January 2021

ERRATA
to the 2018 and Prior Editions of
the National Design Specification® (NDS®) for Wood Construction

Page Revision

91 Revise footnote 1 in Table 12.5.1D as follows:

1. The ℓ/D ratio used to determine the minimum edge distance spacing
between rows shall be the lesser of:
(a) length of fastener in wood main member/D = ℓm/D
(b) total length of fastener in wood side member(s)/D = ℓs /D

222 Catoctin Circle, SE, Suite 201 ▪ Leesburg, VA 20175 ▪ 202-463-2766 ▪ www.awc.org ▪ [email protected]
 July 2020

ERRATA
to the 2018 and Prior Editions of
the National Design Specification® (NDS®) for Wood Construction

Page Revision
166 Clarifies that the following calculations in Example E.7 Sample Solution of Row of Bolts
is intended for a single-row bolted connection with a 3-1/2 thick main member and 1-
1/2 thick side member:

E.7 Sample Solution of Row of Bolts

Calculate the net section area tension and row ZNT=Ftt [w – nrow Dh]
tear-out adjusted ASD design capacities for the ZNT= (788 psi)(1.5)[3.5 – 1(0.5625)] = 3,470 lbs
single-shear single-row bolted connection
represented in Figure E2.

Main and Side Members:

#2 grade Hem-Fir 2x4 lumber. See NDS
Supplement Table 4A – Visually Graded
Dimension Lumber for reference design values.
Adjustment factors CD, CT, CM, and Ci are assumed
to equal 1.0 in this example for calculation of
adjusted design values.
Ft = 525 psi (CF) = 525(1.5) = 788 psi
Fv = 150 psi

Connection Details:
Adjusted For side member, adjusted ASD Row Tear-Out
Bolt diameter, D: 1/2 in.
Bolt hole diameter, Dh: 0.5625 in. Capacity, ZRT:
Adjusted ASD bolt design value, Z||: 550 lbs ZRTi=niFvtscritical
(See NDS Table 12A for 3-1/2main member ZRT1= 3(150 psi)(1.5)(2) = 1,350 lbs
thickness and 1-1/2 side member thickness. For
this trial design, the group action factor, Cg, is In this sample calculation, the adjusted ASD
taken as 1.0). connection capacity is limited to 1,350 pounds by row
tear-out, ZRT.
Adjusted ASD Connection Capacity, n Z||:
nZ||= (3 bolts)(550 lbs) = 1,650 lbs

Adjusted For side member, adjusted ASD Net

Section Area Tension Capacity, ZNT:

222 Catoctin Circle, SE, Suite 201 ▪ Leesburg, VA 20175 ▪ 202-463-2766 ▪ www.awc.org ▪ [email protected]
 March 2019

ERRATA
to the 2015 Edition of
the National Design Specification® (NDS®) for Wood Construction
(all versions)

Page Revision
165 Revise the following calculations in Example E.8 Sample Solution of Row of Split Rings
(remainder of example is unchanged):

E.8 Sample Solution of Row of Split Rings

Calculate the net section area tension and row Adjusted ASD Connection Capacity, nP:
tear-out adjusted ASD design capacities for the
nP = (2 split rings)(2,730 lbs) = 5,460 lbs
single-shear single- row split ring connection
represented in Figure E3.
Adjusted ASD Net Section Area Tension Capacity,
Main and Side Members: ZNT:
#2 grade Southern Pine 2x4 lumber. See NDS
Supplement Table 4B – Visually Graded Southern ZNT  Ft Anet
Pine Dimension Lumber for reference design
values. Adjustment factors CD, CT, CM, and Ci are ZNT = Ft [A2x4 – Abolt-hole – Asplit ring projected area]
assumed to equal 1.0 in this example for ZNT = (825 675 psi)[5.25 in.2 – 1.5" (0.5625") – 1.1 in.2]
calculation of adjusted design values. = 2,728 2,232 lbs
Ft = 825 675 psi
Fv = 175 psi
Adjusted ASD Row Tear-Out Capacity, ZRT:

Main member thickness, tm: 1.5 in. F A

Side member thickness, ts: 1.5 in. ZRTi  ni v critical
2
Main and side member width, w: 3.5 in.
ZRT1 = [(2 connectors)(175 psi)/2](21.735 in.2)
Connection Details: = 3,804 lbs
Split ring diameter, D: 2.5 in. (see Appendix K for
connector dimensions) where:
Adjusted ASD split ring design value, P: 2,730 lbs Acritical = 21.735 in.2 (See Figures E4 and E5)
(see Table 13.2A. For this trial design, the group
action factor, Cg, is taken as 1.0). In this sample calculation, the adjusted ASD
connection capacity is limited to 2,728 2,232 pounds
by net section area tension capacity, ZNT

222 Catoctin Circle, SE, Suite 201 ▪ Leesburg, VA 20175 ▪ 202-463-2766 ▪ www.awc.org ▪ [email protected]
 May 2018

ERRATA
to the 2015 Edition of
the National Design Specification® (NDS®) for Wood Construction

Page Revision
40 Revise Krs as described in Equation (5.4-3) as follows (replace de with dc):

Krs = empirical radial stress factor

= 0.29(dedc/Rm) + 0.32 tan1.2 T

222 Catoctin Circle, SE, Suite 201 ▪ Leesburg, VA 20175 ▪ 202-463-2766 ▪ www.awc.org ▪ [email protected]
 February 2018

ERRATA
to the 2015 Edition of
the National Design Specification® (NDS®) for Wood Construction

Page Revision
62 Revise Table 10.4.1.1 as follows:

Table 10.4.1.1 Shear Deformation

Adjustment Factors, KS

Loading End Fixity KS

Pinned 11.5
Uniformly Distributed
Fixed 57.6
Pinned 14.4
Line Load at midspan
Fixed 57.6
Line Load at quarter points Pinned 10.5

Constant Moment Pinned 0

Uniformly Distributed Cantilevered 4.8

Line Load at free-end Cantilevered 3.6

Pinned 11.8
Column Buckling
Fixed 23.7 47.4

242 Revise Table C10.4.1.1 as follows:

222 Catoctin Circle, SE, Suite 201 ▪ Leesburg, VA 20175 ▪ 202-463-2766 ▪ www.awc.org ▪ [email protected]
 February 2018

ADDENDUM
to the 2015 Edition of
the Commentary to the National Design Specification® (NDS®) for Wood Construction
(web and printed versions dated prior to 10-16)

A summary of those sections that have been updated are provided below with a brief
description of the change. Corrected pages are also attached for convenience.

Chapter C3: Design Provisions and Equations

• C3.4.2 –clarification on reference design shear, Vr.

Chapter C10: Cross-Laminated Timber

• Editorial clarifications in sections C10.1.1 Application, C10.1.2 Definition,
C10.1.4 Specification, and C10.2 Reference Design Values.

• The last paragraph in section C10.4.1 Deflection is expanded to address

the shear deformation adjustment factor, KS, to include the corresponding
beam constants, kb and ks for each of the loading cases shown in NDS
Table 10.4.1.1.

Chapter C12: Mechanical Connections

• In section C12.5.2 End Grain Factor, Ceg, new commentary (C12.5.2.3) is
added.

• Editorial clarification in table headings (Tables C12.1.5.7 and C12.1.6.6).

• Editorial clarification in sections C12.2.2 Wood Screws (C12.2.1.5,

C12.2.2.3, and C12.2.2.4), C12.2.3 Nails and Spikes (C12.2.3.5 and
C12.2.3.6), C12.3.3 Dowel Bearing Strength (C12.3.3.4, C12.3.3.5, and
C12.3.3.6), C12.3.5 Dowel Bearing Length (C12.3.5.2 and C12.3.5.3),
C12.3.6 Dowel Bending Yield Strength, and C12.5.1 Geometry Factor, C
(C12.5.1.4).

Chapter C16: Fire Design of Wood Members

• The term “fire resistance” replaces “fire endurance” throughout Chapter
16 Commentary.

222 Catoctin Circle, SE, Suite 201 ▪ Leesburg, VA 20175 ▪ 202-463-2766 ▪ www.awc.org ▪ [email protected]
Addendum to 2015 NDS Commentary
January 2017
Page 2

• Editorial clarification in section C16.1 General (Beams and Columns, Flame

Spread, and Fire Retardant Treatments sections).

• The definition for “tgi” is revised to clarify that it is the time for the char
front to reach the glued interface for each lamination.

• The last paragraph in section C16.2.4 Special Provisions for Structural

Glued Laminated Timber Beams is expanded and includes two new
figures, Figures C16A and C16B, which illustrate the core, inner and outer
tension, and inner and outer compression laminations for unbalanced and
balanced layups respectively.

Commentary References
• Several references are updated (2, 125, 126, 129, and 140) and a new
reference is added (198).
 NATIONAL DESIGN SPECIFICATION FOR WOOD CONSTRUCTION 199

lins that are adequately attached to the compression edge, plicable where loads are applied to the compression edge of
the unsupported length, 𝓁u, of the bending member is the the bending member, the most conservative loading case.
distance between purlins (61). The bending member must Formulas given in the footnote for load conditions not
also be braced at points of bearing. Consistent with the covered by the formulas in the body of the table represent
intent of NDS 3.3.3.4, the nonmandatory phrase “and/or the most limiting formula for the 𝓁ud range from those
lateral displacement” was removed and does not appear given for specified load conditions. For more information
in 2012 and later editions of the Specification because the on the derivation of these formulas, see TR14 (138).
requirement to “prevent rotation” at points of bearing is C3.3.3.6 The beam slenderness ratio, RB, is comparable
not optional and prevention of lateral displacement does to the slenderness ratio for solid columns, 𝓁ed, in terms C
not necessarily prevent rotation. of its effect on bending member design strength.
C3.3.3.5 Formulas are provided for determining the C3.3.3.7 Limiting the beam slenderness ratio, RB, to
effective span length, 𝓁e, from the unsupported length, a maximum value of 50 is a good practice recommenda-

COMMENTARY: DESIGN PROVISIONS AND EQUATIONS

𝓁u, for different loading and support conditions (138). tion intended to preclude design of bending members with
The 𝓁e values for small span-to-depth ratios, 𝓁ud < 7 are high buckling potential. This limit parallels the limit on
limited to address unrealistically large 𝓁e values that oth- slenderness ratio for columns, 𝓁ed (60).
erwise would be calculated for these short, deep bending C3.3.3.8 The beam stability factor equation is appli-
members (60). cable to all beam slenderness ratios (RB). This equation
The constants in the formulas for effective length in provides a means of combining the bending design stress
NDS Table 3.3.3 include a 15 percent increase in 𝓁u to (Fb*) with the critical buckling design stress (FbE) to esti-
account for the possibility of imperfect torsional restraint mate an “effective” bending design value.
at lateral supports. The formulas given in the table are ap- C3.3.3.10 See C3.9.2 on biaxial bending.

