Missouri University of Science and Technology

Scholars' Mine
International Specialty Conference on Cold- (2010) - 20th International Specialty Conference
Formed Steel Structures on Cold-Formed Steel Structures

Nov 3rd

Screw Connections Subject to Tension Pull-out

and Shear Forces
R. M. Francka

R. A. LaBoube
Missouri University of Science and Technology, laboube@mst.edu

Follow this and additional works at: http://scholarsmine.mst.edu/isccss

Part of the Structural Engineering Commons

Recommended Citation
Francka, R. M. and LaBoube, R. A., "Screw Connections Subject to Tension Pull-out and Shear Forces" (2010). International Specialty
Conference on Cold-Formed Steel Structures. 5.
http://scholarsmine.mst.edu/isccss/20iccfss/20iccfss-session11/5

This Article - Conference proceedings is brought to you for free and open access by Scholars' Mine. It has been accepted for inclusion in International
Specialty Conference on Cold-Formed Steel Structures by an authorized administrator of Scholars' Mine. This work is protected by U. S. Copyright
Law. Unauthorized use including reproduction for redistribution requires the permission of the copyright holder. For more information, please contact
scholarsmine@mst.edu.
 Twentieth International Specialty Conference on Cold-Formed Steel Structures
St. Louis, Missouri, U.S.A., November 3 & 4, 2010

Screw Connections Subject

to Tension Pull-Out and Shear Forces

R. M. Francka1 and R. A. LaBoube2

Abstract

Currently, the behavior of screw connections subject to combined tension pull-

out and shear forces is not well understood. An experimental study was
conducted at Missouri University of Science and Technology to better
understand the relationship or interaction between these forces. The test
program evaluated four parameters that may influence the behavior of pure
tension and pure shear in screw connections: the thickness of the sheet not in
contact with the screw head, the ultimate strength of the steel, the ductility of the
steel, and the screw diameter. Based on the behavior observed and analysis of
the test data, this work formulated new design recommendations for use in
calculating the design capacity of screw connections subject to this potential
limit state.

INTRODUCTION

Screws are a practical and economical means to connect cold-formed steel

structural members. They provide a rapid and effective way of connecting
members subject to tension, shear, or combined tension and shear forces. For
example, common construction methods often use clip angles to connect bracing
members or joists to supporting rim joists (Figure 1). These clip angles may be
subject to simultaneous tension and shear forces.

In 1946, the American Iron and Steel Institute (AISI) began leading the building
industry with the release of its first edition of the Specification for the Design of
Light Gage Steel Structural Members (AISI, 1946). The most recent edition, the
North American Specification for the Design of Cold-Formed Steel Structural
Members (AISI S100), was released in 2007.

1
Former graduate student, Missouri S&T, Rolla, MO
2
Curators Distinguished Teaching Professor, Missouri S&T, Rolla, MO

635
 636

Currently, the specification includes provisions that assess the design strength of
a screw connection subject to pure tension, pure shear, and combined tension
pull-over and shear forces. Additional guidance is required to determine the
design capacity when screw connections are subject to both combined tension
pull-out and shear forces.

Figure 1 Screw Connections Potentially Subject to Pull-out and Shear Forces

LITERATURE REVIEW

Several research studies provide the foundation for this research study (Pekoz,
1990; Zwick and LaBoube, 2006). Pekoz investigated screw connections
subject to pure tension pull-out and pure shear forces alone. Zwick and
LaBoube studied the consequence of combined pull-over and shear loading on
screw connections. Additional information pertaining to these studies and the
behavior of screw connections is given by Yu and LaBoube (2010).

These research studies form the basis for the design provisions of the North
American Specification for the Design of Cold-Formed Steel Structural
Members (AISI S100, 2007). The AISI S100 nominal strength, Pn, are as
follows:

For shear alone the nominal shear strength shall be calculated as follows:
If t2/t1 ≤ 1.0, Pns shall be taken as the smaller of

Pns = 4.2(t 23 d)1/ 2 Fu 2 (1)

Pns  2.7t1dFu1 (2)
Pns  2.7t 2 dFu 2 (3)

If t2/t1 ≥ 2.5, Pns shall be taken as the smaller of

 637

Pns  2.7t1dFu1 (4)

Pns  2.7t 2 dFu 2 (5)

If 1.0 < t2/t1 < 2.5, Pns shall be calculated by linear interpolation
between the above two cases.

