Riveted Cold Form Steel Stru Articles Test

Standard and/or project Stage TC
ISO 1051:1999
Rivet shank diameters
90.93 TC 2
ISO 14588:2000
Blind rivets -- Terminology and definitions
90.93 TC 2
ISO 14589:2000
Blind rivets -- Mechanical testing
90.93 TC 2
ISO 15973:2000
Closed end blind rivets with break pull mandrel and protruding head -- AlA/St
90.93 TC 2
ISO 15974:2000
Closed end blind rivets with break pull mandrel and countersunk head -- AlA/St
90.93 TC 2
ISO 15975:2002
Closed end blind rivets with break pull mandrel and protruding head -- AI/AIA
90.93 TC 2
ISO 15976:2002
Closed end blind rivets with break pull mandrel and protruding head -- St/St
90.93 TC 2
ISO 15977:2002
Open end blind rivets with break pull mandrel and protruding head -- AIA/St
90.93 TC 2
ISO 15978:2002
Open end blind rivets with break pull mandrel and countersunk head -- AIA/St
90.93 TC 2
ISO 15979:2002
Open end blind rivets with break pull mandrel and protruding head -- St/St
90.93 TC 2
ISO 15980:2002
Open end blind rivets with break pull mandrel and countersunk head -- St/St
90.93 TC 2
ISO 15981:2002
Open end blind rivets with break pull mandrel and protruding head -- AIA/AIA
90.93 TC 2
ISO 15982:2002
Open end blind rivets with break pull mandrel and countersunk head -- AIA/AIA
90.93 TC 2
ISO 15983:2002
Open end blind rivets with break pull mandrel and protruding head -- A2/A2
90.93 TC 2
ISO 15984:2002 90.93 TC 2
Standard and/or project Stage TC
Open end blind rivets with break pull mandrel and countersunk head -- A2/A2
ISO 16582:2002
Open end blind rivets with break pull mandrel and protruding head -- Cu/St or Cu/Br or Cu/SSt
90.93 TC 2
ISO 16583:2002
Open end blind rivets with break pull mandrel and countersunk head -- Cu/St or Cu/Br or Cu/SSt
90.93 TC 2
ISO 16584:2002
Open end blind rivets with break pull mandrel and protruding head -- NiCu/St or NiCu/SSt
90.93 TC 2
ISO 16585:2002
Closed end blind rivets with pull mandrel and protruding head -- A2/SSt
90.93 TC 2
Mechanical Design
Second edition
Peter R. N. Childs
BSc (Hons), DPhil, CEng, FIMechE, MIED, ILTM, Mem ASME
University of Sussex, UK
Newnes Mechanical
Engineers Pocket Book
Third edition
Roger L. Timings
Module10: Special Topics
Learning Unit-1:
M10.1 Mechanical Testing of Composites
The mechanical testing oI composite structures to obtain parameters such as strength and
stiffiless is a time consuming and oIten diIIicult process. It is, however, an essential process, and
can be somewhat simpliIied by the testing oI simple structures, such as flat coupons. The data
obtained Irom these tests can then be directly related with varying degrees oI simplicity and
accuracy to any structural shape. The test methods outlined in this section merely represent a
small selection available to the composites scientist. Some, such as the tensile coupon test, are
widely recognised as standards, whereas there are dozens oI diIIerent tests Ior the measurement
oI shear properties.
M10.1.1 Tensile Testing
Tensile testing utilizes the classical coupon test geometry as shown below and consists oI two
regions: a central region called the gauge length, within which Iailure is expected to occur, and
the two end regions which are clamped into a grip mechanism connected to a test machine.
Figure M10.1.1 Typical tensile composite test specimen (all dimensions in mm)
These ends are usually tabbed with a material such as aluminum, to protect the specimen Irom
being crushed by the grips. This test specimen can be used Ior longitudinal, transverse, cross-
ply and angle- ply testing. It is a good idea to polish the specimen sides to remove surIace
Ilaws, especially Ior transverse tests.
M10.1.2 Compressive Testing
This is much more problematical. The results obtained are essentially dependent on the type oI
compression Iixture used. Also, the gauge length is conical, as iI it is too long, the specimen will
buckle and flex, resulting in premature Iailure. II it is too short, then the proximity oI the tabs
will adversely aIIect the stress state, resulting in artiIicially high values. The most widely used
compressive test technique is the Celanese fixture, shown below. Cylindrical in design, a small
specimen sits within a set oI trapezoidal grips, encased in collars and an alignment shell. The
gauge length depends on the type oI test material and varies between 12.7mm Ior longitudinal
specimens and 6mm Ior transverse specimens. Again, it is a good idea to tab the specimens.
Figure M10.1.2 Celanese compressive fixture and specimen (all dimensions in mm)
M10.1.3 Intra-laminar Shear Testing
Most oI the numerous shear test methods in existence measure intra-laminar shear properties,
rather than inter-laminar ones. In theory, however, they should be the same in a perIectly
consolidated material. A very popular test is the asymmetric four-point bend test; the essential
Ieatures oI which are shown overleaI. The specimen is 80mm long; 20mm wide and around 3mm
thick, depending on the number oI plies used in the test laminate. Two notches, 4mm deep, are
cut where shown and application oI the load will cause shear Iailure along the notch roots ('axial
splits') Iollowed by wholesale intralaminar shearing in the centre section. Strain gauges are
bonded at in the specimen centre, enabling accurate shear modulus measurement (usually
even in poorly consolidated materials)
45
$
13 12
G G =
Figure M10.1.3 Schematic representation of the asymmetric four-point bend shear fixture
M10.1.4 Inter-laminar Shear Testing
The most common test Ior measuring shear delamination is the short beam shear test shown
below, where a small specimen (30mm long) is loaded in three-point bending until a
delamination Iorms in the centre plane at one end oI the specimen. P`.
Figure M10.1.4 The short beam shear test
Such a test is diIIicult to optimize as Iailure will oIten occur by crushing under the central
bending nose. In this test, the shear strength,
12
t is given as,
12
3
( 10.1.1)
4
=
P
M
bh
t
where P` is the applied load, b` is the specimen width and h` is the specimen thickness.
M10.1.5 (a) Mode-I Fracture Toughness
As a rule, Mode-I delamination in composites is measured using the Double Cantilever Beam
(DCB) test method shown below, rather than the Compact Tension test geometry used Ior
mode-I cracking oI most other materials.
Figure M10.1.5 An in situ Double Cantilever Beam (DCB) test
The specimens are long, thin coupons (typically 150mm x 25mm x 3mm), tabbed at one end with
aluminium hinges or 'T' tabs and with TeIlon insert at the same end in the centre plane oI the
specimen to represent a delamination oI known length. The test is a lengthy aIIair, as the
specimen is loaded and subsequently unloaded many times aIter a small increment oI crack
growth. ReIerring to the load-displacement plot overleaI, crack growth is detected as a decrease
in the trace. Nine load- unload curves are shown, with both the peak load (P) and the
corresponding crack length
( )
1 9
...... a a recorded. From this, basic beam theory can then be used
to calculate a value oI
Ic
G -the mode I Iracture toughness -as a Iunction oI crack length,
according to equation (M10.1.2), as,
2
3
( 10.1.2)
2
=
P C
G M
Ic
aw
where C` is the specimen compliance, given as the inverse gradient oI the loading portion oI
each curve, and w` is the specimen width. This is the basic analytical equation oI which
numerous modiIications exist.
Figure M10.1.6 Typical load-displacement trace for the Double Cantilever Beam (DCB)
test
It is Iound in most composite materials that
Ic
G will vary with crack length quite signiIicantly.
This phenomenon is known as the 'R curve effect', and is a critical design parameter.
M10.1.5 (b) Mode-II Fracture Toughness
Most mode n testing is conducted using the End Notch Flexure (ENF) test. This test uses
identical coupons to the Double Cantilever Beam (DCB) test (minus tabs) and produces shear
delamination under three-point bending, as shown Figure M10.1.7,
Figure M10.1.7 An in situ End Notch Flexure (ENF) test
The specimen is placed under a compressive load, continuing until the crack propagates (again,
detectable by a drop in the load-displacement trace). II propagation occurs at a load , then
c
P
Ic
G
is given by equation (M10.1.3),
2 2
9
( 10.1.3)
2 3
16
=
P a
c
G M
IIc
E w h
f
Unlike the Double Cantilever Beam (DCB) test, only one load cycle is required. As a general
guide, Thermosets will exhibit stable mode-I growth and unstable mode-n growth, whereas the
opposite is true Ior thermoplastics.
Learning Unit-2: M10.2
M10.2 1oining Of Composites
M10.2.1 Introduction
As in metal structures, local reinIorcement is generally required where any hole (or cut-out) is
placed in a structural part. Analogous to metal parts, reinIorcement can be bonded or Iastened to
the structure. Local reinIorcement may also be required in the vicinity oI joints, either bonded or
bolted and in locations where concentrated loads are introduced into the structure.
Joints oIten occur in transitions between major composite parts and a metal Ieature or Iitting. In
aircraIt, such a situation is represented by articulated fittings on control surIaces as well as on
wing and tail components, which require the ability to pivot the element during various stages oI
operation. Tubular elements such as power shaIting oIten use metal end Iittings Ior connection to
power sources or Ior articulation where changes in direction are needed. In addition, assembly oI
the structure Irom its constituent parts will involve either bonded or mechanically Iastened joints
or both.
Joints represent one oI the greatest challenges in the design oI structures in general and in
composite structures in particular. The reason Ior this is that joints entail interruptions oI the
geometry oI the structure and oIten, material discontinuities, which almost always produce
local highly, stressed areas, except Ior certain idealized types oI adhesive joints such as scarI
joints between similar materials.
One oI the important Iactors aIIecting design oI composite materials is the load carrying
capability oI the composite joints. The two commonly used types oI load carrying joints, made oI
composite laminates, are:
- Mechanically Iastened joints.
- Adhesive or bonded joints.
The classiIication oI technological Ieatures as indicated in Figure M10.2.1 is conventional in
appearance, but the diIIerence lies in the content oI joint manuIacturing with due regard Ior the
special properties oI composite materials.
Figure M10.2.1 The classification of technological features
In principle, adhesive joints are structurally more eIIicient than mechanically Iastened joints
because they provide better opportunities Ior eliminating stress concentrations; Example:
advantage can be taken oI ductile response oI the adhesive to reduce stress peaks. Mechanically
fastened joints tend to use the available material ineIIiciency. Sizeable regions exist where the
material near the fastener is nearly unloaded, which must be compensated Ior by regions oI high
stress to achieve a particular required average load. In many cases, however, mechanically
Iastened joints cannot be avoided because oI requirements Ior disassembly oI the joint Ior
replacement oI damaged structure or to achieve access to underlying structure. Adhesive joints
tend to lack structural redundancy and are highly sensitive to manuIacturing deIiciencies
including poor bonding technique, poor Iit oI mating parts and sensitivity oI the adhesive to
temperature and environmental eIIects such as moisture. Assurance oI bond quality has been a
continuing problem in adhesive joints. While non-destructive evaluation techniques (ultrasonic
and X-ray inspection) may reveal gaps in the bond, there is no present technique, which can
guarantee that a bond, which appears to be intact does, in Iact, have adequate load transIer
capability. Thus mechanically Iastened joints tend to be preIerred over bonded construction in
highly critical and saIety related applications such as primary aircraIt structural components,
especially in large commercial transports, since assurance oI the required level oI structural
integrity is easier to be guaranteed in mechanically Iastened assemblies. As a rule, bonded joints
prove to be more eIIicient Ior lightly loaded/non-Ilight critical aircraIt structures whereas
mechanically Iastened joints are more eIIicient Ior highly loaded structures. Bonded construction
tends to be more prevalent in smaller aircraIt.
Figure M10.2.2 oIIers a list oI the most common requirement oI the joint design. It should be
kept in mind that some oI these requirements might become design variables in the course oI the
design process.
Figure M10.2.2 The most common requirement of the joint design
Geometry oI the members being joined, Ior example, could be altered locally to Iacilitate joint
design. Reliability includes an array oI requirements, one oI which is the implication oI joint
Iailure on system perIormance.
Type oI joining to be used requires careIul consideration oI several parameters with a knowledge
oI the service that the joint is expected to provide. Load carrying joints usually have an overlap
conIiguration.
Various advantages and disadvantages oI bonded and mechanically Iastened joints are as
Iollows:
Advantages Disadvantages
Bonded 1oints
- No stress concentration in
adherents.
- StiII connection.
- Excellent fatigue properties.
- No Iretting problems.
- Sealed against corrosion.
- Damage tolerant.
- Small weight penalties.
- Fewer pieces, lower weight,
good load distribution.
- Limits to thickness that can be joined
with simple joint conIiguration.
- Inspection diIIiculty.
- Prone to environmental degradation.
- Requires high level oI process
control.
- Sensitive to peel and through-
thickness stresses.
- Residual stress problems when
joining dissimilar metals.
- Disassembly is impossible without
component damage.
- Requires surIace preparation.
Table M10.2.1 Advantage and disadvantages of bonded fastened joints
M10.2.2 Mechanically Fastened 1oints
The behaviour oI composites in bolted joints diIIers considerably Irom that oI metals. The brittle
nature oI composites necessitates more detailed analysis to quantiIy the level oI various stress
peaks. This is due to the Iact that stress concentrations dictate part static strength to a larger
extent than in metals. As a result, composite joint design is more sensitive to edge distances and
hole spacings than metal joint designs.
Mechanically Iastened joints can be divided into two groups, viz. single row and multi-row
designs. Typical lightly loaded non-critical joints require a single row oI Iasteners. The root joint
oI a wing, or a control surIace, is an example oI a highly loaded joint where the entire load acting
on the aerodynamic surface is distributed into another structure. In such a case, the bolt pattern
design consisting oI several rows distributes the load Ior more eIIicient transIer.
Advantages Disadvantages
Mechanically Fastened 1oints
- Positive connection.
- No thickness limitations.
- Considerable stress
concentration.
- Simple process.
- Simple inspection procedure.
- Simple joint conIiguration.
- Not environmentally sensitive.
- Provides through-thickness
reinIorcement and not sensitive to peel
stresses.
- No residual stress problems.
- No surIace preparation oI component
required.
- Disassembly possible without
component damage.
- High tolerance to repeated loads.
- Relatively compliant
connection.
- Relatively poor Iatigue
properties.
- Hole Iormation may cause
damage to composite.
- Prone to Iretting.
- Prone to corrosion.
- Large weight penalty.
Table M10.2.2 Advantage and disadvantages of mechanically fastened joints
M10.2.2.1 Design Considerations
The behaviour oI mechanically Iastened joints is inIluenced by:
a) Material parameters.
b) ConIigurational parameters.
c) Fastener parameters, e.g.
i. Fastener type (screw, bolt, rivet).
ii. Fastener size.
iii. Clamping Iorce.
iv. Washer size.
v. Hole size.
vi. Tolerance.
The primary design considerations Ior bolted joints include joint strength, Iastener type, local
reinIorcement, joint conIiguration, holes and pre-load: The process begins with the determination
oI a conIiguration Ior the joint. Single lap joints are normally adequate Ior thin laminates (up to
about 5 mm in thickness). Fastener bending and initial bearing Iailure are primary areas oI
concern. Double lap joints are better Ior cyclic loads and generally stronger.
The use oI mechanical Iasteners to join composite structures is bound by certain constraints,
which do not exist in the design oI metallic joints. Care must be taken to select Iasteners that are
appropriate with the type oI composite structures. Special types oI Iasteners are available Ior use
on composites. These Iasteners develop the Iull bearing capability oI the composite without
encountering local failure modes and are not susceptible to corrosion.
Fastener selection usually raises issues requiring decisions concerning laminate reinIorcement,
hole sizes and their location, drilling, Iastener installation and inspection. Table M10.2.3 given
below identiIies various issues and proven design approaches to each issue. The Table indicates
that the complexity oI designing bolted joints arises Irom two primary sources, namely, (a)
composite laminates cannot re-distribute high local loads by yielding and plasticity; (b)
composites are more easily damaged by drilling and Iastener installation than metals.
Issue Approach
Drilling damage. - Closely controlled manuIacturing operations.
- Inspection oI drilled holes.
High local stresses - Larger Iastener diameter.
- Insert (bushing).
- Increased laminate thickness (locally).
Preload relaxation - Larger Iastener head.
- Washers (one or both sides).
- Limit on installation torque.
Countersunk head - Avoid, iI possible.
- Increased laminate thickness (locally).
Damage induced by installation
oI blind Iasteners and drive
rivets
- Specially designed blind rivets.
- VeriIy joint strength with tests.
Table M10.2.3 The various issues and design approaches to each issue
Design oI local reinIorcement oI the laminate to resist local stresses is an important step in the
design oI bolted joint. II reinIorcement is required, a proven approach is to increase laminate
thickness by addition oI plies placed at +45 and 90 to the primary load direction. A quasi-
isotropic laminate provides the best bearing strength in any continuous Iibre composite. Design
oI mechanically Iastened joints has always been guided by the principle that the material being
joined should Iail beIore the Iastener, and this is the practice with composites. The major
structural limitation in designing mechanically Iastened joints is the insuIIicient through-
thickness strength oI the laminates. This has given rise to the term pull-through strength.
Another area oI concern is the bearing stress, which a Iastener applies to the edge oI the hole in a
composite laminate as its axis rotates due to secondary bending oI the joint. This condition can
impose a severe limitation on a joint with limited stiIIness. Further, composite's inability to
support installation stresses oI Iormed Iasteners, such as solid rivets or blind Iasteners poses
another problem. In addition to surface damage, sub-surface damage to the laminate may
occur. For this reason, use oI these types oI Iastener is avoided.
M10.2.2.2 Failure Criteria
As in metallic joints, modes oI Iailure in bolted joints oI advanced composites are as Iollows:
a) Tension or tearing failure related to the net area through the Iastener hole. The narrower
the laminate, the more likely the chances oI tensile Iailure.
b) Shear out Iailure related to the shear areas emanating Irom the hole edge parallel to the
load and determined by the end distance.
c) Bearing failure based on the projected area oI the hole. Determined by the diameter oI
the hole. Bearing strength is greater than the compressive strength oI the composite.
d) Cleavage failure is a mixed mode Iailure involving tension and bending.
The above Iailure modes are shown in Figure M10.2.3 and Figure M10.2.4. However, in
practice, mixed modes oI Iailure oIten occur. The allowable stresses in each oI these modes are a
Iunction oI the Iollowing:
a) Geometry oI the joint including the hole size, plate width and distance oI the hole Irom
the edge oI the plate.
b) The clamping area and pressure.
c) The Iibre orientations ply sequence.
d) The moisture content and exposure temperature.
e) The nature oI stressing, e.g. tension or compression, sustained or cyclic and any out-oI-
plane loads causing bending.
Figure M10.2.3 The failure modes
Accepted design practice is to select edge distances, plate thicknesses, and Iastener diameters so
that oI all the probable Iailures would be the net section and the bearing. It is recommended that
highly loaded structural joints be designed to Iail in a bearing mode to avoid the catastrophic
failures associated with net section Iailures.
The Iailure stresses will depend on the degree of anisotropy at the hole and hence on the local
Iibre orientation. Laminates containing a signiIicant proportion oI +45 Iibres have high shear
strength and low stress concentrations at the hole. ThereIore, they are relatively insensitive to
edge distance.
M10.2.2.3 Fastener Selection
Fastener requirements Ior joining composite structures diIIer Irom those joining metallic
structures. Fastener selection considerations Ior joining composites include corrosion
compatibility, Iastener material, strength, stiIIness, head conIiguration, importance oI clamp-up,
lightning protection, etc.
- Corrosion Compatibility: Neither Iibre glass nor aramid Iibre reinIorced composites
cause corrosion problems when used with most Iastener materials. Composites reinIorced
with carbon Iibres are quite cathodic when used with materials such as aluminium or
cadmium. Presence oI galvanic corrosion between metallic Iasteners and non-metallic
composite laminates has eliminated several commonly used alloys Irom consideration.
Conventional plating materials are also not being used because oI compatibility
problems. The choice oI Iastener materials Ior composite joints has been limited to those
alloys, which do not produce galvanic reactions. The practice Iollowed in aircraIt
industry is to coat the Iasteners with anti-corrosion agent to alleviate galvanic corrosion.
- Fastener material: The materials currently used in design include alloys oI titanium and
certain corrosion resistant stainless steels with aluminium being eliminated. The choice is
obviously governed by the make up oI the composite materials being joined, weight, cost,
and operational environment. Titanium alloy Ti-6Al-4V is the most common alloy used
with carbon Iibre reinIorced composite structures.
- Bolt bending: Due to increased inter-laminar shear between the composite plies, bending
oI the bolt occurs more easily. High modulus and high tensile strength Iastener material is
desired where bending may occur. Susceptibility oI bolt bending in composite structure
introduces higher reaction loads on the Iastener head, which requires more careIul
consideration oI head conIiguration. Bending should also be considered in multiple
component Iasteners such as blind Iasteners. A threaded core bolt resists bending much
better than a smooth bore pull-type blind Iastener as shown in Figure M10.2.5.
Figure M10.2.5 The Comparison of bending resistance between the threaded core bolt and
the smooth bore pull-type blind fastener
- Head configuration: Composites are sensitive to high bearing loads than are metals.
This means Iastener heads should be designed with as much bearing surIace area as
practicable. The larger area improves pull-through and delamination resistance in
composites, while reducing over-turning Iorces Irom bolt bending. Countersunk or
flush head fasteners are Irequently used on exterior surIaces oI the aircraIt where
aerodynamic smoothness is required. Countersunk Iasteners Ior composites include
tension head Iasteners having the large head depths and shear head Iasteners having
smaller head depths with head angles ranging Irom 1000 to 1300 as shown in Figure
M10.2.6.
Figure M10.2.6 Countersunk fasteners for composites with different head depths
- Countersunk fasteners: tend to bear against the surrounding element more unevenly
through the thickness than protruding head Iasteners do. Tension head Iasteners are
generally preIerred over shear head Iasteners due to greater strength against head pull-
through. However, iI the joint element is so thin that the countersunk depth is greater than
70 oI the element thickness, the tendency towards uneven bearing pressure in tension
head Iasteners is too great and shear head Iasteners are recommended in this case.
Caution should be observed in the use oI 1300 countersunk head Iasteners. Although this
type oI Iastener increases the bearing area oI the Iastener and permits it to be used in thin
laminates, pull-through strength can be adversely aIIected. Although; close tolerance Iit
Iasteners are desirable Ior use with composites, interference fit fasteners cannot be used
due to potential delamination oI plies at the Iastener hole.
- Clamp up: When tolerance Iit holes are used, high clamp up appears to be beneIicial Ior
joint strength and Iatigue liIe. The clamping Iorces, however, must be spread out over a
suIIicient area so that the compressive strength oI the resin system is not exceeded and
the composite crushed.
M10.2.3 Bonded 1oints
As stated previously, adhesive joints are capable oI high structural eIIiciency and constitute a
resource Ior structural weight saving because oI the potential Ior elimination oI stress
concentrations which cannot be achieved with mechanically Iastened joints. However, due to
lack oI reliable inspection methods and a requirement Ior close dimensional tolerances in
Iabrication, aircraIt designers have generally avoided bonded construction in primary structure.
In a structural adhesive joint, the load in one component must be transIerred through the
adhesive layer to another component. The eIIiciency with which this can be done depends on the
joint design, the adhesive characteristics and the adhesive/substrate interIace. In order to transIer
the load through adhesive, the substrates (or adherend) are overlapped to place the adhesive in
shear. Figure M10.2.7 shows some typical joint designs Ior adhesively bonded joints.
Figure M10.2.7 The typical joint designs for adhesively bonded joints
In general, adhesive joints are characterized by high stress concentrations in the adhesive layer.
These originate, in the case oI shear stresses, because oI unequal axial straining oI the adherends,
and in case oI peel stresses, because oI eccentricity in the load path. Considerable ductility is
associated with shear response oI typical adhesives, which is beneIicial in minimizing the eIIect
oI shear stress joint strength. The response oI typical adhesives to peel stresses tends to be much
more brittle than that to shear stresses. Reduction oI peel stresses is desirable Ior achieving good
joint perIormance.
The criteria Ior selecting an adhesive must be considered in view oI the joint design. The joint
design must ensure that the adhesive is loaded in shear as Iar as possible. Tension, cleavage and
peel loading as shown in Figure M10.2.8 should be avoided when using adhesives.
Figure M10.2.8 Tension, cleavage and peel loading in adhesives
M10.2.3.1 Design Considerations
The major considerations in the design oI a bonded joint can be grouped into Iive categories.
These include joint strength, environmental resistance, joint geometry, selection oI the adhesive
system and processing. The Iirst step in the process oI adhesive joint is to determine a
dimensional conIiguration, which minimizes tensile and peel stresses. Once this is accomplished,
the next task is to select an adhesive system, which best satisIies static strength, Iatigue liIe and
environmental requirements. The third step is the development oI process speciIications Ior the
joint to include details Ior surIace preparation, curing the joint and maintaining pressure during
cure, iI necessary. The joint strength is typically veriIied analytically or by structural tests or
both. Bonded joint strength vis-a-vis adherends thickness diIIerent types oI joints is represented
in Figure M10.2.9.
From the standpoint oI joint reliability, it is vital to avoid adhesive layer to be the weak link in
the joint. This means that whenever possible, the joint should be designed to ensure that the
adherends Iail beIore the bond layer. This is because Iailure in the adherends is Iibre controlled,
while Iailure in the adhesive is resin dominated and thus subject eIIects oI voids and other
deIects, thickness variations, environmental eIIects, processing variations, deIiciencies in surIace
preparation and other Iactors that are not always adequately controlled. This is a signiIicant
challenge since adhesives are inherently much weaker than the composite or metallic elements
being joined. However, the objective can be accomplished by recognizing the limitations oI the
joint geometry being considered and placing appropriate restrictions on the thickness dimensions
oI the joint Ior the each geometry. In each type oI joint, the adherend thickness may be increased
as an approach to achieve higher load capacity. When the adherends are relatively thin, results oI
stress analysis show that Ior all types oI joints, the stresses in the bond will be small enough to
guarantee that the adherends will reach their load capacity beIore Iailure can occur in the bond.
As the thickness oI adherends increases, the bond stresses become relatively larger until a point
is reached at which bond Iailure occurs at a lower load than that Ior which the adherends Iail.
This leads to the general principle that Ior a given joint type, the adherend thicknesses should be
restricted to an appropriate range relative to the bond layer thickness. Because oI processing
considerations and deIect sensitivity oI the bond material, bond layer thicknesses are generally
limited to a range oI 0.125 to 0.40 mm.
Figure M10.2.9 Bonded joint strength Vs adherends thickness for different types of joints
M10.2.3.2 Failure Criteria
A number oI Iailure modes may occur in bonded composite joints because oI their anisotropic
nature. In the adherends, Iailure can be tensile, inter-laminar or transverse. There may be
cohesive Iailure also, which can occur in the adhesive. Various Iailure modes are shown in
Figure M10.2.10 and Figure M10.2.11.
M10.2.3.3 Effect of 1oint Geometry
Single and double lap joints with uniIormly thick adherends are the least eIIicient joints. These
