Detailing_CH03 6/2/09 8:16 AM Page 16

CHAPTER 3

not be used in cases where it would be subject to torsion loads. tion of the beam, unless the tee stem is designed to meet the
Because of its flexibility, it is unsuitable for resisting axial requirements for single-plate connections.
tension loads in the beam. The tee connection has good resistance to axial compres-
Figure 3-14 illustrates a single-angle connection. Recom- sion load in the beam, but relatively poor resistance to axial
mended practice is to connect the angle to the supporting tension. Resistance to torsion is fair even when using the min-
member in the shop, either by bolting through standard holes imum length of tee allowed by the beam web.
or by welding. Connecting the single angle to the supported The tee is usually shop attached to the supporting member
member in the shop causes this connection to be far more (Figure 3-15). However, on rare occasions it can be shop at-
difficult to erect. This type of connection often can be used tached to the supported beam, but this negates some of the
where interferences prohibit the use of a double-angle connec- advantages of the tee connection. Bolts or weld can be used to
tion. The field connection is made by bolting through the attach to the supporting member. Standard round holes or
angle leg into the web of the filler beam. Horizontal short horizontal short slots can be used in both the flange and web
slots can be used in the field-connected leg to accommodate of the tee.
fabrication and erection tolerances. The slots also accom- This connection is both safe and fast to erect. However, it
modate the beam end rotation associated with cambering. is not the most economical connection because of the extra
In the Manual Part 10, the steel detailer will find addi- shop work involved in preparing the tee.
tional information, an example and Table 10-10 for designing In the Manual Part 10, the steel detailer will find addi-
single-angle connections. tional information for use in selecting tee connections.

Tee Connections
A tee connection is primarily a shear connection. The tee
FORCES IN WELDS
can either be cut from a W or HP shape or fabricated by
welding two plates together (useful for skewed connections). Welds may be loaded in shear, tension, compression or a com-
Only W and HP shapes that fit criteria for web and flange bination of these, depending on the direction and point of ap-
thicknesses, flange width and other dimensions are suitable plication of external forces and the arrangement of joint
for use as tee connections. To provide for stability during components. Available strengths for welds in buildings are
erection, the recommended practice is to furnish a tee having given in AISC Specification Section J2.4.
a minimum length equal to one-half the T-dimension of the The nomenclature commonly used in design computa-
supported beam. The maximum length must be compatible tions involving strengths of welds employs the following
with the T-dimension of an uncoped beam and the remaining symbols:
web depth, exclusive of fillets, of a coped beam.
The flange thickness of tees should be held to a minimum φRn  Design strength of weld or base metal using
to permit the flexure necessary to accommodate the end rota- LRFD, kips

Figure 3-14. Single-angle connection.

3-16 • Detailing for Steel Construction

Detailing_CH03 6/2/09 8:16 AM Page 17

COMMON CONNECTION DETAILS

Rn/Ω  Allowable strength of weld or base metal Refer to the Manual Part 8 for coverage of eccentric loading
using ASD, kips and its effects on weld joints. See Figure 3-16 for nomencla-
D  Number of sixteenths of an inch in fillet weld ture of the common terms related to fillet and groove welds.
size
FBM  Nominal strength of the base metal to be Forces in Concentrically Loaded Fillet Welds
welded per unit area, ksi The critical force in a fillet weld is always considered to be a
Fw  Nominal strength of the weld metal per unit shear force and may occur in one of two directions: (1) paral-
area, ksi lel to the axis of the weld or (2) transverse to the axis of the
φrn  Design strength of a 1-in. length of weld of a weld. Therefore, tension, compression and moment forces
given size using LRFD, kips acting upon a fillet welded joint are always resolved on the
rn /Ω  Allowable strength of a 1-in. length of weld basis of shear in the weld throat.
of a given size using ASD, kips Figure 3-17a shows a fillet welded lap joint with welds A
l  Effective length of weld, in. loaded in parallel shear and weld B loaded in transverse shear.
t  Material thickness, in. If loads Ru or Ra are increased enough to exceed the total
w  Weld size (leg size for fillet welds and depth strength of these welds, rupture will occur in the planes of
of filling for plug and slot welds), in. least resistance. As shown in Figure 3-17b, this is assumed
ABM  Cross-sectional area of the base material, in.2 to be in the weld throats, where the least cross-sectional area
Aw  Effective cross-sectional area of the weld, is present.
in.2 For design purposes, the effective area of a fillet weld is cal-
Ru  Required strength (load, force) using LRFD, culated as the product of its effective throat and its effective
kips length. The effective throat area1 equals 0.707wl. The deep
Ra  Required strength (load, force) using ASD, penetration of fillet welds made by the submerged arc process
kips is recognized in AISC Specification Section J2.2.
φ  Resistance factor The nominal strength of a fillet weld is equal to the prod-
Ω  Safety factor uct of its effective throat area and the nominal strength of the