Where d = nominal screw diameter, Pns = nominal shear strength per screw, t1 =
thickness of member in contact with screw head or washer, t2 = thickness of
member not in contact with screw head or washer, Fu1 = tensile strength of
member in contact with screw head or washer, Fu2 = tensile strength of member
not in contact with screw head or washer.

For tension alone the nominal pull-out strength, Pnot, shall be calculated
as follows:

Pnot  0.85t c dFu 2 (6)

Where d = nominal screw diameter, tc = lesser of the depth of penetration and

thickness t2, Pnot = nominal pull-out strength per screw, Fu2 = tensile strength of
member not in contact with screw head or washer.

For tension alone the nominal pull-over strength Pnov, shall be calculated as
follows:

Pnov  1.5t1d w ' Fu1 (7)

Where t1 = thickness of member in contact with screw head or washer, Fu1 =

tensile strength of member in contact with screw head or washer, dw’ = effective
pull-over diameter determined in accordance with (a), (b), or (c) as follows:
(a) for a round head, a hex head, or hex washer head screw with an
independent and solid steel washer beneath the screw head
dw’ = dh +2tw +t1 ≤ dw
where
dh = screw head diameter or hex washer integral washer
diameter
tw = steel washer thickness
dw = steel washer diameter
(b) for a round head, a hex head, or hex washer head screw without an
independent washer beneath the screw head:
dw’ = dh but not larger than ½ in.
 638

(c) for a domed (non-solid and independent) washer beneath the screw
head, it is permissible to use dw’ as calculated in (a), with dh as the
washer diameter, tw as the thickness of the material of the washer,
and t1 as previously defined. dw’ cannot exceed 5/8 in.
Alternatively, pull-over design values for domed washers, shall be
permitted to be determined by test in accordance with Chapter F of
AISI S100.

For screw connections subject to combined shear and tension pull-out the
following nominal strength relationship applies

Q T 1.10 (8)
 0.71 
Pns Pnov 

Where Q = required allowable shear strength of connection, T =

required allowable tension strength of connection, Pns = nominal shear
strength per screw = 2.7 t1d Fu1, Pnov = nominal pull-over strength of
connection = 1.5 t1dw Fu1, dw = larger of screw head diameter or washer
diameter.

Equation 8 is valid for connections that meet the following limits: 0.0285 in. ≤ t1
≤ 0.0455 in., No. 12 and No. 14 self-drilling screws with or without washers,
dw≤ 0.75 in., Fu1 ≤ 70 ksi, and t2/t1 ≥2.5.

EXPERIMENTAL INVESTIGATION

Parameters evaluated in this study were the thickness of the sheet not in contact
with the screw head or washer, the tensile strength of the material, the ductility
of the material, and the screw diameter.

The mechanical properties of the sheet steel used in this investigation were
determined by performing tensile coupon tests in accordance with ASTM A 370
(2007). Table 1 summarizes the results of the coupon tests and lists these
properties: uncoated sheet thickness, yield stress, tensile strength and percent
elongation. The notations N and L indicate the normal- and low-ductility steels,
respectively.
 639

Table 1 Material Properties

Uncoated Yield Tensile
Elongation
Thickness Stress Strength
Specimen Fu/Fy
t Fy Fu
%
(in) (ksi) (ksi)
20N 0.0297 41.41 48.295 1.166 42.58
18N 0.0394 29.25 47.315 1.618 38.38
16N 0.0521 62.205 75.49 1.214 29.69
14N 0.0724 68.39 74.32 1.087 34.38
20L 0.0327 102.75 105.99 1.032 2.34
18L 0.0375 91.175 91.18 1.000 1.17
16L 0.0508 84.25 89.645 1.064 3.91
14L 0.0675 117 120.565 1.030 2.73

Test Fixture. The test fixture consisted of a welded T-section and a rotating
arm. This test fixture was essentially the same fixture as previously used by
Stirnemann and LaBoube (2008). Welded T-sections were fabricated at 15°,
30°, 60°, and 75°. These variations in the angle of orientation induced different
combinations of tension and shear forces, thus providing a range of data to
define the interaction of tension pull-out and shear forces. The majority of the
tests used three angles; fifteen degrees, thirty degrees, and sixty degrees, but a
few tests were also completed at seventy-five degrees (Figure 2).

