Beruflich Dokumente
Kultur Dokumente
Steel Construction
Journal of the Australian Steel Institute
Two existing ASI publications on connections contain recommended design models for the bracing cleat
connection. Both these recommended design models are satisfactory for bracing members subject to
tension but the latest research indicates that these design models are unconservative for bracing members
subject to compression.
ASI has issued updated design models for a range of connections in Connections Design Series—Part 1–
2007 and Part 2–2009, which do not contain a design model for the bracing cleat connection. This
publication is intended to offer a recommended design model for review by interested parties before its
future inclusion in Connections Design Series—Part 1. The format of this publication is similar to the Design
Guides in the Connections Design Series.
For the purposes of this publication, light bracing cleat connections are defined as unstiffened cleats that
connect light bracing members to beams or columns. Typical applications include roof and wall bracing in
one-way rigid/one-way braced frames such as industrial building structures. This publication may also be
used to design unstiffened cleats which connect diagonal members in trusses to main chord members.
Bracing cleats for such applications may be cleats cut from plate or cut from square edge flat bar. Cleats
with only a single bracing member connected are the primary focus of the publication, but guidance is also
given for applications where multiple bracing members connect to the same gusset plate.
Electronic copies of Steel Construction are available from the members’ section of the ASI website. These PDFs
may be freely downloaded by members for their personal use. Financial corporate members of the ASI may add
these PDFs to their company intranets but in the event of resignation from the ASI, the PDFs must be deleted. The
ASI permits members to quote excerpts from Steel Construction in their technical reports provided the journal is
referenced as the source.
Page No.
1 Introduction ...................................................................................................................................................... 3
14 References......................................................................................................................................................46
Appendix A
Comparison of experimental results by Stock with design capacities from recommended
Design Model of Section 10 ............................................................................................................................48
Appendix B
Limcon outputs for design examples ...............................................................................................................49
R. T. COLLINS
Engineering Systems Pty Limited
1 INTRODUCTION
For the purposes of this publication, light bracing cleat connections are defined as unstiffened cleats that
connect light bracing members to beams or columns. Typical applications include roof and wall bracing in
one-way rigid/one-way braced frames such as industrial building structures. This publication may also be
used to design unstiffened cleats which connect diagonal members in trusses to main chord members.
The publication does not apply to heavy bracing members and their end connections (which are discussed
in Section 12) or to knee braces in portal frame buildings.
Light bracing members are deemed to include the following member types (see Table 1):
—flat bars
—rods
—angles (single or double)
—circular hollow section members up to 219 mm diameter
—square or rectangular hollow section members up to 200 mm maximum dimension
Bracing cleats for such applications may be cleats either cut from Grade 250 or Grade 350 plate to
AS/NZS 3678 (Ref. 9) or cut from square edge flat bar Grade 300 to AS/NZS 3679.1 (Ref. 10).
Light bracing cleats may be one of three forms shown in Figure 1 as follows:
Bracing cleats may be subject to tension or compression force applied through the connected brace
member.
A recommended design model for light bracing cleats was published by the Australian Institute of Steel
Construction in 1994 (Sections 4.11 and 5.11 of Reference 1—known commonly as the ‘Green Book’). The
recommended design model in Reference 1 ignored the effect of eccentricity on the design of the cleat and
contained what subsequent research showed to be an unconservative assumption for the buckling strength
of a cleat subject to compression.
The Australian Institute of Steel Construction also published in 1996 a publication dealing with light bracing
cleats with hollow section bracing members connected to them (Sections 6 to 9 of Reference 2—known
commonly as the ‘Blue Book’). The bracing members explicitly considered were as follows (see Figure 2):
—flattened end circular hollow section bolted to the bracing cleat (Fig. 2(a));
—cleat welded to cap plate to form a tee, cap plate welded in turn to a hollow section brace member
(Fig 2(b));
—cleat fitted into a slot in a hollow section member and welded along the slot (Fig. 2(c)).
The basis of the design model for all member types in Reference 2 was the Kitipornchai, Al-Bermani and
Murray paper (Reference 3) which had recommended taking account of the eccentricity in such connections
of brace members to cleats through a specifically formulated yield line approach based on test results. An
effective length factor of 0.5 was used in the design method.
Subsequently, as more information became available, the Australian Steel Institute drew attention to the
fact that the design method based on Reference 3 may be unconservative (Ref. 4 of March 2004) and then
issued an Advisory Note (Ref. 5 of December 2005) regarding the bracing cleat connection. Engineering
Systems also issued similar cautionary advice (Ref. 17).
‘… The problem arises because of the assumption that “the connection may be treated as two
eccentrically connected cleat components whose ends are fixed and prevented from sway”.
The connection types to which this advisory applies are the slotted tube, welded tee end and flattened
end connections. Unless restrained against sidesway, each of these connections deflects laterally as
it is loaded in compression, developing a plastic hinge in each plate at a fraction of the section
compression capacity. The real capacity of the connection is very much less than would be computed
assuming the presence of lateral restraint or the absence of eccentricity. The problem is exacerbated
The small eccentricity occurring when a stiff member is connected to a gusset plate (e.g. channel web
bolted to gusset plate) has traditionally been ignored in the design of simple bracing connections and
this is permissible in some cases because most of the eccentricity moment acts on the stiff bracing
member and only a small bending moment acts on the flexible cleat component. There is an important
difference between this situation and that with hollow section bracing connections. In these
connections there are two flexible components bolted together—“eccentrically connected cleats”. The
problem is that the eccentricity moment is shared between the two flexible plates and plastic hinges
develop at a very low load unless there is lateral restraint. The cleat assembly deflects sideways
during loading. Eccentrically connected cleats should not be designed as a concentric column even
when a large effective length factor is used. It is necessary to apply the existing design code rules for
combined bending and compression (AS 4100 Section 8—Members Subject to Combined Actions).
Software is available to perform the necessary code checks.
Recommendations
• Do not use an eccentric hollow section bracing connection for a short compression member
unless it is stiffened against sidesway. A concentric connection should be used if there is no
sidesway stiffening.
• Design eccentric hollow section bracing connections taking eccentricity into account by rigorous
application of design code rules for combined bending and compression—do not use the
method in ‘Design of Structural Steel Hollow Section Connections.’
This publication is intended to provide a revised recommended design model for the bracing cleat
connection that addresses the issue of eccentricity in cleats subject to both tension and compression and
also addresses the issue of the buckling capacity of the cleat when subject to compression. The
recommendations in this publication are formulated in terms of the provisions of AS 4100 (Ref. 6).
The issuing of this publication means that all Sections of Reference 1 will then have been updated (the
updating of the other sections of Reference 1 having been discussed in References 7 and 8) and means
that Sections 6 to 9 in Reference 2 can continue to be used provided the design of the bracing cleat is
modified using the recommendations of this publication.
TABLE 1
LIGHT BRACING MEMBERS—END CONNECTION DETAILS
Hollow sections—fitted
fin plate bolted to cleat
Hollow sections—
fabricated tee section
bolted to cleat
1 Bracing cleats should be detailed as approximately rectangular shapes in order to reduce marking-off
and cutting times.
