See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/265498674

The Development of a New Design Procedure for Conventional Single-Plate

Shear Connections

Article  in  Engineering Journal (New York) · June 2011

CITATIONS READS

6 2,147

2 authors, including:

William A. Thornton

40 PUBLICATIONS   577 CITATIONS   

SEE PROFILE

All content following this page was uploaded by William A. Thornton on 15 December 2015.

The user has requested enhancement of the downloaded file.

 The Development of a New Design Procedure for
Conventional Single-Plate Shear Connections
LARRY S. MUIR and WILLIAM A. THORNTON

ABSTRACT
Conventional single-plate shear connections are common and economical connections. The design procedure outlined in the 13th edition AISC
Steel Construction Manual, relies on the bolt shear values given in the 2005 AISC Specification for Structural Steel Buildings. The nominal bolt
shear values listed in Specification Table J3.2 have historically been 20% lower than the theoretical bolt values. This reduction was provided
to account for uneven force distribution among the bolts in end-loaded connections, such as bolted lap splices. The reduction served the
secondary function of providing an additional factor of safety for all bolted connections designed in accordance with the Specification. The
design procedure for conventional single-plate shear connections contained in the 13th edition Manual relied on this reduction to justify the
practice of neglecting eccentricity in the bolt group for most configurations. The 2010 AISC Specification increases the nominal bolt shear
values, necessitating a revised design procedure for single-plate shear connections in the 14th edition AISC Manual. This paper outlines the
revised procedure.

Keywords: single-plate shear connections.

S ingle-plate shear connections consist of a single plate

welded to the supporting beam or column and field bolt-
ed to the supported beam. Two different configurations of
revised design eccentricities and further limitation on plate
thickness for deeper connections using standard holes.

single-plate shear connections will be recognized in the 14th NEED FOR REVISED DESIGN PROCEDURE
edition of the AISC Steel Construction Manual: the conven-
The need to reevaluate and revise the design procedure con-
tional configuration and the extended configuration. The ex-
tained in the 13th edition Manual arose from an increase
tended configuration is a more general configuration in that
in the nominal bolt shear values provided in AISC’s 2010
it allows greater variation in the distance between the weld
Specification for Structural Steel Buildings. The nominal
and the bolts, the number of bolts, and the plate thicknesses
bolt shear values listed in Specification Table J3.2 have his-
used. The conventional configuration limits the distance be-
torically been 20% lower than the theoretical bolt values.
tween the weld and the bolts to a maximum of 3 in., allows
This reduction was provided to account for uneven force dis-
between 2 and 12 bolts in a single vertical line, and limits
tribution among the bolts in end-loaded connections, such
the ratio of the plate thickness to the bolt diameter.
as bolted lap splices. The reduction served the secondary
Conventional single-plate shear connections (Figure 1) are
common and economical connections. They provide simple
and economical fabrication and erection, and because bolted
connections are only used in the connection to the supported
member, there is no safety concern over the use of shared
bolts through the web of the support during erection.
The design procedure contained in the 14th edition of the
Manual will be similar to that contained in the 13th edition
(AISC, 2005a), but with a few key differences, including

Larry S. Muir, Structural Steel Consultant, Atlanta, GA (corresponding). E-mail:

larrymuir@larrymuir.com
William A. Thornton, Consultant, Cives Steel Company, Roswell, GA. E-mail:
bthornton@cives.com Fig. 1. Single-plate connection.
(Fig. 10-11 in the AISC Manual, 13th ed.)

ENGINEERING JOURNAL / SECOND QUARTER / 2011 / 141

141-152_EJ2Q_2011_2010-19R.indd 141 7/20/11 11:34 AM

 function of providing an additional factor of safety for Table 1. Recommended Design Parameters for
all bolted connections designed in accordance with the Conventional Single-Plate Shear Connections
Specification.
N Hole Type eb Maximum tw or tp
The design procedure for conventional single-plate shear
connections contained in the 13th edition Manual relied on SSL a/2 None
2–5
this reduction to justify the practice of neglecting eccentric- STD a/2 db/2 + z in.
ity in the bolt group for most configurations. Reanalysis SSL a/2 db/2 + z in.
has shown that neglecting the eccentricity is no longer ap- 6–12
STD a db/2 − z in.
propriate, considering the increased 2010 Specification bolt
strengths.