C3.4 Bending Members - Shear

C3.4.1 Strength in Shear Parallel ponents, such as trusses, are specifically excluded because
to Grain (Horizontal Shear) of field experience that indicated the procedures might not
be adequate for shear design of top-hung parallel chord
C3.4.1.1 Shear strength perpendicular to the grain, also trusses and similar components that contained load-bearing
referred to as cross-grain or vertical shear, refers to shear web and top chord connections near points of support.
stresses in the radial-tangential plane tending to cut the Shear design of built-up components is required to be
wood fibers perpendicular to their long axis. The strength based on testing, theoretical analysis, and/or documented
of wood in this plane is very high relative to shear strength experience due to the complexity of determining the effects
parallel to grain, or horizontal shear, which refers to shear of stress concentrations, the influence of embedded metal
stresses in the longitudinal-radial or longitudinal-tangential connectors, and questions regarding the applicability of
plane tending to slide one fiber past another along their the general practice of ignoring loads close to supports.
long axes. As both parallel and perpendicular to grain shear
occur simultaneously, parallel to grain shear strength is C3.4.2 Shear Design Equations
always the limiting case. Therefore, reference shear design
values, Fv, are horizontal or parallel to grain shear stresses. Actual shear stress parallel to grain, fv, in a circular
Shear in the tangential-longitudinal or radial-lon- bending member may be determined as:
gitudinal plane tending to roll one fiber over another
perpendicular to their long axes is termed rolling shear. fv = 4V / 3A (C3.4.2-1)
Rolling shear, which occurs in structural plywood appli- where:
cations as shear in the plane of the plies, is not a design
V = shear force, lbs
consideration in most lumber or timber product applica-
A = cross-sectional area of circular member, in.
tions.
C3.4.1.2 Shear design provisions in NDS 3.4 are For cross-laminated timber, the reference design shear,
limited to solid flexural members such as sawn lumber, Vr, is provided by the CLT manufacturer and is designated
structural glued laminated timber, structural composite in NDS Chapter 10 as Fs(Ib/Q)eff.
lumber, and mechanically laminated timber. Built-up com-
Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
 NATIONAL DESIGN SPECIFICATION FOR WOOD CONSTRUCTION 239

C10 CROSS-LAMINATED
TIMBER
C10.1 General C

C10.1.1 Application thickness of 5/8 in. and 2 in., respectively, are based on

COMMENTARY: CROSS-LAMINATED TIMBER

APA PRG 320.
The general requirements given in Chapters 1, 2, and 10.1.3.2 CLT panel thickness is measured perpen-
3 of the Specification are applicable to cross-laminated dicular to the plane of the panel and is limited to 20 in.
timber (CLT) except where indicated otherwise. Chapter in accordance with APA PRG-320. CLT panel length is
10 of the Specification contains provisions that specifically measured parallel to the major strength direction and
apply to CLT manufactured in accordance with APA PRG CLT panel width is measured perpendicular to the major
320 (197). The provisions of NDS Chapter 10 contain only strength direction.
the basic requirements applicable to engineering design
of CLT. Specific requirements, such as CLT design values C10.1.4 Specification
and the wet service factor are available from the CLT
manufacturer. The specific manufacturer’s CLT product should be
specified, including standard grade, where used, or the CLT
C10.1.2 Definitions configuration based on lamination grades, thicknesses, and
layup (See C10.2). Standard grades of CLT consisting of
The definition for cross-laminated timber is based on specific lamination grades, thicknesses, and layups are
the definition in APA PRG 320 (197). provided in Annex A of APA PRG-320.

C10.1.3 Standard Dimensions C10.1.5 Service Conditions

10.1.3.1 Lamination thickness refers to the narrow face CLT design values are based on dry service conditions
of a lamination perpendicular to the lamination length (face (moisture content in service less than 16%). For other
perpendicular to the glueline). Minimum and maximum conditions, the manufacturer should be consulted.

C10.2 Reference Design Values

Reference design values for specific grades and layups
of CLT are provided in APA PRG-320. CLT design capaci-
ties are a function of the manufacturer’s CLT layup and
properties associated with the lamination grades. The user
should contact the CLT manufacturer for design values and
section properties for specific CLT products.

Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
 NATIONAL DESIGN SPECIFICATION FOR WOOD CONSTRUCTION 241

C10.4 Special Design Considerations

C10.4.1 Deflection deformation equations are combined for each load case
to provide an adjustment that provides an “apparent” EIeff
C10.4.1.1 When cross-laminated panels are loaded value. The beam deflection, δ, for rectangular members
in out-of-plane bending, the shear deformation can be a is estimated as:
significant portion of the total deformation. The provisions
3
C
of NDS 10.4.1 provide a method of calculating the “ap- kbWL 6k sWL
parent” stiffness, (EI)app, from the properties provided in =δ + (C10.4.1-4)
EI eff 5GAeff
PRG 320 and from the CLT manufacturer.

COMMENTARY: CROSS-LAMINATED TIMBER

Where effective bending stiffness values (EIeff) and Where,
effective shear stiffness values (GAeff) are provided by
δ = beam deflection
the CLT manufacturer, the apparent bending stiffness can
be approximated as: kb, ks = beam constants based on beam loading,
support conditions, and measurement
EI eff location
(EI)app = (C10.4.1-1)
K s EI eff W = total load on the beam
1+ 2
GAeff L
Setting the beam deformation equations equal to the
Where only effective bending stiffness values (EIeff) bending deformation equation assuming an “apparent”
are provided, the apparent bending stiffness can be ap- stiffness, EIapp yields:
proximated using NDS Equation 10.4-1 where Ieff and
Aeff are provided by the CLT manufacturer. Design values kbWL3 kbWL3 6ksWL
for strong axis and weak axis modulus of elasticity, E, are = + (C10.4.1-5)
EI app EI eff 5GAeff
provided in PRG 320. PRG 320 assumes the modulus of
rigidity, G, equals E/16. The value for Ieff and Aeff can be Solving for EIapp:
calculated as:
EI eff EI eff
Ieff = EIeff / E (C10.4.1-2) EI app = =
6k s EI eff K s EI eff (C10.4.1-6)
1+ 1+
5kbGAeff L 2
GAeff L 2
Aeff = 16 GAeff / E (C10.4.1-3)
To estimate (EI)app-min the value for (EI)app is adjusted From this derivation, it can be seen that Ks = 6ks/
per provisions of NDS Appendix D and Appendix H and (5kb).
the coefficient of variation of 0.10 from PRG-320:
For the case of Column Buckling Moment, the deriva-
(EI)app-min = (EI)app (1-1.645(0.10))(1.03)/1.66 = 0.518(EI)app tion is more complex. For a column pinned at each end,
the bending deformation is estimated to be δb = A sin(πx)
and the shear deformation is estimated to be δs = Aπ2
Shear deformation adjustment factors, Ks, provided
sin(πx) and at midheight simplifies to δb = A and δs = Aπ2.
in NDS Table 10.4.1.1 and Table C10.4.1.1 are based
For a column fixed at each end, the bending deformation
on relationships for the specified loading and end-fixity
is estimated to be δb=B and the shear deformation is esti-
conditions from the Wood Handbook [183]. The value
mated to be δs=4Bπ2.
of Ks is derived when the beam deformation and shear

Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
242 NDS COMMENTARY: CROSS-LAMINATED TIMBER

Table C10.4.1.1 Shear Deformation Adjustment Factors

Loading End Fixity kb ks Ks

Pinned 5/384 1/8 11.5
Uniformly Distributed
Fixed 1/384 1/8 57.6
Pinned 1/48 1/4 14.4
Line Load at midspan
Fixed 1/192 1/4 57.6
Line Load at quarter points Pinned 11/768 1/8 10.5
Constant Moment - 1/12 0 0
Uniformly Distributed Cantilevered 1/8 1/2 4.8
Line Load at free-end Cantilevered 1/3 1 3.6
Pinned A Aπ2 11.8
Column Buckling
Fixed 2B 4Bπ2 23.7

Cross-laminated timber panels are designed as

one‐way slabs using a beam analogy. Terminology in
NDS Table 10.4.1.1 and Table C10.4.1.1 has been modi-
fied from typical beam terminology to address the panel
width. For example, a point load at midspan of a beam
is called a “line load at midspan” to indicate that the load
is assumed to be applied across the panel (perpendicular
to the span) at the midspan of the cross-laminated timber.
Additional loading and end-fixity conditions are available
in the literature.

Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
248 NDS COMMENTARY: DOWEL-TYPE FASTENERS

C12.1.4.7 Edge distances, end distances, and fastener

Table C12.1.5.7 Recommended
spacing requirements have been consolidated for dowel-
Minimum Spacing for Wood Screws
type fasteners in NDS 12.5.
Wood Side Members
C12.1.5 Wood Screws
Not
Prebored Prebored
C12.1.5.1 ANSI/ASME Standard B18.6.1 is the Edge distance 2.5D 2.5D
quality reference standard for wood screws. It provides End distance
standard wood screw dimensions (see NDS Appendix L) - tension load parallel to grain 15D 10D
but does not specify metal having specific strength proper- - compression load parallel to grain 10D 5D
ties. The designer is responsible for specifying the metal Spacing between fasteners in a row
strength of the wood screws that are to be used. Bending - parallel to grain 15D 10D
- perpendicular to grain 10D 5D
yield strength of the wood screw (see NDS Appendix I) is
Spacing between rows of fasteners
a required input variable to the lateral design value yield - in-line 5D 3D
limit equations of NDS 12.3.1. Additionally, the actual - staggered 2.5D 2.5D
tensile stress in the wood screw at the root diameter must Steel Side Members
be checked when designing wood screw connections for Not
withdrawal (see C11.2.3 of Specification). Prebored Prebored
C12.1.5.2 Lead hole requirements for wood screws are Edge distance 2.5D 2.5D
based on early research involving flat head wood screws End distance
up to 24 gage and 5 inches in length in seven species, - tension load parallel to grain 10D 5D
including southern pine, cypress, and oak (43). - compression load parallel to grain 5D 3D
The provision allowing the insertion of wood screws Spacing between fasteners in a row
- parallel to grain 10D 5D
without a lead hole in species with G ≤ 0.5 when the screw 5D 2.5D
- perpendicular to grain
is subject to withdrawal loads parallels the provision for Spacing between rows of fasteners
3/8 inch and smaller diameter lag screws (see C12.1.4.3). - in line 3D 2.5D
C12.1.5.3 Wood screws resisting lateral loads are - staggered 2.5D 2.5D
required to have shank and threaded portion lead holes
based on early lateral load tests of wood screws (70, 181, C12.1.6 Nails and Spikes
184). Lead holes are required for all wood screws subject
to lateral loads regardless of wood specific gravity. C12.1.6.1 ASTM F 1667 Standard Specification for
C12.1.5.4 Wood screws tests (43, 70, 181) are based Driven Fasteners: Nails, Spikes, and Staples provides
on inserting the screw by turning rather than driving with standard nail, spike, threaded hardened-steel nail, and
a hammer. post-frame ring shank nail dimensions (see NDS Appendix
C12.1.5.5 A lubricant is sometimes used to facilitate L) but does not specify metal of particular strength proper-
screw insertion and avoid screw damage. Tests have shown ties for these driven fasteners. The designer is responsible
that the lubricant has no significant effect on reference for specifying the metal strength of the driven fasteners
design values (43, 70, 184). that are to be used. Bending yield strength of the driven
C12.1.5.6 Minimum length of wood screw penetration fastener (see NDS Appendix I) is a required input variable
requirements, including the length of the tapered tip, are to the lateral design value yield limit equations of NDS
provided to ensure that fasteners can achieve the refer- 12.3.1. Additionally, the actual tensile stress in the driven
ence design value calculated using the yield equations in fastener must be checked when designing driven fastener
NDS 12.3.1. connections for withdrawal (see C11.2.3 of Specification).
C12.1.5.7 Edge distances, end distances, and fastener C12.1.6.4 Toe-nailing procedures consisting of slant
spacing requirements have been consolidated across all driving of nails at a 30° angle from the face of the attached
diameters for dowel-type fasteners in NDS 12.5. For diam- member with an end distance (distance between end of
eters less than ¼", specific requirements are not provided; side member and initial point of entry) of one-third the
however, Table C12.1.5.7 may be used to establish wood nail length are based on lateral and withdrawal tests of
screw placement recommendations. Designers should note nailed joints in frame wall construction (181, 118). The
that wood species, moisture content, and grain orientation toenail factors of NDS 12.5.4.1 and NDS 12.5.4.2 presume
will impact spacing effects between fasteners in a row. use of these driving procedures and the absence of exces-
sive splitting. If such splitting does occur, predrilling or
Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
 NATIONAL DESIGN SPECIFICATION FOR WOOD CONSTRUCTION 249

