is given by;

105

= 0.08115

The Euler buckling

2Ex2

stress
]12

x 2.01 X 2 121.53

134.32

N/MM2

From equation 4.9 286.52 + (1 + 0.08115) 134.32 215.87 N/rnm2 = cc = 2 Substituting acrit into equation 4.8 gives the critical for the middle tube as; buckling stress

Ocrit

134.32 x 286.52 = 215.87 + (215.872 - 134.32 x 286.52)V2 = 125.79 N/inn2.

of the soft outside and inner tubes in the middle tube will and the stress member both

acrit
At

failure

in compression 2 The stress be 125.79 Nlmm have yielded

diagram at failure

is shown In Figure

4.24.

198

13161-

5N

303-53

Nlm

125

- 7S

N/MM2 I 2E33-03N/m

Compression

Compressio

Outer

tube

Middle

tube

Inner

tube

4.24. Diagram Stress For Model Soft Figure Is Governed By Buckling Of The Middle Collapse

Member At Tube.

Failure,

Where

Multiplying

the

tube

stresses

given

in Figure

4.24 by the failure load

corresponding tube areas gives an average theoretical for the four soft members of 13161.5 N. Similar

have also been undertaken to obtain the calculations buckling load for the remaining two two estimates of the flexural buckling groups of model soft members. In addition the flexural loads have been calculated buckling using the values of critical stress for both the outer and middle tubes obtained from BS 5950 Table 4.11 gives a summary of the Part 1, Table 27A, (1985). buckling loads for the groups of soft members estimated flexural together with experimental 4.11 it is evident presented In Table that the model soft members under Investigation buckle when the stiffness of the outside tube decreases to zero. middle tubes average investigation. the buckling loads obtained from the From the results

Pre-strained

the middle tube in of pre-straining tension, twenty seven nominally identical tubular samples have been Four samples prepared from two stock lengths of the middle tube. have been used to determine the yield stress and tensile behaviour of the material, six value of the flexural samples have been used to obtain an average buckling load of the specimens and seventeen
199

To determine the influence

c7) c cm
.-a

(L) -a :jCj. %A 41 0 CD

8-

E: le .uZ
Z

Co 0 4- -0 < -0 1- -r=

0) c3 Cn -(1) Ln V) du

ti)

.00 M (n r

CL)4.4

v (A 400 Co 0 Co 3: '; 0 (L) . r, be 0) m vi C%i

FEr(V >e U x:

Ir CD V-t
G C)

_A 1 c ".

to

0 Lo rA

to 4.4 -, q 4) a)
0 44

43 0 .

c 1-1

pw 2 *r-4

> q a) (L) 40 s 4 0 .H co :q:: "j 4-) 4.4

4-) 14 0 C

a 4j CU #U 4-

C:)

44 41 0 La 4.3

C-4

0 J8
4J 4J P. 44 9 0o

g sz
-00

4J
. r. ei G)

Ul
C%i _A to --4 M m 4.0.

Icno a) w 0 : 4.) ;4 0)

0 du

4- M,0C c -0 0) Ln #0 413. - $- -C_i :3 0) EE

:i C> 4A r-M"
ei

u2

CY) 4--0 r4. -0 40

--4 jo

a '0

r= CM 4r-f r-A W-4

(V

m
1-

CO -_U, VI

zi

G) JD 113 0) 3: M- Z ra >0
3t 4 40 0

0 4J 0 0sC. A ! -' 9s:

4-1 000 45 4) 0

41

a)

r-A 44 "ow
40 c: 0..

12,8'0
4-) CH 4Q')

CD Q) cu m Lr) m :3 (U et cm X -0 c: 4.-b VI LO 40 0 415tn - vi

4-) -1

02 M _j

vi OKJ .C7) tA 4(A 41 c: <

:3

r-

0) OD Co
; c"i

#0 0 -

Co

to
r

r
-4
r

0M

0 4-4

Cc) C)
4-) 4-4 C61-4 0 0 U)

Q)

0
.

0) r:

-, Z -JZ
u= Z

02 ---'

je

Pei

e -M . (U -! U
J-:3

(V

de

8"o,, >0

I 4

CD
C>

,-

40)

F-

3:. -4

a)
00

4-)

D) r-q

0
to

cn fe
u

ai J-s(U
l=

4-b 0

02

4(V 0 vi du (U 0 du S- im
w

17. r
alt

A cd

0
_i

j2 :3
4J

V)

.M

vi

.9

0)

m.

>

*C)

0 je G) U m :30 F-Cov)4-

4-

(U X cu tA 00 0 1EE m ai tn .- - C= tA he du V) 41 r- 4-3 U tu :3MC 0. a 0

" du C%i u) -

--CD r

CM

cli

CD

cu

C%i
E

3 rp 0: ,

bo 403

4.) . Cd 4.4 0 rA

8 0 rs
902 41 $4

0 4-4
EQ

91

EA (D

r-

#0
0 to

dqj

V) Lr)

C% CD V-1

m Lt) LO

EE c21. 0) c2. u= :1 >x LU CO -2

Q)

ra o4

4-)
-4 M

>

'. 4

4-) cc Q) a)
;4

a) r-I

cd A

0) >4
.,

(a
rQ-),

.> q om4.4 #--I 4VI C) Ft41-4r- .0

r-4 ID

a)

-H a) 4-)

rn

4. ) U)

E-'

!I

U) Q) 0) kw

Q) 4-)

c)

du

giE: n u

m :x

iv >-

(1) eie

200

in tension by varying amounts before samples have been pre-strained in compression. Tables 4.12 and 4.13 being stressed to failure load and ultimate load values of yield give the experimental obtained from the samples tested from each of the two parent B and C show the load displacement in obtained from three of the samples pre-strained relationships From the results given in Tables 4.12 and 4.13 it is tension. tubes. Figures A, the tubular samples in tension reduces evident that pre-straining buckling loads. The specimens taken from the first their flexural parent tube, tube L, show a 10% decrease in the average flexural buckling in the specimens of load due to a tensile pre-extension 2.0 mm. A smaller decrease of 4% has been obtained for the corresponding samples taken from the second parent tube, tube K, In addition, both test groups show larger decreases buckH-ng loads corresponding directly in flexural with increasing A tensile to the samples. applied amounts of pre-strain in decreases in the flexural of 6 rnm resulted pre-extension buckling load of 25% and 30% for the two corresponding samples, for Figure 4.25C shows that taken from tubes L and K. 12 hardens. mrn or greater the material strain of pre-extensions This in turn raises the yield stress of the material and prevents buckling load. further decreases in the flexural
aging A total of eleven nominally identical test samples have been

4.25

(Table 4.13).

Strain

length from stock one prepared been used to determine changes aging.

These samples have of middle tube. in yield stress and ductility due to

have Two been used as control samples specimens and strain In tension without have been tested to failure any aging, while the have in been tension, samples pre-strained aged and nine remaining Figures in tension. 4.26 A, B and C show the then failed load-displacement samples and two relationships of the obtained from one of 4.14 the control the with aged specimens. Table

gives

experimental the ultimate the lengths failure

values of first loads obtained of the load

loads and second yield from the test specimens. and the percentage

together

In addition, elongation at

plateaux

are also

given.

A comparison of the load-displacement relationships given In Figures 4.26 A, B and C shows that a prominent yield point and load load-displacement to the tensile plateau return relationship
201

.2
CL4,

c c %9 19 to m m cm cm
rCo cli Pcli cli cli

9 C:)
C)

li
tli

LA

(D

c (D
cm

N cli
14

Ui
tn -4

19
le

N
m

Lr!
Ln

'0

-4

C%

P-

r :

0 W

2 -0

4, M

tA (L) s-

0 ei

Co --0 Co r, %0 fl; ri
CD m
"

C>

Ul)

Ul) %D %0

%D

el NO c f-4

cli

CA

l
C\i

Ln CYN

14
CD LO

t
f-) r-

CM

cli

CY, l &

cm

r cli

v
rclo
-4

Z tm
(:
-4

F4

c2

Z. 41 tn c

Ulb
ei

0 -e

cli

-: r

m f t

1 Co

-e m

-t CD

r, CD

-t m

Co m

-4 -w

r4

-: r

44 D) 0 H

.0 , 0

Z , ci m cli

4-) w
-4 to r_ U*) CY% 0 r d" 14

ri)

a)

-i

c 0

-4 ei

c A; .4 to e tn C>
m M

1 c %0 CYN
cli

-4 m

c; %0

ei -4 A M

%0 cli

c w

r CD m

C> m

Z (n

'. 1

c; m .-4 M

t Co Co
cu

(4 Ln
m

C 2 =D .

=O z2 U

Co

=:. .0) 0 0,

=> c;

rc0 0)

=D c

W
c0

9
(L) c2.

: 2.
; c

C
c0

2
w a

=I,
c cu 0 o (U CL n r 4N

0
ej c2.

gj c21

(L)
C) w

a) 41 r

4) 41 C

c0

c0

E
u

ti e

fi

0, 4,

vi

(L) 41

r- u 0
c

E gi 0 Zu
c

cu u 0

cc

.5

-E

C-

CL)
AZ , (L) , CL) , tu

0)
%A -2

ci
%A -:; A-

tu
%A -:; 0

ci
-Z 0

(i)
-Z

c
c

4)
.-

c
c

4)

'o
4, -:;

c (U 441 x wZ C)

c CL)C
x0 w 4, w

c (U 44 x0 (11 2

cm ai4x0 iv 2 (L) o -0 2

c tu

0..

c 0)4x0 cu

c 0)4x0 cu Z

b.x0 w 4,

G) m 41 44. x
w0

cu *, 4x w0

Ik 14 m Cf)

41

41

41 %A i (U %A tn tn cu tc2

CL

r_

r=

r= 0
ILJ

&A A cu 1CL

W IM om 0

4, 0-0

4, om

2 1

Z 0 4,

, 0m 4,0

*0

10

0-0 2 42

4, 0 0,0

4,0 0 0 2M 3

4,4) 010 -" 4

4-4 0 4-) co CIL) Cc $

. r-

r_ r%0 m

4,

0 4-)

c e rm clim C.- C.10 cm

c Co

(D

c m

r= CNJ c
-4

-4 cm cm

C, 4 XD CD LD cm

C-3 cm

r0 ;4 o

202

cs 8 [a

'0

s.. a) 's
vi 1 .2
4j

-& lw

c r,
c>

19 Co
C)

Ci CY,

cm CD CY

M CD cm

m CY r4

-4 CD

Co Co

r,

sZD

-'.

2 .4 (U -4 Ir cn

0% Co

Co C

4 m

-4 %D

-4 %0 fl)

-&

14

14

4J
.0g

cn cli

%0 CY

%D cli

cli

cli

r4

cli

r4

0%

E; 0

ril

4-4
0

m m cli

to cli cli

CY cli r4

tn m

C) m

-e C, 4

Co m cm

EO
0 4-)

c9

Ui

4A c

C) m

0% CY

r en CD
m

(n g" 0%
CY

Ln
Co CY

tn in (D
4n

(0 Co c
cl i

cyl (: Co
cli

(Li -0
r-

(L)

Qi

cu

4)

0) 10

; c
c ; cu 1c). ic

; c
cu s2. g

c
(U sCL ti c w 1CL

2 1 .
c 2 . c0 cu (U s-

2 . 1
.
c 2 4 . ei 1CL cw 1c0 tL, u

4. )

0) 30

(U sCD. ti

c0 0,

c0 4,

r_ 0 CL, 41

c0 cu

c0 0,

c0 CL, u Z

JE
Q c (L) sc 0 1c (U sc 4 41

jE
tu 3c C,

Z
(L) c (U sc 1c 4, 1-

-2
w 4. x0 0) 4, cu x0
a) ,

ai ). di

tu

w x
w

CL, X0

cu ,

cu x0

(U ,

CU4x0
ai ,

w 42 x0
tu ,

Q)
0-

XO

OM 0

0-

4-4 0

CL 9

92. E

cu

C, 4

ti

41 Z

Nd KD

be cD CD

be (D (D

cli LD

be cm KD

Ne cm

Nd CD

5z 0 0

203

10.0
10.0

20-0

END DISPLACEMENT(mm)

20-0

0-0

0-0

50-0

cn z

FIGURE

4.25A

GTC4L

-J

20-0

10.0 10.0
110.0

10.0

20-0

END DISPLACEMENT(mm)
20-0

2 6
r-I C3 -j
w j 9. 0 6

30-0

40.0

SAMPLE No. GTC7L

$0.0

FIGURE

4.25C

FIGURE

4-25B
SAMPLE No. GTC9L

Load-Displacement Behaviour Of Pre-Sfrained 4.25 AB and C. Figure The figure Tubes. shows the load-displacement relationships obtained in tension by varying from three test samples which were pre-strained in failure loaded to before Sample number GTC4L compression. amounts by 2-Omm failed in in tension and compression under a pre-strained was Sample number GTC7L was pre-strained force of 2736AN (Figure 4.25A). in compression under a force of 1878.6N in tension by 8. Ommand failed 4.25B). Sample number GTC9L was pre-strained in tension (Figure by in compression 12. Omm and failed under a force of 1967.2N (Figure buckling load of the samples which were The average flexural 4.25C). 3041.7N in (Table 4.12). tension was not pre-strained 204

dr% Z

Y.

60.0 50-0

EI

40.0

-i LLI 30-0 -i b-i (4 20-0 Z LLJ lo-c

E No. SAG2

(mm) END DISPLACEMENT FIGURE

Z 60-0 50.0 M 1 Ld
0-4

4.26A

40-0 30-0
20-0

-j

0 z

No SAG3 -

LLJ

10-0

(mm) END DISPLACEMENT FIGURE 4.26B

01% . Z 60.0 x

50.0 40-0 LLJ 30-0 20-0 10-0

z LLJ

No. SAG10

END DISPLACEMENT (mm) FIGURE 4,26C

Load-Dislacement Relationship of Strain Ag Figure 4.26A B and C. load-displacement The Figures show the tensile behaviour Middle Tubes. identical from three in middle tube samples tested nominally obtained Figure 4.26A shows the behaviour of one of the control tension. samples 4.26B shows the load-displacement Figure SAG2 which was not aged. from SAG3 in tension by 1.92% which was pre-strained sample relationship in tension to failure. Test sample before being aged and restrained by 11.2% before in tension being aged and SAGJO was pre-strained (Figure failure in 4.26C). tension to restrained 205

Ch

Kn

r-I 0 P., 4-)

momW 4-)

r. C: .,. i 0
0

$4 4-)

c9

pl:

Il

Cd

4-)

rs 4-)

$4 -4

CD

0,

Co

ig

%2 a

lO

2;

u 4 a)
41

g:

Z
M

o
M

i
"Z

CO

c') V-4 44 0

$4

4.) Lo LO 4-) -'

0 44 0 w 'o

" C M

W.

I :, M

R ,
1

C , M
j

: , c 1 l
1 wo

2 IM
OZ M

42 Me

mi

0
00:! 4. c; -3 In 44 0 0 4-) Cd 4-4

2 IM ZU
C l, k 122 -WC w Fw c= c22 -" CM 229
-z Z

5MZ
-c

z22 8-8 -L"22 A W C =" -tu _M 40

sh

J
z22 C-8 W
C

zum -um
AW Zum

-Aly
-AW Zum

Z. WCW-

222 8-8 -tu 21

g E ng en Erz c 2Z v: z22 z23 422

A
IM

EWS

3- 9 5 W. Z!,-g . U

g" g -

Z
-.

C"CfZU ' --0"

-!

M M:

WAV Zum

Zum

hm Cm ! je Z!
.e uW u" 20

iF-

2 ZU -W
O e C

mZ

MZ

CM

41

.0

IM

91

C) 0 4-3 ;12

2 2

206

Also the by experimental values of the aged specimen. an exhibited in 4.14, indicate Table that this stress stress given yield second increases with a corresponding increase in the amount of initial tensile given to the specimens pre-strain lesser extent, there is also an increase in of the specimens corresponding strength Both test to the samples. given pre-strain before aging. the ultimate to the increase To a tensile in

the load-displacement

4.26 B C the in Figures and and experimental shown relationships increases both the 4.14 that these in Table in show values given strength are accompanied second yield stress and ultimate tensile by a decrease in the ductility exhibited by the aged specimens. DISCUSSION in this have been chapter outlined behaviour the the determine to of novel characteristics undertaken The results device. show that a obtained force-limiting been has a constructed capable of exhibiting member compression The investigations steady failure. All load plateau followed by a reserve of strength before

of the tests reported on both the soft members and the however been have tubes with upmost care, undertaken component for further Relatively large discussion. cause give results some 5%, have been reported in the for the tubes. calculated component modulus elastic of values to be due to the difficulty These variations are most likely the tube thicknesses in in turn wall which -measuring encountered in the the inacuracies calculating cross-sectional areas of cause determining inacuracles the The associated with member members. differences, in the region of in themselves the also present calculation of areas cross-sectional However, the recorded readings of both strain the member stresses. determined load using a combination of precisely more are and 1.0% these the accuracy of results are and within strain gauges of the true readings. investigation has been related to A major part of this loads behaviour buckling for buckling the and critical assessing both the full size and model soft members. A proposal describing buckling behaviour and ultimate capacity of the the theoretical included in introduction has been the to tube soft members triple This proposal postulates that both the inner and the chapter. in before the stiffness tubes compression yield should of the outer
207

mi dd Ie However, based given apparent the soft is both before

tube

dec re ases

to of

zero the

c aus i ng theoretical with the

the

Soft results

member of

to

b uc kIe. load both It is of of tube, the it to

a comparison on these Table that

ultimate results,

assumptions 4.11 shows soft

experimental to be

in

these

proposals

unsound. stiffness

the tube

model

members buckled to zero. stability to provide For

when the the of the

outside

has decreased be governed the middle

stability middle

member to necessary the inner for

by the tube to

sufficient them buckling soft

restraint prematurely which

and outer

tubes

prevent

squashing

in compression. from three different the outer and

The model diameter

members have and

have

been fabricated gaps of 0.08

tubes, tubes as

clearance 0.235 in m. -n

m,,, n between the

and middle inner tubes,

existing 4.27.1

between

middle

shown

Figure

OUTERTUB DIA. 9-52

MIDDLE INNER TUBE TUBE DIA. 7,94

DIA. 6-35

O-OE Q-235, i

pii%4

P-724

Tube Clearances In The Model Soft 4.27. Member. The Figure Figure in between the three the tubes millimetres which exist clearances shows fabrication the The diameters in tube the of soft member. given at used from thirty are average valves the top of the diagram obtained six both for inner the tubes and from one hundred and outer and measurements for the middle tube. fifty three measurements As the diameter outer in and of compression the middle tubes diameter clearance possible a] lowing the the and two for load tube increase. of gap the on the soft member the is increased of the the

decreases After tube

whi le the wi II

diameter tube

inner the the

middle decrease the

has yielded further, and outer to 0.315 mm to longer to at then shed the its the load

tension

even inner

increasing tubes. -n each rn: Occur possible outer critical stiffness resulting It

existing

between

is

this

clearance lateral This to the

gap to movement that

increase of it 0.630 is

side,

a total tubes. tube

between for tube

indicates offer outside

no

middle load.

I ater aI tube tube and soft

restraint wi II fail

consequently

buckling of this

Once the is the

outside negative

has buckled it must

member of

in the

failure

complete 208

Tiember.

details Reviewing the fabrication shown in Figure 4.11 it is questionable to the outside any restraint tubes are behaving elastically. tube the to outside welded

of the type 3 soft member if the middle tube provides tube even when the three component

The top of the middle tube Is plug and when the outside tube becomes at this connection. to the inner tube. in the inner tube

both tubes occur will of rotation unstable, However, the middle tube does offer restraint At failure of the soft member, the strain indicates

the stress in this member to be double the theoretical flexural buckling stress of the tube. This buckling stress for the length to be inner tube has been calculated assuming the effective 0.7 times the actual- length and the tube to be laterally unrestrained at its centre.

size soft members, the middle tube is replaced by flexural These strips have negligible four steel strips. stiffness in but the they tension they prevent stressed will are when even its buckling load from buckling tube critical at provided middle by the outside tube. restrained are laterally they in turn, size soft members and the model soft Consequently, both the full In the full members fall The flexural pre-strain resulting when the outside tube buckles.

investigation undertaken to assess decreases in the load of the middle tubes due to tensile buckling have shown only relatively small decreases of about 7% from pre-extensions of 2 mm. Because both the full size

failed have the tube buckles, outside when member soft and model load of the middle tube due to in the buckling the reduction loads have the the collapse no of effect on soft will pre-straining diagram for 4.23 the the Figure shows stress model soft members. from It be Figure failure. this that the can seen at members Is In the fact of soft member stress in the buckling load of tensile, and consequently any reductions the middle tube cannot influence the ultimate capacity of the soft members. of strain aging on the middle tube may however have an effect on the long term behaviour of the soft members. The limited experimental investigation undertaken has shown Increases to occur in both the yield and ultimate stress when the member is in tension It is a after period of aging. envisaged re-stralned that soft members incorporated into
209

In the middle tube at failure

The influence

space truss

structures

will

remain

fully

overload will load-limiting yielded

under working loads and any significant cause the soft member to yield and behave like a Once the middle tube of the member has device. elastic

then the soft member is prone to the effects of Subsequent overloading of the structure aging. strain and soft member will raise the value of the yield stress in the middle tube in tension the characteristics of its load soft member and alter The load displacement characteristics given in Figure plateau. 4.26B indicates that provided the initial pre-strain of the middle to cause the material to strain harden, member is insufficient of the in he limit t load will cause a sharp aging rise re-straining followed by a return to the normal load-displacement behaviour. has occurred to strain harden the However, if sufficient pre-strain after increase the re-straining after aging will middle tube material, load limit the of the soft member. Any significant of magnitude load offered by the soft member increase in the predicted limit distribution the in turn stress alter within the encompassing will space truss and may change the collapse mode of the structure.

210

CHAPTER 5 OF DOUBLE-LAYER BEHAVIOUR THE EXPERIMENTAL SPACE TRUSSES

INTRODUCTION improving viability of space truss have been behaviour, four model double-layer space truss structures The first two model structures fabricated and tested to collapse. ductile load-displacement tested have been designed to exhibit To investigate the by tensile in to the extensive allowing yield occur characteristics bottom chord members. The last two structures tested have also been in the bottom chord members, but in designed to permit yield force-limiting the devices described in of novel several addition, the preceding chapter have also been incorporated into these structures. The experimental testing of model structures cannot by itself for formulation basis the of new or updated design a constitute is In but experimentation used conjunction when with a methods, the be analysis concepts can extended, with mathematical verified due caution, cover the design of a wide range of related The primary this objectives of experimental structures. data of the investigation experimental were to obtain reliable load-displacement response of these structures and to use these data of the assumptions made In the idealised to assess the validity to collapse analysis program outlined in Chapter two.

theoretical

EXPERIMENTAL Fabrication of Models model structures tested were al I

square-on-square The double-layer space structures with a mansard edge detail. 1.80 bottom five square on a metres plan with grid of were models in directions bays each of the two principal and a top grid square direction. In Each structure was 254.56 mm deep bays four each of in length the the members each structure of same all and making fixing the angle of inclination of the members, used to join the top 451 horizontal to the Figure 5.1 shows bottom at chords, plane. and

The four

211

97
CD

22
CD

se

33

99
CD

ea
m

55
C-0 co

r-

93
r U

21 r, (D

Si

32

95 V)

13

as C)

ii

i, 7
20 so
(S)

l as 31 sl 12 S2
co ED

0
tj

65

ig Lo (D

as

30 C) LD

e7

as

52

TOP CHORD MEMBERS

z
7 el

is

82

29

83

51

WEB MEMBERS

56

17

57

'28

58

-9

!59
Ln N

50

so
M

61

51
q

is

52

2_7_

53 T

38

54

is

55 cr)

,a

is co

15

47

26

1 -18 37 1

AS

18 r C-4

50 C,

is

v
3 il

14

42

25

A3

s :?

17

115

58

c
2 36
CD N

BOTTOM CHORD MEMBERS

13 37 24 38 35 39 6 le
Le CN

57

31

12

32

23

33

-1

34

'5

33

5s

Me

Node And Member 5.1. Fiqure of shows a plan and elevation bottom members with chord in Each each member numbers.

The Figure Numbers For Models 1 To 4. the top Chord members, web' members and their corresponding node and member had a length structure of 339.5mm. 212

the Also shown on of model structures. of one elevation and plan a the Figure are the node and member numbers which apply to each of the four structures. of constructing perfectly pinned joints about the three which allow equal freedom of rotation axes, the model structures were fabricated with rigid principal Although this violates one of the basic assumptions used in joints. to have the collapse analysis program, it was considered beneficial instead of an rigidity assessment of the joint failing joint rigidity between the two extremes of the idealizations. and fixed-joint perfectly-pinned a realistic indeterminate
Member Manufacture:

To overcome the

difficulty

from six different The four model structures were fabricated Five of the member types were annealed, cold-drawn, member types. from tubes the supplier to comply with ordered steel mild seamless, BS 6323 the requirement given in the British Standard Specification (1982). The sixth member type was a solid, bright steel bar ordered to comply with the requirements given in BS 970, Part 1, (1983). Table 5.1 summarises the member types and gives the nominal outside diameters, wall thickness and member areas. Member types classified in Table 5.1 as T2, T3 and T4 have been used in the fabrication of the model soft members incorporated double-layer grid models. into the third and fourth

A suff ic lent quantity of each tube and bar was purchased in an from that type to each came member only one batch of ensure attempt in fabrication tube bar the Each separate and used of the material. from taken colour coded and samples was were each structures model stock length to determine delivered tube was steel bar in the lengths long solid supplied was stock while of metres 2.440 metres. All the members were prepared from straight undamaged Each of the long annealed tubes were f irst cut In tubes and bars. individual the then half members were cut and accurately and length lathe 339.5 Two to in the of mm. a a of members cut machined from each stock length of tubes T1 and T2 (Table 5.1) were tested to One of the members cut from each stbck length in tension. failure (Table T6 5.1) bar T5 tube and were also tested in tension but, of in addition, two members from each of these tubes and bars were also
213

the material The annealed properties. in random lengths approximately 6.0 to 7.0

Member type

Outside Diameter

Wall thickness (mm)

(mm)
T1 T2 T3 T4 4.76 6.35 7.94 9.52 0.91 1.22 0.56 0.71

Cross-sectional area (MM2)

Use

11.00 19.66 12.98 19.65

Tension members Models 1,2 and 3. Tension membersmodels 4. Soft memberinner tube. Soft Membermiddle tube. Soft Memberouter tube. Webmembersmodels 1.2,3 and 4. Top chord membersmodels 3 and 4. Top chord membersmodels 1 and 2.

TS

9.52

0.91

24.61

T6

10.00

Solid

78.54

The Table lists Member Types. Table 5.1. the six different member double-layer types usFd in the construction of the four grid Member types TI to T5 were annealed, structures. cold-drawn tubes complying Standard the British with seamless, . mild steel BS6323 (1982). Specification Member type T6 was a solid bright mild to comply with the requirements steel bar specified given in British BS970, Part 1 (1983). The properties Standard Specification given in the Table are the mean values obtained from the test samples.

Linear variable differential transformer transducers positioned at node numbers: -

Mechanical positioned

dial qauges at node numbers: -

Models 1,2, 3 and 4

31,37,26,25,27,15,38, 14.28,4,39,3,17,5.32, 20,33,9,22,10,21,16,11.

Models 1 and 2 only.

7,18,8,19,42,53,29,30, 36,24,23,12,13,35,34, 2,46,57,47,

Positions Of L. V. D. T. Transducers And Mechanical Dial Table 5.2. In Models 1,2,3 And 4. The Table gives the--no-creGauges U displicements numbers at which the vertical were measured using differential transformer linear variable transducers and mechanical The dial gauges were used as a precautionary dial gauges. measure in case voltage readingsfrom the L. V. D. T. transducers were lost due to the power supply. to interuptions The mechanical gauges were in models 1 and 2 node displacements only used to measure certain but these readings were not used in the-experimental investigation.
214

tested

to

failure

In

compression. using of

member tests adopted model, carefully for

were undertaken

These ninety-five individual an identical procedure to that the model soft

the component tubes After in the preceding chapter. the diameter and wall measured using a vernier

member, outlined the members were prepared for each thickness of each component was calliper.

Joint manufacture:

The joints

structures were very similar (1981). Figure 5.2 shows details of a typical joint, all of which The machined from 28.58 mm square mild steel bars. were carefully held in a jig while each face was machined flat joints were rigidly and each member-locating hole drilled using a milling machine. The to a depth of 8.0 mm member-locating holes were accurately drilled of 8.0 mm of longitudinal giving the members the possibility movement when they are positioned between these two end nodes, prior to welding.
Model assembly: on Each of jig steel (1981). the four models were accurately assembled for the experimental manufactured work The jig

used in each of the four model space to those used and described by Collins

from rectangular was fabricated hollow steel sections and was pivoted about the axis passing through its centre of gravity, so that the whole jig plus model could be The nodes were fixed if necessary. inverted to the Jig using special supports node accurately time-consuming height plane. base to position the which were adjusted at their in the correct location. This proved to be a very because small procedure alterations made to the in the horizontal node also changed its position bottom levelled layer corner nodes were first to each other accurately with check The remaining working from

a special reported by Collins

of the The four and

relative nodes. measurements made between diagonal ly-opposite bottom chord nodes were then positioned and levelled, positioned the four corner

nodes using a 2.0 metre long steel straight edge. The top central node and top corner nodes were positioned using a Jig which bolted onto the top of four bottom nodes and provided the After key top nodes had been depth. the five grid correct the remaining top nodes were positioned and levelled, levelled The and again using the steel straight positioned edge. and level of all of the nodes were rechecked and small position The nodes for the first inaccuracies corrected. model took almost accurately three days to position before the overall geometry of the model

215

14.29

'1 5A

II LI

-41/:
cq

C', co

_L.

Co Ln c6 r4

--4V'--

t
PLAN
IF ip

r
cI
SECTION A-A
TOP LAYER JOINT LOCATING SLOT

-lo

C)

rl jr I 8 ELEVATION

A. LOCATING -2B. 2 BOTTOM

HOLE

LAYER JOINT

LOCATING

SLOT

rigure 28.58mm outlined each and machine.

5.2.