(a) 15° (b) 30° (c) 60° (d) 75°

Figure 2. Angles of Rotation
 640

Test Specimen. Each test specimen consisted of a 12 in. x 12 in. deck section
screwed to a 6 in. x 2 in. or 3 in. flat sheet (Figures 3 and 4). Details pertaining
to the test specimen parameters and fabrication can be found in Francka and
LaBoube (2009).

Figure 3 Typical Test Specimen Configuration

Figure 4 Illustration of Flat Sheet Attached to Test Fixture Plate

Test Procedure. Each prepared test specimen was mounted in an MTS 880
Material Test System (Figure 2). A computer data acquisition system recorded
the load and displacement during each test. Load and displacement were
recorded for each test at eight intervals per second to ensure that the maximum
load was recorded.

During the initial testing, distortion of the flat sheet was observed thus the
stiffness of flat sheet in the test specimen was further evaluated. Normal-
ductility test specimens were stiffened using a brake press. Each of the long
sides was bent to form ½ in. edge stiffeners (Figure 5).
 641

Figure 5 Flat Sheet with Edge Stiffeners

Table 2 summarizes the tests performed to assess the contribution of the stiffer
sheet on the connection strength.

Table 2 Comparison of Stiffened versus Unstiffened Specimens

Ultimate Strength (lbf)
Unstiffened Specimen Average Stiffened Specimen Average
15° 468.8 494.3 481.6 374.3 452.4 413.4
Rotation
Angle of

30° 352.7 353.6 353.2 321.1 365.3 343.2

60° 320.9 337.6 329.3 317.7 323.5 320.6

Tilting of the screw and tearing were the failure modes observed in both
stiffened and unstiffened specimens (Figure 6). Based on Table 2 and a
comparison of the load versus deflection curves (Figure 7), the stiffness of the
test specimens did not affect the ultimate strength of connections subject to
combined tension pull-out and shear.

Figure 6 Comparison of Stiffened versus Unstiffened Failure Modes

 642

400
350
300
250
Load (lbf)

20N14‐30‐1S
200
20N14‐30‐2S
150
20N14‐30‐1
100 20N14‐30‐2
50
0
0 0.2 0.4 0.6 0.8
Deformation (in.)

Figure 7 Load versus Deflection of 30° Test Specimens

TEST RESULTS

A total of eighty-four tests were performed. Thirty-nine were normal-ductility

test specimens, and thirty-six were low-ductility test specimens.

Each test specimen was tested until failure. If the screw failed, the test was
classified as inconclusive for purposes of this study and removed from the
results. Screw failures occurred only in angles introduced to larger shear
components, specifically 15° and 30°.

A typical load versus displacement curve is shown in Figure 8. The peaks of the
curve represent the points at which the threads of the screw were pulled through
the hole. As each layer of threads caught the sheet, the connection gained
strength until it reached the peak strength of those threads and so on and so
forth. The ultimate strength of the connection, Pu, was defined as the highest
load carried during loading, regardless of deformation.
 643

The typical failure mode observed in all tests was a combination of screw pull-
out (tension failure), tilting of the screw (shear failure), and bearing of the sheet
(shear failure). However, the normal- and low-ductility specimens did perform
differently with respect to deformation and strength.

140
120
100
Load (lbf)

80
60
40
20
0
0.00 0.10 0.20 0.30 0.40 0.50

Deformation (in)

Figure 8 Example Load versus Deformation Curve

Normal-Ductility Specimens. All normal-ductility specimens experienced

plastic deformation. Figure 9 shows a typical normal-ductility specimen after
testing. Given the same sheet thickness and screw diameter, the normal-
ductility steel deformed more than the low-ductility steel, and tearing of the
sheet was more prominent. Figure 10 shows a normal-ductility specimen (18N)
and a low-ductility specimen (18L). The distortion of the sheet was typical of
all normal-ductility and low-ductility specimens. The distortion was not an
effect of eccentricity, but rather of the combination of the tension pull-out and
shears forces.

