Seismic Performance Characterization of Wood-Sheathed

and Cold-Formed Steel Framed Floor and Roof

Diaphragm Structures
Violetta Nikolaidou, S.M.ASCE 1; Patrick Latreille 2; Dimitrios G. Lignos 3; and Colin A. Rogers 4
Downloaded from ascelibrary.org by University Of Sydney on 03/23/23. Copyright ASCE. For personal use only; all rights reserved.

Abstract: This paper describes a research program involving wood-sheathed and cold-formed steel (CFS) framed diaphragm assemblies.
The diaphragm’s response to in-plane monotonic and reversed cyclic lateral loading is investigated in an effort to characterize the seismic
performance of this assembly. The work presented herein focuses on the response to loading of the isolated diaphragm subsystem and serves
as a complementary study to a research project involving the dynamic testing of full-scale two-story CFS framed buildings, known as the
CFS–Network for Earthquake Engineering Simulation (NEES) project. Laboratory testing included eight 3.66 × 6.1-m diaphragm speci-
mens, that is, four configurations, comprising oriented strand board (OSB) sheathing screw connected to CFS C-Channel joists. The response
to loading is directly related to screw pattern and size, the use of panel edge blocking, and the type of sheathing. By means of a comparison of
design and experimental shear strength and stiffness values, the provisions of the AISI S400 standard were shown to be in need of improve-
ment regarding the number of listed diaphragm configurations. Deflection predications at the design load level were considered to be
reasonable. DOI: 10.1061/(ASCE)ST.1943-541X.0001962. © 2017 American Society of Civil Engineers.
Author keywords: Cold-formed steel; Diaphragm; In-plane loading; Test program; Shear response; Metal and composite structures.

Introduction of CFS framed buildings, because little research exists in which the
diaphragm response is the focal point of the work (NAHRBC 1999;
A typical construction practice of cold-formed steel (CFS) struc- LGSEA 1998). A shear wall is effectively considered as a vertical
tures is the stud wall system with vertical members forming the cantilevered diaphragm (APA 2007); thus, the structural similarity
walls, and sheathing installed to provide shear resistance to lateral between shear walls and diaphragms enables preliminary assess-
loads (shear walls). A typical floor and roof system comprises ments of the diaphragm response through use of the shear wall
discretely or continuously braced CFS joists overlaid with wood studies. However, the major role of the diaphragm in distributing
sheathing, again to provide shear resistance to lateral loads (dia- the lateral forces to the shear walls, and the structural difference of
phragms). The seismic design of CFS framed structures focuses the diaphragm’s multiple sheathing panels call for an explicit char-
mainly on the lateral response of the shear walls as the primary acterization of its seismic response. The design provisions available
component of the lateral force–resisting system (LFRS), without for CFS framed diaphragms (AISI 2015, 2016; NIST et al. 2016;
explicitly accounting for the diaphragm’s contribution to the overall CSA 2016) are based largely on experimental work on wood as-
seismic response of the structure. Extensive experimental and semblies (Tissell and Elliot 2004; APA 2007); moreover, the North
numerical work realized for the lateral response of shear walls, American standard for the seismic design of cold-formed steel struc-
for example, Dolan and Easterling (2000), Serrette et al. (2002), tural systems, AISI (2015), contains no seismic design procedure
Branston et al. (2006), Pan and Shan (2011), Shamim et al. for CFS framed diaphragms for use in Canada. As such, there exists
(2013), and Peterman et al. (2016a, b), among others, provides a need for this shortcoming to be addressed in order to ensure the
a starting point in the effort to characterize the diaphragm behavior construction of better, safer, and cost-effective CFS structures.
under in-plane loading and its contribution to the seismic response The present design process for diaphragms is solely governed
1
by the selection of suitable connections between the sheathing and
Ph.D. Candidate, Dept. of Civil Engineering and Applied Mechanics, the framing, as well as between the diaphragm and the shear walls,
McGill Univ., 817 Sherbrooke St. West, Montreal, QC, Canada H3A 0C3.
in order to ensure adequate shear strength and stiffness. Currently,
E-mail: [email protected]
2
Master’s Student, Dept. of Civil Engineering and Applied Mechanics,
in the AISI S400 standard design shear strength values are provided
McGill Univ., 817 Sherbrooke St. West, Montreal, QC, Canada H3A 0C3. based on analytical work by the Light Gauge Steel Engineers
E-mail: [email protected] Association (LGSEA 1998) (Table F2.4-1, AISI 2015). These de-
3 sign values are dependent on the field and perimeter screw spacing,
Associate Professor, School of Architecture, Civil and Environmental
Engineering, Swiss Federal Institute in Lausanne (EPFL), GC B3 485 but not on the screw size, and are available only for a limited num-
(Bâtiment GC), Station 18, CH-1015 Lausanne, Switzerland. E-mail: ber of plywood-sheathed/CFS framed diaphragm configurations
[email protected] based on the methodology included in Tissell and Elliot (2004)
4
Associate Professor, Dept. of Civil Engineering and Applied for wood framing. Moreover, Serrette and Chau’s (2003) work
Mechanics, McGill Univ., 817 Sherbrooke St. West, Montreal, QC, Canada
yielded a deflection equation for simply supported diaphragms,
H3A 0C3 (corresponding author). E-mail: [email protected]
Note. This manuscript was submitted on February 2, 2017; approved on
which is included in the AISI S400 standard (Eq. C-F2.4.3-1,
August 9, 2017; published online on December 15, 2017. Discussion AISI 2015). Shear strength and stiffness values were also made
period open until May 15, 2018; separate discussions must be submitted available by the National Association of Home Builders Research
for individual papers. This paper is part of the Journal of Structural En- Center (NAHBRC 1999), which carried out four monotonic tests
gineering, © ASCE, ISSN 0733-9445. on CFS framed/oriented strand board (OSB)-sheathed diaphragms,

© ASCE 04017215-1 J. Struct. Eng.