Joint mild square by Collins sloping

joint DetailsEach was manufactured bar steel using a similar procedure joints (1981). The held were rigidly face hole and was prepared using

from a to that in a jig a milling

216

was within acceptable limits, with the location node within 0.25 mil of its true grid position. structure After

of each

the nodes had been accurately located on the jig the Each member was members were inserted into their correct positions. checked to ensure that it was free to move along its axis before welding commenced. The members were welded to the nodes using four symmetrical ly-placed spot welds, made using a versatile mig welder. A strict operational procedure was adopted for the welding of each The f irst members to be welded were the of the model structures. eight members surrounding the top central joint, node 31 in Figure 5.1. Welding then progressed in a clockwise sequence with the welding of the members at the bottom nodes 26, 37,36 and 25 . (Figure followed by the members at the top nodes 20,21,32 etc. the members were welded to the joints, each spot degree of checked to ensure that a suitable weld was visually penetration of the joint had been achieved and that the tube walls Any tubes which were of the members had not been burnt through. damagedduring the welding sequence were cut out and replaced by new members. Three members were replaced during the fabrication of the four structures which required a total of three thousand two hundred 5.1). After all separate spot welds.
After model structures removed from the jig by the the they were were f inally checked, by first the perimeter releasing Once the structures nodes. were

carefully nodes followed

remaining free from the jig they were placed on a flat lifted table top and edge. No rechecked for alignment and level using the steel straight joint in had occurred in any of the models, position change apparent however one of the thin bottom edge members in model 1 (number 33, Figure 5.1) had a discernible bow, indicating some joint movement This member was subsequently member. the fabrication initial of the steel

stresses. and the presence of residual cut out and replaced by a new tubular Although structures it fit, present during is care apparent was taken in

to prevent

member stresses that

due to

some residual resulting

in each of the structures,

lack of member distribution stress was from the heat generated

welding of the members. These residual stress distributions the complete structure can only be decreased by stress-relieving and it was not possible to undertake this the unfortunately, within budget. existing

217

Each combination

model of

structure

was

the members given

constructed in Table 5.1.

from

a Figures

different 5.3,5.4,

5.5 and 5.6, show the member types and colour codes used for models The soft members used in model 3 were 1,2,3 and 4 respectively. 4.11) while the soft members used in type 1 soft members (Figure model 4 were the type 2 soft members also shown in Figure 4.11. Equipment Loading structure hydraulic system: _ models The system consisted used to of a 50 unit load kN all each of hydraulic the four actuator, space

manufactured its own load cell and contained into a closed transducer, displacement which were both incorporated by the control The loading loop control unit. system monitored the actuator would impose pre-set system was arranged so that to the opportunity displacements providing on the structures, investigate Both the the rate by to post-ultimate strength behaviour and magnitude of the actuator the D. E. C. LSI-11/2 micro-processor. incorporated into digital converter displacements of the models. displacements were Using the the 16-bit

power pack and a control The hydraulic Howden Ltd. actuator

a by R. D. P.

controlled analogue

0.05 be imposed could mm small as microcomputer, -as ' by the actuator over its ful I working range of 100.0 mm. Before the complete system was calibrated by any tests were undertaken, The actuator load cell used in conjunction R. D. P. -Howden Ltd. with the control Institution to comply with the British was certified (1985). 1610 in BS Grade 1 requirements given unit test Standards

models the. actuator was bolted to a sliding joint which was in turn bolted onto the loading frame placed beneath joint The sliding the structure. of a series of parallel consisted horizontal plates separated by hardened steel balls which permitted To load the 11.0 in direction. Test models horizontal free movement of any mm a joint 1 and 4 were loaded through the actuator at the top central This was achieved by bridging (node 31, Figures 5.1). over the top fixed joint the above and attached to the a steel using plate of The steel plate by two 500 mm long threaded steel rods. 10 top mn diameter hardened steel ball of a on which was rested in the top of the node. Figure 5.7 seated in a small cup positioned in device during the testing the operation of model 1. shows actuator

218

a,
r6
t-

Ali

PV

AV

r6
%_q

T6
% T oa

T6
> -

10
WV WY TG 0 R TG

:g WY T6

WY TG
tj 0 tD (D

wle WY T6
%5 ID

0:

R TG
0 0

R T6
ct %D to CD 2 -

R T6
%D 0 %D > , Ix ta 2 1. -

W %D > rG
%D

1 0 %D 2 e-

rro r6

1 c 0 'D co '? >-

8 TG 1

a T6 TG
w e 3

TOP CHORD MEMBERS

g 0 4
G
T6

G
TG

G
Tro

G
T6

WEB MEMBERS

t.J

T > Z:
TI Tt
=> - Z: w ri TI

T1 ll Z w w TI T1
Z: CD w

T1

T1 cz

w w
7> --

w w> rl r1
> Z: w ri TI

w TI T1
W w

w
0 Z %Z

w TI T1
w, w

w
0 cz --

TI
CD

TI Ti
CDz: li

TI T1
CD CD

> Z: T1
>

T1
CD

G T1
, Z - Z
>

G ri r1
.I ,:
Z: >

I
7ZI

G TI T1
Z: CD Z:

TI T1

I
CDZ

T1 TI

BOTTOM CHORD MEMBERS

co :

IIIIIIIIII

Member Types With Colour Code Identification Used For Figure 5.3. Of Model 1. The Figure shows the member types and The Construction the corresponding colour codes used to identify each member used in the fabrication of model 1. Tube types: T1 T5 T6 tube; 4.76mm outside small hollow 0.91mm wall thickness. large hollow 9.52mrn outside tube; 0.91mm wall thickness. solid bar; 10. Ommdiameter. diameter; diameter;

Member Colour Code: G= Green; Y= Yellow; W= White; B= Blue; R= YR = Yellow Red; BY = Blue Yellow; WR = Red; WY = White Yellow; White Red; YG = Yellow Green; W3 = White Blue; WG= White Green. 219

AV, / ,Z
ID o ,

fl, , 'j
0 m

RVW
cl

R",
c 2

ER r6
ID Ic ID ,r c e n ,

EG Tro T ro
tc .

EG rG TG
e tc

ER TG T r. l
(ec e Z , cz

BGW TJ
ID 4 :n

evi T r.
-3 3: 1
0 cc

ew 76
.p .v
e in

BC, v , G

co ID
pQ P2 -,

0 7 9
c ID

Tr. TG to 'D e e uo co cc
.

BG EG TG -2 5 , ID ,
1 1.4

E5G EG TG

-I- -IBR 76 J 2c . CD-

"' z

Co

TOP CHORD MEMBERS

$o

WEB MEMBERS

C, TI

I.,

r li

F vi

PIW

TI

TI

TI

TI

rI :0 aw

ri r1 cc
cc -

TI 71
cr CC CC

TI
C, a ED ED

TI
C, cc m m

TI co
CD

S4 rI CD 5G TI CD cc sa Ti T1 > ED i-

a" TI cc EG BG ri T CC, CC) EG 71 cr C BG

t-

rI
C

81-J rI 0 6G rI ri cr CC) CD

SG a r. TI

51 BG rI

cc

eG rl

BG TI cf IL eG

BG TI cc ) C:

BOTTOM CHORD MEMBERS

BG

BG

IIIIIIIIII

Code Types Colour Identification With Used For Member 5.4. Figure Fu-ction Of Model 2. The Figure shows the member types and T-h-To-nst! each member used in tLFe-coTrrespond1nFg colFour (codes used to identify 2. fabrication model of the Tube types: T1 TS T6 smaII hol 0.91mm wall large hol 0.91m,n wall solid bar; 4.76mm outside low tube; thickness. low tube; 9.52mm outside thickness. 10.0ran diameter. diameter; diameter;

Member Colour Code: BG Blue Green; BW = Blue White; SR = Blue Red; Blue Yellow White; B8 = Blue Blue; 3GW BYW BY = Blue Yellow; Blue Green White; BBW Blue Blue White. 220

0(-

p0Q

(-,

T5
(D 1. )

SM2
cc V') (n v

SM2
CL u3

T5
CD
n

CD

PG pG T5 r5 r PG T5 cc m n V) RG T5 o ', in RG
T5

RG T5 75 In Co co v a RG T5 CD co n PG 75
%n V) Lr) cz Lo CD CC-

PG T5 T T) CD co (D PG T5 c cc n PG T5 75
41 (Z Lo m C

Pci T5 T cz co v. V. CY RG T5 (X PG G T5
e n co L.

l coI Co cc

TOP CHORD MEMBERS

R
SM2 5 M2

PG
T.*

WEB MEMBERS

D9Q2a

ca

0A0A

ri

rI

rI a

TI

TI cl,

0 cc
ko ct

rr
Re TI Re T Re T

Re TI cj

PB TI

Re TI
(2 0.1
py

RB ri

RB F

P13 TI

pa TI

py

py

RY T1 CC Cf Ry TI -? cr

Ti

rl

71

py 71 cr TI cc RY

BOTTOM CHORD MEMBERS

RY T

Py ri

py ri

ply

RY

py

P LIj

IIFIIIIITI

Types With Colour Code Identification Member 5.5. Used For -igure r Model 3. The Figure shows the member fy-p-e-s-ani-ff The Construction7fE-Iourcodes the corresponding used to identify each member used in the fabrication of model 3. Tube types: 0.91: nm small hollow tube; 4.76 outside diameter; wal I thickness I at,ge ho II ow tube; 9.52 ou ts i de di a-me T5 ter; 0.9 1mm wall thickness T6 solid bar; 10.0m-n diameter Member Colour Code: R= Red; RG = Red Green; RB = Red Blue; RBG RBW = Red Red Blue Green; RBB = Red Black Blue; RY = Red Yellow; Blue White; RW= Red White. SM2--Soft Member Typel T1 221

vr.

Fk

vI

. -I

T5
u' -

61-15

5M3 ,:, 'n Y6 5 T5

T5 co
>-

YG5 T5

II- YG8
r5

T5

YGB T5

YCB T5

YGB TS > - L, L,

YGB 75

to

n
YGB T5

, ) - u)
YGB T5

)- o>
YGO TS YCB T5

n ' Z

TOP CHORD MEMBERS

co YG 13 F5 y SM3 y SM3 13G Y T5

WEB MEMBERS

Y G.Q T2
CD CD

YGQ T2

YGIZ T2

YGR T2

YGQ T2
C D I ID m Y114P Y1,4P

I co
Y-Jp Yvp T2 T2 YWR YWR T2 T2 cr cr YWA YWA T2 T2
CD M C\l ,

rJ

YWR r2 r2

YWR YWR T2 T2 cr cr YWR YWR T2 T2

co CD CD

72 72
'JI- C\j cli Cc Cc

>
YWR YWR T2 T2 to Nj Q GQY T? co r's r's GRY T2 CO YGA
.C

C4 co co -

cr cr YLVQ YLVQ T2 T2 ')- N m M GRY 72

YK'Q YP, 'Q T2 T2

to I-

GPY T2
', N co co CY CY

GRY GRY T2 N CD CDN cc cc

CO COcy cr cr

GRY GRY T2 '3CD CD

,"I "1 CD CDa a

GQY G; zy T2 C\j co Cj ce ce VGR

14

Gay Gay TZ > CY cr, cr cr y6p

14

GRV GRY T2 > CD C4 cr cr YGQ

i, _,

Gy GRY
i cr I

BOTTOM CHORD MEMBERS

72

cli co CD Q
I

YGQ
14

Member Types With Colour Code Identification Used For Figure 5.6. The ConsUr--uction Of Model 4. The Figure shoqs the me-mE-e-r types and the corresponding colour codes. used to identify each member used in the fabrication of model 4. T2 = hol low tube; 1. diameter; 1.22min Tube types: ). 35mm outside wall thickness. large hol low tube; T5 9.52mrn outside diameter; 0.91-nm wal I thickness. T6 solid bar; 10. Ommdiameter. Member Co I our Code: BBI 1ack; ue; Y= Ye II ow; YG3 = Ye II ow Green 13 GY Green Yellow; BRY Black Red Yellow; WBY= White Black Yellow; Yel low Blue; YG = Yel low Yel low Red; YW = Yel low White; Y3 YR 31ue BI ack Ye II ow; YGR = Green; BGY =BI ue Green Ye II ow; BBY i te Red; 6RY = Gr ey Red Ye II ow; YBG Ye II ow Gr ey Red; YWR= Ye II ow Wh Red Blue Green; RBY Yellow; Blue GBY = Green Blue Yellow. Yellow = = SM3 = Soft Member Type 2 222

The photograph Attachment Used For Mode 171. Loadin 5.7. Figure displacements to transmit to deviice rom the actuator used sho4s the The 15 wi thick steel the top central plate rested node of !nodel 1. in 10 diameter hardened b3ll top seated steel which was mm a a of on in the top of the node. small cuQ positioned

N1011els Fot 2 b. 8Loadin, Rf-3ifl SY, tem Used Figure atil . ho4_77e 1j)II folpheam t. t) eqjajjv sy,; q jcej f-, photoqY'aF77. fr)tjy(19 Faci the 3. ?I, 43 and 41 ) :)f no]e Is of and ,

3.

The

top two the I oaded nd viaah at, lened at s tee I ba IId Pdr- aIIe l Uie other t0qet 11 I)e Io, b r- idgingpIates connected ib, ) vein, e r- one q Each )f four, these devices to atLaOied qorser)at-, jtejy each node. tkqo pat' aI le I be a!n-, qh -'I f out- ends Ci wer e tur, rl 10 3dej at the of 11 The by one horzontal their, centre ; ectio(I c r, o S,; beam. cr-oss Io aded i ts )a II be arn was at cen tt P- thr- otj qh 3nd 1)1ate tn the +-, , I it wnd ir--c tIv 3tt-ichod

223

loaded at the four top Test models 2 and 3 were symmetrically Each of the four joints were joints 19,21,43 and 41 (Figure 5.1). loaded at the top via a hardened steel ball and two parallel bridging plates connected together one above and the other below to four devices Each these attached separately were of each node. beams, which were in turn loaded at the four ends of two parallel their centre by one horizontal cross beam. The 'I' section cross beam was loaded at its centre through a small ball and plate All of the three beams in the indirectly attached to the actuator. loading system were arranged to be simply supported so that equal The displacements were applied to each of the four top nodes. complete loading Figure 5.8. Test frame: system, used in the test on model 2 is shown in

frame used in the experimental investigation Butterworth frame by both form the test used of was a modified The frame consisted of eight universal (1975) and Collins (1981). run) welded and bolted together to beams (1611 x 51/2' x 26 lbs/ft The models were supported at their four form a square grillage. The test frame four test the on above corners (100 100 x5 mm). x columns, steel square hollow section

Each of the four columns had 15 an thick end plates butt-welded in location, bolted, the then to The correct were them. columns to the test frame using four M20 bolts for each connection. As mentioned previously each of the four models Support conditions: joints, four 1,6,61 its bottom corner numbers and at was supported All of the four supports were designed to 56 (Figure 5.1). free the three about and rotation constraint vertical provide joint 1 the However, to support at constrained was axes. principal in horizontal the the translation plane, while supports any prevent designed horizontal 56 to translation 6 allow joints were and at 61 The node support at was not constrained only. axis along one translation to horizontal and only was constrained any against Figure 5.9 displacement. the details the shows of vertical prevent flat This two 6. support of plates consisted node at support bearing by a race containing separated and surfaced-hardened Two bearing races were also fixed parallel to hardened steel balls. device in the these the side walls of and constrained all other each The bottom plate of the only. movement to one direction horizontal four 12 diameter positioned on Wing was adjustable support mm sl
224

Q) c M5 77 C)

0. =F0
(L)

-)

c: j
N 7C (1)

Q) c

a-

7) > c

Ln

Ln c0

-73
C3

ul)

Ln 0 (71

0V)

1-

i Ol Qo LI; j)

-3

eLi

CT
-Lir, -C7 -, (1) >

225

fixed to the top plate of the support column. The end node of the model was supported off the top plate of the sliding support by into the node. This end means of a spherical end piece which fitted bolts piece was machined, so that the centre of 'rotat-lon of the spherical surface coincided with both the centre of the joint and the point of intersection of the three members meeting at the node. These end pieces were identical to the spherical pieces used by Collins (1981) and were designed to prevent the vertical bending moments on the model structures. Displacement measurements displacements of fourty-three The vertical nodes were measured 1 and 2. This was achieved by using for both model structures (L. V. D. T. ) linear differential transformer twenty-three variable The mechanical gauges. in a precautionary as measure case voltage readings used were gauges from the transducers were lost due to interrupt. ions to the power transducers mechanical transducer lost However, during the first no readings were supply. two tests so the mechanical gauges were not used duri'ng the third For these grids the twenty-three L. V. D.T. and fourth model tests. transducers were the sole means of measuring node displacements. Ten D5/2000 and twelve D5/4000 transducers, with working ranges 200 100 mmrespectively and were used to measure the vertical mm of displacements of displacements of the unloaded nodes. The vertical loaded node was measured using the an L. V. D.T. transducer The twenty-two D5 transducers were incorporated into the actuator. beneath the nodes using the adjustable in position holders These transducer holders supported the developed by Dianat (1979). 10.0 transducers bolted the body and onto a of mm thick mild main just frame. the test The moving above supported plate steel held armatures of the transducers using thin steel wires. All were suspended from the monitored nodes and twenty dial support reaction inducing

investigation used in this were Ltd. have linearity to better than a certified Before each of the four model 0.20% over their working range. The tests commenced, every D5 transducer was calibrated separately. ten D5/2000 transducers were all switched in turn through one in They 1 if I calibrated conjunction with this amplifier were er. amp 100 in their range working of mm steps of 5 mm. The twelve over of the transducers by R.D. P. -Electronics

226

were switched in turn through another amplifier from 0 to 200 = in increments of 10 mm. and they were calibrated Several voltage readings were taken at each displacement increment D5/4000 transducers for factor the each transducer was calculated using calibration and line through the method of least squares to obtain the best straight the data points. Table 5.2 gives the position of both the L. V. D.T. transducers and dial gauges used to measure node deflections In each of the four test models.

Strain measurement
strain gauges have been used to in four the All structures. each of model strains of member measure the gauges used had a resistance of 350 Ohms and were manufactured Type EA/06/125BZ-350 gauges by Micro-Measurements Incorporated. Precision electrical resistance first for three test the structures and used were CEA/06/125UN-350 gauges were used for the fourth structure. types of gauges are manufactur. ed using a constantan alloy foil type Both in a

form CEA the If sated gauges containing with at ure-compen se -temper larger and integral copper coated terminals making thein slightly EA Table 5.3 the than to gauges. gives the strain solder easier lengths for both and resistances gauge types. gauge sizes, gauge
the member axial force fixed two bending strain gauges were onto each of the moments, and 1 in both 2. Two models and gauges were also used members monitored in selected tension members in models 3 and to measure the strains In order to of in these members to be monitored instead fixed to three them two, had of gauges so that a structures forces both the bending axial of and assessment accurate more in be these All the could members made. of gauges existing moments 4, however the compression were carefully fixed to the members after the the models fabricated. 200 and every D. Both of The gauges were cemented to gauge was protected these products were Incorporated. after All members using installation using the were M-Bond M-Coat by were determine the magnitude

recommended and manufactured of strain gauges

Micro-Measurements fixed

two the the and of where member gauges were used mid-length at for each member, they were placed diametrically opposite each other bottom the jig to mark the top the of and member using a special on Where three strain location. gauges were fixed to one member, they

227

-0 (D
4- 0

,a sC) ci
4A 'a r-

14pt

(Y)

ro =

D. 0

+1

0.010
0=

E Ol c to

00 0)

S-

110

lb-Z
Lfl)

4-j

(L)

LO CY)

VI -0 4-J 0) .C S- - S- rM =- -0 4- o S(1) ca. E Qj 4--)

(1)

m
C; +1 C) c: ) r-4 ce) ON al -t:r LCI C Lf) C) C

C)
+1 LO 4.0 C C

I'd E
r. X 0 SC3. to 0) 4j 41
(U 45 C;

W 4-)

Ln cli

M 4- '0
r_ to 4--) C: tio

C)

(1) 4-4 4-) ed L) :3 Ln Su a)

Q)

Ln CY)

0.

(A (1)40 '0

0 E to ., W 4--) ed CE
M., -

LU u

C: Woco V) u

CL C: )- .0 eo U 4-1

4- (A Q) (1)
(A (3) > a r- 0 a)

(0 (A o)
> r

(I) cl. 0 s- -

4A :3

.6C: 0x r. CA a ai *0 CX 0 W (u = a
r.

CA Q) =3

CD

C: a) . CD. (L) SM 0 :3
u C) (0

ux =3 W S- 4 4- J: ) %n
a

m +1
-

C: o to kn =p

m0
S4--l

Ln
(Y)

=3

UE
4-

Irt

la-k
Lil)

4-3

(D

Ln
C; +1 C) co r-4 Olt r*_ r*

C)
+1 LO ILO

93 E-=
0 S-

cli

I r14 co LO

1-4

C. eo oj C) (u E .1-i L) cu C: fKj E 44-J

4fO 4-) 1

rO 4S(1) a -a

(31 Cl 10 L)
LA C3.

:3 =D

a) Ln .0

LO

CY)

Lr) Ln LO CD . CL Lr; 14 14 C f2
.

LLJ
4-S IM C: ) LC) CY) 4--l Ia

0a
CD. r_ to 0 Ca.

LLJ

<

CL) 0 V) u0

4-) V)

4-J

(L)

c: 0

S- 14 *C3 0 4j ca C: r= &A (0

Lr)
0 0

Ln 4- C%J
T3 'a (0 -4 A tO

aj

4) u C)

4-

.00.1-1

fo (1)(= c SUJ

"C3 -

to
4J

u Ln ci

L) a
fo 9

CLS

4-)

CLI a
W C-0

Ln %A CU of

(1) M :3 610 CD

to S(1) > Cl

a (D

#0 S(U > C)

u 0.

r. #0

V)

228

were accurately positioned equidistant the member, also at mid-length.

around the circumference

of

Each of the strain gauges was connected with its own high to form a quarter bridge system. Both the precision dummyresistor were connected to the measurement strain gauge and dummy resistor system incorporated in the data logger using the five wire system shown in Figure 5.10. The measuring system provides twin, constantcurrent energising for each of the strain gauges and this was only switched through each gauge during the actual measurement period. This minimised the heating of the gauge and also decreased lead wire errors to a minimum. output voltage signals from each of the strain Ncomputer program was written gauged were stored on floppy discs. to convert these voltages into strains using a constant value for All of the corresponding to the The values of strain strain gauge under consideration. particular forces to the bending then obtain member axial and used moments were interpretation, to ease using the Hewlett plotted, which were the energising current Packard T221A Graphics Plotter. Table 5.4 gives a list of the members which have been straingauged with their corresponding strain gauge numbers for each of the four test models. Test procedure Before each of the model structures was tested to fai lure the four adjustable supports located at the top of the four tubular levelled accurately using a steel straight-edge were columns support was completed the model structure four top the on of supports and the steel positioned carefully was the transducer armatures were attached to the supporting wires The held in transducer bodies the transducer nodes. monitored and a precision supports were then plumbed and the strain gauges wired into the to the data logger. Each transducer was boxes attached switching independently calibrated and set up to operate over the middle of The actuator and loading beam system were its working range. to the level. When this and the gauge factor

nodes and the structure correct given a small displacement to check the operation of the test equipment. vertical , In addition the voltage readings from each strain gauge were checked connected

229

S+ B2 sBi

DL RE

S+ - SIGNAL LEAD (+Ve)

SINK CURRENT ENERGISING (-Ve) B2 = CONSTANT S- - SIGNAL LEAD (-Ve)

CURRENT ENERGISING SINK (-Ve) Bl = CONSTANT G- GUARD AND ENERGISINGSOURCE (+Ve)

Figure 5.10. used to obtain was connected by energised mi IIi amps.

Strain Gauge Circuit. The diagram shows the circuit a quarter bl'idge operating system. Each strain gauge with its own high precision dummy resistor and was a switched twin -constant current supply set at 5

230

Model

Model 2 Member Numbers 63 62 61 82 101 184 1 2 3 7 8 9 12 13 37 38 32 33 Strain Gauge Numbers 9,10 35,36 25,26 1,2 29,30 23,24 31,32 17,13 7,8 21,22 37,38 39,40 19,20 11,12 S. 6 33,34 3,4 27,28

Model 3 Member Numbers 70 69 66 65 62 61 101 199 82 13 8 7 3 2 1 Strain Gauge Numbers 20,18,17 33,34,23 21.11,12 26,25,24 7,40,8 36,35,22 29,3,30 5,6,19 4,1,2 37.38 13,14 39,9 15,16 27.28 31,32

Model 4 Member Numbers 70 69 66 65 62 61 101 199 82 13 8 7 3 2 1 Strain Gauge Numbers 20,18,17 33,34,23 21,11,12 26,25,24 7,40,8 36,35,22 29,3.30 5,6 4,1,2 37,38 13,14 39,9 15,16 27,28 31,32 19

Member Numbers 58 53 48 43 38 33 3 8 13 18 23 28 118 134 150 98 94 90 86 82

Strain Gauge Numbers 39,40 37,38 35,36 33,34 31,32 29,30 27,28 25,26 23,24 21,22 19,20 17,18 15,16 13,14 11,12 9,10 7,8 5,6 3,4 1,2

Table 5.4

-3 and 4. Te Ta -e correlates '19-0-6e] StrRtures-1,2, member numbers four test the for of each model numbers gauge with- strain to in Two the were strain gauges used measure strain structures. 1 2 three to and and were used models gauges of members selected in 3 the selected models compression of strain members measure and 4.

Table 5.4.

Member Numbers And Associated

Strain

Gauge Numbers For

231

The gauges removed and replaced with new gauges. and any faulty reading from the strain gauges were also checked to ensure that the For the last structure tested, model was deforming symmetrically. load cells were placed under each of the four supports during the Each of these load cells load ing sequence. pre-test forces independently monitored to check that equal reactive obtained for each support. Each of control with per minute. node by the structure the four an initial This rate was were

structures was tested under displacement . displacement of the loaded node of 0.1 mm of displacement was imposed on the loaded response of the

was apparent the increased the displacement to 1.0 mm per actuator of was rate of displacement the total After vertical of actuator of a minute. displacement increased 100 further the to rate was mm approximately 2.0 mmper minute.
The voltage signals from the actuator, transducers and strain floppy disc recorded on and every thirty read seconds were gauges During load-displacement the the test. throughout each test behaviour plotter emanating of the structure by monitoring from the actuator. was continuously both the load plotted and on a Bryant x-y displacement signals

actuator was linear.

while the load-displacement When non-linear behaviour

RESULTS
Component members

ive members have been tested to determine of the different component materials both the physical properties individual behaviour the load-displacement tension and the of and The these tests have been majority of members. compression (Table T1, T5 T6 types 5.1) tube and on and complement undertaken A total of ninety-f the undertaken on tube types T2, T3 and T4 used in the To fabrication of the model soft members, reported in Chapter 4. the different in the physical properties of assessing assist tests

from the a small sample e, ach of six member types members, component its determine Also to constitutents. chemical an was analysed from the taken each of six member types and was sample additional Table of the different steels. used to determine the microstructure 5.5 compares the chemical composition of Samples taken from each of

232

c;

00
CD

Ln

CD

_A

cm

3d . u Cm L2 c ," m CD 3: w Cy%

C CD

c C:
)

c
c;

Co

nr

CNJ
cli -0-A ro u03: (A (0 0) C

kCU CD 3:
C%

c:,

CD CD C-- 0

c c;
m

CA (1) r0) c = J_- 4-3 r15

E -j eo -4. ra

Ln

CNJ

'0 to
0

0a

E
rz c

lu be . r_
3: 41

CD CD

c;

C: ) c;

CD c

C)_-E 4U a) Ln
r

(D

(7% Ln -0

tA (L) C:
:;

4M 0. (A
coo

a
-w C.

c (0
0

c;

. -4 r-

cv Co C% ci c

10c

CL)

(L)

0 Ln M CU . -0

F-

CD
!2

CD

C)

C:)

CD

S. CL -se - - ,a C, > E"0 4- u

u 0 4--) r0)
cl 10

(A

c; r. be Ln CD cz; 3: rc; m c -4 CD

E 0 C) C:
(D c;

>1 C: (A (1) =
-

to r, : 0 CD

S- 0 Cl. tA
Qj 0
1,5 U0 -

=
z

'a

4A 0 w %A .CL) .-

(L) 4j +j -0 L.,) 0
"

C--j 0 fo (1) CO c cn CL E

4Ln

E(1)

V-4 .0 L9 LO CD Uli (D 4- tA

0 4CU

U, `0 z ccco
3: 0 Co $- M CD-4 m $.. 0
vi (D Ln 41 1(D CA

4-) - 3:

a r-4

CD
tA C>

0 kA 'a (D (1) 4-) a C: ro

Wo -0 to cl F= 0 Lc;

LL. le
40

(D

CD

U -

to #10wo (0 CL Eo-

to . S"m ro o S 4-) wo -C

(U 1- 0" 0- 0 :3m V) 4- 4 NO

4.1 1; u tn M IA ZU c M r= =3 -9 0 u c

cu 4J c 0 0) In 0) c

41 c 0

41 r_ (L)

u
c 0

%0 m

41 Z

-2 0 a) E 3c (3.-- ;Z 2 cu r= m0 10 L.) c-) -

0 m 9-3

u V)

CL Ci-

m tA

233

with the specified (1982). in Appendix A BS 6323, 1 Part of composition given chemical Table 5.6 allows a similar comparison to be made between the chemical composition of a sample taken from the solid bar, type T6 the steel and the specified chemical composition for this type of (1983). 1, Part in 970, From the results of the BS steel given in Tables 5.5 and 5.6 it is evident that presented chemical analysis of the six materials comply with the composition given in the relevant British Standard Specification. all Figure requirements (Tab Ie5.1),

five

tubes,

T1 to

T5 (Table

5.1)

5.11 shows photographs of the microstructure obtained from each of the six samples. These samples were first polished and The then an optical microscope. photographs under viewed and etched the the of material at a magnification size of two grain show The black crystals present in hundred times the actual grain size. in the photographs, are crystals of pearlite all and amounts varying their prominence gives a direct indication of the carbon content of the steel specimens. A high density of pearlite crystals shown in the photographs of the samples taken from both member types T5 and T6 correspond with the carbon content of 0.08% by weight (Tables 5.5 from the 5.6) chemical analysis of these two samples. obtained and The photographs of the specimens taken from member types T2 and T3 both show a decrease in the density of the pearlite crystals when from types the T5 Even member photographs and T6. with compared larger decreases in the occurrence of pearlite crystals are shown in the photographs of the samples taken from member types TI and T4. These visual estimates of the concentration of pearlite crystals and hence the carbon content are also supported by the values obtained from the chemical analysis given in Tables 5.5 and 5.6. The six photographs shown in Figure 5.11 give a good indication different the This In turn the size of material. grain physical of the ductility indication the of yield strength and of an gives different materials which tend to improve as the grain size of a decreases. Table 5.7 gives the grain size of material particular (1948). in ASTMS the with samples measured accordance six each of The samples taken from the member types T5 and T6 (Table 5.1) have In addition they also the smallest grain size of the six samples. have the highest carbon content and should therefore exhibit higher higher than the corresponding stress ultimate values and yield four by the remaining samples. given values stress

234

Chemical Composition ladle analysis % allowable. Carbon content Silicon content

Specified properties 220MO7bar BS 970, Part 1 (1983). 0.15 max 0.40 max 0.90 to 1.3 0.070 max 0.20 to 0.30

for

Tube T6 10 mm outside diameter solid 0.083 0.03 1.12 0.052 0.205

bar.

Manganese content Phosphorus content Sulphur content

Table 5.6. composi't'l'on properties composition independent

The Table compares the chemical Steel Bar Composition. 75T'Mined trom bar t pe T6, with the corresponding (1983). 1 in BS970, Part The chemical specified was obtained from a ladle analysis undertaken by an testing laboratory.

Tube Type (Table 5 1) T1 T2 T3 T4 T5 T6

Average ASTM Number 7.5 5.5 5.5 7.5 8.5 8.5

Average Number of 2 Grains per mm

1536 384 384 1536 3072 3072

Grain Size Of Tube Test Samples. The Table gives an Table 5.7. for the grain size of each of the six samples. The valu4 average grain size is specified using the ASTM index number-which is based on the formula:
Number of grains per square inch = 2N-1 (at a magnification of 100) where N is the ASTM index number, (ASTM, 1948).

235

% A

TUBE

TI

TUBE

T4

07.
a

. '.

.,,

".

I.

0 1,:,

'T

',,

TUBE

T2

TUBE

T5

:-+
li:
__"__i

y
--

lzm:

TUBE

T3

BAR

TS

l"Oure 5 11 Steel Microstructure The photograph shows the steel microstructure of each of the six tube types (Table 5 1). From the individual photographs an estimation of the material grain size can be obtained. The black crystals present in varying amounts in all the pholographs, are crystals of pearlite and their prominence gives a direct indication of the carbon content of the steel specimens. The black strip shown at the top of the photographs of the specimens taken from tubes T4 and T5 is bakelite material used to back the samples. The photographs were taken at a magnification of two hundred. TubeT1 4 76 mm outside diameter 0.91 mm wall thickness. TubeT2.6 35 mm outside diameter. 1,22 mm wall thickness. TubeT3 7 94 mm outside diameter 0 91 mm wall thickness TubeT4 9 52 mrn outside diameter. 0.71 mm wall thickness. TubeT5 9 52 mm outside diameter. 0 91 mm wall thickness. BarT6 10,0 mm outside diameter Solid bar.