14TH EDITION MANUAL DESIGN PROCEDURE

4. The horizontal distance Leh must be equal to or great-
Because what follows is largely a discussion of the rationale er than 2db for both the plate and the beam web. Note
underlying the design procedure for single-plate shear con- that Leh is measured to the center of the hole or slot.
nections contained in the 14th edition Manual, it is appropri-
ate to present the procedure and then present the individual 5. Either the plate or the beam web must satisfy the max-
considerations is greater detail. imum thickness requirement given in Table 1.
The procedure presented in this paper has been adopted
into the 14th edition Manual as a method of designing con- Design Checks
ventional single-plate shear connections that is applicable The bolts and plate must be checked for the required shear
over the entire range of support rigidities. This procedure with an eccentricity equal to eb, as given in Table 1.
can be used to determine the strength of single-plate shear Plate buckling will not govern for the conventional
connections, which meet the dimensional limitations set configuration.
forth in the procedure.
PREVIOUS RESEARCH
Shared Provisions
A number of research programs have investigated the be-
The conventional configuration and the extended configu- havior of single-plate shear connections. A brief summary
ration of the single-plate shear connection share some at- includes:
tributes and requirements. Although this paper specifically
addresses the design of the conventional configuration, the • White (1965) conducted tests involving single-plate
requirements that apply to both configurations are presented shear connections to HSS columns.
together to better reflect the organization as contained in the • Lipson (1968) conducted tests of both single angle and
Manual. These shared requirements include: single-plate connections.
1. The use of either group A (ASTM A325 or F1852) or • Richard et al. (1980) conducted tests on single-plate
group B (ASTM A490 or F2280) bolts is acceptable. shear connections using stub beams.
2. The use of snug-tightened, pretensioned, or slip-critical • Hormby et al. (1984) conducted tests on single-plate
bolts is acceptable. connections with Grade 50 steel and composite
3. The use of material with either Fy = 36 ksi or Fy = 50 construction.
ksi is acceptable. • Astaneh et al. (1988) conducted tests on single-
4. The weld size shall be stp. plate shear connections with standard holes to rigid
supports.
Dimensional Limitations • Astaneh and Porter (1990) conducted tests on single-
1. Only a single vertical row of bolts is permitted. The plate shear connections with short-slotted holes to
number of bolts in the connection, n, is limited to rigid supports.
2 to 12. • Astaneh and Shaw (1992) conducted tests on single-
2. The distance from the bolt line to the weld line, a, plate shear connections to support girder webs.
must be equal to or less than 32 in. • Sarkar and Wallace (1992) conducted tests on single-
3. Standard or short-slotted holes are permitted to be plate shear connections to rigid supports.
used as noted in Table 1.

142 / ENGINEERING JOURNAL / SECOND QUARTER / 2011

141-152_EJ2Q_2011_2010-19R.indd 142 7/20/11 11:34 AM

 • Creech (2005) conducted tests on single-plate shear Rotational Demand
connections to rigid and flexible supports.
Before procedures for ensuring ductility can be derived, a
• Baldwin Metzger (2006) conducted tests on conven- target end rotation to be accommodated must first be de-
tional and extended single-plate shear connections to fined. Among researchers and AISC committees, 0.03 ra-
rigid supports. dian has become a de facto standard; 0.03 radian is roughly
equal to the end rotation of a beam whose span is 24 times
The analysis reflected in this paper centered on the tests
its depth, is loaded with the maximum uniform design load
conducted by Astaneh and colleagues, Sarkar and Wallace,
and is commonly accepted to be a reasonable upper bound
Creech, and Baldwin Metzger. Richard et al. did not test
for beam end rotation. This level of end rotation is unlikely
their connections to failure. Therefore, while their tests aid
to occur in many instances, but in order to provide a de-
in the understanding of the behavior of single-plate shear
sign procedure that can be applied to a wide range of practi-
connections, they do not provide much information regard-
cal conditions, AISC has adopted 0.03 radian as the target
ing the ultimate strength of these connections. The purpose
rotation.
of White’s tests was to determine the effect of connections
Beyond establishing a target magnitude for the rota-
on hollow structural section (HSS) columns, so though his
tion, a center of rotation must also be established. If 0.03
findings were considered they were not integral to the analy-
radian of rotation occurs at the top bolt of the connection,
sis. The Hormby et al. tests were not included because the
the movement of the bottom bolt relative to the plate that
design procedure was intended to be used for both compos-
must be accommodated is about twice that required if the
ite and noncomposite construction.
rotation occurs about the center of the bolt group. Research
has shown that the center of rotation can vary considerably
ESTABLISHING A DESIGN MODEL
throughout the loading of the connection. However, when
The inherent rigidity of single-plate connections has been the connection is made to a rigid support, the center of rota-
a concern for designers worried about considerable, unan- tion will coincide roughly with the center of the bolt group
ticipated moments that could be developed in the connec- as the load approaches ultimate. It was observed that “during
tions, which could precipitate a sudden rupture of either the test, neutral axis remained very close to the mid-height
the weld or the bolts. Further, Section B3.6a of the AISC of the single plate.” (Astaneh et al., 1988) Even when the
Specification, requires that simple shear connections have mid-depth of the connection and beam are not coincident,
sufficient rotational capacity to accommodate the required the center of rotation remains near the mid-depth of the con-
beam end rotation. The potential of developing moments in nection. (Hormby et al., 1984). No comments are provided in
the connection beyond those resulting from the eccentric- the research regarding the center of rotation of single-plate
ity between the support and the bolts, the need to accom- connections supporting composite beams; however, Horm-
modate simple beam end rotations, and ductility concerns by et al. (1984) noted, with regard to the use of single-plate
become primary considerations in the development of a de- connections used with composite beams, that “since full-
sign model for the single plate shear connection. Much of scale beam tests of off-axis bolt groups resulted in essential-
the research and resulting design procedures developed over ly the same moment-rotation and center of rotation response
the years has concentrated on predicting and/or controlling as symmetrical connections, it is concluded that the behav-
this behavior. ior of the single plate is not affected by its location relative
The design model used in the 14th edition Manual builds to beam’s neutral axis.”
on previous work. First, ductility requirements relating the The preceding findings relative to the location of the cen-
strength of the plate (or beam web) to the bolt and weld ter of rotation are based on the observed behavior of single-
strengths are set. These requirements are based on a combi- plate connections to rigid supports. There is a lack of similar
nation of theoretically derived ratios and empirical results. data pertaining to flexible supports; however, because the
The goal of the ductility requirements is to accommodate a simple beam end rotation can be accommodated through
target end rotation of 0.03 radian without rupture of any of movement of the flexible support, instead of plowing of the
the elements. The mechanisms used to achieve this ductility bolts, the location of the center of rotation is of less impor-
are discussed in greater detail later. tance for these conditions.
Having ensured sufficient ductility, the bolts are then de- Theoretically, the single-plate shear connection with stan-
signed to resist the required beam end reaction at an em- dard holes can accommodate the requisite beam end rota-
pirically derived effective eccentricity. Finally the plate is tions through a combination of plate flexural yielding; bolt
assumed to be subjected to loads consistent with the other deformation; bolt plowing; and, in the case of a connection
elements, the beam end reaction applied at the effective to a flexible support, through support rotation. Bolt plowing
eccentricity. is the local yielding of the plate or beam web, which occurs