2 For single members, use a square edge flat bar as the cleat component wherever possible. Where
several bracing members connect to the same cleat (as in Fig. 1), use approximately rectangular
shaped plates wherever possible.
3 Welds to supporting members should be 6 or 8 mm fillet welds, both sides of bracing cleat, wherever
possible. Check the economics of using 10 mm fillet welds with the fabricator before using or else use
a full penetration butt weld.
4 Bolts should be M16/M20/M24 bolts in 8.8/S bolting category. 2 mm oversize holes should be used in
general. Where oversize or slotted holes are used, the provisions of AS 4100 Clause 14.3.5.2 in
respect of the use of the applicable bolt category and the use of plate washers need to be complied
with. See also References 11 and 12 for a discussion of the requirements of Clause 14.3.5.2 of
AS 4100 in respect of hole sizes.
5 Centreline of bolt group, centreline of bracing cleat and centre of weld should all coincide wherever
possible for the single member bracing cleat.
6 Members to which bracing is connected are often the first members erected in order to establish
stability for the overall frame. The bracing connections should hence be simple so as to allow rapid
installation of the bracing members.
7 Because bracing cleats project from supporting members, damage may occur during transport and
erection. For light bracing cleats, the cleats need to be able to be returned to their correct position if
damaged. If they cannot be easily returned to their original position, they should be removed and
replaced.
8 Minimum edge distances should be as follows—refer to Figure 4 and Table 9.6.2 of AS 4100
ae3 — Flat bar component M16 bolt 20 mm
M20 bolt 25 mm
M24 bolt 30 mm
since a square edge flat bar will have a rolled edge
— Plate component M16 bolt 28 mm
M20 bolt 35 mm
M24 bolt 42 mm
since it will not be known during the design phase whether the plate will be sheared, hand
flame cut, machine flame cut, sawn or planed along the cleat
ae1 — Flat bar or plate component M16 bolt 28 mm
M20 bolt 35 mm
M24 bolt 42 mm
since it will not be known during the design phase whether the flat bar or plate will be
sheared, hand flame cut, machine flame cut, sawn or planed when cut to length
NOTE: It is recommended that more than the minimum edge distance be used for bracing cleats to tension
members with only one row of bolts because the bolt tearout limit is associated with quite large deformations
and any accidental reduction of the edge distance during fabrication or erection could result in failure of the
connection.
Clause 9.1.4 (b) (iii) of AS 4100 specifies that the minimum design action at the ends of axially loaded
tension or compression members shall be ‘0.30 times the member design capacity except that for threaded
rod acting as a bracing member with turnbuckles, the minimum tensile force shall be equal to the member
design capacity’. This requirement hence defines the minimum design force on a bracing cleat at the end of
a bracing member.
‘Members or components meeting at a joint shall be arranged to transfer the design actions between
the parts and, wherever practicable, with their centroidal axes meeting at a point. Where there is
eccentricity at joints, the members and components shall be designed for the design bending
moments which result … Eccentricity between the centroidal axes of angle members and the gauge
lines for their bolted end connections may be neglected in statically loaded members, but must be
considered in members and connection components subject to fatigue loading.’
From this Clause, it is clear that any eccentricity between the bracing cleat and the bracing member must
be designed for when designing the member and the bracing cleat, irrespective of whether the member is in
tension or compression. However, the Clause allows that angle bracing members may be designed by
ignoring the eccentricity between the centroid of the member and the centroid of any bolt group at the end
connection. This is because generally the centroid is too close to the unconnected leg to allow bolts to fit.
Shear lag effects due to not all elements of the brace member being connected at the end connection are
allowed for in the case of tension members in Section 7 of AS 4100.
Note that eccentricity about both axes of the cleat may need to be considered—eccentricity about the minor
axis due to the bracing member or its end connection coming in on one side of the bracing cleat Fig. 5(b),
and eccentricity about the major axis due to the centroid of the bracing member or its end connection not
coinciding with the centreline of the bracing cleat—although usually the two coincide about the major axis,
see Figure 5(a).
For eccentricity about the minor axis of the bracing cleat, the following applies in terms of the requirements
of AS 4100 as discussed in Section 4 (refer to Table 1 for details of end connections to members):
concentric loading — double angles as bracing members, one each side of bracing cleat;
eccentric loading — single angle as a bracing member connected one side of bracing cleat;
— rod with fin plate connected to one side of bracing cleat;
— flat bar as bracing member connected to one side of bracing cleat;
— flattened end CHS bracing connected to one side of bracing cleat;
— fitted fin plate to hollow section bracing member connected to one side;
— tee plate end connection to hollow section bracing member with projecting
plate connected to one side of bracing cleat.
The notional eccentricity involved for the eccentrically loaded connections is taken to be that shown in
Figure 5 (b), being the distance from the centre of the bracing cleat to the centre of any connection plate to
the bracing member, except for the single angle bracing member where the eccentricity is taken as the
distance from the centroidal axis of the angle to the centre of the bracing cleat (see Figure 6). The angle
bracing member should be designed as eccentrically loaded in compression and designed for tension using
the k t term of Section 7 of AS 4100.
Conventional practice has been to design the bracing cleats for no eccentricity when subject to either
tension or compression (see Reference 1). As noted in Section 4, this is not in accordance with AS 4100
Section 9.
(a) British practice as described in Section 9 of Reference 13 for angle, channel and tee brace members
subject to compression or tension is to ignore the theoretical eccentricity provided that the members
themselves are designed in accordance with BS 5950-1 Clause 4.7.10. For hollow section bracing
members, reference is made to CIDECT publications.
(b) American Institute of Steel Construction Specification (Reference 14) Clause J1.7 states that the
centre of gravity of bolt or weld groups should coincide with the centre of gravity of the member
unless provision is made for any eccentricity that would otherwise result. This is not required to be
(c) The AISC et al Hollow Section Connections Manual (Ref. 15) recommends that no eccentricity be
allowed for the welded tee and slotted plate connection in hollow section brace members if the brace
member is subject to tension. However, the same members when subject to compression are
recommended to be designed for any eccentricity between the centroids of the two plates involved, as
in Figure 5(b). This is the same approach as was used at Reference 2.
(d) More recent work on end connections to bracing members subject to compression (References 16 to
21), all recommend that bracing cleats subject to compression be designed for any eccentricity.
(i) for bracing cleats subject to tension, no eccentricity is allowed for in design based primarily on
References 13 and 15 and past practice in Australia (Refs 1 and 2) and since any eccentricity present
does not cause a stability issue such as occurs in compression;
(ii) for bracing cleats subject to compression, the eccentricity suggested by Figure 5(b) and Figure 6 is
designed for, based on advice given in References 15 to 21.