1 Although this expression does not apply to the throat areas of fillet
The following discussion deals with forces in concentri-
welds in certain types of skewed work, or fillet welds with unequal
cally loaded welded joints in which the welds are assumed legs, its use in most cases will provide conservative results. However,
to be loaded uniformly throughout their lengths. Eccentrically if a critically stressed fillet weld has one leg substantially longer or
loaded welded joints, in which the welds are loaded in vary- shorter than the other, the actual throat size should be determined and
ing amounts along their lengths, are not discussed in this text. used in stress computations.

Figure 3-15. Tee connection.

Detailing for Steel Construction • 3-17

Detailing_CH03 6/2/09 8:16 AM Page 18

CHAPTER 3

Figure 3-16. Common fillet and groove weld nomenclature.

3-18 • Detailing for Steel Construction

Detailing_CH03 6/2/09 8:16 AM Page 19

COMMON CONNECTION DETAILS

weld metal, Fw , permitted by the specifications. Similarly, let weld strengths may be simplified by memorizing the avail-
the strength of a concentrically loaded fillet weld group is the able shear strength value of a 1/16-in. fillet weld, per in. of
sum of the strengths of each fillet weld in the group. Nominal length, for the most frequently used electrode strength levels.
strengths for the weld metal, Fw , for fillet welds permitted For example, the available shear strength value for a 1/16-
by AISC Specification Section J2.2 are shown in Table J2.5 in. fillet weld, 1-in. long, with FEXX  70 ksi, is:
of that specification.
Assuming that the Fw value is appropriate to the steel being φrn  0.75  (0.60  70)  (0.707  1/16  1)
welded and normal throat areas are applicable, the strength  1.392 kips/in. (LRFD)
of a fillet weld in shear may be expressed as:
rn /Ω  [(0.60  70)  (0.707  1/16  1)]/2.00
 0.928 kips/in. (ASD)
φRn  φ  Fw  Aw (LRFD)
φRn  0.75  (0.60 FEXX)  0.707wl (LRFD) Using the letter D to denote the number of sixteenths of
an inch in the leg size, the value per inch of any size of fillet
Rn / Ω  (Fw  A w)/Ω (ASD) weld (where FEXX  70 ksi) can be expressed as φrn  1.392D
Rn / Ω  [(0.60 FEXX)  0.707wl]/Ω (ASD) (LRFD) and rn /Ω  0.928D (ASD).