J. Struct. Eng., 2018, 144(2): 04017215

Downloaded from ascelibrary.org by University Of Sydney on 03/23/23. Copyright ASCE. For personal use only; all rights reserved.

Fig. 1. Cold-formed steel–NEES building (images courtesy of Dr. Kara Peterman, University of Massachusetts Amherst): (a) elevation; (b) ground
floor

and studied the individual sheathing-to-framing connection re- (Fig. 1). The tests were conducted in the Jamieson Structures
sponse. The launch of the CFS–Network for Earthquake Engineer- Laboratory at McGill University following a cantilever test method,
ing Simulation (NEES) project in 2010 was in response to the need detailed according to the provisions of the AISI S907 standard
for advanced seismic design procedures of CFS structures. This (2013) for diaphragm testing, with the overall specimen dimensions
major research project involved the dynamic testing of a full-scale being 3.66 × 6.1 m (Nikolaidou et al. 2015). This paper concludes
two-story CFS framed building (Fig. 1), which was conducted by with a comparison between the measured test values and the
researchers at Johns Hopkins University (Peterman 2014). Particu- calculated shear strength and deflection values following the AISI
lar emphasis was placed on the characterization of the isolated CFS S400 North American standard (2015) for the seismic design of
framed/wood-sheathed shear walls (Peterman et al. 2016a, b; Liu cold-formed steel structural systems.
et al. 2012), whereas the diaphragms in this structure were not
specifically instrumented such that their load-deformation response
could be measured; nor based on observations, were they reported Cold-Formed Steel Framed Diaphragm Test Program
to have surpassed the elastic range.
The research presented herein aims to provide insight into the The research program required the design and construction of a set-
complex nature of the seismic response of the diaphragm subsys- up to accommodate the diaphragm tests (Figs. 2 and 3). It consisted
tem. Eight OSB-sheathed/CFS framed roof and floor diaphragms of a pin-connected self-reacting braced frame with wide-flange
were tested, using either monotonic or reversed cyclic loading. The (W-shape) sections as the main beams and double angle sections
objective was to characterize the diaphragm response to in-plane as the bracing. The design aimed for the frame to remain elastic
loading and to obtain information for an isolated diaphragm’s seis- during the test and to have adequate stiffness to exhibit the mini-
mic performance to supplement the data acquired in the CFS-NEES mum possible deformation, that is, span-length/1125. A 450-kN
project. To this end, the diaphragm configurations were based on (tension)/650-kN (compression) actuator, hinged at both ends,
the floor and roof configurations used in the CFS-NEES building was attached to a force distribution beam, which was in turn bolted

Fig. 2. Cold-formed steel diaphragm test set-up

© ASCE 04017215-2 J. Struct. Eng.

J. Struct. Eng., 2018, 144(2): 04017215

 and shorten) perpendicular to the direction of the applied loading.
The specimen was fixed along the other side to the frame. Selected
photographs of the test specimen and set-up are provided in Fig. 4.
The roof and floor diaphragms of the CFS-NEES building had
the following characteristics: steel thickness 1.37 versus 2.46 mm,
#8 versus #10 sheathing screws, and OSB panel thickness 11.1
versus 18.2 mm with tongue and groove (T&G) edges, respectively
(Table 1). Neither of the diaphragms included edge blocking, that
is, CFS framing under all of the OSB panel edges. Following the
CFS-NEES building design, the first two diaphragm test configura-
tions incorporated these construction details (Fig. 5). Subsequently,
a construction parameter was altered in each configuration: (1) full
panel edge blocking was added to the roof configuration (full height
blocking with joist sections, as shown in Fig. 6), where the full
Downloaded from ascelibrary.org by University Of Sydney on 03/23/23. Copyright ASCE. For personal use only; all rights reserved.

Fig. 3. Cold-formed steel diaphragm test specimen and set-up

perimeter of each OSB panel was fastened to the underlying steel
framing, and (2) a larger sheathing screw size (#12) was used in the
floor configuration. The objective was to investigate the effect of
to one side of each diaphragm specimen. The support of the dis- these two parameters on the shear strength and stiffness of the dia-
tribution beam comprised a roller system at three locations, which phragm. Monotonic and reversed cyclic loading was used for each
allowed it to move freely. Thus, in combination with the hinged of the four configurations. The bare CFS framing without the
actuator, the diaphragm specimen could also move (lengthen sheathing was also tested under monotonic loading in order for

Fig. 4. Diaphragm test specimens (images by Colin A. Rogers): (a) unblocked framing prior to installation of OSB sheathing; (b) completed
diaphragm with roof sheathing

Table 1. Basic Floor and Roof Diaphragm Configurations

Component Section (mm) Length (mm)
Roof diaphragm
Joists 305S51-137M 3,505
Rim joists 305T51-173M 6,480
Web stiffeners L 38 × 38 × 1.37 250
Joist bracing 305S41-137M 560
Joist bracing connectors L 38 × 102 × 1.37 250
Straps 38 × 1.37 6,300
#8 sheathing self-drilling (152=305 mm spacing) — 50
#10 steel-to-steel flat head self-drilling — 20
#10 steel-to-steel hex head cap self-drilling — 25
OSB panels (24=16 rated) 2,400 × 1,200 × 11 —
Floor diaphragm
Joists 350S64-246M 3,505
Rim joists 350T64-246M 6,480
Web stiffeners L 38 × 38 × 1.37 280
Joist bracing 305S51-137M 550
Joist bracing connectors L 38 × 102 × 1.37 250
Straps 38 × 1.37 6,300
#10 sheathing self-drilling (152=305 mm spacing) — 44
#10 steel-to-steel flat head self-drilling — 20
#10 steel-to-steel hex head cap self-drilling — 25
OSB panels (48=24 rated T&G) 2,400 × 1,200 × 18 —

© ASCE 04017215-3 J. Struct. Eng.