ENGINEERING JOURNAL / SECOND QUARTER / 2011 / 143

141-152_EJ2Q_2011_2010-19R.indd 143 7/20/11 11:34 AM

 at the bolt holes causing elongation of the holes. The rela- Table 2. Results of Bolt Plowing Tests
tively short distance between the bolts and the welds, which
Plate Fy Plate Fu
is an integral feature of the conventional single-plate shear Bolt tp (in.) Δ (in.)
(ksi) (ksi)
connections, allows only a small area over which yielding
can occur, so plate yielding is usually discounted as a means ¾-in. A325-N 4 0.65 47.5 65.9
of accommodating the simple beam end rotation. Support ¾-in. A325-N c 0.6 47.3 65.5
rotation is not relied on to accommodate simple beam end ¾-in. A325-N a 0.3 47.6 67.1
rotations, because this mechanism cannot be applied to rigid
¾-in. A490-N c 0.7 47.3 65.5
supports and because other effects associated with support
rotation can lead to serviceability problems. ¾-in. A490-N a 0.4 47.6 67.1
Both bolt deformation and bolt plowing are dependent
on determining not so much the end rotation, but rather the
movement of the bolts relative to the plate that occurs as
a result of the end rotation. For a given center of rotation, thickness of either the plate or the beam web was limited to
there is a direct relationship between the beam end rotation one-half the diameter of the bolt plus z in.
and the amount of movement the bolt group must accom- To evaluate this requirement, it is instructive to look at the
modate either through deformation in the bolt itself and/or results of tests in which a single bolt was essentially plowed
plowing through the joined materials. Assuming a center of through various thicknesses of plates. Sarkar and Wallace
rotation at the center of the bolt group, the relative horizontal (1992) ran a series of such tests. The results are provided in
movement of the bolts and the connected materials can be Table 2.
approximated as: These results indicate that for a plate thickness equal to
one-half the bolt diameter, approximately 0.3 in. of bolt
( n − 1) b plowing, Δ, can occur. Limiting the plate thickness to ap-
δ = ( 0.03 radian ) (1)
2 proximately half the bolt diameter can, therefore, be ex-
pected to accommodate end rotations for up to a seven-row
This results in a maximum relative horizontal movement of connection. Further limiting the plate thickness to half the
0.495 in. for a 12-row connection. Deformation of this mag- bolt diameter minus z in. can accommodate the larger de-
nitude would essentially exhaust the capacity of the bolt—as formations required for deeper connections. Because the
can be shown using the load-deformation relationship given tests were run with w-in. bolts, it is believed that the results
on page 7-6 of the AISC 13th edition Manual (which can be can be safely extrapolated to larger-diameter and -strength
used to calculate the force on a bolt given a deformation, in bolts. Bolts less than w-in. diameter are rarely used in struc-
this instance a deformation of about 0.5 in.): tural connections.
Providing short-slotted holes in the plate can also help to
( ) ( )
0.55 0.55
R = Rult 1− e−10 Δ = Rult 1 − e −10(0.5) accommodate the simple beam end rotation. The short slots
(2) will provide between 4 in. and c in. of horizontal move-
= 0.996 Rult ment in typical connections, before any deformation must
occur in the bolts, the plate or the beam web. This 4 in.
To accommodate even the modest deformational demands
alone is enough to provide the 0.03 radian of rotation for a
of a two-bolt connection, nearly 60% of the bolt capacity
five-row connection. In other words, when short slots are
would be exhausted. Therefore, deformation in the bolt alone
provided in a single-plate shear connection of five rows or
cannot be counted on to accommodate the simple beam end
less, the ratio between plate thickness and bolt diameter is
rotation.
immaterial.
Eliminating plate flexural yielding and support rotation as
As stated previously, 0.03 radian is considered a conserva-
a possible means of accommodating the beam end rotations
tive upper bound for the end rotation. It might be reasonable
leaves only the combined effects of bolt deformation and
to relax the plate thickness to bolt diameter requirements
bolt plowing. In order for the bolts to plow, however, there
when the end rotation is known to be less than 0.03 radian,
must be an upper limit placed on the stiffness and strength
such as when the beam span is short relative to the beam
of the plate and/or the beam web relative to that of the bolts.
depth or when the beam is sized based on serviceability
Prior to the 13th edition Manual, the plate thickness was
rather than strength criteria. Moving from ASTM A325
limited to one-half the diameter of the bolt plus z in. In
bolts to A490 bolts might also offer some relief from this
the 13th edition Manual, the possibility that deformations
requirement. Because all of the tests directly related to bolt
could occur in either the plate or the beam web was formally
plowing utilized A325 bolts, the plowing behavior of A490
recognized, and the requirement was changed such that the