joints are suitable primarily Ior thin structures with low running loads i.e. load per unit width. OI
these, single lap joints are the least capable because the eccentricity oI this type oI geometry
generates signiIicant bending oI the adherends that magniIies the peel stresses. Peel stresses are
also present in the case oI symmetric double lap and double strap joints and become a limiting
Iactor on joint perIormance when the adherends are relatively thick.
In case oI single lap joint, acceptable eIIiciencies can be achieved provided the overlap-to-
thickness ratio is suIIiciently large, oI the order oI 50:1. As stated above, the eIIiciency oI a
single lap joint is limited by the peel stresses. Peel stresses can be reduced to some extent by
tapering oI adherends (as shown in Figure M10.2.12) but the main method is to use large
overlap-to-thickness ratio. No tapering is needed at ends oI the overlap where the adherends butt
together because the transverse normal stresses at that location is compressive and small in
magnitude.
Figure M10.2.12 Reduction of the peel stresses by overlapping of tapered adherends
For thickness above 1.5 to 2 mm, double lap conIiguration is used to transIer the strength oI the
adherends. The optimum overlap-to-thickness ratio in this case is 30: 1. Peel stresses are also
present in these types oI joints but they are not as severe as the stresses in the case oI single lap
joints.
Stepped and scarf joints are used Ior highly loaded adherends, and also where the thickness is
more than 6.5 mm. ScarI joints are theoretically the most eIIicient joints having the potential Ior
complete elimination oI stress concentrations. However, practical scarI joints may be less
durable because oI a tendency towards creep failure associated with a uniIorm distribution oI
shear stress along the length oI the joint. As a result, scarI joint tends to be used only Ior repairs
oI very thin structures. Stepped joints represent a practical solution oI bonding thick members.
Stepped joints have been extensively used where adherends are subjected to high load intensities.
High loads can be transIerred iI number oI short steps oI small rise (thickness increment) in each
step is used, while maintaining suIIicient overall length oI the joint. Factors inIluencing the
joining strength are number oI steps, length and thickness oI each step.
M10.2.3.4 Behaviour of Composite Adherends
Composite adherends are considerably more aIIected by inter-laminar shear stresses than metals.
ThereIore, there is a signiIicant need to account Ior such eIIects in stress analysis oI adhesively
bonded composites. Transverse shear deIormations oI the adherends have an eIIect analogous to
thickening oI the bond layer and result in a lowering oI both shear and peel stress peaks. In
addition, because the resins used Ior adherend matrices tend to be less ductile than typical
adhesives, and are weakened by stress concentrations due to the presence oI the Iibres, the
limiting element in the joint may be the inter-laminar shear and transverse tensile strengths oI the
adherends rather than the bond strength.
In the case oI single lap joints, bending Iailures oI the adherends may occur because oI high
moments at the ends oI the overlap. For metal adherends, bending Iailures take the Iorm oI
plastic bending and hinge Iormation, while Ior composite adherends the bending Iailures are
brittle in nature. In the case oI double lap joints, peel stress build up in thicker adherends can
cause the types oI inter-laminar Iailures in the adherends.
The eIIect oI the stacking sequence oI the laminates making up the adherends in composite joints
is signiIicant. For example, 900 layers placed adjacent to the bond layer theoretically act largely
as additional thicknesses oI bond material, leading to lower peak stresses, while layers next to
the bond layer give stiIIer adherend response with higher stress peaks. In practice, 900 layers
next to the bond layer tend to weaken the joint due to transverse cracking, as such; advantage
cannot be taken oI the reduced peak stresses.
In contrast with metal adherends, composite adherends are subject to moisture diffusion effects.
ThereIore, response oI the adhesive to moisture may be more signiIicant issue Ior composite
joints.
M10.2.3.5 Effects of Bond Defects
DeIects in adhesive joints, which are oI concern, include surface preparation deficiencies,
voids and porosity, and thickness variations in the bond layer. OI the various deIects, which are
oI interest, surIace preparation deIiciencies are probably oI greatest concern. These are
particularly troublesome because there are no current non-destructive evaluation techniques,
which can detect low interIacial strength between the bond and the adherends.
For joints, which are designed to ensure that the adherends are the critical elements, tolerance to
the presence oI porosity and other types oI deIect is considerable. Porosity is usually associated
with over-thickened areas oI the bond, which tend to occur away Irom the edges oI the joint
where most oI the load transIer takes place. ThereIore, it is a relatively benign effect, especially
iI peel stresses are minimized by adherend tapering. II peel stresses are signiIicant, as in the case
oI over-thick adherends, porosity may grow catastrophically and lead to non-damage-tolerant
joint perIormance. In the case oI bond thickness variations, these usually take place in the Iorm
oI thinning due to excess resin bleed at the joint edges, leading to overstressing oI the adhesive in
the vicinity oI the edges. Inside tapering oI the adherends at the joint edges can be used to
compensate Ior this condition. Bond thicknesses should be limited to ranges oI 0.12-0.24 mm to
prevent signiIicant porosity Irom developing. Common practice involves the use oI Iilm
adhesives containing scrim cloth, some Iorms oI which help to maintain bond thicknesses. It is
also common practice to use mat carriers oI chopped Iibres to prevent a direct path Ior access by
moisture to the interior oI the bond.
M10.2.3.6 Surface Pre-Treatment Prior to Bonding
SurIace pre-treatment requires removal oI contaminants such as oils, mold lubricants or general
dirt. Techniques used Ior surIace pre-treatment are:
1. Peel ply method in which one ply oI Iabric should be installed at the bonding
surIace and removed just prior to bonding thereby exposing the clean bondable
surIace. In this technique, a closely woven nylon or polyester cloth is used as the
outer layer oI the composite during lay-up. This ply is torn or peeled away just
beIore bonding. The basic idea is that the tearing or peeling process Iractures the
resin-matrix coating and exposes a clean, virgin roughened surIace Ior the
bonding process.
2. Abrasion and solvent cleaning to remove abrasion products Iollowed by a
solvent wipe. Abrasion increases the surIace energy oI the surIaces to be bonded
and removes any residual contamination. The abrading operation should be
conducted with care to avoid exposing or rupturing oI the reinIorcing Iibres Irom
the surIace.
A typical cleaning sequence would be to remove the peel ply and then lightly abrade the surIace
with a dry grit blast. AIter grit blasting, any remaining residue on the surIace may be removed
by dry vacuuming or wiping with a clean, dry cheese cloth.
M10.2.3.7 1oint Manufacture
Bonded joints can be made by gluing together pre-cured laminates with a suitable adhesive.
Alternatively, bonded joints can be made by Iorming joints during the manuIacturing process in
which the joint and the laminate are cured at the same time (co-cured).
M10.2.3.8 Adhesive Selection
The selection oI adhesives is based on the strength requirements over the expected service
temperature range and the type oI equipment available Ior bonding. DiIIerent types oI adhesives
are available which provide diIIerent ranges oI adhesive bonding shear and peel strengths at
various service temperatures. Adhesives Ior structural bonding can be categorized into three
main physical Iorms in which they are used - (a) films, (b) pastes and (c) foams. Although Iilms
are easier to handle and provide a more uniIorm bond line thickness than paste adhesives, lack oI
reIrigerated storage equipment sometimes necessitates use oI paste adhesives. Foam adhesives
are used Ior stabilizing and splicing pieces oI honeycomb core. The criteria Ior selection oI
adhesive are as Iollows:
- The adhesive must be compatible with the adherends and able to retain its required
strength when exposed to in-service stresses and environment.
- The joint should be designed to ensure Iailure in one oI the adherends rather than Iailure
within the adhesive bond line.
- Thermal expansion oI dissimilar materials must be considered. Due to large thermal
expansion diIIerence between graphite composite and aluminium, adhesively bonded
joints between these two materials are likely to Iail during cool down Irom elevated
temperature cures as a result oI the thermal stresses induced by their diIIerential
expansion coeIIicients.
- Proper joint design should be used avoiding tension, peel or cleavage loading. II peel
Iorces cannot be avoided, a lower modulus adhesive having high peel strength should be
used.
- SurIace preparation should be conducted careIully, avoiding contamination oI the bond
line with moisture, oil, etc.
- The adhesive should be stored at the recommended temperature.
- Use oI adhesives that evolve volatiles during cure should be avoided.
- The recommended pressure and proper alignment Iixtures should be used. The bonding
pressure should be great enough to ensure that -the adherends are in intimate contact with
each other.
- Traveller coupons should always be made Ior testing.
- The exposed edge oI the bond joint should be protected with an appropriate sealing
compound.
The major advantages oI Iilm adhesives are that they are easier to apply and do not require
mixing equipment. They have more uniIorm viscosity and composition and provide more bond
line thickness uniIormity in a joint than do paste adhesives. The major disadvantages oI Iilm
adhesives are that reIrigeration is required Ior storage. In addition, Iilm adhesives are more
expensive than pastes and require heat and pressure to achieve satisIactory bonds.
Paste adhesives have a long shelI liIe and do not require reIrigeration. However, they have to be
mixed beIore application, which introduces possible human error oI incomplete mixing or
improper weighing. Further, paste adhesives have lower strength properties than Iilm adhesives
especially Ior elevated temperature service.
Foam adhesives are used in honeycomb repair to Iill gaps in splice areas or between edge
members and honeycomb core. In addition, they are used to Iill voids and eliminate moisture
paths through splice areas.
M10.2.4 Test Verification
In addition to joint coupon testing, which is perIormed to obtain baseline data, element testing
should be perIormed to veriIy joint analysis, Iailure mode, and location. This is particularly
important Ior primary connections and where the load transIer is complex. The purpose oI testing
is to obtain assurance that the joint behaves in the predicted manner or where analysis is
inadequate.
The bolted joint element or sub-component tests are usually perIormed at ambient conditions to
Iully characterize load transIer details. Tests at other than ambient conditions are necessary in
cases where the low or elevated temperatures with associated moisture contents substantially
change the load distributions.
M10.2.5 Typical 1oint Designs
A variety oI mechanical, adhesive-mechanical and combined joints are shown in the Iollowing
Table M10.2.3.
Table M10.2.4 A variety of mechanical, adhesive-mechanical and combined joints
M10.3 Environmental Effects on Composites
Composite usage has increased enormously mainly due to the advantages oI lightweight, speciIic
strength and stiIIness, dimensional stability, tailor-ability of properties such as coeIIicient oI
thermal expansion and high thermal conductivity. Environmental eIIects on these properties may
compromise a structure and must be considered during the design process.
This module deals with the major environmental concerns Ior the composite designer, problems
encountered with these environments in the past and some materials or protective systems
eIIectively used.
DiIIerent environmental Iactors along with their eIIect on composites are brieIly discussed in the
subsequent paras.
M10.3.1 Biological Attack
Biological attack on composite materials may consist oI Iungal growth or marine Iouling. Fungal
growth does not appear to be as damaging as the wet conditions that promote growth. Fungicide
has been mixed in with resins to retard this growth. Even though marine organisms will grow on
composite surIaces, mechanical properties do not appear to be aIIected and the Iouling can be
removed by scraping. Composites with graphite Iibres have been used in medical applications Ior
both internal and external purposes. Internal composite structures such as artiIicial joints or
plates Ior bone Iracture support must be bio-compatible or the material may degrade over time.
External composite designs (such as artiIicial limbs or orthotic braces) may experience impact
damage, Ilexural and torsional loading during use.
M10.3.2 Fatigue
Fatigue, either through mechanical loads or acoustic vibrations, can cause crack growth or local
deIect Iormation. Fatigue design depends not only on the load but also on the use temperature
range and amount oI moisture present. Very cold temperatures (below -50C) may increase the
stiIIness oI some composite materials thereby increasing the susceptibility to Iatigue damage.
M10.3.3 Fluids
M10.3.3.1 (a).Moisture
Moisture is present in the operational environment in which a composite is manuIactured and
throughout its useIul liIe. Water acts as a plasticiser when absorbed by the matrix, soItening the
material and reducing some properties oI the laminate. Moisture may also migrate along the
Iibre-matrix interIace thereby aIIecting the adhesion. Moisture in composites reduces matrix
dominated properties such as transverse strength, fracture toughness and impact resistance.
Lowering oI the glass transition temperature may also occur in epoxy and polyimide resins with
an increase in absorbed moisture (as shown in Figure M10.3.1). Debonding can occur due to
Iormation oI discontinuous bubbles and cracking in the matrix. Mechanical properties can be
reduced even Iurther iI heat is present or iI the composite is under-cured or has a large amount
oI voids.
Moisture is absorbed into the composite until a saturation point is reached. This has been
described as a non-Fickian process, meaning the rate oI relaxation in the material due to water
absorption is comparable to the diIIusion rate oI water. As the material properties change, such
as decrease in glass transition temperature, the diffusion process changes. The mechanical
properties degrade in relation to the amount oI moisture absorbed, with no Iurther deterioration
aIter saturation is reached. Strength reductions in polyester laminates have been Iound to be 10-
15 while epoxy resins are less vulnerable.
Figure M10.3.1 The relation between Moisture content and Glass transition temperature
Fibre glass composites with either polyester or epoxy resins have been used extensively in
marine structural applications due to their strength to weight characteristics and resistance to the
marine environment. Glass reinIorcement is preIerred over carbon Iibres due to carbon's
electrical conductivity, which may result in severe dissimilar metals galvanic corrosion with sea
water acting as an electrolyte. This is because carbon along with metallic alloys is in the
electromotive series oI alloys commonly used in aircraIt structures. A galvanic cell can thus be
Iormed in the presence oI moisture or any other electrolyte between carbon and contacting metal.
Carbon, which is the cathodic end oI the series and act as a noble metal, is impervious to
corrosion itselI but will accelerate corrosion in the adjacent less noble metal. Special corrosion
control techniques are employed when CFRP components are placed in contact with aluminium
components in aircraIt assemblies. A Iibre glass/epoxy ply is laid up and cocured with the
carbon/epoxy plies. A Iaying surIace sealant is applied between the two components. Aluminium
parts are anodized, primed and painted prior to assembly (as shown in Figure M10.3.2).
M10.3.3.1 (b) Aircraft fluids
The aircraIt Iluid environment consists oI Iuel, hydraulic Iluid, lubricants, de-icing compounds
and water. Polysulphone has been Iound to be sensitive to phosphate ester based hydraulic Iluids.
Some polymer resins such as PEEK may have lower glass transition temperatures aIter
exposure to Iluids with a high aromatic content. The Iuel-water immersion appeared to be the
most damaging, reducing the tensile strength oI graphite/epoxy and Kevlar composites by 11
and 25 respectively.
Figure M10.3.2 Aluminium parts anodized, primed and painted prior to assembly
M10.3.3.1 (c) Automotive Fluids
The automotive Iluid environment consists oI gasoline, oil, battery acid, brake Iluid, transmission
Iluid and coolant. Most oI the composites in a moist high temperature (150) environment
exhibited micro-cracking. The amount oI moisture absorbed, as measured by weight gain, is
directly related to the change in mechanical properties. Salt water, antiIreeze and gasoline
produce most pronounced eIIects on composites.
M10.3.3.1 (d) Other Fluids
Liquids accidentally spilled on composite surIaces may also aIIect the mechanical properties.
Methylene chloride Iound in paint strippers may cause severe damage to epoxy resins and a
number oI other polymers. Solvents, bases and weak acids at room temperature do not appear to
aIIect graphite/epoxies and Kevlar/epoxies.
M10.3.3.2 (a) Weathering
Warm, moist climate may aIIect the perIormance oI composites. Decrease oI 10-20 in tensile
strength has been noted Ior Iibre glass/polyester and Iibre glass/epoxy where the surIace resin
has been eroded away due to extended weathering. Erosion due to rain, snow or ice impact may
be a problem Ior some aircraIt parts, such as radomes or leading edge parts. Coatings such as
polyurethane may be used to make composite parts more resistant to this type oI erosion. EIIect
oI weathering on composites depends on the type oI material used and whether a protective
coating was intact. Studies have indicated that where the paint was intact, the material retained
more than 90 oI its original strength and 80-90 oI modulus. Where the paint had been eroded
away, the composite retained only 68 oI its original strength.
Composite components are required to qualiIy the moisture tests, which broadly include
condition oI structure beIore and during static and fatigue tests by moisture saturation. Static
tests are carried out Iollowing immersion oI composite parts in Iluids like Iuel, hydraulic Iluids,
cleaning agents, de-icing Iluids, etc.
M10.3.3.2 (b) Effect of Contaminants on Weight and Balance
Fluids absorbed by or otherwise introduced into a structure induce weight gains and may cause
out-oI-balance conditions in Ilight control surIaces. Contamination detected should always be
evacuated, the leakage paths identiIied, repaired and the structure re-sealed.
M10.3.3.2 (c) Effects of Contaminants on Structural Integrity
Dimensional swelling oI the resin matrix generally results Irom exposure to high humidity at
high temperatures, exposure to many aircraIt Iluids, to chemical paint strippers and to a variety
oI common solvents. Absorbed moisture lowers the glass transition temperature oI a laminate
and may be conducive to additional micro cracking within the matrix, which in turn increases the
potential Ior additional moisture absorption. Absorbed chemicals mayor may not aIIect the
structural or mechanical properties oI the composite, but generally render the aIIected part un-
repairable.
M10.3.4 Hail
Hail strike to composite structures leads to impact dall1age. For this purpose, composite
structure having a skin thickness oI 0.8 mm is protected at design stage to withstand a 2-inch
hailstone on the ground with a Iree Iall velocity oI 33 m/s and energy 35 Joules (Figure
M10.3.3). There is no major diIIerence between sandwich and monolithic structures Ior the
same skin thickness as they are equally resistant.
Figure M10.3.3 Impact dall1age due to Hail strike in composite structures
M10.3.5 Foreign Object Damage
This type oI damage is caused due to Ioreign objects striking the surIace oI composites causing
possible localised damage or delamination, etc. It includes ballistic damage, damage Irom sand,
dust, stones and more oIten Irom bird strike. The impact resistant oI composite materials can be
controlled by the choice oI reinIorcement and matrix. The matrix can be altered by addition oI
plasticisers, which increase the strain to Iailure. In addition, radomes and leading edges are
designed to protect the structural parts Irom bird impact damage. Some oI these protections are
shown in Figure M10.3.4.
Figure M10.3.4 The protections of surface of Composite materials by design of radomes
and leading edges
M10.3.6 Electro Magnetic Effects
Electromagnetic pulse (EMP) effects and electromagnetic interference (EMI) effects are
caused by various sources like lightning, precipitation static (p-static) or corona discharge. Its
eIIects can be catastrophic. Non-conductive composites provide little shielding eIIectiveness,
while conductive composites like carbon/epoxy provide varying degree oI protection. Fuel tanks,
electrical equipment, etc. require isolation Irom static discharge. Lightning protection schemes
can sometimes serve the dual purpose oI providing lighting protection and p-static protection.
The objective is to bleed oII the static charge prior to any signiIicant build-up that could cause a
Iire or explosion or precipitate electromagnetic interIerence with the on-board electrical
equipment. The shielding eIIectiveness oI composites can be improved by metal coatings.
M10.3.7 Temperature Effects
Temperature eIIects on composite materials include cryogenic temperatures, elevated
temperatures and thermal cycling between these extremes. Cryogenic temperatures do not
appear to aIIect the mechanical properties oI graphite/epoxies or graphite/polyimides
signiIicantly. Elevated temperatures Ior a prolonged period oI time can seriously aIIect the
properties oI a composite, with even greater eIIect iI moisture is present. Loss oI stiIIness with
temperature and ageing is indicated in Figure M10.3.5. Susceptibility to matrix softening is not
only dependent on the resin but also the lay-up. Temperature eIIects are not limited to the matrix
materials. Extended operation at 350C (660F) and 450C (840F) can cause oxidation oI low
modulus PAN-based fibres and high modulus PAN- or Pitch-based fibres, respectively.
Oxidation resistance can be improved with higher purity Iibres. Thermal cycling conditions are
common Ior a number oI applications, including aircraIt and spacecraIt. Thermal cycling may
induce micro-cracking in some composites thereby resulting in reduction oI compressive and
shear strength.
Figure M10.3.5 The effect of temperature and ageing on stiffness of composite Materials
Protection against temperature eIIects can be achieved at the design stage itselI by:
- Selection oI resin system with high glass transition temperature.
- Potential degradation taken into account in the analysis and Iatigue test.
- Protection against moisture exposure.
M10.3.8 Overheat Conditions
Heat generated by lightning strikes has been known to vaporize matrix resins and create large
areas oI delamination and Iibre Iracturing on composite rudders, ailerons, wing and stabilizer
tips, nose domes and nacelle cowling. When exposed to hot gases over long periods, polymeric
resin binders can become completely destroyed through a process oI thermo-oxidation.
Preventive methods may consist oI application oI heat resistant ablative coatings.
M10.3.9 Effect of Ultra Violet Radiation
Ultraviolet radiation is a band oI light Irom 300 to 4000 A. Ultra-violet radiation may cause
degradation through molecular weight change and cross-linking in the resin system. However,
this damage is generally limited to darkening oI the resin in the surIace layer. Coatings, such as
thermal control tape, have been used to protect composite materials Irom degradation.
M10.3.10 Protective Coatings
When an environmentally resistant composite material cannot be utilized, protection oI the
material through the use oI coatings is necessary. A variety oI coatings have been developed Ior
protecting composites Irom various environments. Standard marine paints, pigmented gel
coatings and polyurethanes have been used to prevent ultraviolet damage and weathering erosion
oI marine composites.
M10.3.11 Hygrothermal Stresses and Strains in a Lamina
Composite materials are generally processed at high temperature and then cooled down to room
temperatures. For polymeric matrix composites, this temperature diIIerence is in the range oI 200
to 300 C, while Ior ceramic matrix composites, it may be as high as 1000 C. Due to mismatch
oI the coeIIicients oI thermal expansion oI the Iibre and matrix, residual stresses result in a
lamina when it is cooled down. Also, it induces expansional strains in the lamina. In addition,
most polymeric matrix composites can absorb or deabsorb moisture. This moisture change leads
to swelling strains and stresses similar to those due to thermal expansion. Laminate where lamina
where lamina are placed at diIIerent angles have residual stresses in each lamina due to diIIering
hygrothermal expansion oI each lamina. The hygrothermal strains are not equal in a lamina in the
longitudinal and transverse directions since the elastic constants and the thermal and moisture
expansion coeIIicients oI the Iiber and matrix are diIIerent. In the Iollowing sections, stress-
strain relationships are developed Ior unidirectional and angle lamina subjected to hygrothermal
loads.
Figure M10.3.6 Maximum normal tensile stresses in the x-direction as a Iunction oI angle oI
lamina using Maximum Stress Iailure theory.
Figure M10.3.7 Maximum normal tensile stress in the x-direction as a Iunction oI angle oI
lamina using Maximum Strain Iailure theory.
lamina using Tsai-Hill Iailure theory.
lamina using Tsai-Hill Iailure theory.
M10.3.11.1 Hygrothermal Stress-Strain Relationships for a Unidirectional Lamina
For a unidirectional lamina, the stress-strain relationship with temperature and moisture
diIIerence gives,
(M10.3.1)
where the subscripts T and C are used to denote temperature and swelling, respectively. Note
that the temperature and moisture changes do not have any shearing strain terms, since no
shearing strains are induced in the material axes. The thermal-induced strains are given by,
(M10.3.2)
where
1
o and
2
o are the longitudinal and transverse coeIIicients oI thermal expansion,
respectively, and is the temperature change. The moisture- induced strains are given by, C A
(M10.3.3)
where
1
| and
2
| are the longitudinal and transverse coeIIicients oI swelling, respectively, and
is the weight oI moisture absorption per unit weight oI the lamina. T A
Equation (M10.3.1) can be inverted to give
(M10.3.4)
M10.3.11.2 Hygrothermal Stress-Strain Relationships for an Angle Lamina
The stress-strain relationship Ior an angle lamina takes the Iollowing Iorm:
(M10.3.5)
Where
(M10.3.6)
and .
(M10.3.7)
The terms
x
o ,
v
o , and
xv
o .y are the coeIIicients oI thermal expansion Ior an angle lamina and
are given in terms oI the coeIIicients oI thermal expansion Ior a unidirectional lamina as
(M10.3.8)
Similarly,
x
| ,
v
| and
xv
| are the coeIIicients oI moisture expansion Ior an angle lamina and are
given in terms oI the coeIIicients oI moisture expansion Ior a unidirectional lamina as
(M10.3.9)
From Equation (M10.3.1), iI there are no constraints placed on a lamina, no mechanical strains
will be induced in it. This also implies, then, no mechanical stresses are induced. But in a
laminate, even iI there are no constraints on the laminate, the diIIerence in the thermal/moisture
expansion coeIIicients oI the various layers induces diIIerent thermal/moisture expansions in
each layer. This diIIerence results in residual stresses and will be explained Iully in section
M10.3.11.4.
M10.3.11.4 HYGROTHERMAL EFFECTS IN A LAMINATE
In Section M10.3.11.3, the hygrothermal strains were calculated Ior an angle and uni-directional
lamina subjected to a temperature change, T A , and moisture content change, C A . As
mentioned, iI the lamina is Iree to expand, no residual mechanical stresses would develop in the
lamina at the macro-mechanical level. How- ever, in a laminate with various plies oI diIIerent
angle or material, each individual lamina is not Iree to deIorm. This results in residual stresses in
the laminate.
M10.3.11.4.1 Hygrothermal Stresses and Strains
Sources oI hygrothermal loads include cooling down Irom processing temperatures, operating
temperatures diIIerent Irom processing temperatures, and humid environment such as an aircraIt
Ilying at high altitudes. Each ply in a laminate gets stressed by the deIormation diIIerences oI
adjacent lamina. Only the strains which are in excess oI or less than the hygrothermal strains in
the unrestricted lamina produce the residual stresses. These strain diIIerences are called
mechanical strains and the stresses caused by them are called mechanical stresses.
The mechanical strains induced by hygrothermal loads alone,
(M10.3.10)
where the superscript 'M' represents the mechanical strains, 'T' stands Ior the Iree expansion
thermal strain, and 'C' reIers to the Iree expansion moisture strains. Using stress-strain
Equation
2 2 2
/ 1/
22
E S o c = = , the hygrothermal stresses in a lamina are then given by
(M10.3.11)
where TC` stands Ior combined thermal and moisture eIIects. Hygrothermal stresses induce zero
resultant Iorces and moments in the laminate in the laminate and hence in the n-ply laminate
shown in Figure M10.3.10,
Figure M10.3.10 Coordinate locations oI plies in a laminate
(M10.3.12)
(M10.3.13)
From Equations (M10.3.11) to (M10.3.13),
(M10.3.14a)
and
(M10.3.14b)
On substituting Equations (M10.3.10) and (M10.3.15), they give
(M10.3.15)
(M10.3.16)
The Iour arrays on the right-hand side oI the above Equations (M10.3.16) and (M10.3.17) are
given by,
(M10.3.17)
(M10.3.18)
(M10.3.19)
(M10.3.20)
The loads in Equations (M10.3.18) to (M10.3.25) are called Iictitious hygrothermal loads and are
known. One can calculate the midplane strains and curvatures by combining Equations
(M10.3.17) and (M10.3.18), which is
(M10.3.21)
Using Equation,
0
0
0
x x x
v v
: :
:
v
:
c c k
c c k
c c k