Rn  0.424FEXX wl, where FEXX is the electrode classi- Limitations on Length and Size of Fillet Welds
fication number (Grade) of the electrode The minimum effective length of a fillet weld, when used
alone and not as part of a continuing joint boundary, is lim-
Table 3-1 lists values of φrn and rn /Ω for fillet welds of ited to four times the nominal size. Thus, the shortest 5/16-
various sizes, based on nominal throat areas, for FEXX val- in. weld that can be considered fully effective to transmit
ues of 60, 70 and 80 ksi weld electrodes. Also included in loads is 4  5/16  11/4 in. Conversely, its maximum effective
this table are φrn values based on deep penetration throat size can be no greater than 1/4 of the weld length (see AISC
areas for welds made by the submerged arc process. Table Specification Section J2.2b).
3.1 in the American Welding Society Structural Welding Code- Intermittent fillet welds likewise are subject to the preced-
Steel (AWS D1.1) (AWS, 2008), lists the base and filler metals ing rules, with the added requirement that the increment length
that are permitted to be used with (“match”) each of the elec- must be not less than 11/2 in. (see AISC Specification Section
trodes stated earlier. AWS D1.1 is discussed in Chapter 4. AISC J2.2b).
Specification Table J2.5 states that filler metal one strength AISC Specification Section J2.2b places a limitation on
level less than matching filler metal is permitted to be used for the maximum size of fillet welds against the edges of con-
groove welds between the webs and flanges of built-up sec- nected parts of a joint. For material less than 1/4-in. thick, the
tions transferring shear loads or in applications where high re- weld size may equal the thickness of the material. For mate-
straint is a concern. The steel detailer is not responsible for rial 1/4-in. or more in thickness, the maximum size weld against
ensuring that this requirement is met, but should be aware of it. an edge shall be 1/16-in. less than the thickness of material,
When the available shear strength of fillet welds is based unless the drawing is noted specifically to build up the weld
on normal throat areas, as is usually the case, determining fil- to achieve full throat size. This limitation recognizes that the

Figure 3-17. Fillet weld shear failure.

Detailing for Steel Construction • 3-19

Detailing_CH03 6/2/09 8:16 AM Page 20

CHAPTER 3

Table 3-1. Design Strength of Fillet Welds (φ rn or rn / Ω), kips/in.

LRFD1 ASD1
φ rn rn / Ω
Weld Size
(in.) Grade 70 Grade 80 Grade 70 Grade 80
FEXX FEXX FEXX FEXX
70 ksi 80 ksi 70 ksi 80 ksi
1/8 2.78 3.18 1.86 2.12
3/16 4.18 4.77 2.78 3.18
1/4 5.57 6.36 3.71 4.24
5/16 6.96 7.95 4.64 5.30
3/8 8.35 9.55 5.57 6.36
7/16 9.74 11.1 6.50 7.42
1/2 11.1 12.7 7.42 8.48
9/16 12.5 14.3 8.35 9.54
5/8 13.9 15.9 9.28 10.6
3/4 16.7 19.1 11.1 12.7
1 22.3 25.4 14.8 17.0
1Values are for static loading only. See AISC Specification Appendix 3 for connections subject to high cycle loading.

upper corner of the welded edge tends to melt down into the 1/4-in. weld might suffice to transfer the force across the joint.
weld, as shown in Figure 3-18(a), reducing the leg dimen- However, where dissimilar thicknesses are encountered, the
sion and weld throat. Also, the toes of most rolled shapes weld size need not exceed the thickness of the thinner part, un-
are rounded as shown in Figure 3-18(b) and the actual thick- less a larger size is required by stress calculations. These size
ness of the weld is less than the nominal thickness, t, of the limitations have been established to avoid cracks that might
member. occur in the welds because of internal forces created by the
The AISC Specification also specifies the minimum size of rapid and often uneven cooling after welding. For quick refer-
fillet weld to be used in a joint. Section J2.2b provides Table J2.4 ence, the minimum fillet weld size required for a given material
showing minimum fillet weld sizes. The minimum weld size is thickness of the thinner part connected in a joint, based on
specified according to the thickness of the thinner of the two AISC Specification Section J2.2b, is given in Table 3-2.
parts joined. For example, if two 7/8-in. plates are joined, the AISC Specification Section J2.2b describes limitations on
minimum size of fillet weld permitted is 5/16 in., even though a fillet welds, which deal with lengths of fillet welds, proportions

Figure 3-18. Resultant fillet weld dimensions.