a smaller nail should be used. The vertically-projected

Table C12.1.6.6 Recommended
length is used as the side member bearing length in yield
Minimum Spacing for Nails
limit equations when calculating lateral capacity of a toe-
nailed connection. Wood Side Members
C12.1.6.5 Minimum length of penetration require-
Not
ments, including the length of the tapered tip, are provided Prebored Prebored
to ensure that driven fasteners can achieve the design value Edge distance 2.5D 2.5D
calculated using the yield equations in NDS 12.3.1. The End distance
exception for clinching in double-shear connections is ap- - tension load parallel to grain 15D 10D C
plicable to 0.148" (12d common, 20d box, or 16d sinker - compression load parallel to grain 10D 5D
nails) or smaller diameter nails. Spacing between fasteners in a row
C12.1.6.6 Edge distances, end distances, and fastener - parallel to grain 15D 10D

COMMENTARY: DOWEL-TYPE FASTENERS

- perpendicular to grain 10D 5D
spacing requirements have been consolidated across all
Spacing between rows of fasteners
diameters for dowel-type fasteners in NDS Table 12.5.1A - in-line 5D 3D
through 12.5.1F. For diameters less than ¼", specific re- - staggered 2.5D 2.5D
quirements are not provided; however, Table C12.1.6.6 Steel Side Members
may be used to establish nail placement recommendations. Not
Designers should note that wood species, moisture content, Prebored Prebored
and grain orientation will impact spacing effects between Edge distance 2.5D 2.5D
fasteners in a row. End distance
- tension load parallel to grain 10D 5D
C12.1.7 Drift Bolts and Drift Pins - compression load parallel to grain 5D 3D
Spacing between fasteners in a row
- parallel to grain 10D 5D
C12.1.7.1 Drift bolts and drift pins are unthreaded rods 5D 2.5D
- perpendicular to grain
used to join large structural members where a smooth sur- Spacing between rows of fasteners
face without protruding metal parts is desired. The designer - in line 3D 2.5D
is responsible for specifying the metal strength of the drift - staggered 2.5D 2.5D
bolt or pin that is to be used. Bending yield strength of the
drift bolt or pin (see NDS Appendix I) is a required input C12.1.8 Other Dowel-Type
variable to the reference lateral design value yield limit Fasteners
equations of NDS 12.3.1.
C12.1.7.2 Additional penetration into the members is While specific installation instructions are not pro-
required to resist withdrawal of the drift bolt or pin. vided for all types of dowel-type fasteners, the provisions
C12.1.7.3 Edge distances, end distances, and fastener for withdrawal in NDS 12.2 and the generic yield equa-
spacing requirements have been consolidated across all tions in NDS 12.3.1 for lateral design apply. The designer
diameters for dowel-type fasteners in NDS Table 12.5.1A is responsible for determining the proper installation re-
through 12.5.1F. quirements and for specifying the metal strength of these
fasteners.

C12.2 Reference Withdrawal Design Values

C12.2.1 Lag Screws where:
W = reference withdrawal design value per inch
C12.2.1.1 NDS Equation 12.2-1 was used to establish of thread penetration into main member, lbs
the lag screw reference withdrawal design values given
KW = 1800
in NDS Table 12.2A. This equation was derived from the
following equation based on research (181, 100): G = specific gravity of main member based on
oven dry weight and volume, where 0.31 ≤ G
W = KW G3/2 D3/4 (C12.2.1-1) ≤ 0.73

Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
250 NDS COMMENTARY: DOWEL-TYPE FASTENERS

D = lag screw diameter (equivalent to timber panels (e.g. both end grain and side grain are pres-
unthreaded shank diameter for full body ent in the edge of CLT panels), and the ability to maintain
diameter lag screws), in., where 0.25" ≤ D ≤ minimum edge distances for larger diameter lag screws
1.25" installed in the narrow face of a lamination. For cases
The value of KW represents approximately one-fourth where the narrow face of the laminations is large, such as
(one-fifth increased by 20 percent) of the average constant 2 in., and the lag screw diameter is small such as ¼" and
at oven dry weight and volume obtained from ultimate where installation is in side grain only with adequate edge
load tests of joints made with five different species and distance, application of the 0.75 factor may not be war-
seven sizes of lag screw (100), increased by 20 percent; or ranted where strength reducing conditions associated with
placement in end grain or with inadequate edge distances
 7500  for side grain are not present.
K W = 1.2   (C12.2.1-2)
 5 
C12.2.2 Wood Screws
The twenty percent increase was introduced as part
of the World War II emergency increase in wood design C12.2.2.1 NDS Equation 12.2-2 was used to establish
values, and then subsequently codified as 10 percent for the the wood screw reference withdrawal design values given
change from permanent to normal loading and 10 percent in NDS Table 12.2B. This equation was based on testing
for experience (see C2.3.2). of cut thread wood screws in seven wood species (43):
When the reference withdrawal capacity of a lag screw
is determined by multiplying the reference unit design W = KW G2 D (C12.2.2-1)
value by the length of penetration of the threaded portion
where:
into the side grain of the main member, the length of the
tapered tip of the screw is not to be included. This tapered W = reference withdrawal design value per inch
portion at the tip of the lag screw was not considered as part of thread penetration in the main member,
lbs
of the effective penetration depth in the original joint tests
(100) and in the development of equation C12.2.1-1. In ad- KW = 2850
dition, the thickness of any washer used between the lag
G = specific gravity of main member based on
screw head and the cleat or side member should be taken
oven dry weight and volume, where 0.31 ≤ G
into account when determining the length of penetration ≤ 0.73
of the threaded portion in the main member. Standard
lag screw dimensions, including minimum thread length D = wood screw thread diameter, in., where
and length of tapered tip, are given in Appendix L of the 0.138" ≤ D ≤ 0.372"
Specification. The value of KW represents one-fifth (one-sixth in-
C12.2.1.2 The unit reference withdrawal design value creased by 20 percent) of the average constant at oven
in lbs/in. is multiplied by the depth of thread penetration dry weight and volume obtained from ultimate load tests
into a wood member to calculate the fastener reference of joints (43) made with seven different species and cut-
withdrawal in pounds. thread wood screw; or
C12.2.1.3 Reference withdrawal design values for lag
screws are reduced 25 percent when the screw is inserted in 14250 (C12.2.2-2)
KW =
the end grain (radial-tangential plane) of the main member 6
rather than the side grain (radial-longitudinal or tangential- The twenty percent increase was introduced as part
longitudinal plane) based on lag screw joint tests (100). of the World War II emergency increase in wood design
Because of the greater possibility of splitting when subject values, and then subsequently codified as 10 percent for the
to lateral load, it has been recommended that insertion change from permanent to normal loading and 10 percent
of lag screws in end grain surfaces be avoided (181, 96). for experience (see C2.3.2).
C12.2.1.4 (see C11.2.3). Wood screw reference withdrawal design values are
C12.2.1.5 The required use of the end grain factor of based on tests of cut thread wood screws. The shank or
0.75 for lag screws installed into the narrow edge of CLT body diameter of a cut thread screw is the same as the
panels conservatively assumes the lag screw will be subject outside diameter of the thread. The shank or body diameter
to strength reductions associated with installation in end of the rolled thread screw is the same as the root diameter.
grain. This assumption was judged practical to address For the same nominal diameter of screw, both screw thread
varying grain orientations in the edge of cross-laminated types have the same threads per inch, the same outside
Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
 NATIONAL DESIGN SPECIFICATION FOR WOOD CONSTRUCTION 251

thread diameter, and the same thread depth. If the tensile D = shank diameter of the nail or spike, in.,
strength of the screw is adequate and the lead hole provi- where 0.099" ≤ D ≤ 0.375"
sions based on root diameter are used, the withdrawal The value of KW represents one-fifth (one-sixth in-
resistance of rolled thread screws is considered equivalent creased by 20 percent) of the average constant at oven
to that of cut thread screws (182, 163). dry weight and volume obtained from ultimate load tests
The ANSI/ASME B18.6.1 standard states that the (184), increased by 20 percent; or
thread length is approximately two-thirds of the nominal
screw length.  6900 
C12.2.2.2 The unit reference withdrawal design value
KW = 1.2 
 6 
(C12.2.3-2)
C
in lbs/in. is multiplied by the depth of thread penetration
The twenty percent increase was introduced as part
into a wood member to calculate the fastener reference
of the World War II emergency increase in wood design
withdrawal in pounds.