236

A and B show typical tensile load-strain obtained from tube types TI and T2 (Table relationships 5.1). The small diameter tube, (4.76 mm) type T1, was used for the lower chord tensile members in model structures 1,2 and 3 while the (6.35 larger diameter tube, mm) type T2, was used for the slightly Tension bottom Both of the members in model 4 only. chord tensile load-strain relationships given in Figure 5.12 show a definite yield by a constant load plateau. followed However, the point exhibited by the type T2 members gives a higher value relationship for the ultimate-stress to yield-stress ratio than the type T1

behaviour:

Figures

5.12

members. 5.13 A and B show the tensile load-strain behaviour (Table T5 for types T6 5.1). The tubular T5 and member obtained members have been used to carry both tensile and compressive forces In particular in all of the four model structures. all of the top in 3 4, from the eight soft members models and apart members chord have been fabricated from this tube size. in each structure, The Figures members type T6 have been used for the top chord in both models 1 and'2. The typical load-strain members compression behaviour obtained from the tubular T5 members shows both an upper followed load by lower stress a plateau and an increase in yield and solid bar due to strain hardening (Figure 5.13A). capacity carrying However, the tensile characteristics of the 10.0 mm diameter solid bar, member type T6, d id not exh lb it a def in Ite yield po Int and failure of the test samples occurred after a short load plateau, at (Figure 4.7% 5.13B). total strain of an average load Table 5.8 gives the values of the elastic modulus, upper yield the stress and percentage elongation of the gauge ultimate stress, length after fracture, obtained from the tensile testing of member Table 5.9 summarlses these results types T1, T2, T5 and T6. and the the of experimental mean values of yield stress, compares ultimate stress and percentage elongation with the minimum values. British Standard Specifications in the BS 6323, Part relevant given It can be seen from the values 1 (1982); BS 970, Part 1 (1983). in Table 5.9 that the yield stress of each 'of the tubular bar is the solid significantly plus greater than the minimum members 170 N/mm2 and 400 N/mm2 respectively. stresses of yield specified given However, the average percentage elongation at failure of each of the different member types is less than the minimum specified values. in gauge length The apparent discrepancy is due to the difference
237

lo

7.5

2.5

5 PERCENTAGE STRAIN

10

rTr,TjRE -S--122

SMALL

HOLLOW

TUBE

(WHITE)

--i

2468

le

12 .

Is

18

28

22

Strain Percentage

MODEL 4 1Za i, rTr, uRF, -,

RWY

le Load-Displacement Tensi B. Relationships 5.12A 13 Fi i and (1 ure Abit ua 7_ypes 7u 71 The From 27 Min And 2. T Figure shows typical e rom UFF I- e Wtained from tube types T1 and behaviour loa2-displacement tensile (4.76mn) type T1 was used for the lower diameter tube The T2. small 1,2 in 3 tensile structures members and model while the chord (6.35mm) diameter type T2, larger for tube the was used slightly All of the tensile bottom chord tensile members in model 4 only. displacement tested were under control at an initial specimens 0.063% This of per minute. rate was increased to 0.63% rate strain 4% had in the after strain approximately occurred minute per specimens.
238

40
: k:

30
6-4

1-4 0

20

10

5 FIGURE 5.13A PERCENTAGE STRAIN

10

15

LARGE HOLLOW WHITE

A (TENSION)

40

30

20

10

10

FTGURE-5-. -U3.

PERCENTAGE

STRAIN

SOLID RED WHITE

(TENSION)

Tensi e- Loa Disp acement Re at onships B. -AnT767. i The Figure typical Types lb sh Obtained From lube from T5 tube type obtained and ehaviour acement isp ensi e oa in been T5 have the top chord of tubular The used T6. members bar in four The 4 3 all also as members models. web and and models in the top 1 T6 have been bar used chord models of and members solid All of the tensile specimens were tested under displacement 2. 0.063% initial This rate of per minute. an strain rate at control 4% 0.63% had to increased after approximately strain minute per was in the specimens. occurred
239

tu c 1.1
'" OM c3, u

. .2Z',
CL, ru 1.2:, cm ci

'. 0,

M Ln Cy m r ei -! ei fi 99

r-: c c

'! 9 - ! -1, - - -! '-: -!

e4 , cy

wm Co

ch. w cn -0

C"

m
CD cy, Ln (L) = 0 m -. %0 e r

E.

..... , c ; 4 ce m440, ul %0 %0 r o in mmmmmmmmMmm

cy, CY U-b w ..........

Co u

CD CD Ln

ei

Co -&

:r

km c) cn Co " "CO Co 0--D Co - re MMmmm

4 ; Z c ei ei c; c r4 c. t9 v c ri -; r ei -; c; -; c: ,; c ei A tn -; -;. -: vi AD %0 r v %0 me c> c2 cn 0, c2, c2 c> cr Co (3 %0 tn Ulb tn pl mmm -e r -r w r emmme ;rm en r ul vi ab Ln Ul n Ul Ln tn Ln rr 4

"4

tu

0% CD -e

Co %0 cv

r4

-w 91 C-

tn

Mm

C>

Kn Ln fl

Vt

CP Z;: sw .

CD r,

te

Co r

93% r

teb P% CO CYNr -ir %0 CY Cm

r4

c r4 _: Co CO r
"""

44

IC %0 Co %D m r4 cu cy Cv

........ d= e c-. -r Lrb "

0, "

Cy

co to c> Ln tn 0,
to v4

r. f .............

mr

cn U)

r4

ej o

%0 r r, CY cli

CD l r "

CY% Co M

en

40 Ln f

cr i

(A ..

":

19 (D

V Ln CY CY ...

LA c; c ri

-;

Co rCY "

".

mm"

Co Co CO Co m" CIJ Cm

%t>o r,

CD cn m cn
(D

r,: kn........ cy, cm Co Ci

Ul ul " M : r " M M r M

C C> c l i o

ve m

CN r_

.11.

tu 4

csj " M pl

w (U raa,
0 0 w=-

c, C,

) (3 CD CDCDCD(D C) CDc> c> (D CDC> CDCJ m (D C> cn C: CDc-- (D c> 4-- cn 431 IJ r4 CIJ- cli CM " r4 "C I J CJ CM CY CIJC cy CV

t cg 1, fi , m rm

C! 9111 W! c 1, c W! 1, vi P c9 ll: ll! ul c 1, N Il 19 1, ul c c 1, r 't -! -! -! 43,cy. r, cn m e CD(3 r r, C) r, tn -& "mr m r -e in n tr) 40 r,. r, Co ul g" -,
0 CylCY, CACDCDCD(D CDC> c cn M CD(21CD " CYC CY r4 IJ -4 CM

CD Dt CD C3

(, % c, % CO CO cn

C3NCl% l>

cn M

C% VN 0% ch

CA CD, 0' CY%

CN 0% CY, 934 %0 to

%0 to

to

to

4D o le

ID

-e

-r

-e

r -r

-d- r

rrr

-4

-4

c:, (D c> C:) CD ci 0

0% VN CM cn l> crt 00 CO CK)CO CO CO CO CO CO CO e d= CD CD CD CD C) c2 CD c21 CD C> CD tD g= M In C, CYN

u1,9,9 %9'9'9%9
r ir tli qr

64!
.11zcI 4 )-,

(p

W Lrb --,

CY CY "

cy

"J

rq

eq

cy

MM

rn

g"

MMree:

re

&rr

-4.

-4 4

-4

. 4.

-1

-4

4 -4

.4

-4

Z C
4)

O= 3:

15, s0b -: CL

! w 5 . CL

.. '-.
.-

CYr4 - cm 1-- >%0%0%0 CY ri o %0tooo - -r4 r4 m n tn Nn u Vb a - -4 4 -, ->bbkb&tF->t-b)9->)-b-1-bbeb)b-. b-. kbb-. Fb- -. >- -- - k- b- F-- 1--

o 10 Ic 39 Ir

cu

.. c ww
12

tu

41 -. lu: 2 -

` 3: i 4) c cm
0

c to Co 1l ; wo, 3; 1u2212
0, ww ti w 4) ,2 .2c'. u cil CO

-C

1. 4-

C"i 0 06 M

39 (1) tu 1t.... - tu -KI 1- - 0

cm ti

C -..o
m.

ll c > >. m -0

z3 cu cu
-0 lu

, 0,2

tx 9x : 222c
tu cu

11 v--

4) -2 -C 29 >c Ct dz

ic

cx

'm -

-c Co
i>

'

-3 :, -. f Ir
C! c.

2-wuuuuuuuu&1t?

-w---Z; , zz
v55

-WMM.

Mg> to

g; 222s;
l -"

;; 2; W0,IU "C w 12 12
WXW
4 I 4 A, l l "

-1 1:, lel, 1t 1153 MIU er-0-0>- >>IU >IU

.5-; t4,4.

9,2 22Z--

00

-. -i , ' Z , ,Z I tu 4, gi w Us .22-e s VE: 2

Co cm Co
--

....................

1
; t,

c c o! c C%

4 c Il c Il c cn c c c l'! c

r4 .... 14 14 d c ch cn cn 0% c c rrr:::

tu f-j rj e

-------------000000000

(D . Im w0 9WW

11

c dD% (71 c c c 0 c> c> C, r ga eg c2. c>

c r=
-0 mo 644-) -

.......... cz r2 c2 in t5 en cm cri cmcm im tz C, ci gn r2 CD Z c5 e2 Ca a accIcCa Cc c Ca c> c QQ0000

r r rl Il Il pl: r,: r, ": l' p l' p l' - r - r r r-:

c2 r2 cm e3 c3 cm c5 c2 e2 e5 ac a al: a a CL a
rfi fi ei e't

c3 :;;; :;:;;
lc>C)CDC>C)C>C>CDC>o

:; ::;

so ll,

CD

;rrrr

-e -r ee

-r le

rr

-r

tr

:rre

%0%D %D %0%D -4 -4 W-r e to to to %0%D

-0 4 q -4 -4 0%0%cyb0%Ch 43N

CU

925 c to e en o

V)

cu
c
=

2,42 22 22 " cycy "Z imcyAAmmmmm

C to : r: tm t2 :Z Co Co Co

CJ C Co C a2 -c

Co

cm9

en Co cm c!) >- >. ce o c m cm C o ce i x die t x 0, r e cx Co C

c' tD LD ce ce :x M CD >- >. Crxr w

'gx ,2

l>II, I-C c en c Co K en > t-1) LD 9.

tr -3 en ix

.r

-1 La Co Co a .. >- > -X' -r -r -r -4

:- d

b-

en
t' t^ VI 'k

c Cel >m >. "

Co -cen c en to cm u i -i i -j
-1 %A

240

c:;

fn

Oi

C
Ln ,

to

C,

co

E %D I41
-4

m V C! C'i C5 .
(A

-T

00 U) Ln " .

C)

C V) 0

!! .
00 s- u CL

;Zcao l
cli %_ C) C> CD co

%_0,a= 0) cn 10

"m

;r

-0 C rC; 31.4m
In 4- "0 co

S_tn

0 cn C, 41) r_ _d

.
4 0:

EE= Ln
0) C4 U! -9 14

! .

-! 'I
C) cli

en C)
H . 0

;;z -
m 0

qCO Ulrl
Lo cr 0

i2 -! -! 'A

9 m -:
0

(L) .4-3 ro

'a =

:3<a 4-J
u>^ (a

^0C W 4A S. -

(L) 4-3 cy) 0 a to

rO C:

flo

CL a)
. C)

cli z r
4

to

ro 4. )

41 a
CU U

a)
.to

0
'a

00
4-)Io u (u

t/l Ln = SCU 4C" 0) '% F>-, r AA S.-0 0)

r*cn V) (A CO
CL)

&W0 Ca.

lw (u Ln tA fo 1 .0 0
(A (1) 0 W

C3 .

a _W

C3

.00 CL
U;

S- 4--j a) r- 4-

2 C: +-,fo *r) ro
> LO cu
10

E ro :3
E-= .
-

a) a

S- S(0 U

I. - cli 0)

cli rl Lr) cc
C> " C-i . in
V)

co cli

14 . Cl

Ol

" Pl .

CD
C

c)

0-0

'A .-0 (L) M (lj 4-

0) r- -(0 ." a) => 40> C WS0 S. -

>

V;

V;

4-J

Mc

IL,

cn

2 . F

C! C Ln In

"! U C)

ll;:

1%

CA

Ir

C)0
V)

r, cli

C) cli it. C
t1l WI

0 S. a--

En S. C3. = Q- eo -0 m=E V) = :3
CX 4A c CA o0a a

V,

tn

SW

a a)

S-

>

t7. -0

4u0 >
C:
co

CU 4I(L) U

CA r- CL x

W (1) (D
E E

mc
C, cli 01W

co
be

Cl!
cli

;z -.0 w 4 t9 C Ol 11

z;; CD
ZE 4c)

0= .6-) I'd (L)

W
+j 4J b

4-

m tn C) " Ln Cj -

0% cli " rl cli .

-4

F_ ko N .

u L) (1) (1) -f-

En .4-3 a a) CL 0 (L) _

a) tCL (L) a) M4

CD

C
41)

(1) to -wa -a " ej

C3

::z
0 CD

Z,
z c C;

CA CU C; C\J CL,-M ko ,-.) -0 to(/) +-) CU (U 1=3

10 0 4V) C: C0 4, fo w Ul) &C: :3 a)

.T

>
ma CU -4o rd Cu
-C F.0 41

";
s-

L 4-

tm

, "4 W %0 -4

CD

in)

C;

(11 +3 ca.
0a -0 a 'a
aa (L)

(A

(U Qj 40 CU

a
.0. 4-3 fo

10

V)
.0 s-

0)
(1) c -

C: 0
rO M

LA
Gi

.0 (1) (1) 4-3 =

#A (A LA

: . 4--1 - pl_
4J L)

cxc ro cm f-= C71 m 0 -6--) It% cm U (L) 0) C: 05

:5 .0 490 ro

Wa

4!
41 S_ 0)

11 In LA_ m
- -, CL. to

v00
a) 'A 0 41

(L) tO S- 4-j 4-.) .0

(A (L) SCL M4

oa

a) o S-r0) "a

cy) = fo C7,

(0 cu Sro

a ro -W 4A

CL c 0X

4) r; - %0
V7 ,A CO IT -ell -4

S_ CL
10 ! . 4-

r-".
0 u

P_

1 0

"'o 0
C;

C, 4-0 V) M(D -4 V)

E 5-1CC) " ca 0%

-0.11=
4). C S- w 11 4.0 sCL0

a) CL

10 CU

a) S_(4 Qj
M

C',

cm

11 C,
c; 4-

Cl. CL E
u

0C0
V 4-

C71uli

0 E=

CL

u Cl)

2:

241

in the exper ! mental specifications and material used All of the experimental values were based on a gauge testing. length of 331 mm while the minimum specified values are calculated So, 5.65 So is length the of where original cross a gauge using 1n the sectional
The presented predicted

length. the of gauge area

mean values in Table from both for the yield stress and ultimate stress

5.9 also the

agree with chemical

the relative analysishave the

member strength and grain size grain the stress.

comparison. size and highest highest

Member types carbon

T5 and T6 which of the six 0.5% proof

smallest exhibit

content

samples also

average yield

stress,

stress

and ultimate

A total of twenty tubular type T5 members Compression behaviour: (Table 5.1) and tWenty-seven solid type T6 members have been tested Figures 5.14 A and B show typical in compression. to failure from T5 T6 type obtained and relationships compressive Both "brittle type" types member exhibited respectively. members load by behaviour buckling characterised rapid shedding after strut load of member type T6 is Although the ultimate buckling. load-strain load times than the greater ultimate of member approximately behaviour of both member types Is very type T5, the post-buckling loads be by the This seen comparing average residual can similar. three 15%, total types, measured both at a strain of which are member of load of the both almost one tenth of the corresponding ultimate members. load Table 5.10 gives the value of elastic modulus, ultimate buckling for fourty-seven flexural the stress members and critical Member types T5 have a slenderness ratio of tested In compression. slenderness ratio of 77.16 calculated using an 52.50and a transition (Table 331.70 N/MM2 5.9). Because tensile stress of yield average slenderness the actual slenderness ratio is less than the transition fail buckling by these should plastic members at a stress ratio, the The mean the to of stress yield material. compressive close for buckling the flexural from the stress members, obtained critical in 5.10, is 319.24 Table N/MM2 with a standard presented results deviation compares favourably with an buckling flexural for these members of stress critical estimated 290.72 N/MM2, calculated using the "Perry" formula represented by (Chapter 4) 4.1 4.9, to with an assumed initial member equations of 27.91 N/mm2. bow of L/1000 and a yield stress of 331.70 N/MM2.
242

This

12.5

10

7.5

2.5

-30 PERCENTAGE Finum

-2.5 STRAIN

-,

5 14A .

LARGE HOLLOW RED

25

20
it: 15

10

-30 PERCENTAGE rTGURE 5-14B

-15 STRAIN

SOLID RED

Compressive Load-Displacement Relationships 5.14A and B. Figures load compressive For lube Types T5 And T6. The Figure shows typical from tube type T5 and bar T6. obtained s displacement relationshi 'brittle type' strut-buckling behaviour Both member types exhibited load buckling. 1 by Al shedding after rapid of the characterised displacement initial tested under control at an were specimens This increased 0.063% 0.63% to minute. rate of per was rate strain 4% in had the strain after approximately occurred minute per specimens. 243

1.

to 00 co qr ko CM Ln 0 M LM 00 00 LO 0 P cn C%J
4) C) en

010

-& "

00 Ln Ln 00 cl %Mq*- q:r %g)

-x

:1
U.

4,9 4 ; ; 4 4,9 ; 4 ; 4 aZ 4 (n C) n C%j MM 0 cn CD CD co cn en rn rl Lr) c4,4,4 c; qr (n csi m C" mm en (n ON n < c4 c,,; c4 ; a; a; c) c3 co "j en c (, c r : c,;, c4 w r =;, ,; c; . ,;, c c4 c;, c; e4 .4: en cn en wmm cw M" en " C-i cn m cn fn C" (n m en C. ) mm In cn en mm cn (n en en en Mm -j C- en cn en Cj C.)

c Ln

4 00 en P r I,, en co cn CO r %D

co Ln

lip

Cj CCom 0, -Cjcj C-To r U-)Como %0cj%n 0 -Co co C50 m co CO C4r Ln co " r co Z r%a -Ir CA cmG LAcp in to 55 G co CO CO bm n" in to ul co Ln m '. gn W, in to %a to *, co A in e4 LA %r mmm 7 co ZZ %o %o wi Ln %r LA %a m r- t, rl 00 co (Ij cv cli cli C%j cv " cli cli - cv C-4 f- r r- co rl co co r, r 0 f, Cj --cw Cj --C- -

ci V!C;c! aI "! rl" 1P "! 111: C 17! P ", li m cn -W n cli 00 co Ln tn C5r rl %0

V! 1 9 9 "! Li 19 19 Ili Li 1 C 't ll 1 47! P 9 9r, 9 C -! ! ! Lf Le .t C4 en 1"00 M L-0 n asm qo f-. mr en c%Ln tA n %0 CO

0c) Cj en ,, C "

-0Z! L, 2:

(Ij

1 .... llC I............ n C) C> C n (a CI C 4r nn

C> C, cli cli cli cli C4 "
cli -

mr q

-W %0 C4 f

crt m

tn

cv F

Ln r%

47% c)

cw P.- qr

P 0i

co

co LM crt 0

Ln C% r

co

C%j cm 47, -4

Lm to

CA r

Lm

cli

nn 04 C, cli cl) .....

n CD

19 8 8V c) CDc, CRc, 2 X ( cli ""cv cv

, I. Ch CO(n CO00 CO d. . n C 47,cn 47,47,cn cn cn n C, C, 'm C. cli cli - cli "

E CU CU (L)

a'm t21-- EZ c LM4ei .0" ' 1-

e tu
Zu C

; >- .-%.
-j2 du =0.

009.e 'J '3,lu*Z tu =j tu

0.0 t'h

'c'0,0,0

11 cu

cli

.;;

cli

"

r -*
cli

r . 3. -It 4v cli

rtV. cli

-T Cli

.r

-&

! ul Lf ! ul ul LI 1L L l ul ! "I "! 1 ll1 llll It ll LI 1

d. ", -gr -** cli Cli ---. r r C4 r 00 co 00 co r, 00 00 co co r r- r, r00 co 00 co r r f- 11

co rl

00 co 00 00 Do co r, P r r, I-

! Le ! ,I w u! Le ! v! f! L, ! Lr !w !I v! r!
00 00 co co r- r co 00 co 00

10 1. DI
0Z

vi

Co>

:Z9.9

-5 .--.
CL

-. , '-.

.Z

c 'D 'D Ln c wv -W %'Dko w co to C7. rm

%a tit b -r -r to %0 %0 co co %0 40 Ln 40 %* to qr to ko P, tn tn

d. C'i P.. to to

r -W to C'i qs.

2-u -Z ZO u
,

.5C

IZ

g2mw UM. 5
e .c, . ',

-4

4 -

I-

-'

'r

4 -4

14 4

1-4 -4

-4

.4 4

.4 4

"

Cy

0, ' g> *Z!! '. :2 >'. 'Z '. Z '2 u >,, ra= o

Ln

W CL V)

Ln Ln LM Ln Ln Ln LnLn Ln %q Ln L" in to to 'D c 'D ka to %a .0 Ln Ln Ln Ln Ln LM

49)c

'D

'D to 'D 'D 00 %0 %*

0 %0 'a

t .mV-

Zur 03.
ll Cli 2w 0)
.2 L, cccC . 1c CIJ 5- W fu CD ww , W 0) Cv w -C 00 uc cc 3r )r Ir 0; co a; -C . w 1. CD

4) 1101 4,4, a, ZZ
=C c0C2. Dc :9

.W

lu , .2 z .2 w w -Z u j cc , ,!!
sZ 0 u . 1; 4;

:f --f 31: 31: ,

; Ig

llci

41 If 1-10 1-1 36 :T 'u ', -1 'a) a; . 1: 1: w' 4; c? .2 .2 2 .2c co 7;

lu 0 2 Rn! o ;; . Dr amCDco
-i ",

CL,

3: 4) 0)

000c

cc
41C am LA

v;

........ I -

5 4 a!

4,

III

2 4; . z-

:; .0www :it "' ct: 2 :; 39 - D.- 2-- ZE "'t 41 f !II , aIca c 3; ); 3;: co CC 4u -C 0* 000c x : 1,4L,
==
Q; s- %- S- wv

:: :::: ; :2jA;
41 4) w--ww
'S

40 -

,.

39 )c 3:

;....... . , 39 1:

3c

.,

co 5 44 0! V! S 11, II-

cc

2-- cl

WZ

1 cl cl .

S. w 9);;;;

-0 -0 :3=--, =Ww
I.

v -

1.. '..

-Z

'..

;; 1% Cli

G) 2

41)

ci

39 X3
0! -4 a!

39 "a 10 10 -a
- s;; ; .; ,;

.0 .0------m
Soooooooooooooooooo

- C, "!

C 17: 0 1 0: 0: 11: 0! 0! . CLCiCLCLCCLCLcicLcicLc"iCCtCt

- - C 17: C%l Ci

CCtctclc MMM 0n0nC. Ono n r4 cm 99 ; 9999999 Cl C C, C. 00n0nqnnnqn 99 99C! 99RC! R C! C! C! R C! C! C! C) n

n,

0. C%j

C,

n0nn

C,

(D

Sj - !no

C'. 0

C4 C

C C', 0,0,0,

al a- cy. (A (n 0, (n V

ch 0

CD 0n0n0

CD n

-------

C5
(U

d) -f :9
:9

ac

co

oc 02

LP

(.0;0CD x ij

.I
4

-j -i -i -i

-i 4 4

- 1!9 to 0 co co cm co cm j CA cl: cc
1

-It co -C cc -C -C

c< cc
rrQ gn

<wc

-i

: : it 1: I. ca 00 03 co ce (m 0 mw000 cc V V) 6 2. LI VI V
* -W
-4 -4

cc <C<

932 -:9 co u

c 313-"n V,
Ln

-C

<

co :9T -C co -C = CC DV) wM 3c V) C.,

tn 4") V) V)

a* c amu L., w A cc ce L') b. V) wl V) Ln It =

CA

cv

244

id bar members, type T6, have a diameter so] mm 64.28 transition and a slenderness ratio of of ratio slenderness 61.90 calculated using the average 0.5% proof stress of 515.11 2With the actual slenderness ratio of this member so close to N/mm The 10.0 flexural the critical buckling ratio, less than the compressive yield stress stress will be significantly flexural The mean critical buckling stress for the of the material. investigation is 312.68 members, obtained from the, experimental 2( 2 25.73 Table 5.10). deviation N/mm N/mm with a standard of its transition slenderness Because the tensile (Figure critical 5.13B) load-strain relationships of member type T6 the theoretical

point, yield shows no definite load for these members is buckling flexural

Et, tangent the corresponding modulus, using its load. is it critical supporting member when by: is then given stress at

calculated to the stress in the The tangent modulus

=2

Et
L2 r)

.............

(5.1)

(Shanley,

1947).

load is unknown, this equation can As the theoretical critical iterative by' the tensile be an process solved using only (Figure for 5.13B). the material relationship stress-strain Substituting 5.1 becomes: at = 0.00238 Et ............. (5.2) for the known slenderness ratio of 64.28 equation

To solve equation 5.2 iteratively a value of the stress at Is to this tangent the modulus corresponding chosen stress and assumed level is calculated from the experimental data, by averaging the from dual the The values of extenometer. obtained readings strain Et into then the values of are substituted corresponding and at is Table 5.11 the 5.2 the equality satisfied. shows until equation iterative sequence using the test data obtained from the solid blue-white-B specimen.

245

Chosen
at(N/mm 2)

Et from test data (N/mm? -) 1.8526xlOs 1.7544105 1.6728xjO5

at from equation 5.2 (N/mm2) 440.94 417.56 398.12

375.80 381.97 399.87

Table Critical

Iterative Procedure Used to Obtain The Theoretical 5.11. Buckling Stress Of MemBer Types-767. Flexural

To solve equation 5.2 (at = 0.00238Et) iteratively values of at are assumed and corresponding values of the tangent modulus Et are tensile from the by stress-strain relationship exhibited obtained The values of at and Et are then substituted into the material. is the 5.2 satisfied. equality until equation

From Table 5.11 at lies between for the the

can be seen that the tangent modulus stress two values of 399.87 and 398.12 N/mm2. Us ing tangent modulus buckling buckling stress load of formula of 399.0 N/MM2 gives the member of a flexural 31337.0

it

a mean value for value mean N. which design for

the

The tangent an initially

modulus perfect this

column begins

gives the to deflect

value must be decreased can be achieved by using the "Perry" (Calladine, for the stress value yield ay and a modified Calladine proposed that the modified value of yield stress, purposes This imperfections. to a tangent

load at axial laterally and to allow for formula 1973). a* which from the

is ME located Et = modulus corresponds An curve. estimate of the member critical stress-strain material load accounting for an initial column curvature buckling can then be in "Perry" formula. the for by substituting a* cy obtained

data obtained from the member tests, the to tangent Et a of corresponding modulus value = a* stress mean ME was found to be 477.63 N/mm2. Using this value to represent In equations 4.1 to 4.9, the average member the yi eld stress member bow of L/1000 was buckling assuming an initial stress, 2 This Nlmm 277.75 be to yields a buck] Ing load of calculated 21814.43 N which will be a lower bound to the mean value of 24557.59 Using the experimental
N obtained from the experimental investigation. 246

Model 1 The first designed permitting to double-layer exhibit space truss model tested to collapse was load-displacement behaviour by a ductile tensile to occur in the lower chord yield

extensive members. All of the bottom chord members in this structure were type T1 members (Table 5.1) and all of the top chord members were type T6 members. The web members, connecting together the top and bottom chords were all type T5 members (Table 5.1) except four diagonal web m' embers in the centre of the structure four corner web members, al I of which were the so] id bar, members. The structure was loaded only at the top central for the and the type T6 node by

the 50 kN actuator working under displacement control. The vertical displacements were measured throughout the tests at 43 nodes and in addition the strain was measured in twenty members. Table 5.2 lists at which the displacements were recorded members which have been strain gauged. Figure 5.15 shows the model structure under test.
imposed central load acting on the point incremented truss from zero, the structure was gradually space model identical deflections behaved elastically exhibiting almost at Linear Behaviour: As the symmetrically deflections truss, theoretical the theoretical with an limit of positioned several their nodes. nodes, theoretical load elastic its Table 5.12 compares positioned the measured space 'was the both symmetrically values 7458 of behaviour. in the

the nodes in the structure the and Table 5.4 lists

supporting,

imposed of

when the structure N corresponding to Table 5.13 gives

and experimental

nodes monitored by transducers, The lower load of 4052 N is elastic

for all of the values of deflection for two values of the imposed load. approximately In the middle of the

load of 7458 N corresponds to the range, while the higher Both Tables 5.12 and 5.13 show all range. end of the linear elastic to be within 11% of their theoretical of the deflections values, but the measured deflections of the boundary with nodes generally and the measured deflections less than their theoretical the slightly nodes central of values. However, the measured deflections at the loaded central node, number greater values 31, go against this general trend, with is the the experimental corresponding probably due to stresses node and 2% than apDroximately greater measurements This apparent discrepancy theoretical values. I smal occurring ional it add than their theoretical

d ispl acements ar is 1ng f rom bear 1ng between the threaded bars and nuts and also the 247

Iz :; uu 4&-

LO 9.0 ` 9 C%j cli

0M

E
9

to
uu ui

41

C)

C7% CD C) r

(A -a a +-) 00au o . .

Lm 4. ) (0 (Ii .

ro a)

(1) a 4-J

4cz

4-) U W 4-, (U 9=

tA cu >> " m cl tA

a E+j (L) S. "0 u (1) 0 ca. 0 CL -0 ( -

-iX
Z,

a) @ a
0 :3

Ln

co

ON

7;
(U cl x LLJ
M

>C

a-

A
#0 41 01 C cli f% cli 'Ir! C5 9 -4 -: r ON to C cli ll 04 ON LC)

uj

77v
0

E
j-

U. U 4:3,0

4) 0

to

LM

Ln

C5 cli

CN

C3

Fr .; .1. u
-0

(A 73 00 Ln M 1)

>

(1) Cm j L) 3: 4- -r4-1

E A
r_ 0 42

4- 1) CD CZ
C-i
cr -4 %0
LC!

E-= r . x MC
c:

(L) _r3 0-w

(0

LO

to -. 1

co m

-4 r C%j

-d' m

'o

4J u

-i m

Cj

I-r!

C
(0 :L 0 u0 rd ,c u

(L).

"

40

41

4C) Q) Cl

c3 .,.

f cli

CC) cli

r-

cli Cl)

cli

, r (2) P-4
LM

(1)

CL

(FJ

uj

E qw
1.4

.0 1-4 m 0%
Ln

s-

Z: 0, b-J 4- M

0 -c -4--1 -b-J

CD

Cl)

CC)

. -j <
Fui cl 0 Lu

m
z'

: ; uu

Cli Cl)

ll C%l

1 -4

C%i

T %- 4-

Q) (1) im

%A

L. (U

-4
Ln cli

en

CK;

C
Ln C5 cli %NJ

cli W
z 0 r

ON m C14

cli

co

r1.4

248

Total

Imposed Load 4052 N Experimental Vertical Displacements(mm) 2.007 l. e64 1.572 1.598 1.335 1.345 1.300 1.313 1.144 1.107 1.113 1.119 0.791 0.798 1.506 1.588 1.274 1.291 1.163 1.152 1.325 1.113 0.731 Node Numbers 31 37 26 25 27 15 38 14 28 4 39 3 17 5 32 20 33 9 22 10 21 16 11

Total

Imposed Load 7458 N Experimental Vertical Displacements(mm) 3.668 3.145 2.994 3.001 2.561 2.527 2.509 2.450 2.225 2.070 2.170 2.062 2.516 1.474 2.938 2.969 2.471 2.412 2.234 2.159 2.559 2.099 1.349

Node Numbers 31 37 26 25 27 15 38 14 28 4 39 3 17 5 32 20 33 9 22 10 21 26 11

Theoretical Vertical Displacements(mm) 1.960 1.746 1.746 1.746 1.430 1.430 1.430 1.430 1.111 1.111 1.111 1.111 0.748 0.748 1.686 1.686 1.336 1.336 1.161 1.161 1.494 1.145 0.679

Theoretical Vertical Displacements(mm) 3.607 3.224 3.214 3.214 2.631 2.631 2.631 2.631 2.044 2.044 2.044 2.044 1.376 1.376 3.104 3.104 2.459 2.459 2.137 2.137 2.750 2.207 1.249

Comparison Of Theoretical Table 5.13. Node-s-7 Model I. Rot"- A 11 I nitored 140-1

And Experimental

Deflections

Ihe lable both gives ind- experimentally 5eoretical displacements for measured vertical by transducers for the two values the monitored nodes of of aII The lower load value of 4052N causes the structure imposed load. to in the middle be approximately of its elastic range of behaviour 7458N load higher to the theoretical the of corresponds end of while elastic range. the linear

The Table shows all of the experimentally measured deflections to be within 11% of their theoretical values but with the measured of the boundary nodes genernally deflections greater than their theoretical values and he measured deflections of the central nodes less than their theoretical values. slightly

249

I!
1) Fci, 0 If) I'-)

,x0

250

hardened steel ball, all of which were used in the loading system node and the recording transducer. positioned between the deflecting the member numbers and also values of the for the model structure at the end of the member stress ratios Hnear elastic range. The stress ratio is a ratio of member stress to yield stress for tension members and a ratio of member stress to flexural buckling stress for compression members. From the critical Figure are the most highly stressed compression members with a stress ratio of 0.13 and and 48 are the most highly stressed tension members 13,18,43 members, with a stress ratio of 1.00. As the structure magnitude of was increased, the the central point load acting on the load-displacement behaviour figure it is evident that members 62,63,78-etc. 5.16 gives

actual by the model space truss followed the very closely exhibited behaviour predicted by the analysis procedure. Figure theoretical between the applied load and both the 5.17 shows the relationship displacements of the top theoretical and experimental vertical The drop in appi led load and central node of the test model. in the experimental results, recovery occurring subsequent elastic I imit of the structure, just beyond the elastic was due to a power to the failure at the hydraulic pump supplying oil occurring actuator.
Theoretical structure from the complete chord of Non-Linear predicted centre of that the Behaviour: yield The theoretical in the tension analysis the of the

members would spread sides until the two bottom yield across of

grid outwards towards had formed symmetrically lines yield The f irst the truss structure. set and 48 (Figure the 5.17)

members to

13,18,43 were members N 7458 on acting of increased, 28,33

members 8,23,38 and 58 at the corresponding At this stage, If

under an Imposed load As the Imposed load was structure. followed by members 3, and 53 yielded loads of 8701 strain hardening N and 9447 N Is ignored in the

respectively. tension

to they are assumed remain continuously and on a members, load plateau, the theoretical stiffness, with zero elastic constant becomes a mechanism and the numerical analysis cannot structure hardening if However, of the tension strain members is continue. taken elastic into consideration, the members possess can proceed. stiffness and the analysis a relatively small The next members to

251

0.10 97

0.13 98 cli CD
CD CD %D c" , C t .