The ultimate strength, Pu, was determined from the recorded data. Based on the
angle of the test, the ultimate tension and ultimate shear forces Put and Puv,
respectively, were calculated using basic trigonometry. Table A.1 of Francka
and LaBoube (2009) provides all of the test data.
 644

Figure 9 Typical Normal-Ductility Flat Sheet after Testing

Figure 10 Comparison of Normal- and Low-Ductility Flat Sheets

Low-Ductility Specimens. Low-ductility specimens typically experienced less

plastic deformation than the normal-ductility specimens. Figure 10 shows a
low-ductility specimen (18L) and a normal-ductility specimen (18N). The low-
ductility specimens had less deformation, and tilting of the screw was more
prominent due to the resistance of the steel to allow tearing to occur (Figures 10
11). The same distortion effects observed in the normal-ductility specimens
were apparent in the low-ductility tests, but they were typically less prominent.
Many low-ductility specimens never reached inelasticity, and deformation was
not permanent.
 645

Figure 11 Typical Low-Ductility Flat Sheet after Testing

The ultimate strength, Pu, was determined as for normal-ductility. The tension
and shear components were also determined. Table A.2 of Francka and
LaBoube (2009) contains the complete test data information.

DATA ANALYSIS USING AISI EQUATIONS

Using design Equations 1 through 6, the nominal strengths were calculated for
tension pull-out, Pnot, and shear, Pns. The ultimate load applied to each test
specimen was evaluated for its tension and shear components, Put and Puv,
respectively. These ultimate strength components were then normalized using
the nominal strength equations to form the ratios Put/Pnot and Puv/Pns. Francka
and LaBoube (2009) presents complete details pertaining to the analysis results
for the normal- and low-ductility test data, respectively.

Evaluation of Screw Diameter. Influence of the screw diameter was

investigated to assess its’ impact on the connection capacity. All of the tests
performed for the 30° angle configuration used a broad range of screw sizes
(No. 8, 10, 12, and 14). Figures 12 and 13 show a graph of the normalized shear
strength, Puv/Pns, versus the normalized tension pull-out strength, Put/Pnot.
 646

0.90
0.85
0.80
0.75
0.70 No. 8 Screw
Put/Pnot

0.65 No. 10 Screw
0.60 No. 12 Screw
0.55 No. 14 Screw
0.50
0.45
0.40
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Puv/Pns

Figure 12 Evaluation of Screw Size - Normal Ductility

Based on the distribution of the data for all screw diameters at 30° although
screw diameter affected the overall strength of the connection, it did not
influence the interaction of the combined loading. These conclusions justified a
reduction in the number of tests required for this study. The other test angle
configurations were tested using only one screw size. At 60° No. 10 screws
were used. At 15°, however, No. 14 screws were used due to the large shear
loads being induced. For tests performed at 75° degrees used No. 8 and No. 10
screws were used.
 647

0.70

0.65

0.60

0.55
No. 8 Screw
Put/Pnot

0.50 No. 10 Screw
0.45 No. 12 Screw
No. 14 Screw
0.40

0.35

0.30
0.2 0.3 0.4 0.5 0.6 0.7

Puv/Pns
Figure 13 Evaluation of Screw Size – Low Ductility

Shear versus Tension Pull-out. To illustrate the interaction between tension

pull-out and shear forces within a screw connection, Figure 14 provides the
ratios of ultimate strength to nominal strength, Puv/Pns versus Put/Pnot. As
illustrated, a relationship is apparent between the normalized tension pull-out
and shear forces. Clearly the normal-ductility test specimens performed at a
higher normalized capacity than did the low-ductility test specimens.
 648

1.4

1.2

1.0
Normal 
Put/Pnot

0.8 Ductility

0.6 Low Ductility

0.4

0.2

0.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

Puv/Pns
Figure 14 Pull-out and Shear Interaction using AISI Equations

DEVELOPMENT OF INTERACTION EQUATION

Figures 15 and 16 summarize the test data and the normalized relationships
between the shear force and the tension pull-out force. Several nonlinear and
linear interaction equations were investigated to achieve a desirable mean,
standard deviation, and coefficient of variation. An adjustment factor, L, was
implemented to reflect the behavior of the low-ductility steel test specimens.
The following presents the best-fit cases for a tri-linear and nonlinear interaction
equation.
 649