(iii) the resulting design moment due to the eccentricity is divided between the bracing cleat and the
attaching cleat on the brace member or the brace member itself (if there is no cleat) on the basis of
the relative stiffness of each. If the bracing cleat and the attaching cleat on the brace member (as in
Figures 2(b) and (c)) are of similar length and thickness, this will result in a 50% assignment to each
cleat. In the case of angle brace members and flattened hollow section brace members, the stiffness
of the brace member is much greater than the stiffness of the bracing cleat and a minimum of 10% of
the design moment due to the eccentricity is recommended to be used, based on advice given in
Reference 17 (which is in turn based on modelling done using Microstran).
The end connections to both types of bracing members typically comprise bolts through the members
themselves into the bracing cleat. The design checks required are then:
which all can be based on the guidance given in Sections 9 and 10 of this publication.
The end connections to the rod type of bracing members typically comprise bolts through the fin plate at the
end of the rod into the bracing cleat. The design checks required are then:
—bearing capacity of the bolts in the fin plate to the bracing member (tension only)
—end tearout capacity of the fin plate to the bracing member (tension only)
—block shear capacity of the fin plate to the bracing member (tension only)
—design capacity in tension of the fin plate in terms of Section 7 of AS 4100
—design capacity of the fillet welds connecting the rod to the fin plate
The first four design checks can be based on the recommended design model of Sections 9 of this
publication. Rod members are only used in tension so that Section 10 of this publication does not apply to
these members.
A recommended design model for this type of end connection is given in Section 7 of Reference 2. This
recommended design model may be used except for Section 7.3.1.3 for a member subject to compression
for which the recommended design model in Section 10 of this publication should be substituted.
Welded Tee Connection to Hollow Section (see Table 1 and Figure 2(b))
A recommended design model for this type of end connection is given in Section 8 of Reference 2. This
recommended design model may be used except for Section 8.3.2.3 for a member subject to compression
for which the recommended design model in Section 10 of this publication should be substituted.
Fin Plate to Slotted Hollow Section (see Table 1 and Figure 2(c))
A recommended design model for this type of end connection is given in Section 9 of Reference 2. This
recommended design model may be used except for Section 9.3.1.3 for a member subject to compression
for which the recommended design model in Section 10 of this publication should be substituted.
However, the isolated bracing cleat forms a significant proportion of the overall number of bracing
connections in steel structures and a design model for this form of connection can form the basis for more
complex gusset plate connections, as will be discussed in Section 11.
The recommended design model for a bracing cleat subject to a design tension force is primarily based on
the provisions contained in AS 4100 (Ref. 6) since suitable provisions are available.
The design checks for bolts in shear, bolts bearing on the bracing cleat and plate tearout in the bracing
cleat are based on Clause 9.3.2.1 of AS 4100 and Section 3 of Handbook 1 (Ref. 11). The design of the
bracing cleat subject to axial tension is based on Clause 7.2 of AS 4100 and Section 5.4 of Handbook 1. As
noted in Section 5 of this publication, the bracing cleat is not designed for any eccentricity when subject to
tension force, and neither is the weld connecting the bracing cleat to the support.
The design check for block shear failure of the bracing cleat is based on the recommendations in
Section 5.4 of Handbook 1 (Ref 11), while the design check for the weld group is based on Section 11 of
Handbook 1.
The design check for local effects on the supporting member have been taken from the following sources :
(a) for a bracing cleat connected to the web of an I-section—the discussion in Section 5.11 of
Reference 1;
(b) for a bracing cleat connected to the face of a rectangular or square hollow section—the
recommendations in Section 6.2.1 of Reference 2.
The recommended design model for a bracing cleat subject to a design compression force is also primarily
based on the provisions contained in AS 4100 (Ref. 6) where suitable provisions are available.
The major issues in designing the bracing cleat for combined bending and compression are:
The recommended design model in Reference 1 did not allow for any eccentricity and used an effective
length of 0.70 × the distance from the bolt closest to the support, based on guidance in Reference 16. This
effective length was based on the assumption of fixed/fixed supports in Fig. 4.6.3.2 of AS 4100 (Ref 6).
Reference 2 adopted a design approach based on analytical and experimental work by Kitipornchai et al
(Ref. 3). Kitipornchai et al identified that the connection at the bracing cleat could fail in either a sway mode
or in a non-sway mode and concluded that for most practical connections, the connection at the bracing
cleat approximated a fixed-fixed connection. They derived six possible collapse mechanisms involving
plastic hinges which led to the formulation of the design model which was used in Reference 2. The
Reference 2 stated for their design models to be valid that ‘the design of the bracing cleat eccentrically
loaded in compression assumes that both ends of the connections are fixed against sway’. The eccentricity
used in the design models was the distance between the centres of the two cleats and the effective length
used was 0.5 × the ‘distance between supports of the connection’ which was taken as the distance between
the face of the supporting member and the face of the in-coming bracing member.
Subsequently, some concern about the recommended design model of Reference 2 became evident,
particularly about whether the fixed-fixed/no sway assumption was correct. Advisory information was issued
by Engineering Systems (Ref. 17) and the Australian Steel Institute (Refs 4 and 5). Reference 5 also
suggested that Section 8 of AS 4100 be used to account for the effects of combined bending and
compression.
Following from these concerns, Clifton and el Sarraf (Ref. 18) proposed a design model for the bracing cleat
for combined bending and compression on the basis of a failure mode involving sway. The design model
was based on the provisions of NZ 3404 (Ref. 19) which has a number of sections common to those in
AS 4100, including Sections 6 and 8. This design model amplifies the design moment to allow for second-
order effects, increases the theoretical design eccentricity by 3 mm and uses an effective length factor of
1.0.
A relatively straightforward design method is given in Reference 15 (AISC et al method) which uses the
following assumptions:
(1) design allows for eccentric loading, the resulting bending moment being split equally between the
bracing cleat and the cleat attached to the bracing member;
(4) design for compression is based on the column curve for hollow sections used in Reference 23;
(5) design for combined bending and compression is based on a linear interaction equation taken from
Reference 23.
Recently, Stock (Ref. 20) carried out an experimental study of hollow section braces connected to bracing
cleats using fabricated tee-section end connections (Fig. 1(b)). Stock compared the experimental capacities
obtained from his testing with the design capacities obtained using the above AISC et al design model (Ref.
15) and reported that the ratio of experimental values to the predicted design capacities ranged from 1.05 to
1.39, with a mean of 1.20. This comparison was carried out with a capacity factor of 1.0 so that the mean
value increases to 1.33 for a capacity factor of 0.9. The AISC et al design method (Ref. 15) is hence
conservative based on this evidence.
In 2008, Khoo, Perera and Albermani (Ref. 21) conducted twelve (12) full-scale tests on hollow section
struts with eccentric and concentric end connections. They confirmed the sway collapse mechanism for
eccentric cleats and proposed a two-step design method. Their method was compared with other design
methods and results from finite element analysis and was found to be accurate. They concluded the
following:
(i) the Reference 2 design model grossly overestimated the connection capacity since it excludes the
sway mechanism;
(ii) the Reference 18 method was excessively conservative because it increased the theoretical
eccentricity by 3 mm and used the elastic section modulus rather than the plastic section modulus
when evaluating the design moment capacity of the bracing cleat due to concerns about bracing
cleats used in seismic applications;
(iv) the two-step design method proposed by Khoo et al was found to give good agreement with the
experimental results.