144 / ENGINEERING JOURNAL / SECOND QUARTER / 2011

141-152_EJ2Q_2011_2010-19R.indd 144 7/20/11 11:34 AM

 bolts has not been established, but it seems reasonable to or
believe that due to its greater strength, an A490 bolt would
be capable of plowing through a greater length or thickness 2n
eb = − a for connections using short-slotted holes (4)
of plate before rupturing. 3

Edge Distance Requirements When attached to a flexible support, the effective eccen-
tricity, eb, could not be less than the distance from the
The AISC Manual design procedures for single-plate con- weld group to the bolt group. Assuming a practical range
nections had not included a provision requiring that the of 22  in. to 32  in. for a, the 3rd edition LRFD Manual
horizontal edge distance be twice the bolt diameter prior to equations would predict an effective eccentricity on the bolt
the 13th edition. This requirement comes from the original group of between about 5 to 267% of the distance from the
Richard et al. (1980) research. It was included in the AISC weld group to the bolt group. Where the predicted effective
book Engineering for Steel Construction (1984). It was not eccentricity exceeded the a dimension, this was presumably
included in the Astaneh et al (1988, 1989) procedure, which done to account for potentially large moments occurring at
was the basis of the procedure in the 9th edition (ASD), 2nd the support, which could result in a larger moment at the bolt
edition (LRFD), and 3rd edition (LRFD) Manuals. Rather group. Though this large effective eccentricity will occur
than requiring a horizontal edge distance twice the bolt di- early in the loading history, the tests indicate that the reduc-
ameter, Astaneh et al.’s procedure recommended a horizon- tion in stiffness due to bolt plowing reduced the eccentricity
tal edge distance 1.5 times the bolt diameter. at ultimate loads as was intended.
The intent of the twice the bolt diameter requirement Reanalysis of existing data and further testing (Creech,
seems to be to ensure that the bolts will bear without tear- 2005; Baldwin Metzger, 2006) led to a less conservative re-
ing through the edge of the material. However, bolt tear-out quirement for the 13th edition Manual in which, for most
never occurred in any of the testing nor was any tearing be- cases, eccentricity was neglected. Though the tests did not
tween the edge of the hole and the edge of the plate observed indicate that there was no eccentricity on the bolt group,
that might indicate bolt tear-out was imminent. The maxi- the 20% reduction in the bolt strength inherent in the 2005
mum relative horizontal movement required to develop the Specification allowed the conclusion that the eccentricity
simple beam end rotation of 0.03 radian is 0.495 in., as previ- could safely be neglected.
ously discussed. Based on this fact, and considering the fact Because many of the tests were configured such that the
that it was not required by the Astaneh et al. work (which bolts governed the capacity of the connection, there is a rela-
was the basis for single-plate shear connection design in the tive wealth of data on which to base the design procedure for
United States for 20 years), the edge distance requirement of the bolts. Of 31 tests considered here, the bolts governed the
twice the bolt diameter would seem to be overly conserva- strength of 20 of the connections. These 20 connections also
tive and unnecessary. contained a good mix of connection depths, hole types and
support rigidities.
DESIGN OF THE BOLT GROUP The approach taken in developing a design methodology
It is intuitive to assume that the bolt group in a single-plate for the bolt group followed the historical precedent of de-
shear connection, being offset from the face of the support, termining the effective eccentricity to which the bolt group
will experience some eccentricity. The effective, or design, was subjected. Only the effective eccentricity at ultimate
eccentricity, however, is not necessarily equal to the distance load was considered in developing the design procedure, al-
from the weld group to the bolt group, as might be assumed. though effective eccentricities were often reported through-
A significant end moment might develop when a stiff plate out the loading. There is no evidence in the testing that these
connection is attached to a rigid support. In such cases, the larger effective eccentricities applied in conjunction with
inflection point of the beam might be moved considerably lesser loads can govern the strength of a single-plate shear
into the span, resulting in an effective eccentricity higher connection meeting the dimensional requirements laid out
than the distance from the weld group to the bolt group. in the procedure. A summary of the test data and the analy-
Conversely, the presence of short slots or bolt plowing might sis is provided in Table 3.
reduce the effective eccentricity on the bolt group. Both of In Table 3, the predicted bolt group strength without ec-
these possibilities were reflected in the design procedures centricity values (column 11) were calculated by multiplying
used prior to the 13th edition Manual. The LRFD 3rd edi- the number of bolts in the connection (column 2) by the bolt
tion Manual (AISC, 2001), for instance, calculated the effec- shear strength (column 4). Where the bolt shear strengths
tive eccentricity on the bolt group as: were measured and reported in the available reports, these
values were used. Where measured bolt strengths were not
eb = ( n − 1) − a for connections using standard holes (3) reported, the bolt strength was assumed to be 26.5 kips for
ASTM A325-N bolts and 33.2 kips for ASTM A325-X and

ENGINEERING JOURNAL / SECOND QUARTER / 2011 / 145

141-152_EJ2Q_2011_2010-19R.indd 145 7/20/11 11:34 AM

 Table 3. Summary of Test Data
(2) (4) (6) (7) (8) (9) (11)
(1) (3) (5) (10) (12) (13) (14)
Bolts Bolt Weld Plate Plate Tested Predicted Bolt
Hole a θ Test/ Eff. e = Failure
Test ID (All bolts are Strength Size Fyb Fu b Strength Group Strength
Type (in.) (rad.) Predict %a Mode
¾-in. dia.) (kips) (in.) (ksi) (ksi) (kips) w/o Ecc. (kips)
Astaneh and Colleagues