= +
` ` `

) )
)
One can calculate the global strains in any ply oI the laminate. These global strains are
the actual strains in the laminate. However, it is the diIIerence between the actual strains
and the Iree expansion strains, which results in mechanical stresses. The mechanical
strains in the ply are given by Equation (10.3.10) as
th
k
(M10.3.22)
The mechanical stresses in the ply are then calculated by
th
k
(M10.3.23)
The Iictitious hygrothermal loads represent the loads in Equations (M10.3.17) to (M10.3.20)
which one can apply mechanically to induce the same stresses and strains as by the hygrothermal
load. Hence iI both mechanical and hygrothermal loads are applied, one can add the mechanical
loads to the Iictitious hygrothermal loads to Iind the ply-by-ply stresses and strains in the
laminate, or one can separately apply the mechanical and hygrothermal loads and then add the
resulting stresses and strains Irom the solution oI the two problems.
M10.4 Recycling Of Composite Materials
Recycling is one oI the biggest issues Iacing the composites industry, particularly Ior large-
volume applications. Increasingly stringent environmental regulations are likely to restrict the
use oI composites in Iavour oI materials that can be recycled cost eIIectively. This section deals
with composite matrices that cannot be commercially depolymerised into monomer.
M10.4.1 Categories of Scrap Composites
Scrap composites can be conveniently divided into three categories:
1. Scrap in the Iorm oI offcuts, rejects, sprues etc. arising in the manuIacture oI composite
products. Increasingly, this waste material is used in primary recycling by blending it as
Iiller or reinIorcement with virgin plastic oI the same chemical origin. This route is
subject to the careIul control oI the levels oI contamination in the comminuted composite
and to the deterioration in physical properties which may be caused by repeated thermal
and mechanical processing. Some thermosets have been successIully recycled since the
mid 1980`s without adverse eIIect on quality.
2. Single grades oI contaminated plastic collected Irom consumers or processors may
provide Ieedstock Ior primary or secondary recycling, subject to the Ieasibility oI
contaminant removal.
3. Mixtures oI two or more grades oI composite compounds arise as industrial or consumer
scraps. This category oI scrap poses a substantial problem in composites recycling,
because oI the problems associated with automatic identiIication, separation and
determination and control oI composition. Tertiary recycling is expected to be the more
appropriate route Ior recycling.
M10.4.2 Recycling Methods for: Thermoplastic, Thermoset and Metal Matrix Composites
M10.4.2.1 Thermoplastic Matrix Composites
Thermoplastic composites scrap arising in the Iirst two categories provides Ior more
straightIorward recycling than thermoset composites, principally because the thermoplastic can
be melted. Fibre attrition and degradation oI the matrix polymers lead to reuse applications
with less-demanding physical property requirements. A good example oI large volume
thermoplastic composite recycling is long-glass-Iibre-mat reinIorced polypropylene (GMT). The
oIIcuts Irom GMT sheet used Ior thermoIorming oI products particularly in the manuIacture oI
automotive parts can be subsequently used aIter comminution and used over again as raw
material Ior semi-Iinished sheet. Additionally, the oIIcuts or reject parts can be ground and used
Ior extrusion or Ior injection moulding.
M10.4.2.2 Thermoset Matrix Composites
The recycling oI thermoset composites presents great diIIiculties, centered around the
irreversibility oI cross-linking, the Iibre attrition associated with comminution, and a polymer
content that may be less than 30 oI the total weight. The bulk oI the material is oIten glass-
Iibre reinIorcement or Iiller, including Iire retardants and resin dilutants. In tertiary recycling, the
problem is not just one oI recycling the polymer.
Greater stability in the supply oI scrap is associated with large sources oI standardized composite
scrap arising Irom cooperative industry ventures. For example, ERCOM, consortium oIIour large
European composite manuIacturers in partnership with a number oI leading raw materials
suppliers, shred components manuIactured Irom polyester and vinyl-ester-based sheet and bulk
modelling compounds to a range oI well-deIined particle sizes. The resultant Iibre and powder
Iractions can be used in the production oI new bulk-moulding compound (SMC/BMC)
components, and can also be used as reinIorcing material Ior thermoplastics and other materials.
This type oI initiative is providing Ior large-scale utilization oI composites scrap and the
incentive Ior the development oI more comprehensive secondary and tertiary recycling
operations. Considerable imagination is evident in the research and development work which is
in progress to identiIy proIitable routes Ior the recycling oI composite materials. These include
recycling machinery development, with attention to comminutive procedures, and new product
development that has to overcome the technical, cost and aesthetic advantages oI competing
virgin materials. Recent research relates to proposed radical and potentially very substantial
disposal options Ior composites scrap. Combustion with heat recovery is proposed as a route Ior
utilizing the energy content oI the matrix polymer. The behaviour oI a range oI composites
during combustion, the Iorm oI the ash product and the emissions during combustion has been
investigated systematically.
Two industrial processes that utilize the energy content and the inert materials arising in the ash
have been proposed. In cement manuIacture, thermoset composites may be burned in a cement
kiln to utilize their energy content and the mineral materials utilized in the cement klinker.
Alternatively, polymeric materials Iilled with calcium carbonate may be oI use as a Iuel
substitute and sulphur oxide removing agents in coal-Iired Iluidized bed combustion. Large
volume applications Ior selected recycled composites are being developed as shot-blasting media
Ior selective paint removal, and Ior use as soil conditioners.
Thermoset composite materials represent a large percentage oI composites manuIacturing,
particularly in the automotive sector, and bulk-moulding compound (BMC/SMC's) have been the
subject oI considerable research and success in recycling.
M10.4.2.3 Metal Matrix Composites
The term metal matrix composite (MMC) encompasses a wide range oI materials. Common to
all these is a continuous metallic matrix material and the reinIorcing phase is usually a ceramic.
MMC's are categorized according to the morphology oI the reinIorcement i.e. continuous
fibers, whiskers, or particles, as shown in Figure M10.4.2.3.
Figure M10.4.2.3 Morphology of the Reinforcement
The high formability oI the metallic matrix gives MMC's an advantage over polymeric
materials, and they are oIten available in stock form (e.g. billets, rods and tubes), Irom which
components are shaped and Iormed during secondary operations. Another advantage is that
despite being anisotropic, excellent axial perIormance can be combined with transverse
properties which are more than satisIactory.
M10.4.2.3.1 History
Modem MMC's Iirst appeared in the 1920's with the production oI aluminium/alumina
dispersion hardened systems. The 1950's saw the development oI precipitation hardened
materials and in both these materials, the presence oI small particles impedes dislocation
movement and enhances the toughness oI the base metal. Also, small percentages oI Iiller are
required (15) to obtain this enhancement. Creep is eIIectively suppressed in these materials as
dislocations must climb over the dispersoids by diffusive processes and this result in creep rates
decreasing with increasing dispersoids size.
Evolution continued through the 1970's with the introduction oI dual phase steels, which are
eIIectively particulate MMC's consisting oI up to 20 martensite within a soIt ferrite matrix.
This Iorm oI steel is regarded as the Iorerunner oI modern MMC technology. Fibrous MMC's
were developed in the 1960's and were based on tungsten or boron Iibres embedded within a
copper or aluminium matrix. The morphology oI these materials is similar to that oI polymer
matrix composites however, interest declined as production costs escalated. Nowadays, interest
in titanium matrix composite has provided something oI a renaissance in this Iorm oI material.
The most modem developments in MMC technology have been in the use oI whisker
reinIorcements. The combination oI good transverse properties, low cost, high workability and
signiIicant increases in perIormance over unreinIorced alloys has made tern currently the most
commercially attractive system Ior many diIIerent applications.
M10.4.2.3.2 Fabrication Processes
In modem MMC technology, there are Iour main routes taken in the manuIacture oI an MMC
component:
1. Primary liquid phase processing: This classiIication encompasses techniques such as
squeeze casting and squeeze inIiltration; spray deposition; slurry casting and reactive
processing.
2. Primary solid state processing: This includes powder blending and pressing; diIIusion
bonding oI Ioils and physical vapour deposition.
3. Secondary processing: e.g. extrusion and drawing; rolling; Iorging; isostatic pressing;
superplastic processing and sheet Iorming.
4. Machining and forming processes: such as electrical, mechanical and Iluid-jet cutting,
and joining. These processes are rather complex and outside the scope oI this course.
M10.4.2.3.3 Applications
Perhaps one oI the earliest MMC applications was in the Space Shuttle, in which a structural
component in the cargo bay section Iramework was made Irom a 60 boron monofilament
reinforced aluminium-based composite. Other major applications are based primarily on the
enhanced stiIIness and creep characteristics oI MMC materials e.g. engine components, drive
shaIts, bicycle Irames and cross-booms on yachts.
Figure M10.4.3.2.3.3 (a) Photograph of a satellite boom/waveguide structure fabricated
from an aluminium-carbon fibre MMC. The fibres are aligned parallel to the axis of the
boom.
Figure M10.4.3.2.3.3 (b) Photograph of a rotor brake disc, made from cast Al-
10wtSi / 20vol SiC.
Duralcan
Figure M10.4.3.2.3.3 (b) Photograph of the Stump-jumper M2 bicycle, the frame is made
from 6061 Al-10wt alumina
Duralcan
Proceedings of the 6
th
Asia-Pacific Structural Engineering and Construction Conference
(APSEC 2006), 5 6 September 2006, Kuala Lumpur, Malavsia
A-246
TYPICAL TESTS ON COLD-FORMED STEEL
STRUCTURES
Mahmood Md Tahir, Tan Cher Siang, Shek Poi Ngian
Steel Technology Centre, Universiti Teknologi Malaysia, UTM Skudai, 81310 Johor,
Malaysia.
E-mail: chersianghotmail.com
ABSTRACT: Cold-Iormed steel has been recently brought into Malaysian construction.
It is a steelwork technology that has high potential to be developed in Malaysia, that can
oIIers advantages such as Iast erection, lightweight, clean and easier construction. This
paper reported a series oI research studies carried out in UTM Ior locally produced cold-
Iormed steel sections and rooI truss system. The research work included study on the
member capacities Ior lipped C-section and Hat-section, a Iull-scale test Ior rooI truss
system and the cold-Iormed steel tek-screw connection capacities. All the studies were
based on the requirements oI British Standard BS 5950 Part 5 1998. The actual capacities
oI the proposed sections were ratio Irom 1.09 - 2.21 compared to the design strength. The
Iull-scale experimental test on rooI truss also achieved two times oI the estimated design
capacities. The tek-screw connection capacities in resistance to shear Iorce and pull-out
Iorce ranged 76 - 141 higher than the design requirements. The connection also
withstand 10000 times oI dynamic load that veriIied its perIormance in long-term
serviceability. The results oI the experimental tests on the proposed cold-Iormed steel
section and rooI truss system showed good agreement to the requirements oI BS5950
Part 5 1998.
Keywords - Cold-Iormed steel, member capacities, rooI truss, connection.
1. INTRODUCTION
Industrialized Building System (IBS) has been promoted diligently by CIDB, Malaysia since
Year 2003. Besides reduced dependency on Ioreign labour, the simpliIied construction
solutions oIIer better control oI quality, increased productivity and Iaster completion, less
wastage and cleaner environment. Through industrialization oI construction, huge amount oI
work has been shiIted to the Iactory and leaving the construction sites tidier and saIer
(Sumadi, 2001).
In support oI the ongoing process oI implementation oI IBS in the construction industry,
the research and development have been identiIied to Iocus in the area oI open-building,
lightweight materials, joints and sealants, services, and IT and robotics (Nuruddin, 2003). The
application oI light steel Iraming design using cold-Iormed steel is one oI the developments oI
lightweight material.
Light steel Iraming design is generally based on the use oI standard C or Z shaped steel
sections (see Figure 1) produced by cold rolling Irom strip steel. Cold Iormed sections are
generically diIIerent Irom hot rolled steel sections (e.g. Universal Beams and Universal
Column), which are used in Iabricated steelwork. The steel coil used in cold Iormed sections
is relatively thin, typically 0.5 to 3.2 mm, and is galvanized Ior corrosion protection (Grubb,
2001).
th
A-247
Fig 1. Common shapes for cold formed steel
Cold-Iormed steel sections can be widely used in many sectors oI construction, including
mezzanine Iloors, industrial buildings, commercial buildings and hotels and are gaining
greater acceptance in the residential sector (see Figure 2). However, the application oI light
steel Iraming has not been widely developed in Malaysia, due to the lack oI research study on
the local practice Ior such system. This paper gives a series oI studies on cold-Iormed steel
section that carried in Universiti Teknologi Malaysia (UTM). It summaries the experimental
works on cold-Iormed steel member capacities, rooI truss system and connection capacities.
In the end it provides suggestions Ior Iuture work to enhance the practical use oI light steel
Iraming in Malaysia.
Fig 2 Cold formed steel framing (light steel framing) for residential house in UK
(Adopted from Grubb, 2001)
th
A-248
2. THE BENEFITS OF COLD-FORMED STEEL AND THE LIGHT STEEL
FRAMING
Generally, cold-Iormed steel sections have several advantages over hot rolled steel sections,
timber sections and concrete. The main aspects are listed as Iollows (Rogan, 1998; Thong,
2003):
1. No insect and fungal infection: The problems such as rotten or discomposed due to
insect and Iungal inIection are eliminated; thereIore the material curing and
maintenance costs, which are necessary Ior the timber and concrete construction,
could be eliminated as well.
2. Consistency and accuracy of profile: The nature oI the manuIacturing process: cold
rolling - enables the desired proIile maintained and repeated Ior as long as it is
required, in a very close tolerance. Moreover, the very little tool wears and the cold
rolling process is ideally suited to computerized operation which assists to the
maintenance oI accuracy.
3. Jersatility of profile shape: Almost any desired cross-sectional shape can be
produced by cold rolling, such as T-section, Z-section, Sigma-section etc.
4. It could be pre-galvanized or pre-coated: The steel material may be galvanized or
coated by plastic materials either to enhance its resistance to corrosion or as an
attractive surIace Iinish.
5. Jariety of connection and jointing methods: All conventional methods oI connecting
components, e.g. riveting, bolting, welding, and adhesives are suitable Ior cold-
Iormed section. The tek-screw is concerned in this study since it is Iastest and easiest
way which is available in local.
6. Speedy in construction, and suit for site erection: Generally the steel construction
has eliminated the curing time which is inevitable in concrete construction; thereIore
it is Iar Iaster than concrete construction. The cold Iormed steel may more
advantageous than the hot rolled steel since it can be cut and erected with very light
machine and even only man power.
7. Increase in yield strength due to cold forming: The cold Iorming process introduces
local work hardening in the strip being Iormed in the vicinity oI the Iormed corners.
This local work hardening may results an increment oI ultimate yield strength about
25 Irom its virgin strength.
8. Minimization of material: Since the material used can be very thin in comparison to
the lower thickness limits oI hot rolled steel sections, it allows the material usage Ior
a given strength or stiIIness requirement to be much less than that oI the smallest hot
rolled sections. The material thickness, or even the cross-sectional geometries, could
be controlled to achieve the structural Ieatures with minimum material weight.
9. High profitably: In cold rolled process, the manuIacturing costs oI cold rolled steel
section, mainly involve the initial modal oI purchasing the rolling machine and the
costs oI steel strip material later. The machinery cost only expensed once then it
could be covered back in the continuous production. The cold Iormed steel rooI truss
system, which is mainly interested by local industries, is normally short in
th
A-249
construction, and involve only some light erection tools. ThereIore, the investment
required is not high and the return back is sooner than most oI the constructional part.
3. EXPERIMENTAL TESTS ON COLD-FORMED STEEL IN UTM
The Steel Technology Centre (STC), Universiti Teknologi Malaysia has carried a series oI
analytical and experimental research on cold-Iormed steel structures. The studies include:
x Member capacities oI cold-Iormed lipped C-section and Hat-section
x Full-scale cold-Iormed rooI truss system
x Tek-screw connection capacities Ior UNI-Interlocked rooI truss system.
The research works is discussed separately in the Iollowing sub titles.
3.1 Study on the Member Capacities of Cold-formed Lipped C-section and Hat-section
A research was carried by Steel Technology Centre on cold-Iormed steel Lipped C-section
and Hat-section (Thong, 2003; Tahir, 2005). It aimed to provide a complete design oI the
proposed Lipped C-section and hat-section (see Figure 3), which is anticipated to be applied
as members oI conventional school rooI truss Ior JKR.
These sections have been tested Ior their material yield strength, tension strength,
compression strength Ior both short and slender member and bending capacity. The
connection capacity tests included pull-out test and shear test. The capacities estimation and
tests are done in accordance to BS5950 Part 5: 1987 (BSI, 1987; Chung, 1993). The
experimental layout was shown in Figure 4, and the summary oI the results was given in
Table 1.
Fig 3 Cold-formed Lipped C-section (Left) and Hat-section (Right)
th
A-250
(a) Coupon test (b) Tension test (c) Compression test (short
member)
(d) Compression test (slender
member)
(e) Bending test (I) Connection pull-out test
(g) Connection shear test
Fig 4 Experimental Test Lavout for the Member Capacities of the Proposed Cold-formed
Section
th
A-251
Table 1 Comparison of the experimental test results to the design calculation for member
capacities
Type of Test Type of Specimen Design
Strength
Experimental
Result
Ratio of
Exp/Des`
Coupon test 0.6mm plate
1.0mm plate
300 N/mm
2
250 N/mm
2
344 N/mm
2
300 N/mm
2
1.15
1.20
Tension test Lipped C 804038 20.66 kN 26.0 kN 1.26
Compression test
Ior short member
Lipped C 804038 33.70 kN 38.0 kN 1.13
Compression test
Ior slender
member
Lipped C 804038 22.84 kN 25.0 kN 1.09
Bending test Hat section
Lipped C 804038
0.26 kNm
0.97 kNm
0.38 kNm
1.50 kNm
1.46
1.55
Connection pull-
out test
1 screw - Lipped C
2 screws - Lipped C
1 screw - Hat
2 screws - Hat
0.73 kN
1.46 kN
0.53 kN
1.06 kN
1.50 kN
3.20 kN
1.00 kN
2.20 kN
2.05
2.19
1.89
2.08
Connection shear
test
2 screws - 0.6mm plate
1.90 kN
3.80 kN
3.40 kN
6.80 kN
3.00 kN
7.00 kN
7.50 kN
12.50 kN
1.58
1.84
2.21
1.84
* Exp/Des Experimental Result / Design Strength
The test results showed that actual resistance oI the cold-Iormed steel were higher than
the design strength, ranged 1.09 - 1.55 Ior section capacities, and 1.58 - 2.21 Ior connection
capacities. The study concluded that the proposed cold-Iormed steel section give good
agreement compared to the design strength. It is suggested that a Iull-scale test on the
proposed rooI truss to be carried out, in order to gain Iurther understanding on the global truss
behaviour and to build conIidence in utilisation oI the system.
3.2 Full-scale Test on the Cold-formed Roof Truss System
Lotez Engineering has proposed the Mega Truss Svstem which is intended to be laid out,
Iabricated and handled like wood trusses, except that they are built using screw guns rather
than plated methods. A Iull scale testing Ior the plane truss system was carried out at
Laboratory oI Structure and Material, UTM, Malaysia under the supervision oI Steel
Technology Centre (Tahir, 2004). The objectives oI the test are:
1. To conIorm the adequacy oI intended use oI the proposed system.
2. To deIine the actual behaviour oI the truss system in comparison oI theoretical
estimation.
3. To Iit the requirements claimed by Public Works Department oI Malaysia, Ior the
application in conventional school buildings.
The proposed truss is spanned over 7.8m over two supports; and cantilevered at a span oI
2.3m Irom each end. The truss height Irom the support is 2.078m. The cold-Iormed steel
section in used was lipped C-channel. More speciIication oI the truss system and the section
properties is given in Table 2 and Table 3.
The rooI truss was designed based on BS 5950-1990-Part 5 (BSI, 1998). The design
loading were reIerred to BS 6399 (BSI, 1988). The loadings were to be transIerred Irom the
th
A-252
purlin to the truss` span at 600mm spacing, as shown in Figure 5. From the design analysis, it
was estimated that each point load transIerred Irom the purlin should not less than 8.4kN.
Table 2 Specification of the Roof Truss Svstem
Section C - Lipped Channel (MC 10016),
102mm u 51mm u 1.6mm thick
Material High Tensile Galvanized Steel
Yield Stress 450 N/mm
2
Zinc Coating oI Z 275
Connection Class 2 - SelI Tapping Screws
Truss Spacing 1.5m
RooI Pitch 25
o
Table 3 Section properties of the lipped C-channel and purlin
Section Design
strength,
p
v
(N/mm
2
)
EIIective
aea, A
eff
(mm
2
)
Second
moment
oI area, I
(cm
4
)
Elastic
modulus,
Z (cm
3
)
Moment
capacity,
M
c
(kNm)
Tension
capacities,
P
T
(kN)
Compression
capacity, P
c
(kN)
Lipped
C
450 297.2 146.35 11.2 4.92 148.05 133.74
Purlin 550 77.63 1.16 0.6 3.22 45.1 42.7
th
A-253
LC 1
CH 5
CH 6
CH 7
CH 8
CH 13
CH 10
CH 11
CH 15
CH 14
LC 2
CH 12
LC 3
LC 4
Hydraulic
jack and load
cell set here
Stopper
Support Support
Load
distributor
CH9
(a) Load configuration of the experimental test
(b) Experimental Lavout
Fig 5 Full-scale test of the roof truss svstem
th
A-254
Fig 6 Full-scale test results
The Iull-scale test on the truss system was carried out successIully and was proven to be
valid by computer modelling using MultiFrame 3D. From the study, some conclusions can be
made:
1. All the analysis models are valid to represent the actual behaviour oI the truss.
2. The loading condition in the test is worse than the designed condition. ThereIore iI
the truss able to resist the test condition, it should be more than enough to resist the
design condition.
3. The truss is able to resist a loading oI 16.8kN at each loaded point (or total load oI
67.2kN) in the test condition. It is two times higher than the design load. ThereIore it
can be concluded that the capacity oI truss is very saIe to be implemented.
4. There are no visible Iailure observed during the yielding oI the truss, thereIore the
components are expected have a ductile behaviour. The ductile behaviour is
Iavourable in term oI saIety since it delays the signiIicant collapse oI the structure.
From the Iacts stated above, it is concluded that the truss is having the adequate strength
to resist the design loads, and it is saIe to be implemented in the intended uses.
3.3 Connection Tests for UNI-Interlocked Roof Truss System
Three sets oI connection tests were perIormed on the product oI Tong Yong Sdn. Bhd. - cold
Iormed sections oI UNI-Interlocked steel rooI truss system at Laboratory oI Structural and
Material, UTM, on April 2004 (Thong, 2004). The main purpose oI the tests was to veriIy the
validity oI design method provided by BS 5950 Part 5. Code of practice for design of cold
formed thin gauge sections (BSI, 1998) in estimating the actual behaviour oI the connection
in the truss system. One oI the tests - the Iastener dynamic test was aimed to show the
durability oI the connections under long term dynamic loads. The other two tests were shear
test and pull-out test.
All specimens were 0.55mm thick. The dimension oI the specimens were shown in Figure
7. They were tested by using DARTEC MACHINE with maximum capacity oI 250kN. The
loading sequence used was in accordance with European Recommendations (ECCS, 1983).
th
A-255
The European Recommendations stated that the rate oI loading shall not exceed 1kN/min and
the rate oI straining shall not exceed 1mm/min during the test.
As Ior Iastener dynamic test, the connection oI rooI truss was inherently subjected to both
the dead load and dynamic load especially the vibration oI Ian. According to Rhodes in
Design of Cold Formed Steel Members (1991), it was proven that the strength oI shear
connections were rarely aIIected by less 10000 cycles oI repeated loading. In this test, 10,000
cycles oI cyclic loading (varied between 0 to +6kN, by 0.5Hz Irequency) was imposed onto
the connection. The test conIiguration is shown in Iigure 8.
(a) Shear Test (b) Fastener Dynamic Test
(C) Pull-out Test
Fig 7 Connection tests on cold-formed steel section with thickness of 0.55mm
Screw
Screws
All thickness 0.55mm
220 mm
30 mm 75 mm
300 mm
Screws
20mm
15mm
50mm
th
A-256
Fig 8 Configuration of fastener dvnamic load test and the example of collected data
Table 4 Summarv of experimental test results
Type of Test Type of Specimen Experimental
Result (per
Unit Capacity)
Average
Value of
Exp. Result
Design
Strength
Typical model - 2
screws
Typical model - 4
screws
3.43 kN
3.00 kN
3.22 kN
Connection shear
test
On-site model - 4
screws
3.47 kN 3.47 kN
1.44 kN
Connection pull-
out test
1 layer 1 screw
1 layer 2 screws
2 layers 1 screw
2 layers 2 screws
1.40 kN
1.33 kN
1.91 kN
1.55 kN
1.55 kN 0.88kN
Fastener dynamic
test
-
The connections were able to sustain its
strength aIter 10000 cycles oI cyclic loading.
The shear test average result is 3.22kN Ior typical model and 3.47kN Ior on-site model.
These values are 124 - 141 greater than the design strength. Connection pull-out test
average value is 1.55kN, which is 76 greater than their design strength. As Ior dynamic test,
the connections were able to sustain its strength aIter 10000 cycles oI cyclic loading, thus its
perIormance is proven Ior long-term service.
4. CONCLUSION
A series oI experimental tests on cold-Iormed steel members and structures has been carried
out successIully in UTM. The result oI the tests can be concluded as Iollow:
1. For experiment study on the member capacities oI cold-Iormed lipped C-section and
Hat-section, all tests result is higher than the design value. The ratio oI diIIerence lay
between 1.09 and 2.21. The test proven that the proposed sections are suitable Ior the
design oI rooI truss system.
th
A-257
2. For the Iull-scale test oI rooI truss system, the truss was able to sustain the total load
up to 67.2kN (or point loads oI 16.8kN) which is two time higher than the predicted
capacity (8.4kN).
3. For the connection tests, the actual strength oI the connection in resisting shear Iorce
and pull-out Iorce is greater than the design strength. The percentage oI diIIerence is
76 - 141. The connection was able to withstand 10000 cycles load, which also
proven its perIormance Ior long-term serviceability.
Currently most oI the laboratory research and construction practice oI cold-Iormed steel
in Malaysia were Iocus to rooI truss system. It is suggested that studies on the member
capacities oI thicker cold-Iormed sections (range 1.0mm to 3.2mm) can be carried out in order
to widen the application area oI steel structures, such as wall stub, Iloor batten, beam and
column. The studies on the partial strength connection Ior beam-column cold-Iormed steel
structure is also proposed, that it may leads to optimization oI the structural robustness and
steel weight saving. The cold-Iormed steel sections can be Iurther applied into sub-structures
such as staircases, Ioundation system, retaining walls and scaIIolding.
5. ACKNOWLEDGEMENT
The works reported herein were undertaken in association with Tong Yong Sdn. Bhd and
Lotez Engineering Sdn. Bhd. The technical and laboratory works was mainly carried out by
Mr Thong Chin Mun (master student oI UTM). The technical, Iinancial and material
contributions Irom the above parties are grateIully acknowledged.
6. REFERENCES
British Standard Institution (BSI), 1987, BS5950 Part 5. Code of Practice for Design of Cold
Formed Thin Gauge Sections, British Standard Institution, UK.
British Standard Institution (BSI), 1988, BS6399 Part 3. Code of Practice for Imposed Roof
Loads, British Standard Institution, UK.
British Standard Institution (BSI), 1998, BS5950 Part 5. Code of Practice for Design of Cold
Formed Thin Gauge Sections, British Standard Institution, UK.
Chung, K. F., 1993, Building design using cold formed steel sections. Worked examples, The
Steel Construction Institute, UK.
European Convention Ior Constructional Steelwork (ECCS) , 1983, Publication No. 21
(1983): European recommendations Ior steel construction: the design and testing oI
connections in steel sheeting and sections, European Convention Ior Constructional
Steelwork, UK.
Grubb, P.J., Gorgolewski, M.T., and Lawson, R.M., 2001, Building Design using Cold
Formed Steel Sections. Light Steel Framing in Residential Construction, The Steel
Construction Institute, UK.
Lawson, R. M., Chung, K.F. and Popo-Ola, S. O., 2002, Building Design using Cold Formed
Steel Sections. Structural Design to BS5950-5. 1998 Section Properties and Load Tables,
The Steel Construction Institute, UK.
Rhodes, J., 1991, Design of Cold Formed Steel Members, Elsevier Science Publisher, UK:
London.
Rogan, A. L. and Lawson, R. M., 1998, Jalue and Benefit Assessment of Light Steel Framing
in Housing, The Steel Construction Institute, UK.
Nuruddin, M. M., 2003, Research and Development (R&D) for IBS, in Buletin Bulanan
Jurutera Bil 2003 No 6., Institute oI Engineers, Malaysia.
Sumadi et al. 2001, Promotion Strategies and Future Research & Development Needs on IBS,
National Seminar on IBS, CIDB, 17th Sept 2001.
th
A-258
Tahir, M.M., Saad, S., Saleh, A.L., Tan, C.K., 2003, Full-scale Testing for Roof Truss Svstem
Using Cold-formed Steel Sections, in 2nd International ConIerence on Construction in the
21th Century, Korea.
Tahir, M.M., Thong, C.M. and Tan, C.S., 2005, Cold-formed Steel Section Developed for
Roof Truss Svstem, Jurnal Teknologi Vol 42(B), Pernebit UTM, Malaysia, pp11-28.
Thong, C.M., 2003, Development of New Cold-formed Steel Sections for Roof Truss Svstem,
Master Thesis, UTM.
Thong, C.M., 2004, Test Report of UNI-Interlocked Roof Truss Svstem (Connection),
Technical Report Submitted to UTM.
Trebilcock, P. J., 1994, Building Design Using Cold Formed Steel Sections. An Architects
Guide, The Steel Construction Institute, UK.
length of 230mm. Specify a suitable spring for
this application.
11.4 A helical spring is required to exert a force of
2000N at a length of 200mm and 1500N at a
length of 250mm. Specify a suitable spring for
this application.
11.5 A Belleville spring is required to give a
constant force of 50 5N over a deflection
of 0.2 mm.The spring must fit within a 40 mm
diameter hole. A carbon spring steel with
uts
1700MPa has been proposed.