COMMENTARY: DOWEL-TYPE FASTENERS

values, and then subsequently codified as 10 percent for the
C12.2.2.3 Early tests of wood screws in withdrawal
change from permanent to normal loading and 10 percent
from end grain surfaces of oak, southern pine, maple, and
for experience (see C2.3.2).
cypress gave somewhat erratic results relative to those for
For 8d, 10d, 16d, and 20d threaded hardened nails,
withdrawal from side grain (43). These irregular results
reference withdrawal design values are the same as those
were attributed to the tendency of the screw to split the
for common wire nails of the same pennyweight class,
wood in the end grain configuration. Average ratios of
although the wire diameters are slightly different (0.120,
end grain withdrawal resistance to side grain withdrawal
0.135, 0.148, and 0.177 inches for threaded hardened nails
resistance ranged from 52 to 108 percent (43). Because of
versus 0.131, 0.148, 0.162, and 0.192 inches for common
this variability, structural loading of wood screws in with-
nails, respectively). Threaded hardened nail sizes of 20d,
drawal from end grain has been prohibited. Wood screws
30d, 40d, 50d, and 60d all have the same diameter (0.177
installed in end grain are suitable for lateral resistance;
inches) and, therefore, use the same reference withdrawal
however, to clarify that screws are permitted to be installed
design value. Threaded hardened nail sizes of 70d, 80d,
in end-grain but should not be assigned withdrawal design
and 90d all have the same diameter (0.207 inches) and
values, the end-grain adjustment factor, Ceg, is set to zero
use the same reference withdrawal design value as a 40d
for withdrawal loading.
common nail.
C12.2.2.4 Similar to the provisions of 12.2.2.3, wood
Clinching. Withdrawal resistance of smooth-shank
screws installed in end-grain of cross-laminated timber
nails can be significantly increased by clinching (29).
laminations should not be assigned withdrawal design
C12.2.3.2 The unit reference withdrawal design value
values (i.e. Ceg=0.0). ). There is no reduction in withdrawal
in lbs/in. for nails and spikes is multiplied by the depth of
resistance for wood screws installed in the side grain of
thread penetration into a wood member to calculate the
laminations at cross-laminated timber panel edges.
fastener reference withdrawal in pounds.
C12.2.2.5 See C11.2.3.
C12.2.3.3 In the 2012 edition of the Specification,
provisions were added for post-frame ring shank nails in
C12.2.3 Nails and Spikes
accordance with ASTM F1667. The withdrawal design
value equation (NDS Equation 12.2-4) is based on research
C12.2.3.1 NDS Equation 12.2-3 was used to establish
conducted at the Forest Products Laboratory. The constant
the nail and spike reference withdrawal design values
of 1800 incorporates a 20% reduction to account for ef-
given in NDS Table 12.2C. This equation was based on
fects of galvanized coatings for Southern pine rather than
research (94, 95):
the average reduction of 15% from all species (basswood,
W = KW G5/2 D (C12.2.3-1) SPF, Douglas fir, southern pine, and white oak) tested in
the study (196).
where: C12.2.3.4 The unit reference withdrawal design value
W = nail or spike withdrawal design value per inch in lbs/in. for post-frame ring shank nails is multiplied by
of penetra­tion in main member, lbs the depth of thread penetration into the side grain of a wood
KW = 1380
member to calculate the fastener reference withdrawal in
pounds.
G = specific gravity of main member based on C12.2.3.5 Reduction of withdrawal design values up
oven dry weight and volume, where 0.31 ≤ G to 50 percent have been reported for nails driven in end
≤ 0.73 grain surfaces (radial-tangential plane) as compared to
Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
252 NDS COMMENTARY: DOWEL-TYPE FASTENERS

side grain (radial-longitudinal or tangential-longitudinal not specifically included in the Specification, the following
planes) surfaces (184, 118). When coupled with the effects equation has been used where friction and workmanship
of seasoning in-service after fabrication, such reductions can be maintained (184, 181):
are considered too great for reliable design. On this basis,
structural loading of nails in withdrawal from end grain has W = 1200 G2 D (C12.2.4-1)
been prohibited. Nails installed in end grain are suitable for where:
lateral resistance; however, to clarify that nails are permit-
W = drift bolt or drift pin reference withdrawal
ted to be installed in end-grain but should not be assigned design value per inch of penetration, lbs
withdrawal design values, the end-grain adjustment factor,
Ceg, is set to zero for withdrawal loading. G = specific gravity based on oven dry weight
C12.2.3.6 Similar to the provisions of 12.2.3.5, nails and volume
installed in end-grain of cross-laminated timber lamina- D = drift bolt or drift pin diameter, in.
tions should not be assigned withdrawal design values (i.e.
Equation C12.2.4-1 assumes the fastener is driven into
Ceg=0.0). There is no reduction in withdrawal resistance
a prebored hole having a diameter 1/8 inch less than the
for nails installed in the side grain of laminations at cross-
fastener diameter (184). The reference withdrawal design
laminated timber panel edges.
values calculated with Equation C12.2.4-1 are approxi-
mately one-fifth average ultimate test values (184, 181).
C12.2.4 Drift Bolts and Drift Pins

C12.2.4.1 While specific provisions for determining

withdrawal design values for round drift bolts or pins are

C12.3 Reference Lateral Design Values

Reference lateral design values for dowel-type fasten- Bolts: Reference lateral design values for bolted
ers (bolts, lag screws, wood screws, nails and spikes) are connections are indexed to proportional limit estimates
based on a yield limit model which specifically accounts from bolted connection tests (44, 57, 146, 162) at reference
for the different ways these connections can behave under conditions (seasoned dry, normal load duration).
load. These behavior patterns or modes (see NDS Appen- Lag Screws: Reference lateral design values for
dix I) are uniform bearing in the wood under the fastener, lag screw connections are indexed to average short-term
rotation of the fastener in the joint without bending, and proportional limit test values (100) divided by 1.875. The
development of one or more plastic hinges in the fastener 1.875 factor is based on an original reduction factor of
(67, 122). Equations have been developed for each mode 2.25, increased 20 percent for normal loading and experi-
relating the joint load to the maximum stresses in the wood ence. The twenty percent increase was introduced as part
members and in the fastener (67, 121). The capacity of the of the World War II emergency increase in wood design
connection under each yield mode is keyed to the bearing values, and then subsequently codified as 10 percent for the
strength of the wood under the fastener and the bending change from permanent to normal loading and 10 percent
strength of the fastener, with the lowest capacity calculated for experience (see C2.3.2).
for the various modes being taken as the reference design Wood Screws: Reference lateral design values for
value for the connection. wood screw connections are indexed to average short-term
Although the yield limit model represents significantly proportional limit test values (184, 70) divided by 1.33. The
different methodology than that used previously to estab- 1.33 factor is based on an original reduction factor of 1.6,
lish fastener design values, the relative effects of various increased 20 percent for normal loading and experience.
joint variables shown by both procedures are generally The twenty percent increase was introduced as part of the
similar (85, 86, 89, 121). Short-term design values obtained World War II emergency increase in wood design values,
from application of the yield limit equations have been and then subsequently codified as 10 percent for the change
reduced to the design value levels published in previous from permanent to normal loading and 10 percent for
editions of the Specification for connections made with experience (see C2.3.2). Lateral design values for wood
the same species and member sizes. screw connections at reference conditions (seasoned dry,

Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
 NATIONAL DESIGN SPECIFICATION FOR WOOD CONSTRUCTION 253

normal load duration) are about one-fifth of maximum tions in NDS 12.3.1. Assumptions used in the yield limit
tested capacities (184). equations to develop the tables are provided in the table
Nails & Spikes: Reference lateral design values for headings and footnotes.
nailed connections are indexed to average short-term pro-
portional limit test values (184, 50) divided by 1.33. The C12.3.3 Dowel Bearing Strength
1.33 factor is based on an original reduction factor of 1.6,
increased 20 percent for normal loading and experience. The C12.3.3.1 The limiting wood stresses used in the
twenty percent increase was introduced as part of the World yield limit equations are based on the load at which the
War II emergency increase in wood design values, and then load-deformation curve from a fastener embedment test C
subsequently codified as 10 percent for the change from intersects a line represented by the initial tangent modulus
permanent to normal loading and 10 percent for experience offset 5 percent of the fastener diameter (120). This nomi-
(see C2.3.2). Lateral design values for nailed connections nal yield point is intermediate between the proportional

COMMENTARY: DOWEL-TYPE FASTENERS

at reference conditions (seasoned dry, normal load dura- limit and maximum loads for the material.
tion) are about one-fifth of maximum tested capacities for The effect of specific gravity on dowel bearing strength
softwoods and one-ninth of maximum tested capacities for was established from 3/4-inch dowel embedment tests on
hardwoods (184, 50). Douglas fir, southern pine, spruce-pine-fir, sitka spruce,
red oak, yellow poplar, and aspen. Diameter effects were
C12.3.1 Yield Limit Equations evaluated from tests of 1/4, 1/2, 3/4, 1, and 1-1/2 inch dow-
els in southern pine using bolt holes 1/16-inch larger than
The yield limit equations for single shear connec- the dowel diameter. Diameter was found to be a significant
tions (NDS Equations 12.3-1 to 12.3-6) and for double variable only in perpendicular to grain loading. Bearing
shear connections (NDS Equations 12.3-7 to 12.3-10) specimens were 1/2-inch or thicker such that width and
were developed from European research (121, 78) and number of growth rings did not influence results (158).
have been confirmed by tests of connections made with The specific gravity values given in NDS Table
domestic species (21, 20, 88, 120, 121, 122). The limiting 12.3.3A for each specie or species group are those used
yield modes covered by these equations are bearing in the to establish dowel bearing strength values, Fe, tabulated in
main or side members (Mode I), fastener rotation without NDS Table 12.3.3. These specific gravity values represent
bending (Mode II), development of a plastic hinge in the average values from in-grade lumber test programs or are
fastener in main or side member (Mode III) and develop- based on information from ASTM D 2555.
ment of plastic hinges in the fastener in both main and side The equations provided in footnote 2 of NDS Table
members (Mode IV) (see NDS Appendix I). 12.3.3 were used to calculate tabulated values in NDS
The reduction term, Rd, in NDS Equations 12.3-1 Table 12.3.3. These equations were derived from test data
through 12.3-10, reduces the values calculated using using methods described in ASTM D 5764 (158, 18).
the yield limit equations to approximate estimates of C12.3.3.2 Dowel bearing strengths for wood structural
the nominal proportional limit design values in previous panels using a dowel diameter of less than or equal to ¼
editions of the Specification (157). For fasteners loaded inch are provided in NDS Table 12.3.3B and are based on
perpendicular to grain with diameters equal to or greater research conducted by APA-The Engineered Wood Asso-
than 0.25 inches, the reduction term is increased 25% (Kθ ciation (25). Dowel bearing values for larger diameters in
= 1.25) to match previous design values for connections wood structural panels are available in APA 825E.
loaded perpendicular to grain. C12.3.3.3 Dowel bearing strengths for structural
For detailed technical information on lateral design composite lumber are determined for each product using
equations, see AWC’s Technical Report 12: General Dowel equivalency methods described in ASTM D5456 (16).
Equations for Calculating Lateral Connection Values (137) C12.3.3.4 For fasteners with D≥1/4" installed in
available at www.awc.org. end-grain, the dowel bearing strength is equal to the per-
pendicular to grain value, Fe┴.
C12.3.2 Common Connection C12.3.3.5 For fasteners in the wide face of CLT, dowel
Conditions bearing strength is based on the dowel bearing strength of
the layer at the shear plane. The orientation of the layer at
Reference lateral design values, Z, for connections the shear plane may either be parallel or perpendicular to
with bolts, lag screws, wood screws, nails and spikes are the major strength axis of the panel and should be part of
calculated for common connection conditions and assumed the specification of the cross-laminated timber panel (see
fastener bending yield strengths using the yield limit equa- Commentary C10.1.4). For connections where the load-
Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
254 NDS COMMENTARY: DOWEL-TYPE FASTENERS