Stock (Ref. 20) proposed a design method using Clause 8.4.2.2 of AS 4100 for combined bending and
compression, using eccentricity as discussed in Section 5 split equally between the two connected plates
and an effective length factor of 1.0 with no moment amplification for second-order effects. Using averaged
test results for each of four specimen types tested by him, he obtained a ratio of experimental results to
design capacity for this approach of 1.07—a slightly conservative result. This result was achieved with a
capacity factor of 0.9 .
Stock also proposed an advanced method for considering combined bending and compression using
AS 4100. This advanced method accounts for sway using a second-order moment amplification term based
on Clause 4.4.2 of AS 4100 and an effective length factor of 1.0. A closed form expression for the critical
compression force is then obtained by using Clauses 6.2 and 8.4.2.2 of AS 4100.
It should be noted that the moment amplification method given in Section 4 of AS 4100 is essentially for
frame-type beam and column structures and its application to small-scale problems such as the end
connection of a brace member has to be considered questionable.
The method adopted in the recommended design model in Section 10 of this publication has the following
features:
(b) an effective length of 1.20x the distance between the supporting member and the end of the bracing
member is used based on one end being rotation fixed, translation fixed (supporting member) and one
end being rotation fixed, translation fixed (from Figure 4.6.3.2 of AS 4100);
(e) an αb of 0.5 is used to select the column curve for design for axial compression;
(f) Clause 8.4.4.1 of AS 4100 is used to determine the out-of-plane design capacity for the bracing cleat
subject to combined bending and compression.
The proposed method is believed to be sufficiently conservative and is similar to the AISC et al method
(Ref. 15), the Limcon method (Ref. 22) and the Stock simplified method (Ref. 20) in many respects. A
comparison of the results obtained from the proposed design model and the experimental results obtained
by Stock are contained in Appendix A of this publication.
The design checks for bolts in shear and bolts bearing on the bracing cleat are based on Clause 9.3.2.1 of
AS 4100 and Section 3 of Handbook 1 (Ref. 11). In the case of compression on the bracing cleat, only
tearout between bolt holes need be considered in terms of Clause 9.3.2.4 but no check is included as this is
considered an unlikely and non-critical failure possibility.
The design of the bracing cleat subject to axial compression is based on Clauses 6.2 and 6.3 of AS 4100
(using a value for αb of 0.5 for ‘other sections’ in Table 6.3.3(1) of AS 4100) and Section 5.4 of Handbook 1.
As noted above, the bracing cleat is designed for eccentricity when subject to compression force, resulting
in a bending moment about the minor axis of the bracing cleat. Clause 8.4.4.1 of AS 4100 is used for the
assessment of the out-of-plane design capacity of the bracing cleat subject to axial compression and
bending.
The design check for the weld group is based on Section 11 of Handbook 1 for a fillet weld group subject to
a compressive axial force and bending moment due to the eccentricity on the bracing cleat which is
assumed to also apply to the weld.
The design check for local effects on the supporting member are based as follows:
(a) for a bracing cleat connected to the web of an I-section—the discussion in Section 5.11 of
Reference 1;
(b) for a bracing cleat connected to the face of a rectangular or square hollow section—the
recommendations in Section 6.2.1 of Reference 2.
The recommended design model in this publication meets the requirements of Clause 9.1.4 of AS 4100 that
any design method be based on a recognised method and experimental evidence. The philosophy adopted
in this publication is the same as that espoused in Handbook 1 (Ref. 11) being as follows:
(i) take into account overall connection behaviour, carry out an appropriate analysis in order to
determine a realistic distribution of forces within the connection;
(ii) ensure that each component or fastener in each action path has sufficient capacity to transmit
the applied action;
(iii) recognise that this procedure can only give a connection where equilibrium is capable of being
achieved but where compatibility is unlikely to be satisfied, and therefore ensure that the
connection elements are capable of ductile behaviour.
Design is based on determining N des (the design capacity of the bracing cleat) which is the minimum of the
design capacities N a, N b, N c, N d, N e, N f obtained from DESIGN CHECK Nos 1 to 6 inclusive.
9.6 DESIGN CHECK NO 6—Design capacity of supporting member locally at cleat location (N f)
Design is based on determining N des (the design capacity of the bracing cleat) which is the minimum of the
design capacities N g, N h, N i , N j obtained from DESIGN CHECK No.s 7 to 10 inclusive.
10.2 DESIGN CHECK NO 8—Design capacity of bracing cleat under eccentric axial compression
(N h)
10.4 DESIGN CHECK NO 10—Design capacity of supporting member locally at cleat location (N j)
NOTE: The design checks of both Part A and B do not take account of the parameters of the connected bracing
member (if a flat bar or angle) or of the end connection to the bracing member. The design of the member or its end
connection must consider a number of aspects which may affect the overall design capacity of the total connection
(see Section 6).
The design models contained within this publication are considered to be applicable only to connections
which are essentially statically loaded. Connections subject to dynamic loads, earthquake loads or fatigue
applications may require additional considerations.
where
Na = nbkr(φVdf)
nb = total number of bolts in bracing cleat
kr = reduction factor to account for the length of a bolted lap splice connection—given in Table 11 of
Handbook 1 (Ref. 11) but generally 1.0 for normal connections for light bracing members
φVdf = design capacity of a single bolt in shear or bearing on the bracing cleat
= [φVfn or φV fx; 0.9 × 3.2tidffui]min
fui = tensile strength of cleat component (see Table 2)
ti = thickness of cleat component
df = bolt diameter
φVfn = design capacity in shear for bolt with threads included in the shear plane
= 59.3 kN for M16 bolt, 8.8/S category
92.6 kN for M20 bolt, 8.8/S category
133 kN for M24 bolt, 8.8/S category
φVfx = design capacity in shear for bolt with threads excluded from shear plane
= 82.7 kN for M16 bolt, 8.8/S category
129 kN for M20 bolt, 8.8/S category
186 kN for M24 bolt, 8.8/S category
NOTE: Threads are normally assumed to be included in the shear plane unless specifically detailed so that they can
be considered excluded (refer to Design Guide 1—Ref. 12). Design capacities of bolts in shear taken from Table 10
of Handbook 1 (Ref. 11).
where
Nb = nbk r(φVen)
nb = total number of bolts in bracing cleat
kr = reduction factor to account for the length of a bolted lap splice connection—given in Table 11 of
Handbook 1 (Ref. 11) but generally 1.0 for normal connections
φVen = 0.9aetifui
fui = tensile strength of cleat component (see Table 2)
ti = thickness of cleat component
ae = minimum distance from edge of a hole to the edge of the cleat measured in the direction of the
axial tension force plus half the bolt diameter (d f)
= [ae1; ae2 ]min
= ae1
ae1 = (ae – 1) (Figures 8(a), 8(b), 8(c))
ae2 = (s p – 0.5dh – 1) (Figures 8(a), 8(c))
dh = hole diameter s p = bolt pitch
ae = distance from centre-line of hole to edge (Figures 8(a), 8(b), 8(c))
where
Nc = φ[A ntfui + 0.6Agvf yi] Based on discussion in Section 5.4 of Handbook 1 (Ref. 11)
φ = 0.75 (see Handbook 1—Reference 11)
fui = tensile strength of cleat component (see Table 2)
fyi = yield stress of cleat component (see Table 2)
Agv = gross area subject to shear (see Figure 10)
Agt = gross area subject to tension
Ant = net area subject to tension = Agt – allowance for holes (diameter = dh) (see Figure 10)
ti = thickness of cleat component
The general definitions of areas Ant, A gv are shown in Figure 9 (taken from Figure 50 in Handbook 1).