11.6 A Belleville spring is required to give a constant
force of 10 1N over a deflection of 0.15mm.
The spring must fit within a 16mm diameter
hole. A carbon spring steel with
uts
1700MPa has been proposed.

11.7 A helical spring is required for a pogo stick to
exert a force of 975N at a length of 140mm and
300N at a length of 220mm. Specify a suitable
spring for this application if the spring material
is ASTM A232.
Springs
250
Learning objectives achievement
Can you specify the principal parameters for a helical Section 11.2
compression spring?
Can you specify the principal parameters for a helical Section 11.3
extension spring?
Can you determine the principal dimensions of a Belleville spring Section 11.6
washer for a given duty?
d
251
12
FASTENING AND POWER SCREWS
A fastener is a device used to connect or join
two or more components. Traditional forms of
fastening include nuts, bolts, screws and rivets. In
addition, welding and adhesives can be used to
form permanent joins between components. The
aim of this chapter is to introduce a wide selec-
tion of fastening techniques. Whilst considering
threaded fasteners the subject of power screws,
which are used for converting rotary motion to
linear motion, is also considered here.
LEARNING OBJECTIVES
At the end of this chapter you should be able to
determine the:
preload and tightening torque for a threaded

fastener;
lifting and lowering torques for a power

screw;
safe load for a variety of riveted joints;
thickness or overlap for simple adhesive lap

joint.
12.1 Introduction to
permanent and
non-permanent fastening
The joining of components is a frequent necessity
in the design of products. For example, the Boeing
747 has over 2.5 million fasteners. Fastening tech-
niques can also be a major feature in design. Cur-
rent styling for automotive vehicles dictates an
absence of the means of fastening components
together under a cursory inspection.The range of
fastening techniques is extensive including adhe-
sives, welding, brazing, soldering, threaded and
unthreaded fasteners, special purpose fasteners
and friction joints. Some of these techniques are
permanent in nature and some allow the joint to
be dismantled. A variety of basic types of join is
illustrated in Figure 12.1.
Design considerations include:
whether the joint should be permanent or

non-permanent
cost
loads in the fastener or power screw and the

associated components
life
tooling
assembly
tolerances
aesthetics
size
corrosion
Threaded fasteners are introduced in Section 12.2,
power screws in Section 12.3, rivets in Section
12.4, adhesives in Section 12.5 and welding and
the related subjects of soldering and brazing in
Section 12.6.
12.2 Threaded fasteners
There is a large variety of fasteners available using
a threaded form to produce connection of com-
ponents.The common element of screw fasteners
is a helical thread that causes the screw to advance
into a component or nut when rotated. Screw
threads can be either left-handed or right-handed
Fastening and power screws
252
depending on the direction of rotation desired for
advancing the thread as illustrated in Figure 12.2.
Generally, right-hand threads are normally used
with left-handed threads being reserved for spe-
cialist applications.The detailed aspects of a thread
and the specialist terminology used is illustrated
in Figure 12.3 and defined in Table 12.1.
Specific thread forms, angle of helix, etc., vary
according to specific standards. Common standards
developed include UNS (unified national standard
series threads) and ISO threads. Both of these use
a 60 included angle, but are not interchangeable.
The form for an ISO metric thread for a nut is
illustrated in Figure 12.4. In practice, the root of the
nut and the crest of the mating bolt are rounded.
Both male and female ISO threads are subject to
manufacturing tolerances, which are detailed in BS
3643. A coarse series thread and fine series threads
Butt joint (edge)
Butt joint (tee)
Lap joint
Double lap joint
Cylindrical butt joint
Flange joint
Combination flange
and lap joint
Cylindrical lap joint
Face joint
Cylindrical flange joint
Figure 12.1 A variety of types of join.
Crest
Nut
Root
Bolt
Pitch P
Thread axis
Thread
angle
R
o
o
t

d
i
a
m
e
t
e
r
E
f
f
e
c
t
i
v
e

d
i
a
m
e
t
e
r
M
a
j
o
r

d
i
a
m
e
t
e
r
F
l
a
n
k
D
e
p
t
h
Pitch
_____
2
Figure 12.3 Specialist terminology used for describing
threads.
(a) (b)
Figure 12.2 (a) Right-hand thread; (b) Left-hand thread.
Chapter 12
253
are defined in the ISO standard, but fine series
threads tend to be more expensive and may not be
readily available from all stockists.Table 12.2 gives
the standard sizes for a selection of ISO coarse
series hexagon bolts, screws and nuts. ISO metric
threads are designated by the letter M, followed by
the nominal diameter and the pitch required, for
example, M6 1.5.
The unified system of screw threads was origin-
ally introduced in the United Kingdom, Canada
and the United States in order to provide a com-
mon standard for use by the three countries.Types
of unified threads in common use include the
unified coarse pitch thread series (UNC) and the
unified fine pitch thread series (UNF). Pertinent
dimensions for selected UNC and UNF threads
are given in Tables 12.3 and 12.4. Unified threads
are specified by notation, in the case of a 1/2 bolt,
in the form 1/2 in, 13UNC or 1/2 in, 20UNF
depending on whether a coarse or a fine thread is
being used.
The range of threaded fasteners available is
extensive including nuts and bolts, machine screws,
set screws and sheet metal screws. A variety of
machine screws is illustrated in Figure 12.5.
With such an array of types of fastener, the task
of selection of the appropriate type for a given
application can be time-consuming.
Washers can be used either under the bolt head
or the nut or both in order to distribute the clamp-
ing load over a wide area and to provide a bearing
surface for rotation of the nut.The most basic form
of a washer is a simple disc with a hole through
which the bolt or screw passes. There are, how-
ever, many additional types with particular attri-
butes such as lock washers, which have projections
that deform when compressed, producing add-
itional forces on the assembly, decreasing the pos-
sibility that the fastener assembly, will loosen in
service.Various forms of washer are illustrated in
Figure 12.6.
Selection of a particular fastener will depend
upon many different criteria such as:
strength at the operating temperatures

concerned
weight
cost
corrosion resistance
magnetic properties
life expectancy
assembly considerations.
Threaded fasteners tend to be used such that
they are predominantly loaded in tension. An
example is the bolt shown in Figure 12.7 used to
fasten a flanged joint. As the fastener is tightened
the tension on the bolt increases. It might be
Table 12.1 Thread terminology
Term Description
Pitch The thread pitch is the distance between
corresponding points on adjacent threads.
Measurements must be made parallel to the
thread axis
Outside The outside or major diameter is the
diameter diameter over the crests of the thread
measured at right angles to the thread axis
Crest The crest is the most prominent part of
thread, either external or internal
Root The root lies at the bottom of the groove
between two adjacent threads
Flank The flank of a thread is the straight side of the
thread between the root and the crest
Root The root, minor or core diameter is the
diameter smallest diameter of the thread measured at
right angles to the thread axis
Effective The effective diameter is the diameter on
diameter which the width of the spaces is equal to the
width of the threads. It is measured at right
angles to the thread axis
Lead The lead of a thread is the axial movement of
the screw in one revolution
Pitch P
60
1.4433P
Pitch
line
Nut
Bolt
P_
8
P_
4
2
_P
6 _ H
H _ 4
_ 3 8
H
H
Figure 12.4 ISO metric thread.
254
envisaged that the strength of a threaded fastener
would be limited by the area of its minor diameter.
Testing, however, shows that the tensile strength is
better defined using an area based on an average of
the minor and pitch diameters.
(12.1)
For UNS threads,
(12.2)
For ISO threads
(12.3)
and
d
r
d 1.226869p (12.4)
The stress in a threaded rod due to a tensile load is
(12.5)
Theoretically one might think that when a nut
engages a thread, all the threads in engagement
would share the load. However, inaccuracies in
thread spacing cause virtually all the load to be
taken by the first pair of threads.

t
t

F
A
d d p
p
0.649519
d d
N
d d
N
p r

0.649519
and
1.299038

A d d
t p r
2

16
(

)
Table 12.2 Selected dimensions for a selection of British Standard ISO Metric Precision Hexagon
Bolts. BS 3692:1967
Nominal size Pitch of
and thread thread (coarse
Width across flats Height of head
Tapping Clearance
diameter pitch series) Max Min Max Min drill drill
M1.6 0.35 3.2 3.08 1.225 0.975 1.25 1.65
M2 0.4 4.0 3.88 1.525 1.275 1.60 2.05
M2.5 0.45 5.0 4.88 1.825 1.575 2.05 2.60
M3 0.5 5.5 5.38 2.125 1.875 2.50 3.10
M4 0.7 7.0 6.85 2.925 2.675 3.30 4.10
M5 0.8 8.0 7.85 3.650 3.35 4.20 5.10
M6 1 10.0 9.78 4.15 3.85 5.00 6.10
M8 1.25 13.0 12.73 5.65 5.35 6.80 8.20
M10 1.5 17.0 16.73 7.18 6.82 8.50 10.20
M12 1.75 19.0 18.67 8.18 7.82 10.20 12.20
M14 2 22.0 21.67 9.18 8.82 12.00 14.25
M16 2 24.0 23.67 10.18 9.82 14.00 16.25
M18 2.5 27.0 26.67 12.215 11.785 15.50 18.25
M20 2.5 30.0 29.67 13.215 12.785 17.50 20.25
M22 2.5 32.0 31.61 14.215 13.785 19.50 22.25
M24 3 36.0 35.38 15.215 14.785 21.00 24.25
M27 3 41.0 40.38 17.215 16.785 24.00 27.25
M30 3.5 46.0 45.38 19.26 18.74 26.50 30.50
M33 3.5 50.0 49.38 21.26 20.74 29.50 33.50
M36 4 55.0 54.26 23.26 22.74 32.00 36.50
M39 4 60.0 59.26 25.26 24.74 35.00 39.50
M42 4.5 65.0 64.26 26.26 25.74 37.50 42.50
M45 4.5 70.0 69.26 28.26 27.74 40.50 45.50
M48 5 75.0 74.26 30.26 29.74 43.00 48.75
M52 5 80.0 79.26 33.31 32.69 47.00 52.75
M56 5.5 85.0 84.13 35.31 34.69 50.50 56.75
M60 5.5 90.0 89.13 38.31 37.69 54.50 60.75
M64 6 95.0 94.13 40.31 39.69 58.00 64.75
M68 6 100.0 99.13 43.31 42.96 62.00 68.75
All dimensions in mm.
Bolts are normally tightened by applying
torque to the head or nut, which causes the bolt
to stretch. The stretching results in bolt tension,
known as preload, which is the force that holds a
joint together.Torque is relatively easy to measure
using a torque meter during assembly so this is
the most frequently used indicator of bolt ten-
sion. High preload tension helps to keep bolts
tight, increases the strength of a joint, generates
friction between parts to resist shear and improves
the fatigue resistance of bolted connections. The
recommended preload for reusable connections
can be determined by
F
i
0.75A
t
p
(12.6)
and for permanent joints by
F
i
0.9A
t
p
(12.7)
where A
t
is the tensile stress area of the bolt (m
2
);
p
, proof strength of the bolt (N/m
2
).
Material properties for steel bolts are given in
SAE standard J1199 and by bolt manufacturers.
If detailed information concerning the proof
strength is unavailable then it can be approxi-
mated by
p
0.85
y
(12.8)
Once the preload has been determined the
torque required to tighten the bolt can be esti-
mated from
T KF
i
d (12.9)
where T is wrench torque (N m); K, constant;
F
i
, preload (N); d, nominal bolt diameter (m).
The value of K depends on the bolt material
and size. In the absence of data from manufactur-
ers or detailed analysis, values for K are given in
Table 12.5 for a variety of materials and bolt sizes.
Chapter 12
255
Table 12.3 American Standard thread dimensions for
UNC screw threads
Nominal Tensile
Size major Threads stress area
designation diameter (in) per inch (in
2
)
0 0.0600
1 0.0730 64 0.00263
2 0.0860 56 0.00370
3 0.0990 48 0.0487
4 0.1120 40 0.00604
5 0.1250 40 0.00796
6 0.1380 32 0.00909
8 0.1640 32 0.0140
10 0.1900 24 0.0175
12 0.2160 24 0.0242
Fractional sizes
1/4 0.2500 20 0.0318
5/16 0.3125 18 0.0524
3/8 0.3750 16 0.0775
7/16 0.4375 14 0.1063
1/2 0.5000 13 0.1419
9/16 0.5625 12 0.182
5/8 0.6250 11 0.226
3/4 0.7500 10 0.334
7/8 0.8750 9 0.462
1 1.000 8 0.606
1
1
8 1.125 7 0.763
1
1
4 1.250 7 0.969
1
3
8 1.375 6 1.155
1
1
2 1.500 6 1.405
1
3
4 1.750 5 1.90
2 2.000 4.5 2.50
Table 12.4 American Standard thread dimensions for
UNF screw threads
Nominal Tensile
Size major Threads stress area
designation diameter (in) per inch (in
2
)
0 0.0600 80 0.00180
1 0.0730 72 0.00278
2 0.0860 64 0.00394
3 0.0990 56 0.00523
4 0.1120 48 0.00661
5 0.1250 44 0.00830
6 0.1380 40 0.01015
8 0.1640 36 0.01474
10 0.1900 32 0.0200
12 0.2160 28 0.0258
Fractional sizes
1/4 0.2500 28 0.0364
5/16 0.3125 24 0.0580
3/8 0.3750 24 0.0878
7/16 0.4375 20 0.1187
1/2 0.5000 20 0.1599
9/16 0.5625 18 0.203
5/8 0.6250 18 0.256
3/4 0.7500 16 0.373
7/8 0.8750 14 0.509
1 1.000 12 0.663
1
1
8 1.125 12 0.856
1
1
4 1.250 12 1.073
1
3
8 1.375 12 1.315
1
1
2 1.500 12 1.581
256
(a)
(b)
(c)
(d)
(e)
(f)
(g)
(h)
(i)
Figure 12.5 Various machine screw styles. (a) Flat
countersunk head. (b) Slotted truss head. (c) Slotted
pan head. (d) Slotted fillister head. (e) Slotted oval
countersunk. (f) Round head. (g) Hex. (h) Hex washer.
(i) Slotted hexagon head.
Figure 12.6 Washers.
Figure 12.7 Flanged joint.
Example 12.1
An M10 bolt has been selected for a re-useable
application.The proof stress of the low carbon
steel bolt material is 310MPa. Determine the
recommended preload on the bolt and the
torque setting.
Solution
From Table 12.2, the pitch for a coarse series
M10 bolt is 1.5mm.
For a reusable connection, the recommended
preload is
F
i
0.75A
t
p
13.48kN
From Table 12.5, K 0.2.The torque required
to tighten the bolt is given by
T KF d 0.2 13.48 10 0.01
26.96 Nm
i
3

A
t
2 2

16
8.16) 57.99 mm

( . 9 026
d
r
10 1.226869 1.5 8.160 mm
d
p
10 0.649519 1.5 9.026 mm
The principal applications of threaded fasteners,
such as bolts and nuts is clamping components
together. In such situations the bolt is predomi-
nantly in tension. Both the bolt and the clamped
components will behave as elastic members, pro-
vided material limits are not exceeded. If a load is
applied to a bolted joint that is above the clamp-
ing load, then the behaviour of the joint itself
needs to be considered. As the bolt stretches, the
compressive load on the joint will decrease, alle-
viating some of the load on the bolt. If a very stiff
bolt is used to clamp a flexible member, such as a
soft gasket, most of the additional force, above the
clamping load, is taken by the bolt and the bolt
should be designed to take the clamping force
and any additional force. Such a joint can be clas-
sified as a soft joint. If, however, the bolt is rela-
tively flexible compared to the joint, then nearly
all the externally applied load will initially go
towards decreasing the clamping force until the
components separate.The bolt will then carry all
the external load.This kind of joint is classified as
a hard joint.
Practical joints normally fall between the two
extremes of hard and soft joints. The clamped
components of a typical hard joint have a stiffness
of approximately three times that of the bolt. An
externally applied load will be shared by the bolt
and the clamped components according to the
relative stiffnesses, which can be modelled by
(12.10)
(12.11)
where F
b
is final force in the bolt (N); F
i
, initial
clamping load (N); k
b
, stiffness of the bolt (N/m);
k
c
, stiffness of the clamped components (N/m);
F
e
, externally applied load (N); F
c
, final force on
the clamped components (N).
F F
k
k k
F
c i
c
b c
e

F F
k
k k
F
b i
b
b c
e

Chapter 12
257
Table 12.5 Values for the constant K for determining
the torque required to tighten a bolt
Conditions K
1
4 to 1inch mild steel bolts 0.2
Non-plated black finish steel bolts 0.3
Zinc plated steel bolts 0.2
Lubricated steel bolts 0.18
Cadmium plated steel bolts 0.16
Source: Oberg et al., 1996.
Example 12.2
A set of six M8 bolts is used to provide a
clamping force of 20kN between two compo-
nents in a machine. If the joint is subjected to
an additional load of 18kN after the initial
preload of 8.5kN per bolt has been applied,
determine the stress in the bolts. The stiffness
of the clamped components can be assumed to
be three times that of the bolt material. The
proof stress of the low carbon steel bolt mater-
ial is 310MPa.
Solution
Taking k
c
3k
b
,
As F
c
is greater than zero, the joint remains
tight.The tensile stress area for the M8 bolt can
be determined from
A
t
2 2

16
6.466) 36.61 mm

( . 7 188
d
r
8 1.226869 1.25 6.466 mm
d
p
8 0.649519 1.25 7.188 mm
F F
k
k k
F F
k
k k
F
F F
c i
c
b c
e i
b
b b
e
i e

3
3

3
4
8500
3(18000/6)
4

6250 N

F F
k
k k
F F
k
k k
F
F F
b i
b
b c
e i
b
b b
e
i e

3

1
4
8500
18000/6
4

9250 N

12.3 Power screws

Power screws, which are also known as lead screws,
are used to convert rotary motion into linear move-
ment.With suitably sized threads they are capable of
large mechanical advantage and can lift or move
large loads. Applications include screw jacks and
traverses in production machines. Although suited
to fasteners thread forms, such as the ISO metric
standard screw threads, UNC and UNF series
described in Section 12.2 may not be strong enough
for power screw applications. Instead square, Acme
and buttress thread forms have been developed and
standardized; see Figures 12.8 to 12.10, respectively.
Some of the principal dimensions for standard
Acme threads are given in Table 12.6.
Self-locking in power screws refers to the con-
dition in which a screw cannot be turned by the
application of axial force on the nut. This is very
useful in that a power screw that is self-locking will
hold its position and load unless a torque is applied.
As an example most screw jacks for cars are self-
locking and do not run down when the handle is
let go. The opposite condition to self-locking is a
screw that can be back-driven.This means that an
axial force applied to the nut will cause the screw
to turn. A product application of this is the Yankee
screwdriver, which has a high-lead thread on its
barrel that is attached to the screwdriver bit. The
handle acts as the nut and when pushed the barrel
will turn driving the bit round. Back-driveable
power screws are a useful form of turning linear
motion into rotary motion.
258
The stress in each bolt is given by
This is 82 per cent of the proof stress.The bolts
are therefore safe.

9250
36.61 10
252.7 MPa
b
t
6
F
A
P _ 2
2
_
Pitch P
P
Figure 12.8 Square thread.
29
Pitch P
D
R
C
Figure 12.9 Acme thread.
7
45
H
Pitch P
F
R
S
H
Figure 12.10 Buttress thread.

Table 12.6 Principal dimensions for ACME threads
Major Thread Pitch Minor Tensile
diam- Threads pitch diam- diam- stress
eter (in) per inch (in) eter (in) eter (in) area (in
2
)
0.25 16 0.063 0.219 0.188 0.032
0.313 14 0.071 0.277 0.241 0.053
0.375 12 0.083 0.333 0.292 0.077
0.438 12 0.083 0.396 0.354 0.110
0.500 10 0.100 0.450 0.400 0.142
0.625 8 0.125 0.563 0.500 0.222
0.750 6 0.167 0.667 0.583 0.307
0.875 6 0.167 0.792 0.708 0.442
1.000 5 0.200 0.900 0.800 0.568
1.125 5 0.200 1.025 0.925 0.747
1.250 5 0.200 1.150 1.050 0.950
1.375 4 0.250 1.250 1.125 1.108
1.500 4 0.250 1.375 1.250 1.353
1.750 4 0.250 1.625 1.500 1.918
2.000 4 0.250 1.875 1.750 2.580
2.250 3 0.333 2.083 1.917 3.142
2.500 3 0.333 2.333 2.167 3.976
2.750 3 0.333 2.583 2.417 4.909
3.000 2 0.500 2.750 2.500 5.412
3.500 2 0.500 3.250 3.000 7.670
4.000 2 0.500 3.750 3.500 10.321
4.500 2 0.500 4.250 4.000 13.364
5.000 2 0.500 4.750 4.500 16.800
A screw thread can be considered to be an
inclined plane wrapped around a cylinder to form
a helix. The forces on a single thread of a power
screw are illustrated for the case of lifting a load
and lowering a load in Figure 12.11.
The inclination of the plane is called the lead
angle .
(12.12)
In the case of lifting a load, summing the forces
gives
(12.13)
(12.14)
So
(12.15)
So
(12.16)
where is the coefficient of friction between the
screw and the nut.The coefficient of friction can
typically be taken as 0.15.
Solving Eqs 12.14 and 12.16 to give an expres-
sion for F gives
(12.17)
The screw torque required to lift the load is
given by
(12.18)
It is convenient to rewrite this equation in terms
of the lead L, so substituting with Eq. 12.12 and
rearranging gives:
(12.19)
The thrust collar also contributes a friction torque
that must be accounted for. The torque required
to turn the thrust collar is given by
(12.20)
where d
c
is diameter of the thrust collar;
c
, coef-
ficient of friction in the thrust bearing.
The total torque to lift the load for a square
thread is
(12.21)
Similar analysis can be performed for lowering
a load, in which case:
(12.22)
For Acme threads the equivalent torque relation-
ship for lifting a load is:
(12.23)
and for lowering:
(12.24)
The work done on a power screw is the product
of the torque and angular displacement. For one
revolution of a screw,
W
in
2T (12.25)
T
Pd d L
d L
P
d
d
p p
p
c
c

cos )
2( cos )

2
(
T
Pd d L
d L
P
d
u
p p
p
c
c

cos )
2( cos )

2
(
T
Pd d L
d L
P
d
d
p p
p
c
c

)
2( )

2
(
T
Pd d L
d L
P
d
u
p p
p
c
c

)
2( )

2
(
T P
d
c c
c

2
T
Pd d L
d L
s
p p
p

)
2( )

(
T F
d Pd
s
p p

2

cos sin )
2(cos sin )

(
F
P

( cos sin )
cos sin

N
P

cos sin

F N f P
N N P
y
0 cos sin
cos sin

F N ( cos sin )
F F f N
F N N
x
0 sin
cos sin

cos
tan
p

L
d
Chapter 12
259
y
x
F
P
N
L lead
f
d
p
F
d
p
f
N
P
L
(a) (b)
Figure 12.11 Force analysis at the interface of a lead
screw and nut (a) lifting a load; (b) lowering a load.
The work delivered for one revolution is
W
out
PL (12.26)
The efficiency of the system defined as work
out/work in is
(12.27)
Substituting for the torque for an Acme thread
using Eq. 12.23, neglecting collar friction, gives
(12.28)
Simplifying using Eq. 12.12 gives
(12.29)
For a square thread, 0, so cos 1, giving
(12.30)
Standard Acme screws have lead angles between
about 2 and 5. Assuming a coefficient of friction
of 0.15 gives the efficiency between 18 and 36 per
cent.This shows the disadvantage of power screws.
Higher efficiencies can, however, be attained by
reducing the friction and one of the ways of doing
this has been found to be the use of ball screws
although these potentially add to the cost of
production.
A screw will self-lock if
(12.31)
or
tancos (12.32)
For a square thread, where cos 1,
(12.33)
or
tan (12.34)
It should be noted that Eqs 12.31 to 12.34 are for
a statically loaded screw. Dynamic loading, such as
vibration, can reduce the effective friction and
cause a screw to back-drive.
In the design of a power screw, consideration
should also be given to buckling of the screw and
choice of material for the screw and nut.

p
L
d

cos
p
L
d

1 tan
1 cot

cos tan
cos cot

( cos )
( cos )
p
p p
L d L
d d L

2
PL
T
260
Example 12.3
A self-locking power screw is required for a
screw jack. An initial proposal is to use a single
start 1.255 Acme power screw.The axial load
is 4000N and collar mean diameter is 1.75in.
Determine the lifting and lowering torques,
the efficiency of the power screw and whether
the design proposal is self-locking.
Solution
Single start thread, so the lead, L, is equal to the
pitch, P.
N 5 teeth per inch
Assume sliding friction, 0.15.
The torque to lift the load is given by
Eq. 12.23.
T
4000 0.02921

0.02921 0.00508 cos 14.5
0.02921cos 14.5 0.15 0.00508
0.15 4000
0.04445
2
58.42
0.01376 4.9182 10
7.62 10
13.34
12.39 13.34 25.73 N m
3
4
2
0 15
0 08884
.
.
d
c
1.75 44.45 mm
d
p
1.15 29.21 mm
L 0.2 5.08 mm
P
N