J. Struct. Eng., 2018, 144(2): 04017215

Downloaded from ascelibrary.org by University Of Sydney on 03/23/23. Copyright ASCE. For personal use only; all rights reserved.

Fig. 5. Illustration of (a) CFS framing; (b) wood panel sheathing

connect the CFS framing, whereas Table 2 includes the nomencla-

ture followed for the specimens.
The Consortium of Universities for Research in Earthquake
Engineering (CUREE) displacement-controlled loading protocol
for ordinary ground motions (Krawinkler et al. 2000), which rep-
resents an earthquake excitation with a probability of exceedance of
10% in 50 years, was selected for the reversed cyclic tests (i.e., Fig. 8
depicts the loading protocol for the roof blocked Specimen 8).
A specific loading protocol for CFS framed diaphragms was not
available; because the CUREE protocol had been extensively used
for the testing of the CFS framed shear walls, and was relied on in
the development of the AISI S400 standard, it was decided to also
use it for this study. The effect of cumulative damage is taken into
Fig. 6. Modification to the roof diaphragm configurations; full CFS account with the repetition of multiple small deformation amplitude
frame blocking loading cycles followed by larger deformation amplitudes. The
protocol is based on a postpeak reference displacement obtained
from the monotonic test at 80% of the ultimate load. A displace-
ment rate of 2.5 mm=min for the roof and 5 mm=min for the floor
its contribution to be accounted for separately. In total, the four di- configuration was applied during the monotonic loading, whereas
aphragm configurations, each tested with two loading protocols, the cyclic loading followed a displacement rate that started with
and the two bare frame tests resulted in a laboratory program com- 15 mm=min and increased to 60 mm=min after 60 mm of displace-
prising 10 tests. ment for both the roof and floor configurations.
The material used for the fabrication of the joists and tracks was Regarding the instrumentation used, lateral displacement and
ASTM A653 (ASTM 2015) Grade 50 (i.e., nominal yield stress shear deformation as well as local in-plane displacement were
Fy ¼ 345 MPa) steel. Moreover, Fig. 5 demonstrates the following captured using four string potentiometers (254 and 508 mm total
two features of the diaphragm specimens: a double CFS joist sec- stroke) and 12 LVDTs (15 mm stroke), as shown in Fig. 9. The
tion as a chord element to represent the presence of a wall in actual diaphragm response was also captured by the internal LVDT
conditions (increased stiffness), and a 152.4-mm sheathing exten- and load cell of the actuator. Vishay Model 5100B (Vishay Pre-
sion at the fixed connection location, as per the CFS-NEES build- cision Group, Wendell, North Carolina) scanners and the Vishay
ing design for ledger framing. This led to an out-to-out width of the System 5000 StrainSmart software were used to record the mea-
CFS frame of 3,505 mm. Fig. 7 illustrates the connections used to sured data.

Fig. 7. Cold-formed steel framing connections (images by Colin A. Rogers): (a) joist-to-track connections; (b) blocking-to-joist connections

© ASCE 04017215-4 J. Struct. Eng.

J. Struct. Eng., 2018, 144(2): 04017215

 Table 2. Specimen Nomenclature (50 mm gauge length) were extracted from the CFS sections con-
Specimen Description sidering the different steel thicknesses (roof rim joist, roof joist,
floor rim joist, floor joist). An average value was obtained for each
1-RF-M Roof bare steel frame monotonic
tensile material property from three coupons for each case. Strain
2-FF-M Floor bare steel frame monotonic
3-RU-M Roof unblocked monotonic
gauges and an extensometer were used to measure Young’s
4-RU-C Roof unblocked cyclic modulus and elongation values. The nominal yield stress and
5-F#10-M Floor #10 screws monotonic tensile strength of the ASTM A653 Grade 50 steel was 345 and
6-F#10-C Floor #10 screws cyclic 450 MPa, respectively. Table 3 summarizes the results from the
7-RB-M Roof blocked monotonic tensile coupon tests. Sharp yielding behavior was observed for
8-RB-C Roof blocked cyclic all the coupon specimens with increased yield stress values ex-
9-F#12-M Floor #12 screws monotonic pected owing to the fabrication process of cold-formed steel (cold
10-F#12-C Floor #12 screws cyclic work of forming). For the moisture content measurements, samples
from the OSB panels were placed for 24 h in a constant oven tem-
perature of 103°C in order for the oven-dry mass to be obtained
Downloaded from ascelibrary.org by University Of Sydney on 03/23/23. Copyright ASCE. For personal use only; all rights reserved.

(ASTM 2015, Method B). Three round specimens per panel

(76.2 mm in diameter) were extracted from selected panels immedi-
ately after testing and their weight was measured. Low moisture
content in the range of 4 to 5% was obtained, as expected, owing
to the fabrication process of the OSB panels.