3-20 • Detailing for Steel Construction

Detailing_CH03 6/2/09 8:16 AM Page 21

COMMON CONNECTION DETAILS

Table 3-2. Minimum Permissible Fillet Weld Sizes

Rn/Ω  [(0.60  70)(1.0  0.50sin1.590°)
 (0.707wl)]/2.00 (ASD)
Material Thickness Minimum Size
of Thinner Part of Fillet Weld1, With l  1 in.,
Joined, in. (mm) in. (mm) Rn/Ω  22.3w kips/in. (ASD)
To 1/4 (6) inclusive 1/8 (3)
The available tensile rupture strengths of the plate per inch
Over 1/4 (6) to 1/2 (13) 3/16 (5) are:
Over 1/2 (13) to 3/4 (19) 1/ 4 (6) φRn  φt Fut, kips/in. (LRFD)
Over 3/4 (19) 5/16 (8) Rn /Ωt  Fut/Ωt, kips/in. (ASD)
1Leg dimension of fillet welds. Single pass wleds must be used.
Note: See Section J2.2b for maximum size of fillet welds. The available tensile yield strengths of the plate per inch are:

φRn  φt Fyt, kips/in. (LRFD)

of lap joints, end returns of fillet welds, and fillet welds in Rn /Ωt  Fyt/Ωt, kips/in. (ASD)
holes and slots. The steel detailer must become familiar with
all of these limitations in the AISC Specification. The weld size to develop the full tensile strength of the
plate can be determined by equating these two values.
Strength of Connected Material For Fy  36 ksi steel:
The transfer of factored load from one element to another is
dependent on both the design strength of the weld and the φRn  0.75  58  t  33.4w (LRFD)
design strength of the materials to resist factored loads. Rn /Ωt  58  t/2.00  22.3w (ASD)
A simple example is illustrated in Figure 3-19. The plate N
is checked for available tension strength. Note that the weld w  1.30t
is transversely loaded. Using AISC Specification Equation
φRn  0.90  36  t  33.4w (LRFD)
J2-4 with E70XX electrodes, the weld available strengths per
1-in. length are: Rn /Ωt  36  t/1.67  22.3w (ASD)

φRn  φ  (0.60  FEXX) (1.0  0.50sin1.5θ)(Aw ) (LRFD) w  0.97t

φRn  0.75  (0.60  70)(1.0  0.50sin1.590°) The smallest of these two values will control the design;
 (0.707wl) (LRFD) therefore, w  0.97t. For two lines of fillet weld as shown in
With l  1 in., Figure 3-19:

φRn  33.4w kips/in. (LRFD) w  1/2  0.97t  0.48t for each line of weld
A different case is shown in Figure 3-20. Here, an angle is
Rn/Ω  [(0.60  FEXX) (1.0  0.50sin1.5θ)(Aw )]/Ω welded to a plate or the stem of a tee. Given that the avail-
(ASD) able shear strength, φRn or Rn/Ω, of the weld exceeds the re-
quired axial strength, Pu or Pa , the plate (or tee stem) must be
checked for tension yielding on the Whitmore section. This
section is identified by the width of the plate, Lw, considered
to be resisting the load. The value of the width, Lw, must be
equal to or less than the actual width of the plate, L, avail-
able for resisting the load. Thus, the available tensile strength
from AISC Specification Section J4.1 is:

φRn  φ  Fy  Ag eff ≥ Pu, with φ  0.90 (LRFD)

Rn/Ω  (Fy  Ag eff)/Ω ≥ Pa , with Ω  1.67 (ASD)
Fy  specified minimum yield stress of the type of
steel being used, ksi
where
Figure 3-19. Plate loaded in tension. Ag eff  (t  Lw) ≤ (t  L), in.2