1

1
5
0.2
12.4 Rivets
Rivets are non-threaded fasteners that are usually
manufactured from steel or aluminium.They con-
sist of a preformed head and shank, which is
inserted into the material to be joined and the
second head that enables the rivet to function as a
fastener is formed on the free end by a variety of
means known as setting.A conventional rivet before
and after setting is illustrated in Figure 12.12.
Rivets are widely used to join components in
aircraft, boilers, ships and boxes and other enclos-
ures. Rivets tend to be much cheaper to install
than bolts and the process can be readily auto-
mated with single riveting machines capable of
installing thousands of rivets an hour.
Rivets can be made from any ductile material,
such as carbon steel, aluminium and brass. A var-
iety of coatings is available to improve corrosion
resistance. Care needs to be taken in the selection
of material and coating in order to avoid the pos-
sibility of corrosion by galvanic action. In general,
a given size rivet will be not as strong as the
equivalent threaded fastener.
The two main types of rivet are tubular and
blind and each type are available in a multitude of
varieties. The advantage of blind rivets is that they
require access to only one side of the joint. A fur-
ther type of rivet with potentially many overall
advantages, from the production perspective, is self-
piercing rivets that do not require predrilled holes.
The rivet is driven into the target materials with
high force, piercing the top sheets and spreading
outwards into the bottom sheet of material under
the influence of an upsetting die to form the joint.
Factors in the design and specification of rivets
include the size, type and material for the rivet,
the type of join, and the spacing between rivets.
There are two main types of riveted joint: lap-
joints and butt-joints (Figure 12.13). In lap joints
the components to be joined overlap each other,
while for butt joints an additional piece of material
is used to bridge the two components to be joined
which are butted up against each other. Rivets
can fail by shearing through one cross-section
known as single shear, shearing through two
cross-sections known as double shear, and crush-
ing. Riveted plates can fail by shearing, tearing
and crushing.
For many applications, the correct use of rivets
is safety critical and their use is governed by con-
struction codes. For information and data con-
cerning joints for pressure vessels reference to the
appropriate standards should be made, such as the
ASME boiler code.
Chapter 12
261
The torque to lower the load is given by
Eq. 12.24.
The design will be self-locking if
0.15, so the design is self-locking.

cos
0.00508
0.02921
cos 14.5
0.05359
p
L
d

2

4000 0.00508
2 12.39
0.261

4000 0.00508
2 25.73
0.126
screw
both
PL
T
T 58.42
0.01376 4.9182 10
0.08884 7.62 10
13.34
5.765 13.34 19.11 Nm
3
4
Before setting
After setting
Figure 12.12 Conventional rivet before and after setting.
Riveted joints can be designed using a simple
procedure (Oberg et al., 1996) assuming that:
the load is carried equally by the rivets;
no combined stresses act on a rivet to cause

failure;
the shearing stress in a rivet is uniform across the

cross-section;
the load that would cause failure in single

shear would have to be double to cause failure
in double shear;
the bearing stress of the rivet and plate is dis-

tributed equally over the projected area of the
rivet;
the tensile stress is uniform in the section of

metal between the rivets.
The allowable stress for a rivet is generally
defined in the relevant standard. For example, the
ASME boiler code lists an ultimate tensile stress
for rivets of 379MPa, ultimate shearing stress of
303MPa and an ultimate compressive or bearing
stress of 655MPa. Design stresses are usually 20
per cent of these values, i.e. for tensile, shear and
bearing stresses the design limits are 75, 60 and
131MPa, respectively.
For a single lap joint, the safe tensile load based
on shear is given by
L nA
r
d
(12.35)
on compressive or bearing stress is given by
L nA
b
c
(12.36)
A
b
td (12.37)
on tensile stress is given by
L A
p
t
(12.38)
where L is load (N); A
r
, cross-sectional area of
rivet (m
2
); A
b
, projected bearing area of rivet
(m
2
); A
p
, cross-sectional area of plate between
rivet holes (m
2
); t, thickness of plate (m); d, diam-
eter of rivet (m);
d
, allowable shear stress (N/m
2
);
c
, allowable bearing or compressive stress
(N/m
2
);
t
, allowable tensile stress (N/m
2
).
The efficiency of a riveted joint is given by
(12.39)
A selection of rivets specified in BS 4620:1970
and ANSI B18.1.2-1972 is included in Tables 12.7
to 12.10 as examples. Rivets, however, are avail-
able as stock items from specialist manufacturers
and suppliers in a much wider variety than the
small selection presented in Tables 12.7 to 12.10.
For any given application the relevant standard
should be referenced and a range of manufactur-
ers products considered.

least safe load
ultimate tensile strength of
unperforated section
262
t
p
t
p
p
P
t
t
c
t
c
Single riveted lap-joint
Double riveted lap-joint
Double riveted butt-joint
Figure 12.13 Some types of riveted joints.
Chapter 12
263
Table 12.7 British Standard hot-forged rivets for general engineering purposes
Head dimensions
60 csk and
raised csk head Snap head Universal head
Nom shank Tol on Nom Height of Nom Nom Nom Nom
diam d diam d diam D raise W diam D depth K diam D depth K Rad R Rad r
(14) 21 2.8 22 9 28 5.6 42 8.4
16 0.43 24 3.2 25 10 32 6.4 48 9.6
(18) 27 3.6 28 11.5 36 7.2 54 11
20 30 4.0 32 13 40 8.0 60 12
(22) 0.52 33 4.4 36 14 44 8.8 66 13
24 36 4.8 40 16 48 9.6 72 14
(27) 40 5.4 43 17 54 10.8 81 16
30 45 6.0 48 19 60 12.0 90 18
(33) 0.62 50 6.6 53 21 66 13.2 99 20
36 55 7.2 58 23 72 14.4 108 22
(39) 59 7.8 62 25 78 15.6 117 23
Extracted from BS 4620:1970. Note see the standard for full ranges. All dimensions are in millimetres. Sizes shown in
parentheses are non-preferred.
Hot-forged rivets
60 csk and raised csk head
d
K
D
602.5
L
W
Universal head Snap head
D
R
L
d
D
r
d
K
L
K
Example 12.4
Determine the safe tensile, shear and bearing
loads and the efficiency for a 300mm section of
single-riveted lap joint made from 1/4 plates
using six 16-mm diameter rivets. Assume that
the drilled holes are 1.5mm larger in diameter
than the rivets. The values for the design limits
for tensile, shear and bearing stress can be taken
as 75, 60 and 131MPa, respectively.
Solution
The safe tensile load, L, based on shear of the
rivets is given by
The safe tensile load based on bearing or com-
pressive stress is given by
L 6 0.016 6.35 10 131 10
79.86 kN
3 6

L nA 6
0.016
4
75 10 90.48 kN
r d
6
2

264
Table 12.8 British Standard cold-forged rivets for general engineering purposes
Head dimensions
90 csk
head Snap head Universal head Flat head
Nom shank Tol on Nom Nom Nom Nom Nom Nom Nom
diam d diam d diam D diam D depth K diam D depth K Rad R Rad r diam D depth K
1 0.07 2 1.8 0.6 2 0.4 3.0 0.6 2 0.25
1.2 2.4 2.1 0.7 2.4 0.5 3.6 0.7 2.4 0.3
1.6 3.2 2.8 1.0 3.2 0.6 4.8 1.0 3.2 0.4
2 4 3.5 1.2 4 0.8 6.0 1.2 4 0.5
2.5 5 4.4 1.5 5 1.0 7.5 1.5 5 0.6
3 6 5.3 1.8 6 1.2 9.0 1.8 6 0.8
(3.5) 0.09 7 6.1 2.1 7 1.4 10.5 2.1 7 0.9
4 8 7 2.4 8 1.6 12 2.4 8 1.0
5 10 8.8 3.0 10 2.0 15 3.0 10 1.3
6 12 10.5 3.6 12 2.4 18 3.6 12 1.5
(7) 0.11 14 12.3 4.2 14 2.8 21 4.2 14 1.8
8 16 14 4.8 16 3.2 24 4.8 16 2
10 20 18 6.0 20 4.0 30 6 20 2.5
12 0.14 24 21 7.2 24 4.8 36 7.2
(14) 25 8.4 28 5.6 42 8.4
16 28 9.6 32 6.4 48 9.6
Extracted from BS 4620:1970. Note see the standard for full ranges.
All dimensions are in millimetres. Sizes shown in parentheses are non-preferred.
CoId-forged rivets
K
L
d
D
902.5
d
L
K
D
d
L
K
D
L
K
d
D
R
r
Snap head Universal head Flat head 90 csk head
The safe load based on tensile load is given by
L A
p
t
.
The area of the plate between the rivet
holes, A
p
is given by
A
p
3
3
3 2
0.00635 (0.3 6(16 10
1.5 10 ))
1.238 10 m

The safe tensile load would be the least of the
three values determined, i.e. L 79.86kN.
The efficiency is 56 per cent.

79.86 10
0.3 75 10
0.56
3
6
0 00635 .
L A
t
1.238 10 131 10
162.2 kN
p
3 6

Chapter 12
265
Table 12.9 Selected American National Standard large button, high button, cone and pan head rivets
Head diam A (in) Height H (in) Head diam A (in) Height H (in)
Mfd Driven Mfd Driven Mfd Driven Mfd Driven
Nom body diam D
a
note 1 note 2 note 1 note 2 note 1 note 2 note 1 note 2
Button head High button head (acorn)
1/2 0.875 0.922 0.375 0.344 0.781 0.875 0.500 0.375
5/8 1.094 1.141 0.469 0.438 0.969 1.062 0.594 0.453
3/4 1.312 1.375 0.562 0.516 1.156 1.250 0.688 0.531
7/8 1.531 1.594 0.656 0.609 1.344 1.438 0.781 0.609
1 1.750 1.828 0.750 0.688 1.531 1.625 0.875 0.688
1
1
8 1.969 2.062 0.844 0.781 1.719 1.812 0.969 0.766
1
1
4 2.188 2.281 0.938 0.859 1.906 2.000 1.062 0.844
1
3
8 2.406 2.516 1.031 0.953 2.094 2.188 1.156 0.938
1
1
2 2.625 2.734 1.125 1.031 2.281 2.375 1.250 1.000
1
5
8 2.844 2.969 1.219 1.125 2.469 2.562 1.344 1.094
1
3
4 3.062 3.203 1.312 1.203 2.656 2.750 1.438 1.172
Cone head Pan head
1/2 0.875 0.922 0.438 0.406 0.800 0.844 0.350 0.328
5/8 1.094 1.141 0.547 0.516 1.000 1.047 0.438 0.406
3/4 1.312 1.375 0.656 0.625 1.200 1.266 0.525 0.484
7/8 1.531 1.594 0.766 0.719 1.400 1.469 0.612 0.578
1 1.750 1.828 0.875 0.828 1.600 1.687 0.700 0.656
1
1
8 1.969 2.063 0.984 0.938 1.800 1.891 0.788 0.734
1
1
4 2.188 2.281 1.094 1.031 2.000 2.094 0.875 0.812
1
3
8 2.406 2.516 1.203 1.141 2.200 2.312 0.962 0.906
1
1
2 2.625 2.734 1.312 1.250 2.400 2.516 1.050 0.984
1
5
8 2.844 2.969 1.422 1.344 2.600 2.734 1.138 1.062
1
3
4 3.062 3.203 1.531 1.453 2.800 2.938 1.225 1.141
Extracted from ANSI B18.1.21972, R1989. Note see the standard for full ranges.
All dimensions are given in inches.
a
Tolerance for diameter of body is plus and minus from nominal and for
1
2-in. size equals 0.020, 0.022; for sizes
5
8 to 1-in, incl,
equals 0.030, 0.025; for sizes 1
1
8 and 1
1
4-in equals 0.035, 0.027; for sizes 1
3
8 and 1
1
2-in equals 0.040, 0.030; for sizes 1
5
8 and 1
3
4-in
equals 0.040, 0.037.
Note 1: Basic dimensions of head as manufactured.
Note 2: Dimensions of manufactured head after driving and also of driven head.
Note 3: Slight flat permissible within the specified head-height tolerance.
The following formulae give the basic dimensions for manufactured shapes: Button head, A1.750D; H0.750D; G0.885D. High button head, A
1.500D0.031; H0.750D0.125; F 0.750D0.281; G0.750D 0.281. Cone head, A1.750D; B0.938D; H0.875D. Pan head,
A1.600D; B1.000D; H0.700D. Length L is measured parallel to the rivet axis, from the extreme end to the bearing surface plane for flat bear-
ing surface head type rivets, or to the intersection of the head top surface with the head diameter for countersunk head type rivets.
A
D
G
H
L
L
H
D
G
A
L
H
D
A
0.094
F
0.500
B
D
L
B
A
H
Button head High button Cone head Pan head
see note 3
266
Table 12.10 Selected American National Standard large flat and oval countersunk rivets
Body diameter D (in) Head diam A (in) Head depth H (in)
Oval crown Oval crown
Nominal
a
Max. Min. Max.
b
Min.
c
Ref. height
a
C (in) radius
a
G (in)
1/2 0.500 0.520 0.478 0.936 0.872 0.260 0.095 1.125
5/8 0.625 0.655 0.600 1.194 1.112 0.339 0.119 1.406
3/4 0.750 0.780 0.725 1.421 1.322 0.400 0.142 1.688
7/8 0.875 0.905 0.850 1.647 1.532 0.460 0.166 1.969
1 1. 000 1.030 0.975 1.873 1.745 0.520 0.190 2.250
1
1
8 1.125 1.160 1.098 2.114 1.973 0.589 0.214 2.531
1
1
4 1.250 1.285 1.223 2.340 2.199 0.650 0.238 2.812
1
3
8 1.375 1.415 1.345 2.567 2.426 0.710 0.261 3.094
1
1
2 1.500 1.540 1.470 2.793 2.652 0.771 0.285 3.375
1
5
8 1.625 1.665 1.588 3.019 2.878 0.831 0.309 3.656
1
3
4 1.750 1.790 1.713 3.262 3.121 0.901 0.332 3.938
Extracted from ANSI B18.1.21972, R1989. Note see the standard for full ranges.
All dimensions are given in inches.
a
Basic dimension as manufactured. For tolerances see table footnote on Table 12.9.
b
Sharp edged head.
c
Rounded or flat edged irregularly shaped head (heads are not machined or trimmed).
Flat and oval countersunk head
D
L
Q
H
A
C
G
H
L
D
A
Q
Flat csk head
Oval csk head
Example 12.5
Determine the maximum safe tensile load that
can be supported by a 1m section of double
riveted butt joint with 15mm thick main
plates and two 8mm thick cover plates. There
are six rivets in each of the outer rows and
seven rivets in each of the inner rows. The
rivets are all 20mm in diameter. Assume that
the drilled holes are 1.5mm larger in diameter
than the rivets.The values for the design limits
for tensile, shear and bearing stress can be
taken as 75, 60 and 131MPa, respectively.
Solution
In analysing a double riveted joint, it is only
necessary to analyse one side due to symmetry.
The safe tensile load based on double shear-
ing of the rivets is equal to the number of rivets
288 Mechanical Engineers Pocket Book
4.2 Riveted joints
4.2.1 Typical rivet heads and shanks
2D 2D
Round or
snap head
L
e
n
g
t
h
Pan head Mushroom head
60
D
D
fD
3
2
2D
2D 2D
L
e
n
g
t
h
90
90 countersunk
head
D
fD
4
Flat head Conoidal head
Plain or
solid shank
Semi-tubular
shank
Tubular
shank
or eyelet
Split or
bifurcated
shank
Drive
shank
for soft
materials
4.2.2 Typical riveted lap joints
Single row lap joint
Double row (chain) lap joint
Double row (zigzag) lap joint
3D 3D
3
D
L
a
p

6
D
P
i
t
c
h
Pitch
1
.
5
D
3D
1.5D
3D 3D
3D 1.5D
1
.
5
D
L
a
p

6
D
3D 3D
Pitch
3
D
P
i
t
c
h
D rivet shank diameter
3D 1.5D
1
.
5
D
L
a
p

3
D
3D 3D
Pitch
Fastenings 289
4.2.3 Typical riveted butt joints
Single strap chain riveted butt joint (single row)
Note: This joint may also be double row riveted, chain or zigzag. The strap width 12D
when double riveted (pitch between rows 3D).
Double strap chain riveted butt joint (double row)
Note: This joint may also be double row zigzag riveted (see Section 4.2.2) or it may be
single riveted as above.
1.5D 3D 3D 3D
1
.
5
D
3
D
1
.
5
D
1
.
5
D
3
D
Pitch
P
i
t
c
h
P
i
t
c
h
S
t
r
a
p

o
r

c
o
v
e
r

p
l
a
t
e

w
i
d
t
h

=

1
2
D
Pitch
Strap or
cover plate
1
.
5
D
1
.
5
D
S
t
r
i
p

w
i
d
t
h

6
D
1
.
5
D
1.5D 3D D Rivet shank diameter 3D 3D
Fastenings 291
4.2.4 Proportions for hole diameter and rivet length
1.0625D
D
1.5D
Dimensions in mm
Nominal Tolerance Nominal Tolerance Nominal Tolerance Tolerance
shank on head on head on head on
diameter
a
diameter diameter diameter depth depth length
d d D D K K L
1 1.8 0.6 0.2
1.2 2.1 0.2 0.7 0.0
1.6 0.07 2.8 1.0
2.0
3.5 0.24 1.2 0.24
2.5
4.4 1.5 0.0
3.0
5.3 1.8 0.5
0.0
(3.5) 6.1 0.29 2.1 0.29
4 0.09 7.0 2.4 0.0
5 8.8 3.0
6
10.5 3.6
0.35
0.35 0.0
(7) 12.3 4.2 0.8
8 0.11 14.0 4.8 0.0
10
18.0 6.0
12 21.0 0.42 7.2 0.42 1.0
(14) 0.14 25.0 8.4 0.0 0.0
16 28.0 9.6
a
Rivet sizes shown in brackets are non-preferred.
For further information see BS 4620: 1970.
4.2.5 Cold forged snap head rivets
d
L
K
D
With d 16 mm or smaller
D 1.75d
K 0.6d
L length
Fastenings 293
Dimensions in mm
Nominal Tolerance Nominal Tolerance Nominal Tolerance Tolerance
shank on head on head on head on length
diameter
a
diameter diameter diameter depth depth L
d d D D K K
(14) 22 9 1.0
16 0.43 25 1.25 10 1.00 0.0
18 28 11.5 0.0
20 32 13 1.5 1.6
(22) 0.52 36 1.8 14 0.0 0.0
24 40 16
(27) 43 17 2.0
30 48 2.5 19 0.0
(33) 0.62 53 21 3.0
0.0
36 58 3.0 23 2.5
39 62 25 0.0
a
Rivet sizes shown in brackets are non-preferred.
For further information see BS 4620: 1970.
4.2.6 Hot forged snap head rivets
D
L
K
d
With d 14 mm or larger
D 1.6d
K 0.65d
L length
2
9
4
M
e
c
h
a
n
i
c
a
l

E
n
g
i
n
e
e
r
s

P
o
c
k
e
t

B
o
o
k
Dimensions in mm
Nominal Nominal length* L
shank
3 4 5 6 8 10 12 14 16 (18) 20 (22) 25 (28) 30 (32) 35 (38) 40 45 50 55
diameter
a
d
1.0
1.2
1.6
2.0
2.5
3.0
(3.5)
4.0
5.0
6.0
a
Sizes and lengths shown in brackets are non-preferred and should be avoided if possible. The inclusion of dimensional data is not intended to imply that all the products
described are stock production sizes. The purchaser should consult the manufacturer concerning lists of stock production sizes.
For the full range of head types and sizes up to and including 39mm diameter by 160 mm shank length see BS 4620: 1970.
4.2.7 Tentative range of nominal lengths associated with shank diameters
Fastenings 295
4.2.8 POP
rivets
POP
or blind riveting is a technique which enables a mechanical fastening to be made when

access is limited to only one side of the parts to be assembled, although reliability, predictabil-
ity, reduction of assembly costs and simplicity in operation, mean that blind rivets are also
widely used where access is available to both sides of an assembly. POP
and other brands of

blind riveting systems have two elements, the blind rivet and the chosen setting tool.
The blind rivet is a two-part mechanical fastener. It comprises of a headed tubular
body mounted on a mandrel which has self-contained features that create (when pulled
into the body during setting) both an upset of the blind end and an expansion of the
tubular body, thus joining together the component parts of the assembly. The setting tool
basically comprises an anvil which supports the head of the rivet and jaws which grip
and pull the mandrel to cause it to set the rivet before the mandrel breaks at a pre-
determined load.
Many different styles of blind rivet are available but the most widely used is the Open End
Rivet Body type defined in BS EN ISO 14588: 2000 as: A blind rivet having a body hollow
throughout its length and able to use an standard mandrel. The principle of blind rivet-
ing using open style rivets is shown in the following figure.
With the mandrel held in the setting tool, the rivet body is inserted into the pre-punched
or pre-drilled component. Operation of the setting tool pulls the mandrel head into the
rivet body causing it to expand on the blind side of the assembly, whilst drawing the com-
ponents together and eliminating any gaps between them as it does so. At a predeter-
mined point, when the blind side head is fully formed, continued operation of the setting
tool causes the mandrel to break, the spent portion of the mandrel is pulled clear and the
installation of the rivet is complete.
The Closed End Rivet Body type is defined in BS EN ISO 14588: 2000 as: A blind rivet body
which is closed and remains closed after setting. This type is also commonly known as
sealed. The closed end rivet prevents ingress of vapour or moisture through the bore of
the installed rivet and also ensures mandrel head retention, particularly important in elec-
trical equipment, for example.
POP
blind rivets are available in a variety of materials, body styles and head forms to pro-
vide fastening options for a broad spectrum of assembly and environmental requirements
from brittle and fragile materials such as acrylic plastics through to stainless steel. A sum-
mary is shown in Section 4.2.9. This figure and the following tables are taken from the
publications of Emhart Teknologies (Tucker Fasteners Ltd.) from whom further informa-
tion can be obtained. This companys address is listed in Appendix 3.
4.2.9 POP
range guide
Head style
The low-profile domed head is suitable for most applications but, where soft or brittle
materials are fastened to a rigid backing, the large flat head variety should be considered.
The 120 countersunk head style should be used wherever a flush surface is required.
Mandrel types
POP
open type rivets are normally supplied with Break stem mandrels (code BS)
designed to retain the mandrel head when the rivet is set. Break head (code BH) man-
drels, designed to eject the mandrel head from the rivet body, can be supplied for most
open rivets and are particularly useful in the pre-assembly of electrical circuit boards.
Standard open type rivet
Closed end type rivet
Peel type rivet
MGR
LSR aluminium
Wide range of rivets suitable for use where applications do
not have high load bearing requirements.
For situations where fastening has to be watertight,
pressure tight, or where mandrel retention is a requirement.
Suitable for joining plastics, rubber, wood, GRP or
laminates.
Wide grip range. Ideal for use with inconsistent holes.
Rivet with load spreading characteristics for use in soft,
friable or brittle materials.
Fastenings 297
Finish
G
Rivet body standard finishes: Steel and nickel-copper rivet bodies are normally supplied
zinc plated.
G
Paint and other finishes: Rivets with differing surface finishes and paint colours can be
provided on request. Aluminium alloy rivets are available anodized and dyed, matt or
gloss for aesthetic and environmental reasons.
4.2.10 Good fastening practice
Blind riveting is a highly reliable and proven method of fixing material together perman-
ently. To achieve a superior fastening, the following principles should be considered.
Workpiece materials
When materials of different thickness or strengths are being joined, the stronger material
if possible should be on the blind side. For example, if plastic and metal are to be joined,
the plastic sheet should be beneath the rivet head and the metal component should be
on the blind side.
Hole size and preparation
Achieving a good joint depends on good hole preparation, preferably punched and, if
necessary, de-burred to the sizes recommended in the POP
blind rivet data tables.

Rivet diameter
As a guide for load-bearing joints, the rivet diameter should be at least equal to the thick-
ness of the thickest sheet and not more than three times the thickness of the sheet imme-
diately under the rivet head. Refer to data tables for rivet strength characteristics.
Edge distance
Rivet holes should be drilled or punched at least two diameters away from an edge but
no more than 24 diameters from that edge.
Rivet pitch
As a guide to the distance between the rivets in load-carrying joint situations, this dis-
tance should never exceed three rivet diameters. In butt construction it is advisable to
include a reinforcing cover strip, fastening it to the underlying sheet by staggered rivets.
Rivet material
Choosing rivets of the correct material normally depends on the strength needed in the
riveted joint. When this leads to rivets of material different to the sheets being joined it is
important to be aware that electrolytic action may cause corrosion. (See Section 4.2.12.)
Setting and safety
The type of setting tool is usually selected to suit the production environment. The tool
must be cleared of spent materials before setting the next rivet and, in the case of power
operated tools, must not be operated without the mandrel deflector or mandrel collec-
tion system being in position. Safety glasses or goggles should always be worn.
4.2.11 Selection of POP
(or blind) rivets

Joint strength
First assess the tensile strength and the shear-load strength required by the joint, both of
which can be achieved by the correct number and spacing of fastenings, and by choosing
a rivet with a body of the correct material and diameter. The strength columns on the data
pages enable a rivet of the correct strength to be chosen.
Joint thickness
The next stage is to work out the combined thickness of the materials to be joined,
remembering to allow for any air gaps or intermediate layers such as sealants. Then iden-
tify the selected rivet in the size with the necessary grip by consulting the data page. It is
important to do this because a rivet with the incorrect grip range cannot satisfactorily grip
the back of the workpiece or assembly.
Corrosion acceleration (nature of materials)
Finally, follow the general rule that the rivet chosen should have the same physical and
mechanical properties as the workpiece. A marked difference in properties may cause joint
failure through metal fatigue or galvanic corrosion. Corrosion is accelerated by certain
combinations of materials and environments. Generally, avoid contact between dissimilar
metals. The significance of the letters A, B, C and D in the following chart is as follows:
A The corrosion of the metal considered is not accelerated by the contact metal.
B The corrosion of the metal being considered may be slightly accelerated by the contact
metal.
C The corrosion of the metal considered may be markedly accelerated by the contact
metal.
D When moisture is present, this combination of metal considered and contact metals is
inadvisable, even under mild conditions, without adequate protective measures.
Where two symbols are given (for instance B or C) the acceleration is likely to change with
changes in the environmental conditions or the condition of the metal.
Rivet material Contact metal
Nickel Stainless Copper Steel Aluminium Zinc
Copper Steel and
Alloy Alloys
Nickel Copper Alloy A A A A A
Stainless Steel A A A A A
Copper B or C B or C A A A
Steel C C C B A
Aluminium and Alloys C B or C D B or C A
Zinc C C C C C
4.2.12 Design guidelines
Soft materials to hard
A large flange rivet can be used with the flange on the side of the soft material.
Alternatively, POP
LSR type rivets spreads the clamping loads over a wide area so as to
avoid damage to soft materials.
Plastics and brittle materials
For fragile plastics and brittle materials, POP
riveting offers a variety of application solu-

tions. Soft-set/All Aluminium rivets offer low setting loads, whereas both the Peel and
Large flange
Hard
Soft
POP

LSR
Fastenings 299
LSR ranges afford enhanced support to the materials being joined. For stronger plastics,
standard POP
rivets open or closed end products may be used.