Diaphragm Test Results

The hysteretic and monotonic shear force versus deformation re-

sponse was obtained for all diaphragm configurations, starting with
the monotonic testing of the bare CFS frame without the sheathing.
A maximum displacement of 45 mm was targeted for the bare CFS
framing loading to ensure that the specimen would remain in the
elastic range. These tests revealed that the shear strength and stiff-
ness contribution of the bare CFS frame is negligible, as indicated
in Fig. 10. Fig. 11 shows the typical overall shear deformations
of a wood-sheathed/ CFS framed diaphragm; in this case, Test
10-F #12-C. Subsequently, the monotonic results for Specimens
3-RU-M, 7-RB-M, 5-F#10-M, and 9-F#12-M are presented in
Fig. 8. Consortium of Universities for Research in Earthquake Engi- Figs. 12(a) and 13(b) in the form of a comparison between shear
neering (CUREE) loading protocol for Specimen 8-RB-C (blocked) force versus rotation response curves. Fig. 12(b) includes the
blocked versus unblocked roof diaphragm configuration reversed
cyclic results (8-RB-C versus 4-RU-C), whereas Fig. 13(b) con-
tains a comparison of the floor with #12 sheathing screws versus
Material Properties the floor with #10 screws (10-F#12-C versus 6-F#10-C), respec-
tively. Referring to Figs. 12(b) and 13(b), the equivalent monotonic
Tensile coupon tests and moisture content measurements were curve is superimposed for Specimens 7 and 9, respectively. It is
conducted for the steel and wood material used in the experiments, shown that there is no difference between the diaphragm’s mono-
respectively. The tensile coupon tests were based on the ASTM tonic and the cyclic response up to the ultimate shear strength level;
A370 standard (ASTM 2016), whereas the secondary oven- as expected, the postpeak cyclic curve deteriorates more quickly
drying method of the ASTM D4442 standard Method B (ASTM owing to the cumulative damage of the repeated displacement
2015) was used for the moisture content measurements. Coupons cycles. This cumulative damage also results in the lower resistance

Fig. 9. Instrumentation of diaphragm test specimens

© ASCE 04017215-5 J. Struct. Eng.

J. Struct. Eng., 2018, 144(2): 04017215

 Table 3. Tensile Properties of Steel
Specimens E (MPa) Fy (MPa) εy (mm=mm) Fu (MPa) εu (mm=mm) Fu =Fy Elongation (%) Number
RJ—roof 188,595 387 0.0040 466 0.1717 1.20 27.5 3
RJ—floor 224,149 398 0.0028 474 0.1822 1.19 31.8 3
J—roof 189,049 391 0.0037 471 0.1959 1.20 28.7 3
J—floor 210,854 394 0.0036 462 0.1695 1.17 29.3 1
J—roof B 200,568 385 0.0015 466 0.0673 1.21 14.8 3
J—floor #12 202,097 410 0.0018 477 0.0858 1.16 14.6 3
Note: B = blocked; J = joist; RJ = rim joist; #12 = size #12 sheathing screws.
Downloaded from ascelibrary.org by University Of Sydney on 03/23/23. Copyright ASCE. For personal use only; all rights reserved.

Fig. 10. Shear force versus rotation response for the bare CFS frame: (a) 1-RF-M; (b) 2-FF-M

attained for the negative displacement cycles. The damage to the Roof Configuration Test Results
specimens, a result of the in-plane shear loading, is illustrated in
The failure modes observed during the testing of Specimens 3-RU-M
Figs. 14–16. Table 4 summarizes the corresponding data for all and 4-RU-C were the screws tearing out or pulling through the wood
the tests. Details of the behavior exhibited by the specimens are after wood bearing had occurred [Fig. 14(a)]. Tilting of the screws
provided in the following paragraphs. It should also be mentioned was present as a desirable ductile deformation mode. Damage con-
that the displacement shown in the graphs was obtained from the centrated mostly in the middle row of the panels, where fewer
string potentiometer (N-S SSP , Fig. 9) recording the displacement screws were used (unblocked diaphragm, 304 mm screw spacing).
of the specimen in the north-south direction. The rotation was ob- Toward the end of the test, lift-off of the OSB panels was triggered
tained by dividing this displacement with the end members’ length, in the intermediate panel locations along their edges where the
3,505 mm. sheathing was no longer attached to the framing, as illustrated in
Fig. 14(b).
Adding panel edge blocking to the roof diaphragm configura-
tion (Specimens 7-RB-M and 8-RB-C) had a profound effect on the
diaphragm response (Fig. 12). This configuration yielded a 130%
increase in maximum shear strength and a 70% increase in shear
stiffness compared with the unblocked case (Table 4). The blocked
roof diaphragm configuration exhibited the highest shear strength
and stiffness overall in this experimental program even though the
OSB was thinner and the sheathing screws smaller than for the floor
configuration. The benefit of attaching the full perimeter of each
OSB panel to the underlying CFS framing was demonstrated. In
this case, similar failure modes to the unblocked case [Fig. 15(a)]
were observed (tear out and pull through) accompanied by sheared
fasteners mostly in areas where the fasteners penetrated two layers
of steel (joist-to-rim joist connection locations). After the peak load
was reached, the damage concentrated in the sheathing screw
connections along the fixed edge of the test set-up [Fig. 15(b)].
Owing to the 152-mm extension of the OSB in that location, as
Fig. 11. Example overall shear deformations of typical diaphragm test
explained earlier, a shorter width panel was connected to the steel
specimen (image by Colin A. Rogers)
framing; thus, fewer screws were used, which potentially led to the

© ASCE 04017215-6 J. Struct. Eng.

J. Struct. Eng., 2018, 144(2): 04017215

Downloaded from ascelibrary.org by University Of Sydney on 03/23/23. Copyright ASCE. For personal use only; all rights reserved.