141-152_EJ2Q_2011_2010-19R.indd 146
1 1 (7)-A325-N STD 26.5a 3 4 35.3 61 160 0.026 186 0.860 1.02 Bolt
2 2 (5)-A325-N STD 26.5a 3 4 35.3 61 137 0.054 133 1.03 0 Bolt
3 3 (3)-A325-N STD 26.5a 3 4 35.3 61 94 0.056 79.5 1.18 0 Bolt
4 4 (5)-A490-N STD 33.2a 2.75 R 35.3 61 130 0.053 166 0.783 1.07 Bolt
5 5 (3)-A490-N STD 33.2a 2.75 R 35.3 61 79 0.061 99.6 0.994 0.585 Weld and bolt
Baldwin Metzger
6 31C-3a (3)-A325-N STD 27.0 3 x 68.1 97.5 81 0.032 81 — — Beam
7 4B1C-3a (4)-A325-N STD 27.0 3 x 68.1 97.5 110 0.027 108 — — Beam
8 51C-3a (5)-A325-N STD 31.0 3 x 68.1 97.5 146 0.030 155 0.942 0.407 Bolt
9 71C-3a (7)-A325-N STD 27.0 3 x 68.1 97.5 173 0.018 189 0.915 0.740 Bolt
Creech
10 S1-RSS-3-A325-N (3)-A325-N SSL 30.3 3 — 39.6 62.1 78.8 0.036 90.9 — — Beam
11 S2-RST-3-A325-N (3)-A325-N STD 30.3 3 — 39.6 62.1 90.7 0.027 90.9 0.998 0 Bolt
12 S3-FSS-3-A325-N (3)-A325-N SSL 30.3 3 — 39.6 62.1 71.8 0.039 90.9 0.790 0.55 Bolt
13 S4-FST-3-A325-N (3)-A325-N STD 30.3 3 — 39.6 62.1 61.4 0.023 90.9 0.675 0.82 Bolt

146 / ENGINEERING JOURNAL / SECOND QUARTER / 2011

14 S5-FSS-3-A325-N-SR (3)-A325-N SSL 30.3 3 — 39.6 62.1 75.6 0.031 90.9 0.832 0.43 Bolt
15 S6-FST-2-A325-N (2)-A325-N STD 30.3 3 — 39.6 62.1 44.2 0.012 60.6 0.729 0.44 Bolt
16 S7-FSS-2-A325-N (2)-A325-N SSL 30.3 3 — 39.6 62.1 45.5 0.011 60.6 0.751 0.40 Bolt
17 S8-FSS-2-A325-N-SR (2)-A325-N SSL 30.3 3 — 39.6 62.1 47.9 0.013 60.6 0.790 0.34 Bolt
18 S9-RSS-7-A325-N (7)-A325-N SSL 30.3 3 — 44.4 66.3 166.5 0.028 212 — — Beam
19 S10-FSS-7-A325-N-SR (7)-A325-N SSL 30.3 3 — 44.4 66.3 202.5 0.027 212 0.955 0.40 Bolt
Sarkar and Wallace
20 1a-North End (2)-A325-X STD 33.2a 3.5 — 47.4 — 41.7 0.033 66.6 — — Pl yielding
21 1a-South End (2)-A325-X STD 33.2a 3.5 — 47.4 — 64.3 0.025 (0.043) 66.6 — — Pl distortion
22 1b-North End (2)-A325-N STD 26.5a 3.5 — 47.4 — 60.8 0.028 53 1.15 0 Weld
23 1b-South End (2)-A325-N STD 26.5a 3.5 — 47.4 — 51.8 0.033 53 0.977 0.032 Weld
24 2a-North End (4)-A325-N STD 26.5a 3.5 — 47.4 — 93 (66.4) 0.028 (0.32) 106 0.877 0.41 No failure (bolt)
25 2a-South End (4)-A325-N STD 26.5a 3.5 — 47.4 — 93 (81.6) 0.32 (0.38) 106 0.877 0.41 No failure (bolt)
26 2b-North End (4)-A325-N SSL 26.5a 3.5 — 47.4 — 129 0.042 106 1.22 0 Bolt
27 2b-South End (4)-A325-N SSL 26.5a 3.5 — 47.4 — 129 0.042 106 1.22 0 Bolt
28 3a-North End (6)-A325-N STD 26.5a 3.5 — 47.4 — 119 (102) 0.027 (0.014) 159 0.748 1.14 Bolt
29 3a-South End (6)-A325-N STD 26.5a 3.5 — 47.4 — 119 (109) 0.027 (0.019) 159 0.748 1.14 Bolt
30 3b-North End (6)-A325-N SSL 26.5a 3.5 — 47.4 — 168 0.030 159 1.06 0 Bolt
31 3b-South End (6)-A325-N SSL 26.5a 3.5 — 47.4 — 194 –(0.030) 159 — — No failure
Notes:
a. Theoretical bolts strength with no end load reduction, based on nominal strength from the appropriate ASTM standard
b. All plates a in. thick measured yield and ultimate stresses as shown.
c. Indicates that the data are not available