A nt = (l t – (n h – 1)d h )t i
A gv = l vt i
n h = no. of holes in vertical line
d h = hole diameter
Specific formulae for Ant, Agv for typical bracing cleat configurations are shown in Figure 10.
A nt = (a e3 – 0.5d h )t i A nt = (a e3 – 0.5d h )t i
A gv = (a e + s p )t i A gv = (a e + 2s p )t i
A nt = (s g – d h )t i A nt = (s g – d h )t i
A gv = (a e + s p )t i A gv = (a e + 2s p )t i
where
Nd = [Ndy; Ndf]min
Ndy = design capacity in axial tension – yield check
= 0.9bitifyi
Ndf = design capacity in axial tension – fracture check
= 0.9 × 0.85 × 1.0 × (bi – ngdh)tifui
bi = width of component
fyi = yield stress of component (Table 2)
ti = thickness of component
fui = tensile strength of component (Table 2)
ng = number of holes of diameter dh across component at gauge sg (either 1 or 2 in Figure 10)
dh = hole diameter
Bracing cleats may be cut from Grade 250 plate or may be cut from a Grade 300 flat bar. Yield stress (fyi )
and tensile strength (fui ) for usual cleat thickness (ti ) are as set out in Table 2.
TABLE 2
BRACING CLEAT MATERIAL STRENGTHS
FILLET WELDS
where
Ne = 2Lwφvw (fillet welds both sides of cleat)
Lw = length of weld to support (Figure 4 = bi/cosθ, bi = component width)
φvw = design capacity of fillet weld per unit length of weld (Table 23 of Handbook 1, also see
below)
Welds are fillet welds and, for economy, should be sized to be single pass welds if possible—this generally
means 6 mm or 8 mm fillet welds, although some welding procedures will allow 10 mm single pass fillet
welds to be deposited. Check individual situations using 10 mm fillet welds with fabricator before specifying.
Welds would normally be weld category SP.
BUTT WELDS
No design check is necessary provided weld complies with AS 4100 (Ref. 6) and AS/NZS 1554.1 (Ref. 24)
and weld is SP category. Compliance with these two Standards ensures that weld metal strength matches
or exceeds the strength of bracing cleat. Supporting member should be Grade 300 for Grade 250/300
bracing cleat and Grade 350 for Grade 350 bracing cleat for assumption to be valid.
No assessment necessary.
9.6.2 Connection to web of I-section column from one side or web of I-section beam from one side
where
f yc t c2 [bi + 12t c ]
Nf = 0. 9
d1 cos θ
d1 = clear depth between flanges of I-section
tc = web thickness of I-section
fyc = yield stress of I-section web
bi = width of component
ti = thickness of component
θ = angle of inclination defined in Figure 4
Based on discussion in Section 5.11 of Reference 1, which is in turn based on simplified elastic analysis in
Blodgett (Ref. 25).
Design force is any out-of-balance tension force on the two bracing cleats assuming both cleats come in at
same location and are approximately of equal lengths.
where
Nf = [Nf1; Nf2]min
Nf1 = design capacity due to face shear capacity
= 0.9 × 2 × (0.6fycbvtc)/sinθ
Nf2 = design capacity due to face yielding capacity
=
(
f yc t c2 2η + 4 1 − β ) for
bc
≤ 30
1− β cos θ tc
fyc = yield stress of hollow section face
bc = width of hollow section face to which bracing cleat is attached
bv = width of projection of bracing cleat to face
= bi / cosθ (Fig. 13)
tc = thickness of face of hollow section
θ = angle of inclination (Fig. 13)
β = ti/bc
η = bv/bc
ti = thickness of bracing cleat
bi = width of bracing cleat
where
Nf = [Nf1; Nf2]min
Nf1 = design capacity due to shear capacity
= 0.9 × 2 × (0.6fycbvtc)/sinθ
Nf2 = design capacity due to yielding capacity
2 (1 + 0.25η)f (n′) provided η ≤ 4.0
= 5.0fyc t c
cos θ
fyc = yield stress of circular hollow section
bv = width of projection of bracing cleat
= bi / cosθ (Fig. 14)
tc = thickness of circular hollow section
θ = angle of inclination (Fig. 14)
do = diameter of CHS section
η = bv / d o
f (η′) = 1.0 (CHS column unloaded or in axial tension)
= 1 + 0.3n ′ − 0.3n ′ 2 for n ′ < 0 (CHS column in axial compression)
⎛ N 0* p M 0* ⎞
n′ = ⎜⎜
− + ⎟
⎝ N s′ M s′ ⎟⎠
N 0* p = design preload (i.e. additional axial compression in CHS column at connection other than that
required to maintain equilibrium with bracing cleat expressed as an absolute value)
N s′ = fyAg (for CHS column)
where
Ng = nbkr(φVdf)
nb = total number of bolts in bracing cleat
kr = reduction factor to account for the length of a bolted lap splice connection—given in Table 11 of
Handbook 1 (Ref. 11) but generally 1.0 for normal connections for light bracing members
φVdf = design capacity of a single bolt in shear or bearing in the bracing cleat
= [φVfn or φVfx; 0.9 × 3.2tidffui]min
fui = tensile strength of cleat component (see Table 2)
ti = thickness of cleat component
df = bolt diameter
φVfn = design capacity in shear for bolt with threads included in the shear plane
= 59.3 kN for M16 bolt, 8.8/S category
92.6 kN for M20 bolt, 8.8/S category
133 kN for M24 bolt, 8.8/S category
φVfx = design capacity in shear for bolt with threads excluded from shear plane
= 82.7 kN for M16 bolt, 8.8/S category
129 kN for M20 bolt, 8.8/S category
186 kN for M24 bolt, 8.8/S category
NOTE: Threads are normally assumed to be included in the shear plane unless specifically detailed so that they can
be considered excluded (refer to Design Guide 1—Ref. 12). Design capacities of bolts in shear taken from Table 10
of Handbook 1 (Ref. 11).
where
φM syi
Nh =
[e d + φM syi / N cc ]
Based on Clause 8.4.4.1 of AS 4100 for bracing cleat subject to compression force and bending moment
about minor axis.