Fig. 12. Force versus deformation response for roof specimens: (a) 3-RU-M (unblocked) and 7-RB-M (blocked); (b) 4-RU-C (unblocked) and
8-RB-C, 7-RB-M (blocked)

Fig. 13. Force versus deformation response for floor specimens: (a) 5-F#10-M and 9-F#12-M; (b) 6-F#10-C and 10-F#12-C, 9-F#12-M

Fig. 14. Deformation for the roof unblocked diaphragm configurations 3-RU-M and 4-RU-C (images by Colin A. Rogers): (a) screw edge tear out;
(b) lift-off of OSB panels

© ASCE 04017215-7 J. Struct. Eng.

J. Struct. Eng., 2018, 144(2): 04017215

Downloaded from ascelibrary.org by University Of Sydney on 03/23/23. Copyright ASCE. For personal use only; all rights reserved.

Fig. 15. Deformation for blocked roof diaphragm configuration Specimens 7-RB-M and 8-RB-C (images by Colin A. Rogers): (a) screw edge tear
out; (b) postultimate bending action of steel framing (OSB panels removed for posttest photograph)

Fig. 16. Deformation for floor diaphragm configurations with #10 screws, 5-F#10-M and 6-F#10-C (images by Colin A. Rogers): (a) screw edge
shear failure; (b) relative displacement between panels; (c) panel edge contact effect

Table 4. General Results from the Monotonic (M) and Reversed Cyclic (C) Tests
Specimens Su (kN=m) Δnet;o:4u (mm) Δnet;u (mm) θnet;u (rad × 10−3 ) Rigidity, K (kN=mm)
3-RU-M 5.6 9 41.5 11.8 1.53
5-F#10-M 7.9 6.1 30 8.6 3.15
7-RB-M 13 12 62.1 17.7 2.64
9-F#12-M 11.8 9.4 60.7 17.3 3.07
4-RU-C 5.5= − 5.1 7.5= − 6.9 41.2= − 30.6 11.8= − 8.7 1.79=1.81
6-F#10-C 7.6= − 7.1 5.8= − 6.7 30.8= − 23.4 8.8= − 6.7 3.18=2.57
8-RB-C 13= − 10.7 13.1= − 11.1 65.5= − 45 18.7 − 12.8 2.48=2.34
10-F#12-C 11.8= − 11 8.8= − 9.1 57.1= − 40.7 16.3= − 11.6 3.29=2.93

concentration of sheathing connection failures. Ultimately, the framing in that location because there was no more diaphragm ac-
sheathing connections in these edge panels failed, resulting in a tion, Fig. 15(b)]. This bending action of the steel framing is the cause
transfer of force through the underlying steel framing by means for the constant level of the shear force after approximately 25 mrad
of bending action [cantilever moment frame action of the steel rotation indicated in the response curves in Figs. 12(a and b).

© ASCE 04017215-8 J. Struct. Eng.

J. Struct. Eng., 2018, 144(2): 04017215

 Floor Configuration Test Results wall response [AISI S400 (AISI 2015)]: (1) linear elastic bending
(1st term); (2) linear elastic shear deformation (2nd term); (3) non-
During the testing of Specimens 5-F#10-M and 6-F#10-C, a steeper linear empirical component (3rd term); and (4) overturning
decline of the shear strength versus deformation curve (Fig. 13) was anchorage/chord splice deformation.
observed compared with Specimen 3-RU-M and 4-RU-C, attrib-
uted to the fact that the #10 sheathing screws were primarily failing
2
2vh3 ω ω vh 5 v h
in shear or remaining vertical while the wood sheathing was tearing δ¼ þ 1 2 þ ω41 ω2 ω3 ω4 þ δv ð1Þ
3Es Ac b ρGtsheathing β b
out. This sheathing screw behavior suggested that the #10 screws
[5-F#10-M and 6-F#10-C, Fig. 16(a)] used thus far for this type of

2 Pn
floor configuration were not appropriate based on the sheathing and 0.052vL3 ω1 ω2 vL 5 v j¼1 Δci X i
steel thickness if a more ductile failure mode were desired. More- δ¼ þ þ ω1 ω2 ðaÞ
4
þ
Es Ac b ρGtsheathing 2β 2b
over, at approximately 35 mrad [Figs. 13(a and b)], most of the
screws in the interface of the panel rows and field had failed leading ð2Þ
to the CFS framing underneath taking most of the load. The load
Downloaded from ascelibrary.org by University Of Sydney on 03/23/23. Copyright ASCE. For personal use only; all rights reserved.