7/20/11 11:34 AM
 A490-N bolts. Inherent in these values are the same as- There were 17 tests for which bolt strengths were not re-
sumptions made in the 2010 AISC Specification that the ported; of these, the bolt group governed the strengths of
bolt strengths equal the minimum specified tensile strength 12 tests. Therefore, nearly half of the tests potentially un-
given in the ASTM standards, the ratio of bolt shear strength derestimate the effect of the eccentricity. However, in 7 of
to bolt tensile strength is 0.62 and ratio of effective thread the 12 cases (tests 2, 3, 24, 25, 26, 27 and 30), the calcu-
root area to shank area is 0.80. lated effective eccentricity is more than 20% lower than the
Once a predicted strength neglecting eccentricity was es- recommended eccentricity used in the 14th edition Manual
tablished, the effect of the eccentricity could be determined. design procedure, a considerably larger margin than the re-
Calculating the ratio of the tested strength (column 9) to ported overstrength of the tested bolts. In 3 of the remain-
the predicted strength (column 11) provided the efficiency ing tests (tests 1, 28 and 29), where the predicted effective
of the bolt group in resisting the applied shear (column eccentricity exceeds the recommended eccentricity used in
12), which is essentially the C-value from the eccentrically the 14th edition Manual design procedure, the 14th edition
loaded bolt group tables in Part 7 of the Manual. Using the Manual design procedure limits the plate thickness to less
instantaneous center of rotation method described in Part 7, than the tested configuration to increase the ductility of the
an effective eccentricity corresponding to the bolt group ef- bolt group. In the final 2 of the 12 tests for which no bolt
ficiency could be determined and expressed as a percentage data were available (tests 4 and 5), the rotational demand on
of the a dimension (column 13). the connection during testing was approximately twice the
It can be seen that seven of the tests (tests 2, 3, 11, 22, expected simple beam end rotation.
26, 27 and 30) indicate that the strength of the bolt group is The use of slip-critical connections should also be ad-
best predicted by neglecting the eccentricity. The number dressed. The design procedure contained in the 14th edition
of bolts for these tests ranged from two to six installed in Manual follows the precedent set by previous editions of the
both standard and short-slotted holes. There are eight tests Manual in allowing slip-critical connection design values to
(tests 8, 14, 15, 16, 17, 19, 24 and 25) that indicate that the be used with single-plate shear connections. Because only
strength of the bolt group is best predicted by assuming an standard and short-slotted holes are allowed and accommo-
eccentricity equal to 0 to 50% of the a distance. The number dation of the end rotation is required, the use of slip-critical
of bolts for these tests ranged from two to seven installed in connections would never be required per the Specification
both standard and short-slotted holes. Six of the tests (tests for these connections, and AISC discourages the use of slip-
1, 4, 9, 13, 28 and 29) indicate that the strength of the bolt critical connections unless required by the Specifications.
group is best predicted by assuming an eccentricity equal However, the use of slip-critical connection design values
to 74 to 114% of the a distance. In every case, where the was not felt to be detrimental to the performance of the con-
best predictor of bolt group strength was based on an effec- nection, so they have been allowed. Even when designed
tive eccentricity exceeding one half the a distance, standard using slip-critical design values, the bolts in a single-plate
holes were used. Four of the tests were either six- or seven- shear connection will likely slip into bearing when large end
row connections. rotations are required.
The remaining two tests were treated as outliers. One of
the tests was a three-row connection in which the support DESIGN OF THE WELD GROUP
girder was yielded during testing, which was considered
Just as the ratio between the bolt diameter and plate thick-
unusual. The other outlier was a test in which the applied
ness is intended to allow ductile redistribution of moments
rotation was 0.053 radian, considerably more than the target.
and accommodation of the simple beam end rotation, the
Other connections were subjected to similarly large rota-
weld is also sized to promote ductile behavior. The Manual
tions but did not show an increase in the effective eccentric-
design procedure requires that the weld size be equal to s of
ity. It should be noted that when these two data points are
the plate thickness. A derivation of the weld requirement has
compared to the 14th edition Manual design procedure there
been provided by Muir and Hewitt (2009), so only a brief
is still good agreement, even though an eccentricity less than
discussion will be provided here. The derivation assumes
that predicted by the test is used in the 14th edition Manual
that the plate must yield prior to weld rupture to ensure duc-
design procedure. This can be explained in part by the 10%
tile behavior. Though most single-plate connections tested
reduction in bolt value inherent in the 2010 Specification.
had a weld size equal to at least w of the plate thickness,
In cases where bolt strengths were not reported, it is likely
Baldwin Metzger (2006) ran several single-plate connec-
that the actual bolt strengths were greater than the nominal
tion tests, both extended and conventional configurations,
strengths used to calculate the predicted strength of the con-
with welds sized to one-half the plate thickness, which con-
nection. Underestimating the predicted strength of the con-
firmed the suitability of the current s of the plate thickness
nection would lead to an overestimation of the bolt group
requirement.
efficiency and a lower corresponding effective eccentricity.