Tri-Linear Interaction Equation. The proposed tri-linear interaction equation,

Equation 9, was derived using the data shown in Figure 14. The complete data
summary can be found in Tables B.1 and B.2 of Francka and LaBoube (2009).
The mean value and coefficient of variation used to determine appropriate
resistance and safety factors (  for LRFD and LSD, and Ω for ASD) are also
presented in Tables B.1 and B.2. Figure 15 illustrates the correlation between
the test data and Equation 9, for both with the normal- and low-ductility test
data. Based on Figure 15,

when: Puv/Pns ≥ 0.15 and Put/Pnot ≥ 0.15

Puv P (9)
 ut  1.15
LPns LPnot

Where L = 1.0, for Fu/Fy ≥ 1.087, L = 0.75, for Fu/Fy ≤ 1.064,

Pns = 4.2(t 23 d)1/ 2 Fu 2 nominal shear strength of connection, Eq. 1
Pnot  0.85t c dFu 2 nominal pull-out strength of connection, Eq. 6

1.4
1.2
Normal Ductility
1.0
Put/Pnot

0.8 Low Ductility
0.6
0.4 Equation 9 (Normal 
Ductility)
0.2
Equation   9         
0.0 (Low Ductility)
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

Puv/Pns
Figure 15 Tri-Linear Equation Interaction Relationship
 650

Nonlinear Interaction Equation. The proposed nonlinear interaction equation

is Equation 10 and is shown on Figure 16. It was derived using the data shown
on Figure 14 and data from Tables B.1 and B.2 of Francka and LaBoube (2009).
The mean value and coefficient of variation were used to determine appropriate
resistance and safety factors. This evaluation can be found in Tables D.1 and
D.2 of Francka and LaBoube (2009).
1.15 1.15
 Puv   P  (10)
    ut   1.0
 LPns   LPnot 

where:
L = 1.0, for Fu/Fy ≥ 1.087,
L = 0.80, for Fu/Fy ≤ 1.064,
Pns = 4.2(t23 d)1/ 2 Fu 2 nominal shear strength of connection, Eq. 1
Pnot  0.85t c dFu 2 nominal pull-out strength of connection, Eq. 6

1.4
Normal Ductility
1.2

1.0 Low Ductility
Put/Pnt

0.8
Equation 10       
0.6 (Normal Ductility)
0.4
Equation 10          
0.2 (Low Ductility)

0.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
Puv/Pnv
Figure 16 Nonlinear Equation Interaction Relationship
 651

CONCLUSIONS

This study assessed the interaction relationship between tension pull-out and
shear forces in screw connections of cold-formed structural steel structural
members. A total of eighty-four tests were performed. Based the evaluation of
the test data interaction equations were proposed for use in designing screw
connections subject to this limit state.

BIBLIOGRAPHY

American Iron and Steel Institute. (1946). Specification for the Design of Light
Gage Steel Structural Members (1946 ed.). New York, NY: AISI.

North American Specification for the Design of Cold-Formed Steel Structural

Members, AISI S100-2007, American Iron and Steel Institute, Washington D.C.

ASTM International. (2007), “A370-071 Standard Test Methods and Definitions

for Mechanical Testing of Steel Products,”: ASTM International, West
Conshohocken, PA.

Francka, R.M. and LaBoube, R.A. (2009), “Screw Connections Subject to

Tension Pull-Out and Shear Forces,” Final Report, Wei-Wen Yu Center for
Cold-Formed Steel Structures, Rolla, MO.

Pekoz, T. (1990), “Design of Cold-Formed Steel Screw Connections”

Proceedings of the Tenth Specialty Conference on Cold-Formed Steel
Structures”, Wei-Wen Yu Center for Cold-Formed Steel Structures, Rolla, MO.

Stirnemann, L.K. and LaBoube, R.A. (2008), “Behavior of Arc Spot Weld
Connections Subjected to Combined Shear and Tension Forces,” Proceedings of
the 19th International Specialty Conference on Cold-Formed Steel Structures,
Wei-Wen Yu Center for Cold-Formed Steel Structures, Rolla, MO.

Yu, W.W. and LaBoube, R.A. (2010). Cold-Formed Steel Design (4th ed.), John
Wiley and Sons Inc., New York, NY

Zwick, K., & LaBoube, R. A. (2006), “Self-Drilling Screw Connections

Subjected to Combined Shear and Tension”, Final Report, Wei-Wen Yu Center
for Cold-Formed Steel Structures, University of Missouri-Rolla, Civil,
Architectural, and Environmental Engineering, Rolla, MO