where
φM syi = 0.9f yi bi t i2 / 4
ed = kee
(I / L )cleat ≥ 0.10
ke =
(I / L )cleat + (I / L )angle + (I / L )cleat at other end
Icleat = bi t i3 / 12
Lcleat = di in Figure 15
ed = kee
(I / L )cleat
≥ 0.10
ke =
(I / L )cleat + (I / L )chs + (I / L )cleat at other end
Icleat = bi t i3 / 12
Lcleat = di in Figure 16
(d) Fabricated tee or fitted fin plate end connection to hollow section brace member (Figure 17)
FIGURE 17 END DETAILING FOR FABRICATED TEE AND FITTED FIN PLATE
END CONNECTIONS TO HOLLOW SECTION BRACE MEMBERS
ed = 0.5e
e = (ti + tj) / 2
where
Ncc = 0.9αcNs ≤ 0.9 Ns (AS 4100 Clause 6.3.3)
Ns = nominal section capacity in axial compression
= bitifyi (AS 4100 Clause 6.2 taking kf = 1.0 and
assuming all holes are filled with bolts)
bi = width of cleat component
ti = thickness of cleat component
fyi = yield stress of cleat component (see Table 2)
r = t i2 / 12
(b) Hollow section as bracing member. End connection either type shown in Figure 2
—flattened end circular hollow section
—tee section welded to cap plate
—fin plate fitted to slot in hollow section
Assumption is that both ends are fixed and can sway laterally
FIGURE 18 EFFECTIVE LENGTH OF CLEAT
FILLET WELDS
where
2Lw φv w
Ni = ⎡ 2ed ⎤
⎢1 + ⎥
⎣ ti ⎦
Welds are fillet welds and, for economy, should be sized to be single pass welds if possible—this generally
means 6 mm or 8 mm fillet welds, although some welding procedures will allow 10 mm single pass fillet
welds to be deposited. Check individual situations using 10 mm fillet welds with fabricators before
specifying. Welds would normally be weld category SP.
BUTT WELDS
No design check is necessary provided weld complies with AS 4100 (Ref. 6) and AS/NZS 1554.1 (Ref. 24)
and weld is SP category. Compliance with these two Standards ensures that weld metal strength matches
or exceeds the strength of bracing cleat. Supporting member should be Grade 300 for Grade 250/300
bracing cleat and Grade 350 for Grade 350 bracing cleat for assumption to be valid.
No assessment necessary.
10.4.2 Connection to web of I-section column from one side or web of I-section beam from one side
where
f yc t c2 [bi + 12t c ]
Nj = 0. 9
d1 cos θ
d1 = clear depth between flanges of I-section
tc = web thickness of I-section
fyc = yield stress of I-section web
bi = width of component
ti = thickness of component
θ = angle of inclination defined in Figure 4
Based on discussion in Section 5.11 of Reference 1, which is in turn based on simplified elastic analysis in
Blodgett (Ref. 25).
Design force is any out-of-balance compression force on the two bracing cleats assuming both cleats come
in at same location and are approximately of equal lengths.
where
Nj = [Nj1; Nj2]min
Nj1 = design capacity due to face shear capacity
= 0.9 × 2 × (0.6fycbvtc)/sinθ
Nj2 = design capacity due to face yielding capacity
=
(
f yc t c2 2η + 4 1 − β ) for
bc
≤ 30
1− β cos θ tc
fyc = yield stress of hollow section face
bc = width of hollow section face to which bracing cleat is attached
bv = width of projection of bracing cleat to face
= bi/ cosθ (Fig. 21)
tc = thickness of face of hollow section
θ = angle of inclination (Fig. 21)
β = ti/bc
η = bv/bc
ti = thickness of bracing cleat
bi = width of bracing cleat
where
Nj = [Nj1; Nj2]min
Nj1 = design capacity due to shear capacity
= 0.9 × 2 × (0.6fycbvtc)/sinθ
Nj2 = design capacity due to face yielding
2 (1 + 0.25η)f (n ′) provided η ≤ 4.0
= 5.0f yc t c
cos θ
fyc = yield stress of circular hollow section
bv = width of projection of bracing cleat to face
= bi / cosθ (Fig. 22)
tc = thickness of circular hollow section
θ = angle of inclination (Fig. 22)
do = diameter of CHS section
η = bv/do
f (η′) = 1.0 (CHS column unloaded or in axial tension)
= 1 + 0 . 3n ′ − 0 . 3n ′ 2 for n ′ < 0 (CHS column in axial compression)
⎛ N 0* p M 0* ⎞
n′ = − ⎜⎜ + ⎟
⎟
⎝ N s
′ M ′
s ⎠
N 0* p = design preload (i.e. additional axial compression in CHS column at connection other than that
required to maintain equilibrium with bracing cleat expressed as an absolute value)
N s′ = fyAg (for CHS column)
Whitmore (Ref. 26) investigated the stress distribution in gusset plates and concluded that an effective
section obtained by a dispersion of 30 degrees from the first to the last line of bolts (as in Fig. 23) could be
used. Stresses in the gusset plate can be calculated by dividing the axial force in the bracing member by
the Whitmore effective cross-sectional area. Support for the Whitmore approach may be found in
References 27 to 29, as reviewed in Reference 1.
The design approach recommended by Whitmore fits easily into the recommended design models of
Sections 9 and 10, since each brace member can be considered as being connected to a bracing cleat
whose area is equal to the ‘Whitmore section’ multiplied by the cleat thickness. For the case of three
members attached to a single gusset plate, each with at least two rows of bolts in the end connection—as in
Figure 24—the Whitmore sections are readily determined using Figure 23. If there is any overlap resulting
from the projection at 30 degrees, the Whitmore sections so obtained should be reduced to avoid any such
overlap.
Note that the Whitmore recommendation of a 30 degree spread cannot be applied directly to the common
two bolt end connections used at the ends of many light bracing members, but the Whitmore section would
still be defined as the distance between the bolt columns in such cases.
Figure 25 shows some typical end connections where two or three bracing members connect to the same
gusset plate. A suggested assignment of effective areas is made so that for each bracing member, part of
the gusset plate is assigned to allow the use of the Recommended Design Models of Sections 9 and 10 of
this publication.
Another alternative is to carry out a finite element analysis of the gusset plate—a far more involved
procedure.
For the weld group to the supporting member, the brace member forces need to be resolved into six (6)
possible design actions as shown in Figure 26 so that the weld group and the supporting member are
subject to:
—a resultant moment M y* due to eccentricity of Fx* and Fz* force components relative to weld group
centroid.
Revised DESIGN CHECKS 5 and 9 can then be derived from Section 4.10 of Handbook 1 (Ref. 11) as
follows:
(v ) + (v ) + (v )
*
≤ φv w = φ(0.6fuw t t )
v res = * 2 * 2 * 2
x y z
v y* = 0 since Fy* = 0, M z* = 0
Fz* M y* (± 0.5L w )
v z* = − since M x* = 0
2L w I wy
where: I wy = L3w / 6
Existing DESIGN CHECKS 6 and 10 can be used for the supporting member by using the resultant value of
Fz* from Figure 27.