ing direction is parallel to grain for the layer at the shear bearing length should be reduced by multiplying the bear-
plane, the dowel bearing strength is the parallel to grain ing length in each crossing layer (perpendicular to grain)
dowel bearing strength, Fe║. For connections where the by the ratio of Fe┴/Fe║. For connections where the loading
loading direction is perpendicular to grain for the layer at direction is perpendicular to grain for the layer at the shear
the shear plane, the dowel bearing strength is the perpen- plane, the dowel bearing length can conservatively remain
dicular to grain dowel bearing strength, Fe┴. The influence unadjusted or it can be increased in the crossing layers
of different dowel bearing strengths of crossing layers on (parallel to grain) by the ratio of Fe║/Fe┴. Actual penetration
cross-laminated connection design values is accounted for lengths should be used for checking minimum penetra-
by adjustment of the bearing length in the crossing layers tion requirements. For connections loaded at an angle to
(See NDS 12.3.5.2). grain, the procedures in NDS Appendix J for developing
C12.3.3.6 For fasteners with D≥1/4" that are installed design values based on parallel and perpendicular to grain
into the edge of cross-laminated timber, the dowel bearing design values should be used with these “effective” bear-
strength is assumed to be the same as for fasteners installed ing lengths. Methods of installation should avoid placing
into end-grain (See NDS 12.3.3.4) which conservatively fasteners in gaps between adjacent boards in a lamination,
addresses varying grain orientations and the ability to especially where they might occur in the lamination at the
maintain minimum edge distances within the narrow face shear plane.
of a cross-laminated timber lamination. For fasteners with C12.3.5.3 An analysis provided in Technical Report
D<1/4", the same dowel bearing strength, Fe, applies for 12 (137) shows that the NDS requirement closely ap-
either parallel or perpendicular to grain loading. proximates results from the more detailed evaluation of the
influence of a tapered tip on bearing resistance. For wood
C12.3.4 Dowel Bearing Strength at screws, nails and spikes, the length of the tapered tip is
an Angle to Grain not generally standardized, but for purposes of accounting
for the tip length in the bearing length calculation, E, is
NDS Equation 12.3-11 (and Equation J-2 in NDS Ap- permitted to be taken as 2 diameters (2D). For lag screws,
pendix J) is used to calculate the dowel bearing strength E is permitted to be taken from NDS Appendix L, Table L2.
for a main or side member loaded at an angle to grain. This
equation is a form of the bearing angle to grain equation C12.3.6 Dowel Bending Yield
(NDS Equation J-1). The equation is entered with the Strength
parallel and perpendicular dowel bearing strengths for the
member and the reference bolt design value is determined The dowel bending yield strength, Fyb, of fasteners
from the yield limit equations using Feθ as the dowel bear- such as nails (79), wood screws, lag screws, and bolts are
ing strength for the main or side member. given in NDS Appendix I. For most steel fasteners, Fyb
The reference design value obtained from the yield equal to 45,000 psi is a conservative value and is equiva-
limit equations using dowel bearing strength at an angle lent to the bolt strength reported in the original bolt test
to grain is similar to that obtained from using parallel to research (146).
grain and perpendicular to grain Z values in NDS Equation
J-3 to obtain a Zθ design value for the connection (157). C12.3.7 Dowel Diameter
Determining a Zθ design value using this latter approach
can be used as an alternative to calculating Feθ for use in The reduced moment resistance in the threaded por-
each yield limit equation and allows the use of tabulated tion of dowel-type fasteners can be accounted for by use
Z values from the Specification. of root diameter, Dr, in calculation of reference lateral
design values. Use of diameter, D, is permitted when the
C12.3.5 Dowel Bearing Length threaded portion of the fastener is sufficiently far away
from the connection shear plane(s). For more information,
C12.3.5.2 For fasteners with D≥1/4", crossing layers see NDS Appendix I.5.
in cross-laminated timbers will have different dowel bear- Reference lateral design values for reduced body diam-
ing strengths than the layer at the shear plane due to the eter lag screw and rolled thread wood screw connections
difference in grain orientation. The influence of varying are based on root diameter, Dr, to account for the reduced
dowel bearing strengths in crossing layers on connection diameter of these fasteners. These values, while conserva-
design values is addressed by use of an “effective” bear- tive, can also be used for full-body diameter lag screws and
ing length. For connections where the loading direction is cut thread wood screws. For bolted connections, reference
parallel to grain for the layer at the shear plane, the dowel lateral design values are based on diameter, D.
Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
258 NDS COMMENTARY: DOWEL-TYPE FASTENERS

recognition of the combined effect of bolt hole oversizing 118). When coupled with the effects of seasoning in service
and alignment resulting in 1/32" movement. after fabrication, such reductions are considered too great
Special detailing can be utilized in cases where dis- for reliable design. It is considered to be on this basis that
tances between outer rows of bolts exceed the limits in loading of nails and spikes in withdrawal from end grain
Table 12.5.1F, such as use of multiple splice plates or a has been prohibited.
single splice plate with slotted holes to allow shrinkage. C12.5.2.2 The use of a 0.67 adjustment factor on refer-
Such an example of multiple splice plates is shown in ence lateral design values for lag screws, wood screws, nails,
Figure C12.5.1.3. or spikes driven in the end grain is based on early research
on joints made with softwood species (181, 184).
Figure C12.5.1.3 Connection C12.5.2.3 For fasteners with D≥1/4", the use of a 0.67
Illustrating Use of Multiple Splice adjustment factor for fasteners installed in the edge of a
Plates cross-laminated timber panel is based on the assumption
that fasteners will be installed into end grain of the cross-
laminated timber lamination (see Commentary C12.3.3.6)
regardless of whether installation is actually into end grain.
Testing of large fasteners installed into end-grain or between
laminations with end-grain and side-grain indicated that
C12.5.1.4 For fasteners installed in the edge of the 0.67 adjustment factor was sufficiently conservative,
cross-laminated timber panels, special end distance, edge even when gaps were present (198). For smaller diameter
distance and fastener spacing conditions are provided in fasteners with D<1/4", the end grain factor in 12.5.2.2 ap-
NDS Table 12.5.1G, while all other requirements fol- plies where installation is into end grain (see Commentary
low the general provisions of NDS 12.5.1. For fasteners C12.5.2.2) of a cross-laminated timber lamination.
installed in the wide face of cross-laminated panels, end
distances, edge distances, and fastener spacing require- C12.5.3 Diaphragm Factor, Cdi
ments should follow the requirements for other wood
products in NDS 12.5.1. Placement of fasteners in gaps Diaphragms are large, flat structural units acting like a
should be avoided. deep relatively thin beam or girder. Horizontal wood dia-
phragms consist of floor or roof decks acting as webs, and
C12.5.2 End Grain Factor, Ceg lumber, structural glued laminated timber, SCL, or I-joist
members acting as the flanges. Such assemblies distribute
C12.5.2.1 Reducing reference withdrawal design val- horizontal forces acting on the flanges to vertical resisting
ues for lag screws 25 percent when the screw is inserted in elements (103). Shear walls consisting of wall sheathing
the end grain (radial-tangential plane) of the main member materials attached to top and bottom plates and vertical
rather than the side grain (radial-longitudinal or tangential- framing members also are diaphragms. Such shear walls
longitudinal plane) is based on lag screw joint tests (100). or vertical diaphragms act to transfer loads from horizontal
Early tests of wood screws in withdrawal from end diaphragms down to the supporting foundation. The dia-
grain surfaces of oak, southern pine, maple, and cypress phragm factor, Cdi, applies to both horizontal and vertical
gave somewhat erratic results relative to those for with- diaphragms (144, 145).
drawal from side grain (43). These irregular results were
attributed to the tendency of the screw to split the wood C12.5.4 Toe-Nail Factor, Ctn
in the end grain configuration. Average ratios of end grain
withdrawal resistance to side grain withdrawal resistance C12.5.4.1 The 0.67 adjustment of reference with-
ranged from 52 to 108 percent (43). Because of this vari- drawal design values for toenailing is based on the results
ability, structural loading of wood screws in withdrawal of joint tests comparing slant driving and straight driving
from end grain has been prohibited. Where splitting is (184) and of typical toenailed and end nailed joints used in
avoided, use of an end grain to side grain withdrawal frame wall construction (118) where the attached member
design value ratio of 75 percent has been suggested (184, is pulled directly away from the main member. It is ap-
183). plicable to joints fabricated at all levels of seasoning. This
Reduction of withdrawal design values up to 50 percent includes multiple nail joints fabricated of unseasoned wood
have been reported for nails driven in end grain surfaces and then loaded after seasoning (184, 183, 118). Toenail-
(radial-tangential plane) as compared to side grain (radial- ing with cross slant driving can produce stronger joints
longitudinal or tangential-longitudinal planes) surfaces (184, than end or face nailing. For example, a stud to plate joint
Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
258 NDS COMMENTARY: DOWEL-TYPE FASTENERS

recognition of the combined effect of bolt hole oversizing 118). When coupled with the effects of seasoning in service
and alignment resulting in 1/32" movement. after fabrication, such reductions are considered too great
Special detailing can be utilized in cases where dis- for reliable design. It is considered to be on this basis that
tances between outer rows of bolts exceed the limits in loading of nails and spikes in withdrawal from end grain
Table 12.5.1F, such as use of multiple splice plates or a has been prohibited.
single splice plate with slotted holes to allow shrinkage. C12.5.2.2 The use of a 0.67 adjustment factor on refer-
Such an example of multiple splice plates is shown in ence lateral design values for lag screws, wood screws, nails,
Figure C12.5.1.3. or spikes driven in the end grain is based on early research
on joints made with softwood species (181, 184).
Figure C12.5.1.3 Connection C12.5.2.3 For fasteners with D≥1/4", the use of a 0.67
Illustrating Use of Multiple Splice adjustment factor for fasteners installed in the edge of a
Plates cross-laminated timber panel is based on the assumption
that fasteners will be installed into end grain of the cross-
laminated timber lamination (see Commentary C12.3.3.6)
regardless of whether installation is actually into end grain.
Testing of large fasteners installed into end-grain or between
laminations with end-grain and side-grain indicated that
C12.5.1.4 For fasteners installed in the edge of the 0.67 adjustment factor was sufficiently conservative,
cross-laminated timber panels, special end distance, edge even when gaps were present (198). For smaller diameter
distance and fastener spacing conditions are provided in fasteners with D<1/4", the end grain factor in 12.5.2.2 ap-
NDS Table 12.5.1G, while all other requirements fol- plies where installation is into end grain (see Commentary
low the general provisions of NDS 12.5.1. For fasteners C12.5.2.2) of a cross-laminated timber lamination.
installed in the wide face of cross-laminated panels, end
distances, edge distances, and fastener spacing require- C12.5.3 Diaphragm Factor, Cdi
ments should follow the requirements for other wood
products in NDS 12.5.1. Placement of fasteners in gaps Diaphragms are large, flat structural units acting like a
should be avoided. deep relatively thin beam or girder. Horizontal wood dia-
phragms consist of floor or roof decks acting as webs, and
C12.5.2 End Grain Factor, Ceg lumber, structural glued laminated timber, SCL, or I-joist
members acting as the flanges. Such assemblies distribute
C12.5.2.1 Reducing reference withdrawal design val- horizontal forces acting on the flanges to vertical resisting
ues for lag screws 25 percent when the screw is inserted in elements (103). Shear walls consisting of wall sheathing
the end grain (radial-tangential plane) of the main member materials attached to top and bottom plates and vertical
rather than the side grain (radial-longitudinal or tangential- framing members also are diaphragms. Such shear walls
longitudinal plane) is based on lag screw joint tests (100). or vertical diaphragms act to transfer loads from horizontal
Early tests of wood screws in withdrawal from end diaphragms down to the supporting foundation. The dia-
grain surfaces of oak, southern pine, maple, and cypress phragm factor, Cdi, applies to both horizontal and vertical
gave somewhat erratic results relative to those for with- diaphragms (144, 145).
drawal from side grain (43). These irregular results were
attributed to the tendency of the screw to split the wood C12.5.4 Toe-Nail Factor, Ctn
in the end grain configuration. Average ratios of end grain
withdrawal resistance to side grain withdrawal resistance C12.5.4.1 The 0.67 adjustment of reference with-
ranged from 52 to 108 percent (43). Because of this vari- drawal design values for toenailing is based on the results
ability, structural loading of wood screws in withdrawal of joint tests comparing slant driving and straight driving
from end grain has been prohibited. Where splitting is (184) and of typical toenailed and end nailed joints used in
avoided, use of an end grain to side grain withdrawal frame wall construction (118) where the attached member
design value ratio of 75 percent has been suggested (184, is pulled directly away from the main member. It is ap-
183). plicable to joints fabricated at all levels of seasoning. This
Reduction of withdrawal design values up to 50 percent includes multiple nail joints fabricated of unseasoned wood
have been reported for nails driven in end grain surfaces and then loaded after seasoning (184, 183, 118). Toenail-
(radial-tangential plane) as compared to side grain (radial- ing with cross slant driving can produce stronger joints
longitudinal or tangential-longitudinal planes) surfaces (184, than end or face nailing. For example, a stud to plate joint
Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
 NATIONAL DESIGN SPECIFICATION FOR WOOD CONSTRUCTION 275