0.13 99
o C t ,

0.10 100 CV CD C
c

0.02 93
C,

0.07 91,
rCD

0.07 85
0)
CD CD U') C; r, 0.09

0.02 96
rl Cl)
G 0.01

V) cs to 0.01

r,

C iz 0.09

89
C, CS %D

so 9 ,0 CD U3 0.07 66
FD CD LO

91. a) 9 r, 0 CD N 0.07 87 C

92

cli

C CD , N
(14

M
M

0.02 85
a

0.02 68 Q 0.10

CD

CD

( 0.13 83

; C

0.10

81

0.13 82

0-91

0-73

-71-; CD c Z

56
A CD Z CD

57
CD 'l n CD

58

59
CD 9 U. ) CDr4

60

-C;

q1 CD

0.00 51

0.52 52

0.77 53
CD l iCo ( m

0.52 -5di
u

0.00 b -
CDCM

CD-

C:)

0.00 116
cr) r n CD

0.28 A7
CD

1.00 A8
CD -

0.28 ti u

0.00

r Co
CD

9 rv
1.00
43
CD r ru

cCo
-

t n
0.28
A 41

cr)

CDr4

r
0. ea
di 5

CD

0.00
tf 1 Cl) 1 G
N l N

0.28
A2

OD r N

N LO cy

(1) r- N

CD

CD

CD

0.00 36
CD 9 w CD

0.52 37
CD 1: c -

0.77 38
CD c 0 CD -

0.52 ju
CD CD c CY

0.00 .4u w CD cv 0.51 BOTTOM GRIT)

0.51 31

0.73 32

0.79 33

0.73 31,

critical model of supported For tension Critical

Ratios And Member Stress Numbers For Model diagram The the Yield. shows member numbers and st both bottom for the top and ratios chord stress members loaded the top 1. The was at model central node and bottom the four the nodes of corner chord. at members Stress (bottom Ratio = yield stress chord): member

ical

axial

stress

For

compression Critical

members Stress Ratio

(top =

chord): member flexural 252

axial buckling

stress stress

20

15

CL -.a--

cs -6-> C) F--

50

100

150

200

in mm. Vertico.L Disptacement

Node Displacement Applied Load Vs Central Total For Figure 5.17. diagram shows both the theoretical Ie and experimental Mode II. loaded top the 1. displacements of central node of model verticaTdisplacement imposed a vertical on node 31 of The actuator first 26min displacement for the of after which the 'J. 1m, minute per n ten fold to 1.0mm per minute. 'When deflection rate was increased displacement of the actuator the total was approximately vertical increased further displacement to 2. Omm per 100m, the rate was n in load drop The applied and subsequent elastic recovery minute. just beyond limit in the the results, experimental elastic occurring due failure to the a power was occurring at the structure, of The model structure oil to the actuator. hydraulic pump supplying truss a gin-jointed soace as and the theoretical analysed was form of the Dual Load displacements using a modified were obtained in for changes model geometry. accounting method load-displacement Points on the theoretical curve. in the-elastic A= Mid-point response. limit. Yield of members 13,18,43,43. B= Elastic C= Yield of members 8,23,38,53. D= Yield of members 3,28,33,58. E= Yield of members 2,4,57,59, etc. F= Yield of members 7,9,52,54, etc. for members 7,9,52,54, G= Change in stiffness etc. H= Buckling of compression members 70,71,90,91.
253

yield

are members 2,4,57,59 etc. and then members 7,9,52,54 Yielding etc. at loads of 11435.6 N and 12430 N respectively. continues in the bottom chord members until the imposed load reaches a value of 13325 N when the compression members 70,71,90 and 91 become unstable and buckle. This last value of should theoretically load corresponding to the buckling of the first compression members

was very dependent on the value of post-yield stiffness given to the last group of tension members to yield. At this point in the of the entire structure was very small in analysis, the stiffness stiffness comparison with its original and the force residuals analysis computer program at first generated in the non-linear tended to decrease and then converged towards a minimum, but later if the number of analysis and diverged cycles was Due to this instability increased. in the analysis numerical from the small stiffness this of the structure, program, resulting increased value, taken when the force minimum, should be accepted with caution. last theoretical
Experimental the structure Non-Linear Behaviour: As the

residuals

were at a

imposed load

acting

on

was increased, yield spread through the bottom chord but the structure did not begin to fall by the members as predicted, buckling by the of one of the compression members as indicated theoretical bottom chord analysis. tension Instead, the first ember to 5.18). fall was the member 43 (Point A, Figure This member an imposed load of

was supporting ruptured when the model structure Failure 12699.13 N. by the tensile of member 43 was followed failure of member 38 and then the failure of tension member 33 (Points B and C respectively, Figure 5.18). As the bottom chord of began to open up due to the three failed tension structure in the stress member 58 reversed and this member failed in members, (Point D, Figure 5.18). At this point, the structure compression imposed load 3410.73 of supporting an N, of was capable the approximately 27% of the 5.18) short time after (Point E, Figure structure 5.18). the structure

by the structure. A maximum load carried in tension member 58 had buckled, member 48 failed and the to imposed 2156.35 the load N decreased by the supported (Point F, figure

immediately After

member 48 had failed, slowly began to recover

imposed load supported by (Points F to G, Figure 5.18).

The top chord of the model structure had deflected significantly and the additional the structure load as a catenary. was supporting The testing of the structure was halted when the displacement capacity 254

LOAD APPLIED CURVES TOTAL DISPLACEMENT vs NODE OBTAINED FROM 1 MODEL

28

SULT

15

:Z

C CD 1--

so

lea

158

283

Verticat DispLacement in mm.

Load-Displacement Experimental Behaviour Obtained From Fiqure 5.18. Figure load-displacemenE shows the experimental Moaei I-TK-efrom Model 1. The structure behaviour obtained was tested under imposed vertical displacement displacement control with an initial (A 0.1mm, The 31 to E) shown on the per minute. points of on node Figure, of individual correspond with the failure members leading to of the structure. the complete collapse

Point A B C D E

Tensile failure of Tensile failure of Tensile failure of Compression failure Tension failure of

member43 member 38 member 33 of member 58 member48 the to

Points F to G show a gradual recovery in the stiffness of displacements allow the structure structure as excessive vertical behave as a catenary.

255

limit its deflection the total the and at vertical was actuator of 200 At this loaded the mm. was stage the structure node of 3697.44 load N. imposed of supported an and C show how the theoretical and in 18,23 for the the tension strain and members values experimental The Figures show 3 varied with an increase in the external load. from both of the strain the recorded strain gauges which were Figures 5.19 A, B diametrically opposite. each other on each and centrally positioned member. Examination of the numerical results of the recorded strain in these members followed the proposed that indicate yield first, followed by theoretical sequence with member 18 yielding in is Figure 5.19 it From 3. the 23 shown plots and members in bending these that effects are negligible members when apparent Unfortunately, the members in the they have yielded in tension. fourth and fifth groups to yield were not strain-gauged so it was if determine behaviour their they to or yielded monitor not possible in the predicted theoretical sequence. Figures 5.20 A, B and C show how both the theoretical and values of axial stress in compression members 98,94

experimental increasing load. Is It go external applied with change and to note that as the external imposed load increased and interesting flow, the plastic the bottom chord members yielded and exhibited force in the most heavily stressed compression members Increased at increase than the In the of rate greater external much a rate for Figure 5.20 A 5% increase in that load. shows a applied imposed load there was a 14% theoretical and 12% the external increase in the axial force

by carried compression experimental C behaviour 5.20 for 90, Figure 98. shows similar a member member in load first the increase 5% external after yield caused a where a increase in the axial 16% theoretical and 15% experimental by However, the these member. carried sharp compression increases in axial force, corresponding to small increments In the imposed load exhibited by members 98 and 90, were not shown In the behaviour of compression member number 94. Figure 5.20 B shows that force the axial force in member 94 decreases significantly, with increases three sets of tension members in the imposed load after the first have yielded, forming the two complete yield lines across the bottom chord of the structure.

256

LOAD APPLIED STRAIN CWVES-' TOTAL vs MEMBER OBTAINED FROM 1 MODEL

15

(Positive Strains are TensiLe)

..............

Cl_

NO.18 MEMBER RESULTS EXPERIMENTAL No.21 Strain Gauge No.22 Strain Gauge -------RESULT ANALYTICAL

F'TQURE 5-19A

23 Percentage Strain

9 C:

MEMBER NO.23 EXPERIMENTAL RESULTS Strain Gauge No. 19 Strain Gauge No. 20 ANALYTICAL RESULT

FIGURE

5.19B

23

Percentage Strain

v-.

... 0

k21

,n

MEMBER NO.3 EXPERIMENTAL RESULTS No.27 Strain Gauge No.28 Strain Gauge ANALTTICAL RESULT

FIGURE

5.19C

12343

Strain Percentage

Applied Load Vs Fiqure 5.19A,'B and C. Experimental And Theoretical Member strain Curves Ubtalned For Model 1. lhe I-igures show how-77-Fe in the tension theoretical and experimental values of the strain 23 increase 3,18, in the external applied and vary with an members load. The experimental values were obtained from each of two strain diametrically opposite each other at the centre of placed gauges each member.
257

(Positive Stressesare Tensite)

28

15
8 o le

-10

-28

-36

-10 -58 -08 Axiat Stress (N/sq. mm)

-78

-60

-lK

FIGURE

5.20A (Positive Stresses are TensiLe)

201 91 MEMBER NO. RESULT EXPERIMENTAL RESULT ANALYTICAL 0-----

15

r
-0

58 FIGURE

is

30

20

is

-10

-20

-30

-40

-58

5-20B

Axial Stress (N/sq.mm)

(Positive Stresses are Tensite)

20

15

-c
-ti la

e: .

-10

. 20

-3e

-io

-58

-0

-78

-0

-106

AxidtStress (N/sq. ra) FTGURE 5.20C

And Theoretical Applied Load Vs d C. Experimental Model rurves 1. T For igures show how botF Member Axial Stress in theoretical s of the compression va stress and experimental increasing load. 98,94 90 applied with external and change members for by the The each stress member averaging was obtained axial two diametrically from strain gauges obtained placed readings strain of each member. each other at the centre opposite 258

B and C give the experimental load-strain obtained from each of the two strain gauges fixed at relationships the centre, diametrically opposite each other, on the three top chord compression members 98,94 and 90. It can be seen from these Figures 5.21 A, in the structure was accompanied by bending in these compression members. The maximumstrain recorded frorn strain on member 90 was almost 0.186% gauge number 5 positioned corresponding to a stress of 379 N/MM2. This st ress level is below ' (515.11 the mean value for the 0.5% proof stress of the material that these members were capable of N/mm2 Table 5.10), indicating figures that yielding supporting failure. Model 2
The second double-layer also was load-displacement designed to space exhibit truss a model tested to collapse

a small additional

axial

load and bending moment before

ductile

post-elastic

The second model structure response. was almost bracing to the first identical model except that the four diagonal (Table T5 the type in the of structure centre were members members type T6 members used in the first instead of the solid 5.1) model structure. at four top chord The second model space truss nodes, numbers 19,21,41 was loaded and 43. simultaneously The four point

imposed load was applied through the simply supported beam system, kN 5.8, 50 the Figure in using actuator working under shown The vertical displacements displacement control. were measured at 43 nodes (Table 5.2) Figure (Table 5.4). Behaviour: under and the strain was measured in eighteen members 5.22 shows the model structure In order load, to the determine if the under test. structure behaved several were and

Linear

symmetrically nodes

positioned symmetrically both 5.14 Table gives measured. for displacements the measured theoretically structure is 2000 N. of each relatively exhibit loaded identical by four

displacements vertical of in the model structure the theoretical groups vertical

displacements

nodes which of should displacements when the positioned point loads

symmetrically

obtained between the measured displacement 13.5% occurring of nodes 17 and Smaller differences 5. obtained from similar nodes in other groups in general, from that, a symmetrical indicate respon. se, was obtained 259

The measured displacements given in Table 5.14 show In the experimental occurring small differences values from nodes in the same group. The largest difference of

20

WEERNO.98 EXPERIMENTAL RESLLTS

28

--------

StrainGouge No.9 StrainGouge No.10

2 0 -i -0
-2: --Z

1_f__f

la .

II

"

is

"5

-3

-2

.11

23

Percentage Strain(ICE-01)

FIGURE

5.21A

20

mom NO.91 EXPERIML RESLLTS

28

StrainGauge No.7 StrainGauge No.8

le

, 3.

-2

.1123

FTGURE

5.21B

Percentage StrainME-el)

STRAIN TOTAL LOAD CURVES APPLIED vs MEMBER OBTAINED FROn I MODEL

20 29

MEMBER NO.90 EXPERIMENTAL RESLUS

15 RI -0

..... ..

StrainGauge No.5 StrainGauge No.6

Is

Iq

-3

-2

-t

PT(; uRE

5.21C

StminOKAD Pwcentage

B And C. 5.21A, Experimental Figure Applied Load VS Member Strain Rro-m-7odel 1. The Figure Curves Obtained Shows how the strain 7 Tr 98, members and 90 varies compression with an increasing load. The experimental applied externally values from were obtained diametrically gauges placed each of the two strain opposite each The Figure at the centre of each member. other shows that yielding in the strqctur. e was accompanied by bending- in these compression Tensile strains are shown as Positive. members . 260

mst

-71 -

Figure

5.22

Model

2 Under Te-, t.

261

IF E c 0
M 4 uu (U .AJ LO

-cr
Lr! cli

LO

S- 4cu (U

cn
Ci uu .-W 41 S.. 4cu (V en cli V)

C)
cli

m 4-3 00CU

tA 4j

-VW to +j

*r-. 0 4-3 U

(D. 1- m E=- -) a0

>> 4- -o (L) 0) 0. Ln to .0 -i-.) to -

CL CL V,

) .1-.

j zc

w -0
0

=3 -R

C3,

en

tn

C (31

LA

.0

E
LLJ LLJ
C%j Cj CIJ m

ro u

to (:)

Scu

aj (L)
0) > =3 (1) 4-J '&j -0 UU o :3=

cli
cli

f-I

CYN

t-

CL X. LLJ C%J -a
C) cli -4

cn cli

QJ E-= S. &- 4-3 s1.., tA C urE

u
d) -0 -R :3 0

-4--)

(1)

Ln
4

-tt

cli

%0

tn

4D
cli

CN

C5
-4

4 40 W

C0 'A I. 0) > c= fT5

4-J

>,-C3 0 C:
-0

48a

UW (L)
C) S. - C) C) o

aEr

E A
c 0 .,o 41

(L) 04so

c.
co
cl; CD Kr r-: r

00

Ul) co C

LO r LO

(D -4 C

to r, cl ;

r, rcn C

0 010 Cl. a
CL

E=-4-- to (02. 0 Q)
to

fo co C',
10 u to .,4-3
0

S- 44)

Ez 00 C-)
u U.,

u
(1) A-0LA E

(0

r0) -0 :3

co C%j

r-4

cli m

4" (Y)

cli C%j

W 41

Cn. 4-J

u u fo

E0 Lr;
L/) 4-1

E., (L)
(0 40 '0 OJ (u

F. L. -IJ 4-

LLJ
>. uu S.- 4w (1) m 0 C> cli

c Ln U') -4
Lf!

r-

-: r

Cj cli

C"! cli

Oi C-i

ll

C14

cli

C C%i

ui LU

E (L) u 0 (U C.; en (71 Lr) cli CL C3 it

LC)
CS

tD
rz

Ln r.:

cli 4

C>

(n 4

C)

Wl cli

262

the

model, provided the

that

the

imposed within

load the and

acting elastic

on the range.

structure Table

maintained 5.15

node deflections the theoretical

both gives for all deflection

of the nodes monitored displacements given in Table 5.15 correspond to two separate values load. The first load imposed total the value of 8000 N of displacements theoretical which are still within with corresponds the elastic while the second load value response of the structure,

experimental by transducers.

of values The node

end of the linear-elastic of 10240 N corresponds to the theoretical A comparison of the theoretical behaviour of the structure. and in Table 5.15 displacements the given show all of measured obtained when the structure values, was supporting a experimental to 18% of their be within 8000 N, theoretical load total of values. large discrepancy between the theoretical However, this occurring and experimental lower chord on other displacements theoretical the displacements (3,4 was only apparent and 5), which were in three of to the each their

edge nodes,

adjacent 9% of the

south side of the model structure. nodes were within of the remaining with the measured deflections of 4% greater than their theoretical of values. the central

The measured boundary and 5% less

values,

nodes approximately the measured deflections than their theoretical

values,

nodes approximately

The node displacements, applied a total supporting of the I ! near elastic theoretical

measured when the model structure was load of 10240 N corresponding to the end than their range, aII were greater

displacements The values. measured of the corresponding (3,4 25% lower edge nodes and 5) were approximately three chord theoretical their the than values, while measured greater 16% greater As all of the measured displacements values. than their theoretical it is theoretical the than that the values evident greater were it the limit structure exceeded was of when was supporting elastic deflections of the remaining nodes were, on average, the total imposed load of 10240 N.

Figure 5.23 gives both the member numbers and the values of the for the the structure model at ratios end of the stress member It can be seen from the Figure that members linear elastic range. 3,28,33 a stress just sufficient and 58 are the most heavily stressed tension members with that the force in the member Is ratio of 1.02, indicating to

However, the cause the member to yield. in the. In model structure members are understressed compression .
263

Total

Imposed Load 10240 N Experimental Vertical Displacements(mm) 9.803 4.441 4.265 4.342 3.730 3.875 3.795 4.048 3.613 3.836 3.701 3.956 2.291 2.531 3.839 4.421 3.879 4.177 3.514 3.501 7.140 3.162 2.086

Total

Imposed Load 8000 N Experimental Vertical Displacements(mm) 7.674 2.824 2.732 2.739 2.408 2.410 2.394 2.545 2.478 2.623 2.404 2.651 1.575 1.797 2.410 2.837 2.476 2.703 2.377 2.311 5.252 2.032 1.411

Node Numbers S 31 37 26 25 27 15 38 14 28 4 39 3 17 5 32 20 33 9 22 10 21 16 11

Theoretical Vertical Displacements(mm) 3.780 3.660 3.660 3.660 3.333 3.333 3.333 3.333 2.860 2.860 2.860 2.860 1.893 1.893 3.646 3.646 3.231 3.231 2.848 2.848 3.518 2.777 1.689

Node Mrnbers 31 37 26 25 27 15 38 14 28 4 39 3 17 5 ss 32 20 33 9 22 10 21 16 11

-Theoretical Vertical Displacements(mm) 2.949 2.855 2.855 2.855 2.601 2.601 2.601 2.601 2.232 2.232 2.232 2.232 1.477 1.477 2.844 2.844 2.521 2.521 2.222 2.222 2.745 2.167 1.317

Key:

Into Displacement Measured By An LVDT Transdu&er Incorporated The Actuator. Of Displacement Readings Not Accurate Due To The Interference The Cross-Beam (Transducers Not Attached To The Nodes But Attached To The Cross-Beam).

Of Theoretical And Experimental Deflections Comparison 5.15. Table The Table both Nodes in Model 2. 'For---KTT- Monitored gives experimentally displacements for measuFed vertical theoretical and for by transducers two the the monitored values of nodes of all load 800ON corresponds The first load. value of imposed with the elastic are still response within of the displacements which load 10240N the to the of second value corresponds while structure behaviour limit of the structure. of the elastic theoretical

The Table shows all of the experimental values obtained when the total load 800ON 18% to be a of supporting was of within structure In addition the measured deflections theoretical values. their of than their theoretical greater the boundary nodes were generally deflections the the of measured and central nodes were values theoretical less their than values. slightly
264

S112
97
0 CD0

0-1
98

rl

A-1

rz rl'

V2

99
CDr-, (Z N

100
CD 0

0.02 4f 93 cs
CD

0.05

0.05 85
C; D

-9

0.02 96
7)

CD

63 a (S im
(D (o

0.02 90 r9 c
CD N

0.02 91
C CD N

0.01 92
:0 CD N

0 a
70

0.02 85 4 rIli

0.03 86
C) V, tD

o. es G7 -

r%,

-e

0.02 68
(N cz N.

l tr)
CD Lo

: f% CZ ,1
0.16
U15

L-

0.12
81

0. IS

0.12

ueA- 17 1 2 -n,

91-70

A- q7

56

57

58

59

60 Cs t,

0.00 51 C:

0.70 52

0.69 53

0.70 5

0.03 55 C7)
V

0.00 6
Cs q) ..

0. "11 "17
0 m CD

0.38 6 lei

0.411
418

0.00
00

M c

0) C7,'m .

N -

I-,co CD
0.00 111 0.41
412

-'o
0.38
413

(i
0.41
144

9 n 14

O. ea
A15
0 r -4 CD"y

-rC)

oi cs

0.00 36
C CD D

0.70 37

2.69 T8
(S) 9

0.70 -- -Tg

0. e.0 Ilu

BOTTOM

GRTD

CD

CD

0.70

0.97
134

1.02
113

0.97
0 4 ja

01

Stress Ratios And Member Numbers For Model 2 Critical Fiqure 5.23. The diagrams show the member numbers and critical At Firs t Yield. the both top bottom 2. for and chord members of model ratios stress at four top chord nodes and was The model was loaded symmetrically four the bottom Loaded the corner of nodes chord. at supported * by shown nodes For tension Critical (bottom chord): members member stress axial Stress Ratio -yield stress

For compression members (top chord): member axial stress Stress Ratio Critical flexural buckling stress 265

etc. comparison with the tension members, with the members 62,63 being the most heavily stressed compression members with a stress Other compression members are also stressed at a ratio of 0.16. four diagonal level bracing members having a the corner with similar stress ratio Theoretical analysis of of 0.15. Non-Linear the non-linear pin-jointed indicated that under an increasing structure, in the tension members would point load, yield Behaviour: The

symmetrical four spread from the outside bottom chord edge members towards the centre The first set of members to yield were members 3, of the structure. (Figure 5.23) 58 28,33 which yielded when the structure was and load imposed 10240 N. As the imposed load was total of a supporting etc. became critical, yielding when load of 10800 N. The next group of 8,38,23 53, to members and which yielded when were yield members load of 12240 N, followed by the structure was supporting a total etc. which became critical when the imposed load was increased to 12480 N. When the imposed load reached 14648 etc. yielded, followed by the four N, tension members 1,5,31,35 13,18,43 tension bottom chord members, and 48 which central load imposed increased further by the 56 N. was a yielded when members 7,9,37,39 of the four central bottom chord tension members was almost Immediately followed by yielding of the adjacent members 12,14,47 imposed load the had When 1504 37 N the reached a value of etc. and four corner diagonal bracing members 103,120,197 and 182 had also in At this the in tension. tension analysis stage no other yielded However, the the became analysis critical. continued with members Yield behaviour Ideallsed each group of members conforming to their When the Imposed load acting on the load-displacement relationship. of increased, members 2,4,32,34 the structure supported a total

17596 N, the post strain-hardening axial stiffness reached structure insufficient tension lower to maintain the edge members was chord of the theoretical the procedure, and numerical analysis of stability of the model structure Experimental was unable to continue. Behaviour: Figure 5.24 shows both the from testing to

Non-Linear

experimental truss, the space model second collapse

load-displacement

behaviour,

load-displacement relationship it can be seen from the Figure that there the theoretical of behaviour and experimental (A to B, Figure 5.24)

resulting together with the theoretical obtained from the analysis program.

is close agreement between displacements over the linear range and even over the early portion

266

20

LM

15

0 0 -J

-a

NODE NO.25 EXPERIMENTAL RESULT RESULT --*--ANALYTICAL

-J -I-) 0 I-

A 50 103 150 200

in mm. VerticaL DispLacement

And Theoretical Experimental Load-Displacement 5.24. Figure The Figure shows both tHe experimentaT-anU Behaviour Of Model 2. load-displacement behaviour of model 2. The structure -i-heoretical(19,21,24 four top loaded chord simultaneously at nodes and was displacement 43) by imposing a vertical of O. 1mm per minute on each This displacement rate was continued until of the four nodes. displacement 25mm of vertical has occurred, whereupon approximately the displacement rate was increased ten fold to 1.0mm per minute. displacement of the actuator of approximately After a total vertical further to 2. Omm per 100mm the displacement rate was increased minute. load-displacement Points shown on the theoretical curve. A-B = Elastic response. B= Yield of members 3,28,33,58. C= Yield of members 2,4,32,34, etc. 0= Yield of members 8,38,23,53. E= Yield of members 7,9,37,39, etc. F= Yield of members 1,5,31,35, etc. G= Yield of members 13,18,43,48,12,14,47,37, etc. H= Yield of members 103,120,197,182. I-L = Change in axial stiffness of yielding members. M= Loss of numerical stability. load-displacement Points shown on the experimental curve. limit. B= Elastic G= Yield of members 12,14,47,37, etc. N= Buckling of corner compression member 118. P After failure of member 118, bottom chord members 5 and 56 come into contact with the support. due to members 5 and 56 supporting Increase in stiffness P-Q the force in bending. corner reaction in bending. Q Members 5 and 56 fail R The "knee" of the buckled corner compression member 118 came into contact with the corner support. in the model, the structure S Due to excessive deformation slipped off the corner Support and the test was halted.

267

(B G, ! Figure 5.24). to the Point G (Figure non-I near range of 5.24) is the theoretical point at which the penultimate group of ) (12,14,47,37 etc. yield, and corresponds with the onset members flow occuring in the model structure. The of extensive plastic non-linear range of behaviour G to N (Figure 5.24) is associated with the strain hardening of the yielded tension members. Point N signifies buckling of the model structure which wai caused by of the corner diagonal compression member (number 118, Figure 5.1) when the model structure was supporting a total imposed load of 18598 N. The failure the failure

of the corner compression member led to the rapid loss of load-carrying capacity of the model structure with the imposed load decreasing to 3739 N (Point P, Figure 5.24). Af ter the failed member had deformed the two corner bottom chord tension (numbers 56, Figure 5 5.23) came into contact with the and members base of the corner support of the model and the structure (P to Q, Figure 5.24). When the temporarily regained stiffness imposed load had increased to 5179 N, bending of the two corner tension members occurred and the imposed load decreased to 3635 N (Q Further deformation of the failed to R, Figure 5.24). member 'kneel the in the the The of member sitting on support. resulted immediately increased and the model the structure of stiffness imposed load of 19171 N (R to S, Figure finally supported a total Excessive deformation of the structure 5.24). In one resulted corner terminating of its slipping off support, effectively (Point S, Figure 5.24). the experimental investigation the model 5.25 A to

G show how both the theoretical and In the tension of strain values vary members, as the perimental ex ' imposed load acting on the model structure The was increased. in the Figures 5.25 A to G were obtained plotted strains recorded Figures from both of the strain gauges fixed to each member, apart from the (Figure 5.25A), for 3 member where one of the gauges strain recorded during the member early part of the test. separated Examination of the Figures 5.25 A to G, and the numerical results of indicate that the tension members yielded in the recorded strain, by the theoretical the sequence predicted exact analysis. almost However, a comparison of Figures 5.25 A and 5.25 B which give the for load-percentage strain relationships members 3 and 2, from the that show respectively, indicated as vice versa member 2 yielded by the theoretical
268

before

member 3 and not Figure 5.23 analysis.

25
....

2B

3 NO. IIEMBR RESULTS DMIMENTAL No.8 StrainGauge No.8 StrainGauge RESULT 0 ANALTTICAL ....

. .............................
15 p ...

.....................

FIGURE

5.2-5A

234 Percentage Strain

25

HEW NO.2 RESULTS DMIrIENTAL No. 17 Strain Gauge No. 18 Strain Gauge RESULT .... 0.... MALTTICAL

28

R................
15

0 W............. ......
FIGURE r 5.25R

-0

.............. iI

2345 Percentage Strain

25

20
...... 15 ...... . .... T ........ ......

.... ........................

OBER NO.8 EXPERIMENTAL RESULTS Strain Gauge No. 37 Strain Gauge No. 38 ANALYTICAL RESULT

FIGURE

5-9-5jr-

2345

Percentage Strain
25 HM NO.9 DMIWAL RESUL7S Strain Gouge No. 39 Strain Gauge No. M RESLLT .... 0... ANUTICAL A * ............

28

R.
-0 Is ........... .......... ...........

. *. ..................

FIGURE

5-25T)

234 Percentage Strain

269

(Positive Strains are TensiLe)

25 1 mm NO. RESULTS EVERIWAL No.31 StrainGauge No.32 StrainGauge
RESULT .... 0.... MALYTICAL

20

R
$ -J
Oj

13 le

2,34

Strain Percentage (Positive Strains are TensiLe) 25 13 MW NO. E)MML RESLETS StrainGauge No.11 StrainGauge No.12
RESLLT .... .... ANLYTICAL ....................

28

R
-0

FIGURF

5-25F

234 Percentage Strain

(Positive 23

Strains are Tensile) MEHK-R NO.12 EXPERIWAL RESLLTS Strain Gouge No. 19 Strain Gauge No. 28 RESLLT .... 0.... ANALTTICAL ..........

20

R
13 9 -0 .................

..............

2: .

23 Percentage Strain

Experimental Applied Load Vs 5.25A to G. and Theoretical Fiqure -TT7e Model 2. Ci Ine Igures how es For Member Strain show in tension values of the strain theoretical and experimental members in the external and 12, vary with 3,2,8,9,1,13 an increase load. The experimental from each of values were obtained applied diametrically the two strain placed opposite gauges each other at for member the centre of. each member, apart from the strain recorded 5.25A) from the member 3 (Figure where one of the gauges separated the early during part of the test.