Heavy bracing connections using I-sections as part of the bracing system (as in Fig. 28) do not fall within
the scope of this publication.
The American Institute of Steel Construction sponsored extensive research into this type of connection,
which included that reported at Reference 29, and recommend the use of the Uniform Force Method (see
Refs 30, 31). The essence of the Uniform Force Method is to select the geometry of the connection such
that bending moments do not exist at three connection interfaces:
The connection can then be designed for shear force and tension or compression as appropriate. Full
details of the method may be found at Reference 30.
Design parameters
a e2 = 70 – 22/2 –1= 58 mm a e1 = 35 – 1 = 34 mm
a e2 = 70 – 22/2 –1= 58 mm a e1 = 35 – 1 = 34 mm a e3 = 45 mm
df = 20 mm d h = 22 mm s p = 70 mm
φV fn = 92.6 kN k r = 1.0 nb = 3
Member design capacity = minimum of: (AS 4100, Clause 7.2 k t = 0.85)
= 392 kN
Adopt N t* = 120 kN
SHS Column:
150 × 150 × 6.0 SHS
Grade C350
Component:
90 × 10 mm Flat bar
Grade 300
–235 mm long at CL
Bolts:
2-M20 8.8/S Category
22 dia. Holes
Welds:
6 mm E48XX/WSOX
Continuous fillet weld both
sides of component
Category SP
Brace:
100 × 100 × 10 EA
Grade 300
Holed for bolts to brace cleat
Single angle brace members are not commonly specified for compression braces. However, an angle is
used in this example to illustrate the methodology of the design model.
Angle brace 100 × 100 × 10 EA, Grade 300, actual thickness = 9.5 mm
effective length = 4.0 m I yy = 696 × 10 3 mm3
φN cy = 68 kN (concentric loading) distance to centroid = 28.2 mm
Welds 6 mm fillet weld, both sides of cleat, weld category SP, E48XX weld metal
L w = 117 mm
φv w = 0.978 kN/mm
Adopt N c* = 60 kN
⎛ 156 ⎞ 320
λn = ⎜ ⎟× = 61
⎝ 2 . 89 ⎠ 250
2100 × (61 − 13.5 )
αa = =20.6
612 − 15.3 × 61 + 2050
λ = 61 +20.6 × 0.5 = 71.3
η = 0.00326 × (71.3 – 13.5) = 0.1885 ≥ 0
2
⎛ 71.3 ⎞
⎜ ⎟ + 1 + 0.1885
⎝ 90 ⎠
ξ = 2 = 1.447
⎛ 71.3 ⎞
2×⎜ ⎟
⎝ 90 ⎠
⎧ ⎡ ⎛ 2 ⎫
⎪ 90 ⎞ ⎤⎪
αc = 1.447 ⎨1 − ⎢1 − ⎜ ⎟ ⎥ ⎬ = 0.740
⎪⎩ ⎢⎣ ⎝ 1.447 × 71.3 ⎠ ⎥⎦ ⎪
⎭
Ns = (90 × 10) × 320 / 103 = 288 kN
Ncc = 0.9 × 0.740 × 288 = 192 kN
e = 28.2 + 5.0 = 33.2 mm
Icleat = 90 × 10 / 12
3
= 7500 mm4
(I/L)cleat = (7500 / 235) = 31.9 (assume same at other end)
(I/L)angle = (696 × 10 / 4000)
3
= 174
31.9
ke = = 0.134 ≥ 0.10
31.9 + 174 + 31.9
ed = 0.134 × 33.2 = 4.45 mm
φMsyi = 0.9 × 320 × 90 × 10 / (4 × 10 )
2 3
= 648 kNmm
648
Nh =
[4.45 + 648 / 192] = 82.8 kN
Nj2 =
350 × 6.0 2
×
(2 × 0.780 + 4 × 1 − 0.0667 ) = 95.6 kN
(1 − 0.0667 ) cos 40 o × 10 3
Nj = 95.6 kN
Welds 6 mm fillet weld, both sides of cleat, weld category SP, E48XX weld metal
L w = 90 mm φv w = 0.978 kN/mm
Cap plate to SHS 90 × 90 × 10 square edge flat bar (fillet welded to SHS)
⎛ 216 ⎞ 320
λn = ⎜ ⎟× = 84.6
⎝ 2.89 ⎠ 250
2100 × (84.6 − 13.5 )
αa = =18.9
84.6 2 − 15.3 × 84.6 + 2050
λ = 84.6 +18.9 × 0.5 = 94.0
η = 0.00326 × (94.0 – 13.5) = 0.263 ≥ 0
2
⎛ 94 ⎞
⎜ ⎟ + 1 + 0.263
⎝ 90 ⎠
ξ = 2 = 1.079
⎛ 94 ⎞
2×⎜ ⎟
⎝ 90 ⎠
⎧ ⎡ ⎛ 2 ⎫
⎪ 90 ⎞ ⎤⎪
αc = 1 . 079 ⎨ 1 − ⎢ 1 − ⎜ ⎟ ⎥⎬ = 0.581
⎪⎩ ⎢⎣ ⎝ 1.079 × 94 ⎠ ⎦⎥ ⎪
⎭
Ns = (90 × 10) × 320 / 103 = 288 kN
Ncc = 0.9 × 0.581 × 288 = 151 kN
e = (10 + 10) / 2 = 10 mm
ed = 10 / 2 = 5 mm
φMsyi = 0.9 × 320 × 90 × 10 / (4 × 10 )
2 3
= 648 kNmm
648
Nh =
[5 + 648 / 151] = 69.7 kN
3 Kitipornchai, S., Al-Bermani, F.G.A. and Murray, N.W., 'Eccentrically connected cleat plates in
compression', Journal of Structural Engineering, ASCE, Vol. 119 No 3 1993, pp767–781.
5 AUSTRALIAN STEEL INSTITUTE, Advisory Note, 'Design method for eccentrically loaded cleats not
to be used', Author Munter, S., Steel Construction Vol. 39 No 2 Dec 2005, p.16 (ASI Web Site
Release March 2006).
7 Hogan, T.J. and Munter, S.A., 'ASI limit states connection design series—Part 1—2007, Background
and summary details', AUSTRALIAN STEEL INSTITUTE, Steel Construction, Vol. 41 No 2, Dec 2007.
8 Hogan, T.J., 'ASI limit states connection design series—Part 2—2009, Background and summary',
AUSTRALIAN STEEL INSTITUTE, Steel Construction, Vol. 43 No 1, July 2009.
12 AUSTRALIAN STEEL INSTITUTE, 'Design Guide 1: Bolting in steel connections', Author Hogan, T.J.,
Contributing Author and Editor Munter, S.A., 2007.
16 Thornton, W.A., 'Bracing connections for heavy construction', Engineering Journal, AMERICAN
INSTITUTE OF STEEL CONSTRUCTION, Third quarter 1984, pp 139–148.