increasing and then stabilizing during these final excursions Eq. (1): The shear wall deflection equation refers to the deflec-
showed that contact/bearing action along the edges of the inter- tion of a blocked CFS framed/wood-sheathed shear wall. The δ v
mediate panels provided additional resistance, taking also into ac- variable referring to the anchorage deformation was obtained using
count the T&G characteristic of the OSB panels, which prevented the data of the string potentiometer in the E-W direction (E-W SSP,
lift-off of panels even though the panel edges were not blocked. A Fig. 9), which provided the chord member deformation, because no
finite element model of the floor diaphragm specimen (5-F#10-M anchorage details were included in the diaphragm specimens.
and 6-F#10-C) described in Chatterjee (2016) revealed that the Eq. (2): The diaphragm deflection equation refers to the deflec-
level of static friction force developed during testing in the inter- tion of a blocked CFS framed/wood-sheathed simply supported
mediate panel locations was 0.003 kN=mm, which provides a mini- diaphragm. As such, the total shear load applied was assumed
mum level of contact force being present of 0.0075 kN=mm, to be 2V and the total length of the diaphragm L ¼ 2 × 3505 ¼
assuming an average coefficient of friction for a wood-to-wood sur- 7,010 mm, because the deflection obtained for a cantilever under
face of 0.4 (Giancoli 2009). Further, the T&G panels facilitated the point load P is equal to the one obtained by a simply supported
construction process (walking on top of the diaphragm) and, thus, beam at midspan with double the cantilever length and under a load
would be a useful improvement for the design of the roof diaphragm. 2P. The Δci variable referring to the chord splice deformation was
The floor configuration, composed of greater thickness steel and obtained using the data of the string potentiometer in the E-W
sheathing, was expected to return higher shear strength and stiffness direction (E-W SSP, Fig. 8), as for Eq. (1). The splice was assumed
values compared with the roof configuration, as presented in Table 4. to be in the middle of the chord; thus, X i ¼ 3,505 mm. It should be
Fig. 16 illustrates the failure mode and panel edge contact effect noted that the shear modulus values used in the calculations
for the 5-F#10-M and 6-F#10-C specimens described herein. (i.e., G ¼ 1,317 N=mm2 for the roof specimens) were obtained
The larger screw size (#12 versus #10) for the floor configura- from TECO’s document “Design capacities for oriented strand
tion (9-F#12-M and 10-F#12-C) resulted in an overall increase of board” (TECO 2008). Further, an amplification factor of 2.5 is
50% in shear strength (Table 4). Screw tilting was present before suggested for the diaphragm deflection equation [Eq. C-F2.4.3-1
shearing or pulling out of the steel owing to shear and tensile forces AISI S400 (AISI 2015)] when the diaphragm is unblocked. Such
developing between the CFS framing and OSB panels. Several joist factor does not exist for the shear wall deflection [Eq. E1.4.1.4-1
flanges were distorted owing to these applied uplift forces of the AISI S400 (AISI 2015)] equation because a shear wall is always
panels. Although there was an evident increase in strength due blocked. However, because both equations yield similar results and
to the #12 sheathing screws, based on Fig. 13, the overall force refer to a diaphragm in this paper, the 2.5 factor is applied to both.
versus deformation response was similar in shape for the two dia- Table 5 provides the results for Eqs. (1) and (2) compared with
phragms, but the observed response of the sheathing connection the observed values from testing corresponding to the design level
seemed to be more ductile because shear fracture of the screws of 40 and 60% of the shear strength. It is shown that Eqs. (1) and (2)
did not take place in the 9-F#12-M and C specimens. provide similar results and that in almost all the cases the error
between calculated and observed data is close to 20% or lower.
Further, looking at the error percentages of Table 5 and the force
Diaphragm Design Predictions versus deformation curves of Figs. 12 and 13, it can be observed
that the error is reduced when the level of force considered for
The AISI S400 standard (AISI 2015) provides a diaphragm deflec- calculation corresponds to the near linear part of the curve, which
tion equation for simply supported span lengths (Eq. C-F2.4.3-1) indicates that Eqs. (1) and (2) can confidently be used to calculate
and a shear wall deflection equation (Eq. E1.4.1.4-1). Given the deflection at the design shear strength level, but may not produce as
cantilever approach used in the test, it was deemed appropriate accurate results for the peak shear strength level. Included in Table 5
for the shear wall deflection equation to be used in order to acquire is the relative error of calculated displacements with respect to mea-
design deflection values for the diaphragm configurations. Ulti- surements. It should be noted that a different process was followed
mately, it was revealed that both equations provide similar results, compared with the one presented in Nikolaidou et al. (2017), in
given the appropriate assumptions, and are presented in this paper which the focus was to compare the deflection design values at
[Eq. (1) for cantilever shear wall and Eq. (2) for simply supported the ultimate shear strength level with an equivalent elastic displace-
diaphragm in this paper, respectively]. Design deflection values ment, δ elastic , provided by the experimental data at ultimate assum-
were acquired for the design shear strength level following both ing elastic response of the diaphragm. This effort led to this updated
the Canadian and U.S. code; a resistance factor ϕ of 0.6 for load process where only the design-level shear strength was considered
and resistance factor design (60% of strength) and a safety factor Ω and appropriate assumptions were made for both deflection equa-
of 2.5 for allowable strength design (40% of strength) was consid- tions leading to more reasonable results.
ered, respectively [AISI S400 standard (AISI 2015)]. Eqs. (1) and Table 6 lists the nominal shear resistance values, VAISI , as ob-
(2) translate into the following components of the diaphragm/shear tained from Table F2.4-1 of the AISI S400 standard (AISI 2015) to

© ASCE 04017215-9 J. Struct. Eng.

J. Struct. Eng., 2018, 144(2): 04017215

 Table 5. Design Deflection Values Using Eqs. (1) and (2)
Deflection 3-RU-M and 4-RU-C 5-F#10-M and 6-F#10-C 7-RB-M and 8-RB-C 9-F#12-M and 10-F#12-C
At 40% strength
δ Observed (mm) 8.18 5.61 10.6 7.82
δ Calculated (mm), Eq. (1) 8.46 4.68 8.67 9.59
% error 3.4 16.6 18.2 22.7
δ Calculated (mm), Eq. (2) 8.59 4.52 8.06 8.83
% error 5.1 19.4 23.9 12.9
At 60% strength
δ Observed (mm) 13.52 9.61 17.4 15.32
δ Calculated (mm), Eq. (1) 13.72 10.54 14.4 17.32
% error 1.5 9.7 17.2 13
δ Calculated (mm), Eq. (2) 13.48 9.81 12.53 15.03
% error 0.3 2.1 28 1.9
Downloaded from ascelibrary.org by University Of Sydney on 03/23/23. Copyright ASCE. For personal use only; all rights reserved.