Channel section material
An extended nosepiece can be used to reach to the bottom of a narrow channel section (A)
(see figure below). A longer mandrel rivet should be used and the maximum nosepiece
diameter should be equal to that of the rivet flange. Alternatively, the rivet can be set from
the other side (B) or the channel widened to accept a standard rivet and setting tool (C).
Thick/thin sheet
When materials of different thickness are to be fastened, it is best to locate the thicker
plate at the fastened side (A) (see figure below). When the hole diameter in the thinner
plate is large, a large flange rivet should be used (B). When the thinner plate is located at
the fastened side, either use a backing washer (C) or ensure that the diameter of the hole
in the thicker plate is smaller than the one in the thinner plate (D).
Blind holes and slots
The setting of a POP
or blind rivet against the side of a blind hole, or into and against a
milled slot, intersecting hole or internal cavity, is possible because of the expansion of the
rivet body on installation.
A B C D
A B C
Extended
nosepiece
POP

LSR Large flange Peel rivet
Soft set (PAD)
Aluminium (AD/ABS)
Pivot fastening
Use of a special nosepiece will provide a small gap between the rivet flange and the assem-
bly, so providing for pivot action.
Hole diameters
Whilst standard hole diameters are the rivet body diameter plus 0.1 mm, component hole
sizes may not always be this accurate, for example in pre-punched components. In cases
when the hole on the fastened (blind) side is larger, POP
MGR rivets should be selected

because of its superior hole filling characteristics. POP
LSR and POP
Peel rivets are also

possible alternative solutions in these circumstances, especially when working with friable
or fragile materials. Alternatively, when then larger hole is on the flange side, a large
flange rivet should be chosen.
When, however, good hole filling and retained mandrels are used to give high shear and
vibration resistance POP
F series rivets should be specified. (See Section 4.2.13.)

POP

'F' series rivet
MGR rivet Large flange
Pivot nosepiece
Gap
Part
A B C
Fastenings 301
Elevated temperature performance
Elevated temperature strengths will vary from the figures quoted in the data tables. The
following curves are offered for guidance.
99.5% PURE ALUMINIUM
Elevated temperature
100
90
80
70
60
50
100
90
80
70
60
50
50 100 150 C
50 100 150 200 C
50 100 150 200 C
%
stated
strength
value
%
stated
strength
value
100
90
80
70
60
50
%
stated
strength
value
50 100 150 200 C
100
90
80
70
60
50
%
stated
strength
value
100 200 300 400 C
100
90
80
70
60
50
%
stated
strength
value
100 200 300 400 C
100
90
80
70
60
50
%
stated
strength
value
50 100 300 400 C
100
90
80
70
60
50
%
stated
strength
value
50 100 150 200 C
100
90
80
70
60
50
%
stated
strength
value
ALUMINIUM 2.5% MAGNESIUM ALLOY
ALUMINIUM 3.5% MAGNESIUM ALLOY
ALUMINIUM 5% MAGNESIUM ALLOY STAINLESS STEEL
NICKEL COPPER ALLOY
MILD STEEL
COPPER
3
0
2
M
e
c
h
a
n
i
c
a
l

E
n
g
i
n
e
e
r
s

P
o
c
k
e
t

B
o
o
k
4.2.13 POP
F series
F Series: Domed head
Carbon steel (AISI 1006) mandrel: carbon steel
7.608.25 3.803.98
0.901.40
L
9.209.85 4.704.88
1.151.65
L
Rivet code
Tensile
strength
(N)
Shear
strength
(N)
Mandrel
diameter
(mm)
Hole
diameter
(mm)
Nominal
rivet
diameter
(mm)
Grip
range
(mm) (mm)
L
Length
9810.6 1.53.5
12.213.0 3.56.0
1.53.5
3.56.0
11.312.1
14.215.0
16.317.1 6.08.5
4.0
4.8
4.14.2
4.95.0
2800 3000
3000
3600
4600
5700
2800
4250
4000
3550
2.64
3.20
FSD 4010 BS
FSD 4012 BS
FSD 4812 BS
FSD 4815 BS
FSD 4817 BS
F
a
s
t
e
n
i
n
g
s
3
0
3
F Series: Domed head
Aluminium 3.5% Magnesium alloy (5052) mandrel: carbon steel
Note: Shear and tensile strengths are typical values. Joint strengths will be dependent upon the following application criteria:
1. Hole size, 2. Materials to be fastened, 3. Application material thicknesses
It is recommended users conduct their own test(s) to determine suitability for their application(s).
9.209.85 4.704.88
1.151.65
L
Rivet code
Tensile
strength
(N)
Shear
strength
(N)
Mandrel
diameter
(mm)
Hole
diameter
(mm)
Nominal
rivet
diameter
(mm)
Grip
range
(mm) (mm)
L
Length
1.53.5
3.56.0
10.311.1
13.214.0
15.616.4 6.08.5
4.8 4.95.0
1850
2100
3100
2050
2200
2200
3.20
FSD 4811 BS
FSD 4814 BS
FSD 4816 BS
3
0
4
M
e
c
h
a
n
i
c
a
l

E
n
g
i
n
e
e
r
s

P
o
c
k
e
t

B
o
o
k
The following symbols are used in the tables in Sections 4.2.144.2.16.
This is the recommended This is the recommended The shear and tensile figures This is the nominal diameter
maximum thickness of the diameter of the drilled/ quoted are indicative of the of the mandrel and is shown to
materials to be riveted together punched hole which for performance of the rivet under assist in the selection of the correct
assuming a hole diameter reliable rivet setting should standard test conditions. Actual POP
rivet tool nosepiece

indicated as nominal in the be burr free. performance will depend on essential for correct rivet setting
specification tables. The rivet will in most material types and thickness. For performance.
The thickness should include circumstances set satisfactorily safety, testing in the application is
any gap between materials in holes up to 0.1 mm greater recommended for critical
prior to setting. than the nominal quoted. assemblies.
Nominal diameter hole Tensile/shear
performance
Nominal mandrel diameter Maximum riveting
thickness
F
a
s
t
e
n
i
n
g
s
3
0
5
4.2.14 Open type aluminium 3.5% magnesium alloy
Material composition: aluminium 3.5% magnesium alloy; mandrel: carbon steel
TAPD 31 BS
TAPD 33 BS
TAPD 36 BS
SNAD 3050 BS
SNAD 3065 BS
SNAD 3080 BS
SNAD 3100 BS
SNAD 3120 BS
Rivet
code
carbon
steel
mandrel
Nominal
(mm)
Rivet dimensions and limits
(mm)
Nominal
(mm)
Max.
riveting
thickness
(mm)
2.4
3.0
L
Nominal
rivet
body
length
(mm)
Nominal
rivet
dia.
(mm) (N) (N)
3.5
7.5
6.5
2.5
3.1
3.5
5.0
5.0 2.0
4.8
0.8
2.4
8.0
9.0
6.0
5.0
10.0
12.0
3.082.90
1.10 max.
6.35.7
L
2.482.30 5.004.00
0.90 max.
L
550
1000 800
400 1.42
1.83
continued
3
0
6
M
e
c
h
a
n
i
c
a
l

E
n
g
i
n
e
e
r
s

P
o
c
k
e
t

B
o
o
k
6.656.05 3.283.10
1.10 max.
L
Rivet
code
carbon
steel
mandrel
(N) (N)
Nominal
(mm)
Nominal
(mm)
(mm)
Nominal
rivet
dia.
(mm)
Max.
riveting
thickness
(mm)
L
Nominal
rivet
body
length
(mm)
15.0 12.0
4.5 1.6
3.2 6.0
8.0
9.7
11.5 7.9
13.5 9.5
15.0
17.0
3.2 3.3 1200 800 1.83
12.7
11.1
4.8
6.4
SNAD 3150 BS
TAPD 42 BS
TAPD 44 BS
TAPD 46 BS
TAPD 48 BS
TAPD 410 BS
TAPD 412 BS
TAPD 414 BS
TAPD 416 BS
Section 4.2.14 (continued)
F
a
s
t
e
n
i
n
g
s
3
0
7
10.009.00 3.283.10
1.50 max.
L
14.3
16.7
20.7
3.2
7.9
9.5
3.2
4.0
3.3 1200 800 1.83
3.2
4.8
6.4
7.9
18.5
20.0
24.0
6.2
8.0
9.7
11.5
13.5
7.0
8.5
10.5
12.2
17.0 12.7
6.4
4.8
TAPD 418 BS
TAPD 421 BS
TAPD 425 BS
TAPD 44 BSLF9.5
TAPD 46 BSLF9.5
TAPD 48 BSLF9.5
TAPD 410 BSLF9.5
TAPD 412 BSLF9.5
TAPD 416 BSLF9.5
TAPD 54 BS
TAPD 56 BS
TAPD 58 BS
TAPD 510 BS
continued
3
0
8
M
e
c
h
a
n
i
c
a
l

E
n
g
i
n
e
e
r
s

P
o
c
k
e
t

B
o
o
k
Rivet
code
carbon
steel
mandrel
Rivet dimensions and limits (mm)
(N) (N)
Nominal
(mm)
Nominal
(mm)
Nominal
rivet
dia.
(mm)
Max.
riveting
thickness
(mm)
L
Nominal
rivet
body
length
(mm)
14.0 9.5
15.7 11.1
12.7 17.5
18.5
20.2
22.5 17.4
24.7 19.8
8.5
10.5
4.0 4.1 1910 1330 2.29
6.4
4.8
13.5
15.9
TAPD 514 BS
TAPD 512 BS
TAPD 516 BS
TAPD 517 BS
TAPD 520 BS
TAPD 522 BS
TAPD 525 BS
TAPD 56 BSLF12
TAPD 58 BSLF12
8.227.62 4.083.90
1.35 max.
L
F
a
s
t
e
n
i
n
g
s
3
0
9
12.2 7.9
9.5 14.0
15.7 11.1
17.5 12.7
19.4
7.5
9.2 4.8
11.0 6.4
4.0
4.8 4.9
4.1 1910 1330 2.29
2800 2020 2.64
14.7 9.5
16.5 11.1
13.5 19.0
25.5 19.8
12.7 7.9
3.2
14.3 TAPD 518 BSLF12
TAPD 516 BSLF12
TAPD 514 BSLF12
TAPD 512 BSLF12
TAPD 510 BSLF12
TAPD 64 BS
TAPD 66 BS
TAPD 68 BS
TAPD 610 BS
TAPD 612 BS
TAPD 617 BS
TAPD 614 BS
TAPD 625 BS
TAPD 633 BS 32.0 26.2
4.083.90
1.60 max.
L
12.15011.50
4.884.70
1.60 max.
L
9.809.20
continued
3
1
0
M
e
c
h
a
n
i
c
a
l

E
n
g
i
n
e
e
r
s

P
o
c
k
e
t

B
o
o
k
Rivet
code
carbon
steel
mandrel
(N) (N)
Nominal
(mm)
(mm)
Nominal
(mm)
Nominal
rivet
dia.
(mm)
Max.
riveting
thickness
(mm)
L
Nominal
rivet
body
length
(mm)
38.2 32.8
44.5 39.1
9.20
11.0
12.7 7.9
14.5 9.5
16.5
4.8
4.8
4.9
4.9
2710 1950 2.64
2800 2020 2.64
11.1
4.8
6.4
TAPD 6175 BS
a
TAPD 6150 BS
a
TAPD 66 BSLF14
TAPD 68 BSLF14
TAPD 610 BSLF14
TAPD 612 BSLF14
TAPD 614 BSLF14
4.884.70
2.00 max.
L
14.3013.70
8.978.37 4.884.90
1.20 max.
L
F
a
s
t
e
n
i
n
g
s
3
1
1
19.0
25.5 19.8
3.5 6.5
8.0 4.5
10.0 6.0
12.0
14.0
16.0 11.5
18.0 13.5
5.0
6.0
5.1 2600 2200 2.64
22.0 17.5
25.0 20.5
3.5 8.0
20.0 15.5
9.5
7.5
13.5
SNAD 5120 BS
SNAD 5100 BS
SNAD 5080 BS
SNAD 5065 BS
TAPD 625 BSLF14
SNAD 5140 BS
SNAD 5160 BS
SNAD 5180 BS
SNAD 5200 BS
SNAD 5220 BS
SNAD 6080 BS
SNAD 5250 BS
TAPD 617 BSLF14
5.084.85
1.60 max.
L
9.308.70
continued
3
1
2
M
e
c
h
a
n
i
c
a
l

E
n
g
i
n
e
e
r
s

P
o
c
k
e
t

B
o
o
k
Rivet
code
carbon
steel
mandrel
(N) (N)
Nominal
(mm)
(mm)
Nominal
(mm)
Nominal
river
dia.
(mm)
Max.
riveting
thickness
(mm) (mm)
L
Nominal
river
body
length
12.311.7 6.085.85
2.10 max.
L
13.0012.40 6.486.25
2.10 max.
L
3800 3000 3.20
4525 3200 3.66
10.0
12.0
14.0
16.0
18.0
20.0
12.7
12.7
19.8
19.5
26.2
11.5
13.5
15.5
6.4
9.5
7.5
5.5
6.0
6.4
6.1
6.5
SNAD 6100 BS
SNAD 6120 BS
SNAD 6140 BS
SNAD 6160 BS
SNAD 6180 BS
SNAD 6200 BS
TAPD 88 BS
TAPD 816 BS
TAPD 824 BS
F
a
s
t
e
n
i
n
g
s
3
1
3
4.2.15 Open type carbon steel
Material composition: carbon steel; mandrel: carbon steel
Rivet
code
carbon
steel
mandrel
(N) (N)
Nominal
(mm)
Nominal
(mm)
(mm)
Nominal
rivet
dia.
(mm)
Max.
riveting
thickness
(mm)
L
Nominal
rivet
body
length
(mm)
5.845.24 2.882.90
1.32 max.
L
6.35.7 3.082.90
1.1 max.
L
930 715 1.83
1400 1100 1.83
5.3
5.0
6.5
8.0
7.0
9.0
10.0
12.0
2.0
3.5
5.0
2.9 2.8
3.0
2.9
3.1
SNSD 3050 BS
SNSD 3080 BS
TSPD 33 BS
SNSD 3100 BS
SNSD 3120 BS
SNSD 3065 BS
continued
3
1
4
M
e
c
h
a
n
i
c
a
l

E
n
g
i
n
e
e
r
s

P
o
c
k
e
t

B
o
o
k
Nominal
rivet
dia.
Rivet
code
carbon
steel
mandrel
(N) (N)
Nominal
(mm)
(mm)
Nominal
(mm)
Max.
riveting
thickness
(mm)
L
Nominal
rivet
body
length.
(mm) (mm)
6.656.05 3.283.10
1.10 max.
L
1550 1150 1.93
15.0
4.5
6.0
8.0
9.5
11.5
13.5
1.6 5.0
4.8
6.4
7.9
9.5
3.2
1.6
12.0
3.2 3.3
SNAD 3150 BS*
TSPD 412 BS
TSPD 48 BS
TSPD 42 BS
TSPD 44 BS
TSPD 46 BS
TSPD 410 BS
TSPD 52 BS
3.2 7.0 TSPD 54 BS
F
a
s
t
e
n
i
n
g
s
3
1
5
8.127.72 4.083.85
1.35 max.
L
9.829.22 4.884.65
1.60 max.
L
2500 1730 2.29
3500 2620 2.9
8.5 4.8
6.4 10.5
12.2 7.9
14.0 9.5
6.5 2.4
9.0 4.8
11.0 6.4
12.7 7.9
14.5 9.5
16.5 11.1
18.3 12.7
19.0 13.5
7.5 3.2
4.0
4.8
4.1
4.9
TSPD 56 BS
TSPD 58 BS
TSPD 510 BS
TSPD 512 BS
TSPD 63 BS
TSPD 64 BS
TSPD 66 BS
TSPD 68 BS
TSPD 610 BS
TSPD 612 BS
TSPD 614 BS
TSPD 616 BS
TSPD 617 BS
15.9 11.1
12.7 17.6
TSPD 514 BS
TSPD 516 BS
continued
3
1
6
M
e
c
h
a
n
i
c
a
l

E
n
g
i
n
e
e
r
s

P
o
c
k
e
t

B
o
o
k
Rivet
code
carbon
steel
mandrel
(N) (N)
Nominal
(mm)
(mm)
Nominal
(mm)
Nominal
rivet
dia.
(mm)
Max.
riveting
thickness
(mm)
L
Nominal
rivet
body
length
(mm)
9.308.70 5.084.85
1.6 max.
L
3790 2880 2.9 5.0 5.1
6.5 2.5 SNSD 5065 BS
8.0 4.0 SNSD 5080 BS
10.0 6.0 SNSD 5100 BS
12.0 8.0 SNSD 5120 BS
14.0 10.0 SNSD 5140 BS
16.0 12.0 SNSD 5160 BS
18.0 14.0 SNSD 5180 BS
4.0 10.0 SNSD 6100 BS
F
a
s
t
e
n
i
n
g
s
3
1
7
12.311.7 6.085.85
2.1 max.
L
11.1510.55 6.516.40
1.55 max.
L
5500 4200 3.65
6900 5000 3.86
6.0 12.0
6.0
6.4 6.5
6.1
SNSD 6120 BS
8.0 14.0 SNSD 6140 BS
10.0 16.0 SNSD 6160 BS
12.0 18.0 SNSD 6180 BS
3.8 9.5 TSPD 8095 BS
7.6 13.0 TSPD 8130 BS
12.7 18.5 TSPD 8185 BS
a
Body 5% magnesium alloy.
3
1
8
M
e
c
h
a
n
i
c
a
l

E
n
g
i
n
e
e
r
s

P
o
c
k
e
t

B
o
o
k
4.2.16 Closed end type aluminium 5% magnesium alloy
Material composition: aluminium 5% magnesium alloy; mandrel: carbon steel or stainless steel
Rivet
code
carbon
steel
mandrel
Rivet
code
stainless
steel
mandrel
(N) (N)
Nominal
(mm)
(mm)
Nominal
(mm)
Nominal
rivet
dia.
(mm)
Max.
riveting
thickness
(mm)
L
Nominal
rivet
body
length
(mm)
6.305.70 3.283.10
1.10 max.
L
8.227.62 4.083.90
1.50 max.
L
1400 1110 1.63
2220 1640 2.18
6.0
7.5
9.0
11.0
12.0
8.0
9.5
6.4 11.0
6.4
7.9
3.2
4.8
4.8
3.2
1.6
3.2
4.0
3.3
4.1
AD 42 SB
AD 44 SB
AD 46 SB
AD 48 SB
AD 410 SB
AD 54 SB
AD 56 SB
AD 58 SB
AD 42 SS
AD 44 SS
AD 46 SS
AD 48 SS
AD 410 SS
AD 54 SS
AD 56 SS
AD 58 SS
F
a
s
t
e
n
i
n
g
s
3
1
9
9.859.20 4.884.70
1.75 max.
L
3110 2260 2.64
4800 4000 3.66
7.9 12.5
4.8
6.4
4.9
6.5
AD 510 SB
3.2 8.30 AD 64 SB
4.8 10.0 AD 66 SB
6.4 11.5 AD 68 SB
7.9 13.0 AD 610 SB
9.5 14.5 AD 612 SB
12.7 18.0 AD 616 SB
15.9 22.0 AD 620 SB
AD 510 SS
AD 64 SS
AD 66 SS
AD 68 SS
AD 610 SS
AD 612 SS
AD 616 SS
AD 620 SS
6.4 13.0 AD 84 H
9.5 16.0 AD 86 H
13.3412.06 6.486.32
2.51 max.
L
4.2.17 Blind rivet nuts
Blind rivet nuts are especially designed to offer a means of providing a stronger threaded
joint in sheet and other materials and, like blind rivets, only require access to one side of
the workpiece for installation. They are not only a form of captive nut but, unlike some
other types, they also allow components to be riveted together as well providing a screw
thread anchorage. The principle of their use is shown below.
The POP
blind rivet nut is screwed onto the threaded mandrel of the setting tool and is
then inserted into the drilled or punched hole.
The tool is operated, retracting the mandrel. The unthreaded part of the nut expands
on the blind side of the workpiece to form a collar and applies a powerful clenching
force that rivets the components firmly together.
With the POP
nut firmly in position the tool mandrel is simply unscrewed from the nut.
The threaded portion is now ready to act as a secure anchorage point.
Blind rivet nuts are generally available in thread sizes M3 to M12 in numerous combin-
ations of head style, body form and material. Three head styles are available, flat head,
90 countersunk and, for thin gauge materials, a small flange which provides a near flush
appearance without the need for countersinking.
Standard bodies are round with open or closed ends. The closed end prevents the ingress
of moisture or vapour through the bore of the nut. Where a higher torque resistance is
required, body forms may be fully or partially hexagonal (set in a hexagonal hole), or
splined (set in a round hole).
Fastenings 321
4.2.18 POP
Nut Threaded Inserts: application

POP
Nut Threaded Inserts provide a simple and effective way to join materials with the
benefit of an internal thread, in a variety of applications.
POP
Nut Threaded Inserts are the perfect solution for providing high-quality, load-
bearing threads in various materials where alternative methods cannot maintain torque and
pull out loads. POP
Nut Threaded Inserts are suitable for single sheets down to 0.5mm.
POP
Nut Threaded Inserts enable components, which are assembled later in the production
cycle, to be adjusted.
POP
Nut Threaded Inserts are ideally suited to applications where access is only available
from one side of the workpiece.
4.2.19 POP
Nut Threaded Inserts: installation

POP
Nut Threaded Inserts are easily installed from one side of the workpiece without
damaging surrounding surfaces of previously finished or delicate components and are
suitable for use with all materials in todays manufacturing environment:
G
Available in a variety of materials
G
Wide range of styles
G
Complete range of Hand and Power setting tools
G
Bulk and small pack available.
Install/screw the POP
Nut Threaded Insert to the tools threaded mandrel.

Operating the tool then retracts the mandrel. The unthreaded part of the nut then com-
presses to form a collar on the blind side of the workpiece, applying a powerful clenching
force that firmly joins the components together.
Fastenings 323
With the POP
Nut Threaded Insert firmly in position, the tool mandrel is simply

demounted/unscrewed from the insert.
The threaded part of the insert then acts as a secure anchorage point for subsequent
assembly work.
3
2
4
M
e
c
h
a
n
i
c
a
l

E
n
g
i
n
e
e
r
s

P
o
c
k
e
t

B
o
o
k
4.2.20 POP
Nut: Steel
Flat head open end (with knurls)
Thread Description Length Grip Hole Barrel Flange Flange Bulk box Small pack
d (mm) (mm) dia. (mm) dia. (mm) dia. (mm) thickness (mm) quantity quantity
L e D B S
M4
PSFON430 10.0 0.33.0
6.0 5.9 9.0 1.0
10000 500
PSFON440 11.5 3.14.0 10000 500
M5
PSFON530 12.0 0.33.0
7.0 6.9 10.0 1.0
8000 500
PSFON540 15.0 3.14.0 5000 500
M6
PSFON630 14.5 0.53.0
9.0 8.9 12.0 1.5
4000 500
PSFON645 16.0 3.14.5 4000 500
M8
PSFON830 16.0 0.53.0
11.0 10.9 15.0 1.5
2000 250
PSFON855 18.5 3.15.5 2000 250
M10
PSFON1030 17.0 0.53.0
12.0 11.9 16.0 2.0
1500 200
PSFON1060 22.0 3.06.0 1500 200
M12 PSFON1240 23.0 1.04.0 16.0 15.9 22.0 2.0 1500 200
L
B d D
S
e
F
a
s
t
e
n
i
n
g
s
3
2
5
L e D B S
M5 PSFCN530 18.0 0.33.0 7.0 6.9 10.0 1.0 5000 500
M6 PSFCN630 20.5 0.53.0 9.0 8.9 12.0 1.5 2000 500
M8 PSFCN830 25.0 0.53.0 11.0 10.9 15.0 1.5 1500 250
L
B
S
d D
e
Flat head closed end (with knurls)
3
2
6
M
e
c
h
a
n
i
c
a
l

E
n
g
i
n
e
e
r
s

P
o
c
k
e
t

B
o
o
k
Flat head open end hexagonal
L e M
1
M B S
M4 PSFOH430 11.5 0.53.0 6.1 6.0 9.3 1.0 10 000 500
M5 PSFOH530 13.5 0.53.0 7.1 7.0 10.3 1.0 8 000 500
M6 PSFOH630 15.5 0.53.0 9.1 9.0 12.3 1.5 4 000 500
M8 PSFOH830 17.5 0.53.0 11.1 11.0 14.3 1.5 2000 250
M10 PSFOH1040 22.0 1.04.0 13.1 13.0 16.3 2.0 1 500 200
L
B
S
d
M
e
M
1
F
a
s
t
e
n
i
n
g
s
3
2
7
Countersunk head open end (with knurls)
L e D B S
M4 PSKON435 11.5 2.03.5 6.0 5.9 9.0 1.5 10 000 500
M5 PSKON540 13.5 2.04.0 7.0 6.9 10.0 1.5 8 000 500
M6 PSKON645 16.0 2.04.5 9.0 8.9 12.0 1.5 4 000 500
M8 PSKON845 19.0 2.04.5 11.0 10.9 14.0 1.5 2000 250
M10 PSKON1045 21.0 2.04.5 12.0 11.9 14.7 1.5 1500 200
L
B
S
d D 90
e
For further information see the wide range of publications concerning POP
riveting issued by Emhart Teknologies (Tucker Fasteners Ltd) See Appendix 3.

ECCS TC7 TWG 7.10
Connections in Cold-formed Steel Structures

The Testing of Connections with Mechanical

Fasteners in Steel Sheeting and Sections

2
nd
Edition, 2009

ECCS Publication - The Testing of Connections with Mechanical Fasteners in Steel Sheeting and Sections
FREE Download Publications www.eccspublications.eu

The Testing of Connections with Mechanical Fasteners in Steel Sheeting and Sections

2

The Testing of Connections with Mechanical Fasteners
in Steel Sheeting and Sections

N124, 2
nd
edition, 2009

Published by:
ECCS European Convention Ior Constructional Steelwork
publicationssteelconstruct.com
www.steelconstruct.com

All rights reserved. No parts oI this publication may be reproduced, stored in a retrieval sys-
tem, or transmitted in any Iorm or by any means, electronic, mechanical, photocopying, re-
cording or otherwise, without the prior permission oI the copyright owner.

ECCS assumes no liability regarding the use Ior any application oI the material and inIorma-
tion contained in this publication.

Copyright 2009 ECCS European Convention Ior Constructional Steelwork

ISBN: 92-9147-000-91
Printed in Multicomp, Lda - Mem Martins, Portugal

Preface

3

PREFACE

This document intends to provide guidance on the testing oI mechanical Iasteners, used
to Iorm structural connections in cold-Iormed steel sheeting and sections. It updates an earlier
document European Recommendation Ior Steel Construction: The Design and Testing oI
connections in Steel Sheeting and Sections; Publication no. 21 oI the European Convention
Ior Constructional Steelwork, 1983. There are major diIIerences between the current docu-
ment and the previous publication. Since that document was published, Eurocodes have been
introduced which provide more detailed guidance Ior the structural design oI connections.
Whereas the earlier document provided guidance on the design oI connections, this has been
omitted in the present document as it is now covered by the Eurocodes.

This document has been prepared by the ECCS working group TWG 7.10 and approved
by the Technical Committee TC7. The Iinal draIt was circulated to TC7 in May 2007.