ENGINEERING JOURNAL / SECOND QUARTER / 2011 / 147

141-152_EJ2Q_2011_2010-19R.indd 147 7/20/11 11:34 AM

 DESIGN OF THE PLATE L Fy
λ=
Only two of the tests listed in Table 3 were governed by 2
⎛ L⎞
the strength of the plate, and the governing limit states were t p 47, 000 + 112, 000 ⎜ ⎟
listed as shear yielding and shear distortion (Sarkar and ⎝ 2a ⎠
Wallace, 1992). However, it is clear that the plate must have 36 50
sufficient strength to resist the required design loads, as re- =
2
quired by the Specification. The applicable limit states from ⎛ 36 ⎞
0.25 47, 500 + 112, 000 ⎜⎜ ⎟⎟
⎝ 2 ( 3.5 ) ⎠
Section J4 of the Specification are shear yielding (Equation
J4-3) shear rupture (Equation J4-4) and flexural yielding.
Flexural yielding should be checked using the plastic section = 0.587 < 0.7
modulus (Mohr and Murray, 2008). Block shear may also
Therefore, buckling will not govern.
be a governing limit state if the horizontal edge distance
does not exceed the vertical edge distance. Buckling will
COMPARISON TO TEST RESULTS
not occur in the plate, because the distance from the weld to
the bolt group cannot exceed 32 in. This can be proven as A comparison of the 14th edition Manual design proce-
follows (Muir and Thornton, 2004): dure to the test results is given in Figures 2 and 3. Figure 2
compares the 14th edition Manual design procedure to test
Assuming a = 3.5 in. (the maximum permissible dimen-
run with an a dimension equal to 3 in. Figure 3 makes the
sion), L = 36 in. for a 12-row connection, Fy = 50 ksi, and
comparison to test run with an a dimension equal to 32 in.
tp = 0.25 in.:
There appears to be good agreement, and there is only one
data point for which the 14th edition Manual procedure ap-
pears slightly nonconservative.

Fig. 2. Comparison of design procedure to test results for a = 3.0 in.

148 / ENGINEERING JOURNAL / SECOND QUARTER / 2011

141-152_EJ2Q_2011_2010-19R.indd 148 7/20/11 11:34 AM

 DESIGN EXAMPLE

Given:
Beam: W24×76 (A572 Grade 50); t w = 0.44 in.
Bolts: Six d-in. A325-N (STD holes)
Plate: A572 Grade 50; tp = 0.375 in.; dp = 18 in.
Verify that the plate satisfies the requirements for design as a conventional single-plate shear connection:
Verify number of bolts: 2 < n = 6 < 12
Verify distance between the bolt and the weld: a < 3.5 in.
Verify plate or web thickness: tp = 0.375 in. < db/2 – z in. = (0.875 in.)/2 – z in. = 0.375 in.
Verify horizontal edge distance: Leh = 1.75 in. > 2db = 2(0.875 in.) = 1.75 in.
Determine shear strength of a single bolt:
d2 ⎡ (0.875 in.)2 ⎤
φrb = φ b π Fnv = 0.75 ⎢ ⎥ π ( 54 ksi ) = 24.4 kips
4 ⎢⎣ 4 ⎥⎦

Determine bolt bearing strength of plate per bolt:

First determine clear distance, Lc, to edge of plate:
dh 0.9375 in.
Lc = Le − = 1.5 in. − = 1.03 in.
2 2
(
φrbrg = min φ 2.4 dp t p Fu , φ1.2 Lc t p Fu )
= min ( 0.75(2.4)(18 in.)(0.3775 in.)(58 ksi), 0.75(1.2)(1.03 in.)(0.375 in.)(58 ksi))
= min ( 38.4 kips, 22.6 kips )

Fig. 3. Comparison of design procedure to test results for a = 3.5 in.

ENGINEERING JOURNAL / SECOND QUARTER / 2011 / 149

141-152_EJ2Q_2011_2010-19R.indd 149 7/20/11 11:34 AM

 It should be noted that calculating the bearing strength based on lesser of the horizontal and vertical edge distances is conserva-
tive and other more exact methods are also acceptable. Because the thickness of the beam web is greater than the thickness of
the plate, bearing on the plate will govern.
Determine strength of bolt group:
Because there are six rows of bolts in standard holes, the eccentricity shall be taken as the full distance between the bolts
and the weld, 3 in.
From Table 7-7, C = 4.98

(
φRb = min φrb, φrbrg C )
= 22.6 kips(4.98)
= 113 kips > 100 kips o.k.
Determine the shear yielding strength of the plate:
φRvy = φ 0.6 Fy d p t p
= 1.0(0.6)(50 ksi)(18 in.)(0.375 in.)
= 203 kips > 100 kips o.k.
Determine the shear rupture strength of the plate:
An = t p ⎡⎣ d p − n ( db + 0.125 in. ) ⎤⎦
= 0.375 in. ⎡⎣18 in. − 6 ( 0.875 in.+0.1225 in. ) ⎤⎦

= 4.5 in.2
φRvr = φ 0.6 Fu Anet
= 0.75(0.6)(65 ksi)(4.5 in.2 )
= 132 kips > 100 kips o.k.
Determine flexural strength of the plate:
As noted earlier, the check should be performed using the gross plastic section modulus and buckling of the plate will not
govern.
tp d 2p (0.375 in.)(18 in.)2
Zg = = = 30.4 in.3
4 4
φFy Z g
φR f =
e
0.9(50 ksi)(30.4 in.3 )
=
3 in.
= 456 kips

Size the weld:

w = stp = s(0.375 in.) = 0.234 in.
Therefore, use a 4-in. weld each side.
Because the horizontal edge distance is greater than the vertical edge distance block shear will not govern.