C16 FIRE DESIGN OF WOOD

MEMBERS
C16.1 General
C
The design provisions in the Specification are intended or to prevent the spread of fire through them. Such ratings
for use in allowable stress design (ASD). These provisions are arrived at either by calculation or experiment for both

COMMENTARY: FIRE DESIGN OF WOOD MEMBERS

do not address procedures for rehabilitating a structure or members and assemblies. The standard fire exposure is
an assembly following fire damage. defined in ASTM E119. A 1‑hour fire‑resistance rating for
wall, floor, and floor‑ceiling assemblies incorporating nom-
Introduction to Fire Design inal 2" structural lumber can be accomplished through the
The model building codes in the U.S. cover virtually use of fire-resistive membranes such as gypsum wallboard.
every safety-related topic related to the construction of However, fastening of these surface materials is critical
buildings, and fire-related issues comprise a surprisingly for assembly performance and is carefully specified. For
large portion of the model codes. Designing for fire safety some wood assemblies, 2‑hour ratings have been achieved.
is a complex and multifaceted issue. The following infor- Experimental ratings are available for several generic
mation provides an overview of the subject. assemblies. Ratings for proprietary assemblies are typical-
To provide fire safety in any structure, many ap- ly supplied by the producers. Typically rated floor-ceiling
proaches are considered. This involves a combination of assemblies for various products are provided in AWC’s
(1) preventing fire occurrence by reducing potential igni- DCA No. 3–Fire Rated Wood Floor and Wall Assemblies,
tion sources, (2) controlling fire growth, and (3) providing available at www.awc.org.
protection to life and property. All need systematic atten-
tion to provide a high degree of economical fire safety. The Analytically Rated
building design professional can control fire growth within In lieu of experimentally rating the fire resistance of
the structure by generating plans that include features such members and assemblies, major building codes will accept
as protecting occupants, confining fire in compartment engineering calculations of the expected fire resistance,
areas, and incorporating fire suppression and smoke or based upon engineering principles and material properties.
heat venting devices at critical locations. This applies to the rating of previously untested members
Controlling construction features to facilitate rapid or assemblies, or in cases where it is desired to substitute
egress, protection of occupants in given areas, and pre- one material or component for another. Although cal-
venting fire growth or spread are regulated by codes as a culation procedures may be conservative, they have the
function of building occupancy. If the design professional advantage of quickly rating an assembly or member and
rationally blends protection solutions for these items with allowing interpolation or some extrapolation of expected
the potential use of a fire‑suppression system (sprinklers, performance. Additional details regarding the analytical
for example), economical fire protection can be achieved. approach are provided in AWC’s DCA No. 4 ‑ Component
Although attention could be given to all protection Additive Method (CAM) for Calculating and Demonstrat-
techniques available to the building design professional, ing Assembly Fire Resistance, available at www.awc.org.
this discussion is limited to the provisions that limit fire
growth and limit spread of fire to neighboring compart- Beams and Columns
ments or buildings. Heavy timber construction has traditionally been
recognized to provide a fire‑resistant building. This is
Fire-Rated Assemblies primarily due to the large size of the members, the con-
The previous section explained that some occupan- nection details, and the lack of concealed spaces. Such
cies require the use of fire‑rated assemblies or members a construction type has often satisfied the fire-resistive
to prevent collapse or fire spread from one compartment requirement in all building codes by simple prescription.
of a building to another or from one building to another. Although heavy timber construction has not been “rated”
Members and assemblies are rated for their ability ei- in the United States, Canada has assigned it a 45‑minute
ther to continue to carry design loads during fire exposure fire‑resistance rating.
Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
276 NDS COMMENTARY: FIRE DESIGN OF WOOD MEMBERS

Using calculations, exposed columns and beams of Flame Spread

sawn lumber, glued-laminated timber, laminated veneer Regulation of materials used on interior building
lumber, parallel strand lumber, and laminated strand surfaces (and sometimes exterior surfaces) of other than
lumber can be designed for desired fire-resistance ratings. one‑ and two‑family structures is provided to minimize
Additional details regarding the analytical approach are the danger of rapid flame spread. ASTM E84 gives the
provided in NDS Chapter 16 and AWC’s DCA No. 2–De- method used to obtain the flame‑spread property for regula-
sign of Fire-Resistive Exposed Wood Members, available tory purposes of paneling materials. The flame spread test
at www.awc.org. apparatus is calibrated such that red oak flooring samples
that have been equilibrated to specified conditions will
Fireblocking and Draftstopping yield a flame spread index of 100, and noncombustible
In all construction types, no greater emphasis can be fiber-cement board samples will yield a flame spread index
placed on the control of construction to reduce the fire of 0. A list of accredited flame‑spread ratings for various
growth hazard than the placement of fire and draft stops commercial woods and wood products is given in AWC’s
in concealed spaces. The spread of fire and smoke through DCA No. 1–Flame Spread Performance of Wood Products,
these concealed openings within large rooms or between available at www.awc.org.
rooms is a continuous cause of major life and property
loss. As a result, most building codes contain requirements Fire Retardant Treatments
for fireblocking and draftstopping within building plans. It is possible to make wood highly resistant to the
Fire-blocking considered acceptable are (1) two-inch spread of fire by pressure impregnating it with an approved
nominal lumber, (2) two thicknesses of one-inch nominal chemical formulation. Wood including wood treated with
lumber, and (3) two thicknesses of 3/4‑inch plywood, with a fire‑retardant chemical, will char if exposed to fire;
staggered joints. however, the most significant contribution of chemicals is
Draftstopping does not require fire resistance of fire- a reduction in the spread of fire. Wood that has absorbed
blocking. Therefore, draftstopping material is not required adequate amounts of a fire‑retardant chemical will cease
to be as thick. Typical draftstop materials and their mini- to burn when the source of ignition is removed.
mum thicknesses are (1) l/2‑inch gypsum wallboard and Two general methods of improving resistance of wood
(2) 3/8‑inch plywood. Building codes consider an area to fire are (1) impregnation with an effective chemical,
between draftstops of 1,000 square feet as reasonable. and (2) coating the surface with a layer of intumescent
Concealed spaces consisting of open-web floor truss paint. For interiors or locations protected from weather,
components in protected floor‑ceiling assemblies are an impregnation treatments can be considered permanent
important location to draftstop parallel to the component. and have considerable value in preventing ignition. Sur-
Areas of 500 square feet in single‑family dwellings and face applications offer the principal means of increasing
1,000 square feet in other buildings are recommended, fire‑retardant properties of existing structures. However,
and separation of areas between family compartments is surface applications may require periodic renewal if their
absolutely necessary. Critical draftstop locations are in the effectiveness is to be maintained. In the past, the only ef-
concealed spaces in floor‑ceiling assemblies and in attics fective chemicals were water soluble, making fire‑retardant
of multifamily dwellings when separation walls do not treatments inadaptable to weather exposure. Impregnated
extend to the roof sheathing above. fire retardants that are resistant to both high humidity and
Other important locations to fireblock in wood-frame exterior exposures are becoming increasingly available
construction are in the following concealed spaces: on the market for treated lumber and plywood products.
1. Stud walls and partitions at ceiling and floor In addition, factory-applied, fire-retardant coatings may
levels. be available for other products recognized within this
2. Intersections between concealed horizontal standard. See product-specific recommendations regarding
and vertical spaces such as soffits. proper treatments and specifications.
3. Top and bottom of stairs between stair string-
ers.
4. Openings around vents, pipes, ducts, chim-
neys (and fireplaces at ceiling and floor levels)
with noncombustible fire stops.

Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
 NATIONAL DESIGN SPECIFICATION FOR WOOD CONSTRUCTION 277

C16.2 Design Procedures for Exposed Wood Members

The mechanics-based design procedures in the Speci- model addressed accelerated charring which occurs early
fication for exposed wood members are based on research in the fire exposure by applying a power factor to the char
described in AWC’s Technical Report 10: Calculating the depth, xchar, to adjust for char rate nonlinearity:
Fire Resistance of Exposed Wood Members (136). The
design procedure calculates the capacity of exposed wood t = m(xchar)1.23 (C16.2-1)
members using basic wood engineering mechanics. Actual where: C
mechanical and physical properties of the wood are used
t = exposure time (min.)
and member capacity is directly calculated for a given
period of time. Section properties are computed assuming m = char slope (min./in.1.23)