The members oI working group TWG 7.10 who contributed to the document are:

A. Belica Luxembourg
A. Arnedo Pena Spain
H. HoImeyer Netherlands
R. Hettmann Germany
J. Kesti Finland
L. Sokol France
K. Kathage Germany
Z. Nagy Romania
R. Podleschny Germany
H. Saal Germany
W. Siokola Austria
T. Toma Netherlands
T. Vrany Czech Republic
R. Pedreschi (chairman) United Kingdom

Facing the Iirst edition, in the present second edition some additions and corrections were
made by an ad-hoc working group consisting oI

K. Kathage Germany
Th. Misiek Germany
H. Saal Germany

The Iinal draIt was circulated to TC7 in September 2009.



4


Contents

5

CONTENTS
1. INTRODUCTION ........................................................................................................ 7
1.1 Aims oI the document ................................................................................................ 7
1.2 Scope .......................................................................................................................... 7
1.2.1 Types oI connections .......................................................................................... 7
1.2.2 Types oI material ................................................................................................ 8
2. DEFINITIONS ............................................................................................................. 9
2.1 Failure modes oI connections ..................................................................................... 9
2.1.1 Failure modes oI connections Ior Iasteners loaded in shear .............................. 9
2.1.1.1 Bearing Iailure oI connection .................................................................... 10
2.1.1.2 Net section tension Iailure ......................................................................... 11
2.1.1.3 Shear oI Iastener ........................................................................................ 11
2.1.1.4 End Iailure oI connection .......................................................................... 12
2.1.1.5 Tilting and pull-out Iailure oI connection ................................................. 12
2.1.2 Failure modes Ior Iasteners loaded in tension .................................................. 12
2.1.2.1 Pull-out Iailure .......................................................................................... 13
2.1.2.2 Fastener Iailure .......................................................................................... 13
2.1.2.3 Pull-over and pull-through ........................................................................ 13
2.2 Characteristic resistance oI connection under static load ........................................ 14
2.3 Derivation oI an empirical Iormula Ior characteristic resistance ............................. 15
2.4 Characteristic resistance oI connections under repeated load determined by test. .. 16
2.5 Alternative method oI estimating the design resistance under repeated load .......... 18
2.6 DeIinition oI Iailure resistance and requirements Ior minimum deIormation ......... 18
2.6.1 DeIinition oI Iailure resistance ......................................................................... 18
2.6.2 DeIormation requirements in tension tests ....................................................... 19
2.6.2.1 Requirements to ensure suIIicient deIormation capacity .......................... 19
2.6.2.2 Requirements to ensure a sound structure in serviceability limit state ..... 19
2.6.3 DeIormation requirements in shear tests .......................................................... 20
2.6.3.1 Requirements to ensure suIIicient deIormation capacity .......................... 20
2.6.3.2 Requirements to ensure a sound structure in service state ........................ 20
2.7 Design resistance oI connections under static load .................................................. 21
2.8 Design resistance oI connections under combined loading ..................................... 22
2.9 Shear Ilexibility ........................................................................................................ 22
3. TEST PROCEDURES ............................................................................................... 25
3.1 General requirements ............................................................................................... 25
3.1.1 Properties oI steel samples ............................................................................... 25
3.1.2 Fastener properties ........................................................................................... 25
3.1.3 Testing equipment ............................................................................................ 25
3.1.4 ManuIacture oI the test samples ....................................................................... 26
3.1.5 Number oI tests ................................................................................................ 26
3.2 Shear tests ................................................................................................................. 27


6

3.2.1 Standard shear test ............................................................................................ 27
3.2.1.1 Single Iastener test .................................................................................... 27
3.2.1.2 Two Iastener test ....................................................................................... 28
3.2.1.3 Measurement oI deIormation .................................................................... 28
3.2.2 Pull out by shear/Iastener Iailure test ............................................................... 29
3.2.3 Test procedure Ior shear ................................................................................... 29
3.3 Tension test .............................................................................................................. 30
3.3.1 Standard tension test arrangements .................................................................. 30
3.3.2 Pull-over/pull-trough resistance ....................................................................... 31
3.3.3 Test method oI determining pull-over/pull-through resistance Ior special
apllications ........................................................................................................ 32
3.3.4 Pull-over/pull-trough resistance Ior special applications obtained Irom standard
tests ................................................................................................................... 33
3.3.5 Pull-out resistance ............................................................................................ 34
3.3.6 Test procedure Ior tension ................................................................................ 36
3.4 Test report and recording oI test results ................................................................... 36
BIBLIOGRAPHY ........................................................................................................ 39


Introduction

7

1. INTRODUCTION

This document replaces the earlier publication by the European Convention oI Construc-
tional Steelwork, European Recommendations Ior the Testing oI Connections in ProIiled
Steel Sheeting and Sections, Publication no. 21, Iirst published in 1977 and revised in 1983.
Since this publication the use oI proIiled steel and cold-Iormed steel has grown considerably,
particularly in the use oI cold-Iormed steel sections in structural Iraming. The current docu-
ment has been restructured to conIorm with guidance in current Eurocodes, which have also
been introduced since 1983. The document has been prepared by the Technical Working
Group 7.10, part oI the ECCS Technical Committee TC7, Cold-Iormed steel structures.

1.1 Aims of the document

This document intends to provide inIormation and guidance on the testing oI connections
using mechanical Iasteners in cold-Iormed steel structures and steel cladding to determine the
design resistance and Ilexibility. Guidance is provided on the test methods, the interpretation
oI the results and the preparation oI test reports. A list oI current reIerences is provided in the
appendices.

1.2 Scope

Important deIinitions:
- Fastener: the connecting element in Iastening;
- Fastening: the interaction oI a Iastener with the surrounding material;
- Connection: a group oI one or more Iastenings.

1.2.1 Types oI connections

The recommendations oI this document cover conventional mechanical connections us-
ing Iasteners, such as selI tapping and selI drilling screws, bolts, rivets and powder-actuated
Iasteners (powder-actuated Iasteners are also reIerred to as cartridge Iired pins).
For Iurther guidance on Iastener speciIication, reIer to ECCS publication no. 42 Me-
chanical Fasteners (currently under review) Ior use in steel sheeting and sections.
Fasteners speciIically excluded Irom the recommendations oI this document are:
- Iasteners Ior use in composite sandwich cladding panels, these are considered in sepa-
rate documents (ECCS no. 115: European Recommendations Ior Sandwich Panels part
I Design, 2001 and ECCS no. 127: Preliminary European Recommendations Ior
Testing and Design oI Fasteners Ior Sandwich Panels, 2009);
- non-standard Iorms oI mechanical Iastener such as selI-piercing rivets, mechanical
clinching and rosettes. These will be the subject oI subsequent Iurther guidance docu-
ments;
- Iasteners in composite structures between steel and other materials;


8

- high strength Iriction grip bolts.
- Iastenings made by gluing or welding.

1.2.2 Types oI material

The recommendations in this document reIer to steel sheets and sections commonly used
in proIiled metal cladding and cold-Iormed steel structural sections and hot rolled steel sec-
tions used as the substrate Ior mechanical Iasteners. The sheet materials may be unprotected
or protected by a suitable coating such as hot dipped galvanising or organic coatings.


Definitions

9

2. DEFINITIONS

This section provides deIinitions oI Iailure modes, test parameters, results and outputs.

2.1 Failure modes of connections

The tests described in this document consider connections subjected to two diIIerent
Iorce actions, shear and tension. These Iorce actions relate to the predominant Iorce in the Ias-
tener itselI. The Iailure mode oI the connection may actually be by tension oI the connected
steel whilst the Iastener itselI is subjected to shear. The Iailure mode oI a particular connec-
tion oI either at least two sheets or at least one sheet and the substrate by a mechanical Ias-
tener is inIluenced by a number oI Iactors:
- type oI Iastener;
- the method oI installation oI the Iastener;
- the tensile strength and thickness oI the steel being connected;
- the lay-up oI the sheets/sections in relation to the Iastener;
- shape oI the proIiled sheet in pull-over tests (see 2.1.2.3).
The range oI tests described in this document considers both the Iorce action on the Ias-
tener itselI and the Iailure mode oI the connection. When carrying out tests on connections it
is important and necessary to record the Iailure mode. This section describes typical Iailure
modes Ior connections. The Iigures in this section are illustrated using screws; however the
Iailure modes are applicable to all types oI Iasteners considered by this document.

2.1.1 Failure modes oI connections Ior Iasteners loaded in shear

II the Iastener is loaded in shear, Iailure oI a connection may occur in either the Iastener
itselI, in the steel sheets being connected or in the attachment oI the Iastener to the steel
sheets. The characteristics oI the various Iailure modes are summarised in table 2.1.



10

Table 2.1: Summary oI Iailure characteristics Ior connections tested in shear
Failure
mode
Elongation
of hole
Piling of
material
Necking/
tearing
Pull-out of
fastener
Location of
failure
bearing
Iailure

either one or
both sheets
net section
Iailure

either one or
both sheets
shear oI
Iastener
Iastener
end
Iailure

either one or
both sheets
tilting and
pull-out

between sheet
and Iastener

2.1.1.1 Bearing failure of connection

Fig. 2.1: Bearing Iailure

Typical characteristics:
- elongation oI holes;
- piling oI material in Iront oI Iastener;
- out oI plane curling oI sheet;
- no necking oI sheet;
- possible diagonal cracks originating Irom material adjacent to Iastener;
- Iailure occurs in either one or both sheets.


Definitions

11

2.1.1.2 Net section tension failure

Fig. 2.2: Net section tension Iailure

- necking across width oI sheet;
- necking across thickness;

2.1.1.3 Shear of fastener

Fig. 2.3: Shear oI Iastener

- Iailure by Iracture or necking oI Iastener.



12

2.1.1.4 End failure of connection

Fig. 2.4: End Iailure oI connection

- elongation oI holes;
- longitudinal tear in the direction oI applied Iorce;
- piling oI material;

2.1.1.5 Tilting and pull-out failure of connection

Fig. 2.5: Tilting and pull-out Iailure

- curling oI sheets;
- Iailure occurs by pull-out oI Iastener Irom lower sheet.

2.1.2 Failure modes Ior Iasteners loaded in tension

II the Iastener is loaded in tension, Iailure oI a connection may occur either in the Iastener
itselI, the steel sheet/section or the attachment oI the Iastener to the steel sheet/section.
Curling pull-out

Definitions

13

2.1.2.1 Pull-out failure

Fig. 2.6: Pull-out Iailure

- thread stripping in screws or bolts and/or in the substrate;
- distortion oI head oI Iastener;
- distortion oI steel substrate;
- Iailure occurs by dislocation oI Iastener Irom substrate.

2.1.2.2 Fastener failure

Fig. 2.7: Fastener Iailure

- Iailure occurs by Iracture oI Iastener or necking oI Iastener shank.

2.1.2.3 Pull-over and pull-through

This type oI Iailure may occur in the connection between sheet and substrate (see 3.3.2).



14

Fig. 2.8: Pull-over and pull-through Iailure

- tearing and distortion oI sheet around head oI Iastener - pull-over;
- Iastener remains connected to the substrate;
- Iailure occurs in material adjacent to the head oI the Iastener;
- some steel may be leIt under the Iastener;
- proIile may disconnect completely Irom Iastener pull-through.

2.2 Characteristic resistance of connection under static load

The characteristic resistance R
k
oI a Iastening Iollows Irom a statistical evaluation oI test
results:

s k R R
m k
=

where:
R
m
mean value oI the adjusted test results R
adf
oI the Iailure resistance deIined in
clause 2.6.1 obtained Irom a minimum oI Iive tests:
Pull through separation
Irom Iastener
Original Iorm oI sheet
Distortion and tearing
Substrate
Some steel may be leIt under the Iastener

Definitions

15

k the coeIIicient which depends on the number oI test observations, the type oI dis-
tribution, the Iractile part and conIidence level chosen. The appropriate value oI k
will be Iound in EN 1990 (Annex D)
s standard deviation
R
adf
is the adjusted test result

R
adf,i
R
obs,i
/
R

in which
R
is the resistance adjustment coeIIicient given by:

|
|
\
|
|
|
\
|
=
cor
cor obs
u
obs u
R
t
t
f
f
. .
o

where:
f
u.obs
the actual measured ultimate resistance oI the material
f
u
nominal ultimate resistance
t
obs.cor
the actual measured material core thickness
t
cor
core thickness oI the material
o 1, iI f
u.obs
~ f
u
and

o 0, iI f
u.obs
s f
u
and
R
obs
is the observed test result

The observed values may be used without application oI (2.2) iI Iailure occurs in the Ias-
tener. The values Ior the material where Iailure occurred should be used with (2.3).
The actual measured basic ultimate resistance f
u.obs
should not deviate by more than 25
Irom the nominal f
u
, basic ultimate resistance i.e. f
u.obs
~ 0,75 f
u
.
The actual measured thickness t
obs
should not exceed the nominal material thickness t
nom

by more than 12. The core thickness, t
cor
, is the bare metal thickness, i.e. the actual thick-
ness minus any protective coatings such as galvanising, (According to EN 1993-1-3 clause
3.2.3 with the usual coating oI 275 g/m
2
, the core thickness can be taken as t
nom
0,04).

2.3 Derivation of an empirical formula for characteristic resistance

For tests Ior which either the thickness t or tensile strength f
u
oI the sheet where Iailure
occurs is varied systematically, an empirical relationship may be derived Ior the resistance oI
the Iastener, provided:
- the same mode oI sheeting Iailure occurs, and
- at least three distinct values oI f
u
t are considered, and
- at least Iive sets oI test observations are available Ior each value oI f
u
t
The empirical Iormula Ior characteristic resistance R
k
, is given by:

2
1
c
c
u k
t f R =



16

CoeIIicients c
1
and c
2
may be estimated Irom regression analysis oI the test data to give
the straight line oI best Iit passing through the origin Ior a plot oI ultimate load per Iastener
against the Iunction f
u
t
c
2
.
The empirical Iormula so derived may be used Ior the prediction oI Iastening resistance
Ior sheeting oI thickness within the range oI Iive sets oI tests.
Similar extrapolation limits shall be applied when predicting Iastening resistances oI diI-
Ierent tensile strengths oI sheeting oI the same thickness. The empirical Iormula may be ap-
plied to both shear and tension test results.
For tests where the Iailure occurs in the sheeting Ior which either the thickness t or the
resistance I
u
oI the sheeting where the Iailure occurs is varied systematically, the characteris-
tic resistance R
k
at intermediate values may be obtained by linear interpolation provided
- the same mode oI sheeting Iailure occurs, and
- Ior pull-over/pull-trough tests with varied thickness t oI the sheet to be Iixed, the
range Ior interpolation is limited to
At
max
0,125 mm Ior t s 0,75 mm
At
max
0,25 mm Ior 0,75 mm t s 2,00 mm
- Ior pull out tests with varied thickness t
1
oI the substrate, the range Ior interpolation
is limited to
At
1,max
0,25 mm Ior t
1
s 0,75 mm
At
1,max
0,5 mm Ior 0,75 mm t
1
s 2,00 mm
At
1,max
2,0 mm Ior t
1
> 2,00 mm iI tests Ior t
1
2,00 mm are perIormed, too, and
At
1,max
1,0 mm Ior t
1
> 2,00 mm iI no tests Ior t
1
2,00 mm are perIormed
- Ior shear tests with varied thickness, the above mentioned ranges oI interpolation
Ior t and t
1
apply.
- Ior tests with varied tensile strengths, the range oI interpolation is limited to
AI
u,max
50 N/mm and
AI
u,max
/ I
u,max
25
Extrapolation to higher thicknesses is not allowed.

2.4 Characteristic resistance of connections under repeated load determined by
testing

The characteristic resistance oI Iasteners subjected to repeated loads can be determined
by the test methods described in this section.
II the Palmgren-Miner-Rule is not applicable, then the actual load spectrum has to be
considered in the tests.
The behaviour oI a Iastening under repeated load shall be shown in a S-N curve (Whler
curve). The characteristic S-N curve should be determined as Iollows:
- carry out the repeated loading tests Ior the deIined load spectrum. In constant amplitude
tests this range is deIined by minimum and maximum load. II the load direction is not
alternating a minimum load oI 10 oI the maximum load is recommended.
- tests should be carried out Ior at least three diIIerent levels oI maximum load.
- Ior each load level at least Iive tests should be carried out.
- Specimen Iailure may be deIined by either a resistance or deIormation criterion.

Definitions

17

Resistance criterion:
- inability to attain the maximum load shall constitute Iailure Ior the test.
DeIormation criterion:
- shear test: deIormation Iailure will be considered to occur when a nominated
maximum slip is exceeded;
- tension test: serviceability Iailure may, when relevant, be taken as the observation
oI cracking around the Iastener.
It may be assumed that a Iorce which is sustained 10
7
times will not cause Iailure at
larger number oI Iorce repetitions (endurance limit).

- Iailure occurs when the test sample can no longer sustain the maximum load or exceeds
the maximum deIormation. The number oI cycles oI load at which this occurs should be
recorded.
- the characteristic number oI cycles to Iailure Ior a particular level oI maximum load Iol-
lows Irom:

log (N
c
) (log N)
m
- ks

where:
N
c
the characteristic number oI cycles to Iailure at a certain Iorce level
N the number oI cycles to Iailure oI a test.
(log N)
m
E log N divided by the number oI tests
s the standard deviation oI log N
k a statistical Iactor which depends on the number oI test observations, the type
oI distribution, the Iractile part and conIidence level chosen (EN 1990, Annex
D).

The characteristic S-N curve can be determined by connecting the log (N
c
)-points by
straight lines.

Comments.
If there is a potential of unscrewing due to repeated loading special tests mav be necessarv to
demonstrate that unscrewing does not occur. The test procedures in this document do not cover this In
such cases manufacturers recommendations or alternative connections mav be useful.

In principle the same type oI specimens can be used as Ior the static test (see section 3).
For the maximum load oI the highest load level a value oI (2/3) R
k
is recommended
where R
k
is the characteristic static resistance oI the Iastening.
In the above R
k
is the characteristic static resistance oI the Iastening. The minimum load
should be close to zero, not more than 10 oI the maximum load.
Loading Irequency:
- Ior tension test maximal 5,0. Hz;
- Ior shear test a maximum oI 5,0 Hz is recommended; higher Irequencies may be used
provided resonance is avoided and that there is no inIluence on the material properties
by the eIIect oI hysteresis load.


18

max
min
time
load

Fig. 2.9: Limits Ior repeated loads

The load ranges are appropriate Ior most applications. To determine the endurance limit
oI the Iastener it may be necessary to reduce the lowest maximum load level.

2.5 Alternative method of estimating the design resistance under repeated load

Comments.
Alternative methods for the determination of the design resistance of fastenings to take account of
repeated loads are available.
In manv situations repeated loads are most likelv due to temperature changes and provided the
ductilitv requirements of 2.8.2 and 2.8.3 are met then it mav not be necessarv to carrv out repeated
load tests.
Design resistance of connection subfected to repeated wind load will be given bv

2
3
2
M
k
d
R
R

=

where.
R
d
design resistance of a fastening loaded bv wind which takes into account the influence of
repeated load
R
k
characteristic static resistance according to Chapter 2.1
M2
partial factor for resistance

2.6 Definition of failure resistance and requirements for minimum deformation

2.6.1 DeIinition oI Iailure resistance

The ultimate resistance oI the Iastening shall be taken as the maximum load recorded dur-
ing the test. With double Iastener connections the ultimate load per Iastener shall be taken as
one-halI the ultimate load oI the connection. In shear tests, it is recommended to deIine a Iail-

Definitions

19

ure load as peak load in a deIormation oI 3 mm. The maximum load within 3 mm deIormation
is illustrated in Fig. 2.10.

Fig. 2.10: Failure limits Ior tests

Comments.
In certain situations, where, under extreme loading condition , for example, earthquake loads the
deformation requirement of 3 mm mav not be applied and the peak load mav be taken as the ultimate
load, provided there is sufficient post vield deformation.

2.6.2 DeIormation requirements in tension tests

2.6.2.1 Requirements to ensure sufficient deformation capacitv

The deIormation at ultimate resistance oI the Iastening has to be suIIicient to ensure re-
distribution and equalisation oI Iorces in connections and to avoid consideration oI secondary
Iorces.

2.6.2.2 Requirements to ensure a sound structure in serviceabilitv limit state

An acceptable serviceability state will be achieved when the recovery oI deIlection aIter
loading up to service load (approximate characteristic resistance multiplied by 0,6) and re-
moval oI the loading has not led to excessive deIormation.

Comments.
To avoid the need to consider secondarv forces the deformation at ultimate resistance of the fas-
tening should be greater than 3 mm.
This limit will generallv be reached when all of the following conditions are fulfilled for profiled
sheeting.
- single sheet thickness < 1,5 mm
- vield resistance of the sheet material > 240 N/mm
2

- difference between the width of the sheet flange through which is fastened and the diameter
of head of the fastener or washer > 14 mm.
- the characteristic resistance of the fastener or anchorage of the fastener is sufficient (no
pull-out and fastener-failure).
3mm 3mm
R
max

R
max



20

The deIormation requirements have also to be satisIied Ior the most critical combination
oI thickness and resistance within the tolerance limits oI the sheet material and Iastener. For
this reason the characteristic resistance oI a connection with insuIIicient deIormation capacity
at Iailure should be greater than the characteristic resistance oI the Iastening.
When:
R
k1
is the characteristic resistance belonging to a Iailure mode with insuIIicient deIorma-
tion capacity.
R
k2
is the characteristic resistance belonging to a Iailure mode with suIIicient deIormation
capacity.
Then it is desirable that:

R
k1
> 1,0R
k2

The Iactor 1,0 is suIIicient because at a tension Iorce with a value about R
k2
the Iastening
still possesses suIIicient deIormation capacity. ThereIore variations in yield stress oI the ap-
plied sheet materials and the scatter in test results are not so signiIicant. When the occasion
arises the corrected values (according to clause 2.2) Ior R
k1
and R
k2
have to be taken. A de-
Iormation less than 3 mm may be suIIicient, provided secondary Iorces are considered.

2.6.3 DeIormation requirements in shear tests

To prevent brittle Iailure (minimum required deIormation) and to ensure a sound struc-
ture under serviceability conditions (allowable deIormation) suIIicient deIormation is required
prior to Iailure. SuIIicient ductility oI the connection is especially important in structures aI-
Iected by thermal stresses (e.g. claddings).

2.6.3.1 Requirements to ensure sufficient deformation capacitv

The deIormation at Iailure oI the Iastening has to be suIIicient to ensure redistribution
and equalisation oI Iorces in connections and to avoid consideration oI secondary Iorces.
A Iastening to the substructure needs a larger deIormation capacity than a seam Iastening
when secondary Iorces and Iorces caused by temperature variation are not considered.

2.6.3.2 Requirements to ensure a sound structure in service state

An acceptable serviceability state will be achieved iI the recovery oI deIlection aIter
loading up to service load (approximate characteristic resistance multiplied by 0,6) and re-
moval oI the loading has not led to excessive permanent deIormation. At the level oI 0,6 oI
the characteristic load the sample should still exhibit linear elastic behaviour.

Comments.
In cladding applications, deformation less that 0,5 mm at failure of the connection might be con-
sidered as a fastening with insufficient deformation capacitv. To avoid the need to consider secondarv
forces in seam fastenings in profiled sheeting a deformation at failure of 0,5mm is sufficient. To avoid

Definitions

21

consideration of secondarv forces and forces caused bv temperature variation for profiled sheeting to
substructure fastenings the following should be adopted. The deformation at failure of the connection
is preferablv larger than 3 mm without excessive hole deformation of the substructure. (The most criti-
cal fastening for this requirement is that with four sheet lavers-combined longitudinal and transverse
lap foint. In clause 3.2 a test specimen is shown to check the requirement). Furthermore when ultimate
load is reached before a deformation of 3 mm then the remaining resistance R
r
should be greater than
the design resistance R
d
(The design resistance is defined in clause 2.4).

Fig. 2.11: DeIormation limits at Iailure

These requirements have also to be satisIied Ior the most critical combination oI thick-
ness and resistance within the tolerance limits oI the sheet material and Iastener. For this rea-
son the characteristic resistance Ior a Iailure mode with insuIIicient deIormation capacity
should be greater than the characteristic resistance Ior a Iailure mode with suIIicient deIorma-
tion.
When:
R
k1
is the characteristic resistance belonging to a Iailure mode with insuIIicient deIorma-
tion capacity.
R
k2
is the characteristic resistance belonging to a Iailure mode with suIIicient deIormation
capacity.
Then it is desirable that:

R
k1
~ 1,3 R
k2

When occasion arises the corrected values (according to clause 2.4) Ior R
k1
and R
k2
have
to be taken. The intention oI the Iactor 1,3 is to take into account the variation in yield stress
oI the applied sheet materials in reality and the scatter in test results Ior R
k1
and R
k2
.
A deIormation capacity less than 3 mm can also be suIIicient, but then secondary Iorces
and Iorces caused by temperature variation should be considered.

2.7 Design resistance of connections under static load

The design resistance oI Iastenings under static load is deIined as the characteristic static
resistance divided by an appropriate partial saIety Iactor:

applied load
R
r
remaining resistance
R
d
design resistance
R
d

R
r

3mm deIormation


22

2 M
k
d
R
R
=

where:
R
d
design resistance oI a Iastening
R
k
characteristic resistance oI a Iastening under static load (clause 2.2)
M2
partial Iactor Ior resistance

Comments.
The partial factor for resistance takes into account.
- statistical base of R
k

and accepted risks,
- uncertaintv in material properties and geometric tolerances.
According to Annex D, EN 1990, partial factor of resistance,
M2
should be taken from the appro-
priate Eurocode provided there is sufficient similaritv between the tests and the usual field of applica-
tion of the partial factor as used in calculations. According to EN 1993-1-3, partial factor of resis-
tance mav be chosen
M2
1,25 for fastenings.

2.8 Design resistance of connections under combined loading

II in practice a Iastening can be loaded by shear and tension at the same time, then the
behaviour oI the Iastening under the combined load has to be considered.

Comments.
Design formulae for combined loadings are given in EN 1993-1-3.

2.9 Shear flexibility

The shear Ilexibility oI a Iastening, c
h
, shall be determined Irom:

c
h
=
1
R
d
/
1
Ea
h
n

where:
a
h
the slip oI a Iastening ( corrected with the elongation oI the test specimen over the
measuring length) at a load equivalent to R
d
/
1
R
d
design resistance oI a Iastening
1
an appropriate Iactor
n number oI test specimens

Comments.
The shear flexibilitv shall be determined according to clause 3.2 for the connection of sheet to
sheet using either the two fasteners or one fastener test.

Definitions

23

For fastening sheets to the substructure with the one-fastener test the appropriate value of
1
is
1,5 the corresponding partial load factor for wind action.
The shear flexibilitv is based on the averaged measured deformation obtained from shear tests.
The level of applied shear force in the connection at which the deformation is taken is the maximum
force at working load level i.e. the design resistance of the fastener divided bv the partial factor of
safetv for the particular load in question. In the two fastener shear test the applied load should be di-
vided bv two to determine the force at working load in a single fastener.


24


Test procedures

25

3. TEST PROCEDURES

3.1 General requirements

3.1.1 Properties oI steel samples

Connection resistance depends on the resistance and the thickness oI the steel sheetings
or sections that are connected by the Iastener, but is limited by the load-bearing resistance oI
the Iastener.
For each batch oI steel that is used, tests shall be carried out to determine the Iollowing
material properties, f
v
(f
p0.2
), f
u
and A.
Where:
f
v
, f
p0,2
yield stress, 0,2 prooI stress
f
u
ultimate tensile strength
A percentage elongation at Iracture
For each steel type at least three tensile tests shall be carried out in accordance with the
technical delivery conditions.
The sheet thickness oI the base material exclusive oI galvanising and coating shall be
measured to an accuracy oI 0,05 mm.