Table 6. Nominal Shear Resistance Values Using Table F2.4-1 of AISI S400 (AISI 2015)
Shear resistance 3-RU-M and 4-RU-C 5-F#10-M and 6-F#10-C 7-RB-M and 8-RB-C 9-F#12-M and 10-F#12-C
V AISI (kN=m) 7.37 8.10 11.10 8.10
V TEST (kN=m) 5.6 7.9 13 11.8

be used in design, and the measured shear resistance values, VTEST , equations [Eqs. (1) and (2)] of the AISI S400 standard (AISI
provided from the tests for each diaphragm configuration presented 2015) were in close proximity with the experimental values
herein. Table F2.4-1 refers only to plywood sheathing and does not for the design level shear strength of the specimens. However,
account for the effect of the sheathing screw size; thus, meaningful regarding design shear strength values, the AISI S400 standard
design predictions cannot be made for the specific tested diaphragm (AISI 2015) at present does not include values for the case of
specimens. Nonetheless, these are the only design shear strength OSB panels, and the size of the screws is not considered as an
values available at present in the AISI S400 standard (AISI 2015) influential parameter in the design shear strength calculations.
for the tests included in this paper. As such, relevant design shear strength values could not be
obtained.
Additional experimental and numerical work is required in
Conclusions order for complete information about the CFS framed diaphragm
response to be available to professional engineers. Studies should
A total of 10 CFS framed/OSB-sheathed diaphragm tests were focus on varying parameters, such as screw spacing, load direction,
completed in the experimental program described in this paper. panel blocking type and panel type, as well as implementing
The research focused on four main diaphragm configurations for nonstructural components, such as gypsum panels.
which various parameters were altered, such as the steel section
and the OSB thickness, the screw size, and the use of panel edge
blocking. The objective was to characterize the in-plane force ver- Acknowledgments
sus deformation response of the CFS framed/wood-sheathed dia-
phragm under monotonic and reversed cyclic loading. The main The authors would like to thank the American Iron and Steel
findings are summarized as follows: Institute (AISI) for financially supporting this research project.
• Panel edge blocking substantially increases the diaphragm shear Additional support was obtained from the Canadian Sheet Steel
strength and stiffness, with values of 130 and 70% obtained, Building Institute (CSSBI) and the Natural Sciences and Engineer-
respectively, for the roof configuration. ing Research Council of Canada (NSERC). A special thank you is
• Changing the sheathing screw size from #10 to #12 does not also extended to Bailey Metal Products Ltd., Simpson Strong-Tie
have a measurable effect on the shape of the overall diaphragm Co., Inc., Ontario Tools and Fasteners Ltd, ArcelorMittal, and Con-
load versus displacement response despite the fact that it leads to structions Proco, Inc. for the materials and tools that were provided.
a somewhat more ductile sheathing-to-framing screw connec-
tion behavior. It does cause, however, a considerable increase
in the diaphragm shear strength (50%). Notation
• As tested, the CFS floor and roof framing without the sheathing
The following symbols are used in this paper:
does not contribute to the shear strength and stiffness of the
Ac = gross cross-sectional area of chord member (mm2 );
diaphragm.
• Tongue and groove sheathing panels improve both the construc- b = width of the shear wall/diaphragm (parallel to loading)
tion process and the performance of the diaphragm. As such, (mm);
their further implementation also for roof diaphragms should Es = modulus of elasticity of steel 203,000 MPa;
be considered. G = shear modulus of sheathing material (MPa);
• In an effort to obtain design shear and deflection values, the h = wall height (mm);
AISI S400 standard (2015) was used. The design deflection va- K = rigidity of diaphragm specimen calculated at 40% shear
lues calculated using the shear wall and diaphragm deflection strength (kN/mm);

© ASCE 04017215-10 J. Struct. Eng.

J. Struct. Eng., 2018, 144(2): 04017215

 L = diaphragm length perpendicular to direction of load CSA (Canadian Standards Association). (2016). “North American specifi-
(mm); cation for the design of cold-formed steel structural members.” CSA
S136, Rexdale, Canada.
n = number of chord splices in diaphragm (considering both
Dolan, J. D., and Easterling, W. S. (2000). “Monotonic and cyclic tests
diaphragm chords); of light-frame shear walls with various aspect ratios and tie-down
Su = shear strength of diaphragm specimen (kN/m); restraints.” Rep. No. TE-2000-001, Virginia Polytechnic Institute and
s = maximum fastener spacing at panel edges (mm); State Univ., Blacksburg, VA.
tsheathing = nominal panel thickness (mm); Giancoli, D. (2009). Physics for scientists and engineers with modern
tstud = nominal framing thickness (mm); physics, 4th Ed., Pearson Education, London.
Krawinkler, H., Parisi, F., Ibarra, L., Ayoub, A., and Medina, R. (2000).
V = total in-plane load applied to the diaphragm (N); “Development of a testing protocol for wood frame structures.” Rep.
v = shear demand (V/b) (N/mm); W-02 Covering Task 1.3.2, Consortium of Universities for Research
X i = distance between the ith chord splice and the nearest in Earthquake Engineering, Richmond, CA.
support (mm); LGSEA (Light Gauge Steel Engineers Association). (1998). “Lateral load
α = 1 for a uniformly fastened diaphragm; resisting elements: Diaphragm design values.” Tech Note 558b-1,
Downloaded from ascelibrary.org by University Of Sydney on 03/23/23. Copyright ASCE. For personal use only; all rights reserved.