17 LIMCON V3, 'Limit states design of steel connections', ENGINEERING SYSTEMS web site
<www.steel-connections.com>.
18 Clifton, C. and el Sarraf, R., 'Eccentric cleats in compression and columns in moment resisting
connections', NEW ZEALAND HEAVY ENGINEERING RESEARCH ASSOCIATION (HERA), 2007.
19 NEW ZEALAND STANDARDS ASSOCIATION, NZS 3404:Part 1:1997, 'Steel Structures Standard'.
21 Khoo, X., Perera, M. and Albermani, F., ‘Design of eccentrically connected cleat plates in
compression', International Journal of Advanced Steel Construction, Vol. 6 No. 2, June 2010.
23 AMERICAN INSTITUTE OF STEEL CONSTRUCTION, 'AISC Specification for the design of steel
hollow structural sections', April 1997.
25 Blodgett, O., 'Design of welded structures', The JAMES F LINCOLN ARC WELDING FOUNDATION,
Fifth printing, 1972.
27 Vasarhelyi, D.D., 'Tests of gusset plate models', Journal of the structural division, ASCE, Vol 97
No ST2, Feb 1971, p. 665.
28 Richard, R.M. et al, 'Analytical models for steel connections', Proceedings W.H. Munse Symposium
on behaviour of metal structures—Research to practice, ASCE, 1983, p. 128.
29 Bjorhovde, R. and Chakrabarti, S.K., 'Tests of full-size gusset plate connections', Journal of structural
engineering, ASCE, Vol 111 No 3, March 1985, p. 667.
31 Thornton, W.A., 'On the analysis and design of bracing connections', National steel construction
conference proceedings, AISC, 1991, Paper 26, pp 26.1–26.33.
Design Ratio
Average
Specimen CHS Experiment Average for 0.9*Average capacity 0.9* average
design
identification Details fail load specimen for specimen design test to design
model
(Ref. 20) (Ref. 20) (Note 1) type type model model
capacity
(Note 2) capacity
kN kN kN kN kN
A1 60.3X3.2 27.7 18.9
A2 60.4X3.4 25.3 26.17 23.55 18.9 18.9 1.25
A3 60.5X3.2 25.5 18.9
B1 60.5X3.2 41.2 30.3
B2 60.4X3.1 40.3 40.33 36.30 30.6 30.4 1.19
B3 60.5X3.3 39.5 30.3
C1 60.7X3.1 32.5 33.00 29.70 24.1 24.25 1.22
C2 60.3X3.2 33.5 24.4
D1 60.3X3.2 32.8 34.95 31.46 21.4 21.35 1.47
D2 60.3X3.1 37.1 21.3
NOTES:
1 Obtained from page 26 of Reference 20.
2 Obtained using Limcon.
B1.1 General
This Appendix contains the output from the Limcon computer program for the design examples, detailed
calculations for which are included in this publication. Limcon undertakes design checks set out in the ASI
design model and lists the capacity and capacity ratio for each limit state. A detail diagram is included at the
beginning of the Limcon output. A virtual reality image of the connection can be displayed on the computer
screen to facilitate checking.
NOTE: To express the result of each limit state check Limcon uses the capacity ratio. This is the ratio of the design
capacity to the design action effect and the minimum capacity ratio for all limit states must not be less than 1.0.
In general, Limcon does not have sufficient information to compute member design capacities so it must use
section design capacities when evaluating minimum action criteria. This means that if minimum action
percentages are specified in Limcon they will be conservative to the extent that the section capacity
exceeds the member capacity. For these examples the minimum action percentages have been set to zero.
The method used in Limcon for evaluating eccentric connections in compression, which is not suitable for
hand calculation, is slightly less conservative than the method recommended in this publication.
Nevertheless, it has been in use for some time and has been shown to agree well with experimental results
(Ref. 21). Limcon retains this method but also shows the result for the method recommended in this paper
(Design Check No. 8).
Limcon can evaluate connections of the type discussed in Section 11. The KTG connection comprises two
or three braces connected to a rectangular gusset plate welded to a chord or column. This type of
connection is used in trusses and also for the chevron brace connection in vertical bracing systems.
A new version of Limcon can evaluate connections of the type discussed in Section 12. The UFBR
connection comprises a beam connected to a column with a shear type connection and one or two diagonal
braces connected to rectangular gusset plates welded to the beam and bolted to the column.
Refer to the Technical Notes chapter of the Limcon manual for more information on Limcon calculations
(Reference 22).
N *t
ti ≥ (73)
0.9f yi α
b fc1 − d h
y= b fc1 (74)
2
and the value of α is calculated as follows:
sp
α = min(αa, αb ) when y <
2
sp
= αb when y < and y > ab
2
sp Figure 42 Yield line pattern (b) H sections
= min(αc, αd , αe) when y ≥
2
where:
2
2b fcl − 2b fcl d h + 4 y 2
αa = (Fig. 41)
2sy
b (b − d h )(a b + y ) + 2(y + a b )a b y
α b = fcl fcl (Fig. 42)
2sa b y
2
b fcl − d h b fcl + 2 y c2 + s p y c
αc = (Fig. 43)
2sy c
b fcl s − d h s + 2 y d2 + s p y d − d h y d
αd = (Fig. 44)
sy d
b fcl s − 2d h s + 4a 2b + 2a b s p − 2a b d h Figure 43 Yield line pattern (c) H sections
αe = (Fig. 45)
2a b s
y c = min (a b , y )
⎛ b fcl − d h ⎞⎟
y d = min⎜ a b , s
⎜ 2 ⎟
⎝ ⎠
a b = distance from bolt hole to inside face of flange
EXPLANATION
The original text had an incorrect Figure 41—copy of Fig. 39—and all other figures 42–46 were
incorrectly numbered being out by one. In this Corrigenda, the correct figures have the correct numbers
consistent with the original text. As a consequence of this Corrigenda, there is no longer a Figure 46.
In the expressions for α under equation (74), the first and third expressions have been revised from
“max(αa, αb )” and “max (αc, αd , αe)” respectively to “min(αa, αb )” and “min (αc, αd , αe)” respectively.
(2) On page 26, the text under equation (80), the first and third expressions need to be revised from
“max(αa, αb )” and “max (αc, αd , αe)” respectively to “min(αa, αb )” and “min (αc, αd , αe )” respectively.
The text should read as follows:
“and the value of α is calculated as follows:
sp
α = min(αa, αb ) when y <
2
sp
= αb when y < and y > ab
2
sp
= min(αc, αd , αe) when y ≥ ”
2
(3) On page 27, the text under equation (83), the expression for α should be changed to read “min” instead of
“max”. The text should read:
l l
“α = min(αa, αb ) when y ≤ i = αb when y > i ”
2 2
(4) On page 28, the text under equation (86), the expression for αshould be changed to read “min” instead of
“max”. The text should read:
li - s p
α = min(αa, αb ) when y ≤
2
li - s p
= αb when y >
2