COMMENTARY: FIRE DESIGN OF WOOD MEMBERS

an effective char rate, beff, at a given time, t. Reductions xchar = char depth (in.)
of strength and stiffness of wood in the heated zone ad-
jacent to the char layer are addressed by accelerating the However, application of this model is limited since the
char rate 20%. Average member strength properties are char slope (min./in.1.23), m, is species-specific and limited
approximated from existing accepted procedures used to data exists for different wood species fit to the model. In
calculate design properties. Finally, wood members are addition, the model is limited to one-dimensional slabs.
designed using accepted engineering procedures found in To develop a two-dimensional, nonlinear char rate
NDS for allowable stress design. The design procedures model, one-dimensional non-linear char rate model was
presented in Chapter 16 are not intended to be used for modified to enable values for the slope factor, m, to be
design and retrofit of a structure after a fire event. estimated using nominal char rate values (in./hr.), βn. The
nominal char rate values, βn, are calculated using measured
C16.2.1 Char Rate char depth at approximately one hour. Substituting and
solving for the char depth, xchar, in terms of time, t:
C16.2.1.1 To estimate the reduced cross-sectional di- xchar = βnt0.813 (C16.2-2)
mensions, the location of the char base must be determined
as a function of time on the basis of empirical charring rate To account for rounding at the corners and reduction
data. The char layer can be assumed to have zero strength of strength and stiffness of the heated zone, the nominal
and stiffness. The physical shape of the remaining section char rate value, βn, is increased 20% in NDS Equation
and its load-carrying capacity should be adjusted to ac- 16.2-1. For sawn timbers and structural glued laminated
count for rounding at the corners, and for loss of strength timbers, the nominal char rate, βn, has been assumed to be
and stiffness in the heated zone. In design there are vari- approximately 1.5 inches/hr. This assumption has been
ous documented approaches to account for these results: shown to provide good, repeatable results as discussed in
• additional reduction of the remaining section; Technical Report 10: Calculating the Fire Resistance of
• uniform reduction of the maximum strength Exposed Wood Members (136). Recent analysis and testing
and stiffness; or also reported in Technical Report 10 have verified that the
• more detailed analysis with subdivision of char model provides reliable estimates for sawn lumber,
the remaining section into several zones at structural composite lumber identified as laminated veneer
different temperatures. lumber, parallel strand lumber, laminated strand lumber,
Extensive char rate data is available for one-dimensional and cross-laminated timber.
wood slabs. Data is also available for two-dimensional C16.2.1.2 For sawn lumber and timbers, structural
timbers, but most of this data is limited to larger cross- glued-laminated timbers, laminated veneer lumber, parallel
sections. Evaluation of linear char rate models using strand lumber, and laminated strand lumber, the depth of
one-dimensional char rate data suggests that charring of char can be directly estimated using NDS Equation 16.2-1
wood is slightly nonlinear, and estimates using linear mod- and assuming a nominal char rate, βn, of 1.5 inches/hr. The
els tend to underestimate char depth for short time periods effective depth of char, achar, can be calculated as:
(< 60 minutes) and overestimate char depth for longer time
achar = 1.8t 0.813
periods (> 60 minutes). To account for char rate nonlinear-
ity, a nonlinear, one-dimensional char rate model based on C16.2.1.3 For cross-laminated timber, falloff of lami-
the results of 40 one-dimensional wood slab charring tests nations has been noted in some full-scale tests. The falloff
of various species was developed (154). This non-linear appears to occur as the char front reaches the glueline. To
Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
 278 NDS COMMENTARY: FIRE DESIGN OF WOOD MEMBERS

model this effect, the time required for the char front to h 
1.23

reach the glueline of each lamination, starting from the t gi =  lam 

time the char front reaches the prior lamination, can be  βn 
calculated as: and
t
h 
1.23
nlam =
t gi ,i =  lam ,i  t gi
 βn  Effective char depths, achar, for cross-laminated timber
where: with equal lamination depths have been calculated in NDS
tgi,i = time for char front to reach glued interface Table 16.2.1B.
for each lamination (hr.) C16.2.1.4 For sawn lumber and timbers, structural
glued-laminated timbers, laminated veneer lumber, par-
hlam = lamination thickness (in.)
allel strand lumber, and laminated strand lumber, section
properties can be calculated using standard equations for
The number of laminations that could potentially area, section modulus, and moment of inertia using reduced
falloff is estimated by subtracting each tgi from the total cross-sectional dimensions. The dimensions are reduced by
time until the last partial lamination is determined. The achar calculated per C16.2.1.2 for each surface exposed to
value of nlam is the maximum value in which the following fire. Cross-sectional properties for a member exposed on
equation is true: all four sides are shown in Table C16.2-1. Other exposures
nlam can be calculated using this method.
t − ∑t gi ,i ≥ 0 C16.2.1.5 For cross-laminated timbers, reduced sec-
i =1
tion properties are calculated using the effective char depth,
where: achar, calculated per C16.2.1.3; however, due to the propri-
nlam = number of laminations charred (rounded to etary nature of cross-laminated timber layups, the impact
lowest integer) of charring should be checked with the manufacturer.
The values of tgi,i and nlam determined in the above are Alternatively, the capacity of the cross-laminated timber
used to calculate the effective char depth, achar: can be conservatively determined by limiting the design
to a panel with the same number of full-depth laminations
remaining after a given fire resistance time.
 nlam  nlam  
0.813

=achar 1.2  ∑hlam ,i + βn  t − ∑t gi ,i   C16.2.2 Member Strength

=  i 1 =  i 1  
For cross-laminated timber manufactured with lamina- Generally, average unheated member strength can be
tions of equal thickness, calculation of the effective char approximated from tests or by using design stresses de-
depth, achar, can be simplified as follows: rived from actual member strength data. To approximate
an average member strength using a reference design
(
achar 1.2 ( nlam )( hlam ) + βn t − ( nlam ) ( t gi ) ) 
0.813
= value, the reference design value can be multiplied by an
  adjustment factor, K, to adjust from a 5% exclusion value
where: allowable design value to an average ultimate value. In the
2012 edition of the Specification, the sequence of required
adjustments was clarified to ensure that Fb* is multiplied by

Table C16.2-1 Cross-Sectional Properties for Four-Sided Exposure

Cross-sectional Property Four-Sided Example

Area of the cross section, in.
2
A(t) = (B – 2βeff t)(D – 2βeff t)
Section Modulus in the major-axis direction, in. 3
S(t) = (B – 2βeff t)(D – 2βeff t)2/6
Section Modulus in the minor-axis direction, in.3 S(t) = (B – 2βeff t)2(D – 2βeff t)/6
Moment of Inertia in the major-axis direction, in.4 I(t) = (Dmin – 2βeff t)(Dmax – 2βeff t)3/12
Moment of Inertia in the minor-axis direction, in.4 I(t) = (Dmin – 2βeff t)3(Dmax – 2βeff t)/12
Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
 NATIONAL DESIGN SPECIFICATION FOR WOOD CONSTRUCTION 279

the adjustment factor, K, prior to calculation of the beam

Figure C16A Typical Unbalanced Beam
stability factor, CL (Equation 3.3-6). Similarly, that Fc* is
Layup
multiplied by the adjustment factor, K, prior to calculation
of the column stability factor, CP (Equation 3.7-1).
The adjustment factor, K, has two components, the
inverse of the applicable design value adjustment factor,
1/k, and the inverse of the variability adjustment factor,
c. To develop general design procedures for solid-sawn
lumber, structural glued laminated timber, and structural C
composite lumber, the following design value adjustment
factors and estimates of COV were used to conservatively
develop an allowable design stress to average ultimate

COMMENTARY: FIRE DESIGN OF WOOD MEMBERS

strength adjustment factor, K, shown in Table C16.2-2.

C16.2.3 Design of Members

The induced stress cannot exceed the average member

capacity of a wood member exposed to fire for a given
time, t. The average member capacity can be estimated
using cross-sectional properties reduced for fire exposure
and average ultimate strength properties derived from
reference design values.
Figure C16B Typical Balanced Beam
C16.2.4 Special Provisions for Layup
Structural Glued Laminated Timber
Beams

The outer laminations of glued laminated timber

bending members in Table 5A of the NDS Supplement are
typically higher strength laminations. When the beam is
exposed to fire, these laminations are the first to be charred.
In order to maintain the ultimate capacity of the beam when
these laminations are completely charred, core laminations
should be replaced with the higher strength laminations
in the beam layup. For unbalanced beams, only the core
laminations adjacent to the tension side lamination need
to be replaced as shown in Figure C16A(b) and C16A(c),
respectively. For balanced beams, the core laminations
adjacent to the outer laminations on both sides need to
be replaced as shown in Figure C16B(b) and C16B(c),
respectively.

C16.2.5 Provisions for Timber

Decks

Sides of individual timber decking members are

shielded from full fire exposure by adjacent members col-
lectively acting as a joint. Partial exposure can occur as
members shrink and joints between members open. The
degree of exposure is a function of the view angle of the
Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
280 NDS COMMENTARY: FIRE DESIGN OF WOOD MEMBERS

Table C16.2-2 Allowable Design Stress to Average Ultimate Strength

Adjustment Factors

F 1/k c Assumed COV K

Bending Strength Fb 2.1 1
1 – 1.645 COVb 0.16 2
2.85
Tensile Strength Ft 2.1 1 1 – 1.645 COVt 0.16 2 2.85
Compression Strength Fc 1.9 1
1 – 1.645 COVc 0.16 2
2.58
Buckling Strength E05 1.66 3
1 – 1.645 COVE 0.11 3
2.03
1. Taken from Table 10 of ASTM D245 Standard Practice for Establishing Structural Grades and Related Allowable Properties for Visually Graded Lumber.
2. Taken from Table 4-6 of 1999 Wood Handbook.
3. Taken from Appendices D and H of National Design Specification for Wood Construction.

radiant flame and the ability of hot volatile gases to pass

through the joints. When the joint is completely open, such
as can occur with butt-jointed timber decking, hot gases will
carry into the joint and the sides of the decking members
will char. This charring can be conservatively approximated
assuming the sides of a member along the joint char at the
effective char rate. When the joint is open but covered by
sheathing, as with butt-jointed timber decking covered with
wood structural panels, passage of hot gases is limited, and
tests have shown that charring can be approximated as-
suming a partial exposure char rate along the joint equal to
one-third of the effective char rate. For joints which are not
open, as with tongue-and-groove timber decking, tests have
shown that charring of the sides of members is negligible
and can be ignored.

Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
 NATIONAL DESIGN SPECIFICATION FOR WOOD CONSTRUCTION 281

REFERENCES
1. Aplin, E. N., Factors Affecting the Stiffness and 12. ASTM Standard D 3200-74 (2012), Standard
Strength of Metal Plate Connector Joints, Informa- Specification and Test Method for Establishing
tion Report OP-X-57, Ottawa, Ontario, Department Recommended Design Stresses for Round Timber
of Environment, Canadian Forestry Service, Eastern Construction Poles, ASTM, West Conshohocken,
Forest Products Laboratory, 1973. PA, 2012. C
2. ANSI Standard A190.1-2012, Structural Glued 13. ASTM Standard D 3737-12, Standard Practice for
Laminated Timber, APA-The Engineered Wood As- Establishing Allowable Properties for Structural

COMMENTARY: REFERENCES
sociation, Tacoma, WA 2012. Glued Laminated Timber (Glulam), ASTM, West
Conshohocken, PA, 2012.
3. ASCE Standard 7-10, Minimum Design Loads for
Buildings and Other Structures, American Society 14. ASTM Standard D 4761-13, Standard Test Meth-
of Civil Engineers, Reston, VA, 2010. ods for Mechanical Properties of Lumber and
Wood-Based Structural Material, ASTM, West Con-
4. ASTM Standard D 2915-10, Standard Practice for shohocken, PA, 2013.
Sampling and Data Analysis for Structural Wood and
Wood Based Products, ASTM, West Conshohocken, 15. ASTM Standard D 5055-13, Standard Specification
PA, 2010. for Establishing and Monitoring Structural Capaci-
ties of Prefabricated Wood I-Joists, ASTM, West
5. ASTM Standard D 143-14, Standard Test Methods Conshohocken, PA, 2013.
for Small Clear Specimens of Timber, ASTM, West
Conshohocken, PA, 2014. 16. ASTM Standard D 5456-14, Standard Specification
for Evaluation of Structural Composite Lumber Prod-
6. ASTM Standard D 1760-01, Standard Specification ucts, ASTM, West Conshohocken, PA, 2014.
for Pressure Treatment of Timber Products (With-
drawn 2006), ASTM, West Conshohocken, PA, 2001. 17. ASTM Standard D 5457-12, Standard Specification
for Computing Reference Resistance of Wood-Based
7. ASTM Standard D 1990-14, Standard Practice for Es- Materials and Structural Connections for Load and
tablishing Allowable Properties for Visually Graded Resistance Factor Design, ASTM, West Conshohock-
Dimension Lumber from In-Grade Tests of Full-Size en, PA, 2012.
Specimens, ASTM, West Conshohocken, PA, 2014.
18. ASTM Standard D 5764-97a (2013), Standard Test
8. ASTM Standard D 245-06 (2011), Standard Practice Method for Evaluating Dowel-Bearing Strength of
for Establishing Structural Grades and Related Allow- Wood and Wood-Base Products, ASTM, West Con-
able Properties for Visually Graded Lumber, ASTM, shohocken, PA, 2013.
West Conshohocken, PA, 2011.
19. ASTM Standard D 6555-03 (2014), Standard Guide
9. ASTM Standard D 2555-06 (2011), Standard Practice for Evaluating System Effects in Repetitive-Member
for Establishing Clear Wood Strength Values, ASTM, Wood Assemblies, ASTM, West Conshohocken, PA,
West Conshohocken, PA, 2011. 2014.