270

gives the member stress ratios for members 2 and 3 at the end of the I Mear elastic The small range, as 0.97 and 1.02 respectively. difference between these two values indicates that either member Is to yield. From a comparison of the likely to be the first strain relationships given in Figures 5.25 C to G, it can be seen that the initial yielding of members 2 and 3 was followed by the yielding of members 8,9,1,13 and 12, load-percentage corresponding with the sequence outlined of the model structure. by the theoretical analysis

Figures 5.25 F and G also show that a small amount of bending In was induced in both members 13 and 12 before they yielded As yielding of these two members progressed, the influence tension. of bending moments acting on the members decreased. and in of axial stress compression members 61,62 values experimental increasing load. 101 The Figures external with applied change and 5.26 A and B show that as soon as tensile yield has occurred in the fixed increase in the imposed load a external structure, model in the In resulted a larger increase occurring structure acting on A 47% the cornpression force carried by both members 61 and 62. load carried by the model structure increase in the external 82% theoretical 68% in and a experimental Increase In the a resulted axial force carried by compression response was even more pronounced 47% increase in the load, 62, a external after first where member 99% in theoretical 96% a and a experimental Increase resulted yield, in the axial compression force carried by the member. 5.26 C shows that the theoretical axial stress levels diagonal in the corner compression member are In close occurring The axial stress agreement with the experimentally measured values. from averaging the measured strains recorded obtained were readings Figure from both of the strain gauges fixed to the member. Figures 5.27 A, load-strain B and C give the experimental relationships obtained from each of the two strain gauges fixed at the centre, diametrically opposite each other, on the three compression members 61,62 and 101. It can be seen from these Figures that a relatively in bending these members while the model present of was amount small As was responding elastically. yielding structure bottom chord tension members the bending strains
271

Figures 5.26 A, B and C indicate

how both the theoretical

Th is non- I ! near member 61. in the behaviour exhibited by

occurred recorded

in the In the

(Positive Stressesare Tensile)

25 1

3
1 's, 13 e: . ,EL Q

61 OBERNO. DMIENTALRESULT RESULT ANALYTICAL

toI 'a
I-

-206

-ice

-128

-00

-40

Is

Be

120

Ice

no

FIGURE

5.26A

Axid Stmss (N/sq.m)

(Positive Stresses are Tensile)

201 MMERNO.62 RESLLT EXPERIMWAL RESLLT AMYTICAL 0-----

9 15 -0

"1

-4

8:

0 I-

-280

-168

-120

-ft

-is

Is

FIGURE

5.26B

AxialSUvss (N/sq. mm)

(Positive Stresses are Tensile)
20,

as

128

IN

200

mm NO. 101 EXPERRENTAL RESILT ANALYTICAL RESLLT

is .0

-208

-168

-120

-06

-40

48

128

ice

200

FIGURE

5.26C

Axid Stress (N/sq. m)

Theoretical C. And And Experimental Z6A, 13 Applied Load EigLUyC_5, 2. The Figure Curves Stress For-Model Axial how shows theoretical in and experimental the values of stress both 62 61,, 101 increasing and change members with compression load. The axial for stress was obtained applied each external the from by readings two strain averaging obtained strain member diametrically opposite the each other placed at centre of gauges member. each 272

(Positive Strains are TensiLe)

23

61 tMR NO. 23 WRIMNTAL RESLETS Strain Gauge 25 No. Strain Gauge No. 28
26

20

,a 9

13

..

'.

,.

15

-3

-2

.1122

Percentage Strain(1 "D FIGURE 5.27A

(Positive Strains ore Tensile) 23. 20 .

HM NO. G2 EMIWAL RESLLTS 25 StrainGauge No.35

Strain Gauge No. 3G

21

15

4: is

-1

123

Percentage Strain(I "D

FIGURE

5.27B

(Positive Strains are Tensite)

25

NO. 101 . EVERUMAL RESIVS Strain Gauge No. 29 Strain Gauge No. 38
.

rm

25

20

21

-3

-2

.1123 Percentaqe Strain ( ICE-01)

5.27C FIGURE Applied Load Vs Experimental de-T Z. The Figure shows how Lurves Obtained Fro 101 61, and varies members with compression load. The experimental values were applied externally diametrically two the strain placed gauges of each From the Figure the centre member. of at each other in the bottom the tensile that chord yielding of by bending in these compression members. accompanied 273

Member Strain 7n the strain an increasing from obtained opposite each it is evident structure was

increased. The strain compression members significantly recorded from gauge 30 positioned on the corner compression member 101 indicated that when the adjacent corner compression member (118) three causing collapse of the entire model structue, yield had 101. in The recorded strain of 0.169% measured member occurred not in member 101 corresponds to a stress of 347 N/MM2 indicating that , load and the member was capable of supporting additional axial bending moment before failure. buckled, Model 3 space truss model tested to failue, incorporated both tension members that were permitted to deform and eight soft members, which were also allowed to plastically All of the bottom deform plastically when stressed in compression. The third in 3 bracing members model were nominally identical to web and chord However, the the members used in the second model structure. majority of top chord compression members were the tubular type T5 (Table 5.1) instead of the solid bar members used in both models 1 and 2. The eight soft members used in the third structure were all The soft members were placed type 1 soft members (Figure 4.11). in the top chord of the structure symmetrically (Figure four the 5.5). sides of each of centre in pairs at the This position was the most heavily and theoretically the top double-layer

chosen so that the soft members would replace in the structure members compression stressed forces favourable to occur within redistribution of a permit chord compression members.
model structure was simultaneously (19,21,41 43) in identical and an chord nodes 2. The displacements for vertical model adopted twenty-three (Table nodes 5.2) using The third

loaded at four procedure to

top that at

were measured

only the L. V. D. T. transducers. the strain In addition, was measured in fifteen members (Table 5.4) for the bottom two each six gauges of strain chord tension using for the three each of nine compression members. gauges and members, Figure 5.28 shows the model structure under test.

An assessment of the Hnear behaviour of the increasing imposed load, an supporting can be made by structure and experimentally measured vertical comparing both the theoretical in the model deflections of several nodes, symmetrically positioned Table 5.16 both the theoretical gives and experimental structure. Linear Behaviour: displacements of groups of nodes symmetrically
274

positioned

in the

a
275

E c LM C%i ? 00
0! 4, w

S- 4-

Q)

-0 0 Z,

9 4 :3

c ? - M 4,

00 4 C9

Ln 0% C

Ln m C

C: t- 4(A = c +J 000m 040 .Im %A 4.J (L)

(L) 0

V)

41) (V
4(1) cz

-4-

ui

cm

> Emu (1) >

u ex3 -ci

fo

:3 4j

-j

-C

0) 42 a

co (1)

C7% m

Lr) cli

(L)

s- m -, ; m CL4.1CD 4. J m m (A . (0 ) 25 4.. - it
c: h0 m

W CD Z>

=U

:3 0 -r-

ei r=

ai

JD g4- w =4fo

cc
LO 0-

E
c2 -4--, M (V 400 2: tA 0) lu 4A cm to fo

LAJ

1.

2
-4

cli

-It

()%

C7%

LU '11 c: <0 fe

CU 4s-CD
i W Rl 0

C%i

C-i

C%j C%j

M cu 0u tn 4..) - 41 c2. (A r= fzj

Lr)

cli

cli

C) -4

%A >

(U

,0
-

4tj. %A4-

E
Ln (7%

CD
C cli Ln rP, CD cli to

0
c2. W00 xCC-Z3 ." E-

z
0,4j

4j I-, W

ll -4

cli

9
-4

ll

(%j

cli

0 CL 0 u

(U

rC%i

co cli

rCl)

en C%j

to

(D

) 41 (D 4..
r= um

Lei

2 o 4-A
S. - 4Q) C:) ui

cli

ON LO
LC) r

qr

C-i

V! eq

Cl!

m
vi

U ")

1 CQ

to

qcr

m
-4

.0 4 d2

Q)

cl;

C m

Ln cli ks Cl Co

r_

Lr;

0 =

cv

T -. ; 0 "

IM

cli z r

Ln z r -4 1

; cl en

47% ; C V)

276

model.

The measured displacements 1'n Table 5.16 show given between the differences small relatively measured vertical displacements of symmetrically positioned nodes, but relatively large differences between the theoretical and experimental values of In general, the theoretical displacements 15% than the corresponding greater approximately both the theoretical Table 5.17 gives d1sp I acements. for all of the nodes monitored experimental values of deflection for two different transducers, values of the -imposed load. in the middle of f irst load value of 4000 N is approximately were exper imental and by The the

displacements.

load the second value of 7934 N corresponds with while elastic range, the theoretical end of the linear elastic range. When the structure imposed load total a of 7984 N, the majority of the was supporting 10% less than the were approximately indicating that the elastic limit of values, measured experimentally the structures may have been exceeded supporting this load. Figure 5.29 gives both the member numbers and the values of the for the ratios model structure supporting a total stress member theoretical It can be seen from the Figure that the imposed load of 7984 N. tension heavily stressed members are the lower chord edge most 58, 3,8,33 have and which a ratio of actual stress numbers member The adjacent members 2,4,57,59 to yield stress of 1.05. etc., that they are also close to have a stress ratio of 0.98 indicating 0.28 In The Figure 5.29 for the ratio of stress given yielding. 62,63 indicates that these members are etc., numbers member soft theoretically capable of supporting almost a four fold increase In before buckling. loading compression Theoretical behaviour Non-Linear of model 3 for model 2. Behaviour: is The theoretical non-linear behaviour displacements

to the theoretical similar Using the non-linear described pin-jointed analysis (Figure 3,33,58 that 28 it 5.29) predicted members and was program , first to the the be members yield when structure was would four load total symmetrical point of 7984 N. When the a supporting imposed load was increased to 8640 N, members 2,4,57,59 total in became (Figure 5.29) tension. The critical and yielded etc. followed by the these was members almost simultaneous edge of yield (Figure 7,8,9,52,53,54 5.29) when the members etc. of yield imposed total load a supporting of was structure had increased load acting on the structure force in the eight soft members had theoretical that these members would no longer indicating
277

10320 N. to

When the

10880 N, the increased to 3700 N remain elastic but

Total

Imposed Load 4000 N Experimental Vertical Displacements(mm) 4.658 1.820 2.027 1.935 2.037 2.048 4.042 1.747 1.818 1.856 1.923 1.923 1.921 2.196 2.097 2.066 1.173 1.895 2.194 Node Nmbers SO:31 37 26 25 32 20 21 16 38 27 15 14 33 9 10 22 11 39 28 4 3 1.575 I 17 5

Total

Imposed Load 7S84 N Experimental -Vertical Displacemcits(mm) 8.613 4.411 4.802 4.752 4.806 4.929 7.907 3.915 4.361 4.331 4.496 4.722 4.559 5.071 4.534 4.543 2.629 4.245 4.722

Node Xnbers 31 37 26 25 32 20 21 16 38 27 15 14 33 9 10 22 21 39 28 4 3 17 5

Theoretical Vertical Displacements(mm) 2.370 2.359 2.359 2.359 2.343 2.343 2.257 1.969 2.267 2.267 2.267 2.267 2.256 2.256 1.943 1.943 1.121 2.126 2.126 2.126 2.126 1.275 1.275

Theoretical Vertical Displacements(mm) 4.731 4.709 4.709 4.709 4.677 4.677 4.504 3.731 4.524 4.524 4.524 4.524 4.502 4.502 3.877 3.877 2.238 4.243 4.243 4.243 4.2243 2.545 2.545

3.092 I

Key:

Transducer Armature Fixed To The Cross Beam. Displacement Measured By An LNMTTransducer Incorporated 7be Actuator. Fault. Random Readings Obtained Due To An Amplifier

Into

And Experimental Of Theoretical Deflections Comparison 5.17. Table both the go-inito-r-e! Mode 1 3. Ihe ]able Nodes I gives Al I For -of displacements for values vertical experimental and theoretical for different two by transducers, values the monitored nodes of all first load 400ON is The load. imposed of approximately value of the in the middle of the elastic range while the second load value of linear theoretical the the end of elastic with 7984N corresponds total imposed load the When was supporting a structure of range. displacement were less than 7984N the majority of the theoretical indicating that the elastic limit values measured the experimentally been load. have this exceeded supporting the may structure of
278

121

M-)R

S-PA

)I

37

sa

99 n CD CN
o C t .

100

.r
o. 12
93

L"co

F,
0.12 S5
C

C %D

0.26 91, to C
ED 9 CD N , C, to

0.26 85
CD (

cs LD

, CD N

0.05
0 C,
Lo cm ko 10 CD to

o. eG 90

(D IR

0.06 91
0

LD C C

0.05 92
r N

a: cr

N.

0.12 4K 85
LO

0.26 66
Lo CS) M CP tZ,

0.26 87

-0.12 68

C
0.28 83 0.21 B

0.21 81

0.28 82

AR4m

qQ

I-

AN

A- qR

A- W4

56
Ir

57
CD CD CD in

58
CD

59
CD CD t) ( -y .

60
cz

in
cs

(X cr

0.00 51
r

0.57 52 rl
CD m

CD 3 m -

0.53 53
T

0.57 511
CD 'l 3) a rv r a -4

0. Ple 55
Co cr,

cs

0.00 16
Lr

0. Ple

0. OS

0.09

A7

ji 6 Co to

dis

0.00
z5 0 to
CD14

p2
O. e

vCo CD
di 1
r-

Z
A,4

A2
Co Co N v CD

di 5
rCD

CM

9 rli
0.57 37 0.53 38

c CD
0.57 39

m %j
0.00 410

9 cr,

..
BOTTOM GRTD

0.00 36
-9

CD

CD

CD

CD

0.54
.51

0.98
jw

1. a5
a 13

0.98
41

0.54
33

The diagrams show the member numbers and critical -At FirsU-Ti-eld-for both the top and bottom chord members of model 3. Tt-ress ratios at four top chord nodes and was The model was loaded symmetrically four bottom the The the corner of nodes chord. at supported (Figure I 4.11), type eight soft members member contains structure 99. 98 Loaded 62,63,82,83,78,799 and nodes shown by numbers For tension Critical (bottom chord): members axial stress member Stress Ratio = yield stress chord, including soft members): member axial stress flexural buckling stress 279

Fiqure 5.29.

Critical

Stress Ratios And Member Numbers For Model 3

For compression members (top Stress Ratio Critical

load in in exhibiting a constant plateau compression, yield would their load-displacement response. At this stage in the analysis the in load increase an without the yield of any carried structure deformation of the soft members, but with the plastic additional of the yielded tension members and the gradual strain-hardening load increased imposed the to When the acting on structure members. 12000 N the corner, bottom chord tension members 1,5,31,35 etc. This immediately 5.29) became critical and also yielded. in the analysis program resulting from led to numerical instability in all of the bottom chord edge tension the loss of stiffness theoretical terminated, the the When analysis was middle members. tube of the soft members had yielded a total of 9.78 mm, just short in 1 Type the 10.0 the allowed soft members of movement mm of (Figure (Figure 4.11) before they close to form the triple members. Experimental experimental 3 to model Non-Linear col. lapse, Behaviour: relationships in Figure 5.30 shows the from testing tube compression

load-displacement

theoretical addition obtained from the analysis program. load-displacement relationship It can be seen from the Figure that there is only close agreement

obtained to the

and experimental values for the Initial between the theoretical load-displacement The the relationship. experimental of portion load-displacement behaviour plotted in Figure 5.30, shows that after load on the acting structure applied exceeded 8000 N, the the decreased stiffness of structure as approximatley This Initial ductility the bottom chord tension members yielded. interrupted by the the by structure model was premature exhibited (Figure 62 5.29), when the the soft member, number buckling of (Point load imposed 13816 total N A, of a supporting was structure the total The buckling of this soft member was Immediately capacity of the structure followed by a reduction in load-carrying force failed from Into the transfer of member the adjacent rapid and from buckled load the This shedding soft member resulted members. In the failure of the adjacent top chord compression member, number 66 (Figure 5.29), when the structure was supporting a total load of Figure 5.30). Failure of member 66 11328 N (Point B, Figure 5.30). further decrease in the load carrying capacity of the In three bottom the of occurring stress of a reversal (Figure 48 42,43 5.29). These and numbers member resulted structure In a and chord tension three members

(Points C, D and E, Figure 5.30), in buckled quick succession load decrease imposed the 7669 to to the structure acting on causing N. The buckling of these three lower chord members was followed by
280

15

NODE NO.25 EXPERIMENTAL RESULT ANALYTICAL RESULT

10

--p

50

100

150

VerticaL DispLacement in mm.

Theoretical Experimental And Load-Displacemen 5.30. Fiqure Eheoretical both the Of Model 3. lhe Figure shows Behaviour and displacement 25 in. 3. The of node vertical model experimental four (19,21, loaded top simultaneously at chord was nodes structure displacement a vertical 24 and 43) by imposing of O. 1mm per minute This four displacement the nodes. rate of was continued on each 25mm ical displacement had of vert occurred, approximately until * increased displacement ten fold 1.0min the to rate was per whereupon displacement ' total After a vertical of the actuator. of minute. displacement increased 100mm the further to rate was approximately 2. Omm per minute. load-displacement Points shown on the theoretical curve. in the elastic Pi = Mid-point response. P2 = Yield of members 3,33,58,28. P3 = Yield etc. of members 2,4,57,59, P4 = Change in stiffness of yielded members P5 = Yield etc. of members 7,8,9,52,53,54, P6 = Change in stiffness of yielded members P7 = Yield members. of soft P8 = Yield etc. of members 1,5,31,35, with ensuing numerical lity. instabi load-displacement Points shown on the experimental curve. buckling A= Premature of sof mber 62. B= Buckling of member 66. C= Buckling of lower chord member 42. D= Buckling of lower chord member 43. E= Buckling of lower chord member 48. F= Buckling of lower chord member 33. in load carrying due F-G = Decrease capacity to failure of member 33. H= Buckling of diagonal web member 127. Model slipped halted. I= off support and test

281

The imposed load of the structure. an increase in the stiffness increased the to the outside lower slowly structure until applied (Point in 33, tension F, Figure 5.30) ruptured number member, chord total load 12191 N. The the supporting a was of structure when in a rapid decrease in tensile failure of member 33 resulted first imposed the load-carrying capacity of the structure, with the total load dropping to 5116 N (Point G, Figure 5.30), followed by a return When the imposed load had increased in stiffness of the structure. to 8673 N there was a small lateral deflection of the solid diagonal (Point 127 the load H, under south positioned east point web member load was carried by the structure No additional Figure 5.30). and the test was halted when the model slipped (Point 5.30). Figure I, support Figures 5.31 A to off the north west corner

F show how both the theoretical and in lower the strains measured vary chord tension experimentally load is increased. imposed the the acting on structure as members The experimental values recorded from both of the strain gauges have been for 7 to plotted, member except each member attached (Figure 5.31C), where one of the strain gauges developed a fault during the early part of the experimental investigation. A comparison of the strains recorded for members 3 and 2 given show that member 2 yielded in Figures 5.31 A and B respectively, before member 3 and not vice versa as predicted by the theoretical Figures 5.31 C, However, D a comparison of and E shows analysis. that the next member to yield was member 7 followed by the yield of 1. finally the This 8 yield of member yielding sequence and member from the sequence obtained did correspond with the predicted The strains recorded for member 13 given In analysis. theoretical Figure 5.31F especially greater than Member 13 remained elastic throughout the behaviour is elastic. finally test but the to the yielded, contrary of part early imposed the load acting on the total when theoretical analysis, 13323 N. was structure show the experimental the theoretical values values, to be considerably when the member

A comparison of the measured strains recorded from each pair of 1,2,3,8 (Figures 13 to 5.31 E, and members attached gauges strain In these members are B, A, D and F) indicates that bending strains behaving the When the are members elastically. when negligible, bending the become have strains yielded, more significant members especially in members I and 8 (Figures 5.31 E and D).
282

(Positive Strains are TensiLe)

28

3 NO. hEMBER
79
KLO

15

RESULTS EVERIMENTAL No. 15 Strain Gauge No. 1G Strain Gauge RESULT WLYTICAL

CZ.

la

I FIGURE 5.31A 20 .

23

Strain Percentage

NO.2 MEMBER
-cl . 15

-0

RESULTS EXPERIMENT& No.27 StrainGauge No.28 StrainGauge ANALYTIC& RESULT

c21.

le

23

Strain Percentage NO.7 MEMBER EXPERIMENTAL RESULTS No. 39 Strain Gauge ANALYTICAL RESULT

20
1.0

79 13 cu 15 0. 2.
18

123

Percentage Strain ( IOE-01)

FIGURE 5.31C 283

(Positive Strains are TensiLe)

28 _0 _0 15

NO.8 MEMBER RESULTS EXPERIMENTAL No. 13 Strain Gauge No. H Strain Gauge RESULT ANALTTICAL

12

FIG URE 5 . 31D

20

PercentageStrain ( IOE-01)

:2 -0

15

MEMBER NO.I EXPERIMENTAL RESULTS No. 31 Strain Gauge No. 32 Strain Gauge ANALYTICAL RESULT

10
0

v
235

FIG URE 5 . 31 E
20

PercentageStrain ( IOE-01)

15

MEMBER NO.13 EXPERIMENTAL RESULTS No. 38 Strain Gauge No. 37 Strain Gauge ANALYTICAL RESULT

-2;

1 FIC, uRE 5.31F Strain(I OE-0 1) Percentage

And Theoretical Experimental Applied 5.31A To F. Load Vs Fiqure The rigures R-e-TeF77TF-ain 7u-rves For-7, -o-del 3. show how -t-To-r-Mi-caT the in tension 3,2,7,8, strain of values members experimental and in the external load. The 1 and 13 vary with an increase applied from obtained each of two strain values were gauges experimental diametrically the each opposite other at centre of each placed from (Figure for the 7 5.31C) strain recorded apart member member, functioning correctly. gauge was not one where 284

and experimental values of axial stress vary in compression members 61, (Figure 5.29) under an increasing imposed load 70 62,65,66,69 and It can be seen from the Figures that acting on the model structure. between the are large discrepancies occurring although there the theoretical do theoretical and experimental results, results predict the general trend in member behaviour.

Figures

5.32 A to

H indicate

how both

the

theoretical

Figures 5.32 A and B show the changes in axial stress occurring in the outside tube of soft member 62 and in th'e preceeding member (Figure 5.29). Both 61 of the theoretical and experimental number behaviours plotted in Figures 5.32A and B show that when in bottom tension the chord members the stress in occurs yielding increases 61 62 and at a rate greater than the members compression load-stress rate of increase in the imposed load. A 50% increase in the by the structure, load after yield had occurred in carried external in a 88% theoretical tension and a 78% member 2, resulted in increase force the axial carried by compression experimental, behaviour by Similar the soft member 62, 61. exhibited was member in imposed load increase the 50% acting on the model a where first 90% theoretical yield, caused a after and 240% structure in force by the increase the carried soft member. This experimental large discrepancy occurring between the theoretical and experimental (Figure by the 5.32A) carried stress soft member axial of values indicates that either the initial stiffness of the soft member was in the theoretical this analysis, or particular soft underestimated to permitting relative movement occur between the not member was outside and middle tubes of the assembly. 5.32 C and D show the changes in axial stress which imposed 65 66 load the in and as members acting on the occur Is increased. Both the theoretical experimental and structure Figures in Figures C D 5.32 that and show as yielding of given relationships the bottom chord tension members progressed, the axial compressive in 65 66 decreases. However, as the members and present stress load, tension members begin to strain harden and carry additional behaviour reversed and the compressive stress in members 65 and 66 increased significantly, with small increases In the external load acting on the structure. this
Figures load-stress 5.32 E and F show, both the experimental and theoretical response for top chord compression members 69 and 70.

285

(Positive Stresses are TensiLe)

23

MEMBER NO.62

28

15 -J

10

-20

-is

-so

-88

-100

-120

-lie

-1 Go

-198

-280

FIGURE-5.32A

Axid Stress(N/sq. mm) KEMBER NO.61

25

28

12 c 15 cl, . c, -i
0.

-28

-48

--w

-M

-100

-128

-14a

-Im

-198

-2W

FIGURE

5.32B

Axiat Stress(N/sq. mm)

OVER NO.65

25

i
20

15

-<

la

-18 FIGURE 5.32C

-28

-38

-48 -53 -Ga Axid Stress (N/sq.mm)

-70

-80

-se

-100

286

(Positive
25

Stresses are TensiLe) NO.6G MEMBER

----i nrnill T

23

15

-<

10

-20

-io

-60

-80

-100

-128

-140

-160

-180

-280

FIGURE

5.32D

Axiat Stress (N/sq.mm)

25

20

15

10

100 FIGJJRE

88

io

20

-28

-18

-IGO

5-32E

AxiaL Stress (N/sq.min)

-80

-100

MEMBER NO.70

108 FIGUBE

80

GO

io

20

-20

-ia

-Go

-88

-100

5.32F

AxiaLStress(N/sq. mm) 287

(Positive

Stresses are TensiLe) NO.199 MEMBER .... . ....... ..

C -t C C C

-10

-20

-30

-10

-58

-60

-70

-80

-90

-100

Axial Stress (N/sq.mm) FIGURE 5.32G

25

MEMBER NO.101 ---- --------

28

-0

-<

-10

-1210

-30

--io

-50

-60

-78

-80

-90

-100

FIGURE

5.32H

Axid Stress(N/sq. mm)

Theoretical Experimental H. Applied to Load Vs 5.32A and Figure For Curves Model 3. The Figures Stress Axial -Member show how F'ot-F lues v in experimental of theoretical and axial stress the 62,61,66,65,70,69,199 101 members and change with compression load. The axial stress was obtained applied increasing for external the by from three strain readings averaging obtained member each icaIIV tr the symme around placed gauges circumference at strain of each member. mid-length
288

in both of The compressive stress acting decreased as yield occurred in the bottom

members also chord of the model As yielding of the tension members progressed, the axial structure. 70 69 became This behaviour was in tensile. both and members stress

these

reversed with the premature buckling of soft member62, which caused load from failed the transfer of member onto the adjacent rapid a compression members. load-stress behaviour obtained The experimental and theoretical from the corner web members 199 and 101 are shown in Figures 5.32G Both of these Figures show good agreement and H respectively. between the theoretical and experimental results and indicate that the first compression member the members remained elastic until buckled. D show the load-strain relationships to from each of the three strain gauges attached obtained 62,65,66 70 A and comparison of respectively. members compression the three sets of strain readings recorded from the outside tube of (Figure indicates bending 5.33A) 62 that a moment of member soft Figures 5.33 A to kN 5.4 mm was present at mid-length approximately the buckling member occurred. of premature of the tube when

Figure 5.33 C shows a similar behaviour for compression member 66, which was the second member to buckle during the test on the for The both 65 measured strains recorded members model structure. bending in D) that (Figures 5.33 B these 70 strains show and and members were negligible until the test model under load. large vertical deflections occurred In

Model 4 The fourth model square-on-square double-layer designed to exhibit failure to also was tested The load-displacement structure response. post-elastic members located space truss a ductile Incorporated

in the top two chord, with soft members eight soft (Figure the centre of about each side symmetrically, positioned The soft members used in model 4 were all type 2 soft members 5.6). (Figure 4.11) which were an improved version of the Type I soft fifty-two 3. The in top model remaining chord members used members (Table T5 5.1) identical type tubular to the members the were 3. in the top chord of model used members The web bracing members

289

(Positive Strains are TensiLe)

25 OBER NO.62 RESULTS EXPERIMENTAL No. 7 Strain Gauge 25

20 -

No.40 Strain Gauge Strain Gauge No.8-

28

,a 01
--0 c2.
.<-

15 .e %.
la -

15

--BUCKLING

LOAD a 13-816 kN
la

5-

)AI

.5

-3

-2

-1

Strain ( 10E-01) Percentage vTGuRE 5.33A

25 -

28 .

MEMBER NO. 65 25 EXPERIMENTAL RESULTS ' StrainGauge No.2G StrainGauge No.25 StrainGauge No.21 20 13

,a cl
-0

13

-3

-2

-1

12

Strain ( IOE-01) Percentage

290

(Positive Strains are TensiLe)

25 . NO.66 MEMBER EXPERIMENT& RESLLTS No. 21 Strain Gauge 25

28

No. 11 Strain Gauge No. 12 Strain Gauge

20

BUCKLING LOAD z 11-328 M

Is -.

ca. C%

le

-3 FIGURE 5-33C

-2

-1 Percentage Strain

23 IOE-01)

25

20

MEMBER NO.70 25 EXPERIMENTAL RESULTS No. 20 Strain Gauge N9.18 Strain Gauge 17 Strain Gauge No*. 28

15

ca.

le

la

-3 F_TGURF . 5.33D

1 -1 Strain ( IOE-01) Percentage

Load Vs Strain Experimental Applied Member 0. To Figure in Obtained FFOmModel 3. The Vigure shows how the strain Curves _5.33A 70 increasing 62,65,67 varies an and with compression members from load. The values experimental were obtained applied externally ly the three the symmetrical around positioned gauges of each From it is Figure the member. of each mid-length at circumference 62 buckled the the that structure soft member number when evident 13.8kN load imposed that 66 of and also member an supporting was 11.3kN. imposed load of an buckled under Tensile strains are shown as positive. 291

identical to the in in 4 both also web members were used model used bottom 3, however the 2 chord members of model 4 were and models type T2 members (Table 5.1) which had both a larger external diameter and cross-sectional area than the bottom chord members used in each of the three preceding models. model structure was loaded only at the top central identical (number that for 31) to an procedure adopted using node displacements were measu'red at twenty-three model 1. The vertical (Table V. D. T. 5.2), and the strain was L. transducers using nodes The fourth Three strain members. gauges were used to measured in fifteen in the each of the nine compression members, while strain measure two gauges were used for each of the six lower chord tension members All of these measurements were recorded at thirty (Table 5.4). the throughout intervals experimental second 5.34 shows the model structure under test. investigation. Figure

Linear Behaviour: Before the fourth model was tested to collapse, a tests the structure to determine carried out were on of number small if a symmetrical response was exhibited by the model when loaded In first load the test, the range. a elastic cell was within four the the The each of corner nodes at supports. under positioned first increased load from was zero to approximately point central 5000 N and then decreased to zero again. Readings from each load load both Table 5.18 the taken at values. gives average were cell loads recorded in each of the four load cells when the structure was load 4985 It N. be from point of central can a seen supporting Table 5.18 that there is close agreement between the loads recorded load 61. 1,6 However, the by load the recorded and supports at 2.4% less 56 than load the was approximately support at cell 6 61. both and nodes at recorded
test, one strain gauge was attached For the second preliminary to each of the eight soft members, and also to each of the four in The these each of strain members was members. web corner increased from load the slowly zero to approximately was as recorded 5000 N. Table 5.19 gives the strain recorded in each of the eight four the corner web members, when the model and members soft load supporting a central point was of 4859.4 Newtons. structure The strains recorded in web members 118 and 119 were within 2% of their theoretical

However, the strain values. recorded In web 184 18% 101 26% less approximately and were greater and members.
292

L LIMt Im, ,, , I,Ile

IBM

qm

,i.:

$
1) (1) -fl -n

F.
C)

(1

4-A I
IL

I' I
293

Model 4

in Y direction. the boundary four vertical All restrained nodes Note

Total

Imposed At

Load Cell

Readings

In Newtons

Load Applied

Node 31 (Newtons)

Node Nuaibers

61

56

4985.0

1246.0

1256.0

1256.0

1226.0

For Model 4. 5.18. Readings Load Cell The Table Table the gives -c-e-FTloads from load each recorded average Positioned under the four imposed load of 4985N was nodes of model 4, when a total supports to the structure The individual at the top central applied node. from load load the of recorded cells values Positioned under nodes however the load recorded 1,6 agreement, and 61 are in close under 2.4% less than the loads node 56 was approximately at nades recorded 6 and 61. 294

'N19 79 78 N/*

Ce) COI

CD
o

CY

63_

62

1 Co

Member Numbers

Member Numbers

Member Strain Under A Point Load Of 4859.4N On The Structure Acting Theoretical Measured 1.2375 1.0966 1.0611 7.7837 5.8388 4.4637 1.3316 8.1735 4.9514 2.1431 1.3745 1.2567 x x x
10-4 10-4 10-4

101 118 119 184 62 63 98 99 79 18 83 82

1.05079 1.05079 1.05079 1.05079 2.2349 2.2349 2.2349 2.2349 2.2349 2.2349 2.2349 2.2349

x 10-4 x x x X X X X x X X
10-4 10-4 10-4 10-4 10-4 10-4 10-4 10-4 10-4 10-4 10-4

x 10-5 x 10-5 x 10-5 x

10-4

x 10-5 x 10-5 x 10-5 x 10-5 x 10-5

The Table Strain Readings From Model 4. the 5.19. Table gives in the each members and the four of eight sof-t recorded strains was supporting a central corner web members when the model structure in web members load of 4859.4 Newtons. The strains recorded point 2% of their 118 and 119 are within However the theoretical values. ls% in web members 101 and 184 are approximately recorded strains than the corresponding theoretical and 26% less respectively greater from gauges attached The strain to the recorded values. readings in the measured strain soft eight members showed large variations in these members. The strains in the outside recorded occurring tube soft members 78,82 and 83 were sufficiently small to indicate that these soft members were not functioning correctly.
295

than respectively, differences I arge

the

corresponding

theoretical the

values. theoretical

These and

occurring

between

in members 101 and 184, are most measured strains, experimentally in these likely to be due to the presence of bending strains members, the magnitude of which could not be determined because only one strain gauge was used for each of these given bars. in the recorded large variations members. eight soft members also In the measured strain in the The strains in Table 5.19 show between these

occurring

each of the less than the theoretical eight members was significantly value, in the outside tube of members 78,82 and 83 with the strain that these soft members were not small to indicate sufficiently outside correctly. have been due to the tube, due to functioning The malfunction of these soft members may inner and middle tubes binding on the outside between the three Insufficient tolerance existing by

The measured strain

tube of

tubes,

the middle tube may have been constrained the bottom welds used to attach the soft member to its end node. or movement of Table 5.20 of gives both the theoretical and

groups of nodes symmetrically to a central point load of 15844.4 N acting on model corresponding The Table shows that there are small differences the structure. occurring positioned in the vertical but nodes, measured for symmetrically there that large also are relatively between the measured values of displacements theoretical values. In general, the measured 8% less than the approximately displacements

displacements

experimental in the positioned

occurring variations and the corresponding displacements corresponding

were found to be displacements. theoretical vertical theoretical for values all

of

Table 5.21 gives both the displacements the vertical to two different

and experimental of the monitored

values nodes on the

of Imposed load acting load value of 8015 N is approximately The first in the structure. range, while the second load value of 15844 N middle of the elastic in the elastic limit behaviour of corresponds with the theoretical corresponding the structure. The displacements imposed central a total supporting 16% of their theoretical values. displacements theoretical were less than the corresponding vertical that the actual stiffness of the model structure values, indicating was greater analysis.
296

measured when the structure was load of 15844 N, were all within The majority of the measured

than

the

stiffness

predicted

from

the

theoretical

E 90
c: i
41

S-

c
uu

Uli
c71 Lei

-4 Cm

r m

(D 4- tA -C _C 0 to 3: r 0 -4-c 4-) 0 L) ." 001.