Detailing for Steel Construction • 3-21

Detailing_CH03 6/2/09 8:16 AM Page 22

CHAPTER 3

Lw  [2  l  tan (30°)]  d, in. as a limit state. A limit state is a condition in which a structure
or component becomes unfit for service and is judged either
Secondly, shear yielding of the base metal along the toe and to be no longer useful for its intended function (serviceability
heel of each fillet weld line must be checked. Thus, the available limit state) or to be unsafe (strength limit state). The block
shear yield strength of the plate from AISC Specification Section shear strength is determined by the sum of the shear strength
J4.2 is: on a failure path(s) and the tensile strength on a perpendicu-
lar segment. AISC Specification Section J4.3 provides an
φRn  φ  (0.60  Fy)  Ag ≥ Pu, with φ  1.00 (LRFD) equation for determining nominal block shear rupture strength,
Rn. The basis for the equation is the existence of two possible
Rn/Ω  [(0.60  Fy)  Ag]/Ω ≥ Pa, with Ω  1.50 (ASD)
nominal shear strengths:
where Figure 3-21a — Rupture strength on the tensile section
Ag  4  l  t, in.2 (The factor 4 refers to the number of (Fu  Ant) along with shear rupture on the
shear planes) shear planes (0.60  Fu  Anv ).
Figure 3-21b — Rupture strength on the tensile section
Thirdly, the plate (or tee stem) must be checked for shear (Fu  Ant) along with shear yielding on
rupture of the base metal along the toe and heel of each weld the shear areas (0.60  Fy  Agv).
line. Thus, the available shear rupture strength from AISC
Specification Section J4.2 is: The steel detailer is referred to the CD attached to the
Manual for an example of the calculations of the available
φRn  φ  0.60  Fu)  An ≥ Pu, with φ  0.75 (LRFD) strengths illustrated in Figures 3-20 and 3-21. Note that the
Rn/Ω  [(0.60  Fu)  An ]/Ω ≥ Pa, with Ω  2.00 (ASD) rupture strength on the tensile section is compared to the rup-
ture and yield strength on the shear areas—the smaller sum of
where these values determines the block shear rupture strength of
the plate (or tee stem). For an additional explanation of the
Fu  specified minimum tensile strength of the type of
rupture phenomenon, the steel detailer should read AISC
steel being used, ksi
Specification Commentary Section J4.
An  4  l  t, in.2 (The factor 4 refers to the number of For reasons of shear lag, AISC Specification Section D3.3
shear planes) limits the l/d ratio (Figure 3-21) to obtain the maximum ef-
fective tension area. Shear lag occurs when less than the
Figure 3-21 illustrates failure of the plate (or tee stem) by total cross-sectional area of a member is connected, result-
block shear rupture. Any potential way for a connection to ing in a decrease in efficiency of the section in the region
fail (block shear rupture, bearing, tearout, etc.) is referred to of the connection.

Figure 3-20. Whitmore section yielding of an angle welded to a plate or tee stem.

3-22 • Detailing for Steel Construction

Detailing_CH03 6/2/09 8:16 AM Page 23

COMMON CONNECTION DETAILS

The foregoing discussion does not apply, necessarily, to The following table lists the values of tw for ASTM A36
the case of double angles welded to the end of the web of a (Fu  58 ksi) and A992 (Fu 65 ksi) steels for varying values
girder. In this case the load transferred into the web is more of D:
complex and depends on several factors, such as the distance
from the connection to the girder reaction and the relative
D tw for A36 tw for A992
length of the connection angle fillet weld to the depth of the
girder. The usual practice is to assume only one shear plane. 3 0.32 0.29
Using nominal strengths, 4 0.43 0.38
5 0.53 0.48
Weld fracture Plate fracture 6 0.64 0.57
7 0.75 0.67
0.60 × FEXX × D × 2
1
≤ 0.60 × Fu × tw 8 0.85 0.76
16 × 2 2

FEXX  classification strength of the weld metal, ksi

for each weld length. The minimum girder web thickness, tw, D  number of sixteenths of an inch of weld size
is calculated as follows: Fu  specified minimum tensile strength of the type of
Longitudinal shear in the weld is compared to the shear in steel being used, ksi
the girder web plate with θ  0 (angle of loading measured tw  thickness of web plate, in.
from the weld longitudinal axis, degrees)
A similar comparison between transverse shear in the
Where weld and tension in the web plate (θ  90°) provides a
plate only slightly thinner. Therefore, to check the required
0.0884 × FEXX × D minimum web plate thickness, use the value for tw given
tw ≥ previously. If the weld is not loaded fully, tw may be re-
Fu
duced proportionately.
with
Forces in Complete-Joint-Penetration Groove Welds
AISC Specification Section J2.4 establishes the available
FEXX = 70;
tensile and compressive strength of complete-joint-penetra-
tw ≥ 6.19 × D tion (CJP) groove welds as being equal to the available
Fu strength of the base metal, providing a “match” of base and

(a) (b)

Figure 3-21. Block shear rupture of a plate or a tee stem.

Detailing for Steel Construction • 3-23