Washington, DC.
β = 2.35 for plywood and 1.91 for OSB for SI units
Liu, P., Peterman, K. D., Yu, C., and Schafer, B. W. (2012). “Cold-formed
(N=mm1.5 );
steel shear walls in ledger-framed buildings.” Annual Stability Conf.,
Δci = deformation value associated with ith chord splice (mm); Structural Stability Research Council, Curran Associates, Inc.,
Δnet;o4u = displacement value of diaphragm specimen at 40% shear Dutchess, New York.
strength (mm); NAHB Research Center. (1999). “Innovative residential floor construction:
Δnet;u = displacement value of diaphragm specimen at ultimate Horizontal diaphragm values for cold-formed steel framing.” U.S. Dept.
shear strength (mm); of Housing and Urban Development, Upper Marlboro, MD.
Nikolaidou, V., Latreille, P., Rogers, C. A., and Lignos, D. G. (2015).
δ = calculated in-plane deflection (mm);
“Characterization of CFS framed diaphragm behavior.” Rep. CM-432,
δ v = vertical deformation of anchorage/attachment details American Iron and Steel Institute, Washington, DC.
(mm); Nikolaidou, V., Latreille, P., Rogers, C. A., and Lignos, D. G. (2017).
θnet;u = rotation of diaphragm specimen at ultimate strength, “Characterization of cold-formed steel framed/wood-sheathed floor
Δnet;u =3,505 mm (rad × 10−3 ); and roof diaphragm structures.” Proc., 16th World Conf. on Earthquake
ρ = 1.85 for plywood and 1.05 for OSB, term for different Engineering, Chilean Association of Seismology and Earthquake
sheathing material type; Engineering, Santiago Chile, Santiago, Chile.
NIST, Madsen, R. L., Castle, T. A., and Schafer, B. W. (2016). “Seismic
ω1 = s=152.4 (for s in mm); design of cold-formed steel lateral load resisting systems: A guide for
ω2 = 0.838=tstud (for tstud in mm);
p practicing engineers.” GCR 16-917-38, Gaithersburg, MD.
ω3 = ½ðh=bÞ=2; and Pan, C. L., and Shan, M. Y. (2011). “Monotonic shear tests of cold-formed
ω4 = 1 for wood with structural panels. steel wall frames with sheathing.” Thin-Walled Struct., 49(2), 363–370.
Peterman, K. D. (2014). “Behavior of full-scale cold-formed steel buildings
under seismic excitations.” Ph.D. thesis, Johns Hopkins Univ.,
References Baltimore.
Peterman, K. D., Stehman, M. J., Madsen, R. L., Buonopane, S. G., Nakata,
AISI (American Iron and Steel Institute). (2013). “Test standard for N., and Schafer, B. W. (2016a). “Experimental seismic response of a
cantilever test method for cold-formed steel diaphragms.” AISI S907, full-scale cold-formed steel-framed building. I: System-level response.”
Washington, DC. J. Struct. Eng., 10.1061/(ASCE)ST.1943-541X.0001577, 04016127.
AISI (American Iron and Steel Institute). (2015). “North American Peterman, K. D., Stehman, M. J., Madsen, R. L., Buonopane, S. G., Nakata,
standard for seismic design of cold-formed steel structural systems.” N., and Schafer, B. W. (2016b). “Experimental seismic response of a
AISI S400, Washington, DC. full-scale cold-formed steel-framed building. II: Subsystem-level
AISI (American Iron and Steel Institute). (2016). “North American response.” J. Struct. Eng., 10.1061/(ASCE)ST.1943-541X.0001578,
specification for the design of cold-formed steel structural members.” 04016128.
AISI S100, Washington, DC. Serrette, R. L., Morgan, K. A., and Sorhouet, M. A. (2002). “Performance
APA (The Engineered Wood Association). (2007). “Diaphragms and shear of cold-formed steel-framed shear walls: Alternative configurations.”
walls.” Design/Construction Guide Form No. L350A, Tacoma, WA. Rep. No. LGSRG-06-02, Dept. of Civil Engineering, Santa Clara, CA.
ASTM. (2015a). “Standard specification for steel sheet, zinc-coated Serrette, R. L., and Chau, K. (2003). “Estimating the response of
(galvanized) or zinc-iron alloy-coated (galvannealed) by the hot-dip cold-formed steel-frame shear walls.” Dept. of Civil Engineering,
process.” ASTM A653, West Conshohocken, PA. Santa Clara, CA.
ASTM. (2015b). “Standard test methods for direct moisture content Shamim, I., DaBreo, J., and Rogers, C. A. (2013). “Dynamic testing of
measurement of wood and wood base materials.” ASTM D4442, West single- and double-story steel-sheathed cold-formed steel-framed
Conshohocken, PA. shear walls.” J. Struct. Eng., 10.1061/(ASCE)ST.1943-541X.0000594,
ASTM. (2016). “Standard test methods and definitions for mechanical 807–817.
testing of steel products.” ASTM A370, West Conshohocken, PA. StrainSmart version 4.7 [Computer software]. Vishay Precision Group,
Branston, A. E., Boudreault, F. A., Chen, C. Y., and Rogers, C. A. (2006). Inc., Wendell, NC.
“Light-gauge steel-frame wood structural panel shear wall design TECO (Timberco, Inc.). (2008). “TECHTIP. Design capacities for oriented
method.” Can. J. Civ. Eng., 33(7), 872–889. strand board.” 〈www.pfsteco.com/techtips/pdf/tt_osbdesigncapacities〉
Chatterjee, A. (2016). “Structural system reliability with application to light (Jun. 27, 2017).
steel-framed buildings.” Ph.D. thesis, Virginia Polytechnic Institute and Tissell, J. R., and Elliot, J. R. (2004). “Plywood diaphragms.” Research
State Univ., Blacksburg, VA. Rep. 138, American Plywood Association, Tacoma, WA.

© ASCE 04017215-11 J. Struct. Eng.

J. Struct. Eng., 2018, 144(2): 04017215