-rM0M -4-3 u tA WWoU CDL G) :3

-
Ln -i- t(U C) -CD M

%D

c r N:

1: OD

vi

1-4cu cu

4a) cm

(0 3: > s.. -> -M ro 0 -6.1 tvi "a e r- C) 40 F- (1) .-0 c7, MV) 4- c c:. c: o
'm (1) f- -J L3 c3

r M0

10
LU cl:

L)
ci

ui Ci-

x LU
(3, b -cr cn U- ) to r-

m 41 u

1-4

-.LI

112: C

1C3 cC wVG; < 0 L)

%LA s.CD Mc

cm -C3 0 (1) -+. )

Q)

> 00

>
to Lt) CD 0% ci

CL
CM - U) po G) ro %_ +1 C) >

ei 0 m=

4-3 c: -

CM L) r- -4-3 CD ,> 'U 44.

m0 (L) Ez
CD c: 0
LO rrCYN tm m r C:) C" Co r

c
2 4.3 %0

r,

Z
to

C!
Ulb

.1;

% ci- -0 c:
c2 > 00 uu1.

115 >-, CO Z Ln 2 44sC) -

-t ---

cu S-

%0

cu

tA

(0

CL

E KZ
w w -0 c> fCo rm r-4 M

cm 0

s-

a) tA >. > m4

>,0 (2) 0) CD
-J c -J

(1) m s- r= r=-mu

cu -0 - ..0 o0m (0 (D c: -- F-

(0

cu 'm ro -., .1. CL -4--3 o

n5 (0 a) -c +i E

::

Icr

0%

Ln

Co

r,

u0 zW

M-?
4,

r,

fi

fi

t'

C!

S- 4-

c(;

Ln
o

%0

ti,

C: )

c>

.: cli

C.;

.:

-;

cli

-:

297

Total

Imposed Load $015 N

Total

Imposed Load 15844 N

Node Nu; nbers 31 37 26 25 32 20 21 38 27 14 15 16 33 9 22 10 11 39 28 3 4 5 17

Theoretical Vertical Displacements(mm) 4.651 4.222 4.222 4.222 4.039 4.039 3.658 3.715 3.715 3.715 3.715 2.959 3.553 3.553 3.021 3.021 1.689 3.233 3.233 3.233 3.233 1.910 1.910

Experimental Vertical Displacements(mm) 7.326 3.999 4.219 4.273 3.951 4.240 3.763 3.202 3.455 3.950 4.214 3.244 3.039 3.600 2.471 3.235 2.453 2.377

Node Numbers 31 37 26 25 32 20 21 38 27 14 15 26 33 9 22 10 11 39 28 3 4 5

Theoretical Vertical Displacements(mm) 9.192 8.344 8.344 8.344 7.983 7.983 7.230 7.343 7.343 7.343 7.343 5.849 7.022 7.022 5.970 5.970 3.339 6.392 6.392 6.392 6.392 3.715 3.775

Experimental Vertical Displacements(mm) 12.594 7.796 7.866 8.034 7.743 7.849 7.062 6.595 6.852 7.202 7.513 6.035 6.430 6.459 5.248 5.876 5.491 5.372

1.432

17

3.376

Key:

Random Readings

Obtained

Due To An Amplifier

Fault.

And Experimental Deflections Comparison Of Theoretical Table 5.21. The Table gives both the itored Nodes In Model 4. For All displacements for theoretical values of vertical and experimental for two different values all of the nodes monitored by transducers, is 8015N load first The load. imposed of approximately the value of in the middle of the elastic range while the second load value of limit in the elastic 15844N corresponds with the theoretical range of behaviour of the structure. to an imposed load of The measured displacements corresponding 15844N are all within 16% of their theoretical The majority values. displacements less than the are all of the measured vertical indicating the theoretical that values, corresponding actual than the stiffness of the model structure was greater stiffness analysis. predicted from the theoretical
298

Figure 5.35 gives both the member numbers and the theoretical values of the ratios of axial stress to yield stress for tension members and axial stress to buckling stress for the compression members, when the structure was supporting a total imposed load of The member stress ratios given in the Figure for the soft etc., give the ratios of member axial membersnumbers 62,63,98,99 stress to the yield stress of the middle tube of the soft member. It can be seen from Figure 5.35 that the most heavily stressed tension members are the lower chord edge members 3,8,33 and 58 The adjacent members, numbers 2, which have a stress ratio of 1.0. 4,57,59 etc. are the with a stress ratio of Figure 5.35 for the soft supporting a total load next most heavily stressed tension members 0.92. The stress ratio of 0.92 shown in members indicate that when the structure is of 15844 N the soft members are also close 15844 N.

The most heavily stressed compression members are the to yielding. top chord members66,67,94,95 etc., which have a stress ratio of 0.72, indicating that these should be the first membersto buckle.
Theoretical behaviour program ratios Non-Linear Behaviour: The theoretical non-]! near

of model 4 was investigated in Chapter two. outlined

each member in the that members 3,33,58 and indicates and response of the structure, 28 would be the first members to yield when the structure was imposed point load of 15844 N. When the total a central supporting for load was increased to 16800 N the soft members 62,63,98,99 etc. (Figure 5.35) became critical with the middle tube of the soft in tension when the soft members were supporting member yielding a Yielding of the soft compression force of 3700 N. followed by the yielding of tension members 8,53,23 load of 17840 N. the structure a total was supporting
As edge the imposed 2,4,57,59 this group until of the load was etc. increased (Figure and load to 18000 5.35) the reached N, yielded preceding a value etc., 66,67 the

using the non-Hnear analysis Figure 5.35 shows the stress model at the end of the elastic

members was and 38 when

lower in

chord

members of

tension. groups of N

Yielding members causing Failure immediately the

members imposed

continued the of

of to

18320 buckle. etc.

compression the followed and the top

members chord

66,67,94,95, members the of 28

compression in

was of

by

a reduction

load-carrying the bottom

capacity chord 5.35). shed tension As load,

structure

elastic

unloading and

members buckled

3,33,58,28,8,53,23 compression members

(Figure etc.

the the

66,67,94,95

299

4A

A-Q?

A qIP

40k

97

98
N

99
V, C CD11

100
CD CD

cs
0.21 93 0.72 0.72 85

0.21 cli
(s)

cr) CD N

0.12 89
c

rq

0.51 so
CD

0.51 91.
CD

0.12 92
co CD

CD to

0.21
C
is

0.72 86
CD to CD

0.72 87
Q

0.21 88
CD

TOP

GRID

O. Aa

13.92
tse

0.92
U-3

O. ia
b4l

0.49 56 56
60 C -

0.92 57 57
I -

1 00 . 38 38
C

0.92

59
-4 a: 'Y .

O. is 60 60
(S

0.00 51

0.53

0.72

0.53 !

a. el3,

52

53

0.00
d, 6
ts

0.04
7
(D

0.76
418
tD

0.0
8
N

0.00
u
CD

t co
O. ea 0.01 0.76

t m
0.01

t -)
0.00

142
cs

113
r" 1

115

cli
O. eo 36

V
0.53 37

IR
0.72 38 0.53
CD

V '1
0.00

Vv

BOTTOM
o
l (S) li

CRTD

GO CD 0 -

a) -

(S)

0.149

0.92

1.00
. 1.3

0.92
"1 ",.

at

Is

49

O. -is
aQ

Critical Stress Ratios And Member Numbers For Model 4 Figure 5.35. The diagrams show the member numbers and critical At First Tield. for both the top and botton chord members of model 4. stress ratios The model was loaded at the top central node and supported at the four corner bottom chord nodes. The structure contains eight type 2 soft members (Figure 4.11), mem6el- numbers 62,63,82,83,78,79, 98 and 99. For tension Critical members (bottom chord): axial stress member Stress Ratio = yield stress

For compression members (top chord, excluding soft members): stress axial inember Critical Stress Ratio = flexural buckling stress For soft members: Critical Stress member axial yield stress stress tube

Ratio

of middle

300

adjacent compression

soft

members

'closed-up'

to carried

form

the

triple load.

tube

members, and consequently

additional

analysis, the stiffness of the stage in the theoretical increased slightly and the space truss was capable of structure load. imposed As the point load acting on the additional supporting structure was increased, the buckled compression members continued to shed load and the stress in the tension members, which were returned to the yield value. When previously unloading elastically, At this imposed load acting on the structure had reached 20776 N the to compression force carried by the soft members was sufficient After the soft members had buckled, the cause them to buckle. of the structure decreased and load was transferred from stiffness the the buckled membersonto the adjacent top chord compression members These members inturn buckled, 70,71,90 and 91 (Figure 5.35). in the complete loss of stiffness and the collapse of the resulting structure.
Experimental theoretical Non-Linear Behaviour: Figure 5.36 shows behaviour both the

load-displacement and experimental of model is only close It can be seen from the Figure that there 4. for the and theoretical results agreement between the experimental behaviour. Under test, the part of the load-displacement elastic fourth

failed locally, with the premature buckling model structure The failure of this web web members. of one of the four central (Figure 149 5.1) occurred when the structure number was member (Point A, Figure 5.36), and was load 15917 N total of a supporting of the three adjacent members The buckling of the four central web members led 150,151 and 152. displacement to the vertical of the top central node (node 31, Figure 5.35), which inturn caused the premature buckling of the four immediately central 5.35). The top chord compression members 70,71,90 and 91 (Figure followed by the buckling

buckling' of the four central almost simultaneous web loss top four the caused chord members, a rapid central and members imposed In load-carrying with the central capacity of the structure, The load shedding associated with the load decreasing to 4973 N. web members was terminated of the four central when the deflection to'allow of the top central node was sufficient vertical top chord members 70,71,90 the four buckled central and 91 to act The imposed load carried as a catenary and carry load in tension. failure by the structure slowly increased to 6021 N, causing
301

the compression

P7 28
P5 Is Is P4 41 P3A "*** I", % it, P2f %% a P6 %% 4#Pl

A-o

P8

NODE NO. 31 EXPERIMENTAL RESULT ANALYTICAL RESULT

cl Or
10 Qj r
CL

I
1

G
B C, D IE

-50

lea

150

200

in mm. VerticaL DispLacement

Experimental 5.36. Load-Displacement Fiqure and Theoretical Behaviour of Model 4. The Figure shows both the theoretical and displacement The of node 31 in model 4. vert experimental displacement was loaded at node 31 by imposing a vertical structure 27mm of displacement after which of O. 1mm per minute for the first the displacement rate was increased ten fold to 1.0mm per minute. displacement of the actuator of approximately After a total vertical further 100mm the displacement to 2. Omm per rate was increased minute. Points P1 P2 P3 P4 P5 P6 P7 P8 load-displacement curve. on the theoretical 3,33,58,28. Yield of membeFsYield of soft members 62,63,98,99. Yield of members 8,53,23,38. Yield of members 2,4,57,59, etc. Buckling of members 66,67,94,95, etc. = Soft members 'closed up'. = Soft members buckle. = Buckling of members 70,71,90,91.

load-displacement Points shown on the experimental curve. Premature buckling oTmember 149, followed by web members A= 150,151,152 and top chord members 70,71,90,91. Buckling of member 95. B= C= Buckling of members 74. Buckling of members 66. D= E= Tensile rupture of top chord member 90. F= Buckling of bottom chord member 33. G= Tensile rupture of top chord member 71. G-H = Loss of stiffness due to rupture of memb'er 71. 302

(Point B, Figure 5.36), followed by members 74 95 buckle to member Figure 5.36). The buckling of and 66 (Point C and 0 respectively, the top chord compression member 66 was shortly followed by the tensile rupture of the top chord member 90 (Point E, Figure 5.36) capacity of the sudden decrease In load-carrying load 90, the total With the of carried by member rupture structure. the structure decreased to 4222 N, and the force carried by this 70,71 into the top transferred members adjacent chord member was causing a and 91. As the vertical displacement of the top central node increased, began to the load-carrying slowly capacity of the structure imposed load the 33 buckled bottom The when chord member recover. (Point increased 6526 N F, Figure had to the by structure carried 5.36) and the top chord central member, number 71, failed in tension (Point 7114 further to N G, load increased had Imposed the when The rupture of member 71 caused a rapid loss in Figure 5.36). load 1996 imposed decreasing to the the structure, with stiffness of N (Point H, Figure 5.36). At this stage, the load-carrying capacity of the structure very (number 31), the top increased node and the as central slowly into down it, the to pulled were attached remaining members The experimental investigation was terminated when the structure. displacement of node 31 had reached 200 mmwhich was the vertical displacement limit of the actuator used to load the structure. F show how both the theoretical and vary in six of the bottom chord tension experimental strains indicates Figures that none of these the A of comparison members. failed the had prematurely and, as structure yielded when members 3 by theoretical the analysis, member was the most heavily predicted 2,8,13,7 followed by members and 1 member, stressed Figures 5.37 A to Figure 5.37 A shows a very close agreement between respectively. the theoretical and experimental axial strain values for member 1. it can be seen from this Figure, by comparing the In addition, strains recorded from gauges 31 and 32 attached to the top and bottom of the member, that only small bending strains were present Figures 5.37 B and C show that in in this member at mid-length. both members 2 and 3, the theoretical values of axial strain load value, were larger than the for any particular calculated These Figures also corresponding experimentally measured values. show that bending strains at mid-length
303

in members2 and 3 are still

NO.I M[BEER RESULTS EXPERIMENIAL No. 31 Strain Gauge Strain Gauge No. 32 RESULT ANALYTICAL

-v

FIGURE

5.37A

ia

12

Percentage Strain ( 10E-02)

25

2e
0 a 0 -J

-------

NO.2 MEMER RESULTS EXPERIMENT& No. 27 Strain GGUge No. 28 Strain 6auge ANALYTICAL RESULT

'5 0a l0

FIGURE
5

5.37B

18 Percentage Strain ( IOE-02) NO. 3 MEMBER

25

12

16

^10

28

-------

RESULTS EVERIMENTAL No. 15 Strain Gouge Strain Gcvge No. IG RESULT WLYTICAL

15

FIGURE

5.37C

-4

12

16

70

24

PerCentOge Strain ( 10E-02) MEMBER NO. 7

25

EXPERIMENTAL RESULTS Strain Gauge No. 39 Strain Gouge No. 9 RESULT ANALYTICAL

2$

a 0

r, -i Ia 12

Percentage Strain ( IOE-02)

304

(Positive Strains are TensiLe) NO.8 MEMBER EXPERIMENTAL RESULTS No. 13 Strain Gauge Strain Gauge No. H RESULT ANALYTICAL
.0 -----------------0 ---------------4w

25

zt. -Z -0

20

15

FIGURE

5.37E

12

1ro

21)

Strain ( IOE-02) Percentage

(Positive Strains are Tensite) 25 NO.13 MEMBER RESULTS EXPERIMENTAL No.38 Strain Gauge 37 Strain Gauge 11o. ANALYTICAL RESULT --- 41--Ar

20

15
ra-

If

-8

-4

18

12

16

Strain ( IOE-02) Percentage

Experimental And heoretical Applied Load Vs he FI gur es show 7i_o_w___tFe_ Member Strain Curves For Mode1 4.1 f strain in tension members 1, and experimental values theoretical 13 vary with an increase in the external 2,3,7 and applied ,8 The experimental load. values were obtained from each of two strain ly diametrical opposite each other, mid-length placed gauges of each member.
305

increasing in magnitude the closer the nevertheless small, location in is the structure. to the Figures 5.37 D mid-side member and E show close agreement between the theoretical and experimental for both members 7 and 8. However, load-axial strain relationships the two load-strain obtained from strain gauge 37 and relationships in Figure 5.37 F, Indicate that 38 attached to member 13, plotted but agreement between the theoretical and in the member, bending strains strain of axial values experimental were present in this member even when the structure was supporting a small total imposed load. relatively although there close and in 101, of stress vary compression axial values members experimental 199,61,65 and 69 under the action of an increasing displacement Figures 5.38 A imposed on the top central node of the structure. load-axial for the the two corner B stress relationship give and 199 101 respectively. and members the theoretical between agreement theoretical values of axial Both of these Figures show close and experimental results with the Figure 5.38 A to E indicates how both the theoretical is

than the stress slightly greater for the same value of total load acting obtained values experimental is little However, there the very structure. agreement between on the theoretical and experimental values of axial stress obtained from members 61,65 and 69 shown in Figures 5.38 C, D and E. These three members were in the top chord of the structure and, after the (Figure 149 5.1), buckling of member exhibit a behaviour premature different markedly from their expected behaviour. Figures 5.39 A to C show both the theoretical and experimental load the total between applied acting on the structure relationship in the compression members 62,66 strain axial and 70 and 5.39 Figure A the load relates applied acting on the respectively. in the the to strain axial outside tube of the soft structure large it be from discrepancy the can seen and existing member, and experimental results that this soft member between theoretical from the the start of experimental testing malfunctioning was of load-axial The 4. experimental strain relationship obtained model from member66 (Figure 5.39 B) shows that this member was Influenced to 5000 N, and also that it by the pre-loading of the structure 149 buckled prematLrely. unloading elastic when member experienced The experimental load-axial strain behaviour exhibited by member 70 that the strain (Figure 5.39 C) indicates In this member was decreasing just before member 149 buckled and, also that when member
306

(positive Stresses are TensiLe)

25

28

-0 0
--i c3.
CZ. -< C:9

15

la

-18

-20

-30

-io

-58

-M

-70

-88

-90

-188

FIGURE 5.38A

Axid Stress (N/sq.mm)

20

--e-

le

-10

-28

-30

-40

-50

-ca

-70

. 88

. 90

-100

FIGURE 5.38B

AxiaL Stress (N/sq.mm)

. $sow-*

,a
-0

-20 FIGURE

--ie 5.38C

-M

-88 -108 -120 Axid Stress (N/sq. mm)

-lie

-ice

-180

-200

307

(Positive Stresses are TensiLe) 65 NO. MEMBER

25 C: YDr-PTMPJTAI PPqlg T

28

--p co

-20 -18 FIGURE 5.38D

-38

-68 -50 -io Axid Stress (N/sq. mm)

-78

-88

-90

-108

NO.69 MEMBER
251 ------ 3i
"0 13

RESULT EXPERIMENTAL RESULT ANALYTICAL

20

--------------V

-0 ICU 06

50

is

38

20

Is

-10

-20

-30

-is

-58

Axid Stress (N/sq.mm) FIGURE 5.38E

Applied Load Vs Experimental And Theoretical Figure 5.38A to E. Member Axial StrFs Curves For Model 4. The Figure shows how-7567F of stress in compression the theoretical and experimental vs 101,199,61,65 69 increasing and change with external members The axial stress was obtained for each member by applied load. averaging the strain reading obtained from the three strain gauges around the circumference at mid-length of each placed symmetrically member.
308

149 had failed, member 70 unloaded elastically strained in tension.

and eventually

became

Figures 5.40 A to H show the load-strain relationships obtained from each of the three strain gauges attached to compression members 101,199,61,65,69,62,66 A comparison Of and 70 respectively. obtained for each member gives relationships an indication of the amount of bending occurring within the member. Figures 5.40 A and B show that bending of t6e two corner web occurred from the start of the members, 101 and 199 respectively, test and that bending strains had reached significant values when the three load-strain of the model occurred with the premature buckling of member 149. Even larger bending strains were experienced by members 61,66 (Figures Figure C, G H). 5.40 F 5.40 70 and gives the and behaviour for each of the three gauges attached to the load-strain The Figure shows that a outside tube of soft member number 62. failure 850 Nmm was present at moment of approximately 149 buckled. This bending the tube member moment when of mid-length forming the three tubes the hindered have of movement relative would the soft member and, in addition, would have had a detrimental behaviour exhibited by this influence on the load-displacement However, it is not envisaged that the presence of this member. small bending bending moment caused this particular member to malfunction, in the structure yielded the seven soft members other of several 'closed-up'
DISCUSSION Component Members

as and

during the experimental

testing

of the structure.

Both the tensi le and compressive behaviour of the individual of the test models had a component members used in the fabrication load-displacement response of influence on the entire significant. The tests undertaken on the structures. component members reported in this Chapter were used to determine, the various parameters required to as possible, as accurately describe numerically the complete load-displacement response of the each of test A consideration of the mean values with the given in Table 5.9, which were corresponding standard deviations from the experimental calculated results given in Tables 5.8 indicate that the spread in the values, of both yield and ultimate less than stress obtained from the tensile tests, was significantly members. individual the four

309

1--

(Positive Strains are Tensile)

23 IIEWER NO.G2 RESULTS 25 EXPERMITAL

26 .
v 11 ...........

. .,

'A %\\ -,

STRAIN AXIAL MYTICAL RESULT ---

28

-W

Is -

-I I'

Is

'I
5.

-12

-4

48

:
12

FTC,USE -5-12A

Percentage Strain 18HO 25

20

GG MMER NO. EXPERIWAL RESULTS STRAIN AXIAL ANALYTICAL RESLLT

15

-12

-0

12

rT MJRF

-9

3 9B

Strain Percentage

IBE-02)

23

20

70 tMER NO. EXPERIMENTAL RESULTS AXIAL STRAIN RESLLT ANUTICAL --- 0---

-0

0 0

-12

-8

-4

48

12

VTr, URE

5-3c)r

Strain( ICE-02) Percentage

Experimental And Theoretical Applied Load Vs Figures 5.39A to C. The Figures show how Member Axial Strain Curves For Model 4.

malfunctioning 4.310 model

in the theoretical and experimental values of axial strain and 70 change with incr. easing external compression inembers 62,66 The axial strain was obtained for each member by applied load. gauges averaging the strain readings obtained from three strain placed symmetrically around the circumference at mid-length of each member. Figure 5.39A relates the applied load acting on the structure to the axial strain in the outside tube of the soft member. It can be from seen the large discrepancy existing between theoretical and Figure experimental that results this in soft this member was shown
from the start of the experimental testing of

FIGURE

5.40A

0
15

NO.101 MEMBER EXPERIMENTAL RESULTS Strain Gauge No. 29 Strain Gauge No. 3 No. 30 Strain Gauge

10-

-12

-8

-1

12

10E-02) Percentage Strain (Positive Strains are Tensite)

FIGURE

5.40B

2 5 1: 20 0

C5 0 Cl-

15-

MEMBER NO.199 EXPERIMENTAL RESULTS Strain Gauge No. 5 Strain Gauge No. 6 Strain Gauge No. 19

10

-12

-8

-1

18

12

Percentage Strain

( 10E-02)

(Positive Strains are Tensie) 2

FIGURE

5.40C

20

MEMBER NO.61 RESULTS EXPERIMENTAL

Strain GaugeNo. 36 Strain GaugeNo. 35 Strain GaugeNo. 22

0
15010- 2-

-12

-8

-1

18

12

Percentage Strain ( 10E-02)

(Positive Strains are TensiLe)

311

: 2-'

FIGURE

5 40D .

20

15

CL 01cl (Positive

NO.65 MEMBER EXPERIMENTAL RESULTS Strain Gauge No. 26 Strain Gauge No. 25 Strain Gauge No. 24
Strains are TensiLe)

le

-12

-8

-4

18

12

Strain Percentage

10E-02) NO.69 MEMBER RESULTS EXPERIMENTAL 0

Strain GaugeNo. 33 Strain GaugeNo. 31 Strain GaugeNo. 23
(Positive Strains are TensiLe)

25

FTGURE

5.40E 201 c,

-12

-8

-1

18

12

Percentage Strain ( 10E-02)

NO. 62 MEMBER EXPERIMENTAL RESULTS FTGURE 5.40F - 20

Strain Gauge No. 7 -0

CL Strain GaugeNo. Q Strain GaugeNo. 8 (Positive Strains are TensiLe) 0

15

-12

-1

18

12

Percentage Strain ( IOE-02) 312

25

20
-0

NO.66 MFJIBER EXPERIMENTAL RESULTS Strain Gauge No.21 Strain Gauge No. 11 Strain Gauge No. 12

is

le

-12

-8

-1

48

12

Percentage Strain ( IOE-02)

25-

20

cl

NO.70 MEMBER EXPERIMENTAL RESULTS No.20 Strain Gauge No. 18 Strain Gauge Strain Gauge No. 17

15

10

-12

-8

-1

12

Percentage Strain ( IOE-02) (Positive Strains are TensiLe)

5.40A To H. Experimental Aoplied Load Vs Member Strain Figure Mode1 4. The Figure shows how the strain Curves Obtained in 61,65,69,62,66 compression members 101,179-, and 70 varies with The experimental externally applied load. an increasing values were from each of the three gauges positioned obtained symmetrically at mid-length around the circumference of each member.
313

flexural buck] ing stress the spread in the values of the critical However, the mean experimental obtained from the compression tests. buckling stress, obtained from both types of values of the flexural (Table T6 5.1), showed good agreement T5 and compression members values, calculated assuming an with the corresponding theoretical For both of these member initial sinosodial member bow of 1/1000. types the theoretical values of the buckling stress were less than indicating that the assumed the mean of the experimental values, member bow of L/1000 was probably greater than the average bow of the test samples. An initial member bow of L/1000 was used difficulty because of the in the theoretical calculations experienced in measuring accurately both the deviation of the test initial line, and also the wall thickness of the samples from a 'true' various tubes. These are both extremely important parameters, which if be to results are experimental used measurement precise require assumptions. to assess the validity of theoretical Behaviour of Models 1 and 2 designed Test models 1 and 2 were almost identical structures, imposed increasing load, the that of an action extensive under so tensile yield would occur in the bottom chord members before any of To fulfil these design requirements, the top chord members buckled. the top chord compression members were significantly stronger than the bottom chord tension members, with the flexural buckling load of the compression members over eight times the yield load, and six times the ultimate load of the tension members. This unusual ratio led to the required behaviour, of member strengths with the load-displacement response of both of these structures showing after first yield. considerable ductility The experimentally observed elastic behaviour of each of the double-layer grids was both linear and fully symmetric. displacements of symmetrically The measured vertical positioned in close agreement with each other and also In reasonably were nodes Elastic behaviour: In both values. good agreement with their corresponding theoretical the measured vertical deflections model structures of the boundary nodes were approximately 5% greater than their theoretical values, of the central nodes were approximately and the measured deflections 5% less than their theoretical This pattern occurring In values. the difference between the experimental and theoretical deflections was unexpected. It was envisaged that, because the theoretical

314

instead be to the of rigidly pin-jointed structure analysis assumed deflections the theoretical of all of the monitor jointed, vertical the measured values. experimentally exceed marginally would nodes This apparent discrepancy between the experimental and theoretical in between differences due be to displacement rigidity may values of The boundary the boundary and middle regions of the structure. into half framing five in by held one of members position nodes are are held in the joint whereas the remaining nodes in the structure position by eight members symmetrically was obtained positioned between around the node. the

experimentally agreement In the theoretical strain. general of axial values measured and than the slightly greater theoretical strain were of axial values due be to the difference The may values. measured corresponding in both structures, which were not bending present strains small Close considered Inelastic in the theoretical behaviour: A analysis. the theoretical and of comparison I from both behaviour load-displacement models obtained experimental the that 5.24 5.17 in Figures shows 2, respectively, and given and the actual of prediction an accurate provides program analytical both behaviour The behaviour. of structures non-linear structural in in a moderately good agreement resulting symmetric, general, was, between the theoretical nodes. there was also very good agreement between the In addition, in the strain axial recorded of values theoretical measured and However, that the tension the monitored. were members majority of (Figure 2, 2,3 8 5.25) in model and of members strains measured yield occurred path just after first in the model. This discrepancy in the theoretical and experimental from have the yield resulted may strain axial of values One the these quarter of members. particular of characteristics bottom chord members of model 2 were cut from one tube, type deviated from the theoretical brown-green, small an extremely The before actual post-yield strain stiffness post-yield of this member, measured from tensile specimen tests, was stiffness in the the less than value used analytical mean significantly program, which was calculated from all of the tensile specimen tests which exhibited hardening. undertaken on tube stiffness post-yield type The use of to model the displacement behaviour
315

and measured vertical

displacements

of the

(Figure

5.4)

T1 (Table

5.1).

a larger of these

program would lead to the members in the analysis in these members, which is underestimation of the strain occurring load-displacement of the post-yield shown in the early portion behaviour given in Figures 5.25. tension A comparison of the theoretical and experimental values of axial stress in the top chord compression members of both models 1 and 2, indicates that in general there was a reasonable agreement between these two sets of values throughout the. non-linear range of Examination of the test data shows that in both of these behaviour. structures the experimental values of axial strain were greater than the theoretical values for the top chord edge compression members, values for the middle top chord and less than the theoretical These findings are compatible members. with the compression during the elastic exhibited response which the model structures load-displacement behaviour, where the measured portion of their displacements of the nodes were slightly greater than the vertical for the boundary nodes, and theoretical values corresponding slightly less than the theoretical values for the internal nodes. Figures 5.20 (A, B and C) and 5.26 ( A, B and C) show how the axial stress in some of the top chord compression members of models 1 and 2 alter non-linearly as tensile yield progresses throughout to consider the bottom chord tension members. It is interesting that exists between the top and bottom chord further the Interaction axial forces and the changes which occur in the force distribution is increased. Figure as the imposed load acting on the structure 5.41 gives the theoretical values of axial force occurring in the taken through model 1. for each of members at four cross-sections A comparison of these values of the imposed load. diagrams given in Figure 5.41 shows that as yield progresses In the bottom chord tension members, from the centre members outwards two different towards the edge members, the axial force in the top* chord above the yielded tension members, compression members, positioned increases by a larger percentage than. the increases occurring in the other top chord compression members. This behaviour was not expected. Instead it was envisaged that as bottom chord tension members yielded and carried a constant force, the force in the compression chord members above the yielded It members would also rema1n almost constant. was considered that increases in the imposed load would be carried by the adjacent elastic top and bottom chord members. However, both tension
316

131 131 Ga z
: 4Z Be

0 cc w
cc w to

L A!

sl

Cl

el

41

4A z
ILI

19

99

A0 -

L S 1 4
01
L,
.N

. ;
c

0
01 C.
r,

AA U

4- -.

cc -= -0 u " , 1; s- N; -- (U 7s
(L'

u
0.

" 0

1c cu "J" rs 0
.