10. ASTM Standard D 25-12, Standard Specification for 20. Auune, P. and M. Patton-Mallory, Lateral Load-
Round Timber Piles, ASTM, West Conshohocken, Bearing Capacity of Nailed Joints Based on the
PA, 2012. Yield Theory: Experimental Verification, Research
Paper FPL 470, Madison, WI, U.S. Department of
11. ASTM Standard D 2899-12, Standard Practice for Agriculture, Forest Service, Forest Products Labora-
Establishing Allowable Stresses for Round Timber tory, 1986.
Piles, ASTM, West Conshohocken, PA, 2012.
Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
 NATIONAL DESIGN SPECIFICATION FOR WOOD CONSTRUCTION 287

119. Snodgrass, J. D. and W. W. Gleaves, Effect of End- 130. Standard Specifications for Structural Glued Lami-
Distance on Strength of Single-Bolt Joints, Corvallis, nated Timber of Hardwood Species, AITC 119-96,
OR, Oregon State University, Oregon Forest Research American Institute of Timber Construction, Centen-
Center, 1960. nial, CO, 1996.

120. Soltis, L. A., European Yield Model for Wood Con- 131. Standard Specifications for Structural Glued Lami-
nections, New York, NY, American Society of Civil nated Timber of Softwood Species, AITC 117-2004,
Engineers, Proceedings of Structures Congress 1991, American Institute of Timber Construction, Centen-
Indianapolis, Indiana, 60-63, 1991. nial, CO, 2004. C
121. Soltis, L. A. and T. L. Wilkinson, Bolted Connec- 132. Stieda, C. K. A., Stress Concentrations in Notched
tion Design, General Research Report FPL-GRT-54, Timber Beams, Contribution No. P-49, Vancouver,

COMMENTARY: REFERENCES
Madison, WI, U.S. Department of Agriculture, Forest British Columbia, Department of Forestry of Canada,
Service, Forest Products Laboratory, 1987. Vancouver Laboratory, Forest Products Research
Branch, 1964.
122. Soltis, L. A. and T. L. Wilkinson, Timber Bolted Con-
nection Design, New York, NY, American Society of 133. Structural Glued Laminated Timber, Commercial
Civil Engineers, Proceedings of Structural Congress Standard CS 253-63, U.S. Department of Com-
1987, Orlando, Florida, 205-220, 1987. merce, National Bureau of Standards, Washington,
DC, 1963.
123. Soltis, L. A., W. Nelson, and J. L. Hills, Creep of
Structural Lumber, New York, NY, American Society 134. Structural Glued Laminated Timber, Voluntary
of Mechanical Engineers, Proceedings of 3rd Joint Product Standard PS 56-73, U.S. Department of Com-
ASCE/ASME Mechanics Conference, San Diego, merce, National Bureau of Standards, Washington,
216-221, 1989. DC, 1973.

124. Special Design Provisions for Wind and Seismic 135. Suddarth, S. K., Test Performance of 1-1/2 Inch Bolts
(SDPWS-05), American Wood Council, Leesburg, in Glulam–Row Effects and Effect of Subsequent
VA, 2005. Drying, Portland, OR, Wood Products Information
Center, Wood Design Focus, Components, Vol. 1,
125. Specification for Structural Steel Buildings (ANSI/ No.1, 1990.
AISC 360-10), American Institute of Steel Construc-
tion (AISC), Chicago, IL, 2010. 136. Technical Report 10–Calculating the Fire Resistance
of Exposed Wood Members, American Forest & Pa-
126. North American Standard for Cold-Formed Steel per Association, Washington, DC, 2015.
Framing, American Iron and Steel Institute (AISI),
Washington, DC, 2007. 137. Technical Report 12–General Dowel Equations for
Calculating Lateral Connection Values, American
127. Stamm, A. J., Wood and cellulose science, New York, Forest & Paper Association, Washington, DC, 2015.
NY, Ronald Press, 1964.
138. Technical Report 14–Designing for Lateral-Torsional
128. Standard Appearance Grades for Structural Glued Stability in Wood Members, American Forest & Paper
Laminated Timber, AITC 110-2001, American Insti- Association, Washington DC, 2003.
tute of Timber Construction, Centennial, CO, 2001.
139. Tiemann, H. D., Some Results of Dead Load Bending
129. Standard Specifications for Highway Bridges, Ameri- Tests by Means of a Recording Deflectometer, Phila-
can Association of State Highway and Transportation delphia, PA, Proceedings of the American Society for
Officials (AASHTO), Washington, DC, 2002. Testing Materials, Vol. 9: 534-548, 1909.

140. Timber Construction Manual, American Institute of

Timber Construction (AITC), John Wiley & Sons,
2012.
Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
290 NDS COMMENTARY: REFERENCES

179. Wolfe, R. W. and T. LaBissoniere, Structural Per- 190. Zahn, J. J., Interaction of Crushing and Buckling
formance of Light-Frame Roof Assemblies, II, in Wood Columns and Beams, Madison, WI, U.S.
Conventional Truss Assemblies, Research Paper Department of Agriculture, Forest Service, Forest
FPL-RP-499, Madison, WI, U.S. Department of Products Laboratory, 1990.
Agriculture, Forest Service, Forest Products Labora-
tory, 1991. 191. Zahn, J. J., Progress Report to NFPA on Column
Research at FPL, Madison, WI, U.S. Department of
180. Wood Construction Data 1 (WCD 1) Details for Con- Agriculture, Forest Service, Forest Products Labora-
ventional Wood Frame Construction, Washington, tory, 1989.
DC, American Forest & Paper Association, 2001.
192. Zahn, J. J., Proposed Design Formula for Wood
181. Wood Handbook, Agriculture Handbook No. 72, Beam Columns, Paper No. 87-4002, Baltimore, MD,
Washington, D.C, U.S. Department of Agriculture, Proceedings of American Society of Agricultural
Forest Service, Forest Products Laboratory, 1955. Engineers, 1987.

182. Wood Handbook, Agriculture Handbook No. 72, 193. Zahn, J. J., Strength of Lumber Under Combined
Washington, D.C, U.S. Department of Agriculture, Bending and Compression, Research Paper FPL 391,
Forest Service, Forest Products Laboratory, 1974. Madison, WI, U.S. Department of Agriculture, Forest
Service, Forest Products Laboratory, 1982.
183. Wood Handbook, Agriculture Handbook No. 72,
Washington, D.C, U.S. Department of Agriculture, 194. Zahn, J. J., Strength of Multiple-Member Structures,
Forest Service, Forest Products Laboratory, 1987. Research Paper FPL 139, Madison, WI, U.S. Depart-
ment of Agriculture, Forest Service, Forest Products
184. Wood Handbook, Washington, DC, U.S. Depart- Laboratory, 1970.
ment of Agriculture, Forest Service, Forest Products
Laboratory, 1935. 195. Wilkinson, T. L., Moisture Cycling of Trussed Rafter
Joints, Research Paper FPL 67, Madison, WI, U.S.
185. Wood, L. W., Behavior of Wood Under Continued Department of Agriculture, Forest Service, Forest
Loading, New York, NY, Engineering News Record, Products Laboratory, 1966.
Vol. 139, No. 24: 108-111, 1947.
196. Rammer, D.R., Withdrawal Strength Design Expres-
186. Wood, L. W., Formulas for Columns with Side Loads sion of Bright Annularly Threaded Nails Conforming
and Eccentricity, Report No. R1782, Madison, WI, to ASTM F1667 Table 45, Madison, WI, U.S. Depart-
U.S. Department of Agriculture, Forest Service, For- ment of Agriculture, Forest Service, Forest Products
est Products Laboratory, 1950. Laboratory, 2010.

187. Zahn, J. J., Combined-Load Stability Criterion for 197. ANSI/APA PRG 320-2011, Standard for Perfor-
Wood Beam-Columns, New York, NY, American mance-Rated Cross-Laminated Timber, APA-The
Society of Civil Engineers, Journal of Structural Engineered Wood Association, Tacoma, WA, 2011.
Engineering, Vol. 114, No. 11: 2612-2618, 1986.
198. Uibel, T and H. J. Blass. Edge Joints with Dowel
188. Zahn, J. J., Design Equation for Multiple-Fastener Type Fasteners in Cross Laminated Timber. Pro-
Wood Connections, Madison, WI, U.S. Department ceedings of International Council for Research and
of Agriculture, Forest Service, Forest Products Labo- Innovation in Building Construction, Working Com-
ratory, 1991. mission W18 – Timber Structures, Bled, Slovenia,
August 2007.
189. Zahn, J. J., Design of Wood Members Under Com-
bined Load, New York, NY, American Society of
Civil Engineers, Journal of Structural Engineering,
Vol. 112, No. 9: 2109-2126, 1986.

Copyright © American Wood Council. Downloaded/printed pursuant to License Agreement. No reproduction or transfer authorized.
AMERICAN WOOD COUNCIL
 August 2015

ERRATA
to the 2015 Edition of
the National Design Specification® (NDS®) for Wood Construction
(web version dated 11-14)

Page Revision
62 Revise Table 10.4.1.1 as follows:

Table 10.4.1.1 Shear Deformation

Adjustment Factors, KS

Loading End Fixity KS

Pinned 11.5
Uniformly Distributed
Fixed 57.6
Pinned 14.4
Line Load at midspan
Fixed 57.6
Line Load at quarter points Pinned 10.5

Constant Moment Pinned 11.8 0

Uniformly Distributed Cantilevered 4.8

Line Load at free-end Cantilevered 3.6

Pinned 11.8
Column Buckling
Fixed 23.7

152 Revise the first sentence of Section 16.2.1.3 as follows:

16.2.1.3 For cross-laminated timber manufactured with laminations of equal

thickness, the effective char depth, achar, shall be calculated as follows:

222 Catoctin Circle, SE, Suite 201 ▪ Leesburg, VA 20175 ▪ 202-463-2766 ▪ www.awc.org ▪ [email protected]