3.1.2 Fastener properties

Nominal Iastener and washer dimensions shall be veriIied to an accuracy oI 0,01 mm.
Tests should also be carried out in accordance with the European Recommendations Ior
Steel Construction Publication no. 42, (currently under review) in order to determine the me-
chanical properties oI the Iastener.

Comments.
For washers, these dimensions refer to the metal and to the sealing material.
For screws and blind rivets, tension tests and shear tests have to be carried out in order to de-
termine the tension resistance and the shear resistance of the fastener.
For screws, drill-drive, thread forming and torque tests have to be carried out.

3.1.3 Testing equipment

The testing apparatus shall be such that the rate oI loading can be controlled, and constant
loads maintained.
The testing apparatus should be calibrated. Applied loads should be measured to an accu-
racy oI at least 1.
Support systems shall be such that the original direction oI loading shall be maintained
throughout the test.



26

3.1.4 ManuIacture oI the test samples

Fasteners shall be Iixed within 1 mm oI the positions speciIied on the test pieces.
All test samples shall be made in a consistent manner with the Iasteners applied in accor-
dance with the producer`s recommendation or the procedure to be adopted on site. This ap-
plies Ior:
- diameter oI pre-drilled holes;
- depth setting control or alternatively tightening torque, depending on the speciIica-
tions oI the manuIacturer Ior threaded Iasteners;
- type oI tool and powder load Ior powder actuated Iasteners;
- strength (steel grade) and thickness oI substrate.

Comments.
Care must be taken to ensure that laboratorv test samples and inspection procedures are not in-
compatible with site practice. Usuallv, the fasteners are mounted bv control of the setting depth.

With the specimens Ior the testing oI riveted connections under shear the mandrel has to
be removed by drilling or pressing out.

Comments.
The mandrel is not a part of the load-bearing svstem of the connection. The mandrel mav get lost
during the lifetime of the connection (for e.g. mav pop out or corrode). Also the location of the break-
ing point of the mandrel mav change depending on tools and production lots. If the mandrel is not re-
moved the test results mav not be used for the design because of the aforementioned reasons and inde-
pendent of this mav not be used for connections with larger clamping si:e.

3.1.5 Number oI tests

The characteristic Iastening resistances are determined using a statistical evaluation oI
test results.
The minimum number oI tests Irom which the characteristic resistance is calculated is in-
Iluenced by the variability oI the results, and may be determined as Iollows:
a) II the test series includes only one nominal thickness or includes several thicknesses
according to clause 2.3, a minimum oI ten tests shall be carried out. II any one oI
these tests results in an ultimate load that diIIers Irom the mean ultimate load by
more than 10 oI the mean at least Iive Iurther tests shall be carried out. Test results
shall be evaluated in accordance with section 2.
b) II the test series includes several nominal thicknesses where the diIIerence between
the actual thickness is at least 0,1 mm, or several nominal tensile strengths where the
diIIerence between the actual tensile strengths is at least 30 N/mm
2
, at least Iive tests
shall be carried out Ior each thickness x strength value. The total number oI tests in
the series shall be at least nine. Test results shall be evaluated in accordance with
section 2.
Test results may only be disregarded iI a reason Ior the deviation is explained and docu-
mented. Such a non-representative result obtained Irom any one test, may be replaced by

Test procedures

27

those Irom two or more equivalent tests. Any test so rejected shall be reported, and the reason
Ior the rejection clearly stated.

Comments.
The minimum number of five tests allows the variabilitv of the fastening to be established. The
characteristic fastening resistance will generallv increase if the number of tests increases, as the frac-
tile factor k (k
o
according to EN 1990) for the test series will be improved. To obtain better results ten
tests per test series is recommended.

3.2 Shear tests

3.2.1 Standard shear test

Two test arrangements are recommended, described in Fig. 3.1 and Fig. 3.2, using either
one or two Iasteners per test sample.

Comments.
Mean ultimate load values per fastener for single fastener connections are normallv little differ-
ent from those obtained for double fastener connections. Differing characteristic resistances mav
however be obtained following statistical evaluation of test results.

3.2.1.1 Single fastener test

2
5
0
m
m
2
5
0
m
m
e
1
e
1
L
0
Extensometer
50 mm

Fig. 3.1: Shear test specimen with one Iastener

II the thickness oI either oI the steel pieces Iorming the connection exceeds 2 mm then
packing pieces or shims are recommended to be used in order to apply an axial load to the test
specimen.


28

3.2.1.2 Two fastener test

In cases where curling or distortion oI the connection is expected (Ior example Ior con-
nections oI two thin sheets such as side lap connections) the test with two Iasteners (Fig. 3.2)
may provide results with a lower variation as the inIluence oI the curling and the distortion oI
the connection is reduced.

2
5
0
m
m
2
5
0
m
m
e
1
e
1
L
0
Extensometer
s
1
50 mm

Fig. 3.2: Shear test specimen with two Iasteners

Table 3.1: Specimen dimensions
Fastener
diameter
d |mm|
Specimen |mm|
w L
0
e
1
p
1
6,5 60 150 30 60
~ 6,5 10d 20d30 5d 10d
Tolerance + 2 + 6 + 1 + 1

3.2.1.3 Measurement of deformation

DeIormation should be measured using an appropriate extensometer.
A single extensometer may be used provided it can be ensured that it is connected on the
central axis oI the connection. The extensometer shall be a single unit Iixed across the width
oI the tests specimens. Alternatively two separate extensometers, one Iixed to each side
(width or thickness) oI the specimen shall be used. In this case deIormation shall be taken as
the average oI the recordings oI the two extensometers.
The required accuracy oI measurement is given in clause 3.2.3.


Test procedures

29

Comments.
Shear flexibilitv (i.e. the shear deformation per unit shear load) obtained from double fastener
tests will be greater than shear flexibilitv values obtained from single fastener tests. If it is necessarv
to calculate the shear flexibilitv, following clause 2.10, for a single fastener from a double fastener
test then the flexibilitv should be based on R
d
is taken as 50 of the design load obtained for the dou-
ble fastener connection and the deformation as recorded during the test.
Deformations obtained bv measurement of the relative travel of the grips of the testing machine
shall not be used for the determination of the deformation of the connection unless.
a) there is no slip of the specimen in the grips of the testing device, and
b) deformation readings are corrected to account for the elastic extension of the specimen be-
vond the specified extensometer length.

3.2.2 Pull out by shear/Iastener Iailure test

For connections with more than one steel sheet pull-out by shear as well as Iastener Iail-
ure can govern. For the resistance and deIormation requirements see section 2.

2
5
0
m
m
2
5
0
m
m
e
1
e
1
L
0
Extensometer
2 x t
1
1 x t
2
50 mm

Fig. 3.3: Test specimen Ior pull out by shear/Iastener Iailure

Comments.
In the case of fastening more than one laver, the deformations decrease. For powder actuated
fasteners and screws with short penetration depth also the pull-out resistance mav decrease.

3.2.3 Test procedure Ior shear

The rate oI loading shall not exceed 1 kN/min and the rate oI deIormation shall not ex-
ceed 1 mm/min.


30

During the test either the rate oI deIormation or the rate oI load application must be con-
trolled to ensure that both oI the above limits are not exceeded.
DeIormation shall be measured to an accuracy oI + 0,2 mm Ior Iastenings loaded in shear.
The test should be continued beyond 3 mm oI relative displacement. The inclination oI
the Iastener should be measured at the unloaded specimens. The elongation oI the holes
should be determined with the disassembled specimen as the diIIerence oI the measured di-
ameter in Iorce direction and that normal to Iorce direction.

Comments.
The specified loading rate is necessarv to ensure adequate time within the range of slip.
For the design of diaphragms and deformations requirements it is essential to have an accurate
knowledge of the shear flexibilitv of the fastening (i.e. the shear deformation per unit shear load).
An inclination of the fastener of more than 10 degrees can be regarded as a failure of the sub-
structure. If a hole elongation of more than 0,1 mm is measured in a sheet, failure can be regarded as
occurred in this sheet.

3.3 Tension test

3.3.1 Standard tension test arrangements

The test Iixture Ior the standard tension test is shown in Fig. 3.4 in principal. The dimen-
sions oI the sheeting are the same Ior all tension tests (see Fig. 3.5, 3.7 and 3.10). DiIIerent
Iailure modes are obtained by using diIIerent material thicknesses.

Fig. 3.4: Set-up and Iixture Ior the standard tension test


Test procedures

31

3.3.2 Pull-over/pull-trough resistance

The tests to determine the pull-over/pull-trough resistance should be carried out with
specimens according to Fig. 3.5.

Fig. 3.5: Pull-over/pull-trough test specimen

The Ilat sheeting, Irom which the specimen is Iabricated, shall be that actually used in
practice Ior the proIiled sheeting. The support material shall be suIIiciently thick, to resist pull
out Iailure oI the Iastener.
For cartridge Iired pins and blind rivets the presence oI the support material is a neces-
sary requirement Ior their proper installation. ThereIore Ior these Iasteners the support mate-
rial must be included in the pull-over/pull-trough test according to Fig. 3.5. DeIormation oI
the support may require to chose an appropriate geometry oI the support, see Table 3.2.
For selI-drilling and selI-tapping screws an alternative pull-over/pull-trough test is shown
in Fig. 3.6, Ior these Iasteners, the support material is not essential Ior this type oI Iailure.

Fig. 3.6: Pull-over/pull-trough test without support material

Comments.
The formed sheeting of the specimen with its fixed dimensions together with the rigid clamping of
its webs serves onlv as a model of real profiled sheeting. This model will give satisfactorv results for
manv steel sheeting profiles.

60


32

3.3.3 Test method oI determining pull-over/pull-through resistance Ior special apllications

The pull-over/pull-trough resistance Ior the special applications oI Fig. 3.7 can be deter-
mined by tests according to Fig. 3.8 and Fig. 3.9. As a simple and conservative alternative the
results Irom the standard tests according to clause 3.3.2 may be reduced by application oI the
coeIIicients given in clause 3.3.4.

Fig. 3.7: Special applications

The test according to Fig. 3.8 takes into account that the pull-over/pull-trough resistance
can be reduced considerably iI the Iastener is inclined, owing to asymmetrical deIormation oI
the support member, such as in case oI thin (t 3 mm) C-, L- or Z-shaped members (condi-
tion 4 oI Fig. 3.7). The standardised dimensions oI the support member shown in Fig. 3.8 will
give satisIactory results Ior most C-, L- or Z-shaped members used in practise.
The alternative tension test according to Fig. 3.9 is appropriate Ior the applications speci-
Iied in conditions 1, 2 and 3 oI Fig. 3.7. For this test the dimensions oI the sheeting and the
support member as well as the adjustment oI the Iasteners should comply with the real condi-
tions on site. The alternative tension test can also be used in cases where detailed inIormation
about the sheet deIormation is required.


Test procedures

33

Fig. 3.8: Pull-over/pull-trough test specimen with Ilexible support member

Fig. 3.9: Alternative pull-over/pull-trough test
Comments.
The test span L
:
depends on the geometrv of the profile. A value of L
:
6b is recommended, where b
is the width of the connected flange of the profile. If failure of the profile occurs, the test span should
be decreased to L
:
3 M
u
/F
:
where M
u
is the ultimate moment capacitv for a single corrugation of the
test sheet and F
:
is the estimated ultimate fastening load.

3.3.4 Pull-over/pull-trough resistance Ior special applications obtained Irom standard tests

The pull-over/pull-trough resistance determined by tests according to clause 3.3.2 is not
appropriate Ior the connection conditions speciIied in Fig. 3.10. The pull-over/pull-trough re-
sistance Ior the applications speciIied in Fig. 3.10 can be estimated by multiplying the pull-
60



34

over/pull-trough resistance, determined by tests according to clause 3.3.2 with the coeIIicients
given in Fig 3.10.

Fig. 3.10: Special applications - CoeIIicients Ior the pull-over/pull-trough resistance

3.3.5 Pull-out resistance

Generally the tests to determine the pull-out resistance oI the Iastener Irom the substrate
should be carried out with specimens according to Fig. 3.11.


Test procedures

35

Fig. 3.11: Pull out test specimen

The support shall be chosen according to the table 3.2.

Table 3.2: Types oI supports
Thickness oI support
material
t
1
6 mm t
1
6 mm
type oI support to be
used in practice
all types hot rolled sections
cold Iormed sections,
hollow sections,
sheeting (when testing
sheeting to sheeting
connections)
standardised support
to be used in test
hot rolled Ilat steel hot rolled angle cold Iormed channel

30

For pull out tests with selI-drilling and selI-tapping screws an alternative test is shown in
Fig. 3.12. In this test the support shall also be chosen according to the table above and the
penetration depth oI the Iastener into the support material should comply with the actual con-
ditions oI application in practice.

40
t
1 40
70
t
1

60
t
1

60


36

Fig. 3.12: Pull out test without sheeting specimen

Comments.
The depth of penetration of the fastener into the substrate has a mafor influence on the resistance
of the connection. The penetration depth itself depends on the total thickness of the sheeting connected
to the support. This thickness is a maximum (4t) for the sheeting fastening where longitudinal and
transverse lap foints coincide. The influence of this should be taken into account bv using specimens in
accordance with Fig. 3.11 and 3.12. It should be noted that asvmmetrical deformation of thin sup-
ported members e.g. C-, L- or Z-shaped members mav increase the pull out resistance, especiallv in
the case of self-tapping screws. Such increase shall be avoided.

3.3.6 Test procedure Ior tension

The rate oI loading shall not exceed 1 kN/min and the rate oI deIormation shall not ex-
ceed 5 mm/min.
DeIormation shall be measured to an accuracy oI + 0,5 mm Ior Iastenings loaded in ten-
sion.

3.4 Test report and recording of test results

The test report shall include the Iollowing data as applicable:
- A reIerence to this recommendation
- Type oI test and test specimen;
- Details oI Iastener and washers, dimensions, material, identiIication data, including
drawings;
- Details oI Iastener application, including pre-drilled hole diameter, tightening torque,
powder load, setting tools;
- Details oI sheeting, including dimensions oI elements, identiIication data, mechanical
properties and thickness according to clause 3.1.1;

Test procedures

37

- For tensile tests: tabulation oI ultimate load values and Iailure mode Ior each speci-
men;
- For tensile tests: tabulation oI maximum load values, load at 3 mm deIormation, de-
Iormation at maximum load as well as hole elongation and inclination oI the Iastener
as characteristics oI the Iailure mode Ior each specimen;
- Load-deIormation curves Ior each specimen;
- Note oI calibration oI the used test equipment or calibration certiIicates;
- Other inIormation as appropriate.


38


Bibliographv

39

BIBLIOGRAPHY

AISI, (1992), Test Methods for Mechanicallv Fastened Cold-formed Steel Connections, Re-
port CF 92-1, American Iron and Steel Institute, Washington, 17 pages.
AISI, (2001), Specification for the Design of Cold-Formed Steel Structural Members, edition.
Washington DC, 162 pages.
Baur S.W. & Suaris W., (2000), Evaluation of cold-formed steel connections attached with
pneumaticallv driven pins. Proceedings of the 15th International Specialtv Conference on
Cold-Formed Steel Design and Construction, University oI Missouri-Rolla, October, pp.
619-633.
Beck, H., and Engelhardt, M., (2002), Net Section Efficiencv of Steel Coupons with power ac-
tuated Fasteners, ASCE, Journal oI Structural Engineering, 128:1, pp. 12-21.
Birch, R.S., and Alves, M., (2000), Dvnamic Failure of Structural Joint Svstems, Thin Walled
Structures, Vol. 36, pp 137-154.
Boston R.M. & Pask J.W., (1978), Structural Fasteners and their application. Published by
BCSA, London.
Bryan, E.R., (1993), The design of bolted foints in cold-formed steel sections, Thin-walled
Structures, Vol. 16, pp. 239-262.
Bryan E.R., Sedlacek G., Toma A.W. & Weynand K., (1990), Evaluation of test results on
connections in thin-walled sheetings and members in order to obtain strength functions
and suitable model factors. Background report to Eurocode 3 Annex A, August.
Davies J.M. (2000), Light gauge construction. developments in applications, Progress in
Structural Engineering and Materials, January-March, Vol. 2 (1), pp. 26-33.
ECCS TC 7 (1983), The design and testing of connections in steel sheeting and sections,
European Recommendations for Steel Construction. ECCS publication no. 21. Brussels.
ECCS TC 7, (1983), Mechanical fasteners for use in steel sheeting and sections. European
Recommendations for Steel Construction. ECCS publication no. 42. Brussels.
ECCS TC 7 & CIB W56, (2001), European Recommendations for sandwich panels, Part 1.
Design. CIB Report publication 257/ ECCS publication no. 115, Ch.7, pp. 117-131. Rot-
terdam/Brussels.
ECCS TC 7 & CIB W56, (2009), Preliminarv European Recommendations for testing and
design of fastenings for sandwich panels, CIB Report publication 320/ ECCS publication
no. 127, Rotterdam/Brussels.
EN 1993-1-3, (2006) (European standard). Eurocode 3. Design of steel structures. Part 1.3.
General rules, Supplementarv rules for cold-formed thin gauge members and sheeting.
Ch.8, Design oI joints. Published by CEN (Comite Europeen de Normalisation), Brussels.
ENV 1993-1-3 (European pre-standard), (2001), Eurocode 3. Design of steel structures. Part
1.3. General rules, Supplementarv rules for cold formed thin gauge members and sheeting.
Ch.8. Published by CEN (Comite Europeen de Normalisation), Brussels.
ENV 1993-1-4 (European pre-standard), (1996), Eurocode 3. Design of steel structures. Part
1.4. General rules, Supplementarv rules for stainless steels. Published by CEN, Brussels.
Fan L.X., Rondal J. & Cescotto S., (1997), Numerical simulation of lap screw connections,
Thin-walled structures, Vol. 29 (1-4), pp. 235-241.


40

Groberndt H. & Kniese A., (1975), Untersuchung ber Querkraft- und Zugbeanspruchungen
sowie Folgerungen ber kombinierte Beanspruchungen von Schraubenverbindungen bei
Stahlprofilblech-Konstruktionen, Stahlbau 44, pp. 289-300 and 344-351.
Hancock G.J., (1997), Light gauge construction, Progress in Structural Engineering and Ma-
terials, September, Vol. 1 (1), pp. 25-30.
Holz R. & Kniese A., (1988), Stahltrape:profile mit Obergurtbefestigung, Stahlbau 57, pp.
71-79.
Hettmann R. & Saal H., (1999), Tragverhalten von Jerbindungen des Leichtbaus, 11. DASt-
Forschungskolloquium Karlsruhe.
Hettmann R., (2007), Statische Aus:ugstragfhigkeit gewindeformender Schrauben in
Metallkonstruktionen, Berichte der Versuchsanstalt Ir Stahl, Holz und Steine der
Universitt Fridericiana in Karlsruhe, 5. Folge HeIt 16.
Klee S. & Seeger T., (1979), Jorschlag :ur vereinfachten Ermittlung von :ulssigen Krften
fr Befestigungen von Stahltrape:blechen, VerIIentlichung des Instituts Ir Statik und
Stahlbau der Technischen Hochschule Darmstadt.
LaBoube R.A & Sokol, M.A., (2002), Behaviour of screw connections in residential con-
struction, ASCE Journal oI Structural Engineering 128:1, pp. 115-118.
LaBoube R.A., Yu W.W. & Carril J.L., (1995), Serviceabilitv limit state for cold-formed steel
bolted connections, In Proceedings oI the Third International Workshop: Connections in
Steel Structures III (Behaviour, Strength and Design). Trento, Italy, pp. 403-411.
LaBoube R.A., Yu W.W. & Jones M.L., (1998), Spacing of connections in compression
flanges of built-up cold-formed beams, Proceedings oI the 14th International Specialty
ConIerence on Cold-Formed Steel Design and Construction, University oI Missouri-Rolla,
October, pp. 563-578.
Lixin Fan., (1996), Contribution to steel sheet connections of screws, blind rivets and car-
tridge fired pins, Doctoral thesis at the University oI Liege, Mecanique des materiaux Sta-
bilite des constructions Mecanique des structures.
Maiola C.H., Malite M., Gonalves R.M. & Neto J.M., (2002), On the structural behaviour of
bolted connections in thin sheets and cold-formed steel members, Proceedings oI the 3rd
European ConIerence on Steel Structures, Coimbra, Portugal, September, pp. 957-966.
Mahaarachchi D. & Mahendran M., (2002), Improved numerical modelling of crest-fixed
steel cladding svstems subfect to pull-through failures, Proceedings oI the 3rd European
ConIerence on Steel Structures, Coimbra, Portugal, September, pp. 645-655.
Mahaarachchi, D. & Mahendran, M., (2000), Pull-through failures of crest-fixed steel clad-
dings, Proceedings oI the 15th International Speciality ConIerence on Cold-Formed Steel
Design and Construction, University oI Missouri Rolla, U.S.A., pp. 635-646.
Mahendran, M., (1997), Review of Current Test Methods for Screwed Connections, ASCE
Journal oI Structural Engineering, 123:3, pp. 321-325.
Mahendran M. & Tang R.B., (1998), Pull-Out Strength of Steel Roof and Wall Cladding Svs-
tems, ASCE Journal oI Structural Engineering 124:10, pp. 1192-1201.
Mahendran, M. & Mahaarachchi, D., (2000), Cvclic pull-out strength of steel roof and wall
cladding svstems, Proceedings oI the 15th International Specialty ConIerence on Cold-
Formed Steel Design and Construction 19.-20. October 2000. University oI Missouri
Rolla, U.S.A., pp. 647-658.

Bibliographv

41

Misiek Th. & Saal H., (2008), Durchknpftragfhigkeit der Jerbindungen von
Aluminiumtrape:profilen und Aluminiumwellprofilen bei Befestigung im anliegenden Gurt,
Stahlbau 77, pp. 515-523.
Nicolaas T. & Toma A.W., (2000), Design tools and new applications of cold-formed steel in
buildings, New connection methods, (ECSC research project: 7210-PG-059). TNO report
2000-CON-R4028, DelIt.
Rogers C.A. & Hancock G.J., (1998), Bolted connection tests of thin G550 and G300 sheet
steels, ASCE Journal oI Structural Engineering 124:7, pp. 798-808.
Rogers C.A. & Hancock G.J., (1999), Screwed connection tests of thin G550 and G300 sheet
steels, ASCE Journal oI Structural Engineering 125:2, pp.128-136.
Rogers C.A. & Hancock G.J., (1998), Bearing design of thin sheet steel screwed connections,
Proceedings oI the 14th International Specialty ConIerence on Cold-Formed Steel Design
and Construction, University oI Missouri-Rolla, October 1998, pp. 481-494.
Rogers C.A. & Hancock G.J., (2000), Failure modes of bolted-sheet-steel connections loaded
in shear, ASCE Journal oI Structural Engineering 126:3, pp. 288-296.
Pekoz T., (2001), Design of cold-formed steel screw connections, Proceedings oI the 10th In-
ternational Specialty ConIerence on Cold-Formed Steel Design and Construction, Univer-
sity oI Missouri-Rolla, pp. 575-588.
prEN 1990: (2002), (provisional European standard). Eurocode 0. Basis of structural design.
Annex D (informative). Design assisted bv testing, Published by CEN (Comite Europeen
de Normalisation), Brussels.
S136-94, (1994), Cold-formed steel structural members, Canadian Standards Association,
Rexdale (Toronto), Ontario, Canada.
Serrette, R. & Lopez, (1996), Performance of self-tapping screws in lap-shear metal-to-metal
connections, Proceedings oI the 13th International Speciality ConIerence on Cold-Formed
Steel Design and Construction 17.-18. October 1996. University oI Missouri Rolla, U.S.A.,
pp. 615-622.
Standards Australia (SA), (1996), AS4600 Cold-formed steel structures code, Sydney, Aus-
tralia.
Stark J.W.B. & Toma A.W., (1982), Connections in thin-walled structures, In: Developments
in Thin-Walled Structures-1, edited by Rhodes J. and Walker A.C. Ch. 5. Applied Science
Publishers LTD., Essex, pp. 159-203.
Steenhuis C.M., Stark J.W.B. & Gresnigt A.M., (1997), Cost-effective connections, Progress
in Structural Engineering and Materials, September, Vol. 1 (1), pp. 18-24.
Strnad, M., (1981), Fatigue strength of screwed fastenings in thin sheet steel components, The
Structural Engineer, Vol. 59B, no.3, pp. 33-39.
Tan, S.H., Rhodes, J., (1994), Semi-rigid connections in cold-formed, thin-walled structures,
Journal oI Construction Steel Research 28, pp. 279-293.
Toma, A., Sedlacek, G. & Weynand, K., (1993), Connections in cold-formed steel, Thin-
walled Structures, 16, pp. 219-237.
Toma T. & Wittemann K., (1994), Design of cold-formed purlins and rails restrained bv
sheeting, Journal oI Constructional Steel Research, Vol. 31 (2-3), pp. 149-168.
Walter W., (2002), Flexible nodes in calculations of thin-walled structures, Proceedings oI
the 3rd European ConIerence on Steel Structures, Coimbra, Portugal, pp. 1189-1198.


42

Wieland H., (1988), Korrosionsprobleme in der Profilblech- und Flachdach-
Befestigungstechnik, BeIestigungstechnik Bau oI SFS Stadler BeIestigungs- und
UmIormtechnik AG. Heerbrugg, Switzerland, pp. 1-41.
Yu Wei-Wen, (2000), Cold-formed steel design, New York: John Wiley & Sons, Inc., pp.
430-525.
ZadanIarrokh F. & Bryan E.R., (1992), Testing and design of bolted connections in cold-
formed steel sections, Proceedings oI the 11th International Specialty ConIerence on Cold-
Formed Steel Design and Construction, University oI Missouri-Rolla, pp. 625-662.


Riveted Cold Form Steel Stru Articles Test

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Riveted Cold Form Steel Stru Articles Test

Uploaded by

Copyright:

Available Formats

Standard and/or project Stage TC

1700MPa has been proposed.

1700MPa has been proposed.

preload and tightening torque for a threaded

lifting and lowering torques for a power

safe load for a variety of riveted joints;

thickness or overlap for simple adhesive lap

whether the joint should be permanent or

loads in the fastener or power screw and the

strength at the operating temperatures

12.3 Power screws

Figure 12.10 Buttress thread.

the load is carried equally by the rivets;

no combined stresses act on a rivet to cause

the shearing stress in a rivet is uniform across the

the load that would cause failure in single

the bearing stress of the rivet and plate is dis-

the tensile stress is uniform in the section of

or blind riveting is a technique which enables a mechanical fastening to be made when

and other brands of

blind rivet data tables.

(or blind) rivets

riveting offers a variety of application solu-

rivets open or closed end products may be used.

MGR rivets should be selected

LSR and POP

Peel rivets are also

F series rivets should be specified. (See Section 4.2.13.)

rivet tool nosepiece

Nut Threaded Inserts: application

Nut Threaded Inserts: installation

Nut Threaded Insert to the tools threaded mandrel.

Nut Threaded Insert firmly in position, the tool mandrel is simply

riveting issued by Emhart Teknologies (Tucker Fasteners Ltd) See Appendix 3.

The Testing of Connections with Mechanical

You might also like