4, 0 ;0
0 scu 4-

-; -

c c
0 0 0 0

.
40

Uw

cc:,

0
'6 ,0

1?

21

-. .0

0. -mc

C.
.--. qu 4- v,

C;

u
-*tt al Zo

(4.

V
0

:9 10 CL 0 2, to -* -. It)
E

1 0s .. . 0 "o -Z 0

01

20
6
0 ", N
,

N , 0 ,
mT 01
., 0

C',

.+

a, 3
41,

.21

'a. E-

I-

qu

' W s

w U;z
20 p 5z ;z

0
ab

'j.
-

0 'D V,S% 61 )

. t,

p &. ,

I
I

" .
V

" . 0.

w
(U
a

r- -

I" E

i 02

.0
1

02

e ,.
&I

ti, m
1

4p 0

I
! c :1, ,W
CIw0

aw W u S..

c =, 00

FE4- C
Iu
*

C',

1 'r 0m ' cv:;

%Q 6

.t

-w oz 2.:! 4 = :5- .-1. -Z . .

z

m f4

is mM -

=, @

tm

wI,
to L6 0

-7,
1,6 116

w
8c lot M a
cu A co o 2

tu
w
ILI ,";

l e" aj -D

cu c

wm

12
cc

12

;n

=-

P,

ej .0
3c
aw Iz .0

C Go

3a

'9 1
'a

<,
0

to
'Am .,,

ko

.0 .7 CD

- 11

-0

i
CD
N N

S.-

3: M01. 5

cx

IZ
4 ;,

s i.

'2 Z., 0

tf)

D S.

r)

0)

317

and experimental results are in close agreement in describing this behaviour, and show that although the tension to the yielded tension members, which are both adjacent and parallel proportion of the additional tensile members, do carry a significant load, the top chord compression members, above the yielded tension the theoretical larger proportion of the additional members, carry a significantly compression load, compared with that carried by the adjacent parallel top chord compression members. forces of axial taken through model occurring In the members at four cross-sections It can be 2 for each of two different values of the imposed load. seen, by comparing the member forces obtained for each of the two gives values load cases, that as tensile yielding occurs in the bottom chord edge me. -nbers and progresses towards the centre of the bottom chord, the compression members above the yielded tension members carry all of In addition, the compression the additional compression force. force in the adjacent parallel This behaviour top chord members actually decreases. Figure 5.42 the theoretical

is similar to that exhibited by model I shown in However, in model 1, which is loaded by a central Figure 5.41. in the initially yield results understressed point load, tensile Figure 5.41) positioned top chord members (70,71,90,91 centrally In carrying the majority of the additional compression chord force. model 2, the large additional compression force, resulting from both an increase in the imposed load, and tensile yielding, was carried by the top chord edge members (62,63,82,83 etc. Figure 5.42) which were also members. Behaviour of Models 3 and 4 3 and 4 were designed to model structures load-displacement by post-yield a duct I le exhibit response, tensile yield to occur In the bottom chord members In permitting the yielding with of eight conjunction model soft members, in the top compression chord of each positioned structure. Test model 3 was similar to models 1 and 2, In that the strength of the compression members was greater than the strength of the tension members. The compression members used in model 3 (type symmetrically T5, flexural load a critical buckling times the yield load, and almost twice approximately the ultimate load of the tension members. However, larger tension exhibited two and a half
318

initially

the

most heavily

stressed

compression

Both of

the test

Table

5.1)

LLJ oc
64 84 4.4

sa

6z

LZ

1 9? 1
z

"I U 04 CAL ea I' r, Z/. 14 I. U& 5u

I'z

cc

-'

ZZ

cr
1? , ui uj

61

ol

Ll

$I z < 4) -0 =w fA :2M M
)

qu
GO

%A C

';

co

l A
'a ACS

91

11

w
cla -i

(LJ
It I

Cj U
10

1
0 %0 N

:A '
In p %D

1 0 0

w
. A

' *-, t . .. 22 cc c1 '. "; Et cr 8 -t' : 4.4e-E

CL

0 5

u0
4-

0 .0 'z
CL

-0-0

" t7.0) c"

E- Zu

%I-

c
4)

00 Lm16 c; I
'D

2"

T'O

1 0)

;e
a

a
r.
1 '0 I. A i

lz. 10 -9 10
0 " 1

IA

'R
It

./

0, In
11
0

+- 6 'm

q wIt

v
0 +1 %0

1 I C4.:

-I

iz

'

. -

4 .. u"I, ")
"1
5 > C

0q
. -r

'L.

- W "L, ,, z?, - .r
4? ' Cc,
s

W" 0,. 1.

w
ILI

cc
3 0N .0 iT -I I,

U -C0

Iw

0 10

-, '

%P f, e) el

v 0 1

4 .

".

1-

L =3= 47,

'0 -0"*;

c= 10

to
I . .Z < I

R V, 'a
2 V)

.. .1

0
0 .0 rM

--l 0- 0+

". v

0"-

-0 .

-I

"t -A J. ..

sz "
cy

00

02

ta

--

'0

:5"

', (Us

' j' .

:z0 .2,

m0 clj
N

aj
"N' 0

a)

e
4) u% L6 -

a" , = . ,
c .0

In

0)

0 , -. '0 c 0.0 MmI- '-,

r rl

.1
1

,-, WOO., . ,
U0"cS. _0

Uc

ED.

'00

'0

'

% 0

-ma

31 '00

11
0 i
CD 0.

a! ?" "I'l '.

"bi

0'w

10

'0
0
V

.I
uI z

"'. -" -. cc .z 4=I

-0

C r -

r c 'A '0
.0 In

rm

V)

; -+ ,

W-

1 -

10

0)

c.< 0 A A*

319

members (type T2, Table 5.1)

were used in the fabrication of model 4, providing a yield load approximately 70% of the buckling load (Table 5.1) compression members, also used in T5 by the exhibited

model 4. Elastic 6ehavlour both the theoretical and 5.36 give and load-displacement response obtained-from models 3 and 4 It is evident from these Figures that close agreement 5.30 and theoretical parts of post-elastic, results only for the the load-displacement

Figures experimental

respectively. exists between the experimental and early elastic, behaviour. In general

the elastic range of behaviour, displacements of the nodes for both structures theoretical vertical displacements. This the than corresponding measured were greater in the theoretical due the to be assumption, used partly may to the nodes of the that the members were pin-jointed analysis, throughout the structure, the axial but may also be due to possible variations occurring in into both of the soft members incorporated stiffness , The experimental values of axial stiffness obtained

structures. from the compression tests undertaken on the model soft members, (Table 4.9), in Chapter 4 show significant variations, reported 12% than the highest the greater approximately value measured with identical from nominally members. obtained value smallest in the axial stiffness Consequently, similar variations occurring in soft members, used in either any of the e.ight nominally identical influence the measured values of vertical would models, displacements of the nodes in these structures. undertaken on Model 4, to assess the indicated that in spite of early elastic response of the structure, almost equal values of reactive forces, recorded from each load cell large under each corner support of the structure, positioned The preliminary tests in the outside tubes of some of were measured variations the eight soft members. This Indicated that several of these soft With hindsight it would as designed. members were not functioning have been sensible to remove and replace these members at this stage, but it was considered that their poor performance was due to Interference arising would gradually between the three close fitting tubes, which reduce and disappear as the load acting on the soft in strain

320

members Increased. theoretical analysis,

To it

improve would

both be

the

model

behaviour to test

and the each to soft

beneficial

but in conjunction with its end nodes, member separately, if each member was functioning correctly prior to yielding, to determine Independently the axial stiffness

assess

and also

of each device.

Throughout the elastic range of behaviour, the measured values In both the tension and compression members recorded strain axial of in 4 less than the 3 was, general, slightly and of models theoretical strain obtained from the analysis values of axial Very small bending strains were recorded in the majority of members and because these were not considered in the analysis arising between program, this may have led to the small differences the measured and theoretical values of axial strain. program.
Inelastic behaviour

behaviour of both models 3 and 4 was, in general, Yield did occur in analysis. not as predicted by the theoretical three separate groups of tension members In model 3, as indicated by The inelastic program, but the desired non-]! near behaviour of the by the premature buckling of one of the structure was interrupted soft members. Soft member, number 62, buckled when the structure imposed load 13816 total N. With this value of of a supporting was the force applied to the soft imposed load acting on the structure, the analysis to cause the middle tube of the member would have been sufficient in tension, causing the vertical to yield gap between the member However, premature buckling of inner and outer tubes to decrease. flow had this soft member occurred before sufficient plastic accumulated in the middle tube to allow the three tubes of the soft Close to examination of the soft member during the close. member of model 3 revealed that small rotations exper ! mental testing The occurred to the end nodes of the soft member prior to buckling. of the end nodes imposed bending moments on the soft rotation by the local member, and led to buckling of the member precipitated failure and rotation of the extended portion of the inner tube, (Figure 4.11). to the top This inherent weakness In node close in the type 2 (Figure 4.11) soft bending stiffness occurring members, was minimised by increasing the section modulus of the device close to the top node, by incorporating a collar located over the top of the Inner tube (Type 3, soft member, Figure 4.11). As an to changing the inember design details, it would have alternative
321

been possible incorporating is particularly rotation

to

isolate

the

soft design

member from a 'pinned' any axis,

bending

pinned-joints difficult of the when loaded

at each end of the member. to joint

moments by However, it

equal

member about in tension

which allows and has similar

characteristics

or compression.

The non-linear behaviour of model 4 was also interrupted by the premature buck] ing of one of the compression members. The central (Number 149, Figure 5.1) buckled member when the compression web imposed load 15917 total N a point of supporting was structure . in the member at in theoretical of stress value axial a resulting 2. This value is approximately 29% less than failure of 227.37 N/mm buckling stress obtained from the the mean value of the flexural In twenty T5 samples tested prior to the model test (Table 5.10). (227.37 149 is N/MM2) buckling the stress of member equal addition, to the mean value of the buckling 2), minus 3.29 times (319.24 N/mm 2) N/mm of the test group, indicating stress of the twenty the standard deviation samples (27.91

that there was a probability of less than one in one thousand of member 149 failing at this low However, the theoretical in stress. value of stress axial of value ignoring by the possibility 149 that bending obtained was member in in this develop turn, have a member which will, can strains load of buckling the detrimental on the effect member. arising between the theoretical Nevertheless, the large difference 149, stress of member and the mean value critical obtained from the twenty test samples suggests that it is most likely that this member was damaged during fabrication of the model, flexural buckling probably during the welding of the member to the end nodes. the influence To model mathematically which the premature load-displacement buckling of member 149 had on the theoretical the model was re-analysed, behaviour of the structure, assumming a flexural buckling load of member 149. 29% reduction in the critical Figure 5.43 compares the theoretical -load displacement behaviour, in the buckling load of member 149, this reduction assuming obtained behaviour obtained from the with the experimental load-displkement test on model 4. Is evident, by comparing the theoretical load-displac. ement behaviour of the model structure assuming no imperfections in member 149, (given in Figure 5.36), with the theoretical behaviour of the It buckling load of member model assuming a reduction in the flexural 149 (Figure 5.43), that imperfections occurring in the compression
322

20

15

NO.31 NODE RESULT EXPERIMENTAL RESULT ANALYTICAL

-10 cl 0 -j -0 le CL)

50'

100

150

200

in mm. DispLacement VerticaL

Comparison 5.43. Of Experimental And Theoretical Figure 5Load-Displace ent Behaviour Of Model 4 Assuming An ImpeFfection The Figure Member 149. the lo-aU shows both experimental displacement behaviour behaviour of model 4 and the theoretical buckling load of the of 29% in the flexural assuming a reduction web member number 149. central It is evident by comparing the theoretical load-displacement behaviour in the of the model structure assuming no reductions buckling load of member 149 given in Figure 5.36, with the flexural behaviour of the model assuming a 29% reduction in the theoretical buckling load of member 149, that imperfections flexural occurring in certain have a members in these structures compression will influence behaviour of the structure. on the overall significant

323

have a significant influence members of these structures will on the behaviour the This is of structure. overall an important which should be taken into account when assessing both consideration load and collapse behaviour the ultimate 'compression of chord critical' double-layer grids.

General Comments Table 5.22 gives both the theoretical and experimental values of the ultimate loads supported by each of the four test models in description to a brief of the collapse mode of each addition It is evident from the results given in Table 5.22, and structure. from the experimental work reported in this Chapter, that attempts to improve the ductility of square-on-square double-layer grids, by in the lower chord members, has been tensile permitting yield However, the attempts to improve both the completely successful. double-layer square-on-square within ductility by and the post-elastic of the structure, grids, incorporating soft members into the compression chords, has not been The disappointing so rewarding. results emanating from the two tests on double-layer grid structures incorporating the soft members elastic have been due to both the poor performance of the type 2 and type 3 soft members (Figure 4.11) when subjected to both axial forces and introduced bending moments, and also to imperfections into the structure during fabrication. force distribution

324

4-1 44 0 w a) 0 4-) r.
4-) a)

Ea

44 0

r1 d -! ()

4.) 4.4
a)
4-) C) . 0 r i

0 w

., 110

pq to

44 0

r-I 0 ()

co -4

PQ

P4

$4

P4

::s.

>

4-4 0 4-) 0 0 P-i

H 4-)

0 r-q 0

-4 4-) 4-1 r-l

(D E ,
ri

>-, M 0 4.JM JD 0c

44

4-) 0
F4

4-) COU 5 (d
r-i P4 a E >

1-4 4-)

4-) 0

F-4 4-)

r-I

74

-4

0) rI 44

1--

ro 0-

4 r-q 0 0) 91

(0
C! LO

0) 4

0 Cf) ro > &i

F4

94

=) 4CD

rM

X 9: 4-) Z a) 4J
r-q P

C) C13

00 (n LO 00

CD 4 00 CY)
4

LO
7-4

v) ,V c:
0i to

-C3 =

4-

F=
W

C\j
CU

rulo 10
.4-) CL '453 4(v 0o =

4-)

4-) z

U. 04 CY)

0 41

to (n LO t-

0 C) o N

0) uw -0 fo = cc% -a 14-j (L) t4-)

cr)

325

CHAPTER 6 GENERAL DISCUSSION INTRODUCTION the behaviour studying of square-on-square double-layer space trusses have found that although these structures may be highly statically indeterminate, complete types of structure collapse of certain may occur, under an Imposed load, with the failure of one or a small number The of compression members (Schmidt, et a], 1976; Smith, 1984b). double-layer in this space truss square-on-square studied indeterminate with shown in Figure 5.1 is statically twenty-five However, if the redundancies (Affan, et al, 1986). are designed to have equal compression members in the structure strength, failure of one of the four corner web members or one of chord, mid-side, edge compression members would be A typical to cause a complete collapse of the structure. sufficient has a space trusss compression member used in a double-layer to the transition ratio close slenderness slenderness and the top consequently these members when supporting an increasing compression load buckle plastically If a with a sudden release of energy. is to remain stable after buckling of double-layer truss structure the most heavily stressed compression member, the remaining members in the structure must be capable of absorbing the total energy released from the failed member. If the members surrounding the members are unable to support this sudden increase in load, progressive buckling and collapse of the structure will ensue. The collapse mode of the structure will be similar to the failure mode compression members, where the collapse exhibited by the individual failed is sudden, showing no ductility Supple et al, in the load-displacement response. (1981) along with other investigators have shown the effect which the post-buckling response of the individual the behaviour compression members has on collapse of If compression members which space trusses. square-on-square a gradual load shedding after buckling are incorporated into is introduced into the load-displacement a space truss, ductility response of the structure (Saka, et a], 1984). exhibit investigation Increasing Several Investigators

326

reported by Schmidt, et al, (1979) have shown that some model double-layer space trusses have failed at loads below their theoretical values even when the theoretical analysis buckling loads of the component members has used the actual critical Investigations compression tests undertaken on the individual Schmidt et a], have suggested that this discrepancy members. arising between the experimental and theoretical ultimate loads of force certain model space trusses is a direct result of an initial obtained distribution existing within the structure due to the lack of the members. The real possibility of a significant reduction ultimate load of square-on-square trusses due to the initial fit of members calls into question the magnitude of the true of in the lack of factor fit from

on certain truss of safety against collaps e existing designed to comply with the present codes of practice.

structures

lack of fit of members The problems associated with the initial are I lkely to be present in varying amounts in all statically However, tne magnitude of the detrimental indeterminant structures. force distribution has on a space truss is effect which an initial the position directly related to the configuration of the structure, of the support points, and tne type of imposed load acting on the Square-on-square double-layer space trusses supported at structure. load, possess a uniformly distributed a] I boundary nodes carrying to cause the reserves of strengtn aoove tne load level sufficient Consequently, progressive first compression member to buckle. is most unlikely and adverse initial collapse of these structures from member lack of fit have less force distributions occurring reserve strength index of the structure. effect on the structural This index can be used as a measure of the reserve of strength in the structure, and may be defined as the ratio of the existing to the load applied to the strength of the structure ultimate (Feng, 1988). structure
Analytical Methods to of be assess the ul t ! mate strength and for efficient a space truss is a prerequisite to The ab Ie

The ab 111ty collapse structural behaviour design.

has been written to algorithm which both the collapse load and collapse determine behaviour of space in Chapter 2, is a modified form of the dual load trusses, outlined

327

The proposed by Schmidt, et al, (1980b). method of analysis first behaviour method has been extended to deal with the elastic-plastic by tension members, the 'brittle type' exhibited post-buckling response exhibited by compression members, and changes in geometry deflects. The non-linear member which occur as the structure behaviour has been modelled using several linear approximations to represent the load-displacement response of the individual members. is similar to the approach adopted by other This technique investigators (Collins, 1981; Howells, algorithm uses a greater number of the memberbehaviour than have been used in previous work. This in turn has increased the 1985), however the present linear approximations to model

procedure undertaken to ensure that in the structure conforms with the equilibrium both the structure and of requirements approximation of the member behaviour. these checks are performed interactively

complexity of the checking the deformation of each member and compatibility I ! near ization the

In the present algorith; n, so that the analyst may

paths and reduce to a minimull the possible equilibrium pre-select This has proved to be large amount of computational work required. time consuming for the analyst who may have to try several load paths for the failed members before combinations of alternative is found. To fully automate this checking unique solution which would procedure it would be necessary to develop a rationale load-paths to of certain combinations members and priority give of fulfilling which were considered to give the highest possibility the However, it is and compatability requirements. equilibrium the numerical that the priorities likely used to minimise solution would change computations required to find the correct the failed of each each member new or when member, after linearized to the on the next changed from one region At present, Smith (1984a) Is the only load-displacement path. investigator who has reported an algorithm capable of undertaking Smith's algorithm, complex checking procedure automatically. in Chapter 2, partitions the members into three distinct outlined sets depending on their behaviour, and checks each set in turn to and compatability requirements are met. ensure that equilibrium this
is possible above and described It that the numerical in detail checking procedures mentioned in Chapter 2, could be simplified by

the failure

328

describing

the post-buck] ing behaviour of the compression members, behaviour of the tension members, by continuous and the post-yield A modified form of the Ramberg-Osgood (1943) algebraic functions.

for the description of the tensile equation would be suitable derived by Paris (1954) and behaviour and the expression initially (1981) Supple Collins by and adequately describes the extended flexural post-buckling behaviour These non-Hnear expressions can then be solved iteratively the main set of linear of tubular compression members. individual member behaviour describing in conjunction equations. with the solution of

equilibrium

Methods of Improving Space Truss Behaviour

At behaviour improving work which different present, relatively few published papers on space truss

investigations to to study ways of relate undertaken the load-displacement The response of double-layer grids. has been carried areas. (1978) have obtained model even with space a ratio a ductile trusses load-displacement with out to date has concentrated on five

Schmidt et a], response chords. chord ductile by

constructing They found that of 28 to

weak tensile of compressive to tensile collapse limited the

areas range

1 compressive

chord

This has led to the of behaviour of the truss. (Hanaor, 1979) that every double-layer space truss misconception by buckling The collapse of the compression members. must finally double-layer during truss tested first the space present had a ratio to tensile of compressive chord area of 8 to 1, and collapsed with the progressive rupture of approximately several of the lower chord tension members. investigation

the

space trusses encompassed the removal of selected web members from the structure (Marsh, 1986). The theoretical investigation reported by Marsh shows that removal of certain web members can improve the force

An alternative approach undertaken to introduce load-displacement response of double-layer

ductility

into

the chords and increase the load-carrying within the diagrams The structure. of given in Figure 5.41 and capacity 5.42 show how the axial forces change in both the web and chord occurs in the lower chord members of members as tensile yielding It can be seen from the Figures test models 1 and 2 respectively. that large force changes occur in the web members and this gives an distribution indication of the important role
329

that

web members undertake

would be to extend the preliminary beneficial work undertaken by Marsh and investigate both theoretically the and experimentally load-displacement response of double-layer structures which have structure. both a yielding tensile bottom chord and selected web members removed. have investigated the researchers possibility of improving both the load-displacement response and the ultimate load into carrying capacity of double-layer space trusses by introducing 1979; the structure an initial state of pre-stress (HoInIckI-Szulc, Spillers, et a], 1984; Hanaor, et a], 1986). For this concept to be Other effective it is essential that and compressive pre-stress indeterminate truss structure. force distribution an Initial the correct is imparted This operation combination into the of tensile statically by

In transferring

forces

throughout

the

It

may be complicated

of unknown magnitude existing within the structure due to various imperfections such as the lack of fit Kowa] et al, of members and small var. lations in support positions. (1975) have measured random force distributions occurring within 'zerb stressed web members made models where theoretically were supporting forces approximately equal to 8% of the forces carried by the most heavily strained web members. Even larger force variations initial were measured in the chord members. Hence, carefully any initial detrimental distribution structure. response of have concentrated the double-layer space trusses on modifying behaviour exhibited by typical brittle post-buckling compression (1977) in these Schmidt used structures. members et al, and Marsh (1984) Fard have used excentr ical ly- loaded compression members and in the load-displacement a small amount of ductility space trusses but this has always been response of double-layer in the load-carrying accompanied by a reduction capacity of the structure. to obtain has successfully into compression member behaviour by incorporating However, artificial ductility been introduced into the member Other attempts to improve the load-displacement pre-stress effects introduced arising from resulting into the structure from an existing initial imperfections must negate any random force within the

330

(Schmidt, device force-limiting et a], 1979; Hanaor, et a], a devices have been reviewed in 1980). A wide range of structural Chapter 3 with the aim of assessing their potential use as The ideal pre-requisite force-limiting devices. which a structural device must possess if it perfect elastic-plastic both time and loading sequence. is to be of use as a load I Wter, is a load-displacement response independent of

devices considered in Chapter 3 are Several of the structural load-displacement capable of producing the required elastic-plastic loaded tubes provide response when tested in compression. Laterally load-displacement response exhibiting a smooth post-yield increase in stiffness as the tube crushes (De Runtz, et This mechanism could be adopted Reddy et al, 1979). device by positioning load-limiting a short portion in a box, at the end constrained possibly longitudinal The axis of the protected. to the long itud inal axis of perpendicular load and length of the ensuing operational to suit the design requirements tailored tube diameter and wall thickness. of a gradual a], 1963; to

form a of tube, the member to be

short tube would be the member, and the load plateau could be choice of

by a suitable

An acceptable compressive load-displacement behaviour was also (Johnson, W-frame from deformation a progressive plastic of obtained by the W-frame The load-displacement 1972). response exhibited (Figure 3.15) shows a long, almost constant, load plateau after the by a rapid return in the of the frame as the two long legs of the W-frame meet. stiffness This increase in stiffness occurring after the load plateau would first plastic In a force-limiting device reserve of strength device constructed using the W-frame mechanism. The load-limiting in the form of a circular bellow which would be could be fabricated positioned along the axis of the member, between the member end and provide a useful The actual fabrication process would require to ensure that the formation and rotation consideration hinges would not be affected by welds positioned plastic the node. seams, or by changes in material technique adopted. Both the lateral ly-loaded properties careful of the hinge has formed followed

at the due to the fabrication

tube and the W-frame would produce force-limiting devices which exhibited only one full operational However other devices described in Chapter 3 are capable of cycle. providing an acceptable load-displacement
331

response over many cycles

initially truss load-limiter" The "rolling exhibits a of operation. load-displacement fluctuating response (Figure 3.17) but settles load plateau after two or down to produce a slightly undulating Although it would be three cycles of operation (U. S. Army, 1971). to possible force-limiting form of the torus as a modified use a slightly device in space trusses, the h.lgh initial cost of the (Johnson, device and the difficulties experienced during fabrication et a], 1975) severely limit the possibility. Robinson ideally (1977) suited exhibits to the

damper designed by The extrusion load-displacement characteristics cyclic device. force-limiting a of requirements by pushing or pul I ing a bulged shaft internally lined tube with constricted modification, load-limiting be incorporated into

The damper, which operates through the centre of a lead,

could, with little a space truss to provide a

selected compression members. protecting potential Unfortunately the damper was only tested under a continuous cycle of tension and compression, and no information was presented on the to support a continuously dampers ability applied constant axial load plateau value. Most alloys load smaller than the operational (Polakowski, 1966), from lead creep, al, so et extensively suffer of is most likely that the bulged shaft would move slowly through the lead surround when the damper was supporting a load well below load. Both the creep behaviour of the damper and its the limit load-displacement response at elevated temperatures would require if these devices were to be be used as investigation further it load-limiters in space truss Device structures.

Novel Force-Limiting

has been devoted to A large part of the present investigation the design, testing and improvement of a novel force-limiting The gradual development of both the full device. size and model in Chapter 4. It can be seen from soft members has been outlined the compressive load-strain relationship obtained from a full size (Figure is 4.2) device that the the overall response of member soft is It to the possible significantly alter acceptable. behaviour of the soft member by changing the load-displacement length and cross-sectional The areas of the component parts. initial reduced axial stiffness exhibited by the device is directly

332

tubes used in the soft by the member, and the magnitude of the load plateau is controlled ' area of the middle tube in the yield stress and cross-sectional in the full the model soft members or the four middle strips size related of soft members. The tensile testing of the individual tubes and strips also there is good agreement

to the

lengths

the

individual

reported in Chapter 4 shows that although between the values of yield stress obtained from nominally identical samples, the mean value of yield stress are significantly greater in the the than values min1mu.,, ii given each of relevant This is to be expected as it is necessary for the specifications. steel manufacturers to of the yield probability ensure that there is only a very small below the minimum specified stress falling

However, any increase or variations occurring in the yield stress of the middle tube or strips will have a commanding effect on load of the soft member.and as a result change the the operational value. the collapse behaviour of the space it would To reduce this possibility truss incorporating be necessary to calculate the yield load of the soft member based on The 'target strength' the 'target strength' of the steel material. is usually calculated by the manufacturers by adding to the minimum force distribution and possibly the device. value three times the standard deviation of test obtained from tensile coupon tests undertaken to ensure results adequate quality control of the material. specified yield characteristics of the soft member may also be adversely affected by strain aging of the middle tube or If the soft member is subjected to cyles of loading middle strips. and unloading with the compression cycle of sufficient magnitude to cause the middle member to yield in tension, ' then Increases in the yield stress of the middle member will occur if there is sufficient time in between the load cycles for the middle member to strain age. This will in turn increase the limit-load of the soft member the force distribution the encompassing within and may alter If the soft members are designed so that the load structure. The load-displacement

is supporting working plateau is not reached when the structure in the imposed load acting on the loads then normal fluctuations have any detrimental not will structure effect on the load-displacement behaviour of the soft member.

333

load of both the ful I size and ultimate from several carefully model soft members has been -obtained in Chapter 4. A theoretical tests reported compression controlled tube soft of the model triple analysis members step-by-step An estimate of the that these members should buckle flexurally when the to compression load acting on the 'closed up' member is sufficient in the middle tube. However, the experimental cause instability indicates results buckl Ing analysis may not be entirely obtained from the compression tests indicate that the soft members buckled when the outside tube, and not the middle tube, becomes It was assumed in the theoretical analysis that the unstable. outside tube would be continuously restrained by the middle tube but the end details of the soft member and the small gap existing between the two tubes may have caused this assumption to be invalid. loads of the soft members have been calculated using mean values of the yield stress and the measured bow of the members. However, no allowance has been made for Initial in tube wall thickness or residual stresses occurring variations The theoretical buckling The tube wall thickness from the fabrication procedure. resulting to measure, and any variations occurring along the proved difficult In the tube Variations length of the tube could not be quantified. influence on the buckling load wall thickness do have a significant 5% in A decrease the wall theoretical the of members. soft of thickness of the outside tube of the model soft member, equivalent in thickness of 0.035 mm, would result in a 4.3% to a reduction buckling load of the soft member. decrease in the flexural size and model soft members of both the full involved connecting the tubes together using plug welds. This would in the members which are likely to be induce residual stresses greater in the model soft members due to the larger ratio of weld The fabrication The model soft members are sufficiently area to tube size. have been annealed individually and it would prove by studying the supplement the present investigation flexural buckling small to useful to values of Indicate that the assumptions made in the theoretical correct. - The ulti. mate loads

loads obtained from annealed model soft members.

To fully further member

the failure mechanism of the novel soft is necessary. It would be experimental investigation understand

334

to accurately measure the strain changes in all most beneficial three tubes so that the validity in of the assumptions outlined Chapter 4 could be investigated. It is relatively simple to measure the strain in the outside tube but recording the strain changes in the middle and inner tubes may prove more arduous. It would be in the inner tube of the full possible to measure the strain size soft member by attaching strain gauges to the inside of the tube but it would be difficult to measure strain changes in the middle tube without influencing the load-displacement
With Tensile

response of the member.