150 / ENGINEERING JOURNAL / SECOND QUARTER / 2011

141-152_EJ2Q_2011_2010-19R.indd 150 7/20/11 11:34 AM

 CONCLUSION REFERENCES
The design procedure for conventional single-plate shear AISC (1984), Engineering for Steel Construction, American
connections contained in the 14th edition of the AISC Man- Institute of Steel Construction, Chicago, IL.
ual has been revised to accommodate a change in the nomi- AISC (2001), Manual of Steel Construction: Load and Re-
nal bolt strength presented in the AISC Specification. The sistance Factor Design, 3rd ed., American Institute of
new procedure represents a safe and economical approach Steel Construction, Chicago, IL.
to these connections based on rational design methods and
AISC (2005a), Steel Construction Manual, 13th ed., Ameri-
confirmed by testing.
can Institute of Steel Construction, Chicago, IL.
ACKNOWLEDGMENTS AISC (2005b), Specification for Structural Steel Buildings,
ANSI/AISC 360-05, American Institute of Steel Con-
The authors would like to acknowledge the participation of struction, Chicago, IL.
the AISC Committee on Manuals and Textbooks in develop-
AISC (2010), Specification for Structural Steel Buildings,
ing the design procedure presented in this paper with special
ANSI/AISC 360-10, American Institute of Steel Con-
thanks to Charlie Carter, Patrick Fortney, William Lindley
struction, Chicago, IL.
and Clifford Schwinger, who, along with the authors, com-
prised the task group that worked to distill the available data Astaneh, A., Call, S.M. and McMullin, K.M. (1988), “De-
into a clear and user-friendly procedure. sign of Single Plate Framing Connections,” Report No.
UCB/SEMM-88/12, Department of Civil Engineering,
SYMBOLS University of California, Berkeley.
Astaneh, A., Call, S.M. and McMullin, K.M. (1989), “De-
Fy Minimum specified yield stress, ksi
sign of Single Plate Shear Connections,” Engineering
Leh Horizontal edge distance, in. Journal, AISC, Vol. 26, No. 1.
Lev Vertical edge distance, in. Astaneh, A., McMullin, K.M. and Call, S.M. (1993), “Be-
havior and Design of Steel Single Plate Shear Connec-
R Simple beam end reaction, kips tions,” Journal of Structural Engineering, ASCE, Vol.
Rult Ultimate shear strength of the bolt, kips 119, No. 8, pp. 2421–2440.
Astaneh, A. and Porter, K.A. (1990), “Design of Single
a Distance from the face of the support to the vertical
Plate Connections with Snug-Tight Bolts in Short Slot-
line of bolts
ted Holes,” Report No. UCB/SEMM-90/23, Department
b Spacing between rows of bolts of Civil Engineering, University of California, Berkeley.
d Plate depth Astaneh, A. and Shaw, A.L. (1992), “Experimental Study of
Single-Plate Steel Beam-to-Girder Connections,” Report
db Bolt diameter No. UCB/SEMM-92/13, Department of Civil Engineer-
eb Effective (design) eccentricity of the bolt group ing, University of California, Berkeley.
Baldwin Metzger, K.A. (2006). “Experimental Verification
n Number of rows
of a New Single Plate Shear Connection Design Model,”
tp Plate thickness, in. Master’s Thesis, Virginia Tech, Blacksburg, VA.
tw Beam web thickness, in. Creech, D. (2005), “Behavior of Single Plate Shear Connec-
tions with Rigid and Flexible Supports,: Master’s Thesis,
w Weld leg size, in. North Carolina State University, Raleigh, NC.
δ Relative horizontal movement of the bolts and the Hormby, D.E., Richard, R.M. and Kreigh, J.D. (1984), “Sin-
connected materials, in. gle-Plate Framing Connections with Grade-50 Steel and
Composite Construction,” Engineering Journal, AISC,
Δ Deformation of the bolt, in.
Vol. 21, No. 3.
λ Slenderness parameter, dimensionless Lipson, S.L. (1968), “Single-Angle and Single-Plate Beam
Framing Connections,” Canadian Structural Engineering
Conference, Toronto, ON, Canada, pp. 141–162.
Mohr, B.A., and Murray, T.M. (2008), “Bending Strength
of Steel Bracket and Splice Plates,” Engineering Journal,
AISC, Vol. 45, No. 2. 

ENGINEERING JOURNAL / SECOND QUARTER / 2011 / 151

141-152_EJ2Q_2011_2010-19R.indd 151 7/20/11 11:34 AM

 Muir, L.S. and Hewitt, C. (2009), “Design of Unstiffened Richard, R.M., Kreigh, J.D. and Hormby, D.E. (1982), “De-
Extended Single-Plate Shear Connections,” Engineering sign of Single Plate Framing Connections with A307
Journal, AISC, Vol. 46, No. 2.  Bolts,” Engineering Journal, AISC, Vol. 19, No. 4.
Muir, L.S. and Thornton, W.S. (2004). “A Direct Method Sarkar, D. and Wallace, B. (1992), “Design of Single Plate
for Obtaining the Plate Buckling Coefficient for Double Framing Connections,” Draft Research Report, Report
Coped Beams,” Engineering Journal, AISC, Vol. 41, No. FSEL/AISC 92-01, University of Oklahoma, Nor-
No. 3. man, OK.
Richard, R.M., Gillet, P.E., Kreigh, J.D. and Lewis, B.A. White, R.N. (1965), “Framing Connections for Square and
(1980), “The Analysis and Design of Single Plate Fram- Rectangular Structural Tubing,” Engineering Journal,
ing Connections,” Engineering Journal, AISC, Vol. 17, AISC, Vol. 2, No. 3.
No. 2.

152 / ENGINEERING JOURNAL / SECOND QUARTER / 2011

141-152_EJ2Q_2011_2010-19R.indd
View publication stats 152 7/20/11 11:34 AM