Space Truss Structures

Chord Yield,

published work exists reporting the behaviour in the lower of double-layer space truss in which tensile yielding (1978; 1980a) have is Schmidt a], permitted. et members chord tested to collapse three space trusses with weak tensile chords, two load-displacement a ductile response before of which exhibited Relatively In the present investigation two model square-on-square collapse. designed to allow extensive space trusses have been carefully tensile yielding of certain bottom chord members. The load tests were completely successful with each undertaken on both structures structure addition structure analysis It it is ductility before collapse. In exhibiting extensive load-displacement the experimental response of each was accurately modelled by the non-Hnear collapse program. is evident from the present experimental investigation that into the load-displacement to design ductility possible

little

space trusses, and consequently avoid the response of double-layer 'brittle' if collapse behaviour exhibited by some space trusses. double-layer space trusses are to be designed to allow tensile yield lower in further chord members., certain points require In order to ascertain correctly the magnitude of the consideration. imposed load acting on the structure, to which is just sufficient cause yield to occur in some of the tension members, it Is most important that the yield Is accurately stress of the material values of yield stress obtained from test coupons cut from both square and round steel tubes 43C and 50C were significantly higher than the minimu.n Consequently, if yield stress given In BS4360 (1986). known. Experimental forty-two of grades values of it is not

335

possible tensile subsequent stress relevant

or

economic tests

to

obtain the

the

material

yield

stress analysis

from and

coupon design

, of the structure greater

t heoretical

collapse

substantially specification.

must be based on a value of yield than the minimum value given in the

The type of member may also influence the amount of tensile in a space truss structure. To determine yield capacity available tensile the ductility tests of different member types, were undertaken on twenty-one full size square and circular steel hollow The sections of grade 43C and 50C material (Parke et a], 1985). load-displacement relationships obtained from the tests indicated hollow sections of both grades of that although the rectangular deformation and ductility than material exhibited greater plastic ductility in all of the circular members, sufficient was available the tubes tested tensile
the collapse

to

enable prior

them to

be used

in

a space truss

exhibiting
If

yield

to collapse.
of a double-layer then the analysis of* the space truss procedure truss space can be adopted can be

behaviour only

to tensile yield restricted the ultimate to determine The non-Hnear simplified.

capacity elastic-plastic

behaviour

of the tension

linear members can be modelled using only two or three different If strain hardening of ranges depending on the accuracy required. the tension members is ignored then only two separate linear ranges and perfectly are necessary to cover the elastic plastic response of The analysis of the structure may proceed by first ' to identify the most heavily a linear analysis stressed undertaking The imposed load is then increased by an amount tension members. to cause the first sufficient group of tension members to yield. the member. The behaviour both reducing of these members may then axial stiffness to be effectively zero and in modellea addition load by by and the their

applying, opposite

at both of the end nodes of the yielded forces in magnitude to the equivalent

members, equal yield of

must now be checked to members. All of the members in the structure that the ensure that they are not overstressed and in particular yielded structure following simplified. members have been are response. in the conforming Because by the only with two their load-displacement prescribed members in the phases the

yielded

approximated

linear

procedure

analysis program has been considerably The imposed load is incremented again until the next

336

The group of tension members yielded and require a modification. further load increments applied to the can continue with analysis structure until either a compression member becomes overstressed. or tension members have yielded to reduce the structure to a sufficient mechanism. approach has been adopted for a series of preliminary des ign invest ! gat ions undertaken to compare the I imit state des ign of double-layer square-on-square space trusses with the design of in which tensile yield of the lower chord members these structures This A specimen structure was has been permitted (Parke et a], 1984). first designed in the NODUS system (NODUS,1981) to comply with the This structure design rules given in BS 5950 (1985). was then modifled by reducing the size of the bottom chord members and joints in the Nodus system could the that node size available smallest sb The be used throughout the bottom chord of the space truss. structure was analysed as described previously but the analysis was first terminated compression members bec &me when the not overstressed. the strength of these compression members but increasing the thickness by wall not the diameter enhanced was For several compression members the tube tubes. of the individual wall thickness was increased to the maximum value that could be This increasing the the strength of node. without accommodated design constraint nodes was imposed because the cost of the NODUS in a double-layer space structure forms a substantial portion of the total cost of the structural framework. The analysis an enhanced compression of the stiffness matrix of the structure overstressed or singularity The structure was then re-analysed to assess both the had occurred. deflections of the structure and to determine if any of the bottom when either chord members had yielded loading. serviceability
A comparison weight tensile of the two in the order reduction yield to occur in the

Instead,

terminated

was then member became

when the

structure

was supporting

design

approaches

indicated

that

of 10% could be achieved by allowing lower chord members of the structure

limit loading conditions. The design of the state ultimate under in which tensile yield was permitted was normally governed structure limit by the serviceability state requirements and not by the ultimate reductions camber the capacity of in weight could the structure structure. be achieved and by 337 allowing However, by even greater introducing a slight percentage of

into

a small

the

tension

condition. structure

loading members to yield under the serviceability At present, the plastic deformation of any member in a

loading is not permitted in the supporting serviceability (1985) BS5950, design and Draft Eurocode, EC3, codes relevant (1984), detailed but collapse analysis of square-on-square double-layer space trusses in which tensile yield is permitted shows Members which are to be unduly conservative. this restriction is and deform plastically when the structure ' loading will permit a moderate amount of supporting serviceability the structure before the to occur within force re-distribution allowed to yield strain members eventually yielded at the serviceability effects harden. limit Because these state they will members have be prone to the

of strain aging when the structure supports increases In the increase both the yield stress loading. This will serviceability and ultimate tensile strength of the member but should not decrease load capacity of the complete structure.
Soft Members of which both the model soft by

the ultimate

Space Truss Structures The

Incorporating

behaviour collapse post-elastic double-layer space trusses square-on-square members described the with test axial theoretical in Chapter 5 was,

incorporated

in general,

not as predicted

Both structures failed collapse analysis program. In the premature buckling of one of the compression members. while supporting an model 3 it was a soft member which failed force approximately determined controlled from test 56% of the average failure load of these compression machine. in ,a undertaken In model 4, the central web tests

members, displacement compression compression from testing

buckled supporting prematurely a member also while load approximately 29% less than the mean value obtained identical The behaviour of twenty nominally specimens. in general the ultimate load capacity is sensitive initial force

both models 3 and 4 shows that

of space trusses which are "compression chord critical" imperfections in the structure to both and any from the fabrication distribution arising process.

The load-displacement modified, by adjusting to suit the requirements a considerable

response of the soft members can be both the individual tube lengths and areas, However, ascertain

of the encompassing space truss. amount of additional work is required to

338

stiffness optimum values for the initial of the soft members, the magnitude and length of the load plateau, and the ratio of the yield to ultimate load of the member. For the soft member to act as a the magnitude of the load plateau should be less than the flexural buckling load of an adjacent equally stressed To ensure a reasonable compression member in the space truss. before an equally stressed of the soft member yielding probability compression member buckles it would be necessary to set the yield below the buckling load of the load of the soft member significantly remaining members. From a consideration of the individual buckling loads of the type T5 compression members given in Table 5.10 a yield load for the soft member of approximately 75% of the mean value of to ensure the buckling load of the T5 members would be sufficient that only one member in one thousand would fall before the soft memberyielded. members used in both models 3 and 4 had an average yield value of 3690 N which is only 47% of the mean buck] ing load of If the value the other compression members used in the structure., The soft of yield load of the model soft members could have been increased to is likely that a more favourable load-displacement 5890 N it response would have been obtained from both test models.
their member to be able to fulfil primary function duct iI ity in the I oad-disp I acement response of a space of creating length of load plateau in the load-displacement truss, a sufficient For the soft behaviour (1980b) of the soft member must truss, be available. collapse Schmidt have investigated space how the theoretical supported et a], behaviour of

load-limiting

device

a squ are-on- square changes member with

variations

which exhibits Schmidt et al, have reported response. length, equal in magnitude to half the yield, would into be sufficient amount of ductility and increase investigation,

boundary nodes, at all in the plateau length of a compression load-displacement an elastic-plastic that total even a small displacement both plateau of the a limited

member at

to

introduce

the post-elastic response of the structure load of the truss. the theoretical An earlier collapse limited to the study of a plane truss, indicated that behaviour of a plastic displacement a compression member In the load plateau length approximately the It

the

load-displacement

structure would require equal to five times the full capacity of the

truss

at yield of the member if 1968). was to be developed (Stevens,

339

Is not unrealistic to assume that the length of the load plateau from the soft ductility in the member to create required behaviour of a double-layer space truss structure will post-elastic depend extensively the on the configuration of the structure, the loading and positions principal support condition. Consequently, to be able to quantify the minimum length of load in a particular plateau required from a soft member positioned of the collapse behaviour of the a detailed investigation structure, space truss would be required. investigation, The soft members used in this described in Chapter 4, are capable of exhibiting a load plateau in their load-displacement response which has a total length of eight times the displacement However, the low soft member at yield. initial to the axial stiffness of the soft member also contributes by the soft member, resulting in an ductility total exhibited equivalent load plateau length of appproximately eighteen times the total failure. The low initial influences throughout positioned elastic stiffness exhibited by the soft members load-displacement the the response of structure If the soft members are the elastic range of behaviour. in the stucture to replace the most heavily stressed displacement exhibited by the T5 compression members at of the

low axial stiffness compression members, the relatively of the soft allows a redistribution members theoretically of axial forces to occur away from the soft member towards the understressed regions In The analytical investigations the space truss. undertaken on both of forces model structures 3 and 4 indicate that this redistribution away from the soft member does occur, but that the adjacent parallel become overstressed as they carry the compression members rapidly members. To transfer axial forces from the highly stressed regions to the understressed areas in the several soft members will be required In the space truss structure, to 'guide' the redistribution of forces throughout the structure. To allow the soft member to have an adequate margin of safety the flexural buckling load of the against premature buckling, 'closed up' soft member should be substantially greater than the yield load of the soft member. If the soft members are designed to is supporting an approximate increase of yield when the structure 25% in the serviceability loading, then a ratio of yield load to buckling buckling load of 1.7 should be sufficient to ensure that of the soft members does not occur. 340 premature load transferred from the soft

a limited number of soft members are incorporated into a space truss, then to assess the load-displacement behaviour of the full to a collapse, non-linear analysis up of the structure if is However, soft members are positioned required. structure in which across one bay of the top chord of a structure right If yield of the bottom chord members is also permitted, then provided the yield load of the soft members Is approximately equal to the yield load of the lower chord tension members, an ultimate can readily be calculated. moment of resistance for the structure tensile lines' in the members could be positioned to form 'yield load An the carrying capacity of the truss of assessment structure. could be obtained by equating the applied moment acting along the yield line to the ultimate moment of resistance of the structure, the total yield load of all of the soft calculated by multiplying The soft members positioned the structure. General Comments preceding discussion dominant imposed loading controlling In the it has been implied that the along the yield line by the effective depth of

the design of a space truss was load or a syr. imetrically a vertical uniformly-distributed either it is possible Under these load conditions positioned point load. load-displacement design and a ductile to obtain both an efficient response from the structure by allowing tensile yield in the lower incorporating in by the top chord of soft meTbers and members chord However, if a non-symmetric load case governs the the space truss. then a weak tensile design of the structure chord, permitting During the last tensile yielding, may not prove to be so efficient. have been used extensively to roof decade, space truss structures (Bell in buildings Hong Kong Roof structures et al, 1984). major constructed in this region are exposed to extensive wind suction, forces of 2.5 kN/M2. This value of imposed creating average uplift load is significantly greater than the dead weight of the structure be and consequently bottorn chord members in the structure will This member loading will subject to frequent stress reversals. limit the possible advantages to be gained from tensile yield of the lower chord members, but does not preclude the possibility of obtaining a ductile load-displacement response from the structure by incorporating soft members strategically placed in the lower chord of the space truss. .
341

Both symmetric and non-symmetric imposed loading acting on a to cause a non-symmetric collapse of double-layer truss are likely in case of over-loading. This is especially true for the structure, in the space trusses where imperfections compression chord critical compression members will always result identical members, being nominally in one member, in a group of weaker than the others. Consequently, a collapse analysis of these structures should include a series of separate analyses undertaken to determine how sensitive behaviour of-the a group of is to structure nominally identical

both the ultimate load and collapse the premature buckling of one of compression members.

is not so The effect of compression member imperfections in which important in the collapse analysis of space trusses is lower tension the members allowed chord of extensive yielding before the first the test Both of compression member becomes overstressed. 1 and 2 which allowed tensile yield, model structures of symmetric deflections over a substantial proportion load-displacement one response. This indicates that only

exhibited their post-elastic possibly

the structure, analysis of non-linear to corresponding to, each dominant load case, would be sufficient determine the ultimate load capacity of the space truss provided the load-displacement response of the structure was traced throughout yield c ompression member buckled. region
One of investigation of into the device. the bottom adjacent by the the advantages If of is the relatively several chord to of the of a

the

of

tensile

up to
the low, the

the

point

where the
in

first

soft initial soft

member described elastic axial

this

stiffness truss

members were multi-bay

Incorporated space

continuous

structure decreased bending the the

internal

moments occurring

in the vicinity soft members, this is

then high bending columns, supports would be of the internal permitting a redistribution the centre the is height allowed of of of to forces towards

moments and consequently At present, columns the so that support

axial

span. internal

achieved

by reducing structure

the continuous

deflect could

before

becomes effective. soft

also be obtained

supports. of forces the internal

column members for in the required Not only would this result redistribution but would also limit the maximum force occurring within soft member supports.

by using

A similar the internal

result

342

has involved both the development of the soft member and a study of the behaviour of square-on-square space trusses incorporating soft members. This work has now been the behaviour of square-on-diagonal extended to investigate space The present Two model structures have soft members. tested and the construction and testing of already been successfully a large portion of a full size space truss incorporating soft members is at present in progress. trusses incorporating

investigation

343

Af fan,

A. and Calladine, C. R. (1986). Proceedings, Space Grids, Double-Layer Vol. 3,41-48.

Structural Mechanics Of IASS Symposium, Osaka,

Characteristics Of S. T. S., Johnson, W., Lowe, W.T. (1972). Al-Hassanij Loading, Journal Tubes Under Axial Mechanical Inversion Science,, 14 (6), 370-381. Engineering Allens H. G. and Bulson, P. S. (1980). McGraw-Hill Book Co. Ltd. W.E. and Perry, J. Background To Buckli England, ,

Ayrton,

(1866). On Struts, The Engineer, 62,464. I Ohio, U. S. A., Metals Handbook, ed Lyman T, Cleveland, ASTMS, (1948). The American Society For Metals, 401-403. Baird, J. D., Part I., J. D., Part II., (1963a). Strain Aging Of Steel Iron and Steel, May, 186-192. (1963b). Strain Aging Of Steel Iron and Steel, July, 368-374. -A Critical Review,

Baird,

-A

Critical

Review,

Baktash, P., Marsh, C., Wall With Friction 10 (1), 52-59.

Pall, A. (1983). Seismic Tests On A Model Shear Joints. Engineeri Canadian Journal Of Civil

Limit On The Change Of The Elastic And Hardness J. (1886). Bauschinger, Through Extension And Compression, Through And Steels Of Iron And Through Cycling, Mitteilung And Cooling, Heating aus dem der K. Laboratorium Technischen Technische Mechanisch, Hochschule in Munchen, 13,5,31. Bell, Nodus Spaceframe Roof Construction A. J. and Ho, T. Y. (1984). In 3rd International Proceedings, Conference Hong Kong, On Space 1010-1015. H., Guildford, England, Ed. Nooshin, Structures, (1952). Buckling Strength F. Book Co., Inc. McGraw-Hill Standards Institution, BS449, Part Buildings, Standards Grading Of BSI. Institution, Extensometers (1969). London, Of Metal Structuresq New York,

Bleich,

British

2,

Structural BSI.

Use Of

Steelwork

In

British

(1970). Methods For For Tesing Of Metals,

Calibration And BS3846, London,

British

(1971). Methods Standards Institution, Metals, BS18, Parts 2 and 4, London, BSI. Standards Institution, Sections, BS4848, Part

For

Tensile

Testing

Of

British

2,

(1975). Hot-Rolled London, BSI.

Structural

Steel

344

British

(1982). Standards Institution, Seamless And Welded Steel Tubes Mechanical And General Engineering For Automobile, Purposes, BS6323, Part 1, London, BSI. Standards Institution, Engineering And Allied Standards Institution, BS5950, Part Buildings, (1983). Wrought Steels For Mechanical Purposes, BS970, Part 1, London. BSI. (1985). Structural 11 London, BSI. Use Of Steelwork In

British

British

British British

(1985). Material Testing Machines Standards Institution, Equipment, B91610, Part 1, London, BSI. Force Verification (1986). Institution, Specification Standards BS4360, London, BSI. Steels, Structual Analysis And Stability University of Surrey. For

And

Weldable

Nonlinear J. W. (1975). Butterworth, Systems, Ph. D. Thesis, Skeletal C. R. (1973). Calladine, Int. Imperfection,

Of Elastic The Effect

Buckling Of Columns: Inelastic 15,593-604. J. Mech. Sci. Analysis And Design

Of

Yield-Line J. P. (1969). Cannon, J., 6,124-129. AISC Engineering Chajes,

Of Grids,

S. J., G., A., Britvec, and Winter, Sheet Steels. On Structual Cold-Straining 89(ST2), 1-32. Division, Structual

(1963). Effects Of Proceedings, ASCE Load Friction Proceedings

Clark,

Plate R. H., Bradshaw, J. B. (1973). P. J., Bassett, And Potential. Devices - Their Application Control I. C. E., 55(2), 335-352. I. M. (1981). Collapse Analysis PH. D. Thesis, University of Surrey.

Collins,

Of Double-Layer

Grids,

Congr'ds Int. Des Pi6ces Comprim6es. R6sistance A. (1889). Consid6re, iii, Univ. Int. de 1889, Paris, Des Proc4des De Const, Exposition 371. Coppa, A. P. March, (1968). 130-140. New Ways Of Shock Absorbtion, Machine Design,

B. A. (1949). Dislocation Theory Of Yielding A. H., Bilby, Cottrell, Strain Aging Of Iron, Proc. Phys. Soo., 62(A)o 49-62.

And

(1977). Of The Enviroment, Department Of Transport. Department Recommended Standard For Structual Testing Of Steel Practices Models. Transport TRRL Laboratory, and Road Research Supplementary Report, 254. De Runtz, J. A., Hodge, P. G. (1963). Crushing Plates, J. Appl. Mech., 30,391-395. Of A Tube Between Rigid

345

Dianat,

N. (1979).
University

' Elastoplastic
Of Surrey.

Behaviour

Of Flat

Grids,

Ph. D. Thesis,

Dwyer,

T. V. T. J. and Galambos, Proceedings, Beam. Columns, 153-168. (1984). Constructional

(1965). Plastic ASCE, Structural

Behaviour Division,

Of Tubular 91 (ST4),

ECCS.

Procedures, Testing (36). Steelwork, W. J. and Walker, London, Granada

European

Convention

For

Ellinas,

C. P., Supple, Structures, Offshore

A. C. - (1984). Books. Technical

Buckling

Of

Ueber Die Knickfestigkeit F. (1889). Engesser, d. Arch. U. Ing. -Ver. zu Hannover, xxxv, Zeits. (1893). Ueber Die Berechnung Auf F. Engesser, St;ibe Aus Schweissund Flusseisen. Beanspruchter Ing. -U. Arch. -Ver. Wien., 506. Euler, L. (1759). Berlin Berlin Sur La Force De Colonnes, 252. Annge, T. xiii,

Gerader 455.

SfEibe.

Knickfestigkeit d. Oest. Zeits. de

Memoires de lAcadgmie

(1984). CommonUnified Eurocode Commission Of The European Communities, Draft Eurocode No. 3, Luxembourg. Rules For Steel Structures, Ezra, Ezra, A. A. (1968)'. Program For The Exploitation CR-106648. Annual Report, NASA-, Of Unused NASA Patents.

An Assessment Of Energy Absorbing Devices A. A., Fay, R. J. (1972). Dynami Use In Aircraft Impact Situations. For Prospective Eds. Herrmann, G., Perrone, N., Pergamon Response Of Structures, Press, 225-246. Redundancy And Its Effect The Theory Of Structural Vol. 28, No. Design, Computers and Structuresp In Space On 1,

Feng, Y. (1988). Structural 15-24.

(1979). Devices Force Limiting A. Hanaor, Ph. D. Thesis, University of Melbourne.

Trusses,

Space Truss Studies With Force Limiting Hanaor, A., Schmidt, C. (1980). 106 (ST11), 2313-2329. Division, Devices. ASCE J. Structural Levy# R. (1985). A., Imposed Lack Of Fit As A Means Of Hanaor, 1,147-154. Enhancing Space Truss Design, Space Structures, Investigation N. (1986). Of Prestressing Hanaor, A., Levy, R., Rizzuto, In Double-Layer Grids, Ed., Heki, K., Proceedings IASS Symposi And Space Frames, Osaka, Japan, 3,73-80. On Membrane Structures J. (1979). Prestressing Of Truss And Frame Structures, Holnicki-Szulc, Division, ASCE J. Structural 105 (ST3), 601-616.

346

H. A. Howells, Thin-Walled Johnson,

(1985). members, Impact

Collapse Behaviour Of Ph. D. Thesis, University Strength Of Materialsq Analysis Of Load Limiter,

S-pace Trusses of Surrey. London, Arnold.

With

W. (1972).

W. (1973). Johnson, The Device: 15,357-366.

An Elementary Rolling Torus

Absorbing An Energy J. Mech. Int. Sci.

S. R.; Singh, W. 9 Reid, Johnson, Torus Load Limiter, Rolling Kirk,

L. B. (1975). E)Terimental Study 17,603-615. Int. J. Mech. Sci. s

Of The

(1977). Design Of A Metal J. A., Int. Subway System. U. S. Capital R. M. Flexuret (1979). Int. J. Critical Mech. Sci.,

Skinning For Energy Absorber 19,595-602. J. Mech. Sci., Of Round Tubes

The

Korol,

Buckling Strains 21,719-730.

In

Kowal,

W. (1975). An Attempt Of Measurement of Random Z. and Seidel, Proc. 2nd. Forces In Bars Of A Regular Space Structure, Internal Ed. Supple, W.J., University Of Conf. On Space Structures, Intnl. England, 762-766. Surrey, Guildford,
(1770). Des Colonnes. Sur La Figure et de Math. de la Soo. Roy. de Turin. Llelanges de

J. L. Lagrange, Philosophie

Tamrle,
Lev

Mgmoire Sur La Flexion Du Bois E. (1846). T. iv, 1. Ann. des travaux pjb. de Belgiguet

(deuxime

partie),

(1978). Report Of The Engineering Inc. Associates, Zetlin Of Hartford The The Cause Of The Collapse Concerning Investigation 18,1978, June 12,1978. Coliseum Space Truss Roof On January

Lin, Loo,

T. H. (1950). Inelastic 17 (3), 159-172.

Column Buckling,

J.

Aeronautical

Sci.,

The Behaviour Y. C., Upfold, R. W., Banks, D. M. (1977). Of An Earthquake Energy Absorber For Bridge Decks, Sixth Australasian And Materials, Of Structures Conference On The Mechanics New Zealandp 365-371. University Christchurch, of Canterbury,
R. S., Torsional Structural J. R., Trans. Loomis, Buckling Division, R. W. (1980). R. H., Loomis# R. W. 9 LoOMiB. Coliseumf Proceedings, Study Of Hartford AS 106 (ST1), 211-231. Aging And The Yield Point In Steelf

Loomis,

Low,

Gensamer, A. I. M. E.,

M. (1944). 158,207-249.

Lusasy

(1984). Finite Element Stress London, Finite Element Analysis. Z. S. (1965). Steel

Analysis

System

User's

Manual,

Makowski,

Space Stucturest

London,

Michael

Joseph.

347

Z. S. (1987). Makowski, Colloquium Intnl. Ed. Lam, T. T. and 1-38. Science, Marsh,

Proc. Space Structures Buildings, For Sports Buildings, For Sports On Space Structures Applied, Zhilian, Y., Beijing, China# Elsevier

By Member Removall Space Truss Performance Improving C. (1986). IASS Symposium on Membrane Structures K., Proceedings Heki, Ed., and Space Frames, Osaka, Japan, 3,215-220. Using Of Space Trusses Optimisation C. and Fardy M. R. (1984). 3rd. Intnl. Proc. Diagonals, Of Eccentric Behaviour Non-Linear H., University Ed. Nooshin, Conf. On Space Structures, of Surrey, 669-671. England, Guildford,

Marsh,

On A More Advanced Design De Goyet, V. (1986). C., De Ville Massonnet, Recent Research Advances Structures Steel Of Steel Bar Structures, M. N., London, Ed. PavloviC, To Design, Applications And Their 3-20. Science, Applied Elsevier And Experimental Automatic A. (1975). G., Tosto, M., Prete, Mezzina, Behaviour, In Elasto-Plastic For A Model Of Space Grid Analysis Structures, Conference International Second on Space 570-588. England, Surrey, W. J., Guildford, Ed. Supple, A Design Guide For Architects Nodus Space Frames (1981). British Steel Corporation. Tubes Division, Formex Configuration H. (1984). Nooshin, Applied Elsevier England, Engineering, Internal Orowan, E. (1966). And Grube, Rassweiler Paris, P. C. (1954). 43-49. 21 (1), Limit Stresses Elsevier. Design of Processing Science. and Engineers,

In

Structural

And Fatigue

Of Metals,

Amsterdam,

Columns,

J.

Aeronautical

Sci.,

Parke,

H. B. G. A. R., and Walker, 3rd Proc. Grids, Double-Layer H., University Nooshin, Ed. 528-532.

(1984). State A Limit Design Of Conf. Intnl. on Space Structures, Guildford, England, of Surrey,

Parke,

Report D. I. (1985). G. A. R. and Retief, Strips, Hollow Tubes And Loongitudinal 1985. April University of Surrey, K. J. (1971). Directional 6,30 Of Strain Analysis) Effects 181-184.

Tests On Steel Of Tensile Engineering, Dept. Civil

Pascoe,

Of Prestrain

In

Steel,

Journal

N. H., Polakowskil Engineering

E. J. (1966). Strength and Structure and Ripling, U. S. A. Prentice-Hall, Inc. Materials, 0 Of Stress-Strain

Of

Description Ramberg, W., arid Osgood, W. (1943). By Three Parameters, NACA, TN 902.

Curves

348

B. (1967). Experimental Rawlings, Equipment For Impulsive J. Instr. Engrs. Aust., No. 2206,59-66. Structures, Reddy, T. Y., Reid, Tube-Systems 21,187-199. S. R. (1979). Lateral With Side Constraints, Compression Int. J.

Testing

Of

Of Tubes And Mech. Sci. ,

Phenomena Associated With The Crushing Reddy, T. Y., Reid, S. R. (1980). Plateso Of Metal Tubes Between Rigid International Journal Of 16,545. Solids And Structures, Redwood, R. G. (1964). Plates, J. Appl. Reid, Discussion On: Crushing Mech., 31,357-358. Of A Tube Between Rigid

Hardening Effects Of Strain On The S. R., Reddy, T. Y. (1978). International Compression Of Tubes Between Rigid Plates, Lateral 14p 213. Journal Of Solids And Structuresp Laterally Compressed Metal Tubes As Impact Energy S. R. (1983). Jones, N., Wierzbickil T., Structural Ed. Absorbsj 1-43. London, Butterworths, Crashworthiness, (1977). Damper, Sixth Properties Of An Extrusion Conference On The Mechanics Of Structures And Christchurchl New Zealand, University of Canterbury,

Reid,

W. H., Robinson, Australasian Materials. 372-375. Rolfe,

Effect Of State-ofS. T., Haak, R. P., and Grossr J. H. (1968). On The Bauschinger Effect. Journal Of Stress And Yield Criterion Trans. Am. Soc. Mech. Engrs., 90,403-408. Basic Engineering, 4th Edition, L<)ndon,

For Engineers, Metallurgy E. C. (1982). Rollason, Edward Arnold, The Chaucer Press Ltd. Ross, D. A., and Chen, W. F. (1976). Fritz Columns, Lehigh University,

Preliminary Tests On Tubular Engng. Lab. p Report 393.5.

The Effect Of Joints On The Stregth Of Space Saka, T., Heki, K. (1984). Third Conference Trusses, International On Space Structures, Ed. Nooshin, H., Guildfordq Surreys England, 417-422. Saka, Capacity T. 9 Heki, K. (1986). The Load Carrying Of Inclined Square Mesh Grids Constructed By A Bolted Jointing Systems Ed. Heki, K., Proceedings 1ASS Symposium On Membrane Structures And Space Frames, Osakaj Japan, 3,89-96. Design Engineer, Methods For Parallel-Chord 50 (8), 295-302.

L. C. (1972). Alternative Schmidt, Space Trusses, The Structural

Member Buckling Characteristics Schmidt, - L. C. (1976). And Space Truss Behaviour, IASS World Congress on Space Enclosures. Concordia Montreal. University,

349

Schmidt, L. C., Morgant P. R. , arxi Clarkson# Type Strut Buckling, With Brittle ASCE, 102 (ST7)t 1479-1492. Division,

J. A. (1976). Space Trusses Journal of the Structural

B. R. (1977). Of Schmidt, L. C., Morgan, P. R., Coulthard, The Influence On The Inelastic Eccentricity And Continuity Behaviour Of A Space Austlsn. Conf. Mech. Str. Sixth Univs. Truss, and Matls., New Zealandp 274-281. Christchurch, Canterbury,
P. R., L. C., Morgan, Schmidt, and Range, In Inelastic Design 1915-1919. ASCE, 104 (ST12), Clarksont Journal J. A. (1978). of Structural Space Truss Engineering,

L. C., Schmidtl Trusses.

Hanaor, A. (1979). Force Division, ASCE J. Structural

Devices In Limiting 105 (ST5), 939-951.

Space

Schmidt, L. C., Morgan, P. R., O'Meagher, A. J. p and Cogan, K. (1980a). Space Truss, Proc. Instn. Load Behaviour Of A FAl-Scale Ultimate Civ. Engrs., 69 (2), 97-109. A Method For L. C., and Gregg, B. M. (1980b). Schmidt, Ranges International In The Post-Buckling Analysis 15,237-247. Numerical Methods In Engineering, F. R. Shanley, 14,261-267. (1947). Inelastic Column Theory, J. Space Truss Journal For

Aero.

Sc.,

Residual Stresses And Tubular Compression Memberst Shermanj D. (1971). 97 (ST3)9 891-904. Divisions, ASCE Structural Proceedings, Smith, Buckling Strength And C. S., Kirkwood, W., and Swan, J. W. (1979). Members Including Bracing Of Tubular Behaviour Post-Collapse Conference On Behaviour, of Second International Damage Effects, College, London. Paper 70, Imperial Off-Shore Structures, (1980). H. I. Hartford Coliseum Roof Collapse: E. A-. j Epstein, Learned, Sequence And Lessons Civil Collapse Structural Engineering ASCE, 59-62. E. A. (1983). Buckling Of Four Equal-Leg Angle Cruciform. Columns, 439-450. 109 (M), Division, Proceedings, ASCE Structural E. A. (1984a). Space Truss Nonlinear Analysis, Engineering, ASCE, 110 (ST4), 688-705. Structural Journal Of

Smith,

Smith, Smith,

Smith,

In Double-Layer Grid Space Trusses, E. A. (1984b). Ductility Proceedings, 3rd International Conference Space Structures, Ed. Nooshin, H., Guildford, England, 510-515. On The Analysis Of Experimental Observations In Stability, Proc. Roy. Soc. A., 135,601-616.

R. V. (1932). Southwell, Problems Of Elastic

(1973). Space Decks Ltd. Static Load Tests Carried Out On A 20.4M_ 18. OM Space Frame Roof Constructed From 900 Zone Roof Units, Report No. 96.

350

Spillers,

W.R., Levy, With Prestress, 677-687.

R.

(1984). ASCE J.

Truss Design: Two Loading Conditions Structural Engineering, 110 (ST4),

(1968). L. K. Plastic Stevens, Heyman, J., F. A., and Leckie, University Press, 627-646. Supple, W.J. , and Collins, 1. Tubular Struts, Engineering

Design and Engineering

Trussed Plasticit --

Frames, Ed. Cambridge 0 Behaviour Of

(1980). Post-Critical 2,225-229. Structures,

Supple,

(1981). W. J. , and Collins, 1. Limit State Analysis Of Double-Layer Grids, Analysis, Design and Construction of DoubleEd. Z. S. Makowski, Layer Grids, Applied Science Pulishers, London, 93-117.

Tardif,

C. S. (1956). H. P., Ball, The Effect of Teurper-Rolling Strain-Aging Journal Iron and Steel of Low-Carbon Steel, 182,9-19.

on the Inst..,

Investigation Of The Causes Of Hartford Thornton, C. H. (1984). Conference Collapse, Ed. Nooshin, H., Third International Guildford, Surrey, England, 636-641. Structures, Tsuboi,

Coliseum on Space

The Space Frame For The Symbol Zone Y., Kawaguchi, M. (1971). Symposium on Tension Structures Of Expo' 70, Proc. IASS Pacific and Space Frames, Part 11,893-904.
358, Eustis Development

U. S.

(1971). Design Crash Survival Guide, AD 733 Army. U. S. Army, Air Mobility Research Directorate, and Fort Eustis, Virginia, U. S. A. Laboratory, (1948). 62-118. Theory of Limit Desig

J. A. Van den Broek, Sons, New York,

John ,

Wiley

and

Van Laethem, M., de Coen, J., Jiang, D. H., Ranmant, J. P., Rogier,, J. (1975). Of A Double-Layer Grid Space Structure, Stability Second Ed. Supple, W.J., International Conference on Space Structures, Guildfordf Surrey, England, 745-754. P. (1729). van Musschenbroek, Firmorum, Lugduni. von Kgrman, Zeits. Wolf, Introductio ad Cohaerentiam gerader StIgbe. Corporum

T. (1908). Die Knickfestigkeit Leipzig. Jahrg. ix (4), 136.

Physik.

J. P. (1973). Post-Buckled Strength of Large Space Trusses, Journal of the Structural Division, ASCEf 99 (ST7)v 1708-1712.

D. S., and Rebholz,, M. J. (1958). Wolford, Beam And Column Tests Of Welded Steel Tubing With Design Recommendations, Amer. Soc. Testing Material Bulletin 233,45-51. Yeomans, N. F. (1976). Buckling Programme 2D, CIDECT Report Strength 76/40. of Hot-Finished CHS. CIDECT

351

Youngs

T. (1807). A Course of Lectures Mechanical Arts, London, 320-324.

on Natural

Philosophy

and the Cambridge

Young, B. W. (1971). Residual Stresses University) CLJED/C-Struct/TR. 8.

In Hot-Rolled

Sections,

O. C., Valliappan, Zienkiewicz, S. and King, I. P. Analysis Of Rocks As A "No Tension" Material, 18 (1), 56-66.

(1968). Stress Geotechnique,

UNIVERSITY OF Su"RREY LIBRARY

352